diff --git "a/train_job_output.txt" "b/train_job_output.txt"
--- "a/train_job_output.txt"
+++ "b/train_job_output.txt"
@@ -1,4 +1,4 @@
-slurm submission log: 2024-05-08 15:15:19.565748
+slurm submission log: 2024-05-19 09:14:41.566584
 created following sbatch script: 
 
 ###############################
@@ -7,24 +7,23 @@ created following sbatch script:
 
 #SBATCH --account=nlp
 #SBATCH --cpus-per-task=16
-#SBATCH --dependency=afterok:7590695
-#SBATCH --gres=gpu:2
-#SBATCH --job-name=tthrush-job-1731429
-#SBATCH --mem=400G
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-2782455
+#SBATCH --mem=100G
 #SBATCH --nodelist=sphinx2
 #SBATCH --open-mode=append
-#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa/train_job_output.txt
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/train_job_output.txt
 #SBATCH --partition=sphinx
 #SBATCH --time=14-0
 
 # activate your desired anaconda environment
-. /nlp/scr/tthrush/miniconda3/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
 
 # cd to working directory
 cd .
 
 # launch commands
-srun --unbuffered run_as_child_processes 'torchrun --master_port 29506 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/training_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa --output_hub_id pythia-14m_piqa --model_id /pythia-14m --num_train_epochs 1 --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2'
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29517 --nproc_per_node=1 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_more_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1 --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --learning_rate 6e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 1 --per_device_train_batch_size 1024 --seed 1 --num_train_epochs 0.2'
 
 ###############################
 
@@ -34,122 +33,136 @@ submission to slurm complete!
 ###############################
 slurm submission output
 
-Submitted batch job 7590696
+Submitted batch job 7631089
 
 
 
 ###############################
 
+slurm submission log: 2024-05-19 09:16:20.758126
+created following sbatch script: 
+
 ###############################
-start time: 2024-05-08 16:43:55.476578
-machine: sphinx2
-conda env: pretraining-coreset-selection
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-914626
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29517 --nproc_per_node=1 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_more_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1 --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --learning_rate 6e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 1 --per_device_train_batch_size 1024 --seed 1 --num_train_epochs 0.2'
+
 ###############################
-running following processes
 
-	torchrun --master_port 29506 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/training_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa --output_hub_id pythia-14m_piqa --model_id /pythia-14m --num_train_epochs 1 --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2
+submission to slurm complete!
 
 
 ###############################
-command outputs: 
+slurm submission output
+
+Submitted batch job 7631150
+
+
+
+###############################
+
+slurm submission log: 2024-05-19 09:25:08.150536
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-758301
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29517 --nproc_per_node=1 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_more_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1 --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --learning_rate 6e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 1 --per_device_train_batch_size 768 --seed 1 --num_train_epochs 0.2'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7631221
+
+
+
+###############################
+
+slurm submission log: 2024-05-19 09:27:21.164411
+created following sbatch script: 
+
+###############################
 
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-4168844
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29517 --nproc_per_node=1 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_more_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1 --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --learning_rate 6e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 1 --per_device_train_batch_size 640 --seed 1 --num_train_epochs 0.2'
+
+###############################
 
-[2024-05-08 16:43:59,360] torch.distributed.run: [WARNING] 
-[2024-05-08 16:43:59,360] torch.distributed.run: [WARNING] *****************************************
-[2024-05-08 16:43:59,360] torch.distributed.run: [WARNING] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 
-[2024-05-08 16:43:59,360] torch.distributed.run: [WARNING] *****************************************
-05/08/2024 16:44:05 - INFO - __main__ - Script parameters ScriptArguments(dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/training_data/piqa', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa', output_hub_id='pythia-14m_piqa', hf_hub_token=True, model_id='/pythia-14m', per_device_train_batch_size=256, num_train_epochs=1, learning_rate=0.001, gradient_accumulation_steps=2, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
-05/08/2024 16:44:05 - INFO - __main__ - Script parameters ScriptArguments(dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/training_data/piqa', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa', output_hub_id='pythia-14m_piqa', hf_hub_token=True, model_id='/pythia-14m', per_device_train_batch_size=256, num_train_epochs=1, learning_rate=0.001, gradient_accumulation_steps=2, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
-Traceback (most recent call last):
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/utils/hub.py", line 398, in cached_file
-    resolved_file = hf_hub_download(
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/huggingface_hub/utils/_validators.py", line 110, in _inner_fn
-    validate_repo_id(arg_value)
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/huggingface_hub/utils/_validators.py", line 164, in validate_repo_id
-    raise HFValidationError(
-huggingface_hub.utils._validators.HFValidationError: Repo id must use alphanumeric chars or '-', '_', '.', '--' and '..' are forbidden, '-' and '.' cannot start or end the name, max length is 96: '/pythia-14m'.
-
-The above exception was the direct cause of the following exception:
-
-Traceback (most recent call last):
-  File "/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_llm.py", line 202, in <module>
-    train_model()
-  File "/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_llm.py", line 166, in train_model
-    config = AutoConfig.from_pretrained(script_args.model_id)
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py", line 1138, in from_pretrained
-    config_dict, unused_kwargs = PretrainedConfig.get_config_dict(pretrained_model_name_or_path, **kwargs)
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/configuration_utils.py", line 631, in get_config_dict
-    config_dict, kwargs = cls._get_config_dict(pretrained_model_name_or_path, **kwargs)
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/configuration_utils.py", line 686, in _get_config_dict
-    resolved_config_file = cached_file(
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/utils/hub.py", line 462, in cached_file
-Traceback (most recent call last):
-      File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/utils/hub.py", line 398, in cached_file
-raise EnvironmentError(
-OSError: Incorrect path_or_model_id: '/pythia-14m'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
-    resolved_file = hf_hub_download(
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/huggingface_hub/utils/_validators.py", line 110, in _inner_fn
-    validate_repo_id(arg_value)
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/huggingface_hub/utils/_validators.py", line 164, in validate_repo_id
-    raise HFValidationError(
-huggingface_hub.utils._validators.HFValidationError: Repo id must use alphanumeric chars or '-', '_', '.', '--' and '..' are forbidden, '-' and '.' cannot start or end the name, max length is 96: '/pythia-14m'.
-
-The above exception was the direct cause of the following exception:
-
-Traceback (most recent call last):
-  File "/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_llm.py", line 202, in <module>
-    train_model()
-  File "/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_llm.py", line 166, in train_model
-    config = AutoConfig.from_pretrained(script_args.model_id)
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/models/auto/configuration_auto.py", line 1138, in from_pretrained
-    config_dict, unused_kwargs = PretrainedConfig.get_config_dict(pretrained_model_name_or_path, **kwargs)
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/configuration_utils.py", line 631, in get_config_dict
-    config_dict, kwargs = cls._get_config_dict(pretrained_model_name_or_path, **kwargs)
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/configuration_utils.py", line 686, in _get_config_dict
-    resolved_config_file = cached_file(
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/transformers/utils/hub.py", line 462, in cached_file
-    raise EnvironmentError(
-OSError: Incorrect path_or_model_id: '/pythia-14m'. Please provide either the path to a local folder or the repo_id of a model on the Hub.
-[2024-05-08 16:45:49,480] torch.distributed.elastic.multiprocessing.api: [ERROR] failed (exitcode: 1) local_rank: 0 (pid: 2312304) of binary: /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/bin/python
-Traceback (most recent call last):
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/bin/torchrun", line 8, in <module>
-    sys.exit(main())
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 347, in wrapper
-    return f(*args, **kwargs)
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/torch/distributed/run.py", line 812, in main
-    run(args)
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/torch/distributed/run.py", line 803, in run
-    elastic_launch(
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 135, in __call__
-    return launch_agent(self._config, self._entrypoint, list(args))
-  File "/nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/lib/python3.10/site-packages/torch/distributed/launcher/api.py", line 268, in launch_agent
-    raise ChildFailedError(
-torch.distributed.elastic.multiprocessing.errors.ChildFailedError: 
-============================================================
-train_llm.py FAILED
-------------------------------------------------------------
-Failures:
-[1]:
-  time      : 2024-05-08_16:45:49
-  host      : sphinx2.stanford.edu
-  rank      : 1 (local_rank: 1)
-  exitcode  : 1 (pid: 2312305)
-  error_file: <N/A>
-  traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
-------------------------------------------------------------
-Root Cause (first observed failure):
-[0]:
-  time      : 2024-05-08_16:45:49
-  host      : sphinx2.stanford.edu
-  rank      : 0 (local_rank: 0)
-  exitcode  : 1 (pid: 2312304)
-  error_file: <N/A>
-  traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
-============================================================
-###############################
-end time: 2024-05-08 16:45:55.593886
-elapsed time: 0:02:00.117308
-slurm submission log: 2024-05-09 07:34:40.871568
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7631284
+
+
+
+###############################
+
+slurm submission log: 2024-05-19 09:28:17.764747
 created following sbatch script: 
 
 ###############################
@@ -158,23 +171,23 @@ created following sbatch script:
 
 #SBATCH --account=nlp
 #SBATCH --cpus-per-task=16
-#SBATCH --gres=gpu:2
-#SBATCH --job-name=tthrush-job-4443020
-#SBATCH --mem=400G
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-3776623
+#SBATCH --mem=100G
 #SBATCH --nodelist=sphinx2
 #SBATCH --open-mode=append
-#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa/train_job_output.txt
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/train_job_output.txt
 #SBATCH --partition=sphinx
 #SBATCH --time=14-0
 
 # activate your desired anaconda environment
-. /nlp/scr/tthrush/miniconda3/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
 
 # cd to working directory
 cd .
 
 # launch commands
-srun --unbuffered run_as_child_processes 'torchrun --master_port 29506 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/training_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --num_train_epochs 1 --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2'
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29517 --nproc_per_node=1 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_more_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1 --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --learning_rate 6e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 1 --per_device_train_batch_size 512 --seed 1 --num_train_epochs 0.2'
 
 ###############################
 
@@ -184,461 +197,525 @@ submission to slurm complete!
 ###############################
 slurm submission output
 
-Submitted batch job 7591656
+Submitted batch job 7631348
+
+
 
+###############################
 
+slurm submission log: 2024-05-19 09:29:20.389708
+created following sbatch script: 
 
 ###############################
 
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-4689387
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29517 --nproc_per_node=1 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_more_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1 --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --learning_rate 6e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 1 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 0.2'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7631409
+
+
+
+###############################
+
+slurm submission log: 2024-05-19 09:30:33.987595
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-1409388
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29517 --nproc_per_node=1 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_more_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1 --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --learning_rate 6e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 1 --per_device_train_batch_size 340 --seed 1 --num_train_epochs 0.2'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7631469
+
+
+
+###############################
+
+slurm submission log: 2024-05-19 09:31:39.572545
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-1319359
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29517 --nproc_per_node=1 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_more_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1 --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --learning_rate 6e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 1 --per_device_train_batch_size 320 --seed 1 --num_train_epochs 0.2'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7631529
+
+
+
+###############################
+
+slurm submission log: 2024-05-19 09:34:33.193118
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-742694
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29517 --nproc_per_node=1 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_more_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1 --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --learning_rate 6e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 1 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 0.2'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7631593
+
+
+
+###############################
+
+slurm submission log: 2024-05-19 09:45:20.459710
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-3186286
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29517 --nproc_per_node=1 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_more_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1 --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --learning_rate 6e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 1 --per_device_train_batch_size 320 --seed 1 --num_train_epochs 0.2'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7631663
+
+
+
+###############################
+
+/var/lib/slurm/slurmd/job7631663/slurm_script: line 15: /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh: No such file or directory
+
+CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'.
+To initialize your shell, run
+
+    $ conda init <SHELL_NAME>
+
+Currently supported shells are:
+  - bash
+  - fish
+  - tcsh
+  - xonsh
+  - zsh
+  - powershell
+
+See 'conda init --help' for more information and options.
+
+IMPORTANT: You may need to close and restart your shell after running 'conda init'.
+
+
 ###############################
-start time: 2024-05-09 13:07:36.318275
+start time: 2024-05-19 10:14:28.271980
 machine: sphinx2
 conda env: pretraining-coreset-selection
 ###############################
 running following processes
 
-	torchrun --master_port 29506 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/training_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --num_train_epochs 1 --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2
+	torchrun --master_port 29517 --nproc_per_node=1 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_more_data/piqa --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1 --output_hub_id pythia-14m_piqa --model_id EleutherAI/pythia-14m --learning_rate 6e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 1 --per_device_train_batch_size 320 --seed 1 --num_train_epochs 0.2
 
 
 ###############################
 command outputs: 
 
 
-[2024-05-09 13:07:41,155] torch.distributed.run: [WARNING] 
-[2024-05-09 13:07:41,155] torch.distributed.run: [WARNING] *****************************************
-[2024-05-09 13:07:41,155] torch.distributed.run: [WARNING] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 
-[2024-05-09 13:07:41,155] torch.distributed.run: [WARNING] *****************************************
-05/09/2024 13:07:50 - INFO - __main__ - Script parameters ScriptArguments(dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/training_data/piqa', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa', output_hub_id='pythia-14m_piqa', hf_hub_token=True, model_id='EleutherAI/pythia-14m', per_device_train_batch_size=256, num_train_epochs=1, learning_rate=0.001, gradient_accumulation_steps=2, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
-05/09/2024 13:07:50 - INFO - __main__ - Script parameters ScriptArguments(dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/training_data/piqa', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms/pythia-14m_piqa', output_hub_id='pythia-14m_piqa', hf_hub_token=True, model_id='EleutherAI/pythia-14m', per_device_train_batch_size=256, num_train_epochs=1, learning_rate=0.001, gradient_accumulation_steps=2, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
-
  0%|          | 0/10682 [00:00<?, ?it/s][rank1]:[W reducer.cpp:1360] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
-[rank0]:[W reducer.cpp:1360] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
-
  0%|          | 1/10682 [01:14<222:22:40, 74.95s/it]
  0%|          | 2/10682 [01:55<162:29:44, 54.77s/it]
  0%|          | 3/10682 [02:37<144:59:33, 48.88s/it]
  0%|          | 4/10682 [03:23<141:29:28, 47.70s/it]
  0%|          | 5/10682 [03:58<127:50:08, 43.10s/it]
  0%|          | 6/10682 [04:30<117:16:41, 39.55s/it]
  0%|          | 7/10682 [04:58<105:29:33, 35.58s/it]
  0%|          | 8/10682 [05:24<96:38:47, 32.60s/it] 
  0%|          | 9/10682 [05:47<87:21:57, 29.47s/it]
  0%|          | 10/10682 [06:14<85:23:07, 28.80s/it]
  0%|          | 11/10682 [06:41<83:32:23, 28.18s/it]
  0%|          | 12/10682 [07:10<84:14:40, 28.42s/it]
  0%|          | 13/10682 [07:31<78:10:31, 26.38s/it]
  0%|          | 14/10682 [07:53<73:27:55, 24.79s/it]
  0%|          | 15/10682 [08:10<66:38:18, 22.49s/it]
  0%|          | 16/10682 [08:25<60:33:46, 20.44s/it]
  0%|          | 17/10682 [08:40<55:15:36, 18.65s/it]
  0%|          | 18/10682 [08:57<53:36:08, 18.10s/it]
  0%|          | 19/10682 [09:12<50:46:46, 17.14s/it]
  0%|          | 20/10682 [09:26<48:32:26, 16.39s/it]
  0%|          | 21/10682 [09:39<45:36:18, 15.40s/it]
  0%|          | 22/10682 [09:50<41:40:46, 14.08s/it]
  0%|          | 23/10682 [10:02<39:51:48, 13.46s/it]
  0%|          | 24/10682 [10:15<39:17:39, 13.27s/it]
  0%|          | 25/10682 [10:30<40:16:20, 13.60s/it]
                                                     
{'loss': 10.8095, 'grad_norm': 0.9220895767211914, 'learning_rate': 2.3386342376052384e-05, 'epoch': 0.0}
-
  0%|          | 25/10682 [10:30<40:16:20, 13.60s/it]
  0%|          | 26/10682 [10:41<38:31:52, 13.02s/it]
  0%|          | 27/10682 [10:50<34:41:56, 11.72s/it]
  0%|          | 28/10682 [11:01<33:45:12, 11.41s/it]
  0%|          | 29/10682 [11:11<32:27:47, 10.97s/it]
  0%|          | 30/10682 [11:21<32:08:32, 10.86s/it]
  0%|          | 31/10682 [11:30<30:21:01, 10.26s/it]
  0%|          | 32/10682 [11:37<27:22:42,  9.25s/it]
  0%|          | 33/10682 [11:43<24:49:22,  8.39s/it]
  0%|          | 34/10682 [11:50<23:08:00,  7.82s/it]
  0%|          | 35/10682 [11:56<21:23:23,  7.23s/it]
  0%|          | 36/10682 [12:02<20:34:49,  6.96s/it]
  0%|          | 37/10682 [12:08<20:00:47,  6.77s/it]
  0%|          | 38/10682 [12:15<19:49:37,  6.71s/it]
  0%|          | 39/10682 [12:20<18:43:47,  6.34s/it]
  0%|          | 40/10682 [12:24<16:36:04,  5.62s/it]
  0%|          | 41/10682 [12:29<15:58:30,  5.40s/it]
  0%|          | 42/10682 [12:35<16:14:58,  5.50s/it]
  0%|          | 43/10682 [12:41<16:32:25,  5.60s/it]
  0%|          | 44/10682 [12:46<16:22:21,  5.54s/it]
  0%|          | 45/10682 [12:51<15:47:04,  5.34s/it]
  0%|          | 46/10682 [12:56<15:07:02,  5.12s/it]
  0%|          | 47/10682 [13:00<14:56:46,  5.06s/it]
  0%|          | 48/10682 [13:05<14:35:46,  4.94s/it]
  0%|          | 49/10682 [13:09<13:14:29,  4.48s/it]
  0%|          | 50/10682 [13:13<13:01:15,  4.41s/it]
                                                     
{'loss': 10.6257, 'grad_norm': 0.7397403717041016, 'learning_rate': 4.677268475210477e-05, 'epoch': 0.0}
-
  0%|          | 50/10682 [13:13<13:01:15,  4.41s/it]
  0%|          | 51/10682 [13:16<11:58:10,  4.05s/it]
  0%|          | 52/10682 [13:20<11:49:34,  4.01s/it]
  0%|          | 53/10682 [13:24<11:47:16,  3.99s/it]
  1%|          | 54/10682 [13:28<12:19:25,  4.17s/it]
  1%|          | 55/10682 [13:32<12:05:34,  4.10s/it]
  1%|          | 56/10682 [13:36<11:37:51,  3.94s/it]
  1%|          | 57/10682 [13:39<10:56:23,  3.71s/it]
  1%|          | 58/10682 [13:42<10:16:08,  3.48s/it]
  1%|          | 59/10682 [13:46<10:37:55,  3.60s/it]
  1%|          | 60/10682 [13:49<10:22:39,  3.52s/it]
  1%|          | 61/10682 [13:52<9:35:26,  3.25s/it] 
  1%|          | 62/10682 [13:56<10:06:34,  3.43s/it]
  1%|          | 63/10682 [14:00<10:26:19,  3.54s/it]
  1%|          | 64/10682 [14:03<10:03:07,  3.41s/it]
  1%|          | 65/10682 [14:06<9:46:16,  3.31s/it] 
  1%|          | 66/10682 [14:08<9:14:04,  3.13s/it]
  1%|          | 67/10682 [14:11<9:08:14,  3.10s/it]
  1%|          | 68/10682 [14:14<8:51:47,  3.01s/it]
  1%|          | 69/10682 [14:17<8:46:01,  2.97s/it]
  1%|          | 70/10682 [14:20<8:22:32,  2.84s/it]
  1%|          | 71/10682 [14:22<8:15:43,  2.80s/it]
  1%|          | 72/10682 [14:24<7:35:08,  2.57s/it]
  1%|          | 73/10682 [14:27<7:16:49,  2.47s/it]
  1%|          | 74/10682 [14:29<7:12:32,  2.45s/it]
  1%|          | 75/10682 [14:31<7:07:33,  2.42s/it]{'loss': 10.418, 'grad_norm': 0.7278160452842712, 'learning_rate': 7.015902712815715e-05, 'epoch': 0.01}
                                                    
-
  1%|          | 75/10682 [14:31<7:07:33,  2.42s/it]
  1%|          | 76/10682 [14:33<6:45:38,  2.29s/it]
  1%|          | 77/10682 [14:36<6:33:49,  2.23s/it]
  1%|          | 78/10682 [14:38<6:39:46,  2.26s/it]
  1%|          | 79/10682 [14:40<6:28:47,  2.20s/it]
  1%|          | 80/10682 [14:42<6:48:42,  2.31s/it]
  1%|          | 81/10682 [14:45<6:46:17,  2.30s/it]
  1%|          | 82/10682 [14:47<6:24:10,  2.17s/it]
  1%|          | 83/10682 [14:48<5:59:51,  2.04s/it]
  1%|          | 84/10682 [14:50<5:48:18,  1.97s/it]
  1%|          | 85/10682 [14:52<5:29:34,  1.87s/it]
  1%|          | 86/10682 [14:53<5:20:07,  1.81s/it]
  1%|          | 87/10682 [14:55<5:04:23,  1.72s/it]
  1%|          | 88/10682 [14:57<5:27:44,  1.86s/it]
  1%|          | 89/10682 [14:59<5:48:11,  1.97s/it]
  1%|          | 90/10682 [15:01<5:34:22,  1.89s/it]
  1%|          | 91/10682 [15:03<5:45:26,  1.96s/it]
  1%|          | 92/10682 [15:05<5:32:43,  1.89s/it]
  1%|          | 93/10682 [15:06<5:14:50,  1.78s/it]
  1%|          | 94/10682 [15:08<4:52:49,  1.66s/it]
  1%|          | 95/10682 [15:09<4:32:55,  1.55s/it]
  1%|          | 96/10682 [15:11<4:29:05,  1.53s/it]
  1%|          | 97/10682 [15:12<4:24:19,  1.50s/it]
  1%|          | 98/10682 [15:13<4:20:57,  1.48s/it]
  1%|          | 99/10682 [15:15<4:11:00,  1.42s/it]
  1%|          | 100/10682 [15:16<4:09:08,  1.41s/it]{'loss': 10.137, 'grad_norm': 0.7353196144104004, 'learning_rate': 9.354536950420954e-05, 'epoch': 0.01}
                                                     
-
  1%|          | 100/10682 [15:16<4:09:08,  1.41s/it]
  1%|          | 101/10682 [15:18<4:10:48,  1.42s/it]
  1%|          | 102/10682 [15:19<4:02:46,  1.38s/it]
  1%|          | 103/10682 [15:20<3:46:05,  1.28s/it]
  1%|          | 104/10682 [15:21<3:45:52,  1.28s/it]
  1%|          | 105/10682 [15:22<3:43:47,  1.27s/it]
  1%|          | 106/10682 [15:24<3:36:50,  1.23s/it]
  1%|          | 107/10682 [15:25<3:31:23,  1.20s/it]
  1%|          | 108/10682 [15:26<3:35:20,  1.22s/it]
  1%|          | 109/10682 [15:27<3:35:03,  1.22s/it]
  1%|          | 110/10682 [15:29<3:43:16,  1.27s/it]
  1%|          | 111/10682 [15:30<3:50:54,  1.31s/it]
  1%|          | 112/10682 [15:31<3:56:02,  1.34s/it]
  1%|          | 113/10682 [15:32<3:42:59,  1.27s/it]
  1%|          | 114/10682 [15:34<3:47:51,  1.29s/it]
  1%|          | 115/10682 [15:35<3:39:12,  1.24s/it]
  1%|          | 116/10682 [15:36<3:45:26,  1.28s/it]
  1%|          | 117/10682 [15:37<3:35:34,  1.22s/it]
  1%|          | 118/10682 [15:38<3:22:07,  1.15s/it]
  1%|          | 119/10682 [15:39<3:09:00,  1.07s/it]
  1%|          | 120/10682 [15:40<3:13:47,  1.10s/it]
  1%|          | 121/10682 [15:42<3:22:24,  1.15s/it]
  1%|          | 122/10682 [15:43<3:07:56,  1.07s/it]
  1%|          | 123/10682 [15:44<3:00:50,  1.03s/it]
  1%|          | 124/10682 [15:45<2:59:32,  1.02s/it]
  1%|          | 125/10682 [15:45<2:53:03,  1.02it/s]
                                                     {'loss': 9.7938, 'grad_norm': 0.7511488795280457, 'learning_rate': 0.00011693171188026193, 'epoch': 0.01}
-
  1%|          | 125/10682 [15:45<2:53:03,  1.02it/s]
  1%|          | 126/10682 [15:46<2:44:55,  1.07it/s]
  1%|          | 127/10682 [15:47<2:48:22,  1.04it/s]
  1%|          | 128/10682 [15:48<2:43:29,  1.08it/s]
  1%|          | 129/10682 [15:49<2:45:46,  1.06it/s]
  1%|          | 130/10682 [15:50<2:56:52,  1.01s/it]
  1%|          | 131/10682 [15:51<2:47:44,  1.05it/s]
  1%|          | 132/10682 [15:52<2:48:37,  1.04it/s]
  1%|          | 133/10682 [15:53<2:45:46,  1.06it/s]
  1%|▏         | 134/10682 [15:54<2:41:54,  1.09it/s]
  1%|▏         | 135/10682 [15:55<2:37:29,  1.12it/s]
  1%|▏         | 136/10682 [15:56<2:33:38,  1.14it/s]
  1%|▏         | 137/10682 [15:56<2:31:46,  1.16it/s]
  1%|▏         | 138/10682 [15:57<2:31:10,  1.16it/s]
  1%|▏         | 139/10682 [15:58<2:27:55,  1.19it/s]
  1%|▏         | 140/10682 [15:59<2:30:41,  1.17it/s]
  1%|▏         | 141/10682 [16:00<2:33:09,  1.15it/s]
  1%|▏         | 142/10682 [16:01<2:26:39,  1.20it/s]
  1%|▏         | 143/10682 [16:01<2:32:24,  1.15it/s]
  1%|▏         | 144/10682 [16:02<2:36:41,  1.12it/s]
  1%|▏         | 145/10682 [16:03<2:32:01,  1.16it/s]
  1%|▏         | 146/10682 [16:04<2:33:41,  1.14it/s]
  1%|▏         | 147/10682 [16:05<2:33:07,  1.15it/s]
  1%|▏         | 148/10682 [16:06<2:31:14,  1.16it/s]
  1%|▏         | 149/10682 [16:07<2:29:54,  1.17it/s]
  1%|▏         | 150/10682 [16:08<2:28:33,  1.18it/s]{'loss': 9.416, 'grad_norm': 0.7498224377632141, 'learning_rate': 0.0001403180542563143, 'epoch': 0.01}
                                                     
-
  1%|▏         | 150/10682 [16:08<2:28:33,  1.18it/s]
  1%|▏         | 151/10682 [16:09<2:43:45,  1.07it/s]
  1%|▏         | 152/10682 [16:10<2:45:12,  1.06it/s]
  1%|▏         | 153/10682 [16:11<2:43:42,  1.07it/s]
  1%|▏         | 154/10682 [16:11<2:39:28,  1.10it/s]
  1%|▏         | 155/10682 [16:12<2:34:15,  1.14it/s]
  1%|▏         | 156/10682 [16:13<2:27:50,  1.19it/s]
  1%|▏         | 157/10682 [16:14<2:19:58,  1.25it/s]
  1%|▏         | 158/10682 [16:14<2:13:05,  1.32it/s]
  1%|▏         | 159/10682 [16:15<2:11:26,  1.33it/s]
  1%|▏         | 160/10682 [16:16<2:08:01,  1.37it/s]
  2%|▏         | 161/10682 [16:16<2:09:41,  1.35it/s]
  2%|▏         | 162/10682 [16:17<2:10:32,  1.34it/s]
  2%|▏         | 163/10682 [16:18<2:08:37,  1.36it/s]
  2%|▏         | 164/10682 [16:19<2:10:34,  1.34it/s]
  2%|▏         | 165/10682 [16:20<2:19:08,  1.26it/s]
  2%|▏         | 166/10682 [16:20<2:09:29,  1.35it/s]
  2%|▏         | 167/10682 [16:21<2:09:57,  1.35it/s]
  2%|▏         | 168/10682 [16:22<2:08:01,  1.37it/s]
  2%|▏         | 169/10682 [16:22<2:04:47,  1.40it/s]
  2%|▏         | 170/10682 [16:23<1:57:50,  1.49it/s]
  2%|▏         | 171/10682 [16:24<1:53:40,  1.54it/s]
  2%|▏         | 172/10682 [16:24<1:58:23,  1.48it/s]
  2%|▏         | 173/10682 [16:25<1:52:52,  1.55it/s]
  2%|▏         | 174/10682 [16:25<1:52:27,  1.56it/s]
  2%|▏         | 175/10682 [16:26<1:54:40,  1.53it/s]
                                                     
{'loss': 9.0342, 'grad_norm': 0.7531564831733704, 'learning_rate': 0.00016370439663236668, 'epoch': 0.02}
-
  2%|▏         | 175/10682 [16:26<1:54:40,  1.53it/s]
  2%|▏         | 176/10682 [16:27<1:54:48,  1.53it/s]
  2%|▏         | 177/10682 [16:28<2:00:20,  1.45it/s]
  2%|▏         | 178/10682 [16:28<1:54:01,  1.54it/s]
  2%|▏         | 179/10682 [16:29<1:48:41,  1.61it/s]
  2%|▏         | 180/10682 [16:30<2:06:32,  1.38it/s]
  2%|▏         | 181/10682 [16:30<2:04:20,  1.41it/s]
  2%|▏         | 182/10682 [16:31<2:00:32,  1.45it/s]
  2%|▏         | 183/10682 [16:32<1:58:38,  1.47it/s]
  2%|▏         | 184/10682 [16:32<1:54:43,  1.53it/s]
  2%|▏         | 185/10682 [16:33<1:49:39,  1.60it/s]
  2%|▏         | 186/10682 [16:33<1:46:38,  1.64it/s]
  2%|▏         | 187/10682 [16:34<1:54:33,  1.53it/s]
  2%|▏         | 188/10682 [16:35<1:48:24,  1.61it/s]
  2%|▏         | 189/10682 [16:35<1:48:02,  1.62it/s]
  2%|▏         | 190/10682 [16:36<1:53:35,  1.54it/s]
  2%|▏         | 191/10682 [16:37<1:51:01,  1.57it/s]
  2%|▏         | 192/10682 [16:37<1:48:44,  1.61it/s]
  2%|▏         | 193/10682 [16:38<1:46:23,  1.64it/s]
  2%|▏         | 194/10682 [16:38<1:44:30,  1.67it/s]
  2%|▏         | 195/10682 [16:39<1:47:07,  1.63it/s]
  2%|▏         | 196/10682 [16:40<1:49:48,  1.59it/s]
  2%|▏         | 197/10682 [16:40<1:49:37,  1.59it/s]
  2%|▏         | 198/10682 [16:41<1:49:02,  1.60it/s]
  2%|▏         | 199/10682 [16:42<1:49:53,  1.59it/s]
  2%|▏         | 200/10682 [16:42<1:51:12,  1.57it/s]
                                                     
{'loss': 8.6748, 'grad_norm': 0.7312736511230469, 'learning_rate': 0.00018709073900841907, 'epoch': 0.02}
-
  2%|▏         | 200/10682 [16:42<1:51:12,  1.57it/s]
  2%|▏         | 201/10682 [16:43<1:45:53,  1.65it/s]
  2%|▏         | 202/10682 [16:43<1:42:13,  1.71it/s]
  2%|▏         | 203/10682 [16:44<1:39:28,  1.76it/s]
  2%|▏         | 204/10682 [16:44<1:35:19,  1.83it/s]
  2%|▏         | 205/10682 [16:45<1:42:22,  1.71it/s]
  2%|▏         | 206/10682 [16:46<1:40:42,  1.73it/s]
  2%|▏         | 207/10682 [16:46<1:43:59,  1.68it/s]
  2%|▏         | 208/10682 [16:47<1:39:49,  1.75it/s]
  2%|▏         | 209/10682 [16:47<1:37:00,  1.80it/s]
  2%|▏         | 210/10682 [16:48<1:44:56,  1.66it/s]
  2%|▏         | 211/10682 [16:49<1:46:57,  1.63it/s]
  2%|▏         | 212/10682 [16:49<1:43:20,  1.69it/s]
  2%|▏         | 213/10682 [16:50<1:40:54,  1.73it/s]
  2%|▏         | 214/10682 [16:50<1:37:28,  1.79it/s]
  2%|▏         | 215/10682 [16:51<1:50:54,  1.57it/s]
  2%|▏         | 216/10682 [16:52<1:58:08,  1.48it/s]
  2%|▏         | 217/10682 [16:52<1:55:46,  1.51it/s]
  2%|▏         | 218/10682 [16:53<1:50:45,  1.57it/s]
  2%|▏         | 219/10682 [16:53<1:41:12,  1.72it/s]
  2%|▏         | 220/10682 [16:54<1:37:24,  1.79it/s]
  2%|▏         | 221/10682 [16:54<1:34:51,  1.84it/s]
  2%|▏         | 222/10682 [16:55<1:31:01,  1.92it/s]
  2%|▏         | 223/10682 [16:55<1:28:39,  1.97it/s]
  2%|▏         | 224/10682 [16:56<1:27:02,  2.00it/s]
  2%|▏         | 225/10682 [16:56<1:27:54,  1.98it/s]{'loss': 8.3498, 'grad_norm': 0.677618145942688, 'learning_rate': 0.00021047708138447147, 'epoch': 0.02}
                                                     
-
  2%|▏         | 225/10682 [16:56<1:27:54,  1.98it/s]
  2%|▏         | 226/10682 [16:57<1:27:47,  1.99it/s]
  2%|▏         | 227/10682 [16:57<1:30:21,  1.93it/s]
  2%|▏         | 228/10682 [16:58<1:32:59,  1.87it/s]
  2%|▏         | 229/10682 [16:58<1:31:00,  1.91it/s]
  2%|▏         | 230/10682 [16:59<1:29:18,  1.95it/s]
  2%|▏         | 231/10682 [16:59<1:28:42,  1.96it/s]
  2%|▏         | 232/10682 [17:00<1:26:47,  2.01it/s]
  2%|▏         | 233/10682 [17:00<1:26:58,  2.00it/s]
  2%|▏         | 234/10682 [17:01<1:25:47,  2.03it/s]
  2%|▏         | 235/10682 [17:01<1:24:37,  2.06it/s]
  2%|▏         | 236/10682 [17:02<1:25:21,  2.04it/s]
  2%|▏         | 237/10682 [17:02<1:27:18,  1.99it/s]
  2%|▏         | 238/10682 [17:03<1:28:27,  1.97it/s]
  2%|▏         | 239/10682 [17:04<1:39:38,  1.75it/s]
  2%|▏         | 240/10682 [17:04<1:51:21,  1.56it/s]
  2%|▏         | 241/10682 [17:05<1:46:48,  1.63it/s]
  2%|▏         | 242/10682 [17:06<1:41:47,  1.71it/s]
  2%|▏         | 243/10682 [17:06<1:35:14,  1.83it/s]
  2%|▏         | 244/10682 [17:07<1:33:46,  1.86it/s]
  2%|▏         | 245/10682 [17:07<1:33:25,  1.86it/s]
  2%|▏         | 246/10682 [17:07<1:28:29,  1.97it/s]
  2%|▏         | 247/10682 [17:08<1:25:35,  2.03it/s]
  2%|▏         | 248/10682 [17:08<1:24:17,  2.06it/s]
  2%|▏         | 249/10682 [17:09<1:24:45,  2.05it/s]
  2%|▏         | 250/10682 [17:09<1:25:31,  2.03it/s]{'loss': 8.0505, 'grad_norm': 0.668661892414093, 'learning_rate': 0.00023386342376052386, 'epoch': 0.02}
                                                     
-
  2%|▏         | 250/10682 [17:09<1:25:31,  2.03it/s]
  2%|▏         | 251/10682 [17:10<1:24:26,  2.06it/s]
  2%|▏         | 252/10682 [17:10<1:25:41,  2.03it/s]
  2%|▏         | 253/10682 [17:11<1:24:55,  2.05it/s]
  2%|▏         | 254/10682 [17:11<1:25:11,  2.04it/s]
  2%|▏         | 255/10682 [17:12<1:23:19,  2.09it/s]
  2%|▏         | 256/10682 [17:12<1:20:57,  2.15it/s]
  2%|▏         | 257/10682 [17:13<1:21:02,  2.14it/s]
  2%|▏         | 258/10682 [17:13<1:19:28,  2.19it/s]
  2%|▏         | 259/10682 [17:14<1:24:15,  2.06it/s]
  2%|▏         | 260/10682 [17:14<1:21:35,  2.13it/s]
  2%|▏         | 261/10682 [17:15<1:20:57,  2.15it/s]
  2%|▏         | 262/10682 [17:15<1:24:30,  2.06it/s]
  2%|▏         | 263/10682 [17:16<1:24:02,  2.07it/s]
  2%|▏         | 264/10682 [17:16<1:21:49,  2.12it/s]
  2%|▏         | 265/10682 [17:17<1:23:08,  2.09it/s]
  2%|▏         | 266/10682 [17:17<1:23:32,  2.08it/s]
  2%|▏         | 267/10682 [17:17<1:20:54,  2.15it/s]
  3%|▎         | 268/10682 [17:18<1:19:23,  2.19it/s]
  3%|▎         | 269/10682 [17:18<1:18:51,  2.20it/s]
  3%|▎         | 270/10682 [17:19<1:20:08,  2.17it/s]
  3%|▎         | 271/10682 [17:19<1:20:05,  2.17it/s]
  3%|▎         | 272/10682 [17:20<1:18:47,  2.20it/s]
  3%|▎         | 273/10682 [17:20<1:17:34,  2.24it/s]
  3%|▎         | 274/10682 [17:21<1:16:39,  2.26it/s]
  3%|▎         | 275/10682 [17:21<1:16:52,  2.26it/s]
                                                     
{'loss': 7.7573, 'grad_norm': 0.5522019267082214, 'learning_rate': 0.00025724976613657625, 'epoch': 0.03}
-
  3%|▎         | 275/10682 [17:21<1:16:52,  2.26it/s]
  3%|▎         | 276/10682 [17:22<1:19:21,  2.19it/s]
  3%|▎         | 277/10682 [17:22<1:20:01,  2.17it/s]
  3%|▎         | 278/10682 [17:22<1:18:02,  2.22it/s]
  3%|▎         | 279/10682 [17:23<1:19:54,  2.17it/s]
  3%|▎         | 280/10682 [17:23<1:18:22,  2.21it/s]
  3%|▎         | 281/10682 [17:24<1:16:57,  2.25it/s]
  3%|▎         | 282/10682 [17:24<1:16:00,  2.28it/s]
  3%|▎         | 283/10682 [17:25<1:16:29,  2.27it/s]
  3%|▎         | 284/10682 [17:25<1:16:30,  2.27it/s]
  3%|▎         | 285/10682 [17:26<1:16:31,  2.26it/s]
  3%|▎         | 286/10682 [17:26<1:18:30,  2.21it/s]
  3%|▎         | 287/10682 [17:26<1:17:10,  2.24it/s]
  3%|▎         | 288/10682 [17:27<1:16:23,  2.27it/s]
  3%|▎         | 289/10682 [17:27<1:16:20,  2.27it/s]
  3%|▎         | 290/10682 [17:28<1:15:43,  2.29it/s]
  3%|▎         | 291/10682 [17:28<1:16:54,  2.25it/s]
  3%|▎         | 292/10682 [17:29<1:16:07,  2.27it/s]
  3%|▎         | 293/10682 [17:29<1:15:16,  2.30it/s]
  3%|▎         | 294/10682 [17:29<1:14:32,  2.32it/s]
  3%|▎         | 295/10682 [17:30<1:14:03,  2.34it/s]
  3%|▎         | 296/10682 [17:30<1:13:37,  2.35it/s]
  3%|▎         | 297/10682 [17:31<1:14:02,  2.34it/s]
  3%|▎         | 298/10682 [17:31<1:13:45,  2.35it/s]
  3%|▎         | 299/10682 [17:32<1:13:02,  2.37it/s]
  3%|▎         | 300/10682 [17:32<1:14:35,  2.32it/s]{'loss': 7.4799, 'grad_norm': 0.6841521859169006, 'learning_rate': 0.0002806361085126286, 'epoch': 0.03}
-
                                                     

  3%|▎         | 300/10682 [17:32<1:14:35,  2.32it/s]
  3%|▎         | 301/10682 [17:32<1:16:43,  2.25it/s]
  3%|▎         | 302/10682 [17:33<1:16:49,  2.25it/s]
  3%|▎         | 303/10682 [17:33<1:16:24,  2.26it/s]
  3%|▎         | 304/10682 [17:34<1:15:55,  2.28it/s]
  3%|▎         | 305/10682 [17:34<1:14:50,  2.31it/s]
  3%|▎         | 306/10682 [17:35<1:14:22,  2.33it/s]
  3%|▎         | 307/10682 [17:35<1:14:47,  2.31it/s]
  3%|▎         | 308/10682 [17:36<1:14:51,  2.31it/s]
  3%|▎         | 309/10682 [17:36<1:14:08,  2.33it/s]
  3%|▎         | 310/10682 [17:36<1:13:31,  2.35it/s]
  3%|▎         | 311/10682 [17:37<1:15:42,  2.28it/s]
  3%|▎         | 312/10682 [17:37<1:14:48,  2.31it/s]
  3%|▎         | 313/10682 [17:38<1:15:50,  2.28it/s]
  3%|▎         | 314/10682 [17:38<1:14:22,  2.32it/s]
  3%|▎         | 315/10682 [17:39<1:13:27,  2.35it/s]
  3%|▎         | 316/10682 [17:39<1:12:57,  2.37it/s]
  3%|▎         | 317/10682 [17:39<1:12:28,  2.38it/s]
  3%|▎         | 318/10682 [17:40<1:12:06,  2.40it/s]
  3%|▎         | 319/10682 [17:40<1:11:35,  2.41it/s]
  3%|▎         | 320/10682 [17:41<1:12:21,  2.39it/s]
  3%|▎         | 321/10682 [17:41<1:14:25,  2.32it/s]
  3%|▎         | 322/10682 [17:41<1:14:27,  2.32it/s]
  3%|▎         | 323/10682 [17:42<1:26:11,  2.00it/s]
  3%|▎         | 324/10682 [17:43<1:23:03,  2.08it/s]
  3%|▎         | 325/10682 [17:43<1:20:56,  2.13it/s]
                                                     
{'loss': 7.2347, 'grad_norm': 0.4638720154762268, 'learning_rate': 0.00030402245088868103, 'epoch': 0.03}
-
  3%|▎         | 325/10682 [17:43<1:20:56,  2.13it/s]
  3%|▎         | 326/10682 [17:43<1:18:08,  2.21it/s]
  3%|▎         | 327/10682 [17:44<1:31:11,  1.89it/s]
  3%|▎         | 328/10682 [17:45<1:43:46,  1.66it/s]
  3%|▎         | 329/10682 [17:45<1:34:12,  1.83it/s]
  3%|▎         | 330/10682 [17:46<1:27:27,  1.97it/s]
  3%|▎         | 331/10682 [17:46<1:22:58,  2.08it/s]
  3%|▎         | 332/10682 [17:47<1:41:33,  1.70it/s]
  3%|▎         | 333/10682 [17:47<1:32:39,  1.86it/s]
  3%|▎         | 334/10682 [17:48<1:28:01,  1.96it/s]
  3%|▎         | 335/10682 [17:48<1:22:51,  2.08it/s]
  3%|▎         | 336/10682 [17:49<1:19:17,  2.17it/s]
  3%|▎         | 337/10682 [17:49<1:16:56,  2.24it/s]
  3%|▎         | 338/10682 [17:50<1:15:56,  2.27it/s]
  3%|▎         | 339/10682 [17:50<1:15:35,  2.28it/s]
  3%|▎         | 340/10682 [17:50<1:14:50,  2.30it/s]
  3%|▎         | 341/10682 [17:51<1:14:32,  2.31it/s]
  3%|▎         | 342/10682 [17:51<1:13:32,  2.34it/s]
  3%|▎         | 343/10682 [17:52<1:12:42,  2.37it/s]
  3%|▎         | 344/10682 [17:52<1:11:36,  2.41it/s]
  3%|▎         | 345/10682 [17:52<1:11:12,  2.42it/s]
  3%|▎         | 346/10682 [17:53<1:10:48,  2.43it/s]
  3%|▎         | 347/10682 [17:53<1:10:12,  2.45it/s]
  3%|▎         | 348/10682 [17:54<1:09:34,  2.48it/s]
  3%|▎         | 349/10682 [17:54<1:09:48,  2.47it/s]
  3%|▎         | 350/10682 [17:54<1:10:14,  2.45it/s]
                                                     
{'loss': 7.0099, 'grad_norm': 0.5867111086845398, 'learning_rate': 0.00032740879326473337, 'epoch': 0.03}
-
  3%|▎         | 350/10682 [17:54<1:10:14,  2.45it/s]
  3%|▎         | 351/10682 [17:55<1:11:19,  2.41it/s]
  3%|▎         | 352/10682 [17:55<1:12:25,  2.38it/s]
  3%|▎         | 353/10682 [17:56<1:11:24,  2.41it/s]
  3%|▎         | 354/10682 [17:56<1:10:20,  2.45it/s]
  3%|▎         | 355/10682 [17:57<1:10:46,  2.43it/s]
  3%|▎         | 356/10682 [17:57<1:10:32,  2.44it/s]
  3%|▎         | 357/10682 [17:57<1:10:25,  2.44it/s]
  3%|▎         | 358/10682 [17:58<1:09:56,  2.46it/s]
  3%|▎         | 359/10682 [17:58<1:10:27,  2.44it/s]
  3%|▎         | 360/10682 [17:59<1:10:09,  2.45it/s]
  3%|▎         | 361/10682 [17:59<1:09:31,  2.47it/s]
  3%|▎         | 362/10682 [17:59<1:10:29,  2.44it/s]
  3%|▎         | 363/10682 [18:00<1:10:24,  2.44it/s]
  3%|▎         | 364/10682 [18:00<1:10:14,  2.45it/s]
  3%|▎         | 365/10682 [18:01<1:09:41,  2.47it/s]
  3%|▎         | 366/10682 [18:01<1:09:28,  2.47it/s]
  3%|▎         | 367/10682 [18:01<1:09:01,  2.49it/s]
  3%|▎         | 368/10682 [18:02<1:09:06,  2.49it/s]
  3%|▎         | 369/10682 [18:02<1:09:13,  2.48it/s]
  3%|▎         | 370/10682 [18:03<1:09:32,  2.47it/s]
  3%|▎         | 371/10682 [18:03<1:09:23,  2.48it/s]
  3%|▎         | 372/10682 [18:03<1:09:30,  2.47it/s]
  3%|▎         | 373/10682 [18:04<1:09:13,  2.48it/s]
  4%|▎         | 374/10682 [18:04<1:09:51,  2.46it/s]
  4%|▎         | 375/10682 [18:05<1:09:43,  2.46it/s]{'loss': 6.8104, 'grad_norm': 0.5779871344566345, 'learning_rate': 0.0003507951356407858, 'epoch': 0.04}
                                                     
-
  4%|▎         | 375/10682 [18:05<1:09:43,  2.46it/s]
  4%|▎         | 376/10682 [18:05<1:09:47,  2.46it/s]
  4%|▎         | 377/10682 [18:05<1:09:39,  2.47it/s]
  4%|▎         | 378/10682 [18:06<1:09:15,  2.48it/s]
  4%|▎         | 379/10682 [18:06<1:09:22,  2.48it/s]
  4%|▎         | 380/10682 [18:07<1:09:18,  2.48it/s]
  4%|▎         | 381/10682 [18:07<1:09:10,  2.48it/s]
  4%|▎         | 382/10682 [18:07<1:08:39,  2.50it/s]
  4%|▎         | 383/10682 [18:08<1:08:27,  2.51it/s]
  4%|▎         | 384/10682 [18:08<1:08:29,  2.51it/s]
  4%|▎         | 385/10682 [18:09<1:08:17,  2.51it/s]
  4%|▎         | 386/10682 [18:09<1:07:55,  2.53it/s]
  4%|▎         | 387/10682 [18:09<1:09:05,  2.48it/s]
  4%|▎         | 388/10682 [18:10<1:10:20,  2.44it/s]
  4%|▎         | 389/10682 [18:10<1:09:21,  2.47it/s]
  4%|▎         | 390/10682 [18:11<1:09:22,  2.47it/s]
  4%|▎         | 391/10682 [18:11<1:08:53,  2.49it/s]
  4%|▎         | 392/10682 [18:12<1:11:05,  2.41it/s]
  4%|▎         | 393/10682 [18:12<1:10:25,  2.44it/s]
  4%|▎         | 394/10682 [18:12<1:10:56,  2.42it/s]
  4%|▎         | 395/10682 [18:13<1:09:41,  2.46it/s]
  4%|▎         | 396/10682 [18:13<1:10:58,  2.42it/s]
  4%|▎         | 397/10682 [18:14<1:09:46,  2.46it/s]
  4%|▎         | 398/10682 [18:14<1:09:12,  2.48it/s]
  4%|▎         | 399/10682 [18:14<1:08:45,  2.49it/s]
  4%|▎         | 400/10682 [18:15<1:08:41,  2.49it/s]{'loss': 6.6411, 'grad_norm': 0.7801389694213867, 'learning_rate': 0.00037418147801683815, 'epoch': 0.04}
                                                     
-
  4%|▎         | 400/10682 [18:15<1:08:41,  2.49it/s]
  4%|▍         | 401/10682 [18:15<1:08:28,  2.50it/s]
  4%|▍         | 402/10682 [18:16<1:08:21,  2.51it/s]
  4%|▍         | 403/10682 [18:16<1:09:17,  2.47it/s]
  4%|▍         | 404/10682 [18:16<1:08:50,  2.49it/s]
  4%|▍         | 405/10682 [18:17<1:08:17,  2.51it/s]
  4%|▍         | 406/10682 [18:17<1:08:32,  2.50it/s]
  4%|▍         | 407/10682 [18:18<1:08:05,  2.52it/s]
  4%|▍         | 408/10682 [18:18<1:08:05,  2.51it/s]
  4%|▍         | 409/10682 [18:18<1:08:11,  2.51it/s]
  4%|▍         | 410/10682 [18:19<1:07:44,  2.53it/s]
  4%|▍         | 411/10682 [18:19<1:07:19,  2.54it/s]
  4%|▍         | 412/10682 [18:20<1:08:42,  2.49it/s]
  4%|▍         | 413/10682 [18:20<1:08:18,  2.51it/s]
  4%|▍         | 414/10682 [18:20<1:08:11,  2.51it/s]
  4%|▍         | 415/10682 [18:21<1:07:32,  2.53it/s]
  4%|▍         | 416/10682 [18:21<1:07:23,  2.54it/s]
  4%|▍         | 417/10682 [18:22<1:07:55,  2.52it/s]
  4%|▍         | 418/10682 [18:22<1:07:47,  2.52it/s]
  4%|▍         | 419/10682 [18:22<1:07:50,  2.52it/s]
  4%|▍         | 420/10682 [18:23<1:07:56,  2.52it/s]
  4%|▍         | 421/10682 [18:23<1:07:37,  2.53it/s]
  4%|▍         | 422/10682 [18:23<1:07:37,  2.53it/s]
  4%|▍         | 423/10682 [18:24<1:07:49,  2.52it/s]
  4%|▍         | 424/10682 [18:24<1:07:30,  2.53it/s]
  4%|▍         | 425/10682 [18:25<1:07:37,  2.53it/s]
                                                     
{'loss': 6.497, 'grad_norm': 0.6400334239006042, 'learning_rate': 0.0003975678203928906, 'epoch': 0.04}
-
  4%|▍         | 425/10682 [18:25<1:07:37,  2.53it/s]
  4%|▍         | 426/10682 [18:25<1:07:11,  2.54it/s]
  4%|▍         | 427/10682 [18:25<1:07:12,  2.54it/s]
  4%|▍         | 428/10682 [18:26<1:07:16,  2.54it/s]
  4%|▍         | 429/10682 [18:26<1:08:10,  2.51it/s]
  4%|▍         | 430/10682 [18:27<1:07:40,  2.52it/s]
  4%|▍         | 431/10682 [18:27<1:07:27,  2.53it/s]
  4%|▍         | 432/10682 [18:27<1:07:28,  2.53it/s]
  4%|▍         | 433/10682 [18:28<1:07:38,  2.53it/s]
  4%|▍         | 434/10682 [18:28<1:07:03,  2.55it/s]
  4%|▍         | 435/10682 [18:29<1:07:03,  2.55it/s]
  4%|▍         | 436/10682 [18:29<1:07:07,  2.54it/s]
  4%|▍         | 437/10682 [18:29<1:06:59,  2.55it/s]
  4%|▍         | 438/10682 [18:30<1:07:17,  2.54it/s]
  4%|▍         | 439/10682 [18:30<1:07:23,  2.53it/s]
  4%|▍         | 440/10682 [18:31<1:07:23,  2.53it/s]
  4%|▍         | 441/10682 [18:31<1:07:29,  2.53it/s]
  4%|▍         | 442/10682 [18:31<1:07:04,  2.54it/s]
  4%|▍         | 443/10682 [18:32<1:06:37,  2.56it/s]
  4%|▍         | 444/10682 [18:32<1:06:44,  2.56it/s]
  4%|▍         | 445/10682 [18:33<1:07:50,  2.51it/s]
  4%|▍         | 446/10682 [18:33<1:07:42,  2.52it/s]
  4%|▍         | 447/10682 [18:33<1:07:20,  2.53it/s]
  4%|▍         | 448/10682 [18:34<1:07:13,  2.54it/s]
  4%|▍         | 449/10682 [18:34<1:08:12,  2.50it/s]
  4%|▍         | 450/10682 [18:35<1:07:28,  2.53it/s]
                                                     
{'loss': 6.3792, 'grad_norm': 0.8415307998657227, 'learning_rate': 0.00042095416276894293, 'epoch': 0.04}
-
  4%|▍         | 450/10682 [18:35<1:07:28,  2.53it/s]
  4%|▍         | 451/10682 [18:35<1:07:22,  2.53it/s]
  4%|▍         | 452/10682 [18:35<1:06:54,  2.55it/s]
  4%|▍         | 453/10682 [18:36<1:06:28,  2.56it/s]
  4%|▍         | 454/10682 [18:36<1:07:01,  2.54it/s]
  4%|▍         | 455/10682 [18:36<1:06:35,  2.56it/s]
  4%|▍         | 456/10682 [18:37<1:06:43,  2.55it/s]
  4%|▍         | 457/10682 [18:37<1:07:58,  2.51it/s]
  4%|▍         | 458/10682 [18:38<1:07:12,  2.54it/s]
  4%|▍         | 459/10682 [18:38<1:06:58,  2.54it/s]
  4%|▍         | 460/10682 [18:38<1:06:34,  2.56it/s]
  4%|▍         | 461/10682 [18:39<1:06:29,  2.56it/s]
  4%|▍         | 462/10682 [18:39<1:06:12,  2.57it/s]
  4%|▍         | 463/10682 [18:40<1:06:01,  2.58it/s]
  4%|▍         | 464/10682 [18:40<1:06:58,  2.54it/s]
  4%|▍         | 465/10682 [18:41<1:17:17,  2.20it/s]
  4%|▍         | 466/10682 [18:41<1:13:55,  2.30it/s]
  4%|▍         | 467/10682 [18:42<1:22:13,  2.07it/s]
  4%|▍         | 468/10682 [18:42<1:17:20,  2.20it/s]
  4%|▍         | 469/10682 [18:42<1:13:38,  2.31it/s]
  4%|▍         | 470/10682 [18:43<1:11:19,  2.39it/s]
  4%|▍         | 471/10682 [18:43<1:10:41,  2.41it/s]
  4%|▍         | 472/10682 [18:44<1:09:57,  2.43it/s]
  4%|▍         | 473/10682 [18:44<1:08:46,  2.47it/s]
  4%|▍         | 474/10682 [18:44<1:08:03,  2.50it/s]
  4%|▍         | 475/10682 [18:45<1:07:18,  2.53it/s]{'loss': 6.279, 'grad_norm': 0.853901743888855, 'learning_rate': 0.0004443405051449954, 'epoch': 0.04}
-
                                                     

  4%|▍         | 475/10682 [18:45<1:07:18,  2.53it/s]
  4%|▍         | 476/10682 [18:45<1:07:00,  2.54it/s]
  4%|▍         | 477/10682 [18:46<1:06:53,  2.54it/s]
  4%|▍         | 478/10682 [18:46<1:06:48,  2.55it/s]
  4%|▍         | 479/10682 [18:46<1:06:19,  2.56it/s]
  4%|▍         | 480/10682 [18:47<1:06:11,  2.57it/s]
  5%|▍         | 481/10682 [18:47<1:05:57,  2.58it/s]
  5%|▍         | 482/10682 [18:47<1:05:53,  2.58it/s]
  5%|▍         | 483/10682 [18:48<1:05:49,  2.58it/s]
  5%|▍         | 484/10682 [18:48<1:05:42,  2.59it/s]
  5%|▍         | 485/10682 [18:49<1:05:56,  2.58it/s]
  5%|▍         | 486/10682 [18:49<1:06:10,  2.57it/s]
  5%|▍         | 487/10682 [18:49<1:06:04,  2.57it/s]
  5%|▍         | 488/10682 [18:50<1:06:16,  2.56it/s]
  5%|▍         | 489/10682 [18:50<1:05:54,  2.58it/s]
  5%|▍         | 490/10682 [18:51<1:07:43,  2.51it/s]
  5%|▍         | 491/10682 [18:51<1:07:15,  2.53it/s]
  5%|▍         | 492/10682 [18:51<1:06:34,  2.55it/s]
  5%|▍         | 493/10682 [18:52<1:06:14,  2.56it/s]
  5%|▍         | 494/10682 [18:52<1:05:57,  2.57it/s]
  5%|▍         | 495/10682 [18:53<1:05:53,  2.58it/s]
  5%|▍         | 496/10682 [18:53<1:05:43,  2.58it/s]
  5%|▍         | 497/10682 [18:53<1:05:40,  2.58it/s]
  5%|▍         | 498/10682 [18:54<1:05:30,  2.59it/s]
  5%|▍         | 499/10682 [18:54<1:05:48,  2.58it/s]
  5%|▍         | 500/10682 [18:54<1:05:26,  2.59it/s]{'loss': 6.1821, 'grad_norm': 1.0189615488052368, 'learning_rate': 0.0004677268475210477, 'epoch': 0.05}
                                                     
-
  5%|▍         | 500/10682 [18:54<1:05:26,  2.59it/s]
  5%|▍         | 501/10682 [18:55<1:05:39,  2.58it/s]
  5%|▍         | 502/10682 [18:55<1:05:33,  2.59it/s]
  5%|▍         | 503/10682 [18:56<1:05:29,  2.59it/s]
  5%|▍         | 504/10682 [18:56<1:05:22,  2.59it/s]
  5%|▍         | 505/10682 [18:56<1:05:14,  2.60it/s]
  5%|▍         | 506/10682 [18:57<1:05:45,  2.58it/s]
  5%|▍         | 507/10682 [18:57<1:05:36,  2.58it/s]
  5%|▍         | 508/10682 [18:58<1:05:23,  2.59it/s]
  5%|▍         | 509/10682 [18:58<1:05:19,  2.60it/s]
  5%|▍         | 510/10682 [18:58<1:05:11,  2.60it/s]
  5%|▍         | 511/10682 [18:59<1:05:09,  2.60it/s]
  5%|▍         | 512/10682 [18:59<1:05:20,  2.59it/s]
  5%|▍         | 513/10682 [18:59<1:05:18,  2.59it/s]
  5%|▍         | 514/10682 [19:00<1:05:06,  2.60it/s]
  5%|▍         | 515/10682 [19:00<1:05:19,  2.59it/s]
  5%|▍         | 516/10682 [19:01<1:05:11,  2.60it/s]
  5%|▍         | 517/10682 [19:01<1:05:03,  2.60it/s]
  5%|▍         | 518/10682 [19:01<1:05:04,  2.60it/s]
  5%|▍         | 519/10682 [19:02<1:05:05,  2.60it/s]
  5%|▍         | 520/10682 [19:02<1:05:08,  2.60it/s]
  5%|▍         | 521/10682 [19:03<1:05:02,  2.60it/s]
  5%|▍         | 522/10682 [19:03<1:04:58,  2.61it/s]
  5%|▍         | 523/10682 [19:03<1:05:04,  2.60it/s]
  5%|▍         | 524/10682 [19:04<1:05:01,  2.60it/s]
  5%|▍         | 525/10682 [19:04<1:05:25,  2.59it/s]{'loss': 6.088, 'grad_norm': 0.9269179701805115, 'learning_rate': 0.0004911131898971, 'epoch': 0.05}
-
                                                     

  5%|▍         | 525/10682 [19:04<1:05:25,  2.59it/s]
  5%|▍         | 526/10682 [19:04<1:05:17,  2.59it/s]
  5%|▍         | 527/10682 [19:05<1:05:05,  2.60it/s]
  5%|▍         | 528/10682 [19:05<1:04:54,  2.61it/s]
  5%|▍         | 529/10682 [19:06<1:04:50,  2.61it/s]
  5%|▍         | 530/10682 [19:06<1:04:46,  2.61it/s]
  5%|▍         | 531/10682 [19:06<1:04:50,  2.61it/s]
  5%|▍         | 532/10682 [19:07<1:04:54,  2.61it/s]
  5%|▍         | 533/10682 [19:07<1:05:55,  2.57it/s]
  5%|▍         | 534/10682 [19:08<1:05:48,  2.57it/s]
  5%|▌         | 535/10682 [19:08<1:05:30,  2.58it/s]
  5%|▌         | 536/10682 [19:08<1:05:24,  2.59it/s]
  5%|▌         | 537/10682 [19:09<1:05:06,  2.60it/s]
  5%|▌         | 538/10682 [19:09<1:04:55,  2.60it/s]
  5%|▌         | 539/10682 [19:09<1:04:53,  2.60it/s]
  5%|▌         | 540/10682 [19:10<1:05:43,  2.57it/s]
  5%|▌         | 541/10682 [19:10<1:05:30,  2.58it/s]
  5%|▌         | 542/10682 [19:11<1:05:14,  2.59it/s]
  5%|▌         | 543/10682 [19:11<1:05:00,  2.60it/s]
  5%|▌         | 544/10682 [19:11<1:04:55,  2.60it/s]
  5%|▌         | 545/10682 [19:12<1:04:50,  2.61it/s]
  5%|▌         | 546/10682 [19:12<1:04:48,  2.61it/s]
  5%|▌         | 547/10682 [19:13<1:04:46,  2.61it/s]
  5%|▌         | 548/10682 [19:13<1:04:43,  2.61it/s]
  5%|▌         | 549/10682 [19:13<1:04:32,  2.62it/s]
  5%|▌         | 550/10682 [19:14<1:04:31,  2.62it/s]
                                                     {'loss': 6.0064, 'grad_norm': 1.5148539543151855, 'learning_rate': 0.0005144995322731525, 'epoch': 0.05}
-

  5%|▌         | 550/10682 [19:14<1:04:31,  2.62it/s]
  5%|▌         | 551/10682 [19:14<1:04:34,  2.61it/s]
  5%|▌         | 552/10682 [19:14<1:04:32,  2.62it/s]
  5%|▌         | 553/10682 [19:15<1:04:33,  2.61it/s]
  5%|▌         | 554/10682 [19:15<1:04:30,  2.62it/s]
  5%|▌         | 555/10682 [19:16<1:04:33,  2.61it/s]
  5%|▌         | 556/10682 [19:16<1:04:28,  2.62it/s]
  5%|▌         | 557/10682 [19:16<1:04:35,  2.61it/s]
  5%|▌         | 558/10682 [19:17<1:04:41,  2.61it/s]
  5%|▌         | 559/10682 [19:17<1:04:35,  2.61it/s]
  5%|▌         | 560/10682 [19:18<1:04:22,  2.62it/s]
  5%|▌         | 561/10682 [19:18<1:04:34,  2.61it/s]
  5%|▌         | 562/10682 [19:18<1:04:41,  2.61it/s]
  5%|▌         | 563/10682 [19:19<1:04:27,  2.62it/s]
  5%|▌         | 564/10682 [19:19<1:04:22,  2.62it/s]
  5%|▌         | 565/10682 [19:19<1:04:20,  2.62it/s]
  5%|▌         | 566/10682 [19:20<1:04:20,  2.62it/s]
  5%|▌         | 567/10682 [19:20<1:04:19,  2.62it/s]
  5%|▌         | 568/10682 [19:21<1:04:27,  2.62it/s]
  5%|▌         | 569/10682 [19:21<1:04:25,  2.62it/s]
  5%|▌         | 570/10682 [19:21<1:04:25,  2.62it/s]
  5%|▌         | 571/10682 [19:22<1:04:21,  2.62it/s]
  5%|▌         | 572/10682 [19:22<1:04:22,  2.62it/s]
  5%|▌         | 573/10682 [19:22<1:04:21,  2.62it/s]
  5%|▌         | 574/10682 [19:23<1:04:17,  2.62it/s]
  5%|▌         | 575/10682 [19:23<1:04:16,  2.62it/s]{'loss': 5.9353, 'grad_norm': 1.2231987714767456, 'learning_rate': 0.0005378858746492049, 'epoch': 0.05}
                                                     
-
  5%|▌         | 575/10682 [19:23<1:04:16,  2.62it/s]
  5%|▌         | 576/10682 [19:24<1:04:15,  2.62it/s]
  5%|▌         | 577/10682 [19:24<1:04:11,  2.62it/s]
  5%|▌         | 578/10682 [19:24<1:04:09,  2.62it/s]
  5%|▌         | 579/10682 [19:25<1:04:09,  2.62it/s]
  5%|▌         | 580/10682 [19:25<1:04:03,  2.63it/s]
  5%|▌         | 581/10682 [19:26<1:04:02,  2.63it/s]
  5%|▌         | 582/10682 [19:26<1:04:21,  2.62it/s]
  5%|▌         | 583/10682 [19:26<1:04:38,  2.60it/s]
  5%|▌         | 584/10682 [19:27<1:04:24,  2.61it/s]
  5%|▌         | 585/10682 [19:27<1:04:18,  2.62it/s]
  5%|▌         | 586/10682 [19:27<1:05:28,  2.57it/s]
  5%|▌         | 587/10682 [19:28<1:05:07,  2.58it/s]
  6%|▌         | 588/10682 [19:28<1:04:51,  2.59it/s]
  6%|▌         | 589/10682 [19:29<1:04:32,  2.61it/s]
  6%|▌         | 590/10682 [19:29<1:04:42,  2.60it/s]
  6%|▌         | 591/10682 [19:29<1:04:34,  2.60it/s]
  6%|▌         | 592/10682 [19:30<1:04:30,  2.61it/s]
  6%|▌         | 593/10682 [19:30<1:04:21,  2.61it/s]
  6%|▌         | 594/10682 [19:31<1:04:12,  2.62it/s]
  6%|▌         | 595/10682 [19:31<1:04:05,  2.62it/s]
  6%|▌         | 596/10682 [19:31<1:04:03,  2.62it/s]
  6%|▌         | 597/10682 [19:32<1:04:01,  2.63it/s]
  6%|▌         | 598/10682 [19:32<1:03:57,  2.63it/s]
  6%|▌         | 599/10682 [19:32<1:03:52,  2.63it/s]
  6%|▌         | 600/10682 [19:33<1:03:56,  2.63it/s]
                                                     {'loss': 5.8604, 'grad_norm': 0.9680529236793518, 'learning_rate': 0.0005612722170252572, 'epoch': 0.06}
-

  6%|▌         | 600/10682 [19:33<1:03:56,  2.63it/s]
  6%|▌         | 601/10682 [19:33<1:03:58,  2.63it/s]
  6%|▌         | 602/10682 [19:34<1:04:02,  2.62it/s]
  6%|▌         | 603/10682 [19:34<1:04:07,  2.62it/s]
  6%|▌         | 604/10682 [19:34<1:04:16,  2.61it/s]
  6%|▌         | 605/10682 [19:35<1:04:20,  2.61it/s]
  6%|▌         | 606/10682 [19:35<1:04:05,  2.62it/s]
  6%|▌         | 607/10682 [19:35<1:03:55,  2.63it/s]
  6%|▌         | 608/10682 [19:36<1:03:54,  2.63it/s]
  6%|▌         | 609/10682 [19:36<1:03:49,  2.63it/s]
  6%|▌         | 610/10682 [19:37<1:03:52,  2.63it/s]
  6%|▌         | 611/10682 [19:37<1:03:48,  2.63it/s]
  6%|▌         | 612/10682 [19:37<1:03:45,  2.63it/s]
  6%|▌         | 613/10682 [19:38<1:03:47,  2.63it/s]
  6%|▌         | 614/10682 [19:38<1:03:49,  2.63it/s]
  6%|▌         | 615/10682 [19:39<1:03:43,  2.63it/s]
  6%|▌         | 616/10682 [19:39<1:03:47,  2.63it/s]
  6%|▌         | 617/10682 [19:39<1:03:56,  2.62it/s]
  6%|▌         | 618/10682 [19:40<1:03:54,  2.62it/s]
  6%|▌         | 619/10682 [19:40<1:03:52,  2.63it/s]
  6%|▌         | 620/10682 [19:40<1:03:55,  2.62it/s]
  6%|▌         | 621/10682 [19:41<1:03:50,  2.63it/s]
  6%|▌         | 622/10682 [19:41<1:03:49,  2.63it/s]
  6%|▌         | 623/10682 [19:42<1:03:50,  2.63it/s]
  6%|▌         | 624/10682 [19:42<1:03:53,  2.62it/s]
  6%|▌         | 625/10682 [19:42<1:03:48,  2.63it/s]
                                                     
{'loss': 5.7941, 'grad_norm': 1.0056424140930176, 'learning_rate': 0.0005846585594013096, 'epoch': 0.06}
-
  6%|▌         | 625/10682 [19:42<1:03:48,  2.63it/s]
  6%|▌         | 626/10682 [19:43<1:03:51,  2.62it/s]
  6%|▌         | 627/10682 [19:43<1:05:06,  2.57it/s]
  6%|▌         | 628/10682 [19:44<1:05:10,  2.57it/s]
  6%|▌         | 629/10682 [19:44<1:04:46,  2.59it/s]
  6%|▌         | 630/10682 [19:44<1:04:26,  2.60it/s]
  6%|▌         | 631/10682 [19:45<1:04:15,  2.61it/s]
  6%|▌         | 632/10682 [19:45<1:04:04,  2.61it/s]
  6%|▌         | 633/10682 [19:45<1:04:14,  2.61it/s]
  6%|▌         | 634/10682 [19:46<1:04:14,  2.61it/s]
  6%|▌         | 635/10682 [19:46<1:04:06,  2.61it/s]
  6%|▌         | 636/10682 [19:47<1:14:22,  2.25it/s]
  6%|▌         | 637/10682 [19:47<1:11:40,  2.34it/s]
  6%|▌         | 638/10682 [19:48<1:09:20,  2.41it/s]
  6%|▌         | 639/10682 [19:48<1:07:38,  2.47it/s]
  6%|▌         | 640/10682 [19:48<1:06:22,  2.52it/s]
  6%|▌         | 641/10682 [19:49<1:05:24,  2.56it/s]
  6%|▌         | 642/10682 [19:49<1:05:01,  2.57it/s]
  6%|▌         | 643/10682 [19:49<1:04:44,  2.58it/s]
  6%|▌         | 644/10682 [19:50<1:04:33,  2.59it/s]
  6%|▌         | 645/10682 [19:50<1:04:19,  2.60it/s]
  6%|▌         | 646/10682 [19:51<1:04:08,  2.61it/s]
  6%|▌         | 647/10682 [19:51<1:03:56,  2.62it/s]
  6%|▌         | 648/10682 [19:51<1:03:53,  2.62it/s]
  6%|▌         | 649/10682 [19:52<1:03:47,  2.62it/s]
  6%|▌         | 650/10682 [19:52<1:03:43,  2.62it/s]{'loss': 5.7309, 'grad_norm': 1.095314860343933, 'learning_rate': 0.0006080449017773621, 'epoch': 0.06}
                                                     
-
  6%|▌         | 650/10682 [19:52<1:03:43,  2.62it/s]
  6%|▌         | 651/10682 [19:53<1:03:44,  2.62it/s]
  6%|▌         | 652/10682 [19:53<1:03:43,  2.62it/s]
  6%|▌         | 653/10682 [19:53<1:03:39,  2.63it/s]
  6%|▌         | 654/10682 [19:54<1:03:33,  2.63it/s]
  6%|▌         | 655/10682 [19:54<1:03:34,  2.63it/s]
  6%|▌         | 656/10682 [19:54<1:03:28,  2.63it/s]
  6%|▌         | 657/10682 [19:55<1:03:28,  2.63it/s]
  6%|▌         | 658/10682 [19:55<1:14:19,  2.25it/s]
  6%|▌         | 659/10682 [19:56<1:11:03,  2.35it/s]
  6%|▌         | 660/10682 [19:56<1:08:54,  2.42it/s]
  6%|▌         | 661/10682 [19:57<1:07:09,  2.49it/s]
  6%|▌         | 662/10682 [19:57<1:05:56,  2.53it/s]
  6%|▌         | 663/10682 [19:57<1:05:16,  2.56it/s]
  6%|▌         | 664/10682 [19:58<1:04:41,  2.58it/s]
  6%|▌         | 665/10682 [19:58<1:04:21,  2.59it/s]
  6%|▌         | 666/10682 [19:58<1:04:10,  2.60it/s]
  6%|▌         | 667/10682 [19:59<1:04:05,  2.60it/s]
  6%|▋         | 668/10682 [19:59<1:04:08,  2.60it/s]
  6%|▋         | 669/10682 [20:00<1:03:57,  2.61it/s]
  6%|▋         | 670/10682 [20:00<1:03:43,  2.62it/s]
  6%|▋         | 671/10682 [20:00<1:03:33,  2.63it/s]
  6%|▋         | 672/10682 [20:01<1:03:32,  2.63it/s]
  6%|▋         | 673/10682 [20:01<1:03:33,  2.62it/s]
  6%|▋         | 674/10682 [20:01<1:03:25,  2.63it/s]
  6%|▋         | 675/10682 [20:02<1:03:42,  2.62it/s]
                                                     
{'loss': 5.6729, 'grad_norm': 1.151089072227478, 'learning_rate': 0.0006314312441534145, 'epoch': 0.06}
-
  6%|▋         | 675/10682 [20:02<1:03:42,  2.62it/s]
  6%|▋         | 676/10682 [20:02<1:03:51,  2.61it/s]
  6%|▋         | 677/10682 [20:03<1:03:45,  2.62it/s]
  6%|▋         | 678/10682 [20:03<1:03:40,  2.62it/s]
  6%|▋         | 679/10682 [20:03<1:03:31,  2.62it/s]
  6%|▋         | 680/10682 [20:04<1:03:26,  2.63it/s]
  6%|▋         | 681/10682 [20:04<1:03:28,  2.63it/s]
  6%|▋         | 682/10682 [20:05<1:03:19,  2.63it/s]
  6%|▋         | 683/10682 [20:05<1:03:29,  2.62it/s]
  6%|▋         | 684/10682 [20:05<1:03:29,  2.62it/s]
  6%|▋         | 685/10682 [20:06<1:03:25,  2.63it/s]
  6%|▋         | 686/10682 [20:06<1:03:34,  2.62it/s]
  6%|▋         | 687/10682 [20:06<1:03:29,  2.62it/s]
  6%|▋         | 688/10682 [20:07<1:03:29,  2.62it/s]
  6%|▋         | 689/10682 [20:07<1:03:18,  2.63it/s]
  6%|▋         | 690/10682 [20:08<1:03:20,  2.63it/s]
  6%|▋         | 691/10682 [20:08<1:03:22,  2.63it/s]
  6%|▋         | 692/10682 [20:08<1:03:16,  2.63it/s]
  6%|▋         | 693/10682 [20:09<1:03:25,  2.63it/s]
  6%|▋         | 694/10682 [20:09<1:03:14,  2.63it/s]
  7%|▋         | 695/10682 [20:09<1:03:11,  2.63it/s]
  7%|▋         | 696/10682 [20:10<1:03:04,  2.64it/s]
  7%|▋         | 697/10682 [20:10<1:03:07,  2.64it/s]
  7%|▋         | 698/10682 [20:11<1:03:04,  2.64it/s]
  7%|▋         | 699/10682 [20:11<1:02:58,  2.64it/s]
  7%|▋         | 700/10682 [20:11<1:02:59,  2.64it/s]{'loss': 5.6247, 'grad_norm': 1.3073188066482544, 'learning_rate': 0.0006548175865294667, 'epoch': 0.07}
                                                     
-
  7%|▋         | 700/10682 [20:11<1:02:59,  2.64it/s]
  7%|▋         | 701/10682 [20:12<1:03:03,  2.64it/s]
  7%|▋         | 702/10682 [20:12<1:02:59,  2.64it/s]
  7%|▋         | 703/10682 [20:12<1:03:00,  2.64it/s]
  7%|▋         | 704/10682 [20:13<1:03:01,  2.64it/s]
  7%|▋         | 705/10682 [20:13<1:02:59,  2.64it/s]
  7%|▋         | 706/10682 [20:14<1:02:56,  2.64it/s]
  7%|▋         | 707/10682 [20:14<1:03:00,  2.64it/s]
  7%|▋         | 708/10682 [20:14<1:02:59,  2.64it/s]
  7%|▋         | 709/10682 [20:15<1:02:53,  2.64it/s]
  7%|▋         | 710/10682 [20:15<1:03:03,  2.64it/s]
  7%|▋         | 711/10682 [20:16<1:03:06,  2.63it/s]
  7%|▋         | 712/10682 [20:16<1:03:02,  2.64it/s]
  7%|▋         | 713/10682 [20:16<1:03:00,  2.64it/s]
  7%|▋         | 714/10682 [20:17<1:03:01,  2.64it/s]
  7%|▋         | 715/10682 [20:17<1:03:00,  2.64it/s]
  7%|▋         | 716/10682 [20:17<1:02:58,  2.64it/s]
  7%|▋         | 717/10682 [20:18<1:02:56,  2.64it/s]
  7%|▋         | 718/10682 [20:18<1:02:53,  2.64it/s]
  7%|▋         | 719/10682 [20:19<1:02:51,  2.64it/s]
  7%|▋         | 720/10682 [20:19<1:02:49,  2.64it/s]
  7%|▋         | 721/10682 [20:19<1:02:41,  2.65it/s]
  7%|▋         | 722/10682 [20:20<1:02:35,  2.65it/s]
  7%|▋         | 723/10682 [20:20<1:02:37,  2.65it/s]
  7%|▋         | 724/10682 [20:20<1:02:41,  2.65it/s]
  7%|▋         | 725/10682 [20:21<1:02:53,  2.64it/s]
                                                     {'loss': 5.5698, 'grad_norm': 0.9318245649337769, 'learning_rate': 0.0006782039289055192, 'epoch': 0.07}
-

  7%|▋         | 725/10682 [20:21<1:02:53,  2.64it/s]
  7%|▋         | 726/10682 [20:21<1:03:00,  2.63it/s]
  7%|▋         | 727/10682 [20:22<1:03:02,  2.63it/s]
  7%|▋         | 728/10682 [20:22<1:02:50,  2.64it/s]
  7%|▋         | 729/10682 [20:22<1:02:44,  2.64it/s]
  7%|▋         | 730/10682 [20:23<1:02:38,  2.65it/s]
  7%|▋         | 731/10682 [20:23<1:02:38,  2.65it/s]
  7%|▋         | 732/10682 [20:23<1:02:45,  2.64it/s]
  7%|▋         | 733/10682 [20:24<1:02:50,  2.64it/s]
  7%|▋         | 734/10682 [20:24<1:02:40,  2.65it/s]
  7%|▋         | 735/10682 [20:25<1:02:35,  2.65it/s]
  7%|▋         | 736/10682 [20:25<1:02:41,  2.64it/s]
  7%|▋         | 737/10682 [20:25<1:02:35,  2.65it/s]
  7%|▋         | 738/10682 [20:26<1:02:32,  2.65it/s]
  7%|▋         | 739/10682 [20:26<1:02:42,  2.64it/s]
  7%|▋         | 740/10682 [20:27<1:02:47,  2.64it/s]
  7%|▋         | 741/10682 [20:27<1:02:55,  2.63it/s]
  7%|▋         | 742/10682 [20:27<1:02:51,  2.64it/s]
  7%|▋         | 743/10682 [20:28<1:02:53,  2.63it/s]
  7%|▋         | 744/10682 [20:28<1:02:49,  2.64it/s]
  7%|▋         | 745/10682 [20:28<1:02:48,  2.64it/s]
  7%|▋         | 746/10682 [20:29<1:02:53,  2.63it/s]
  7%|▋         | 747/10682 [20:29<1:02:44,  2.64it/s]
  7%|▋         | 748/10682 [20:30<1:02:44,  2.64it/s]
  7%|▋         | 749/10682 [20:30<1:02:40,  2.64it/s]
  7%|▋         | 750/10682 [20:30<1:02:38,  2.64it/s]
                                                     
{'loss': 5.5258, 'grad_norm': 1.0060827732086182, 'learning_rate': 0.0007015902712815716, 'epoch': 0.07}
-
  7%|▋         | 750/10682 [20:30<1:02:38,  2.64it/s]
  7%|▋         | 751/10682 [20:31<1:02:38,  2.64it/s]
  7%|▋         | 752/10682 [20:31<1:02:34,  2.64it/s]
  7%|▋         | 753/10682 [20:31<1:02:29,  2.65it/s]
  7%|▋         | 754/10682 [20:32<1:02:31,  2.65it/s]
  7%|▋         | 755/10682 [20:32<1:02:38,  2.64it/s]
  7%|▋         | 756/10682 [20:33<1:02:40,  2.64it/s]
  7%|▋         | 757/10682 [20:33<1:02:44,  2.64it/s]
  7%|▋         | 758/10682 [20:33<1:02:37,  2.64it/s]
  7%|▋         | 759/10682 [20:34<1:02:29,  2.65it/s]
  7%|▋         | 760/10682 [20:34<1:02:24,  2.65it/s]
  7%|▋         | 761/10682 [20:34<1:02:26,  2.65it/s]
  7%|▋         | 762/10682 [20:35<1:02:22,  2.65it/s]
  7%|▋         | 763/10682 [20:35<1:02:30,  2.64it/s]
  7%|▋         | 764/10682 [20:36<1:02:29,  2.64it/s]
  7%|▋         | 765/10682 [20:36<1:02:22,  2.65it/s]
  7%|▋         | 766/10682 [20:36<1:02:22,  2.65it/s]
  7%|▋         | 767/10682 [20:37<1:02:21,  2.65it/s]
  7%|▋         | 768/10682 [20:37<1:02:18,  2.65it/s]
  7%|▋         | 769/10682 [20:37<1:02:24,  2.65it/s]
  7%|▋         | 770/10682 [20:38<1:02:23,  2.65it/s]
  7%|▋         | 771/10682 [20:38<1:02:23,  2.65it/s]
  7%|▋         | 772/10682 [20:39<1:02:38,  2.64it/s]
  7%|▋         | 773/10682 [20:39<1:02:38,  2.64it/s]
  7%|▋         | 774/10682 [20:39<1:02:34,  2.64it/s]
  7%|▋         | 775/10682 [20:40<1:02:27,  2.64it/s]{'loss': 5.4727, 'grad_norm': 1.0262162685394287, 'learning_rate': 0.0007249766136576241, 'epoch': 0.07}
                                                     
-
  7%|▋         | 775/10682 [20:40<1:02:27,  2.64it/s]
  7%|▋         | 776/10682 [20:40<1:02:28,  2.64it/s]
  7%|▋         | 777/10682 [20:41<1:02:29,  2.64it/s]
  7%|▋         | 778/10682 [20:41<1:02:23,  2.65it/s]
  7%|▋         | 779/10682 [20:41<1:02:16,  2.65it/s]
  7%|▋         | 780/10682 [20:42<1:02:18,  2.65it/s]
  7%|▋         | 781/10682 [20:42<1:02:13,  2.65it/s]
  7%|▋         | 782/10682 [20:42<1:02:14,  2.65it/s]
  7%|▋         | 783/10682 [20:43<1:02:14,  2.65it/s]
  7%|▋         | 784/10682 [20:43<1:02:24,  2.64it/s]
  7%|▋         | 785/10682 [20:44<1:02:25,  2.64it/s]
  7%|▋         | 786/10682 [20:44<1:02:16,  2.65it/s]
  7%|▋         | 787/10682 [20:44<1:02:18,  2.65it/s]
  7%|▋         | 788/10682 [20:45<1:02:13,  2.65it/s]
  7%|▋         | 789/10682 [20:45<1:02:08,  2.65it/s]
  7%|▋         | 790/10682 [20:45<1:02:04,  2.66it/s]
  7%|▋         | 791/10682 [20:46<1:02:10,  2.65it/s]
  7%|▋         | 792/10682 [20:46<1:02:11,  2.65it/s]
  7%|▋         | 793/10682 [20:47<1:02:12,  2.65it/s]
  7%|▋         | 794/10682 [20:47<1:02:09,  2.65it/s]
  7%|▋         | 795/10682 [20:47<1:02:06,  2.65it/s]
  7%|▋         | 796/10682 [20:48<1:02:07,  2.65it/s]
  7%|▋         | 797/10682 [20:48<1:02:11,  2.65it/s]
  7%|▋         | 798/10682 [20:48<1:02:10,  2.65it/s]
  7%|▋         | 799/10682 [20:49<1:02:20,  2.64it/s]
  7%|▋         | 800/10682 [20:49<1:02:21,  2.64it/s]{'loss': 5.4339, 'grad_norm': 1.0497294664382935, 'learning_rate': 0.0007483629560336763, 'epoch': 0.07}
                                                     
-
  7%|▋         | 800/10682 [20:49<1:02:21,  2.64it/s]
  7%|▋         | 801/10682 [20:50<1:02:34,  2.63it/s]
  8%|▊         | 802/10682 [20:50<1:02:27,  2.64it/s]
  8%|▊         | 803/10682 [20:50<1:02:22,  2.64it/s]
  8%|▊         | 804/10682 [20:51<1:02:19,  2.64it/s]
  8%|▊         | 805/10682 [20:51<1:02:17,  2.64it/s]
  8%|▊         | 806/10682 [20:51<1:02:12,  2.65it/s]
  8%|▊         | 807/10682 [20:52<1:02:12,  2.65it/s]
  8%|▊         | 808/10682 [20:52<1:02:13,  2.64it/s]
  8%|▊         | 809/10682 [20:53<1:02:14,  2.64it/s]
  8%|▊         | 810/10682 [20:53<1:02:08,  2.65it/s]
  8%|▊         | 811/10682 [20:53<1:02:08,  2.65it/s]
  8%|▊         | 812/10682 [20:54<1:02:10,  2.65it/s]
  8%|▊         | 813/10682 [20:54<1:02:14,  2.64it/s]
  8%|▊         | 814/10682 [20:54<1:02:11,  2.64it/s]
  8%|▊         | 815/10682 [20:55<1:02:08,  2.65it/s]
  8%|▊         | 816/10682 [20:55<1:02:09,  2.65it/s]
  8%|▊         | 817/10682 [20:56<1:02:09,  2.64it/s]
  8%|▊         | 818/10682 [20:56<1:02:08,  2.65it/s]
  8%|▊         | 819/10682 [20:56<1:02:10,  2.64it/s]
  8%|▊         | 820/10682 [20:57<1:02:08,  2.64it/s]
  8%|▊         | 821/10682 [20:57<1:02:12,  2.64it/s]
  8%|▊         | 822/10682 [20:58<1:02:10,  2.64it/s]
  8%|▊         | 823/10682 [20:58<1:02:11,  2.64it/s]
  8%|▊         | 824/10682 [20:58<1:02:12,  2.64it/s]
  8%|▊         | 825/10682 [20:59<1:02:08,  2.64it/s]{'loss': 5.3987, 'grad_norm': 1.1469128131866455, 'learning_rate': 0.0007717492984097287, 'epoch': 0.08}
                                                     
-
  8%|▊         | 825/10682 [20:59<1:02:08,  2.64it/s]
  8%|▊         | 826/10682 [20:59<1:02:06,  2.64it/s]
  8%|▊         | 827/10682 [20:59<1:02:05,  2.64it/s]
  8%|▊         | 828/10682 [21:00<1:02:14,  2.64it/s]
  8%|▊         | 829/10682 [21:00<1:02:11,  2.64it/s]
  8%|▊         | 830/10682 [21:01<1:02:11,  2.64it/s]
  8%|▊         | 831/10682 [21:01<1:02:08,  2.64it/s]
  8%|▊         | 832/10682 [21:01<1:02:05,  2.64it/s]
  8%|▊         | 833/10682 [21:02<1:02:07,  2.64it/s]
  8%|▊         | 834/10682 [21:02<1:02:10,  2.64it/s]
  8%|▊         | 835/10682 [21:03<1:13:31,  2.23it/s]
  8%|▊         | 836/10682 [21:03<1:18:32,  2.09it/s]
  8%|▊         | 837/10682 [21:04<1:13:30,  2.23it/s]
  8%|▊         | 838/10682 [21:04<1:10:01,  2.34it/s]
  8%|▊         | 839/10682 [21:04<1:07:39,  2.42it/s]
  8%|▊         | 840/10682 [21:05<1:05:57,  2.49it/s]
  8%|▊         | 841/10682 [21:05<1:04:50,  2.53it/s]
  8%|▊         | 842/10682 [21:05<1:03:55,  2.57it/s]
  8%|▊         | 843/10682 [21:06<1:03:22,  2.59it/s]
  8%|▊         | 844/10682 [21:06<1:02:54,  2.61it/s]
  8%|▊         | 845/10682 [21:07<1:02:38,  2.62it/s]
  8%|▊         | 846/10682 [21:07<1:02:31,  2.62it/s]
  8%|▊         | 847/10682 [21:07<1:02:15,  2.63it/s]
  8%|▊         | 848/10682 [21:08<1:02:14,  2.63it/s]
  8%|▊         | 849/10682 [21:08<1:02:03,  2.64it/s]
  8%|▊         | 850/10682 [21:08<1:01:56,  2.65it/s]
                                                     
{'loss': 5.365, 'grad_norm': 1.0227476358413696, 'learning_rate': 0.0007951356407857812, 'epoch': 0.08}
-
  8%|▊         | 850/10682 [21:08<1:01:56,  2.65it/s]
  8%|▊         | 851/10682 [21:09<1:02:01,  2.64it/s]
  8%|▊         | 852/10682 [21:09<1:01:57,  2.64it/s]
  8%|▊         | 853/10682 [21:10<1:01:52,  2.65it/s]
  8%|▊         | 854/10682 [21:10<1:01:55,  2.65it/s]
  8%|▊         | 855/10682 [21:10<1:01:49,  2.65it/s]
  8%|▊         | 856/10682 [21:11<1:01:54,  2.64it/s]
  8%|▊         | 857/10682 [21:11<1:01:54,  2.65it/s]
  8%|▊         | 858/10682 [21:12<1:01:47,  2.65it/s]
  8%|▊         | 859/10682 [21:12<1:01:50,  2.65it/s]
  8%|▊         | 860/10682 [21:12<1:01:56,  2.64it/s]
  8%|▊         | 861/10682 [21:13<1:02:01,  2.64it/s]
  8%|▊         | 862/10682 [21:13<1:01:56,  2.64it/s]
  8%|▊         | 863/10682 [21:13<1:01:57,  2.64it/s]
  8%|▊         | 864/10682 [21:14<1:01:53,  2.64it/s]
  8%|▊         | 865/10682 [21:14<1:01:54,  2.64it/s]
  8%|▊         | 866/10682 [21:15<1:01:51,  2.64it/s]
  8%|▊         | 867/10682 [21:15<1:01:45,  2.65it/s]
  8%|▊         | 868/10682 [21:15<1:01:40,  2.65it/s]
  8%|▊         | 869/10682 [21:16<1:01:42,  2.65it/s]
  8%|▊         | 870/10682 [21:16<1:01:38,  2.65it/s]
  8%|▊         | 871/10682 [21:16<1:01:39,  2.65it/s]
  8%|▊         | 872/10682 [21:17<1:01:45,  2.65it/s]
  8%|▊         | 873/10682 [21:17<1:01:51,  2.64it/s]
  8%|▊         | 874/10682 [21:18<1:01:50,  2.64it/s]
  8%|▊         | 875/10682 [21:18<1:01:45,  2.65it/s]
                                                     
{'loss': 5.3273, 'grad_norm': 1.2115955352783203, 'learning_rate': 0.0008185219831618334, 'epoch': 0.08}
-
  8%|▊         | 875/10682 [21:18<1:01:45,  2.65it/s]
  8%|▊         | 876/10682 [21:18<1:01:55,  2.64it/s]
  8%|▊         | 877/10682 [21:19<1:01:49,  2.64it/s]
  8%|▊         | 878/10682 [21:19<1:01:48,  2.64it/s]
  8%|▊         | 879/10682 [21:19<1:01:48,  2.64it/s]
  8%|▊         | 880/10682 [21:20<1:01:44,  2.65it/s]
  8%|▊         | 881/10682 [21:20<1:01:38,  2.65it/s]
  8%|▊         | 882/10682 [21:21<1:01:46,  2.64it/s]
  8%|▊         | 883/10682 [21:21<1:01:41,  2.65it/s]
  8%|▊         | 884/10682 [21:21<1:01:39,  2.65it/s]
  8%|▊         | 885/10682 [21:22<1:01:33,  2.65it/s]
  8%|▊         | 886/10682 [21:22<1:01:34,  2.65it/s]
  8%|▊         | 887/10682 [21:22<1:01:34,  2.65it/s]
  8%|▊         | 888/10682 [21:23<1:01:38,  2.65it/s]
  8%|▊         | 889/10682 [21:23<1:01:39,  2.65it/s]
  8%|▊         | 890/10682 [21:24<1:01:34,  2.65it/s]
  8%|▊         | 891/10682 [21:24<1:12:27,  2.25it/s]
  8%|▊         | 892/10682 [21:25<1:09:08,  2.36it/s]
  8%|▊         | 893/10682 [21:25<1:06:48,  2.44it/s]
  8%|▊         | 894/10682 [21:25<1:05:18,  2.50it/s]
  8%|▊         | 895/10682 [21:26<1:04:10,  2.54it/s]
  8%|▊         | 896/10682 [21:26<1:03:21,  2.57it/s]
  8%|▊         | 897/10682 [21:26<1:02:41,  2.60it/s]
  8%|▊         | 898/10682 [21:27<1:02:20,  2.62it/s]
  8%|▊         | 899/10682 [21:27<1:01:59,  2.63it/s]
  8%|▊         | 900/10682 [21:28<1:01:49,  2.64it/s]{'loss': 5.2891, 'grad_norm': 0.8936060070991516, 'learning_rate': 0.0008419083255378859, 'epoch': 0.08}
                                                     
-
  8%|▊         | 900/10682 [21:28<1:01:49,  2.64it/s]
  8%|▊         | 901/10682 [21:28<1:01:55,  2.63it/s]
  8%|▊         | 902/10682 [21:28<1:01:42,  2.64it/s]
  8%|▊         | 903/10682 [21:29<1:01:39,  2.64it/s]
  8%|▊         | 904/10682 [21:29<1:01:36,  2.65it/s]
  8%|▊         | 905/10682 [21:29<1:01:31,  2.65it/s]
  8%|▊         | 906/10682 [21:30<1:01:25,  2.65it/s]
  8%|▊         | 907/10682 [21:30<1:01:25,  2.65it/s]
  9%|▊         | 908/10682 [21:31<1:01:21,  2.65it/s]
  9%|▊         | 909/10682 [21:31<1:01:30,  2.65it/s]
  9%|▊         | 910/10682 [21:31<1:01:25,  2.65it/s]
  9%|▊         | 911/10682 [21:32<1:01:32,  2.65it/s]
  9%|▊         | 912/10682 [21:32<1:01:39,  2.64it/s]
  9%|▊         | 913/10682 [21:33<1:01:35,  2.64it/s]
  9%|▊         | 914/10682 [21:33<1:01:30,  2.65it/s]
  9%|▊         | 915/10682 [21:33<1:01:37,  2.64it/s]
  9%|▊         | 916/10682 [21:34<1:01:30,  2.65it/s]
  9%|▊         | 917/10682 [21:34<1:01:27,  2.65it/s]
  9%|▊         | 918/10682 [21:34<1:01:30,  2.65it/s]
  9%|▊         | 919/10682 [21:35<1:01:25,  2.65it/s]
  9%|▊         | 920/10682 [21:35<1:01:19,  2.65it/s]
  9%|▊         | 921/10682 [21:36<1:01:15,  2.66it/s]
  9%|▊         | 922/10682 [21:36<1:01:12,  2.66it/s]
  9%|▊         | 923/10682 [21:36<1:01:19,  2.65it/s]
  9%|▊         | 924/10682 [21:37<1:01:27,  2.65it/s]
  9%|▊         | 925/10682 [21:37<1:01:21,  2.65it/s]
                                                     
{'loss': 5.2492, 'grad_norm': 0.9921799302101135, 'learning_rate': 0.0008652946679139383, 'epoch': 0.09}
-
  9%|▊         | 925/10682 [21:37<1:01:21,  2.65it/s]
  9%|▊         | 926/10682 [21:37<1:01:18,  2.65it/s]
  9%|▊         | 927/10682 [21:38<1:01:12,  2.66it/s]
  9%|▊         | 928/10682 [21:38<1:01:12,  2.66it/s]
  9%|▊         | 929/10682 [21:39<1:01:11,  2.66it/s]
  9%|▊         | 930/10682 [21:39<1:01:12,  2.66it/s]
  9%|▊         | 931/10682 [21:39<1:01:09,  2.66it/s]
  9%|▊         | 932/10682 [21:40<1:01:13,  2.65it/s]
  9%|▊         | 933/10682 [21:40<1:01:31,  2.64it/s]
  9%|▊         | 934/10682 [21:40<1:01:24,  2.65it/s]
  9%|▉         | 935/10682 [21:41<1:01:22,  2.65it/s]
  9%|▉         | 936/10682 [21:41<1:01:25,  2.64it/s]
  9%|▉         | 937/10682 [21:42<1:01:21,  2.65it/s]
  9%|▉         | 938/10682 [21:42<1:01:15,  2.65it/s]
  9%|▉         | 939/10682 [21:42<1:01:13,  2.65it/s]
  9%|▉         | 940/10682 [21:43<1:01:08,  2.66it/s]
  9%|▉         | 941/10682 [21:43<1:01:03,  2.66it/s]
  9%|▉         | 942/10682 [21:43<1:00:59,  2.66it/s]
  9%|▉         | 943/10682 [21:44<1:01:02,  2.66it/s]
  9%|▉         | 944/10682 [21:44<1:01:15,  2.65it/s]
  9%|▉         | 945/10682 [21:45<1:01:19,  2.65it/s]
  9%|▉         | 946/10682 [21:45<1:01:20,  2.65it/s]
  9%|▉         | 947/10682 [21:45<1:01:15,  2.65it/s]
  9%|▉         | 948/10682 [21:46<1:01:10,  2.65it/s]
  9%|▉         | 949/10682 [21:46<1:01:05,  2.66it/s]
  9%|▉         | 950/10682 [21:46<1:01:06,  2.65it/s]{'loss': 5.2264, 'grad_norm': 0.9657398462295532, 'learning_rate': 0.0008886810102899908, 'epoch': 0.09}
                                                     
-
  9%|▉         | 950/10682 [21:46<1:01:06,  2.65it/s]
  9%|▉         | 951/10682 [21:47<1:01:08,  2.65it/s]
  9%|▉         | 952/10682 [21:47<1:01:03,  2.66it/s]
  9%|▉         | 953/10682 [21:48<1:01:03,  2.66it/s]
  9%|▉         | 954/10682 [21:48<1:01:10,  2.65it/s]
  9%|▉         | 955/10682 [21:48<1:01:09,  2.65it/s]
  9%|▉         | 956/10682 [21:49<1:01:08,  2.65it/s]
  9%|▉         | 957/10682 [21:49<1:01:04,  2.65it/s]
  9%|▉         | 958/10682 [21:49<1:01:07,  2.65it/s]
  9%|▉         | 959/10682 [21:50<1:01:29,  2.64it/s]
  9%|▉         | 960/10682 [21:50<1:01:22,  2.64it/s]
  9%|▉         | 961/10682 [21:51<1:01:18,  2.64it/s]
  9%|▉         | 962/10682 [21:51<1:01:10,  2.65it/s]
  9%|▉         | 963/10682 [21:51<1:01:05,  2.65it/s]
  9%|▉         | 964/10682 [21:52<1:01:03,  2.65it/s]
  9%|▉         | 965/10682 [21:52<1:00:58,  2.66it/s]
  9%|▉         | 966/10682 [21:53<1:00:57,  2.66it/s]
  9%|▉         | 967/10682 [21:53<1:00:57,  2.66it/s]
  9%|▉         | 968/10682 [21:53<1:00:59,  2.65it/s]
  9%|▉         | 969/10682 [21:54<1:00:54,  2.66it/s]
  9%|▉         | 970/10682 [21:54<1:01:02,  2.65it/s]
  9%|▉         | 971/10682 [21:54<1:00:59,  2.65it/s]
  9%|▉         | 972/10682 [21:55<1:00:57,  2.66it/s]
  9%|▉         | 973/10682 [21:55<1:01:09,  2.65it/s]
  9%|▉         | 974/10682 [21:56<1:01:07,  2.65it/s]
  9%|▉         | 975/10682 [21:56<1:01:11,  2.64it/s]{'loss': 5.1874, 'grad_norm': 1.0826326608657837, 'learning_rate': 0.000912067352666043, 'epoch': 0.09}
                                                     
-
  9%|▉         | 975/10682 [21:56<1:01:11,  2.64it/s]
  9%|▉         | 976/10682 [21:56<1:01:11,  2.64it/s]
  9%|▉         | 977/10682 [21:57<1:01:08,  2.65it/s]
  9%|▉         | 978/10682 [21:57<1:01:11,  2.64it/s]
  9%|▉         | 979/10682 [21:57<1:01:03,  2.65it/s]
  9%|▉         | 980/10682 [21:58<1:01:03,  2.65it/s]
  9%|▉         | 981/10682 [21:58<1:00:58,  2.65it/s]
  9%|▉         | 982/10682 [21:59<1:00:55,  2.65it/s]
  9%|▉         | 983/10682 [21:59<1:00:49,  2.66it/s]
  9%|▉         | 984/10682 [21:59<1:00:43,  2.66it/s]
  9%|▉         | 985/10682 [22:00<1:00:43,  2.66it/s]
  9%|▉         | 986/10682 [22:00<1:00:39,  2.66it/s]
  9%|▉         | 987/10682 [22:00<1:00:48,  2.66it/s]
  9%|▉         | 988/10682 [22:01<1:00:45,  2.66it/s]
  9%|▉         | 989/10682 [22:01<1:00:44,  2.66it/s]
  9%|▉         | 990/10682 [22:02<1:00:47,  2.66it/s]
  9%|▉         | 991/10682 [22:02<1:00:43,  2.66it/s]
  9%|▉         | 992/10682 [22:02<1:00:40,  2.66it/s]
  9%|▉         | 993/10682 [22:03<1:00:38,  2.66it/s]
  9%|▉         | 994/10682 [22:03<1:00:42,  2.66it/s]
  9%|▉         | 995/10682 [22:03<1:00:49,  2.65it/s]
  9%|▉         | 996/10682 [22:04<1:00:43,  2.66it/s]
  9%|▉         | 997/10682 [22:04<1:00:46,  2.66it/s]
  9%|▉         | 998/10682 [22:05<1:00:47,  2.66it/s]
  9%|▉         | 999/10682 [22:05<1:00:43,  2.66it/s]
  9%|▉         | 1000/10682 [22:05<1:00:40,  2.66it/s]
                                                      
{'loss': 5.1658, 'grad_norm': 1.0371633768081665, 'learning_rate': 0.0009354536950420954, 'epoch': 0.09}
-
  9%|▉         | 1000/10682 [22:05<1:00:40,  2.66it/s]
  9%|▉         | 1001/10682 [22:06<1:00:53,  2.65it/s]
  9%|▉         | 1002/10682 [22:06<1:00:46,  2.65it/s]
  9%|▉         | 1003/10682 [22:06<1:00:47,  2.65it/s]
  9%|▉         | 1004/10682 [22:07<1:00:51,  2.65it/s]
  9%|▉         | 1005/10682 [22:07<1:00:48,  2.65it/s]
  9%|▉         | 1006/10682 [22:08<1:00:45,  2.65it/s]
  9%|▉         | 1007/10682 [22:08<1:00:43,  2.66it/s]
  9%|▉         | 1008/10682 [22:08<1:00:41,  2.66it/s]
  9%|▉         | 1009/10682 [22:09<1:00:38,  2.66it/s]
  9%|▉         | 1010/10682 [22:09<1:00:36,  2.66it/s]
  9%|▉         | 1011/10682 [22:09<1:00:32,  2.66it/s]
  9%|▉         | 1012/10682 [22:10<1:00:39,  2.66it/s]
  9%|▉         | 1013/10682 [22:10<1:00:39,  2.66it/s]
  9%|▉         | 1014/10682 [22:11<1:00:36,  2.66it/s]
 10%|▉         | 1015/10682 [22:11<1:00:34,  2.66it/s]
 10%|▉         | 1016/10682 [22:11<1:00:33,  2.66it/s]
 10%|▉         | 1017/10682 [22:12<1:00:32,  2.66it/s]
 10%|▉         | 1018/10682 [22:12<1:00:44,  2.65it/s]
 10%|▉         | 1019/10682 [22:12<1:00:45,  2.65it/s]
 10%|▉         | 1020/10682 [22:13<1:00:41,  2.65it/s]
 10%|▉         | 1021/10682 [22:13<1:00:40,  2.65it/s]
 10%|▉         | 1022/10682 [22:14<1:00:41,  2.65it/s]
 10%|▉         | 1023/10682 [22:14<1:00:40,  2.65it/s]
 10%|▉         | 1024/10682 [22:14<1:00:51,  2.64it/s]
 10%|▉         | 1025/10682 [22:15<1:00:49,  2.65it/s]
                                                      {'loss': 5.1371, 'grad_norm': 0.9833744764328003, 'learning_rate': 0.0009588400374181479, 'epoch': 0.1}
-
 10%|▉         | 1025/10682 [22:15<1:00:49,  2.65it/s]
 10%|▉         | 1026/10682 [22:15<1:00:44,  2.65it/s]
 10%|▉         | 1027/10682 [22:15<1:00:46,  2.65it/s]
 10%|▉         | 1028/10682 [22:16<1:00:38,  2.65it/s]
 10%|▉         | 1029/10682 [22:16<1:00:32,  2.66it/s]
 10%|▉         | 1030/10682 [22:17<1:00:29,  2.66it/s]
 10%|▉         | 1031/10682 [22:17<1:00:23,  2.66it/s]
 10%|▉         | 1032/10682 [22:17<1:00:34,  2.66it/s]
 10%|▉         | 1033/10682 [22:18<1:00:28,  2.66it/s]
 10%|▉         | 1034/10682 [22:18<1:00:31,  2.66it/s]
 10%|▉         | 1035/10682 [22:18<1:00:27,  2.66it/s]
 10%|▉         | 1036/10682 [22:19<1:00:21,  2.66it/s]
 10%|▉         | 1037/10682 [22:19<1:00:23,  2.66it/s]
 10%|▉         | 1038/10682 [22:20<1:00:25,  2.66it/s]
 10%|▉         | 1039/10682 [22:20<1:00:24,  2.66it/s]
 10%|▉         | 1040/10682 [22:20<1:00:23,  2.66it/s]
 10%|▉         | 1041/10682 [22:21<1:00:28,  2.66it/s]
 10%|▉         | 1042/10682 [22:21<1:13:41,  2.18it/s]
 10%|▉         | 1043/10682 [22:22<1:09:48,  2.30it/s]
 10%|▉         | 1044/10682 [22:22<1:07:02,  2.40it/s]
 10%|▉         | 1045/10682 [22:23<1:05:01,  2.47it/s]
 10%|▉         | 1046/10682 [22:23<1:03:42,  2.52it/s]
 10%|▉         | 1047/10682 [22:23<1:02:39,  2.56it/s]
 10%|▉         | 1048/10682 [22:24<1:15:04,  2.14it/s]
 10%|▉         | 1049/10682 [22:24<1:10:42,  2.27it/s]
 10%|▉         | 1050/10682 [22:25<1:07:33,  2.38it/s]
                                                      
{'loss': 5.1089, 'grad_norm': 0.965286374092102, 'learning_rate': 0.0009822263797942, 'epoch': 0.1}
-
 10%|▉         | 1050/10682 [22:25<1:07:33,  2.38it/s]
 10%|▉         | 1051/10682 [22:25<1:05:28,  2.45it/s]
 10%|▉         | 1052/10682 [22:25<1:03:54,  2.51it/s]
 10%|▉         | 1053/10682 [22:26<1:02:49,  2.55it/s]
 10%|▉         | 1054/10682 [22:26<1:02:08,  2.58it/s]
 10%|▉         | 1055/10682 [22:27<1:01:35,  2.61it/s]
 10%|▉         | 1056/10682 [22:27<1:01:23,  2.61it/s]
 10%|▉         | 1057/10682 [22:27<1:01:00,  2.63it/s]
 10%|▉         | 1058/10682 [22:28<1:00:46,  2.64it/s]
 10%|▉         | 1059/10682 [22:28<1:00:38,  2.65it/s]
 10%|▉         | 1060/10682 [22:28<1:00:34,  2.65it/s]
 10%|▉         | 1061/10682 [22:29<1:00:32,  2.65it/s]
 10%|▉         | 1062/10682 [22:29<1:00:33,  2.65it/s]
 10%|▉         | 1063/10682 [22:30<1:00:28,  2.65it/s]
 10%|▉         | 1064/10682 [22:30<1:00:25,  2.65it/s]
 10%|▉         | 1065/10682 [22:30<1:00:19,  2.66it/s]
 10%|▉         | 1066/10682 [22:31<1:00:18,  2.66it/s]
 10%|▉         | 1067/10682 [22:31<1:00:43,  2.64it/s]
 10%|▉         | 1068/10682 [22:31<1:00:33,  2.65it/s]
 10%|█         | 1069/10682 [22:32<1:00:29,  2.65it/s]
 10%|█         | 1070/10682 [22:32<1:00:22,  2.65it/s]
 10%|█         | 1071/10682 [22:33<1:00:17,  2.66it/s]
 10%|█         | 1072/10682 [22:33<1:00:19,  2.65it/s]
 10%|█         | 1073/10682 [22:33<1:00:20,  2.65it/s]
 10%|█         | 1074/10682 [22:34<1:00:18,  2.66it/s]
 10%|█         | 1075/10682 [22:34<1:00:20,  2.65it/s]{'loss': 5.0885, 'grad_norm': 0.8102666139602661, 'learning_rate': 0.0009999990387768293, 'epoch': 0.1}
                                                      
-
 10%|█         | 1075/10682 [22:34<1:00:20,  2.65it/s]
 10%|█         | 1076/10682 [22:34<1:00:15,  2.66it/s]
 10%|█         | 1077/10682 [22:35<1:00:13,  2.66it/s]
 10%|█         | 1078/10682 [22:35<1:00:05,  2.66it/s]
 10%|█         | 1079/10682 [22:36<1:00:04,  2.66it/s]
 10%|█         | 1080/10682 [22:36<1:00:05,  2.66it/s]
 10%|█         | 1081/10682 [22:36<1:00:02,  2.67it/s]
 10%|█         | 1082/10682 [22:37<1:00:02,  2.66it/s]
 10%|█         | 1083/10682 [22:37<1:00:00,  2.67it/s]
 10%|█         | 1084/10682 [22:37<59:57,  2.67it/s]  
 10%|█         | 1085/10682 [22:38<59:57,  2.67it/s]
 10%|█         | 1086/10682 [22:38<59:54,  2.67it/s]
 10%|█         | 1087/10682 [22:39<59:57,  2.67it/s]
 10%|█         | 1088/10682 [22:39<59:56,  2.67it/s]
 10%|█         | 1089/10682 [22:39<59:54,  2.67it/s]
 10%|█         | 1090/10682 [22:40<59:54,  2.67it/s]
 10%|█         | 1091/10682 [22:40<59:50,  2.67it/s]
 10%|█         | 1092/10682 [22:40<59:51,  2.67it/s]
 10%|█         | 1093/10682 [22:41<59:50,  2.67it/s]
 10%|█         | 1094/10682 [22:41<59:54,  2.67it/s]
 10%|█         | 1095/10682 [22:42<59:52,  2.67it/s]
 10%|█         | 1096/10682 [22:42<59:56,  2.67it/s]
 10%|█         | 1097/10682 [22:42<59:56,  2.67it/s]
 10%|█         | 1098/10682 [22:43<59:55,  2.67it/s]
 10%|█         | 1099/10682 [22:43<59:55,  2.66it/s]
 10%|█         | 1100/10682 [22:43<59:53,  2.67it/s]
                                                    
{'loss': 5.0649, 'grad_norm': 0.8110109567642212, 'learning_rate': 0.0009999743408927195, 'epoch': 0.1}
-
 10%|█         | 1100/10682 [22:43<59:53,  2.67it/s]
 10%|█         | 1101/10682 [22:44<59:59,  2.66it/s]
 10%|█         | 1102/10682 [22:44<59:56,  2.66it/s]
 10%|█         | 1103/10682 [22:45<59:59,  2.66it/s]
 10%|█         | 1104/10682 [22:45<1:00:07,  2.66it/s]
 10%|█         | 1105/10682 [22:45<1:00:06,  2.66it/s]
 10%|█         | 1106/10682 [22:46<59:57,  2.66it/s]  
 10%|█         | 1107/10682 [22:46<59:53,  2.66it/s]
 10%|█         | 1108/10682 [22:46<59:47,  2.67it/s]
 10%|█         | 1109/10682 [22:47<59:47,  2.67it/s]
 10%|█         | 1110/10682 [22:47<59:49,  2.67it/s]
 10%|█         | 1111/10682 [22:48<59:46,  2.67it/s]
 10%|█         | 1112/10682 [22:48<59:46,  2.67it/s]
 10%|█         | 1113/10682 [22:48<59:49,  2.67it/s]
 10%|█         | 1114/10682 [22:49<59:47,  2.67it/s]
 10%|█         | 1115/10682 [22:49<59:55,  2.66it/s]
 10%|█         | 1116/10682 [22:49<59:53,  2.66it/s]
 10%|█         | 1117/10682 [22:50<59:50,  2.66it/s]
 10%|█         | 1118/10682 [22:50<59:44,  2.67it/s]
 10%|█         | 1119/10682 [22:51<59:46,  2.67it/s]
 10%|█         | 1120/10682 [22:51<59:44,  2.67it/s]
 10%|█         | 1121/10682 [22:51<59:46,  2.67it/s]
 11%|█         | 1122/10682 [22:52<59:49,  2.66it/s]
 11%|█         | 1123/10682 [22:52<59:46,  2.67it/s]
 11%|█         | 1124/10682 [22:52<59:43,  2.67it/s]
 11%|█         | 1125/10682 [22:53<59:40,  2.67it/s]{'loss': 5.0474, 'grad_norm': 1.1373594999313354, 'learning_rate': 0.0009999162690918172, 'epoch': 0.11}
                                                    
-
 11%|█         | 1125/10682 [22:53<59:40,  2.67it/s]
 11%|█         | 1126/10682 [22:53<59:56,  2.66it/s]
 11%|█         | 1127/10682 [22:54<59:55,  2.66it/s]
 11%|█         | 1128/10682 [22:54<59:46,  2.66it/s]
 11%|█         | 1129/10682 [22:54<59:42,  2.67it/s]
 11%|█         | 1130/10682 [22:55<59:52,  2.66it/s]
 11%|█         | 1131/10682 [22:55<59:47,  2.66it/s]
 11%|█         | 1132/10682 [22:55<59:44,  2.66it/s]
 11%|█         | 1133/10682 [22:56<59:47,  2.66it/s]
 11%|█         | 1134/10682 [22:56<59:49,  2.66it/s]
 11%|█         | 1135/10682 [22:57<59:50,  2.66it/s]
 11%|█         | 1136/10682 [22:57<59:54,  2.66it/s]
 11%|█         | 1137/10682 [22:57<59:48,  2.66it/s]
 11%|█         | 1138/10682 [22:58<59:47,  2.66it/s]
 11%|█         | 1139/10682 [22:58<59:44,  2.66it/s]
 11%|█         | 1140/10682 [22:58<59:40,  2.67it/s]
 11%|█         | 1141/10682 [22:59<59:40,  2.66it/s]
 11%|█         | 1142/10682 [22:59<59:37,  2.67it/s]
 11%|█         | 1143/10682 [23:00<59:38,  2.67it/s]
 11%|█         | 1144/10682 [23:00<59:38,  2.67it/s]
 11%|█         | 1145/10682 [23:00<59:45,  2.66it/s]
 11%|█         | 1146/10682 [23:01<59:41,  2.66it/s]
 11%|█         | 1147/10682 [23:01<59:41,  2.66it/s]
 11%|█         | 1148/10682 [23:01<59:40,  2.66it/s]
 11%|█         | 1149/10682 [23:02<59:34,  2.67it/s]
 11%|█         | 1150/10682 [23:02<59:32,  2.67it/s]
                                                    
{'loss': 5.0206, 'grad_norm': 0.8969134092330933, 'learning_rate': 0.0009998248272504882, 'epoch': 0.11}
-
 11%|█         | 1150/10682 [23:02<59:32,  2.67it/s]
 11%|█         | 1151/10682 [23:03<59:43,  2.66it/s]
 11%|█         | 1152/10682 [23:03<59:45,  2.66it/s]
 11%|█         | 1153/10682 [23:03<59:39,  2.66it/s]
 11%|█         | 1154/10682 [23:04<59:34,  2.67it/s]
 11%|█         | 1155/10682 [23:04<59:36,  2.66it/s]
 11%|█         | 1156/10682 [23:04<59:33,  2.67it/s]
 11%|█         | 1157/10682 [23:05<59:31,  2.67it/s]
 11%|█         | 1158/10682 [23:05<59:33,  2.67it/s]
 11%|█         | 1159/10682 [23:06<59:37,  2.66it/s]
 11%|█         | 1160/10682 [23:06<59:36,  2.66it/s]
 11%|█         | 1161/10682 [23:06<59:37,  2.66it/s]
 11%|█         | 1162/10682 [23:07<59:36,  2.66it/s]
 11%|█         | 1163/10682 [23:07<59:37,  2.66it/s]
 11%|█         | 1164/10682 [23:08<59:40,  2.66it/s]
 11%|█         | 1165/10682 [23:08<59:36,  2.66it/s]
 11%|█         | 1166/10682 [23:08<59:41,  2.66it/s]
 11%|█         | 1167/10682 [23:09<59:35,  2.66it/s]
 11%|█         | 1168/10682 [23:09<59:33,  2.66it/s]
 11%|█         | 1169/10682 [23:09<59:28,  2.67it/s]
 11%|█         | 1170/10682 [23:10<59:29,  2.66it/s]
 11%|█         | 1171/10682 [23:10<59:26,  2.67it/s]
 11%|█         | 1172/10682 [23:11<59:27,  2.67it/s]
 11%|█         | 1173/10682 [23:11<59:36,  2.66it/s]
 11%|█         | 1174/10682 [23:11<59:30,  2.66it/s]
 11%|█         | 1175/10682 [23:12<59:26,  2.67it/s]{'loss': 4.9931, 'grad_norm': 1.1357430219650269, 'learning_rate': 0.0009997000214725905, 'epoch': 0.11}
-
                                                    

 11%|█         | 1175/10682 [23:12<59:26,  2.67it/s]
 11%|█         | 1176/10682 [23:12<59:33,  2.66it/s]
 11%|█         | 1177/10682 [23:12<59:26,  2.67it/s]
 11%|█         | 1178/10682 [23:13<59:23,  2.67it/s]
 11%|█         | 1179/10682 [23:13<59:27,  2.66it/s]
 11%|█         | 1180/10682 [23:14<59:27,  2.66it/s]
 11%|█         | 1181/10682 [23:14<59:27,  2.66it/s]
 11%|█         | 1182/10682 [23:14<59:26,  2.66it/s]
 11%|█         | 1183/10682 [23:15<59:32,  2.66it/s]
 11%|█         | 1184/10682 [23:15<59:29,  2.66it/s]
 11%|█         | 1185/10682 [23:15<59:27,  2.66it/s]
 11%|█         | 1186/10682 [23:16<59:23,  2.66it/s]
 11%|█         | 1187/10682 [23:16<59:21,  2.67it/s]
 11%|█         | 1188/10682 [23:17<59:27,  2.66it/s]
 11%|█         | 1189/10682 [23:17<59:23,  2.66it/s]
 11%|█         | 1190/10682 [23:17<59:25,  2.66it/s]
 11%|█         | 1191/10682 [23:18<59:22,  2.66it/s]
 11%|█         | 1192/10682 [23:18<59:23,  2.66it/s]
 11%|█         | 1193/10682 [23:18<59:20,  2.66it/s]
 11%|█         | 1194/10682 [23:19<59:20,  2.67it/s]
 11%|█         | 1195/10682 [23:19<59:23,  2.66it/s]
 11%|█         | 1196/10682 [23:20<59:20,  2.66it/s]
 11%|█         | 1197/10682 [23:20<59:26,  2.66it/s]
 11%|█         | 1198/10682 [23:20<59:20,  2.66it/s]
 11%|█         | 1199/10682 [23:21<59:18,  2.66it/s]
 11%|█         | 1200/10682 [23:21<59:17,  2.67it/s]{'loss': 4.9766, 'grad_norm': 0.7919648885726929, 'learning_rate': 0.000999541860089067, 'epoch': 0.11}
-
                                                    

 11%|█         | 1200/10682 [23:21<59:17,  2.67it/s]
 11%|█         | 1201/10682 [23:21<59:20,  2.66it/s]
 11%|█▏        | 1202/10682 [23:22<59:22,  2.66it/s]
 11%|█▏        | 1203/10682 [23:22<59:21,  2.66it/s]
 11%|█▏        | 1204/10682 [23:23<59:26,  2.66it/s]
 11%|█▏        | 1205/10682 [23:23<59:26,  2.66it/s]
 11%|█▏        | 1206/10682 [23:23<59:18,  2.66it/s]
 11%|█▏        | 1207/10682 [23:24<59:20,  2.66it/s]
 11%|█▏        | 1208/10682 [23:24<59:14,  2.67it/s]
 11%|█▏        | 1209/10682 [23:24<59:19,  2.66it/s]
 11%|█▏        | 1210/10682 [23:25<59:15,  2.66it/s]
 11%|█▏        | 1211/10682 [23:25<59:11,  2.67it/s]
 11%|█▏        | 1212/10682 [23:26<59:09,  2.67it/s]
 11%|█▏        | 1213/10682 [23:26<59:06,  2.67it/s]
 11%|█▏        | 1214/10682 [23:26<59:04,  2.67it/s]
 11%|█▏        | 1215/10682 [23:27<59:04,  2.67it/s]
 11%|█▏        | 1216/10682 [23:27<59:11,  2.67it/s]
 11%|█▏        | 1217/10682 [23:27<59:10,  2.67it/s]
 11%|█▏        | 1218/10682 [23:28<59:12,  2.66it/s]
 11%|█▏        | 1219/10682 [23:28<59:16,  2.66it/s]
 11%|█▏        | 1220/10682 [23:29<59:13,  2.66it/s]
 11%|█▏        | 1221/10682 [23:29<59:15,  2.66it/s]
 11%|█▏        | 1222/10682 [23:29<59:14,  2.66it/s]
 11%|█▏        | 1223/10682 [23:30<59:13,  2.66it/s]
 11%|█▏        | 1224/10682 [23:30<59:15,  2.66it/s]
 11%|█▏        | 1225/10682 [23:30<59:14,  2.66it/s]
                                                    
{'loss': 4.9557, 'grad_norm': 0.8168599009513855, 'learning_rate': 0.0009993503536573894, 'epoch': 0.11}
-
 11%|█▏        | 1225/10682 [23:30<59:14,  2.66it/s]
 11%|█▏        | 1226/10682 [23:31<59:17,  2.66it/s]
 11%|█▏        | 1227/10682 [23:31<59:10,  2.66it/s]
 11%|█▏        | 1228/10682 [23:32<59:10,  2.66it/s]
 12%|█▏        | 1229/10682 [23:32<59:05,  2.67it/s]
 12%|█▏        | 1230/10682 [23:32<59:08,  2.66it/s]
 12%|█▏        | 1231/10682 [23:33<59:06,  2.66it/s]
 12%|█▏        | 1232/10682 [23:33<59:06,  2.66it/s]
 12%|█▏        | 1233/10682 [23:33<59:07,  2.66it/s]
 12%|█▏        | 1234/10682 [23:34<59:04,  2.67it/s]
 12%|█▏        | 1235/10682 [23:34<59:20,  2.65it/s]
 12%|█▏        | 1236/10682 [23:35<59:18,  2.65it/s]
 12%|█▏        | 1237/10682 [23:35<59:13,  2.66it/s]
 12%|█▏        | 1238/10682 [23:35<59:09,  2.66it/s]
 12%|█▏        | 1239/10682 [23:36<59:08,  2.66it/s]
 12%|█▏        | 1240/10682 [23:36<59:05,  2.66it/s]
 12%|█▏        | 1241/10682 [23:36<59:02,  2.67it/s]
 12%|█▏        | 1242/10682 [23:37<59:01,  2.67it/s]
 12%|█▏        | 1243/10682 [23:37<58:58,  2.67it/s]
 12%|█▏        | 1244/10682 [23:38<58:59,  2.67it/s]
 12%|█▏        | 1245/10682 [23:38<59:00,  2.67it/s]
 12%|█▏        | 1246/10682 [23:38<58:58,  2.67it/s]
 12%|█▏        | 1247/10682 [23:39<59:10,  2.66it/s]
 12%|█▏        | 1248/10682 [23:39<59:13,  2.66it/s]
 12%|█▏        | 1249/10682 [23:39<59:08,  2.66it/s]
 12%|█▏        | 1250/10682 [23:40<59:04,  2.66it/s]
                                                    
{'loss': 4.9388, 'grad_norm': 0.7113502621650696, 'learning_rate': 0.0009991255149608528, 'epoch': 0.12}
-
 12%|█▏        | 1250/10682 [23:40<59:04,  2.66it/s]
 12%|█▏        | 1251/10682 [23:40<59:06,  2.66it/s]
 12%|█▏        | 1252/10682 [23:41<59:07,  2.66it/s]
 12%|█▏        | 1253/10682 [23:41<59:02,  2.66it/s]
 12%|█▏        | 1254/10682 [23:41<58:59,  2.66it/s]
 12%|█▏        | 1255/10682 [23:42<58:58,  2.66it/s]
 12%|█▏        | 1256/10682 [23:42<58:55,  2.67it/s]
 12%|█▏        | 1257/10682 [23:42<58:51,  2.67it/s]
 12%|█▏        | 1258/10682 [23:43<58:57,  2.66it/s]
 12%|█▏        | 1259/10682 [23:43<58:58,  2.66it/s]
 12%|█▏        | 1260/10682 [23:44<58:52,  2.67it/s]
 12%|█▏        | 1261/10682 [23:44<59:02,  2.66it/s]
 12%|█▏        | 1262/10682 [23:44<58:56,  2.66it/s]
 12%|█▏        | 1263/10682 [23:45<58:56,  2.66it/s]
 12%|█▏        | 1264/10682 [23:45<58:52,  2.67it/s]
 12%|█▏        | 1265/10682 [23:45<58:52,  2.67it/s]
 12%|█▏        | 1266/10682 [23:46<58:55,  2.66it/s]
 12%|█▏        | 1267/10682 [23:46<58:57,  2.66it/s]
 12%|█▏        | 1268/10682 [23:47<58:57,  2.66it/s]
 12%|█▏        | 1269/10682 [23:47<58:55,  2.66it/s]
 12%|█▏        | 1270/10682 [23:47<58:59,  2.66it/s]
 12%|█▏        | 1271/10682 [23:48<58:54,  2.66it/s]
 12%|█▏        | 1272/10682 [23:48<58:59,  2.66it/s]
 12%|█▏        | 1273/10682 [23:48<58:57,  2.66it/s]
 12%|█▏        | 1274/10682 [23:49<58:53,  2.66it/s]
 12%|█▏        | 1275/10682 [23:49<58:48,  2.67it/s]{'loss': 4.9207, 'grad_norm': 0.7155961394309998, 'learning_rate': 0.0009988673590077238, 'epoch': 0.12}
-
                                                    

 12%|█▏        | 1275/10682 [23:49<58:48,  2.67it/s]
 12%|█▏        | 1276/10682 [23:50<58:52,  2.66it/s]
 12%|█▏        | 1277/10682 [23:50<58:52,  2.66it/s]
 12%|█▏        | 1278/10682 [23:50<58:48,  2.67it/s]
 12%|█▏        | 1279/10682 [23:51<58:46,  2.67it/s]
 12%|█▏        | 1280/10682 [23:51<58:46,  2.67it/s]
 12%|█▏        | 1281/10682 [23:51<58:44,  2.67it/s]
 12%|█▏        | 1282/10682 [23:52<58:43,  2.67it/s]
 12%|█▏        | 1283/10682 [23:52<58:40,  2.67it/s]
 12%|█▏        | 1284/10682 [23:53<58:41,  2.67it/s]
 12%|█▏        | 1285/10682 [23:53<58:43,  2.67it/s]
 12%|█▏        | 1286/10682 [23:53<58:47,  2.66it/s]
 12%|█▏        | 1287/10682 [23:54<58:49,  2.66it/s]
 12%|█▏        | 1288/10682 [23:54<58:51,  2.66it/s]
 12%|█▏        | 1289/10682 [23:54<58:55,  2.66it/s]
 12%|█▏        | 1290/10682 [23:55<58:52,  2.66it/s]
 12%|█▏        | 1291/10682 [23:55<58:51,  2.66it/s]
 12%|█▏        | 1292/10682 [23:56<58:43,  2.66it/s]
 12%|█▏        | 1293/10682 [23:56<58:43,  2.66it/s]
 12%|█▏        | 1294/10682 [23:56<58:43,  2.66it/s]
 12%|█▏        | 1295/10682 [23:57<58:38,  2.67it/s]
 12%|█▏        | 1296/10682 [23:57<58:44,  2.66it/s]
 12%|█▏        | 1297/10682 [23:57<58:40,  2.67it/s]
 12%|█▏        | 1298/10682 [23:58<58:38,  2.67it/s]
 12%|█▏        | 1299/10682 [23:58<58:36,  2.67it/s]
 12%|█▏        | 1300/10682 [23:59<58:43,  2.66it/s]
                                                    {'loss': 4.9043, 'grad_norm': 0.6650585532188416, 'learning_rate': 0.0009985759030302366, 'epoch': 0.12}
-

 12%|█▏        | 1300/10682 [23:59<58:43,  2.66it/s]
 12%|█▏        | 1301/10682 [23:59<58:45,  2.66it/s]
 12%|█▏        | 1302/10682 [23:59<58:47,  2.66it/s]
 12%|█▏        | 1303/10682 [24:00<58:53,  2.65it/s]
 12%|█▏        | 1304/10682 [24:00<58:49,  2.66it/s]
 12%|█▏        | 1305/10682 [24:00<58:51,  2.66it/s]
 12%|█▏        | 1306/10682 [24:01<58:48,  2.66it/s]
 12%|█▏        | 1307/10682 [24:01<58:47,  2.66it/s]
 12%|█▏        | 1308/10682 [24:02<58:42,  2.66it/s]
 12%|█▏        | 1309/10682 [24:02<58:39,  2.66it/s]
 12%|█▏        | 1310/10682 [24:02<58:39,  2.66it/s]
 12%|█▏        | 1311/10682 [24:03<58:38,  2.66it/s]
 12%|█▏        | 1312/10682 [24:03<58:41,  2.66it/s]
 12%|█▏        | 1313/10682 [24:03<58:40,  2.66it/s]
 12%|█▏        | 1314/10682 [24:04<58:44,  2.66it/s]
 12%|█▏        | 1315/10682 [24:04<58:37,  2.66it/s]
 12%|█▏        | 1316/10682 [24:05<58:44,  2.66it/s]
 12%|█▏        | 1317/10682 [24:05<58:45,  2.66it/s]
 12%|█▏        | 1318/10682 [24:05<58:47,  2.65it/s]
 12%|█▏        | 1319/10682 [24:06<58:39,  2.66it/s]
 12%|█▏        | 1320/10682 [24:06<58:34,  2.66it/s]
 12%|█▏        | 1321/10682 [24:06<58:33,  2.66it/s]
 12%|█▏        | 1322/10682 [24:07<58:33,  2.66it/s]
 12%|█▏        | 1323/10682 [24:07<58:31,  2.67it/s]
 12%|█▏        | 1324/10682 [24:08<58:33,  2.66it/s]
 12%|█▏        | 1325/10682 [24:08<58:33,  2.66it/s]{'loss': 4.8967, 'grad_norm': 1.208127737045288, 'learning_rate': 0.0009982511664834452, 'epoch': 0.12}
-
                                                    

 12%|█▏        | 1325/10682 [24:08<58:33,  2.66it/s]
 12%|█▏        | 1326/10682 [24:08<58:39,  2.66it/s]
 12%|█▏        | 1327/10682 [24:09<58:33,  2.66it/s]
 12%|█▏        | 1328/10682 [24:09<58:37,  2.66it/s]
 12%|█▏        | 1329/10682 [24:09<58:35,  2.66it/s]
 12%|█▏        | 1330/10682 [24:10<58:37,  2.66it/s]
 12%|█▏        | 1331/10682 [24:10<58:35,  2.66it/s]
 12%|█▏        | 1332/10682 [24:11<58:31,  2.66it/s]
 12%|█▏        | 1333/10682 [24:11<58:27,  2.67it/s]
 12%|█▏        | 1334/10682 [24:11<58:26,  2.67it/s]
 12%|█▏        | 1335/10682 [24:12<58:24,  2.67it/s]
 13%|█▎        | 1336/10682 [24:12<58:30,  2.66it/s]
 13%|█▎        | 1337/10682 [24:12<58:26,  2.67it/s]
 13%|█▎        | 1338/10682 [24:13<58:24,  2.67it/s]
 13%|█▎        | 1339/10682 [24:13<58:21,  2.67it/s]
 13%|█▎        | 1340/10682 [24:14<58:31,  2.66it/s]
 13%|█▎        | 1341/10682 [24:14<58:28,  2.66it/s]
 13%|█▎        | 1342/10682 [24:14<58:32,  2.66it/s]
 13%|█▎        | 1343/10682 [24:15<58:28,  2.66it/s]
 13%|█▎        | 1344/10682 [24:15<58:29,  2.66it/s]
 13%|█▎        | 1345/10682 [24:15<58:29,  2.66it/s]
 13%|█▎        | 1346/10682 [24:16<58:29,  2.66it/s]
 13%|█▎        | 1347/10682 [24:16<58:31,  2.66it/s]
 13%|█▎        | 1348/10682 [24:17<58:29,  2.66it/s]
 13%|█▎        | 1349/10682 [24:17<58:29,  2.66it/s]
 13%|█▎        | 1350/10682 [24:17<58:26,  2.66it/s]{'loss': 4.8817, 'grad_norm': 0.8315516114234924, 'learning_rate': 0.0009978931710439229, 'epoch': 0.13}
                                                    
-
 13%|█▎        | 1350/10682 [24:17<58:26,  2.66it/s]
 13%|█▎        | 1351/10682 [24:18<58:29,  2.66it/s]
 13%|█▎        | 1352/10682 [24:18<58:24,  2.66it/s]
 13%|█▎        | 1353/10682 [24:18<58:21,  2.66it/s]
 13%|█▎        | 1354/10682 [24:19<58:19,  2.67it/s]
 13%|█▎        | 1355/10682 [24:19<58:16,  2.67it/s]
 13%|█▎        | 1356/10682 [24:20<58:24,  2.66it/s]
 13%|█▎        | 1357/10682 [24:20<58:22,  2.66it/s]
 13%|█▎        | 1358/10682 [24:20<58:23,  2.66it/s]
 13%|█▎        | 1359/10682 [24:21<58:20,  2.66it/s]
 13%|█▎        | 1360/10682 [24:21<58:21,  2.66it/s]
 13%|█▎        | 1361/10682 [24:21<58:26,  2.66it/s]
 13%|█▎        | 1362/10682 [24:22<58:22,  2.66it/s]
 13%|█▎        | 1363/10682 [24:22<58:16,  2.67it/s]
 13%|█▎        | 1364/10682 [24:23<58:15,  2.67it/s]
 13%|█▎        | 1365/10682 [24:23<58:12,  2.67it/s]
 13%|█▎        | 1366/10682 [24:23<58:15,  2.67it/s]
 13%|█▎        | 1367/10682 [24:24<58:15,  2.66it/s]
 13%|█▎        | 1368/10682 [24:24<58:18,  2.66it/s]
 13%|█▎        | 1369/10682 [24:24<58:15,  2.66it/s]
 13%|█▎        | 1370/10682 [24:25<58:19,  2.66it/s]
 13%|█▎        | 1371/10682 [24:25<58:18,  2.66it/s]
 13%|█▎        | 1372/10682 [24:26<58:16,  2.66it/s]
 13%|█▎        | 1373/10682 [24:26<58:20,  2.66it/s]
 13%|█▎        | 1374/10682 [24:26<58:18,  2.66it/s]
 13%|█▎        | 1375/10682 [24:27<58:14,  2.66it/s]{'loss': 4.8707, 'grad_norm': 0.6487434506416321, 'learning_rate': 0.000997501940608316, 'epoch': 0.13}
                                                    
-
 13%|█▎        | 1375/10682 [24:27<58:14,  2.66it/s]
 13%|█▎        | 1376/10682 [24:27<58:23,  2.66it/s]
 13%|█▎        | 1377/10682 [24:28<58:18,  2.66it/s]
 13%|█▎        | 1378/10682 [24:28<58:15,  2.66it/s]
 13%|█▎        | 1379/10682 [24:28<58:17,  2.66it/s]
 13%|█▎        | 1380/10682 [24:29<58:16,  2.66it/s]
 13%|█▎        | 1381/10682 [24:29<58:09,  2.67it/s]
 13%|█▎        | 1382/10682 [24:29<58:06,  2.67it/s]
 13%|█▎        | 1383/10682 [24:30<58:05,  2.67it/s]
 13%|█▎        | 1384/10682 [24:30<58:08,  2.67it/s]
 13%|█▎        | 1385/10682 [24:31<58:07,  2.67it/s]
 13%|█▎        | 1386/10682 [24:31<58:18,  2.66it/s]
 13%|█▎        | 1387/10682 [24:31<58:12,  2.66it/s]
 13%|█▎        | 1388/10682 [24:32<58:11,  2.66it/s]
 13%|█▎        | 1389/10682 [24:32<58:06,  2.67it/s]
 13%|█▎        | 1390/10682 [24:32<58:16,  2.66it/s]
 13%|█▎        | 1391/10682 [24:33<58:12,  2.66it/s]
 13%|█▎        | 1392/10682 [24:33<58:12,  2.66it/s]
 13%|█▎        | 1393/10682 [24:34<58:07,  2.66it/s]
 13%|█▎        | 1394/10682 [24:34<58:03,  2.67it/s]
 13%|█▎        | 1395/10682 [24:34<58:01,  2.67it/s]
 13%|█▎        | 1396/10682 [24:35<58:00,  2.67it/s]
 13%|█▎        | 1397/10682 [24:35<58:04,  2.66it/s]
 13%|█▎        | 1398/10682 [24:35<58:05,  2.66it/s]
 13%|█▎        | 1399/10682 [24:36<58:02,  2.67it/s]
 13%|█▎        | 1400/10682 [24:36<58:13,  2.66it/s]
                                                    
{'loss': 4.8555, 'grad_norm': 0.6858766674995422, 'learning_rate': 0.0009970775012917489, 'epoch': 0.13}
-
 13%|█▎        | 1400/10682 [24:36<58:13,  2.66it/s]
 13%|█▎        | 1401/10682 [24:37<58:14,  2.66it/s]
 13%|█▎        | 1402/10682 [24:37<58:08,  2.66it/s]
 13%|█▎        | 1403/10682 [24:37<58:04,  2.66it/s]
 13%|█▎        | 1404/10682 [24:38<58:08,  2.66it/s]
 13%|█▎        | 1405/10682 [24:38<58:04,  2.66it/s]
 13%|█▎        | 1406/10682 [24:38<58:01,  2.66it/s]
 13%|█▎        | 1407/10682 [24:39<58:01,  2.66it/s]
 13%|█▎        | 1408/10682 [24:39<58:01,  2.66it/s]
 13%|█▎        | 1409/10682 [24:40<58:06,  2.66it/s]
 13%|█▎        | 1410/10682 [24:40<58:03,  2.66it/s]
 13%|█▎        | 1411/10682 [24:40<58:14,  2.65it/s]
 13%|█▎        | 1412/10682 [24:41<58:09,  2.66it/s]
 13%|█▎        | 1413/10682 [24:41<58:07,  2.66it/s]
 13%|█▎        | 1414/10682 [24:41<58:06,  2.66it/s]
 13%|█▎        | 1415/10682 [24:42<58:01,  2.66it/s]
 13%|█▎        | 1416/10682 [24:42<57:59,  2.66it/s]
 13%|█▎        | 1417/10682 [24:43<57:56,  2.66it/s]
 13%|█▎        | 1418/10682 [24:43<58:07,  2.66it/s]
 13%|█▎        | 1419/10682 [24:43<58:02,  2.66it/s]
 13%|█▎        | 1420/10682 [24:44<58:00,  2.66it/s]
 13%|█▎        | 1421/10682 [24:44<57:55,  2.66it/s]
 13%|█▎        | 1422/10682 [24:44<57:55,  2.66it/s]
 13%|█▎        | 1423/10682 [24:45<57:57,  2.66it/s]
 13%|█▎        | 1424/10682 [24:45<57:56,  2.66it/s]
 13%|█▎        | 1425/10682 [24:46<57:55,  2.66it/s]{'loss': 4.8315, 'grad_norm': 0.7808846831321716, 'learning_rate': 0.00099661988142608, 'epoch': 0.13}
-
                                                    

 13%|█▎        | 1425/10682 [24:46<57:55,  2.66it/s]
 13%|█▎        | 1426/10682 [24:46<57:58,  2.66it/s]
 13%|█▎        | 1427/10682 [24:46<57:57,  2.66it/s]
 13%|█▎        | 1428/10682 [24:47<58:03,  2.66it/s]
 13%|█▎        | 1429/10682 [24:47<58:05,  2.65it/s]
 13%|█▎        | 1430/10682 [24:47<58:00,  2.66it/s]
 13%|█▎        | 1431/10682 [24:48<57:55,  2.66it/s]
 13%|█▎        | 1432/10682 [24:48<57:56,  2.66it/s]
 13%|█▎        | 1433/10682 [24:49<57:52,  2.66it/s]
 13%|█▎        | 1434/10682 [24:49<57:52,  2.66it/s]
 13%|█▎        | 1435/10682 [24:49<57:51,  2.66it/s]
 13%|█▎        | 1436/10682 [24:50<57:47,  2.67it/s]
 13%|█▎        | 1437/10682 [24:50<57:48,  2.67it/s]
 13%|█▎        | 1438/10682 [24:50<57:47,  2.67it/s]
 13%|█▎        | 1439/10682 [24:51<57:52,  2.66it/s]
 13%|█▎        | 1440/10682 [24:51<57:47,  2.67it/s]
 13%|█▎        | 1441/10682 [24:52<57:47,  2.67it/s]
 13%|█▎        | 1442/10682 [24:52<57:47,  2.66it/s]
 14%|█▎        | 1443/10682 [24:52<57:48,  2.66it/s]
 14%|█▎        | 1444/10682 [24:53<57:46,  2.66it/s]
 14%|█▎        | 1445/10682 [24:53<57:49,  2.66it/s]
 14%|█▎        | 1446/10682 [24:53<57:48,  2.66it/s]
 14%|█▎        | 1447/10682 [24:54<57:42,  2.67it/s]
 14%|█▎        | 1448/10682 [24:54<57:44,  2.67it/s]
 14%|█▎        | 1449/10682 [24:55<57:44,  2.67it/s]
 14%|█▎        | 1450/10682 [24:55<57:41,  2.67it/s]
                                                    
{'loss': 4.8258, 'grad_norm': 0.5850582122802734, 'learning_rate': 0.0009961291115580116, 'epoch': 0.14}
-
 14%|█▎        | 1450/10682 [24:55<57:41,  2.67it/s]
 14%|█▎        | 1451/10682 [24:55<57:42,  2.67it/s]
 14%|█▎        | 1452/10682 [24:56<57:40,  2.67it/s]
 14%|█▎        | 1453/10682 [24:56<57:44,  2.66it/s]
 14%|█▎        | 1454/10682 [24:56<57:42,  2.67it/s]
 14%|█▎        | 1455/10682 [24:57<57:48,  2.66it/s]
 14%|█▎        | 1456/10682 [24:57<57:45,  2.66it/s]
 14%|█▎        | 1457/10682 [24:58<57:46,  2.66it/s]
 14%|█▎        | 1458/10682 [24:58<57:43,  2.66it/s]
 14%|█▎        | 1459/10682 [24:58<57:45,  2.66it/s]
 14%|█▎        | 1460/10682 [24:59<57:39,  2.67it/s]
 14%|█▎        | 1461/10682 [24:59<57:40,  2.66it/s]
 14%|█▎        | 1462/10682 [24:59<57:38,  2.67it/s]
 14%|█▎        | 1463/10682 [25:00<57:34,  2.67it/s]
 14%|█▎        | 1464/10682 [25:00<57:34,  2.67it/s]
 14%|█▎        | 1465/10682 [25:01<57:35,  2.67it/s]
 14%|█▎        | 1466/10682 [25:01<57:36,  2.67it/s]
 14%|█▎        | 1467/10682 [25:01<57:35,  2.67it/s]
 14%|█▎        | 1468/10682 [25:02<57:34,  2.67it/s]
 14%|█▍        | 1469/10682 [25:02<57:45,  2.66it/s]
 14%|█▍        | 1470/10682 [25:02<57:40,  2.66it/s]
 14%|█▍        | 1471/10682 [25:03<57:39,  2.66it/s]
 14%|█▍        | 1472/10682 [25:03<57:34,  2.67it/s]
 14%|█▍        | 1473/10682 [25:04<57:38,  2.66it/s]
 14%|█▍        | 1474/10682 [25:04<57:41,  2.66it/s]
 14%|█▍        | 1475/10682 [25:04<57:44,  2.66it/s]
                                                    
{'loss': 4.8133, 'grad_norm': 0.7012140154838562, 'learning_rate': 0.0009956052244470509, 'epoch': 0.14}
-
 14%|█▍        | 1475/10682 [25:04<57:44,  2.66it/s]
 14%|█▍        | 1476/10682 [25:05<57:45,  2.66it/s]
 14%|█▍        | 1477/10682 [25:05<57:38,  2.66it/s]
 14%|█▍        | 1478/10682 [25:05<57:34,  2.66it/s]
 14%|█▍        | 1479/10682 [25:06<57:32,  2.67it/s]
 14%|█▍        | 1480/10682 [25:06<57:39,  2.66it/s]
 14%|█▍        | 1481/10682 [25:07<57:34,  2.66it/s]
 14%|█▍        | 1482/10682 [25:07<57:39,  2.66it/s]
 14%|█▍        | 1483/10682 [25:07<57:39,  2.66it/s]
 14%|█▍        | 1484/10682 [25:08<57:34,  2.66it/s]
 14%|█▍        | 1485/10682 [25:08<57:34,  2.66it/s]
 14%|█▍        | 1486/10682 [25:08<57:30,  2.66it/s]
 14%|█▍        | 1487/10682 [25:09<57:35,  2.66it/s]
 14%|█▍        | 1488/10682 [25:09<57:33,  2.66it/s]
 14%|█▍        | 1489/10682 [25:10<57:35,  2.66it/s]
 14%|█▍        | 1490/10682 [25:10<57:32,  2.66it/s]
 14%|█▍        | 1491/10682 [25:10<57:32,  2.66it/s]
 14%|█▍        | 1492/10682 [25:11<57:31,  2.66it/s]
 14%|█▍        | 1493/10682 [25:11<57:28,  2.66it/s]
 14%|█▍        | 1494/10682 [25:11<57:37,  2.66it/s]
 14%|█▍        | 1495/10682 [25:12<57:31,  2.66it/s]
 14%|█▍        | 1496/10682 [25:12<57:39,  2.66it/s]
 14%|█▍        | 1497/10682 [25:13<57:35,  2.66it/s]
 14%|█▍        | 1498/10682 [25:13<57:32,  2.66it/s]
 14%|█▍        | 1499/10682 [25:13<57:31,  2.66it/s]
 14%|█▍        | 1500/10682 [25:14<57:34,  2.66it/s]{'loss': 4.7966, 'grad_norm': 0.5960355997085571, 'learning_rate': 0.0009950482550633215, 'epoch': 0.14}
                                                    
-
 14%|█▍        | 1500/10682 [25:14<57:34,  2.66it/s]
 14%|█▍        | 1501/10682 [25:14<57:36,  2.66it/s]
 14%|█▍        | 1502/10682 [25:14<57:32,  2.66it/s]
 14%|█▍        | 1503/10682 [25:15<57:30,  2.66it/s]
 14%|█▍        | 1504/10682 [25:15<57:27,  2.66it/s]
 14%|█▍        | 1505/10682 [25:16<57:24,  2.66it/s]
 14%|█▍        | 1506/10682 [25:16<57:22,  2.67it/s]
 14%|█▍        | 1507/10682 [25:16<57:23,  2.66it/s]
 14%|█▍        | 1508/10682 [25:17<57:23,  2.66it/s]
 14%|█▍        | 1509/10682 [25:17<57:24,  2.66it/s]
 14%|█▍        | 1510/10682 [25:17<57:32,  2.66it/s]
 14%|█▍        | 1511/10682 [25:18<57:28,  2.66it/s]
 14%|█▍        | 1512/10682 [25:18<57:25,  2.66it/s]
 14%|█▍        | 1513/10682 [25:19<57:21,  2.66it/s]
 14%|█▍        | 1514/10682 [25:19<57:28,  2.66it/s]
 14%|█▍        | 1515/10682 [25:19<57:23,  2.66it/s]
 14%|█▍        | 1516/10682 [25:20<57:19,  2.66it/s]
 14%|█▍        | 1517/10682 [25:20<57:20,  2.66it/s]
 14%|█▍        | 1518/10682 [25:20<57:18,  2.66it/s]
 14%|█▍        | 1519/10682 [25:21<57:18,  2.66it/s]
 14%|█▍        | 1520/10682 [25:21<57:16,  2.67it/s]
 14%|█▍        | 1521/10682 [25:22<57:21,  2.66it/s]
 14%|█▍        | 1522/10682 [25:22<57:19,  2.66it/s]
 14%|█▍        | 1523/10682 [25:22<57:14,  2.67it/s]
 14%|█▍        | 1524/10682 [25:23<57:23,  2.66it/s]
 14%|█▍        | 1525/10682 [25:23<57:29,  2.65it/s]{'loss': 4.7864, 'grad_norm': 0.6349527835845947, 'learning_rate': 0.0009944582405852314, 'epoch': 0.14}
-
                                                    

 14%|█▍        | 1525/10682 [25:23<57:29,  2.65it/s]
 14%|█▍        | 1526/10682 [25:23<57:34,  2.65it/s]
 14%|█▍        | 1527/10682 [25:24<57:29,  2.65it/s]
 14%|█▍        | 1528/10682 [25:24<57:27,  2.66it/s]
 14%|█▍        | 1529/10682 [25:25<57:25,  2.66it/s]
 14%|█▍        | 1530/10682 [25:25<57:21,  2.66it/s]
 14%|█▍        | 1531/10682 [25:25<57:23,  2.66it/s]
 14%|█▍        | 1532/10682 [25:26<57:31,  2.65it/s]
 14%|█▍        | 1533/10682 [25:26<57:27,  2.65it/s]
 14%|█▍        | 1534/10682 [25:26<57:19,  2.66it/s]
 14%|█▍        | 1535/10682 [25:27<57:21,  2.66it/s]
 14%|█▍        | 1536/10682 [25:27<57:18,  2.66it/s]
 14%|█▍        | 1537/10682 [25:28<57:19,  2.66it/s]
 14%|█▍        | 1538/10682 [25:28<57:16,  2.66it/s]
 14%|█▍        | 1539/10682 [25:28<57:18,  2.66it/s]
 14%|█▍        | 1540/10682 [25:29<57:17,  2.66it/s]
 14%|█▍        | 1541/10682 [25:29<57:14,  2.66it/s]
 14%|█▍        | 1542/10682 [25:29<57:14,  2.66it/s]
 14%|█▍        | 1543/10682 [25:30<57:11,  2.66it/s]
 14%|█▍        | 1544/10682 [25:30<57:09,  2.66it/s]
 14%|█▍        | 1545/10682 [25:31<57:09,  2.66it/s]
 14%|█▍        | 1546/10682 [25:31<57:05,  2.67it/s]
 14%|█▍        | 1547/10682 [25:31<57:02,  2.67it/s]
 14%|█▍        | 1548/10682 [25:32<57:06,  2.67it/s]
 15%|█▍        | 1549/10682 [25:32<57:10,  2.66it/s]
 15%|█▍        | 1550/10682 [25:32<57:05,  2.67it/s]
                                                    
{'loss': 4.7715, 'grad_norm': 0.6615831255912781, 'learning_rate': 0.0009938352203969901, 'epoch': 0.15}
-
 15%|█▍        | 1550/10682 [25:32<57:05,  2.67it/s]
 15%|█▍        | 1551/10682 [25:33<57:15,  2.66it/s]
 15%|█▍        | 1552/10682 [25:33<57:11,  2.66it/s]
 15%|█▍        | 1553/10682 [25:34<57:09,  2.66it/s]
 15%|█▍        | 1554/10682 [25:34<57:03,  2.67it/s]
 15%|█▍        | 1555/10682 [25:34<57:06,  2.66it/s]
 15%|█▍        | 1556/10682 [25:35<57:02,  2.67it/s]
 15%|█▍        | 1557/10682 [25:35<57:02,  2.67it/s]
 15%|█▍        | 1558/10682 [25:35<56:59,  2.67it/s]
 15%|█▍        | 1559/10682 [25:36<56:54,  2.67it/s]
 15%|█▍        | 1560/10682 [25:36<56:58,  2.67it/s]
 15%|█▍        | 1561/10682 [25:37<56:58,  2.67it/s]
 15%|█▍        | 1562/10682 [25:37<57:01,  2.67it/s]
 15%|█▍        | 1563/10682 [25:37<57:00,  2.67it/s]
 15%|█▍        | 1564/10682 [25:38<56:59,  2.67it/s]
 15%|█▍        | 1565/10682 [25:38<57:01,  2.66it/s]
 15%|█▍        | 1566/10682 [25:38<56:58,  2.67it/s]
 15%|█▍        | 1567/10682 [25:39<56:56,  2.67it/s]
 15%|█▍        | 1568/10682 [25:39<56:56,  2.67it/s]
 15%|█▍        | 1569/10682 [25:40<56:59,  2.67it/s]
 15%|█▍        | 1570/10682 [25:40<56:57,  2.67it/s]
 15%|█▍        | 1571/10682 [25:40<56:59,  2.66it/s]
 15%|█▍        | 1572/10682 [25:41<56:59,  2.66it/s]
 15%|█▍        | 1573/10682 [25:41<56:57,  2.67it/s]
 15%|█▍        | 1574/10682 [25:41<56:57,  2.67it/s]
 15%|█▍        | 1575/10682 [25:42<56:56,  2.67it/s]
                                                    
{'loss': 4.7737, 'grad_norm': 0.6493037939071655, 'learning_rate': 0.0009931792360859792, 'epoch': 0.15}
-
 15%|█▍        | 1575/10682 [25:42<56:56,  2.67it/s]
 15%|█▍        | 1576/10682 [25:42<57:07,  2.66it/s]
 15%|█▍        | 1577/10682 [25:43<57:01,  2.66it/s]
 15%|█▍        | 1578/10682 [25:43<57:04,  2.66it/s]
 15%|█▍        | 1579/10682 [25:43<57:00,  2.66it/s]
 15%|█▍        | 1580/10682 [25:44<56:57,  2.66it/s]
 15%|█▍        | 1581/10682 [25:44<56:55,  2.66it/s]
 15%|█▍        | 1582/10682 [25:44<56:54,  2.66it/s]
 15%|█▍        | 1583/10682 [25:45<56:58,  2.66it/s]
 15%|█▍        | 1584/10682 [25:45<56:58,  2.66it/s]
 15%|█▍        | 1585/10682 [25:46<56:55,  2.66it/s]
 15%|█▍        | 1586/10682 [25:46<56:54,  2.66it/s]
 15%|█▍        | 1587/10682 [25:46<56:50,  2.67it/s]
 15%|█▍        | 1588/10682 [25:47<56:50,  2.67it/s]
 15%|█▍        | 1589/10682 [25:47<56:49,  2.67it/s]
 15%|█▍        | 1590/10682 [25:48<56:54,  2.66it/s]
 15%|█▍        | 1591/10682 [25:48<56:49,  2.67it/s]
 15%|█▍        | 1592/10682 [25:48<56:54,  2.66it/s]
 15%|█▍        | 1593/10682 [25:49<56:52,  2.66it/s]
 15%|█▍        | 1594/10682 [25:49<56:51,  2.66it/s]
 15%|█▍        | 1595/10682 [25:49<56:51,  2.66it/s]
 15%|█▍        | 1596/10682 [25:50<56:54,  2.66it/s]
 15%|█▍        | 1597/10682 [25:50<56:52,  2.66it/s]
 15%|█▍        | 1598/10682 [25:51<56:50,  2.66it/s]
 15%|█▍        | 1599/10682 [25:51<56:47,  2.67it/s]
 15%|█▍        | 1600/10682 [25:51<56:49,  2.66it/s]
                                                    
{'loss': 4.7611, 'grad_norm': 0.6049810647964478, 'learning_rate': 0.0009924903314399776, 'epoch': 0.15}
-
 15%|█▍        | 1600/10682 [25:51<56:49,  2.66it/s]
 15%|█▍        | 1601/10682 [25:52<56:51,  2.66it/s]
 15%|█▍        | 1602/10682 [25:52<56:53,  2.66it/s]
 15%|█▌        | 1603/10682 [25:52<56:54,  2.66it/s]
 15%|█▌        | 1604/10682 [25:53<56:55,  2.66it/s]
 15%|█▌        | 1605/10682 [25:53<56:51,  2.66it/s]
 15%|█▌        | 1606/10682 [25:54<56:55,  2.66it/s]
 15%|█▌        | 1607/10682 [25:54<56:50,  2.66it/s]
 15%|█▌        | 1608/10682 [25:54<56:51,  2.66it/s]
 15%|█▌        | 1609/10682 [25:55<56:47,  2.66it/s]
 15%|█▌        | 1610/10682 [25:55<56:44,  2.66it/s]
 15%|█▌        | 1611/10682 [25:55<56:42,  2.67it/s]
 15%|█▌        | 1612/10682 [25:56<56:43,  2.67it/s]
 15%|█▌        | 1613/10682 [25:56<56:49,  2.66it/s]
 15%|█▌        | 1614/10682 [25:57<56:46,  2.66it/s]
 15%|█▌        | 1615/10682 [25:57<56:50,  2.66it/s]
 15%|█▌        | 1616/10682 [25:57<56:46,  2.66it/s]
 15%|█▌        | 1617/10682 [25:58<56:53,  2.66it/s]
 15%|█▌        | 1618/10682 [25:58<56:52,  2.66it/s]
 15%|█▌        | 1619/10682 [25:58<57:05,  2.65it/s]
 15%|█▌        | 1620/10682 [25:59<56:58,  2.65it/s]
 15%|█▌        | 1621/10682 [25:59<56:56,  2.65it/s]
 15%|█▌        | 1622/10682 [26:00<56:50,  2.66it/s]
 15%|█▌        | 1623/10682 [26:00<56:46,  2.66it/s]
 15%|█▌        | 1624/10682 [26:00<56:45,  2.66it/s]
 15%|█▌        | 1625/10682 [26:01<56:45,  2.66it/s]
                                                    
{'loss': 4.7484, 'grad_norm': 0.5885592699050903, 'learning_rate': 0.0009917685524442382, 'epoch': 0.15}
-
 15%|█▌        | 1625/10682 [26:01<56:45,  2.66it/s]
 15%|█▌        | 1626/10682 [26:01<56:43,  2.66it/s]
 15%|█▌        | 1627/10682 [26:01<56:44,  2.66it/s]
 15%|█▌        | 1628/10682 [26:02<56:41,  2.66it/s]
 15%|█▌        | 1629/10682 [26:02<56:40,  2.66it/s]
 15%|█▌        | 1630/10682 [26:03<56:40,  2.66it/s]
 15%|█▌        | 1631/10682 [26:03<56:44,  2.66it/s]
 15%|█▌        | 1632/10682 [26:03<56:43,  2.66it/s]
 15%|█▌        | 1633/10682 [26:04<56:44,  2.66it/s]
 15%|█▌        | 1634/10682 [26:04<56:41,  2.66it/s]
 15%|█▌        | 1635/10682 [26:04<56:39,  2.66it/s]
 15%|█▌        | 1636/10682 [26:05<56:40,  2.66it/s]
 15%|█▌        | 1637/10682 [26:05<56:33,  2.67it/s]
 15%|█▌        | 1638/10682 [26:06<56:31,  2.67it/s]
 15%|█▌        | 1639/10682 [26:06<56:30,  2.67it/s]
 15%|█▌        | 1640/10682 [26:06<56:27,  2.67it/s]
 15%|█▌        | 1641/10682 [26:07<56:25,  2.67it/s]
 15%|█▌        | 1642/10682 [26:07<56:25,  2.67it/s]
 15%|█▌        | 1643/10682 [26:07<56:27,  2.67it/s]
 15%|█▌        | 1644/10682 [26:08<56:28,  2.67it/s]
 15%|█▌        | 1645/10682 [26:08<56:33,  2.66it/s]
 15%|█▌        | 1646/10682 [26:09<56:29,  2.67it/s]
 15%|█▌        | 1647/10682 [26:09<56:32,  2.66it/s]
 15%|█▌        | 1648/10682 [26:09<56:30,  2.66it/s]
 15%|█▌        | 1649/10682 [26:10<56:27,  2.67it/s]
 15%|█▌        | 1650/10682 [26:10<56:26,  2.67it/s]{'loss': 4.7436, 'grad_norm': 0.597842812538147, 'learning_rate': 0.0009910139472784174, 'epoch': 0.15}
                                                    
-
 15%|█▌        | 1650/10682 [26:10<56:26,  2.67it/s]
 15%|█▌        | 1651/10682 [26:10<56:32,  2.66it/s]
 15%|█▌        | 1652/10682 [26:11<56:28,  2.66it/s]
 15%|█▌        | 1653/10682 [26:11<56:30,  2.66it/s]
 15%|█▌        | 1654/10682 [26:12<56:28,  2.66it/s]
 15%|█▌        | 1655/10682 [26:12<56:26,  2.67it/s]
 16%|█▌        | 1656/10682 [26:12<56:25,  2.67it/s]
 16%|█▌        | 1657/10682 [26:13<56:26,  2.66it/s]
 16%|█▌        | 1658/10682 [26:13<56:35,  2.66it/s]
 16%|█▌        | 1659/10682 [26:13<56:36,  2.66it/s]
 16%|█▌        | 1660/10682 [26:14<56:33,  2.66it/s]
 16%|█▌        | 1661/10682 [26:14<56:35,  2.66it/s]
 16%|█▌        | 1662/10682 [26:15<56:32,  2.66it/s]
 16%|█▌        | 1663/10682 [26:15<56:27,  2.66it/s]
 16%|█▌        | 1664/10682 [26:15<56:25,  2.66it/s]
 16%|█▌        | 1665/10682 [26:16<56:23,  2.67it/s]
 16%|█▌        | 1666/10682 [26:16<56:25,  2.66it/s]
 16%|█▌        | 1667/10682 [26:16<56:22,  2.67it/s]
 16%|█▌        | 1668/10682 [26:17<56:24,  2.66it/s]
 16%|█▌        | 1669/10682 [26:17<56:22,  2.66it/s]
 16%|█▌        | 1670/10682 [26:18<56:23,  2.66it/s]
 16%|█▌        | 1671/10682 [26:18<56:22,  2.66it/s]
 16%|█▌        | 1672/10682 [26:18<56:24,  2.66it/s]
 16%|█▌        | 1673/10682 [26:19<56:25,  2.66it/s]
 16%|█▌        | 1674/10682 [26:19<56:29,  2.66it/s]
 16%|█▌        | 1675/10682 [26:19<56:27,  2.66it/s]{'loss': 4.7303, 'grad_norm': 0.7069166898727417, 'learning_rate': 0.0009902265663133602, 'epoch': 0.16}
                                                    
-
 16%|█▌        | 1675/10682 [26:19<56:27,  2.66it/s]
 16%|█▌        | 1676/10682 [26:20<56:25,  2.66it/s]
 16%|█▌        | 1677/10682 [26:20<56:22,  2.66it/s]
 16%|█▌        | 1678/10682 [26:21<56:24,  2.66it/s]
 16%|█▌        | 1679/10682 [26:21<56:22,  2.66it/s]
 16%|█▌        | 1680/10682 [26:21<56:20,  2.66it/s]
 16%|█▌        | 1681/10682 [26:22<56:14,  2.67it/s]
 16%|█▌        | 1682/10682 [26:22<56:16,  2.67it/s]
 16%|█▌        | 1683/10682 [26:22<56:18,  2.66it/s]
 16%|█▌        | 1684/10682 [26:23<56:19,  2.66it/s]
 16%|█▌        | 1685/10682 [26:23<56:22,  2.66it/s]
 16%|█▌        | 1686/10682 [26:24<56:22,  2.66it/s]
 16%|█▌        | 1687/10682 [26:24<56:19,  2.66it/s]
 16%|█▌        | 1688/10682 [26:24<56:27,  2.65it/s]
 16%|█▌        | 1689/10682 [26:25<56:28,  2.65it/s]
 16%|█▌        | 1690/10682 [26:25<56:29,  2.65it/s]
 16%|█▌        | 1691/10682 [26:25<56:24,  2.66it/s]
 16%|█▌        | 1692/10682 [26:26<56:21,  2.66it/s]
 16%|█▌        | 1693/10682 [26:26<56:18,  2.66it/s]
 16%|█▌        | 1694/10682 [26:27<56:15,  2.66it/s]
 16%|█▌        | 1695/10682 [26:27<56:14,  2.66it/s]
 16%|█▌        | 1696/10682 [26:27<56:11,  2.67it/s]
 16%|█▌        | 1697/10682 [26:28<56:09,  2.67it/s]
 16%|█▌        | 1698/10682 [26:28<56:10,  2.67it/s]
 16%|█▌        | 1699/10682 [26:28<56:13,  2.66it/s]
 16%|█▌        | 1700/10682 [26:29<56:12,  2.66it/s]
                                                    
{'loss': 4.7248, 'grad_norm': 0.6332598328590393, 'learning_rate': 0.0009894064621077375, 'epoch': 0.16}
-
 16%|█▌        | 1700/10682 [26:29<56:12,  2.66it/s]
 16%|█▌        | 1701/10682 [26:29<56:34,  2.65it/s]
 16%|█▌        | 1702/10682 [26:30<56:26,  2.65it/s]
 16%|█▌        | 1703/10682 [26:30<56:21,  2.66it/s]
 16%|█▌        | 1704/10682 [26:30<56:19,  2.66it/s]
 16%|█▌        | 1705/10682 [26:31<56:15,  2.66it/s]
 16%|█▌        | 1706/10682 [26:31<56:15,  2.66it/s]
 16%|█▌        | 1707/10682 [26:31<56:11,  2.66it/s]
 16%|█▌        | 1708/10682 [26:32<56:09,  2.66it/s]
 16%|█▌        | 1709/10682 [26:32<56:10,  2.66it/s]
 16%|█▌        | 1710/10682 [26:33<56:15,  2.66it/s]
 16%|█▌        | 1711/10682 [26:33<56:12,  2.66it/s]
 16%|█▌        | 1712/10682 [26:33<56:15,  2.66it/s]
 16%|█▌        | 1713/10682 [26:34<56:15,  2.66it/s]
 16%|█▌        | 1714/10682 [26:34<56:07,  2.66it/s]
 16%|█▌        | 1715/10682 [26:34<56:09,  2.66it/s]
 16%|█▌        | 1716/10682 [26:35<56:09,  2.66it/s]
 16%|█▌        | 1717/10682 [26:35<56:07,  2.66it/s]
 16%|█▌        | 1718/10682 [26:36<56:08,  2.66it/s]
 16%|█▌        | 1719/10682 [26:36<56:06,  2.66it/s]
 16%|█▌        | 1720/10682 [26:36<56:09,  2.66it/s]
 16%|█▌        | 1721/10682 [26:37<56:12,  2.66it/s]
 16%|█▌        | 1722/10682 [26:37<56:07,  2.66it/s]
 16%|█▌        | 1723/10682 [26:37<56:04,  2.66it/s]
 16%|█▌        | 1724/10682 [26:38<56:02,  2.66it/s]
 16%|█▌        | 1725/10682 [26:38<56:03,  2.66it/s]
                                                    
{'loss': 4.7152, 'grad_norm': 0.6330534219741821, 'learning_rate': 0.0009885536894045377, 'epoch': 0.16}
-
 16%|█▌        | 1725/10682 [26:38<56:03,  2.66it/s]
 16%|█▌        | 1726/10682 [26:39<56:07,  2.66it/s]
 16%|█▌        | 1727/10682 [26:39<56:03,  2.66it/s]
 16%|█▌        | 1728/10682 [26:39<56:01,  2.66it/s]
 16%|█▌        | 1729/10682 [26:40<56:05,  2.66it/s]
 16%|█▌        | 1730/10682 [26:40<56:07,  2.66it/s]
 16%|█▌        | 1731/10682 [26:40<56:07,  2.66it/s]
 16%|█▌        | 1732/10682 [26:41<56:04,  2.66it/s]
 16%|█▌        | 1733/10682 [26:41<56:06,  2.66it/s]
 16%|█▌        | 1734/10682 [26:42<56:05,  2.66it/s]
 16%|█▌        | 1735/10682 [26:42<56:04,  2.66it/s]
 16%|█▋        | 1736/10682 [26:42<56:00,  2.66it/s]
 16%|█▋        | 1737/10682 [26:43<55:57,  2.66it/s]
 16%|█▋        | 1738/10682 [26:43<55:59,  2.66it/s]
 16%|█▋        | 1739/10682 [26:43<55:54,  2.67it/s]
 16%|█▋        | 1740/10682 [26:44<56:02,  2.66it/s]
 16%|█▋        | 1741/10682 [26:44<55:59,  2.66it/s]
 16%|█▋        | 1742/10682 [26:45<56:03,  2.66it/s]
 16%|█▋        | 1743/10682 [26:45<56:00,  2.66it/s]
 16%|█▋        | 1744/10682 [26:45<56:04,  2.66it/s]
 16%|█▋        | 1745/10682 [26:46<56:07,  2.65it/s]
 16%|█▋        | 1746/10682 [26:46<56:06,  2.65it/s]
 16%|█▋        | 1747/10682 [26:47<56:00,  2.66it/s]
 16%|█▋        | 1748/10682 [26:47<56:00,  2.66it/s]
 16%|█▋        | 1749/10682 [26:47<56:01,  2.66it/s]
 16%|█▋        | 1750/10682 [26:48<56:02,  2.66it/s]{'loss': 4.7053, 'grad_norm': 0.6206636428833008, 'learning_rate': 0.0009876683051274124, 'epoch': 0.16}
                                                    
-
 16%|█▋        | 1750/10682 [26:48<56:02,  2.66it/s]
 16%|█▋        | 1751/10682 [26:48<56:05,  2.65it/s]
 16%|█▋        | 1752/10682 [26:48<56:02,  2.66it/s]
 16%|█▋        | 1753/10682 [26:49<55:58,  2.66it/s]
 16%|█▋        | 1754/10682 [26:49<56:03,  2.65it/s]
 16%|█▋        | 1755/10682 [26:50<55:58,  2.66it/s]
 16%|█▋        | 1756/10682 [26:50<1:08:12,  2.18it/s]
 16%|█▋        | 1757/10682 [26:51<1:04:28,  2.31it/s]
 16%|█▋        | 1758/10682 [26:51<1:01:55,  2.40it/s]
 16%|█▋        | 1759/10682 [26:51<1:00:09,  2.47it/s]
 16%|█▋        | 1760/10682 [26:52<58:49,  2.53it/s]  
 16%|█▋        | 1761/10682 [26:52<57:58,  2.56it/s]
 16%|█▋        | 1762/10682 [26:52<57:20,  2.59it/s]
 17%|█▋        | 1763/10682 [26:53<1:09:10,  2.15it/s]
 17%|█▋        | 1764/10682 [26:53<1:05:09,  2.28it/s]
 17%|█▋        | 1765/10682 [26:54<1:02:20,  2.38it/s]
 17%|█▋        | 1766/10682 [26:54<1:00:27,  2.46it/s]
 17%|█▋        | 1767/10682 [26:55<59:02,  2.52it/s]  
 17%|█▋        | 1768/10682 [26:55<58:02,  2.56it/s]
 17%|█▋        | 1769/10682 [26:55<57:22,  2.59it/s]
 17%|█▋        | 1770/10682 [26:56<56:52,  2.61it/s]
 17%|█▋        | 1771/10682 [26:56<56:31,  2.63it/s]
 17%|█▋        | 1772/10682 [26:56<56:14,  2.64it/s]
 17%|█▋        | 1773/10682 [26:57<56:05,  2.65it/s]
 17%|█▋        | 1774/10682 [26:57<55:57,  2.65it/s]
 17%|█▋        | 1775/10682 [26:58<55:50,  2.66it/s]{'loss': 4.6918, 'grad_norm': 0.5729764699935913, 'learning_rate': 0.0009867503683768772, 'epoch': 0.17}
                                                    
-
 17%|█▋        | 1775/10682 [26:58<55:50,  2.66it/s]
 17%|█▋        | 1776/10682 [26:58<56:01,  2.65it/s]
 17%|█▋        | 1777/10682 [26:58<55:53,  2.66it/s]
 17%|█▋        | 1778/10682 [26:59<55:46,  2.66it/s]
 17%|█▋        | 1779/10682 [26:59<55:44,  2.66it/s]
 17%|█▋        | 1780/10682 [26:59<55:42,  2.66it/s]
 17%|█▋        | 1781/10682 [27:00<55:44,  2.66it/s]
 17%|█▋        | 1782/10682 [27:00<55:47,  2.66it/s]
 17%|█▋        | 1783/10682 [27:01<55:46,  2.66it/s]
 17%|█▋        | 1784/10682 [27:01<55:49,  2.66it/s]
 17%|█▋        | 1785/10682 [27:01<55:47,  2.66it/s]
 17%|█▋        | 1786/10682 [27:02<55:42,  2.66it/s]
 17%|█▋        | 1787/10682 [27:02<55:41,  2.66it/s]
 17%|█▋        | 1788/10682 [27:02<55:39,  2.66it/s]
 17%|█▋        | 1789/10682 [27:03<55:36,  2.67it/s]
 17%|█▋        | 1790/10682 [27:03<55:34,  2.67it/s]
 17%|█▋        | 1791/10682 [27:04<55:34,  2.67it/s]
 17%|█▋        | 1792/10682 [27:04<55:36,  2.66it/s]
 17%|█▋        | 1793/10682 [27:04<55:36,  2.66it/s]
 17%|█▋        | 1794/10682 [27:05<55:33,  2.67it/s]
 17%|█▋        | 1795/10682 [27:05<55:38,  2.66it/s]
 17%|█▋        | 1796/10682 [27:05<55:33,  2.67it/s]
 17%|█▋        | 1797/10682 [27:06<55:32,  2.67it/s]
 17%|█▋        | 1798/10682 [27:06<55:37,  2.66it/s]
 17%|█▋        | 1799/10682 [27:07<55:36,  2.66it/s]
 17%|█▋        | 1800/10682 [27:07<55:33,  2.66it/s]
                                                    
{'loss': 4.6926, 'grad_norm': 0.5709038972854614, 'learning_rate': 0.0009857999404263656, 'epoch': 0.17}
-
 17%|█▋        | 1800/10682 [27:07<55:33,  2.66it/s]
 17%|█▋        | 1801/10682 [27:07<55:35,  2.66it/s]
 17%|█▋        | 1802/10682 [27:08<55:37,  2.66it/s]
 17%|█▋        | 1803/10682 [27:08<55:36,  2.66it/s]
 17%|█▋        | 1804/10682 [27:08<55:35,  2.66it/s]
 17%|█▋        | 1805/10682 [27:09<55:34,  2.66it/s]
 17%|█▋        | 1806/10682 [27:09<55:35,  2.66it/s]
 17%|█▋        | 1807/10682 [27:10<55:35,  2.66it/s]
 17%|█▋        | 1808/10682 [27:10<55:35,  2.66it/s]
 17%|█▋        | 1809/10682 [27:10<55:42,  2.65it/s]
 17%|█▋        | 1810/10682 [27:11<55:37,  2.66it/s]
 17%|█▋        | 1811/10682 [27:11<55:37,  2.66it/s]
 17%|█▋        | 1812/10682 [27:11<55:32,  2.66it/s]
 17%|█▋        | 1813/10682 [27:12<55:28,  2.66it/s]
 17%|█▋        | 1814/10682 [27:12<55:26,  2.67it/s]
 17%|█▋        | 1815/10682 [27:13<55:23,  2.67it/s]
 17%|█▋        | 1816/10682 [27:13<55:23,  2.67it/s]
 17%|█▋        | 1817/10682 [27:13<55:25,  2.67it/s]
 17%|█▋        | 1818/10682 [27:14<55:25,  2.67it/s]
 17%|█▋        | 1819/10682 [27:14<55:23,  2.67it/s]
 17%|█▋        | 1820/10682 [27:14<55:24,  2.67it/s]
 17%|█▋        | 1821/10682 [27:15<55:22,  2.67it/s]
 17%|█▋        | 1822/10682 [27:15<55:21,  2.67it/s]
 17%|█▋        | 1823/10682 [27:16<55:22,  2.67it/s]
 17%|█▋        | 1824/10682 [27:16<55:19,  2.67it/s]
 17%|█▋        | 1825/10682 [27:16<55:26,  2.66it/s]{'loss': 4.6828, 'grad_norm': 0.5669360160827637, 'learning_rate': 0.0009848170847181403, 'epoch': 0.17}
                                                    
-
 17%|█▋        | 1825/10682 [27:16<55:26,  2.66it/s]
 17%|█▋        | 1826/10682 [27:17<55:26,  2.66it/s]
 17%|█▋        | 1827/10682 [27:17<55:22,  2.66it/s]
 17%|█▋        | 1828/10682 [27:17<55:20,  2.67it/s]
 17%|█▋        | 1829/10682 [27:18<55:20,  2.67it/s]
 17%|█▋        | 1830/10682 [27:18<55:21,  2.67it/s]
 17%|█▋        | 1831/10682 [27:19<55:15,  2.67it/s]
 17%|█▋        | 1832/10682 [27:19<55:15,  2.67it/s]
 17%|█▋        | 1833/10682 [27:19<55:32,  2.65it/s]
 17%|█▋        | 1834/10682 [27:20<55:29,  2.66it/s]
 17%|█▋        | 1835/10682 [27:20<55:24,  2.66it/s]
 17%|█▋        | 1836/10682 [27:20<55:23,  2.66it/s]
 17%|█▋        | 1837/10682 [27:21<55:17,  2.67it/s]
 17%|█▋        | 1838/10682 [27:21<55:17,  2.67it/s]
 17%|█▋        | 1839/10682 [27:22<55:22,  2.66it/s]
 17%|█▋        | 1840/10682 [27:22<55:18,  2.66it/s]
 17%|█▋        | 1841/10682 [27:22<55:16,  2.67it/s]
 17%|█▋        | 1842/10682 [27:23<55:15,  2.67it/s]
 17%|█▋        | 1843/10682 [27:23<55:14,  2.67it/s]
 17%|█▋        | 1844/10682 [27:23<55:13,  2.67it/s]
 17%|█▋        | 1845/10682 [27:24<55:10,  2.67it/s]
 17%|█▋        | 1846/10682 [27:24<55:07,  2.67it/s]
 17%|█▋        | 1847/10682 [27:25<55:09,  2.67it/s]
 17%|█▋        | 1848/10682 [27:25<55:07,  2.67it/s]
 17%|█▋        | 1849/10682 [27:25<55:05,  2.67it/s]
 17%|█▋        | 1850/10682 [27:26<55:04,  2.67it/s]
                                                    
{'loss': 4.6772, 'grad_norm': 0.5577237606048584, 'learning_rate': 0.0009838018668590573, 'epoch': 0.17}
-
 17%|█▋        | 1850/10682 [27:26<55:04,  2.67it/s]
 17%|█▋        | 1851/10682 [27:26<55:17,  2.66it/s]
 17%|█▋        | 1852/10682 [27:26<55:16,  2.66it/s]
 17%|█▋        | 1853/10682 [27:27<55:18,  2.66it/s]
 17%|█▋        | 1854/10682 [27:27<55:20,  2.66it/s]
 17%|█▋        | 1855/10682 [27:28<55:19,  2.66it/s]
 17%|█▋        | 1856/10682 [27:28<55:16,  2.66it/s]
 17%|█▋        | 1857/10682 [27:28<55:14,  2.66it/s]
 17%|█▋        | 1858/10682 [27:29<55:15,  2.66it/s]
 17%|█▋        | 1859/10682 [27:29<55:12,  2.66it/s]
 17%|█▋        | 1860/10682 [27:29<55:09,  2.67it/s]
 17%|█▋        | 1861/10682 [27:30<55:07,  2.67it/s]
 17%|█▋        | 1862/10682 [27:30<55:07,  2.67it/s]
 17%|█▋        | 1863/10682 [27:31<55:12,  2.66it/s]
 17%|█▋        | 1864/10682 [27:31<55:11,  2.66it/s]
 17%|█▋        | 1865/10682 [27:31<55:11,  2.66it/s]
 17%|█▋        | 1866/10682 [27:32<55:05,  2.67it/s]
 17%|█▋        | 1867/10682 [27:32<55:13,  2.66it/s]
 17%|█▋        | 1868/10682 [27:32<55:10,  2.66it/s]
 17%|█▋        | 1869/10682 [27:33<55:11,  2.66it/s]
 18%|█▊        | 1870/10682 [27:33<55:10,  2.66it/s]
 18%|█▊        | 1871/10682 [27:34<55:08,  2.66it/s]
 18%|█▊        | 1872/10682 [27:34<55:08,  2.66it/s]
 18%|█▊        | 1873/10682 [27:34<55:04,  2.67it/s]
 18%|█▊        | 1874/10682 [27:35<55:01,  2.67it/s]
 18%|█▊        | 1875/10682 [27:35<55:05,  2.66it/s]
                                                    
{'loss': 4.6718, 'grad_norm': 0.5988249778747559, 'learning_rate': 0.000982754354616187, 'epoch': 0.18}
-
 18%|█▊        | 1875/10682 [27:35<55:05,  2.66it/s]
 18%|█▊        | 1876/10682 [27:36<55:34,  2.64it/s]
 18%|█▊        | 1877/10682 [27:36<55:28,  2.65it/s]
 18%|█▊        | 1878/10682 [27:36<55:20,  2.65it/s]
 18%|█▊        | 1879/10682 [27:37<55:19,  2.65it/s]
 18%|█▊        | 1880/10682 [27:37<55:11,  2.66it/s]
 18%|█▊        | 1881/10682 [27:37<55:17,  2.65it/s]
 18%|█▊        | 1882/10682 [27:38<55:17,  2.65it/s]
 18%|█▊        | 1883/10682 [27:38<55:13,  2.66it/s]
 18%|█▊        | 1884/10682 [27:39<55:10,  2.66it/s]
 18%|█▊        | 1885/10682 [27:39<55:08,  2.66it/s]
 18%|█▊        | 1886/10682 [27:39<55:08,  2.66it/s]
 18%|█▊        | 1887/10682 [27:40<55:03,  2.66it/s]
 18%|█▊        | 1888/10682 [27:40<54:58,  2.67it/s]
 18%|█▊        | 1889/10682 [27:40<54:58,  2.67it/s]
 18%|█▊        | 1890/10682 [27:41<54:58,  2.67it/s]
 18%|█▊        | 1891/10682 [27:41<54:56,  2.67it/s]
 18%|█▊        | 1892/10682 [27:42<55:02,  2.66it/s]
 18%|█▊        | 1893/10682 [27:42<55:53,  2.62it/s]
 18%|█▊        | 1894/10682 [27:42<55:35,  2.63it/s]
 18%|█▊        | 1895/10682 [27:43<55:33,  2.64it/s]
 18%|█▊        | 1896/10682 [27:43<55:20,  2.65it/s]
 18%|█▊        | 1897/10682 [27:43<55:10,  2.65it/s]
 18%|█▊        | 1898/10682 [27:44<55:03,  2.66it/s]
 18%|█▊        | 1899/10682 [27:44<55:03,  2.66it/s]
 18%|█▊        | 1900/10682 [27:45<55:02,  2.66it/s]{'loss': 4.658, 'grad_norm': 0.6356169581413269, 'learning_rate': 0.0009816746179122904, 'epoch': 0.18}
-
                                                    

 18%|█▊        | 1900/10682 [27:45<55:02,  2.66it/s]
 18%|█▊        | 1901/10682 [27:45<55:03,  2.66it/s]
 18%|█▊        | 1902/10682 [27:45<55:00,  2.66it/s]
 18%|█▊        | 1903/10682 [27:46<54:57,  2.66it/s]
 18%|█▊        | 1904/10682 [27:46<54:56,  2.66it/s]
 18%|█▊        | 1905/10682 [27:46<54:54,  2.66it/s]
 18%|█▊        | 1906/10682 [27:47<54:57,  2.66it/s]
 18%|█▊        | 1907/10682 [27:47<54:53,  2.66it/s]
 18%|█▊        | 1908/10682 [27:48<54:53,  2.66it/s]
 18%|█▊        | 1909/10682 [27:48<54:54,  2.66it/s]
 18%|█▊        | 1910/10682 [27:48<54:51,  2.66it/s]
 18%|█▊        | 1911/10682 [27:49<54:53,  2.66it/s]
 18%|█▊        | 1912/10682 [27:49<54:52,  2.66it/s]
 18%|█▊        | 1913/10682 [27:49<54:50,  2.66it/s]
 18%|█▊        | 1914/10682 [27:50<54:51,  2.66it/s]
 18%|█▊        | 1915/10682 [27:50<54:50,  2.66it/s]
 18%|█▊        | 1916/10682 [27:51<54:49,  2.66it/s]
 18%|█▊        | 1917/10682 [27:51<54:44,  2.67it/s]
 18%|█▊        | 1918/10682 [27:51<54:45,  2.67it/s]
 18%|█▊        | 1919/10682 [27:52<54:46,  2.67it/s]
 18%|█▊        | 1920/10682 [27:52<54:52,  2.66it/s]
 18%|█▊        | 1921/10682 [27:52<54:52,  2.66it/s]
 18%|█▊        | 1922/10682 [27:53<54:57,  2.66it/s]
 18%|█▊        | 1923/10682 [27:53<54:56,  2.66it/s]
 18%|█▊        | 1924/10682 [27:54<54:50,  2.66it/s]
 18%|█▊        | 1925/10682 [27:54<54:49,  2.66it/s]{'loss': 4.6573, 'grad_norm': 0.5952888131141663, 'learning_rate': 0.0009805627288211524, 'epoch': 0.18}
-
                                                    

 18%|█▊        | 1925/10682 [27:54<54:49,  2.66it/s]
 18%|█▊        | 1926/10682 [27:54<54:55,  2.66it/s]
 18%|█▊        | 1927/10682 [27:55<54:51,  2.66it/s]
 18%|█▊        | 1928/10682 [27:55<54:51,  2.66it/s]
 18%|█▊        | 1929/10682 [27:55<54:45,  2.66it/s]
 18%|█▊        | 1930/10682 [27:56<54:45,  2.66it/s]
 18%|█▊        | 1931/10682 [27:56<54:43,  2.67it/s]
 18%|█▊        | 1932/10682 [27:57<54:42,  2.67it/s]
 18%|█▊        | 1933/10682 [27:57<54:43,  2.66it/s]
 18%|█▊        | 1934/10682 [27:57<54:45,  2.66it/s]
 18%|█▊        | 1935/10682 [27:58<54:44,  2.66it/s]
 18%|█▊        | 1936/10682 [27:58<54:46,  2.66it/s]
 18%|█▊        | 1937/10682 [27:58<54:44,  2.66it/s]
 18%|█▊        | 1938/10682 [27:59<54:40,  2.67it/s]
 18%|█▊        | 1939/10682 [27:59<54:42,  2.66it/s]
 18%|█▊        | 1940/10682 [28:00<54:41,  2.66it/s]
 18%|█▊        | 1941/10682 [28:00<54:42,  2.66it/s]
 18%|█▊        | 1942/10682 [28:00<54:42,  2.66it/s]
 18%|█▊        | 1943/10682 [28:01<54:40,  2.66it/s]
 18%|█▊        | 1944/10682 [28:01<54:37,  2.67it/s]
 18%|█▊        | 1945/10682 [28:01<54:38,  2.66it/s]
 18%|█▊        | 1946/10682 [28:02<54:37,  2.67it/s]
 18%|█▊        | 1947/10682 [28:02<54:41,  2.66it/s]
 18%|█▊        | 1948/10682 [28:03<54:45,  2.66it/s]
 18%|█▊        | 1949/10682 [28:03<54:44,  2.66it/s]
 18%|█▊        | 1950/10682 [28:03<54:43,  2.66it/s]{'loss': 4.6464, 'grad_norm': 0.5554062724113464, 'learning_rate': 0.0009794187615627694, 'epoch': 0.18}
                                                    
-
 18%|█▊        | 1950/10682 [28:03<54:43,  2.66it/s]
 18%|█▊        | 1951/10682 [28:04<54:47,  2.66it/s]
 18%|█▊        | 1952/10682 [28:04<54:43,  2.66it/s]
 18%|█▊        | 1953/10682 [28:04<54:40,  2.66it/s]
 18%|█▊        | 1954/10682 [28:05<54:34,  2.67it/s]
 18%|█▊        | 1955/10682 [28:05<54:30,  2.67it/s]
 18%|█▊        | 1956/10682 [28:06<54:28,  2.67it/s]
 18%|█▊        | 1957/10682 [28:06<54:31,  2.67it/s]
 18%|█▊        | 1958/10682 [28:06<54:28,  2.67it/s]
 18%|█▊        | 1959/10682 [28:07<54:28,  2.67it/s]
 18%|█▊        | 1960/10682 [28:07<54:27,  2.67it/s]
 18%|█▊        | 1961/10682 [28:07<54:28,  2.67it/s]
 18%|█▊        | 1962/10682 [28:08<54:31,  2.67it/s]
 18%|█▊        | 1963/10682 [28:08<54:33,  2.66it/s]
 18%|█▊        | 1964/10682 [28:09<54:33,  2.66it/s]
 18%|█▊        | 1965/10682 [28:09<54:30,  2.67it/s]
 18%|█▊        | 1966/10682 [28:09<54:26,  2.67it/s]
 18%|█▊        | 1967/10682 [28:10<54:27,  2.67it/s]
 18%|█▊        | 1968/10682 [28:10<54:25,  2.67it/s]
 18%|█▊        | 1969/10682 [28:10<54:26,  2.67it/s]
 18%|█▊        | 1970/10682 [28:11<54:22,  2.67it/s]
 18%|█▊        | 1971/10682 [28:11<54:22,  2.67it/s]
 18%|█▊        | 1972/10682 [28:12<54:22,  2.67it/s]
 18%|█▊        | 1973/10682 [28:12<54:27,  2.67it/s]
 18%|█▊        | 1974/10682 [28:12<54:24,  2.67it/s]
 18%|█▊        | 1975/10682 [28:13<54:34,  2.66it/s]{'loss': 4.6452, 'grad_norm': 0.5772939920425415, 'learning_rate': 0.0009782427924983968, 'epoch': 0.18}
                                                    
-
 18%|█▊        | 1975/10682 [28:13<54:34,  2.66it/s]
 18%|█▊        | 1976/10682 [28:13<54:35,  2.66it/s]
 19%|█▊        | 1977/10682 [28:13<54:37,  2.66it/s]
 19%|█▊        | 1978/10682 [28:14<54:36,  2.66it/s]
 19%|█▊        | 1979/10682 [28:14<54:34,  2.66it/s]
 19%|█▊        | 1980/10682 [28:15<54:30,  2.66it/s]
 19%|█▊        | 1981/10682 [28:15<54:29,  2.66it/s]
 19%|█▊        | 1982/10682 [28:15<54:30,  2.66it/s]
 19%|█▊        | 1983/10682 [28:16<54:26,  2.66it/s]
 19%|█▊        | 1984/10682 [28:16<54:23,  2.67it/s]
 19%|█▊        | 1985/10682 [28:16<54:24,  2.66it/s]
 19%|█▊        | 1986/10682 [28:17<54:27,  2.66it/s]
 19%|█▊        | 1987/10682 [28:17<54:23,  2.66it/s]
 19%|█▊        | 1988/10682 [28:18<54:21,  2.67it/s]
 19%|█▊        | 1989/10682 [28:18<54:24,  2.66it/s]
 19%|█▊        | 1990/10682 [28:18<54:27,  2.66it/s]
 19%|█▊        | 1991/10682 [28:19<54:28,  2.66it/s]
 19%|█▊        | 1992/10682 [28:19<54:28,  2.66it/s]
 19%|█▊        | 1993/10682 [28:19<54:25,  2.66it/s]
 19%|█▊        | 1994/10682 [28:20<54:23,  2.66it/s]
 19%|█▊        | 1995/10682 [28:20<54:20,  2.66it/s]
 19%|█▊        | 1996/10682 [28:21<54:22,  2.66it/s]
 19%|█▊        | 1997/10682 [28:21<54:23,  2.66it/s]
 19%|█▊        | 1998/10682 [28:21<54:26,  2.66it/s]
 19%|█▊        | 1999/10682 [28:22<54:21,  2.66it/s]
 19%|█▊        | 2000/10682 [28:22<54:24,  2.66it/s]
                                                    
{'loss': 4.6327, 'grad_norm': 0.561221182346344, 'learning_rate': 0.00097703490012545, 'epoch': 0.19}
-
 19%|█▊        | 2000/10682 [28:22<54:24,  2.66it/s]
 19%|█▊        | 2001/10682 [28:22<54:27,  2.66it/s]
 19%|█▊        | 2002/10682 [28:23<54:21,  2.66it/s]
 19%|█▉        | 2003/10682 [28:23<54:22,  2.66it/s]
 19%|█▉        | 2004/10682 [28:24<54:20,  2.66it/s]
 19%|█▉        | 2005/10682 [28:24<54:27,  2.66it/s]
 19%|█▉        | 2006/10682 [28:24<54:21,  2.66it/s]
 19%|█▉        | 2007/10682 [28:25<54:15,  2.67it/s]
 19%|█▉        | 2008/10682 [28:25<54:14,  2.67it/s]
 19%|█▉        | 2009/10682 [28:25<54:18,  2.66it/s]
 19%|█▉        | 2010/10682 [28:26<54:12,  2.67it/s]
 19%|█▉        | 2011/10682 [28:26<54:10,  2.67it/s]
 19%|█▉        | 2012/10682 [28:27<54:09,  2.67it/s]
 19%|█▉        | 2013/10682 [28:27<54:14,  2.66it/s]
 19%|█▉        | 2014/10682 [28:27<54:11,  2.67it/s]
 19%|█▉        | 2015/10682 [28:28<54:09,  2.67it/s]
 19%|█▉        | 2016/10682 [28:28<54:16,  2.66it/s]
 19%|█▉        | 2017/10682 [28:28<54:13,  2.66it/s]
 19%|█▉        | 2018/10682 [28:29<54:11,  2.66it/s]
 19%|█▉        | 2019/10682 [28:29<54:14,  2.66it/s]
 19%|█▉        | 2020/10682 [28:30<54:13,  2.66it/s]
 19%|█▉        | 2021/10682 [28:30<54:13,  2.66it/s]
 19%|█▉        | 2022/10682 [28:30<54:09,  2.67it/s]
 19%|█▉        | 2023/10682 [28:31<54:10,  2.66it/s]
 19%|█▉        | 2024/10682 [28:31<54:08,  2.67it/s]
 19%|█▉        | 2025/10682 [28:31<54:05,  2.67it/s]
                                                    {'loss': 4.6301, 'grad_norm': 0.5352743268013, 'learning_rate': 0.0009757951650722661, 'epoch': 0.19}
-
 19%|█▉        | 2025/10682 [28:31<54:05,  2.67it/s]
 19%|█▉        | 2026/10682 [28:32<54:06,  2.67it/s]
 19%|█▉        | 2027/10682 [28:32<54:06,  2.67it/s]
 19%|█▉        | 2028/10682 [28:33<54:08,  2.66it/s]
 19%|█▉        | 2029/10682 [28:33<54:04,  2.67it/s]
 19%|█▉        | 2030/10682 [28:33<54:09,  2.66it/s]
 19%|█▉        | 2031/10682 [28:34<54:02,  2.67it/s]
 19%|█▉        | 2032/10682 [28:34<54:09,  2.66it/s]
 19%|█▉        | 2033/10682 [28:34<54:11,  2.66it/s]
 19%|█▉        | 2034/10682 [28:35<54:06,  2.66it/s]
 19%|█▉        | 2035/10682 [28:35<54:05,  2.66it/s]
 19%|█▉        | 2036/10682 [28:36<54:03,  2.67it/s]
 19%|█▉        | 2037/10682 [28:36<54:00,  2.67it/s]
 19%|█▉        | 2038/10682 [28:36<53:55,  2.67it/s]
 19%|█▉        | 2039/10682 [28:37<53:55,  2.67it/s]
 19%|█▉        | 2040/10682 [28:37<53:53,  2.67it/s]
 19%|█▉        | 2041/10682 [28:37<53:52,  2.67it/s]
 19%|█▉        | 2042/10682 [28:38<53:51,  2.67it/s]
 19%|█▉        | 2043/10682 [28:38<53:51,  2.67it/s]
 19%|█▉        | 2044/10682 [28:39<53:58,  2.67it/s]
 19%|█▉        | 2045/10682 [28:39<53:56,  2.67it/s]
 19%|█▉        | 2046/10682 [28:39<54:03,  2.66it/s]
 19%|█▉        | 2047/10682 [28:40<54:03,  2.66it/s]
 19%|█▉        | 2048/10682 [28:40<54:00,  2.66it/s]
 19%|█▉        | 2049/10682 [28:40<53:58,  2.67it/s]
 19%|█▉        | 2050/10682 [28:41<53:56,  2.67it/s]{'loss': 4.6243, 'grad_norm': 0.6106463670730591, 'learning_rate': 0.0009745236700927205, 'epoch': 0.19}
-
                                                    

 19%|█▉        | 2050/10682 [28:41<53:56,  2.67it/s]
 19%|█▉        | 2051/10682 [28:41<54:01,  2.66it/s]
 19%|█▉        | 2052/10682 [28:42<53:56,  2.67it/s]
 19%|█▉        | 2053/10682 [28:42<53:54,  2.67it/s]
 19%|█▉        | 2054/10682 [28:42<53:51,  2.67it/s]
 19%|█▉        | 2055/10682 [28:43<53:52,  2.67it/s]
 19%|█▉        | 2056/10682 [28:43<53:47,  2.67it/s]
 19%|█▉        | 2057/10682 [28:43<53:46,  2.67it/s]
 19%|█▉        | 2058/10682 [28:44<53:50,  2.67it/s]
 19%|█▉        | 2059/10682 [28:44<53:51,  2.67it/s]
 19%|█▉        | 2060/10682 [28:45<53:53,  2.67it/s]
 19%|█▉        | 2061/10682 [28:45<53:49,  2.67it/s]
 19%|█▉        | 2062/10682 [28:45<53:46,  2.67it/s]
 19%|█▉        | 2063/10682 [28:46<53:43,  2.67it/s]
 19%|█▉        | 2064/10682 [28:46<53:44,  2.67it/s]
 19%|█▉        | 2065/10682 [28:46<53:44,  2.67it/s]
 19%|█▉        | 2066/10682 [28:47<53:43,  2.67it/s]
 19%|█▉        | 2067/10682 [28:47<53:42,  2.67it/s]
 19%|█▉        | 2068/10682 [28:48<53:42,  2.67it/s]
 19%|█▉        | 2069/10682 [28:48<53:44,  2.67it/s]
 19%|█▉        | 2070/10682 [28:48<53:43,  2.67it/s]
 19%|█▉        | 2071/10682 [28:49<53:47,  2.67it/s]
 19%|█▉        | 2072/10682 [28:49<53:49,  2.67it/s]
 19%|█▉        | 2073/10682 [28:49<53:48,  2.67it/s]
 19%|█▉        | 2074/10682 [28:50<53:51,  2.66it/s]
 19%|█▉        | 2075/10682 [28:50<53:49,  2.67it/s]{'loss': 4.62, 'grad_norm': 0.5305244326591492, 'learning_rate': 0.0009732205000607044, 'epoch': 0.19}
                                                    
-
 19%|█▉        | 2075/10682 [28:50<53:49,  2.67it/s]
 19%|█▉        | 2076/10682 [28:51<53:50,  2.66it/s]
 19%|█▉        | 2077/10682 [28:51<53:45,  2.67it/s]
 19%|█▉        | 2078/10682 [28:51<53:44,  2.67it/s]
 19%|█▉        | 2079/10682 [28:52<53:41,  2.67it/s]
 19%|█▉        | 2080/10682 [28:52<53:40,  2.67it/s]
 19%|█▉        | 2081/10682 [28:52<53:39,  2.67it/s]
 19%|█▉        | 2082/10682 [28:53<53:35,  2.67it/s]
 20%|█▉        | 2083/10682 [28:53<53:33,  2.68it/s]
 20%|█▉        | 2084/10682 [28:54<53:31,  2.68it/s]
 20%|█▉        | 2085/10682 [28:54<53:34,  2.67it/s]
 20%|█▉        | 2086/10682 [28:54<53:36,  2.67it/s]
 20%|█▉        | 2087/10682 [28:55<53:35,  2.67it/s]
 20%|█▉        | 2088/10682 [28:55<53:38,  2.67it/s]
 20%|█▉        | 2089/10682 [28:55<53:37,  2.67it/s]
 20%|█▉        | 2090/10682 [28:56<53:41,  2.67it/s]
 20%|█▉        | 2091/10682 [28:56<53:42,  2.67it/s]
 20%|█▉        | 2092/10682 [28:57<53:42,  2.67it/s]
 20%|█▉        | 2093/10682 [28:57<53:38,  2.67it/s]
 20%|█▉        | 2094/10682 [28:57<53:42,  2.67it/s]
 20%|█▉        | 2095/10682 [28:58<53:39,  2.67it/s]
 20%|█▉        | 2096/10682 [28:58<53:35,  2.67it/s]
 20%|█▉        | 2097/10682 [28:58<53:33,  2.67it/s]
 20%|█▉        | 2098/10682 [28:59<53:33,  2.67it/s]
 20%|█▉        | 2099/10682 [28:59<53:42,  2.66it/s]
 20%|█▉        | 2100/10682 [29:00<53:39,  2.67it/s]
                                                    
{'loss': 4.6109, 'grad_norm': 0.5552241802215576, 'learning_rate': 0.000971885741964458, 'epoch': 0.2}
-
 20%|█▉        | 2100/10682 [29:00<53:39,  2.67it/s]
 20%|█▉        | 2101/10682 [29:00<53:43,  2.66it/s]
 20%|█▉        | 2102/10682 [29:00<53:50,  2.66it/s]
 20%|█▉        | 2103/10682 [29:01<53:44,  2.66it/s]
 20%|█▉        | 2104/10682 [29:01<53:44,  2.66it/s]
 20%|█▉        | 2105/10682 [29:01<53:40,  2.66it/s]
 20%|█▉        | 2106/10682 [29:02<53:36,  2.67it/s]
 20%|█▉        | 2107/10682 [29:02<53:35,  2.67it/s]
 20%|█▉        | 2108/10682 [29:03<53:33,  2.67it/s]
 20%|█▉        | 2109/10682 [29:03<53:30,  2.67it/s]
 20%|█▉        | 2110/10682 [29:03<53:27,  2.67it/s]
 20%|█▉        | 2111/10682 [29:04<53:32,  2.67it/s]
 20%|█▉        | 2112/10682 [29:04<53:29,  2.67it/s]
 20%|█▉        | 2113/10682 [29:04<53:34,  2.67it/s]
 20%|█▉        | 2114/10682 [29:05<53:31,  2.67it/s]
 20%|█▉        | 2115/10682 [29:05<53:33,  2.67it/s]
 20%|█▉        | 2116/10682 [29:06<53:34,  2.67it/s]
 20%|█▉        | 2117/10682 [29:06<53:33,  2.67it/s]
 20%|█▉        | 2118/10682 [29:06<53:30,  2.67it/s]
 20%|█▉        | 2119/10682 [29:07<53:28,  2.67it/s]
 20%|█▉        | 2120/10682 [29:07<53:30,  2.67it/s]
 20%|█▉        | 2121/10682 [29:07<53:32,  2.66it/s]
 20%|█▉        | 2122/10682 [29:08<53:32,  2.66it/s]
 20%|█▉        | 2123/10682 [29:08<53:28,  2.67it/s]
 20%|█▉        | 2124/10682 [29:09<53:27,  2.67it/s]
 20%|█▉        | 2125/10682 [29:09<53:28,  2.67it/s]
                                                    
{'loss': 4.6095, 'grad_norm': 0.5105506181716919, 'learning_rate': 0.0009705194849007652, 'epoch': 0.2}
-
 20%|█▉        | 2125/10682 [29:09<53:28,  2.67it/s]
 20%|█▉        | 2126/10682 [29:09<53:31,  2.66it/s]
 20%|█▉        | 2127/10682 [29:10<53:34,  2.66it/s]
 20%|█▉        | 2128/10682 [29:10<53:36,  2.66it/s]
 20%|█▉        | 2129/10682 [29:10<53:38,  2.66it/s]
 20%|█▉        | 2130/10682 [29:11<53:32,  2.66it/s]
 20%|█▉        | 2131/10682 [29:11<53:32,  2.66it/s]
 20%|█▉        | 2132/10682 [29:12<53:28,  2.66it/s]
 20%|█▉        | 2133/10682 [29:12<53:26,  2.67it/s]
 20%|█▉        | 2134/10682 [29:12<53:29,  2.66it/s]
 20%|█▉        | 2135/10682 [29:13<53:25,  2.67it/s]
 20%|█▉        | 2136/10682 [29:13<53:22,  2.67it/s]
 20%|██        | 2137/10682 [29:13<53:20,  2.67it/s]
 20%|██        | 2138/10682 [29:14<53:23,  2.67it/s]
 20%|██        | 2139/10682 [29:14<53:20,  2.67it/s]
 20%|██        | 2140/10682 [29:15<53:24,  2.67it/s]
 20%|██        | 2141/10682 [29:15<53:21,  2.67it/s]
 20%|██        | 2142/10682 [29:15<53:25,  2.66it/s]
 20%|██        | 2143/10682 [29:16<53:23,  2.67it/s]
 20%|██        | 2144/10682 [29:16<53:19,  2.67it/s]
 20%|██        | 2145/10682 [29:16<53:22,  2.67it/s]
 20%|██        | 2146/10682 [29:17<53:20,  2.67it/s]
 20%|██        | 2147/10682 [29:17<53:23,  2.66it/s]
 20%|██        | 2148/10682 [29:18<53:19,  2.67it/s]
 20%|██        | 2149/10682 [29:18<53:17,  2.67it/s]
 20%|██        | 2150/10682 [29:18<53:14,  2.67it/s]
                                                    
{'loss': 4.6067, 'grad_norm': 0.4936760663986206, 'learning_rate': 0.0009691218200690053, 'epoch': 0.2}
-
 20%|██        | 2150/10682 [29:18<53:14,  2.67it/s]
 20%|██        | 2151/10682 [29:19<53:33,  2.65it/s]
 20%|██        | 2152/10682 [29:19<53:28,  2.66it/s]
 20%|██        | 2153/10682 [29:19<53:25,  2.66it/s]
 20%|██        | 2154/10682 [29:20<53:28,  2.66it/s]
 20%|██        | 2155/10682 [29:20<53:22,  2.66it/s]
 20%|██        | 2156/10682 [29:21<53:26,  2.66it/s]
 20%|██        | 2157/10682 [29:21<53:22,  2.66it/s]
 20%|██        | 2158/10682 [29:21<53:19,  2.66it/s]
 20%|██        | 2159/10682 [29:22<53:24,  2.66it/s]
 20%|██        | 2160/10682 [29:22<53:17,  2.67it/s]
 20%|██        | 2161/10682 [29:22<53:15,  2.67it/s]
 20%|██        | 2162/10682 [29:23<53:12,  2.67it/s]
 20%|██        | 2163/10682 [29:23<53:19,  2.66it/s]
 20%|██        | 2164/10682 [29:24<53:14,  2.67it/s]
 20%|██        | 2165/10682 [29:24<53:13,  2.67it/s]
 20%|██        | 2166/10682 [29:24<53:15,  2.66it/s]
 20%|██        | 2167/10682 [29:25<53:16,  2.66it/s]
 20%|██        | 2168/10682 [29:25<53:13,  2.67it/s]
 20%|██        | 2169/10682 [29:25<53:15,  2.66it/s]
 20%|██        | 2170/10682 [29:26<53:20,  2.66it/s]
 20%|██        | 2171/10682 [29:26<53:17,  2.66it/s]
 20%|██        | 2172/10682 [29:27<53:14,  2.66it/s]
 20%|██        | 2173/10682 [29:27<53:13,  2.66it/s]
 20%|██        | 2174/10682 [29:27<53:16,  2.66it/s]
 20%|██        | 2175/10682 [29:28<53:15,  2.66it/s]{'loss': 4.5964, 'grad_norm': 0.560893714427948, 'learning_rate': 0.0009676928407650656, 'epoch': 0.2}
-
                                                    

 20%|██        | 2175/10682 [29:28<53:15,  2.66it/s]
 20%|██        | 2176/10682 [29:28<53:16,  2.66it/s]
 20%|██        | 2177/10682 [29:28<53:17,  2.66it/s]
 20%|██        | 2178/10682 [29:29<53:13,  2.66it/s]
 20%|██        | 2179/10682 [29:29<53:16,  2.66it/s]
 20%|██        | 2180/10682 [29:30<53:11,  2.66it/s]
 20%|██        | 2181/10682 [29:30<53:12,  2.66it/s]
 20%|██        | 2182/10682 [29:30<53:07,  2.67it/s]
 20%|██        | 2183/10682 [29:31<53:07,  2.67it/s]
 20%|██        | 2184/10682 [29:31<53:08,  2.67it/s]
 20%|██        | 2185/10682 [29:31<53:06,  2.67it/s]
 20%|██        | 2186/10682 [29:32<53:08,  2.66it/s]
 20%|██        | 2187/10682 [29:32<53:06,  2.67it/s]
 20%|██        | 2188/10682 [29:33<53:06,  2.67it/s]
 20%|██        | 2189/10682 [29:33<53:03,  2.67it/s]
 21%|██        | 2190/10682 [29:33<53:02,  2.67it/s]
 21%|██        | 2191/10682 [29:34<52:58,  2.67it/s]
 21%|██        | 2192/10682 [29:34<52:58,  2.67it/s]
 21%|██        | 2193/10682 [29:34<52:56,  2.67it/s]
 21%|██        | 2194/10682 [29:35<52:58,  2.67it/s]
 21%|██        | 2195/10682 [29:35<53:02,  2.67it/s]
 21%|██        | 2196/10682 [29:36<52:59,  2.67it/s]
 21%|██        | 2197/10682 [29:36<53:09,  2.66it/s]
 21%|██        | 2198/10682 [29:36<53:12,  2.66it/s]
 21%|██        | 2199/10682 [29:37<53:06,  2.66it/s]
 21%|██        | 2200/10682 [29:37<53:07,  2.66it/s]{'loss': 4.5926, 'grad_norm': 0.5969334840774536, 'learning_rate': 0.0009662326423751136, 'epoch': 0.21}
-
                                                    

 21%|██        | 2200/10682 [29:37<53:07,  2.66it/s]
 21%|██        | 2201/10682 [29:37<53:09,  2.66it/s]
 21%|██        | 2202/10682 [29:38<53:03,  2.66it/s]
 21%|██        | 2203/10682 [29:38<53:05,  2.66it/s]
 21%|██        | 2204/10682 [29:39<53:05,  2.66it/s]
 21%|██        | 2205/10682 [29:39<53:00,  2.67it/s]
 21%|██        | 2206/10682 [29:39<52:58,  2.67it/s]
 21%|██        | 2207/10682 [29:40<52:56,  2.67it/s]
 21%|██        | 2208/10682 [29:40<52:58,  2.67it/s]
 21%|██        | 2209/10682 [29:40<53:00,  2.66it/s]
 21%|██        | 2210/10682 [29:41<53:02,  2.66it/s]
 21%|██        | 2211/10682 [29:41<53:02,  2.66it/s]
 21%|██        | 2212/10682 [29:42<52:57,  2.67it/s]
 21%|██        | 2213/10682 [29:42<52:56,  2.67it/s]
 21%|██        | 2214/10682 [29:42<52:50,  2.67it/s]
 21%|██        | 2215/10682 [29:43<52:48,  2.67it/s]
 21%|██        | 2216/10682 [29:43<52:45,  2.67it/s]
 21%|██        | 2217/10682 [29:43<52:46,  2.67it/s]
 21%|██        | 2218/10682 [29:44<52:47,  2.67it/s]
 21%|██        | 2219/10682 [29:44<52:46,  2.67it/s]
 21%|██        | 2220/10682 [29:45<52:46,  2.67it/s]
 21%|██        | 2221/10682 [29:45<52:45,  2.67it/s]
 21%|██        | 2222/10682 [29:45<52:52,  2.67it/s]
 21%|██        | 2223/10682 [29:46<52:51,  2.67it/s]
 21%|██        | 2224/10682 [29:46<52:54,  2.66it/s]
 21%|██        | 2225/10682 [29:46<52:53,  2.67it/s]
                                                    
{'loss': 4.5879, 'grad_norm': 0.5549899935722351, 'learning_rate': 0.000964741322369231, 'epoch': 0.21}
-
 21%|██        | 2225/10682 [29:46<52:53,  2.67it/s]
 21%|██        | 2226/10682 [29:47<53:11,  2.65it/s]
 21%|██        | 2227/10682 [29:47<53:07,  2.65it/s]
 21%|██        | 2228/10682 [29:48<53:00,  2.66it/s]
 21%|██        | 2229/10682 [29:48<53:00,  2.66it/s]
 21%|██        | 2230/10682 [29:48<52:55,  2.66it/s]
 21%|██        | 2231/10682 [29:49<52:51,  2.66it/s]
 21%|██        | 2232/10682 [29:49<52:48,  2.67it/s]
 21%|██        | 2233/10682 [29:49<52:44,  2.67it/s]
 21%|██        | 2234/10682 [29:50<52:44,  2.67it/s]
 21%|██        | 2235/10682 [29:50<52:45,  2.67it/s]
 21%|██        | 2236/10682 [29:51<52:48,  2.67it/s]
 21%|██        | 2237/10682 [29:51<52:46,  2.67it/s]
 21%|██        | 2238/10682 [29:51<52:55,  2.66it/s]
 21%|██        | 2239/10682 [29:52<52:55,  2.66it/s]
 21%|██        | 2240/10682 [29:52<52:50,  2.66it/s]
 21%|██        | 2241/10682 [29:52<52:48,  2.66it/s]
 21%|██        | 2242/10682 [29:53<52:48,  2.66it/s]
 21%|██        | 2243/10682 [29:53<52:45,  2.67it/s]
 21%|██        | 2244/10682 [29:54<52:44,  2.67it/s]
 21%|██        | 2245/10682 [29:54<52:41,  2.67it/s]
 21%|██        | 2246/10682 [29:54<52:41,  2.67it/s]
 21%|██        | 2247/10682 [29:55<52:47,  2.66it/s]
 21%|██        | 2248/10682 [29:55<52:43,  2.67it/s]
 21%|██        | 2249/10682 [29:55<52:46,  2.66it/s]
 21%|██        | 2250/10682 [29:56<52:45,  2.66it/s]
                                                    
{'loss': 4.5878, 'grad_norm': 0.4979619085788727, 'learning_rate': 0.0009632189802949055, 'epoch': 0.21}
-
 21%|██        | 2250/10682 [29:56<52:45,  2.66it/s]
 21%|██        | 2251/10682 [29:56<52:51,  2.66it/s]
 21%|██        | 2252/10682 [29:57<52:53,  2.66it/s]
 21%|██        | 2253/10682 [29:57<52:46,  2.66it/s]
 21%|██        | 2254/10682 [29:57<52:42,  2.66it/s]
 21%|██        | 2255/10682 [29:58<52:42,  2.66it/s]
 21%|██        | 2256/10682 [29:58<52:42,  2.66it/s]
 21%|██        | 2257/10682 [29:58<52:38,  2.67it/s]
 21%|██        | 2258/10682 [29:59<52:36,  2.67it/s]
 21%|██        | 2259/10682 [29:59<52:34,  2.67it/s]
 21%|██        | 2260/10682 [30:00<52:35,  2.67it/s]
 21%|██        | 2261/10682 [30:00<52:33,  2.67it/s]
 21%|██        | 2262/10682 [30:00<52:36,  2.67it/s]
 21%|██        | 2263/10682 [30:01<52:37,  2.67it/s]
 21%|██        | 2264/10682 [30:01<52:36,  2.67it/s]
 21%|██        | 2265/10682 [30:01<52:42,  2.66it/s]
 21%|██        | 2266/10682 [30:02<52:41,  2.66it/s]
 21%|██        | 2267/10682 [30:02<52:37,  2.67it/s]
 21%|██        | 2268/10682 [30:03<52:37,  2.66it/s]
 21%|██        | 2269/10682 [30:03<52:34,  2.67it/s]
 21%|██▏       | 2270/10682 [30:03<52:32,  2.67it/s]
 21%|██▏       | 2271/10682 [30:04<52:30,  2.67it/s]
 21%|██▏       | 2272/10682 [30:04<52:30,  2.67it/s]
 21%|██▏       | 2273/10682 [30:04<52:32,  2.67it/s]
 21%|██▏       | 2274/10682 [30:05<52:33,  2.67it/s]
 21%|██▏       | 2275/10682 [30:05<52:33,  2.67it/s]{'loss': 4.5761, 'grad_norm': 0.5065141320228577, 'learning_rate': 0.0009616657177703876, 'epoch': 0.21}
-
                                                    

 21%|██▏       | 2275/10682 [30:05<52:33,  2.67it/s]
 21%|██▏       | 2276/10682 [30:06<52:38,  2.66it/s]
 21%|██▏       | 2277/10682 [30:06<52:35,  2.66it/s]
 21%|██▏       | 2278/10682 [30:06<52:35,  2.66it/s]
 21%|██▏       | 2279/10682 [30:07<52:36,  2.66it/s]
 21%|██▏       | 2280/10682 [30:07<52:33,  2.66it/s]
 21%|██▏       | 2281/10682 [30:07<52:32,  2.67it/s]
 21%|██▏       | 2282/10682 [30:08<52:30,  2.67it/s]
 21%|██▏       | 2283/10682 [30:08<52:30,  2.67it/s]
 21%|██▏       | 2284/10682 [30:09<52:31,  2.67it/s]
 21%|██▏       | 2285/10682 [30:09<52:35,  2.66it/s]
 21%|██▏       | 2286/10682 [30:09<52:28,  2.67it/s]
 21%|██▏       | 2287/10682 [30:10<52:28,  2.67it/s]
 21%|██▏       | 2288/10682 [30:10<52:25,  2.67it/s]
 21%|██▏       | 2289/10682 [30:10<52:27,  2.67it/s]
 21%|██▏       | 2290/10682 [30:11<52:32,  2.66it/s]
 21%|██▏       | 2291/10682 [30:11<52:29,  2.66it/s]
 21%|██▏       | 2292/10682 [30:12<52:28,  2.66it/s]
 21%|██▏       | 2293/10682 [30:12<52:27,  2.67it/s]
 21%|██▏       | 2294/10682 [30:12<52:26,  2.67it/s]
 21%|██▏       | 2295/10682 [30:13<52:25,  2.67it/s]
 21%|██▏       | 2296/10682 [30:13<52:26,  2.67it/s]
 22%|██▏       | 2297/10682 [30:13<52:22,  2.67it/s]
 22%|██▏       | 2298/10682 [30:14<52:20,  2.67it/s]
 22%|██▏       | 2299/10682 [30:14<52:18,  2.67it/s]
 22%|██▏       | 2300/10682 [30:15<52:20,  2.67it/s]{'loss': 4.5728, 'grad_norm': 0.5257205963134766, 'learning_rate': 0.0009600816384779067, 'epoch': 0.22}
                                                    
-
 22%|██▏       | 2300/10682 [30:15<52:20,  2.67it/s]
 22%|██▏       | 2301/10682 [30:15<52:24,  2.67it/s]
 22%|██▏       | 2302/10682 [30:15<52:20,  2.67it/s]
 22%|██▏       | 2303/10682 [30:16<52:22,  2.67it/s]
 22%|██▏       | 2304/10682 [30:16<52:24,  2.66it/s]
 22%|██▏       | 2305/10682 [30:16<52:27,  2.66it/s]
 22%|██▏       | 2306/10682 [30:17<52:26,  2.66it/s]
 22%|██▏       | 2307/10682 [30:17<52:25,  2.66it/s]
 22%|██▏       | 2308/10682 [30:18<52:22,  2.66it/s]
 22%|██▏       | 2309/10682 [30:18<52:19,  2.67it/s]
 22%|██▏       | 2310/10682 [30:18<52:19,  2.67it/s]
 22%|██▏       | 2311/10682 [30:19<52:16,  2.67it/s]
 22%|██▏       | 2312/10682 [30:19<52:13,  2.67it/s]
 22%|██▏       | 2313/10682 [30:19<52:15,  2.67it/s]
 22%|██▏       | 2314/10682 [30:20<52:15,  2.67it/s]
 22%|██▏       | 2315/10682 [30:20<52:17,  2.67it/s]
 22%|██▏       | 2316/10682 [30:21<52:15,  2.67it/s]
 22%|██▏       | 2317/10682 [30:21<52:10,  2.67it/s]
 22%|██▏       | 2318/10682 [30:21<52:09,  2.67it/s]
 22%|██▏       | 2319/10682 [30:22<52:12,  2.67it/s]
 22%|██▏       | 2320/10682 [30:22<52:09,  2.67it/s]
 22%|██▏       | 2321/10682 [30:22<52:11,  2.67it/s]
 22%|██▏       | 2322/10682 [30:23<52:09,  2.67it/s]
 22%|██▏       | 2323/10682 [30:23<52:10,  2.67it/s]
 22%|██▏       | 2324/10682 [30:24<52:13,  2.67it/s]
 22%|██▏       | 2325/10682 [30:24<52:16,  2.66it/s]
                                                    
{'loss': 4.5666, 'grad_norm': 0.5462834239006042, 'learning_rate': 0.00095846684815675, 'epoch': 0.22}
-
 22%|██▏       | 2325/10682 [30:24<52:16,  2.66it/s]
 22%|██▏       | 2326/10682 [30:24<52:21,  2.66it/s]
 22%|██▏       | 2327/10682 [30:25<52:14,  2.67it/s]
 22%|██▏       | 2328/10682 [30:25<52:12,  2.67it/s]
 22%|██▏       | 2329/10682 [30:25<52:09,  2.67it/s]
 22%|██▏       | 2330/10682 [30:26<52:14,  2.66it/s]
 22%|██▏       | 2331/10682 [30:26<52:10,  2.67it/s]
 22%|██▏       | 2332/10682 [30:27<52:07,  2.67it/s]
 22%|██▏       | 2333/10682 [30:27<52:09,  2.67it/s]
 22%|██▏       | 2334/10682 [30:27<52:07,  2.67it/s]
 22%|██▏       | 2335/10682 [30:28<52:06,  2.67it/s]
 22%|██▏       | 2336/10682 [30:28<52:08,  2.67it/s]
 22%|██▏       | 2337/10682 [30:28<52:04,  2.67it/s]
 22%|██▏       | 2338/10682 [30:29<52:06,  2.67it/s]
 22%|██▏       | 2339/10682 [30:29<52:04,  2.67it/s]
 22%|██▏       | 2340/10682 [30:30<52:03,  2.67it/s]
 22%|██▏       | 2341/10682 [30:30<52:00,  2.67it/s]
 22%|██▏       | 2342/10682 [30:30<52:00,  2.67it/s]
 22%|██▏       | 2343/10682 [30:31<51:59,  2.67it/s]
 22%|██▏       | 2344/10682 [30:31<52:05,  2.67it/s]
 22%|██▏       | 2345/10682 [30:31<52:02,  2.67it/s]
 22%|██▏       | 2346/10682 [30:32<52:06,  2.67it/s]
 22%|██▏       | 2347/10682 [30:32<52:07,  2.66it/s]
 22%|██▏       | 2348/10682 [30:33<52:07,  2.66it/s]
 22%|██▏       | 2349/10682 [30:33<52:02,  2.67it/s]
 22%|██▏       | 2350/10682 [30:33<52:06,  2.67it/s]
                                                    
{'loss': 4.5693, 'grad_norm': 0.5251888036727905, 'learning_rate': 0.0009568214545962046, 'epoch': 0.22}
-
 22%|██▏       | 2350/10682 [30:33<52:06,  2.67it/s]
 22%|██▏       | 2351/10682 [30:34<52:21,  2.65it/s]
 22%|██▏       | 2352/10682 [30:34<52:11,  2.66it/s]
 22%|██▏       | 2353/10682 [30:34<52:08,  2.66it/s]
 22%|██▏       | 2354/10682 [30:35<52:05,  2.66it/s]
 22%|█��▏       | 2355/10682 [30:35<52:01,  2.67it/s]
 22%|██▏       | 2356/10682 [30:36<51:59,  2.67it/s]
 22%|██▏       | 2357/10682 [30:36<52:01,  2.67it/s]
 22%|██▏       | 2358/10682 [30:36<51:59,  2.67it/s]
 22%|██▏       | 2359/10682 [30:37<52:04,  2.66it/s]
 22%|██▏       | 2360/10682 [30:37<52:05,  2.66it/s]
 22%|██▏       | 2361/10682 [30:37<52:04,  2.66it/s]
 22%|██▏       | 2362/10682 [30:38<52:04,  2.66it/s]
 22%|██▏       | 2363/10682 [30:38<52:02,  2.66it/s]
 22%|██▏       | 2364/10682 [30:39<52:02,  2.66it/s]
 22%|██▏       | 2365/10682 [30:39<52:04,  2.66it/s]
 22%|██▏       | 2366/10682 [30:39<52:04,  2.66it/s]
 22%|██▏       | 2367/10682 [30:40<52:00,  2.66it/s]
 22%|██▏       | 2368/10682 [30:40<51:56,  2.67it/s]
 22%|██▏       | 2369/10682 [30:40<51:59,  2.66it/s]
 22%|██▏       | 2370/10682 [30:41<51:55,  2.67it/s]
 22%|██▏       | 2371/10682 [30:41<51:59,  2.66it/s]
 22%|██▏       | 2372/10682 [30:42<51:55,  2.67it/s]
 22%|██▏       | 2373/10682 [30:42<51:56,  2.67it/s]
 22%|██▏       | 2374/10682 [30:42<51:53,  2.67it/s]
 22%|██▏       | 2375/10682 [30:43<51:52,  2.67it/s]
                                                    
{'loss': 4.5645, 'grad_norm': 0.5100898742675781, 'learning_rate': 0.0009551455676283627, 'epoch': 0.22}
-
 22%|██▏       | 2375/10682 [30:43<51:52,  2.67it/s]
 22%|██▏       | 2376/10682 [30:43<52:11,  2.65it/s]
 22%|██▏       | 2377/10682 [30:43<52:03,  2.66it/s]
 22%|██▏       | 2378/10682 [30:44<52:04,  2.66it/s]
 22%|██▏       | 2379/10682 [30:44<51:58,  2.66it/s]
 22%|██▏       | 2380/10682 [30:45<51:57,  2.66it/s]
 22%|██▏       | 2381/10682 [30:45<51:56,  2.66it/s]
 22%|██▏       | 2382/10682 [30:45<51:53,  2.67it/s]
 22%|██▏       | 2383/10682 [30:46<51:52,  2.67it/s]
 22%|██▏       | 2384/10682 [30:46<51:52,  2.67it/s]
 22%|██▏       | 2385/10682 [30:46<51:50,  2.67it/s]
 22%|██▏       | 2386/10682 [30:47<51:45,  2.67it/s]
 22%|██▏       | 2387/10682 [30:47<51:47,  2.67it/s]
 22%|██▏       | 2388/10682 [30:48<51:45,  2.67it/s]
 22%|██▏       | 2389/10682 [30:48<51:45,  2.67it/s]
 22%|██▏       | 2390/10682 [30:48<51:41,  2.67it/s]
 22%|██▏       | 2391/10682 [30:49<51:42,  2.67it/s]
 22%|██▏       | 2392/10682 [30:49<51:43,  2.67it/s]
 22%|██▏       | 2393/10682 [30:49<51:41,  2.67it/s]
 22%|██▏       | 2394/10682 [30:50<51:39,  2.67it/s]
 22%|██▏       | 2395/10682 [30:50<51:38,  2.67it/s]
 22%|██▏       | 2396/10682 [30:51<51:42,  2.67it/s]
 22%|██▏       | 2397/10682 [30:51<51:40,  2.67it/s]
 22%|██▏       | 2398/10682 [30:51<51:43,  2.67it/s]
 22%|██▏       | 2399/10682 [30:52<51:44,  2.67it/s]
 22%|██▏       | 2400/10682 [30:52<51:46,  2.67it/s]
                                                    
{'loss': 4.5579, 'grad_norm': 0.5397598147392273, 'learning_rate': 0.0009534392991207893, 'epoch': 0.22}
-
 22%|██▏       | 2400/10682 [30:52<51:46,  2.67it/s]
 22%|██▏       | 2401/10682 [30:52<52:05,  2.65it/s]
 22%|██▏       | 2402/10682 [30:53<51:56,  2.66it/s]
 22%|██▏       | 2403/10682 [30:53<51:53,  2.66it/s]
 23%|██▎       | 2404/10682 [30:54<51:47,  2.66it/s]
 23%|██▎       | 2405/10682 [30:54<51:48,  2.66it/s]
 23%|██▎       | 2406/10682 [30:54<51:44,  2.67it/s]
 23%|██▎       | 2407/10682 [30:55<51:45,  2.66it/s]
 23%|██▎       | 2408/10682 [30:55<51:41,  2.67it/s]
 23%|██▎       | 2409/10682 [30:55<51:41,  2.67it/s]
 23%|██▎       | 2410/10682 [30:56<51:37,  2.67it/s]
 23%|██▎       | 2411/10682 [30:56<51:35,  2.67it/s]
 23%|██▎       | 2412/10682 [30:57<51:38,  2.67it/s]
 23%|██▎       | 2413/10682 [30:57<51:36,  2.67it/s]
 23%|██▎       | 2414/10682 [30:57<51:41,  2.67it/s]
 23%|██▎       | 2415/10682 [30:58<51:41,  2.67it/s]
 23%|██▎       | 2416/10682 [30:58<51:37,  2.67it/s]
 23%|██▎       | 2417/10682 [30:58<51:38,  2.67it/s]
 23%|██▎       | 2418/10682 [30:59<51:39,  2.67it/s]
 23%|██▎       | 2419/10682 [30:59<51:37,  2.67it/s]
 23%|██▎       | 2420/10682 [31:00<51:34,  2.67it/s]
 23%|██▎       | 2421/10682 [31:00<51:38,  2.67it/s]
 23%|██▎       | 2422/10682 [31:00<51:33,  2.67it/s]
 23%|██▎       | 2423/10682 [31:01<51:35,  2.67it/s]
 23%|██▎       | 2424/10682 [31:01<51:31,  2.67it/s]
 23%|██▎       | 2425/10682 [31:01<51:34,  2.67it/s]{'loss': 4.5508, 'grad_norm': 0.5710335969924927, 'learning_rate': 0.0009517027629690559, 'epoch': 0.23}
-
                                                    

 23%|██▎       | 2425/10682 [31:01<51:34,  2.67it/s]
 23%|██▎       | 2426/10682 [31:02<51:36,  2.67it/s]
 23%|██▎       | 2427/10682 [31:02<51:36,  2.67it/s]
 23%|██▎       | 2428/10682 [31:03<51:41,  2.66it/s]
 23%|██▎       | 2429/10682 [31:03<51:40,  2.66it/s]
 23%|██▎       | 2430/10682 [31:03<51:38,  2.66it/s]
 23%|██▎       | 2431/10682 [31:04<51:34,  2.67it/s]
 23%|██▎       | 2432/10682 [31:04<51:34,  2.67it/s]
 23%|██▎       | 2433/10682 [31:04<51:34,  2.67it/s]
 23%|██▎       | 2434/10682 [31:05<51:35,  2.66it/s]
 23%|██▎       | 2435/10682 [31:05<51:32,  2.67it/s]
 23%|██▎       | 2436/10682 [31:06<51:33,  2.67it/s]
 23%|██▎       | 2437/10682 [31:06<51:30,  2.67it/s]
 23%|██▎       | 2438/10682 [31:06<51:24,  2.67it/s]
 23%|██▎       | 2439/10682 [31:07<51:28,  2.67it/s]
 23%|██▎       | 2440/10682 [31:07<51:32,  2.67it/s]
 23%|██▎       | 2441/10682 [31:07<51:35,  2.66it/s]
 23%|██▎       | 2442/10682 [31:08<51:34,  2.66it/s]
 23%|██▎       | 2443/10682 [31:08<51:31,  2.67it/s]
 23%|██▎       | 2444/10682 [31:09<51:32,  2.66it/s]
 23%|██▎       | 2445/10682 [31:09<51:31,  2.66it/s]
 23%|██▎       | 2446/10682 [31:09<51:29,  2.67it/s]
 23%|██▎       | 2447/10682 [31:10<51:27,  2.67it/s]
 23%|██▎       | 2448/10682 [31:10<51:25,  2.67it/s]
 23%|██▎       | 2449/10682 [31:10<51:22,  2.67it/s]
 23%|██▎       | 2450/10682 [31:11<51:24,  2.67it/s]
                                                    
{'loss': 4.5417, 'grad_norm': 0.5384529829025269, 'learning_rate': 0.0009499360750891371, 'epoch': 0.23}
-
 23%|██▎       | 2450/10682 [31:11<51:24,  2.67it/s]
 23%|██▎       | 2451/10682 [31:11<51:25,  2.67it/s]
 23%|██▎       | 2452/10682 [31:12<51:29,  2.66it/s]
 23%|██▎       | 2453/10682 [31:12<51:25,  2.67it/s]
 23%|██▎       | 2454/10682 [31:12<51:23,  2.67it/s]
 23%|██▎       | 2455/10682 [31:13<51:27,  2.66it/s]
 23%|██▎       | 2456/10682 [31:13<51:26,  2.66it/s]
 23%|██▎       | 2457/10682 [31:13<51:25,  2.67it/s]
 23%|██▎       | 2458/10682 [31:14<51:25,  2.67it/s]
 23%|██▎       | 2459/10682 [31:14<51:25,  2.67it/s]
 23%|██▎       | 2460/10682 [31:15<51:23,  2.67it/s]
 23%|██▎       | 2461/10682 [31:15<51:26,  2.66it/s]
 23%|██▎       | 2462/10682 [31:15<51:21,  2.67it/s]
 23%|██▎       | 2463/10682 [31:16<51:19,  2.67it/s]
 23%|██▎       | 2464/10682 [31:16<51:26,  2.66it/s]
 23%|██▎       | 2465/10682 [31:16<51:23,  2.66it/s]
 23%|██▎       | 2466/10682 [31:17<51:25,  2.66it/s]
 23%|██▎       | 2467/10682 [31:17<51:26,  2.66it/s]
 23%|██▎       | 2468/10682 [31:18<51:27,  2.66it/s]
 23%|██▎       | 2469/10682 [31:18<51:25,  2.66it/s]
 23%|██▎       | 2470/10682 [31:18<51:21,  2.66it/s]
 23%|██▎       | 2471/10682 [31:19<1:02:38,  2.18it/s]
 23%|██▎       | 2472/10682 [31:19<59:15,  2.31it/s]  
 23%|██▎       | 2473/10682 [31:20<56:49,  2.41it/s]
 23%|██▎       | 2474/10682 [31:20<55:11,  2.48it/s]
 23%|██▎       | 2475/10682 [31:21<53:59,  2.53it/s]{'loss': 4.5394, 'grad_norm': 0.4804738461971283, 'learning_rate': 0.0009481393534096735, 'epoch': 0.23}
                                                    
-
 23%|██▎       | 2475/10682 [31:21<53:59,  2.53it/s]
 23%|██▎       | 2476/10682 [31:21<53:13,  2.57it/s]
 23%|██▎       | 2477/10682 [31:21<52:39,  2.60it/s]
 23%|██▎       | 2478/10682 [31:22<52:12,  2.62it/s]
 23%|██▎       | 2479/10682 [31:22<51:59,  2.63it/s]
 23%|██▎       | 2480/10682 [31:22<51:41,  2.64it/s]
 23%|██▎       | 2481/10682 [31:23<51:36,  2.65it/s]
 23%|██▎       | 2482/10682 [31:23<51:29,  2.65it/s]
 23%|██▎       | 2483/10682 [31:24<1:02:46,  2.18it/s]
 23%|██▎       | 2484/10682 [31:24<59:17,  2.30it/s]  
 23%|██▎       | 2485/10682 [31:25<56:59,  2.40it/s]
 23%|██▎       | 2486/10682 [31:25<55:12,  2.47it/s]
 23%|██▎       | 2487/10682 [31:25<53:58,  2.53it/s]
 23%|██▎       | 2488/10682 [31:26<53:02,  2.57it/s]
 23%|██▎       | 2489/10682 [31:26<52:32,  2.60it/s]
 23%|██▎       | 2490/10682 [31:26<52:04,  2.62it/s]
 23%|██▎       | 2491/10682 [31:27<51:47,  2.64it/s]
 23%|██▎       | 2492/10682 [31:27<51:39,  2.64it/s]
 23%|██▎       | 2493/10682 [31:28<51:24,  2.65it/s]
 23%|██▎       | 2494/10682 [31:28<51:19,  2.66it/s]
 23%|██▎       | 2495/10682 [31:28<51:13,  2.66it/s]
 23%|██▎       | 2496/10682 [31:29<51:21,  2.66it/s]
 23%|██▎       | 2497/10682 [31:29<51:20,  2.66it/s]
 23%|██▎       | 2498/10682 [31:29<51:16,  2.66it/s]
 23%|██▎       | 2499/10682 [31:30<51:11,  2.66it/s]
 23%|██▎       | 2500/10682 [31:30<51:05,  2.67it/s]{'loss': 4.5407, 'grad_norm': 0.45588549971580505, 'learning_rate': 0.0009463127178640994, 'epoch': 0.23}
-
                                                    

 23%|██▎       | 2500/10682 [31:30<51:05,  2.67it/s]
 23%|██▎       | 2501/10682 [31:31<51:10,  2.66it/s]
 23%|██▎       | 2502/10682 [31:31<51:08,  2.67it/s]
 23%|██▎       | 2503/10682 [31:31<51:05,  2.67it/s]
 23%|██▎       | 2504/10682 [31:32<51:02,  2.67it/s]
 23%|██▎       | 2505/10682 [31:32<51:01,  2.67it/s]
 23%|██▎       | 2506/10682 [31:32<51:00,  2.67it/s]
 23%|██▎       | 2507/10682 [31:33<51:02,  2.67it/s]
 23%|██▎       | 2508/10682 [31:33<51:07,  2.66it/s]
 23%|██▎       | 2509/10682 [31:34<51:05,  2.67it/s]
 23%|██▎       | 2510/10682 [31:34<51:07,  2.66it/s]
 24%|██▎       | 2511/10682 [31:34<51:07,  2.66it/s]
 24%|██▎       | 2512/10682 [31:35<51:06,  2.66it/s]
 24%|██▎       | 2513/10682 [31:35<51:02,  2.67it/s]
 24%|██▎       | 2514/10682 [31:35<51:01,  2.67it/s]
 24%|██▎       | 2515/10682 [31:36<51:02,  2.67it/s]
 24%|██▎       | 2516/10682 [31:36<51:01,  2.67it/s]
 24%|██▎       | 2517/10682 [31:37<51:01,  2.67it/s]
 24%|██▎       | 2518/10682 [31:37<50:56,  2.67it/s]
 24%|██▎       | 2519/10682 [31:37<50:54,  2.67it/s]
 24%|██▎       | 2520/10682 [31:38<50:55,  2.67it/s]
 24%|██▎       | 2521/10682 [31:38<50:52,  2.67it/s]
 24%|██▎       | 2522/10682 [31:38<50:54,  2.67it/s]
 24%|██▎       | 2523/10682 [31:39<50:53,  2.67it/s]
 24%|██▎       | 2524/10682 [31:39<50:57,  2.67it/s]
 24%|██▎       | 2525/10682 [31:40<50:56,  2.67it/s]
                                                    
{'loss': 4.5316, 'grad_norm': 0.49940308928489685, 'learning_rate': 0.0009444562903826375, 'epoch': 0.24}
-
 24%|██▎       | 2525/10682 [31:40<50:56,  2.67it/s]
 24%|██▎       | 2526/10682 [31:40<51:05,  2.66it/s]
 24%|██▎       | 2527/10682 [31:40<50:59,  2.67it/s]
 24%|██▎       | 2528/10682 [31:41<50:57,  2.67it/s]
 24%|██▎       | 2529/10682 [31:41<50:59,  2.66it/s]
 24%|██▎       | 2530/10682 [31:41<50:56,  2.67it/s]
 24%|██▎       | 2531/10682 [31:42<50:53,  2.67it/s]
 24%|██▎       | 2532/10682 [31:42<50:54,  2.67it/s]
 24%|██▎       | 2533/10682 [31:43<50:51,  2.67it/s]
 24%|██▎       | 2534/10682 [31:43<50:53,  2.67it/s]
 24%|██▎       | 2535/10682 [31:43<50:53,  2.67it/s]
 24%|██▎       | 2536/10682 [31:44<50:54,  2.67it/s]
 24%|██▍       | 2537/10682 [31:44<50:54,  2.67it/s]
 24%|██▍       | 2538/10682 [31:44<51:01,  2.66it/s]
 24%|██▍       | 2539/10682 [31:45<51:01,  2.66it/s]
 24%|██▍       | 2540/10682 [31:45<51:00,  2.66it/s]
 24%|██▍       | 2541/10682 [31:46<51:01,  2.66it/s]
 24%|██▍       | 2542/10682 [31:46<50:56,  2.66it/s]
 24%|██▍       | 2543/10682 [31:46<50:54,  2.66it/s]
 24%|██▍       | 2544/10682 [31:47<50:51,  2.67it/s]
 24%|██▍       | 2545/10682 [31:47<50:50,  2.67it/s]
 24%|██▍       | 2546/10682 [31:47<50:51,  2.67it/s]
 24%|██▍       | 2547/10682 [31:48<50:50,  2.67it/s]
 24%|██▍       | 2548/10682 [31:48<50:49,  2.67it/s]
 24%|██▍       | 2549/10682 [31:49<50:52,  2.66it/s]
 24%|██▍       | 2550/10682 [31:49<50:52,  2.66it/s]
                                                    
{'loss': 4.5397, 'grad_norm': 0.53085857629776, 'learning_rate': 0.0009425701948841596, 'epoch': 0.24}
-
 24%|██▍       | 2550/10682 [31:49<50:52,  2.66it/s]
 24%|██▍       | 2551/10682 [31:49<50:54,  2.66it/s]
 24%|██▍       | 2552/10682 [31:50<50:54,  2.66it/s]
 24%|██▍       | 2553/10682 [31:50<50:55,  2.66it/s]
 24%|██▍       | 2554/10682 [31:50<50:52,  2.66it/s]
 24%|██▍       | 2555/10682 [31:51<50:50,  2.66it/s]
 24%|██▍       | 2556/10682 [31:51<50:48,  2.67it/s]
 24%|██▍       | 2557/10682 [31:52<50:47,  2.67it/s]
 24%|██▍       | 2558/10682 [31:52<50:45,  2.67it/s]
 24%|██▍       | 2559/10682 [31:52<50:47,  2.67it/s]
 24%|██▍       | 2560/10682 [31:53<50:45,  2.67it/s]
 24%|██▍       | 2561/10682 [31:53<50:44,  2.67it/s]
 24%|██▍       | 2562/10682 [31:53<50:41,  2.67it/s]
 24%|██▍       | 2563/10682 [31:54<50:46,  2.66it/s]
 24%|██▍       | 2564/10682 [31:54<50:47,  2.66it/s]
 24%|██▍       | 2565/10682 [31:55<50:42,  2.67it/s]
 24%|██▍       | 2566/10682 [31:55<50:45,  2.67it/s]
 24%|██▍       | 2567/10682 [31:55<50:42,  2.67it/s]
 24%|██▍       | 2568/10682 [31:56<50:45,  2.66it/s]
 24%|██▍       | 2569/10682 [31:56<50:43,  2.67it/s]
 24%|██▍       | 2570/10682 [31:56<50:43,  2.67it/s]
 24%|██▍       | 2571/10682 [31:57<50:39,  2.67it/s]
 24%|██▍       | 2572/10682 [31:57<50:39,  2.67it/s]
 24%|██▍       | 2573/10682 [31:58<50:38,  2.67it/s]
 24%|██▍       | 2574/10682 [31:58<50:38,  2.67it/s]
 24%|██▍       | 2575/10682 [31:58<50:38,  2.67it/s]{'loss': 4.5293, 'grad_norm': 0.4714706242084503, 'learning_rate': 0.0009406545572679153, 'epoch': 0.24}
-
                                                    

 24%|██▍       | 2575/10682 [31:58<50:38,  2.67it/s]
 24%|██▍       | 2576/10682 [31:59<50:40,  2.67it/s]
 24%|██▍       | 2577/10682 [31:59<50:46,  2.66it/s]
 24%|██▍       | 2578/10682 [31:59<50:43,  2.66it/s]
 24%|██▍       | 2579/10682 [32:00<50:40,  2.66it/s]
 24%|██▍       | 2580/10682 [32:00<50:44,  2.66it/s]
 24%|██▍       | 2581/10682 [32:01<50:39,  2.67it/s]
 24%|██▍       | 2582/10682 [32:01<50:44,  2.66it/s]
 24%|██▍       | 2583/10682 [32:01<50:41,  2.66it/s]
 24%|██▍       | 2584/10682 [32:02<50:43,  2.66it/s]
 24%|██▍       | 2585/10682 [32:02<50:39,  2.66it/s]
 24%|██▍       | 2586/10682 [32:02<50:36,  2.67it/s]
 24%|██▍       | 2587/10682 [32:03<50:36,  2.67it/s]
 24%|██▍       | 2588/10682 [32:03<50:40,  2.66it/s]
 24%|██▍       | 2589/10682 [32:04<50:38,  2.66it/s]
 24%|██▍       | 2590/10682 [32:04<50:33,  2.67it/s]
 24%|██▍       | 2591/10682 [32:04<50:40,  2.66it/s]
 24%|██▍       | 2592/10682 [32:05<50:38,  2.66it/s]
 24%|██▍       | 2593/10682 [32:05<50:41,  2.66it/s]
 24%|██▍       | 2594/10682 [32:05<50:37,  2.66it/s]
 24%|██▍       | 2595/10682 [32:06<50:32,  2.67it/s]
 24%|██▍       | 2596/10682 [32:06<50:36,  2.66it/s]
 24%|██▍       | 2597/10682 [32:07<50:32,  2.67it/s]
 24%|██▍       | 2598/10682 [32:07<50:30,  2.67it/s]
 24%|██▍       | 2599/10682 [32:07<50:26,  2.67it/s]
 24%|██▍       | 2600/10682 [32:08<50:27,  2.67it/s]{'loss': 4.5212, 'grad_norm': 0.5515668988227844, 'learning_rate': 0.0009387095054051274, 'epoch': 0.24}
-
                                                    

 24%|██▍       | 2600/10682 [32:08<50:27,  2.67it/s]
 24%|██▍       | 2601/10682 [32:08<50:30,  2.67it/s]
 24%|██▍       | 2602/10682 [32:08<50:28,  2.67it/s]
 24%|██▍       | 2603/10682 [32:09<50:27,  2.67it/s]
 24%|██▍       | 2604/10682 [32:09<50:27,  2.67it/s]
 24%|██▍       | 2605/10682 [32:10<50:31,  2.66it/s]
 24%|██▍       | 2606/10682 [32:10<50:27,  2.67it/s]
 24%|██▍       | 2607/10682 [32:10<50:29,  2.67it/s]
 24%|██▍       | 2608/10682 [32:11<50:31,  2.66it/s]
 24%|██▍       | 2609/10682 [32:11<50:28,  2.67it/s]
 24%|██▍       | 2610/10682 [32:11<50:27,  2.67it/s]
 24%|██▍       | 2611/10682 [32:12<50:26,  2.67it/s]
 24%|██▍       | 2612/10682 [32:12<50:24,  2.67it/s]
 24%|██▍       | 2613/10682 [32:13<50:26,  2.67it/s]
 24%|██▍       | 2614/10682 [32:13<50:25,  2.67it/s]
 24%|██▍       | 2615/10682 [32:13<50:26,  2.67it/s]
 24%|██▍       | 2616/10682 [32:14<50:24,  2.67it/s]
 24%|██▍       | 2617/10682 [32:14<50:22,  2.67it/s]
 25%|██▍       | 2618/10682 [32:14<50:30,  2.66it/s]
 25%|██▍       | 2619/10682 [32:15<50:26,  2.66it/s]
 25%|██▍       | 2620/10682 [32:15<50:31,  2.66it/s]
 25%|██▍       | 2621/10682 [32:16<50:32,  2.66it/s]
 25%|██▍       | 2622/10682 [32:16<50:28,  2.66it/s]
 25%|██▍       | 2623/10682 [32:16<50:28,  2.66it/s]
 25%|██▍       | 2624/10682 [32:17<50:27,  2.66it/s]
 25%|██▍       | 2625/10682 [32:17<50:25,  2.66it/s]{'loss': 4.5195, 'grad_norm': 0.5317202806472778, 'learning_rate': 0.0009367351691304569, 'epoch': 0.25}
                                                    
-
 25%|██▍       | 2625/10682 [32:17<50:25,  2.66it/s]
 25%|██▍       | 2626/10682 [32:17<50:45,  2.65it/s]
 25%|██▍       | 2627/10682 [32:18<50:39,  2.65it/s]
 25%|██▍       | 2628/10682 [32:18<50:32,  2.66it/s]
 25%|██▍       | 2629/10682 [32:19<50:25,  2.66it/s]
 25%|██▍       | 2630/10682 [32:19<50:23,  2.66it/s]
 25%|██▍       | 2631/10682 [32:19<50:17,  2.67it/s]
 25%|██▍       | 2632/10682 [32:20<50:20,  2.67it/s]
 25%|██▍       | 2633/10682 [32:20<50:17,  2.67it/s]
 25%|██▍       | 2634/10682 [32:20<50:31,  2.65it/s]
 25%|██▍       | 2635/10682 [32:21<50:23,  2.66it/s]
 25%|██▍       | 2636/10682 [32:21<50:29,  2.66it/s]
 25%|██▍       | 2637/10682 [32:22<50:28,  2.66it/s]
 25%|██▍       | 2638/10682 [32:22<50:24,  2.66it/s]
 25%|██▍       | 2639/10682 [32:22<50:23,  2.66it/s]
 25%|██▍       | 2640/10682 [32:23<50:20,  2.66it/s]
 25%|██▍       | 2641/10682 [32:23<50:17,  2.66it/s]
 25%|██▍       | 2642/10682 [32:23<50:14,  2.67it/s]
 25%|██▍       | 2643/10682 [32:24<50:12,  2.67it/s]
 25%|██▍       | 2644/10682 [32:24<50:14,  2.67it/s]
 25%|██▍       | 2645/10682 [32:25<50:13,  2.67it/s]
 25%|██▍       | 2646/10682 [32:25<50:15,  2.66it/s]
 25%|██▍       | 2647/10682 [32:25<50:12,  2.67it/s]
 25%|██▍       | 2648/10682 [32:26<50:15,  2.66it/s]
 25%|██▍       | 2649/10682 [32:26<50:13,  2.67it/s]
 25%|██▍       | 2650/10682 [32:26<50:14,  2.66it/s]{'loss': 4.5159, 'grad_norm': 0.49796804785728455, 'learning_rate': 0.0009347316802333359, 'epoch': 0.25}
                                                    
-
 25%|██▍       | 2650/10682 [32:26<50:14,  2.66it/s]
 25%|██▍       | 2651/10682 [32:27<50:16,  2.66it/s]
 25%|██▍       | 2652/10682 [32:27<50:14,  2.66it/s]
 25%|██▍       | 2653/10682 [32:28<50:09,  2.67it/s]
 25%|██▍       | 2654/10682 [32:28<50:07,  2.67it/s]
 25%|██▍       | 2655/10682 [32:28<50:07,  2.67it/s]
 25%|██▍       | 2656/10682 [32:29<50:06,  2.67it/s]
 25%|██▍       | 2657/10682 [32:29<50:06,  2.67it/s]
 25%|██▍       | 2658/10682 [32:29<50:06,  2.67it/s]
 25%|██▍       | 2659/10682 [32:30<50:11,  2.66it/s]
 25%|██▍       | 2660/10682 [32:30<50:12,  2.66it/s]
 25%|██▍       | 2661/10682 [32:31<50:08,  2.67it/s]
 25%|██▍       | 2662/10682 [32:31<50:09,  2.67it/s]
 25%|██▍       | 2663/10682 [32:31<50:09,  2.66it/s]
 25%|██▍       | 2664/10682 [32:32<50:08,  2.67it/s]
 25%|██▍       | 2665/10682 [32:32<50:04,  2.67it/s]
 25%|██▍       | 2666/10682 [32:32<50:01,  2.67it/s]
 25%|██▍       | 2667/10682 [32:33<50:01,  2.67it/s]
 25%|██▍       | 2668/10682 [32:33<50:02,  2.67it/s]
 25%|██▍       | 2669/10682 [32:34<49:58,  2.67it/s]
 25%|██▍       | 2670/10682 [32:34<49:56,  2.67it/s]
 25%|██▌       | 2671/10682 [32:34<49:57,  2.67it/s]
 25%|██▌       | 2672/10682 [32:35<49:56,  2.67it/s]
 25%|██▌       | 2673/10682 [32:35<50:01,  2.67it/s]
 25%|██▌       | 2674/10682 [32:35<49:59,  2.67it/s]
 25%|██▌       | 2675/10682 [32:36<50:06,  2.66it/s]{'loss': 4.5182, 'grad_norm': 0.5547919273376465, 'learning_rate': 0.0009326991724491712, 'epoch': 0.25}
                                                    
-
 25%|██▌       | 2675/10682 [32:36<50:06,  2.66it/s]
 25%|██▌       | 2676/10682 [32:36<50:23,  2.65it/s]
 25%|██▌       | 2677/10682 [32:37<50:17,  2.65it/s]
 25%|██▌       | 2678/10682 [32:37<50:08,  2.66it/s]
 25%|██▌       | 2679/10682 [32:37<50:02,  2.67it/s]
 25%|██▌       | 2680/10682 [32:38<50:04,  2.66it/s]
 25%|██▌       | 2681/10682 [32:38<50:00,  2.67it/s]
 25%|██▌       | 2682/10682 [32:38<49:59,  2.67it/s]
 25%|██▌       | 2683/10682 [32:39<49:55,  2.67it/s]
 25%|██▌       | 2684/10682 [32:39<49:57,  2.67it/s]
 25%|██▌       | 2685/10682 [32:40<49:55,  2.67it/s]
 25%|██▌       | 2686/10682 [32:40<49:55,  2.67it/s]
 25%|██▌       | 2687/10682 [32:40<49:59,  2.67it/s]
 25%|██▌       | 2688/10682 [32:41<50:00,  2.66it/s]
 25%|██▌       | 2689/10682 [32:41<49:56,  2.67it/s]
 25%|██▌       | 2690/10682 [32:41<50:02,  2.66it/s]
 25%|██▌       | 2691/10682 [32:42<50:02,  2.66it/s]
 25%|██▌       | 2692/10682 [32:42<49:59,  2.66it/s]
 25%|██▌       | 2693/10682 [32:43<49:56,  2.67it/s]
 25%|██▌       | 2694/10682 [32:43<49:55,  2.67it/s]
 25%|██▌       | 2695/10682 [32:43<49:53,  2.67it/s]
 25%|██▌       | 2696/10682 [32:44<49:53,  2.67it/s]
 25%|██▌       | 2697/10682 [32:44<49:51,  2.67it/s]
 25%|██▌       | 2698/10682 [32:44<49:51,  2.67it/s]
 25%|██▌       | 2699/10682 [32:45<49:52,  2.67it/s]
 25%|██▌       | 2700/10682 [32:45<49:49,  2.67it/s]{'loss': 4.5226, 'grad_norm': 0.5646228790283203, 'learning_rate': 0.0009306377814504165, 'epoch': 0.25}
-
                                                    

 25%|██▌       | 2700/10682 [32:45<49:49,  2.67it/s]
 25%|██▌       | 2701/10682 [32:46<49:53,  2.67it/s]
 25%|██▌       | 2702/10682 [32:46<49:56,  2.66it/s]
 25%|██▌       | 2703/10682 [32:46<49:59,  2.66it/s]
 25%|██▌       | 2704/10682 [32:47<49:55,  2.66it/s]
 25%|██▌       | 2705/10682 [32:47<49:51,  2.67it/s]
 25%|██▌       | 2706/10682 [32:47<49:51,  2.67it/s]
 25%|██▌       | 2707/10682 [32:48<49:50,  2.67it/s]
 25%|██▌       | 2708/10682 [32:48<49:52,  2.66it/s]
 25%|██▌       | 2709/10682 [32:49<49:47,  2.67it/s]
 25%|██▌       | 2710/10682 [32:49<49:46,  2.67it/s]
 25%|██▌       | 2711/10682 [32:49<49:49,  2.67it/s]
 25%|██▌       | 2712/10682 [32:50<49:51,  2.66it/s]
 25%|██▌       | 2713/10682 [32:50<49:49,  2.67it/s]
 25%|██▌       | 2714/10682 [32:50<49:56,  2.66it/s]
 25%|██▌       | 2715/10682 [32:51<49:52,  2.66it/s]
 25%|██▌       | 2716/10682 [32:51<49:54,  2.66it/s]
 25%|██▌       | 2717/10682 [32:52<49:55,  2.66it/s]
 25%|██▌       | 2718/10682 [32:52<49:52,  2.66it/s]
 25%|██▌       | 2719/10682 [32:52<49:47,  2.67it/s]
 25%|██▌       | 2720/10682 [32:53<49:43,  2.67it/s]
 25%|██▌       | 2721/10682 [32:53<49:42,  2.67it/s]
 25%|██▌       | 2722/10682 [32:53<49:40,  2.67it/s]
 25%|██▌       | 2723/10682 [32:54<49:42,  2.67it/s]
 26%|██▌       | 2724/10682 [32:54<49:43,  2.67it/s]
 26%|██▌       | 2725/10682 [32:55<49:41,  2.67it/s]
                                                    
{'loss': 4.5091, 'grad_norm': 0.6684327721595764, 'learning_rate': 0.0009285476448375166, 'epoch': 0.26}
-
 26%|██▌       | 2725/10682 [32:55<49:41,  2.67it/s]
 26%|██▌       | 2726/10682 [32:55<49:48,  2.66it/s]
 26%|██▌       | 2727/10682 [32:55<49:45,  2.66it/s]
 26%|██▌       | 2728/10682 [32:56<49:49,  2.66it/s]
 26%|██▌       | 2729/10682 [32:56<49:46,  2.66it/s]
 26%|██▌       | 2730/10682 [32:56<49:47,  2.66it/s]
 26%|██▌       | 2731/10682 [32:57<49:45,  2.66it/s]
 26%|██▌       | 2732/10682 [32:57<49:45,  2.66it/s]
 26%|██▌       | 2733/10682 [32:58<49:45,  2.66it/s]
 26%|██▌       | 2734/10682 [32:58<49:41,  2.67it/s]
 26%|██▌       | 2735/10682 [32:58<49:43,  2.66it/s]
 26%|██▌       | 2736/10682 [32:59<49:42,  2.66it/s]
 26%|██▌       | 2737/10682 [32:59<49:38,  2.67it/s]
 26%|██▌       | 2738/10682 [32:59<49:37,  2.67it/s]
 26%|██▌       | 2739/10682 [33:00<49:35,  2.67it/s]
 26%|██▌       | 2740/10682 [33:00<49:36,  2.67it/s]
 26%|██▌       | 2741/10682 [33:01<49:37,  2.67it/s]
 26%|██▌       | 2742/10682 [33:01<49:42,  2.66it/s]
 26%|██▌       | 2743/10682 [33:01<49:37,  2.67it/s]
 26%|██▌       | 2744/10682 [33:02<49:45,  2.66it/s]
 26%|██▌       | 2745/10682 [33:02<49:42,  2.66it/s]
 26%|██▌       | 2746/10682 [33:02<49:40,  2.66it/s]
 26%|██▌       | 2747/10682 [33:03<49:37,  2.66it/s]
 26%|██▌       | 2748/10682 [33:03<49:36,  2.67it/s]
 26%|██▌       | 2749/10682 [33:04<49:34,  2.67it/s]
 26%|██▌       | 2750/10682 [33:04<49:31,  2.67it/s]
                                                    
{'loss': 4.4979, 'grad_norm': 0.5191640853881836, 'learning_rate': 0.0009264289021297221, 'epoch': 0.26}
-
 26%|██▌       | 2750/10682 [33:04<49:31,  2.67it/s]
 26%|██▌       | 2751/10682 [33:04<49:35,  2.67it/s]
 26%|██▌       | 2752/10682 [33:05<49:33,  2.67it/s]
 26%|██▌       | 2753/10682 [33:05<49:33,  2.67it/s]
 26%|██▌       | 2754/10682 [33:05<49:33,  2.67it/s]
 26%|██▌       | 2755/10682 [33:06<49:33,  2.67it/s]
 26%|██▌       | 2756/10682 [33:06<49:33,  2.67it/s]
 26%|██▌       | 2757/10682 [33:07<49:35,  2.66it/s]
 26%|██▌       | 2758/10682 [33:07<49:38,  2.66it/s]
 26%|██▌       | 2759/10682 [33:07<49:37,  2.66it/s]
 26%|██▌       | 2760/10682 [33:08<49:36,  2.66it/s]
 26%|██▌       | 2761/10682 [33:08<49:34,  2.66it/s]
 26%|██▌       | 2762/10682 [33:08<49:34,  2.66it/s]
 26%|██▌       | 2763/10682 [33:09<49:33,  2.66it/s]
 26%|██▌       | 2764/10682 [33:09<49:34,  2.66it/s]
 26%|██▌       | 2765/10682 [33:10<49:30,  2.66it/s]
 26%|██▌       | 2766/10682 [33:10<49:25,  2.67it/s]
 26%|██▌       | 2767/10682 [33:10<49:29,  2.67it/s]
 26%|██▌       | 2768/10682 [33:11<49:26,  2.67it/s]
 26%|██▌       | 2769/10682 [33:11<49:31,  2.66it/s]
 26%|██▌       | 2770/10682 [33:11<49:30,  2.66it/s]
 26%|██▌       | 2771/10682 [33:12<49:29,  2.66it/s]
 26%|██▌       | 2772/10682 [33:12<49:28,  2.66it/s]
 26%|██▌       | 2773/10682 [33:13<49:30,  2.66it/s]
 26%|██▌       | 2774/10682 [33:13<49:29,  2.66it/s]
 26%|██▌       | 2775/10682 [33:13<49:25,  2.67it/s]
                                                    
{'loss': 4.5025, 'grad_norm': 0.4669909179210663, 'learning_rate': 0.0009242816947557767, 'epoch': 0.26}
-
 26%|██▌       | 2775/10682 [33:13<49:25,  2.67it/s]
 26%|██▌       | 2776/10682 [33:14<49:29,  2.66it/s]
 26%|██▌       | 2777/10682 [33:14<49:29,  2.66it/s]
 26%|██▌       | 2778/10682 [33:14<49:24,  2.67it/s]
 26%|██▌       | 2779/10682 [33:15<49:21,  2.67it/s]
 26%|██▌       | 2780/10682 [33:15<49:22,  2.67it/s]
 26%|██▌       | 2781/10682 [33:16<49:22,  2.67it/s]
 26%|██▌       | 2782/10682 [33:16<49:18,  2.67it/s]
 26%|██▌       | 2783/10682 [33:16<49:20,  2.67it/s]
 26%|██▌       | 2784/10682 [33:17<49:18,  2.67it/s]
 26%|██▌       | 2785/10682 [33:17<49:26,  2.66it/s]
 26%|██▌       | 2786/10682 [33:17<49:22,  2.67it/s]
 26%|██▌       | 2787/10682 [33:18<49:18,  2.67it/s]
 26%|██▌       | 2788/10682 [33:18<49:20,  2.67it/s]
 26%|██▌       | 2789/10682 [33:19<49:22,  2.66it/s]
 26%|██▌       | 2790/10682 [33:19<49:24,  2.66it/s]
 26%|██▌       | 2791/10682 [33:19<49:23,  2.66it/s]
 26%|██▌       | 2792/10682 [33:20<49:23,  2.66it/s]
 26%|██▌       | 2793/10682 [33:20<49:19,  2.67it/s]
 26%|██▌       | 2794/10682 [33:20<49:20,  2.66it/s]
 26%|██▌       | 2795/10682 [33:21<49:19,  2.67it/s]
 26%|██▌       | 2796/10682 [33:21<49:21,  2.66it/s]
 26%|██▌       | 2797/10682 [33:22<49:18,  2.66it/s]
 26%|██▌       | 2798/10682 [33:22<49:23,  2.66it/s]
 26%|██▌       | 2799/10682 [33:22<49:23,  2.66it/s]
 26%|██▌       | 2800/10682 [33:23<49:25,  2.66it/s]
                                                    
{'loss': 4.4932, 'grad_norm': 0.4643097519874573, 'learning_rate': 0.0009221061660444758, 'epoch': 0.26}
-
 26%|██▌       | 2800/10682 [33:23<49:25,  2.66it/s]
 26%|██▌       | 2801/10682 [33:23<49:26,  2.66it/s]
 26%|██▌       | 2802/10682 [33:23<49:18,  2.66it/s]
 26%|██▌       | 2803/10682 [33:24<49:17,  2.66it/s]
 26%|██▌       | 2804/10682 [33:24<49:15,  2.67it/s]
 26%|██▋       | 2805/10682 [33:25<49:13,  2.67it/s]
 26%|██▋       | 2806/10682 [33:25<49:10,  2.67it/s]
 26%|██▋       | 2807/10682 [33:25<49:08,  2.67it/s]
 26%|██▋       | 2808/10682 [33:26<49:06,  2.67it/s]
 26%|██▋       | 2809/10682 [33:26<49:06,  2.67it/s]
 26%|██▋       | 2810/10682 [33:26<49:11,  2.67it/s]
 26%|██▋       | 2811/10682 [33:27<49:09,  2.67it/s]
 26%|██▋       | 2812/10682 [33:27<49:12,  2.67it/s]
 26%|██▋       | 2813/10682 [33:28<49:11,  2.67it/s]
 26%|██▋       | 2814/10682 [33:28<49:09,  2.67it/s]
 26%|██▋       | 2815/10682 [33:28<49:09,  2.67it/s]
 26%|██▋       | 2816/10682 [33:29<49:06,  2.67it/s]
 26%|██▋       | 2817/10682 [33:29<49:08,  2.67it/s]
 26%|██▋       | 2818/10682 [33:29<49:08,  2.67it/s]
 26%|██▋       | 2819/10682 [33:30<49:08,  2.67it/s]
 26%|██▋       | 2820/10682 [33:30<49:08,  2.67it/s]
 26%|██▋       | 2821/10682 [33:31<49:10,  2.66it/s]
 26%|██▋       | 2822/10682 [33:31<49:07,  2.67it/s]
 26%|██▋       | 2823/10682 [33:31<49:10,  2.66it/s]
 26%|██▋       | 2824/10682 [33:32<49:08,  2.67it/s]
 26%|██▋       | 2825/10682 [33:32<49:05,  2.67it/s]
                                                    
{'loss': 4.4924, 'grad_norm': 0.5130152702331543, 'learning_rate': 0.0009199024612151008, 'epoch': 0.26}
-
 26%|██▋       | 2825/10682 [33:32<49:05,  2.67it/s]
 26%|██▋       | 2826/10682 [33:32<49:21,  2.65it/s]
 26%|██▋       | 2827/10682 [33:33<49:21,  2.65it/s]
 26%|██▋       | 2828/10682 [33:33<49:16,  2.66it/s]
 26%|██▋       | 2829/10682 [33:34<49:16,  2.66it/s]
 26%|██▋       | 2830/10682 [33:34<49:14,  2.66it/s]
 27%|██▋       | 2831/10682 [33:34<49:11,  2.66it/s]
 27%|██▋       | 2832/10682 [33:35<49:04,  2.67it/s]
 27%|██▋       | 2833/10682 [33:35<49:04,  2.67it/s]
 27%|██▋       | 2834/10682 [33:35<49:04,  2.67it/s]
 27%|██▋       | 2835/10682 [33:36<49:01,  2.67it/s]
 27%|██▋       | 2836/10682 [33:36<49:00,  2.67it/s]
 27%|██▋       | 2837/10682 [33:37<49:03,  2.67it/s]
 27%|██▋       | 2838/10682 [33:37<49:02,  2.67it/s]
 27%|██▋       | 2839/10682 [33:37<49:05,  2.66it/s]
 27%|██▋       | 2840/10682 [33:38<49:03,  2.66it/s]
 27%|██▋       | 2841/10682 [33:38<49:04,  2.66it/s]
 27%|██▋       | 2842/10682 [33:39<49:02,  2.66it/s]
 27%|██▋       | 2843/10682 [33:39<49:00,  2.67it/s]
 27%|██▋       | 2844/10682 [33:39<49:02,  2.66it/s]
 27%|██▋       | 2845/10682 [33:40<49:02,  2.66it/s]
 27%|██▋       | 2846/10682 [33:40<48:58,  2.67it/s]
 27%|██▋       | 2847/10682 [33:40<49:00,  2.66it/s]
 27%|██▋       | 2848/10682 [33:41<48:57,  2.67it/s]
 27%|██▋       | 2849/10682 [33:41<48:56,  2.67it/s]
 27%|██▋       | 2850/10682 [33:42<48:56,  2.67it/s]
                                                    
{'loss': 4.4856, 'grad_norm': 0.4416806995868683, 'learning_rate': 0.0009176707273677235, 'epoch': 0.27}
-
 27%|██▋       | 2850/10682 [33:42<48:56,  2.67it/s]
 27%|██▋       | 2851/10682 [33:42<48:59,  2.66it/s]
 27%|██▋       | 2852/10682 [33:42<48:57,  2.67it/s]
 27%|██▋       | 2853/10682 [33:43<48:59,  2.66it/s]
 27%|██▋       | 2854/10682 [33:43<48:56,  2.67it/s]
 27%|██▋       | 2855/10682 [33:43<48:57,  2.66it/s]
 27%|██▋       | 2856/10682 [33:44<48:55,  2.67it/s]
 27%|██▋       | 2857/10682 [33:44<49:00,  2.66it/s]
 27%|██▋       | 2858/10682 [33:45<48:59,  2.66it/s]
 27%|██▋       | 2859/10682 [33:45<48:58,  2.66it/s]
 27%|██▋       | 2860/10682 [33:45<48:55,  2.66it/s]
 27%|██▋       | 2861/10682 [33:46<48:53,  2.67it/s]
 27%|██▋       | 2862/10682 [33:46<48:52,  2.67it/s]
 27%|██▋       | 2863/10682 [33:46<48:51,  2.67it/s]
 27%|██▋       | 2864/10682 [33:47<48:51,  2.67it/s]
 27%|██▋       | 2865/10682 [33:47<48:52,  2.67it/s]
 27%|██▋       | 2866/10682 [33:48<48:47,  2.67it/s]
 27%|██▋       | 2867/10682 [33:48<48:50,  2.67it/s]
 27%|██▋       | 2868/10682 [33:48<48:50,  2.67it/s]
 27%|██▋       | 2869/10682 [33:49<48:49,  2.67it/s]
 27%|██▋       | 2870/10682 [33:49<48:50,  2.67it/s]
 27%|██▋       | 2871/10682 [33:49<48:51,  2.66it/s]
 27%|██▋       | 2872/10682 [33:50<48:50,  2.67it/s]
 27%|██▋       | 2873/10682 [33:50<48:47,  2.67it/s]
 27%|██▋       | 2874/10682 [33:51<48:47,  2.67it/s]
 27%|██▋       | 2875/10682 [33:51<48:44,  2.67it/s]{'loss': 4.4881, 'grad_norm': 0.5349512696266174, 'learning_rate': 0.0009154111134733883, 'epoch': 0.27}
                                                    
-
 27%|██▋       | 2875/10682 [33:51<48:44,  2.67it/s]
 27%|██▋       | 2876/10682 [33:51<48:45,  2.67it/s]
 27%|██▋       | 2877/10682 [33:52<48:47,  2.67it/s]
 27%|██▋       | 2878/10682 [33:52<48:48,  2.66it/s]
 27%|██▋       | 2879/10682 [33:52<48:46,  2.67it/s]
 27%|██▋       | 2880/10682 [33:53<48:49,  2.66it/s]
 27%|██▋       | 2881/10682 [33:53<48:50,  2.66it/s]
 27%|██▋       | 2882/10682 [33:54<48:47,  2.66it/s]
 27%|██▋       | 2883/10682 [33:54<48:47,  2.66it/s]
 27%|██▋       | 2884/10682 [33:54<48:42,  2.67it/s]
 27%|██▋       | 2885/10682 [33:55<48:42,  2.67it/s]
 27%|██▋       | 2886/10682 [33:55<48:44,  2.67it/s]
 27%|██▋       | 2887/10682 [33:55<48:47,  2.66it/s]
 27%|██▋       | 2888/10682 [33:56<48:46,  2.66it/s]
 27%|██▋       | 2889/10682 [33:56<48:43,  2.67it/s]
 27%|██▋       | 2890/10682 [33:57<48:45,  2.66it/s]
 27%|██▋       | 2891/10682 [33:57<48:39,  2.67it/s]
 27%|██▋       | 2892/10682 [33:57<48:42,  2.67it/s]
 27%|██▋       | 2893/10682 [33:58<48:40,  2.67it/s]
 27%|██▋       | 2894/10682 [33:58<48:42,  2.66it/s]
 27%|██▋       | 2895/10682 [33:58<48:42,  2.66it/s]
 27%|██▋       | 2896/10682 [33:59<48:40,  2.67it/s]
 27%|██▋       | 2897/10682 [33:59<48:36,  2.67it/s]
 27%|██▋       | 2898/10682 [34:00<48:35,  2.67it/s]
 27%|██▋       | 2899/10682 [34:00<48:37,  2.67it/s]
 27%|██▋       | 2900/10682 [34:00<48:37,  2.67it/s]
                                                    
{'loss': 4.4799, 'grad_norm': 0.45078521966934204, 'learning_rate': 0.0009131237703641679, 'epoch': 0.27}
-
 27%|██▋       | 2900/10682 [34:00<48:37,  2.67it/s]
 27%|██▋       | 2901/10682 [34:01<48:40,  2.66it/s]
 27%|██▋       | 2902/10682 [34:01<48:40,  2.66it/s]
 27%|██▋       | 2903/10682 [34:01<48:40,  2.66it/s]
 27%|██▋       | 2904/10682 [34:02<48:38,  2.66it/s]
 27%|██▋       | 2905/10682 [34:02<48:38,  2.66it/s]
 27%|██▋       | 2906/10682 [34:03<48:39,  2.66it/s]
 27%|██▋       | 2907/10682 [34:03<48:38,  2.66it/s]
 27%|██▋       | 2908/10682 [34:03<48:39,  2.66it/s]
 27%|██▋       | 2909/10682 [34:04<48:38,  2.66it/s]
 27%|██▋       | 2910/10682 [34:04<48:36,  2.67it/s]
 27%|██▋       | 2911/10682 [34:04<48:38,  2.66it/s]
 27%|██▋       | 2912/10682 [34:05<48:38,  2.66it/s]
 27%|██▋       | 2913/10682 [34:05<48:37,  2.66it/s]
 27%|██▋       | 2914/10682 [34:06<48:35,  2.66it/s]
 27%|██▋       | 2915/10682 [34:06<48:35,  2.66it/s]
 27%|██▋       | 2916/10682 [34:06<48:30,  2.67it/s]
 27%|██▋       | 2917/10682 [34:07<48:29,  2.67it/s]
 27%|██▋       | 2918/10682 [34:07<48:32,  2.67it/s]
 27%|██▋       | 2919/10682 [34:07<48:35,  2.66it/s]
 27%|██▋       | 2920/10682 [34:08<48:34,  2.66it/s]
 27%|██▋       | 2921/10682 [34:08<48:36,  2.66it/s]
 27%|██▋       | 2922/10682 [34:09<48:33,  2.66it/s]
 27%|██▋       | 2923/10682 [34:09<48:33,  2.66it/s]
 27%|██▋       | 2924/10682 [34:09<48:33,  2.66it/s]
 27%|██▋       | 2925/10682 [34:10<48:31,  2.66it/s]
                                                    
{'loss': 4.4843, 'grad_norm': 0.4466519355773926, 'learning_rate': 0.0009108088507230948, 'epoch': 0.27}
-
 27%|██▋       | 2925/10682 [34:10<48:31,  2.66it/s]
 27%|██▋       | 2926/10682 [34:10<48:32,  2.66it/s]
 27%|██▋       | 2927/10682 [34:10<48:32,  2.66it/s]
 27%|██▋       | 2928/10682 [34:11<48:28,  2.67it/s]
 27%|██▋       | 2929/10682 [34:11<48:28,  2.67it/s]
 27%|██▋       | 2930/10682 [34:12<48:28,  2.67it/s]
 27%|██▋       | 2931/10682 [34:12<48:29,  2.66it/s]
 27%|██▋       | 2932/10682 [34:12<48:24,  2.67it/s]
 27%|██▋       | 2933/10682 [34:13<48:29,  2.66it/s]
 27%|██▋       | 2934/10682 [34:13<48:28,  2.66it/s]
 27%|██▋       | 2935/10682 [34:13<48:28,  2.66it/s]
 27%|██▋       | 2936/10682 [34:14<48:28,  2.66it/s]
 27%|██▋       | 2937/10682 [34:14<48:29,  2.66it/s]
 28%|██▊       | 2938/10682 [34:15<48:26,  2.66it/s]
 28%|██▊       | 2939/10682 [34:15<48:22,  2.67it/s]
 28%|██▊       | 2940/10682 [34:15<48:18,  2.67it/s]
 28%|██▊       | 2941/10682 [34:16<48:18,  2.67it/s]
 28%|██▊       | 2942/10682 [34:16<48:18,  2.67it/s]
 28%|██▊       | 2943/10682 [34:16<48:17,  2.67it/s]
 28%|██▊       | 2944/10682 [34:17<48:17,  2.67it/s]
 28%|██▊       | 2945/10682 [34:17<48:17,  2.67it/s]
 28%|██▊       | 2946/10682 [34:18<48:14,  2.67it/s]
 28%|██▊       | 2947/10682 [34:18<48:15,  2.67it/s]
 28%|██▊       | 2948/10682 [34:18<48:15,  2.67it/s]
 28%|██▊       | 2949/10682 [34:19<48:17,  2.67it/s]
 28%|██▊       | 2950/10682 [34:19<48:16,  2.67it/s]
                                                    
{'loss': 4.4712, 'grad_norm': 0.45295438170433044, 'learning_rate': 0.0009084665090739699, 'epoch': 0.28}
-
 28%|██▊       | 2950/10682 [34:19<48:16,  2.67it/s]
 28%|██▊       | 2951/10682 [34:19<48:18,  2.67it/s]
 28%|██▊       | 2952/10682 [34:20<48:20,  2.67it/s]
 28%|██▊       | 2953/10682 [34:20<48:21,  2.66it/s]
 28%|██▊       | 2954/10682 [34:21<48:20,  2.66it/s]
 28%|██▊       | 2955/10682 [34:21<48:19,  2.66it/s]
 28%|██▊       | 2956/10682 [34:21<48:18,  2.67it/s]
 28%|██▊       | 2957/10682 [34:22<48:14,  2.67it/s]
 28%|██▊       | 2958/10682 [34:22<48:16,  2.67it/s]
 28%|██▊       | 2959/10682 [34:22<48:18,  2.66it/s]
 28%|██▊       | 2960/10682 [34:23<48:20,  2.66it/s]
 28%|██▊       | 2961/10682 [34:23<48:21,  2.66it/s]
 28%|██▊       | 2962/10682 [34:24<48:20,  2.66it/s]
 28%|██▊       | 2963/10682 [34:24<48:21,  2.66it/s]
 28%|██▊       | 2964/10682 [34:24<48:20,  2.66it/s]
 28%|██▊       | 2965/10682 [34:25<48:22,  2.66it/s]
 28%|██▊       | 2966/10682 [34:25<48:20,  2.66it/s]
 28%|██▊       | 2967/10682 [34:25<48:19,  2.66it/s]
 28%|██▊       | 2968/10682 [34:26<48:16,  2.66it/s]
 28%|██▊       | 2969/10682 [34:26<48:11,  2.67it/s]
 28%|██▊       | 2970/10682 [34:27<48:07,  2.67it/s]
 28%|██▊       | 2971/10682 [34:27<48:09,  2.67it/s]
 28%|██▊       | 2972/10682 [34:27<48:08,  2.67it/s]
 28%|██▊       | 2973/10682 [34:28<48:08,  2.67it/s]
 28%|██▊       | 2974/10682 [34:28<48:10,  2.67it/s]
 28%|██▊       | 2975/10682 [34:28<48:12,  2.66it/s]{'loss': 4.4764, 'grad_norm': 0.45623284578323364, 'learning_rate': 0.0009060969017710476, 'epoch': 0.28}
-
                                                    

 28%|██▊       | 2975/10682 [34:28<48:12,  2.66it/s]
 28%|██▊       | 2976/10682 [34:29<48:21,  2.66it/s]
 28%|██▊       | 2977/10682 [34:29<48:18,  2.66it/s]
 28%|██▊       | 2978/10682 [34:30<48:18,  2.66it/s]
 28%|██▊       | 2979/10682 [34:30<48:15,  2.66it/s]
 28%|██▊       | 2980/10682 [34:30<48:11,  2.66it/s]
 28%|██▊       | 2981/10682 [34:31<48:11,  2.66it/s]
 28%|██▊       | 2982/10682 [34:31<48:10,  2.66it/s]
 28%|██▊       | 2983/10682 [34:31<48:10,  2.66it/s]
 28%|██▊       | 2984/10682 [34:32<48:07,  2.67it/s]
 28%|██▊       | 2985/10682 [34:32<48:04,  2.67it/s]
 28%|██▊       | 2986/10682 [34:33<48:07,  2.67it/s]
 28%|██▊       | 2987/10682 [34:33<48:09,  2.66it/s]
 28%|██▊       | 2988/10682 [34:33<48:11,  2.66it/s]
 28%|██▊       | 2989/10682 [34:34<48:10,  2.66it/s]
 28%|██▊       | 2990/10682 [34:34<48:15,  2.66it/s]
 28%|██▊       | 2991/10682 [34:34<48:11,  2.66it/s]
 28%|██▊       | 2992/10682 [34:35<48:08,  2.66it/s]
 28%|██▊       | 2993/10682 [34:35<48:07,  2.66it/s]
 28%|██▊       | 2994/10682 [34:36<48:03,  2.67it/s]
 28%|██▊       | 2995/10682 [34:36<48:00,  2.67it/s]
 28%|██▊       | 2996/10682 [34:36<48:01,  2.67it/s]
 28%|██▊       | 2997/10682 [34:37<48:01,  2.67it/s]
 28%|██▊       | 2998/10682 [34:37<48:01,  2.67it/s]
 28%|██▊       | 2999/10682 [34:37<48:02,  2.67it/s]
 28%|██▊       | 3000/10682 [34:38<48:02,  2.66it/s]{'loss': 4.4714, 'grad_norm': 0.5410290360450745, 'learning_rate': 0.0009037001869885992, 'epoch': 0.28}
                                                    
-
 28%|██▊       | 3000/10682 [34:38<48:02,  2.66it/s]
 28%|██▊       | 3001/10682 [34:38<48:09,  2.66it/s]
 28%|██▊       | 3002/10682 [34:39<48:06,  2.66it/s]
 28%|██▊       | 3003/10682 [34:39<48:04,  2.66it/s]
 28%|██▊       | 3004/10682 [34:39<48:05,  2.66it/s]
 28%|██▊       | 3005/10682 [34:40<48:02,  2.66it/s]
 28%|██▊       | 3006/10682 [34:40<48:01,  2.66it/s]
 28%|██▊       | 3007/10682 [34:40<47:56,  2.67it/s]
 28%|██▊       | 3008/10682 [34:41<47:55,  2.67it/s]
 28%|██▊       | 3009/10682 [34:41<47:55,  2.67it/s]
 28%|██▊       | 3010/10682 [34:42<47:52,  2.67it/s]
 28%|██▊       | 3011/10682 [34:42<47:52,  2.67it/s]
 28%|██▊       | 3012/10682 [34:42<48:10,  2.65it/s]
 28%|██▊       | 3013/10682 [34:43<48:05,  2.66it/s]
 28%|██▊       | 3014/10682 [34:43<47:58,  2.66it/s]
 28%|██▊       | 3015/10682 [34:43<48:01,  2.66it/s]
 28%|██▊       | 3016/10682 [34:44<47:54,  2.67it/s]
 28%|██▊       | 3017/10682 [34:44<47:58,  2.66it/s]
 28%|██▊       | 3018/10682 [34:45<47:55,  2.67it/s]
 28%|██▊       | 3019/10682 [34:45<47:52,  2.67it/s]
 28%|██▊       | 3020/10682 [34:45<47:51,  2.67it/s]
 28%|██▊       | 3021/10682 [34:46<47:51,  2.67it/s]
 28%|██▊       | 3022/10682 [34:46<47:52,  2.67it/s]
 28%|██▊       | 3023/10682 [34:46<47:50,  2.67it/s]
 28%|██▊       | 3024/10682 [34:47<47:47,  2.67it/s]
 28%|██▊       | 3025/10682 [34:47<47:45,  2.67it/s]
                                                    
{'loss': 4.4733, 'grad_norm': 0.4982568621635437, 'learning_rate': 0.0009012765247103541, 'epoch': 0.28}
-
 28%|██▊       | 3025/10682 [34:47<47:45,  2.67it/s]
 28%|██▊       | 3026/10682 [34:48<47:48,  2.67it/s]
 28%|██▊       | 3027/10682 [34:48<47:48,  2.67it/s]
 28%|██▊       | 3028/10682 [34:48<47:49,  2.67it/s]
 28%|██▊       | 3029/10682 [34:49<47:49,  2.67it/s]
 28%|██▊       | 3030/10682 [34:49<47:55,  2.66it/s]
 28%|██▊       | 3031/10682 [34:49<47:55,  2.66it/s]
 28%|██▊       | 3032/10682 [34:50<47:50,  2.66it/s]
 28%|██▊       | 3033/10682 [34:50<47:49,  2.67it/s]
 28%|██▊       | 3034/10682 [34:51<47:48,  2.67it/s]
 28%|██▊       | 3035/10682 [34:51<47:45,  2.67it/s]
 28%|██▊       | 3036/10682 [34:51<47:45,  2.67it/s]
 28%|██▊       | 3037/10682 [34:52<47:42,  2.67it/s]
 28%|██▊       | 3038/10682 [34:52<47:41,  2.67it/s]
 28%|██▊       | 3039/10682 [34:52<47:41,  2.67it/s]
 28%|██▊       | 3040/10682 [34:53<47:41,  2.67it/s]
 28%|██▊       | 3041/10682 [34:53<47:43,  2.67it/s]
 28%|██▊       | 3042/10682 [34:54<47:48,  2.66it/s]
 28%|██▊       | 3043/10682 [34:54<47:45,  2.67it/s]
 28%|██▊       | 3044/10682 [34:54<47:43,  2.67it/s]
 29%|██▊       | 3045/10682 [34:55<47:45,  2.67it/s]
 29%|██▊       | 3046/10682 [34:55<47:42,  2.67it/s]
 29%|██▊       | 3047/10682 [34:55<47:43,  2.67it/s]
 29%|██▊       | 3048/10682 [34:56<47:39,  2.67it/s]
 29%|██▊       | 3049/10682 [34:56<47:38,  2.67it/s]
 29%|██▊       | 3050/10682 [34:57<47:37,  2.67it/s]{'loss': 4.4705, 'grad_norm': 0.47466275095939636, 'learning_rate': 0.000898826076718821, 'epoch': 0.29}
                                                    
-
 29%|██▊       | 3050/10682 [34:57<47:37,  2.67it/s]
 29%|██▊       | 3051/10682 [34:57<47:39,  2.67it/s]
 29%|██▊       | 3052/10682 [34:57<47:38,  2.67it/s]
 29%|██▊       | 3053/10682 [34:58<47:36,  2.67it/s]
 29%|██▊       | 3054/10682 [34:58<47:34,  2.67it/s]
 29%|██▊       | 3055/10682 [34:58<47:36,  2.67it/s]
 29%|██▊       | 3056/10682 [34:59<47:37,  2.67it/s]
 29%|██▊       | 3057/10682 [34:59<47:35,  2.67it/s]
 29%|██▊       | 3058/10682 [35:00<47:40,  2.67it/s]
 29%|██▊       | 3059/10682 [35:00<47:37,  2.67it/s]
 29%|██▊       | 3060/10682 [35:00<47:37,  2.67it/s]
 29%|██▊       | 3061/10682 [35:01<47:37,  2.67it/s]
 29%|██▊       | 3062/10682 [35:01<47:38,  2.67it/s]
 29%|██▊       | 3063/10682 [35:01<47:39,  2.66it/s]
 29%|██▊       | 3064/10682 [35:02<47:37,  2.67it/s]
 29%|██▊       | 3065/10682 [35:02<47:37,  2.67it/s]
 29%|██▊       | 3066/10682 [35:03<47:34,  2.67it/s]
 29%|██▊       | 3067/10682 [35:03<47:36,  2.67it/s]
 29%|██▊       | 3068/10682 [35:03<47:33,  2.67it/s]
 29%|██▊       | 3069/10682 [35:04<47:34,  2.67it/s]
 29%|██▊       | 3070/10682 [35:04<47:33,  2.67it/s]
 29%|██▊       | 3071/10682 [35:04<47:37,  2.66it/s]
 29%|██▉       | 3072/10682 [35:05<47:39,  2.66it/s]
 29%|██▉       | 3073/10682 [35:05<47:36,  2.66it/s]
 29%|██▉       | 3074/10682 [35:06<47:33,  2.67it/s]
 29%|██▉       | 3075/10682 [35:06<47:31,  2.67it/s]
                                                    
{'loss': 4.4653, 'grad_norm': 0.48210328817367554, 'learning_rate': 0.0008963490065844889, 'epoch': 0.29}
-
 29%|██▉       | 3075/10682 [35:06<47:31,  2.67it/s]
 29%|██▉       | 3076/10682 [35:06<47:38,  2.66it/s]
 29%|██▉       | 3077/10682 [35:07<47:36,  2.66it/s]
 29%|██▉       | 3078/10682 [35:07<47:35,  2.66it/s]
 29%|██▉       | 3079/10682 [35:07<47:30,  2.67it/s]
 29%|██▉       | 3080/10682 [35:08<47:29,  2.67it/s]
 29%|██▉       | 3081/10682 [35:08<47:27,  2.67it/s]
 29%|██▉       | 3082/10682 [35:09<47:27,  2.67it/s]
 29%|██▉       | 3083/10682 [35:09<47:32,  2.66it/s]
 29%|██▉       | 3084/10682 [35:09<47:33,  2.66it/s]
 29%|██▉       | 3085/10682 [35:10<47:37,  2.66it/s]
 29%|██▉       | 3086/10682 [35:10<47:36,  2.66it/s]
 29%|██▉       | 3087/10682 [35:10<47:33,  2.66it/s]
 29%|██▉       | 3088/10682 [35:11<47:32,  2.66it/s]
 29%|██▉       | 3089/10682 [35:11<47:30,  2.66it/s]
 29%|██▉       | 3090/10682 [35:12<47:27,  2.67it/s]
 29%|██▉       | 3091/10682 [35:12<47:24,  2.67it/s]
 29%|██▉       | 3092/10682 [35:12<47:25,  2.67it/s]
 29%|██▉       | 3093/10682 [35:13<47:45,  2.65it/s]
 29%|██▉       | 3094/10682 [35:13<47:40,  2.65it/s]
 29%|██▉       | 3095/10682 [35:13<47:35,  2.66it/s]
 29%|██▉       | 3096/10682 [35:14<47:37,  2.65it/s]
 29%|██▉       | 3097/10682 [35:14<47:36,  2.66it/s]
 29%|██▉       | 3098/10682 [35:15<47:30,  2.66it/s]
 29%|██▉       | 3099/10682 [35:15<47:36,  2.65it/s]
 29%|██▉       | 3100/10682 [35:15<47:33,  2.66it/s]{'loss': 4.4698, 'grad_norm': 0.44848933815956116, 'learning_rate': 0.0008938454796549082, 'epoch': 0.29}
                                                    
-
 29%|██▉       | 3100/10682 [35:15<47:33,  2.66it/s]
 29%|██▉       | 3101/10682 [35:16<47:35,  2.65it/s]
 29%|██▉       | 3102/10682 [35:16<47:31,  2.66it/s]
 29%|██▉       | 3103/10682 [35:16<47:29,  2.66it/s]
 29%|██▉       | 3104/10682 [35:17<47:28,  2.66it/s]
 29%|██▉       | 3105/10682 [35:17<47:24,  2.66it/s]
 29%|██▉       | 3106/10682 [35:18<47:28,  2.66it/s]
 29%|██▉       | 3107/10682 [35:18<47:22,  2.66it/s]
 29%|██▉       | 3108/10682 [35:18<47:19,  2.67it/s]
 29%|██▉       | 3109/10682 [35:19<47:19,  2.67it/s]
 29%|██▉       | 3110/10682 [35:19<47:29,  2.66it/s]
 29%|██▉       | 3111/10682 [35:19<47:25,  2.66it/s]
 29%|██▉       | 3112/10682 [35:20<47:24,  2.66it/s]
 29%|██▉       | 3113/10682 [35:20<47:24,  2.66it/s]
 29%|██▉       | 3114/10682 [35:21<47:24,  2.66it/s]
 29%|██▉       | 3115/10682 [35:21<47:23,  2.66it/s]
 29%|██▉       | 3116/10682 [35:21<47:24,  2.66it/s]
 29%|██▉       | 3117/10682 [35:22<47:21,  2.66it/s]
 29%|██▉       | 3118/10682 [35:22<47:20,  2.66it/s]
 29%|██▉       | 3119/10682 [35:22<47:16,  2.67it/s]
 29%|██▉       | 3120/10682 [35:23<47:14,  2.67it/s]
 29%|██▉       | 3121/10682 [35:23<47:15,  2.67it/s]
 29%|██▉       | 3122/10682 [35:24<47:12,  2.67it/s]
 29%|██▉       | 3123/10682 [35:24<47:14,  2.67it/s]
 29%|██▉       | 3124/10682 [35:24<47:16,  2.66it/s]
 29%|██▉       | 3125/10682 [35:25<47:13,  2.67it/s]{'loss': 4.4544, 'grad_norm': 0.4247789978981018, 'learning_rate': 0.000891315663043654, 'epoch': 0.29}
                                                    
-
 29%|██▉       | 3125/10682 [35:25<47:13,  2.67it/s]
 29%|██▉       | 3126/10682 [35:25<47:18,  2.66it/s]
 29%|██▉       | 3127/10682 [35:25<47:17,  2.66it/s]
 29%|██▉       | 3128/10682 [35:26<47:16,  2.66it/s]
 29%|██▉       | 3129/10682 [35:26<47:12,  2.67it/s]
 29%|██▉       | 3130/10682 [35:27<47:10,  2.67it/s]
 29%|██▉       | 3131/10682 [35:27<47:11,  2.67it/s]
 29%|██▉       | 3132/10682 [35:27<47:09,  2.67it/s]
 29%|██▉       | 3133/10682 [35:28<47:11,  2.67it/s]
 29%|██▉       | 3134/10682 [35:28<47:07,  2.67it/s]
 29%|██▉       | 3135/10682 [35:28<47:07,  2.67it/s]
 29%|██▉       | 3136/10682 [35:29<47:08,  2.67it/s]
 29%|██▉       | 3137/10682 [35:29<47:13,  2.66it/s]
 29%|██▉       | 3138/10682 [35:30<47:09,  2.67it/s]
 29%|██▉       | 3139/10682 [35:30<47:12,  2.66it/s]
 29%|██▉       | 3140/10682 [35:30<47:10,  2.66it/s]
 29%|██▉       | 3141/10682 [35:31<47:07,  2.67it/s]
 29%|██▉       | 3142/10682 [35:31<47:10,  2.66it/s]
 29%|██▉       | 3143/10682 [35:31<47:05,  2.67it/s]
 29%|██▉       | 3144/10682 [35:32<47:07,  2.67it/s]
 29%|██▉       | 3145/10682 [35:32<47:10,  2.66it/s]
 29%|██▉       | 3146/10682 [35:33<47:09,  2.66it/s]
 29%|██▉       | 3147/10682 [35:33<47:04,  2.67it/s]
 29%|██▉       | 3148/10682 [35:33<47:03,  2.67it/s]
 29%|██▉       | 3149/10682 [35:34<47:00,  2.67it/s]
 29%|██▉       | 3150/10682 [35:34<47:06,  2.67it/s]{'loss': 4.46, 'grad_norm': 0.4564974308013916, 'learning_rate': 0.0008887597256191701, 'epoch': 0.29}
-
                                                    

 29%|██▉       | 3150/10682 [35:34<47:06,  2.67it/s]
 29%|██▉       | 3151/10682 [35:34<47:08,  2.66it/s]
 30%|██▉       | 3152/10682 [35:35<47:06,  2.66it/s]
 30%|██▉       | 3153/10682 [35:35<47:06,  2.66it/s]
 30%|██▉       | 3154/10682 [35:36<47:03,  2.67it/s]
 30%|██▉       | 3155/10682 [35:36<47:03,  2.67it/s]
 30%|██▉       | 3156/10682 [35:36<47:00,  2.67it/s]
 30%|██▉       | 3157/10682 [35:37<47:00,  2.67it/s]
 30%|██▉       | 3158/10682 [35:37<46:58,  2.67it/s]
 30%|██▉       | 3159/10682 [35:37<46:58,  2.67it/s]
 30%|██▉       | 3160/10682 [35:38<46:55,  2.67it/s]
 30%|██▉       | 3161/10682 [35:38<46:54,  2.67it/s]
 30%|██▉       | 3162/10682 [35:39<47:17,  2.65it/s]
 30%|██▉       | 3163/10682 [35:39<47:16,  2.65it/s]
 30%|██▉       | 3164/10682 [35:39<47:15,  2.65it/s]
 30%|██▉       | 3165/10682 [35:40<47:07,  2.66it/s]
 30%|██▉       | 3166/10682 [35:40<47:09,  2.66it/s]
 30%|██▉       | 3167/10682 [35:40<47:05,  2.66it/s]
 30%|██▉       | 3168/10682 [35:41<47:26,  2.64it/s]
 30%|██▉       | 3169/10682 [35:41<47:18,  2.65it/s]
 30%|██▉       | 3170/10682 [35:42<47:10,  2.65it/s]
 30%|██▉       | 3171/10682 [35:42<47:06,  2.66it/s]
 30%|██▉       | 3172/10682 [35:42<47:00,  2.66it/s]
 30%|██▉       | 3173/10682 [35:43<46:57,  2.67it/s]
 30%|██▉       | 3174/10682 [35:43<46:54,  2.67it/s]
 30%|██▉       | 3175/10682 [35:43<46:55,  2.67it/s]{'loss': 4.4497, 'grad_norm': 0.4883781373500824, 'learning_rate': 0.0008861778379934982, 'epoch': 0.3}
-
                                                    

 30%|██▉       | 3175/10682 [35:43<46:55,  2.67it/s]
 30%|██▉       | 3176/10682 [35:44<46:56,  2.66it/s]
 30%|██▉       | 3177/10682 [35:44<46:59,  2.66it/s]
 30%|██▉       | 3178/10682 [35:45<47:01,  2.66it/s]
 30%|██▉       | 3179/10682 [35:45<47:03,  2.66it/s]
 30%|██▉       | 3180/10682 [35:45<47:05,  2.66it/s]
 30%|██▉       | 3181/10682 [35:46<46:59,  2.66it/s]
 30%|██▉       | 3182/10682 [35:46<47:00,  2.66it/s]
 30%|██▉       | 3183/10682 [35:46<46:55,  2.66it/s]
 30%|██▉       | 3184/10682 [35:47<46:54,  2.66it/s]
 30%|██▉       | 3185/10682 [35:48<57:15,  2.18it/s]
 30%|██▉       | 3186/10682 [35:48<54:04,  2.31it/s]
 30%|██▉       | 3187/10682 [35:48<51:53,  2.41it/s]
 30%|██▉       | 3188/10682 [35:49<50:24,  2.48it/s]
 30%|██▉       | 3189/10682 [35:49<49:18,  2.53it/s]
 30%|██▉       | 3190/10682 [35:49<48:36,  2.57it/s]
 30%|██▉       | 3191/10682 [35:50<48:02,  2.60it/s]
 30%|██▉       | 3192/10682 [35:50<47:43,  2.62it/s]
 30%|██▉       | 3193/10682 [35:51<47:28,  2.63it/s]
 30%|██▉       | 3194/10682 [35:51<47:27,  2.63it/s]
 30%|██▉       | 3195/10682 [35:51<47:14,  2.64it/s]
 30%|██▉       | 3196/10682 [35:52<47:03,  2.65it/s]
 30%|██▉       | 3197/10682 [35:52<47:00,  2.65it/s]
 30%|██▉       | 3198/10682 [35:53<57:07,  2.18it/s]
 30%|██▉       | 3199/10682 [35:53<54:05,  2.31it/s]
 30%|██▉       | 3200/10682 [35:53<51:58,  2.40it/s]{'loss': 4.4524, 'grad_norm': 0.48857221007347107, 'learning_rate': 0.000883570172510888, 'epoch': 0.3}
                                                    
-
 30%|██▉       | 3200/10682 [35:53<51:58,  2.40it/s]
 30%|██▉       | 3201/10682 [35:54<50:23,  2.47it/s]
 30%|██▉       | 3202/10682 [35:54<49:14,  2.53it/s]
 30%|██▉       | 3203/10682 [35:55<48:32,  2.57it/s]
 30%|██▉       | 3204/10682 [35:55<47:57,  2.60it/s]
 30%|███       | 3205/10682 [35:55<47:34,  2.62it/s]
 30%|███       | 3206/10682 [35:56<47:18,  2.63it/s]
 30%|███       | 3207/10682 [35:56<47:07,  2.64it/s]
 30%|███       | 3208/10682 [35:56<46:59,  2.65it/s]
 30%|███       | 3209/10682 [35:57<46:52,  2.66it/s]
 30%|███       | 3210/10682 [35:57<46:46,  2.66it/s]
 30%|███       | 3211/10682 [35:58<46:41,  2.67it/s]
 30%|███       | 3212/10682 [35:58<46:41,  2.67it/s]
 30%|███       | 3213/10682 [35:58<46:37,  2.67it/s]
 30%|███       | 3214/10682 [35:59<46:36,  2.67it/s]
 30%|███       | 3215/10682 [35:59<46:33,  2.67it/s]
 30%|███       | 3216/10682 [35:59<46:34,  2.67it/s]
 30%|███       | 3217/10682 [36:00<46:34,  2.67it/s]
 30%|███       | 3218/10682 [36:00<46:40,  2.67it/s]
 30%|███       | 3219/10682 [36:01<46:40,  2.67it/s]
 30%|███       | 3220/10682 [36:01<46:41,  2.66it/s]
 30%|███       | 3221/10682 [36:01<46:42,  2.66it/s]
 30%|███       | 3222/10682 [36:02<46:41,  2.66it/s]
 30%|███       | 3223/10682 [36:02<46:40,  2.66it/s]
 30%|███       | 3224/10682 [36:02<46:36,  2.67it/s]
 30%|███       | 3225/10682 [36:03<46:35,  2.67it/s]{'loss': 4.4376, 'grad_norm': 0.44689470529556274, 'learning_rate': 0.0008809369032362939, 'epoch': 0.3}
                                                    
-
 30%|███       | 3225/10682 [36:03<46:35,  2.67it/s]
 30%|███       | 3226/10682 [36:03<46:38,  2.66it/s]
 30%|███       | 3227/10682 [36:04<46:40,  2.66it/s]
 30%|███       | 3228/10682 [36:04<46:36,  2.67it/s]
 30%|███       | 3229/10682 [36:04<46:32,  2.67it/s]
 30%|███       | 3230/10682 [36:05<46:34,  2.67it/s]
 30%|███       | 3231/10682 [36:05<46:33,  2.67it/s]
 30%|███       | 3232/10682 [36:05<46:37,  2.66it/s]
 30%|███       | 3233/10682 [36:06<46:36,  2.66it/s]
 30%|███       | 3234/10682 [36:06<46:36,  2.66it/s]
 30%|███       | 3235/10682 [36:07<46:34,  2.67it/s]
 30%|███       | 3236/10682 [36:07<46:36,  2.66it/s]
 30%|███       | 3237/10682 [36:07<46:35,  2.66it/s]
 30%|███       | 3238/10682 [36:08<46:34,  2.66it/s]
 30%|███       | 3239/10682 [36:08<46:33,  2.66it/s]
 30%|███       | 3240/10682 [36:08<46:30,  2.67it/s]
 30%|███       | 3241/10682 [36:09<46:30,  2.67it/s]
 30%|███       | 3242/10682 [36:09<46:29,  2.67it/s]
 30%|███       | 3243/10682 [36:10<46:28,  2.67it/s]
 30%|███       | 3244/10682 [36:10<46:29,  2.67it/s]
 30%|███       | 3245/10682 [36:10<46:33,  2.66it/s]
 30%|███       | 3246/10682 [36:11<46:31,  2.66it/s]
 30%|███       | 3247/10682 [36:11<46:28,  2.67it/s]
 30%|███       | 3248/10682 [36:11<46:31,  2.66it/s]
 30%|███       | 3249/10682 [36:12<46:29,  2.66it/s]
 30%|███       | 3250/10682 [36:12<46:29,  2.66it/s]
                                                    
{'loss': 4.4506, 'grad_norm': 0.4307403266429901, 'learning_rate': 0.0008782782059437559, 'epoch': 0.3}
-
 30%|███       | 3250/10682 [36:12<46:29,  2.66it/s]
 30%|███       | 3251/10682 [36:13<46:30,  2.66it/s]
 30%|███       | 3252/10682 [36:13<46:29,  2.66it/s]
 30%|███       | 3253/10682 [36:13<46:28,  2.66it/s]
 30%|███       | 3254/10682 [36:14<46:27,  2.67it/s]
 30%|███       | 3255/10682 [36:14<46:27,  2.66it/s]
 30%|███       | 3256/10682 [36:14<46:29,  2.66it/s]
 30%|███       | 3257/10682 [36:15<46:26,  2.66it/s]
 30%|███       | 3258/10682 [36:15<46:24,  2.67it/s]
 31%|███       | 3259/10682 [36:16<46:30,  2.66it/s]
 31%|███       | 3260/10682 [36:16<46:27,  2.66it/s]
 31%|███       | 3261/10682 [36:16<46:31,  2.66it/s]
 31%|███       | 3262/10682 [36:17<46:35,  2.65it/s]
 31%|███       | 3263/10682 [36:17<46:32,  2.66it/s]
 31%|███       | 3264/10682 [36:17<46:27,  2.66it/s]
 31%|███       | 3265/10682 [36:18<46:24,  2.66it/s]
 31%|███       | 3266/10682 [36:18<46:26,  2.66it/s]
 31%|███       | 3267/10682 [36:19<46:23,  2.66it/s]
 31%|███       | 3268/10682 [36:19<46:23,  2.66it/s]
 31%|███       | 3269/10682 [36:19<46:22,  2.66it/s]
 31%|███       | 3270/10682 [36:20<46:22,  2.66it/s]
 31%|███       | 3271/10682 [36:20<46:24,  2.66it/s]
 31%|███       | 3272/10682 [36:20<46:27,  2.66it/s]
 31%|███       | 3273/10682 [36:21<46:24,  2.66it/s]
 31%|███       | 3274/10682 [36:21<46:19,  2.67it/s]
 31%|███       | 3275/10682 [36:22<46:26,  2.66it/s]
                                                    {'loss': 4.4458, 'grad_norm': 0.45058199763298035, 'learning_rate': 0.0008755942581046657, 'epoch': 0.31}
-

 31%|███       | 3275/10682 [36:22<46:26,  2.66it/s]
 31%|███       | 3276/10682 [36:22<46:25,  2.66it/s]
 31%|███       | 3277/10682 [36:22<46:23,  2.66it/s]
 31%|███       | 3278/10682 [36:23<46:20,  2.66it/s]
 31%|███       | 3279/10682 [36:23<46:22,  2.66it/s]
 31%|███       | 3280/10682 [36:23<46:21,  2.66it/s]
 31%|███       | 3281/10682 [36:24<46:21,  2.66it/s]
 31%|███       | 3282/10682 [36:24<46:19,  2.66it/s]
 31%|███       | 3283/10682 [36:25<46:17,  2.66it/s]
 31%|███       | 3284/10682 [36:25<46:18,  2.66it/s]
 31%|███       | 3285/10682 [36:25<46:13,  2.67it/s]
 31%|███       | 3286/10682 [36:26<46:20,  2.66it/s]
 31%|███       | 3287/10682 [36:26<46:21,  2.66it/s]
 31%|███       | 3288/10682 [36:26<46:28,  2.65it/s]
 31%|███       | 3289/10682 [36:27<46:22,  2.66it/s]
 31%|███       | 3290/10682 [36:27<46:17,  2.66it/s]
 31%|███       | 3291/10682 [36:28<46:13,  2.66it/s]
 31%|███       | 3292/10682 [36:28<46:11,  2.67it/s]
 31%|███       | 3293/10682 [36:28<46:15,  2.66it/s]
 31%|███       | 3294/10682 [36:29<46:14,  2.66it/s]
 31%|███       | 3295/10682 [36:29<46:14,  2.66it/s]
 31%|███       | 3296/10682 [36:29<46:11,  2.66it/s]
 31%|███       | 3297/10682 [36:30<46:09,  2.67it/s]
 31%|███       | 3298/10682 [36:30<46:07,  2.67it/s]
 31%|███       | 3299/10682 [36:31<46:09,  2.67it/s]
 31%|███       | 3300/10682 [36:31<46:09,  2.67it/s]{'loss': 4.4476, 'grad_norm': 0.43921715021133423, 'learning_rate': 0.0008728852388759212, 'epoch': 0.31}
                                                    
-
 31%|███       | 3300/10682 [36:31<46:09,  2.67it/s]
 31%|███       | 3301/10682 [36:31<46:11,  2.66it/s]
 31%|███       | 3302/10682 [36:32<46:16,  2.66it/s]
 31%|███       | 3303/10682 [36:32<46:13,  2.66it/s]
 31%|███       | 3304/10682 [36:32<46:12,  2.66it/s]
 31%|███       | 3305/10682 [36:33<46:08,  2.66it/s]
 31%|███       | 3306/10682 [36:33<46:07,  2.66it/s]
 31%|███       | 3307/10682 [36:34<46:07,  2.66it/s]
 31%|███       | 3308/10682 [36:34<46:06,  2.67it/s]
 31%|███       | 3309/10682 [36:34<46:05,  2.67it/s]
 31%|███       | 3310/10682 [36:35<46:01,  2.67it/s]
 31%|███       | 3311/10682 [36:35<46:00,  2.67it/s]
 31%|███       | 3312/10682 [36:35<45:59,  2.67it/s]
 31%|███       | 3313/10682 [36:36<46:05,  2.66it/s]
 31%|███       | 3314/10682 [36:36<46:04,  2.67it/s]
 31%|███       | 3315/10682 [36:37<46:07,  2.66it/s]
 31%|███       | 3316/10682 [36:37<46:04,  2.66it/s]
 31%|███       | 3317/10682 [36:37<46:06,  2.66it/s]
 31%|███       | 3318/10682 [36:38<46:02,  2.67it/s]
 31%|███       | 3319/10682 [36:38<46:00,  2.67it/s]
 31%|███       | 3320/10682 [36:38<45:59,  2.67it/s]
 31%|███       | 3321/10682 [36:39<45:54,  2.67it/s]
 31%|███       | 3322/10682 [36:39<45:55,  2.67it/s]
 31%|███       | 3323/10682 [36:40<45:56,  2.67it/s]
 31%|███       | 3324/10682 [36:40<45:58,  2.67it/s]
 31%|███       | 3325/10682 [36:40<45:57,  2.67it/s]{'loss': 4.4355, 'grad_norm': 0.4638637900352478, 'learning_rate': 0.0008701513290879668, 'epoch': 0.31}
                                                    
-
 31%|███       | 3325/10682 [36:40<45:57,  2.67it/s]
 31%|███       | 3326/10682 [36:41<46:03,  2.66it/s]
 31%|███       | 3327/10682 [36:41<46:05,  2.66it/s]
 31%|███       | 3328/10682 [36:41<46:02,  2.66it/s]
 31%|███       | 3329/10682 [36:42<46:05,  2.66it/s]
 31%|███       | 3330/10682 [36:42<46:03,  2.66it/s]
 31%|███       | 3331/10682 [36:43<46:03,  2.66it/s]
 31%|███       | 3332/10682 [36:43<45:57,  2.67it/s]
 31%|███       | 3333/10682 [36:43<45:55,  2.67it/s]
 31%|███       | 3334/10682 [36:44<45:53,  2.67it/s]
 31%|███       | 3335/10682 [36:44<45:53,  2.67it/s]
 31%|███       | 3336/10682 [36:44<45:51,  2.67it/s]
 31%|███       | 3337/10682 [36:45<45:50,  2.67it/s]
 31%|███       | 3338/10682 [36:45<45:52,  2.67it/s]
 31%|███▏      | 3339/10682 [36:46<45:47,  2.67it/s]
 31%|███▏      | 3340/10682 [36:46<45:55,  2.66it/s]
 31%|███▏      | 3341/10682 [36:46<45:52,  2.67it/s]
 31%|███▏      | 3342/10682 [36:47<45:56,  2.66it/s]
 31%|███▏      | 3343/10682 [36:47<45:54,  2.66it/s]
 31%|███▏      | 3344/10682 [36:47<45:55,  2.66it/s]
 31%|███▏      | 3345/10682 [36:48<45:52,  2.67it/s]
 31%|███▏      | 3346/10682 [36:48<45:50,  2.67it/s]
 31%|███▏      | 3347/10682 [36:49<45:55,  2.66it/s]
 31%|███▏      | 3348/10682 [36:49<45:53,  2.66it/s]
 31%|███▏      | 3349/10682 [36:49<45:52,  2.66it/s]
 31%|███▏      | 3350/10682 [36:50<45:48,  2.67it/s]
                                                    
{'loss': 4.424, 'grad_norm': 0.4883844256401062, 'learning_rate': 0.0008673927112327234, 'epoch': 0.31}
-
 31%|███▏      | 3350/10682 [36:50<45:48,  2.67it/s]
 31%|███▏      | 3351/10682 [36:50<45:52,  2.66it/s]
 31%|███▏      | 3352/10682 [36:50<45:50,  2.66it/s]
 31%|███▏      | 3353/10682 [36:51<45:56,  2.66it/s]
 31%|███▏      | 3354/10682 [36:51<45:50,  2.66it/s]
 31%|███▏      | 3355/10682 [36:52<45:47,  2.67it/s]
 31%|███▏      | 3356/10682 [36:52<45:53,  2.66it/s]
 31%|███▏      | 3357/10682 [36:52<45:52,  2.66it/s]
 31%|███▏      | 3358/10682 [36:53<45:56,  2.66it/s]
 31%|███▏      | 3359/10682 [36:53<45:55,  2.66it/s]
 31%|███▏      | 3360/10682 [36:53<45:53,  2.66it/s]
 31%|███▏      | 3361/10682 [36:54<45:49,  2.66it/s]
 31%|███▏      | 3362/10682 [36:54<45:47,  2.66it/s]
 31%|███▏      | 3363/10682 [36:55<45:44,  2.67it/s]
 31%|███▏      | 3364/10682 [36:55<45:42,  2.67it/s]
 32%|███▏      | 3365/10682 [36:55<45:41,  2.67it/s]
 32%|███▏      | 3366/10682 [36:56<45:39,  2.67it/s]
 32%|███▏      | 3367/10682 [36:56<45:51,  2.66it/s]
 32%|███▏      | 3368/10682 [36:56<45:50,  2.66it/s]
 32%|███▏      | 3369/10682 [36:57<45:51,  2.66it/s]
 32%|███▏      | 3370/10682 [36:57<45:49,  2.66it/s]
 32%|███▏      | 3371/10682 [36:58<45:46,  2.66it/s]
 32%|███▏      | 3372/10682 [36:58<45:44,  2.66it/s]
 32%|███▏      | 3373/10682 [36:58<45:42,  2.66it/s]
 32%|███▏      | 3374/10682 [36:59<45:42,  2.66it/s]
 32%|███▏      | 3375/10682 [36:59<45:39,  2.67it/s]
                                                    
{'loss': 4.4352, 'grad_norm': 0.4242475926876068, 'learning_rate': 0.000864609569451406, 'epoch': 0.32}
-
 32%|███▏      | 3375/10682 [36:59<45:39,  2.67it/s]
 32%|███▏      | 3376/10682 [36:59<45:42,  2.66it/s]
 32%|███▏      | 3377/10682 [37:00<45:37,  2.67it/s]
 32%|███▏      | 3378/10682 [37:00<45:38,  2.67it/s]
 32%|███▏      | 3379/10682 [37:01<45:36,  2.67it/s]
 32%|███▏      | 3380/10682 [37:01<45:38,  2.67it/s]
 32%|███▏      | 3381/10682 [37:01<45:40,  2.66it/s]
 32%|███▏      | 3382/10682 [37:02<45:35,  2.67it/s]
 32%|███▏      | 3383/10682 [37:02<45:44,  2.66it/s]
 32%|███▏      | 3384/10682 [37:02<45:40,  2.66it/s]
 32%|███▏      | 3385/10682 [37:03<45:39,  2.66it/s]
 32%|███▏      | 3386/10682 [37:03<45:38,  2.66it/s]
 32%|███▏      | 3387/10682 [37:04<45:41,  2.66it/s]
 32%|███▏      | 3388/10682 [37:04<45:41,  2.66it/s]
 32%|███▏      | 3389/10682 [37:04<45:39,  2.66it/s]
 32%|███▏      | 3390/10682 [37:05<45:36,  2.66it/s]
 32%|███▏      | 3391/10682 [37:05<45:33,  2.67it/s]
 32%|███▏      | 3392/10682 [37:05<45:34,  2.67it/s]
 32%|███▏      | 3393/10682 [37:06<45:32,  2.67it/s]
 32%|███▏      | 3394/10682 [37:06<45:38,  2.66it/s]
 32%|███▏      | 3395/10682 [37:07<45:35,  2.66it/s]
 32%|███▏      | 3396/10682 [37:07<45:40,  2.66it/s]
 32%|███▏      | 3397/10682 [37:07<45:37,  2.66it/s]
 32%|███▏      | 3398/10682 [37:08<45:34,  2.66it/s]
 32%|███▏      | 3399/10682 [37:08<45:37,  2.66it/s]
 32%|███▏      | 3400/10682 [37:08<45:34,  2.66it/s]{'loss': 4.4347, 'grad_norm': 0.49447283148765564, 'learning_rate': 0.0008618020895222331, 'epoch': 0.32}
-
                                                    

 32%|███▏      | 3400/10682 [37:08<45:34,  2.66it/s]
 32%|███▏      | 3401/10682 [37:09<45:36,  2.66it/s]
 32%|███▏      | 3402/10682 [37:09<45:34,  2.66it/s]
 32%|███▏      | 3403/10682 [37:10<45:33,  2.66it/s]
 32%|███▏      | 3404/10682 [37:10<45:28,  2.67it/s]
 32%|███▏      | 3405/10682 [37:10<45:31,  2.66it/s]
 32%|███▏      | 3406/10682 [37:11<45:28,  2.67it/s]
 32%|███▏      | 3407/10682 [37:11<45:30,  2.66it/s]
 32%|███▏      | 3408/10682 [37:11<45:31,  2.66it/s]
 32%|███▏      | 3409/10682 [37:12<45:30,  2.66it/s]
 32%|███▏      | 3410/10682 [37:12<45:28,  2.66it/s]
 32%|███▏      | 3411/10682 [37:13<45:30,  2.66it/s]
 32%|███▏      | 3412/10682 [37:13<45:28,  2.66it/s]
 32%|███▏      | 3413/10682 [37:13<45:29,  2.66it/s]
 32%|███▏      | 3414/10682 [37:14<45:30,  2.66it/s]
 32%|███▏      | 3415/10682 [37:14<45:27,  2.66it/s]
 32%|███▏      | 3416/10682 [37:14<45:24,  2.67it/s]
 32%|███▏      | 3417/10682 [37:15<45:22,  2.67it/s]
 32%|███▏      | 3418/10682 [37:15<45:21,  2.67it/s]
 32%|███▏      | 3419/10682 [37:16<45:24,  2.67it/s]
 32%|███▏      | 3420/10682 [37:16<45:22,  2.67it/s]
 32%|███▏      | 3421/10682 [37:16<45:30,  2.66it/s]
 32%|███▏      | 3422/10682 [37:17<45:26,  2.66it/s]
 32%|███▏      | 3423/10682 [37:17<45:29,  2.66it/s]
 32%|███▏      | 3424/10682 [37:17<45:27,  2.66it/s]
 32%|███▏      | 3425/10682 [37:18<45:28,  2.66it/s]
                                                    
{'loss': 4.4311, 'grad_norm': 0.44754794239997864, 'learning_rate': 0.0008589704588480251, 'epoch': 0.32}
-
 32%|███▏      | 3425/10682 [37:18<45:28,  2.66it/s]
 32%|███▏      | 3426/10682 [37:18<45:27,  2.66it/s]
 32%|███▏      | 3427/10682 [37:19<45:24,  2.66it/s]
 32%|███▏      | 3428/10682 [37:19<45:20,  2.67it/s]
 32%|███▏      | 3429/10682 [37:19<45:18,  2.67it/s]
 32%|███▏      | 3430/10682 [37:20<45:19,  2.67it/s]
 32%|███▏      | 3431/10682 [37:20<45:18,  2.67it/s]
 32%|███▏      | 3432/10682 [37:20<45:20,  2.67it/s]
 32%|███▏      | 3433/10682 [37:21<45:17,  2.67it/s]
 32%|███▏      | 3434/10682 [37:21<45:19,  2.67it/s]
 32%|███▏      | 3435/10682 [37:22<45:16,  2.67it/s]
 32%|███▏      | 3436/10682 [37:22<45:18,  2.67it/s]
 32%|███▏      | 3437/10682 [37:22<45:22,  2.66it/s]
 32%|███▏      | 3438/10682 [37:23<45:20,  2.66it/s]
 32%|███▏      | 3439/10682 [37:23<45:17,  2.66it/s]
 32%|███▏      | 3440/10682 [37:23<45:16,  2.67it/s]
 32%|███▏      | 3441/10682 [37:24<45:16,  2.67it/s]
 32%|███▏      | 3442/10682 [37:24<45:14,  2.67it/s]
 32%|███▏      | 3443/10682 [37:25<45:13,  2.67it/s]
 32%|███▏      | 3444/10682 [37:25<45:11,  2.67it/s]
 32%|███▏      | 3445/10682 [37:25<45:11,  2.67it/s]
 32%|███▏      | 3446/10682 [37:26<45:10,  2.67it/s]
 32%|███▏      | 3447/10682 [37:26<45:11,  2.67it/s]
 32%|███▏      | 3448/10682 [37:26<45:15,  2.66it/s]
 32%|███▏      | 3449/10682 [37:27<45:10,  2.67it/s]
 32%|███▏      | 3450/10682 [37:27<45:15,  2.66it/s]
                                                    {'loss': 4.4301, 'grad_norm': 0.4940510392189026, 'learning_rate': 0.0008561148664436948, 'epoch': 0.32}
-

 32%|███▏      | 3450/10682 [37:27<45:15,  2.66it/s]
 32%|███▏      | 3451/10682 [37:28<45:16,  2.66it/s]
 32%|███▏      | 3452/10682 [37:28<45:15,  2.66it/s]
 32%|███▏      | 3453/10682 [37:28<45:12,  2.66it/s]
 32%|███▏      | 3454/10682 [37:29<45:14,  2.66it/s]
 32%|███▏      | 3455/10682 [37:29<45:12,  2.66it/s]
 32%|███▏      | 3456/10682 [37:29<45:08,  2.67it/s]
 32%|███▏      | 3457/10682 [37:30<45:07,  2.67it/s]
 32%|███▏      | 3458/10682 [37:30<45:05,  2.67it/s]
 32%|███▏      | 3459/10682 [37:31<45:06,  2.67it/s]
 32%|███▏      | 3460/10682 [37:31<45:06,  2.67it/s]
 32%|███▏      | 3461/10682 [37:31<45:10,  2.66it/s]
 32%|███▏      | 3462/10682 [37:32<45:11,  2.66it/s]
 32%|███▏      | 3463/10682 [37:32<45:08,  2.67it/s]
 32%|███▏      | 3464/10682 [37:32<45:10,  2.66it/s]
 32%|███▏      | 3465/10682 [37:33<45:09,  2.66it/s]
 32%|███▏      | 3466/10682 [37:33<45:07,  2.66it/s]
 32%|███▏      | 3467/10682 [37:34<45:06,  2.67it/s]
 32%|███▏      | 3468/10682 [37:34<45:07,  2.66it/s]
 32%|███▏      | 3469/10682 [37:34<45:10,  2.66it/s]
 32%|███▏      | 3470/10682 [37:35<45:05,  2.67it/s]
 32%|███▏      | 3471/10682 [37:35<45:06,  2.66it/s]
 33%|███▎      | 3472/10682 [37:35<45:06,  2.66it/s]
 33%|███▎      | 3473/10682 [37:36<45:06,  2.66it/s]
 33%|███▎      | 3474/10682 [37:36<45:03,  2.67it/s]
 33%|███▎      | 3475/10682 [37:37<45:06,  2.66it/s]{'loss': 4.4119, 'grad_norm': 0.47009092569351196, 'learning_rate': 0.0008532355029236307, 'epoch': 0.33}
                                                    
-
 33%|███▎      | 3475/10682 [37:37<45:06,  2.66it/s]
 33%|███▎      | 3476/10682 [37:37<45:12,  2.66it/s]
 33%|███▎      | 3477/10682 [37:37<45:12,  2.66it/s]
 33%|███▎      | 3478/10682 [37:38<45:09,  2.66it/s]
 33%|███▎      | 3479/10682 [37:38<45:04,  2.66it/s]
 33%|███▎      | 3480/10682 [37:38<45:04,  2.66it/s]
 33%|███▎      | 3481/10682 [37:39<45:05,  2.66it/s]
 33%|███▎      | 3482/10682 [37:39<45:03,  2.66it/s]
 33%|███▎      | 3483/10682 [37:40<45:01,  2.66it/s]
 33%|███▎      | 3484/10682 [37:40<44:59,  2.67it/s]
 33%|███▎      | 3485/10682 [37:40<44:58,  2.67it/s]
 33%|███▎      | 3486/10682 [37:41<44:56,  2.67it/s]
 33%|███▎      | 3487/10682 [37:41<44:55,  2.67it/s]
 33%|███▎      | 3488/10682 [37:41<44:54,  2.67it/s]
 33%|███▎      | 3489/10682 [37:42<44:58,  2.67it/s]
 33%|███▎      | 3490/10682 [37:42<44:55,  2.67it/s]
 33%|███▎      | 3491/10682 [37:43<44:58,  2.67it/s]
 33%|███▎      | 3492/10682 [37:43<45:00,  2.66it/s]
 33%|███▎      | 3493/10682 [37:43<44:59,  2.66it/s]
 33%|███▎      | 3494/10682 [37:44<44:56,  2.67it/s]
 33%|███▎      | 3495/10682 [37:44<44:55,  2.67it/s]
 33%|███▎      | 3496/10682 [37:44<44:54,  2.67it/s]
 33%|███▎      | 3497/10682 [37:45<44:51,  2.67it/s]
 33%|███▎      | 3498/10682 [37:45<44:52,  2.67it/s]
 33%|███▎      | 3499/10682 [37:46<44:50,  2.67it/s]
 33%|███▎      | 3500/10682 [37:46<44:49,  2.67it/s]{'loss': 4.4146, 'grad_norm': 0.4550090432167053, 'learning_rate': 0.0008503325604889736, 'epoch': 0.33}
-
                                                    

 33%|███▎      | 3500/10682 [37:46<44:49,  2.67it/s]
 33%|███▎      | 3501/10682 [37:46<44:56,  2.66it/s]
 33%|███▎      | 3502/10682 [37:47<45:00,  2.66it/s]
 33%|███▎      | 3503/10682 [37:47<44:59,  2.66it/s]
 33%|███▎      | 3504/10682 [37:48<44:55,  2.66it/s]
 33%|███▎      | 3505/10682 [37:48<44:58,  2.66it/s]
 33%|███▎      | 3506/10682 [37:48<44:58,  2.66it/s]
 33%|███▎      | 3507/10682 [37:49<45:00,  2.66it/s]
 33%|███▎      | 3508/10682 [37:49<44:55,  2.66it/s]
 33%|███▎      | 3509/10682 [37:49<44:51,  2.67it/s]
 33%|███▎      | 3510/10682 [37:50<44:49,  2.67it/s]
 33%|███▎      | 3511/10682 [37:50<44:50,  2.67it/s]
 33%|███▎      | 3512/10682 [37:51<44:52,  2.66it/s]
 33%|███▎      | 3513/10682 [37:51<44:48,  2.67it/s]
 33%|███▎      | 3514/10682 [37:51<44:47,  2.67it/s]
 33%|███▎      | 3515/10682 [37:52<44:46,  2.67it/s]
 33%|███▎      | 3516/10682 [37:52<44:48,  2.67it/s]
 33%|███▎      | 3517/10682 [37:52<44:46,  2.67it/s]
 33%|███▎      | 3518/10682 [37:53<44:49,  2.66it/s]
 33%|███▎      | 3519/10682 [37:53<44:50,  2.66it/s]
 33%|███▎      | 3520/10682 [37:54<44:47,  2.66it/s]
 33%|███▎      | 3521/10682 [37:54<44:45,  2.67it/s]
 33%|███▎      | 3522/10682 [37:54<44:44,  2.67it/s]
 33%|███▎      | 3523/10682 [37:55<44:50,  2.66it/s]
 33%|███▎      | 3524/10682 [37:55<44:46,  2.66it/s]
 33%|███▎      | 3525/10682 [37:55<44:45,  2.66it/s]{'loss': 4.4151, 'grad_norm': 0.4471551775932312, 'learning_rate': 0.0008474062329147856, 'epoch': 0.33}
-
                                                    

 33%|███▎      | 3525/10682 [37:55<44:45,  2.66it/s]
 33%|███▎      | 3526/10682 [37:56<44:49,  2.66it/s]
 33%|███▎      | 3527/10682 [37:56<44:46,  2.66it/s]
 33%|███▎      | 3528/10682 [37:57<44:45,  2.66it/s]
 33%|███▎      | 3529/10682 [37:57<44:44,  2.67it/s]
 33%|███▎      | 3530/10682 [37:57<44:49,  2.66it/s]
 33%|███▎      | 3531/10682 [37:58<44:43,  2.67it/s]
 33%|███▎      | 3532/10682 [37:58<44:49,  2.66it/s]
 33%|███▎      | 3533/10682 [37:58<44:45,  2.66it/s]
 33%|███▎      | 3534/10682 [37:59<44:44,  2.66it/s]
 33%|███▎      | 3535/10682 [37:59<44:44,  2.66it/s]
 33%|███▎      | 3536/10682 [38:00<44:42,  2.66it/s]
 33%|███▎      | 3537/10682 [38:00<44:40,  2.67it/s]
 33%|███▎      | 3538/10682 [38:00<44:41,  2.66it/s]
 33%|███▎      | 3539/10682 [38:01<44:41,  2.66it/s]
 33%|███▎      | 3540/10682 [38:01<44:39,  2.67it/s]
 33%|███▎      | 3541/10682 [38:01<44:38,  2.67it/s]
 33%|███▎      | 3542/10682 [38:02<44:40,  2.66it/s]
 33%|███▎      | 3543/10682 [38:02<44:40,  2.66it/s]
 33%|███▎      | 3544/10682 [38:03<44:39,  2.66it/s]
 33%|███▎      | 3545/10682 [38:03<44:35,  2.67it/s]
 33%|███▎      | 3546/10682 [38:03<44:44,  2.66it/s]
 33%|███▎      | 3547/10682 [38:04<44:41,  2.66it/s]
 33%|███▎      | 3548/10682 [38:04<44:41,  2.66it/s]
 33%|███▎      | 3549/10682 [38:04<44:38,  2.66it/s]
 33%|███▎      | 3550/10682 [38:05<44:40,  2.66it/s]
                                                    
{'loss': 4.4115, 'grad_norm': 0.4591195285320282, 'learning_rate': 0.0008444567155371176, 'epoch': 0.33}
-
 33%|███▎      | 3550/10682 [38:05<44:40,  2.66it/s]
 33%|███▎      | 3551/10682 [38:05<44:44,  2.66it/s]
 33%|███▎      | 3552/10682 [38:06<44:37,  2.66it/s]
 33%|███▎      | 3553/10682 [38:06<44:34,  2.67it/s]
 33%|███▎      | 3554/10682 [38:06<44:30,  2.67it/s]
 33%|███▎      | 3555/10682 [38:07<44:31,  2.67it/s]
 33%|███▎      | 3556/10682 [38:07<44:28,  2.67it/s]
 33%|███▎      | 3557/10682 [38:07<44:27,  2.67it/s]
 33%|███▎      | 3558/10682 [38:08<44:25,  2.67it/s]
 33%|███▎      | 3559/10682 [38:08<44:30,  2.67it/s]
 33%|███▎      | 3560/10682 [38:09<44:28,  2.67it/s]
 33%|███▎      | 3561/10682 [38:09<44:29,  2.67it/s]
 33%|███▎      | 3562/10682 [38:09<44:27,  2.67it/s]
 33%|███▎      | 3563/10682 [38:10<44:31,  2.67it/s]
 33%|███▎      | 3564/10682 [38:10<44:27,  2.67it/s]
 33%|███▎      | 3565/10682 [38:10<44:27,  2.67it/s]
 33%|███▎      | 3566/10682 [38:11<44:26,  2.67it/s]
 33%|███▎      | 3567/10682 [38:11<44:23,  2.67it/s]
 33%|███▎      | 3568/10682 [38:12<44:22,  2.67it/s]
 33%|███▎      | 3569/10682 [38:12<44:22,  2.67it/s]
 33%|███▎      | 3570/10682 [38:12<44:27,  2.67it/s]
 33%|███▎      | 3571/10682 [38:13<44:25,  2.67it/s]
 33%|███▎      | 3572/10682 [38:13<44:26,  2.67it/s]
 33%|███▎      | 3573/10682 [38:13<44:26,  2.67it/s]
 33%|███▎      | 3574/10682 [38:14<44:22,  2.67it/s]
 33%|███▎      | 3575/10682 [38:14<44:22,  2.67it/s]
                                                    
{'loss': 4.4169, 'grad_norm': 0.4312704801559448, 'learning_rate': 0.000841484205239968, 'epoch': 0.33}
-
 33%|███▎      | 3575/10682 [38:14<44:22,  2.67it/s]
 33%|███▎      | 3576/10682 [38:15<44:25,  2.67it/s]
 33%|███▎      | 3577/10682 [38:15<44:23,  2.67it/s]
 33%|███▎      | 3578/10682 [38:15<44:23,  2.67it/s]
 34%|███▎      | 3579/10682 [38:16<44:24,  2.67it/s]
 34%|███▎      | 3580/10682 [38:16<44:22,  2.67it/s]
 34%|███▎      | 3581/10682 [38:16<44:21,  2.67it/s]
 34%|███▎      | 3582/10682 [38:17<44:20,  2.67it/s]
 34%|███▎      | 3583/10682 [38:17<44:19,  2.67it/s]
 34%|███▎      | 3584/10682 [38:18<44:24,  2.66it/s]
 34%|███▎      | 3585/10682 [38:18<44:20,  2.67it/s]
 34%|███▎      | 3586/10682 [38:18<44:26,  2.66it/s]
 34%|███▎      | 3587/10682 [38:19<44:24,  2.66it/s]
 34%|███▎      | 3588/10682 [38:19<44:22,  2.66it/s]
 34%|███▎      | 3589/10682 [38:19<44:19,  2.67it/s]
 34%|███▎      | 3590/10682 [38:20<44:16,  2.67it/s]
 34%|███▎      | 3591/10682 [38:20<44:15,  2.67it/s]
 34%|███▎      | 3592/10682 [38:21<44:14,  2.67it/s]
 34%|███▎      | 3593/10682 [38:21<44:14,  2.67it/s]
 34%|███▎      | 3594/10682 [38:21<44:11,  2.67it/s]
 34%|███▎      | 3595/10682 [38:22<44:13,  2.67it/s]
 34%|███▎      | 3596/10682 [38:22<44:12,  2.67it/s]
 34%|███▎      | 3597/10682 [38:22<44:14,  2.67it/s]
 34%|███▎      | 3598/10682 [38:23<44:12,  2.67it/s]
 34%|███▎      | 3599/10682 [38:23<44:14,  2.67it/s]
 34%|███▎      | 3600/10682 [38:24<44:14,  2.67it/s]{'loss': 4.4119, 'grad_norm': 0.4856119751930237, 'learning_rate': 0.0008384889004421422, 'epoch': 0.34}
                                                    
-
 34%|███▎      | 3600/10682 [38:24<44:14,  2.67it/s]
 34%|███▎      | 3601/10682 [38:24<44:17,  2.66it/s]
 34%|███▎      | 3602/10682 [38:24<44:19,  2.66it/s]
 34%|███▎      | 3603/10682 [38:25<44:19,  2.66it/s]
 34%|███▎      | 3604/10682 [38:25<44:17,  2.66it/s]
 34%|███▎      | 3605/10682 [38:25<44:16,  2.66it/s]
 34%|███▍      | 3606/10682 [38:26<44:14,  2.67it/s]
 34%|███▍      | 3607/10682 [38:26<44:16,  2.66it/s]
 34%|███▍      | 3608/10682 [38:27<44:15,  2.66it/s]
 34%|███▍      | 3609/10682 [38:27<44:12,  2.67it/s]
 34%|███▍      | 3610/10682 [38:27<44:12,  2.67it/s]
 34%|███▍      | 3611/10682 [38:28<44:12,  2.67it/s]
 34%|███▍      | 3612/10682 [38:28<44:11,  2.67it/s]
 34%|███▍      | 3613/10682 [38:28<44:11,  2.67it/s]
 34%|███▍      | 3614/10682 [38:29<44:08,  2.67it/s]
 34%|███▍      | 3615/10682 [38:29<44:08,  2.67it/s]
 34%|███▍      | 3616/10682 [38:30<44:09,  2.67it/s]
 34%|███▍      | 3617/10682 [38:30<44:10,  2.67it/s]
 34%|███▍      | 3618/10682 [38:30<44:09,  2.67it/s]
 34%|███▍      | 3619/10682 [38:31<44:06,  2.67it/s]
 34%|███▍      | 3620/10682 [38:31<44:05,  2.67it/s]
 34%|███▍      | 3621/10682 [38:31<44:03,  2.67it/s]
 34%|███▍      | 3622/10682 [38:32<44:06,  2.67it/s]
 34%|███▍      | 3623/10682 [38:32<44:07,  2.67it/s]
 34%|███▍      | 3624/10682 [38:33<44:14,  2.66it/s]
 34%|███▍      | 3625/10682 [38:33<44:14,  2.66it/s]{'loss': 4.4105, 'grad_norm': 0.473195880651474, 'learning_rate': 0.0008354710010840066, 'epoch': 0.34}
                                                    
-
 34%|███▍      | 3625/10682 [38:33<44:14,  2.66it/s]
 34%|███▍      | 3626/10682 [38:33<44:26,  2.65it/s]
 34%|███▍      | 3627/10682 [38:34<44:20,  2.65it/s]
 34%|███▍      | 3628/10682 [38:34<44:14,  2.66it/s]
 34%|███▍      | 3629/10682 [38:34<44:09,  2.66it/s]
 34%|███▍      | 3630/10682 [38:35<44:05,  2.67it/s]
 34%|███▍      | 3631/10682 [38:35<44:06,  2.66it/s]
 34%|███▍      | 3632/10682 [38:36<44:08,  2.66it/s]
 34%|███▍      | 3633/10682 [38:36<44:07,  2.66it/s]
 34%|███▍      | 3634/10682 [38:36<44:05,  2.66it/s]
 34%|███▍      | 3635/10682 [38:37<44:03,  2.67it/s]
 34%|███▍      | 3636/10682 [38:37<44:06,  2.66it/s]
 34%|███▍      | 3637/10682 [38:37<44:06,  2.66it/s]
 34%|███▍      | 3638/10682 [38:38<44:08,  2.66it/s]
 34%|███▍      | 3639/10682 [38:38<44:03,  2.66it/s]
 34%|███▍      | 3640/10682 [38:39<44:05,  2.66it/s]
 34%|███▍      | 3641/10682 [38:39<44:04,  2.66it/s]
 34%|███▍      | 3642/10682 [38:39<44:03,  2.66it/s]
 34%|███▍      | 3643/10682 [38:40<44:00,  2.67it/s]
 34%|███▍      | 3644/10682 [38:40<44:01,  2.66it/s]
 34%|███▍      | 3645/10682 [38:40<44:03,  2.66it/s]
 34%|███▍      | 3646/10682 [38:41<44:01,  2.66it/s]
 34%|███▍      | 3647/10682 [38:41<43:58,  2.67it/s]
 34%|███▍      | 3648/10682 [38:42<43:55,  2.67it/s]
 34%|███▍      | 3649/10682 [38:42<43:54,  2.67it/s]
 34%|███▍      | 3650/10682 [38:42<43:56,  2.67it/s]
                                                    
{'loss': 4.4037, 'grad_norm': 0.5131106376647949, 'learning_rate': 0.0008324307086141438, 'epoch': 0.34}
-
 34%|███▍      | 3650/10682 [38:42<43:56,  2.67it/s]
 34%|███▍      | 3651/10682 [38:43<44:03,  2.66it/s]
 34%|███▍      | 3652/10682 [38:43<43:59,  2.66it/s]
 34%|███▍      | 3653/10682 [38:43<43:59,  2.66it/s]
 34%|███▍      | 3654/10682 [38:44<44:01,  2.66it/s]
 34%|███▍      | 3655/10682 [38:44<43:59,  2.66it/s]
 34%|███▍      | 3656/10682 [38:45<43:57,  2.66it/s]
 34%|███▍      | 3657/10682 [38:45<43:55,  2.67it/s]
 34%|███▍      | 3658/10682 [38:45<43:53,  2.67it/s]
 34%|███▍      | 3659/10682 [38:46<43:50,  2.67it/s]
 34%|███▍      | 3660/10682 [38:46<43:51,  2.67it/s]
 34%|███▍      | 3661/10682 [38:46<43:50,  2.67it/s]
 34%|███▍      | 3662/10682 [38:47<43:50,  2.67it/s]
 34%|███▍      | 3663/10682 [38:47<43:49,  2.67it/s]
 34%|███▍      | 3664/10682 [38:48<43:51,  2.67it/s]
 34%|███▍      | 3665/10682 [38:48<43:51,  2.67it/s]
 34%|███▍      | 3666/10682 [38:48<43:48,  2.67it/s]
 34%|███▍      | 3667/10682 [38:49<43:57,  2.66it/s]
 34%|███▍      | 3668/10682 [38:49<43:54,  2.66it/s]
 34%|███▍      | 3669/10682 [38:49<43:51,  2.66it/s]
 34%|███▍      | 3670/10682 [38:50<43:47,  2.67it/s]
 34%|███▍      | 3671/10682 [38:50<43:47,  2.67it/s]
 34%|███▍      | 3672/10682 [38:51<43:48,  2.67it/s]
 34%|███▍      | 3673/10682 [38:51<43:45,  2.67it/s]
 34%|███▍      | 3674/10682 [38:51<43:44,  2.67it/s]
 34%|███▍      | 3675/10682 [38:52<43:41,  2.67it/s]{'loss': 4.4104, 'grad_norm': 0.4209146201610565, 'learning_rate': 0.000829368225975904, 'epoch': 0.34}
                                                    
-
 34%|███▍      | 3675/10682 [38:52<43:41,  2.67it/s]
 34%|███▍      | 3676/10682 [38:52<43:45,  2.67it/s]
 34%|███▍      | 3677/10682 [38:52<43:43,  2.67it/s]
 34%|███▍      | 3678/10682 [38:53<43:44,  2.67it/s]
 34%|███▍      | 3679/10682 [38:53<43:42,  2.67it/s]
 34%|███▍      | 3680/10682 [38:54<43:48,  2.66it/s]
 34%|███▍      | 3681/10682 [38:54<43:46,  2.67it/s]
 34%|███▍      | 3682/10682 [38:54<43:45,  2.67it/s]
 34%|███▍      | 3683/10682 [38:55<43:47,  2.66it/s]
 34%|███▍      | 3684/10682 [38:55<43:45,  2.67it/s]
 34%|███▍      | 3685/10682 [38:55<43:44,  2.67it/s]
 35%|███▍      | 3686/10682 [38:56<43:40,  2.67it/s]
 35%|███▍      | 3687/10682 [38:56<43:38,  2.67it/s]
 35%|███▍      | 3688/10682 [38:57<43:39,  2.67it/s]
 35%|███▍      | 3689/10682 [38:57<43:38,  2.67it/s]
 35%|███▍      | 3690/10682 [38:57<43:37,  2.67it/s]
 35%|███▍      | 3691/10682 [38:58<43:39,  2.67it/s]
 35%|███▍      | 3692/10682 [38:58<43:39,  2.67it/s]
 35%|███▍      | 3693/10682 [38:58<43:40,  2.67it/s]
 35%|███▍      | 3694/10682 [38:59<43:42,  2.66it/s]
 35%|███▍      | 3695/10682 [38:59<43:39,  2.67it/s]
 35%|███▍      | 3696/10682 [39:00<43:41,  2.66it/s]
 35%|███▍      | 3697/10682 [39:00<43:37,  2.67it/s]
 35%|███▍      | 3698/10682 [39:00<43:41,  2.66it/s]
 35%|███▍      | 3699/10682 [39:01<43:42,  2.66it/s]
 35%|███▍      | 3700/10682 [39:01<43:38,  2.67it/s]
                                                    
{'loss': 4.4017, 'grad_norm': 0.48307502269744873, 'learning_rate': 0.0008262837575938593, 'epoch': 0.35}
-
 35%|███▍      | 3700/10682 [39:01<43:38,  2.67it/s]
 35%|███▍      | 3701/10682 [39:01<43:41,  2.66it/s]
 35%|███▍      | 3702/10682 [39:02<43:40,  2.66it/s]
 35%|███▍      | 3703/10682 [39:02<43:38,  2.67it/s]
 35%|███▍      | 3704/10682 [39:03<43:34,  2.67it/s]
 35%|███▍      | 3705/10682 [39:03<43:37,  2.67it/s]
 35%|███▍      | 3706/10682 [39:03<43:35,  2.67it/s]
 35%|███▍      | 3707/10682 [39:04<43:35,  2.67it/s]
 35%|███▍      | 3708/10682 [39:04<43:35,  2.67it/s]
 35%|███▍      | 3709/10682 [39:04<43:35,  2.67it/s]
 35%|███▍      | 3710/10682 [39:05<43:34,  2.67it/s]
 35%|███▍      | 3711/10682 [39:05<43:31,  2.67it/s]
 35%|███▍      | 3712/10682 [39:06<43:32,  2.67it/s]
 35%|███▍      | 3713/10682 [39:06<43:30,  2.67it/s]
 35%|███▍      | 3714/10682 [39:06<43:30,  2.67it/s]
 35%|███▍      | 3715/10682 [39:07<43:33,  2.67it/s]
 35%|███▍      | 3716/10682 [39:07<43:31,  2.67it/s]
 35%|███▍      | 3717/10682 [39:07<43:33,  2.67it/s]
 35%|███▍      | 3718/10682 [39:08<43:37,  2.66it/s]
 35%|███▍      | 3719/10682 [39:08<43:34,  2.66it/s]
 35%|███▍      | 3720/10682 [39:09<43:36,  2.66it/s]
 35%|███▍      | 3721/10682 [39:09<43:37,  2.66it/s]
 35%|███▍      | 3722/10682 [39:09<43:35,  2.66it/s]
 35%|███▍      | 3723/10682 [39:10<43:33,  2.66it/s]
 35%|███▍      | 3724/10682 [39:10<43:32,  2.66it/s]
 35%|███▍      | 3725/10682 [39:10<43:32,  2.66it/s]{'loss': 4.3988, 'grad_norm': 0.43017488718032837, 'learning_rate': 0.0008231775093601584, 'epoch': 0.35}
                                                    
-
 35%|███▍      | 3725/10682 [39:10<43:32,  2.66it/s]
 35%|███▍      | 3726/10682 [39:11<43:31,  2.66it/s]
 35%|███▍      | 3727/10682 [39:11<43:28,  2.67it/s]
 35%|███▍      | 3728/10682 [39:12<43:26,  2.67it/s]
 35%|███▍      | 3729/10682 [39:12<43:25,  2.67it/s]
 35%|███▍      | 3730/10682 [39:12<43:27,  2.67it/s]
 35%|███▍      | 3731/10682 [39:13<43:26,  2.67it/s]
 35%|███▍      | 3732/10682 [39:13<43:31,  2.66it/s]
 35%|███▍      | 3733/10682 [39:13<43:30,  2.66it/s]
 35%|███▍      | 3734/10682 [39:14<43:35,  2.66it/s]
 35%|███▍      | 3735/10682 [39:14<43:32,  2.66it/s]
 35%|███▍      | 3736/10682 [39:15<43:28,  2.66it/s]
 35%|███▍      | 3737/10682 [39:15<43:28,  2.66it/s]
 35%|███▍      | 3738/10682 [39:15<43:30,  2.66it/s]
 35%|███▌      | 3739/10682 [39:16<43:29,  2.66it/s]
 35%|███▌      | 3740/10682 [39:16<43:26,  2.66it/s]
 35%|███▌      | 3741/10682 [39:16<43:28,  2.66it/s]
 35%|███▌      | 3742/10682 [39:17<43:22,  2.67it/s]
 35%|███▌      | 3743/10682 [39:17<43:25,  2.66it/s]
 35%|███▌      | 3744/10682 [39:18<43:25,  2.66it/s]
 35%|███▌      | 3745/10682 [39:18<43:30,  2.66it/s]
 35%|███▌      | 3746/10682 [39:18<43:26,  2.66it/s]
 35%|███▌      | 3747/10682 [39:19<43:26,  2.66it/s]
 35%|███▌      | 3748/10682 [39:19<43:25,  2.66it/s]
 35%|███▌      | 3749/10682 [39:19<43:25,  2.66it/s]
 35%|███▌      | 3750/10682 [39:20<43:22,  2.66it/s]
                                                    
{'loss': 4.3978, 'grad_norm': 0.41869381070137024, 'learning_rate': 0.0008200496886207818, 'epoch': 0.35}
-
 35%|███▌      | 3750/10682 [39:20<43:22,  2.66it/s]
 35%|███▌      | 3751/10682 [39:20<43:31,  2.65it/s]
 35%|███▌      | 3752/10682 [39:21<43:29,  2.66it/s]
 35%|███▌      | 3753/10682 [39:21<43:24,  2.66it/s]
 35%|███▌      | 3754/10682 [39:21<43:20,  2.66it/s]
 35%|███▌      | 3755/10682 [39:22<43:22,  2.66it/s]
 35%|███▌      | 3756/10682 [39:22<43:20,  2.66it/s]
 35%|███▌      | 3757/10682 [39:22<43:19,  2.66it/s]
 35%|███▌      | 3758/10682 [39:23<43:20,  2.66it/s]
 35%|███▌      | 3759/10682 [39:23<43:18,  2.66it/s]
 35%|███▌      | 3760/10682 [39:24<43:15,  2.67it/s]
 35%|███▌      | 3761/10682 [39:24<43:23,  2.66it/s]
 35%|███▌      | 3762/10682 [39:24<43:22,  2.66it/s]
 35%|███▌      | 3763/10682 [39:25<43:20,  2.66it/s]
 35%|███▌      | 3764/10682 [39:25<43:20,  2.66it/s]
 35%|███▌      | 3765/10682 [39:25<43:17,  2.66it/s]
 35%|███▌      | 3766/10682 [39:26<43:13,  2.67it/s]
 35%|███▌      | 3767/10682 [39:26<43:12,  2.67it/s]
 35%|███▌      | 3768/10682 [39:27<43:12,  2.67it/s]
 35%|███▌      | 3769/10682 [39:27<43:11,  2.67it/s]
 35%|███▌      | 3770/10682 [39:27<43:14,  2.66it/s]
 35%|███▌      | 3771/10682 [39:28<43:13,  2.66it/s]
 35%|███▌      | 3772/10682 [39:28<43:16,  2.66it/s]
 35%|███▌      | 3773/10682 [39:28<43:13,  2.66it/s]
 35%|███▌      | 3774/10682 [39:29<43:15,  2.66it/s]
 35%|███▌      | 3775/10682 [39:29<43:13,  2.66it/s]
                                                    
{'loss': 4.3896, 'grad_norm': 0.4877290427684784, 'learning_rate': 0.0008169005041617021, 'epoch': 0.35}
-
 35%|███▌      | 3775/10682 [39:29<43:13,  2.66it/s]
 35%|███▌      | 3776/10682 [39:30<43:13,  2.66it/s]
 35%|███▌      | 3777/10682 [39:30<43:09,  2.67it/s]
 35%|███▌      | 3778/10682 [39:30<43:08,  2.67it/s]
 35%|███▌      | 3779/10682 [39:31<43:07,  2.67it/s]
 35%|███▌      | 3780/10682 [39:31<43:03,  2.67it/s]
 35%|███▌      | 3781/10682 [39:31<43:02,  2.67it/s]
 35%|███▌      | 3782/10682 [39:32<43:06,  2.67it/s]
 35%|███▌      | 3783/10682 [39:32<43:04,  2.67it/s]
 35%|███▌      | 3784/10682 [39:33<43:05,  2.67it/s]
 35%|███▌      | 3785/10682 [39:33<43:05,  2.67it/s]
 35%|███▌      | 3786/10682 [39:33<43:03,  2.67it/s]
 35%|███▌      | 3787/10682 [39:34<43:05,  2.67it/s]
 35%|███▌      | 3788/10682 [39:34<43:06,  2.67it/s]
 35%|███▌      | 3789/10682 [39:34<43:04,  2.67it/s]
 35%|███▌      | 3790/10682 [39:35<43:02,  2.67it/s]
 35%|███▌      | 3791/10682 [39:35<43:00,  2.67it/s]
 35%|███▌      | 3792/10682 [39:36<43:04,  2.67it/s]
 36%|███▌      | 3793/10682 [39:36<43:05,  2.66it/s]
 36%|███▌      | 3794/10682 [39:36<43:01,  2.67it/s]
 36%|███▌      | 3795/10682 [39:37<43:02,  2.67it/s]
 36%|███▌      | 3796/10682 [39:37<42:59,  2.67it/s]
 36%|███▌      | 3797/10682 [39:37<43:03,  2.66it/s]
 36%|███▌      | 3798/10682 [39:38<43:03,  2.66it/s]
 36%|███▌      | 3799/10682 [39:38<43:05,  2.66it/s]
 36%|███▌      | 3800/10682 [39:39<43:00,  2.67it/s]{'loss': 4.3951, 'grad_norm': 0.43008846044540405, 'learning_rate': 0.0008137301661949477, 'epoch': 0.36}
-
                                                    

 36%|███▌      | 3800/10682 [39:39<43:00,  2.67it/s]
 36%|███▌      | 3801/10682 [39:39<43:06,  2.66it/s]
 36%|███▌      | 3802/10682 [39:39<43:01,  2.66it/s]
 36%|███▌      | 3803/10682 [39:40<42:57,  2.67it/s]
 36%|███▌      | 3804/10682 [39:40<42:56,  2.67it/s]
 36%|███▌      | 3805/10682 [39:40<42:58,  2.67it/s]
 36%|███▌      | 3806/10682 [39:41<42:55,  2.67it/s]
 36%|███▌      | 3807/10682 [39:41<42:56,  2.67it/s]
 36%|███▌      | 3808/10682 [39:42<42:53,  2.67it/s]
 36%|███▌      | 3809/10682 [39:42<42:58,  2.67it/s]
 36%|███▌      | 3810/10682 [39:42<42:56,  2.67it/s]
 36%|███▌      | 3811/10682 [39:43<42:55,  2.67it/s]
 36%|███▌      | 3812/10682 [39:43<42:58,  2.66it/s]
 36%|███▌      | 3813/10682 [39:43<42:57,  2.66it/s]
 36%|███▌      | 3814/10682 [39:44<42:56,  2.67it/s]
 36%|███▌      | 3815/10682 [39:44<42:58,  2.66it/s]
 36%|███▌      | 3816/10682 [39:45<42:56,  2.66it/s]
 36%|███▌      | 3817/10682 [39:45<42:55,  2.67it/s]
 36%|███▌      | 3818/10682 [39:45<42:53,  2.67it/s]
 36%|███▌      | 3819/10682 [39:46<42:56,  2.66it/s]
 36%|███▌      | 3820/10682 [39:46<42:51,  2.67it/s]
 36%|███▌      | 3821/10682 [39:46<42:50,  2.67it/s]
 36%|███▌      | 3822/10682 [39:47<42:49,  2.67it/s]
 36%|███▌      | 3823/10682 [39:47<42:49,  2.67it/s]
 36%|███▌      | 3824/10682 [39:48<42:45,  2.67it/s]
 36%|███▌      | 3825/10682 [39:48<42:49,  2.67it/s]
                                                    
{'loss': 4.3956, 'grad_norm': 0.4086023271083832, 'learning_rate': 0.0008105388863445693, 'epoch': 0.36}
-
 36%|███▌      | 3825/10682 [39:48<42:49,  2.67it/s]
 36%|███▌      | 3826/10682 [39:48<42:55,  2.66it/s]
 36%|███▌      | 3827/10682 [39:49<42:52,  2.66it/s]
 36%|███▌      | 3828/10682 [39:49<42:54,  2.66it/s]
 36%|███▌      | 3829/10682 [39:49<42:53,  2.66it/s]
 36%|███▌      | 3830/10682 [39:50<42:54,  2.66it/s]
 36%|███▌      | 3831/10682 [39:50<42:52,  2.66it/s]
 36%|███▌      | 3832/10682 [39:51<42:54,  2.66it/s]
 36%|███▌      | 3833/10682 [39:51<42:51,  2.66it/s]
 36%|███▌      | 3834/10682 [39:51<42:49,  2.67it/s]
 36%|███▌      | 3835/10682 [39:52<42:47,  2.67it/s]
 36%|███▌      | 3836/10682 [39:52<42:45,  2.67it/s]
 36%|███▌      | 3837/10682 [39:52<42:46,  2.67it/s]
 36%|███▌      | 3838/10682 [39:53<42:44,  2.67it/s]
 36%|███▌      | 3839/10682 [39:53<42:48,  2.66it/s]
 36%|███▌      | 3840/10682 [39:54<42:45,  2.67it/s]
 36%|███▌      | 3841/10682 [39:54<42:44,  2.67it/s]
 36%|███▌      | 3842/10682 [39:54<42:45,  2.67it/s]
 36%|███▌      | 3843/10682 [39:55<42:42,  2.67it/s]
 36%|███▌      | 3844/10682 [39:55<42:42,  2.67it/s]
 36%|███▌      | 3845/10682 [39:55<42:43,  2.67it/s]
 36%|███▌      | 3846/10682 [39:56<42:41,  2.67it/s]
 36%|███▌      | 3847/10682 [39:56<42:39,  2.67it/s]
 36%|███▌      | 3848/10682 [39:57<42:38,  2.67it/s]
 36%|███▌      | 3849/10682 [39:57<42:38,  2.67it/s]
 36%|███▌      | 3850/10682 [39:57<42:40,  2.67it/s]{'loss': 4.3901, 'grad_norm': 0.4449955224990845, 'learning_rate': 0.0008073268776325152, 'epoch': 0.36}
                                                    
-
 36%|███▌      | 3850/10682 [39:57<42:40,  2.67it/s]
 36%|███▌      | 3851/10682 [39:58<42:42,  2.67it/s]
 36%|███▌      | 3852/10682 [39:58<42:42,  2.66it/s]
 36%|███▌      | 3853/10682 [39:58<42:43,  2.66it/s]
 36%|███▌      | 3854/10682 [39:59<42:42,  2.66it/s]
 36%|███▌      | 3855/10682 [39:59<42:45,  2.66it/s]
 36%|███▌      | 3856/10682 [40:00<42:44,  2.66it/s]
 36%|███▌      | 3857/10682 [40:00<42:41,  2.66it/s]
 36%|███▌      | 3858/10682 [40:00<42:40,  2.67it/s]
 36%|███▌      | 3859/10682 [40:01<42:37,  2.67it/s]
 36%|███▌      | 3860/10682 [40:01<42:37,  2.67it/s]
 36%|███▌      | 3861/10682 [40:01<42:35,  2.67it/s]
 36%|███▌      | 3862/10682 [40:02<42:33,  2.67it/s]
 36%|███▌      | 3863/10682 [40:02<42:32,  2.67it/s]
 36%|███▌      | 3864/10682 [40:03<42:30,  2.67it/s]
 36%|███▌      | 3865/10682 [40:03<42:30,  2.67it/s]
 36%|███▌      | 3866/10682 [40:03<42:35,  2.67it/s]
 36%|███▌      | 3867/10682 [40:04<42:35,  2.67it/s]
 36%|███▌      | 3868/10682 [40:04<42:44,  2.66it/s]
 36%|███▌      | 3869/10682 [40:04<42:40,  2.66it/s]
 36%|███▌      | 3870/10682 [40:05<42:36,  2.66it/s]
 36%|███▌      | 3871/10682 [40:05<42:33,  2.67it/s]
 36%|███▌      | 3872/10682 [40:06<42:35,  2.66it/s]
 36%|███▋      | 3873/10682 [40:06<42:34,  2.67it/s]
 36%|███▋      | 3874/10682 [40:06<42:35,  2.66it/s]
 36%|███▋      | 3875/10682 [40:07<42:32,  2.67it/s]{'loss': 4.3941, 'grad_norm': 0.4480914771556854, 'learning_rate': 0.0008040943544644116, 'epoch': 0.36}
-
                                                    

 36%|███▋      | 3875/10682 [40:07<42:32,  2.67it/s]
 36%|███▋      | 3876/10682 [40:07<42:35,  2.66it/s]
 36%|███▋      | 3877/10682 [40:07<42:32,  2.67it/s]
 36%|███▋      | 3878/10682 [40:08<42:34,  2.66it/s]
 36%|███▋      | 3879/10682 [40:08<42:37,  2.66it/s]
 36%|███▋      | 3880/10682 [40:09<42:34,  2.66it/s]
 36%|███▋      | 3881/10682 [40:09<42:33,  2.66it/s]
 36%|███▋      | 3882/10682 [40:09<42:31,  2.66it/s]
 36%|███▋      | 3883/10682 [40:10<42:34,  2.66it/s]
 36%|███▋      | 3884/10682 [40:10<42:30,  2.67it/s]
 36%|███▋      | 3885/10682 [40:10<42:28,  2.67it/s]
 36%|███▋      | 3886/10682 [40:11<42:27,  2.67it/s]
 36%|███▋      | 3887/10682 [40:11<42:27,  2.67it/s]
 36%|███▋      | 3888/10682 [40:12<42:27,  2.67it/s]
 36%|███▋      | 3889/10682 [40:12<42:28,  2.67it/s]
 36%|███▋      | 3890/10682 [40:12<42:26,  2.67it/s]
 36%|███▋      | 3891/10682 [40:13<42:23,  2.67it/s]
 36%|███▋      | 3892/10682 [40:13<42:32,  2.66it/s]
 36%|███▋      | 3893/10682 [40:13<42:32,  2.66it/s]
 36%|███▋      | 3894/10682 [40:14<42:27,  2.66it/s]
 36%|███▋      | 3895/10682 [40:14<42:33,  2.66it/s]
 36%|███▋      | 3896/10682 [40:15<42:34,  2.66it/s]
 36%|███▋      | 3897/10682 [40:15<42:30,  2.66it/s]
 36%|███▋      | 3898/10682 [40:15<42:25,  2.66it/s]
 37%|███▋      | 3899/10682 [40:16<42:24,  2.67it/s]
 37%|███▋      | 3900/10682 [40:16<51:43,  2.19it/s]
                                                    {'loss': 4.3866, 'grad_norm': 0.40509456396102905, 'learning_rate': 0.00080084153261525, 'epoch': 0.37}
-

 37%|███▋      | 3900/10682 [40:16<51:43,  2.19it/s]
 37%|███▋      | 3901/10682 [40:17<49:02,  2.30it/s]
 37%|███▋      | 3902/10682 [40:17<47:02,  2.40it/s]
 37%|███▋      | 3903/10682 [40:17<45:35,  2.48it/s]
 37%|███▋      | 3904/10682 [40:18<44:35,  2.53it/s]
 37%|███▋      | 3905/10682 [40:18<43:54,  2.57it/s]
 37%|███▋      | 3906/10682 [40:19<43:26,  2.60it/s]
 37%|███▋      | 3907/10682 [40:19<43:08,  2.62it/s]
 37%|███▋      | 3908/10682 [40:19<42:56,  2.63it/s]
 37%|███▋      | 3909/10682 [40:20<42:46,  2.64it/s]
 37%|███▋      | 3910/10682 [40:20<42:36,  2.65it/s]
 37%|███▋      | 3911/10682 [40:20<42:35,  2.65it/s]
 37%|███▋      | 3912/10682 [40:21<42:33,  2.65it/s]
 37%|███▋      | 3913/10682 [40:21<42:30,  2.65it/s]
 37%|███▋      | 3914/10682 [40:22<42:25,  2.66it/s]
 37%|███▋      | 3915/10682 [40:22<42:20,  2.66it/s]
 37%|███▋      | 3916/10682 [40:22<42:18,  2.66it/s]
 37%|███▋      | 3917/10682 [40:23<42:16,  2.67it/s]
 37%|███▋      | 3918/10682 [40:23<52:22,  2.15it/s]
 37%|███▋      | 3919/10682 [40:24<49:17,  2.29it/s]
 37%|███▋      | 3920/10682 [40:24<47:11,  2.39it/s]
 37%|███▋      | 3921/10682 [40:25<45:41,  2.47it/s]
 37%|███▋      | 3922/10682 [40:25<44:40,  2.52it/s]
 37%|███▋      | 3923/10682 [40:25<43:54,  2.57it/s]
 37%|███▋      | 3924/10682 [40:26<43:26,  2.59it/s]
 37%|███▋      | 3925/10682 [40:26<43:04,  2.61it/s]
                                                    
{'loss': 4.383, 'grad_norm': 0.481227844953537, 'learning_rate': 0.0007975686292149845, 'epoch': 0.37}
-
 37%|███▋      | 3925/10682 [40:26<43:04,  2.61it/s]
 37%|███▋      | 3926/10682 [40:26<42:49,  2.63it/s]
 37%|███▋      | 3927/10682 [40:27<42:42,  2.64it/s]
 37%|███▋      | 3928/10682 [40:27<42:33,  2.65it/s]
 37%|███▋      | 3929/10682 [40:28<42:22,  2.66it/s]
 37%|███▋      | 3930/10682 [40:28<42:18,  2.66it/s]
 37%|███▋      | 3931/10682 [40:28<42:16,  2.66it/s]
 37%|███▋      | 3932/10682 [40:29<42:14,  2.66it/s]
 37%|███▋      | 3933/10682 [40:29<42:15,  2.66it/s]
 37%|███▋      | 3934/10682 [40:29<42:12,  2.66it/s]
 37%|███▋      | 3935/10682 [40:30<42:10,  2.67it/s]
 37%|███▋      | 3936/10682 [40:30<42:11,  2.67it/s]
 37%|███▋      | 3937/10682 [40:31<42:09,  2.67it/s]
 37%|███▋      | 3938/10682 [40:31<42:09,  2.67it/s]
 37%|███▋      | 3939/10682 [40:31<42:09,  2.67it/s]
 37%|███▋      | 3940/10682 [40:32<42:08,  2.67it/s]
 37%|███▋      | 3941/10682 [40:32<42:10,  2.66it/s]
 37%|███▋      | 3942/10682 [40:32<42:08,  2.67it/s]
 37%|███▋      | 3943/10682 [40:33<42:04,  2.67it/s]
 37%|███▋      | 3944/10682 [40:33<42:03,  2.67it/s]
 37%|███▋      | 3945/10682 [40:34<42:01,  2.67it/s]
 37%|███▋      | 3946/10682 [40:34<42:02,  2.67it/s]
 37%|███▋      | 3947/10682 [40:34<42:02,  2.67it/s]
 37%|███▋      | 3948/10682 [40:35<41:59,  2.67it/s]
 37%|███▋      | 3949/10682 [40:35<41:57,  2.67it/s]
 37%|███▋      | 3950/10682 [40:35<42:01,  2.67it/s]{'loss': 4.3852, 'grad_norm': 0.4570035934448242, 'learning_rate': 0.0007942758627340383, 'epoch': 0.37}
                                                    
-
 37%|███▋      | 3950/10682 [40:35<42:01,  2.67it/s]
 37%|███▋      | 3951/10682 [40:36<42:04,  2.67it/s]
 37%|███▋      | 3952/10682 [40:36<42:02,  2.67it/s]
 37%|███▋      | 3953/10682 [40:37<42:00,  2.67it/s]
 37%|███▋      | 3954/10682 [40:37<41:59,  2.67it/s]
 37%|███▋      | 3955/10682 [40:37<42:02,  2.67it/s]
 37%|███▋      | 3956/10682 [40:38<41:59,  2.67it/s]
 37%|███▋      | 3957/10682 [40:38<41:59,  2.67it/s]
 37%|███▋      | 3958/10682 [40:38<41:59,  2.67it/s]
 37%|███▋      | 3959/10682 [40:39<42:04,  2.66it/s]
 37%|███▋      | 3960/10682 [40:39<42:00,  2.67it/s]
 37%|███▋      | 3961/10682 [40:40<42:04,  2.66it/s]
 37%|███▋      | 3962/10682 [40:40<42:05,  2.66it/s]
 37%|███▋      | 3963/10682 [40:40<42:04,  2.66it/s]
 37%|███▋      | 3964/10682 [40:41<42:06,  2.66it/s]
 37%|███▋      | 3965/10682 [40:41<42:06,  2.66it/s]
 37%|███▋      | 3966/10682 [40:41<42:01,  2.66it/s]
 37%|███▋      | 3967/10682 [40:42<41:57,  2.67it/s]
 37%|███▋      | 3968/10682 [40:42<41:55,  2.67it/s]
 37%|███▋      | 3969/10682 [40:43<41:54,  2.67it/s]
 37%|███▋      | 3970/10682 [40:43<41:54,  2.67it/s]
 37%|███▋      | 3971/10682 [40:43<41:56,  2.67it/s]
 37%|███▋      | 3972/10682 [40:44<41:54,  2.67it/s]
 37%|███▋      | 3973/10682 [40:44<41:54,  2.67it/s]
 37%|███▋      | 3974/10682 [40:44<41:54,  2.67it/s]
 37%|███▋      | 3975/10682 [40:45<41:58,  2.66it/s]
                                                    
{'loss': 4.3807, 'grad_norm': 0.4591383934020996, 'learning_rate': 0.00079096345296872, 'epoch': 0.37}
-
 37%|███▋      | 3975/10682 [40:45<41:58,  2.66it/s]
 37%|███▋      | 3976/10682 [40:45<41:57,  2.66it/s]
 37%|███▋      | 3977/10682 [40:46<42:03,  2.66it/s]
 37%|███▋      | 3978/10682 [40:46<42:01,  2.66it/s]
 37%|███▋      | 3979/10682 [40:46<41:57,  2.66it/s]
 37%|███▋      | 3980/10682 [40:47<41:57,  2.66it/s]
 37%|███▋      | 3981/10682 [40:47<41:55,  2.66it/s]
 37%|███▋      | 3982/10682 [40:47<41:55,  2.66it/s]
 37%|███▋      | 3983/10682 [40:48<41:51,  2.67it/s]
 37%|███▋      | 3984/10682 [40:48<41:53,  2.67it/s]
 37%|███▋      | 3985/10682 [40:49<41:52,  2.67it/s]
 37%|███▋      | 3986/10682 [40:49<41:51,  2.67it/s]
 37%|███▋      | 3987/10682 [40:49<41:50,  2.67it/s]
 37%|███▋      | 3988/10682 [40:50<41:53,  2.66it/s]
 37%|███▋      | 3989/10682 [40:50<41:54,  2.66it/s]
 37%|███▋      | 3990/10682 [40:50<41:53,  2.66it/s]
 37%|███▋      | 3991/10682 [40:51<41:56,  2.66it/s]
 37%|███▋      | 3992/10682 [40:51<41:56,  2.66it/s]
 37%|███▋      | 3993/10682 [40:52<41:54,  2.66it/s]
 37%|███▋      | 3994/10682 [40:52<41:56,  2.66it/s]
 37%|███▋      | 3995/10682 [40:52<41:56,  2.66it/s]
 37%|███▋      | 3996/10682 [40:53<41:54,  2.66it/s]
 37%|███▋      | 3997/10682 [40:53<41:50,  2.66it/s]
 37%|███▋      | 3998/10682 [40:53<41:48,  2.66it/s]
 37%|███▋      | 3999/10682 [40:54<41:45,  2.67it/s]
 37%|███▋      | 4000/10682 [40:54<41:49,  2.66it/s]{'loss': 4.3793, 'grad_norm': 0.466601699590683, 'learning_rate': 0.0007876316210265521, 'epoch': 0.37}
                                                    
-
 37%|███▋      | 4000/10682 [40:54<41:49,  2.66it/s]
 37%|███▋      | 4001/10682 [40:55<41:50,  2.66it/s]
 37%|███▋      | 4002/10682 [40:55<41:52,  2.66it/s]
 37%|███▋      | 4003/10682 [40:55<41:50,  2.66it/s]
 37%|███▋      | 4004/10682 [40:56<41:54,  2.66it/s]
 37%|███▋      | 4005/10682 [40:56<41:54,  2.66it/s]
 38%|███▊      | 4006/10682 [40:56<41:50,  2.66it/s]
 38%|███▊      | 4007/10682 [40:57<41:50,  2.66it/s]
 38%|███▊      | 4008/10682 [40:57<41:46,  2.66it/s]
 38%|███▊      | 4009/10682 [40:58<41:43,  2.67it/s]
 38%|███▊      | 4010/10682 [40:58<41:41,  2.67it/s]
 38%|███▊      | 4011/10682 [40:58<41:40,  2.67it/s]
 38%|███▊      | 4012/10682 [40:59<41:37,  2.67it/s]
 38%|███▊      | 4013/10682 [40:59<41:39,  2.67it/s]
 38%|███▊      | 4014/10682 [40:59<41:36,  2.67it/s]
 38%|███▊      | 4015/10682 [41:00<41:38,  2.67it/s]
 38%|███▊      | 4016/10682 [41:00<41:43,  2.66it/s]
 38%|███▊      | 4017/10682 [41:01<41:41,  2.66it/s]
 38%|███▊      | 4018/10682 [41:01<41:39,  2.67it/s]
 38%|███▊      | 4019/10682 [41:01<41:40,  2.66it/s]
 38%|███▊      | 4020/10682 [41:02<41:40,  2.66it/s]
 38%|███▊      | 4021/10682 [41:02<41:40,  2.66it/s]
 38%|███▊      | 4022/10682 [41:02<41:40,  2.66it/s]
 38%|███▊      | 4023/10682 [41:03<41:41,  2.66it/s]
 38%|███▊      | 4024/10682 [41:03<41:39,  2.66it/s]
 38%|███▊      | 4025/10682 [41:04<41:37,  2.67it/s]{'loss': 4.3758, 'grad_norm': 0.4560650587081909, 'learning_rate': 0.0007842805893115124, 'epoch': 0.38}
                                                    
-
 38%|███▊      | 4025/10682 [41:04<41:37,  2.67it/s]
 38%|███▊      | 4026/10682 [41:04<41:38,  2.66it/s]
 38%|███▊      | 4027/10682 [41:04<41:40,  2.66it/s]
 38%|███▊      | 4028/10682 [41:05<41:37,  2.66it/s]
 38%|███▊      | 4029/10682 [41:05<41:40,  2.66it/s]
 38%|███▊      | 4030/10682 [41:05<41:40,  2.66it/s]
 38%|███▊      | 4031/10682 [41:06<41:37,  2.66it/s]
 38%|███▊      | 4032/10682 [41:06<41:40,  2.66it/s]
 38%|███▊      | 4033/10682 [41:07<41:37,  2.66it/s]
 38%|███▊      | 4034/10682 [41:07<41:35,  2.66it/s]
 38%|███▊      | 4035/10682 [41:07<41:32,  2.67it/s]
 38%|███▊      | 4036/10682 [41:08<41:35,  2.66it/s]
 38%|███▊      | 4037/10682 [41:08<41:32,  2.67it/s]
 38%|███▊      | 4038/10682 [41:08<41:32,  2.67it/s]
 38%|███▊      | 4039/10682 [41:09<41:34,  2.66it/s]
 38%|███▊      | 4040/10682 [41:09<41:32,  2.67it/s]
 38%|███▊      | 4041/10682 [41:10<41:30,  2.67it/s]
 38%|███▊      | 4042/10682 [41:10<41:28,  2.67it/s]
 38%|███▊      | 4043/10682 [41:10<41:31,  2.66it/s]
 38%|███▊      | 4044/10682 [41:11<41:30,  2.67it/s]
 38%|███▊      | 4045/10682 [41:11<41:29,  2.67it/s]
 38%|███▊      | 4046/10682 [41:11<41:29,  2.67it/s]
 38%|███▊      | 4047/10682 [41:12<41:27,  2.67it/s]
 38%|███▊      | 4048/10682 [41:12<41:28,  2.67it/s]
 38%|███▊      | 4049/10682 [41:13<41:28,  2.67it/s]
 38%|███▊      | 4050/10682 [41:13<41:28,  2.66it/s]
                                                    
{'loss': 4.3768, 'grad_norm': 0.4785013794898987, 'learning_rate': 0.0007809105815091867, 'epoch': 0.38}
-
 38%|███▊      | 4050/10682 [41:13<41:28,  2.66it/s]
 38%|███▊      | 4051/10682 [41:13<41:27,  2.67it/s]
 38%|███▊      | 4052/10682 [41:14<41:30,  2.66it/s]
 38%|███▊      | 4053/10682 [41:14<41:27,  2.66it/s]
 38%|███▊      | 4054/10682 [41:14<41:26,  2.67it/s]
 38%|███▊      | 4055/10682 [41:15<41:24,  2.67it/s]
 38%|███▊      | 4056/10682 [41:15<41:23,  2.67it/s]
 38%|███▊      | 4057/10682 [41:16<41:24,  2.67it/s]
 38%|███▊      | 4058/10682 [41:16<41:23,  2.67it/s]
 38%|███▊      | 4059/10682 [41:16<41:26,  2.66it/s]
 38%|███▊      | 4060/10682 [41:17<41:26,  2.66it/s]
 38%|███▊      | 4061/10682 [41:17<41:23,  2.67it/s]
 38%|███▊      | 4062/10682 [41:17<41:23,  2.67it/s]
 38%|███▊      | 4063/10682 [41:18<41:22,  2.67it/s]
 38%|███▊      | 4064/10682 [41:18<41:25,  2.66it/s]
 38%|███▊      | 4065/10682 [41:19<41:25,  2.66it/s]
 38%|███▊      | 4066/10682 [41:19<41:24,  2.66it/s]
 38%|███▊      | 4067/10682 [41:19<41:25,  2.66it/s]
 38%|███▊      | 4068/10682 [41:20<41:26,  2.66it/s]
 38%|███▊      | 4069/10682 [41:20<41:27,  2.66it/s]
 38%|███▊      | 4070/10682 [41:20<41:28,  2.66it/s]
 38%|███▊      | 4071/10682 [41:21<41:26,  2.66it/s]
 38%|███▊      | 4072/10682 [41:21<41:23,  2.66it/s]
 38%|███▊      | 4073/10682 [41:22<41:22,  2.66it/s]
 38%|███▊      | 4074/10682 [41:22<41:20,  2.66it/s]
 38%|███▊      | 4075/10682 [41:22<41:19,  2.67it/s]{'loss': 4.3805, 'grad_norm': 0.45421546697616577, 'learning_rate': 0.0007775218225718394, 'epoch': 0.38}
                                                    
-
 38%|███▊      | 4075/10682 [41:22<41:19,  2.67it/s]
 38%|███▊      | 4076/10682 [41:23<41:20,  2.66it/s]
 38%|███▊      | 4077/10682 [41:23<41:19,  2.66it/s]
 38%|███▊      | 4078/10682 [41:23<41:17,  2.67it/s]
 38%|███▊      | 4079/10682 [41:24<41:20,  2.66it/s]
 38%|███▊      | 4080/10682 [41:24<41:17,  2.66it/s]
 38%|███▊      | 4081/10682 [41:25<41:16,  2.67it/s]
 38%|███▊      | 4082/10682 [41:25<41:16,  2.67it/s]
 38%|███▊      | 4083/10682 [41:25<41:13,  2.67it/s]
 38%|███▊      | 4084/10682 [41:26<41:13,  2.67it/s]
 38%|███▊      | 4085/10682 [41:26<41:13,  2.67it/s]
 38%|███▊      | 4086/10682 [41:26<41:14,  2.67it/s]
 38%|███▊      | 4087/10682 [41:27<41:12,  2.67it/s]
 38%|███▊      | 4088/10682 [41:27<41:10,  2.67it/s]
 38%|███▊      | 4089/10682 [41:28<41:08,  2.67it/s]
 38%|███▊      | 4090/10682 [41:28<41:09,  2.67it/s]
 38%|███▊      | 4091/10682 [41:28<41:11,  2.67it/s]
 38%|███▊      | 4092/10682 [41:29<41:09,  2.67it/s]
 38%|███▊      | 4093/10682 [41:29<41:06,  2.67it/s]
 38%|███▊      | 4094/10682 [41:29<41:10,  2.67it/s]
 38%|███▊      | 4095/10682 [41:30<41:09,  2.67it/s]
 38%|███▊      | 4096/10682 [41:30<41:08,  2.67it/s]
 38%|███▊      | 4097/10682 [41:31<41:07,  2.67it/s]
 38%|███▊      | 4098/10682 [41:31<41:06,  2.67it/s]
 38%|███▊      | 4099/10682 [41:31<41:09,  2.67it/s]
 38%|███▊      | 4100/10682 [41:32<41:11,  2.66it/s]
                                                    {'loss': 4.3756, 'grad_norm': 0.453066885471344, 'learning_rate': 0.0007741145387033967, 'epoch': 0.38}
-

 38%|███▊      | 4100/10682 [41:32<41:11,  2.66it/s]
 38%|███▊      | 4101/10682 [41:32<41:11,  2.66it/s]
 38%|███▊      | 4102/10682 [41:32<41:10,  2.66it/s]
 38%|███▊      | 4103/10682 [41:33<41:06,  2.67it/s]
 38%|███▊      | 4104/10682 [41:33<41:06,  2.67it/s]
 38%|███▊      | 4105/10682 [41:34<41:05,  2.67it/s]
 38%|███▊      | 4106/10682 [41:34<41:03,  2.67it/s]
 38%|███▊      | 4107/10682 [41:34<41:03,  2.67it/s]
 38%|███▊      | 4108/10682 [41:35<41:01,  2.67it/s]
 38%|███▊      | 4109/10682 [41:35<41:02,  2.67it/s]
 38%|███▊      | 4110/10682 [41:35<41:01,  2.67it/s]
 38%|███▊      | 4111/10682 [41:36<41:02,  2.67it/s]
 38%|███▊      | 4112/10682 [41:36<41:03,  2.67it/s]
 39%|███▊      | 4113/10682 [41:37<41:01,  2.67it/s]
 39%|███▊      | 4114/10682 [41:37<41:10,  2.66it/s]
 39%|███▊      | 4115/10682 [41:37<41:09,  2.66it/s]
 39%|███▊      | 4116/10682 [41:38<41:06,  2.66it/s]
 39%|███▊      | 4117/10682 [41:38<41:03,  2.66it/s]
 39%|███▊      | 4118/10682 [41:38<41:02,  2.67it/s]
 39%|███▊      | 4119/10682 [41:39<41:02,  2.67it/s]
 39%|███▊      | 4120/10682 [41:39<40:59,  2.67it/s]
 39%|███▊      | 4121/10682 [41:40<40:58,  2.67it/s]
 39%|███▊      | 4122/10682 [41:40<40:56,  2.67it/s]
 39%|███▊      | 4123/10682 [41:40<40:56,  2.67it/s]
 39%|███▊      | 4124/10682 [41:41<40:54,  2.67it/s]
 39%|███▊      | 4125/10682 [41:41<40:58,  2.67it/s]
                                                    
{'loss': 4.3716, 'grad_norm': 0.44294261932373047, 'learning_rate': 0.0007706889573443467, 'epoch': 0.39}
-
 39%|███▊      | 4125/10682 [41:41<40:58,  2.67it/s]
 39%|███▊      | 4126/10682 [41:41<41:04,  2.66it/s]
 39%|███▊      | 4127/10682 [41:42<41:03,  2.66it/s]
 39%|███▊      | 4128/10682 [41:42<41:01,  2.66it/s]
 39%|███▊      | 4129/10682 [41:43<40:56,  2.67it/s]
 39%|███▊      | 4130/10682 [41:43<40:55,  2.67it/s]
 39%|███▊      | 4131/10682 [41:43<40:54,  2.67it/s]
 39%|███▊      | 4132/10682 [41:44<40:57,  2.67it/s]
 39%|███▊      | 4133/10682 [41:44<40:56,  2.67it/s]
 39%|███▊      | 4134/10682 [41:44<40:56,  2.67it/s]
 39%|███▊      | 4135/10682 [41:45<40:55,  2.67it/s]
 39%|███▊      | 4136/10682 [41:45<40:56,  2.66it/s]
 39%|███▊      | 4137/10682 [41:46<40:57,  2.66it/s]
 39%|███▊      | 4138/10682 [41:46<40:54,  2.67it/s]
 39%|███▊      | 4139/10682 [41:46<40:55,  2.66it/s]
 39%|███▉      | 4140/10682 [41:47<40:52,  2.67it/s]
 39%|███▉      | 4141/10682 [41:47<40:55,  2.66it/s]
 39%|███▉      | 4142/10682 [41:47<40:55,  2.66it/s]
 39%|███▉      | 4143/10682 [41:48<40:53,  2.67it/s]
 39%|███▉      | 4144/10682 [41:48<40:50,  2.67it/s]
 39%|███▉      | 4145/10682 [41:49<40:49,  2.67it/s]
 39%|███▉      | 4146/10682 [41:49<40:49,  2.67it/s]
 39%|███▉      | 4147/10682 [41:49<40:48,  2.67it/s]
 39%|███▉      | 4148/10682 [41:50<40:47,  2.67it/s]
 39%|███▉      | 4149/10682 [41:50<40:47,  2.67it/s]
 39%|███▉      | 4150/10682 [41:50<40:45,  2.67it/s]
                                                    
{'loss': 4.3708, 'grad_norm': 0.4292045831680298, 'learning_rate': 0.0007672453071565586, 'epoch': 0.39}
-
 39%|███▉      | 4150/10682 [41:50<40:45,  2.67it/s]
 39%|███▉      | 4151/10682 [41:51<40:56,  2.66it/s]
 39%|███▉      | 4152/10682 [41:51<40:55,  2.66it/s]
 39%|███▉      | 4153/10682 [41:52<40:55,  2.66it/s]
 39%|███▉      | 4154/10682 [41:52<41:01,  2.65it/s]
 39%|███▉      | 4155/10682 [41:52<41:05,  2.65it/s]
 39%|███▉      | 4156/10682 [41:53<40:56,  2.66it/s]
 39%|███▉      | 4157/10682 [41:53<40:55,  2.66it/s]
 39%|███▉      | 4158/10682 [41:53<40:51,  2.66it/s]
 39%|███▉      | 4159/10682 [41:54<40:53,  2.66it/s]
 39%|███▉      | 4160/10682 [41:54<40:54,  2.66it/s]
 39%|███▉      | 4161/10682 [41:55<40:51,  2.66it/s]
 39%|███▉      | 4162/10682 [41:55<40:48,  2.66it/s]
 39%|███▉      | 4163/10682 [41:55<40:46,  2.67it/s]
 39%|███▉      | 4164/10682 [41:56<40:46,  2.66it/s]
 39%|███▉      | 4165/10682 [41:56<40:45,  2.67it/s]
 39%|███▉      | 4166/10682 [41:56<40:51,  2.66it/s]
 39%|███▉      | 4167/10682 [41:57<40:49,  2.66it/s]
 39%|███▉      | 4168/10682 [41:57<40:48,  2.66it/s]
 39%|███▉      | 4169/10682 [41:58<40:48,  2.66it/s]
 39%|███▉      | 4170/10682 [41:58<40:47,  2.66it/s]
 39%|███▉      | 4171/10682 [41:58<40:47,  2.66it/s]
 39%|███▉      | 4172/10682 [41:59<40:44,  2.66it/s]
 39%|███▉      | 4173/10682 [41:59<40:45,  2.66it/s]
 39%|███▉      | 4174/10682 [41:59<40:44,  2.66it/s]
 39%|███▉      | 4175/10682 [42:00<40:41,  2.67it/s]{'loss': 4.376, 'grad_norm': 0.4271419644355774, 'learning_rate': 0.0007637838180080183, 'epoch': 0.39}
                                                    
-
 39%|███▉      | 4175/10682 [42:00<40:41,  2.67it/s]
 39%|███▉      | 4176/10682 [42:00<40:42,  2.66it/s]
 39%|███▉      | 4177/10682 [42:01<40:42,  2.66it/s]
 39%|███▉      | 4178/10682 [42:01<40:42,  2.66it/s]
 39%|███▉      | 4179/10682 [42:01<40:43,  2.66it/s]
 39%|███▉      | 4180/10682 [42:02<40:44,  2.66it/s]
 39%|███▉      | 4181/10682 [42:02<40:49,  2.65it/s]
 39%|███▉      | 4182/10682 [42:02<40:54,  2.65it/s]
 39%|███▉      | 4183/10682 [42:03<40:51,  2.65it/s]
 39%|███▉      | 4184/10682 [42:03<40:48,  2.65it/s]
 39%|███▉      | 4185/10682 [42:04<40:45,  2.66it/s]
 39%|███▉      | 4186/10682 [42:04<40:43,  2.66it/s]
 39%|███▉      | 4187/10682 [42:04<40:44,  2.66it/s]
 39%|███▉      | 4188/10682 [42:05<40:43,  2.66it/s]
 39%|███▉      | 4189/10682 [42:05<40:41,  2.66it/s]
 39%|███▉      | 4190/10682 [42:05<40:39,  2.66it/s]
 39%|███▉      | 4191/10682 [42:06<40:40,  2.66it/s]
 39%|███▉      | 4192/10682 [42:06<40:40,  2.66it/s]
 39%|███▉      | 4193/10682 [42:07<40:47,  2.65it/s]
 39%|███▉      | 4194/10682 [42:07<40:48,  2.65it/s]
 39%|███▉      | 4195/10682 [42:07<40:46,  2.65it/s]
 39%|███▉      | 4196/10682 [42:08<40:43,  2.65it/s]
 39%|███▉      | 4197/10682 [42:08<40:40,  2.66it/s]
 39%|███▉      | 4198/10682 [42:09<40:40,  2.66it/s]
 39%|███▉      | 4199/10682 [42:09<40:37,  2.66it/s]
 39%|███▉      | 4200/10682 [42:09<40:37,  2.66it/s]
                                                    
{'loss': 4.3718, 'grad_norm': 0.43281835317611694, 'learning_rate': 0.0007603047209574854, 'epoch': 0.39}
-
 39%|███▉      | 4200/10682 [42:09<40:37,  2.66it/s]
 39%|███▉      | 4201/10682 [42:10<40:36,  2.66it/s]
 39%|███▉      | 4202/10682 [42:10<40:35,  2.66it/s]
 39%|███▉      | 4203/10682 [42:10<40:33,  2.66it/s]
 39%|███▉      | 4204/10682 [42:11<40:33,  2.66it/s]
 39%|███▉      | 4205/10682 [42:11<40:32,  2.66it/s]
 39%|███▉      | 4206/10682 [42:12<40:35,  2.66it/s]
 39%|███▉      | 4207/10682 [42:12<40:30,  2.66it/s]
 39%|███▉      | 4208/10682 [42:12<40:31,  2.66it/s]
 39%|███▉      | 4209/10682 [42:13<40:34,  2.66it/s]
 39%|███▉      | 4210/10682 [42:13<40:35,  2.66it/s]
 39%|███▉      | 4211/10682 [42:13<40:32,  2.66it/s]
 39%|███▉      | 4212/10682 [42:14<40:31,  2.66it/s]
 39%|███▉      | 4213/10682 [42:14<40:29,  2.66it/s]
 39%|███▉      | 4214/10682 [42:15<40:25,  2.67it/s]
 39%|███▉      | 4215/10682 [42:15<40:26,  2.67it/s]
 39%|███▉      | 4216/10682 [42:15<40:27,  2.66it/s]
 39%|███▉      | 4217/10682 [42:16<40:25,  2.66it/s]
 39%|███▉      | 4218/10682 [42:16<40:23,  2.67it/s]
 39%|███▉      | 4219/10682 [42:16<40:23,  2.67it/s]
 40%|███▉      | 4220/10682 [42:17<40:26,  2.66it/s]
 40%|███▉      | 4221/10682 [42:17<40:27,  2.66it/s]
 40%|███▉      | 4222/10682 [42:18<40:29,  2.66it/s]
 40%|███▉      | 4223/10682 [42:18<40:25,  2.66it/s]
 40%|███▉      | 4224/10682 [42:18<40:23,  2.67it/s]
 40%|███▉      | 4225/10682 [42:19<40:22,  2.67it/s]{'loss': 4.3682, 'grad_norm': 0.4565103054046631, 'learning_rate': 0.0007568082482390682, 'epoch': 0.4}
                                                    
-
 40%|███▉      | 4225/10682 [42:19<40:22,  2.67it/s]
 40%|███▉      | 4226/10682 [42:19<40:31,  2.66it/s]
 40%|███▉      | 4227/10682 [42:19<40:26,  2.66it/s]
 40%|███▉      | 4228/10682 [42:20<40:20,  2.67it/s]
 40%|███▉      | 4229/10682 [42:20<40:21,  2.67it/s]
 40%|███▉      | 4230/10682 [42:21<40:17,  2.67it/s]
 40%|███▉      | 4231/10682 [42:21<40:17,  2.67it/s]
 40%|███▉      | 4232/10682 [42:21<40:17,  2.67it/s]
 40%|███▉      | 4233/10682 [42:22<40:23,  2.66it/s]
 40%|███▉      | 4234/10682 [42:22<40:19,  2.67it/s]
 40%|███▉      | 4235/10682 [42:22<40:19,  2.66it/s]
 40%|███▉      | 4236/10682 [42:23<40:20,  2.66it/s]
 40%|███▉      | 4237/10682 [42:23<40:17,  2.67it/s]
 40%|███▉      | 4238/10682 [42:24<40:15,  2.67it/s]
 40%|███▉      | 4239/10682 [42:24<40:13,  2.67it/s]
 40%|███▉      | 4240/10682 [42:24<40:14,  2.67it/s]
 40%|███▉      | 4241/10682 [42:25<40:12,  2.67it/s]
 40%|███▉      | 4242/10682 [42:25<40:15,  2.67it/s]
 40%|███▉      | 4243/10682 [42:25<40:15,  2.67it/s]
 40%|███▉      | 4244/10682 [42:26<40:17,  2.66it/s]
 40%|███▉      | 4245/10682 [42:26<40:17,  2.66it/s]
 40%|███▉      | 4246/10682 [42:27<40:19,  2.66it/s]
 40%|███▉      | 4247/10682 [42:27<40:18,  2.66it/s]
 40%|███▉      | 4248/10682 [42:27<40:14,  2.66it/s]
 40%|███▉      | 4249/10682 [42:28<40:16,  2.66it/s]
 40%|███▉      | 4250/10682 [42:28<40:15,  2.66it/s]{'loss': 4.3629, 'grad_norm': 0.419894814491272, 'learning_rate': 0.0007532946332467235, 'epoch': 0.4}
                                                    
-
 40%|███▉      | 4250/10682 [42:28<40:15,  2.66it/s]
 40%|███▉      | 4251/10682 [42:28<40:17,  2.66it/s]
 40%|███▉      | 4252/10682 [42:29<40:16,  2.66it/s]
 40%|███▉      | 4253/10682 [42:29<40:14,  2.66it/s]
 40%|███▉      | 4254/10682 [42:30<40:11,  2.67it/s]
 40%|███▉      | 4255/10682 [42:30<40:11,  2.67it/s]
 40%|███▉      | 4256/10682 [42:30<40:09,  2.67it/s]
 40%|███▉      | 4257/10682 [42:31<40:09,  2.67it/s]
 40%|███▉      | 4258/10682 [42:31<40:09,  2.67it/s]
 40%|███▉      | 4259/10682 [42:31<40:07,  2.67it/s]
 40%|███▉      | 4260/10682 [42:32<40:14,  2.66it/s]
 40%|███▉      | 4261/10682 [42:32<40:13,  2.66it/s]
 40%|███▉      | 4262/10682 [42:33<40:13,  2.66it/s]
 40%|███▉      | 4263/10682 [42:33<40:09,  2.66it/s]
 40%|███▉      | 4264/10682 [42:33<40:08,  2.66it/s]
 40%|███▉      | 4265/10682 [42:34<40:06,  2.67it/s]
 40%|███▉      | 4266/10682 [42:34<40:05,  2.67it/s]
 40%|███▉      | 4267/10682 [42:34<40:09,  2.66it/s]
 40%|███▉      | 4268/10682 [42:35<40:10,  2.66it/s]
 40%|███▉      | 4269/10682 [42:35<40:08,  2.66it/s]
 40%|███▉      | 4270/10682 [42:36<40:06,  2.66it/s]
 40%|███▉      | 4271/10682 [42:36<40:06,  2.66it/s]
 40%|███▉      | 4272/10682 [42:36<40:03,  2.67it/s]
 40%|████      | 4273/10682 [42:37<40:06,  2.66it/s]
 40%|████      | 4274/10682 [42:37<40:07,  2.66it/s]
 40%|████      | 4275/10682 [42:37<40:05,  2.66it/s]
                                                    
{'loss': 4.3627, 'grad_norm': 0.4409235715866089, 'learning_rate': 0.0007497641105186759, 'epoch': 0.4}
-
 40%|████      | 4275/10682 [42:37<40:05,  2.66it/s]
 40%|████      | 4276/10682 [42:38<40:08,  2.66it/s]
 40%|████      | 4277/10682 [42:38<40:02,  2.67it/s]
 40%|████      | 4278/10682 [42:39<40:01,  2.67it/s]
 40%|████      | 4279/10682 [42:39<39:58,  2.67it/s]
 40%|████      | 4280/10682 [42:39<39:59,  2.67it/s]
 40%|████      | 4281/10682 [42:40<39:59,  2.67it/s]
 40%|████      | 4282/10682 [42:40<39:58,  2.67it/s]
 40%|████      | 4283/10682 [42:40<39:58,  2.67it/s]
 40%|████      | 4284/10682 [42:41<40:02,  2.66it/s]
 40%|████      | 4285/10682 [42:41<39:58,  2.67it/s]
 40%|████      | 4286/10682 [42:42<39:57,  2.67it/s]
 40%|████      | 4287/10682 [42:42<40:01,  2.66it/s]
 40%|████      | 4288/10682 [42:42<39:57,  2.67it/s]
 40%|████      | 4289/10682 [42:43<39:58,  2.67it/s]
 40%|████      | 4290/10682 [42:43<39:58,  2.66it/s]
 40%|████      | 4291/10682 [42:43<39:55,  2.67it/s]
 40%|████      | 4292/10682 [42:44<39:52,  2.67it/s]
 40%|████      | 4293/10682 [42:44<39:51,  2.67it/s]
 40%|████      | 4294/10682 [42:45<39:54,  2.67it/s]
 40%|████      | 4295/10682 [42:45<39:51,  2.67it/s]
 40%|████      | 4296/10682 [42:45<39:52,  2.67it/s]
 40%|████      | 4297/10682 [42:46<39:50,  2.67it/s]
 40%|████      | 4298/10682 [42:46<39:53,  2.67it/s]
 40%|████      | 4299/10682 [42:46<39:51,  2.67it/s]
 40%|████      | 4300/10682 [42:47<39:53,  2.67it/s]{'loss': 4.3603, 'grad_norm': 0.4029136002063751, 'learning_rate': 0.0007462169157217631, 'epoch': 0.4}
                                                    
-
 40%|████      | 4300/10682 [42:47<39:53,  2.67it/s]
 40%|████      | 4301/10682 [42:47<40:03,  2.65it/s]
 40%|████      | 4302/10682 [42:48<40:01,  2.66it/s]
 40%|████      | 4303/10682 [42:48<39:59,  2.66it/s]
 40%|████      | 4304/10682 [42:48<39:57,  2.66it/s]
 40%|████      | 4305/10682 [42:49<39:54,  2.66it/s]
 40%|████      | 4306/10682 [42:49<39:52,  2.66it/s]
 40%|████      | 4307/10682 [42:49<39:53,  2.66it/s]
 40%|████      | 4308/10682 [42:50<39:50,  2.67it/s]
 40%|████      | 4309/10682 [42:50<39:47,  2.67it/s]
 40%|████      | 4310/10682 [42:51<39:44,  2.67it/s]
 40%|████      | 4311/10682 [42:51<39:47,  2.67it/s]
 40%|████      | 4312/10682 [42:51<39:46,  2.67it/s]
 40%|████      | 4313/10682 [42:52<39:46,  2.67it/s]
 40%|████      | 4314/10682 [42:52<39:48,  2.67it/s]
 40%|████      | 4315/10682 [42:52<39:45,  2.67it/s]
 40%|████      | 4316/10682 [42:53<39:49,  2.66it/s]
 40%|████      | 4317/10682 [42:53<39:50,  2.66it/s]
 40%|████      | 4318/10682 [42:54<39:47,  2.67it/s]
 40%|████      | 4319/10682 [42:54<39:45,  2.67it/s]
 40%|████      | 4320/10682 [42:54<39:48,  2.66it/s]
 40%|████      | 4321/10682 [42:55<39:48,  2.66it/s]
 40%|████      | 4322/10682 [42:55<39:44,  2.67it/s]
 40%|████      | 4323/10682 [42:55<39:43,  2.67it/s]
 40%|████      | 4324/10682 [42:56<39:44,  2.67it/s]
 40%|████      | 4325/10682 [42:56<39:44,  2.67it/s]{'loss': 4.3591, 'grad_norm': 0.46767646074295044, 'learning_rate': 0.0007426532856357045, 'epoch': 0.4}
                                                    
-
 40%|████      | 4325/10682 [42:56<39:44,  2.67it/s]
 40%|████      | 4326/10682 [42:57<39:55,  2.65it/s]
 41%|████      | 4327/10682 [42:57<39:56,  2.65it/s]
 41%|████      | 4328/10682 [42:57<39:52,  2.66it/s]
 41%|████      | 4329/10682 [42:58<39:50,  2.66it/s]
 41%|████      | 4330/10682 [42:58<39:47,  2.66it/s]
 41%|████      | 4331/10682 [42:58<39:43,  2.66it/s]
 41%|████      | 4332/10682 [42:59<39:41,  2.67it/s]
 41%|████      | 4333/10682 [42:59<39:40,  2.67it/s]
 41%|████      | 4334/10682 [43:00<39:42,  2.66it/s]
 41%|████      | 4335/10682 [43:00<39:40,  2.67it/s]
 41%|████      | 4336/10682 [43:00<39:37,  2.67it/s]
 41%|████      | 4337/10682 [43:01<39:37,  2.67it/s]
 41%|████      | 4338/10682 [43:01<39:36,  2.67it/s]
 41%|████      | 4339/10682 [43:01<39:34,  2.67it/s]
 41%|████      | 4340/10682 [43:02<39:35,  2.67it/s]
 41%|████      | 4341/10682 [43:02<39:36,  2.67it/s]
 41%|████      | 4342/10682 [43:03<39:35,  2.67it/s]
 41%|████      | 4343/10682 [43:03<39:38,  2.67it/s]
 41%|████      | 4344/10682 [43:03<39:37,  2.67it/s]
 41%|████      | 4345/10682 [43:04<39:36,  2.67it/s]
 41%|████      | 4346/10682 [43:04<39:36,  2.67it/s]
 41%|████      | 4347/10682 [43:04<39:34,  2.67it/s]
 41%|████      | 4348/10682 [43:05<39:33,  2.67it/s]
 41%|████      | 4349/10682 [43:05<39:33,  2.67it/s]
 41%|████      | 4350/10682 [43:06<39:33,  2.67it/s]
                                                    {'loss': 4.3523, 'grad_norm': 0.40910473465919495, 'learning_rate': 0.0007390734581372948, 'epoch': 0.41}
-

 41%|████      | 4350/10682 [43:06<39:33,  2.67it/s]
 41%|████      | 4351/10682 [43:06<39:37,  2.66it/s]
 41%|████      | 4352/10682 [43:06<39:34,  2.67it/s]
 41%|████      | 4353/10682 [43:07<39:32,  2.67it/s]
 41%|████      | 4354/10682 [43:07<39:33,  2.67it/s]
 41%|████      | 4355/10682 [43:07<39:32,  2.67it/s]
 41%|████      | 4356/10682 [43:08<39:35,  2.66it/s]
 41%|████      | 4357/10682 [43:08<39:35,  2.66it/s]
 41%|████      | 4358/10682 [43:09<39:35,  2.66it/s]
 41%|████      | 4359/10682 [43:09<39:33,  2.66it/s]
 41%|████      | 4360/10682 [43:09<39:32,  2.66it/s]
 41%|████      | 4361/10682 [43:10<39:34,  2.66it/s]
 41%|████      | 4362/10682 [43:10<39:32,  2.66it/s]
 41%|████      | 4363/10682 [43:10<39:29,  2.67it/s]
 41%|████      | 4364/10682 [43:11<39:28,  2.67it/s]
 41%|████      | 4365/10682 [43:11<39:29,  2.67it/s]
 41%|████      | 4366/10682 [43:12<39:26,  2.67it/s]
 41%|████      | 4367/10682 [43:12<39:31,  2.66it/s]
 41%|████      | 4368/10682 [43:12<39:29,  2.66it/s]
 41%|████      | 4369/10682 [43:13<39:26,  2.67it/s]
 41%|████      | 4370/10682 [43:13<39:29,  2.66it/s]
 41%|████      | 4371/10682 [43:13<39:28,  2.67it/s]
 41%|████      | 4372/10682 [43:14<39:25,  2.67it/s]
 41%|████      | 4373/10682 [43:14<39:25,  2.67it/s]
 41%|████      | 4374/10682 [43:15<39:26,  2.67it/s]
 41%|████      | 4375/10682 [43:15<39:24,  2.67it/s]{'loss': 4.3567, 'grad_norm': 0.41932207345962524, 'learning_rate': 0.0007354776721845277, 'epoch': 0.41}
-
                                                    

 41%|████      | 4375/10682 [43:15<39:24,  2.67it/s]
 41%|████      | 4376/10682 [43:15<39:26,  2.67it/s]
 41%|████      | 4377/10682 [43:16<39:24,  2.67it/s]
 41%|████      | 4378/10682 [43:16<39:23,  2.67it/s]
 41%|████      | 4379/10682 [43:16<39:21,  2.67it/s]
 41%|████      | 4380/10682 [43:17<39:20,  2.67it/s]
 41%|████      | 4381/10682 [43:17<39:25,  2.66it/s]
 41%|████      | 4382/10682 [43:18<39:24,  2.66it/s]
 41%|████      | 4383/10682 [43:18<39:24,  2.66it/s]
 41%|████      | 4384/10682 [43:18<39:25,  2.66it/s]
 41%|████      | 4385/10682 [43:19<39:27,  2.66it/s]
 41%|████      | 4386/10682 [43:19<39:25,  2.66it/s]
 41%|████      | 4387/10682 [43:19<39:24,  2.66it/s]
 41%|████      | 4388/10682 [43:20<39:24,  2.66it/s]
 41%|████      | 4389/10682 [43:20<39:23,  2.66it/s]
 41%|████      | 4390/10682 [43:21<39:22,  2.66it/s]
 41%|████      | 4391/10682 [43:21<39:22,  2.66it/s]
 41%|████      | 4392/10682 [43:21<39:22,  2.66it/s]
 41%|████      | 4393/10682 [43:22<39:18,  2.67it/s]
 41%|████      | 4394/10682 [43:22<39:22,  2.66it/s]
 41%|████      | 4395/10682 [43:22<39:20,  2.66it/s]
 41%|████      | 4396/10682 [43:23<39:17,  2.67it/s]
 41%|████      | 4397/10682 [43:23<39:20,  2.66it/s]
 41%|████      | 4398/10682 [43:24<39:23,  2.66it/s]
 41%|████      | 4399/10682 [43:24<39:22,  2.66it/s]
 41%|████      | 4400/10682 [43:24<39:22,  2.66it/s]{'loss': 4.3556, 'grad_norm': 0.4824904799461365, 'learning_rate': 0.0007318661678006427, 'epoch': 0.41}
                                                    
-
 41%|████      | 4400/10682 [43:24<39:22,  2.66it/s]
 41%|████      | 4401/10682 [43:25<39:29,  2.65it/s]
 41%|████      | 4402/10682 [43:25<39:24,  2.66it/s]
 41%|████      | 4403/10682 [43:25<39:19,  2.66it/s]
 41%|████      | 4404/10682 [43:26<39:17,  2.66it/s]
 41%|████      | 4405/10682 [43:26<39:16,  2.66it/s]
 41%|████      | 4406/10682 [43:27<39:14,  2.67it/s]
 41%|████▏     | 4407/10682 [43:27<39:14,  2.67it/s]
 41%|████▏     | 4408/10682 [43:27<39:18,  2.66it/s]
 41%|████▏     | 4409/10682 [43:28<39:16,  2.66it/s]
 41%|████▏     | 4410/10682 [43:28<39:15,  2.66it/s]
 41%|████▏     | 4411/10682 [43:28<39:15,  2.66it/s]
 41%|████▏     | 4412/10682 [43:29<39:15,  2.66it/s]
 41%|████▏     | 4413/10682 [43:29<39:15,  2.66it/s]
 41%|████▏     | 4414/10682 [43:30<39:11,  2.67it/s]
 41%|████▏     | 4415/10682 [43:30<39:12,  2.66it/s]
 41%|████▏     | 4416/10682 [43:30<39:11,  2.66it/s]
 41%|████▏     | 4417/10682 [43:31<39:10,  2.67it/s]
 41%|████▏     | 4418/10682 [43:31<39:07,  2.67it/s]
 41%|████▏     | 4419/10682 [43:31<39:07,  2.67it/s]
 41%|████▏     | 4420/10682 [43:32<39:04,  2.67it/s]
 41%|████▏     | 4421/10682 [43:32<39:02,  2.67it/s]
 41%|████▏     | 4422/10682 [43:33<39:07,  2.67it/s]
 41%|████▏     | 4423/10682 [43:33<39:07,  2.67it/s]
 41%|████▏     | 4424/10682 [43:33<39:19,  2.65it/s]
 41%|████▏     | 4425/10682 [43:34<39:12,  2.66it/s]
                                                    
{'loss': 4.3537, 'grad_norm': 0.4319685697555542, 'learning_rate': 0.0007282391860581049, 'epoch': 0.41}
-
 41%|████▏     | 4425/10682 [43:34<39:12,  2.66it/s]
 41%|████▏     | 4426/10682 [43:34<39:11,  2.66it/s]
 41%|████▏     | 4427/10682 [43:34<39:09,  2.66it/s]
 41%|████▏     | 4428/10682 [43:35<39:06,  2.66it/s]
 41%|████▏     | 4429/10682 [43:35<39:03,  2.67it/s]
 41%|████▏     | 4430/10682 [43:36<39:02,  2.67it/s]
 41%|████▏     | 4431/10682 [43:36<39:02,  2.67it/s]
 41%|████▏     | 4432/10682 [43:36<39:03,  2.67it/s]
 41%|████▏     | 4433/10682 [43:37<39:02,  2.67it/s]
 42%|████▏     | 4434/10682 [43:37<39:05,  2.66it/s]
 42%|████▏     | 4435/10682 [43:37<39:04,  2.66it/s]
 42%|████▏     | 4436/10682 [43:38<39:07,  2.66it/s]
 42%|████▏     | 4437/10682 [43:38<39:03,  2.66it/s]
 42%|████▏     | 4438/10682 [43:39<39:07,  2.66it/s]
 42%|████▏     | 4439/10682 [43:39<39:03,  2.66it/s]
 42%|████▏     | 4440/10682 [43:39<39:00,  2.67it/s]
 42%|████▏     | 4441/10682 [43:40<38:58,  2.67it/s]
 42%|████▏     | 4442/10682 [43:40<38:57,  2.67it/s]
 42%|████▏     | 4443/10682 [43:40<38:59,  2.67it/s]
 42%|████▏     | 4444/10682 [43:41<39:00,  2.66it/s]
 42%|████▏     | 4445/10682 [43:41<38:57,  2.67it/s]
 42%|████▏     | 4446/10682 [43:42<38:55,  2.67it/s]
 42%|████▏     | 4447/10682 [43:42<38:55,  2.67it/s]
 42%|████▏     | 4448/10682 [43:42<38:54,  2.67it/s]
 42%|████▏     | 4449/10682 [43:43<38:57,  2.67it/s]
 42%|████▏     | 4450/10682 [43:43<38:55,  2.67it/s]
                                                    
{'loss': 4.3498, 'grad_norm': 0.5200799703598022, 'learning_rate': 0.0007245969690625126, 'epoch': 0.42}
-
 42%|████▏     | 4450/10682 [43:43<38:55,  2.67it/s]
 42%|████▏     | 4451/10682 [43:43<39:02,  2.66it/s]
 42%|████▏     | 4452/10682 [43:44<39:03,  2.66it/s]
 42%|████▏     | 4453/10682 [43:44<39:05,  2.66it/s]
 42%|████▏     | 4454/10682 [43:45<39:03,  2.66it/s]
 42%|████▏     | 4455/10682 [43:45<38:58,  2.66it/s]
 42%|████▏     | 4456/10682 [43:45<39:00,  2.66it/s]
 42%|████▏     | 4457/10682 [43:46<39:01,  2.66it/s]
 42%|████▏     | 4458/10682 [43:46<38:57,  2.66it/s]
 42%|████▏     | 4459/10682 [43:46<38:56,  2.66it/s]
 42%|████▏     | 4460/10682 [43:47<38:55,  2.66it/s]
 42%|████▏     | 4461/10682 [43:47<38:53,  2.67it/s]
 42%|████▏     | 4462/10682 [43:48<38:51,  2.67it/s]
 42%|████▏     | 4463/10682 [43:48<38:56,  2.66it/s]
 42%|████▏     | 4464/10682 [43:48<38:55,  2.66it/s]
 42%|████▏     | 4465/10682 [43:49<39:02,  2.65it/s]
 42%|████▏     | 4466/10682 [43:49<38:56,  2.66it/s]
 42%|████▏     | 4467/10682 [43:49<38:53,  2.66it/s]
 42%|████▏     | 4468/10682 [43:50<38:52,  2.66it/s]
 42%|████▏     | 4469/10682 [43:50<38:55,  2.66it/s]
 42%|████▏     | 4470/10682 [43:51<38:51,  2.66it/s]
 42%|████▏     | 4471/10682 [43:51<38:50,  2.66it/s]
 42%|████▏     | 4472/10682 [43:51<38:48,  2.67it/s]
 42%|████▏     | 4473/10682 [43:52<38:46,  2.67it/s]
 42%|████▏     | 4474/10682 [43:52<38:48,  2.67it/s]
 42%|████▏     | 4475/10682 [43:52<38:47,  2.67it/s]{'loss': 4.3499, 'grad_norm': 0.4230349063873291, 'learning_rate': 0.0007209397599364363, 'epoch': 0.42}
                                                    
-
 42%|████▏     | 4475/10682 [43:52<38:47,  2.67it/s]
 42%|████▏     | 4476/10682 [43:53<39:00,  2.65it/s]
 42%|████▏     | 4477/10682 [43:53<38:54,  2.66it/s]
 42%|████▏     | 4478/10682 [43:54<38:52,  2.66it/s]
 42%|████▏     | 4479/10682 [43:54<38:53,  2.66it/s]
 42%|████▏     | 4480/10682 [43:54<38:52,  2.66it/s]
 42%|████▏     | 4481/10682 [43:55<38:49,  2.66it/s]
 42%|████▏     | 4482/10682 [43:55<38:44,  2.67it/s]
 42%|████▏     | 4483/10682 [43:55<38:48,  2.66it/s]
 42%|████▏     | 4484/10682 [43:56<38:45,  2.67it/s]
 42%|████▏     | 4485/10682 [43:56<38:42,  2.67it/s]
 42%|████▏     | 4486/10682 [43:57<38:40,  2.67it/s]
 42%|████▏     | 4487/10682 [43:57<38:39,  2.67it/s]
 42%|████▏     | 4488/10682 [43:57<38:41,  2.67it/s]
 42%|████▏     | 4489/10682 [43:58<38:40,  2.67it/s]
 42%|████▏     | 4490/10682 [43:58<38:44,  2.66it/s]
 42%|████▏     | 4491/10682 [43:58<38:44,  2.66it/s]
 42%|████▏     | 4492/10682 [43:59<38:43,  2.66it/s]
 42%|████▏     | 4493/10682 [43:59<38:42,  2.66it/s]
 42%|████▏     | 4494/10682 [44:00<38:42,  2.66it/s]
 42%|████▏     | 4495/10682 [44:00<38:41,  2.66it/s]
 42%|████▏     | 4496/10682 [44:00<38:41,  2.67it/s]
 42%|████▏     | 4497/10682 [44:01<38:39,  2.67it/s]
 42%|████▏     | 4498/10682 [44:01<38:38,  2.67it/s]
 42%|████▏     | 4499/10682 [44:01<38:37,  2.67it/s]
 42%|████▏     | 4500/10682 [44:02<38:35,  2.67it/s]
                                                    
{'loss': 4.3514, 'grad_norm': 0.45323652029037476, 'learning_rate': 0.0007172678028031901, 'epoch': 0.42}
-
 42%|████▏     | 4500/10682 [44:02<38:35,  2.67it/s]
 42%|████▏     | 4501/10682 [44:02<38:41,  2.66it/s]
 42%|████▏     | 4502/10682 [44:03<38:37,  2.67it/s]
 42%|████▏     | 4503/10682 [44:03<38:36,  2.67it/s]
 42%|████▏     | 4504/10682 [44:03<38:36,  2.67it/s]
 42%|████▏     | 4505/10682 [44:04<38:34,  2.67it/s]
 42%|████▏     | 4506/10682 [44:04<38:40,  2.66it/s]
 42%|████▏     | 4507/10682 [44:04<38:39,  2.66it/s]
 42%|████▏     | 4508/10682 [44:05<38:37,  2.66it/s]
 42%|████▏     | 4509/10682 [44:05<38:34,  2.67it/s]
 42%|████▏     | 4510/10682 [44:06<38:32,  2.67it/s]
 42%|████▏     | 4511/10682 [44:06<38:32,  2.67it/s]
 42%|████▏     | 4512/10682 [44:06<38:31,  2.67it/s]
 42%|████▏     | 4513/10682 [44:07<38:31,  2.67it/s]
 42%|████▏     | 4514/10682 [44:07<38:29,  2.67it/s]
 42%|████▏     | 4515/10682 [44:07<38:31,  2.67it/s]
 42%|████▏     | 4516/10682 [44:08<38:29,  2.67it/s]
 42%|████▏     | 4517/10682 [44:08<38:30,  2.67it/s]
 42%|████▏     | 4518/10682 [44:09<38:32,  2.67it/s]
 42%|████▏     | 4519/10682 [44:09<38:31,  2.67it/s]
 42%|████▏     | 4520/10682 [44:09<38:33,  2.66it/s]
 42%|████▏     | 4521/10682 [44:10<38:30,  2.67it/s]
 42%|████▏     | 4522/10682 [44:10<38:28,  2.67it/s]
 42%|████▏     | 4523/10682 [44:10<38:29,  2.67it/s]
 42%|████▏     | 4524/10682 [44:11<38:31,  2.66it/s]
 42%|████▏     | 4525/10682 [44:11<38:28,  2.67it/s]
                                                    
{'loss': 4.3444, 'grad_norm': 0.3843716084957123, 'learning_rate': 0.0007135813427705359, 'epoch': 0.42}
-
 42%|████▏     | 4525/10682 [44:11<38:28,  2.67it/s]
 42%|████▏     | 4526/10682 [44:12<38:29,  2.67it/s]
 42%|████▏     | 4527/10682 [44:12<38:26,  2.67it/s]
 42%|████▏     | 4528/10682 [44:12<38:25,  2.67it/s]
 42%|████▏     | 4529/10682 [44:13<38:23,  2.67it/s]
 42%|████▏     | 4530/10682 [44:13<38:26,  2.67it/s]
 42%|████▏     | 4531/10682 [44:13<38:29,  2.66it/s]
 42%|████▏     | 4532/10682 [44:14<38:26,  2.67it/s]
 42%|████▏     | 4533/10682 [44:14<38:26,  2.67it/s]
 42%|████▏     | 4534/10682 [44:15<38:28,  2.66it/s]
 42%|████▏     | 4535/10682 [44:15<38:28,  2.66it/s]
 42%|████▏     | 4536/10682 [44:15<38:28,  2.66it/s]
 42%|████▏     | 4537/10682 [44:16<38:28,  2.66it/s]
 42%|████▏     | 4538/10682 [44:16<38:28,  2.66it/s]
 42%|████▏     | 4539/10682 [44:16<38:25,  2.66it/s]
 43%|████▎     | 4540/10682 [44:17<38:24,  2.66it/s]
 43%|████▎     | 4541/10682 [44:17<38:23,  2.67it/s]
 43%|████▎     | 4542/10682 [44:18<38:24,  2.66it/s]
 43%|████▎     | 4543/10682 [44:18<38:22,  2.67it/s]
 43%|████▎     | 4544/10682 [44:18<38:26,  2.66it/s]
 43%|████▎     | 4545/10682 [44:19<38:25,  2.66it/s]
 43%|████▎     | 4546/10682 [44:19<38:21,  2.67it/s]
 43%|████▎     | 4547/10682 [44:19<38:26,  2.66it/s]
 43%|████▎     | 4548/10682 [44:20<38:21,  2.67it/s]
 43%|████▎     | 4549/10682 [44:20<38:20,  2.67it/s]
 43%|████▎     | 4550/10682 [44:21<38:18,  2.67it/s]{'loss': 4.3447, 'grad_norm': 0.4108799397945404, 'learning_rate': 0.0007098806259143232, 'epoch': 0.43}
                                                    
-
 43%|████▎     | 4550/10682 [44:21<38:18,  2.67it/s]
 43%|████▎     | 4551/10682 [44:21<38:30,  2.65it/s]
 43%|████▎     | 4552/10682 [44:21<38:27,  2.66it/s]
 43%|████▎     | 4553/10682 [44:22<38:25,  2.66it/s]
 43%|████▎     | 4554/10682 [44:22<38:22,  2.66it/s]
 43%|████▎     | 4555/10682 [44:22<38:19,  2.66it/s]
 43%|████▎     | 4556/10682 [44:23<38:18,  2.67it/s]
 43%|████▎     | 4557/10682 [44:23<38:17,  2.67it/s]
 43%|████▎     | 4558/10682 [44:24<38:22,  2.66it/s]
 43%|████▎     | 4559/10682 [44:24<38:20,  2.66it/s]
 43%|████▎     | 4560/10682 [44:24<38:23,  2.66it/s]
 43%|████▎     | 4561/10682 [44:25<38:23,  2.66it/s]
 43%|████▎     | 4562/10682 [44:25<38:20,  2.66it/s]
 43%|████▎     | 4563/10682 [44:25<38:17,  2.66it/s]
 43%|████▎     | 4564/10682 [44:26<38:16,  2.66it/s]
 43%|████▎     | 4565/10682 [44:26<38:15,  2.66it/s]
 43%|████▎     | 4566/10682 [44:27<38:12,  2.67it/s]
 43%|████▎     | 4567/10682 [44:27<38:14,  2.67it/s]
 43%|████▎     | 4568/10682 [44:27<38:11,  2.67it/s]
 43%|████▎     | 4569/10682 [44:28<38:12,  2.67it/s]
 43%|████▎     | 4570/10682 [44:28<38:11,  2.67it/s]
 43%|████▎     | 4571/10682 [44:28<38:10,  2.67it/s]
 43%|████▎     | 4572/10682 [44:29<38:13,  2.66it/s]
 43%|████▎     | 4573/10682 [44:29<38:13,  2.66it/s]
 43%|████▎     | 4574/10682 [44:30<38:17,  2.66it/s]
 43%|████▎     | 4575/10682 [44:30<38:15,  2.66it/s]{'loss': 4.3406, 'grad_norm': 0.4273871183395386, 'learning_rate': 0.0007061658992620613, 'epoch': 0.43}
                                                    
-
 43%|████▎     | 4575/10682 [44:30<38:15,  2.66it/s]
 43%|████▎     | 4576/10682 [44:30<38:16,  2.66it/s]
 43%|████▎     | 4577/10682 [44:31<38:13,  2.66it/s]
 43%|████▎     | 4578/10682 [44:31<38:12,  2.66it/s]
 43%|████▎     | 4579/10682 [44:31<38:13,  2.66it/s]
 43%|████▎     | 4580/10682 [44:32<38:13,  2.66it/s]
 43%|████▎     | 4581/10682 [44:32<38:11,  2.66it/s]
 43%|████▎     | 4582/10682 [44:33<38:09,  2.66it/s]
 43%|████▎     | 4583/10682 [44:33<38:07,  2.67it/s]
 43%|████▎     | 4584/10682 [44:33<38:05,  2.67it/s]
 43%|████▎     | 4585/10682 [44:34<38:05,  2.67it/s]
 43%|████▎     | 4586/10682 [44:34<38:05,  2.67it/s]
 43%|████▎     | 4587/10682 [44:34<38:05,  2.67it/s]
 43%|████▎     | 4588/10682 [44:35<38:08,  2.66it/s]
 43%|████▎     | 4589/10682 [44:35<38:08,  2.66it/s]
 43%|████▎     | 4590/10682 [44:36<38:06,  2.66it/s]
 43%|████▎     | 4591/10682 [44:36<38:04,  2.67it/s]
 43%|████▎     | 4592/10682 [44:36<38:03,  2.67it/s]
 43%|████▎     | 4593/10682 [44:37<38:01,  2.67it/s]
 43%|████▎     | 4594/10682 [44:37<38:02,  2.67it/s]
 43%|████▎     | 4595/10682 [44:37<37:59,  2.67it/s]
 43%|████▎     | 4596/10682 [44:38<38:02,  2.67it/s]
 43%|████▎     | 4597/10682 [44:38<38:01,  2.67it/s]
 43%|████▎     | 4598/10682 [44:39<38:03,  2.66it/s]
 43%|████▎     | 4599/10682 [44:39<38:07,  2.66it/s]
 43%|████▎     | 4600/10682 [44:39<38:06,  2.66it/s]{'loss': 4.3358, 'grad_norm': 0.46230849623680115, 'learning_rate': 0.0007024374107764318, 'epoch': 0.43}
                                                    
-
 43%|████▎     | 4600/10682 [44:39<38:06,  2.66it/s]
 43%|████▎     | 4601/10682 [44:40<38:11,  2.65it/s]
 43%|████▎     | 4602/10682 [44:40<38:09,  2.66it/s]
 43%|████▎     | 4603/10682 [44:41<38:10,  2.65it/s]
 43%|████▎     | 4604/10682 [44:41<38:06,  2.66it/s]
 43%|████▎     | 4605/10682 [44:41<38:02,  2.66it/s]
 43%|████▎     | 4606/10682 [44:42<38:02,  2.66it/s]
 43%|████▎     | 4607/10682 [44:42<38:01,  2.66it/s]
 43%|████▎     | 4608/10682 [44:42<37:57,  2.67it/s]
 43%|████▎     | 4609/10682 [44:43<37:57,  2.67it/s]
 43%|████▎     | 4610/10682 [44:43<37:57,  2.67it/s]
 43%|████▎     | 4611/10682 [44:44<37:55,  2.67it/s]
 43%|████▎     | 4612/10682 [44:44<37:56,  2.67it/s]
 43%|████▎     | 4613/10682 [44:44<37:57,  2.66it/s]
 43%|████▎     | 4614/10682 [44:45<46:17,  2.19it/s]
 43%|████▎     | 4615/10682 [44:45<43:50,  2.31it/s]
 43%|████▎     | 4616/10682 [44:46<42:05,  2.40it/s]
 43%|████▎     | 4617/10682 [44:46<40:52,  2.47it/s]
 43%|████▎     | 4618/10682 [44:46<40:01,  2.53it/s]
 43%|████▎     | 4619/10682 [44:47<39:26,  2.56it/s]
 43%|████▎     | 4620/10682 [44:47<38:55,  2.60it/s]
 43%|████▎     | 4621/10682 [44:48<38:49,  2.60it/s]
 43%|████▎     | 4622/10682 [44:48<38:32,  2.62it/s]
 43%|████▎     | 4623/10682 [44:48<38:21,  2.63it/s]
 43%|████▎     | 4624/10682 [44:49<38:10,  2.64it/s]
 43%|████▎     | 4625/10682 [44:49<38:07,  2.65it/s]{'loss': 4.3455, 'grad_norm': 0.43842563033103943, 'learning_rate': 0.0006986954093387361, 'epoch': 0.43}
                                                    
-
 43%|████▎     | 4625/10682 [44:49<38:07,  2.65it/s]
 43%|████▎     | 4626/10682 [44:49<38:02,  2.65it/s]
 43%|████▎     | 4627/10682 [44:50<37:59,  2.66it/s]
 43%|████▎     | 4628/10682 [44:50<37:57,  2.66it/s]
 43%|████▎     | 4629/10682 [44:51<37:54,  2.66it/s]
 43%|████▎     | 4630/10682 [44:51<37:53,  2.66it/s]
 43%|████▎     | 4631/10682 [44:51<37:51,  2.66it/s]
 43%|████▎     | 4632/10682 [44:52<37:52,  2.66it/s]
 43%|████▎     | 4633/10682 [44:52<46:14,  2.18it/s]
 43%|████▎     | 4634/10682 [44:53<43:42,  2.31it/s]
 43%|████▎     | 4635/10682 [44:53<41:53,  2.41it/s]
 43%|████▎     | 4636/10682 [44:53<40:39,  2.48it/s]
 43%|████▎     | 4637/10682 [44:54<39:44,  2.54it/s]
 43%|████▎     | 4638/10682 [44:54<39:18,  2.56it/s]
 43%|████▎     | 4639/10682 [44:55<38:50,  2.59it/s]
 43%|████▎     | 4640/10682 [44:55<38:28,  2.62it/s]
 43%|████▎     | 4641/10682 [44:55<38:14,  2.63it/s]
 43%|████▎     | 4642/10682 [44:56<38:07,  2.64it/s]
 43%|████▎     | 4643/10682 [44:56<38:00,  2.65it/s]
 43%|████▎     | 4644/10682 [44:56<37:52,  2.66it/s]
 43%|████▎     | 4645/10682 [44:57<37:48,  2.66it/s]
 43%|████▎     | 4646/10682 [44:57<37:43,  2.67it/s]
 44%|████▎     | 4647/10682 [44:58<37:42,  2.67it/s]
 44%|████▎     | 4648/10682 [44:58<37:40,  2.67it/s]
 44%|████▎     | 4649/10682 [44:58<37:39,  2.67it/s]
 44%|████▎     | 4650/10682 [44:59<37:38,  2.67it/s]{'loss': 4.3341, 'grad_norm': 0.5031459331512451, 'learning_rate': 0.0006949401447322816, 'epoch': 0.44}
-
                                                    

 44%|████▎     | 4650/10682 [44:59<37:38,  2.67it/s]
 44%|████▎     | 4651/10682 [44:59<37:39,  2.67it/s]
 44%|████▎     | 4652/10682 [44:59<37:38,  2.67it/s]
 44%|████▎     | 4653/10682 [45:00<37:36,  2.67it/s]
 44%|████▎     | 4654/10682 [45:00<37:38,  2.67it/s]
 44%|████▎     | 4655/10682 [45:01<37:38,  2.67it/s]
 44%|████▎     | 4656/10682 [45:01<37:40,  2.67it/s]
 44%|████▎     | 4657/10682 [45:01<37:41,  2.66it/s]
 44%|████▎     | 4658/10682 [45:02<37:38,  2.67it/s]
 44%|████▎     | 4659/10682 [45:02<37:35,  2.67it/s]
 44%|████▎     | 4660/10682 [45:02<37:35,  2.67it/s]
 44%|████▎     | 4661/10682 [45:03<37:36,  2.67it/s]
 44%|████▎     | 4662/10682 [45:03<37:35,  2.67it/s]
 44%|████▎     | 4663/10682 [45:04<37:33,  2.67it/s]
 44%|████▎     | 4664/10682 [45:04<37:30,  2.67it/s]
 44%|████▎     | 4665/10682 [45:04<37:30,  2.67it/s]
 44%|████▎     | 4666/10682 [45:05<37:29,  2.67it/s]
 44%|████▎     | 4667/10682 [45:05<37:33,  2.67it/s]
 44%|████▎     | 4668/10682 [45:05<37:32,  2.67it/s]
 44%|████▎     | 4669/10682 [45:06<37:33,  2.67it/s]
 44%|████▎     | 4670/10682 [45:06<37:33,  2.67it/s]
 44%|████▎     | 4671/10682 [45:07<37:31,  2.67it/s]
 44%|████▎     | 4672/10682 [45:07<37:30,  2.67it/s]
 44%|████▎     | 4673/10682 [45:07<37:28,  2.67it/s]
 44%|████▍     | 4674/10682 [45:08<37:28,  2.67it/s]
 44%|████▍     | 4675/10682 [45:08<37:26,  2.67it/s]{'loss': 4.3365, 'grad_norm': 0.4305307865142822, 'learning_rate': 0.0006911718676257091, 'epoch': 0.44}
-
                                                    

 44%|████▍     | 4675/10682 [45:08<37:26,  2.67it/s]
 44%|████▍     | 4676/10682 [45:08<37:28,  2.67it/s]
 44%|████▍     | 4677/10682 [45:09<37:28,  2.67it/s]
 44%|████▍     | 4678/10682 [45:09<37:28,  2.67it/s]
 44%|████▍     | 4679/10682 [45:10<37:25,  2.67it/s]
 44%|████▍     | 4680/10682 [45:10<37:27,  2.67it/s]
 44%|████▍     | 4681/10682 [45:10<37:28,  2.67it/s]
 44%|████▍     | 4682/10682 [45:11<37:25,  2.67it/s]
 44%|████▍     | 4683/10682 [45:11<37:30,  2.67it/s]
 44%|████▍     | 4684/10682 [45:11<37:27,  2.67it/s]
 44%|████▍     | 4685/10682 [45:12<37:23,  2.67it/s]
 44%|████▍     | 4686/10682 [45:12<37:23,  2.67it/s]
 44%|████▍     | 4687/10682 [45:13<37:25,  2.67it/s]
 44%|████▍     | 4688/10682 [45:13<37:23,  2.67it/s]
 44%|████▍     | 4689/10682 [45:13<37:24,  2.67it/s]
 44%|████▍     | 4690/10682 [45:14<37:24,  2.67it/s]
 44%|████▍     | 4691/10682 [45:14<37:21,  2.67it/s]
 44%|████▍     | 4692/10682 [45:14<37:22,  2.67it/s]
 44%|████▍     | 4693/10682 [45:15<37:22,  2.67it/s]
 44%|████▍     | 4694/10682 [45:15<37:24,  2.67it/s]
 44%|████▍     | 4695/10682 [45:16<37:26,  2.67it/s]
 44%|████▍     | 4696/10682 [45:16<37:30,  2.66it/s]
 44%|████▍     | 4697/10682 [45:16<37:26,  2.66it/s]
 44%|████▍     | 4698/10682 [45:17<37:25,  2.67it/s]
 44%|████▍     | 4699/10682 [45:17<37:23,  2.67it/s]
 44%|████▍     | 4700/10682 [45:17<37:22,  2.67it/s]
                                                    
{'loss': 4.3339, 'grad_norm': 0.4541700482368469, 'learning_rate': 0.0006873908295562602, 'epoch': 0.44}
-
 44%|████▍     | 4700/10682 [45:17<37:22,  2.67it/s]
 44%|████▍     | 4701/10682 [45:18<37:25,  2.66it/s]
 44%|████▍     | 4702/10682 [45:18<37:21,  2.67it/s]
 44%|████▍     | 4703/10682 [45:19<37:19,  2.67it/s]
 44%|████▍     | 4704/10682 [45:19<37:17,  2.67it/s]
 44%|████▍     | 4705/10682 [45:19<37:18,  2.67it/s]
 44%|████▍     | 4706/10682 [45:20<37:18,  2.67it/s]
 44%|████▍     | 4707/10682 [45:20<37:19,  2.67it/s]
 44%|████▍     | 4708/10682 [45:20<37:23,  2.66it/s]
 44%|████▍     | 4709/10682 [45:21<37:23,  2.66it/s]
 44%|████▍     | 4710/10682 [45:21<37:23,  2.66it/s]
 44%|████▍     | 4711/10682 [45:22<37:21,  2.66it/s]
 44%|████▍     | 4712/10682 [45:22<37:20,  2.66it/s]
 44%|████▍     | 4713/10682 [45:22<37:17,  2.67it/s]
 44%|████▍     | 4714/10682 [45:23<37:16,  2.67it/s]
 44%|████▍     | 4715/10682 [45:23<37:15,  2.67it/s]
 44%|████▍     | 4716/10682 [45:23<37:14,  2.67it/s]
 44%|████▍     | 4717/10682 [45:24<37:12,  2.67it/s]
 44%|████▍     | 4718/10682 [45:24<37:12,  2.67it/s]
 44%|████▍     | 4719/10682 [45:25<37:13,  2.67it/s]
 44%|████▍     | 4720/10682 [45:25<37:13,  2.67it/s]
 44%|████▍     | 4721/10682 [45:25<37:13,  2.67it/s]
 44%|████▍     | 4722/10682 [45:26<37:11,  2.67it/s]
 44%|████▍     | 4723/10682 [45:26<37:12,  2.67it/s]
 44%|████▍     | 4724/10682 [45:26<37:10,  2.67it/s]
 44%|████▍     | 4725/10682 [45:27<37:09,  2.67it/s]
                                                    
{'loss': 4.3376, 'grad_norm': 0.43473684787750244, 'learning_rate': 0.0006835972829129873, 'epoch': 0.44}
-
 44%|████▍     | 4725/10682 [45:27<37:09,  2.67it/s]
 44%|████▍     | 4726/10682 [45:27<37:12,  2.67it/s]
 44%|████▍     | 4727/10682 [45:28<37:11,  2.67it/s]
 44%|████▍     | 4728/10682 [45:28<37:09,  2.67it/s]
 44%|████▍     | 4729/10682 [45:28<37:07,  2.67it/s]
 44%|████▍     | 4730/10682 [45:29<37:06,  2.67it/s]
 44%|████▍     | 4731/10682 [45:29<37:05,  2.67it/s]
 44%|████▍     | 4732/10682 [45:29<37:04,  2.67it/s]
 44%|████▍     | 4733/10682 [45:30<37:08,  2.67it/s]
 44%|████▍     | 4734/10682 [45:30<37:09,  2.67it/s]
 44%|████▍     | 4735/10682 [45:31<37:06,  2.67it/s]
 44%|████▍     | 4736/10682 [45:31<37:08,  2.67it/s]
 44%|████▍     | 4737/10682 [45:31<37:07,  2.67it/s]
 44%|████▍     | 4738/10682 [45:32<37:07,  2.67it/s]
 44%|████▍     | 4739/10682 [45:32<37:05,  2.67it/s]
 44%|████▍     | 4740/10682 [45:32<37:06,  2.67it/s]
 44%|████▍     | 4741/10682 [45:33<37:05,  2.67it/s]
 44%|████▍     | 4742/10682 [45:33<37:02,  2.67it/s]
 44%|████▍     | 4743/10682 [45:34<37:02,  2.67it/s]
 44%|████▍     | 4744/10682 [45:34<37:00,  2.67it/s]
 44%|████▍     | 4745/10682 [45:34<37:02,  2.67it/s]
 44%|████▍     | 4746/10682 [45:35<37:02,  2.67it/s]
 44%|████▍     | 4747/10682 [45:35<37:04,  2.67it/s]
 44%|████▍     | 4748/10682 [45:35<37:02,  2.67it/s]
 44%|████▍     | 4749/10682 [45:36<37:01,  2.67it/s]
 44%|████▍     | 4750/10682 [45:36<37:02,  2.67it/s]{'loss': 4.3373, 'grad_norm': 0.4037042260169983, 'learning_rate': 0.000679791480919905, 'epoch': 0.44}
                                                    
-
 44%|████▍     | 4750/10682 [45:36<37:02,  2.67it/s]
 44%|████▍     | 4751/10682 [45:37<37:04,  2.67it/s]
 44%|████▍     | 4752/10682 [45:37<37:03,  2.67it/s]
 44%|████▍     | 4753/10682 [45:37<37:02,  2.67it/s]
 45%|████▍     | 4754/10682 [45:38<37:01,  2.67it/s]
 45%|████▍     | 4755/10682 [45:38<37:03,  2.67it/s]
 45%|████▍     | 4756/10682 [45:38<37:01,  2.67it/s]
 45%|████▍     | 4757/10682 [45:39<36:59,  2.67it/s]
 45%|████▍     | 4758/10682 [45:39<36:56,  2.67it/s]
 45%|████▍     | 4759/10682 [45:40<36:57,  2.67it/s]
 45%|████▍     | 4760/10682 [45:40<36:58,  2.67it/s]
 45%|████▍     | 4761/10682 [45:40<37:03,  2.66it/s]
 45%|████▍     | 4762/10682 [45:41<37:01,  2.66it/s]
 45%|████▍     | 4763/10682 [45:41<37:02,  2.66it/s]
 45%|████▍     | 4764/10682 [45:41<37:03,  2.66it/s]
 45%|████▍     | 4765/10682 [45:42<37:00,  2.67it/s]
 45%|████▍     | 4766/10682 [45:42<36:58,  2.67it/s]
 45%|████▍     | 4767/10682 [45:43<36:56,  2.67it/s]
 45%|████▍     | 4768/10682 [45:43<36:56,  2.67it/s]
 45%|████▍     | 4769/10682 [45:43<36:53,  2.67it/s]
 45%|████▍     | 4770/10682 [45:44<36:51,  2.67it/s]
 45%|████▍     | 4771/10682 [45:44<36:50,  2.67it/s]
 45%|████▍     | 4772/10682 [45:44<36:51,  2.67it/s]
 45%|████▍     | 4773/10682 [45:45<36:54,  2.67it/s]
 45%|████▍     | 4774/10682 [45:45<36:55,  2.67it/s]
 45%|████▍     | 4775/10682 [45:46<36:54,  2.67it/s]{'loss': 4.3361, 'grad_norm': 0.43911507725715637, 'learning_rate': 0.000675973677619089, 'epoch': 0.45}
-
                                                    

 45%|████▍     | 4775/10682 [45:46<36:54,  2.67it/s]
 45%|████▍     | 4776/10682 [45:46<36:58,  2.66it/s]
 45%|████▍     | 4777/10682 [45:46<36:55,  2.67it/s]
 45%|████▍     | 4778/10682 [45:47<36:55,  2.66it/s]
 45%|████▍     | 4779/10682 [45:47<36:53,  2.67it/s]
 45%|████▍     | 4780/10682 [45:47<36:53,  2.67it/s]
 45%|████▍     | 4781/10682 [45:48<36:51,  2.67it/s]
 45%|████▍     | 4782/10682 [45:48<36:53,  2.67it/s]
 45%|████▍     | 4783/10682 [45:49<36:52,  2.67it/s]
 45%|████▍     | 4784/10682 [45:49<36:50,  2.67it/s]
 45%|████▍     | 4785/10682 [45:49<36:48,  2.67it/s]
 45%|████▍     | 4786/10682 [45:50<36:48,  2.67it/s]
 45%|████▍     | 4787/10682 [45:50<36:47,  2.67it/s]
 45%|████▍     | 4788/10682 [45:50<36:51,  2.67it/s]
 45%|████▍     | 4789/10682 [45:51<36:49,  2.67it/s]
 45%|████▍     | 4790/10682 [45:51<36:51,  2.66it/s]
 45%|████▍     | 4791/10682 [45:52<36:50,  2.67it/s]
 45%|████▍     | 4792/10682 [45:52<36:47,  2.67it/s]
 45%|████▍     | 4793/10682 [45:52<36:45,  2.67it/s]
 45%|████▍     | 4794/10682 [45:53<36:45,  2.67it/s]
 45%|████▍     | 4795/10682 [45:53<36:44,  2.67it/s]
 45%|████▍     | 4796/10682 [45:53<36:43,  2.67it/s]
 45%|████▍     | 4797/10682 [45:54<36:41,  2.67it/s]
 45%|████▍     | 4798/10682 [45:54<36:42,  2.67it/s]
 45%|████▍     | 4799/10682 [45:55<36:40,  2.67it/s]
 45%|████▍     | 4800/10682 [45:55<36:43,  2.67it/s]
                                                    
{'loss': 4.3282, 'grad_norm': 0.4172917604446411, 'learning_rate': 0.0006721441278537161, 'epoch': 0.45}
-
 45%|████▍     | 4800/10682 [45:55<36:43,  2.67it/s]
 45%|████▍     | 4801/10682 [45:55<36:48,  2.66it/s]
 45%|████▍     | 4802/10682 [45:56<36:47,  2.66it/s]
 45%|████▍     | 4803/10682 [45:56<36:48,  2.66it/s]
 45%|████▍     | 4804/10682 [45:56<36:45,  2.67it/s]
 45%|████▍     | 4805/10682 [45:57<36:44,  2.67it/s]
 45%|████▍     | 4806/10682 [45:57<36:44,  2.67it/s]
 45%|████▌     | 4807/10682 [45:58<36:43,  2.67it/s]
 45%|████▌     | 4808/10682 [45:58<36:42,  2.67it/s]
 45%|████▌     | 4809/10682 [45:58<36:42,  2.67it/s]
 45%|████▌     | 4810/10682 [45:59<36:40,  2.67it/s]
 45%|████▌     | 4811/10682 [45:59<36:39,  2.67it/s]
 45%|████▌     | 4812/10682 [45:59<36:38,  2.67it/s]
 45%|████▌     | 4813/10682 [46:00<36:42,  2.67it/s]
 45%|████▌     | 4814/10682 [46:00<36:42,  2.66it/s]
 45%|████▌     | 4815/10682 [46:01<36:41,  2.66it/s]
 45%|████▌     | 4816/10682 [46:01<36:39,  2.67it/s]
 45%|████▌     | 4817/10682 [46:01<36:42,  2.66it/s]
 45%|████▌     | 4818/10682 [46:02<36:41,  2.66it/s]
 45%|████▌     | 4819/10682 [46:02<36:40,  2.66it/s]
 45%|████▌     | 4820/10682 [46:02<36:38,  2.67it/s]
 45%|████▌     | 4821/10682 [46:03<36:38,  2.67it/s]
 45%|████▌     | 4822/10682 [46:03<36:36,  2.67it/s]
 45%|████▌     | 4823/10682 [46:04<36:35,  2.67it/s]
 45%|████▌     | 4824/10682 [46:04<36:35,  2.67it/s]
 45%|████▌     | 4825/10682 [46:04<36:36,  2.67it/s]{'loss': 4.3321, 'grad_norm': 0.4377874433994293, 'learning_rate': 0.0006683030872510552, 'epoch': 0.45}
                                                    
-
 45%|████▌     | 4825/10682 [46:04<36:36,  2.67it/s]
 45%|████▌     | 4826/10682 [46:05<36:42,  2.66it/s]
 45%|████▌     | 4827/10682 [46:05<36:40,  2.66it/s]
 45%|████▌     | 4828/10682 [46:05<36:40,  2.66it/s]
 45%|████▌     | 4829/10682 [46:06<36:37,  2.66it/s]
 45%|████▌     | 4830/10682 [46:06<36:38,  2.66it/s]
 45%|████▌     | 4831/10682 [46:07<36:40,  2.66it/s]
 45%|████▌     | 4832/10682 [46:07<36:36,  2.66it/s]
 45%|████▌     | 4833/10682 [46:07<36:35,  2.66it/s]
 45%|████▌     | 4834/10682 [46:08<36:34,  2.66it/s]
 45%|████▌     | 4835/10682 [46:08<36:35,  2.66it/s]
 45%|████▌     | 4836/10682 [46:08<36:33,  2.66it/s]
 45%|████▌     | 4837/10682 [46:09<36:31,  2.67it/s]
 45%|████▌     | 4838/10682 [46:09<36:31,  2.67it/s]
 45%|████▌     | 4839/10682 [46:10<36:30,  2.67it/s]
 45%|████▌     | 4840/10682 [46:10<36:31,  2.67it/s]
 45%|████▌     | 4841/10682 [46:10<36:35,  2.66it/s]
 45%|████▌     | 4842/10682 [46:11<36:35,  2.66it/s]
 45%|████▌     | 4843/10682 [46:11<36:35,  2.66it/s]
 45%|████▌     | 4844/10682 [46:11<36:36,  2.66it/s]
 45%|████▌     | 4845/10682 [46:12<36:35,  2.66it/s]
 45%|████▌     | 4846/10682 [46:12<36:36,  2.66it/s]
 45%|████▌     | 4847/10682 [46:13<36:34,  2.66it/s]
 45%|████▌     | 4848/10682 [46:13<36:33,  2.66it/s]
 45%|████▌     | 4849/10682 [46:13<36:30,  2.66it/s]
 45%|████▌     | 4850/10682 [46:14<36:27,  2.67it/s]
                                                    
{'loss': 4.3214, 'grad_norm': 0.42770621180534363, 'learning_rate': 0.0006644508122054031, 'epoch': 0.45}
-
 45%|████▌     | 4850/10682 [46:14<36:27,  2.67it/s]
 45%|████▌     | 4851/10682 [46:14<36:40,  2.65it/s]
 45%|████▌     | 4852/10682 [46:14<36:33,  2.66it/s]
 45%|████▌     | 4853/10682 [46:15<36:31,  2.66it/s]
 45%|████▌     | 4854/10682 [46:15<36:27,  2.66it/s]
 45%|████▌     | 4855/10682 [46:16<36:30,  2.66it/s]
 45%|████▌     | 4856/10682 [46:16<36:28,  2.66it/s]
 45%|████▌     | 4857/10682 [46:16<36:30,  2.66it/s]
 45%|████▌     | 4858/10682 [46:17<36:30,  2.66it/s]
 45%|████▌     | 4859/10682 [46:17<36:27,  2.66it/s]
 45%|████▌     | 4860/10682 [46:17<36:26,  2.66it/s]
 46%|████▌     | 4861/10682 [46:18<36:25,  2.66it/s]
 46%|████▌     | 4862/10682 [46:18<36:23,  2.67it/s]
 46%|████▌     | 4863/10682 [46:19<36:20,  2.67it/s]
 46%|████▌     | 4864/10682 [46:19<36:20,  2.67it/s]
 46%|████▌     | 4865/10682 [46:19<36:19,  2.67it/s]
 46%|████▌     | 4866/10682 [46:20<36:19,  2.67it/s]
 46%|████▌     | 4867/10682 [46:20<36:18,  2.67it/s]
 46%|████▌     | 4868/10682 [46:20<36:25,  2.66it/s]
 46%|████▌     | 4869/10682 [46:21<36:25,  2.66it/s]
 46%|████▌     | 4870/10682 [46:21<36:26,  2.66it/s]
 46%|████▌     | 4871/10682 [46:22<36:24,  2.66it/s]
 46%|████▌     | 4872/10682 [46:22<36:22,  2.66it/s]
 46%|████▌     | 4873/10682 [46:22<36:19,  2.67it/s]
 46%|████▌     | 4874/10682 [46:23<36:17,  2.67it/s]
 46%|████▌     | 4875/10682 [46:23<36:17,  2.67it/s]
                                                    {'loss': 4.3286, 'grad_norm': 0.4632282257080078, 'learning_rate': 0.0006605875598609699, 'epoch': 0.46}
-

 46%|████▌     | 4875/10682 [46:23<36:17,  2.67it/s]
 46%|████▌     | 4876/10682 [46:23<36:18,  2.67it/s]
 46%|████▌     | 4877/10682 [46:24<36:19,  2.66it/s]
 46%|████▌     | 4878/10682 [46:24<36:19,  2.66it/s]
 46%|████▌     | 4879/10682 [46:25<36:21,  2.66it/s]
 46%|████▌     | 4880/10682 [46:25<36:21,  2.66it/s]
 46%|████▌     | 4881/10682 [46:25<36:23,  2.66it/s]
 46%|████▌     | 4882/10682 [46:26<36:22,  2.66it/s]
 46%|████▌     | 4883/10682 [46:26<36:21,  2.66it/s]
 46%|████▌     | 4884/10682 [46:26<36:20,  2.66it/s]
 46%|████▌     | 4885/10682 [46:27<36:20,  2.66it/s]
 46%|████▌     | 4886/10682 [46:27<36:18,  2.66it/s]
 46%|████▌     | 4887/10682 [46:28<36:15,  2.66it/s]
 46%|████▌     | 4888/10682 [46:28<36:15,  2.66it/s]
 46%|████▌     | 4889/10682 [46:28<36:13,  2.66it/s]
 46%|████▌     | 4890/10682 [46:29<36:14,  2.66it/s]
 46%|████▌     | 4891/10682 [46:29<36:15,  2.66it/s]
 46%|████▌     | 4892/10682 [46:29<36:15,  2.66it/s]
 46%|████▌     | 4893/10682 [46:30<36:13,  2.66it/s]
 46%|████▌     | 4894/10682 [46:30<36:10,  2.67it/s]
 46%|████▌     | 4895/10682 [46:31<36:09,  2.67it/s]
 46%|████▌     | 4896/10682 [46:31<36:06,  2.67it/s]
 46%|████▌     | 4897/10682 [46:31<36:11,  2.66it/s]
 46%|████▌     | 4898/10682 [46:32<36:12,  2.66it/s]
 46%|████▌     | 4899/10682 [46:32<36:10,  2.66it/s]
 46%|████▌     | 4900/10682 [46:32<36:10,  2.66it/s]{'loss': 4.3289, 'grad_norm': 0.4335024356842041, 'learning_rate': 0.0006567135880947134, 'epoch': 0.46}
                                                    
-
 46%|████▌     | 4900/10682 [46:32<36:10,  2.66it/s]
 46%|████▌     | 4901/10682 [46:33<36:14,  2.66it/s]
 46%|████▌     | 4902/10682 [46:33<36:10,  2.66it/s]
 46%|████▌     | 4903/10682 [46:34<36:07,  2.67it/s]
 46%|████▌     | 4904/10682 [46:34<36:04,  2.67it/s]
 46%|████▌     | 4905/10682 [46:34<36:02,  2.67it/s]
 46%|████▌     | 4906/10682 [46:35<36:02,  2.67it/s]
 46%|████▌     | 4907/10682 [46:35<36:02,  2.67it/s]
 46%|████▌     | 4908/10682 [46:35<36:04,  2.67it/s]
 46%|████▌     | 4909/10682 [46:36<36:05,  2.67it/s]
 46%|████▌     | 4910/10682 [46:36<36:06,  2.66it/s]
 46%|████▌     | 4911/10682 [46:37<36:07,  2.66it/s]
 46%|████▌     | 4912/10682 [46:37<36:08,  2.66it/s]
 46%|████▌     | 4913/10682 [46:37<36:04,  2.66it/s]
 46%|████▌     | 4914/10682 [46:38<36:05,  2.66it/s]
 46%|████▌     | 4915/10682 [46:38<36:04,  2.66it/s]
 46%|████▌     | 4916/10682 [46:38<36:01,  2.67it/s]
 46%|████▌     | 4917/10682 [46:39<35:59,  2.67it/s]
 46%|████▌     | 4918/10682 [46:39<35:59,  2.67it/s]
 46%|████▌     | 4919/10682 [46:40<35:58,  2.67it/s]
 46%|████▌     | 4920/10682 [46:40<36:00,  2.67it/s]
 46%|████▌     | 4921/10682 [46:40<36:02,  2.66it/s]
 46%|████▌     | 4922/10682 [46:41<36:01,  2.67it/s]
 46%|████▌     | 4923/10682 [46:41<36:00,  2.67it/s]
 46%|████▌     | 4924/10682 [46:41<36:04,  2.66it/s]
 46%|████▌     | 4925/10682 [46:42<36:00,  2.66it/s]
                                                    
{'loss': 4.3284, 'grad_norm': 0.4271387755870819, 'learning_rate': 0.0006528291554991272, 'epoch': 0.46}
-
 46%|████▌     | 4925/10682 [46:42<36:00,  2.66it/s]
 46%|████▌     | 4926/10682 [46:42<36:10,  2.65it/s]
 46%|████▌     | 4927/10682 [46:43<36:06,  2.66it/s]
 46%|████▌     | 4928/10682 [46:43<36:06,  2.66it/s]
 46%|████▌     | 4929/10682 [46:43<36:03,  2.66it/s]
 46%|████▌     | 4930/10682 [46:44<36:00,  2.66it/s]
 46%|████▌     | 4931/10682 [46:44<35:56,  2.67it/s]
 46%|████▌     | 4932/10682 [46:44<35:56,  2.67it/s]
 46%|████▌     | 4933/10682 [46:45<35:55,  2.67it/s]
 46%|████▌     | 4934/10682 [46:45<35:56,  2.67it/s]
 46%|��███▌     | 4935/10682 [46:46<36:00,  2.66it/s]
 46%|████▌     | 4936/10682 [46:46<35:59,  2.66it/s]
 46%|████▌     | 4937/10682 [46:46<35:58,  2.66it/s]
 46%|████▌     | 4938/10682 [46:47<35:56,  2.66it/s]
 46%|████▌     | 4939/10682 [46:47<35:55,  2.66it/s]
 46%|████▌     | 4940/10682 [46:47<35:56,  2.66it/s]
 46%|████▋     | 4941/10682 [46:48<35:55,  2.66it/s]
 46%|████▋     | 4942/10682 [46:48<35:58,  2.66it/s]
 46%|████▋     | 4943/10682 [46:49<35:59,  2.66it/s]
 46%|████▋     | 4944/10682 [46:49<35:56,  2.66it/s]
 46%|████▋     | 4945/10682 [46:49<35:52,  2.66it/s]
 46%|████▋     | 4946/10682 [46:50<35:51,  2.67it/s]
 46%|████▋     | 4947/10682 [46:50<35:50,  2.67it/s]
 46%|████▋     | 4948/10682 [46:50<35:51,  2.67it/s]
 46%|████▋     | 4949/10682 [46:51<35:53,  2.66it/s]
 46%|████▋     | 4950/10682 [46:51<35:50,  2.67it/s]
                                                    
{'loss': 4.3277, 'grad_norm': 0.4087849259376526, 'learning_rate': 0.0006489345213649782, 'epoch': 0.46}
-
 46%|████▋     | 4950/10682 [46:51<35:50,  2.67it/s]
 46%|████▋     | 4951/10682 [46:52<35:55,  2.66it/s]
 46%|████▋     | 4952/10682 [46:52<35:52,  2.66it/s]
 46%|████▋     | 4953/10682 [46:52<35:53,  2.66it/s]
 46%|████▋     | 4954/10682 [46:53<35:50,  2.66it/s]
 46%|████▋     | 4955/10682 [46:53<35:50,  2.66it/s]
 46%|████▋     | 4956/10682 [46:53<35:52,  2.66it/s]
 46%|████▋     | 4957/10682 [46:54<35:51,  2.66it/s]
 46%|████▋     | 4958/10682 [46:54<35:49,  2.66it/s]
 46%|████▋     | 4959/10682 [46:55<35:49,  2.66it/s]
 46%|████▋     | 4960/10682 [46:55<35:47,  2.66it/s]
 46%|████▋     | 4961/10682 [46:55<35:45,  2.67it/s]
 46%|████▋     | 4962/10682 [46:56<35:45,  2.67it/s]
 46%|████▋     | 4963/10682 [46:56<35:46,  2.66it/s]
 46%|████▋     | 4964/10682 [46:56<35:49,  2.66it/s]
 46%|████▋     | 4965/10682 [46:57<35:45,  2.66it/s]
 46%|████▋     | 4966/10682 [46:57<35:42,  2.67it/s]
 46%|████▋     | 4967/10682 [46:58<35:46,  2.66it/s]
 47%|████▋     | 4968/10682 [46:58<35:45,  2.66it/s]
 47%|████▋     | 4969/10682 [46:58<35:47,  2.66it/s]
 47%|████▋     | 4970/10682 [46:59<35:45,  2.66it/s]
 47%|████▋     | 4971/10682 [46:59<35:46,  2.66it/s]
 47%|████▋     | 4972/10682 [46:59<35:42,  2.66it/s]
 47%|████▋     | 4973/10682 [47:00<35:41,  2.67it/s]
 47%|████▋     | 4974/10682 [47:00<35:37,  2.67it/s]
 47%|████▋     | 4975/10682 [47:01<35:43,  2.66it/s]{'loss': 4.3317, 'grad_norm': 0.43323931097984314, 'learning_rate': 0.0006450299456639985, 'epoch': 0.47}
-
                                                    

 47%|████▋     | 4975/10682 [47:01<35:43,  2.66it/s]
 47%|████▋     | 4976/10682 [47:01<35:45,  2.66it/s]
 47%|████▋     | 4977/10682 [47:01<35:45,  2.66it/s]
 47%|████▋     | 4978/10682 [47:02<35:45,  2.66it/s]
 47%|████▋     | 4979/10682 [47:02<35:43,  2.66it/s]
 47%|████▋     | 4980/10682 [47:02<35:40,  2.66it/s]
 47%|████▋     | 4981/10682 [47:03<35:37,  2.67it/s]
 47%|████▋     | 4982/10682 [47:03<35:38,  2.67it/s]
 47%|████▋     | 4983/10682 [47:04<35:39,  2.66it/s]
 47%|████▋     | 4984/10682 [47:04<35:37,  2.67it/s]
 47%|████▋     | 4985/10682 [47:04<35:37,  2.67it/s]
 47%|████▋     | 4986/10682 [47:05<35:35,  2.67it/s]
 47%|████▋     | 4987/10682 [47:05<35:36,  2.67it/s]
 47%|████▋     | 4988/10682 [47:05<35:39,  2.66it/s]
 47%|████▋     | 4989/10682 [47:06<35:37,  2.66it/s]
 47%|████▋     | 4990/10682 [47:06<35:34,  2.67it/s]
 47%|████▋     | 4991/10682 [47:07<35:37,  2.66it/s]
 47%|████▋     | 4992/10682 [47:07<35:38,  2.66it/s]
 47%|████▋     | 4993/10682 [47:07<35:35,  2.66it/s]
 47%|████▋     | 4994/10682 [47:08<35:35,  2.66it/s]
 47%|████▋     | 4995/10682 [47:08<35:33,  2.67it/s]
 47%|████▋     | 4996/10682 [47:08<35:33,  2.67it/s]
 47%|████▋     | 4997/10682 [47:09<35:31,  2.67it/s]
 47%|████▋     | 4998/10682 [47:09<35:32,  2.67it/s]
 47%|████▋     | 4999/10682 [47:10<35:31,  2.67it/s]
 47%|████▋     | 5000/10682 [47:10<35:31,  2.67it/s]
                                                    
{'loss': 4.3245, 'grad_norm': 0.4278932213783264, 'learning_rate': 0.0006411156890315328, 'epoch': 0.47}
-
 47%|████▋     | 5000/10682 [47:10<35:31,  2.67it/s]
 47%|████▋     | 5001/10682 [47:10<35:32,  2.66it/s]
 47%|████▋     | 5002/10682 [47:11<35:37,  2.66it/s]
 47%|████▋     | 5003/10682 [47:11<35:33,  2.66it/s]
 47%|████▋     | 5004/10682 [47:11<35:31,  2.66it/s]
 47%|████▋     | 5005/10682 [47:12<35:28,  2.67it/s]
 47%|████▋     | 5006/10682 [47:12<35:27,  2.67it/s]
 47%|████▋     | 5007/10682 [47:13<35:26,  2.67it/s]
 47%|████▋     | 5008/10682 [47:13<35:24,  2.67it/s]
 47%|████▋     | 5009/10682 [47:13<35:24,  2.67it/s]
 47%|████▋     | 5010/10682 [47:14<35:23,  2.67it/s]
 47%|████▋     | 5011/10682 [47:14<35:22,  2.67it/s]
 47%|████▋     | 5012/10682 [47:14<35:24,  2.67it/s]
 47%|████▋     | 5013/10682 [47:15<35:24,  2.67it/s]
 47%|████▋     | 5014/10682 [47:15<35:24,  2.67it/s]
 47%|████▋     | 5015/10682 [47:16<35:25,  2.67it/s]
 47%|████▋     | 5016/10682 [47:16<35:24,  2.67it/s]
 47%|████▋     | 5017/10682 [47:16<35:25,  2.67it/s]
 47%|████▋     | 5018/10682 [47:17<35:26,  2.66it/s]
 47%|████▋     | 5019/10682 [47:17<35:24,  2.67it/s]
 47%|████▋     | 5020/10682 [47:17<35:21,  2.67it/s]
 47%|████▋     | 5021/10682 [47:18<35:21,  2.67it/s]
 47%|████▋     | 5022/10682 [47:18<35:20,  2.67it/s]
 47%|████▋     | 5023/10682 [47:19<35:19,  2.67it/s]
 47%|████▋     | 5024/10682 [47:19<35:18,  2.67it/s]
 47%|████▋     | 5025/10682 [47:19<35:17,  2.67it/s]
                                                    
{'loss': 4.3261, 'grad_norm': 0.49377310276031494, 'learning_rate': 0.0006371920127491398, 'epoch': 0.47}
-
 47%|████▋     | 5025/10682 [47:19<35:17,  2.67it/s]
 47%|████▋     | 5026/10682 [47:20<35:21,  2.67it/s]
 47%|████▋     | 5027/10682 [47:20<35:20,  2.67it/s]
 47%|████▋     | 5028/10682 [47:20<35:19,  2.67it/s]
 47%|████▋     | 5029/10682 [47:21<35:22,  2.66it/s]
 47%|████▋     | 5030/10682 [47:21<35:21,  2.66it/s]
 47%|████▋     | 5031/10682 [47:22<35:24,  2.66it/s]
 47%|████▋     | 5032/10682 [47:22<35:21,  2.66it/s]
 47%|████▋     | 5033/10682 [47:22<35:19,  2.66it/s]
 47%|████▋     | 5034/10682 [47:23<35:19,  2.67it/s]
 47%|████▋     | 5035/10682 [47:23<35:18,  2.67it/s]
 47%|████▋     | 5036/10682 [47:23<35:17,  2.67it/s]
 47%|████▋     | 5037/10682 [47:24<35:18,  2.67it/s]
 47%|████▋     | 5038/10682 [47:24<35:18,  2.66it/s]
 47%|████▋     | 5039/10682 [47:25<35:18,  2.66it/s]
 47%|████▋     | 5040/10682 [47:25<35:19,  2.66it/s]
 47%|████▋     | 5041/10682 [47:25<35:19,  2.66it/s]
 47%|████▋     | 5042/10682 [47:26<35:22,  2.66it/s]
 47%|████▋     | 5043/10682 [47:26<35:20,  2.66it/s]
 47%|████▋     | 5044/10682 [47:26<35:20,  2.66it/s]
 47%|████▋     | 5045/10682 [47:27<35:16,  2.66it/s]
 47%|████▋     | 5046/10682 [47:27<35:15,  2.66it/s]
 47%|████▋     | 5047/10682 [47:28<35:15,  2.66it/s]
 47%|████▋     | 5048/10682 [47:28<35:13,  2.67it/s]
 47%|████▋     | 5049/10682 [47:28<35:14,  2.66it/s]
 47%|████▋     | 5050/10682 [47:29<35:12,  2.67it/s]
                                                    {'loss': 4.3243, 'grad_norm': 0.4185958504676819, 'learning_rate': 0.0006332591787271516, 'epoch': 0.47}
-

 47%|████▋     | 5050/10682 [47:29<35:12,  2.67it/s]
 47%|████▋     | 5051/10682 [47:29<35:15,  2.66it/s]
 47%|████▋     | 5052/10682 [47:29<35:15,  2.66it/s]
 47%|████▋     | 5053/10682 [47:30<35:13,  2.66it/s]
 47%|████▋     | 5054/10682 [47:30<35:12,  2.66it/s]
 47%|████▋     | 5055/10682 [47:31<35:14,  2.66it/s]
 47%|████▋     | 5056/10682 [47:31<35:17,  2.66it/s]
 47%|████▋     | 5057/10682 [47:31<35:13,  2.66it/s]
 47%|████▋     | 5058/10682 [47:32<35:14,  2.66it/s]
 47%|████▋     | 5059/10682 [47:32<35:12,  2.66it/s]
 47%|████▋     | 5060/10682 [47:33<35:13,  2.66it/s]
 47%|████▋     | 5061/10682 [47:33<35:12,  2.66it/s]
 47%|████▋     | 5062/10682 [47:33<35:09,  2.66it/s]
 47%|████▋     | 5063/10682 [47:34<35:10,  2.66it/s]
 47%|████▋     | 5064/10682 [47:34<35:07,  2.67it/s]
 47%|████▋     | 5065/10682 [47:34<35:07,  2.66it/s]
 47%|████▋     | 5066/10682 [47:35<35:06,  2.67it/s]
 47%|████▋     | 5067/10682 [47:35<35:08,  2.66it/s]
 47%|████▋     | 5068/10682 [47:36<35:04,  2.67it/s]
 47%|████▋     | 5069/10682 [47:36<35:10,  2.66it/s]
 47%|████▋     | 5070/10682 [47:36<35:09,  2.66it/s]
 47%|████▋     | 5071/10682 [47:37<35:10,  2.66it/s]
 47%|████▋     | 5072/10682 [47:37<35:08,  2.66it/s]
 47%|████▋     | 5073/10682 [47:37<35:06,  2.66it/s]
 48%|████▊     | 5074/10682 [47:38<35:04,  2.66it/s]
 48%|████▊     | 5075/10682 [47:38<35:02,  2.67it/s]
                                                    
{'loss': 4.3241, 'grad_norm': 0.43277043104171753, 'learning_rate': 0.0006293174494871916, 'epoch': 0.48}
-
 48%|████▊     | 5075/10682 [47:38<35:02,  2.67it/s]
 48%|████▊     | 5076/10682 [47:39<35:07,  2.66it/s]
 48%|████▊     | 5077/10682 [47:39<35:04,  2.66it/s]
 48%|████▊     | 5078/10682 [47:39<35:04,  2.66it/s]
 48%|████▊     | 5079/10682 [47:40<35:02,  2.66it/s]
 48%|████▊     | 5080/10682 [47:40<35:02,  2.66it/s]
 48%|████▊     | 5081/10682 [47:40<35:02,  2.66it/s]
 48%|████▊     | 5082/10682 [47:41<35:03,  2.66it/s]
 48%|████▊     | 5083/10682 [47:41<35:01,  2.66it/s]
 48%|████▊     | 5084/10682 [47:42<35:03,  2.66it/s]
 48%|████▊     | 5085/10682 [47:42<35:06,  2.66it/s]
 48%|████▊     | 5086/10682 [47:42<35:03,  2.66it/s]
 48%|████▊     | 5087/10682 [47:43<35:03,  2.66it/s]
 48%|████▊     | 5088/10682 [47:43<35:00,  2.66it/s]
 48%|████▊     | 5089/10682 [47:43<35:01,  2.66it/s]
 48%|████▊     | 5090/10682 [47:44<34:58,  2.66it/s]
 48%|████▊     | 5091/10682 [47:44<34:57,  2.67it/s]
 48%|████▊     | 5092/10682 [47:45<34:56,  2.67it/s]
 48%|████▊     | 5093/10682 [47:45<34:58,  2.66it/s]
 48%|████▊     | 5094/10682 [47:45<35:01,  2.66it/s]
 48%|████▊     | 5095/10682 [47:46<34:58,  2.66it/s]
 48%|████▊     | 5096/10682 [47:46<35:00,  2.66it/s]
 48%|████▊     | 5097/10682 [47:46<34:58,  2.66it/s]
 48%|████▊     | 5098/10682 [47:47<35:01,  2.66it/s]
 48%|████▊     | 5099/10682 [47:47<35:00,  2.66it/s]
 48%|████▊     | 5100/10682 [47:48<34:55,  2.66it/s]{'loss': 4.3206, 'grad_norm': 0.4976399540901184, 'learning_rate': 0.0006253670881446494, 'epoch': 0.48}
-
                                                    

 48%|████▊     | 5100/10682 [47:48<34:55,  2.66it/s]
 48%|████▊     | 5101/10682 [47:48<34:57,  2.66it/s]
 48%|████▊     | 5102/10682 [47:48<34:58,  2.66it/s]
 48%|████▊     | 5103/10682 [47:49<34:56,  2.66it/s]
 48%|████▊     | 5104/10682 [47:49<34:54,  2.66it/s]
 48%|████▊     | 5105/10682 [47:49<34:52,  2.66it/s]
 48%|████▊     | 5106/10682 [47:50<34:51,  2.67it/s]
 48%|████▊     | 5107/10682 [47:50<34:50,  2.67it/s]
 48%|████▊     | 5108/10682 [47:51<34:50,  2.67it/s]
 48%|████▊     | 5109/10682 [47:51<34:52,  2.66it/s]
 48%|████▊     | 5110/10682 [47:51<34:52,  2.66it/s]
 48%|████▊     | 5111/10682 [47:52<34:55,  2.66it/s]
 48%|████▊     | 5112/10682 [47:52<34:56,  2.66it/s]
 48%|████▊     | 5113/10682 [47:52<34:54,  2.66it/s]
 48%|████▊     | 5114/10682 [47:53<34:51,  2.66it/s]
 48%|████▊     | 5115/10682 [47:53<34:48,  2.67it/s]
 48%|████▊     | 5116/10682 [47:54<34:48,  2.66it/s]
 48%|████▊     | 5117/10682 [47:54<34:49,  2.66it/s]
 48%|████▊     | 5118/10682 [47:54<34:48,  2.66it/s]
 48%|████▊     | 5119/10682 [47:55<34:46,  2.67it/s]
 48%|████▊     | 5120/10682 [47:55<34:44,  2.67it/s]
 48%|████▊     | 5121/10682 [47:55<34:43,  2.67it/s]
 48%|████▊     | 5122/10682 [47:56<34:42,  2.67it/s]
 48%|████▊     | 5123/10682 [47:56<34:45,  2.67it/s]
 48%|████▊     | 5124/10682 [47:57<34:44,  2.67it/s]
 48%|████▊     | 5125/10682 [47:57<34:48,  2.66it/s]{'loss': 4.3115, 'grad_norm': 0.40484437346458435, 'learning_rate': 0.0006214083583911186, 'epoch': 0.48}
-
                                                    

 48%|████▊     | 5125/10682 [47:57<34:48,  2.66it/s]
 48%|████▊     | 5126/10682 [47:57<34:50,  2.66it/s]
 48%|████▊     | 5127/10682 [47:58<34:45,  2.66it/s]
 48%|████▊     | 5128/10682 [47:58<34:44,  2.67it/s]
 48%|████▊     | 5129/10682 [47:58<34:44,  2.66it/s]
 48%|████▊     | 5130/10682 [47:59<34:43,  2.67it/s]
 48%|████▊     | 5131/10682 [47:59<34:42,  2.67it/s]
 48%|████▊     | 5132/10682 [48:00<34:40,  2.67it/s]
 48%|████▊     | 5133/10682 [48:00<34:39,  2.67it/s]
 48%|████▊     | 5134/10682 [48:00<34:43,  2.66it/s]
 48%|████▊     | 5135/10682 [48:01<34:45,  2.66it/s]
 48%|████▊     | 5136/10682 [48:01<34:46,  2.66it/s]
 48%|████▊     | 5137/10682 [48:01<34:43,  2.66it/s]
 48%|████▊     | 5138/10682 [48:02<34:43,  2.66it/s]
 48%|████▊     | 5139/10682 [48:02<34:44,  2.66it/s]
 48%|████▊     | 5140/10682 [48:03<34:42,  2.66it/s]
 48%|████▊     | 5141/10682 [48:03<34:44,  2.66it/s]
 48%|████▊     | 5142/10682 [48:03<34:40,  2.66it/s]
 48%|████▊     | 5143/10682 [48:04<34:42,  2.66it/s]
 48%|████▊     | 5144/10682 [48:04<34:41,  2.66it/s]
 48%|████▊     | 5145/10682 [48:04<34:40,  2.66it/s]
 48%|████▊     | 5146/10682 [48:05<34:40,  2.66it/s]
 48%|████▊     | 5147/10682 [48:05<34:38,  2.66it/s]
 48%|████▊     | 5148/10682 [48:06<34:38,  2.66it/s]
 48%|████▊     | 5149/10682 [48:06<34:40,  2.66it/s]
 48%|████▊     | 5150/10682 [48:06<34:39,  2.66it/s]{'loss': 4.3198, 'grad_norm': 0.4118918478488922, 'learning_rate': 0.0006174415244767953, 'epoch': 0.48}
                                                    
-
 48%|████▊     | 5150/10682 [48:06<34:39,  2.66it/s]
 48%|████▊     | 5151/10682 [48:07<34:40,  2.66it/s]
 48%|████▊     | 5152/10682 [48:07<34:40,  2.66it/s]
 48%|████▊     | 5153/10682 [48:07<34:37,  2.66it/s]
 48%|████▊     | 5154/10682 [48:08<34:35,  2.66it/s]
 48%|████▊     | 5155/10682 [48:08<34:37,  2.66it/s]
 48%|████▊     | 5156/10682 [48:09<34:35,  2.66it/s]
 48%|████▊     | 5157/10682 [48:09<34:35,  2.66it/s]
 48%|████▊     | 5158/10682 [48:09<34:35,  2.66it/s]
 48%|████▊     | 5159/10682 [48:10<34:35,  2.66it/s]
 48%|████▊     | 5160/10682 [48:10<34:32,  2.66it/s]
 48%|████▊     | 5161/10682 [48:10<34:33,  2.66it/s]
 48%|████▊     | 5162/10682 [48:11<34:29,  2.67it/s]
 48%|████▊     | 5163/10682 [48:11<34:30,  2.67it/s]
 48%|████▊     | 5164/10682 [48:12<34:28,  2.67it/s]
 48%|████▊     | 5165/10682 [48:12<34:31,  2.66it/s]
 48%|████▊     | 5166/10682 [48:12<34:32,  2.66it/s]
 48%|████▊     | 5167/10682 [48:13<34:31,  2.66it/s]
 48%|████▊     | 5168/10682 [48:13<34:29,  2.66it/s]
 48%|████▊     | 5169/10682 [48:13<34:30,  2.66it/s]
 48%|████▊     | 5170/10682 [48:14<34:26,  2.67it/s]
 48%|████▊     | 5171/10682 [48:14<34:25,  2.67it/s]
 48%|████▊     | 5172/10682 [48:15<34:25,  2.67it/s]
 48%|████▊     | 5173/10682 [48:15<34:24,  2.67it/s]
 48%|████▊     | 5174/10682 [48:15<34:23,  2.67it/s]
 48%|████▊     | 5175/10682 [48:16<34:24,  2.67it/s]
                                                    {'loss': 4.3169, 'grad_norm': 0.429974228143692, 'learning_rate': 0.0006134668511928376, 'epoch': 0.48}
-
 48%|████▊     | 5175/10682 [48:16<34:24,  2.67it/s]
 48%|████▊     | 5176/10682 [48:16<34:28,  2.66it/s]
 48%|████▊     | 5177/10682 [48:16<34:29,  2.66it/s]
 48%|████▊     | 5178/10682 [48:17<34:30,  2.66it/s]
 48%|████▊     | 5179/10682 [48:17<34:31,  2.66it/s]
 48%|████▊     | 5180/10682 [48:18<34:28,  2.66it/s]
 49%|████▊     | 5181/10682 [48:18<34:27,  2.66it/s]
 49%|████▊     | 5182/10682 [48:18<34:26,  2.66it/s]
 49%|████▊     | 5183/10682 [48:19<34:23,  2.67it/s]
 49%|████▊     | 5184/10682 [48:19<34:22,  2.67it/s]
 49%|████▊     | 5185/10682 [48:19<34:21,  2.67it/s]
 49%|████▊     | 5186/10682 [48:20<34:20,  2.67it/s]
 49%|████▊     | 5187/10682 [48:20<34:21,  2.67it/s]
 49%|████▊     | 5188/10682 [48:21<34:21,  2.67it/s]
 49%|████▊     | 5189/10682 [48:21<34:23,  2.66it/s]
 49%|████▊     | 5190/10682 [48:21<34:21,  2.66it/s]
 49%|████▊     | 5191/10682 [48:22<34:21,  2.66it/s]
 49%|████▊     | 5192/10682 [48:22<34:23,  2.66it/s]
 49%|████▊     | 5193/10682 [48:22<34:21,  2.66it/s]
 49%|████▊     | 5194/10682 [48:23<34:24,  2.66it/s]
 49%|████▊     | 5195/10682 [48:23<34:25,  2.66it/s]
 49%|████▊     | 5196/10682 [48:24<34:22,  2.66it/s]
 49%|████▊     | 5197/10682 [48:24<34:19,  2.66it/s]
 49%|████▊     | 5198/10682 [48:24<34:18,  2.66it/s]
 49%|████▊     | 5199/10682 [48:25<34:16,  2.67it/s]
 49%|████▊     | 5200/10682 [48:25<34:14,  2.67it/s]
                                                    
{'loss': 4.3146, 'grad_norm': 0.48373499512672424, 'learning_rate': 0.0006094846038536922, 'epoch': 0.49}
-
 49%|████▊     | 5200/10682 [48:25<34:14,  2.67it/s]
 49%|████▊     | 5201/10682 [48:25<34:18,  2.66it/s]
 49%|████▊     | 5202/10682 [48:26<34:18,  2.66it/s]
 49%|████▊     | 5203/10682 [48:26<34:21,  2.66it/s]
 49%|████▊     | 5204/10682 [48:27<34:19,  2.66it/s]
 49%|████▊     | 5205/10682 [48:27<34:19,  2.66it/s]
 49%|████▊     | 5206/10682 [48:27<34:15,  2.66it/s]
 49%|████▊     | 5207/10682 [48:28<34:14,  2.66it/s]
 49%|████▉     | 5208/10682 [48:28<34:12,  2.67it/s]
 49%|████▉     | 5209/10682 [48:28<34:11,  2.67it/s]
 49%|████▉     | 5210/10682 [48:29<34:11,  2.67it/s]
 49%|████▉     | 5211/10682 [48:29<34:12,  2.67it/s]
 49%|████▉     | 5212/10682 [48:30<34:10,  2.67it/s]
 49%|████▉     | 5213/10682 [48:30<34:09,  2.67it/s]
 49%|████▉     | 5214/10682 [48:30<34:08,  2.67it/s]
 49%|████▉     | 5215/10682 [48:31<34:06,  2.67it/s]
 49%|████▉     | 5216/10682 [48:31<34:13,  2.66it/s]
 49%|████▉     | 5217/10682 [48:31<34:14,  2.66it/s]
 49%|████▉     | 5218/10682 [48:32<34:12,  2.66it/s]
 49%|████▉     | 5219/10682 [48:32<34:14,  2.66it/s]
 49%|████▉     | 5220/10682 [48:33<34:11,  2.66it/s]
 49%|██���█▉     | 5221/10682 [48:33<34:11,  2.66it/s]
 49%|████▉     | 5222/10682 [48:33<34:10,  2.66it/s]
 49%|████▉     | 5223/10682 [48:34<34:07,  2.67it/s]
 49%|████▉     | 5224/10682 [48:34<34:07,  2.67it/s]
 49%|████▉     | 5225/10682 [48:34<34:05,  2.67it/s]
                                                    
{'loss': 4.3113, 'grad_norm': 0.41959527134895325, 'learning_rate': 0.000605495048279383, 'epoch': 0.49}
-
 49%|████▉     | 5225/10682 [48:34<34:05,  2.67it/s]
 49%|████▉     | 5226/10682 [48:35<34:03,  2.67it/s]
 49%|████▉     | 5227/10682 [48:35<34:02,  2.67it/s]
 49%|████▉     | 5228/10682 [48:36<34:03,  2.67it/s]
 49%|████▉     | 5229/10682 [48:36<34:03,  2.67it/s]
 49%|████▉     | 5230/10682 [48:36<34:06,  2.66it/s]
 49%|████▉     | 5231/10682 [48:37<34:06,  2.66it/s]
 49%|████▉     | 5232/10682 [48:37<34:08,  2.66it/s]
 49%|████▉     | 5233/10682 [48:37<34:08,  2.66it/s]
 49%|████▉     | 5234/10682 [48:38<34:06,  2.66it/s]
 49%|████▉     | 5235/10682 [48:38<34:05,  2.66it/s]
 49%|████▉     | 5236/10682 [48:39<34:06,  2.66it/s]
 49%|████▉     | 5237/10682 [48:39<34:06,  2.66it/s]
 49%|████▉     | 5238/10682 [48:39<34:03,  2.66it/s]
 49%|████▉     | 5239/10682 [48:40<34:01,  2.67it/s]
 49%|████▉     | 5240/10682 [48:40<34:01,  2.67it/s]
 49%|████▉     | 5241/10682 [48:40<34:04,  2.66it/s]
 49%|████▉     | 5242/10682 [48:41<34:01,  2.66it/s]
 49%|████▉     | 5243/10682 [48:41<34:03,  2.66it/s]
 49%|████▉     | 5244/10682 [48:42<34:01,  2.66it/s]
 49%|████▉     | 5245/10682 [48:42<34:06,  2.66it/s]
 49%|████▉     | 5246/10682 [48:42<34:05,  2.66it/s]
 49%|████▉     | 5247/10682 [48:43<34:03,  2.66it/s]
 49%|████▉     | 5248/10682 [48:43<34:01,  2.66it/s]
 49%|████▉     | 5249/10682 [48:43<34:00,  2.66it/s]
 49%|████▉     | 5250/10682 [48:44<34:01,  2.66it/s]
                                                    
{'loss': 4.3112, 'grad_norm': 0.4051324427127838, 'learning_rate': 0.0006014984507777682, 'epoch': 0.49}
-
 49%|████▉     | 5250/10682 [48:44<34:01,  2.66it/s]
 49%|████▉     | 5251/10682 [48:44<34:03,  2.66it/s]
 49%|████▉     | 5252/10682 [48:45<34:02,  2.66it/s]
 49%|████▉     | 5253/10682 [48:45<33:59,  2.66it/s]
 49%|████▉     | 5254/10682 [48:45<33:57,  2.66it/s]
 49%|████▉     | 5255/10682 [48:46<33:54,  2.67it/s]
 49%|████▉     | 5256/10682 [48:46<33:55,  2.67it/s]
 49%|████▉     | 5257/10682 [48:46<33:54,  2.67it/s]
 49%|████▉     | 5258/10682 [48:47<33:54,  2.67it/s]
 49%|████▉     | 5259/10682 [48:47<33:54,  2.67it/s]
 49%|████▉     | 5260/10682 [48:48<33:55,  2.66it/s]
 49%|████▉     | 5261/10682 [48:48<33:59,  2.66it/s]
 49%|████▉     | 5262/10682 [48:48<34:01,  2.66it/s]
 49%|████▉     | 5263/10682 [48:49<34:01,  2.65it/s]
 49%|████▉     | 5264/10682 [48:49<33:59,  2.66it/s]
 49%|████▉     | 5265/10682 [48:49<33:57,  2.66it/s]
 49%|████▉     | 5266/10682 [48:50<33:53,  2.66it/s]
 49%|████▉     | 5267/10682 [48:50<33:52,  2.66it/s]
 49%|████▉     | 5268/10682 [48:51<33:51,  2.67it/s]
 49%|████▉     | 5269/10682 [48:51<33:51,  2.67it/s]
 49%|████▉     | 5270/10682 [48:51<33:53,  2.66it/s]
 49%|████▉     | 5271/10682 [48:52<33:53,  2.66it/s]
 49%|████▉     | 5272/10682 [48:52<33:52,  2.66it/s]
 49%|████▉     | 5273/10682 [48:52<33:51,  2.66it/s]
 49%|████▉     | 5274/10682 [48:53<33:52,  2.66it/s]
 49%|████▉     | 5275/10682 [48:53<33:53,  2.66it/s]{'loss': 4.3049, 'grad_norm': 0.4835157096385956, 'learning_rate': 0.000597495078126763, 'epoch': 0.49}
                                                    
-
 49%|████▉     | 5275/10682 [48:53<33:53,  2.66it/s]
 49%|████▉     | 5276/10682 [48:54<33:53,  2.66it/s]
 49%|████▉     | 5277/10682 [48:54<33:50,  2.66it/s]
 49%|████▉     | 5278/10682 [48:54<33:48,  2.66it/s]
 49%|████▉     | 5279/10682 [48:55<33:46,  2.67it/s]
 49%|████▉     | 5280/10682 [48:55<33:45,  2.67it/s]
 49%|████▉     | 5281/10682 [48:55<33:45,  2.67it/s]
 49%|████▉     | 5282/10682 [48:56<33:47,  2.66it/s]
 49%|████▉     | 5283/10682 [48:56<33:50,  2.66it/s]
 49%|████▉     | 5284/10682 [48:57<33:55,  2.65it/s]
 49%|████▉     | 5285/10682 [48:57<33:59,  2.65it/s]
 49%|████▉     | 5286/10682 [48:57<33:55,  2.65it/s]
 49%|████▉     | 5287/10682 [48:58<33:51,  2.66it/s]
 50%|████▉     | 5288/10682 [48:58<33:50,  2.66it/s]
 50%|████▉     | 5289/10682 [48:59<33:48,  2.66it/s]
 50%|████▉     | 5290/10682 [48:59<33:46,  2.66it/s]
 50%|████▉     | 5291/10682 [48:59<33:45,  2.66it/s]
 50%|████▉     | 5292/10682 [49:00<33:47,  2.66it/s]
 50%|████▉     | 5293/10682 [49:00<33:45,  2.66it/s]
 50%|████▉     | 5294/10682 [49:00<33:45,  2.66it/s]
 50%|████▉     | 5295/10682 [49:01<33:42,  2.66it/s]
 50%|████▉     | 5296/10682 [49:01<33:43,  2.66it/s]
 50%|████▉     | 5297/10682 [49:02<33:43,  2.66it/s]
 50%|████▉     | 5298/10682 [49:02<33:41,  2.66it/s]
 50%|████▉     | 5299/10682 [49:02<33:44,  2.66it/s]
 50%|████▉     | 5300/10682 [49:03<33:46,  2.66it/s]{'loss': 4.3128, 'grad_norm': 0.42285779118537903, 'learning_rate': 0.000593485197556533, 'epoch': 0.5}
                                                    
-
 50%|████▉     | 5300/10682 [49:03<33:46,  2.66it/s]
 50%|████▉     | 5301/10682 [49:03<33:54,  2.65it/s]
 50%|████▉     | 5302/10682 [49:03<33:49,  2.65it/s]
 50%|████▉     | 5303/10682 [49:04<33:44,  2.66it/s]
 50%|████▉     | 5304/10682 [49:04<33:43,  2.66it/s]
 50%|████▉     | 5305/10682 [49:05<33:39,  2.66it/s]
 50%|████▉     | 5306/10682 [49:05<33:38,  2.66it/s]
 50%|████▉     | 5307/10682 [49:05<33:37,  2.66it/s]
 50%|████▉     | 5308/10682 [49:06<33:36,  2.66it/s]
 50%|████▉     | 5309/10682 [49:06<33:35,  2.67it/s]
 50%|████▉     | 5310/10682 [49:06<33:40,  2.66it/s]
 50%|████▉     | 5311/10682 [49:07<33:37,  2.66it/s]
 50%|████▉     | 5312/10682 [49:07<33:39,  2.66it/s]
 50%|████▉     | 5313/10682 [49:08<33:36,  2.66it/s]
 50%|████▉     | 5314/10682 [49:08<33:39,  2.66it/s]
 50%|████▉     | 5315/10682 [49:08<33:34,  2.66it/s]
 50%|████▉     | 5316/10682 [49:09<33:32,  2.67it/s]
 50%|████▉     | 5317/10682 [49:09<33:31,  2.67it/s]
 50%|████▉     | 5318/10682 [49:09<33:30,  2.67it/s]
 50%|████▉     | 5319/10682 [49:10<33:29,  2.67it/s]
 50%|████▉     | 5320/10682 [49:10<33:28,  2.67it/s]
 50%|████▉     | 5321/10682 [49:11<33:29,  2.67it/s]
 50%|████▉     | 5322/10682 [49:11<33:26,  2.67it/s]
 50%|████▉     | 5323/10682 [49:11<33:28,  2.67it/s]
 50%|████▉     | 5324/10682 [49:12<33:28,  2.67it/s]
 50%|████▉     | 5325/10682 [49:12<33:29,  2.67it/s]
                                                    
{'loss': 4.3033, 'grad_norm': 0.46620646119117737, 'learning_rate': 0.0005894690767316552, 'epoch': 0.5}
-
 50%|████▉     | 5325/10682 [49:12<33:29,  2.67it/s]
 50%|████▉     | 5326/10682 [49:12<33:30,  2.66it/s]
 50%|████▉     | 5327/10682 [49:13<33:30,  2.66it/s]
 50%|████▉     | 5328/10682 [49:13<33:27,  2.67it/s]
 50%|████▉     | 5329/10682 [49:14<41:07,  2.17it/s]
 50%|████▉     | 5330/10682 [49:14<38:49,  2.30it/s]
 50%|████▉     | 5331/10682 [49:15<37:11,  2.40it/s]
 50%|████▉     | 5332/10682 [49:15<36:03,  2.47it/s]
 50%|████▉     | 5333/10682 [49:15<35:17,  2.53it/s]
 50%|████▉     | 5334/10682 [49:16<34:44,  2.57it/s]
 50%|████▉     | 5335/10682 [49:16<34:21,  2.59it/s]
 50%|████▉     | 5336/10682 [49:16<34:09,  2.61it/s]
 50%|████▉     | 5337/10682 [49:17<33:55,  2.63it/s]
 50%|████▉     | 5338/10682 [49:17<33:49,  2.63it/s]
 50%|████▉     | 5339/10682 [49:18<33:42,  2.64it/s]
 50%|████▉     | 5340/10682 [49:18<33:37,  2.65it/s]
 50%|█████     | 5341/10682 [49:18<33:34,  2.65it/s]
 50%|█████     | 5342/10682 [49:19<33:30,  2.66it/s]
 50%|█████     | 5343/10682 [49:19<33:25,  2.66it/s]
 50%|█████     | 5344/10682 [49:19<33:24,  2.66it/s]
 50%|█████     | 5345/10682 [49:20<33:22,  2.67it/s]
 50%|█████     | 5346/10682 [49:20<33:20,  2.67it/s]
 50%|█████     | 5347/10682 [49:21<33:20,  2.67it/s]
 50%|█████     | 5348/10682 [49:21<33:19,  2.67it/s]
 50%|█████     | 5349/10682 [49:21<33:24,  2.66it/s]
 50%|█████     | 5350/10682 [49:22<33:24,  2.66it/s]{'loss': 4.3101, 'grad_norm': 0.4171009063720703, 'learning_rate': 0.0005854469837332512, 'epoch': 0.5}
                                                    
-
 50%|█████     | 5350/10682 [49:22<33:24,  2.66it/s]
 50%|█████     | 5351/10682 [49:22<33:27,  2.66it/s]
 50%|█████     | 5352/10682 [49:22<33:26,  2.66it/s]
 50%|█████     | 5353/10682 [49:23<40:42,  2.18it/s]
 50%|█████     | 5354/10682 [49:23<38:30,  2.31it/s]
 50%|█████     | 5355/10682 [49:24<36:54,  2.41it/s]
 50%|█████     | 5356/10682 [49:24<35:49,  2.48it/s]
 50%|█████     | 5357/10682 [49:25<35:02,  2.53it/s]
 50%|█████     | 5358/10682 [49:25<34:29,  2.57it/s]
 50%|█████     | 5359/10682 [49:25<34:05,  2.60it/s]
 50%|█████     | 5360/10682 [49:26<33:49,  2.62it/s]
 50%|█████     | 5361/10682 [49:26<33:39,  2.64it/s]
 50%|█████     | 5362/10682 [49:26<33:34,  2.64it/s]
 50%|█████     | 5363/10682 [49:27<33:30,  2.65it/s]
 50%|█████     | 5364/10682 [49:27<33:22,  2.66it/s]
 50%|█████     | 5365/10682 [49:28<33:16,  2.66it/s]
 50%|█████     | 5366/10682 [49:28<33:15,  2.66it/s]
 50%|█████     | 5367/10682 [49:28<33:14,  2.67it/s]
 50%|█████     | 5368/10682 [49:29<33:13,  2.67it/s]
 50%|█████     | 5369/10682 [49:29<33:12,  2.67it/s]
 50%|█████     | 5370/10682 [49:29<33:13,  2.67it/s]
 50%|█████     | 5371/10682 [49:30<33:10,  2.67it/s]
 50%|█████     | 5372/10682 [49:30<33:10,  2.67it/s]
 50%|█████     | 5373/10682 [49:31<33:09,  2.67it/s]
 50%|█████     | 5374/10682 [49:31<33:07,  2.67it/s]
 50%|█████     | 5375/10682 [49:31<33:08,  2.67it/s]{'loss': 4.3031, 'grad_norm': 0.41331252455711365, 'learning_rate': 0.0005814191870410933, 'epoch': 0.5}
                                                    
-
 50%|█████     | 5375/10682 [49:31<33:08,  2.67it/s]
 50%|█████     | 5376/10682 [49:32<33:11,  2.66it/s]
 50%|█████     | 5377/10682 [49:32<33:11,  2.66it/s]
 50%|█████     | 5378/10682 [49:32<33:12,  2.66it/s]
 50%|█████     | 5379/10682 [49:33<33:09,  2.66it/s]
 50%|█████     | 5380/10682 [49:33<33:10,  2.66it/s]
 50%|█████     | 5381/10682 [49:34<33:10,  2.66it/s]
 50%|█████     | 5382/10682 [49:34<33:09,  2.66it/s]
 50%|█████     | 5383/10682 [49:34<33:09,  2.66it/s]
 50%|█████     | 5384/10682 [49:35<33:09,  2.66it/s]
 50%|█████     | 5385/10682 [49:35<33:08,  2.66it/s]
 50%|█████     | 5386/10682 [49:35<33:06,  2.67it/s]
 50%|█████     | 5387/10682 [49:36<33:04,  2.67it/s]
 50%|█████     | 5388/10682 [49:36<33:01,  2.67it/s]
 50%|█████     | 5389/10682 [49:37<33:03,  2.67it/s]
 50%|█████     | 5390/10682 [49:37<33:06,  2.66it/s]
 50%|█████     | 5391/10682 [49:37<33:08,  2.66it/s]
 50%|█████     | 5392/10682 [49:38<33:06,  2.66it/s]
 50%|█████     | 5393/10682 [49:38<33:06,  2.66it/s]
 50%|█████     | 5394/10682 [49:38<33:06,  2.66it/s]
 51%|█████     | 5395/10682 [49:39<33:03,  2.67it/s]
 51%|█████     | 5396/10682 [49:39<33:06,  2.66it/s]
 51%|█████     | 5397/10682 [49:40<33:08,  2.66it/s]
 51%|█████     | 5398/10682 [49:40<33:04,  2.66it/s]
 51%|█████     | 5399/10682 [49:40<33:02,  2.67it/s]
 51%|█████     | 5400/10682 [49:41<33:00,  2.67it/s]{'loss': 4.3083, 'grad_norm': 0.4281124174594879, 'learning_rate': 0.0005773859555156823, 'epoch': 0.51}
                                                    
-
 51%|█████     | 5400/10682 [49:41<33:00,  2.67it/s]
 51%|█████     | 5401/10682 [49:41<33:11,  2.65it/s]
 51%|█████     | 5402/10682 [49:41<33:06,  2.66it/s]
 51%|█████     | 5403/10682 [49:42<33:05,  2.66it/s]
 51%|█████     | 5404/10682 [49:42<33:06,  2.66it/s]
 51%|█████     | 5405/10682 [49:43<33:04,  2.66it/s]
 51%|█████     | 5406/10682 [49:43<33:03,  2.66it/s]
 51%|█████     | 5407/10682 [49:43<33:06,  2.66it/s]
 51%|█████     | 5408/10682 [49:44<33:04,  2.66it/s]
 51%|█████     | 5409/10682 [49:44<33:03,  2.66it/s]
 51%|█████     | 5410/10682 [49:44<33:02,  2.66it/s]
 51%|█████     | 5411/10682 [49:45<33:02,  2.66it/s]
 51%|█████     | 5412/10682 [49:45<32:59,  2.66it/s]
 51%|█████     | 5413/10682 [49:46<32:56,  2.67it/s]
 51%|█████     | 5414/10682 [49:46<32:56,  2.67it/s]
 51%|█████     | 5415/10682 [49:46<32:55,  2.67it/s]
 51%|█████     | 5416/10682 [49:47<32:54,  2.67it/s]
 51%|█████     | 5417/10682 [49:47<32:55,  2.67it/s]
 51%|█████     | 5418/10682 [49:48<32:57,  2.66it/s]
 51%|█████     | 5419/10682 [49:48<32:55,  2.66it/s]
 51%|█████     | 5420/10682 [49:48<32:57,  2.66it/s]
 51%|█████     | 5421/10682 [49:49<32:56,  2.66it/s]
 51%|█████     | 5422/10682 [49:49<32:56,  2.66it/s]
 51%|█████     | 5423/10682 [49:49<32:53,  2.66it/s]
 51%|█████     | 5424/10682 [49:50<32:52,  2.67it/s]
 51%|█████     | 5425/10682 [49:50<32:52,  2.67it/s]{'loss': 4.3042, 'grad_norm': 0.4233202040195465, 'learning_rate': 0.0005733475583803009, 'epoch': 0.51}
                                                    
-
 51%|█████     | 5425/10682 [49:50<32:52,  2.67it/s]
 51%|█████     | 5426/10682 [49:51<32:57,  2.66it/s]
 51%|█████     | 5427/10682 [49:51<32:52,  2.66it/s]
 51%|█████     | 5428/10682 [49:51<32:50,  2.67it/s]
 51%|█████     | 5429/10682 [49:52<32:51,  2.66it/s]
 51%|█████     | 5430/10682 [49:52<32:50,  2.67it/s]
 51%|█████     | 5431/10682 [49:52<32:50,  2.66it/s]
 51%|█████     | 5432/10682 [49:53<32:50,  2.66it/s]
 51%|██���██     | 5433/10682 [49:53<32:54,  2.66it/s]
 51%|█████     | 5434/10682 [49:54<32:50,  2.66it/s]
 51%|█████     | 5435/10682 [49:54<32:48,  2.67it/s]
 51%|█████     | 5436/10682 [49:54<32:45,  2.67it/s]
 51%|█████     | 5437/10682 [49:55<32:48,  2.66it/s]
 51%|█████     | 5438/10682 [49:55<32:50,  2.66it/s]
 51%|█████     | 5439/10682 [49:55<32:49,  2.66it/s]
 51%|█████     | 5440/10682 [49:56<32:45,  2.67it/s]
 51%|█████     | 5441/10682 [49:56<32:46,  2.67it/s]
 51%|█████     | 5442/10682 [49:57<32:43,  2.67it/s]
 51%|█████     | 5443/10682 [49:57<32:41,  2.67it/s]
 51%|█████     | 5444/10682 [49:57<32:46,  2.66it/s]
 51%|█████     | 5445/10682 [49:58<32:44,  2.67it/s]
 51%|█████     | 5446/10682 [49:58<32:41,  2.67it/s]
 51%|█████     | 5447/10682 [49:58<32:44,  2.67it/s]
 51%|█████     | 5448/10682 [49:59<32:45,  2.66it/s]
 51%|█████     | 5449/10682 [49:59<32:42,  2.67it/s]
 51%|█████     | 5450/10682 [50:00<32:41,  2.67it/s]
                                                    
{'loss': 4.2981, 'grad_norm': 0.47275081276893616, 'learning_rate': 0.0005693042652030427, 'epoch': 0.51}
-
 51%|█████     | 5450/10682 [50:00<32:41,  2.67it/s]
 51%|█████     | 5451/10682 [50:00<32:43,  2.66it/s]
 51%|█████     | 5452/10682 [50:00<32:41,  2.67it/s]
 51%|█████     | 5453/10682 [50:01<32:41,  2.67it/s]
 51%|█████     | 5454/10682 [50:01<32:40,  2.67it/s]
 51%|█████     | 5455/10682 [50:01<32:40,  2.67it/s]
 51%|█████     | 5456/10682 [50:02<32:38,  2.67it/s]
 51%|█████     | 5457/10682 [50:02<32:38,  2.67it/s]
 51%|█████     | 5458/10682 [50:03<32:40,  2.66it/s]
 51%|█████     | 5459/10682 [50:03<32:38,  2.67it/s]
 51%|█████     | 5460/10682 [50:03<32:37,  2.67it/s]
 51%|█████     | 5461/10682 [50:04<32:40,  2.66it/s]
 51%|█████     | 5462/10682 [50:04<32:39,  2.66it/s]
 51%|█████     | 5463/10682 [50:04<32:39,  2.66it/s]
 51%|█████     | 5464/10682 [50:05<32:36,  2.67it/s]
 51%|█████     | 5465/10682 [50:05<32:35,  2.67it/s]
 51%|█████     | 5466/10682 [50:06<32:33,  2.67it/s]
 51%|█████     | 5467/10682 [50:06<32:35,  2.67it/s]
 51%|█████     | 5468/10682 [50:06<32:36,  2.67it/s]
 51%|█████     | 5469/10682 [50:07<32:36,  2.66it/s]
 51%|█████     | 5470/10682 [50:07<32:34,  2.67it/s]
 51%|█████     | 5471/10682 [50:07<32:36,  2.66it/s]
 51%|█████     | 5472/10682 [50:08<32:33,  2.67it/s]
 51%|█████     | 5473/10682 [50:08<32:30,  2.67it/s]
 51%|█████     | 5474/10682 [50:09<32:33,  2.67it/s]
 51%|█████▏    | 5475/10682 [50:09<32:33,  2.66it/s]
                                                    
{'loss': 4.3049, 'grad_norm': 0.44021129608154297, 'learning_rate': 0.0005652563458788184, 'epoch': 0.51}
-
 51%|█████▏    | 5475/10682 [50:09<32:33,  2.66it/s]
 51%|█████▏    | 5476/10682 [50:09<32:36,  2.66it/s]
 51%|█████▏    | 5477/10682 [50:10<32:32,  2.67it/s]
 51%|█████▏    | 5478/10682 [50:10<32:31,  2.67it/s]
 51%|█████▏    | 5479/10682 [50:10<32:29,  2.67it/s]
 51%|█████▏    | 5480/10682 [50:11<32:28,  2.67it/s]
 51%|█████▏    | 5481/10682 [50:11<32:28,  2.67it/s]
 51%|█████▏    | 5482/10682 [50:12<32:29,  2.67it/s]
 51%|█████▏    | 5483/10682 [50:12<32:29,  2.67it/s]
 51%|█████▏    | 5484/10682 [50:12<32:33,  2.66it/s]
 51%|█████▏    | 5485/10682 [50:13<32:32,  2.66it/s]
 51%|█████▏    | 5486/10682 [50:13<32:30,  2.66it/s]
 51%|█████▏    | 5487/10682 [50:13<32:32,  2.66it/s]
 51%|█████▏    | 5488/10682 [50:14<32:30,  2.66it/s]
 51%|█████▏    | 5489/10682 [50:14<32:31,  2.66it/s]
 51%|█████▏    | 5490/10682 [50:15<32:29,  2.66it/s]
 51%|█████▏    | 5491/10682 [50:15<32:29,  2.66it/s]
 51%|█████▏    | 5492/10682 [50:15<32:29,  2.66it/s]
 51%|█████▏    | 5493/10682 [50:16<32:27,  2.67it/s]
 51%|█████▏    | 5494/10682 [50:16<32:25,  2.67it/s]
 51%|█████▏    | 5495/10682 [50:16<32:25,  2.67it/s]
 51%|█████▏    | 5496/10682 [50:17<32:24,  2.67it/s]
 51%|█████▏    | 5497/10682 [50:17<32:23,  2.67it/s]
 51%|█████▏    | 5498/10682 [50:18<32:22,  2.67it/s]
 51%|█████▏    | 5499/10682 [50:18<32:22,  2.67it/s]
 51%|█████▏    | 5500/10682 [50:18<32:23,  2.67it/s]
                                                    
{'loss': 4.2953, 'grad_norm': 0.4174872636795044, 'learning_rate': 0.0005612040706113396, 'epoch': 0.51}
-
 51%|█████▏    | 5500/10682 [50:18<32:23,  2.67it/s]
 51%|█████▏    | 5501/10682 [50:19<32:26,  2.66it/s]
 52%|█████▏    | 5502/10682 [50:19<32:25,  2.66it/s]
 52%|█████▏    | 5503/10682 [50:19<32:23,  2.66it/s]
 52%|█████▏    | 5504/10682 [50:20<32:23,  2.66it/s]
 52%|█████▏    | 5505/10682 [50:20<32:21,  2.67it/s]
 52%|█████▏    | 5506/10682 [50:21<32:19,  2.67it/s]
 52%|█████▏    | 5507/10682 [50:21<32:19,  2.67it/s]
 52%|█████▏    | 5508/10682 [50:21<32:20,  2.67it/s]
 52%|█████▏    | 5509/10682 [50:22<32:21,  2.66it/s]
 52%|█████▏    | 5510/10682 [50:22<32:21,  2.66it/s]
 52%|█████▏    | 5511/10682 [50:22<32:19,  2.67it/s]
 52%|█████▏    | 5512/10682 [50:23<32:20,  2.66it/s]
 52%|█████▏    | 5513/10682 [50:23<32:20,  2.66it/s]
 52%|█████▏    | 5514/10682 [50:24<32:23,  2.66it/s]
 52%|█████▏    | 5515/10682 [50:24<32:22,  2.66it/s]
 52%|█████▏    | 5516/10682 [50:24<32:18,  2.66it/s]
 52%|█████▏    | 5517/10682 [50:25<32:16,  2.67it/s]
 52%|█████▏    | 5518/10682 [50:25<32:13,  2.67it/s]
 52%|█████▏    | 5519/10682 [50:25<32:15,  2.67it/s]
 52%|█████▏    | 5520/10682 [50:26<32:14,  2.67it/s]
 52%|█████▏    | 5521/10682 [50:26<32:14,  2.67it/s]
 52%|█████▏    | 5522/10682 [50:27<32:13,  2.67it/s]
 52%|█████▏    | 5523/10682 [50:27<32:13,  2.67it/s]
 52%|█████▏    | 5524/10682 [50:27<32:12,  2.67it/s]
 52%|█████▏    | 5525/10682 [50:28<32:16,  2.66it/s]{'loss': 4.301, 'grad_norm': 0.40748897194862366, 'learning_rate': 0.0005571477098950832, 'epoch': 0.52}
                                                    
-
 52%|█████▏    | 5525/10682 [50:28<32:16,  2.66it/s]
 52%|█████▏    | 5526/10682 [50:28<32:23,  2.65it/s]
 52%|█████▏    | 5527/10682 [50:28<32:24,  2.65it/s]
 52%|█████▏    | 5528/10682 [50:29<32:20,  2.66it/s]
 52%|█████▏    | 5529/10682 [50:29<32:18,  2.66it/s]
 52%|█████▏    | 5530/10682 [50:30<32:15,  2.66it/s]
 52%|█████▏    | 5531/10682 [50:30<32:13,  2.66it/s]
 52%|█████▏    | 5532/10682 [50:30<32:12,  2.66it/s]
 52%|█████▏    | 5533/10682 [50:31<32:12,  2.66it/s]
 52%|█████▏    | 5534/10682 [50:31<32:09,  2.67it/s]
 52%|█████▏    | 5535/10682 [50:31<32:11,  2.67it/s]
 52%|█████▏    | 5536/10682 [50:32<32:09,  2.67it/s]
 52%|█████▏    | 5537/10682 [50:32<32:09,  2.67it/s]
 52%|█████▏    | 5538/10682 [50:33<32:11,  2.66it/s]
 52%|█████▏    | 5539/10682 [50:33<32:10,  2.66it/s]
 52%|█████▏    | 5540/10682 [50:33<32:07,  2.67it/s]
 52%|█████▏    | 5541/10682 [50:34<32:11,  2.66it/s]
 52%|█████▏    | 5542/10682 [50:34<32:09,  2.66it/s]
 52%|█████▏    | 5543/10682 [50:34<32:08,  2.66it/s]
 52%|█████▏    | 5544/10682 [50:35<32:07,  2.67it/s]
 52%|█████▏    | 5545/10682 [50:35<32:06,  2.67it/s]
 52%|█████▏    | 5546/10682 [50:36<32:07,  2.66it/s]
 52%|█████▏    | 5547/10682 [50:36<32:05,  2.67it/s]
 52%|█████▏    | 5548/10682 [50:36<32:04,  2.67it/s]
 52%|█████▏    | 5549/10682 [50:37<32:02,  2.67it/s]
 52%|█████▏    | 5550/10682 [50:37<32:02,  2.67it/s]{'loss': 4.2969, 'grad_norm': 0.4338599145412445, 'learning_rate': 0.0005530875344972343, 'epoch': 0.52}
                                                    
-
 52%|█████▏    | 5550/10682 [50:37<32:02,  2.67it/s]
 52%|█████▏    | 5551/10682 [50:37<32:15,  2.65it/s]
 52%|█████▏    | 5552/10682 [50:38<32:14,  2.65it/s]
 52%|█████▏    | 5553/10682 [50:38<32:11,  2.66it/s]
 52%|█████▏    | 5554/10682 [50:39<32:08,  2.66it/s]
 52%|█████▏    | 5555/10682 [50:39<32:13,  2.65it/s]
 52%|█████▏    | 5556/10682 [50:39<32:10,  2.66it/s]
 52%|█████▏    | 5557/10682 [50:40<32:08,  2.66it/s]
 52%|█████▏    | 5558/10682 [50:40<32:04,  2.66it/s]
 52%|█████▏    | 5559/10682 [50:40<32:03,  2.66it/s]
 52%|█████▏    | 5560/10682 [50:41<32:00,  2.67it/s]
 52%|█████▏    | 5561/10682 [50:41<32:02,  2.66it/s]
 52%|█████▏    | 5562/10682 [50:42<32:01,  2.66it/s]
 52%|█████▏    | 5563/10682 [50:42<32:01,  2.66it/s]
 52%|█████▏    | 5564/10682 [50:42<32:00,  2.66it/s]
 52%|█████▏    | 5565/10682 [50:43<32:00,  2.66it/s]
 52%|█████▏    | 5566/10682 [50:43<31:59,  2.67it/s]
 52%|█████▏    | 5567/10682 [50:43<31:59,  2.66it/s]
 52%|█████▏    | 5568/10682 [50:44<31:59,  2.66it/s]
 52%|█████▏    | 5569/10682 [50:44<31:58,  2.67it/s]
 52%|█████▏    | 5570/10682 [50:45<31:58,  2.67it/s]
 52%|█████▏    | 5571/10682 [50:45<31:55,  2.67it/s]
 52%|█████▏    | 5572/10682 [50:45<31:56,  2.67it/s]
 52%|█████▏    | 5573/10682 [50:46<31:54,  2.67it/s]
 52%|█████▏    | 5574/10682 [50:46<31:52,  2.67it/s]
 52%|█████▏    | 5575/10682 [50:46<31:53,  2.67it/s]{'loss': 4.2894, 'grad_norm': 0.4097755253314972, 'learning_rate': 0.0005490238154396129, 'epoch': 0.52}
                                                    
-
 52%|█████▏    | 5575/10682 [50:46<31:53,  2.67it/s]
 52%|█████▏    | 5576/10682 [50:47<31:57,  2.66it/s]
 52%|█████▏    | 5577/10682 [50:47<31:54,  2.67it/s]
 52%|█████▏    | 5578/10682 [50:48<31:52,  2.67it/s]
 52%|█████▏    | 5579/10682 [50:48<31:55,  2.66it/s]
 52%|█████▏    | 5580/10682 [50:48<31:55,  2.66it/s]
 52%|█████▏    | 5581/10682 [50:49<31:59,  2.66it/s]
 52%|█████▏    | 5582/10682 [50:49<31:57,  2.66it/s]
 52%|█████▏    | 5583/10682 [50:49<31:54,  2.66it/s]
 52%|█████▏    | 5584/10682 [50:50<31:52,  2.67it/s]
 52%|█████▏    | 5585/10682 [50:50<31:53,  2.66it/s]
 52%|█████▏    | 5586/10682 [50:51<31:50,  2.67it/s]
 52%|█████▏    | 5587/10682 [50:51<31:48,  2.67it/s]
 52%|█████▏    | 5588/10682 [50:51<31:47,  2.67it/s]
 52%|█████▏    | 5589/10682 [50:52<31:49,  2.67it/s]
 52%|█████▏    | 5590/10682 [50:52<31:50,  2.66it/s]
 52%|█████▏    | 5591/10682 [50:52<31:48,  2.67it/s]
 52%|█████▏    | 5592/10682 [50:53<31:50,  2.66it/s]
 52%|█████▏    | 5593/10682 [50:53<31:50,  2.66it/s]
 52%|█████▏    | 5594/10682 [50:54<31:54,  2.66it/s]
 52%|█████▏    | 5595/10682 [50:54<31:55,  2.66it/s]
 52%|█████▏    | 5596/10682 [50:54<31:52,  2.66it/s]
 52%|█████▏    | 5597/10682 [50:55<31:48,  2.66it/s]
 52%|█████▏    | 5598/10682 [50:55<31:47,  2.66it/s]
 52%|█████▏    | 5599/10682 [50:55<31:47,  2.66it/s]
 52%|█████▏    | 5600/10682 [50:56<31:46,  2.67it/s]{'loss': 4.2926, 'grad_norm': 0.41281068325042725, 'learning_rate': 0.0005449568239805831, 'epoch': 0.52}
                                                    
-
 52%|█████▏    | 5600/10682 [50:56<31:46,  2.67it/s]
 52%|█████▏    | 5601/10682 [50:56<31:50,  2.66it/s]
 52%|█████▏    | 5602/10682 [50:57<31:48,  2.66it/s]
 52%|█████▏    | 5603/10682 [50:57<31:46,  2.66it/s]
 52%|█████▏    | 5604/10682 [50:57<31:45,  2.66it/s]
 52%|█████▏    | 5605/10682 [50:58<31:47,  2.66it/s]
 52%|█████▏    | 5606/10682 [50:58<31:44,  2.66it/s]
 52%|█████▏    | 5607/10682 [50:58<31:43,  2.67it/s]
 52%|█████▏    | 5608/10682 [50:59<31:44,  2.66it/s]
 53%|█████▎    | 5609/10682 [50:59<31:43,  2.66it/s]
 53%|█████▎    | 5610/10682 [51:00<31:42,  2.67it/s]
 53%|█████▎    | 5611/10682 [51:00<31:46,  2.66it/s]
 53%|█████▎    | 5612/10682 [51:00<31:45,  2.66it/s]
 53%|█████▎    | 5613/10682 [51:01<31:43,  2.66it/s]
 53%|█████▎    | 5614/10682 [51:01<31:42,  2.66it/s]
 53%|█████▎    | 5615/10682 [51:01<31:43,  2.66it/s]
 53%|█████▎    | 5616/10682 [51:02<31:40,  2.67it/s]
 53%|█████▎    | 5617/10682 [51:02<31:40,  2.67it/s]
 53%|█████▎    | 5618/10682 [51:03<31:40,  2.66it/s]
 53%|█████▎    | 5619/10682 [51:03<31:40,  2.66it/s]
 53%|█████▎    | 5620/10682 [51:03<31:39,  2.67it/s]
 53%|█████▎    | 5621/10682 [51:04<31:41,  2.66it/s]
 53%|█████▎    | 5622/10682 [51:04<31:39,  2.66it/s]
 53%|█████▎    | 5623/10682 [51:04<31:39,  2.66it/s]
 53%|█████▎    | 5624/10682 [51:05<31:39,  2.66it/s]
 53%|█████▎    | 5625/10682 [51:05<31:37,  2.67it/s]{'loss': 4.2994, 'grad_norm': 0.4388512074947357, 'learning_rate': 0.0005408868315969458, 'epoch': 0.53}
                                                    
-
 53%|█████▎    | 5625/10682 [51:05<31:37,  2.67it/s]
 53%|█████▎    | 5626/10682 [51:06<31:38,  2.66it/s]
 53%|█████▎    | 5627/10682 [51:06<31:38,  2.66it/s]
 53%|█████▎    | 5628/10682 [51:06<31:35,  2.67it/s]
 53%|█████▎    | 5629/10682 [51:07<31:34,  2.67it/s]
 53%|█████▎    | 5630/10682 [51:07<31:33,  2.67it/s]
 53%|█████▎    | 5631/10682 [51:07<31:31,  2.67it/s]
 53%|█████▎    | 5632/10682 [51:08<31:32,  2.67it/s]
 53%|█████▎    | 5633/10682 [51:08<31:32,  2.67it/s]
 53%|█████▎    | 5634/10682 [51:09<31:32,  2.67it/s]
 53%|█████▎    | 5635/10682 [51:09<31:36,  2.66it/s]
 53%|█████▎    | 5636/10682 [51:09<31:33,  2.67it/s]
 53%|█████▎    | 5637/10682 [51:10<31:31,  2.67it/s]
 53%|█████▎    | 5638/10682 [51:10<31:30,  2.67it/s]
 53%|█████▎    | 5639/10682 [51:10<31:32,  2.66it/s]
 53%|█████▎    | 5640/10682 [51:11<31:31,  2.67it/s]
 53%|█████▎    | 5641/10682 [51:11<31:29,  2.67it/s]
 53%|█████▎    | 5642/10682 [51:12<31:27,  2.67it/s]
 53%|█████▎    | 5643/10682 [51:12<31:29,  2.67it/s]
 53%|█████▎    | 5644/10682 [51:12<31:30,  2.67it/s]
 53%|█████▎    | 5645/10682 [51:13<31:29,  2.67it/s]
 53%|█████▎    | 5646/10682 [51:13<31:30,  2.66it/s]
 53%|█████▎    | 5647/10682 [51:13<31:32,  2.66it/s]
 53%|█████▎    | 5648/10682 [51:14<31:30,  2.66it/s]
 53%|█████▎    | 5649/10682 [51:14<31:29,  2.66it/s]
 53%|█████▎    | 5650/10682 [51:15<31:28,  2.66it/s]{'loss': 4.2968, 'grad_norm': 0.4257469177246094, 'learning_rate': 0.0005368141099658168, 'epoch': 0.53}
-
                                                    

 53%|█████▎    | 5650/10682 [51:15<31:28,  2.66it/s]
 53%|█████▎    | 5651/10682 [51:15<31:31,  2.66it/s]
 53%|█████▎    | 5652/10682 [51:15<31:29,  2.66it/s]
 53%|█████▎    | 5653/10682 [51:16<31:27,  2.66it/s]
 53%|█████▎    | 5654/10682 [51:16<31:29,  2.66it/s]
 53%|█████▎    | 5655/10682 [51:16<31:30,  2.66it/s]
 53%|█████▎    | 5656/10682 [51:17<31:28,  2.66it/s]
 53%|█████▎    | 5657/10682 [51:17<31:26,  2.66it/s]
 53%|█████▎    | 5658/10682 [51:18<31:25,  2.66it/s]
 53%|█████▎    | 5659/10682 [51:18<31:31,  2.66it/s]
 53%|█████▎    | 5660/10682 [51:18<31:26,  2.66it/s]
 53%|█████▎    | 5661/10682 [51:19<31:27,  2.66it/s]
 53%|█████▎    | 5662/10682 [51:19<31:26,  2.66it/s]
 53%|█████▎    | 5663/10682 [51:19<31:27,  2.66it/s]
 53%|█████▎    | 5664/10682 [51:20<31:25,  2.66it/s]
 53%|█████▎    | 5665/10682 [51:20<31:26,  2.66it/s]
 53%|█████▎    | 5666/10682 [51:21<31:24,  2.66it/s]
 53%|█████▎    | 5667/10682 [51:21<31:23,  2.66it/s]
 53%|█████▎    | 5668/10682 [51:21<31:20,  2.67it/s]
 53%|█████▎    | 5669/10682 [51:22<31:20,  2.67it/s]
 53%|█████▎    | 5670/10682 [51:22<31:21,  2.66it/s]
 53%|█████▎    | 5671/10682 [51:22<31:21,  2.66it/s]
 53%|█████▎    | 5672/10682 [51:23<31:22,  2.66it/s]
 53%|█████▎    | 5673/10682 [51:23<31:23,  2.66it/s]
 53%|█████▎    | 5674/10682 [51:24<31:22,  2.66it/s]
 53%|█████▎    | 5675/10682 [51:24<31:26,  2.65it/s]{'loss': 4.2926, 'grad_norm': 0.44102925062179565, 'learning_rate': 0.0005327389309464938, 'epoch': 0.53}
                                                    
-
 53%|█████▎    | 5675/10682 [51:24<31:26,  2.65it/s]
 53%|█████▎    | 5676/10682 [51:24<31:26,  2.65it/s]
 53%|█████▎    | 5677/10682 [51:25<31:23,  2.66it/s]
 53%|█████▎    | 5678/10682 [51:25<31:20,  2.66it/s]
 53%|█████▎    | 5679/10682 [51:25<31:20,  2.66it/s]
 53%|█████▎    | 5680/10682 [51:26<31:20,  2.66it/s]
 53%|█████▎    | 5681/10682 [51:26<31:15,  2.67it/s]
 53%|█████▎    | 5682/10682 [51:27<31:15,  2.67it/s]
 53%|█████▎    | 5683/10682 [51:27<31:13,  2.67it/s]
 53%|█████▎    | 5684/10682 [51:27<31:15,  2.67it/s]
 53%|█████▎    | 5685/10682 [51:28<31:14,  2.67it/s]
 53%|█████▎    | 5686/10682 [51:28<31:17,  2.66it/s]
 53%|█████▎    | 5687/10682 [51:28<31:17,  2.66it/s]
 53%|█████▎    | 5688/10682 [51:29<31:19,  2.66it/s]
 53%|█████▎    | 5689/10682 [51:29<31:17,  2.66it/s]
 53%|█████▎    | 5690/10682 [51:30<31:18,  2.66it/s]
 53%|█████▎    | 5691/10682 [51:30<31:17,  2.66it/s]
 53%|█████▎    | 5692/10682 [51:30<31:15,  2.66it/s]
 53%|█████▎    | 5693/10682 [51:31<31:15,  2.66it/s]
 53%|█████▎    | 5694/10682 [51:31<31:13,  2.66it/s]
 53%|█████▎    | 5695/10682 [51:31<31:12,  2.66it/s]
 53%|█████▎    | 5696/10682 [51:32<31:11,  2.66it/s]
 53%|█████▎    | 5697/10682 [51:32<31:13,  2.66it/s]
 53%|█████▎    | 5698/10682 [51:33<31:14,  2.66it/s]
 53%|█████▎    | 5699/10682 [51:33<31:15,  2.66it/s]
 53%|█████▎    | 5700/10682 [51:33<31:14,  2.66it/s]{'loss': 4.291, 'grad_norm': 0.40123507380485535, 'learning_rate': 0.0005286615665623071, 'epoch': 0.53}
-
                                                    

 53%|█████▎    | 5700/10682 [51:33<31:14,  2.66it/s]
 53%|█████▎    | 5701/10682 [51:34<31:17,  2.65it/s]
 53%|█████▎    | 5702/10682 [51:34<31:14,  2.66it/s]
 53%|█████▎    | 5703/10682 [51:34<31:12,  2.66it/s]
 53%|█████▎    | 5704/10682 [51:35<31:13,  2.66it/s]
 53%|█████▎    | 5705/10682 [51:35<31:11,  2.66it/s]
 53%|█████▎    | 5706/10682 [51:36<31:09,  2.66it/s]
 53%|█████▎    | 5707/10682 [51:36<31:08,  2.66it/s]
 53%|█████▎    | 5708/10682 [51:36<31:07,  2.66it/s]
 53%|█████▎    | 5709/10682 [51:37<31:05,  2.67it/s]
 53%|█████▎    | 5710/10682 [51:37<31:03,  2.67it/s]
 53%|█████▎    | 5711/10682 [51:37<31:02,  2.67it/s]
 53%|█████▎    | 5712/10682 [51:38<31:07,  2.66it/s]
 53%|█████▎    | 5713/10682 [51:38<31:06,  2.66it/s]
 53%|█████▎    | 5714/10682 [51:39<31:09,  2.66it/s]
 54%|█████▎    | 5715/10682 [51:39<31:08,  2.66it/s]
 54%|█████▎    | 5716/10682 [51:39<31:06,  2.66it/s]
 54%|█████▎    | 5717/10682 [51:40<31:05,  2.66it/s]
 54%|█████▎    | 5718/10682 [51:40<31:04,  2.66it/s]
 54%|█████▎    | 5719/10682 [51:40<31:04,  2.66it/s]
 54%|█████▎    | 5720/10682 [51:41<31:02,  2.66it/s]
 54%|█████▎    | 5721/10682 [51:41<30:59,  2.67it/s]
 54%|█████▎    | 5722/10682 [51:42<30:58,  2.67it/s]
 54%|█████▎    | 5723/10682 [51:42<31:00,  2.67it/s]
 54%|█████▎    | 5724/10682 [51:42<30:58,  2.67it/s]
 54%|█████▎    | 5725/10682 [51:43<30:58,  2.67it/s]{'loss': 4.282, 'grad_norm': 0.4555445909500122, 'learning_rate': 0.0005245822889824633, 'epoch': 0.54}
                                                    
-
 54%|█████▎    | 5725/10682 [51:43<30:58,  2.67it/s]
 54%|█████▎    | 5726/10682 [51:43<31:04,  2.66it/s]
 54%|█████▎    | 5727/10682 [51:44<31:03,  2.66it/s]
 54%|█████▎    | 5728/10682 [51:44<31:03,  2.66it/s]
 54%|█████▎    | 5729/10682 [51:44<31:01,  2.66it/s]
 54%|█████▎    | 5730/10682 [51:45<31:00,  2.66it/s]
 54%|█████▎    | 5731/10682 [51:45<31:01,  2.66it/s]
 54%|█████▎    | 5732/10682 [51:45<31:01,  2.66it/s]
 54%|█████▎    | 5733/10682 [51:46<31:01,  2.66it/s]
 54%|█████▎    | 5734/10682 [51:46<30:58,  2.66it/s]
 54%|█████▎    | 5735/10682 [51:47<30:57,  2.66it/s]
 54%|█████▎    | 5736/10682 [51:47<30:58,  2.66it/s]
 54%|█████▎    | 5737/10682 [51:47<30:56,  2.66it/s]
 54%|█████▎    | 5738/10682 [51:48<30:55,  2.67it/s]
 54%|█████▎    | 5739/10682 [51:48<30:57,  2.66it/s]
 54%|█████▎    | 5740/10682 [51:48<30:57,  2.66it/s]
 54%|█████▎    | 5741/10682 [51:49<30:56,  2.66it/s]
 54%|█████▍    | 5742/10682 [51:49<30:53,  2.66it/s]
 54%|█████▍    | 5743/10682 [51:50<30:54,  2.66it/s]
 54%|█████▍    | 5744/10682 [51:50<30:53,  2.66it/s]
 54%|█████▍    | 5745/10682 [51:50<30:52,  2.66it/s]
 54%|█████▍    | 5746/10682 [51:51<30:52,  2.66it/s]
 54%|█████▍    | 5747/10682 [51:51<30:50,  2.67it/s]
 54%|█████▍    | 5748/10682 [51:51<30:49,  2.67it/s]
 54%|█████▍    | 5749/10682 [51:52<30:49,  2.67it/s]
 54%|█████▍    | 5750/10682 [51:52<30:48,  2.67it/s]
                                                    
{'loss': 4.29, 'grad_norm': 0.4346594512462616, 'learning_rate': 0.0005205013705038776, 'epoch': 0.54}
-
 54%|█████▍    | 5750/10682 [51:52<30:48,  2.67it/s]
 54%|█████▍    | 5751/10682 [51:53<30:54,  2.66it/s]
 54%|█████▍    | 5752/10682 [51:53<30:54,  2.66it/s]
 54%|█████▍    | 5753/10682 [51:53<30:52,  2.66it/s]
 54%|█████▍    | 5754/10682 [51:54<30:50,  2.66it/s]
 54%|█████▍    | 5755/10682 [51:54<30:53,  2.66it/s]
 54%|█████▍    | 5756/10682 [51:54<30:50,  2.66it/s]
 54%|█████▍    | 5757/10682 [51:55<30:47,  2.67it/s]
 54%|█████▍    | 5758/10682 [51:55<30:45,  2.67it/s]
 54%|█████▍    | 5759/10682 [51:56<30:45,  2.67it/s]
 54%|█████▍    | 5760/10682 [51:56<30:46,  2.67it/s]
 54%|█████▍    | 5761/10682 [51:56<30:45,  2.67it/s]
 54%|█████▍    | 5762/10682 [51:57<30:44,  2.67it/s]
 54%|█████▍    | 5763/10682 [51:57<30:44,  2.67it/s]
 54%|█████▍    | 5764/10682 [51:57<30:44,  2.67it/s]
 54%|█████▍    | 5765/10682 [51:58<30:42,  2.67it/s]
 54%|█████▍    | 5766/10682 [51:58<30:43,  2.67it/s]
 54%|█████▍    | 5767/10682 [51:59<30:42,  2.67it/s]
 54%|█████▍    | 5768/10682 [51:59<30:44,  2.66it/s]
 54%|█████▍    | 5769/10682 [51:59<30:46,  2.66it/s]
 54%|█████▍    | 5770/10682 [52:00<30:43,  2.66it/s]
 54%|█████▍    | 5771/10682 [52:00<30:42,  2.67it/s]
 54%|█████▍    | 5772/10682 [52:00<30:42,  2.67it/s]
 54%|█████▍    | 5773/10682 [52:01<30:39,  2.67it/s]
 54%|█████▍    | 5774/10682 [52:01<30:40,  2.67it/s]
 54%|█████▍    | 5775/10682 [52:02<30:37,  2.67it/s]{'loss': 4.2855, 'grad_norm': 0.45139801502227783, 'learning_rate': 0.0005164190835329964, 'epoch': 0.54}
                                                    
-
 54%|█████▍    | 5775/10682 [52:02<30:37,  2.67it/s]
 54%|█████▍    | 5776/10682 [52:02<30:42,  2.66it/s]
 54%|█████▍    | 5777/10682 [52:02<30:42,  2.66it/s]
 54%|█████▍    | 5778/10682 [52:03<30:42,  2.66it/s]
 54%|█████▍    | 5779/10682 [52:03<30:42,  2.66it/s]
 54%|█████▍    | 5780/10682 [52:03<30:40,  2.66it/s]
 54%|█████▍    | 5781/10682 [52:04<30:40,  2.66it/s]
 54%|█████▍    | 5782/10682 [52:04<30:43,  2.66it/s]
 54%|█████▍    | 5783/10682 [52:05<30:45,  2.65it/s]
 54%|█████▍    | 5784/10682 [52:05<30:42,  2.66it/s]
 54%|█████▍    | 5785/10682 [52:05<30:40,  2.66it/s]
 54%|█████▍    | 5786/10682 [52:06<30:41,  2.66it/s]
 54%|█████▍    | 5787/10682 [52:06<30:40,  2.66it/s]
 54%|█████▍    | 5788/10682 [52:06<30:37,  2.66it/s]
 54%|█████▍    | 5789/10682 [52:07<30:34,  2.67it/s]
 54%|█████▍    | 5790/10682 [52:07<30:34,  2.67it/s]
 54%|█████▍    | 5791/10682 [52:08<30:33,  2.67it/s]
 54%|█████▍    | 5792/10682 [52:08<30:33,  2.67it/s]
 54%|█████▍    | 5793/10682 [52:08<30:33,  2.67it/s]
 54%|█████▍    | 5794/10682 [52:09<30:33,  2.67it/s]
 54%|█████▍    | 5795/10682 [52:09<30:36,  2.66it/s]
 54%|█████▍    | 5796/10682 [52:09<30:39,  2.66it/s]
 54%|█████▍    | 5797/10682 [52:10<30:38,  2.66it/s]
 54%|█████▍    | 5798/10682 [52:10<30:36,  2.66it/s]
 54%|█████▍    | 5799/10682 [52:11<30:33,  2.66it/s]
 54%|█████▍    | 5800/10682 [52:11<30:31,  2.66it/s]{'loss': 4.2878, 'grad_norm': 0.40731754899024963, 'learning_rate': 0.0005123357005676156, 'epoch': 0.54}
                                                    
-
 54%|█████▍    | 5800/10682 [52:11<30:31,  2.66it/s]
 54%|█████▍    | 5801/10682 [52:11<30:33,  2.66it/s]
 54%|█████▍    | 5802/10682 [52:12<30:30,  2.67it/s]
 54%|█████▍    | 5803/10682 [52:12<30:28,  2.67it/s]
 54%|█████▍    | 5804/10682 [52:12<30:41,  2.65it/s]
 54%|█████▍    | 5805/10682 [52:13<30:37,  2.65it/s]
 54%|█████▍    | 5806/10682 [52:13<30:35,  2.66it/s]
 54%|█████▍    | 5807/10682 [52:14<30:31,  2.66it/s]
 54%|█████▍    | 5808/10682 [52:14<30:33,  2.66it/s]
 54%|█████▍    | 5809/10682 [52:14<30:30,  2.66it/s]
 54%|█████▍    | 5810/10682 [52:15<30:27,  2.67it/s]
 54%|█████▍    | 5811/10682 [52:15<30:26,  2.67it/s]
 54%|█████▍    | 5812/10682 [52:15<30:24,  2.67it/s]
 54%|█████▍    | 5813/10682 [52:16<30:24,  2.67it/s]
 54%|█████▍    | 5814/10682 [52:16<30:22,  2.67it/s]
 54%|█████▍    | 5815/10682 [52:17<30:21,  2.67it/s]
 54%|█████▍    | 5816/10682 [52:17<30:20,  2.67it/s]
 54%|█████▍    | 5817/10682 [52:17<30:20,  2.67it/s]
 54%|█████▍    | 5818/10682 [52:18<30:20,  2.67it/s]
 54%|█████▍    | 5819/10682 [52:18<30:23,  2.67it/s]
 54%|█████▍    | 5820/10682 [52:18<30:22,  2.67it/s]
 54%|█████▍    | 5821/10682 [52:19<30:24,  2.66it/s]
 55%|█████▍    | 5822/10682 [52:19<30:23,  2.67it/s]
 55%|█████▍    | 5823/10682 [52:20<30:23,  2.66it/s]
 55%|█████▍    | 5824/10682 [52:20<30:21,  2.67it/s]
 55%|█████▍    | 5825/10682 [52:20<30:21,  2.67it/s]
                                                    
{'loss': 4.2875, 'grad_norm': 0.4336436688899994, 'learning_rate': 0.0005082514941786895, 'epoch': 0.55}
-
 55%|█████▍    | 5825/10682 [52:20<30:21,  2.67it/s]
 55%|█████▍    | 5826/10682 [52:21<30:23,  2.66it/s]
 55%|█████▍    | 5827/10682 [52:21<30:20,  2.67it/s]
 55%|█████▍    | 5828/10682 [52:21<30:24,  2.66it/s]
 55%|█████▍    | 5829/10682 [52:22<30:22,  2.66it/s]
 55%|█████▍    | 5830/10682 [52:22<30:19,  2.67it/s]
 55%|█████▍    | 5831/10682 [52:23<30:17,  2.67it/s]
 55%|█████▍    | 5832/10682 [52:23<30:17,  2.67it/s]
 55%|█████▍    | 5833/10682 [52:23<30:19,  2.66it/s]
 55%|█████▍    | 5834/10682 [52:24<30:20,  2.66it/s]
 55%|█████▍    | 5835/10682 [52:24<30:20,  2.66it/s]
 55%|█████▍    | 5836/10682 [52:24<30:18,  2.67it/s]
 55%|█████▍    | 5837/10682 [52:25<30:18,  2.66it/s]
 55%|█████▍    | 5838/10682 [52:25<30:18,  2.66it/s]
 55%|█████▍    | 5839/10682 [52:26<30:17,  2.66it/s]
 55%|█████▍    | 5840/10682 [52:26<30:16,  2.67it/s]
 55%|█████▍    | 5841/10682 [52:26<30:16,  2.67it/s]
 55%|█████▍    | 5842/10682 [52:27<30:15,  2.67it/s]
 55%|█████▍    | 5843/10682 [52:27<30:14,  2.67it/s]
 55%|█████▍    | 5844/10682 [52:27<30:13,  2.67it/s]
 55%|█████▍    | 5845/10682 [52:28<30:12,  2.67it/s]
 55%|█████▍    | 5846/10682 [52:28<30:12,  2.67it/s]
 55%|█████▍    | 5847/10682 [52:29<30:11,  2.67it/s]
 55%|█████▍    | 5848/10682 [52:29<30:10,  2.67it/s]
 55%|█████▍    | 5849/10682 [52:29<30:11,  2.67it/s]
 55%|█████▍    | 5850/10682 [52:30<30:11,  2.67it/s]{'loss': 4.2798, 'grad_norm': 0.4013449251651764, 'learning_rate': 0.0005041667369921374, 'epoch': 0.55}
                                                    
-
 55%|█████▍    | 5850/10682 [52:30<30:11,  2.67it/s]
 55%|█████▍    | 5851/10682 [52:30<30:13,  2.66it/s]
 55%|█████▍    | 5852/10682 [52:30<30:10,  2.67it/s]
 55%|█████▍    | 5853/10682 [52:31<30:10,  2.67it/s]
 55%|█████▍    | 5854/10682 [52:31<30:09,  2.67it/s]
 55%|█████▍    | 5855/10682 [52:32<30:07,  2.67it/s]
 55%|█████▍    | 5856/10682 [52:32<30:07,  2.67it/s]
 55%|█████▍    | 5857/10682 [52:32<30:06,  2.67it/s]
 55%|█████▍    | 5858/10682 [52:33<30:06,  2.67it/s]
 55%|█████▍    | 5859/10682 [52:33<30:04,  2.67it/s]
 55%|█████▍    | 5860/10682 [52:33<30:08,  2.67it/s]
 55%|█████▍    | 5861/10682 [52:34<30:07,  2.67it/s]
 55%|█████▍    | 5862/10682 [52:34<30:06,  2.67it/s]
 55%|█████▍    | 5863/10682 [52:35<30:05,  2.67it/s]
 55%|█████▍    | 5864/10682 [52:35<30:05,  2.67it/s]
 55%|█████▍    | 5865/10682 [52:35<30:04,  2.67it/s]
 55%|█████▍    | 5866/10682 [52:36<30:03,  2.67it/s]
 55%|█████▍    | 5867/10682 [52:36<30:02,  2.67it/s]
 55%|█████▍    | 5868/10682 [52:36<30:02,  2.67it/s]
 55%|█████▍    | 5869/10682 [52:37<30:01,  2.67it/s]
 55%|█████▍    | 5870/10682 [52:37<30:02,  2.67it/s]
 55%|█████▍    | 5871/10682 [52:38<30:01,  2.67it/s]
 55%|█████▍    | 5872/10682 [52:38<30:00,  2.67it/s]
 55%|█████▍    | 5873/10682 [52:38<30:02,  2.67it/s]
 55%|█████▍    | 5874/10682 [52:39<30:01,  2.67it/s]
 55%|█████▍    | 5875/10682 [52:39<30:02,  2.67it/s]
                                                    {'loss': 4.283, 'grad_norm': 0.42053884267807007, 'learning_rate': 0.0005000817016706441, 'epoch': 0.55}
-

 55%|█████▍    | 5875/10682 [52:39<30:02,  2.67it/s]
 55%|█████▌    | 5876/10682 [52:39<30:05,  2.66it/s]
 55%|█████▌    | 5877/10682 [52:40<30:02,  2.67it/s]
 55%|█████▌    | 5878/10682 [52:40<30:01,  2.67it/s]
 55%|█████▌    | 5879/10682 [52:41<30:00,  2.67it/s]
 55%|█████▌    | 5880/10682 [52:41<30:00,  2.67it/s]
 55%|█████▌    | 5881/10682 [52:41<30:01,  2.66it/s]
 55%|█████▌    | 5882/10682 [52:42<29:58,  2.67it/s]
 55%|█████▌    | 5883/10682 [52:42<29:57,  2.67it/s]
 55%|█████▌    | 5884/10682 [52:42<29:57,  2.67it/s]
 55%|█████▌    | 5885/10682 [52:43<29:58,  2.67it/s]
 55%|█████▌    | 5886/10682 [52:43<29:57,  2.67it/s]
 55%|█████▌    | 5887/10682 [52:44<30:00,  2.66it/s]
 55%|█████▌    | 5888/10682 [52:44<29:58,  2.67it/s]
 55%|█████▌    | 5889/10682 [52:44<29:57,  2.67it/s]
 55%|█████▌    | 5890/10682 [52:45<29:56,  2.67it/s]
 55%|█████▌    | 5891/10682 [52:45<29:56,  2.67it/s]
 55%|█████▌    | 5892/10682 [52:45<29:55,  2.67it/s]
 55%|█████▌    | 5893/10682 [52:46<29:57,  2.66it/s]
 55%|█████▌    | 5894/10682 [52:46<29:56,  2.66it/s]
 55%|█████▌    | 5895/10682 [52:47<29:55,  2.67it/s]
 55%|█████▌    | 5896/10682 [52:47<29:52,  2.67it/s]
 55%|█████▌    | 5897/10682 [52:47<29:50,  2.67it/s]
 55%|█████▌    | 5898/10682 [52:48<29:50,  2.67it/s]
 55%|█████▌    | 5899/10682 [52:48<29:51,  2.67it/s]
 55%|█████▌    | 5900/10682 [52:48<29:52,  2.67it/s]
                                                    
{'loss': 4.2799, 'grad_norm': 0.4086085855960846, 'learning_rate': 0.0004959966608954618, 'epoch': 0.55}
-
 55%|█████▌    | 5900/10682 [52:48<29:52,  2.67it/s]
 55%|█████▌    | 5901/10682 [52:49<29:54,  2.66it/s]
 55%|█████▌    | 5902/10682 [52:49<29:56,  2.66it/s]
 55%|█████▌    | 5903/10682 [52:50<29:57,  2.66it/s]
 55%|█████▌    | 5904/10682 [52:50<29:56,  2.66it/s]
 55%|█████▌    | 5905/10682 [52:50<29:57,  2.66it/s]
 55%|█████▌    | 5906/10682 [52:51<29:53,  2.66it/s]
 55%|█████▌    | 5907/10682 [52:51<29:51,  2.66it/s]
 55%|█████▌    | 5908/10682 [52:51<29:51,  2.66it/s]
 55%|█████▌    | 5909/10682 [52:52<29:48,  2.67it/s]
 55%|█████▌    | 5910/10682 [52:52<29:48,  2.67it/s]
 55%|█████▌    | 5911/10682 [52:53<29:50,  2.67it/s]
 55%|█████▌    | 5912/10682 [52:53<29:47,  2.67it/s]
 55%|█████▌    | 5913/10682 [52:53<29:47,  2.67it/s]
 55%|█████▌    | 5914/10682 [52:54<29:48,  2.67it/s]
 55%|█████▌    | 5915/10682 [52:54<29:47,  2.67it/s]
 55%|█████▌    | 5916/10682 [52:54<29:50,  2.66it/s]
 55%|█████▌    | 5917/10682 [52:55<29:48,  2.66it/s]
 55%|█████▌    | 5918/10682 [52:55<29:47,  2.67it/s]
 55%|█████▌    | 5919/10682 [52:56<29:47,  2.66it/s]
 55%|█████▌    | 5920/10682 [52:56<29:48,  2.66it/s]
 55%|█████▌    | 5921/10682 [52:56<29:46,  2.66it/s]
 55%|█████▌    | 5922/10682 [52:57<29:45,  2.67it/s]
 55%|█████▌    | 5923/10682 [52:57<29:44,  2.67it/s]
 55%|█████▌    | 5924/10682 [52:57<29:43,  2.67it/s]
 55%|█████▌    | 5925/10682 [52:58<29:42,  2.67it/s]
                                                    
{'loss': 4.2863, 'grad_norm': 0.41729795932769775, 'learning_rate': 0.0004919118873482054, 'epoch': 0.55}
-
 55%|█████▌    | 5925/10682 [52:58<29:42,  2.67it/s]
 55%|█████▌    | 5926/10682 [52:58<29:42,  2.67it/s]
 55%|█████▌    | 5927/10682 [52:59<29:44,  2.66it/s]
 55%|█████▌    | 5928/10682 [52:59<29:42,  2.67it/s]
 56%|█████▌    | 5929/10682 [52:59<29:44,  2.66it/s]
 56%|█████▌    | 5930/10682 [53:00<29:44,  2.66it/s]
 56%|█████▌    | 5931/10682 [53:00<29:43,  2.66it/s]
 56%|█████▌    | 5932/10682 [53:00<29:42,  2.66it/s]
 56%|█████▌    | 5933/10682 [53:01<29:41,  2.67it/s]
 56%|█████▌    | 5934/10682 [53:01<29:41,  2.67it/s]
 56%|█████▌    | 5935/10682 [53:02<29:40,  2.67it/s]
 56%|█████▌    | 5936/10682 [53:02<29:39,  2.67it/s]
 56%|█████▌    | 5937/10682 [53:02<29:36,  2.67it/s]
 56%|█████▌    | 5938/10682 [53:03<29:37,  2.67it/s]
 56%|█████▌    | 5939/10682 [53:03<29:37,  2.67it/s]
 56%|█████▌    | 5940/10682 [53:03<29:37,  2.67it/s]
 56%|█████▌    | 5941/10682 [53:04<29:37,  2.67it/s]
 56%|█████▌    | 5942/10682 [53:04<29:36,  2.67it/s]
 56%|█████▌    | 5943/10682 [53:05<29:38,  2.67it/s]
 56%|█████▌    | 5944/10682 [53:05<29:36,  2.67it/s]
 56%|█████▌    | 5945/10682 [53:05<29:37,  2.67it/s]
 56%|█████▌    | 5946/10682 [53:06<29:35,  2.67it/s]
 56%|█████▌    | 5947/10682 [53:06<29:35,  2.67it/s]
 56%|█████▌    | 5948/10682 [53:06<29:34,  2.67it/s]
 56%|█████▌    | 5949/10682 [53:07<29:32,  2.67it/s]
 56%|█████▌    | 5950/10682 [53:07<29:30,  2.67it/s]
                                                    
{'loss': 4.2846, 'grad_norm': 0.42851948738098145, 'learning_rate': 0.0004878276536926525, 'epoch': 0.56}
-
 56%|█████▌    | 5950/10682 [53:07<29:30,  2.67it/s]
 56%|█████▌    | 5951/10682 [53:08<29:36,  2.66it/s]
 56%|█████▌    | 5952/10682 [53:08<29:33,  2.67it/s]
 56%|█████▌    | 5953/10682 [53:08<29:33,  2.67it/s]
 56%|█████▌    | 5954/10682 [53:09<29:34,  2.66it/s]
 56%|█████▌    | 5955/10682 [53:09<29:32,  2.67it/s]
 56%|█████▌    | 5956/10682 [53:09<29:34,  2.66it/s]
 56%|█████▌    | 5957/10682 [53:10<29:33,  2.66it/s]
 56%|█████▌    | 5958/10682 [53:10<29:31,  2.67it/s]
 56%|█████▌    | 5959/10682 [53:11<29:33,  2.66it/s]
 56%|█████▌    | 5960/10682 [53:11<29:33,  2.66it/s]
 56%|█████▌    | 5961/10682 [53:11<29:29,  2.67it/s]
 56%|█████▌    | 5962/10682 [53:12<29:28,  2.67it/s]
 56%|█████▌    | 5963/10682 [53:12<29:29,  2.67it/s]
 56%|█████▌    | 5964/10682 [53:12<29:28,  2.67it/s]
 56%|█████▌    | 5965/10682 [53:13<29:28,  2.67it/s]
 56%|█████▌    | 5966/10682 [53:13<29:27,  2.67it/s]
 56%|█████▌    | 5967/10682 [53:14<29:31,  2.66it/s]
 56%|█████▌    | 5968/10682 [53:14<29:28,  2.67it/s]
 56%|█████▌    | 5969/10682 [53:14<29:32,  2.66it/s]
 56%|█████▌    | 5970/10682 [53:15<29:32,  2.66it/s]
 56%|█████▌    | 5971/10682 [53:15<29:30,  2.66it/s]
 56%|█████▌    | 5972/10682 [53:15<29:28,  2.66it/s]
 56%|█████▌    | 5973/10682 [53:16<29:27,  2.66it/s]
 56%|█████▌    | 5974/10682 [53:16<29:26,  2.67it/s]
 56%|█████▌    | 5975/10682 [53:17<29:24,  2.67it/s]
                                                    
{'loss': 4.2877, 'grad_norm': 0.4320595860481262, 'learning_rate': 0.0004837442325565422, 'epoch': 0.56}
-
 56%|█████▌    | 5975/10682 [53:17<29:24,  2.67it/s]
 56%|█████▌    | 5976/10682 [53:17<29:28,  2.66it/s]
 56%|█████▌    | 5977/10682 [53:17<29:26,  2.66it/s]
 56%|█████▌    | 5978/10682 [53:18<29:25,  2.66it/s]
 56%|█████▌    | 5979/10682 [53:18<29:24,  2.67it/s]
 56%|█████▌    | 5980/10682 [53:18<29:23,  2.67it/s]
 56%|█████▌    | 5981/10682 [53:19<29:22,  2.67it/s]
 56%|█████▌    | 5982/10682 [53:19<29:21,  2.67it/s]
 56%|█████▌    | 5983/10682 [53:20<29:24,  2.66it/s]
 56%|█████▌    | 5984/10682 [53:20<29:24,  2.66it/s]
 56%|█████▌    | 5985/10682 [53:20<29:24,  2.66it/s]
 56%|█████▌    | 5986/10682 [53:21<29:22,  2.66it/s]
 56%|█████▌    | 5987/10682 [53:21<29:23,  2.66it/s]
 56%|█████▌    | 5988/10682 [53:21<29:21,  2.67it/s]
 56%|█████▌    | 5989/10682 [53:22<29:19,  2.67it/s]
 56%|█████▌    | 5990/10682 [53:22<29:17,  2.67it/s]
 56%|█████▌    | 5991/10682 [53:23<29:15,  2.67it/s]
 56%|█████▌    | 5992/10682 [53:23<29:17,  2.67it/s]
 56%|█████▌    | 5993/10682 [53:23<29:16,  2.67it/s]
 56%|█████▌    | 5994/10682 [53:24<29:17,  2.67it/s]
 56%|█████▌    | 5995/10682 [53:24<29:17,  2.67it/s]
 56%|█████▌    | 5996/10682 [53:24<29:19,  2.66it/s]
 56%|█████▌    | 5997/10682 [53:25<29:19,  2.66it/s]
 56%|█████▌    | 5998/10682 [53:25<29:16,  2.67it/s]
 56%|█████▌    | 5999/10682 [53:26<29:14,  2.67it/s]
 56%|█████▌    | 6000/10682 [53:26<29:14,  2.67it/s]
                                                    
{'loss': 4.2773, 'grad_norm': 0.40863239765167236, 'learning_rate': 0.00047966189651337725, 'epoch': 0.56}
-
 56%|█████▌    | 6000/10682 [53:26<29:14,  2.67it/s]
 56%|█████▌    | 6001/10682 [53:26<29:17,  2.66it/s]
 56%|█████▌    | 6002/10682 [53:27<29:18,  2.66it/s]
 56%|█████▌    | 6003/10682 [53:27<29:17,  2.66it/s]
 56%|█████▌    | 6004/10682 [53:27<29:14,  2.67it/s]
 56%|█████▌    | 6005/10682 [53:28<29:15,  2.66it/s]
 56%|█████▌    | 6006/10682 [53:28<29:12,  2.67it/s]
 56%|█████▌    | 6007/10682 [53:29<29:14,  2.66it/s]
 56%|█████▌    | 6008/10682 [53:29<29:12,  2.67it/s]
 56%|█████▋    | 6009/10682 [53:29<29:12,  2.67it/s]
 56%|█████▋    | 6010/10682 [53:30<29:14,  2.66it/s]
 56%|█████▋    | 6011/10682 [53:30<29:12,  2.67it/s]
 56%|█████▋    | 6012/10682 [53:30<29:11,  2.67it/s]
 56%|█████▋    | 6013/10682 [53:31<29:09,  2.67it/s]
 56%|█████▋    | 6014/10682 [53:31<29:08,  2.67it/s]
 56%|█████▋    | 6015/10682 [53:32<29:08,  2.67it/s]
 56%|█████▋    | 6016/10682 [53:32<29:08,  2.67it/s]
 56%|█████▋    | 6017/10682 [53:32<29:08,  2.67it/s]
 56%|█████▋    | 6018/10682 [53:33<29:06,  2.67it/s]
 56%|█████▋    | 6019/10682 [53:33<29:07,  2.67it/s]
 56%|█████▋    | 6020/10682 [53:33<29:05,  2.67it/s]
 56%|█████▋    | 6021/10682 [53:34<29:08,  2.67it/s]
 56%|█████▋    | 6022/10682 [53:34<29:05,  2.67it/s]
 56%|█████▋    | 6023/10682 [53:35<29:07,  2.67it/s]
 56%|█████▋    | 6024/10682 [53:35<29:06,  2.67it/s]
 56%|█████▋    | 6025/10682 [53:35<29:07,  2.67it/s]{'loss': 4.2788, 'grad_norm': 0.43671664595603943, 'learning_rate': 0.0004755809180642286, 'epoch': 0.56}
                                                    
-
 56%|█████▋    | 6025/10682 [53:35<29:07,  2.67it/s]
 56%|█████▋    | 6026/10682 [53:36<29:06,  2.67it/s]
 56%|█████▋    | 6027/10682 [53:36<29:05,  2.67it/s]
 56%|█████▋    | 6028/10682 [53:36<29:05,  2.67it/s]
 56%|█████▋    | 6029/10682 [53:37<29:04,  2.67it/s]
 56%|█████▋    | 6030/10682 [53:37<29:04,  2.67it/s]
 56%|█████▋    | 6031/10682 [53:38<29:04,  2.67it/s]
 56%|█████▋    | 6032/10682 [53:38<29:04,  2.67it/s]
 56%|█████▋    | 6033/10682 [53:38<29:02,  2.67it/s]
 56%|█████▋    | 6034/10682 [53:39<29:03,  2.67it/s]
 56%|█████▋    | 6035/10682 [53:39<29:01,  2.67it/s]
 57%|█████▋    | 6036/10682 [53:39<29:00,  2.67it/s]
 57%|█████▋    | 6037/10682 [53:40<29:03,  2.66it/s]
 57%|█████▋    | 6038/10682 [53:40<29:05,  2.66it/s]
 57%|█████▋    | 6039/10682 [53:41<29:03,  2.66it/s]
 57%|█████▋    | 6040/10682 [53:41<29:00,  2.67it/s]
 57%|█████▋    | 6041/10682 [53:41<29:00,  2.67it/s]
 57%|█████▋    | 6042/10682 [53:42<28:57,  2.67it/s]
 57%|█████▋    | 6043/10682 [53:42<35:31,  2.18it/s]
 57%|█████▋    | 6044/10682 [53:43<33:30,  2.31it/s]
 57%|█████▋    | 6045/10682 [53:43<32:10,  2.40it/s]
 57%|█████▋    | 6046/10682 [53:43<31:10,  2.48it/s]
 57%|█████▋    | 6047/10682 [53:44<30:32,  2.53it/s]
 57%|█████▋    | 6048/10682 [53:44<30:03,  2.57it/s]
 57%|█████▋    | 6049/10682 [53:45<29:44,  2.60it/s]
 57%|█████▋    | 6050/10682 [53:45<29:29,  2.62it/s]
                                                    
{'loss': 4.2773, 'grad_norm': 0.4569658637046814, 'learning_rate': 0.00047150156961954607, 'epoch': 0.57}
-
 57%|█████▋    | 6050/10682 [53:45<29:29,  2.62it/s]
 57%|█████▋    | 6051/10682 [53:45<29:21,  2.63it/s]
 57%|█████▋    | 6052/10682 [53:46<29:12,  2.64it/s]
 57%|█████▋    | 6053/10682 [53:46<29:06,  2.65it/s]
 57%|█████▋    | 6054/10682 [53:46<29:03,  2.65it/s]
 57%|█████▋    | 6055/10682 [53:47<28:59,  2.66it/s]
 57%|█████▋    | 6056/10682 [53:47<28:58,  2.66it/s]
 57%|█████▋    | 6057/10682 [53:48<28:54,  2.67it/s]
 57%|█████▋    | 6058/10682 [53:48<28:54,  2.67it/s]
 57%|█████▋    | 6059/10682 [53:48<28:56,  2.66it/s]
 57%|█████▋    | 6060/10682 [53:49<28:59,  2.66it/s]
 57%|█████▋    | 6061/10682 [53:49<28:55,  2.66it/s]
 57%|█████▋    | 6062/10682 [53:49<28:56,  2.66it/s]
 57%|█████▋    | 6063/10682 [53:50<28:59,  2.66it/s]
 57%|█████▋    | 6064/10682 [53:50<28:56,  2.66it/s]
 57%|█████▋    | 6065/10682 [53:51<28:56,  2.66it/s]
 57%|█████▋    | 6066/10682 [53:51<28:53,  2.66it/s]
 57%|█████▋    | 6067/10682 [53:51<28:51,  2.66it/s]
 57%|█████▋    | 6068/10682 [53:52<35:10,  2.19it/s]
 57%|█████▋    | 6069/10682 [53:52<33:14,  2.31it/s]
 57%|█████▋    | 6070/10682 [53:53<31:53,  2.41it/s]
 57%|█████▋    | 6071/10682 [53:53<30:57,  2.48it/s]
 57%|█████▋    | 6072/10682 [53:53<30:15,  2.54it/s]
 57%|█████▋    | 6073/10682 [53:54<29:54,  2.57it/s]
 57%|█████▋    | 6074/10682 [53:54<29:33,  2.60it/s]
 57%|█████▋    | 6075/10682 [53:55<29:20,  2.62it/s]
                                                    
{'loss': 4.2837, 'grad_norm': 0.49070876836776733, 'learning_rate': 0.0004674241234809751, 'epoch': 0.57}
-
 57%|█████▋    | 6075/10682 [53:55<29:20,  2.62it/s]
 57%|█████▋    | 6076/10682 [53:55<29:19,  2.62it/s]
 57%|█████▋    | 6077/10682 [53:55<29:11,  2.63it/s]
 57%|█████▋    | 6078/10682 [53:56<29:03,  2.64it/s]
 57%|█████▋    | 6079/10682 [53:56<28:58,  2.65it/s]
 57%|█████▋    | 6080/10682 [53:56<28:53,  2.65it/s]
 57%|█████▋    | 6081/10682 [53:57<28:48,  2.66it/s]
 57%|█████▋    | 6082/10682 [53:57<28:45,  2.67it/s]
 57%|█████▋    | 6083/10682 [53:58<28:45,  2.67it/s]
 57%|█████▋    | 6084/10682 [53:58<28:42,  2.67it/s]
 57%|█████▋    | 6085/10682 [53:58<28:41,  2.67it/s]
 57%|█████▋    | 6086/10682 [53:59<28:44,  2.67it/s]
 57%|█████▋    | 6087/10682 [53:59<28:44,  2.66it/s]
 57%|█████▋    | 6088/10682 [53:59<28:43,  2.67it/s]
 57%|█████▋    | 6089/10682 [54:00<28:47,  2.66it/s]
 57%|█████▋    | 6090/10682 [54:00<28:45,  2.66it/s]
 57%|█████▋    | 6091/10682 [54:01<28:46,  2.66it/s]
 57%|█████▋    | 6092/10682 [54:01<28:43,  2.66it/s]
 57%|█████▋    | 6093/10682 [54:01<28:43,  2.66it/s]
 57%|█████▋    | 6094/10682 [54:02<28:41,  2.67it/s]
 57%|█████▋    | 6095/10682 [54:02<28:41,  2.66it/s]
 57%|█████▋    | 6096/10682 [54:02<28:39,  2.67it/s]
 57%|█████▋    | 6097/10682 [54:03<28:37,  2.67it/s]
 57%|█████▋    | 6098/10682 [54:03<28:36,  2.67it/s]
 57%|█████▋    | 6099/10682 [54:04<28:34,  2.67it/s]
 57%|█████▋    | 6100/10682 [54:04<28:37,  2.67it/s]
                                                    
{'loss': 4.2757, 'grad_norm': 0.4120471477508545, 'learning_rate': 0.0004633488518231793, 'epoch': 0.57}
-
 57%|█████▋    | 6100/10682 [54:04<28:37,  2.67it/s]
 57%|█████▋    | 6101/10682 [54:04<28:39,  2.66it/s]
 57%|█████▋    | 6102/10682 [54:05<28:40,  2.66it/s]
 57%|█████▋    | 6103/10682 [54:05<28:39,  2.66it/s]
 57%|█████▋    | 6104/10682 [54:05<28:38,  2.66it/s]
 57%|█████▋    | 6105/10682 [54:06<28:39,  2.66it/s]
 57%|█████▋    | 6106/10682 [54:06<28:37,  2.66it/s]
 57%|█████▋    | 6107/10682 [54:07<28:34,  2.67it/s]
 57%|█████▋    | 6108/10682 [54:07<28:33,  2.67it/s]
 57%|█████▋    | 6109/10682 [54:07<28:33,  2.67it/s]
 57%|█████▋    | 6110/10682 [54:08<28:31,  2.67it/s]
 57%|█████▋    | 6111/10682 [54:08<28:31,  2.67it/s]
 57%|█████▋    | 6112/10682 [54:08<28:31,  2.67it/s]
 57%|█████▋    | 6113/10682 [54:09<28:30,  2.67it/s]
 57%|█████▋    | 6114/10682 [54:09<28:30,  2.67it/s]
 57%|█████▋    | 6115/10682 [54:10<28:31,  2.67it/s]
 57%|█████▋    | 6116/10682 [54:10<28:30,  2.67it/s]
 57%|█████▋    | 6117/10682 [54:10<28:27,  2.67it/s]
 57%|█████▋    | 6118/10682 [54:11<28:29,  2.67it/s]
 57%|█████▋    | 6119/10682 [54:11<28:28,  2.67it/s]
 57%|█████▋    | 6120/10682 [54:11<28:27,  2.67it/s]
 57%|█████▋    | 6121/10682 [54:12<28:26,  2.67it/s]
 57%|█████▋    | 6122/10682 [54:12<28:26,  2.67it/s]
 57%|█████▋    | 6123/10682 [54:13<28:26,  2.67it/s]
 57%|█████▋    | 6124/10682 [54:13<28:25,  2.67it/s]
 57%|█████▋    | 6125/10682 [54:13<28:23,  2.68it/s]
                                                    
{'loss': 4.2735, 'grad_norm': 0.4533524513244629, 'learning_rate': 0.00045927602667567333, 'epoch': 0.57}
-
 57%|��████▋    | 6125/10682 [54:13<28:23,  2.68it/s]
 57%|█████▋    | 6126/10682 [54:14<28:25,  2.67it/s]
 57%|█████▋    | 6127/10682 [54:14<28:25,  2.67it/s]
 57%|█████▋    | 6128/10682 [54:14<28:24,  2.67it/s]
 57%|█████▋    | 6129/10682 [54:15<28:25,  2.67it/s]
 57%|█████▋    | 6130/10682 [54:15<28:26,  2.67it/s]
 57%|█████▋    | 6131/10682 [54:16<28:29,  2.66it/s]
 57%|█████▋    | 6132/10682 [54:16<28:28,  2.66it/s]
 57%|█████▋    | 6133/10682 [54:16<28:25,  2.67it/s]
 57%|█████▋    | 6134/10682 [54:17<28:24,  2.67it/s]
 57%|█████▋    | 6135/10682 [54:17<28:22,  2.67it/s]
 57%|█████▋    | 6136/10682 [54:17<28:23,  2.67it/s]
 57%|█████▋    | 6137/10682 [54:18<28:20,  2.67it/s]
 57%|█████▋    | 6138/10682 [54:18<28:19,  2.67it/s]
 57%|█████▋    | 6139/10682 [54:19<28:18,  2.67it/s]
 57%|█████▋    | 6140/10682 [54:19<28:19,  2.67it/s]
 57%|█████▋    | 6141/10682 [54:19<28:19,  2.67it/s]
 57%|█████▋    | 6142/10682 [54:20<28:20,  2.67it/s]
 58%|█████▊    | 6143/10682 [54:20<28:19,  2.67it/s]
 58%|█████▊    | 6144/10682 [54:20<28:20,  2.67it/s]
 58%|█████▊    | 6145/10682 [54:21<28:24,  2.66it/s]
 58%|█████▊    | 6146/10682 [54:21<28:22,  2.66it/s]
 58%|█████▊    | 6147/10682 [54:22<28:20,  2.67it/s]
 58%|█████▊    | 6148/10682 [54:22<28:20,  2.67it/s]
 58%|█████▊    | 6149/10682 [54:22<28:19,  2.67it/s]
 58%|█████▊    | 6150/10682 [54:23<28:16,  2.67it/s]{'loss': 4.275, 'grad_norm': 0.39726245403289795, 'learning_rate': 0.00045520591990466365, 'epoch': 0.58}
                                                    
-
 58%|█████▊    | 6150/10682 [54:23<28:16,  2.67it/s]
 58%|█████▊    | 6151/10682 [54:23<28:19,  2.67it/s]
 58%|█████▊    | 6152/10682 [54:23<28:18,  2.67it/s]
 58%|█████▊    | 6153/10682 [54:24<28:17,  2.67it/s]
 58%|█████▊    | 6154/10682 [54:24<28:16,  2.67it/s]
 58%|█████▊    | 6155/10682 [54:25<28:14,  2.67it/s]
 58%|█████▊    | 6156/10682 [54:25<28:16,  2.67it/s]
 58%|█████▊    | 6157/10682 [54:25<28:15,  2.67it/s]
 58%|█████▊    | 6158/10682 [54:26<28:15,  2.67it/s]
 58%|█████▊    | 6159/10682 [54:26<28:17,  2.66it/s]
 58%|█████▊    | 6160/10682 [54:26<28:14,  2.67it/s]
 58%|█████▊    | 6161/10682 [54:27<28:13,  2.67it/s]
 58%|█████▊    | 6162/10682 [54:27<28:12,  2.67it/s]
 58%|█████▊    | 6163/10682 [54:28<28:11,  2.67it/s]
 58%|█████▊    | 6164/10682 [54:28<28:13,  2.67it/s]
 58%|█████▊    | 6165/10682 [54:28<28:12,  2.67it/s]
 58%|█████▊    | 6166/10682 [54:29<28:11,  2.67it/s]
 58%|█████▊    | 6167/10682 [54:29<28:10,  2.67it/s]
 58%|█████▊    | 6168/10682 [54:29<28:09,  2.67it/s]
 58%|█████▊    | 6169/10682 [54:30<28:09,  2.67it/s]
 58%|█████▊    | 6170/10682 [54:30<28:08,  2.67it/s]
 58%|█████▊    | 6171/10682 [54:31<28:09,  2.67it/s]
 58%|█████▊    | 6172/10682 [54:31<28:08,  2.67it/s]
 58%|█████▊    | 6173/10682 [54:31<28:07,  2.67it/s]
 58%|█████▊    | 6174/10682 [54:32<28:06,  2.67it/s]
 58%|█████▊    | 6175/10682 [54:32<28:09,  2.67it/s]
                                                    
{'loss': 4.2754, 'grad_norm': 0.42136693000793457, 'learning_rate': 0.00045113880319490264, 'epoch': 0.58}
-
 58%|█████▊    | 6175/10682 [54:32<28:09,  2.67it/s]
 58%|█████▊    | 6176/10682 [54:32<28:10,  2.67it/s]
 58%|█████▊    | 6177/10682 [54:33<28:08,  2.67it/s]
 58%|█████▊    | 6178/10682 [54:33<28:07,  2.67it/s]
 58%|█████▊    | 6179/10682 [54:34<28:05,  2.67it/s]
 58%|█████▊    | 6180/10682 [54:34<28:04,  2.67it/s]
 58%|█████▊    | 6181/10682 [54:34<28:04,  2.67it/s]
 58%|█████▊    | 6182/10682 [54:35<28:07,  2.67it/s]
 58%|█████▊    | 6183/10682 [54:35<28:06,  2.67it/s]
 58%|█████▊    | 6184/10682 [54:35<28:04,  2.67it/s]
 58%|█████▊    | 6185/10682 [54:36<28:05,  2.67it/s]
 58%|█████▊    | 6186/10682 [54:36<28:04,  2.67it/s]
 58%|█████▊    | 6187/10682 [54:37<28:04,  2.67it/s]
 58%|█████▊    | 6188/10682 [54:37<28:03,  2.67it/s]
 58%|█████▊    | 6189/10682 [54:37<28:02,  2.67it/s]
 58%|█████▊    | 6190/10682 [54:38<28:02,  2.67it/s]
 58%|█████▊    | 6191/10682 [54:38<28:00,  2.67it/s]
 58%|█████▊    | 6192/10682 [54:38<27:59,  2.67it/s]
 58%|█████▊    | 6193/10682 [54:39<27:59,  2.67it/s]
 58%|█████▊    | 6194/10682 [54:39<28:00,  2.67it/s]
 58%|█████▊    | 6195/10682 [54:40<27:59,  2.67it/s]
 58%|█████▊    | 6196/10682 [54:40<28:00,  2.67it/s]
 58%|█████▊    | 6197/10682 [54:40<28:01,  2.67it/s]
 58%|█████▊    | 6198/10682 [54:41<28:03,  2.66it/s]
 58%|█████▊    | 6199/10682 [54:41<28:02,  2.66it/s]
 58%|█████▊    | 6200/10682 [54:41<28:01,  2.66it/s]
                                                    
{'loss': 4.2709, 'grad_norm': 0.4262063205242157, 'learning_rate': 0.00044707494803155137, 'epoch': 0.58}
-
 58%|█████▊    | 6200/10682 [54:41<28:01,  2.66it/s]
 58%|█████▊    | 6201/10682 [54:42<28:01,  2.66it/s]
 58%|█████▊    | 6202/10682 [54:42<28:00,  2.67it/s]
 58%|█████▊    | 6203/10682 [54:43<27:59,  2.67it/s]
 58%|█████▊    | 6204/10682 [54:43<27:59,  2.67it/s]
 58%|█████▊    | 6205/10682 [54:43<27:58,  2.67it/s]
 58%|█████▊    | 6206/10682 [54:44<27:56,  2.67it/s]
 58%|█████▊    | 6207/10682 [54:44<27:56,  2.67it/s]
 58%|█████▊    | 6208/10682 [54:44<27:55,  2.67it/s]
 58%|█████▊    | 6209/10682 [54:45<27:56,  2.67it/s]
 58%|█████▊    | 6210/10682 [54:45<27:57,  2.67it/s]
 58%|█████▊    | 6211/10682 [54:46<27:57,  2.67it/s]
 58%|█████▊    | 6212/10682 [54:46<27:59,  2.66it/s]
 58%|█████▊    | 6213/10682 [54:46<27:57,  2.66it/s]
 58%|█████▊    | 6214/10682 [54:47<27:55,  2.67it/s]
 58%|█████▊    | 6215/10682 [54:47<27:54,  2.67it/s]
 58%|█████▊    | 6216/10682 [54:47<27:53,  2.67it/s]
 58%|█████▊    | 6217/10682 [54:48<27:52,  2.67it/s]
 58%|█████▊    | 6218/10682 [54:48<27:52,  2.67it/s]
 58%|█████▊    | 6219/10682 [54:49<27:50,  2.67it/s]
 58%|█████▊    | 6220/10682 [54:49<27:51,  2.67it/s]
 58%|█████▊    | 6221/10682 [54:49<27:50,  2.67it/s]
 58%|█████▊    | 6222/10682 [54:50<27:50,  2.67it/s]
 58%|█████▊    | 6223/10682 [54:50<27:48,  2.67it/s]
 58%|█████▊    | 6224/10682 [54:50<27:48,  2.67it/s]
 58%|█████▊    | 6225/10682 [54:51<27:52,  2.67it/s]{'loss': 4.2662, 'grad_norm': 0.42039230465888977, 'learning_rate': 0.00044301462568205915, 'epoch': 0.58}
-
                                                    

 58%|█████▊    | 6225/10682 [54:51<27:52,  2.67it/s]
 58%|█████▊    | 6226/10682 [54:51<27:51,  2.67it/s]
 58%|█████▊    | 6227/10682 [54:52<27:50,  2.67it/s]
 58%|█████▊    | 6228/10682 [54:52<27:48,  2.67it/s]
 58%|█████▊    | 6229/10682 [54:52<27:49,  2.67it/s]
 58%|█████▊    | 6230/10682 [54:53<27:48,  2.67it/s]
 58%|█████▊    | 6231/10682 [54:53<27:48,  2.67it/s]
 58%|█████▊    | 6232/10682 [54:53<27:46,  2.67it/s]
 58%|█████▊    | 6233/10682 [54:54<27:45,  2.67it/s]
 58%|█████▊    | 6234/10682 [54:54<27:46,  2.67it/s]
 58%|█████▊    | 6235/10682 [54:55<27:45,  2.67it/s]
 58%|█████▊    | 6236/10682 [54:55<27:46,  2.67it/s]
 58%|█████▊    | 6237/10682 [54:55<27:45,  2.67it/s]
 58%|█████▊    | 6238/10682 [54:56<27:46,  2.67it/s]
 58%|█████▊    | 6239/10682 [54:56<27:47,  2.66it/s]
 58%|█████▊    | 6240/10682 [54:56<27:45,  2.67it/s]
 58%|█████▊    | 6241/10682 [54:57<27:45,  2.67it/s]
 58%|█████▊    | 6242/10682 [54:57<27:43,  2.67it/s]
 58%|█████▊    | 6243/10682 [54:58<27:44,  2.67it/s]
 58%|█████▊    | 6244/10682 [54:58<27:44,  2.67it/s]
 58%|█████▊    | 6245/10682 [54:58<27:42,  2.67it/s]
 58%|█████▊    | 6246/10682 [54:59<27:41,  2.67it/s]
 58%|█████▊    | 6247/10682 [54:59<27:41,  2.67it/s]
 58%|█████▊    | 6248/10682 [54:59<27:40,  2.67it/s]
 59%|█████▊    | 6249/10682 [55:00<27:42,  2.67it/s]
 59%|█████▊    | 6250/10682 [55:00<27:41,  2.67it/s]{'loss': 4.2726, 'grad_norm': 0.4319385290145874, 'learning_rate': 0.0004389581071780553, 'epoch': 0.59}
                                                    
-
 59%|█████▊    | 6250/10682 [55:00<27:41,  2.67it/s]
 59%|█████▊    | 6251/10682 [55:01<27:41,  2.67it/s]
 59%|█████▊    | 6252/10682 [55:01<27:43,  2.66it/s]
 59%|█████▊    | 6253/10682 [55:01<27:40,  2.67it/s]
 59%|█████▊    | 6254/10682 [55:02<27:39,  2.67it/s]
 59%|█████▊    | 6255/10682 [55:02<27:37,  2.67it/s]
 59%|█████▊    | 6256/10682 [55:02<27:38,  2.67it/s]
 59%|█████▊    | 6257/10682 [55:03<27:37,  2.67it/s]
 59%|█████▊    | 6258/10682 [55:03<27:37,  2.67it/s]
 59%|█████▊    | 6259/10682 [55:04<27:35,  2.67it/s]
 59%|█████▊    | 6260/10682 [55:04<27:36,  2.67it/s]
 59%|█████▊    | 6261/10682 [55:04<27:35,  2.67it/s]
 59%|█████▊    | 6262/10682 [55:05<27:34,  2.67it/s]
 59%|█████▊    | 6263/10682 [55:05<27:38,  2.67it/s]
 59%|█████▊    | 6264/10682 [55:05<27:36,  2.67it/s]
 59%|███��█▊    | 6265/10682 [55:06<27:38,  2.66it/s]
 59%|█████▊    | 6266/10682 [55:06<27:36,  2.67it/s]
 59%|█████▊    | 6267/10682 [55:07<27:36,  2.67it/s]
 59%|█████▊    | 6268/10682 [55:07<27:34,  2.67it/s]
 59%|█████▊    | 6269/10682 [55:07<27:32,  2.67it/s]
 59%|█████▊    | 6270/10682 [55:08<27:31,  2.67it/s]
 59%|█████▊    | 6271/10682 [55:08<27:31,  2.67it/s]
 59%|█████▊    | 6272/10682 [55:08<27:31,  2.67it/s]
 59%|█████▊    | 6273/10682 [55:09<27:30,  2.67it/s]
 59%|█████▊    | 6274/10682 [55:09<27:30,  2.67it/s]
 59%|█████▊    | 6275/10682 [55:10<27:28,  2.67it/s]
                                                    
{'loss': 4.2651, 'grad_norm': 0.4314216375350952, 'learning_rate': 0.0004349056632972576, 'epoch': 0.59}
-
 59%|█████▊    | 6275/10682 [55:10<27:28,  2.67it/s]
 59%|█████▉    | 6276/10682 [55:10<27:32,  2.67it/s]
 59%|█████▉    | 6277/10682 [55:10<27:30,  2.67it/s]
 59%|█████▉    | 6278/10682 [55:11<27:31,  2.67it/s]
 59%|█████▉    | 6279/10682 [55:11<27:33,  2.66it/s]
 59%|█████▉    | 6280/10682 [55:11<27:32,  2.66it/s]
 59%|█████▉    | 6281/10682 [55:12<27:30,  2.67it/s]
 59%|█████▉    | 6282/10682 [55:12<27:29,  2.67it/s]
 59%|█████▉    | 6283/10682 [55:13<27:31,  2.66it/s]
 59%|█████▉    | 6284/10682 [55:13<27:29,  2.67it/s]
 59%|█████▉    | 6285/10682 [55:13<27:27,  2.67it/s]
 59%|█████▉    | 6286/10682 [55:14<27:25,  2.67it/s]
 59%|█████▉    | 6287/10682 [55:14<27:25,  2.67it/s]
 59%|█████▉    | 6288/10682 [55:14<27:24,  2.67it/s]
 59%|█████▉    | 6289/10682 [55:15<27:25,  2.67it/s]
 59%|█████▉    | 6290/10682 [55:15<27:24,  2.67it/s]
 59%|█████▉    | 6291/10682 [55:16<27:23,  2.67it/s]
 59%|█████▉    | 6292/10682 [55:16<27:26,  2.67it/s]
 59%|█████▉    | 6293/10682 [55:16<27:25,  2.67it/s]
 59%|█████▉    | 6294/10682 [55:17<27:24,  2.67it/s]
 59%|█████▉    | 6295/10682 [55:17<27:24,  2.67it/s]
 59%|█████▉    | 6296/10682 [55:17<27:23,  2.67it/s]
 59%|█████▉    | 6297/10682 [55:18<27:21,  2.67it/s]
 59%|█████▉    | 6298/10682 [55:18<27:21,  2.67it/s]
 59%|█████▉    | 6299/10682 [55:19<27:19,  2.67it/s]
 59%|█████▉    | 6300/10682 [55:19<27:20,  2.67it/s]
                                                    
{'loss': 4.2794, 'grad_norm': 0.43929776549339294, 'learning_rate': 0.00043085756454539745, 'epoch': 0.59}
-
 59%|█████▉    | 6300/10682 [55:19<27:20,  2.67it/s]
 59%|█████▉    | 6301/10682 [55:19<27:27,  2.66it/s]
 59%|█████▉    | 6302/10682 [55:20<27:23,  2.67it/s]
 59%|█████▉    | 6303/10682 [55:20<27:24,  2.66it/s]
 59%|█████▉    | 6304/10682 [55:20<27:20,  2.67it/s]
 59%|█████▉    | 6305/10682 [55:21<27:23,  2.66it/s]
 59%|█████▉    | 6306/10682 [55:21<27:22,  2.66it/s]
 59%|█████▉    | 6307/10682 [55:22<27:22,  2.66it/s]
 59%|█████▉    | 6308/10682 [55:22<27:21,  2.66it/s]
 59%|█████▉    | 6309/10682 [55:22<27:20,  2.67it/s]
 59%|█████▉    | 6310/10682 [55:23<27:18,  2.67it/s]
 59%|█████▉    | 6311/10682 [55:23<27:19,  2.67it/s]
 59%|█████▉    | 6312/10682 [55:23<27:17,  2.67it/s]
 59%|█████▉    | 6313/10682 [55:24<27:15,  2.67it/s]
 59%|█████▉    | 6314/10682 [55:24<27:15,  2.67it/s]
 59%|█████▉    | 6315/10682 [55:25<27:15,  2.67it/s]
 59%|█████▉    | 6316/10682 [55:25<27:15,  2.67it/s]
 59%|█████▉    | 6317/10682 [55:25<27:16,  2.67it/s]
 59%|█████▉    | 6318/10682 [55:26<27:13,  2.67it/s]
 59%|█████▉    | 6319/10682 [55:26<27:16,  2.67it/s]
 59%|█████▉    | 6320/10682 [55:26<27:16,  2.67it/s]
 59%|█████▉    | 6321/10682 [55:27<27:16,  2.66it/s]
 59%|█████▉    | 6322/10682 [55:27<27:14,  2.67it/s]
 59%|█████▉    | 6323/10682 [55:28<27:13,  2.67it/s]
 59%|█████▉    | 6324/10682 [55:28<27:12,  2.67it/s]
 59%|█████▉    | 6325/10682 [55:28<27:12,  2.67it/s]
                                                    
{'loss': 4.2694, 'grad_norm': 0.3928705155849457, 'learning_rate': 0.0004268140811381641, 'epoch': 0.59}
-
 59%|█████▉    | 6325/10682 [55:28<27:12,  2.67it/s]
 59%|█████▉    | 6326/10682 [55:29<27:13,  2.67it/s]
 59%|█████▉    | 6327/10682 [55:29<27:10,  2.67it/s]
 59%|█████▉    | 6328/10682 [55:29<27:10,  2.67it/s]
 59%|█████▉    | 6329/10682 [55:30<27:09,  2.67it/s]
 59%|█████▉    | 6330/10682 [55:30<27:13,  2.66it/s]
 59%|█████▉    | 6331/10682 [55:31<27:11,  2.67it/s]
 59%|█████▉    | 6332/10682 [55:31<27:13,  2.66it/s]
 59%|█████▉    | 6333/10682 [55:31<27:11,  2.66it/s]
 59%|█████▉    | 6334/10682 [55:32<27:10,  2.67it/s]
 59%|█████▉    | 6335/10682 [55:32<27:11,  2.66it/s]
 59%|█████▉    | 6336/10682 [55:32<27:08,  2.67it/s]
 59%|█████▉    | 6337/10682 [55:33<27:09,  2.67it/s]
 59%|█████▉    | 6338/10682 [55:33<27:11,  2.66it/s]
 59%|█████▉    | 6339/10682 [55:34<27:09,  2.67it/s]
 59%|█████▉    | 6340/10682 [55:34<27:07,  2.67it/s]
 59%|█████▉    | 6341/10682 [55:34<27:05,  2.67it/s]
 59%|█████▉    | 6342/10682 [55:35<27:04,  2.67it/s]
 59%|█████▉    | 6343/10682 [55:35<27:04,  2.67it/s]
 59%|█████▉    | 6344/10682 [55:35<27:04,  2.67it/s]
 59%|█████▉    | 6345/10682 [55:36<27:04,  2.67it/s]
 59%|█████▉    | 6346/10682 [55:36<27:06,  2.67it/s]
 59%|█████▉    | 6347/10682 [55:37<27:07,  2.66it/s]
 59%|█████▉    | 6348/10682 [55:37<27:06,  2.66it/s]
 59%|█████▉    | 6349/10682 [55:37<27:05,  2.67it/s]
 59%|█████▉    | 6350/10682 [55:38<27:03,  2.67it/s]
                                                    
{'loss': 4.2679, 'grad_norm': 0.43367069959640503, 'learning_rate': 0.0004227754829831662, 'epoch': 0.59}
-
 59%|█████▉    | 6350/10682 [55:38<27:03,  2.67it/s]
 59%|█████▉    | 6351/10682 [55:38<27:04,  2.67it/s]
 59%|█████▉    | 6352/10682 [55:38<27:03,  2.67it/s]
 59%|█████▉    | 6353/10682 [55:39<27:01,  2.67it/s]
 59%|█████▉    | 6354/10682 [55:39<27:01,  2.67it/s]
 59%|█████▉    | 6355/10682 [55:40<27:01,  2.67it/s]
 60%|█████▉    | 6356/10682 [55:40<26:59,  2.67it/s]
 60%|█████▉    | 6357/10682 [55:40<27:00,  2.67it/s]
 60%|█████▉    | 6358/10682 [55:41<26:59,  2.67it/s]
 60%|█████▉    | 6359/10682 [55:41<27:03,  2.66it/s]
 60%|█████▉    | 6360/10682 [55:41<27:02,  2.66it/s]
 60%|█████▉    | 6361/10682 [55:42<27:00,  2.67it/s]
 60%|█████▉    | 6362/10682 [55:42<27:00,  2.67it/s]
 60%|█████▉    | 6363/10682 [55:43<26:59,  2.67it/s]
 60%|█████▉    | 6364/10682 [55:43<26:58,  2.67it/s]
 60%|█████▉    | 6365/10682 [55:43<26:57,  2.67it/s]
 60%|█████▉    | 6366/10682 [55:44<26:56,  2.67it/s]
 60%|█████▉    | 6367/10682 [55:44<26:54,  2.67it/s]
 60%|█████▉    | 6368/10682 [55:44<26:55,  2.67it/s]
 60%|█████▉    | 6369/10682 [55:45<26:54,  2.67it/s]
 60%|█████▉    | 6370/10682 [55:45<26:57,  2.67it/s]
 60%|█████▉    | 6371/10682 [55:46<26:57,  2.67it/s]
 60%|█████▉    | 6372/10682 [55:46<26:59,  2.66it/s]
 60%|█████▉    | 6373/10682 [55:46<26:57,  2.66it/s]
 60%|█████▉    | 6374/10682 [55:47<26:54,  2.67it/s]
 60%|█████▉    | 6375/10682 [55:47<26:54,  2.67it/s]{'loss': 4.2671, 'grad_norm': 0.4550933241844177, 'learning_rate': 0.0004187420396619156, 'epoch': 0.6}
                                                    
-
 60%|█████▉    | 6375/10682 [55:47<26:54,  2.67it/s]
 60%|█████▉    | 6376/10682 [55:47<26:55,  2.66it/s]
 60%|█████▉    | 6377/10682 [55:48<26:54,  2.67it/s]
 60%|█████▉    | 6378/10682 [55:48<26:52,  2.67it/s]
 60%|█████▉    | 6379/10682 [55:49<26:51,  2.67it/s]
 60%|█████▉    | 6380/10682 [55:49<26:50,  2.67it/s]
 60%|█████▉    | 6381/10682 [55:49<26:51,  2.67it/s]
 60%|█████▉    | 6382/10682 [55:50<26:51,  2.67it/s]
 60%|█████▉    | 6383/10682 [55:50<26:55,  2.66it/s]
 60%|█████▉    | 6384/10682 [55:50<26:54,  2.66it/s]
 60%|█████▉    | 6385/10682 [55:51<26:51,  2.67it/s]
 60%|█████▉    | 6386/10682 [55:51<26:54,  2.66it/s]
 60%|█████▉    | 6387/10682 [55:52<26:52,  2.66it/s]
 60%|█████▉    | 6388/10682 [55:52<26:51,  2.66it/s]
 60%|█████▉    | 6389/10682 [55:52<26:50,  2.67it/s]
 60%|█████▉    | 6390/10682 [55:53<26:50,  2.67it/s]
 60%|█████▉    | 6391/10682 [55:53<26:47,  2.67it/s]
 60%|█████▉    | 6392/10682 [55:53<26:49,  2.67it/s]
 60%|█████▉    | 6393/10682 [55:54<26:49,  2.67it/s]
 60%|█████▉    | 6394/10682 [55:54<26:48,  2.67it/s]
 60%|█████▉    | 6395/10682 [55:55<26:48,  2.67it/s]
 60%|█████▉    | 6396/10682 [55:55<26:46,  2.67it/s]
 60%|█████▉    | 6397/10682 [55:55<26:51,  2.66it/s]
 60%|█████▉    | 6398/10682 [55:56<26:50,  2.66it/s]
 60%|█████▉    | 6399/10682 [55:56<26:51,  2.66it/s]
 60%|█████▉    | 6400/10682 [55:56<26:52,  2.66it/s]{'loss': 4.2622, 'grad_norm': 0.41201719641685486, 'learning_rate': 0.00041471402041183287, 'epoch': 0.6}
                                                    
-
 60%|█████▉    | 6400/10682 [55:56<26:52,  2.66it/s]
 60%|█████▉    | 6401/10682 [55:57<26:56,  2.65it/s]
 60%|█████▉    | 6402/10682 [55:57<26:52,  2.65it/s]
 60%|█████▉    | 6403/10682 [55:58<26:49,  2.66it/s]
 60%|█████▉    | 6404/10682 [55:58<26:46,  2.66it/s]
 60%|█████▉    | 6405/10682 [55:58<26:45,  2.66it/s]
 60%|█████▉    | 6406/10682 [55:59<26:42,  2.67it/s]
 60%|█████▉    | 6407/10682 [55:59<26:41,  2.67it/s]
 60%|█████▉    | 6408/10682 [55:59<26:43,  2.66it/s]
 60%|█████▉    | 6409/10682 [56:00<26:41,  2.67it/s]
 60%|██████    | 6410/10682 [56:00<26:46,  2.66it/s]
 60%|██████    | 6411/10682 [56:01<26:45,  2.66it/s]
 60%|██████    | 6412/10682 [56:01<26:43,  2.66it/s]
 60%|██████    | 6413/10682 [56:01<26:45,  2.66it/s]
 60%|██████    | 6414/10682 [56:02<26:44,  2.66it/s]
 60%|██████    | 6415/10682 [56:02<26:43,  2.66it/s]
 60%|██████    | 6416/10682 [56:02<26:40,  2.67it/s]
 60%|██████    | 6417/10682 [56:03<26:41,  2.66it/s]
 60%|██████    | 6418/10682 [56:03<26:39,  2.67it/s]
 60%|██████    | 6419/10682 [56:04<26:37,  2.67it/s]
 60%|██████    | 6420/10682 [56:04<26:35,  2.67it/s]
 60%|██████    | 6421/10682 [56:04<26:36,  2.67it/s]
 60%|██████    | 6422/10682 [56:05<26:34,  2.67it/s]
 60%|██████    | 6423/10682 [56:05<26:34,  2.67it/s]
 60%|██████    | 6424/10682 [56:05<26:35,  2.67it/s]
 60%|██████    | 6425/10682 [56:06<26:35,  2.67it/s]{'loss': 4.2658, 'grad_norm': 0.44719216227531433, 'learning_rate': 0.0004106916941082745, 'epoch': 0.6}
                                                    
-
 60%|██████    | 6425/10682 [56:06<26:35,  2.67it/s]
 60%|██████    | 6426/10682 [56:06<26:38,  2.66it/s]
 60%|██████    | 6427/10682 [56:07<26:38,  2.66it/s]
 60%|██████    | 6428/10682 [56:07<26:36,  2.66it/s]
 60%|██████    | 6429/10682 [56:07<26:35,  2.67it/s]
 60%|██████    | 6430/10682 [56:08<26:34,  2.67it/s]
 60%|██████    | 6431/10682 [56:08<26:34,  2.67it/s]
 60%|██████    | 6432/10682 [56:08<26:33,  2.67it/s]
 60%|██████    | 6433/10682 [56:09<26:33,  2.67it/s]
 60%|██████    | 6434/10682 [56:09<26:31,  2.67it/s]
 60%|██████    | 6435/10682 [56:10<26:31,  2.67it/s]
 60%|██████    | 6436/10682 [56:10<26:31,  2.67it/s]
 60%|██████    | 6437/10682 [56:10<26:34,  2.66it/s]
 60%|██████    | 6438/10682 [56:11<26:34,  2.66it/s]
 60%|██████    | 6439/10682 [56:11<26:34,  2.66it/s]
 60%|██████    | 6440/10682 [56:11<26:35,  2.66it/s]
 60%|██████    | 6441/10682 [56:12<26:33,  2.66it/s]
 60%|██████    | 6442/10682 [56:12<26:31,  2.66it/s]
 60%|██████    | 6443/10682 [56:13<26:30,  2.67it/s]
 60%|██████    | 6444/10682 [56:13<26:30,  2.67it/s]
 60%|██████    | 6445/10682 [56:13<26:29,  2.67it/s]
 60%|██████    | 6446/10682 [56:14<26:28,  2.67it/s]
 60%|██████    | 6447/10682 [56:14<26:29,  2.66it/s]
 60%|██████    | 6448/10682 [56:14<26:27,  2.67it/s]
 60%|██████    | 6449/10682 [56:15<26:26,  2.67it/s]
 60%|██████    | 6450/10682 [56:15<26:25,  2.67it/s]
                                                    
{'loss': 4.2705, 'grad_norm': 0.44100525975227356, 'learning_rate': 0.0004066753292465861, 'epoch': 0.6}
-
 60%|██████    | 6450/10682 [56:15<26:25,  2.67it/s]
 60%|██████    | 6451/10682 [56:16<26:32,  2.66it/s]
 60%|██████    | 6452/10682 [56:16<26:28,  2.66it/s]
 60%|██████    | 6453/10682 [56:16<26:30,  2.66it/s]
 60%|██████    | 6454/10682 [56:17<26:28,  2.66it/s]
 60%|██████    | 6455/10682 [56:17<26:25,  2.67it/s]
 60%|██████    | 6456/10682 [56:17<26:24,  2.67it/s]
 60%|██████    | 6457/10682 [56:18<26:23,  2.67it/s]
 60%|██████    | 6458/10682 [56:18<26:25,  2.66it/s]
 60%|██████    | 6459/10682 [56:19<26:24,  2.67it/s]
 60%|██████    | 6460/10682 [56:19<26:24,  2.66it/s]
 60%|██████    | 6461/10682 [56:19<26:24,  2.66it/s]
 60%|██████    | 6462/10682 [56:20<26:24,  2.66it/s]
 61%|██████    | 6463/10682 [56:20<26:23,  2.66it/s]
 61%|██████    | 6464/10682 [56:20<26:22,  2.66it/s]
 61%|██████    | 6465/10682 [56:21<26:20,  2.67it/s]
 61%|██████    | 6466/10682 [56:21<26:21,  2.67it/s]
 61%|██████    | 6467/10682 [56:22<26:21,  2.67it/s]
 61%|██████    | 6468/10682 [56:22<26:19,  2.67it/s]
 61%|██████    | 6469/10682 [56:22<26:21,  2.66it/s]
 61%|██████    | 6470/10682 [56:23<26:19,  2.67it/s]
 61%|██████    | 6471/10682 [56:23<26:18,  2.67it/s]
 61%|██████    | 6472/10682 [56:23<26:18,  2.67it/s]
 61%|██████    | 6473/10682 [56:24<26:18,  2.67it/s]
 61%|██████    | 6474/10682 [56:24<26:17,  2.67it/s]
 61%|██████    | 6475/10682 [56:25<26:19,  2.66it/s]{'loss': 4.267, 'grad_norm': 0.4220218360424042, 'learning_rate': 0.00040266519392417936, 'epoch': 0.61}
                                                    
-
 61%|██████    | 6475/10682 [56:25<26:19,  2.66it/s]
 61%|██████    | 6476/10682 [56:25<26:19,  2.66it/s]
 61%|██████    | 6477/10682 [56:25<26:17,  2.66it/s]
 61%|██████    | 6478/10682 [56:26<26:16,  2.67it/s]
 61%|██████    | 6479/10682 [56:26<26:16,  2.67it/s]
 61%|██████    | 6480/10682 [56:26<26:16,  2.67it/s]
 61%|██████    | 6481/10682 [56:27<26:15,  2.67it/s]
 61%|██████    | 6482/10682 [56:27<26:13,  2.67it/s]
 61%|██████    | 6483/10682 [56:28<26:14,  2.67it/s]
 61%|██████    | 6484/10682 [56:28<26:12,  2.67it/s]
 61%|██████    | 6485/10682 [56:28<26:10,  2.67it/s]
 61%|██████    | 6486/10682 [56:29<26:10,  2.67it/s]
 61%|██████    | 6487/10682 [56:29<26:10,  2.67it/s]
 61%|██████    | 6488/10682 [56:29<26:09,  2.67it/s]
 61%|██████    | 6489/10682 [56:30<26:09,  2.67it/s]
 61%|██████    | 6490/10682 [56:30<26:07,  2.67it/s]
 61%|██████    | 6491/10682 [56:31<26:10,  2.67it/s]
 61%|██████    | 6492/10682 [56:31<26:10,  2.67it/s]
 61%|██████    | 6493/10682 [56:31<26:11,  2.67it/s]
 61%|██████    | 6494/10682 [56:32<26:09,  2.67it/s]
 61%|██████    | 6495/10682 [56:32<26:08,  2.67it/s]
 61%|██████    | 6496/10682 [56:32<26:08,  2.67it/s]
 61%|██████    | 6497/10682 [56:33<26:09,  2.67it/s]
 61%|██████    | 6498/10682 [56:33<26:10,  2.66it/s]
 61%|██████    | 6499/10682 [56:34<26:10,  2.66it/s]
 61%|██████    | 6500/10682 [56:34<26:09,  2.66it/s]{'loss': 4.2641, 'grad_norm': 0.43373069167137146, 'learning_rate': 0.00039866155582263653, 'epoch': 0.61}
-
                                                    

 61%|██████    | 6500/10682 [56:34<26:09,  2.66it/s]
 61%|██████    | 6501/10682 [56:34<26:11,  2.66it/s]
 61%|██████    | 6502/10682 [56:35<26:09,  2.66it/s]
 61%|██████    | 6503/10682 [56:35<26:08,  2.66it/s]
 61%|██████    | 6504/10682 [56:35<26:10,  2.66it/s]
 61%|██████    | 6505/10682 [56:36<26:08,  2.66it/s]
 61%|██████    | 6506/10682 [56:36<26:08,  2.66it/s]
 61%|██████    | 6507/10682 [56:37<26:08,  2.66it/s]
 61%|██████    | 6508/10682 [56:37<26:06,  2.66it/s]
 61%|██████    | 6509/10682 [56:37<26:06,  2.66it/s]
 61%|██████    | 6510/10682 [56:38<26:05,  2.67it/s]
 61%|██████    | 6511/10682 [56:38<26:04,  2.67it/s]
 61%|██████    | 6512/10682 [56:38<26:04,  2.67it/s]
 61%|██████    | 6513/10682 [56:39<26:03,  2.67it/s]
 61%|██████    | 6514/10682 [56:39<26:02,  2.67it/s]
 61%|██████    | 6515/10682 [56:40<26:03,  2.66it/s]
 61%|██████    | 6516/10682 [56:40<26:02,  2.67it/s]
 61%|██████    | 6517/10682 [56:40<26:04,  2.66it/s]
 61%|██████    | 6518/10682 [56:41<26:04,  2.66it/s]
 61%|██████    | 6519/10682 [56:41<26:04,  2.66it/s]
 61%|██████    | 6520/10682 [56:41<26:03,  2.66it/s]
 61%|██████    | 6521/10682 [56:42<26:03,  2.66it/s]
 61%|██████    | 6522/10682 [56:42<26:01,  2.66it/s]
 61%|██████    | 6523/10682 [56:43<26:00,  2.67it/s]
 61%|██████    | 6524/10682 [56:43<26:00,  2.67it/s]
 61%|██████    | 6525/10682 [56:43<26:00,  2.66it/s]{'loss': 4.2717, 'grad_norm': 0.4001096785068512, 'learning_rate': 0.0003946646821898418, 'epoch': 0.61}
                                                    
-
 61%|██████    | 6525/10682 [56:43<26:00,  2.66it/s]
 61%|██████    | 6526/10682 [56:44<26:00,  2.66it/s]
 61%|██████    | 6527/10682 [56:44<25:58,  2.67it/s]
 61%|██████    | 6528/10682 [56:44<25:58,  2.66it/s]
 61%|██████    | 6529/10682 [56:45<25:57,  2.67it/s]
 61%|██████    | 6530/10682 [56:45<25:56,  2.67it/s]
 61%|██████    | 6531/10682 [56:46<25:54,  2.67it/s]
 61%|██████    | 6532/10682 [56:46<25:54,  2.67it/s]
 61%|██████    | 6533/10682 [56:46<25:55,  2.67it/s]
 61%|██████    | 6534/10682 [56:47<25:57,  2.66it/s]
 61%|██████    | 6535/10682 [56:47<25:55,  2.67it/s]
 61%|██████    | 6536/10682 [56:47<25:56,  2.66it/s]
 61%|██████    | 6537/10682 [56:48<25:56,  2.66it/s]
 61%|██████    | 6538/10682 [56:48<25:57,  2.66it/s]
 61%|██████    | 6539/10682 [56:49<25:54,  2.67it/s]
 61%|██████    | 6540/10682 [56:49<25:52,  2.67it/s]
 61%|██████    | 6541/10682 [56:49<25:51,  2.67it/s]
 61%|██████    | 6542/10682 [56:50<25:50,  2.67it/s]
 61%|██████▏   | 6543/10682 [56:50<25:51,  2.67it/s]
 61%|██████▏   | 6544/10682 [56:50<25:53,  2.66it/s]
 61%|██████▏   | 6545/10682 [56:51<25:51,  2.67it/s]
 61%|██████▏   | 6546/10682 [56:51<25:52,  2.66it/s]
 61%|██████▏   | 6547/10682 [56:52<25:52,  2.66it/s]
 61%|██████▏   | 6548/10682 [56:52<25:51,  2.66it/s]
 61%|██████▏   | 6549/10682 [56:52<25:48,  2.67it/s]
 61%|██████▏   | 6550/10682 [56:53<25:46,  2.67it/s]
                                                    
{'loss': 4.2625, 'grad_norm': 0.41876399517059326, 'learning_rate': 0.0003906748398221429, 'epoch': 0.61}
-
 61%|██████▏   | 6550/10682 [56:53<25:46,  2.67it/s]
 61%|██████▏   | 6551/10682 [56:53<25:47,  2.67it/s]
 61%|██████▏   | 6552/10682 [56:53<25:48,  2.67it/s]
 61%|██████▏   | 6553/10682 [56:54<25:47,  2.67it/s]
 61%|██████▏   | 6554/10682 [56:54<25:46,  2.67it/s]
 61%|██████▏   | 6555/10682 [56:55<25:46,  2.67it/s]
 61%|██████▏   | 6556/10682 [56:55<25:44,  2.67it/s]
 61%|██████▏   | 6557/10682 [56:55<25:48,  2.66it/s]
 61%|██████▏   | 6558/10682 [56:56<25:46,  2.67it/s]
 61%|██████▏   | 6559/10682 [56:56<25:45,  2.67it/s]
 61%|██████▏   | 6560/10682 [56:56<25:45,  2.67it/s]
 61%|██████▏   | 6561/10682 [56:57<25:46,  2.66it/s]
 61%|██████▏   | 6562/10682 [56:57<25:45,  2.67it/s]
 61%|██████▏   | 6563/10682 [56:58<25:44,  2.67it/s]
 61%|██████▏   | 6564/10682 [56:58<25:43,  2.67it/s]
 61%|██████▏   | 6565/10682 [56:58<25:42,  2.67it/s]
 61%|██████▏   | 6566/10682 [56:59<25:39,  2.67it/s]
 61%|██████▏   | 6567/10682 [56:59<25:38,  2.68it/s]
 61%|██████▏   | 6568/10682 [56:59<25:40,  2.67it/s]
 61%|██████▏   | 6569/10682 [57:00<25:40,  2.67it/s]
 62%|██████▏   | 6570/10682 [57:00<25:39,  2.67it/s]
 62%|██████▏   | 6571/10682 [57:01<25:44,  2.66it/s]
 62%|██████▏   | 6572/10682 [57:01<25:42,  2.66it/s]
 62%|██████▏   | 6573/10682 [57:01<25:43,  2.66it/s]
 62%|██████▏   | 6574/10682 [57:02<25:42,  2.66it/s]
 62%|██████▏   | 6575/10682 [57:02<25:39,  2.67it/s]
                                                    
{'loss': 4.2613, 'grad_norm': 0.4312584698200226, 'learning_rate': 0.00038669229504654146, 'epoch': 0.62}
-
 62%|██████▏   | 6575/10682 [57:02<25:39,  2.67it/s]
 62%|██████▏   | 6576/10682 [57:02<25:39,  2.67it/s]
 62%|██████▏   | 6577/10682 [57:03<25:38,  2.67it/s]
 62%|██████▏   | 6578/10682 [57:03<25:37,  2.67it/s]
 62%|██████▏   | 6579/10682 [57:04<25:40,  2.66it/s]
 62%|██████▏   | 6580/10682 [57:04<25:38,  2.67it/s]
 62%|██████▏   | 6581/10682 [57:04<25:38,  2.67it/s]
 62%|██████▏   | 6582/10682 [57:05<25:37,  2.67it/s]
 62%|██████▏   | 6583/10682 [57:05<25:36,  2.67it/s]
 62%|██████▏   | 6584/10682 [57:05<25:38,  2.66it/s]
 62%|██████▏   | 6585/10682 [57:06<25:40,  2.66it/s]
 62%|██████▏   | 6586/10682 [57:06<25:39,  2.66it/s]
 62%|██████▏   | 6587/10682 [57:07<25:36,  2.67it/s]
 62%|██████▏   | 6588/10682 [57:07<25:35,  2.67it/s]
 62%|██████▏   | 6589/10682 [57:07<25:33,  2.67it/s]
 62%|██████▏   | 6590/10682 [57:08<25:33,  2.67it/s]
 62%|██████▏   | 6591/10682 [57:08<25:32,  2.67it/s]
 62%|██████▏   | 6592/10682 [57:08<25:34,  2.67it/s]
 62%|██████▏   | 6593/10682 [57:09<25:31,  2.67it/s]
 62%|██████▏   | 6594/10682 [57:09<25:31,  2.67it/s]
 62%|██████▏   | 6595/10682 [57:10<25:29,  2.67it/s]
 62%|██████▏   | 6596/10682 [57:10<25:28,  2.67it/s]
 62%|██████▏   | 6597/10682 [57:10<25:29,  2.67it/s]
 62%|██████▏   | 6598/10682 [57:11<25:29,  2.67it/s]
 62%|██████▏   | 6599/10682 [57:11<25:27,  2.67it/s]
 62%|██████▏   | 6600/10682 [57:11<25:29,  2.67it/s]{'loss': 4.2522, 'grad_norm': 0.44955694675445557, 'learning_rate': 0.00038271731370291583, 'epoch': 0.62}
                                                    
-
 62%|██████▏   | 6600/10682 [57:11<25:29,  2.67it/s]
 62%|██████▏   | 6601/10682 [57:12<25:30,  2.67it/s]
 62%|██████▏   | 6602/10682 [57:12<25:30,  2.67it/s]
 62%|██████▏   | 6603/10682 [57:13<25:29,  2.67it/s]
 62%|██████▏   | 6604/10682 [57:13<25:28,  2.67it/s]
 62%|██████▏   | 6605/10682 [57:13<25:28,  2.67it/s]
 62%|██████▏   | 6606/10682 [57:14<25:27,  2.67it/s]
 62%|██████▏   | 6607/10682 [57:14<25:26,  2.67it/s]
 62%|██████▏   | 6608/10682 [57:14<25:25,  2.67it/s]
 62%|██████▏   | 6609/10682 [57:15<25:25,  2.67it/s]
 62%|██████▏   | 6610/10682 [57:15<25:23,  2.67it/s]
 62%|██████▏   | 6611/10682 [57:16<25:24,  2.67it/s]
 62%|██████▏   | 6612/10682 [57:16<25:24,  2.67it/s]
 62%|██████▏   | 6613/10682 [57:16<25:25,  2.67it/s]
 62%|██████▏   | 6614/10682 [57:17<25:23,  2.67it/s]
 62%|██████▏   | 6615/10682 [57:17<25:23,  2.67it/s]
 62%|██████▏   | 6616/10682 [57:17<25:22,  2.67it/s]
 62%|██████▏   | 6617/10682 [57:18<25:21,  2.67it/s]
 62%|██████▏   | 6618/10682 [57:18<25:20,  2.67it/s]
 62%|██████▏   | 6619/10682 [57:19<25:20,  2.67it/s]
 62%|██████▏   | 6620/10682 [57:19<25:19,  2.67it/s]
 62%|██████▏   | 6621/10682 [57:19<25:19,  2.67it/s]
 62%|██████▏   | 6622/10682 [57:20<25:20,  2.67it/s]
 62%|██████▏   | 6623/10682 [57:20<25:19,  2.67it/s]
 62%|██████▏   | 6624/10682 [57:20<25:21,  2.67it/s]
 62%|██████▏   | 6625/10682 [57:21<25:21,  2.67it/s]{'loss': 4.2572, 'grad_norm': 0.42687222361564636, 'learning_rate': 0.0003787501611262762, 'epoch': 0.62}
                                                    
-
 62%|██████▏   | 6625/10682 [57:21<25:21,  2.67it/s]
 62%|██████▏   | 6626/10682 [57:21<25:22,  2.66it/s]
 62%|██████▏   | 6627/10682 [57:22<25:21,  2.66it/s]
 62%|██████▏   | 6628/10682 [57:22<25:21,  2.67it/s]
 62%|██████▏   | 6629/10682 [57:22<25:20,  2.67it/s]
 62%|██████▏   | 6630/10682 [57:23<25:22,  2.66it/s]
 62%|██████▏   | 6631/10682 [57:23<25:22,  2.66it/s]
 62%|██████▏   | 6632/10682 [57:23<25:21,  2.66it/s]
 62%|██████▏   | 6633/10682 [57:24<25:18,  2.67it/s]
 62%|██████▏   | 6634/10682 [57:24<25:16,  2.67it/s]
 62%|██████▏   | 6635/10682 [57:25<25:18,  2.66it/s]
 62%|██████▏   | 6636/10682 [57:25<25:16,  2.67it/s]
 62%|██████▏   | 6637/10682 [57:25<25:16,  2.67it/s]
 62%|██████▏   | 6638/10682 [57:26<25:18,  2.66it/s]
 62%|██████▏   | 6639/10682 [57:26<25:16,  2.67it/s]
 62%|██████▏   | 6640/10682 [57:26<25:18,  2.66it/s]
 62%|██████▏   | 6641/10682 [57:27<25:17,  2.66it/s]
 62%|██████▏   | 6642/10682 [57:27<25:16,  2.66it/s]
 62%|██████▏   | 6643/10682 [57:28<25:13,  2.67it/s]
 62%|██████▏   | 6644/10682 [57:28<25:13,  2.67it/s]
 62%|██████▏   | 6645/10682 [57:28<25:13,  2.67it/s]
 62%|██████▏   | 6646/10682 [57:29<25:13,  2.67it/s]
 62%|██████▏   | 6647/10682 [57:29<25:12,  2.67it/s]
 62%|██████▏   | 6648/10682 [57:29<25:11,  2.67it/s]
 62%|██████▏   | 6649/10682 [57:30<25:10,  2.67it/s]
 62%|██████▏   | 6650/10682 [57:30<25:08,  2.67it/s]{'loss': 4.2593, 'grad_norm': 0.4012000858783722, 'learning_rate': 0.0003747911021290519, 'epoch': 0.62}
-
                                                    

 62%|██████▏   | 6650/10682 [57:30<25:08,  2.67it/s]
 62%|██████▏   | 6651/10682 [57:31<25:13,  2.66it/s]
 62%|██████▏   | 6652/10682 [57:31<25:13,  2.66it/s]
 62%|██████▏   | 6653/10682 [57:31<25:17,  2.66it/s]
 62%|██████▏   | 6654/10682 [57:32<25:14,  2.66it/s]
 62%|██████▏   | 6655/10682 [57:32<25:12,  2.66it/s]
 62%|██████▏   | 6656/10682 [57:32<25:10,  2.67it/s]
 62%|██████▏   | 6657/10682 [57:33<25:09,  2.67it/s]
 62%|██████▏   | 6658/10682 [57:33<25:09,  2.67it/s]
 62%|██████▏   | 6659/10682 [57:34<25:08,  2.67it/s]
 62%|██████▏   | 6660/10682 [57:34<25:07,  2.67it/s]
 62%|██████▏   | 6661/10682 [57:34<25:06,  2.67it/s]
 62%|██████▏   | 6662/10682 [57:35<25:05,  2.67it/s]
 62%|██████▏   | 6663/10682 [57:35<25:05,  2.67it/s]
 62%|██████▏   | 6664/10682 [57:35<25:06,  2.67it/s]
 62%|██████▏   | 6665/10682 [57:36<25:04,  2.67it/s]
 62%|██████▏   | 6666/10682 [57:36<25:04,  2.67it/s]
 62%|██████▏   | 6667/10682 [57:37<25:07,  2.66it/s]
 62%|██████▏   | 6668/10682 [57:37<25:06,  2.66it/s]
 62%|██████▏   | 6669/10682 [57:37<25:05,  2.67it/s]
 62%|██████▏   | 6670/10682 [57:38<25:05,  2.66it/s]
 62%|██████▏   | 6671/10682 [57:38<25:06,  2.66it/s]
 62%|██████▏   | 6672/10682 [57:38<25:04,  2.67it/s]
 62%|██████▏   | 6673/10682 [57:39<25:02,  2.67it/s]
 62%|██████▏   | 6674/10682 [57:39<25:00,  2.67it/s]
 62%|██████▏   | 6675/10682 [57:40<25:00,  2.67it/s]
                                                    {'loss': 4.261, 'grad_norm': 0.4148128032684326, 'learning_rate': 0.0003708404009834161, 'epoch': 0.62}
-
 62%|██████▏   | 6675/10682 [57:40<25:00,  2.67it/s]
 62%|██████▏   | 6676/10682 [57:40<25:01,  2.67it/s]
 63%|██████▎   | 6677/10682 [57:40<25:01,  2.67it/s]
 63%|██████▎   | 6678/10682 [57:41<25:02,  2.67it/s]
 63%|██████▎   | 6679/10682 [57:41<24:59,  2.67it/s]
 63%|██████▎   | 6680/10682 [57:41<25:03,  2.66it/s]
 63%|██████▎   | 6681/10682 [57:42<25:01,  2.66it/s]
 63%|██████▎   | 6682/10682 [57:42<25:01,  2.66it/s]
 63%|██████▎   | 6683/10682 [57:43<25:01,  2.66it/s]
 63%|██████▎   | 6684/10682 [57:43<25:03,  2.66it/s]
 63%|██████▎   | 6685/10682 [57:43<25:03,  2.66it/s]
 63%|██████▎   | 6686/10682 [57:44<25:00,  2.66it/s]
 63%|██████▎   | 6687/10682 [57:44<24:58,  2.67it/s]
 63%|██████▎   | 6688/10682 [57:44<24:58,  2.67it/s]
 63%|██████▎   | 6689/10682 [57:45<24:59,  2.66it/s]
 63%|██████▎   | 6690/10682 [57:45<24:57,  2.67it/s]
 63%|██████▎   | 6691/10682 [57:46<25:01,  2.66it/s]
 63%|██████▎   | 6692/10682 [57:46<24:59,  2.66it/s]
 63%|██████▎   | 6693/10682 [57:46<25:01,  2.66it/s]
 63%|██████▎   | 6694/10682 [57:47<24:59,  2.66it/s]
 63%|██████▎   | 6695/10682 [57:47<24:59,  2.66it/s]
 63%|██████▎   | 6696/10682 [57:47<24:57,  2.66it/s]
 63%|██████▎   | 6697/10682 [57:48<24:57,  2.66it/s]
 63%|██████▎   | 6698/10682 [57:48<24:55,  2.66it/s]
 63%|██████▎   | 6699/10682 [57:49<24:53,  2.67it/s]
 63%|██████▎   | 6700/10682 [57:49<24:53,  2.67it/s]{'loss': 4.2556, 'grad_norm': 0.4058483839035034, 'learning_rate': 0.0003668983214036445, 'epoch': 0.63}
                                                    
-
 63%|██████▎   | 6700/10682 [57:49<24:53,  2.67it/s]
 63%|██████▎   | 6701/10682 [57:49<24:54,  2.66it/s]
 63%|██████▎   | 6702/10682 [57:50<24:54,  2.66it/s]
 63%|██████▎   | 6703/10682 [57:50<24:55,  2.66it/s]
 63%|██████▎   | 6704/10682 [57:50<24:55,  2.66it/s]
 63%|██████▎   | 6705/10682 [57:51<24:55,  2.66it/s]
 63%|██████▎   | 6706/10682 [57:51<24:53,  2.66it/s]
 63%|██████▎   | 6707/10682 [57:52<24:52,  2.66it/s]
 63%|██████▎   | 6708/10682 [57:52<24:51,  2.66it/s]
 63%|██████▎   | 6709/10682 [57:52<24:52,  2.66it/s]
 63%|██████▎   | 6710/10682 [57:53<24:49,  2.67it/s]
 63%|██████▎   | 6711/10682 [57:53<24:50,  2.66it/s]
 63%|██████▎   | 6712/10682 [57:53<24:47,  2.67it/s]
 63%|██████▎   | 6713/10682 [57:54<24:47,  2.67it/s]
 63%|██████▎   | 6714/10682 [57:54<24:46,  2.67it/s]
 63%|██████▎   | 6715/10682 [57:55<24:46,  2.67it/s]
 63%|██████▎   | 6716/10682 [57:55<24:46,  2.67it/s]
 63%|██████▎   | 6717/10682 [57:55<24:46,  2.67it/s]
 63%|██████▎   | 6718/10682 [57:56<24:47,  2.67it/s]
 63%|██████▎   | 6719/10682 [57:56<24:47,  2.67it/s]
 63%|██████▎   | 6720/10682 [57:56<24:48,  2.66it/s]
 63%|██████▎   | 6721/10682 [57:57<24:45,  2.67it/s]
 63%|██████▎   | 6722/10682 [57:57<24:46,  2.66it/s]
 63%|██████▎   | 6723/10682 [57:58<24:44,  2.67it/s]
 63%|██████▎   | 6724/10682 [57:58<24:44,  2.67it/s]
 63%|██████▎   | 6725/10682 [57:58<24:43,  2.67it/s]{'loss': 4.2579, 'grad_norm': 0.4707386791706085, 'learning_rate': 0.00036296512652851323, 'epoch': 0.63}
                                                    
-
 63%|██████▎   | 6725/10682 [57:58<24:43,  2.67it/s]
 63%|██████▎   | 6726/10682 [57:59<24:45,  2.66it/s]
 63%|██████▎   | 6727/10682 [57:59<24:42,  2.67it/s]
 63%|██████▎   | 6728/10682 [57:59<24:41,  2.67it/s]
 63%|██████▎   | 6729/10682 [58:00<24:42,  2.67it/s]
 63%|██████▎   | 6730/10682 [58:00<24:41,  2.67it/s]
 63%|██████▎   | 6731/10682 [58:01<24:43,  2.66it/s]
 63%|██████▎   | 6732/10682 [58:01<24:43,  2.66it/s]
 63%|██████▎   | 6733/10682 [58:01<24:43,  2.66it/s]
 63%|██████▎   | 6734/10682 [58:02<24:41,  2.66it/s]
 63%|██████▎   | 6735/10682 [58:02<24:41,  2.66it/s]
 63%|██████▎   | 6736/10682 [58:02<24:39,  2.67it/s]
 63%|██████▎   | 6737/10682 [58:03<24:38,  2.67it/s]
 63%|██████▎   | 6738/10682 [58:03<24:39,  2.67it/s]
 63%|██████▎   | 6739/10682 [58:04<24:38,  2.67it/s]
 63%|██████▎   | 6740/10682 [58:04<24:37,  2.67it/s]
 63%|██████▎   | 6741/10682 [58:04<24:36,  2.67it/s]
 63%|██████▎   | 6742/10682 [58:05<24:35,  2.67it/s]
 63%|██████▎   | 6743/10682 [58:05<24:34,  2.67it/s]
 63%|████���█▎   | 6744/10682 [58:05<24:38,  2.66it/s]
 63%|██████▎   | 6745/10682 [58:06<24:36,  2.67it/s]
 63%|██████▎   | 6746/10682 [58:06<24:37,  2.66it/s]
 63%|██████▎   | 6747/10682 [58:07<24:39,  2.66it/s]
 63%|██████▎   | 6748/10682 [58:07<24:37,  2.66it/s]
 63%|██████▎   | 6749/10682 [58:07<24:35,  2.67it/s]
 63%|██████▎   | 6750/10682 [58:08<24:33,  2.67it/s]
                                                    {'loss': 4.2563, 'grad_norm': 0.4264725148677826, 'learning_rate': 0.00035904107890373196, 'epoch': 0.63}
-
 63%|██████▎   | 6750/10682 [58:08<24:33,  2.67it/s]
 63%|██████▎   | 6751/10682 [58:08<24:35,  2.66it/s]
 63%|██████▎   | 6752/10682 [58:08<24:34,  2.67it/s]
 63%|██████▎   | 6753/10682 [58:09<24:33,  2.67it/s]
 63%|██████▎   | 6754/10682 [58:09<24:31,  2.67it/s]
 63%|██████▎   | 6755/10682 [58:10<24:31,  2.67it/s]
 63%|██████▎   | 6756/10682 [58:10<24:31,  2.67it/s]
 63%|██████▎   | 6757/10682 [58:10<24:28,  2.67it/s]
 63%|██████▎   | 6758/10682 [58:11<29:55,  2.19it/s]
 63%|██████▎   | 6759/10682 [58:11<28:20,  2.31it/s]
 63%|██████▎   | 6760/10682 [58:12<27:09,  2.41it/s]
 63%|██████▎   | 6761/10682 [58:12<26:21,  2.48it/s]
 63%|██████▎   | 6762/10682 [58:12<25:47,  2.53it/s]
 63%|██████▎   | 6763/10682 [58:13<25:24,  2.57it/s]
 63%|██████▎   | 6764/10682 [58:13<25:08,  2.60it/s]
 63%|██████▎   | 6765/10682 [58:14<24:55,  2.62it/s]
 63%|██████▎   | 6766/10682 [58:14<24:47,  2.63it/s]
 63%|██████▎   | 6767/10682 [58:14<24:39,  2.65it/s]
 63%|██████▎   | 6768/10682 [58:15<24:35,  2.65it/s]
 63%|██████▎   | 6769/10682 [58:15<24:33,  2.66it/s]
 63%|██████▎   | 6770/10682 [58:15<24:32,  2.66it/s]
 63%|██████▎   | 6771/10682 [58:16<24:29,  2.66it/s]
 63%|██████▎   | 6772/10682 [58:16<24:29,  2.66it/s]
 63%|██████▎   | 6773/10682 [58:17<24:32,  2.65it/s]
 63%|██████▎   | 6774/10682 [58:17<24:29,  2.66it/s]
 63%|██████▎   | 6775/10682 [58:17<24:28,  2.66it/s]{'loss': 4.2499, 'grad_norm': 0.4281921684741974, 'learning_rate': 0.00035512644046442116, 'epoch': 0.63}
                                                    
-
 63%|██████▎   | 6775/10682 [58:17<24:28,  2.66it/s]
 63%|██████▎   | 6776/10682 [58:18<24:29,  2.66it/s]
 63%|██████▎   | 6777/10682 [58:18<24:29,  2.66it/s]
 63%|██████▎   | 6778/10682 [58:19<24:26,  2.66it/s]
 63%|██████▎   | 6779/10682 [58:19<24:24,  2.67it/s]
 63%|██████▎   | 6780/10682 [58:19<24:24,  2.67it/s]
 63%|██████▎   | 6781/10682 [58:20<24:22,  2.67it/s]
 63%|██████▎   | 6782/10682 [58:20<24:21,  2.67it/s]
 63%|██████▎   | 6783/10682 [58:20<24:20,  2.67it/s]
 64%|██████▎   | 6784/10682 [58:21<24:20,  2.67it/s]
 64%|██████▎   | 6785/10682 [58:21<24:18,  2.67it/s]
 64%|██████▎   | 6786/10682 [58:21<24:20,  2.67it/s]
 64%|██████▎   | 6787/10682 [58:22<24:19,  2.67it/s]
 64%|██████▎   | 6788/10682 [58:23<29:41,  2.19it/s]
 64%|██████▎   | 6789/10682 [58:23<28:06,  2.31it/s]
 64%|██████▎   | 6790/10682 [58:23<26:59,  2.40it/s]
 64%|██████▎   | 6791/10682 [58:24<26:08,  2.48it/s]
 64%|██████▎   | 6792/10682 [58:24<25:33,  2.54it/s]
 64%|██████▎   | 6793/10682 [58:24<25:10,  2.57it/s]
 64%|██████▎   | 6794/10682 [58:25<24:59,  2.59it/s]
 64%|██████▎   | 6795/10682 [58:25<24:44,  2.62it/s]
 64%|██████▎   | 6796/10682 [58:26<24:37,  2.63it/s]
 64%|██████▎   | 6797/10682 [58:26<24:31,  2.64it/s]
 64%|██████▎   | 6798/10682 [58:26<24:25,  2.65it/s]
 64%|██████▎   | 6799/10682 [58:27<24:20,  2.66it/s]
 64%|██████▎   | 6800/10682 [58:27<24:16,  2.66it/s]
                                                    
{'loss': 4.2612, 'grad_norm': 0.4516674876213074, 'learning_rate': 0.00035122147251762554, 'epoch': 0.64}
-
 64%|██████▎   | 6800/10682 [58:27<24:16,  2.66it/s]
 64%|██████▎   | 6801/10682 [58:27<24:17,  2.66it/s]
 64%|██████▎   | 6802/10682 [58:28<24:14,  2.67it/s]
 64%|██████▎   | 6803/10682 [58:28<24:15,  2.66it/s]
 64%|██████▎   | 6804/10682 [58:29<24:13,  2.67it/s]
 64%|██████▎   | 6805/10682 [58:29<24:13,  2.67it/s]
 64%|██████▎   | 6806/10682 [58:29<24:11,  2.67it/s]
 64%|██████▎   | 6807/10682 [58:30<24:09,  2.67it/s]
 64%|██████▎   | 6808/10682 [58:30<24:08,  2.67it/s]
 64%|██████▎   | 6809/10682 [58:30<24:07,  2.67it/s]
 64%|██████▍   | 6810/10682 [58:31<24:09,  2.67it/s]
 64%|██████▍   | 6811/10682 [58:31<24:09,  2.67it/s]
 64%|██████▍   | 6812/10682 [58:32<24:11,  2.67it/s]
 64%|██████▍   | 6813/10682 [58:32<24:08,  2.67it/s]
 64%|██████▍   | 6814/10682 [58:32<24:09,  2.67it/s]
 64%|██████▍   | 6815/10682 [58:33<24:09,  2.67it/s]
 64%|██████▍   | 6816/10682 [58:33<24:08,  2.67it/s]
 64%|██████▍   | 6817/10682 [58:33<24:08,  2.67it/s]
 64%|██████▍   | 6818/10682 [58:34<24:07,  2.67it/s]
 64%|██████▍   | 6819/10682 [58:34<24:08,  2.67it/s]
 64%|██████▍   | 6820/10682 [58:35<24:08,  2.67it/s]
 64%|██████▍   | 6821/10682 [58:35<24:06,  2.67it/s]
 64%|██████▍   | 6822/10682 [58:35<24:06,  2.67it/s]
 64%|██████▍   | 6823/10682 [58:36<24:07,  2.67it/s]
 64%|██████▍   | 6824/10682 [58:36<24:05,  2.67it/s]
 64%|██████▍   | 6825/10682 [58:36<24:07,  2.67it/s]
                                                    
{'loss': 4.2606, 'grad_norm': 0.42735663056373596, 'learning_rate': 0.00034732643572487243, 'epoch': 0.64}
-
 64%|██████▍   | 6825/10682 [58:36<24:07,  2.67it/s]
 64%|██████▍   | 6826/10682 [58:37<24:06,  2.67it/s]
 64%|██████▍   | 6827/10682 [58:37<24:07,  2.66it/s]
 64%|██████▍   | 6828/10682 [58:38<24:06,  2.66it/s]
 64%|██████▍   | 6829/10682 [58:38<24:06,  2.66it/s]
 64%|██████▍   | 6830/10682 [58:38<24:04,  2.67it/s]
 64%|██████▍   | 6831/10682 [58:39<24:03,  2.67it/s]
 64%|██████▍   | 6832/10682 [58:39<24:04,  2.67it/s]
 64%|██████▍   | 6833/10682 [58:39<24:01,  2.67it/s]
 64%|██████▍   | 6834/10682 [58:40<23:59,  2.67it/s]
 64%|██████▍   | 6835/10682 [58:40<23:59,  2.67it/s]
 64%|██████▍   | 6836/10682 [58:41<23:59,  2.67it/s]
 64%|██████▍   | 6837/10682 [58:41<23:58,  2.67it/s]
 64%|██████▍   | 6838/10682 [58:41<24:00,  2.67it/s]
 64%|██████▍   | 6839/10682 [58:42<24:02,  2.66it/s]
 64%|██████▍   | 6840/10682 [58:42<24:00,  2.67it/s]
 64%|██████▍   | 6841/10682 [58:42<24:02,  2.66it/s]
 64%|██████▍   | 6842/10682 [58:43<24:00,  2.66it/s]
 64%|██████▍   | 6843/10682 [58:43<23:59,  2.67it/s]
 64%|██████▍   | 6844/10682 [58:44<23:58,  2.67it/s]
 64%|██████▍   | 6845/10682 [58:44<23:57,  2.67it/s]
 64%|██████▍   | 6846/10682 [58:44<23:58,  2.67it/s]
 64%|██████▍   | 6847/10682 [58:45<23:56,  2.67it/s]
 64%|██████▍   | 6848/10682 [58:45<23:57,  2.67it/s]
 64%|██████▍   | 6849/10682 [58:45<23:57,  2.67it/s]
 64%|██████▍   | 6850/10682 [58:46<23:55,  2.67it/s]{'loss': 4.2447, 'grad_norm': 0.40772441029548645, 'learning_rate': 0.000343441590084772, 'epoch': 0.64}
-
                                                    

 64%|██████▍   | 6850/10682 [58:46<23:55,  2.67it/s]
 64%|██████▍   | 6851/10682 [58:46<24:03,  2.65it/s]
 64%|██████▍   | 6852/10682 [58:47<24:01,  2.66it/s]
 64%|██████▍   | 6853/10682 [58:47<23:58,  2.66it/s]
 64%|██████▍   | 6854/10682 [58:47<23:58,  2.66it/s]
 64%|██████▍   | 6855/10682 [58:48<23:57,  2.66it/s]
 64%|██████▍   | 6856/10682 [58:48<23:54,  2.67it/s]
 64%|██████▍   | 6857/10682 [58:48<23:55,  2.67it/s]
 64%|██████▍   | 6858/10682 [58:49<23:53,  2.67it/s]
 64%|██████▍   | 6859/10682 [58:49<23:56,  2.66it/s]
 64%|██████▍   | 6860/10682 [58:50<23:54,  2.66it/s]
 64%|██████▍   | 6861/10682 [58:50<23:52,  2.67it/s]
 64%|██████▍   | 6862/10682 [58:50<23:51,  2.67it/s]
 64%|██████▍   | 6863/10682 [58:51<23:51,  2.67it/s]
 64%|██████▍   | 6864/10682 [58:51<23:50,  2.67it/s]
 64%|██████▍   | 6865/10682 [58:51<23:50,  2.67it/s]
 64%|██████▍   | 6866/10682 [58:52<23:49,  2.67it/s]
 64%|██████▍   | 6867/10682 [58:52<23:49,  2.67it/s]
 64%|██████▍   | 6868/10682 [58:53<23:50,  2.67it/s]
 64%|██████▍   | 6869/10682 [58:53<23:49,  2.67it/s]
 64%|██████▍   | 6870/10682 [58:53<23:49,  2.67it/s]
 64%|██████▍   | 6871/10682 [58:54<23:48,  2.67it/s]
 64%|██████▍   | 6872/10682 [58:54<23:46,  2.67it/s]
 64%|██████▍   | 6873/10682 [58:54<23:46,  2.67it/s]
 64%|██████▍   | 6874/10682 [58:55<23:45,  2.67it/s]
 64%|██████▍   | 6875/10682 [58:55<23:45,  2.67it/s]{'loss': 4.2539, 'grad_norm': 0.4391065239906311, 'learning_rate': 0.0003395671949156626, 'epoch': 0.64}
                                                    
-
 64%|██████▍   | 6875/10682 [58:55<23:45,  2.67it/s]
 64%|██████▍   | 6876/10682 [58:56<23:47,  2.67it/s]
 64%|██████▍   | 6877/10682 [58:56<23:46,  2.67it/s]
 64%|██████▍   | 6878/10682 [58:56<23:45,  2.67it/s]
 64%|██████▍   | 6879/10682 [58:57<23:47,  2.66it/s]
 64%|██████▍   | 6880/10682 [58:57<23:46,  2.67it/s]
 64%|██████▍   | 6881/10682 [58:57<23:46,  2.67it/s]
 64%|██████▍   | 6882/10682 [58:58<23:45,  2.67it/s]
 64%|██████▍   | 6883/10682 [58:58<23:43,  2.67it/s]
 64%|██████▍   | 6884/10682 [58:59<23:42,  2.67it/s]
 64%|██████▍   | 6885/10682 [58:59<23:42,  2.67it/s]
 64%|██████▍   | 6886/10682 [58:59<23:42,  2.67it/s]
 64%|██████▍   | 6887/10682 [59:00<23:40,  2.67it/s]
 64%|██████▍   | 6888/10682 [59:00<23:41,  2.67it/s]
 64%|██████▍   | 6889/10682 [59:00<23:41,  2.67it/s]
 65%|██████▍   | 6890/10682 [59:01<23:41,  2.67it/s]
 65%|██████▍   | 6891/10682 [59:01<23:41,  2.67it/s]
 65%|██████▍   | 6892/10682 [59:02<23:45,  2.66it/s]
 65%|██████▍   | 6893/10682 [59:02<23:43,  2.66it/s]
 65%|██████▍   | 6894/10682 [59:02<23:44,  2.66it/s]
 65%|██████▍   | 6895/10682 [59:03<23:42,  2.66it/s]
 65%|██████▍   | 6896/10682 [59:03<23:42,  2.66it/s]
 65%|██████▍   | 6897/10682 [59:03<23:41,  2.66it/s]
 65%|██████▍   | 6898/10682 [59:04<23:39,  2.66it/s]
 65%|██████▍   | 6899/10682 [59:04<23:39,  2.67it/s]
 65%|██████▍   | 6900/10682 [59:05<23:37,  2.67it/s]
                                                    
{'loss': 4.2553, 'grad_norm': 0.4013723134994507, 'learning_rate': 0.0003357035088382997, 'epoch': 0.65}
-
 65%|██████▍   | 6900/10682 [59:05<23:37,  2.67it/s]
 65%|██████▍   | 6901/10682 [59:05<23:38,  2.67it/s]
 65%|██████▍   | 6902/10682 [59:05<23:37,  2.67it/s]
 65%|██████▍   | 6903/10682 [59:06<23:36,  2.67it/s]
 65%|██████▍   | 6904/10682 [59:06<23:34,  2.67it/s]
 65%|██████▍   | 6905/10682 [59:06<23:36,  2.67it/s]
 65%|██████▍   | 6906/10682 [59:07<23:34,  2.67it/s]
 65%|██████▍   | 6907/10682 [59:07<23:33,  2.67it/s]
 65%|██████▍   | 6908/10682 [59:08<23:33,  2.67it/s]
 65%|██████▍   | 6909/10682 [59:08<23:33,  2.67it/s]
 65%|██████▍   | 6910/10682 [59:08<23:35,  2.67it/s]
 65%|██████▍   | 6911/10682 [59:09<23:32,  2.67it/s]
 65%|██████▍   | 6912/10682 [59:09<23:32,  2.67it/s]
 65%|██████▍   | 6913/10682 [59:09<23:31,  2.67it/s]
 65%|██████▍   | 6914/10682 [59:10<23:29,  2.67it/s]
 65%|██████▍   | 6915/10682 [59:10<23:29,  2.67it/s]
 65%|██████▍   | 6916/10682 [59:11<23:31,  2.67it/s]
 65%|██████▍   | 6917/10682 [59:11<23:30,  2.67it/s]
 65%|██████▍   | 6918/10682 [59:11<23:28,  2.67it/s]
 65%|██████▍   | 6919/10682 [59:12<23:29,  2.67it/s]
 65%|██████▍   | 6920/10682 [59:12<23:28,  2.67it/s]
 65%|██████▍   | 6921/10682 [59:12<23:29,  2.67it/s]
 65%|██████▍   | 6922/10682 [59:13<23:28,  2.67it/s]
 65%|██████▍   | 6923/10682 [59:13<23:27,  2.67it/s]
 65%|██████▍   | 6924/10682 [59:14<23:26,  2.67it/s]
 65%|██████▍   | 6925/10682 [59:14<23:26,  2.67it/s]{'loss': 4.2507, 'grad_norm': 0.3779071867465973, 'learning_rate': 0.0003318507897585944, 'epoch': 0.65}
-
                                                    

 65%|██████▍   | 6925/10682 [59:14<23:26,  2.67it/s]
 65%|██████▍   | 6926/10682 [59:14<23:28,  2.67it/s]
 65%|██████▍   | 6927/10682 [59:15<23:28,  2.67it/s]
 65%|██████▍   | 6928/10682 [59:15<23:27,  2.67it/s]
 65%|██████▍   | 6929/10682 [59:15<23:26,  2.67it/s]
 65%|██████▍   | 6930/10682 [59:16<23:25,  2.67it/s]
 65%|██████▍   | 6931/10682 [59:16<23:24,  2.67it/s]
 65%|██████▍   | 6932/10682 [59:17<23:24,  2.67it/s]
 65%|██████▍   | 6933/10682 [59:17<23:24,  2.67it/s]
 65%|██████▍   | 6934/10682 [59:17<23:23,  2.67it/s]
 65%|██████▍   | 6935/10682 [59:18<23:24,  2.67it/s]
 65%|██████▍   | 6936/10682 [59:18<23:23,  2.67it/s]
 65%|██████▍   | 6937/10682 [59:18<23:23,  2.67it/s]
 65%|██████▍   | 6938/10682 [59:19<23:21,  2.67it/s]
 65%|██████▍   | 6939/10682 [59:19<23:20,  2.67it/s]
 65%|██████▍   | 6940/10682 [59:19<23:20,  2.67it/s]
 65%|██████▍   | 6941/10682 [59:20<23:19,  2.67it/s]
 65%|██████▍   | 6942/10682 [59:20<23:21,  2.67it/s]
 65%|██████▍   | 6943/10682 [59:21<23:20,  2.67it/s]
 65%|██████▌   | 6944/10682 [59:21<23:20,  2.67it/s]
 65%|██████▌   | 6945/10682 [59:21<23:18,  2.67it/s]
 65%|██████▌   | 6946/10682 [59:22<23:19,  2.67it/s]
 65%|██████▌   | 6947/10682 [59:22<23:18,  2.67it/s]
 65%|██████▌   | 6948/10682 [59:22<23:19,  2.67it/s]
 65%|██████▌   | 6949/10682 [59:23<23:19,  2.67it/s]
 65%|██████▌   | 6950/10682 [59:23<23:18,  2.67it/s]
                                                    {'loss': 4.2509, 'grad_norm': 0.462848961353302, 'learning_rate': 0.0003280092948503962, 'epoch': 0.65}
-
 65%|██████▌   | 6950/10682 [59:23<23:18,  2.67it/s]
 65%|██████▌   | 6951/10682 [59:24<23:19,  2.67it/s]
 65%|██████▌   | 6952/10682 [59:24<23:18,  2.67it/s]
 65%|██████▌   | 6953/10682 [59:24<23:18,  2.67it/s]
 65%|██████▌   | 6954/10682 [59:25<23:18,  2.66it/s]
 65%|██████▌   | 6955/10682 [59:25<23:16,  2.67it/s]
 65%|██████▌   | 6956/10682 [59:25<23:15,  2.67it/s]
 65%|██████▌   | 6957/10682 [59:26<23:15,  2.67it/s]
 65%|██████▌   | 6958/10682 [59:26<23:16,  2.67it/s]
 65%|██████▌   | 6959/10682 [59:27<23:17,  2.66it/s]
 65%|██████▌   | 6960/10682 [59:27<23:16,  2.67it/s]
 65%|██████▌   | 6961/10682 [59:27<23:15,  2.67it/s]
 65%|██████▌   | 6962/10682 [59:28<23:15,  2.67it/s]
 65%|██████▌   | 6963/10682 [59:28<23:14,  2.67it/s]
 65%|██████▌   | 6964/10682 [59:28<23:13,  2.67it/s]
 65%|██████▌   | 6965/10682 [59:29<23:12,  2.67it/s]
 65%|██████▌   | 6966/10682 [59:29<23:13,  2.67it/s]
 65%|██████▌   | 6967/10682 [59:30<23:12,  2.67it/s]
 65%|██████▌   | 6968/10682 [59:30<23:12,  2.67it/s]
 65%|██████▌   | 6969/10682 [59:30<23:12,  2.67it/s]
 65%|██████▌   | 6970/10682 [59:31<23:11,  2.67it/s]
 65%|██████▌   | 6971/10682 [59:31<23:13,  2.66it/s]
 65%|██████▌   | 6972/10682 [59:31<23:12,  2.66it/s]
 65%|██████▌   | 6973/10682 [59:32<23:12,  2.66it/s]
 65%|██████▌   | 6974/10682 [59:32<23:09,  2.67it/s]
 65%|██████▌   | 6975/10682 [59:33<23:09,  2.67it/s]
                                                    
{'loss': 4.2459, 'grad_norm': 0.4369693100452423, 'learning_rate': 0.00032417928053832723, 'epoch': 0.65}
-
 65%|██████▌   | 6975/10682 [59:33<23:09,  2.67it/s]
 65%|██████▌   | 6976/10682 [59:33<23:10,  2.67it/s]
 65%|██████▌   | 6977/10682 [59:33<23:08,  2.67it/s]
 65%|██████▌   | 6978/10682 [59:34<23:08,  2.67it/s]
 65%|██████▌   | 6979/10682 [59:34<23:09,  2.67it/s]
 65%|██████▌   | 6980/10682 [59:34<23:07,  2.67it/s]
 65%|██████▌   | 6981/10682 [59:35<23:06,  2.67it/s]
 65%|██████▌   | 6982/10682 [59:35<23:05,  2.67it/s]
 65%|██████▌   | 6983/10682 [59:36<23:06,  2.67it/s]
 65%|██████▌   | 6984/10682 [59:36<23:05,  2.67it/s]
 65%|██████▌   | 6985/10682 [59:36<23:03,  2.67it/s]
 65%|██████▌   | 6986/10682 [59:37<23:05,  2.67it/s]
 65%|██████▌   | 6987/10682 [59:37<23:04,  2.67it/s]
 65%|██████▌   | 6988/10682 [59:37<23:05,  2.67it/s]
 65%|██████▌   | 6989/10682 [59:38<23:04,  2.67it/s]
 65%|██████▌   | 6990/10682 [59:38<23:02,  2.67it/s]
 65%|██████▌   | 6991/10682 [59:39<23:03,  2.67it/s]
 65%|██████▌   | 6992/10682 [59:39<23:02,  2.67it/s]
 65%|██████▌   | 6993/10682 [59:39<23:02,  2.67it/s]
 65%|██████▌   | 6994/10682 [59:40<23:02,  2.67it/s]
 65%|██████▌   | 6995/10682 [59:40<23:01,  2.67it/s]
 65%|██████▌   | 6996/10682 [59:40<23:00,  2.67it/s]
 66%|██████▌   | 6997/10682 [59:41<22:59,  2.67it/s]
 66%|██████▌   | 6998/10682 [59:41<22:59,  2.67it/s]
 66%|██████▌   | 6999/10682 [59:42<23:00,  2.67it/s]
 66%|██████▌   | 7000/10682 [59:42<22:58,  2.67it/s]
                                                    
{'loss': 4.2515, 'grad_norm': 0.3876669108867645, 'learning_rate': 0.0003203610024806651, 'epoch': 0.66}
-
 66%|██████▌   | 7000/10682 [59:42<22:58,  2.67it/s]
 66%|██████▌   | 7001/10682 [59:42<22:59,  2.67it/s]
 66%|██████▌   | 7002/10682 [59:43<22:59,  2.67it/s]
 66%|██████▌   | 7003/10682 [59:43<22:59,  2.67it/s]
 66%|██████▌   | 7004/10682 [59:43<22:59,  2.67it/s]
 66%|██████▌   | 7005/10682 [59:44<22:58,  2.67it/s]
 66%|██████▌   | 7006/10682 [59:44<22:57,  2.67it/s]
 66%|██████▌   | 7007/10682 [59:45<22:55,  2.67it/s]
 66%|██████▌   | 7008/10682 [59:45<22:54,  2.67it/s]
 66%|██████▌   | 7009/10682 [59:45<22:54,  2.67it/s]
 66%|██████▌   | 7010/10682 [59:46<22:53,  2.67it/s]
 66%|██████▌   | 7011/10682 [59:46<22:54,  2.67it/s]
 66%|██████▌   | 7012/10682 [59:46<22:54,  2.67it/s]
 66%|██████▌   | 7013/10682 [59:47<22:55,  2.67it/s]
 66%|██████▌   | 7014/10682 [59:47<22:55,  2.67it/s]
 66%|██████▌   | 7015/10682 [59:48<22:57,  2.66it/s]
 66%|██████▌   | 7016/10682 [59:48<22:54,  2.67it/s]
 66%|██████▌   | 7017/10682 [59:48<22:53,  2.67it/s]
 66%|██████▌   | 7018/10682 [59:49<22:52,  2.67it/s]
 66%|██████▌   | 7019/10682 [59:49<22:52,  2.67it/s]
 66%|██████▌   | 7020/10682 [59:49<22:51,  2.67it/s]
 66%|██████▌   | 7021/10682 [59:50<22:50,  2.67it/s]
 66%|██████▌   | 7022/10682 [59:50<22:51,  2.67it/s]
 66%|██████▌   | 7023/10682 [59:51<22:50,  2.67it/s]
 66%|██████▌   | 7024/10682 [59:51<22:50,  2.67it/s]
 66%|██████▌   | 7025/10682 [59:51<22:49,  2.67it/s]
                                                    
{'loss': 4.2528, 'grad_norm': 0.41266462206840515, 'learning_rate': 0.0003165547155522783, 'epoch': 0.66}
-
 66%|██████▌   | 7025/10682 [59:51<22:49,  2.67it/s]
 66%|██████▌   | 7026/10682 [59:52<22:50,  2.67it/s]
 66%|██████▌   | 7027/10682 [59:52<22:50,  2.67it/s]
 66%|██████▌   | 7028/10682 [59:52<22:52,  2.66it/s]
 66%|██████▌   | 7029/10682 [59:53<22:51,  2.66it/s]
 66%|██████▌   | 7030/10682 [59:53<22:50,  2.66it/s]
 66%|██████▌   | 7031/10682 [59:54<22:49,  2.67it/s]
 66%|██████▌   | 7032/10682 [59:54<22:48,  2.67it/s]
 66%|██████▌   | 7033/10682 [59:54<22:47,  2.67it/s]
 66%|██████▌   | 7034/10682 [59:55<22:46,  2.67it/s]
 66%|██████▌   | 7035/10682 [59:55<22:46,  2.67it/s]
 66%|██████▌   | 7036/10682 [59:55<22:46,  2.67it/s]
 66%|██████▌   | 7037/10682 [59:56<22:47,  2.66it/s]
 66%|██████▌   | 7038/10682 [59:56<22:46,  2.67it/s]
 66%|██████▌   | 7039/10682 [59:57<22:45,  2.67it/s]
 66%|██████▌   | 7040/10682 [59:57<22:46,  2.67it/s]
 66%|██████▌   | 7041/10682 [59:57<22:45,  2.67it/s]
 66%|██████▌   | 7042/10682 [59:58<22:47,  2.66it/s]
 66%|██████▌   | 7043/10682 [59:58<22:46,  2.66it/s]
 66%|██████▌   | 7044/10682 [59:58<22:45,  2.66it/s]
 66%|██████▌   | 7045/10682 [59:59<22:43,  2.67it/s]
 66%|██████▌   | 7046/10682 [59:59<22:44,  2.66it/s]
 66%|██████▌   | 7047/10682 [1:00:00<22:43,  2.66it/s]
 66%|██████▌   | 7048/10682 [1:00:00<22:44,  2.66it/s]
 66%|██████▌   | 7049/10682 [1:00:00<22:42,  2.67it/s]
 66%|██████▌   | 7050/10682 [1:00:01<22:42,  2.67it/s]
                                                      
{'loss': 4.2488, 'grad_norm': 0.388588011264801, 'learning_rate': 0.0003127606738276115, 'epoch': 0.66}
-
 66%|██████▌   | 7050/10682 [1:00:01<22:42,  2.67it/s]
 66%|██████▌   | 7051/10682 [1:00:01<22:49,  2.65it/s]
 66%|██████▌   | 7052/10682 [1:00:01<22:46,  2.66it/s]
 66%|██████▌   | 7053/10682 [1:00:02<22:46,  2.66it/s]
 66%|██████▌   | 7054/10682 [1:00:02<22:43,  2.66it/s]
 66%|██████▌   | 7055/10682 [1:00:03<22:43,  2.66it/s]
 66%|██████▌   | 7056/10682 [1:00:03<22:41,  2.66it/s]
 66%|██████▌   | 7057/10682 [1:00:03<22:40,  2.66it/s]
 66%|██████▌   | 7058/10682 [1:00:04<22:39,  2.66it/s]
 66%|██████▌   | 7059/10682 [1:00:04<22:38,  2.67it/s]
 66%|██████▌   | 7060/10682 [1:00:04<22:39,  2.66it/s]
 66%|██████▌   | 7061/10682 [1:00:05<22:37,  2.67it/s]
 66%|██████▌   | 7062/10682 [1:00:05<22:41,  2.66it/s]
 66%|██████▌   | 7063/10682 [1:00:06<22:39,  2.66it/s]
 66%|██████▌   | 7064/10682 [1:00:06<22:40,  2.66it/s]
 66%|██████▌   | 7065/10682 [1:00:06<22:39,  2.66it/s]
 66%|██████▌   | 7066/10682 [1:00:07<22:40,  2.66it/s]
 66%|██████▌   | 7067/10682 [1:00:07<22:38,  2.66it/s]
 66%|██████▌   | 7068/10682 [1:00:07<22:37,  2.66it/s]
 66%|██████▌   | 7069/10682 [1:00:08<22:37,  2.66it/s]
 66%|██████▌   | 7070/10682 [1:00:08<22:35,  2.67it/s]
 66%|██████▌   | 7071/10682 [1:00:09<22:34,  2.67it/s]
 66%|██████▌   | 7072/10682 [1:00:09<22:34,  2.67it/s]
 66%|██████▌   | 7073/10682 [1:00:09<22:34,  2.67it/s]
 66%|██████▌   | 7074/10682 [1:00:10<22:35,  2.66it/s]
 66%|██████▌   | 7075/10682 [1:00:10<22:32,  2.67it/s]
                                                      
{'loss': 4.2457, 'grad_norm': 0.44303110241889954, 'learning_rate': 0.00030897913056372755, 'epoch': 0.66}
-
 66%|██████▌   | 7075/10682 [1:00:10<22:32,  2.67it/s]
 66%|██████▌   | 7076/10682 [1:00:11<22:35,  2.66it/s]
 66%|██████▋   | 7077/10682 [1:00:11<22:33,  2.66it/s]
 66%|██████▋   | 7078/10682 [1:00:11<22:31,  2.67it/s]
 66%|██████▋   | 7079/10682 [1:00:12<22:29,  2.67it/s]
 66%|██████▋   | 7080/10682 [1:00:12<22:31,  2.67it/s]
 66%|██████▋   | 7081/10682 [1:00:12<22:30,  2.67it/s]
 66%|██████▋   | 7082/10682 [1:00:13<22:30,  2.67it/s]
 66%|██████▋   | 7083/10682 [1:00:13<22:31,  2.66it/s]
 66%|██████▋   | 7084/10682 [1:00:14<22:30,  2.66it/s]
 66%|██████▋   | 7085/10682 [1:00:14<22:29,  2.67it/s]
 66%|██████▋   | 7086/10682 [1:00:14<22:28,  2.67it/s]
 66%|██████▋   | 7087/10682 [1:00:15<22:27,  2.67it/s]
 66%|██████▋   | 7088/10682 [1:00:15<22:25,  2.67it/s]
 66%|██████▋   | 7089/10682 [1:00:15<22:24,  2.67it/s]
 66%|██████▋   | 7090/10682 [1:00:16<22:23,  2.67it/s]
 66%|██████▋   | 7091/10682 [1:00:16<22:24,  2.67it/s]
 66%|██████▋   | 7092/10682 [1:00:16<22:25,  2.67it/s]
 66%|██████▋   | 7093/10682 [1:00:17<22:28,  2.66it/s]
 66%|██████▋   | 7094/10682 [1:00:17<22:27,  2.66it/s]
 66%|██████▋   | 7095/10682 [1:00:18<22:27,  2.66it/s]
 66%|██████▋   | 7096/10682 [1:00:18<22:25,  2.66it/s]
 66%|██████▋   | 7097/10682 [1:00:18<22:23,  2.67it/s]
 66%|██████▋   | 7098/10682 [1:00:19<22:22,  2.67it/s]
 66%|██████▋   | 7099/10682 [1:00:19<22:21,  2.67it/s]
 66%|██████▋   | 7100/10682 [1:00:19<22:21,  2.67it/s]
                                                      
{'loss': 4.2402, 'grad_norm': 0.4012932777404785, 'learning_rate': 0.00030521033818340054, 'epoch': 0.66}
-
 66%|██████▋   | 7100/10682 [1:00:20<22:21,  2.67it/s]
 66%|██████▋   | 7101/10682 [1:00:20<22:27,  2.66it/s]
 66%|██████▋   | 7102/10682 [1:00:20<22:25,  2.66it/s]
 66%|██████▋   | 7103/10682 [1:00:21<22:25,  2.66it/s]
 67%|██████▋   | 7104/10682 [1:00:21<22:24,  2.66it/s]
 67%|██████▋   | 7105/10682 [1:00:21<22:22,  2.67it/s]
 67%|██████▋   | 7106/10682 [1:00:22<22:21,  2.67it/s]
 67%|██████▋   | 7107/10682 [1:00:22<22:20,  2.67it/s]
 67%|██████▋   | 7108/10682 [1:00:23<22:21,  2.66it/s]
 67%|██████▋   | 7109/10682 [1:00:23<22:20,  2.66it/s]
 67%|██████▋   | 7110/10682 [1:00:23<22:20,  2.66it/s]
 67%|██████▋   | 7111/10682 [1:00:24<22:19,  2.67it/s]
 67%|██████▋   | 7112/10682 [1:00:24<22:17,  2.67it/s]
 67%|██████▋   | 7113/10682 [1:00:24<22:19,  2.67it/s]
 67%|██████▋   | 7114/10682 [1:00:25<22:18,  2.67it/s]
 67%|██████▋   | 7115/10682 [1:00:25<22:16,  2.67it/s]
 67%|██████▋   | 7116/10682 [1:00:26<22:17,  2.67it/s]
 67%|██████▋   | 7117/10682 [1:00:26<22:15,  2.67it/s]
 67%|██████▋   | 7118/10682 [1:00:26<22:14,  2.67it/s]
 67%|██████▋   | 7119/10682 [1:00:27<22:15,  2.67it/s]
 67%|██████▋   | 7120/10682 [1:00:27<22:16,  2.66it/s]
 67%|██████▋   | 7121/10682 [1:00:27<22:14,  2.67it/s]
 67%|██████▋   | 7122/10682 [1:00:28<22:16,  2.66it/s]
 67%|██████▋   | 7123/10682 [1:00:28<22:14,  2.67it/s]
 67%|██████▋   | 7124/10682 [1:00:29<22:14,  2.67it/s]
 67%|██████▋   | 7125/10682 [1:00:29<22:13,  2.67it/s]{'loss': 4.2507, 'grad_norm': 0.39986035227775574, 'learning_rate': 0.0003014545482582672, 'epoch': 0.67}
                                                      
-
 67%|██████▋   | 7125/10682 [1:00:29<22:13,  2.67it/s]
 67%|██████▋   | 7126/10682 [1:00:29<22:21,  2.65it/s]
 67%|██████▋   | 7127/10682 [1:00:30<22:18,  2.66it/s]
 67%|██████▋   | 7128/10682 [1:00:30<22:15,  2.66it/s]
 67%|██████▋   | 7129/10682 [1:00:30<22:13,  2.66it/s]
 67%|██████▋   | 7130/10682 [1:00:31<22:12,  2.67it/s]
 67%|██████▋   | 7131/10682 [1:00:31<22:12,  2.66it/s]
 67%|██████▋   | 7132/10682 [1:00:32<22:11,  2.67it/s]
 67%|██████▋   | 7133/10682 [1:00:32<22:11,  2.67it/s]
 67%|██████▋   | 7134/10682 [1:00:32<22:11,  2.67it/s]
 67%|██████▋   | 7135/10682 [1:00:33<22:12,  2.66it/s]
 67%|██████▋   | 7136/10682 [1:00:33<22:11,  2.66it/s]
 67%|██████▋   | 7137/10682 [1:00:33<22:11,  2.66it/s]
 67%|██████▋   | 7138/10682 [1:00:34<22:09,  2.67it/s]
 67%|██████▋   | 7139/10682 [1:00:34<22:08,  2.67it/s]
 67%|██████▋   | 7140/10682 [1:00:35<22:07,  2.67it/s]
 67%|██████▋   | 7141/10682 [1:00:35<22:06,  2.67it/s]
 67%|██████▋   | 7142/10682 [1:00:35<22:06,  2.67it/s]
 67%|██████▋   | 7143/10682 [1:00:36<22:05,  2.67it/s]
 67%|██████▋   | 7144/10682 [1:00:36<22:05,  2.67it/s]
 67%|████���█▋   | 7145/10682 [1:00:36<22:05,  2.67it/s]
 67%|██████▋   | 7146/10682 [1:00:37<22:06,  2.67it/s]
 67%|██████▋   | 7147/10682 [1:00:37<22:05,  2.67it/s]
 67%|██████▋   | 7148/10682 [1:00:38<22:05,  2.67it/s]
 67%|██████▋   | 7149/10682 [1:00:38<22:05,  2.67it/s]
 67%|██████▋   | 7150/10682 [1:00:38<22:04,  2.67it/s]{'loss': 4.2423, 'grad_norm': 0.4048359990119934, 'learning_rate': 0.0002977120114920343, 'epoch': 0.67}
                                                      
-
 67%|██████▋   | 7150/10682 [1:00:38<22:04,  2.67it/s]
 67%|██████▋   | 7151/10682 [1:00:39<22:05,  2.66it/s]
 67%|██████▋   | 7152/10682 [1:00:39<22:02,  2.67it/s]
 67%|██████▋   | 7153/10682 [1:00:39<22:02,  2.67it/s]
 67%|██████▋   | 7154/10682 [1:00:40<22:00,  2.67it/s]
 67%|██████▋   | 7155/10682 [1:00:40<22:00,  2.67it/s]
 67%|██████▋   | 7156/10682 [1:00:41<21:59,  2.67it/s]
 67%|██████▋   | 7157/10682 [1:00:41<22:00,  2.67it/s]
 67%|██████▋   | 7158/10682 [1:00:41<22:00,  2.67it/s]
 67%|██████▋   | 7159/10682 [1:00:42<22:00,  2.67it/s]
 67%|██████▋   | 7160/10682 [1:00:42<22:01,  2.67it/s]
 67%|██████▋   | 7161/10682 [1:00:42<22:00,  2.67it/s]
 67%|██████▋   | 7162/10682 [1:00:43<22:02,  2.66it/s]
 67%|██████▋   | 7163/10682 [1:00:43<22:00,  2.67it/s]
 67%|██████▋   | 7164/10682 [1:00:44<21:59,  2.67it/s]
 67%|██████▋   | 7165/10682 [1:00:44<21:58,  2.67it/s]
 67%|██████▋   | 7166/10682 [1:00:44<21:59,  2.66it/s]
 67%|██████▋   | 7167/10682 [1:00:45<21:58,  2.67it/s]
 67%|██████▋   | 7168/10682 [1:00:45<21:57,  2.67it/s]
 67%|██████▋   | 7169/10682 [1:00:45<21:56,  2.67it/s]
 67%|██████▋   | 7170/10682 [1:00:46<21:56,  2.67it/s]
 67%|██████▋   | 7171/10682 [1:00:46<21:55,  2.67it/s]
 67%|██████▋   | 7172/10682 [1:00:47<21:54,  2.67it/s]
 67%|██████▋   | 7173/10682 [1:00:47<21:57,  2.66it/s]
 67%|██████▋   | 7174/10682 [1:00:47<21:54,  2.67it/s]
 67%|██████▋   | 7175/10682 [1:00:48<21:55,  2.67it/s]
                                                      
{'loss': 4.2402, 'grad_norm': 0.39884158968925476, 'learning_rate': 0.00029398297770374336, 'epoch': 0.67}
-
 67%|██████▋   | 7175/10682 [1:00:48<21:55,  2.67it/s]
 67%|██████▋   | 7176/10682 [1:00:48<21:56,  2.66it/s]
 67%|██████▋   | 7177/10682 [1:00:48<21:55,  2.66it/s]
 67%|██████▋   | 7178/10682 [1:00:49<21:54,  2.67it/s]
 67%|██████▋   | 7179/10682 [1:00:49<21:53,  2.67it/s]
 67%|██████▋   | 7180/10682 [1:00:50<21:54,  2.66it/s]
 67%|██████▋   | 7181/10682 [1:00:50<21:52,  2.67it/s]
 67%|██████▋   | 7182/10682 [1:00:50<21:53,  2.67it/s]
 67%|██████▋   | 7183/10682 [1:00:51<21:52,  2.67it/s]
 67%|██████▋   | 7184/10682 [1:00:51<21:51,  2.67it/s]
 67%|██████▋   | 7185/10682 [1:00:51<21:49,  2.67it/s]
 67%|██████▋   | 7186/10682 [1:00:52<21:52,  2.66it/s]
 67%|██████▋   | 7187/10682 [1:00:52<21:50,  2.67it/s]
 67%|██████▋   | 7188/10682 [1:00:53<21:50,  2.67it/s]
 67%|██████▋   | 7189/10682 [1:00:53<21:54,  2.66it/s]
 67%|██████▋   | 7190/10682 [1:00:53<21:52,  2.66it/s]
 67%|██████▋   | 7191/10682 [1:00:54<21:50,  2.66it/s]
 67%|██████▋   | 7192/10682 [1:00:54<21:49,  2.67it/s]
 67%|██████▋   | 7193/10682 [1:00:54<21:49,  2.66it/s]
 67%|██████▋   | 7194/10682 [1:00:55<21:48,  2.67it/s]
 67%|██████▋   | 7195/10682 [1:00:55<21:46,  2.67it/s]
 67%|██████▋   | 7196/10682 [1:00:56<21:46,  2.67it/s]
 67%|██████▋   | 7197/10682 [1:00:56<21:46,  2.67it/s]
 67%|██████▋   | 7198/10682 [1:00:56<21:45,  2.67it/s]
 67%|██████▋   | 7199/10682 [1:00:57<21:44,  2.67it/s]
 67%|██████▋   | 7200/10682 [1:00:57<21:45,  2.67it/s]
                                                      {'loss': 4.2417, 'grad_norm': 0.41083720326423645, 'learning_rate': 0.0002902676958110948, 'epoch': 0.67}
-

 67%|██████▋   | 7200/10682 [1:00:57<21:45,  2.67it/s]
 67%|██████▋   | 7201/10682 [1:00:57<21:45,  2.67it/s]
 67%|██████▋   | 7202/10682 [1:00:58<21:45,  2.66it/s]
 67%|██████▋   | 7203/10682 [1:00:58<21:44,  2.67it/s]
 67%|██████▋   | 7204/10682 [1:00:59<21:44,  2.67it/s]
 67%|██████▋   | 7205/10682 [1:00:59<21:42,  2.67it/s]
 67%|██████▋   | 7206/10682 [1:00:59<21:42,  2.67it/s]
 67%|██████▋   | 7207/10682 [1:01:00<21:43,  2.67it/s]
 67%|██████▋   | 7208/10682 [1:01:00<21:41,  2.67it/s]
 67%|██████▋   | 7209/10682 [1:01:00<21:41,  2.67it/s]
 67%|██████▋   | 7210/10682 [1:01:01<21:40,  2.67it/s]
 68%|██████▊   | 7211/10682 [1:01:01<21:42,  2.67it/s]
 68%|██████▊   | 7212/10682 [1:01:02<21:40,  2.67it/s]
 68%|██████▊   | 7213/10682 [1:01:02<21:41,  2.67it/s]
 68%|██████▊   | 7214/10682 [1:01:02<21:40,  2.67it/s]
 68%|██████▊   | 7215/10682 [1:01:03<21:40,  2.67it/s]
 68%|██████▊   | 7216/10682 [1:01:03<21:42,  2.66it/s]
 68%|██████▊   | 7217/10682 [1:01:03<21:41,  2.66it/s]
 68%|██████▊   | 7218/10682 [1:01:04<21:41,  2.66it/s]
 68%|██████▊   | 7219/10682 [1:01:04<21:41,  2.66it/s]
 68%|██████▊   | 7220/10682 [1:01:05<21:40,  2.66it/s]
 68%|██████▊   | 7221/10682 [1:01:05<21:38,  2.66it/s]
 68%|██████▊   | 7222/10682 [1:01:05<21:37,  2.67it/s]
 68%|██████▊   | 7223/10682 [1:01:06<21:36,  2.67it/s]
 68%|██████▊   | 7224/10682 [1:01:06<21:35,  2.67it/s]
 68%|██████▊   | 7225/10682 [1:01:06<21:35,  2.67it/s]{'loss': 4.2449, 'grad_norm': 0.40563198924064636, 'learning_rate': 0.0002865664138138335, 'epoch': 0.68}
-
                                                      

 68%|██████▊   | 7225/10682 [1:01:06<21:35,  2.67it/s]
 68%|██████▊   | 7226/10682 [1:01:07<21:36,  2.67it/s]
 68%|██████▊   | 7227/10682 [1:01:07<21:37,  2.66it/s]
 68%|██████▊   | 7228/10682 [1:01:08<21:35,  2.67it/s]
 68%|██████▊   | 7229/10682 [1:01:08<21:35,  2.66it/s]
 68%|██████▊   | 7230/10682 [1:01:08<21:35,  2.67it/s]
 68%|██████▊   | 7231/10682 [1:01:09<21:36,  2.66it/s]
 68%|██████▊   | 7232/10682 [1:01:09<21:35,  2.66it/s]
 68%|██████▊   | 7233/10682 [1:01:09<21:34,  2.66it/s]
 68%|██████▊   | 7234/10682 [1:01:10<21:33,  2.67it/s]
 68%|██████▊   | 7235/10682 [1:01:10<21:32,  2.67it/s]
 68%|██████▊   | 7236/10682 [1:01:11<21:30,  2.67it/s]
 68%|██████▊   | 7237/10682 [1:01:11<21:30,  2.67it/s]
 68%|██████▊   | 7238/10682 [1:01:11<21:30,  2.67it/s]
 68%|██████▊   | 7239/10682 [1:01:12<21:30,  2.67it/s]
 68%|██████▊   | 7240/10682 [1:01:12<21:33,  2.66it/s]
 68%|██████▊   | 7241/10682 [1:01:12<21:33,  2.66it/s]
 68%|██████▊   | 7242/10682 [1:01:13<21:34,  2.66it/s]
 68%|██████▊   | 7243/10682 [1:01:13<21:32,  2.66it/s]
 68%|██████▊   | 7244/10682 [1:01:14<21:31,  2.66it/s]
 68%|██████▊   | 7245/10682 [1:01:14<21:29,  2.67it/s]
 68%|██████▊   | 7246/10682 [1:01:14<21:28,  2.67it/s]
 68%|██████▊   | 7247/10682 [1:01:15<21:28,  2.67it/s]
 68%|██████▊   | 7248/10682 [1:01:15<21:26,  2.67it/s]
 68%|██████▊   | 7249/10682 [1:01:15<21:26,  2.67it/s]
 68%|██████▊   | 7250/10682 [1:01:16<21:26,  2.67it/s]
                                                      
{'loss': 4.2442, 'grad_norm': 0.48112356662750244, 'learning_rate': 0.0002828793787771929, 'epoch': 0.68}
-
 68%|██████▊   | 7250/10682 [1:01:16<21:26,  2.67it/s]
 68%|██████▊   | 7251/10682 [1:01:16<21:27,  2.66it/s]
 68%|██████▊   | 7252/10682 [1:01:17<21:26,  2.67it/s]
 68%|██████▊   | 7253/10682 [1:01:17<21:28,  2.66it/s]
 68%|██████▊   | 7254/10682 [1:01:17<21:27,  2.66it/s]
 68%|██████▊   | 7255/10682 [1:01:18<21:24,  2.67it/s]
 68%|██████▊   | 7256/10682 [1:01:18<21:26,  2.66it/s]
 68%|██████▊   | 7257/10682 [1:01:18<21:26,  2.66it/s]
 68%|██████▊   | 7258/10682 [1:01:19<21:24,  2.67it/s]
 68%|██████▊   | 7259/10682 [1:01:19<21:24,  2.66it/s]
 68%|██████▊   | 7260/10682 [1:01:20<21:25,  2.66it/s]
 68%|██████▊   | 7261/10682 [1:01:20<21:24,  2.66it/s]
 68%|██████▊   | 7262/10682 [1:01:20<21:21,  2.67it/s]
 68%|██████▊   | 7263/10682 [1:01:21<21:21,  2.67it/s]
 68%|██████▊   | 7264/10682 [1:01:21<21:20,  2.67it/s]
 68%|██████▊   | 7265/10682 [1:01:21<21:20,  2.67it/s]
 68%|██████▊   | 7266/10682 [1:01:22<21:19,  2.67it/s]
 68%|██████▊   | 7267/10682 [1:01:22<21:22,  2.66it/s]
 68%|██████▊   | 7268/10682 [1:01:23<21:20,  2.67it/s]
 68%|██████▊   | 7269/10682 [1:01:23<21:20,  2.67it/s]
 68%|██████▊   | 7270/10682 [1:01:23<21:20,  2.67it/s]
 68%|██████▊   | 7271/10682 [1:01:24<21:20,  2.66it/s]
 68%|██████▊   | 7272/10682 [1:01:24<21:21,  2.66it/s]
 68%|██████▊   | 7273/10682 [1:01:24<21:19,  2.66it/s]
 68%|██████▊   | 7274/10682 [1:01:25<21:19,  2.66it/s]
 68%|██████▊   | 7275/10682 [1:01:25<21:17,  2.67it/s]{'loss': 4.2376, 'grad_norm': 0.4094863831996918, 'learning_rate': 0.0002792068368154043, 'epoch': 0.68}
                                                      
-
 68%|██████▊   | 7275/10682 [1:01:25<21:17,  2.67it/s]
 68%|██████▊   | 7276/10682 [1:01:26<21:19,  2.66it/s]
 68%|██████▊   | 7277/10682 [1:01:26<21:18,  2.66it/s]
 68%|██████▊   | 7278/10682 [1:01:26<21:18,  2.66it/s]
 68%|██████▊   | 7279/10682 [1:01:27<21:16,  2.67it/s]
 68%|██████▊   | 7280/10682 [1:01:27<21:17,  2.66it/s]
 68%|██████▊   | 7281/10682 [1:01:27<21:16,  2.66it/s]
 68%|██████▊   | 7282/10682 [1:01:28<21:18,  2.66it/s]
 68%|██████▊   | 7283/10682 [1:01:28<21:18,  2.66it/s]
 68%|██████▊   | 7284/10682 [1:01:29<21:18,  2.66it/s]
 68%|██████▊   | 7285/10682 [1:01:29<21:16,  2.66it/s]
 68%|██████▊   | 7286/10682 [1:01:29<21:14,  2.66it/s]
 68%|██████▊   | 7287/10682 [1:01:30<21:15,  2.66it/s]
 68%|██████▊   | 7288/10682 [1:01:30<21:13,  2.66it/s]
 68%|██████▊   | 7289/10682 [1:01:30<21:12,  2.67it/s]
 68%|██████▊   | 7290/10682 [1:01:31<21:12,  2.67it/s]
 68%|██████▊   | 7291/10682 [1:01:31<21:11,  2.67it/s]
 68%|██████▊   | 7292/10682 [1:01:32<21:11,  2.67it/s]
 68%|██████▊   | 7293/10682 [1:01:32<21:10,  2.67it/s]
 68%|██████▊   | 7294/10682 [1:01:32<21:12,  2.66it/s]
 68%|██████▊   | 7295/10682 [1:01:33<21:11,  2.66it/s]
 68%|██████▊   | 7296/10682 [1:01:33<21:11,  2.66it/s]
 68%|██████▊   | 7297/10682 [1:01:33<21:10,  2.66it/s]
 68%|██████▊   | 7298/10682 [1:01:34<21:08,  2.67it/s]
 68%|██████▊   | 7299/10682 [1:01:34<21:09,  2.66it/s]
 68%|██████▊   | 7300/10682 [1:01:35<21:08,  2.67it/s]
                                                      {'loss': 4.2437, 'grad_norm': 0.38206663727760315, 'learning_rate': 0.0002755490330752679, 'epoch': 0.68}
-

 68%|██████▊   | 7300/10682 [1:01:35<21:08,  2.67it/s]
 68%|██████▊   | 7301/10682 [1:01:35<21:09,  2.66it/s]
 68%|██████▊   | 7302/10682 [1:01:35<21:06,  2.67it/s]
 68%|██████▊   | 7303/10682 [1:01:36<21:06,  2.67it/s]
 68%|██████▊   | 7304/10682 [1:01:36<21:06,  2.67it/s]
 68%|██████▊   | 7305/10682 [1:01:36<21:05,  2.67it/s]
 68%|██████▊   | 7306/10682 [1:01:37<21:04,  2.67it/s]
 68%|██████▊   | 7307/10682 [1:01:37<21:04,  2.67it/s]
 68%|██████▊   | 7308/10682 [1:01:38<21:05,  2.67it/s]
 68%|██████▊   | 7309/10682 [1:01:38<21:06,  2.66it/s]
 68%|██████▊   | 7310/10682 [1:01:38<21:06,  2.66it/s]
 68%|██████▊   | 7311/10682 [1:01:39<21:04,  2.67it/s]
 68%|██████▊   | 7312/10682 [1:01:39<21:03,  2.67it/s]
 68%|██████▊   | 7313/10682 [1:01:39<21:02,  2.67it/s]
 68%|██████▊   | 7314/10682 [1:01:40<21:04,  2.66it/s]
 68%|██████▊   | 7315/10682 [1:01:40<21:05,  2.66it/s]
 68%|██████▊   | 7316/10682 [1:01:41<21:02,  2.67it/s]
 68%|██████▊   | 7317/10682 [1:01:41<21:02,  2.67it/s]
 69%|██████▊   | 7318/10682 [1:01:41<21:01,  2.67it/s]
 69%|██████▊   | 7319/10682 [1:01:42<20:59,  2.67it/s]
 69%|██████▊   | 7320/10682 [1:01:42<21:00,  2.67it/s]
 69%|██████▊   | 7321/10682 [1:01:42<21:02,  2.66it/s]
 69%|██████▊   | 7322/10682 [1:01:43<21:00,  2.67it/s]
 69%|██████▊   | 7323/10682 [1:01:43<21:01,  2.66it/s]
 69%|██████▊   | 7324/10682 [1:01:44<21:00,  2.66it/s]
 69%|██████▊   | 7325/10682 [1:01:44<21:00,  2.66it/s]
                                                      {'loss': 4.2337, 'grad_norm': 0.3926284611225128, 'learning_rate': 0.0002719062117197888, 'epoch': 0.69}
-

 69%|██████▊   | 7325/10682 [1:01:44<21:00,  2.66it/s]
 69%|██████▊   | 7326/10682 [1:01:44<20:59,  2.66it/s]
 69%|██████▊   | 7327/10682 [1:01:45<20:57,  2.67it/s]
 69%|██████▊   | 7328/10682 [1:01:45<20:57,  2.67it/s]
 69%|██████▊   | 7329/10682 [1:01:45<20:56,  2.67it/s]
 69%|██████▊   | 7330/10682 [1:01:46<20:57,  2.67it/s]
 69%|██████▊   | 7331/10682 [1:01:46<20:55,  2.67it/s]
 69%|██████▊   | 7332/10682 [1:01:47<20:55,  2.67it/s]
 69%|██████▊   | 7333/10682 [1:01:47<20:56,  2.67it/s]
 69%|██████▊   | 7334/10682 [1:01:47<20:57,  2.66it/s]
 69%|██████▊   | 7335/10682 [1:01:48<20:55,  2.67it/s]
 69%|██████▊   | 7336/10682 [1:01:48<20:55,  2.66it/s]
 69%|██████▊   | 7337/10682 [1:01:48<20:54,  2.67it/s]
 69%|██████▊   | 7338/10682 [1:01:49<20:53,  2.67it/s]
 69%|██████▊   | 7339/10682 [1:01:49<20:52,  2.67it/s]
 69%|██████▊   | 7340/10682 [1:01:50<20:51,  2.67it/s]
 69%|██████▊   | 7341/10682 [1:01:50<20:50,  2.67it/s]
 69%|██████▊   | 7342/10682 [1:01:50<20:50,  2.67it/s]
 69%|██████▊   | 7343/10682 [1:01:51<20:51,  2.67it/s]
 69%|██████▉   | 7344/10682 [1:01:51<20:52,  2.67it/s]
 69%|██████▉   | 7345/10682 [1:01:51<20:49,  2.67it/s]
 69%|██████▉   | 7346/10682 [1:01:52<20:49,  2.67it/s]
 69%|██████▉   | 7347/10682 [1:01:52<20:52,  2.66it/s]
 69%|██████▉   | 7348/10682 [1:01:53<20:50,  2.67it/s]
 69%|██████▉   | 7349/10682 [1:01:53<20:48,  2.67it/s]
 69%|██████▉   | 7350/10682 [1:01:53<20:49,  2.67it/s]{'loss': 4.237, 'grad_norm': 0.4417230486869812, 'learning_rate': 0.00026827861591187944, 'epoch': 0.69}
-
                                                      

 69%|██████▉   | 7350/10682 [1:01:53<20:49,  2.67it/s]
 69%|██████▉   | 7351/10682 [1:01:54<20:51,  2.66it/s]
 69%|██████▉   | 7352/10682 [1:01:54<20:49,  2.66it/s]
 69%|██████▉   | 7353/10682 [1:01:54<20:47,  2.67it/s]
 69%|██████▉   | 7354/10682 [1:01:55<20:47,  2.67it/s]
 69%|██████▉   | 7355/10682 [1:01:55<20:46,  2.67it/s]
 69%|██████▉   | 7356/10682 [1:01:56<20:46,  2.67it/s]
 69%|██████▉   | 7357/10682 [1:01:56<20:45,  2.67it/s]
 69%|██████▉   | 7358/10682 [1:01:56<20:45,  2.67it/s]
 69%|██████▉   | 7359/10682 [1:01:57<20:45,  2.67it/s]
 69%|██████▉   | 7360/10682 [1:01:57<20:44,  2.67it/s]
 69%|██████▉   | 7361/10682 [1:01:57<20:45,  2.67it/s]
 69%|██████▉   | 7362/10682 [1:01:58<20:45,  2.67it/s]
 69%|██████▉   | 7363/10682 [1:01:58<20:47,  2.66it/s]
 69%|██████▉   | 7364/10682 [1:01:59<20:46,  2.66it/s]
 69%|██████▉   | 7365/10682 [1:01:59<20:45,  2.66it/s]
 69%|██████▉   | 7366/10682 [1:01:59<20:44,  2.66it/s]
 69%|██████▉   | 7367/10682 [1:02:00<20:44,  2.66it/s]
 69%|██████▉   | 7368/10682 [1:02:00<20:43,  2.66it/s]
 69%|██████▉   | 7369/10682 [1:02:00<20:42,  2.67it/s]
 69%|██████▉   | 7370/10682 [1:02:01<20:41,  2.67it/s]
 69%|██████▉   | 7371/10682 [1:02:01<20:42,  2.67it/s]
 69%|██████▉   | 7372/10682 [1:02:02<20:42,  2.66it/s]
 69%|██████▉   | 7373/10682 [1:02:02<20:41,  2.67it/s]
 69%|██████▉   | 7374/10682 [1:02:02<20:42,  2.66it/s]
 69%|██████▉   | 7375/10682 [1:02:03<20:41,  2.66it/s]
                                                      {'loss': 4.2389, 'grad_norm': 0.42905041575431824, 'learning_rate': 0.00026466648779812794, 'epoch': 0.69}
-
 69%|██████▉   | 7375/10682 [1:02:03<20:41,  2.66it/s]
 69%|██████▉   | 7376/10682 [1:02:03<20:43,  2.66it/s]
 69%|██████▉   | 7377/10682 [1:02:03<20:42,  2.66it/s]
 69%|██████▉   | 7378/10682 [1:02:04<20:40,  2.66it/s]
 69%|██████▉   | 7379/10682 [1:02:04<20:37,  2.67it/s]
 69%|██████▉   | 7380/10682 [1:02:05<20:37,  2.67it/s]
 69%|██████▉   | 7381/10682 [1:02:05<20:36,  2.67it/s]
 69%|██████▉   | 7382/10682 [1:02:05<20:37,  2.67it/s]
 69%|██████▉   | 7383/10682 [1:02:06<20:36,  2.67it/s]
 69%|██████▉   | 7384/10682 [1:02:06<20:36,  2.67it/s]
 69%|██████▉   | 7385/10682 [1:02:06<20:36,  2.67it/s]
 69%|██████▉   | 7386/10682 [1:02:07<20:35,  2.67it/s]
 69%|██████▉   | 7387/10682 [1:02:07<20:35,  2.67it/s]
 69%|██████▉   | 7388/10682 [1:02:08<20:35,  2.67it/s]
 69%|██████▉   | 7389/10682 [1:02:08<20:34,  2.67it/s]
 69%|██████▉   | 7390/10682 [1:02:08<20:35,  2.66it/s]
 69%|██████▉   | 7391/10682 [1:02:09<20:35,  2.66it/s]
 69%|██████▉   | 7392/10682 [1:02:09<20:33,  2.67it/s]
 69%|██████▉   | 7393/10682 [1:02:09<20:32,  2.67it/s]
 69%|██████▉   | 7394/10682 [1:02:10<20:32,  2.67it/s]
 69%|██████▉   | 7395/10682 [1:02:10<20:31,  2.67it/s]
 69%|██████▉   | 7396/10682 [1:02:11<20:30,  2.67it/s]
 69%|██████▉   | 7397/10682 [1:02:11<20:31,  2.67it/s]
 69%|██████▉   | 7398/10682 [1:02:11<20:30,  2.67it/s]
 69%|██████▉   | 7399/10682 [1:02:12<20:29,  2.67it/s]
 69%|██████▉   | 7400/10682 [1:02:12<20:29,  2.67it/s]{'loss': 4.2353, 'grad_norm': 0.40220603346824646, 'learning_rate': 0.0002610700684926339, 'epoch': 0.69}
                                                      
-
 69%|██████▉   | 7400/10682 [1:02:12<20:29,  2.67it/s]
 69%|██████▉   | 7401/10682 [1:02:12<20:30,  2.67it/s]
 69%|██████▉   | 7402/10682 [1:02:13<20:31,  2.66it/s]
 69%|██████▉   | 7403/10682 [1:02:13<20:31,  2.66it/s]
 69%|██████▉   | 7404/10682 [1:02:14<20:30,  2.66it/s]
 69%|██████▉   | 7405/10682 [1:02:14<20:30,  2.66it/s]
 69%|██████▉   | 7406/10682 [1:02:14<20:29,  2.66it/s]
 69%|██████▉   | 7407/10682 [1:02:15<20:28,  2.67it/s]
 69%|██████▉   | 7408/10682 [1:02:15<20:28,  2.66it/s]
 69%|██████▉   | 7409/10682 [1:02:15<20:27,  2.67it/s]
 69%|██████▉   | 7410/10682 [1:02:16<20:26,  2.67it/s]
 69%|██████▉   | 7411/10682 [1:02:16<20:24,  2.67it/s]
 69%|██████▉   | 7412/10682 [1:02:17<20:24,  2.67it/s]
 69%|██████▉   | 7413/10682 [1:02:17<20:25,  2.67it/s]
 69%|██████▉   | 7414/10682 [1:02:17<20:24,  2.67it/s]
 69%|██████▉   | 7415/10682 [1:02:18<20:25,  2.67it/s]
 69%|██████▉   | 7416/10682 [1:02:18<20:28,  2.66it/s]
 69%|██████▉   | 7417/10682 [1:02:18<20:27,  2.66it/s]
 69%|██████▉   | 7418/10682 [1:02:19<20:27,  2.66it/s]
 69%|██████▉   | 7419/10682 [1:02:19<20:26,  2.66it/s]
 69%|██████▉   | 7420/10682 [1:02:20<20:26,  2.66it/s]
 69%|██████▉   | 7421/10682 [1:02:20<20:25,  2.66it/s]
 69%|██████▉   | 7422/10682 [1:02:20<20:24,  2.66it/s]
 69%|██████▉   | 7423/10682 [1:02:21<20:23,  2.66it/s]
 70%|██████▉   | 7424/10682 [1:02:21<20:22,  2.67it/s]
 70%|██████▉   | 7425/10682 [1:02:21<20:20,  2.67it/s]{'loss': 4.2398, 'grad_norm': 0.36822500824928284, 'learning_rate': 0.00025748959806091423, 'epoch': 0.7}
                                                      
-
 70%|██████▉   | 7425/10682 [1:02:21<20:20,  2.67it/s]
 70%|██████▉   | 7426/10682 [1:02:22<20:21,  2.67it/s]
 70%|██████▉   | 7427/10682 [1:02:22<20:19,  2.67it/s]
 70%|██████▉   | 7428/10682 [1:02:23<20:19,  2.67it/s]
 70%|██████▉   | 7429/10682 [1:02:23<20:19,  2.67it/s]
 70%|██████▉   | 7430/10682 [1:02:23<20:21,  2.66it/s]
 70%|██████▉   | 7431/10682 [1:02:24<20:19,  2.67it/s]
 70%|██████▉   | 7432/10682 [1:02:24<20:18,  2.67it/s]
 70%|██████▉   | 7433/10682 [1:02:24<20:18,  2.67it/s]
 70%|██████▉   | 7434/10682 [1:02:25<20:19,  2.66it/s]
 70%|██████▉   | 7435/10682 [1:02:25<20:17,  2.67it/s]
 70%|██████▉   | 7436/10682 [1:02:26<20:15,  2.67it/s]
 70%|██████▉   | 7437/10682 [1:02:26<20:15,  2.67it/s]
 70%|██████▉   | 7438/10682 [1:02:26<20:14,  2.67it/s]
 70%|██████▉   | 7439/10682 [1:02:27<20:14,  2.67it/s]
 70%|██████▉   | 7440/10682 [1:02:27<20:13,  2.67it/s]
 70%|██████▉   | 7441/10682 [1:02:27<20:14,  2.67it/s]
 70%|██████▉   | 7442/10682 [1:02:28<20:14,  2.67it/s]
 70%|██████▉   | 7443/10682 [1:02:28<20:15,  2.66it/s]
 70%|██████▉   | 7444/10682 [1:02:29<20:15,  2.66it/s]
 70%|██████▉   | 7445/10682 [1:02:29<20:13,  2.67it/s]
 70%|██████▉   | 7446/10682 [1:02:29<20:13,  2.67it/s]
 70%|██████▉   | 7447/10682 [1:02:30<20:12,  2.67it/s]
 70%|██████▉   | 7448/10682 [1:02:30<20:12,  2.67it/s]
 70%|██████▉   | 7449/10682 [1:02:30<20:12,  2.67it/s]
 70%|██████▉   | 7450/10682 [1:02:31<20:12,  2.67it/s]{'loss': 4.2273, 'grad_norm': 0.47050049901008606, 'learning_rate': 0.00025392531550387867, 'epoch': 0.7}
                                                      
-
 70%|██████▉   | 7450/10682 [1:02:31<20:12,  2.67it/s]
 70%|██████▉   | 7451/10682 [1:02:31<20:13,  2.66it/s]
 70%|██████▉   | 7452/10682 [1:02:32<20:11,  2.67it/s]
 70%|██████▉   | 7453/10682 [1:02:32<20:11,  2.67it/s]
 70%|██████▉   | 7454/10682 [1:02:32<20:11,  2.66it/s]
 70%|██████▉   | 7455/10682 [1:02:33<20:11,  2.66it/s]
 70%|██████▉   | 7456/10682 [1:02:33<20:11,  2.66it/s]
 70%|██████▉   | 7457/10682 [1:02:33<20:13,  2.66it/s]
 70%|██████▉   | 7458/10682 [1:02:34<20:11,  2.66it/s]
 70%|██████▉   | 7459/10682 [1:02:34<20:11,  2.66it/s]
 70%|██████▉   | 7460/10682 [1:02:35<20:09,  2.66it/s]
 70%|██████▉   | 7461/10682 [1:02:35<20:11,  2.66it/s]
 70%|██████▉   | 7462/10682 [1:02:35<20:09,  2.66it/s]
 70%|██████▉   | 7463/10682 [1:02:36<20:07,  2.67it/s]
 70%|██████▉   | 7464/10682 [1:02:36<20:07,  2.67it/s]
 70%|██████▉   | 7465/10682 [1:02:36<20:07,  2.66it/s]
 70%|██████▉   | 7466/10682 [1:02:37<20:08,  2.66it/s]
 70%|██████▉   | 7467/10682 [1:02:37<20:07,  2.66it/s]
 70%|██████▉   | 7468/10682 [1:02:38<20:08,  2.66it/s]
 70%|██████▉   | 7469/10682 [1:02:38<20:07,  2.66it/s]
 70%|██████▉   | 7470/10682 [1:02:38<20:11,  2.65it/s]
 70%|██████▉   | 7471/10682 [1:02:39<20:10,  2.65it/s]
 70%|██████▉   | 7472/10682 [1:02:39<24:32,  2.18it/s]
 70%|██████▉   | 7473/10682 [1:02:40<23:11,  2.31it/s]
 70%|██████▉   | 7474/10682 [1:02:40<22:15,  2.40it/s]
 70%|██████▉   | 7475/10682 [1:02:40<21:36,  2.47it/s]
                                                      
{'loss': 4.2365, 'grad_norm': 0.46030333638191223, 'learning_rate': 0.0002503774587418758, 'epoch': 0.7}
-
 70%|██████▉   | 7475/10682 [1:02:40<21:36,  2.47it/s]
 70%|██████▉   | 7476/10682 [1:02:41<21:09,  2.53it/s]
 70%|██████▉   | 7477/10682 [1:02:41<20:49,  2.56it/s]
 70%|███████   | 7478/10682 [1:02:42<20:33,  2.60it/s]
 70%|███████   | 7479/10682 [1:02:42<20:23,  2.62it/s]
 70%|███████   | 7480/10682 [1:02:42<20:16,  2.63it/s]
 70%|███████   | 7481/10682 [1:02:43<20:11,  2.64it/s]
 70%|███████   | 7482/10682 [1:02:43<20:10,  2.64it/s]
 70%|███████   | 7483/10682 [1:02:43<20:08,  2.65it/s]
 70%|███████   | 7484/10682 [1:02:44<20:07,  2.65it/s]
 70%|███████   | 7485/10682 [1:02:44<20:05,  2.65it/s]
 70%|███████   | 7486/10682 [1:02:45<20:03,  2.66it/s]
 70%|███████   | 7487/10682 [1:02:45<20:02,  2.66it/s]
 70%|███████   | 7488/10682 [1:02:45<19:59,  2.66it/s]
 70%|███████   | 7489/10682 [1:02:46<19:58,  2.66it/s]
 70%|███████   | 7490/10682 [1:02:46<19:57,  2.67it/s]
 70%|███████   | 7491/10682 [1:02:46<19:56,  2.67it/s]
 70%|███████   | 7492/10682 [1:02:47<19:56,  2.67it/s]
 70%|███████   | 7493/10682 [1:02:47<19:56,  2.67it/s]
 70%|███████   | 7494/10682 [1:02:48<19:56,  2.66it/s]
 70%|███████   | 7495/10682 [1:02:48<19:54,  2.67it/s]
 70%|███████   | 7496/10682 [1:02:48<19:56,  2.66it/s]
 70%|███████   | 7497/10682 [1:02:49<19:57,  2.66it/s]
 70%|███████   | 7498/10682 [1:02:49<19:56,  2.66it/s]
 70%|███████   | 7499/10682 [1:02:49<19:56,  2.66it/s]
 70%|███████   | 7500/10682 [1:02:50<19:56,  2.66it/s]{'loss': 4.2447, 'grad_norm': 0.45175284147262573, 'learning_rate': 0.00024684626459881156, 'epoch': 0.7}
                                                      
-
 70%|███████   | 7500/10682 [1:02:50<19:56,  2.66it/s]
 70%|███████   | 7501/10682 [1:02:50<20:00,  2.65it/s]
 70%|███████   | 7502/10682 [1:02:51<19:56,  2.66it/s]
 70%|███████   | 7503/10682 [1:02:51<24:15,  2.18it/s]
 70%|███████   | 7504/10682 [1:02:52<22:57,  2.31it/s]
 70%|███████   | 7505/10682 [1:02:52<22:00,  2.41it/s]
 70%|███████   | 7506/10682 [1:02:52<21:22,  2.48it/s]
 70%|███████   | 7507/10682 [1:02:53<20:54,  2.53it/s]
 70%|███████   | 7508/10682 [1:02:53<20:34,  2.57it/s]
 70%|███████   | 7509/10682 [1:02:54<20:20,  2.60it/s]
 70%|███████   | 7510/10682 [1:02:54<20:10,  2.62it/s]
 70%|███████   | 7511/10682 [1:02:54<20:04,  2.63it/s]
 70%|███████   | 7512/10682 [1:02:55<19:58,  2.64it/s]
 70%|███████   | 7513/10682 [1:02:55<19:54,  2.65it/s]
 70%|███████   | 7514/10682 [1:02:55<19:53,  2.65it/s]
 70%|███████   | 7515/10682 [1:02:56<19:51,  2.66it/s]
 70%|███████   | 7516/10682 [1:02:56<19:50,  2.66it/s]
 70%|███████   | 7517/10682 [1:02:57<19:51,  2.66it/s]
 70%|███████   | 7518/10682 [1:02:57<19:49,  2.66it/s]
 70%|███████   | 7519/10682 [1:02:57<19:48,  2.66it/s]
 70%|███████   | 7520/10682 [1:02:58<19:47,  2.66it/s]
 70%|███████   | 7521/10682 [1:02:58<19:46,  2.67it/s]
 70%|███████   | 7522/10682 [1:02:58<19:47,  2.66it/s]
 70%|███████   | 7523/10682 [1:02:59<19:46,  2.66it/s]
 70%|███████   | 7524/10682 [1:02:59<19:44,  2.67it/s]
 70%|███████   | 7525/10682 [1:03:00<19:43,  2.67it/s]{'loss': 4.242, 'grad_norm': 0.42227354645729065, 'learning_rate': 0.00024333196878634163, 'epoch': 0.7}
                                                      
-
 70%|███████   | 7525/10682 [1:03:00<19:43,  2.67it/s]
 70%|███████   | 7526/10682 [1:03:00<19:50,  2.65it/s]
 70%|███████   | 7527/10682 [1:03:00<19:47,  2.66it/s]
 70%|███████   | 7528/10682 [1:03:01<19:45,  2.66it/s]
 70%|███████   | 7529/10682 [1:03:01<19:43,  2.66it/s]
 70%|███████   | 7530/10682 [1:03:01<19:40,  2.67it/s]
 71%|███████   | 7531/10682 [1:03:02<19:41,  2.67it/s]
 71%|███████   | 7532/10682 [1:03:02<19:39,  2.67it/s]
 71%|███████   | 7533/10682 [1:03:03<19:39,  2.67it/s]
 71%|███████   | 7534/10682 [1:03:03<19:39,  2.67it/s]
 71%|███████   | 7535/10682 [1:03:03<19:41,  2.66it/s]
 71%|███████   | 7536/10682 [1:03:04<19:40,  2.66it/s]
 71%|███████   | 7537/10682 [1:03:04<19:40,  2.66it/s]
 71%|███████   | 7538/10682 [1:03:04<19:41,  2.66it/s]
 71%|███████   | 7539/10682 [1:03:05<19:40,  2.66it/s]
 71%|███████   | 7540/10682 [1:03:05<19:38,  2.67it/s]
 71%|███████   | 7541/10682 [1:03:06<19:38,  2.66it/s]
 71%|███████   | 7542/10682 [1:03:06<19:37,  2.67it/s]
 71%|███████   | 7543/10682 [1:03:06<19:36,  2.67it/s]
 71%|███████   | 7544/10682 [1:03:07<19:35,  2.67it/s]
 71%|███████   | 7545/10682 [1:03:07<19:33,  2.67it/s]
 71%|███████   | 7546/10682 [1:03:07<19:35,  2.67it/s]
 71%|███████   | 7547/10682 [1:03:08<19:35,  2.67it/s]
 71%|███████   | 7548/10682 [1:03:08<19:33,  2.67it/s]
 71%|███████   | 7549/10682 [1:03:09<19:35,  2.67it/s]
 71%|███████   | 7550/10682 [1:03:09<19:33,  2.67it/s]
                                                      
{'loss': 4.2439, 'grad_norm': 0.43100959062576294, 'learning_rate': 0.00023983480588813622, 'epoch': 0.71}
-
 71%|███████   | 7550/10682 [1:03:09<19:33,  2.67it/s]
 71%|███████   | 7551/10682 [1:03:09<19:36,  2.66it/s]
 71%|███████   | 7552/10682 [1:03:10<19:34,  2.67it/s]
 71%|███████   | 7553/10682 [1:03:10<19:33,  2.67it/s]
 71%|███████   | 7554/10682 [1:03:10<19:33,  2.66it/s]
 71%|███████   | 7555/10682 [1:03:11<19:34,  2.66it/s]
 71%|███████   | 7556/10682 [1:03:11<19:32,  2.67it/s]
 71%|███████   | 7557/10682 [1:03:12<19:31,  2.67it/s]
 71%|███████   | 7558/10682 [1:03:12<19:31,  2.67it/s]
 71%|███████   | 7559/10682 [1:03:12<19:32,  2.66it/s]
 71%|███████   | 7560/10682 [1:03:13<19:33,  2.66it/s]
 71%|███████   | 7561/10682 [1:03:13<19:32,  2.66it/s]
 71%|███████   | 7562/10682 [1:03:13<19:31,  2.66it/s]
 71%|███████   | 7563/10682 [1:03:14<19:30,  2.66it/s]
 71%|███████   | 7564/10682 [1:03:14<19:30,  2.66it/s]
 71%|███████   | 7565/10682 [1:03:15<19:29,  2.66it/s]
 71%|███████   | 7566/10682 [1:03:15<19:28,  2.67it/s]
 71%|███████   | 7567/10682 [1:03:15<19:27,  2.67it/s]
 71%|███████   | 7568/10682 [1:03:16<19:27,  2.67it/s]
 71%|███████   | 7569/10682 [1:03:16<19:26,  2.67it/s]
 71%|███████   | 7570/10682 [1:03:16<19:26,  2.67it/s]
 71%|███████   | 7571/10682 [1:03:17<19:25,  2.67it/s]
 71%|███████   | 7572/10682 [1:03:17<19:24,  2.67it/s]
 71%|███████   | 7573/10682 [1:03:18<19:23,  2.67it/s]
 71%|███████   | 7574/10682 [1:03:18<19:22,  2.67it/s]
 71%|███████   | 7575/10682 [1:03:18<19:23,  2.67it/s]{'loss': 4.2372, 'grad_norm': 0.4148232042789459, 'learning_rate': 0.00023635500934422239, 'epoch': 0.71}
-
                                                      

 71%|███████   | 7575/10682 [1:03:18<19:23,  2.67it/s]
 71%|███████   | 7576/10682 [1:03:19<19:24,  2.67it/s]
 71%|███████   | 7577/10682 [1:03:19<19:22,  2.67it/s]
 71%|███████   | 7578/10682 [1:03:19<19:24,  2.67it/s]
 71%|███████   | 7579/10682 [1:03:20<19:23,  2.67it/s]
 71%|███████   | 7580/10682 [1:03:20<19:22,  2.67it/s]
 71%|███████   | 7581/10682 [1:03:21<19:22,  2.67it/s]
 71%|███████   | 7582/10682 [1:03:21<19:22,  2.67it/s]
 71%|███████   | 7583/10682 [1:03:21<19:21,  2.67it/s]
 71%|███████   | 7584/10682 [1:03:22<19:21,  2.67it/s]
 71%|███████   | 7585/10682 [1:03:22<19:20,  2.67it/s]
 71%|███████   | 7586/10682 [1:03:22<19:41,  2.62it/s]
 71%|███████   | 7587/10682 [1:03:23<19:38,  2.63it/s]
 71%|███████   | 7588/10682 [1:03:23<19:34,  2.63it/s]
 71%|███████   | 7589/10682 [1:03:24<19:33,  2.63it/s]
 71%|███████   | 7590/10682 [1:03:24<19:33,  2.64it/s]
 71%|███████   | 7591/10682 [1:03:24<19:33,  2.63it/s]
 71%|███████   | 7592/10682 [1:03:25<19:31,  2.64it/s]
 71%|███████   | 7593/10682 [1:03:25<19:29,  2.64it/s]
 71%|███████   | 7594/10682 [1:03:25<19:26,  2.65it/s]
 71%|███████   | 7595/10682 [1:03:26<19:25,  2.65it/s]
 71%|███████   | 7596/10682 [1:03:26<19:22,  2.65it/s]
 71%|███████   | 7597/10682 [1:03:27<19:20,  2.66it/s]
 71%|███████   | 7598/10682 [1:03:27<19:19,  2.66it/s]
 71%|███████   | 7599/10682 [1:03:27<19:17,  2.66it/s]
 71%|███████   | 7600/10682 [1:03:28<19:16,  2.67it/s]
                                                      
{'loss': 4.2379, 'grad_norm': 0.449252188205719, 'learning_rate': 0.00023289281143540065, 'epoch': 0.71}
-
 71%|███████   | 7600/10682 [1:03:28<19:16,  2.67it/s]
 71%|███████   | 7601/10682 [1:03:28<19:21,  2.65it/s]
 71%|███████   | 7602/10682 [1:03:28<19:21,  2.65it/s]
 71%|███████   | 7603/10682 [1:03:29<19:22,  2.65it/s]
 71%|███████   | 7604/10682 [1:03:29<19:20,  2.65it/s]
 71%|███████   | 7605/10682 [1:03:30<19:20,  2.65it/s]
 71%|███████   | 7606/10682 [1:03:30<19:18,  2.66it/s]
 71%|███████   | 7607/10682 [1:03:30<19:19,  2.65it/s]
 71%|███████   | 7608/10682 [1:03:31<19:18,  2.65it/s]
 71%|███████   | 7609/10682 [1:03:31<19:16,  2.66it/s]
 71%|███████   | 7610/10682 [1:03:31<19:14,  2.66it/s]
 71%|███████▏  | 7611/10682 [1:03:32<19:12,  2.67it/s]
 71%|███████▏  | 7612/10682 [1:03:32<19:11,  2.67it/s]
 71%|███████▏  | 7613/10682 [1:03:33<19:11,  2.67it/s]
 71%|███████▏  | 7614/10682 [1:03:33<19:10,  2.67it/s]
 71%|███████▏  | 7615/10682 [1:03:33<19:10,  2.67it/s]
 71%|███████▏  | 7616/10682 [1:03:34<19:10,  2.67it/s]
 71%|███████▏  | 7617/10682 [1:03:34<19:08,  2.67it/s]
 71%|███████▏  | 7618/10682 [1:03:34<19:08,  2.67it/s]
 71%|███████▏  | 7619/10682 [1:03:35<19:11,  2.66it/s]
 71%|███████▏  | 7620/10682 [1:03:35<19:09,  2.66it/s]
 71%|███████▏  | 7621/10682 [1:03:36<19:09,  2.66it/s]
 71%|███████▏  | 7622/10682 [1:03:36<19:07,  2.67it/s]
 71%|███████▏  | 7623/10682 [1:03:36<19:06,  2.67it/s]
 71%|███████▏  | 7624/10682 [1:03:37<19:05,  2.67it/s]
 71%|███████▏  | 7625/10682 [1:03:37<19:05,  2.67it/s]{'loss': 4.2438, 'grad_norm': 0.39902031421661377, 'learning_rate': 0.00022944844326774121, 'epoch': 0.71}
-
                                                      

 71%|███████▏  | 7625/10682 [1:03:37<19:05,  2.67it/s]
 71%|███████▏  | 7626/10682 [1:03:37<19:06,  2.67it/s]
 71%|███████▏  | 7627/10682 [1:03:38<19:04,  2.67it/s]
 71%|███████▏  | 7628/10682 [1:03:38<19:04,  2.67it/s]
 71%|███████▏  | 7629/10682 [1:03:39<19:04,  2.67it/s]
 71%|███████▏  | 7630/10682 [1:03:39<19:05,  2.66it/s]
 71%|███████▏  | 7631/10682 [1:03:39<19:04,  2.67it/s]
 71%|███████▏  | 7632/10682 [1:03:40<19:05,  2.66it/s]
 71%|███████▏  | 7633/10682 [1:03:40<19:04,  2.66it/s]
 71%|███████▏  | 7634/10682 [1:03:40<19:03,  2.67it/s]
 71%|███████▏  | 7635/10682 [1:03:41<19:02,  2.67it/s]
 71%|███████▏  | 7636/10682 [1:03:41<19:01,  2.67it/s]
 71%|███████▏  | 7637/10682 [1:03:42<19:00,  2.67it/s]
 72%|███████▏  | 7638/10682 [1:03:42<19:00,  2.67it/s]
 72%|███████▏  | 7639/10682 [1:03:42<18:59,  2.67it/s]
 72%|███████▏  | 7640/10682 [1:03:43<18:59,  2.67it/s]
 72%|███████▏  | 7641/10682 [1:03:43<18:58,  2.67it/s]
 72%|███████▏  | 7642/10682 [1:03:43<18:57,  2.67it/s]
 72%|███████▏  | 7643/10682 [1:03:44<18:59,  2.67it/s]
 72%|███████▏  | 7644/10682 [1:03:44<18:59,  2.67it/s]
 72%|███████▏  | 7645/10682 [1:03:45<18:59,  2.66it/s]
 72%|███████▏  | 7646/10682 [1:03:45<18:58,  2.67it/s]
 72%|███████▏  | 7647/10682 [1:03:45<18:57,  2.67it/s]
 72%|███████▏  | 7648/10682 [1:03:46<18:55,  2.67it/s]
 72%|███████▏  | 7649/10682 [1:03:46<18:56,  2.67it/s]
 72%|███████▏  | 7650/10682 [1:03:46<18:56,  2.67it/s]{'loss': 4.2303, 'grad_norm': 0.3928167223930359, 'learning_rate': 0.00022602213475715589, 'epoch': 0.72}
                                                      
-
 72%|███████▏  | 7650/10682 [1:03:46<18:56,  2.67it/s]
 72%|███████▏  | 7651/10682 [1:03:47<18:56,  2.67it/s]
 72%|███████▏  | 7652/10682 [1:03:47<18:54,  2.67it/s]
 72%|███████▏  | 7653/10682 [1:03:48<18:53,  2.67it/s]
 72%|███████▏  | 7654/10682 [1:03:48<18:52,  2.67it/s]
 72%|███████▏  | 7655/10682 [1:03:48<18:51,  2.67it/s]
 72%|███████▏  | 7656/10682 [1:03:49<18:54,  2.67it/s]
 72%|███████▏  | 7657/10682 [1:03:49<18:55,  2.66it/s]
 72%|███████▏  | 7658/10682 [1:03:49<18:55,  2.66it/s]
 72%|███████▏  | 7659/10682 [1:03:50<18:55,  2.66it/s]
 72%|███████▏  | 7660/10682 [1:03:50<18:55,  2.66it/s]
 72%|███████▏  | 7661/10682 [1:03:51<18:54,  2.66it/s]
 72%|███████▏  | 7662/10682 [1:03:51<18:53,  2.66it/s]
 72%|███████▏  | 7663/10682 [1:03:51<18:52,  2.66it/s]
 72%|███████▏  | 7664/10682 [1:03:52<18:52,  2.67it/s]
 72%|███████▏  | 7665/10682 [1:03:52<18:51,  2.67it/s]
 72%|███████▏  | 7666/10682 [1:03:52<18:50,  2.67it/s]
 72%|███████▏  | 7667/10682 [1:03:53<18:50,  2.67it/s]
 72%|███████▏  | 7668/10682 [1:03:53<18:50,  2.67it/s]
 72%|███████▏  | 7669/10682 [1:03:54<18:48,  2.67it/s]
 72%|███████▏  | 7670/10682 [1:03:54<18:50,  2.66it/s]
 72%|███████▏  | 7671/10682 [1:03:54<18:49,  2.67it/s]
 72%|███████▏  | 7672/10682 [1:03:55<18:49,  2.66it/s]
 72%|███████▏  | 7673/10682 [1:03:55<18:49,  2.66it/s]
 72%|███████▏  | 7674/10682 [1:03:55<18:48,  2.66it/s]
 72%|███████▏  | 7675/10682 [1:03:56<18:48,  2.66it/s]{'loss': 4.2409, 'grad_norm': 0.4055207073688507, 'learning_rate': 0.0002226141146140523, 'epoch': 0.72}
                                                      
-
 72%|███████▏  | 7675/10682 [1:03:56<18:48,  2.66it/s]
 72%|███████▏  | 7676/10682 [1:03:56<18:49,  2.66it/s]
 72%|███████▏  | 7677/10682 [1:03:57<18:48,  2.66it/s]
 72%|███████▏  | 7678/10682 [1:03:57<18:48,  2.66it/s]
 72%|███████▏  | 7679/10682 [1:03:57<18:46,  2.67it/s]
 72%|███████▏  | 7680/10682 [1:03:58<18:45,  2.67it/s]
 72%|███████▏  | 7681/10682 [1:03:58<18:44,  2.67it/s]
 72%|███████▏  | 7682/10682 [1:03:58<18:44,  2.67it/s]
 72%|███████▏  | 7683/10682 [1:03:59<18:45,  2.66it/s]
 72%|███████▏  | 7684/10682 [1:03:59<18:46,  2.66it/s]
 72%|███████▏  | 7685/10682 [1:04:00<18:46,  2.66it/s]
 72%|███████▏  | 7686/10682 [1:04:00<18:44,  2.66it/s]
 72%|███████▏  | 7687/10682 [1:04:00<18:45,  2.66it/s]
 72%|███████▏  | 7688/10682 [1:04:01<18:44,  2.66it/s]
 72%|███████▏  | 7689/10682 [1:04:01<18:42,  2.67it/s]
 72%|███████▏  | 7690/10682 [1:04:01<18:42,  2.67it/s]
 72%|███████▏  | 7691/10682 [1:04:02<18:40,  2.67it/s]
 72%|███████▏  | 7692/10682 [1:04:02<18:40,  2.67it/s]
 72%|███████▏  | 7693/10682 [1:04:03<18:39,  2.67it/s]
 72%|███████▏  | 7694/10682 [1:04:03<18:40,  2.67it/s]
 72%|███████▏  | 7695/10682 [1:04:03<18:38,  2.67it/s]
 72%|███████▏  | 7696/10682 [1:04:04<18:40,  2.66it/s]
 72%|███████▏  | 7697/10682 [1:04:04<18:40,  2.66it/s]
 72%|███████▏  | 7698/10682 [1:04:04<18:39,  2.66it/s]
 72%|███████▏  | 7699/10682 [1:04:05<18:39,  2.67it/s]
 72%|███████▏  | 7700/10682 [1:04:05<18:37,  2.67it/s]
                                                      
{'loss': 4.2322, 'grad_norm': 0.3998636305332184, 'learning_rate': 0.00021922461032806601, 'epoch': 0.72}
-
 72%|███████▏  | 7700/10682 [1:04:05<18:37,  2.67it/s]
 72%|███████▏  | 7701/10682 [1:04:06<18:38,  2.67it/s]
 72%|███████▏  | 7702/10682 [1:04:06<18:36,  2.67it/s]
 72%|███████▏  | 7703/10682 [1:04:06<18:36,  2.67it/s]
 72%|███████▏  | 7704/10682 [1:04:07<18:35,  2.67it/s]
 72%|███████▏  | 7705/10682 [1:04:07<18:35,  2.67it/s]
 72%|███████▏  | 7706/10682 [1:04:07<18:35,  2.67it/s]
 72%|███████▏  | 7707/10682 [1:04:08<18:37,  2.66it/s]
 72%|███████▏  | 7708/10682 [1:04:08<18:35,  2.67it/s]
 72%|███████▏  | 7709/10682 [1:04:09<18:36,  2.66it/s]
 72%|███████▏  | 7710/10682 [1:04:09<18:35,  2.66it/s]
 72%|███████▏  | 7711/10682 [1:04:09<18:33,  2.67it/s]
 72%|███████▏  | 7712/10682 [1:04:10<18:35,  2.66it/s]
 72%|███████▏  | 7713/10682 [1:04:10<18:36,  2.66it/s]
 72%|███████▏  | 7714/10682 [1:04:10<18:34,  2.66it/s]
 72%|███████▏  | 7715/10682 [1:04:11<18:32,  2.67it/s]
 72%|███████▏  | 7716/10682 [1:04:11<18:33,  2.66it/s]
 72%|███████▏  | 7717/10682 [1:04:12<18:32,  2.66it/s]
 72%|███████▏  | 7718/10682 [1:04:12<18:31,  2.67it/s]
 72%|███████▏  | 7719/10682 [1:04:12<18:30,  2.67it/s]
 72%|███████▏  | 7720/10682 [1:04:13<18:29,  2.67it/s]
 72%|███████▏  | 7721/10682 [1:04:13<18:30,  2.67it/s]
 72%|███████▏  | 7722/10682 [1:04:13<18:29,  2.67it/s]
 72%|███████▏  | 7723/10682 [1:04:14<18:29,  2.67it/s]
 72%|███████▏  | 7724/10682 [1:04:14<18:28,  2.67it/s]
 72%|███████▏  | 7725/10682 [1:04:15<18:28,  2.67it/s]
                                                      
{'loss': 4.2344, 'grad_norm': 0.396217942237854, 'learning_rate': 0.0002158538481528759, 'epoch': 0.72}
-
 72%|███████▏  | 7725/10682 [1:04:15<18:28,  2.67it/s]
 72%|███████▏  | 7726/10682 [1:04:15<18:30,  2.66it/s]
 72%|███████▏  | 7727/10682 [1:04:15<18:30,  2.66it/s]
 72%|███████▏  | 7728/10682 [1:04:16<18:28,  2.67it/s]
 72%|███████▏  | 7729/10682 [1:04:16<18:27,  2.67it/s]
 72%|███████▏  | 7730/10682 [1:04:16<18:27,  2.67it/s]
 72%|███████▏  | 7731/10682 [1:04:17<18:25,  2.67it/s]
 72%|███████▏  | 7732/10682 [1:04:17<18:26,  2.67it/s]
 72%|███████▏  | 7733/10682 [1:04:18<18:25,  2.67it/s]
 72%|███████▏  | 7734/10682 [1:04:18<18:24,  2.67it/s]
 72%|███████▏  | 7735/10682 [1:04:18<18:24,  2.67it/s]
 72%|███████▏  | 7736/10682 [1:04:19<18:27,  2.66it/s]
 72%|███████▏  | 7737/10682 [1:04:19<18:28,  2.66it/s]
 72%|███████▏  | 7738/10682 [1:04:19<18:26,  2.66it/s]
 72%|███████▏  | 7739/10682 [1:04:20<18:26,  2.66it/s]
 72%|███████▏  | 7740/10682 [1:04:20<18:27,  2.66it/s]
 72%|███████▏  | 7741/10682 [1:04:21<18:25,  2.66it/s]
 72%|███████▏  | 7742/10682 [1:04:21<18:24,  2.66it/s]
 72%|███████▏  | 7743/10682 [1:04:21<18:23,  2.66it/s]
 72%|███████▏  | 7744/10682 [1:04:22<18:22,  2.67it/s]
 73%|███████▎  | 7745/10682 [1:04:22<18:21,  2.67it/s]
 73%|███████▎  | 7746/10682 [1:04:22<18:21,  2.67it/s]
 73%|███████▎  | 7747/10682 [1:04:23<18:20,  2.67it/s]
 73%|███████▎  | 7748/10682 [1:04:23<18:22,  2.66it/s]
 73%|███████▎  | 7749/10682 [1:04:24<18:21,  2.66it/s]
 73%|███████▎  | 7750/10682 [1:04:24<18:23,  2.66it/s]
                                                      {'loss': 4.2267, 'grad_norm': 0.42890244722366333, 'learning_rate': 0.00021250205309110155, 'epoch': 0.73}
-

 73%|███████▎  | 7750/10682 [1:04:24<18:23,  2.66it/s]
 73%|███████▎  | 7751/10682 [1:04:24<18:23,  2.66it/s]
 73%|███████▎  | 7752/10682 [1:04:25<18:23,  2.65it/s]
 73%|███████▎  | 7753/10682 [1:04:25<18:22,  2.66it/s]
 73%|███████▎  | 7754/10682 [1:04:25<18:22,  2.66it/s]
 73%|███████▎  | 7755/10682 [1:04:26<18:20,  2.66it/s]
 73%|███████▎  | 7756/10682 [1:04:26<18:19,  2.66it/s]
 73%|███████▎  | 7757/10682 [1:04:27<18:19,  2.66it/s]
 73%|███████▎  | 7758/10682 [1:04:27<18:17,  2.67it/s]
 73%|███████▎  | 7759/10682 [1:04:27<18:17,  2.66it/s]
 73%|███████▎  | 7760/10682 [1:04:28<18:21,  2.65it/s]
 73%|███████▎  | 7761/10682 [1:04:28<18:34,  2.62it/s]
 73%|███████▎  | 7762/10682 [1:04:29<18:46,  2.59it/s]
 73%|███████▎  | 7763/10682 [1:04:29<18:55,  2.57it/s]
 73%|███████▎  | 7764/10682 [1:04:29<18:52,  2.58it/s]
 73%|███████▎  | 7765/10682 [1:04:30<18:42,  2.60it/s]
 73%|███████▎  | 7766/10682 [1:04:30<18:33,  2.62it/s]
 73%|███████▎  | 7767/10682 [1:04:30<18:28,  2.63it/s]
 73%|███████▎  | 7768/10682 [1:04:31<18:24,  2.64it/s]
 73%|███████▎  | 7769/10682 [1:04:31<18:23,  2.64it/s]
 73%|███████▎  | 7770/10682 [1:04:32<18:21,  2.64it/s]
 73%|███████▎  | 7771/10682 [1:04:32<18:19,  2.65it/s]
 73%|███████▎  | 7772/10682 [1:04:32<18:18,  2.65it/s]
 73%|███████▎  | 7773/10682 [1:04:33<18:16,  2.65it/s]
 73%|███████▎  | 7774/10682 [1:04:33<18:15,  2.65it/s]
 73%|███████▎  | 7775/10682 [1:04:33<18:14,  2.66it/s]{'loss': 4.2332, 'grad_norm': 0.39339062571525574, 'learning_rate': 0.00020916944887928359, 'epoch': 0.73}
                                                      
-
 73%|███████▎  | 7775/10682 [1:04:33<18:14,  2.66it/s]
 73%|███████▎  | 7776/10682 [1:04:34<18:16,  2.65it/s]
 73%|███████▎  | 7777/10682 [1:04:34<18:14,  2.65it/s]
 73%|███████▎  | 7778/10682 [1:04:35<18:13,  2.65it/s]
 73%|███████▎  | 7779/10682 [1:04:35<18:13,  2.66it/s]
 73%|███████▎  | 7780/10682 [1:04:35<18:12,  2.66it/s]
 73%|███████▎  | 7781/10682 [1:04:36<18:11,  2.66it/s]
 73%|███████▎  | 7782/10682 [1:04:36<18:10,  2.66it/s]
 73%|███████▎  | 7783/10682 [1:04:36<18:09,  2.66it/s]
 73%|███████▎  | 7784/10682 [1:04:37<18:10,  2.66it/s]
 73%|███████▎  | 7785/10682 [1:04:37<18:09,  2.66it/s]
 73%|███████▎  | 7786/10682 [1:04:38<18:08,  2.66it/s]
 73%|███████▎  | 7787/10682 [1:04:38<18:09,  2.66it/s]
 73%|███████▎  | 7788/10682 [1:04:38<18:07,  2.66it/s]
 73%|███████▎  | 7789/10682 [1:04:39<18:07,  2.66it/s]
 73%|███████▎  | 7790/10682 [1:04:39<18:06,  2.66it/s]
 73%|███████▎  | 7791/10682 [1:04:39<18:06,  2.66it/s]
 73%|███████▎  | 7792/10682 [1:04:40<18:07,  2.66it/s]
 73%|███████▎  | 7793/10682 [1:04:40<18:05,  2.66it/s]
 73%|███████▎  | 7794/10682 [1:04:41<18:05,  2.66it/s]
 73%|███████▎  | 7795/10682 [1:04:41<18:06,  2.66it/s]
 73%|███████▎  | 7796/10682 [1:04:41<18:06,  2.66it/s]
 73%|███████▎  | 7797/10682 [1:04:42<18:05,  2.66it/s]
 73%|███████▎  | 7798/10682 [1:04:42<18:04,  2.66it/s]
 73%|███████▎  | 7799/10682 [1:04:42<18:04,  2.66it/s]
 73%|███████▎  | 7800/10682 [1:04:43<18:04,  2.66it/s]{'loss': 4.2314, 'grad_norm': 0.40094488859176636, 'learning_rate': 0.00020585625797294927, 'epoch': 0.73}
                                                      
-
 73%|███████▎  | 7800/10682 [1:04:43<18:04,  2.66it/s]
 73%|███████▎  | 7801/10682 [1:04:43<18:05,  2.65it/s]
 73%|███████▎  | 7802/10682 [1:04:44<18:03,  2.66it/s]
 73%|███████▎  | 7803/10682 [1:04:44<18:06,  2.65it/s]
 73%|███████▎  | 7804/10682 [1:04:44<18:04,  2.65it/s]
 73%|███████▎  | 7805/10682 [1:04:45<18:05,  2.65it/s]
 73%|███████▎  | 7806/10682 [1:04:45<18:03,  2.65it/s]
 73%|███████▎  | 7807/10682 [1:04:45<18:02,  2.66it/s]
 73%|███████▎  | 7808/10682 [1:04:46<18:01,  2.66it/s]
 73%|███████▎  | 7809/10682 [1:04:46<17:59,  2.66it/s]
 73%|███████▎  | 7810/10682 [1:04:47<17:59,  2.66it/s]
 73%|███████▎  | 7811/10682 [1:04:47<17:58,  2.66it/s]
 73%|███████▎  | 7812/10682 [1:04:47<17:59,  2.66it/s]
 73%|███████▎  | 7813/10682 [1:04:48<17:58,  2.66it/s]
 73%|███████▎  | 7814/10682 [1:04:48<17:58,  2.66it/s]
 73%|███████▎  | 7815/10682 [1:04:48<17:57,  2.66it/s]
 73%|███████▎  | 7816/10682 [1:04:49<17:56,  2.66it/s]
 73%|███████▎  | 7817/10682 [1:04:49<17:57,  2.66it/s]
 73%|███████▎  | 7818/10682 [1:04:50<17:56,  2.66it/s]
 73%|███████▎  | 7819/10682 [1:04:50<17:56,  2.66it/s]
 73%|███████▎  | 7820/10682 [1:04:50<17:56,  2.66it/s]
 73%|███████▎  | 7821/10682 [1:04:51<17:57,  2.66it/s]
 73%|███████▎  | 7822/10682 [1:04:51<17:55,  2.66it/s]
 73%|███████▎  | 7823/10682 [1:04:52<17:55,  2.66it/s]
 73%|███████▎  | 7824/10682 [1:04:52<17:55,  2.66it/s]
 73%|███████▎  | 7825/10682 [1:04:52<17:55,  2.66it/s]
                                                      
{'loss': 4.2214, 'grad_norm': 0.4255618751049042, 'learning_rate': 0.00020256270153176371, 'epoch': 0.73}
-
 73%|███████▎  | 7825/10682 [1:04:52<17:55,  2.66it/s]
 73%|███████▎  | 7826/10682 [1:04:53<17:55,  2.66it/s]
 73%|███████▎  | 7827/10682 [1:04:53<17:55,  2.65it/s]
 73%|███████▎  | 7828/10682 [1:04:53<17:53,  2.66it/s]
 73%|███████▎  | 7829/10682 [1:04:54<17:53,  2.66it/s]
 73%|███████▎  | 7830/10682 [1:04:54<17:55,  2.65it/s]
 73%|███████▎  | 7831/10682 [1:04:55<17:58,  2.64it/s]
 73%|███████▎  | 7832/10682 [1:04:55<17:58,  2.64it/s]
 73%|███████▎  | 7833/10682 [1:04:55<17:54,  2.65it/s]
 73%|███████▎  | 7834/10682 [1:04:56<17:54,  2.65it/s]
 73%|███████▎  | 7835/10682 [1:04:56<17:51,  2.66it/s]
 73%|███████▎  | 7836/10682 [1:04:56<17:49,  2.66it/s]
 73%|███████▎  | 7837/10682 [1:04:57<17:48,  2.66it/s]
 73%|███████▎  | 7838/10682 [1:04:57<17:46,  2.67it/s]
 73%|███████▎  | 7839/10682 [1:04:58<17:44,  2.67it/s]
 73%|███████▎  | 7840/10682 [1:04:58<17:46,  2.67it/s]
 73%|███████▎  | 7841/10682 [1:04:58<17:44,  2.67it/s]
 73%|███████▎  | 7842/10682 [1:04:59<17:42,  2.67it/s]
 73%|███████▎  | 7843/10682 [1:04:59<17:44,  2.67it/s]
 73%|███████▎  | 7844/10682 [1:04:59<17:44,  2.67it/s]
 73%|███████▎  | 7845/10682 [1:05:00<17:44,  2.66it/s]
 73%|███████▎  | 7846/10682 [1:05:00<17:44,  2.66it/s]
 73%|███████▎  | 7847/10682 [1:05:01<17:44,  2.66it/s]
 73%|███████▎  | 7848/10682 [1:05:01<17:42,  2.67it/s]
 73%|███████▎  | 7849/10682 [1:05:01<17:41,  2.67it/s]
 73%|███████▎  | 7850/10682 [1:05:02<17:40,  2.67it/s]{'loss': 4.2278, 'grad_norm': 0.39953678846359253, 'learning_rate': 0.00019928899940476624, 'epoch': 0.73}
                                                      
-
 73%|███████▎  | 7850/10682 [1:05:02<17:40,  2.67it/s]
 73%|███████▎  | 7851/10682 [1:05:02<17:41,  2.67it/s]
 74%|███████▎  | 7852/10682 [1:05:02<17:41,  2.67it/s]
 74%|███████▎  | 7853/10682 [1:05:03<17:39,  2.67it/s]
 74%|███████▎  | 7854/10682 [1:05:03<17:39,  2.67it/s]
 74%|███████▎  | 7855/10682 [1:05:04<17:38,  2.67it/s]
 74%|███████▎  | 7856/10682 [1:05:04<17:40,  2.67it/s]
 74%|███████▎  | 7857/10682 [1:05:04<17:40,  2.66it/s]
 74%|███████▎  | 7858/10682 [1:05:05<17:39,  2.67it/s]
 74%|███████▎  | 7859/10682 [1:05:05<17:40,  2.66it/s]
 74%|███████▎  | 7860/10682 [1:05:05<17:40,  2.66it/s]
 74%|███████▎  | 7861/10682 [1:05:06<17:38,  2.66it/s]
 74%|███████▎  | 7862/10682 [1:05:06<17:38,  2.67it/s]
 74%|███████▎  | 7863/10682 [1:05:07<17:38,  2.66it/s]
 74%|███████▎  | 7864/10682 [1:05:07<17:37,  2.66it/s]
 74%|███████▎  | 7865/10682 [1:05:07<17:37,  2.66it/s]
 74%|███████▎  | 7866/10682 [1:05:08<17:36,  2.67it/s]
 74%|███████▎  | 7867/10682 [1:05:08<17:36,  2.67it/s]
 74%|███████▎  | 7868/10682 [1:05:08<17:35,  2.67it/s]
 74%|███████▎  | 7869/10682 [1:05:09<17:34,  2.67it/s]
 74%|███████▎  | 7870/10682 [1:05:09<17:34,  2.67it/s]
 74%|███████▎  | 7871/10682 [1:05:10<17:33,  2.67it/s]
 74%|███████▎  | 7872/10682 [1:05:10<17:34,  2.66it/s]
 74%|███████▎  | 7873/10682 [1:05:10<17:34,  2.66it/s]
 74%|███████▎  | 7874/10682 [1:05:11<17:33,  2.66it/s]
 74%|███████▎  | 7875/10682 [1:05:11<17:32,  2.67it/s]{'loss': 4.2252, 'grad_norm': 0.3977838456630707, 'learning_rate': 0.00019603537011569566, 'epoch': 0.74}
                                                      
-
 74%|███████▎  | 7875/10682 [1:05:11<17:32,  2.67it/s]
 74%|███████▎  | 7876/10682 [1:05:11<17:33,  2.66it/s]
 74%|███████▎  | 7877/10682 [1:05:12<17:33,  2.66it/s]
 74%|███████▍  | 7878/10682 [1:05:12<17:31,  2.67it/s]
 74%|███████▍  | 7879/10682 [1:05:13<17:30,  2.67it/s]
 74%|███████▍  | 7880/10682 [1:05:13<17:30,  2.67it/s]
 74%|███████▍  | 7881/10682 [1:05:13<17:30,  2.67it/s]
 74%|███████▍  | 7882/10682 [1:05:14<17:29,  2.67it/s]
 74%|███████▍  | 7883/10682 [1:05:14<17:31,  2.66it/s]
 74%|███████▍  | 7884/10682 [1:05:14<17:30,  2.66it/s]
 74%|███████▍  | 7885/10682 [1:05:15<17:31,  2.66it/s]
 74%|███████▍  | 7886/10682 [1:05:15<17:29,  2.66it/s]
 74%|███████▍  | 7887/10682 [1:05:16<17:30,  2.66it/s]
 74%|███████▍  | 7888/10682 [1:05:16<17:29,  2.66it/s]
 74%|███████▍  | 7889/10682 [1:05:16<17:29,  2.66it/s]
 74%|███████▍  | 7890/10682 [1:05:17<17:29,  2.66it/s]
 74%|███████▍  | 7891/10682 [1:05:17<17:28,  2.66it/s]
 74%|███████▍  | 7892/10682 [1:05:17<17:27,  2.66it/s]
 74%|███████▍  | 7893/10682 [1:05:18<17:26,  2.67it/s]
 74%|███████▍  | 7894/10682 [1:05:18<17:25,  2.67it/s]
 74%|███████▍  | 7895/10682 [1:05:19<17:26,  2.66it/s]
 74%|███████▍  | 7896/10682 [1:05:19<17:26,  2.66it/s]
 74%|███████▍  | 7897/10682 [1:05:19<17:24,  2.67it/s]
 74%|███████▍  | 7898/10682 [1:05:20<17:22,  2.67it/s]
 74%|███████▍  | 7899/10682 [1:05:20<17:23,  2.67it/s]
 74%|███████▍  | 7900/10682 [1:05:20<17:22,  2.67it/s]
                                                      
{'loss': 4.2301, 'grad_norm': 0.404459685087204, 'learning_rate': 0.0001928020308484042, 'epoch': 0.74}
-
 74%|███████▍  | 7900/10682 [1:05:20<17:22,  2.67it/s]
 74%|███████▍  | 7901/10682 [1:05:21<17:25,  2.66it/s]
 74%|███████▍  | 7902/10682 [1:05:21<17:23,  2.66it/s]
 74%|███████▍  | 7903/10682 [1:05:22<17:22,  2.67it/s]
 74%|███████▍  | 7904/10682 [1:05:22<17:20,  2.67it/s]
 74%|███████▍  | 7905/10682 [1:05:22<17:21,  2.67it/s]
 74%|███████▍  | 7906/10682 [1:05:23<17:19,  2.67it/s]
 74%|███████▍  | 7907/10682 [1:05:23<17:20,  2.67it/s]
 74%|███████▍  | 7908/10682 [1:05:23<17:19,  2.67it/s]
 74%|███████▍  | 7909/10682 [1:05:24<17:18,  2.67it/s]
 74%|███████▍  | 7910/10682 [1:05:24<17:21,  2.66it/s]
 74%|███████▍  | 7911/10682 [1:05:25<17:23,  2.66it/s]
 74%|███████▍  | 7912/10682 [1:05:25<17:23,  2.66it/s]
 74%|███████▍  | 7913/10682 [1:05:25<17:22,  2.66it/s]
 74%|███████▍  | 7914/10682 [1:05:26<17:21,  2.66it/s]
 74%|███████▍  | 7915/10682 [1:05:26<17:19,  2.66it/s]
 74%|███████▍  | 7916/10682 [1:05:26<17:19,  2.66it/s]
 74%|███████▍  | 7917/10682 [1:05:27<17:18,  2.66it/s]
 74%|███████▍  | 7918/10682 [1:05:27<17:18,  2.66it/s]
 74%|███████▍  | 7919/10682 [1:05:28<17:17,  2.66it/s]
 74%|███████▍  | 7920/10682 [1:05:28<17:16,  2.67it/s]
 74%|███████▍  | 7921/10682 [1:05:28<17:17,  2.66it/s]
 74%|███████▍  | 7922/10682 [1:05:29<17:15,  2.67it/s]
 74%|███████▍  | 7923/10682 [1:05:29<17:18,  2.66it/s]
 74%|███████▍  | 7924/10682 [1:05:29<17:16,  2.66it/s]
 74%|███████▍  | 7925/10682 [1:05:30<17:18,  2.65it/s]
                                                      {'loss': 4.2254, 'grad_norm': 0.41146212816238403, 'learning_rate': 0.00018958919743235897, 'epoch': 0.74}
-
 74%|███████▍  | 7925/10682 [1:05:30<17:18,  2.65it/s]
 74%|███████▍  | 7926/10682 [1:05:30<17:18,  2.65it/s]
 74%|███████▍  | 7927/10682 [1:05:31<17:16,  2.66it/s]
 74%|███████▍  | 7928/10682 [1:05:31<17:15,  2.66it/s]
 74%|███████▍  | 7929/10682 [1:05:31<17:15,  2.66it/s]
 74%|███████▍  | 7930/10682 [1:05:32<17:13,  2.66it/s]
 74%|███████▍  | 7931/10682 [1:05:32<17:12,  2.66it/s]
 74%|███████▍  | 7932/10682 [1:05:32<17:11,  2.67it/s]
 74%|███████▍  | 7933/10682 [1:05:33<17:10,  2.67it/s]
 74%|███████▍  | 7934/10682 [1:05:33<17:10,  2.67it/s]
 74%|███████▍  | 7935/10682 [1:05:34<17:09,  2.67it/s]
 74%|███████▍  | 7936/10682 [1:05:34<17:10,  2.66it/s]
 74%|███████▍  | 7937/10682 [1:05:34<17:09,  2.67it/s]
 74%|███████▍  | 7938/10682 [1:05:35<17:08,  2.67it/s]
 74%|███████▍  | 7939/10682 [1:05:35<17:09,  2.67it/s]
 74%|███████▍  | 7940/10682 [1:05:35<17:08,  2.67it/s]
 74%|███████▍  | 7941/10682 [1:05:36<17:08,  2.67it/s]
 74%|███████▍  | 7942/10682 [1:05:36<17:07,  2.67it/s]
 74%|███████▍  | 7943/10682 [1:05:37<17:07,  2.67it/s]
 74%|███████▍  | 7944/10682 [1:05:37<17:06,  2.67it/s]
 74%|███████▍  | 7945/10682 [1:05:37<17:06,  2.67it/s]
 74%|███████▍  | 7946/10682 [1:05:38<17:05,  2.67it/s]
 74%|███████▍  | 7947/10682 [1:05:38<17:05,  2.67it/s]
 74%|███████▍  | 7948/10682 [1:05:38<17:06,  2.66it/s]
 74%|███████▍  | 7949/10682 [1:05:39<17:04,  2.67it/s]
 74%|███████▍  | 7950/10682 [1:05:39<17:06,  2.66it/s]
                                                      
{'loss': 4.227, 'grad_norm': 0.414027601480484, 'learning_rate': 0.0001863970843282357, 'epoch': 0.74}
-
 74%|███████▍  | 7950/10682 [1:05:39<17:06,  2.66it/s]
 74%|███████▍  | 7951/10682 [1:05:40<17:05,  2.66it/s]
 74%|███████▍  | 7952/10682 [1:05:40<17:06,  2.66it/s]
 74%|███████▍  | 7953/10682 [1:05:40<17:04,  2.66it/s]
 74%|███████▍  | 7954/10682 [1:05:41<17:03,  2.67it/s]
 74%|███████▍  | 7955/10682 [1:05:41<17:01,  2.67it/s]
 74%|███████▍  | 7956/10682 [1:05:41<17:01,  2.67it/s]
 74%|███████▍  | 7957/10682 [1:05:42<17:01,  2.67it/s]
 74%|███████▍  | 7958/10682 [1:05:42<16:59,  2.67it/s]
 75%|███████▍  | 7959/10682 [1:05:43<16:59,  2.67it/s]
 75%|███████▍  | 7960/10682 [1:05:43<16:58,  2.67it/s]
 75%|███████▍  | 7961/10682 [1:05:43<16:58,  2.67it/s]
 75%|███████▍  | 7962/10682 [1:05:44<16:57,  2.67it/s]
 75%|███████▍  | 7963/10682 [1:05:44<16:59,  2.67it/s]
 75%|███████▍  | 7964/10682 [1:05:44<16:58,  2.67it/s]
 75%|███████▍  | 7965/10682 [1:05:45<17:01,  2.66it/s]
 75%|███████▍  | 7966/10682 [1:05:45<17:00,  2.66it/s]
 75%|███████▍  | 7967/10682 [1:05:46<16:59,  2.66it/s]
 75%|███████▍  | 7968/10682 [1:05:46<16:59,  2.66it/s]
 75%|███████▍  | 7969/10682 [1:05:46<16:58,  2.66it/s]
 75%|███████▍  | 7970/10682 [1:05:47<16:58,  2.66it/s]
 75%|███████▍  | 7971/10682 [1:05:47<16:57,  2.67it/s]
 75%|███████▍  | 7972/10682 [1:05:47<16:56,  2.67it/s]
 75%|███████▍  | 7973/10682 [1:05:48<16:54,  2.67it/s]
 75%|███████▍  | 7974/10682 [1:05:48<16:54,  2.67it/s]
 75%|███████▍  | 7975/10682 [1:05:49<16:53,  2.67it/s]
                                                      
{'loss': 4.234, 'grad_norm': 0.4030643105506897, 'learning_rate': 0.00018322590461360383, 'epoch': 0.75}
-
 75%|███████▍  | 7975/10682 [1:05:49<16:53,  2.67it/s]
 75%|███████▍  | 7976/10682 [1:05:49<16:54,  2.67it/s]
 75%|███████▍  | 7977/10682 [1:05:49<16:55,  2.66it/s]
 75%|███████▍  | 7978/10682 [1:05:50<16:54,  2.67it/s]
 75%|███████▍  | 7979/10682 [1:05:50<16:56,  2.66it/s]
 75%|███████▍  | 7980/10682 [1:05:50<16:54,  2.66it/s]
 75%|███████▍  | 7981/10682 [1:05:51<16:54,  2.66it/s]
 75%|███████▍  | 7982/10682 [1:05:51<16:53,  2.66it/s]
 75%|███████▍  | 7983/10682 [1:05:52<16:52,  2.67it/s]
 75%|███████▍  | 7984/10682 [1:05:52<16:51,  2.67it/s]
 75%|███████▍  | 7985/10682 [1:05:52<16:50,  2.67it/s]
 75%|███████▍  | 7986/10682 [1:05:53<16:49,  2.67it/s]
 75%|███████▍  | 7987/10682 [1:05:53<16:48,  2.67it/s]
 75%|███████▍  | 7988/10682 [1:05:53<16:49,  2.67it/s]
 75%|███████▍  | 7989/10682 [1:05:54<16:48,  2.67it/s]
 75%|███████▍  | 7990/10682 [1:05:54<16:50,  2.67it/s]
 75%|███████▍  | 7991/10682 [1:05:55<16:49,  2.67it/s]
 75%|███████▍  | 7992/10682 [1:05:55<16:49,  2.66it/s]
 75%|███████▍  | 7993/10682 [1:05:55<16:49,  2.66it/s]
 75%|███████▍  | 7994/10682 [1:05:56<16:48,  2.66it/s]
 75%|███████▍  | 7995/10682 [1:05:56<16:48,  2.66it/s]
 75%|███████▍  | 7996/10682 [1:05:56<16:47,  2.67it/s]
 75%|███████▍  | 7997/10682 [1:05:57<16:47,  2.67it/s]
 75%|███████▍  | 7998/10682 [1:05:57<16:45,  2.67it/s]
 75%|███████▍  | 7999/10682 [1:05:58<16:45,  2.67it/s]
 75%|███████▍  | 8000/10682 [1:05:58<16:44,  2.67it/s]
                                                      
{'loss': 4.22, 'grad_norm': 0.3845983147621155, 'learning_rate': 0.00018007586996870206, 'epoch': 0.75}
-
 75%|███████▍  | 8000/10682 [1:05:58<16:44,  2.67it/s]
 75%|███████▍  | 8001/10682 [1:05:58<16:45,  2.67it/s]
 75%|███████▍  | 8002/10682 [1:05:59<16:44,  2.67it/s]
 75%|███████▍  | 8003/10682 [1:05:59<16:44,  2.67it/s]
 75%|███████▍  | 8004/10682 [1:05:59<16:44,  2.67it/s]
 75%|███████▍  | 8005/10682 [1:06:00<16:44,  2.66it/s]
 75%|███████▍  | 8006/10682 [1:06:00<16:44,  2.66it/s]
 75%|███████▍  | 8007/10682 [1:06:01<16:42,  2.67it/s]
 75%|███████▍  | 8008/10682 [1:06:01<16:42,  2.67it/s]
 75%|███████▍  | 8009/10682 [1:06:01<16:41,  2.67it/s]
 75%|███████▍  | 8010/10682 [1:06:02<16:42,  2.67it/s]
 75%|███████▍  | 8011/10682 [1:06:02<16:41,  2.67it/s]
 75%|███████▌  | 8012/10682 [1:06:02<16:40,  2.67it/s]
 75%|███████▌  | 8013/10682 [1:06:03<16:39,  2.67it/s]
 75%|███████▌  | 8014/10682 [1:06:03<16:39,  2.67it/s]
 75%|███████▌  | 8015/10682 [1:06:04<16:39,  2.67it/s]
 75%|███████▌  | 8016/10682 [1:06:04<16:39,  2.67it/s]
 75%|███████▌  | 8017/10682 [1:06:04<16:40,  2.66it/s]
 75%|███████▌  | 8018/10682 [1:06:05<16:38,  2.67it/s]
 75%|███████▌  | 8019/10682 [1:06:05<16:40,  2.66it/s]
 75%|███████▌  | 8020/10682 [1:06:05<16:39,  2.66it/s]
 75%|███████▌  | 8021/10682 [1:06:06<16:38,  2.67it/s]
 75%|███████▌  | 8022/10682 [1:06:06<16:38,  2.66it/s]
 75%|███████▌  | 8023/10682 [1:06:07<16:37,  2.66it/s]
 75%|███████▌  | 8024/10682 [1:06:07<16:37,  2.67it/s]
 75%|███████▌  | 8025/10682 [1:06:07<16:36,  2.67it/s]
                                                      
{'loss': 4.2233, 'grad_norm': 0.4067979156970978, 'learning_rate': 0.00017694719066230924, 'epoch': 0.75}
-
 75%|███████▌  | 8025/10682 [1:06:07<16:36,  2.67it/s]
 75%|███████▌  | 8026/10682 [1:06:08<16:37,  2.66it/s]
 75%|███████▌  | 8027/10682 [1:06:08<16:35,  2.67it/s]
 75%|███████▌  | 8028/10682 [1:06:08<16:35,  2.67it/s]
 75%|███████▌  | 8029/10682 [1:06:09<16:34,  2.67it/s]
 75%|███████▌  | 8030/10682 [1:06:09<16:35,  2.66it/s]
 75%|███████▌  | 8031/10682 [1:06:10<16:33,  2.67it/s]
 75%|███████▌  | 8032/10682 [1:06:10<16:35,  2.66it/s]
 75%|███████▌  | 8033/10682 [1:06:10<16:33,  2.67it/s]
 75%|███████▌  | 8034/10682 [1:06:11<16:33,  2.67it/s]
 75%|███████▌  | 8035/10682 [1:06:11<16:31,  2.67it/s]
 75%|███████▌  | 8036/10682 [1:06:11<16:31,  2.67it/s]
 75%|███████▌  | 8037/10682 [1:06:12<16:30,  2.67it/s]
 75%|███████▌  | 8038/10682 [1:06:12<16:29,  2.67it/s]
 75%|███████▌  | 8039/10682 [1:06:13<16:29,  2.67it/s]
 75%|███████▌  | 8040/10682 [1:06:13<16:27,  2.67it/s]
 75%|███████▌  | 8041/10682 [1:06:13<16:28,  2.67it/s]
 75%|███████▌  | 8042/10682 [1:06:14<16:27,  2.67it/s]
 75%|███████▌  | 8043/10682 [1:06:14<16:29,  2.67it/s]
 75%|███████▌  | 8044/10682 [1:06:14<16:28,  2.67it/s]
 75%|███████▌  | 8045/10682 [1:06:15<16:28,  2.67it/s]
 75%|███████▌  | 8046/10682 [1:06:15<16:28,  2.67it/s]
 75%|███████▌  | 8047/10682 [1:06:16<16:27,  2.67it/s]
 75%|███████▌  | 8048/10682 [1:06:16<16:27,  2.67it/s]
 75%|███████▌  | 8049/10682 [1:06:16<16:27,  2.67it/s]
 75%|███████▌  | 8050/10682 [1:06:17<16:27,  2.66it/s]{'loss': 4.2333, 'grad_norm': 0.4031130373477936, 'learning_rate': 0.00017384007553770858, 'epoch': 0.75}
                                                      
-
 75%|███████▌  | 8050/10682 [1:06:17<16:27,  2.66it/s]
 75%|███████▌  | 8051/10682 [1:06:17<16:29,  2.66it/s]
 75%|███████▌  | 8052/10682 [1:06:17<16:28,  2.66it/s]
 75%|███████▌  | 8053/10682 [1:06:18<16:27,  2.66it/s]
 75%|███████▌  | 8054/10682 [1:06:18<16:26,  2.66it/s]
 75%|███████▌  | 8055/10682 [1:06:19<16:25,  2.67it/s]
 75%|███████▌  | 8056/10682 [1:06:19<16:25,  2.67it/s]
 75%|███████▌  | 8057/10682 [1:06:19<16:25,  2.66it/s]
 75%|███████▌  | 8058/10682 [1:06:20<16:24,  2.67it/s]
 75%|███████▌  | 8059/10682 [1:06:20<16:25,  2.66it/s]
 75%|███████▌  | 8060/10682 [1:06:20<16:23,  2.67it/s]
 75%|███████▌  | 8061/10682 [1:06:21<16:23,  2.67it/s]
 75%|███████▌  | 8062/10682 [1:06:21<16:22,  2.67it/s]
 75%|███████▌  | 8063/10682 [1:06:22<16:22,  2.67it/s]
 75%|███████▌  | 8064/10682 [1:06:22<16:21,  2.67it/s]
 76%|███████▌  | 8065/10682 [1:06:22<16:20,  2.67it/s]
 76%|███████▌  | 8066/10682 [1:06:23<16:19,  2.67it/s]
 76%|███████▌  | 8067/10682 [1:06:23<16:19,  2.67it/s]
 76%|███████▌  | 8068/10682 [1:06:23<16:19,  2.67it/s]
 76%|███████▌  | 8069/10682 [1:06:24<16:18,  2.67it/s]
 76%|███████▌  | 8070/10682 [1:06:24<16:20,  2.66it/s]
 76%|███████▌  | 8071/10682 [1:06:25<16:19,  2.66it/s]
 76%|███████▌  | 8072/10682 [1:06:25<16:18,  2.67it/s]
 76%|███████▌  | 8073/10682 [1:06:25<16:18,  2.67it/s]
 76%|███████▌  | 8074/10682 [1:06:26<16:18,  2.66it/s]
 76%|███████▌  | 8075/10682 [1:06:26<16:17,  2.67it/s]
                                                      
{'loss': 4.2143, 'grad_norm': 0.3802649974822998, 'learning_rate': 0.00017075473199874692, 'epoch': 0.76}
-
 76%|███████▌  | 8075/10682 [1:06:26<16:17,  2.67it/s]
 76%|███████▌  | 8076/10682 [1:06:26<16:19,  2.66it/s]
 76%|███████▌  | 8077/10682 [1:06:27<16:17,  2.66it/s]
 76%|███████▌  | 8078/10682 [1:06:27<16:16,  2.67it/s]
 76%|███████▌  | 8079/10682 [1:06:28<16:15,  2.67it/s]
 76%|███████▌  | 8080/10682 [1:06:28<16:14,  2.67it/s]
 76%|███████▌  | 8081/10682 [1:06:28<16:14,  2.67it/s]
 76%|███████▌  | 8082/10682 [1:06:29<16:14,  2.67it/s]
 76%|███████▌  | 8083/10682 [1:06:29<16:16,  2.66it/s]
 76%|███████▌  | 8084/10682 [1:06:29<16:14,  2.66it/s]
 76%|███████▌  | 8085/10682 [1:06:30<16:14,  2.66it/s]
 76%|███████▌  | 8086/10682 [1:06:30<16:14,  2.66it/s]
 76%|███████▌  | 8087/10682 [1:06:31<16:13,  2.66it/s]
 76%|███████▌  | 8088/10682 [1:06:31<16:14,  2.66it/s]
 76%|███████▌  | 8089/10682 [1:06:31<16:14,  2.66it/s]
 76%|███████▌  | 8090/10682 [1:06:32<16:13,  2.66it/s]
 76%|███████▌  | 8091/10682 [1:06:32<16:13,  2.66it/s]
 76%|███████▌  | 8092/10682 [1:06:32<16:12,  2.66it/s]
 76%|███████▌  | 8093/10682 [1:06:33<16:10,  2.67it/s]
 76%|███████▌  | 8094/10682 [1:06:33<16:10,  2.67it/s]
 76%|███████▌  | 8095/10682 [1:06:34<16:08,  2.67it/s]
 76%|███████▌  | 8096/10682 [1:06:34<16:09,  2.67it/s]
 76%|███████▌  | 8097/10682 [1:06:34<16:09,  2.67it/s]
 76%|███████▌  | 8098/10682 [1:06:35<16:09,  2.67it/s]
 76%|███████▌  | 8099/10682 [1:06:35<16:10,  2.66it/s]
 76%|███████▌  | 8100/10682 [1:06:35<16:10,  2.66it/s]
                                                      
{'loss': 4.2224, 'grad_norm': 0.3966319262981415, 'learning_rate': 0.00016769136599599017, 'epoch': 0.76}
-
 76%|███████▌  | 8100/10682 [1:06:35<16:10,  2.66it/s]
 76%|███████▌  | 8101/10682 [1:06:36<16:11,  2.66it/s]
 76%|███████▌  | 8102/10682 [1:06:36<16:08,  2.66it/s]
 76%|███████▌  | 8103/10682 [1:06:37<16:08,  2.66it/s]
 76%|███████▌  | 8104/10682 [1:06:37<16:06,  2.67it/s]
 76%|███████▌  | 8105/10682 [1:06:37<16:05,  2.67it/s]
 76%|███████▌  | 8106/10682 [1:06:38<16:04,  2.67it/s]
 76%|███████▌  | 8107/10682 [1:06:38<16:04,  2.67it/s]
 76%|███████▌  | 8108/10682 [1:06:38<16:03,  2.67it/s]
 76%|███████▌  | 8109/10682 [1:06:39<16:03,  2.67it/s]
 76%|███████▌  | 8110/10682 [1:06:39<16:03,  2.67it/s]
 76%|███████▌  | 8111/10682 [1:06:40<16:02,  2.67it/s]
 76%|███████▌  | 8112/10682 [1:06:40<16:04,  2.66it/s]
 76%|███████▌  | 8113/10682 [1:06:40<16:03,  2.67it/s]
 76%|███████▌  | 8114/10682 [1:06:41<16:02,  2.67it/s]
 76%|███████▌  | 8115/10682 [1:06:41<16:02,  2.67it/s]
 76%|███████▌  | 8116/10682 [1:06:41<16:01,  2.67it/s]
 76%|███████▌  | 8117/10682 [1:06:42<16:01,  2.67it/s]
 76%|███████▌  | 8118/10682 [1:06:42<16:00,  2.67it/s]
 76%|███████▌  | 8119/10682 [1:06:43<16:00,  2.67it/s]
 76%|███████▌  | 8120/10682 [1:06:43<15:59,  2.67it/s]
 76%|███████▌  | 8121/10682 [1:06:43<16:01,  2.66it/s]
 76%|███████▌  | 8122/10682 [1:06:44<15:59,  2.67it/s]
 76%|███████▌  | 8123/10682 [1:06:44<16:01,  2.66it/s]
 76%|███████▌  | 8124/10682 [1:06:44<16:00,  2.66it/s]
 76%|███████▌  | 8125/10682 [1:06:45<16:02,  2.66it/s]{'loss': 4.2191, 'grad_norm': 0.41085362434387207, 'learning_rate': 0.0001646501820129766, 'epoch': 0.76}
                                                      
-
 76%|███████▌  | 8125/10682 [1:06:45<16:02,  2.66it/s]
 76%|███████▌  | 8126/10682 [1:06:45<16:03,  2.65it/s]
 76%|███████▌  | 8127/10682 [1:06:46<16:01,  2.66it/s]
 76%|███████▌  | 8128/10682 [1:06:46<16:00,  2.66it/s]
 76%|███████▌  | 8129/10682 [1:06:46<15:59,  2.66it/s]
 76%|███████▌  | 8130/10682 [1:06:47<15:57,  2.66it/s]
 76%|███████▌  | 8131/10682 [1:06:47<15:56,  2.67it/s]
 76%|███████▌  | 8132/10682 [1:06:47<15:55,  2.67it/s]
 76%|███████▌  | 8133/10682 [1:06:48<15:54,  2.67it/s]
 76%|███████▌  | 8134/10682 [1:06:48<15:54,  2.67it/s]
 76%|███████▌  | 8135/10682 [1:06:49<15:54,  2.67it/s]
 76%|███████▌  | 8136/10682 [1:06:49<15:56,  2.66it/s]
 76%|███████▌  | 8137/10682 [1:06:49<15:56,  2.66it/s]
 76%|███████▌  | 8138/10682 [1:06:50<15:54,  2.66it/s]
 76%|███████▌  | 8139/10682 [1:06:50<15:54,  2.66it/s]
 76%|███████▌  | 8140/10682 [1:06:50<15:53,  2.67it/s]
 76%|███████▌  | 8141/10682 [1:06:51<15:52,  2.67it/s]
 76%|███████▌  | 8142/10682 [1:06:51<15:51,  2.67it/s]
 76%|███████▌  | 8143/10682 [1:06:52<15:52,  2.66it/s]
 76%|███████▌  | 8144/10682 [1:06:52<15:52,  2.66it/s]
 76%|███████▌  | 8145/10682 [1:06:52<15:51,  2.67it/s]
 76%|███████▋  | 8146/10682 [1:06:53<15:51,  2.67it/s]
 76%|███████▋  | 8147/10682 [1:06:53<15:50,  2.67it/s]
 76%|███████▋  | 8148/10682 [1:06:53<15:50,  2.67it/s]
 76%|███████▋  | 8149/10682 [1:06:54<15:49,  2.67it/s]
 76%|███████▋  | 8150/10682 [1:06:54<15:49,  2.67it/s]
                                                      
{'loss': 4.2257, 'grad_norm': 0.3901253640651703, 'learning_rate': 0.00016163138305256598, 'epoch': 0.76}
-
 76%|███████▋  | 8150/10682 [1:06:54<15:49,  2.67it/s]
 76%|███████▋  | 8151/10682 [1:06:55<15:49,  2.67it/s]
 76%|███████▋  | 8152/10682 [1:06:55<15:50,  2.66it/s]
 76%|███████▋  | 8153/10682 [1:06:55<15:48,  2.67it/s]
 76%|███████▋  | 8154/10682 [1:06:56<15:48,  2.67it/s]
 76%|███████▋  | 8155/10682 [1:06:56<15:47,  2.67it/s]
 76%|███████▋  | 8156/10682 [1:06:56<15:45,  2.67it/s]
 76%|███████▋  | 8157/10682 [1:06:57<15:45,  2.67it/s]
 76%|███████▋  | 8158/10682 [1:06:57<15:45,  2.67it/s]
 76%|███████▋  | 8159/10682 [1:06:58<15:45,  2.67it/s]
 76%|███████▋  | 8160/10682 [1:06:58<15:46,  2.67it/s]
 76%|███████▋  | 8161/10682 [1:06:58<15:45,  2.67it/s]
 76%|███████▋  | 8162/10682 [1:06:59<15:44,  2.67it/s]
 76%|███████▋  | 8163/10682 [1:06:59<15:44,  2.67it/s]
 76%|███████▋  | 8164/10682 [1:06:59<15:44,  2.67it/s]
 76%|███████▋  | 8165/10682 [1:07:00<15:43,  2.67it/s]
 76%|███████▋  | 8166/10682 [1:07:00<15:45,  2.66it/s]
 76%|███████▋  | 8167/10682 [1:07:01<15:44,  2.66it/s]
 76%|███████▋  | 8168/10682 [1:07:01<15:43,  2.66it/s]
 76%|███████▋  | 8169/10682 [1:07:01<15:42,  2.67it/s]
 76%|███████▋  | 8170/10682 [1:07:02<15:42,  2.66it/s]
 76%|███████▋  | 8171/10682 [1:07:02<15:41,  2.67it/s]
 77%|███████▋  | 8172/10682 [1:07:02<15:41,  2.67it/s]
 77%|███████▋  | 8173/10682 [1:07:03<15:39,  2.67it/s]
 77%|███████▋  | 8174/10682 [1:07:03<15:39,  2.67it/s]
 77%|███████▋  | 8175/10682 [1:07:04<15:39,  2.67it/s]
                                                      {'loss': 4.2231, 'grad_norm': 0.40253767371177673, 'learning_rate': 0.00015863517062339038, 'epoch': 0.77}
-

 77%|███████▋  | 8175/10682 [1:07:04<15:39,  2.67it/s]
 77%|███████▋  | 8176/10682 [1:07:04<15:39,  2.67it/s]
 77%|███████▋  | 8177/10682 [1:07:04<15:39,  2.67it/s]
 77%|███████▋  | 8178/10682 [1:07:05<15:38,  2.67it/s]
 77%|███████▋  | 8179/10682 [1:07:05<15:39,  2.66it/s]
 77%|███████▋  | 8180/10682 [1:07:05<15:39,  2.66it/s]
 77%|███████▋  | 8181/10682 [1:07:06<15:38,  2.67it/s]
 77%|███████▋  | 8182/10682 [1:07:06<15:37,  2.67it/s]
 77%|███████▋  | 8183/10682 [1:07:07<15:37,  2.66it/s]
 77%|███████▋  | 8184/10682 [1:07:07<15:37,  2.67it/s]
 77%|███████▋  | 8185/10682 [1:07:07<15:35,  2.67it/s]
 77%|███████▋  | 8186/10682 [1:07:08<15:34,  2.67it/s]
 77%|███████▋  | 8187/10682 [1:07:08<18:58,  2.19it/s]
 77%|███████▋  | 8188/10682 [1:07:09<17:57,  2.31it/s]
 77%|███████▋  | 8189/10682 [1:07:09<17:14,  2.41it/s]
 77%|███████▋  | 8190/10682 [1:07:09<16:44,  2.48it/s]
 77%|███████▋  | 8191/10682 [1:07:10<16:22,  2.53it/s]
 77%|███████▋  | 8192/10682 [1:07:10<16:09,  2.57it/s]
 77%|███████▋  | 8193/10682 [1:07:11<15:58,  2.60it/s]
 77%|███████▋  | 8194/10682 [1:07:11<15:49,  2.62it/s]
 77%|███████▋  | 8195/10682 [1:07:11<15:43,  2.64it/s]
 77%|███████▋  | 8196/10682 [1:07:12<15:39,  2.65it/s]
 77%|███████▋  | 8197/10682 [1:07:12<15:36,  2.65it/s]
 77%|███████▋  | 8198/10682 [1:07:12<15:34,  2.66it/s]
 77%|███████▋  | 8199/10682 [1:07:13<15:32,  2.66it/s]
 77%|███████▋  | 8200/10682 [1:07:13<15:31,  2.66it/s]
                                                      
{'loss': 4.22, 'grad_norm': 0.388113409280777, 'learning_rate': 0.00015566174472640188, 'epoch': 0.77}
-
 77%|███████▋  | 8200/10682 [1:07:13<15:31,  2.66it/s]
 77%|███████▋  | 8201/10682 [1:07:14<15:31,  2.66it/s]
 77%|███████▋  | 8202/10682 [1:07:14<15:31,  2.66it/s]
 77%|███████▋  | 8203/10682 [1:07:14<15:30,  2.66it/s]
 77%|███████▋  | 8204/10682 [1:07:15<15:30,  2.66it/s]
 77%|███████▋  | 8205/10682 [1:07:15<15:31,  2.66it/s]
 77%|███████▋  | 8206/10682 [1:07:15<15:30,  2.66it/s]
 77%|███████▋  | 8207/10682 [1:07:16<15:29,  2.66it/s]
 77%|███████▋  | 8208/10682 [1:07:16<15:27,  2.67it/s]
 77%|███████▋  | 8209/10682 [1:07:17<15:27,  2.67it/s]
 77%|███████▋  | 8210/10682 [1:07:17<15:25,  2.67it/s]
 77%|███████▋  | 8211/10682 [1:07:17<15:26,  2.67it/s]
 77%|███████▋  | 8212/10682 [1:07:18<15:24,  2.67it/s]
 77%|███████▋  | 8213/10682 [1:07:18<15:24,  2.67it/s]
 77%|███████▋  | 8214/10682 [1:07:18<15:24,  2.67it/s]
 77%|███████▋  | 8215/10682 [1:07:19<15:23,  2.67it/s]
 77%|███████▋  | 8216/10682 [1:07:19<15:24,  2.67it/s]
 77%|███████▋  | 8217/10682 [1:07:20<15:22,  2.67it/s]
 77%|███████▋  | 8218/10682 [1:07:20<15:24,  2.67it/s]
 77%|███████▋  | 8219/10682 [1:07:20<15:23,  2.67it/s]
 77%|███████▋  | 8220/10682 [1:07:21<15:22,  2.67it/s]
 77%|███████▋  | 8221/10682 [1:07:21<15:22,  2.67it/s]
 77%|███████▋  | 8222/10682 [1:07:21<15:21,  2.67it/s]
 77%|███████▋  | 8223/10682 [1:07:22<18:44,  2.19it/s]
 77%|███████▋  | 8224/10682 [1:07:22<17:44,  2.31it/s]
 77%|███████▋  | 8225/10682 [1:07:23<17:00,  2.41it/s]
                                                      
{'loss': 4.2192, 'grad_norm': 0.3859880566596985, 'learning_rate': 0.0001527113038415231, 'epoch': 0.77}
-
 77%|███████▋  | 8225/10682 [1:07:23<17:00,  2.41it/s]
 77%|███████▋  | 8226/10682 [1:07:23<16:31,  2.48it/s]
 77%|███████▋  | 8227/10682 [1:07:24<16:08,  2.53it/s]
 77%|███████▋  | 8228/10682 [1:07:24<15:54,  2.57it/s]
 77%|███████▋  | 8229/10682 [1:07:24<15:43,  2.60it/s]
 77%|███████▋  | 8230/10682 [1:07:25<15:35,  2.62it/s]
 77%|███████▋  | 8231/10682 [1:07:25<15:28,  2.64it/s]
 77%|███████▋  | 8232/10682 [1:07:25<15:24,  2.65it/s]
 77%|███████▋  | 8233/10682 [1:07:26<15:23,  2.65it/s]
 77%|███████▋  | 8234/10682 [1:07:26<15:21,  2.66it/s]
 77%|███████▋  | 8235/10682 [1:07:27<15:20,  2.66it/s]
 77%|███████▋  | 8236/10682 [1:07:27<15:19,  2.66it/s]
 77%|███████▋  | 8237/10682 [1:07:27<15:18,  2.66it/s]
 77%|███████▋  | 8238/10682 [1:07:28<15:16,  2.67it/s]
 77%|███████▋  | 8239/10682 [1:07:28<15:15,  2.67it/s]
 77%|███████▋  | 8240/10682 [1:07:28<15:14,  2.67it/s]
 77%|███████▋  | 8241/10682 [1:07:29<15:13,  2.67it/s]
 77%|███████▋  | 8242/10682 [1:07:29<15:13,  2.67it/s]
 77%|███████▋  | 8243/10682 [1:07:30<15:13,  2.67it/s]
 77%|███████▋  | 8244/10682 [1:07:30<15:15,  2.66it/s]
 77%|███████▋  | 8245/10682 [1:07:30<15:13,  2.67it/s]
 77%|███████▋  | 8246/10682 [1:07:31<15:14,  2.66it/s]
 77%|███████▋  | 8247/10682 [1:07:31<15:12,  2.67it/s]
 77%|███████▋  | 8248/10682 [1:07:31<15:11,  2.67it/s]
 77%|███████▋  | 8249/10682 [1:07:32<15:11,  2.67it/s]
 77%|███████▋  | 8250/10682 [1:07:32<15:11,  2.67it/s]
                                                      {'loss': 4.2251, 'grad_norm': 0.3866763710975647, 'learning_rate': 0.00014978404491439802, 'epoch': 0.77}
-
 77%|███████▋  | 8250/10682 [1:07:32<15:11,  2.67it/s]
 77%|███████▋  | 8251/10682 [1:07:33<15:12,  2.67it/s]
 77%|███████▋  | 8252/10682 [1:07:33<15:10,  2.67it/s]
 77%|███████▋  | 8253/10682 [1:07:33<15:10,  2.67it/s]
 77%|███████▋  | 8254/10682 [1:07:34<15:08,  2.67it/s]
 77%|███████▋  | 8255/10682 [1:07:34<15:09,  2.67it/s]
 77%|███████▋  | 8256/10682 [1:07:34<15:08,  2.67it/s]
 77%|███████▋  | 8257/10682 [1:07:35<15:09,  2.67it/s]
 77%|███████▋  | 8258/10682 [1:07:35<15:08,  2.67it/s]
 77%|███████▋  | 8259/10682 [1:07:36<15:08,  2.67it/s]
 77%|███████▋  | 8260/10682 [1:07:36<15:08,  2.67it/s]
 77%|███████▋  | 8261/10682 [1:07:36<15:07,  2.67it/s]
 77%|███████▋  | 8262/10682 [1:07:37<15:06,  2.67it/s]
 77%|███████▋  | 8263/10682 [1:07:37<15:05,  2.67it/s]
 77%|███████▋  | 8264/10682 [1:07:37<15:05,  2.67it/s]
 77%|███████▋  | 8265/10682 [1:07:38<15:05,  2.67it/s]
 77%|███████▋  | 8266/10682 [1:07:38<15:04,  2.67it/s]
 77%|███████▋  | 8267/10682 [1:07:39<15:03,  2.67it/s]
 77%|███████▋  | 8268/10682 [1:07:39<15:02,  2.67it/s]
 77%|███████▋  | 8269/10682 [1:07:39<15:03,  2.67it/s]
 77%|███████▋  | 8270/10682 [1:07:40<15:01,  2.67it/s]
 77%|███████▋  | 8271/10682 [1:07:40<15:03,  2.67it/s]
 77%|███████▋  | 8272/10682 [1:07:40<15:01,  2.67it/s]
 77%|███████▋  | 8273/10682 [1:07:41<15:02,  2.67it/s]
 77%|███████▋  | 8274/10682 [1:07:41<15:01,  2.67it/s]
 77%|███████▋  | 8275/10682 [1:07:42<15:01,  2.67it/s]{'loss': 4.2194, 'grad_norm': 0.3850996494293213, 'learning_rate': 0.00014688016334324605, 'epoch': 0.77}
-
                                                      

 77%|███████▋  | 8275/10682 [1:07:42<15:01,  2.67it/s]
 77%|███████▋  | 8276/10682 [1:07:42<15:02,  2.67it/s]
 77%|███████▋  | 8277/10682 [1:07:42<15:00,  2.67it/s]
 77%|███████▋  | 8278/10682 [1:07:43<15:00,  2.67it/s]
 78%|███████▊  | 8279/10682 [1:07:43<14:59,  2.67it/s]
 78%|███████▊  | 8280/10682 [1:07:43<14:59,  2.67it/s]
 78%|███████▊  | 8281/10682 [1:07:44<14:58,  2.67it/s]
 78%|███████▊  | 8282/10682 [1:07:44<14:59,  2.67it/s]
 78%|███████▊  | 8283/10682 [1:07:45<14:57,  2.67it/s]
 78%|███████▊  | 8284/10682 [1:07:45<14:58,  2.67it/s]
 78%|███████▊  | 8285/10682 [1:07:45<14:57,  2.67it/s]
 78%|███████▊  | 8286/10682 [1:07:46<14:58,  2.67it/s]
 78%|███████▊  | 8287/10682 [1:07:46<14:58,  2.67it/s]
 78%|███████▊  | 8288/10682 [1:07:46<14:57,  2.67it/s]
 78%|███████▊  | 8289/10682 [1:07:47<14:56,  2.67it/s]
 78%|███████▊  | 8290/10682 [1:07:47<14:56,  2.67it/s]
 78%|███████▊  | 8291/10682 [1:07:48<14:55,  2.67it/s]
 78%|███████▊  | 8292/10682 [1:07:48<14:55,  2.67it/s]
 78%|███████▊  | 8293/10682 [1:07:48<14:54,  2.67it/s]
 78%|███████▊  | 8294/10682 [1:07:49<14:53,  2.67it/s]
 78%|███████▊  | 8295/10682 [1:07:49<14:53,  2.67it/s]
 78%|███████▊  | 8296/10682 [1:07:49<14:52,  2.67it/s]
 78%|███████▊  | 8297/10682 [1:07:50<14:53,  2.67it/s]
 78%|███████▊  | 8298/10682 [1:07:50<14:53,  2.67it/s]
 78%|███████▊  | 8299/10682 [1:07:51<14:53,  2.67it/s]
 78%|███████▊  | 8300/10682 [1:07:51<14:52,  2.67it/s]{'loss': 4.2143, 'grad_norm': 0.37504732608795166, 'learning_rate': 0.00014399985296581835, 'epoch': 0.78}
                                                      
-
 78%|███████▊  | 8300/10682 [1:07:51<14:52,  2.67it/s]
 78%|███████▊  | 8301/10682 [1:07:51<14:52,  2.67it/s]
 78%|███████▊  | 8302/10682 [1:07:52<14:51,  2.67it/s]
 78%|███████▊  | 8303/10682 [1:07:52<14:51,  2.67it/s]
 78%|███████▊  | 8304/10682 [1:07:52<14:51,  2.67it/s]
 78%|███████▊  | 8305/10682 [1:07:53<14:51,  2.67it/s]
 78%|███████▊  | 8306/10682 [1:07:53<14:50,  2.67it/s]
 78%|███████▊  | 8307/10682 [1:07:54<14:48,  2.67it/s]
 78%|███████▊  | 8308/10682 [1:07:54<14:48,  2.67it/s]
 78%|███████▊  | 8309/10682 [1:07:54<14:48,  2.67it/s]
 78%|███████▊  | 8310/10682 [1:07:55<14:48,  2.67it/s]
 78%|███████▊  | 8311/10682 [1:07:55<14:50,  2.66it/s]
 78%|███████▊  | 8312/10682 [1:07:55<14:48,  2.67it/s]
 78%|███████▊  | 8313/10682 [1:07:56<14:48,  2.67it/s]
 78%|███████▊  | 8314/10682 [1:07:56<14:48,  2.67it/s]
 78%|███████▊  | 8315/10682 [1:07:57<14:48,  2.66it/s]
 78%|███████▊  | 8316/10682 [1:07:57<14:46,  2.67it/s]
 78%|███████▊  | 8317/10682 [1:07:57<14:46,  2.67it/s]
 78%|███████▊  | 8318/10682 [1:07:58<14:45,  2.67it/s]
 78%|███████▊  | 8319/10682 [1:07:58<14:45,  2.67it/s]
 78%|███████▊  | 8320/10682 [1:07:58<14:44,  2.67it/s]
 78%|███████▊  | 8321/10682 [1:07:59<14:43,  2.67it/s]
 78%|███████▊  | 8322/10682 [1:07:59<14:43,  2.67it/s]
 78%|███████▊  | 8323/10682 [1:08:00<14:43,  2.67it/s]
 78%|███████▊  | 8324/10682 [1:08:00<14:44,  2.67it/s]
 78%|███████▊  | 8325/10682 [1:08:00<14:44,  2.66it/s]
                                                      {'loss': 4.2206, 'grad_norm': 0.3773188889026642, 'learning_rate': 0.00014114330604645943, 'epoch': 0.78}
-
 78%|███████▊  | 8325/10682 [1:08:00<14:44,  2.66it/s]
 78%|███████▊  | 8326/10682 [1:08:01<14:45,  2.66it/s]
 78%|███████▊  | 8327/10682 [1:08:01<14:45,  2.66it/s]
 78%|███████▊  | 8328/10682 [1:08:01<14:42,  2.67it/s]
 78%|███████▊  | 8329/10682 [1:08:02<14:41,  2.67it/s]
 78%|███████▊  | 8330/10682 [1:08:02<14:41,  2.67it/s]
 78%|███████▊  | 8331/10682 [1:08:03<14:41,  2.67it/s]
 78%|███████▊  | 8332/10682 [1:08:03<14:41,  2.67it/s]
 78%|███████▊  | 8333/10682 [1:08:03<14:41,  2.67it/s]
 78%|███████▊  | 8334/10682 [1:08:04<14:40,  2.67it/s]
 78%|███████▊  | 8335/10682 [1:08:04<14:39,  2.67it/s]
 78%|███████▊  | 8336/10682 [1:08:04<14:39,  2.67it/s]
 78%|███████▊  | 8337/10682 [1:08:05<14:39,  2.67it/s]
 78%|███████▊  | 8338/10682 [1:08:05<14:38,  2.67it/s]
 78%|███████▊  | 8339/10682 [1:08:06<14:38,  2.67it/s]
 78%|███████▊  | 8340/10682 [1:08:06<14:38,  2.67it/s]
 78%|███████▊  | 8341/10682 [1:08:06<14:37,  2.67it/s]
 78%|███████▊  | 8342/10682 [1:08:07<14:37,  2.67it/s]
 78%|███████▊  | 8343/10682 [1:08:07<14:36,  2.67it/s]
 78%|███████▊  | 8344/10682 [1:08:07<14:35,  2.67it/s]
 78%|███████▊  | 8345/10682 [1:08:08<14:35,  2.67it/s]
 78%|███████▊  | 8346/10682 [1:08:08<14:34,  2.67it/s]
 78%|███████▊  | 8347/10682 [1:08:09<14:33,  2.67it/s]
 78%|███████▊  | 8348/10682 [1:08:09<14:33,  2.67it/s]
 78%|███████▊  | 8349/10682 [1:08:09<14:32,  2.67it/s]
 78%|███████▊  | 8350/10682 [1:08:10<14:33,  2.67it/s]{'loss': 4.2193, 'grad_norm': 0.39353060722351074, 'learning_rate': 0.00013831071326327282, 'epoch': 0.78}
                                                      
-
 78%|███████▊  | 8350/10682 [1:08:10<14:33,  2.67it/s]
 78%|███████▊  | 8351/10682 [1:08:10<14:34,  2.67it/s]
 78%|███████▊  | 8352/10682 [1:08:10<14:32,  2.67it/s]
 78%|███████▊  | 8353/10682 [1:08:11<14:34,  2.66it/s]
 78%|███████▊  | 8354/10682 [1:08:11<14:33,  2.66it/s]
 78%|███████▊  | 8355/10682 [1:08:12<14:34,  2.66it/s]
 78%|███████▊  | 8356/10682 [1:08:12<14:32,  2.66it/s]
 78%|███████▊  | 8357/10682 [1:08:12<14:32,  2.66it/s]
 78%|███████▊  | 8358/10682 [1:08:13<14:32,  2.66it/s]
 78%|███████▊  | 8359/10682 [1:08:13<14:31,  2.67it/s]
 78%|███████▊  | 8360/10682 [1:08:13<14:30,  2.67it/s]
 78%|███████▊  | 8361/10682 [1:08:14<14:29,  2.67it/s]
 78%|███████▊  | 8362/10682 [1:08:14<14:29,  2.67it/s]
 78%|███████▊  | 8363/10682 [1:08:15<14:28,  2.67it/s]
 78%|███████▊  | 8364/10682 [1:08:15<14:29,  2.67it/s]
 78%|███████▊  | 8365/10682 [1:08:15<14:29,  2.67it/s]
 78%|███████▊  | 8366/10682 [1:08:16<14:29,  2.66it/s]
 78%|███████▊  | 8367/10682 [1:08:16<14:29,  2.66it/s]
 78%|███████▊  | 8368/10682 [1:08:16<14:28,  2.66it/s]
 78%|███████▊  | 8369/10682 [1:08:17<14:28,  2.66it/s]
 78%|███████▊  | 8370/10682 [1:08:17<14:27,  2.67it/s]
 78%|███████▊  | 8371/10682 [1:08:18<14:26,  2.67it/s]
 78%|███████▊  | 8372/10682 [1:08:18<14:25,  2.67it/s]
 78%|███████▊  | 8373/10682 [1:08:18<14:24,  2.67it/s]
 78%|███████▊  | 8374/10682 [1:08:19<14:23,  2.67it/s]
 78%|███████▊  | 8375/10682 [1:08:19<14:23,  2.67it/s]{'loss': 4.2172, 'grad_norm': 0.39157170057296753, 'learning_rate': 0.0001355022636953933, 'epoch': 0.78}
                                                      
-
 78%|███████▊  | 8375/10682 [1:08:19<14:23,  2.67it/s]
 78%|███████▊  | 8376/10682 [1:08:19<14:24,  2.67it/s]
 78%|███████▊  | 8377/10682 [1:08:20<14:25,  2.66it/s]
 78%|███████▊  | 8378/10682 [1:08:20<14:25,  2.66it/s]
 78%|███████▊  | 8379/10682 [1:08:21<14:25,  2.66it/s]
 78%|███████▊  | 8380/10682 [1:08:21<14:25,  2.66it/s]
 78%|███████▊  | 8381/10682 [1:08:21<14:24,  2.66it/s]
 78%|███████▊  | 8382/10682 [1:08:22<14:23,  2.66it/s]
 78%|███████▊  | 8383/10682 [1:08:22<14:22,  2.67it/s]
 78%|███████▊  | 8384/10682 [1:08:22<14:21,  2.67it/s]
 78%|███████▊  | 8385/10682 [1:08:23<14:20,  2.67it/s]
 79%|███████▊  | 8386/10682 [1:08:23<14:20,  2.67it/s]
 79%|███████▊  | 8387/10682 [1:08:24<14:20,  2.67it/s]
 79%|███████▊  | 8388/10682 [1:08:24<14:18,  2.67it/s]
 79%|███████▊  | 8389/10682 [1:08:24<14:18,  2.67it/s]
 79%|███████▊  | 8390/10682 [1:08:25<14:17,  2.67it/s]
 79%|███████▊  | 8391/10682 [1:08:25<14:18,  2.67it/s]
 79%|███████▊  | 8392/10682 [1:08:25<14:17,  2.67it/s]
 79%|███████▊  | 8393/10682 [1:08:26<14:19,  2.66it/s]
 79%|███████▊  | 8394/10682 [1:08:26<14:17,  2.67it/s]
 79%|███████▊  | 8395/10682 [1:08:27<14:18,  2.66it/s]
 79%|███████▊  | 8396/10682 [1:08:27<14:17,  2.67it/s]
 79%|███████▊  | 8397/10682 [1:08:27<14:17,  2.66it/s]
 79%|███████▊  | 8398/10682 [1:08:28<14:16,  2.67it/s]
 79%|███████▊  | 8399/10682 [1:08:28<14:14,  2.67it/s]
 79%|███████▊  | 8400/10682 [1:08:28<14:14,  2.67it/s]
                                                      
{'loss': 4.2165, 'grad_norm': 0.40509170293807983, 'learning_rate': 0.0001327181448103661, 'epoch': 0.79}
-
 79%|███████▊  | 8400/10682 [1:08:28<14:14,  2.67it/s]
 79%|███████▊  | 8401/10682 [1:08:29<14:15,  2.66it/s]
 79%|███████▊  | 8402/10682 [1:08:29<14:14,  2.67it/s]
 79%|███████▊  | 8403/10682 [1:08:30<14:14,  2.67it/s]
 79%|███████▊  | 8404/10682 [1:08:30<14:15,  2.66it/s]
 79%|███████▊  | 8405/10682 [1:08:30<14:14,  2.66it/s]
 79%|███████▊  | 8406/10682 [1:08:31<14:15,  2.66it/s]
 79%|███████▊  | 8407/10682 [1:08:31<14:14,  2.66it/s]
 79%|███████▊  | 8408/10682 [1:08:31<14:13,  2.66it/s]
 79%|███████▊  | 8409/10682 [1:08:32<14:13,  2.66it/s]
 79%|███████▊  | 8410/10682 [1:08:32<14:11,  2.67it/s]
 79%|███████▊  | 8411/10682 [1:08:33<14:11,  2.67it/s]
 79%|███████▊  | 8412/10682 [1:08:33<14:10,  2.67it/s]
 79%|███████▉  | 8413/10682 [1:08:33<14:10,  2.67it/s]
 79%|███████▉  | 8414/10682 [1:08:34<14:10,  2.67it/s]
 79%|███████▉  | 8415/10682 [1:08:34<14:09,  2.67it/s]
 79%|███████▉  | 8416/10682 [1:08:34<14:08,  2.67it/s]
 79%|███████▉  | 8417/10682 [1:08:35<14:09,  2.67it/s]
 79%|███████▉  | 8418/10682 [1:08:35<14:08,  2.67it/s]
 79%|███████▉  | 8419/10682 [1:08:36<14:07,  2.67it/s]
 79%|███████▉  | 8420/10682 [1:08:36<14:08,  2.67it/s]
 79%|███████▉  | 8421/10682 [1:08:36<14:07,  2.67it/s]
 79%|███████▉  | 8422/10682 [1:08:37<14:07,  2.67it/s]
 79%|███████▉  | 8423/10682 [1:08:37<14:06,  2.67it/s]
 79%|███████▉  | 8424/10682 [1:08:37<14:05,  2.67it/s]
 79%|███████▉  | 8425/10682 [1:08:38<14:04,  2.67it/s]
                                                      
{'loss': 4.2098, 'grad_norm': 0.39052045345306396, 'learning_rate': 0.00012995854245163207, 'epoch': 0.79}
-
 79%|███████▉  | 8425/10682 [1:08:38<14:04,  2.67it/s]
 79%|███████▉  | 8426/10682 [1:08:38<14:05,  2.67it/s]
 79%|███████▉  | 8427/10682 [1:08:39<14:04,  2.67it/s]
 79%|███████▉  | 8428/10682 [1:08:39<14:03,  2.67it/s]
 79%|███████▉  | 8429/10682 [1:08:39<14:04,  2.67it/s]
 79%|███████▉  | 8430/10682 [1:08:40<14:03,  2.67it/s]
 79%|███████▉  | 8431/10682 [1:08:40<14:04,  2.67it/s]
 79%|███████▉  | 8432/10682 [1:08:40<14:04,  2.67it/s]
 79%|███████▉  | 8433/10682 [1:08:41<14:04,  2.66it/s]
 79%|███████▉  | 8434/10682 [1:08:41<14:03,  2.67it/s]
 79%|███████▉  | 8435/10682 [1:08:42<14:02,  2.67it/s]
 79%|███████▉  | 8436/10682 [1:08:42<14:02,  2.67it/s]
 79%|███████▉  | 8437/10682 [1:08:42<14:01,  2.67it/s]
 79%|███████▉  | 8438/10682 [1:08:43<14:00,  2.67it/s]
 79%|███████▉  | 8439/10682 [1:08:43<13:59,  2.67it/s]
 79%|███████▉  | 8440/10682 [1:08:43<13:58,  2.67it/s]
 79%|███████▉  | 8441/10682 [1:08:44<13:58,  2.67it/s]
 79%|███████▉  | 8442/10682 [1:08:44<13:58,  2.67it/s]
 79%|███████▉  | 8443/10682 [1:08:45<13:58,  2.67it/s]
 79%|███████▉  | 8444/10682 [1:08:45<13:59,  2.67it/s]
 79%|███████▉  | 8445/10682 [1:08:45<13:58,  2.67it/s]
 79%|███████▉  | 8446/10682 [1:08:46<13:57,  2.67it/s]
 79%|███████▉  | 8447/10682 [1:08:46<13:57,  2.67it/s]
 79%|███████▉  | 8448/10682 [1:08:46<13:57,  2.67it/s]
 79%|███████▉  | 8449/10682 [1:08:47<13:56,  2.67it/s]
 79%|███████▉  | 8450/10682 [1:08:47<13:56,  2.67it/s]
                                                      
{'loss': 4.2226, 'grad_norm': 0.403303325176239, 'learning_rate': 0.0001272236408261237, 'epoch': 0.79}
-
 79%|███████▉  | 8450/10682 [1:08:47<13:56,  2.67it/s]
 79%|███████▉  | 8451/10682 [1:08:48<13:57,  2.66it/s]
 79%|███████▉  | 8452/10682 [1:08:48<13:56,  2.66it/s]
 79%|███████▉  | 8453/10682 [1:08:48<13:56,  2.66it/s]
 79%|███████▉  | 8454/10682 [1:08:49<13:55,  2.67it/s]
 79%|███████▉  | 8455/10682 [1:08:49<13:55,  2.67it/s]
 79%|███████▉  | 8456/10682 [1:08:49<13:54,  2.67it/s]
 79%|███████▉  | 8457/10682 [1:08:50<13:54,  2.67it/s]
 79%|███████▉  | 8458/10682 [1:08:50<13:53,  2.67it/s]
 79%|███████▉  | 8459/10682 [1:08:51<13:52,  2.67it/s]
 79%|███████▉  | 8460/10682 [1:08:51<13:54,  2.66it/s]
 79%|███████▉  | 8461/10682 [1:08:51<13:53,  2.66it/s]
 79%|███████▉  | 8462/10682 [1:08:52<13:53,  2.66it/s]
 79%|███████▉  | 8463/10682 [1:08:52<13:52,  2.67it/s]
 79%|███████▉  | 8464/10682 [1:08:52<13:51,  2.67it/s]
 79%|███████▉  | 8465/10682 [1:08:53<13:51,  2.67it/s]
 79%|███████▉  | 8466/10682 [1:08:53<13:50,  2.67it/s]
 79%|███████▉  | 8467/10682 [1:08:54<13:50,  2.67it/s]
 79%|███████▉  | 8468/10682 [1:08:54<13:51,  2.66it/s]
 79%|███████▉  | 8469/10682 [1:08:54<13:50,  2.66it/s]
 79%|███████▉  | 8470/10682 [1:08:55<13:48,  2.67it/s]
 79%|███████▉  | 8471/10682 [1:08:55<13:49,  2.67it/s]
 79%|███████▉  | 8472/10682 [1:08:55<13:47,  2.67it/s]
 79%|███████▉  | 8473/10682 [1:08:56<13:49,  2.66it/s]
 79%|███████▉  | 8474/10682 [1:08:56<13:47,  2.67it/s]
 79%|███████▉  | 8475/10682 [1:08:57<13:47,  2.67it/s]
                                                      {'loss': 4.2198, 'grad_norm': 0.4062618911266327, 'learning_rate': 0.00012451362249196797, 'epoch': 0.79}
-
 79%|███████▉  | 8475/10682 [1:08:57<13:47,  2.67it/s]
 79%|███████▉  | 8476/10682 [1:08:57<13:48,  2.66it/s]
 79%|███████▉  | 8477/10682 [1:08:57<13:48,  2.66it/s]
 79%|███████▉  | 8478/10682 [1:08:58<13:47,  2.66it/s]
 79%|███████▉  | 8479/10682 [1:08:58<13:45,  2.67it/s]
 79%|███████▉  | 8480/10682 [1:08:58<13:45,  2.67it/s]
 79%|███████▉  | 8481/10682 [1:08:59<13:44,  2.67it/s]
 79%|███████▉  | 8482/10682 [1:08:59<13:44,  2.67it/s]
 79%|███████▉  | 8483/10682 [1:09:00<13:42,  2.67it/s]
 79%|███████▉  | 8484/10682 [1:09:00<13:44,  2.67it/s]
 79%|███████▉  | 8485/10682 [1:09:00<13:44,  2.66it/s]
 79%|███████▉  | 8486/10682 [1:09:01<13:43,  2.67it/s]
 79%|███████▉  | 8487/10682 [1:09:01<13:44,  2.66it/s]
 79%|███████▉  | 8488/10682 [1:09:01<13:43,  2.66it/s]
 79%|███████▉  | 8489/10682 [1:09:02<13:42,  2.67it/s]
 79%|███████▉  | 8490/10682 [1:09:02<13:41,  2.67it/s]
 79%|███████▉  | 8491/10682 [1:09:03<13:41,  2.67it/s]
 79%|███████▉  | 8492/10682 [1:09:03<13:41,  2.67it/s]
 80%|███████▉  | 8493/10682 [1:09:03<13:40,  2.67it/s]
 80%|███████▉  | 8494/10682 [1:09:04<13:40,  2.67it/s]
 80%|███████▉  | 8495/10682 [1:09:04<13:39,  2.67it/s]
 80%|███████▉  | 8496/10682 [1:09:04<13:40,  2.67it/s]
 80%|███████▉  | 8497/10682 [1:09:05<13:39,  2.67it/s]
 80%|███████▉  | 8498/10682 [1:09:05<13:39,  2.66it/s]
 80%|███████▉  | 8499/10682 [1:09:06<13:38,  2.67it/s]
 80%|███████▉  | 8500/10682 [1:09:06<13:39,  2.66it/s]{'loss': 4.2241, 'grad_norm': 0.38584184646606445, 'learning_rate': 0.00012182866834630096, 'epoch': 0.8}
                                                      
-
 80%|███████▉  | 8500/10682 [1:09:06<13:39,  2.66it/s]
 80%|███████▉  | 8501/10682 [1:09:06<13:39,  2.66it/s]
 80%|███████▉  | 8502/10682 [1:09:07<13:38,  2.66it/s]
 80%|███████▉  | 8503/10682 [1:09:07<13:37,  2.67it/s]
 80%|███████▉  | 8504/10682 [1:09:07<13:36,  2.67it/s]
 80%|███████▉  | 8505/10682 [1:09:08<13:35,  2.67it/s]
 80%|███████▉  | 8506/10682 [1:09:08<13:34,  2.67it/s]
 80%|███████▉  | 8507/10682 [1:09:09<13:33,  2.67it/s]
 80%|███████▉  | 8508/10682 [1:09:09<13:33,  2.67it/s]
 80%|███████▉  | 8509/10682 [1:09:09<13:35,  2.67it/s]
 80%|███████▉  | 8510/10682 [1:09:10<13:34,  2.67it/s]
 80%|███████▉  | 8511/10682 [1:09:10<13:35,  2.66it/s]
 80%|███████▉  | 8512/10682 [1:09:10<13:34,  2.66it/s]
 80%|███████▉  | 8513/10682 [1:09:11<13:35,  2.66it/s]
 80%|███████▉  | 8514/10682 [1:09:11<13:35,  2.66it/s]
 80%|███████▉  | 8515/10682 [1:09:12<13:34,  2.66it/s]
 80%|███████▉  | 8516/10682 [1:09:12<13:32,  2.66it/s]
 80%|███████▉  | 8517/10682 [1:09:12<13:33,  2.66it/s]
 80%|███████▉  | 8518/10682 [1:09:13<13:32,  2.66it/s]
 80%|███████▉  | 8519/10682 [1:09:13<13:30,  2.67it/s]
 80%|███████▉  | 8520/10682 [1:09:13<13:30,  2.67it/s]
 80%|███████▉  | 8521/10682 [1:09:14<13:30,  2.67it/s]
 80%|███████▉  | 8522/10682 [1:09:14<13:30,  2.67it/s]
 80%|███████▉  | 8523/10682 [1:09:15<13:28,  2.67it/s]
 80%|███████▉  | 8524/10682 [1:09:15<13:29,  2.67it/s]
 80%|███████▉  | 8525/10682 [1:09:15<13:29,  2.67it/s]{'loss': 4.2115, 'grad_norm': 0.3842063248157501, 'learning_rate': 0.00011916895761319264, 'epoch': 0.8}
                                                      
-
 80%|███████▉  | 8525/10682 [1:09:15<13:29,  2.67it/s]
 80%|███████▉  | 8526/10682 [1:09:16<13:30,  2.66it/s]
 80%|███████▉  | 8527/10682 [1:09:16<13:30,  2.66it/s]
 80%|███████▉  | 8528/10682 [1:09:16<13:28,  2.66it/s]
 80%|███████▉  | 8529/10682 [1:09:17<13:28,  2.66it/s]
 80%|███████▉  | 8530/10682 [1:09:17<13:28,  2.66it/s]
 80%|███████▉  | 8531/10682 [1:09:18<13:27,  2.66it/s]
 80%|███████▉  | 8532/10682 [1:09:18<13:26,  2.67it/s]
 80%|███████▉  | 8533/10682 [1:09:18<13:26,  2.66it/s]
 80%|███████▉  | 8534/10682 [1:09:19<13:25,  2.67it/s]
 80%|███████▉  | 8535/10682 [1:09:19<13:24,  2.67it/s]
 80%|███████▉  | 8536/10682 [1:09:19<13:23,  2.67it/s]
 80%|███████▉  | 8537/10682 [1:09:20<13:24,  2.67it/s]
 80%|███████▉  | 8538/10682 [1:09:20<13:25,  2.66it/s]
 80%|███████▉  | 8539/10682 [1:09:21<13:24,  2.66it/s]
 80%|███████▉  | 8540/10682 [1:09:21<13:24,  2.66it/s]
 80%|███████▉  | 8541/10682 [1:09:21<13:23,  2.67it/s]
 80%|███████▉  | 8542/10682 [1:09:22<13:23,  2.66it/s]
 80%|███████▉  | 8543/10682 [1:09:22<13:22,  2.66it/s]
 80%|███████▉  | 8544/10682 [1:09:22<13:22,  2.67it/s]
 80%|███████▉  | 8545/10682 [1:09:23<13:21,  2.67it/s]
 80%|████████  | 8546/10682 [1:09:23<13:21,  2.67it/s]
 80%|████████  | 8547/10682 [1:09:24<13:20,  2.67it/s]
 80%|████████  | 8548/10682 [1:09:24<13:19,  2.67it/s]
 80%|████████  | 8549/10682 [1:09:24<13:20,  2.67it/s]
 80%|████████  | 8550/10682 [1:09:25<13:19,  2.67it/s]{'loss': 4.2102, 'grad_norm': 0.4174042344093323, 'learning_rate': 0.0001165346678316832, 'epoch': 0.8}
                                                      
-
 80%|████████  | 8550/10682 [1:09:25<13:19,  2.67it/s]
 80%|████████  | 8551/10682 [1:09:25<13:19,  2.66it/s]
 80%|████████  | 8552/10682 [1:09:25<13:18,  2.67it/s]
 80%|████████  | 8553/10682 [1:09:26<13:20,  2.66it/s]
 80%|████████  | 8554/10682 [1:09:26<13:21,  2.66it/s]
 80%|████████  | 8555/10682 [1:09:27<13:19,  2.66it/s]
 80%|████████  | 8556/10682 [1:09:27<13:18,  2.66it/s]
 80%|████████  | 8557/10682 [1:09:27<13:18,  2.66it/s]
 80%|████████  | 8558/10682 [1:09:28<13:17,  2.66it/s]
 80%|████████  | 8559/10682 [1:09:28<13:17,  2.66it/s]
 80%|████████  | 8560/10682 [1:09:28<13:15,  2.67it/s]
 80%|████████  | 8561/10682 [1:09:29<13:15,  2.67it/s]
 80%|████████  | 8562/10682 [1:09:29<13:14,  2.67it/s]
 80%|████████  | 8563/10682 [1:09:30<13:14,  2.67it/s]
 80%|████████  | 8564/10682 [1:09:30<13:13,  2.67it/s]
 80%|████████  | 8565/10682 [1:09:30<13:15,  2.66it/s]
 80%|████████  | 8566/10682 [1:09:31<13:14,  2.66it/s]
 80%|████████  | 8567/10682 [1:09:31<13:13,  2.66it/s]
 80%|████████  | 8568/10682 [1:09:31<13:12,  2.67it/s]
 80%|████████  | 8569/10682 [1:09:32<13:13,  2.66it/s]
 80%|████████  | 8570/10682 [1:09:32<13:11,  2.67it/s]
 80%|████████  | 8571/10682 [1:09:33<13:10,  2.67it/s]
 80%|████████  | 8572/10682 [1:09:33<13:10,  2.67it/s]
 80%|████████  | 8573/10682 [1:09:33<13:10,  2.67it/s]
 80%|████████  | 8574/10682 [1:09:34<13:10,  2.67it/s]
 80%|████████  | 8575/10682 [1:09:34<13:09,  2.67it/s]
                                                      
{'loss': 4.2152, 'grad_norm': 0.38696378469467163, 'learning_rate': 0.00011392597484393285, 'epoch': 0.8}
-
 80%|████████  | 8575/10682 [1:09:34<13:09,  2.67it/s]
 80%|████████  | 8576/10682 [1:09:34<13:10,  2.66it/s]
 80%|████████  | 8577/10682 [1:09:35<13:11,  2.66it/s]
 80%|████████  | 8578/10682 [1:09:35<13:10,  2.66it/s]
 80%|████████  | 8579/10682 [1:09:36<13:10,  2.66it/s]
 80%|████████  | 8580/10682 [1:09:36<13:10,  2.66it/s]
 80%|████████  | 8581/10682 [1:09:36<13:09,  2.66it/s]
 80%|████████  | 8582/10682 [1:09:37<13:08,  2.66it/s]
 80%|████████  | 8583/10682 [1:09:37<13:10,  2.66it/s]
 80%|████████  | 8584/10682 [1:09:37<13:09,  2.66it/s]
 80%|████████  | 8585/10682 [1:09:38<13:07,  2.66it/s]
 80%|████████  | 8586/10682 [1:09:38<13:06,  2.67it/s]
 80%|████████  | 8587/10682 [1:09:39<13:05,  2.67it/s]
 80%|████████  | 8588/10682 [1:09:39<13:05,  2.66it/s]
 80%|████████  | 8589/10682 [1:09:39<13:05,  2.67it/s]
 80%|████████  | 8590/10682 [1:09:40<13:04,  2.67it/s]
 80%|████████  | 8591/10682 [1:09:40<13:06,  2.66it/s]
 80%|████████  | 8592/10682 [1:09:40<13:05,  2.66it/s]
 80%|████████  | 8593/10682 [1:09:41<13:05,  2.66it/s]
 80%|████████  | 8594/10682 [1:09:41<13:05,  2.66it/s]
 80%|████████  | 8595/10682 [1:09:42<13:04,  2.66it/s]
 80%|████████  | 8596/10682 [1:09:42<13:03,  2.66it/s]
 80%|████████  | 8597/10682 [1:09:42<13:03,  2.66it/s]
 80%|████████  | 8598/10682 [1:09:43<13:03,  2.66it/s]
 80%|████████  | 8599/10682 [1:09:43<13:02,  2.66it/s]
 81%|████████  | 8600/10682 [1:09:43<13:01,  2.66it/s]
                                                      
{'loss': 4.2171, 'grad_norm': 0.38167816400527954, 'learning_rate': 0.00011134305278348312, 'epoch': 0.81}
-
 81%|████████  | 8600/10682 [1:09:43<13:01,  2.66it/s]
 81%|████████  | 8601/10682 [1:09:44<13:02,  2.66it/s]
 81%|████████  | 8602/10682 [1:09:44<13:02,  2.66it/s]
 81%|████████  | 8603/10682 [1:09:45<13:01,  2.66it/s]
 81%|████████  | 8604/10682 [1:09:45<13:02,  2.66it/s]
 81%|████████  | 8605/10682 [1:09:45<13:01,  2.66it/s]
 81%|████████  | 8606/10682 [1:09:46<13:01,  2.66it/s]
 81%|████████  | 8607/10682 [1:09:46<12:59,  2.66it/s]
 81%|████████  | 8608/10682 [1:09:47<13:00,  2.66it/s]
 81%|████████  | 8609/10682 [1:09:47<12:58,  2.66it/s]
 81%|████████  | 8610/10682 [1:09:47<12:57,  2.66it/s]
 81%|████████  | 8611/10682 [1:09:48<12:56,  2.67it/s]
 81%|████████  | 8612/10682 [1:09:48<12:56,  2.66it/s]
 81%|████████  | 8613/10682 [1:09:48<12:56,  2.67it/s]
 81%|████████  | 8614/10682 [1:09:49<12:55,  2.67it/s]
 81%|████████  | 8615/10682 [1:09:49<12:54,  2.67it/s]
 81%|████████  | 8616/10682 [1:09:50<12:56,  2.66it/s]
 81%|████████  | 8617/10682 [1:09:50<12:55,  2.66it/s]
 81%|████████  | 8618/10682 [1:09:50<12:56,  2.66it/s]
 81%|████████  | 8619/10682 [1:09:51<12:54,  2.66it/s]
 81%|████████  | 8620/10682 [1:09:51<12:55,  2.66it/s]
 81%|████████  | 8621/10682 [1:09:51<12:54,  2.66it/s]
 81%|████████  | 8622/10682 [1:09:52<12:54,  2.66it/s]
 81%|████████  | 8623/10682 [1:09:52<12:53,  2.66it/s]
 81%|████████  | 8624/10682 [1:09:53<12:52,  2.66it/s]
 81%|████████  | 8625/10682 [1:09:53<12:52,  2.66it/s]
                                                      
{'loss': 4.2171, 'grad_norm': 0.40333402156829834, 'learning_rate': 0.00010878607406363367, 'epoch': 0.81}
-
 81%|████████  | 8625/10682 [1:09:53<12:52,  2.66it/s]
 81%|████████  | 8626/10682 [1:09:53<12:51,  2.66it/s]
 81%|████████  | 8627/10682 [1:09:54<12:52,  2.66it/s]
 81%|████████  | 8628/10682 [1:09:54<12:51,  2.66it/s]
 81%|████████  | 8629/10682 [1:09:54<12:50,  2.66it/s]
 81%|████████  | 8630/10682 [1:09:55<12:49,  2.67it/s]
 81%|████████  | 8631/10682 [1:09:55<12:49,  2.67it/s]
 81%|████████  | 8632/10682 [1:09:56<12:50,  2.66it/s]
 81%|████████  | 8633/10682 [1:09:56<12:50,  2.66it/s]
 81%|████████  | 8634/10682 [1:09:56<12:49,  2.66it/s]
 81%|████████  | 8635/10682 [1:09:57<12:48,  2.66it/s]
 81%|████████  | 8636/10682 [1:09:57<12:48,  2.66it/s]
 81%|████████  | 8637/10682 [1:09:57<12:47,  2.66it/s]
 81%|████████  | 8638/10682 [1:09:58<12:47,  2.66it/s]
 81%|████████  | 8639/10682 [1:09:58<12:46,  2.67it/s]
 81%|████████  | 8640/10682 [1:09:59<12:46,  2.66it/s]
 81%|████████  | 8641/10682 [1:09:59<12:46,  2.66it/s]
 81%|████████  | 8642/10682 [1:09:59<12:45,  2.67it/s]
 81%|████████  | 8643/10682 [1:10:00<12:44,  2.67it/s]
 81%|████████  | 8644/10682 [1:10:00<12:46,  2.66it/s]
 81%|████████  | 8645/10682 [1:10:00<12:45,  2.66it/s]
 81%|████████  | 8646/10682 [1:10:01<12:45,  2.66it/s]
 81%|████████  | 8647/10682 [1:10:01<12:45,  2.66it/s]
 81%|████████  | 8648/10682 [1:10:02<12:44,  2.66it/s]
 81%|████████  | 8649/10682 [1:10:02<12:43,  2.66it/s]
 81%|████████  | 8650/10682 [1:10:02<12:42,  2.67it/s]
                                                      
{'loss': 4.2098, 'grad_norm': 0.37596365809440613, 'learning_rate': 0.00010625520936593375, 'epoch': 0.81}
-
 81%|████████  | 8650/10682 [1:10:02<12:42,  2.67it/s]
 81%|████████  | 8651/10682 [1:10:03<12:42,  2.66it/s]
 81%|████████  | 8652/10682 [1:10:03<12:42,  2.66it/s]
 81%|████████  | 8653/10682 [1:10:03<12:42,  2.66it/s]
 81%|████████  | 8654/10682 [1:10:04<12:40,  2.67it/s]
 81%|████████  | 8655/10682 [1:10:04<12:39,  2.67it/s]
 81%|████████  | 8656/10682 [1:10:05<12:39,  2.67it/s]
 81%|████████  | 8657/10682 [1:10:05<12:38,  2.67it/s]
 81%|████████  | 8658/10682 [1:10:05<12:38,  2.67it/s]
 81%|████████  | 8659/10682 [1:10:06<12:38,  2.67it/s]
 81%|████████  | 8660/10682 [1:10:06<12:39,  2.66it/s]
 81%|████████  | 8661/10682 [1:10:06<12:39,  2.66it/s]
 81%|████████  | 8662/10682 [1:10:07<12:38,  2.66it/s]
 81%|████████  | 8663/10682 [1:10:07<12:38,  2.66it/s]
 81%|████████  | 8664/10682 [1:10:08<12:37,  2.66it/s]
 81%|████████  | 8665/10682 [1:10:08<12:37,  2.66it/s]
 81%|████████  | 8666/10682 [1:10:08<12:36,  2.67it/s]
 81%|████████  | 8667/10682 [1:10:09<12:35,  2.67it/s]
 81%|████████  | 8668/10682 [1:10:09<12:35,  2.67it/s]
 81%|████████  | 8669/10682 [1:10:09<12:35,  2.67it/s]
 81%|████████  | 8670/10682 [1:10:10<12:35,  2.66it/s]
 81%|████████  | 8671/10682 [1:10:10<12:34,  2.67it/s]
 81%|████████  | 8672/10682 [1:10:11<12:33,  2.67it/s]
 81%|████████  | 8673/10682 [1:10:11<12:33,  2.67it/s]
 81%|████████  | 8674/10682 [1:10:11<12:32,  2.67it/s]
 81%|████████  | 8675/10682 [1:10:12<12:32,  2.67it/s]
                                                      
{'loss': 4.2091, 'grad_norm': 0.35390347242355347, 'learning_rate': 0.0001037506276287885, 'epoch': 0.81}
-
 81%|████████  | 8675/10682 [1:10:12<12:32,  2.67it/s]
 81%|████████  | 8676/10682 [1:10:12<12:32,  2.67it/s]
 81%|████████  | 8677/10682 [1:10:12<12:32,  2.67it/s]
 81%|████████  | 8678/10682 [1:10:13<12:32,  2.66it/s]
 81%|████████  | 8679/10682 [1:10:13<12:31,  2.67it/s]
 81%|████████▏ | 8680/10682 [1:10:14<12:30,  2.67it/s]
 81%|████████▏ | 8681/10682 [1:10:14<12:30,  2.67it/s]
 81%|████████▏ | 8682/10682 [1:10:14<12:29,  2.67it/s]
 81%|████████▏ | 8683/10682 [1:10:15<12:29,  2.67it/s]
 81%|████████▏ | 8684/10682 [1:10:15<12:30,  2.66it/s]
 81%|█████���██▏ | 8685/10682 [1:10:15<12:30,  2.66it/s]
 81%|████████▏ | 8686/10682 [1:10:16<12:31,  2.66it/s]
 81%|████████▏ | 8687/10682 [1:10:16<12:29,  2.66it/s]
 81%|████████▏ | 8688/10682 [1:10:17<12:29,  2.66it/s]
 81%|████████▏ | 8689/10682 [1:10:17<12:27,  2.66it/s]
 81%|████████▏ | 8690/10682 [1:10:17<12:26,  2.67it/s]
 81%|████████▏ | 8691/10682 [1:10:18<12:27,  2.66it/s]
 81%|████████▏ | 8692/10682 [1:10:18<12:26,  2.66it/s]
 81%|████████▏ | 8693/10682 [1:10:18<12:26,  2.67it/s]
 81%|████████▏ | 8694/10682 [1:10:19<12:25,  2.67it/s]
 81%|████████▏ | 8695/10682 [1:10:19<12:25,  2.67it/s]
 81%|████████▏ | 8696/10682 [1:10:20<12:24,  2.67it/s]
 81%|████████▏ | 8697/10682 [1:10:20<12:25,  2.66it/s]
 81%|████████▏ | 8698/10682 [1:10:20<12:25,  2.66it/s]
 81%|████████▏ | 8699/10682 [1:10:21<12:24,  2.66it/s]
 81%|████████▏ | 8700/10682 [1:10:21<12:25,  2.66it/s]
                                                      
{'loss': 4.2163, 'grad_norm': 0.39793896675109863, 'learning_rate': 0.0001012724960361826, 'epoch': 0.81}
-
 81%|████████▏ | 8700/10682 [1:10:21<12:25,  2.66it/s]
 81%|████████▏ | 8701/10682 [1:10:21<12:25,  2.66it/s]
 81%|████████▏ | 8702/10682 [1:10:22<12:24,  2.66it/s]
 81%|████████▏ | 8703/10682 [1:10:22<12:23,  2.66it/s]
 81%|████████▏ | 8704/10682 [1:10:23<12:23,  2.66it/s]
 81%|████████▏ | 8705/10682 [1:10:23<12:21,  2.67it/s]
 82%|████████▏ | 8706/10682 [1:10:23<12:21,  2.67it/s]
 82%|████████▏ | 8707/10682 [1:10:24<12:21,  2.67it/s]
 82%|████████▏ | 8708/10682 [1:10:24<12:20,  2.66it/s]
 82%|████████▏ | 8709/10682 [1:10:24<12:20,  2.67it/s]
 82%|████████▏ | 8710/10682 [1:10:25<12:19,  2.67it/s]
 82%|████████▏ | 8711/10682 [1:10:25<12:20,  2.66it/s]
 82%|████████▏ | 8712/10682 [1:10:26<12:19,  2.66it/s]
 82%|████████▏ | 8713/10682 [1:10:26<12:20,  2.66it/s]
 82%|████████▏ | 8714/10682 [1:10:26<12:19,  2.66it/s]
 82%|████████▏ | 8715/10682 [1:10:27<12:18,  2.66it/s]
 82%|████████▏ | 8716/10682 [1:10:27<12:17,  2.67it/s]
 82%|████████▏ | 8717/10682 [1:10:27<12:17,  2.67it/s]
 82%|████████▏ | 8718/10682 [1:10:28<12:16,  2.67it/s]
 82%|████████▏ | 8719/10682 [1:10:28<12:15,  2.67it/s]
 82%|████████▏ | 8720/10682 [1:10:29<12:15,  2.67it/s]
 82%|████████▏ | 8721/10682 [1:10:29<12:15,  2.67it/s]
 82%|████████▏ | 8722/10682 [1:10:29<12:14,  2.67it/s]
 82%|████████▏ | 8723/10682 [1:10:30<12:13,  2.67it/s]
 82%|████████▏ | 8724/10682 [1:10:30<12:13,  2.67it/s]
 82%|████████▏ | 8725/10682 [1:10:30<12:13,  2.67it/s]{'loss': 4.2166, 'grad_norm': 0.38921457529067993, 'learning_rate': 9.882098000652034e-05, 'epoch': 0.82}
-
                                                      

 82%|████████▏ | 8725/10682 [1:10:30<12:13,  2.67it/s]
 82%|████████▏ | 8726/10682 [1:10:31<12:14,  2.66it/s]
 82%|████████▏ | 8727/10682 [1:10:31<12:16,  2.66it/s]
 82%|████████▏ | 8728/10682 [1:10:32<12:15,  2.66it/s]
 82%|████████▏ | 8729/10682 [1:10:32<12:13,  2.66it/s]
 82%|████████▏ | 8730/10682 [1:10:32<12:13,  2.66it/s]
 82%|████████▏ | 8731/10682 [1:10:33<12:13,  2.66it/s]
 82%|████████▏ | 8732/10682 [1:10:33<12:12,  2.66it/s]
 82%|████████▏ | 8733/10682 [1:10:33<12:12,  2.66it/s]
 82%|████████▏ | 8734/10682 [1:10:34<12:10,  2.67it/s]
 82%|████████▏ | 8735/10682 [1:10:34<12:10,  2.67it/s]
 82%|████████▏ | 8736/10682 [1:10:35<12:10,  2.66it/s]
 82%|████████▏ | 8737/10682 [1:10:35<12:10,  2.66it/s]
 82%|████████▏ | 8738/10682 [1:10:35<12:10,  2.66it/s]
 82%|████████▏ | 8739/10682 [1:10:36<12:09,  2.66it/s]
 82%|████████▏ | 8740/10682 [1:10:36<12:09,  2.66it/s]
 82%|████████▏ | 8741/10682 [1:10:36<12:08,  2.66it/s]
 82%|████████▏ | 8742/10682 [1:10:37<12:07,  2.67it/s]
 82%|████████▏ | 8743/10682 [1:10:37<12:07,  2.66it/s]
 82%|████████▏ | 8744/10682 [1:10:38<12:07,  2.66it/s]
 82%|████████▏ | 8745/10682 [1:10:38<12:06,  2.67it/s]
 82%|████████▏ | 8746/10682 [1:10:38<12:05,  2.67it/s]
 82%|████████▏ | 8747/10682 [1:10:39<12:04,  2.67it/s]
 82%|████████▏ | 8748/10682 [1:10:39<12:05,  2.67it/s]
 82%|████████▏ | 8749/10682 [1:10:39<12:04,  2.67it/s]
 82%|████████▏ | 8750/10682 [1:10:40<12:04,  2.67it/s]{'loss': 4.2125, 'grad_norm': 0.37359005212783813, 'learning_rate': 9.639624318158335e-05, 'epoch': 0.82}
                                                      
-
 82%|████████▏ | 8750/10682 [1:10:40<12:04,  2.67it/s]
 82%|████████▏ | 8751/10682 [1:10:40<12:05,  2.66it/s]
 82%|████████▏ | 8752/10682 [1:10:41<12:04,  2.66it/s]
 82%|████████▏ | 8753/10682 [1:10:41<12:04,  2.66it/s]
 82%|████████▏ | 8754/10682 [1:10:41<12:04,  2.66it/s]
 82%|████████▏ | 8755/10682 [1:10:42<12:04,  2.66it/s]
 82%|████████▏ | 8756/10682 [1:10:42<12:03,  2.66it/s]
 82%|████████▏ | 8757/10682 [1:10:42<12:02,  2.66it/s]
 82%|████████▏ | 8758/10682 [1:10:43<12:02,  2.66it/s]
 82%|████████▏ | 8759/10682 [1:10:43<12:01,  2.66it/s]
 82%|████████▏ | 8760/10682 [1:10:44<12:01,  2.66it/s]
 82%|████████▏ | 8761/10682 [1:10:44<12:00,  2.67it/s]
 82%|████████▏ | 8762/10682 [1:10:44<12:00,  2.67it/s]
 82%|████████▏ | 8763/10682 [1:10:45<11:59,  2.67it/s]
 82%|████████▏ | 8764/10682 [1:10:45<12:00,  2.66it/s]
 82%|████████▏ | 8765/10682 [1:10:45<12:00,  2.66it/s]
 82%|████████▏ | 8766/10682 [1:10:46<12:00,  2.66it/s]
 82%|████████▏ | 8767/10682 [1:10:46<11:59,  2.66it/s]
 82%|████████▏ | 8768/10682 [1:10:47<11:59,  2.66it/s]
 82%|████████▏ | 8769/10682 [1:10:47<11:58,  2.66it/s]
 82%|████████▏ | 8770/10682 [1:10:47<11:58,  2.66it/s]
 82%|████████▏ | 8771/10682 [1:10:48<11:58,  2.66it/s]
 82%|████████▏ | 8772/10682 [1:10:48<11:57,  2.66it/s]
 82%|████████▏ | 8773/10682 [1:10:48<11:56,  2.66it/s]
 82%|████████▏ | 8774/10682 [1:10:49<11:56,  2.66it/s]
 82%|████████▏ | 8775/10682 [1:10:49<11:55,  2.67it/s]{'loss': 4.2135, 'grad_norm': 0.3786744475364685, 'learning_rate': 9.399844741560781e-05, 'epoch': 0.82}
-
                                                      

 82%|████████▏ | 8775/10682 [1:10:49<11:55,  2.67it/s]
 82%|████████▏ | 8776/10682 [1:10:50<11:55,  2.67it/s]
 82%|████████▏ | 8777/10682 [1:10:50<11:54,  2.67it/s]
 82%|████████▏ | 8778/10682 [1:10:50<11:54,  2.66it/s]
 82%|████████▏ | 8779/10682 [1:10:51<11:54,  2.66it/s]
 82%|████████▏ | 8780/10682 [1:10:51<11:54,  2.66it/s]
 82%|████████▏ | 8781/10682 [1:10:51<11:54,  2.66it/s]
 82%|████████▏ | 8782/10682 [1:10:52<11:52,  2.67it/s]
 82%|████████▏ | 8783/10682 [1:10:52<11:51,  2.67it/s]
 82%|████████▏ | 8784/10682 [1:10:53<11:51,  2.67it/s]
 82%|████████▏ | 8785/10682 [1:10:53<11:50,  2.67it/s]
 82%|████████▏ | 8786/10682 [1:10:53<11:50,  2.67it/s]
 82%|████████▏ | 8787/10682 [1:10:54<11:49,  2.67it/s]
 82%|████████▏ | 8788/10682 [1:10:54<11:50,  2.67it/s]
 82%|████████▏ | 8789/10682 [1:10:54<11:49,  2.67it/s]
 82%|████████▏ | 8790/10682 [1:10:55<11:49,  2.67it/s]
 82%|████████▏ | 8791/10682 [1:10:55<11:48,  2.67it/s]
 82%|████████▏ | 8792/10682 [1:10:56<11:48,  2.67it/s]
 82%|████████▏ | 8793/10682 [1:10:56<11:49,  2.66it/s]
 82%|████████▏ | 8794/10682 [1:10:56<11:49,  2.66it/s]
 82%|████████▏ | 8795/10682 [1:10:57<11:48,  2.66it/s]
 82%|████████▏ | 8796/10682 [1:10:57<11:47,  2.66it/s]
 82%|████████▏ | 8797/10682 [1:10:57<11:46,  2.67it/s]
 82%|████████▏ | 8798/10682 [1:10:58<11:47,  2.66it/s]
 82%|████████▏ | 8799/10682 [1:10:58<11:46,  2.67it/s]
 82%|████████▏ | 8800/10682 [1:10:59<11:45,  2.67it/s]{'loss': 4.2105, 'grad_norm': 0.3708396553993225, 'learning_rate': 9.162775276448015e-05, 'epoch': 0.82}
                                                      
-
 82%|████████▏ | 8800/10682 [1:10:59<11:45,  2.67it/s]
 82%|████████▏ | 8801/10682 [1:10:59<11:45,  2.67it/s]
 82%|████████▏ | 8802/10682 [1:10:59<11:44,  2.67it/s]
 82%|████████▏ | 8803/10682 [1:11:00<11:43,  2.67it/s]
 82%|████████▏ | 8804/10682 [1:11:00<11:45,  2.66it/s]
 82%|████████▏ | 8805/10682 [1:11:00<11:44,  2.66it/s]
 82%|████████▏ | 8806/10682 [1:11:01<11:45,  2.66it/s]
 82%|████████▏ | 8807/10682 [1:11:01<11:46,  2.66it/s]
 82%|████████▏ | 8808/10682 [1:11:02<11:45,  2.66it/s]
 82%|████████�� | 8809/10682 [1:11:02<11:43,  2.66it/s]
 82%|████████▏ | 8810/10682 [1:11:02<11:42,  2.66it/s]
 82%|████████▏ | 8811/10682 [1:11:03<11:42,  2.66it/s]
 82%|████████▏ | 8812/10682 [1:11:03<11:41,  2.67it/s]
 83%|████████▎ | 8813/10682 [1:11:03<11:40,  2.67it/s]
 83%|████████▎ | 8814/10682 [1:11:04<11:40,  2.67it/s]
 83%|████████▎ | 8815/10682 [1:11:04<11:40,  2.66it/s]
 83%|████████▎ | 8816/10682 [1:11:05<11:39,  2.67it/s]
 83%|████████▎ | 8817/10682 [1:11:05<11:39,  2.67it/s]
 83%|████████▎ | 8818/10682 [1:11:05<11:39,  2.66it/s]
 83%|████████▎ | 8819/10682 [1:11:06<11:39,  2.66it/s]
 83%|████████▎ | 8820/10682 [1:11:06<11:40,  2.66it/s]
 83%|████████▎ | 8821/10682 [1:11:06<11:39,  2.66it/s]
 83%|████████▎ | 8822/10682 [1:11:07<11:38,  2.66it/s]
 83%|████████▎ | 8823/10682 [1:11:07<11:37,  2.67it/s]
 83%|████████▎ | 8824/10682 [1:11:08<11:36,  2.67it/s]
 83%|████████▎ | 8825/10682 [1:11:08<11:36,  2.67it/s]
                                                      
{'loss': 4.2028, 'grad_norm': 0.37751805782318115, 'learning_rate': 8.928431747505355e-05, 'epoch': 0.83}
-
 83%|████████▎ | 8825/10682 [1:11:08<11:36,  2.67it/s]
 83%|████████▎ | 8826/10682 [1:11:08<11:37,  2.66it/s]
 83%|████████▎ | 8827/10682 [1:11:09<11:36,  2.66it/s]
 83%|████████▎ | 8828/10682 [1:11:09<11:35,  2.67it/s]
 83%|████████▎ | 8829/10682 [1:11:09<11:34,  2.67it/s]
 83%|████████▎ | 8830/10682 [1:11:10<11:34,  2.67it/s]
 83%|████████▎ | 8831/10682 [1:11:10<11:35,  2.66it/s]
 83%|████████▎ | 8832/10682 [1:11:11<11:34,  2.66it/s]
 83%|████████▎ | 8833/10682 [1:11:11<11:35,  2.66it/s]
 83%|████████▎ | 8834/10682 [1:11:11<11:35,  2.66it/s]
 83%|████████▎ | 8835/10682 [1:11:12<11:34,  2.66it/s]
 83%|████████▎ | 8836/10682 [1:11:12<11:33,  2.66it/s]
 83%|████████▎ | 8837/10682 [1:11:12<11:33,  2.66it/s]
 83%|████████▎ | 8838/10682 [1:11:13<11:32,  2.66it/s]
 83%|████████▎ | 8839/10682 [1:11:13<11:31,  2.66it/s]
 83%|████████▎ | 8840/10682 [1:11:14<11:31,  2.67it/s]
 83%|████████▎ | 8841/10682 [1:11:14<11:31,  2.66it/s]
 83%|████████▎ | 8842/10682 [1:11:14<11:30,  2.66it/s]
 83%|████████▎ | 8843/10682 [1:11:15<11:30,  2.66it/s]
 83%|████████▎ | 8844/10682 [1:11:15<11:31,  2.66it/s]
 83%|████████▎ | 8845/10682 [1:11:15<11:30,  2.66it/s]
 83%|████████▎ | 8846/10682 [1:11:16<11:30,  2.66it/s]
 83%|████████▎ | 8847/10682 [1:11:16<11:30,  2.66it/s]
 83%|████████▎ | 8848/10682 [1:11:17<11:30,  2.66it/s]
 83%|████████▎ | 8849/10682 [1:11:17<11:28,  2.66it/s]
 83%|████████▎ | 8850/10682 [1:11:17<11:28,  2.66it/s]
                                                      {'loss': 4.2129, 'grad_norm': 0.35487285256385803, 'learning_rate': 8.6968297974584e-05, 'epoch': 0.83}
-

 83%|████████▎ | 8850/10682 [1:11:17<11:28,  2.66it/s]
 83%|████████▎ | 8851/10682 [1:11:18<11:28,  2.66it/s]
 83%|████████▎ | 8852/10682 [1:11:18<11:28,  2.66it/s]
 83%|████████▎ | 8853/10682 [1:11:18<11:27,  2.66it/s]
 83%|████████▎ | 8854/10682 [1:11:19<11:26,  2.66it/s]
 83%|████████▎ | 8855/10682 [1:11:19<11:26,  2.66it/s]
 83%|████████▎ | 8856/10682 [1:11:20<11:25,  2.66it/s]
 83%|████████▎ | 8857/10682 [1:11:20<11:25,  2.66it/s]
 83%|████████▎ | 8858/10682 [1:11:20<11:25,  2.66it/s]
 83%|████████▎ | 8859/10682 [1:11:21<11:26,  2.66it/s]
 83%|████████▎ | 8860/10682 [1:11:21<11:25,  2.66it/s]
 83%|████████▎ | 8861/10682 [1:11:21<11:24,  2.66it/s]
 83%|████████▎ | 8862/10682 [1:11:22<11:24,  2.66it/s]
 83%|████████▎ | 8863/10682 [1:11:22<11:23,  2.66it/s]
 83%|████████▎ | 8864/10682 [1:11:23<11:22,  2.66it/s]
 83%|████████▎ | 8865/10682 [1:11:23<11:22,  2.66it/s]
 83%|████████▎ | 8866/10682 [1:11:23<11:21,  2.67it/s]
 83%|████████▎ | 8867/10682 [1:11:24<11:21,  2.66it/s]
 83%|████████▎ | 8868/10682 [1:11:24<11:20,  2.66it/s]
 83%|████████▎ | 8869/10682 [1:11:24<11:20,  2.66it/s]
 83%|████████▎ | 8870/10682 [1:11:25<11:21,  2.66it/s]
 83%|████████▎ | 8871/10682 [1:11:25<11:20,  2.66it/s]
 83%|████████▎ | 8872/10682 [1:11:26<11:19,  2.66it/s]
 83%|█���██████▎ | 8873/10682 [1:11:26<11:19,  2.66it/s]
 83%|████████▎ | 8874/10682 [1:11:26<11:19,  2.66it/s]
 83%|████████▎ | 8875/10682 [1:11:27<11:19,  2.66it/s]
                                                      {'loss': 4.2173, 'grad_norm': 0.38829487562179565, 'learning_rate': 8.467984886028967e-05, 'epoch': 0.83}
-
 83%|████████▎ | 8875/10682 [1:11:27<11:19,  2.66it/s]
 83%|████████▎ | 8876/10682 [1:11:27<11:19,  2.66it/s]
 83%|████████▎ | 8877/10682 [1:11:27<11:19,  2.66it/s]
 83%|████████▎ | 8878/10682 [1:11:28<11:18,  2.66it/s]
 83%|████████▎ | 8879/10682 [1:11:28<11:18,  2.66it/s]
 83%|████████▎ | 8880/10682 [1:11:29<11:17,  2.66it/s]
 83%|████████▎ | 8881/10682 [1:11:29<11:17,  2.66it/s]
 83%|████████▎ | 8882/10682 [1:11:29<11:16,  2.66it/s]
 83%|████████▎ | 8883/10682 [1:11:30<11:15,  2.66it/s]
 83%|████████▎ | 8884/10682 [1:11:30<11:15,  2.66it/s]
 83%|████████▎ | 8885/10682 [1:11:30<11:14,  2.66it/s]
 83%|████████▎ | 8886/10682 [1:11:31<11:14,  2.66it/s]
 83%|████████▎ | 8887/10682 [1:11:31<11:14,  2.66it/s]
 83%|████████▎ | 8888/10682 [1:11:32<11:14,  2.66it/s]
 83%|████████▎ | 8889/10682 [1:11:32<11:12,  2.66it/s]
 83%|████████▎ | 8890/10682 [1:11:32<11:13,  2.66it/s]
 83%|████████▎ | 8891/10682 [1:11:33<11:12,  2.66it/s]
 83%|████████▎ | 8892/10682 [1:11:33<11:11,  2.66it/s]
 83%|████████▎ | 8893/10682 [1:11:34<11:11,  2.67it/s]
 83%|████████▎ | 8894/10682 [1:11:34<11:10,  2.67it/s]
 83%|████████▎ | 8895/10682 [1:11:34<11:11,  2.66it/s]
 83%|████████▎ | 8896/10682 [1:11:35<11:09,  2.67it/s]
 83%|████████▎ | 8897/10682 [1:11:35<11:10,  2.66it/s]
 83%|████████▎ | 8898/10682 [1:11:35<11:09,  2.66it/s]
 83%|████████▎ | 8899/10682 [1:11:36<11:09,  2.66it/s]
 83%|████████▎ | 8900/10682 [1:11:36<11:09,  2.66it/s]
                                                      {'loss': 4.2063, 'grad_norm': 0.35413768887519836, 'learning_rate': 8.24191228890303e-05, 'epoch': 0.83}
-
 83%|████████▎ | 8900/10682 [1:11:36<11:09,  2.66it/s]
 83%|████████▎ | 8901/10682 [1:11:37<13:47,  2.15it/s]
 83%|████████▎ | 8902/10682 [1:11:37<12:58,  2.29it/s]
 83%|████████▎ | 8903/10682 [1:11:38<12:25,  2.39it/s]
 83%|████████▎ | 8904/10682 [1:11:38<12:02,  2.46it/s]
 83%|████████▎ | 8905/10682 [1:11:38<11:45,  2.52it/s]
 83%|████████▎ | 8906/10682 [1:11:39<11:32,  2.56it/s]
 83%|████████▎ | 8907/10682 [1:11:39<11:24,  2.59it/s]
 83%|████████▎ | 8908/10682 [1:11:39<11:17,  2.62it/s]
 83%|████████▎ | 8909/10682 [1:11:40<11:14,  2.63it/s]
 83%|████████▎ | 8910/10682 [1:11:40<11:12,  2.63it/s]
 83%|████████▎ | 8911/10682 [1:11:41<11:09,  2.64it/s]
 83%|████████▎ | 8912/10682 [1:11:41<11:09,  2.65it/s]
 83%|████████▎ | 8913/10682 [1:11:41<11:08,  2.65it/s]
 83%|████████▎ | 8914/10682 [1:11:42<11:07,  2.65it/s]
 83%|████████▎ | 8915/10682 [1:11:42<11:06,  2.65it/s]
 83%|████████▎ | 8916/10682 [1:11:42<11:05,  2.65it/s]
 83%|████████▎ | 8917/10682 [1:11:43<11:03,  2.66it/s]
 83%|████████▎ | 8918/10682 [1:11:43<11:02,  2.66it/s]
 83%|████████▎ | 8919/10682 [1:11:44<11:01,  2.66it/s]
 84%|████████▎ | 8920/10682 [1:11:44<11:01,  2.66it/s]
 84%|████████▎ | 8921/10682 [1:11:44<11:01,  2.66it/s]
 84%|████████▎ | 8922/10682 [1:11:45<10:59,  2.67it/s]
 84%|████████▎ | 8923/10682 [1:11:45<11:00,  2.67it/s]
 84%|████████▎ | 8924/10682 [1:11:45<11:00,  2.66it/s]
 84%|████████▎ | 8925/10682 [1:11:46<11:02,  2.65it/s]
                                                      
{'loss': 4.2094, 'grad_norm': 0.38395610451698303, 'learning_rate': 8.018627096711106e-05, 'epoch': 0.84}
-
 84%|████████▎ | 8925/10682 [1:11:46<11:02,  2.65it/s]
 84%|████████▎ | 8926/10682 [1:11:46<11:01,  2.65it/s]
 84%|████████▎ | 8927/10682 [1:11:47<10:59,  2.66it/s]
 84%|████████▎ | 8928/10682 [1:11:47<10:58,  2.66it/s]
 84%|████████▎ | 8929/10682 [1:11:47<10:59,  2.66it/s]
 84%|████████▎ | 8930/10682 [1:11:48<10:57,  2.66it/s]
 84%|████████▎ | 8931/10682 [1:11:48<10:57,  2.66it/s]
 84%|████████▎ | 8932/10682 [1:11:48<10:56,  2.66it/s]
 84%|████████▎ | 8933/10682 [1:11:49<10:55,  2.67it/s]
 84%|████████▎ | 8934/10682 [1:11:49<10:55,  2.66it/s]
 84%|████████▎ | 8935/10682 [1:11:50<10:55,  2.67it/s]
 84%|████████▎ | 8936/10682 [1:11:50<10:56,  2.66it/s]
 84%|████████▎ | 8937/10682 [1:11:50<10:55,  2.66it/s]
 84%|████████▎ | 8938/10682 [1:11:51<13:26,  2.16it/s]
 84%|████████▎ | 8939/10682 [1:11:51<12:39,  2.29it/s]
 84%|████████▎ | 8940/10682 [1:11:52<12:07,  2.39it/s]
 84%|████████▎ | 8941/10682 [1:11:52<11:44,  2.47it/s]
 84%|████████▎ | 8942/10682 [1:11:52<11:28,  2.53it/s]
 84%|████████▎ | 8943/10682 [1:11:53<11:17,  2.57it/s]
 84%|████████▎ | 8944/10682 [1:11:53<11:09,  2.60it/s]
 84%|████████▎ | 8945/10682 [1:11:54<11:03,  2.62it/s]
 84%|████████▎ | 8946/10682 [1:11:54<11:00,  2.63it/s]
 84%|████████▍ | 8947/10682 [1:11:54<10:56,  2.64it/s]
 84%|████████▍ | 8948/10682 [1:11:55<10:54,  2.65it/s]
 84%|████████▍ | 8949/10682 [1:11:55<10:53,  2.65it/s]
 84%|████████▍ | 8950/10682 [1:11:55<10:51,  2.66it/s]
                                                      
{'loss': 4.208, 'grad_norm': 0.4103846251964569, 'learning_rate': 7.798144214020909e-05, 'epoch': 0.84}
-
 84%|████████▍ | 8950/10682 [1:11:55<10:51,  2.66it/s]
 84%|████████▍ | 8951/10682 [1:11:56<10:50,  2.66it/s]
 84%|████████▍ | 8952/10682 [1:11:56<10:49,  2.67it/s]
 84%|████████▍ | 8953/10682 [1:11:57<10:49,  2.66it/s]
 84%|████████▍ | 8954/10682 [1:11:57<10:48,  2.67it/s]
 84%|████████▍ | 8955/10682 [1:11:57<10:47,  2.67it/s]
 84%|████████▍ | 8956/10682 [1:11:58<10:46,  2.67it/s]
 84%|████████▍ | 8957/10682 [1:11:58<10:47,  2.67it/s]
 84%|████████▍ | 8958/10682 [1:11:58<10:46,  2.67it/s]
 84%|████████▍ | 8959/10682 [1:11:59<10:45,  2.67it/s]
 84%|████████▍ | 8960/10682 [1:11:59<10:45,  2.67it/s]
 84%|████████▍ | 8961/10682 [1:12:00<10:44,  2.67it/s]
 84%|████████▍ | 8962/10682 [1:12:00<10:45,  2.66it/s]
 84%|████████▍ | 8963/10682 [1:12:00<10:44,  2.67it/s]
 84%|████████▍ | 8964/10682 [1:12:01<10:44,  2.67it/s]
 84%|████████▍ | 8965/10682 [1:12:01<10:43,  2.67it/s]
 84%|████████▍ | 8966/10682 [1:12:01<10:44,  2.66it/s]
 84%|████████▍ | 8967/10682 [1:12:02<10:43,  2.67it/s]
 84%|████████▍ | 8968/10682 [1:12:02<10:42,  2.67it/s]
 84%|████████▍ | 8969/10682 [1:12:03<10:41,  2.67it/s]
 84%|████████▍ | 8970/10682 [1:12:03<10:41,  2.67it/s]
 84%|████████▍ | 8971/10682 [1:12:03<10:40,  2.67it/s]
 84%|████████▍ | 8972/10682 [1:12:04<10:39,  2.67it/s]
 84%|████████▍ | 8973/10682 [1:12:04<10:39,  2.67it/s]
 84%|████████▍ | 8974/10682 [1:12:04<10:40,  2.67it/s]
 84%|████████▍ | 8975/10682 [1:12:05<10:39,  2.67it/s]
                                                      {'loss': 4.2113, 'grad_norm': 0.38598573207855225, 'learning_rate': 7.58047835834249e-05, 'epoch': 0.84}
-
 84%|████████▍ | 8975/10682 [1:12:05<10:39,  2.67it/s]
 84%|████████▍ | 8976/10682 [1:12:05<10:39,  2.67it/s]
 84%|████████▍ | 8977/10682 [1:12:06<10:40,  2.66it/s]
 84%|████████▍ | 8978/10682 [1:12:06<10:39,  2.66it/s]
 84%|████████▍ | 8979/10682 [1:12:06<10:37,  2.67it/s]
 84%|████████▍ | 8980/10682 [1:12:07<10:37,  2.67it/s]
 84%|████████▍ | 8981/10682 [1:12:07<10:36,  2.67it/s]
 84%|████████▍ | 8982/10682 [1:12:07<10:36,  2.67it/s]
 84%|████████▍ | 8983/10682 [1:12:08<10:36,  2.67it/s]
 84%|████████▍ | 8984/10682 [1:12:08<10:36,  2.67it/s]
 84%|████████▍ | 8985/10682 [1:12:09<10:35,  2.67it/s]
 84%|████████▍ | 8986/10682 [1:12:09<10:34,  2.67it/s]
 84%|████████▍ | 8987/10682 [1:12:09<10:34,  2.67it/s]
 84%|████████▍ | 8988/10682 [1:12:10<10:34,  2.67it/s]
 84%|████████▍ | 8989/10682 [1:12:10<10:33,  2.67it/s]
 84%|████████▍ | 8990/10682 [1:12:10<10:33,  2.67it/s]
 84%|████████▍ | 8991/10682 [1:12:11<10:33,  2.67it/s]
 84%|████████▍ | 8992/10682 [1:12:11<10:32,  2.67it/s]
 84%|████████▍ | 8993/10682 [1:12:12<10:32,  2.67it/s]
 84%|████████▍ | 8994/10682 [1:12:12<10:31,  2.67it/s]
 84%|████████▍ | 8995/10682 [1:12:12<10:31,  2.67it/s]
 84%|████████▍ | 8996/10682 [1:12:13<10:31,  2.67it/s]
 84%|█████���██▍ | 8997/10682 [1:12:13<10:31,  2.67it/s]
 84%|████████▍ | 8998/10682 [1:12:13<10:31,  2.67it/s]
 84%|████████▍ | 8999/10682 [1:12:14<10:30,  2.67it/s]
 84%|████████▍ | 9000/10682 [1:12:14<10:29,  2.67it/s]
                                                      {'loss': 4.2058, 'grad_norm': 0.3868936002254486, 'learning_rate': 7.365644059145782e-05, 'epoch': 0.84}
-
 84%|████████▍ | 9000/10682 [1:12:14<10:29,  2.67it/s]
 84%|████████▍ | 9001/10682 [1:12:15<10:30,  2.67it/s]
 84%|████████▍ | 9002/10682 [1:12:15<10:30,  2.67it/s]
 84%|████████▍ | 9003/10682 [1:12:15<10:30,  2.66it/s]
 84%|████████▍ | 9004/10682 [1:12:16<10:29,  2.66it/s]
 84%|████████▍ | 9005/10682 [1:12:16<10:28,  2.67it/s]
 84%|████████▍ | 9006/10682 [1:12:16<10:28,  2.67it/s]
 84%|████████▍ | 9007/10682 [1:12:17<10:29,  2.66it/s]
 84%|████████▍ | 9008/10682 [1:12:17<10:28,  2.66it/s]
 84%|████████▍ | 9009/10682 [1:12:18<10:27,  2.67it/s]
 84%|████████▍ | 9010/10682 [1:12:18<10:26,  2.67it/s]
 84%|████████▍ | 9011/10682 [1:12:18<10:26,  2.67it/s]
 84%|████████▍ | 9012/10682 [1:12:19<10:25,  2.67it/s]
 84%|████████▍ | 9013/10682 [1:12:19<10:25,  2.67it/s]
 84%|████████▍ | 9014/10682 [1:12:19<10:24,  2.67it/s]
 84%|████████▍ | 9015/10682 [1:12:20<10:25,  2.67it/s]
 84%|████████▍ | 9016/10682 [1:12:20<10:24,  2.67it/s]
 84%|████████▍ | 9017/10682 [1:12:21<10:24,  2.67it/s]
 84%|████████▍ | 9018/10682 [1:12:21<10:23,  2.67it/s]
 84%|████████▍ | 9019/10682 [1:12:21<10:22,  2.67it/s]
 84%|████████▍ | 9020/10682 [1:12:22<10:23,  2.67it/s]
 84%|████████▍ | 9021/10682 [1:12:22<10:22,  2.67it/s]
 84%|████████▍ | 9022/10682 [1:12:22<10:21,  2.67it/s]
 84%|████████▍ | 9023/10682 [1:12:23<10:21,  2.67it/s]
 84%|████████▍ | 9024/10682 [1:12:23<10:21,  2.67it/s]
 84%|████████▍ | 9025/10682 [1:12:24<10:20,  2.67it/s]
                                                      
{'loss': 4.2016, 'grad_norm': 0.37044039368629456, 'learning_rate': 7.153655656890773e-05, 'epoch': 0.84}
-
 84%|████████▍ | 9025/10682 [1:12:24<10:20,  2.67it/s]
 84%|████████▍ | 9026/10682 [1:12:24<10:21,  2.67it/s]
 85%|████████▍ | 9027/10682 [1:12:24<10:21,  2.66it/s]
 85%|████████▍ | 9028/10682 [1:12:25<10:19,  2.67it/s]
 85%|████████▍ | 9029/10682 [1:12:25<10:20,  2.66it/s]
 85%|████████▍ | 9030/10682 [1:12:25<10:19,  2.67it/s]
 85%|████████▍ | 9031/10682 [1:12:26<10:20,  2.66it/s]
 85%|████████▍ | 9032/10682 [1:12:26<10:18,  2.67it/s]
 85%|████████▍ | 9033/10682 [1:12:27<10:18,  2.67it/s]
 85%|████████▍ | 9034/10682 [1:12:27<10:17,  2.67it/s]
 85%|████████▍ | 9035/10682 [1:12:27<10:17,  2.67it/s]
 85%|████████▍ | 9036/10682 [1:12:28<10:17,  2.67it/s]
 85%|████████▍ | 9037/10682 [1:12:28<10:16,  2.67it/s]
 85%|████████▍ | 9038/10682 [1:12:28<10:15,  2.67it/s]
 85%|████████▍ | 9039/10682 [1:12:29<10:15,  2.67it/s]
 85%|████████▍ | 9040/10682 [1:12:29<10:15,  2.67it/s]
 85%|████████▍ | 9041/10682 [1:12:30<10:13,  2.67it/s]
 85%|████████▍ | 9042/10682 [1:12:30<10:14,  2.67it/s]
 85%|████████▍ | 9043/10682 [1:12:30<10:13,  2.67it/s]
 85%|████████▍ | 9044/10682 [1:12:31<10:14,  2.66it/s]
 85%|████████▍ | 9045/10682 [1:12:31<10:13,  2.67it/s]
 85%|████████▍ | 9046/10682 [1:12:31<10:13,  2.67it/s]
 85%|████████▍ | 9047/10682 [1:12:32<10:12,  2.67it/s]
 85%|████████▍ | 9048/10682 [1:12:32<10:12,  2.67it/s]
 85%|████████▍ | 9049/10682 [1:12:33<10:11,  2.67it/s]
 85%|████████▍ | 9050/10682 [1:12:33<10:10,  2.67it/s]
                                                      {'loss': 4.2098, 'grad_norm': 0.3900246322154999, 'learning_rate': 6.94452730207023e-05, 'epoch': 0.85}
-
 85%|████████▍ | 9050/10682 [1:12:33<10:10,  2.67it/s]
 85%|████████▍ | 9051/10682 [1:12:33<10:11,  2.67it/s]
 85%|████████▍ | 9052/10682 [1:12:34<10:10,  2.67it/s]
 85%|████████▍ | 9053/10682 [1:12:34<10:10,  2.67it/s]
 85%|████████▍ | 9054/10682 [1:12:34<10:09,  2.67it/s]
 85%|████████▍ | 9055/10682 [1:12:35<10:09,  2.67it/s]
 85%|████████▍ | 9056/10682 [1:12:35<10:08,  2.67it/s]
 85%|████████▍ | 9057/10682 [1:12:36<10:08,  2.67it/s]
 85%|████████▍ | 9058/10682 [1:12:36<10:08,  2.67it/s]
 85%|████████▍ | 9059/10682 [1:12:36<10:08,  2.67it/s]
 85%|████████▍ | 9060/10682 [1:12:37<10:09,  2.66it/s]
 85%|████████▍ | 9061/10682 [1:12:37<10:07,  2.67it/s]
 85%|████████▍ | 9062/10682 [1:12:37<10:07,  2.67it/s]
 85%|████████▍ | 9063/10682 [1:12:38<10:06,  2.67it/s]
 85%|████████▍ | 9064/10682 [1:12:38<10:05,  2.67it/s]
 85%|████████▍ | 9065/10682 [1:12:39<10:05,  2.67it/s]
 85%|████████▍ | 9066/10682 [1:12:39<10:05,  2.67it/s]
 85%|████████▍ | 9067/10682 [1:12:39<10:04,  2.67it/s]
 85%|████████▍ | 9068/10682 [1:12:40<10:03,  2.67it/s]
 85%|████████▍ | 9069/10682 [1:12:40<10:04,  2.67it/s]
 85%|████████▍ | 9070/10682 [1:12:40<10:03,  2.67it/s]
 85%|████████▍ | 9071/10682 [1:12:41<10:03,  2.67it/s]
 85%|████████▍ | 9072/10682 [1:12:41<10:02,  2.67it/s]
 85%|████████▍ | 9073/10682 [1:12:42<10:02,  2.67it/s]
 85%|████████▍ | 9074/10682 [1:12:42<10:02,  2.67it/s]
 85%|████████▍ | 9075/10682 [1:12:42<10:01,  2.67it/s]{'loss': 4.2113, 'grad_norm': 0.3510001003742218, 'learning_rate': 6.738272954265156e-05, 'epoch': 0.85}
                                                      
-
 85%|████████▍ | 9075/10682 [1:12:42<10:01,  2.67it/s]
 85%|████████▍ | 9076/10682 [1:12:43<10:01,  2.67it/s]
 85%|████████▍ | 9077/10682 [1:12:43<10:01,  2.67it/s]
 85%|████████▍ | 9078/10682 [1:12:43<10:00,  2.67it/s]
 85%|████████▍ | 9079/10682 [1:12:44<09:59,  2.67it/s]
 85%|████████▌ | 9080/10682 [1:12:44<09:59,  2.67it/s]
 85%|████████▌ | 9081/10682 [1:12:45<09:59,  2.67it/s]
 85%|████████▌ | 9082/10682 [1:12:45<09:58,  2.67it/s]
 85%|████████▌ | 9083/10682 [1:12:45<09:59,  2.67it/s]
 85%|████████▌ | 9084/10682 [1:12:46<09:59,  2.67it/s]
 85%|████████▌ | 9085/10682 [1:12:46<09:58,  2.67it/s]
 85%|████████▌ | 9086/10682 [1:12:46<09:58,  2.67it/s]
 85%|████████▌ | 9087/10682 [1:12:47<09:58,  2.66it/s]
 85%|████████▌ | 9088/10682 [1:12:47<09:58,  2.66it/s]
 85%|████████▌ | 9089/10682 [1:12:48<09:57,  2.67it/s]
 85%|████████▌ | 9090/10682 [1:12:48<09:56,  2.67it/s]
 85%|████████▌ | 9091/10682 [1:12:48<09:55,  2.67it/s]
 85%|████████▌ | 9092/10682 [1:12:49<09:55,  2.67it/s]
 85%|████████▌ | 9093/10682 [1:12:49<09:54,  2.67it/s]
 85%|████████▌ | 9094/10682 [1:12:49<09:54,  2.67it/s]
 85%|████████▌ | 9095/10682 [1:12:50<09:53,  2.67it/s]
 85%|████████▌ | 9096/10682 [1:12:50<09:53,  2.67it/s]
 85%|████████▌ | 9097/10682 [1:12:51<09:52,  2.67it/s]
 85%|████████▌ | 9098/10682 [1:12:51<09:52,  2.67it/s]
 85%|████████▌ | 9099/10682 [1:12:51<09:53,  2.67it/s]
 85%|████████▌ | 9100/10682 [1:12:52<09:52,  2.67it/s]
                                                      
{'loss': 4.2032, 'grad_norm': 0.3725866675376892, 'learning_rate': 6.534906381212979e-05, 'epoch': 0.85}
-
 85%|████████▌ | 9100/10682 [1:12:52<09:52,  2.67it/s]
 85%|████████▌ | 9101/10682 [1:12:52<09:53,  2.66it/s]
 85%|████████▌ | 9102/10682 [1:12:52<09:53,  2.66it/s]
 85%|████████▌ | 9103/10682 [1:12:53<09:52,  2.67it/s]
 85%|████████▌ | 9104/10682 [1:12:53<09:51,  2.67it/s]
 85%|████████▌ | 9105/10682 [1:12:54<09:51,  2.67it/s]
 85%|████████▌ | 9106/10682 [1:12:54<09:50,  2.67it/s]
 85%|████████▌ | 9107/10682 [1:12:54<09:49,  2.67it/s]
 85%|████████▌ | 9108/10682 [1:12:55<09:49,  2.67it/s]
 85%|████████▌ | 9109/10682 [1:12:55<09:49,  2.67it/s]
 85%|████████▌ | 9110/10682 [1:12:55<09:48,  2.67it/s]
 85%|████████▌ | 9111/10682 [1:12:56<09:48,  2.67it/s]
 85%|████████▌ | 9112/10682 [1:12:56<09:48,  2.67it/s]
 85%|████████▌ | 9113/10682 [1:12:57<09:47,  2.67it/s]
 85%|████████▌ | 9114/10682 [1:12:57<09:48,  2.67it/s]
 85%|████████▌ | 9115/10682 [1:12:57<09:48,  2.66it/s]
 85%|████████▌ | 9116/10682 [1:12:58<09:47,  2.66it/s]
 85%|████████▌ | 9117/10682 [1:12:58<09:46,  2.67it/s]
 85%|████████▌ | 9118/10682 [1:12:58<09:45,  2.67it/s]
 85%|████████▌ | 9119/10682 [1:12:59<09:44,  2.67it/s]
 85%|████████▌ | 9120/10682 [1:12:59<09:44,  2.67it/s]
 85%|████████▌ | 9121/10682 [1:13:00<09:44,  2.67it/s]
 85%|████████▌ | 9122/10682 [1:13:00<09:44,  2.67it/s]
 85%|████████▌ | 9123/10682 [1:13:00<09:43,  2.67it/s]
 85%|████████▌ | 9124/10682 [1:13:01<09:44,  2.67it/s]
 85%|████████▌ | 9125/10682 [1:13:01<09:43,  2.67it/s]
                                                      
{'loss': 4.2081, 'grad_norm': 0.3538189232349396, 'learning_rate': 6.334441157888504e-05, 'epoch': 0.85}
-
 85%|████████▌ | 9125/10682 [1:13:01<09:43,  2.67it/s]
 85%|████████▌ | 9126/10682 [1:13:01<09:44,  2.66it/s]
 85%|████████▌ | 9127/10682 [1:13:02<09:44,  2.66it/s]
 85%|████████▌ | 9128/10682 [1:13:02<09:44,  2.66it/s]
 85%|████████▌ | 9129/10682 [1:13:03<09:43,  2.66it/s]
 85%|████████▌ | 9130/10682 [1:13:03<09:42,  2.67it/s]
 85%|████████▌ | 9131/10682 [1:13:03<09:41,  2.67it/s]
 85%|████████▌ | 9132/10682 [1:13:04<09:40,  2.67it/s]
 85%|████████▌ | 9133/10682 [1:13:04<09:39,  2.67it/s]
 86%|████████▌ | 9134/10682 [1:13:04<09:39,  2.67it/s]
 86%|████████▌ | 9135/10682 [1:13:05<09:38,  2.67it/s]
 86%|████████▌ | 9136/10682 [1:13:05<09:38,  2.67it/s]
 86%|████████▌ | 9137/10682 [1:13:06<09:38,  2.67it/s]
 86%|████████▌ | 9138/10682 [1:13:06<09:39,  2.67it/s]
 86%|████████▌ | 9139/10682 [1:13:06<09:38,  2.67it/s]
 86%|████████▌ | 9140/10682 [1:13:07<09:38,  2.66it/s]
 86%|████████▌ | 9141/10682 [1:13:07<09:37,  2.67it/s]
 86%|████████▌ | 9142/10682 [1:13:07<09:37,  2.67it/s]
 86%|████████▌ | 9143/10682 [1:13:08<09:37,  2.67it/s]
 86%|████████▌ | 9144/10682 [1:13:08<09:36,  2.67it/s]
 86%|████████▌ | 9145/10682 [1:13:09<09:36,  2.67it/s]
 86%|████████▌ | 9146/10682 [1:13:09<09:35,  2.67it/s]
 86%|████████▌ | 9147/10682 [1:13:09<09:35,  2.67it/s]
 86%|████████▌ | 9148/10682 [1:13:10<09:34,  2.67it/s]
 86%|████████▌ | 9149/10682 [1:13:10<09:33,  2.67it/s]
 86%|████████▌ | 9150/10682 [1:13:10<09:33,  2.67it/s]
                                                      {'loss': 4.2075, 'grad_norm': 0.37441039085388184, 'learning_rate': 6.1368906655978e-05, 'epoch': 0.86}
-

 86%|████████▌ | 9150/10682 [1:13:10<09:33,  2.67it/s]
 86%|████████▌ | 9151/10682 [1:13:11<09:34,  2.66it/s]
 86%|████████▌ | 9152/10682 [1:13:11<09:33,  2.67it/s]
 86%|████████▌ | 9153/10682 [1:13:12<09:33,  2.67it/s]
 86%|████████▌ | 9154/10682 [1:13:12<09:32,  2.67it/s]
 86%|████████▌ | 9155/10682 [1:13:12<09:32,  2.67it/s]
 86%|████████▌ | 9156/10682 [1:13:13<09:31,  2.67it/s]
 86%|████████▌ | 9157/10682 [1:13:13<09:31,  2.67it/s]
 86%|████████▌ | 9158/10682 [1:13:13<09:31,  2.67it/s]
 86%|████████▌ | 9159/10682 [1:13:14<09:31,  2.67it/s]
 86%|████████▌ | 9160/10682 [1:13:14<09:30,  2.67it/s]
 86%|████████▌ | 9161/10682 [1:13:15<09:29,  2.67it/s]
 86%|████████▌ | 9162/10682 [1:13:15<09:29,  2.67it/s]
 86%|████████▌ | 9163/10682 [1:13:15<09:28,  2.67it/s]
 86%|████████▌ | 9164/10682 [1:13:16<09:28,  2.67it/s]
 86%|████████▌ | 9165/10682 [1:13:16<09:29,  2.66it/s]
 86%|████████▌ | 9166/10682 [1:13:16<09:29,  2.66it/s]
 86%|████████▌ | 9167/10682 [1:13:17<09:29,  2.66it/s]
 86%|████████▌ | 9168/10682 [1:13:17<09:28,  2.66it/s]
 86%|████████▌ | 9169/10682 [1:13:18<09:27,  2.66it/s]
 86%|████████▌ | 9170/10682 [1:13:18<09:26,  2.67it/s]
 86%|████████▌ | 9171/10682 [1:13:18<09:25,  2.67it/s]
 86%|████████▌ | 9172/10682 [1:13:19<09:25,  2.67it/s]
 86%|████████▌ | 9173/10682 [1:13:19<09:24,  2.67it/s]
 86%|████████▌ | 9174/10682 [1:13:19<09:24,  2.67it/s]
 86%|████████▌ | 9175/10682 [1:13:20<09:24,  2.67it/s]
                                                      {'loss': 4.2037, 'grad_norm': 0.36688268184661865, 'learning_rate': 5.94226809108499e-05, 'epoch': 0.86}
-
 86%|████████▌ | 9175/10682 [1:13:20<09:24,  2.67it/s]
 86%|████████▌ | 9176/10682 [1:13:20<09:24,  2.67it/s]
 86%|████████▌ | 9177/10682 [1:13:21<09:24,  2.67it/s]
 86%|████████▌ | 9178/10682 [1:13:21<09:24,  2.67it/s]
 86%|████████▌ | 9179/10682 [1:13:21<09:23,  2.67it/s]
 86%|████████▌ | 9180/10682 [1:13:22<09:24,  2.66it/s]
 86%|████████▌ | 9181/10682 [1:13:22<09:23,  2.66it/s]
 86%|████████▌ | 9182/10682 [1:13:22<09:22,  2.67it/s]
 86%|████████▌ | 9183/10682 [1:13:23<09:22,  2.66it/s]
 86%|████████▌ | 9184/10682 [1:13:23<09:21,  2.67it/s]
 86%|████████▌ | 9185/10682 [1:13:24<09:21,  2.67it/s]
 86%|████████▌ | 9186/10682 [1:13:24<09:20,  2.67it/s]
 86%|████████▌ | 9187/10682 [1:13:24<09:19,  2.67it/s]
 86%|████████▌ | 9188/10682 [1:13:25<09:18,  2.67it/s]
 86%|████████▌ | 9189/10682 [1:13:25<09:18,  2.67it/s]
 86%|████████▌ | 9190/10682 [1:13:25<09:17,  2.67it/s]
 86%|████████▌ | 9191/10682 [1:13:26<09:18,  2.67it/s]
 86%|████████▌ | 9192/10682 [1:13:26<09:17,  2.67it/s]
 86%|████████▌ | 9193/10682 [1:13:27<09:18,  2.67it/s]
 86%|████████▌ | 9194/10682 [1:13:27<09:17,  2.67it/s]
 86%|████████▌ | 9195/10682 [1:13:27<09:18,  2.66it/s]
 86%|████████▌ | 9196/10682 [1:13:28<09:17,  2.67it/s]
 86%|████████▌ | 9197/10682 [1:13:28<09:16,  2.67it/s]
 86%|████████▌ | 9198/10682 [1:13:28<09:15,  2.67it/s]
 86%|████████▌ | 9199/10682 [1:13:29<09:15,  2.67it/s]
 86%|████████▌ | 9200/10682 [1:13:29<09:14,  2.67it/s]{'loss': 4.2087, 'grad_norm': 0.3599756956100464, 'learning_rate': 5.7505864256519716e-05, 'epoch': 0.86}
                                                      
-
 86%|████████▌ | 9200/10682 [1:13:29<09:14,  2.67it/s]
 86%|████████▌ | 9201/10682 [1:13:30<09:15,  2.67it/s]
 86%|████████▌ | 9202/10682 [1:13:30<09:14,  2.67it/s]
 86%|████████▌ | 9203/10682 [1:13:30<09:13,  2.67it/s]
 86%|████████▌ | 9204/10682 [1:13:31<09:13,  2.67it/s]
 86%|████████▌ | 9205/10682 [1:13:31<09:13,  2.67it/s]
 86%|████████▌ | 9206/10682 [1:13:31<09:12,  2.67it/s]
 86%|████████▌ | 9207/10682 [1:13:32<09:12,  2.67it/s]
 86%|████████▌ | 9208/10682 [1:13:32<09:12,  2.67it/s]
 86%|████████▌ | 9209/10682 [1:13:33<09:12,  2.67it/s]
 86%|████████▌ | 9210/10682 [1:13:33<09:11,  2.67it/s]
 86%|████████▌ | 9211/10682 [1:13:33<09:10,  2.67it/s]
 86%|████████▌ | 9212/10682 [1:13:34<09:09,  2.67it/s]
 86%|████████▌ | 9213/10682 [1:13:34<09:09,  2.67it/s]
 86%|████████▋ | 9214/10682 [1:13:34<09:09,  2.67it/s]
 86%|████████▋ | 9215/10682 [1:13:35<09:09,  2.67it/s]
 86%|████████▋ | 9216/10682 [1:13:35<09:08,  2.67it/s]
 86%|████████▋ | 9217/10682 [1:13:36<09:08,  2.67it/s]
 86%|████████▋ | 9218/10682 [1:13:36<09:07,  2.67it/s]
 86%|████████▋ | 9219/10682 [1:13:36<09:07,  2.67it/s]
 86%|████████▋ | 9220/10682 [1:13:37<09:08,  2.67it/s]
 86%|████████▋ | 9221/10682 [1:13:37<09:07,  2.67it/s]
 86%|████████▋ | 9222/10682 [1:13:37<09:07,  2.67it/s]
 86%|████████▋ | 9223/10682 [1:13:38<09:06,  2.67it/s]
 86%|████████▋ | 9224/10682 [1:13:38<09:06,  2.67it/s]
 86%|████████▋ | 9225/10682 [1:13:39<09:06,  2.67it/s]{'loss': 4.2129, 'grad_norm': 0.3621729612350464, 'learning_rate': 5.561858464291258e-05, 'epoch': 0.86}
                                                      
-
 86%|████████▋ | 9225/10682 [1:13:39<09:06,  2.67it/s]
 86%|████████▋ | 9226/10682 [1:13:39<09:06,  2.66it/s]
 86%|████████▋ | 9227/10682 [1:13:39<09:05,  2.67it/s]
 86%|████████▋ | 9228/10682 [1:13:40<09:04,  2.67it/s]
 86%|████████▋ | 9229/10682 [1:13:40<09:03,  2.67it/s]
 86%|████████▋ | 9230/10682 [1:13:40<09:03,  2.67it/s]
 86%|████████▋ | 9231/10682 [1:13:41<09:03,  2.67it/s]
 86%|████████▋ | 9232/10682 [1:13:41<09:02,  2.67it/s]
 86%|████████▋ | 9233/10682 [1:13:42<09:03,  2.67it/s]
 86%|████████▋ | 9234/10682 [1:13:42<09:02,  2.67it/s]
 86%|████████▋ | 9235/10682 [1:13:42<09:02,  2.67it/s]
 86%|████████▋ | 9236/10682 [1:13:43<09:01,  2.67it/s]
 86%|████████▋ | 9237/10682 [1:13:43<09:02,  2.66it/s]
 86%|████████▋ | 9238/10682 [1:13:43<09:01,  2.67it/s]
 86%|████████▋ | 9239/10682 [1:13:44<09:00,  2.67it/s]
 87%|████████▋ | 9240/10682 [1:13:44<09:00,  2.67it/s]
 87%|████████▋ | 9241/10682 [1:13:45<08:59,  2.67it/s]
 87%|████████▋ | 9242/10682 [1:13:45<08:59,  2.67it/s]
 87%|████████▋ | 9243/10682 [1:13:45<08:58,  2.67it/s]
 87%|████████▋ | 9244/10682 [1:13:46<08:59,  2.67it/s]
 87%|████████▋ | 9245/10682 [1:13:46<08:58,  2.67it/s]
 87%|████████▋ | 9246/10682 [1:13:46<08:58,  2.67it/s]
 87%|████████▋ | 9247/10682 [1:13:47<08:58,  2.67it/s]
 87%|████████▋ | 9248/10682 [1:13:47<08:57,  2.67it/s]
 87%|████████▋ | 9249/10682 [1:13:48<08:57,  2.67it/s]
 87%|████████▋ | 9250/10682 [1:13:48<08:56,  2.67it/s]
                                                      
{'loss': 4.2036, 'grad_norm': 0.3613390028476715, 'learning_rate': 5.3760968048319145e-05, 'epoch': 0.87}
-
 87%|████████▋ | 9250/10682 [1:13:48<08:56,  2.67it/s]
 87%|████████▋ | 9251/10682 [1:13:48<08:57,  2.66it/s]
 87%|████████▋ | 9252/10682 [1:13:49<08:56,  2.66it/s]
 87%|████████▋ | 9253/10682 [1:13:49<08:56,  2.66it/s]
 87%|████████▋ | 9254/10682 [1:13:49<08:55,  2.67it/s]
 87%|████████▋ | 9255/10682 [1:13:50<08:54,  2.67it/s]
 87%|████████▋ | 9256/10682 [1:13:50<08:54,  2.67it/s]
 87%|████████▋ | 9257/10682 [1:13:51<08:54,  2.67it/s]
 87%|████████▋ | 9258/10682 [1:13:51<08:53,  2.67it/s]
 87%|████████▋ | 9259/10682 [1:13:51<08:53,  2.67it/s]
 87%|████████▋ | 9260/10682 [1:13:52<08:53,  2.66it/s]
 87%|████████▋ | 9261/10682 [1:13:52<08:53,  2.66it/s]
 87%|████████▋ | 9262/10682 [1:13:52<08:53,  2.66it/s]
 87%|████████▋ | 9263/10682 [1:13:53<08:51,  2.67it/s]
 87%|████████▋ | 9264/10682 [1:13:53<08:51,  2.67it/s]
 87%|████████▋ | 9265/10682 [1:13:54<08:50,  2.67it/s]
 87%|████████▋ | 9266/10682 [1:13:54<08:50,  2.67it/s]
 87%|████████▋ | 9267/10682 [1:13:54<08:49,  2.67it/s]
 87%|████████▋ | 9268/10682 [1:13:55<08:49,  2.67it/s]
 87%|████████▋ | 9269/10682 [1:13:55<08:48,  2.67it/s]
 87%|████████▋ | 9270/10682 [1:13:55<08:48,  2.67it/s]
 87%|████████▋ | 9271/10682 [1:13:56<08:48,  2.67it/s]
 87%|████████▋ | 9272/10682 [1:13:56<08:47,  2.67it/s]
 87%|████████▋ | 9273/10682 [1:13:57<08:48,  2.67it/s]
 87%|████████▋ | 9274/10682 [1:13:57<08:47,  2.67it/s]
 87%|████████▋ | 9275/10682 [1:13:57<08:48,  2.66it/s]{'loss': 4.2063, 'grad_norm': 0.3530200123786926, 'learning_rate': 5.193313847098613e-05, 'epoch': 0.87}
                                                      
-
 87%|████████▋ | 9275/10682 [1:13:57<08:48,  2.66it/s]
 87%|████████▋ | 9276/10682 [1:13:58<08:48,  2.66it/s]
 87%|████████▋ | 9277/10682 [1:13:58<08:47,  2.66it/s]
 87%|████████▋ | 9278/10682 [1:13:58<08:46,  2.67it/s]
 87%|████████▋ | 9279/10682 [1:13:59<08:45,  2.67it/s]
 87%|████████▋ | 9280/10682 [1:13:59<08:45,  2.67it/s]
 87%|████████▋ | 9281/10682 [1:14:00<08:44,  2.67it/s]
 87%|████████▋ | 9282/10682 [1:14:00<08:44,  2.67it/s]
 87%|████████▋ | 9283/10682 [1:14:00<08:44,  2.67it/s]
 87%|████████▋ | 9284/10682 [1:14:01<08:43,  2.67it/s]
 87%|████████▋ | 9285/10682 [1:14:01<08:43,  2.67it/s]
 87%|████████▋ | 9286/10682 [1:14:01<08:43,  2.67it/s]
 87%|████████▋ | 9287/10682 [1:14:02<08:43,  2.66it/s]
 87%|████████▋ | 9288/10682 [1:14:02<08:42,  2.67it/s]
 87%|████████▋ | 9289/10682 [1:14:03<08:42,  2.67it/s]
 87%|████████▋ | 9290/10682 [1:14:03<08:42,  2.67it/s]
 87%|████████▋ | 9291/10682 [1:14:03<08:42,  2.66it/s]
 87%|████████▋ | 9292/10682 [1:14:04<08:41,  2.67it/s]
 87%|████████▋ | 9293/10682 [1:14:04<08:40,  2.67it/s]
 87%|████████▋ | 9294/10682 [1:14:04<08:40,  2.67it/s]
 87%|████████▋ | 9295/10682 [1:14:05<08:40,  2.67it/s]
 87%|████████▋ | 9296/10682 [1:14:05<08:39,  2.67it/s]
 87%|████████▋ | 9297/10682 [1:14:06<08:39,  2.66it/s]
 87%|████████▋ | 9298/10682 [1:14:06<08:38,  2.67it/s]
 87%|████████▋ | 9299/10682 [1:14:06<08:37,  2.67it/s]
 87%|████████▋ | 9300/10682 [1:14:07<08:38,  2.66it/s]{'loss': 4.2019, 'grad_norm': 0.38395193219184875, 'learning_rate': 5.0135217920839137e-05, 'epoch': 0.87}
-
                                                      

 87%|████████▋ | 9300/10682 [1:14:07<08:38,  2.66it/s]
 87%|████████▋ | 9301/10682 [1:14:07<08:38,  2.66it/s]
 87%|████████▋ | 9302/10682 [1:14:07<08:38,  2.66it/s]
 87%|████████▋ | 9303/10682 [1:14:08<08:37,  2.66it/s]
 87%|████████▋ | 9304/10682 [1:14:08<08:37,  2.67it/s]
 87%|████████▋ | 9305/10682 [1:14:09<08:36,  2.66it/s]
 87%|���███████▋ | 9306/10682 [1:14:09<08:36,  2.67it/s]
 87%|████████▋ | 9307/10682 [1:14:09<08:35,  2.67it/s]
 87%|████████▋ | 9308/10682 [1:14:10<08:35,  2.66it/s]
 87%|████████▋ | 9309/10682 [1:14:10<08:35,  2.66it/s]
 87%|████████▋ | 9310/10682 [1:14:10<08:34,  2.67it/s]
 87%|████████▋ | 9311/10682 [1:14:11<08:34,  2.67it/s]
 87%|████████▋ | 9312/10682 [1:14:11<08:33,  2.67it/s]
 87%|████████▋ | 9313/10682 [1:14:12<08:34,  2.66it/s]
 87%|████████▋ | 9314/10682 [1:14:12<08:33,  2.67it/s]
 87%|████████▋ | 9315/10682 [1:14:12<08:32,  2.67it/s]
 87%|████████▋ | 9316/10682 [1:14:13<08:31,  2.67it/s]
 87%|████████▋ | 9317/10682 [1:14:13<08:31,  2.67it/s]
 87%|████████▋ | 9318/10682 [1:14:13<08:31,  2.67it/s]
 87%|████████▋ | 9319/10682 [1:14:14<08:31,  2.67it/s]
 87%|████████▋ | 9320/10682 [1:14:14<08:30,  2.67it/s]
 87%|████████▋ | 9321/10682 [1:14:15<08:29,  2.67it/s]
 87%|████████▋ | 9322/10682 [1:14:15<08:29,  2.67it/s]
 87%|████████▋ | 9323/10682 [1:14:15<08:28,  2.67it/s]
 87%|████████▋ | 9324/10682 [1:14:16<08:28,  2.67it/s]
 87%|████████▋ | 9325/10682 [1:14:16<08:29,  2.67it/s]
                                                      
{'loss': 4.2027, 'grad_norm': 0.3675685524940491, 'learning_rate': 4.836732641133895e-05, 'epoch': 0.87}
-
 87%|████████▋ | 9325/10682 [1:14:16<08:29,  2.67it/s]
 87%|████████▋ | 9326/10682 [1:14:16<08:29,  2.66it/s]
 87%|████████▋ | 9327/10682 [1:14:17<08:29,  2.66it/s]
 87%|████████▋ | 9328/10682 [1:14:17<08:28,  2.66it/s]
 87%|████████▋ | 9329/10682 [1:14:18<08:27,  2.67it/s]
 87%|████████▋ | 9330/10682 [1:14:18<08:26,  2.67it/s]
 87%|████████▋ | 9331/10682 [1:14:18<08:26,  2.67it/s]
 87%|████████▋ | 9332/10682 [1:14:19<08:26,  2.67it/s]
 87%|████████▋ | 9333/10682 [1:14:19<08:26,  2.66it/s]
 87%|████████▋ | 9334/10682 [1:14:19<08:25,  2.67it/s]
 87%|████████▋ | 9335/10682 [1:14:20<08:25,  2.67it/s]
 87%|████████▋ | 9336/10682 [1:14:20<08:24,  2.67it/s]
 87%|████████▋ | 9337/10682 [1:14:21<08:24,  2.66it/s]
 87%|████████▋ | 9338/10682 [1:14:21<08:24,  2.67it/s]
 87%|████████▋ | 9339/10682 [1:14:21<08:23,  2.67it/s]
 87%|████████▋ | 9340/10682 [1:14:22<08:23,  2.66it/s]
 87%|████████▋ | 9341/10682 [1:14:22<08:22,  2.67it/s]
 87%|████████▋ | 9342/10682 [1:14:22<08:22,  2.67it/s]
 87%|████████▋ | 9343/10682 [1:14:23<08:21,  2.67it/s]
 87%|████████▋ | 9344/10682 [1:14:23<08:22,  2.66it/s]
 87%|████████▋ | 9345/10682 [1:14:24<08:21,  2.67it/s]
 87%|████████▋ | 9346/10682 [1:14:24<08:20,  2.67it/s]
 88%|████████▊ | 9347/10682 [1:14:24<08:20,  2.67it/s]
 88%|████████▊ | 9348/10682 [1:14:25<08:19,  2.67it/s]
 88%|████████▊ | 9349/10682 [1:14:25<08:19,  2.67it/s]
 88%|████████▊ | 9350/10682 [1:14:25<08:18,  2.67it/s]
                                                      
{'loss': 4.2025, 'grad_norm': 0.3643516004085541, 'learning_rate': 4.662958195146971e-05, 'epoch': 0.88}
-
 88%|████████▊ | 9350/10682 [1:14:25<08:18,  2.67it/s]
 88%|████████▊ | 9351/10682 [1:14:26<08:19,  2.66it/s]
 88%|████████▊ | 9352/10682 [1:14:26<08:18,  2.67it/s]
 88%|████████▊ | 9353/10682 [1:14:27<08:19,  2.66it/s]
 88%|████████▊ | 9354/10682 [1:14:27<08:18,  2.66it/s]
 88%|████████▊ | 9355/10682 [1:14:27<08:18,  2.66it/s]
 88%|████████▊ | 9356/10682 [1:14:28<08:17,  2.67it/s]
 88%|████████▊ | 9357/10682 [1:14:28<08:17,  2.66it/s]
 88%|████████▊ | 9358/10682 [1:14:28<08:17,  2.66it/s]
 88%|████████▊ | 9359/10682 [1:14:29<08:16,  2.66it/s]
 88%|████████▊ | 9360/10682 [1:14:29<08:16,  2.66it/s]
 88%|████████▊ | 9361/10682 [1:14:30<08:15,  2.66it/s]
 88%|████████▊ | 9362/10682 [1:14:30<08:15,  2.67it/s]
 88%|████████▊ | 9363/10682 [1:14:30<08:15,  2.66it/s]
 88%|████████▊ | 9364/10682 [1:14:31<08:16,  2.65it/s]
 88%|████████▊ | 9365/10682 [1:14:31<08:16,  2.66it/s]
 88%|████████▊ | 9366/10682 [1:14:31<08:15,  2.66it/s]
 88%|████████▊ | 9367/10682 [1:14:32<08:15,  2.66it/s]
 88%|████████▊ | 9368/10682 [1:14:32<08:13,  2.66it/s]
 88%|████████▊ | 9369/10682 [1:14:33<08:13,  2.66it/s]
 88%|████████▊ | 9370/10682 [1:14:33<08:12,  2.67it/s]
 88%|████████▊ | 9371/10682 [1:14:33<08:11,  2.67it/s]
 88%|████████▊ | 9372/10682 [1:14:34<08:11,  2.67it/s]
 88%|████████▊ | 9373/10682 [1:14:34<08:10,  2.67it/s]
 88%|████████▊ | 9374/10682 [1:14:34<08:10,  2.67it/s]
 88%|████████▊ | 9375/10682 [1:14:35<08:09,  2.67it/s]
                                                      
{'loss': 4.2102, 'grad_norm': 0.3679913580417633, 'learning_rate': 4.492210053786228e-05, 'epoch': 0.88}
-
 88%|████████▊ | 9375/10682 [1:14:35<08:09,  2.67it/s]
 88%|████████▊ | 9376/10682 [1:14:35<08:10,  2.66it/s]
 88%|████████▊ | 9377/10682 [1:14:36<08:09,  2.66it/s]
 88%|████████▊ | 9378/10682 [1:14:36<08:09,  2.67it/s]
 88%|████████▊ | 9379/10682 [1:14:36<08:08,  2.67it/s]
 88%|████████▊ | 9380/10682 [1:14:37<08:08,  2.66it/s]
 88%|████████▊ | 9381/10682 [1:14:37<08:08,  2.67it/s]
 88%|████████▊ | 9382/10682 [1:14:37<08:07,  2.67it/s]
 88%|████████▊ | 9383/10682 [1:14:38<08:07,  2.67it/s]
 88%|████████▊ | 9384/10682 [1:14:38<08:06,  2.67it/s]
 88%|████████▊ | 9385/10682 [1:14:39<08:06,  2.67it/s]
 88%|████████▊ | 9386/10682 [1:14:39<08:05,  2.67it/s]
 88%|████████▊ | 9387/10682 [1:14:39<08:06,  2.66it/s]
 88%|████████▊ | 9388/10682 [1:14:40<08:05,  2.66it/s]
 88%|████████▊ | 9389/10682 [1:14:40<08:05,  2.66it/s]
 88%|████████▊ | 9390/10682 [1:14:40<08:05,  2.66it/s]
 88%|████████▊ | 9391/10682 [1:14:41<08:05,  2.66it/s]
 88%|████████▊ | 9392/10682 [1:14:41<08:04,  2.66it/s]
 88%|████████▊ | 9393/10682 [1:14:42<08:04,  2.66it/s]
 88%|████████▊ | 9394/10682 [1:14:42<08:04,  2.66it/s]
 88%|████████▊ | 9395/10682 [1:14:42<08:03,  2.66it/s]
 88%|████████▊ | 9396/10682 [1:14:43<08:02,  2.66it/s]
 88%|████████▊ | 9397/10682 [1:14:43<08:02,  2.67it/s]
 88%|████████▊ | 9398/10682 [1:14:43<08:01,  2.67it/s]
 88%|████████▊ | 9399/10682 [1:14:44<08:01,  2.67it/s]
 88%|████████▊ | 9400/10682 [1:14:44<08:00,  2.67it/s]
                                                      {'loss': 4.1991, 'grad_norm': 0.35802435874938965, 'learning_rate': 4.3244996147050855e-05, 'epoch': 0.88}
-

 88%|████████▊ | 9400/10682 [1:14:44<08:00,  2.67it/s]
 88%|████████▊ | 9401/10682 [1:14:45<08:01,  2.66it/s]
 88%|████████▊ | 9402/10682 [1:14:45<08:01,  2.66it/s]
 88%|████████▊ | 9403/10682 [1:14:45<08:00,  2.66it/s]
 88%|████████▊ | 9404/10682 [1:14:46<08:00,  2.66it/s]
 88%|████████▊ | 9405/10682 [1:14:46<07:59,  2.66it/s]
 88%|████████▊ | 9406/10682 [1:14:46<07:59,  2.66it/s]
 88%|████████▊ | 9407/10682 [1:14:47<07:59,  2.66it/s]
 88%|████████▊ | 9408/10682 [1:14:47<07:58,  2.66it/s]
 88%|████████▊ | 9409/10682 [1:14:48<07:57,  2.67it/s]
 88%|████████▊ | 9410/10682 [1:14:48<07:58,  2.66it/s]
 88%|████████▊ | 9411/10682 [1:14:48<07:57,  2.66it/s]
 88%|████████▊ | 9412/10682 [1:14:49<07:56,  2.67it/s]
 88%|████████▊ | 9413/10682 [1:14:49<07:55,  2.67it/s]
 88%|████████▊ | 9414/10682 [1:14:49<07:55,  2.67it/s]
 88%|████████▊ | 9415/10682 [1:14:50<07:54,  2.67it/s]
 88%|████████▊ | 9416/10682 [1:14:50<07:55,  2.66it/s]
 88%|████████▊ | 9417/10682 [1:14:51<07:54,  2.67it/s]
 88%|████████▊ | 9418/10682 [1:14:51<07:54,  2.67it/s]
 88%|████████▊ | 9419/10682 [1:14:51<07:53,  2.67it/s]
 88%|████████▊ | 9420/10682 [1:14:52<07:53,  2.66it/s]
 88%|████████▊ | 9421/10682 [1:14:52<07:52,  2.67it/s]
 88%|████████▊ | 9422/10682 [1:14:52<07:52,  2.67it/s]
 88%|████████▊ | 9423/10682 [1:14:53<07:52,  2.66it/s]
 88%|████████▊ | 9424/10682 [1:14:53<07:52,  2.66it/s]
 88%|████████▊ | 9425/10682 [1:14:54<07:51,  2.67it/s]
                                                      
{'loss': 4.2019, 'grad_norm': 0.3508157730102539, 'learning_rate': 4.1598380727865315e-05, 'epoch': 0.88}
-
 88%|████████▊ | 9425/10682 [1:14:54<07:51,  2.67it/s]
 88%|████████▊ | 9426/10682 [1:14:54<07:51,  2.66it/s]
 88%|████████▊ | 9427/10682 [1:14:54<07:50,  2.66it/s]
 88%|████████▊ | 9428/10682 [1:14:55<07:50,  2.66it/s]
 88%|████████▊ | 9429/10682 [1:14:55<07:50,  2.66it/s]
 88%|███��████▊ | 9430/10682 [1:14:55<07:49,  2.67it/s]
 88%|████████▊ | 9431/10682 [1:14:56<07:49,  2.66it/s]
 88%|████████▊ | 9432/10682 [1:14:56<07:49,  2.66it/s]
 88%|████████▊ | 9433/10682 [1:14:57<07:48,  2.66it/s]
 88%|████████▊ | 9434/10682 [1:14:57<07:48,  2.66it/s]
 88%|████████▊ | 9435/10682 [1:14:57<07:47,  2.66it/s]
 88%|████████▊ | 9436/10682 [1:14:58<07:47,  2.66it/s]
 88%|████████▊ | 9437/10682 [1:14:58<07:46,  2.67it/s]
 88%|████████▊ | 9438/10682 [1:14:58<07:46,  2.67it/s]
 88%|████████▊ | 9439/10682 [1:14:59<07:45,  2.67it/s]
 88%|████████▊ | 9440/10682 [1:14:59<07:46,  2.66it/s]
 88%|████████▊ | 9441/10682 [1:15:00<07:45,  2.67it/s]
 88%|████████▊ | 9442/10682 [1:15:00<07:44,  2.67it/s]
 88%|████████▊ | 9443/10682 [1:15:00<07:44,  2.67it/s]
 88%|████████▊ | 9444/10682 [1:15:01<07:44,  2.66it/s]
 88%|████████▊ | 9445/10682 [1:15:01<07:43,  2.67it/s]
 88%|████████▊ | 9446/10682 [1:15:01<07:44,  2.66it/s]
 88%|████████▊ | 9447/10682 [1:15:02<07:43,  2.66it/s]
 88%|████████▊ | 9448/10682 [1:15:02<07:43,  2.66it/s]
 88%|████████▊ | 9449/10682 [1:15:03<07:42,  2.66it/s]
 88%|████████▊ | 9450/10682 [1:15:03<07:42,  2.66it/s]
                                                      {'loss': 4.2014, 'grad_norm': 0.34794771671295166, 'learning_rate': 3.998236419395806e-05, 'epoch': 0.88}
-
 88%|████████▊ | 9450/10682 [1:15:03<07:42,  2.66it/s]
 88%|████████▊ | 9451/10682 [1:15:03<07:42,  2.66it/s]
 88%|████████▊ | 9452/10682 [1:15:04<07:41,  2.66it/s]
 88%|████████▊ | 9453/10682 [1:15:04<07:40,  2.67it/s]
 89%|████████▊ | 9454/10682 [1:15:04<07:40,  2.67it/s]
 89%|████████▊ | 9455/10682 [1:15:05<07:39,  2.67it/s]
 89%|████████▊ | 9456/10682 [1:15:05<07:39,  2.67it/s]
 89%|████████▊ | 9457/10682 [1:15:06<07:39,  2.66it/s]
 89%|████████▊ | 9458/10682 [1:15:06<07:39,  2.67it/s]
 89%|████████▊ | 9459/10682 [1:15:06<07:38,  2.67it/s]
 89%|████████▊ | 9460/10682 [1:15:07<07:38,  2.67it/s]
 89%|████████▊ | 9461/10682 [1:15:07<07:37,  2.67it/s]
 89%|████████▊ | 9462/10682 [1:15:07<07:36,  2.67it/s]
 89%|████████▊ | 9463/10682 [1:15:08<07:35,  2.67it/s]
 89%|████████▊ | 9464/10682 [1:15:08<07:36,  2.67it/s]
 89%|████████▊ | 9465/10682 [1:15:09<07:35,  2.67it/s]
 89%|████████▊ | 9466/10682 [1:15:09<07:35,  2.67it/s]
 89%|████████▊ | 9467/10682 [1:15:09<07:34,  2.67it/s]
 89%|████████▊ | 9468/10682 [1:15:10<07:34,  2.67it/s]
 89%|████████▊ | 9469/10682 [1:15:10<07:34,  2.67it/s]
 89%|████████▊ | 9470/10682 [1:15:10<07:34,  2.67it/s]
 89%|████████▊ | 9471/10682 [1:15:11<07:34,  2.66it/s]
 89%|████████▊ | 9472/10682 [1:15:11<07:34,  2.66it/s]
 89%|████████▊ | 9473/10682 [1:15:12<07:33,  2.66it/s]
 89%|████████▊ | 9474/10682 [1:15:12<07:33,  2.67it/s]
 89%|████████▊ | 9475/10682 [1:15:12<07:32,  2.67it/s]
                                                      
{'loss': 4.1986, 'grad_norm': 0.3629344701766968, 'learning_rate': 3.839705441646779e-05, 'epoch': 0.89}
-
 89%|████████▊ | 9475/10682 [1:15:12<07:32,  2.67it/s]
 89%|████████▊ | 9476/10682 [1:15:13<07:32,  2.66it/s]
 89%|████████▊ | 9477/10682 [1:15:13<07:32,  2.66it/s]
 89%|████████▊ | 9478/10682 [1:15:13<07:31,  2.66it/s]
 89%|████████▊ | 9479/10682 [1:15:14<07:31,  2.67it/s]
 89%|████████▊ | 9480/10682 [1:15:14<07:30,  2.67it/s]
 89%|████████▉ | 9481/10682 [1:15:15<07:30,  2.67it/s]
 89%|████████▉ | 9482/10682 [1:15:15<07:29,  2.67it/s]
 89%|████████▉ | 9483/10682 [1:15:15<07:30,  2.66it/s]
 89%|████████▉ | 9484/10682 [1:15:16<07:30,  2.66it/s]
 89%|████████▉ | 9485/10682 [1:15:16<07:29,  2.66it/s]
 89%|████████▉ | 9486/10682 [1:15:16<07:29,  2.66it/s]
 89%|████████▉ | 9487/10682 [1:15:17<07:29,  2.66it/s]
 89%|████████▉ | 9488/10682 [1:15:17<07:28,  2.66it/s]
 89%|████████▉ | 9489/10682 [1:15:18<07:27,  2.66it/s]
 89%|████████▉ | 9490/10682 [1:15:18<07:27,  2.67it/s]
 89%|████████▉ | 9491/10682 [1:15:18<07:27,  2.66it/s]
 89%|████████▉ | 9492/10682 [1:15:19<07:26,  2.67it/s]
 89%|████████▉ | 9493/10682 [1:15:19<07:25,  2.67it/s]
 89%|████████▉ | 9494/10682 [1:15:19<07:25,  2.67it/s]
 89%|████████▉ | 9495/10682 [1:15:20<07:25,  2.67it/s]
 89%|████████▉ | 9496/10682 [1:15:20<07:24,  2.67it/s]
 89%|████████▉ | 9497/10682 [1:15:21<07:25,  2.66it/s]
 89%|████████▉ | 9498/10682 [1:15:21<07:24,  2.66it/s]
 89%|████████▉ | 9499/10682 [1:15:21<07:23,  2.67it/s]
 89%|████████▉ | 9500/10682 [1:15:22<07:23,  2.67it/s]
                                                      {'loss': 4.198, 'grad_norm': 0.34248027205467224, 'learning_rate': 3.6842557216818006e-05, 'epoch': 0.89}
-
 89%|████████▉ | 9500/10682 [1:15:22<07:23,  2.67it/s]
 89%|████████▉ | 9501/10682 [1:15:22<07:23,  2.66it/s]
 89%|████████▉ | 9502/10682 [1:15:22<07:23,  2.66it/s]
 89%|████████▉ | 9503/10682 [1:15:23<07:22,  2.66it/s]
 89%|████████▉ | 9504/10682 [1:15:23<07:22,  2.66it/s]
 89%|████████▉ | 9505/10682 [1:15:24<07:21,  2.66it/s]
 89%|████████▉ | 9506/10682 [1:15:24<07:21,  2.67it/s]
 89%|████████▉ | 9507/10682 [1:15:24<07:20,  2.67it/s]
 89%|████████▉ | 9508/10682 [1:15:25<07:20,  2.67it/s]
 89%|████████▉ | 9509/10682 [1:15:25<07:19,  2.67it/s]
 89%|████████▉ | 9510/10682 [1:15:25<07:19,  2.67it/s]
 89%|████████▉ | 9511/10682 [1:15:26<07:19,  2.66it/s]
 89%|████████▉ | 9512/10682 [1:15:26<07:19,  2.66it/s]
 89%|████████▉ | 9513/10682 [1:15:27<07:19,  2.66it/s]
 89%|████████▉ | 9514/10682 [1:15:27<07:18,  2.66it/s]
 89%|████████▉ | 9515/10682 [1:15:27<07:18,  2.66it/s]
 89%|████████▉ | 9516/10682 [1:15:28<07:18,  2.66it/s]
 89%|████████▉ | 9517/10682 [1:15:28<07:17,  2.66it/s]
 89%|████████▉ | 9518/10682 [1:15:28<07:17,  2.66it/s]
 89%|████████▉ | 9519/10682 [1:15:29<07:16,  2.66it/s]
 89%|████████▉ | 9520/10682 [1:15:29<07:16,  2.66it/s]
 89%|████████▉ | 9521/10682 [1:15:30<07:15,  2.67it/s]
 89%|████████▉ | 9522/10682 [1:15:30<07:15,  2.67it/s]
 89%|████████▉ | 9523/10682 [1:15:30<07:14,  2.67it/s]
 89%|████████▉ | 9524/10682 [1:15:31<07:15,  2.66it/s]
 89%|████████▉ | 9525/10682 [1:15:31<07:14,  2.66it/s]
                                                      {'loss': 4.1985, 'grad_norm': 0.3607853353023529, 'learning_rate': 3.531897635965431e-05, 'epoch': 0.89}
-

 89%|████████▉ | 9525/10682 [1:15:31<07:14,  2.66it/s]
 89%|████████▉ | 9526/10682 [1:15:31<07:15,  2.66it/s]
 89%|████████▉ | 9527/10682 [1:15:32<07:15,  2.65it/s]
 89%|████████▉ | 9528/10682 [1:15:32<07:14,  2.66it/s]
 89%|████████▉ | 9529/10682 [1:15:33<07:13,  2.66it/s]
 89%|████████▉ | 9530/10682 [1:15:33<07:13,  2.66it/s]
 89%|████████▉ | 9531/10682 [1:15:33<07:12,  2.66it/s]
 89%|████████▉ | 9532/10682 [1:15:34<07:11,  2.66it/s]
 89%|████████▉ | 9533/10682 [1:15:34<07:11,  2.66it/s]
 89%|████████▉ | 9534/10682 [1:15:34<07:10,  2.66it/s]
 89%|████████▉ | 9535/10682 [1:15:35<07:10,  2.66it/s]
 89%|████████▉ | 9536/10682 [1:15:35<07:09,  2.67it/s]
 89%|████████▉ | 9537/10682 [1:15:36<07:10,  2.66it/s]
 89%|████████▉ | 9538/10682 [1:15:36<07:09,  2.66it/s]
 89%|████████▉ | 9539/10682 [1:15:36<07:09,  2.66it/s]
 89%|████████▉ | 9540/10682 [1:15:37<07:09,  2.66it/s]
 89%|████████▉ | 9541/10682 [1:15:37<07:09,  2.66it/s]
 89%|████████▉ | 9542/10682 [1:15:37<07:08,  2.66it/s]
 89%|████████▉ | 9543/10682 [1:15:38<07:07,  2.66it/s]
 89%|████████▉ | 9544/10682 [1:15:38<07:07,  2.66it/s]
 89%|████████▉ | 9545/10682 [1:15:39<07:06,  2.66it/s]
 89%|████████▉ | 9546/10682 [1:15:39<07:06,  2.67it/s]
 89%|████████▉ | 9547/10682 [1:15:39<07:05,  2.67it/s]
 89%|████████▉ | 9548/10682 [1:15:40<07:05,  2.67it/s]
 89%|████████▉ | 9549/10682 [1:15:40<07:04,  2.67it/s]
 89%|████████▉ | 9550/10682 [1:15:40<07:05,  2.66it/s]
                                                      
{'loss': 4.2069, 'grad_norm': 0.3613714575767517, 'learning_rate': 3.382641354591731e-05, 'epoch': 0.89}
-
 89%|████████▉ | 9550/10682 [1:15:40<07:05,  2.66it/s]
 89%|████████▉ | 9551/10682 [1:15:41<07:05,  2.66it/s]
 89%|████████▉ | 9552/10682 [1:15:41<07:04,  2.66it/s]
 89%|████████▉ | 9553/10682 [1:15:42<07:04,  2.66it/s]
 89%|███████���▉ | 9554/10682 [1:15:42<07:04,  2.66it/s]
 89%|████████▉ | 9555/10682 [1:15:42<07:02,  2.66it/s]
 89%|████████▉ | 9556/10682 [1:15:43<07:02,  2.66it/s]
 89%|████████▉ | 9557/10682 [1:15:43<07:02,  2.67it/s]
 89%|████████▉ | 9558/10682 [1:15:43<07:01,  2.66it/s]
 89%|████████▉ | 9559/10682 [1:15:44<07:01,  2.67it/s]
 89%|████████▉ | 9560/10682 [1:15:44<07:01,  2.66it/s]
 90%|████████▉ | 9561/10682 [1:15:45<07:00,  2.67it/s]
 90%|████████▉ | 9562/10682 [1:15:45<07:00,  2.66it/s]
 90%|████████▉ | 9563/10682 [1:15:45<06:59,  2.67it/s]
 90%|████████▉ | 9564/10682 [1:15:46<06:59,  2.66it/s]
 90%|████████▉ | 9565/10682 [1:15:46<06:58,  2.67it/s]
 90%|████████▉ | 9566/10682 [1:15:46<06:58,  2.66it/s]
 90%|████████▉ | 9567/10682 [1:15:47<06:58,  2.66it/s]
 90%|████████▉ | 9568/10682 [1:15:47<06:58,  2.66it/s]
 90%|████████▉ | 9569/10682 [1:15:48<06:57,  2.66it/s]
 90%|████████▉ | 9570/10682 [1:15:48<06:57,  2.66it/s]
 90%|████████▉ | 9571/10682 [1:15:48<06:56,  2.66it/s]
 90%|████████▉ | 9572/10682 [1:15:49<06:56,  2.66it/s]
 90%|████████▉ | 9573/10682 [1:15:49<06:56,  2.66it/s]
 90%|████████▉ | 9574/10682 [1:15:49<06:56,  2.66it/s]
 90%|████████▉ | 9575/10682 [1:15:50<06:56,  2.66it/s]
                                                      {'loss': 4.2033, 'grad_norm': 0.36338624358177185, 'learning_rate': 3.2364968406054075e-05, 'epoch': 0.9}
-
 90%|████████▉ | 9575/10682 [1:15:50<06:56,  2.66it/s]
 90%|████████▉ | 9576/10682 [1:15:50<06:56,  2.66it/s]
 90%|████████▉ | 9577/10682 [1:15:51<06:55,  2.66it/s]
 90%|████████▉ | 9578/10682 [1:15:51<06:55,  2.66it/s]
 90%|████████▉ | 9579/10682 [1:15:51<06:55,  2.65it/s]
 90%|████████▉ | 9580/10682 [1:15:52<06:55,  2.65it/s]
 90%|████████▉ | 9581/10682 [1:15:52<06:54,  2.66it/s]
 90%|████████▉ | 9582/10682 [1:15:53<06:53,  2.66it/s]
 90%|████████▉ | 9583/10682 [1:15:53<06:52,  2.66it/s]
 90%|████████▉ | 9584/10682 [1:15:53<06:52,  2.66it/s]
 90%|████████▉ | 9585/10682 [1:15:54<06:51,  2.67it/s]
 90%|████████▉ | 9586/10682 [1:15:54<06:50,  2.67it/s]
 90%|████████▉ | 9587/10682 [1:15:54<06:50,  2.67it/s]
 90%|████████▉ | 9588/10682 [1:15:55<06:50,  2.66it/s]
 90%|████████▉ | 9589/10682 [1:15:55<06:50,  2.66it/s]
 90%|████████▉ | 9590/10682 [1:15:56<06:50,  2.66it/s]
 90%|████████▉ | 9591/10682 [1:15:56<06:49,  2.67it/s]
 90%|████████▉ | 9592/10682 [1:15:56<06:48,  2.67it/s]
 90%|████████▉ | 9593/10682 [1:15:57<06:48,  2.66it/s]
 90%|████████▉ | 9594/10682 [1:15:57<06:47,  2.67it/s]
 90%|████████▉ | 9595/10682 [1:15:57<06:47,  2.67it/s]
 90%|████████▉ | 9596/10682 [1:15:58<06:46,  2.67it/s]
 90%|████████▉ | 9597/10682 [1:15:58<06:47,  2.67it/s]
 90%|████████▉ | 9598/10682 [1:15:59<06:47,  2.66it/s]
 90%|████████▉ | 9599/10682 [1:15:59<06:46,  2.66it/s]
 90%|████████▉ | 9600/10682 [1:15:59<06:45,  2.67it/s]
                                                      
{'loss': 4.2042, 'grad_norm': 0.3582654595375061, 'learning_rate': 3.093473849336781e-05, 'epoch': 0.9}
-
 90%|████████▉ | 9600/10682 [1:15:59<06:45,  2.67it/s]
 90%|████████▉ | 9601/10682 [1:16:00<06:46,  2.66it/s]
 90%|████████▉ | 9602/10682 [1:16:00<06:45,  2.66it/s]
 90%|████████▉ | 9603/10682 [1:16:00<06:45,  2.66it/s]
 90%|████████▉ | 9604/10682 [1:16:01<06:44,  2.66it/s]
 90%|████████▉ | 9605/10682 [1:16:01<06:44,  2.66it/s]
 90%|████████▉ | 9606/10682 [1:16:02<06:44,  2.66it/s]
 90%|████████▉ | 9607/10682 [1:16:02<06:43,  2.66it/s]
 90%|████████▉ | 9608/10682 [1:16:02<06:43,  2.66it/s]
 90%|████████▉ | 9609/10682 [1:16:03<06:43,  2.66it/s]
 90%|████████▉ | 9610/10682 [1:16:03<06:42,  2.66it/s]
 90%|████████▉ | 9611/10682 [1:16:03<06:41,  2.66it/s]
 90%|████████▉ | 9612/10682 [1:16:04<06:41,  2.67it/s]
 90%|████████▉ | 9613/10682 [1:16:04<06:41,  2.66it/s]
 90%|█████████ | 9614/10682 [1:16:05<06:40,  2.66it/s]
 90%|█████████ | 9615/10682 [1:16:05<06:40,  2.66it/s]
 90%|█████████ | 9616/10682 [1:16:06<08:08,  2.18it/s]
 90%|█████████ | 9617/10682 [1:16:06<07:41,  2.31it/s]
 90%|█████████ | 9618/10682 [1:16:06<07:22,  2.40it/s]
 90%|█████████ | 9619/10682 [1:16:07<07:10,  2.47it/s]
 90%|█████████ | 9620/10682 [1:16:07<07:00,  2.53it/s]
 90%|█████████ | 9621/10682 [1:16:07<06:54,  2.56it/s]
 90%|█████████ | 9622/10682 [1:16:08<06:48,  2.59it/s]
 90%|█████████ | 9623/10682 [1:16:08<06:45,  2.61it/s]
 90%|█████████ | 9624/10682 [1:16:09<06:42,  2.63it/s]
 90%|█████████ | 9625/10682 [1:16:09<06:40,  2.64it/s]{'loss': 4.2065, 'grad_norm': 0.34402942657470703, 'learning_rate': 2.9535819277506203e-05, 'epoch': 0.9}
                                                      
-
 90%|█████████ | 9625/10682 [1:16:09<06:40,  2.64it/s]
 90%|█████████ | 9626/10682 [1:16:09<06:39,  2.65it/s]
 90%|█████████ | 9627/10682 [1:16:10<06:37,  2.65it/s]
 90%|█████████ | 9628/10682 [1:16:10<06:37,  2.65it/s]
 90%|█████████ | 9629/10682 [1:16:10<06:37,  2.65it/s]
 90%|█████████ | 9630/10682 [1:16:11<06:36,  2.65it/s]
 90%|█████████ | 9631/10682 [1:16:11<06:35,  2.66it/s]
 90%|█████████ | 9632/10682 [1:16:12<06:35,  2.66it/s]
 90%|█████████ | 9633/10682 [1:16:12<06:34,  2.66it/s]
 90%|█████████ | 9634/10682 [1:16:12<06:34,  2.66it/s]
 90%|█████████ | 9635/10682 [1:16:13<06:33,  2.66it/s]
 90%|█████████ | 9636/10682 [1:16:13<06:33,  2.66it/s]
 90%|█████████ | 9637/10682 [1:16:13<06:32,  2.66it/s]
 90%|█████████ | 9638/10682 [1:16:14<06:31,  2.66it/s]
 90%|█████████ | 9639/10682 [1:16:14<06:30,  2.67it/s]
 90%|█████████ | 9640/10682 [1:16:15<06:30,  2.67it/s]
 90%|█████████ | 9641/10682 [1:16:15<06:30,  2.67it/s]
 90%|█████████ | 9642/10682 [1:16:15<06:29,  2.67it/s]
 90%|█████████ | 9643/10682 [1:16:16<06:30,  2.66it/s]
 90%|█████████ | 9644/10682 [1:16:16<06:29,  2.67it/s]
 90%|█████████ | 9645/10682 [1:16:16<06:29,  2.66it/s]
 90%|█████████ | 9646/10682 [1:16:17<06:29,  2.66it/s]
 90%|█████████ | 9647/10682 [1:16:17<06:28,  2.67it/s]
 90%|█████████ | 9648/10682 [1:16:18<06:28,  2.66it/s]
 90%|█████████ | 9649/10682 [1:16:18<06:27,  2.67it/s]
 90%|█████████ | 9650/10682 [1:16:18<06:27,  2.67it/s]
                                                      
{'loss': 4.21, 'grad_norm': 0.34801849722862244, 'learning_rate': 2.8168304138088295e-05, 'epoch': 0.9}
-
 90%|█████████ | 9650/10682 [1:16:18<06:27,  2.67it/s]
 90%|█████████ | 9651/10682 [1:16:19<06:26,  2.67it/s]
 90%|█████████ | 9652/10682 [1:16:19<06:26,  2.67it/s]
 90%|█████████ | 9653/10682 [1:16:19<06:26,  2.67it/s]
 90%|█████████ | 9654/10682 [1:16:20<06:25,  2.66it/s]
 90%|█████████ | 9655/10682 [1:16:20<06:25,  2.67it/s]
 90%|█████████ | 9656/10682 [1:16:21<06:25,  2.66it/s]
 90%|█████████ | 9657/10682 [1:16:21<06:24,  2.66it/s]
 90%|█████████ | 9658/10682 [1:16:22<07:50,  2.18it/s]
 90%|█████████ | 9659/10682 [1:16:22<07:23,  2.30it/s]
 90%|█████████ | 9660/10682 [1:16:22<07:05,  2.40it/s]
 90%|█████████ | 9661/10682 [1:16:23<06:51,  2.48it/s]
 90%|█████████ | 9662/10682 [1:16:23<06:42,  2.53it/s]
 90%|█████████ | 9663/10682 [1:16:23<06:36,  2.57it/s]
 90%|█████████ | 9664/10682 [1:16:24<06:31,  2.60it/s]
 90%|█████████ | 9665/10682 [1:16:24<06:28,  2.62it/s]
 90%|█████████ | 9666/10682 [1:16:25<06:26,  2.63it/s]
 90%|█████████ | 9667/10682 [1:16:25<06:24,  2.64it/s]
 91%|█████████ | 9668/10682 [1:16:25<06:22,  2.65it/s]
 91%|█████████ | 9669/10682 [1:16:26<06:21,  2.66it/s]
 91%|█████████ | 9670/10682 [1:16:26<06:20,  2.66it/s]
 91%|█████████ | 9671/10682 [1:16:26<06:19,  2.66it/s]
 91%|█████████ | 9672/10682 [1:16:27<06:18,  2.67it/s]
 91%|█████████ | 9673/10682 [1:16:27<06:18,  2.67it/s]
 91%|█████████ | 9674/10682 [1:16:28<06:18,  2.66it/s]
 91%|█████████ | 9675/10682 [1:16:28<06:18,  2.66it/s]
                                                      
{'loss': 4.2057, 'grad_norm': 0.36842605471611023, 'learning_rate': 2.6832284358471516e-05, 'epoch': 0.91}
-
 91%|█████████ | 9675/10682 [1:16:28<06:18,  2.66it/s]
 91%|█████████ | 9676/10682 [1:16:28<06:18,  2.66it/s]
 91%|█████████ | 9677/10682 [1:16:29<06:17,  2.66it/s]
 91%|█████████ | 9678/10682 [1:16:29<06:16,  2.67it/s]
 91%|█████████ | 9679/10682 [1:16:29<06:15,  2.67it/s]
 91%|█████████ | 9680/10682 [1:16:30<06:15,  2.67it/s]
 91%|█████████ | 9681/10682 [1:16:30<06:14,  2.67it/s]
 91%|█████████ | 9682/10682 [1:16:31<06:14,  2.67it/s]
 91%|█████████ | 9683/10682 [1:16:31<06:13,  2.67it/s]
 91%|█████████ | 9684/10682 [1:16:31<06:13,  2.67it/s]
 91%|█████████ | 9685/10682 [1:16:32<06:13,  2.67it/s]
 91%|█████████ | 9686/10682 [1:16:32<06:13,  2.67it/s]
 91%|█████████ | 9687/10682 [1:16:32<06:13,  2.67it/s]
 91%|█████████ | 9688/10682 [1:16:33<06:12,  2.67it/s]
 91%|█████████ | 9689/10682 [1:16:33<06:11,  2.67it/s]
 91%|█████████ | 9690/10682 [1:16:34<06:11,  2.67it/s]
 91%|█████████ | 9691/10682 [1:16:34<06:11,  2.67it/s]
 91%|█████████ | 9692/10682 [1:16:34<06:11,  2.67it/s]
 91%|█████████ | 9693/10682 [1:16:35<06:10,  2.67it/s]
 91%|█████████ | 9694/10682 [1:16:35<06:10,  2.66it/s]
 91%|█████████ | 9695/10682 [1:16:35<06:10,  2.66it/s]
 91%|█████████ | 9696/10682 [1:16:36<06:09,  2.67it/s]
 91%|█████████ | 9697/10682 [1:16:36<06:09,  2.67it/s]
 91%|█████████ | 9698/10682 [1:16:37<06:09,  2.67it/s]
 91%|█████████ | 9699/10682 [1:16:37<06:08,  2.67it/s]
 91%|█████████ | 9700/10682 [1:16:37<06:09,  2.66it/s]
                                                      
{'loss': 4.2065, 'grad_norm': 0.35051053762435913, 'learning_rate': 2.5527849119658387e-05, 'epoch': 0.91}
-
 91%|█████████ | 9700/10682 [1:16:37<06:09,  2.66it/s]
 91%|█████████ | 9701/10682 [1:16:38<06:08,  2.66it/s]
 91%|█████████ | 9702/10682 [1:16:38<06:08,  2.66it/s]
 91%|█████████ | 9703/10682 [1:16:38<06:07,  2.66it/s]
 91%|█████████ | 9704/10682 [1:16:39<06:07,  2.66it/s]
 91%|█████████ | 9705/10682 [1:16:39<06:06,  2.67it/s]
 91%|█████████ | 9706/10682 [1:16:40<06:05,  2.67it/s]
 91%|█████████ | 9707/10682 [1:16:40<06:05,  2.67it/s]
 91%|█████████ | 9708/10682 [1:16:40<06:05,  2.67it/s]
 91%|█████████ | 9709/10682 [1:16:41<06:04,  2.67it/s]
 91%|█████████ | 9710/10682 [1:16:41<06:03,  2.67it/s]
 91%|█████████ | 9711/10682 [1:16:41<06:04,  2.66it/s]
 91%|█████████ | 9712/10682 [1:16:42<06:03,  2.67it/s]
 91%|█████████ | 9713/10682 [1:16:42<06:03,  2.66it/s]
 91%|█████████ | 9714/10682 [1:16:43<06:03,  2.66it/s]
 91%|█████████ | 9715/10682 [1:16:43<06:02,  2.67it/s]
 91%|█████████ | 9716/10682 [1:16:43<06:01,  2.67it/s]
 91%|█████████ | 9717/10682 [1:16:44<06:01,  2.67it/s]
 91%|█████████ | 9718/10682 [1:16:44<06:01,  2.67it/s]
 91%|█████████ | 9719/10682 [1:16:44<06:00,  2.67it/s]
 91%|█████████ | 9720/10682 [1:16:45<06:00,  2.67it/s]
 91%|█████████ | 9721/10682 [1:16:45<05:59,  2.67it/s]
 91%|█████████ | 9722/10682 [1:16:46<05:59,  2.67it/s]
 91%|█████████ | 9723/10682 [1:16:46<05:59,  2.67it/s]
 91%|█████████ | 9724/10682 [1:16:46<05:59,  2.67it/s]
 91%|█████████ | 9725/10682 [1:16:47<05:59,  2.66it/s]
                                                      
{'loss': 4.2073, 'grad_norm': 0.35367998480796814, 'learning_rate': 2.4255085494343522e-05, 'epoch': 0.91}
-
 91%|█████████ | 9725/10682 [1:16:47<05:59,  2.66it/s]
 91%|█████████ | 9726/10682 [1:16:47<05:58,  2.66it/s]
 91%|█████████ | 9727/10682 [1:16:47<05:58,  2.67it/s]
 91%|█████████ | 9728/10682 [1:16:48<05:58,  2.66it/s]
 91%|█████████ | 9729/10682 [1:16:48<05:57,  2.67it/s]
 91%|█████████ | 9730/10682 [1:16:49<05:57,  2.67it/s]
 91%|█████████ | 9731/10682 [1:16:49<05:56,  2.67it/s]
 91%|█████████ | 9732/10682 [1:16:49<05:55,  2.67it/s]
 91%|█████████ | 9733/10682 [1:16:50<05:55,  2.67it/s]
 91%|█████████ | 9734/10682 [1:16:50<05:54,  2.67it/s]
 91%|█████████ | 9735/10682 [1:16:50<05:54,  2.67it/s]
 91%|█████████ | 9736/10682 [1:16:51<05:54,  2.67it/s]
 91%|█████████ | 9737/10682 [1:16:51<05:54,  2.67it/s]
 91%|█████████ | 9738/10682 [1:16:52<05:53,  2.67it/s]
 91%|█████████ | 9739/10682 [1:16:52<05:53,  2.67it/s]
 91%|█████████ | 9740/10682 [1:16:52<05:53,  2.66it/s]
 91%|█████████ | 9741/10682 [1:16:53<05:53,  2.66it/s]
 91%|██���██████ | 9742/10682 [1:16:53<05:53,  2.66it/s]
 91%|█████████ | 9743/10682 [1:16:53<05:52,  2.67it/s]
 91%|█████████ | 9744/10682 [1:16:54<05:52,  2.66it/s]
 91%|█████████ | 9745/10682 [1:16:54<05:51,  2.67it/s]
 91%|█████████ | 9746/10682 [1:16:55<05:51,  2.67it/s]
 91%|█████████ | 9747/10682 [1:16:55<05:50,  2.67it/s]
 91%|█████████▏| 9748/10682 [1:16:55<05:49,  2.67it/s]
 91%|█████████▏| 9749/10682 [1:16:56<05:49,  2.67it/s]
 91%|█████████▏| 9750/10682 [1:16:56<05:48,  2.67it/s]
                                                      
{'loss': 4.2034, 'grad_norm': 0.3431812524795532, 'learning_rate': 2.301407844110154e-05, 'epoch': 0.91}
-
 91%|█████████▏| 9750/10682 [1:16:56<05:48,  2.67it/s]
 91%|█████████▏| 9751/10682 [1:16:56<05:49,  2.66it/s]
 91%|█████████▏| 9752/10682 [1:16:57<05:49,  2.66it/s]
 91%|█████████▏| 9753/10682 [1:16:57<05:49,  2.66it/s]
 91%|█████████▏| 9754/10682 [1:16:58<05:48,  2.66it/s]
 91%|█████████▏| 9755/10682 [1:16:58<05:48,  2.66it/s]
 91%|█████████▏| 9756/10682 [1:16:58<05:47,  2.66it/s]
 91%|█████████▏| 9757/10682 [1:16:59<05:47,  2.66it/s]
 91%|█████████▏| 9758/10682 [1:16:59<05:47,  2.66it/s]
 91%|█████████▏| 9759/10682 [1:16:59<05:46,  2.66it/s]
 91%|█████████▏| 9760/10682 [1:17:00<05:45,  2.67it/s]
 91%|█████████▏| 9761/10682 [1:17:00<05:45,  2.67it/s]
 91%|█████████▏| 9762/10682 [1:17:01<05:44,  2.67it/s]
 91%|█████████▏| 9763/10682 [1:17:01<05:43,  2.67it/s]
 91%|█████████▏| 9764/10682 [1:17:01<05:44,  2.67it/s]
 91%|█████████▏| 9765/10682 [1:17:02<05:43,  2.67it/s]
 91%|█████████▏| 9766/10682 [1:17:02<05:43,  2.67it/s]
 91%|█████████▏| 9767/10682 [1:17:02<05:43,  2.66it/s]
 91%|█████████▏| 9768/10682 [1:17:03<05:43,  2.66it/s]
 91%|█████████▏| 9769/10682 [1:17:03<05:42,  2.66it/s]
 91%|█████████▏| 9770/10682 [1:17:04<05:42,  2.66it/s]
 91%|█████████▏| 9771/10682 [1:17:04<05:42,  2.66it/s]
 91%|█████████▏| 9772/10682 [1:17:04<05:40,  2.67it/s]
 91%|█████████▏| 9773/10682 [1:17:05<05:40,  2.67it/s]
 91%|█████████▏| 9774/10682 [1:17:05<05:40,  2.67it/s]
 92%|█████████▏| 9775/10682 [1:17:05<05:39,  2.67it/s]
                                                      
{'loss': 4.197, 'grad_norm': 0.34784138202667236, 'learning_rate': 2.1804910798715826e-05, 'epoch': 0.92}
-
 92%|█████████▏| 9775/10682 [1:17:05<05:39,  2.67it/s]
 92%|█████████▏| 9776/10682 [1:17:06<05:39,  2.67it/s]
 92%|█████████▏| 9777/10682 [1:17:06<05:39,  2.67it/s]
 92%|█████████▏| 9778/10682 [1:17:07<05:38,  2.67it/s]
 92%|█████████▏| 9779/10682 [1:17:07<05:38,  2.67it/s]
 92%|█████████▏| 9780/10682 [1:17:07<05:38,  2.66it/s]
 92%|█████████▏| 9781/10682 [1:17:08<05:37,  2.67it/s]
 92%|█████████▏| 9782/10682 [1:17:08<05:37,  2.67it/s]
 92%|█████████▏| 9783/10682 [1:17:08<05:37,  2.67it/s]
 92%|█████████▏| 9784/10682 [1:17:09<05:37,  2.66it/s]
 92%|█████████▏| 9785/10682 [1:17:09<05:36,  2.67it/s]
 92%|█████████▏| 9786/10682 [1:17:10<05:35,  2.67it/s]
 92%|█████████▏| 9787/10682 [1:17:10<05:35,  2.67it/s]
 92%|█████████▏| 9788/10682 [1:17:10<05:34,  2.67it/s]
 92%|█████████▏| 9789/10682 [1:17:11<05:34,  2.67it/s]
 92%|█████████▏| 9790/10682 [1:17:11<05:34,  2.67it/s]
 92%|█████████▏| 9791/10682 [1:17:11<05:34,  2.67it/s]
 92%|█████████▏| 9792/10682 [1:17:12<05:33,  2.66it/s]
 92%|█████████▏| 9793/10682 [1:17:12<05:34,  2.66it/s]
 92%|█████████▏| 9794/10682 [1:17:13<05:33,  2.66it/s]
 92%|█████████▏| 9795/10682 [1:17:13<05:33,  2.66it/s]
 92%|█████████▏| 9796/10682 [1:17:13<05:32,  2.67it/s]
 92%|█████████▏| 9797/10682 [1:17:14<05:31,  2.67it/s]
 92%|█████████▏| 9798/10682 [1:17:14<05:32,  2.66it/s]
 92%|█████████▏| 9799/10682 [1:17:14<05:31,  2.66it/s]
 92%|█████████▏| 9800/10682 [1:17:15<05:31,  2.66it/s]
                                                      {'loss': 4.1965, 'grad_norm': 0.34480077028274536, 'learning_rate': 2.0627663280649135e-05, 'epoch': 0.92}
-
 92%|█████████▏| 9800/10682 [1:17:15<05:31,  2.66it/s]
 92%|█████████▏| 9801/10682 [1:17:15<05:30,  2.66it/s]
 92%|█████████▏| 9802/10682 [1:17:16<05:30,  2.66it/s]
 92%|█████████▏| 9803/10682 [1:17:16<05:29,  2.67it/s]
 92%|█████████▏| 9804/10682 [1:17:16<05:29,  2.66it/s]
 92%|█████████▏| 9805/10682 [1:17:17<05:28,  2.67it/s]
 92%|█████████▏| 9806/10682 [1:17:17<05:28,  2.67it/s]
 92%|█████████▏| 9807/10682 [1:17:17<05:27,  2.67it/s]
 92%|█████████▏| 9808/10682 [1:17:18<05:27,  2.67it/s]
 92%|█████████▏| 9809/10682 [1:17:18<05:27,  2.67it/s]
 92%|█████████▏| 9810/10682 [1:17:19<05:26,  2.67it/s]
 92%|█████████▏| 9811/10682 [1:17:19<05:26,  2.67it/s]
 92%|█████████▏| 9812/10682 [1:17:19<05:25,  2.67it/s]
 92%|█████████▏| 9813/10682 [1:17:20<05:25,  2.67it/s]
 92%|█████████▏| 9814/10682 [1:17:20<05:24,  2.67it/s]
 92%|█████████▏| 9815/10682 [1:17:20<05:24,  2.67it/s]
 92%|█████████▏| 9816/10682 [1:17:21<05:24,  2.67it/s]
 92%|█████████▏| 9817/10682 [1:17:21<05:23,  2.67it/s]
 92%|█████████▏| 9818/10682 [1:17:22<05:24,  2.67it/s]
 92%|█████████▏| 9819/10682 [1:17:22<05:23,  2.67it/s]
 92%|█████████▏| 9820/10682 [1:17:22<05:23,  2.66it/s]
 92%|█████████▏| 9821/10682 [1:17:23<05:23,  2.66it/s]
 92%|█████████▏| 9822/10682 [1:17:23<05:22,  2.66it/s]
 92%|█████████▏| 9823/10682 [1:17:23<05:22,  2.67it/s]
 92%|█████████▏| 9824/10682 [1:17:24<05:21,  2.67it/s]
 92%|█████████▏| 9825/10682 [1:17:24<05:21,  2.67it/s]
                                                      
{'loss': 4.1984, 'grad_norm': 0.3383018970489502, 'learning_rate': 1.9482414469655486e-05, 'epoch': 0.92}
-
 92%|█████████▏| 9825/10682 [1:17:24<05:21,  2.67it/s]
 92%|█████████▏| 9826/10682 [1:17:25<05:21,  2.67it/s]
 92%|█████████▏| 9827/10682 [1:17:25<05:20,  2.67it/s]
 92%|█████████▏| 9828/10682 [1:17:25<05:19,  2.67it/s]
 92%|█████████▏| 9829/10682 [1:17:26<05:19,  2.67it/s]
 92%|█████████▏| 9830/10682 [1:17:26<05:19,  2.67it/s]
 92%|█████████▏| 9831/10682 [1:17:26<05:19,  2.66it/s]
 92%|█████████▏| 9832/10682 [1:17:27<05:18,  2.67it/s]
 92%|█████████▏| 9833/10682 [1:17:27<05:18,  2.66it/s]
 92%|█████████▏| 9834/10682 [1:17:28<05:18,  2.66it/s]
 92%|█████████▏| 9835/10682 [1:17:28<05:18,  2.66it/s]
 92%|█████████▏| 9836/10682 [1:17:28<05:17,  2.66it/s]
 92%|█████████▏| 9837/10682 [1:17:29<05:16,  2.67it/s]
 92%|█████████▏| 9838/10682 [1:17:29<05:16,  2.67it/s]
 92%|█████████▏| 9839/10682 [1:17:29<05:16,  2.67it/s]
 92%|█████████▏| 9840/10682 [1:17:30<05:15,  2.67it/s]
 92%|█████████▏| 9841/10682 [1:17:30<05:15,  2.67it/s]
 92%|█████████▏| 9842/10682 [1:17:31<05:14,  2.67it/s]
 92%|█████████▏| 9843/10682 [1:17:31<05:14,  2.67it/s]
 92%|█████████▏| 9844/10682 [1:17:31<05:14,  2.66it/s]
 92%|█████████▏| 9845/10682 [1:17:32<05:14,  2.66it/s]
 92%|█████████▏| 9846/10682 [1:17:32<05:13,  2.67it/s]
 92%|█████████▏| 9847/10682 [1:17:32<05:13,  2.66it/s]
 92%|█████████▏| 9848/10682 [1:17:33<05:13,  2.66it/s]
 92%|█████████▏| 9849/10682 [1:17:33<05:13,  2.66it/s]
 92%|█████████▏| 9850/10682 [1:17:34<05:12,  2.66it/s]{'loss': 4.2093, 'grad_norm': 0.344695508480072, 'learning_rate': 1.8369240812535104e-05, 'epoch': 0.92}
                                                      
-
 92%|█████████▏| 9850/10682 [1:17:34<05:12,  2.66it/s]
 92%|█████████▏| 9851/10682 [1:17:34<05:12,  2.66it/s]
 92%|█████████▏| 9852/10682 [1:17:34<05:11,  2.66it/s]
 92%|█████████▏| 9853/10682 [1:17:35<05:11,  2.66it/s]
 92%|█████████▏| 9854/10682 [1:17:35<05:10,  2.67it/s]
 92%|█████████▏| 9855/10682 [1:17:35<05:09,  2.67it/s]
 92%|█████████▏| 9856/10682 [1:17:36<05:09,  2.67it/s]
 92%|█████████▏| 9857/10682 [1:17:36<05:09,  2.67it/s]
 92%|█████████▏| 9858/10682 [1:17:37<05:09,  2.66it/s]
 92%|█████████▏| 9859/10682 [1:17:37<05:08,  2.66it/s]
 92%|█████████▏| 9860/10682 [1:17:37<05:09,  2.66it/s]
 92%|█████████▏| 9861/10682 [1:17:38<05:08,  2.66it/s]
 92%|█████████▏| 9862/10682 [1:17:38<05:08,  2.66it/s]
 92%|█████████▏| 9863/10682 [1:17:38<05:07,  2.67it/s]
 92%|█████████▏| 9864/10682 [1:17:39<05:07,  2.66it/s]
 92%|█████████▏| 9865/10682 [1:17:39<05:07,  2.66it/s]
 92%|█████████▏| 9866/10682 [1:17:40<05:06,  2.66it/s]
 92%|█████████▏| 9867/10682 [1:17:40<05:05,  2.67it/s]
 92%|█████████▏| 9868/10682 [1:17:40<05:05,  2.67it/s]
 92%|█████████▏| 9869/10682 [1:17:41<05:04,  2.67it/s]
 92%|█████████▏| 9870/10682 [1:17:41<05:04,  2.67it/s]
 92%|█████████▏| 9871/10682 [1:17:41<05:04,  2.66it/s]
 92%|█████████▏| 9872/10682 [1:17:42<05:03,  2.67it/s]
 92%|█████████▏| 9873/10682 [1:17:42<05:03,  2.66it/s]
 92%|█████████▏| 9874/10682 [1:17:43<05:03,  2.66it/s]
 92%|█████████▏| 9875/10682 [1:17:43<05:03,  2.66it/s]
                                                      
{'loss': 4.2014, 'grad_norm': 0.3453322947025299, 'learning_rate': 1.7288216615031272e-05, 'epoch': 0.92}
-
 92%|█████████▏| 9875/10682 [1:17:43<05:03,  2.66it/s]
 92%|█████████▏| 9876/10682 [1:17:43<05:03,  2.66it/s]
 92%|█████████▏| 9877/10682 [1:17:44<05:02,  2.66it/s]
 92%|█████████▏| 9878/10682 [1:17:44<05:02,  2.66it/s]
 92%|█████████▏| 9879/10682 [1:17:44<05:01,  2.66it/s]
 92%|█████████▏| 9880/10682 [1:17:45<05:00,  2.67it/s]
 93%|█████████▎| 9881/10682 [1:17:45<05:00,  2.67it/s]
 93%|█████████▎| 9882/10682 [1:17:46<04:59,  2.67it/s]
 93%|█████████▎| 9883/10682 [1:17:46<04:59,  2.67it/s]
 93%|█████████▎| 9884/10682 [1:17:46<04:59,  2.67it/s]
 93%|█████████▎| 9885/10682 [1:17:47<04:58,  2.67it/s]
 93%|█████████▎| 9886/10682 [1:17:47<04:58,  2.67it/s]
 93%|█████████▎| 9887/10682 [1:17:47<04:57,  2.67it/s]
 93%|█████████▎| 9888/10682 [1:17:48<04:57,  2.67it/s]
 93%|█████████▎| 9889/10682 [1:17:48<04:57,  2.67it/s]
 93%|█████████▎| 9890/10682 [1:17:49<04:56,  2.67it/s]
 93%|█████████▎| 9891/10682 [1:17:49<04:56,  2.67it/s]
 93%|█████████▎| 9892/10682 [1:17:49<04:56,  2.66it/s]
 93%|█████████▎| 9893/10682 [1:17:50<04:55,  2.67it/s]
 93%|█████████▎| 9894/10682 [1:17:50<04:55,  2.67it/s]
 93%|█████████▎| 9895/10682 [1:17:50<04:54,  2.67it/s]
 93%|█████████▎| 9896/10682 [1:17:51<04:54,  2.67it/s]
 93%|█████████▎| 9897/10682 [1:17:51<04:53,  2.67it/s]
 93%|█████████▎| 9898/10682 [1:17:52<04:53,  2.67it/s]
 93%|█████████▎| 9899/10682 [1:17:52<04:53,  2.67it/s]
 93%|█████████▎| 9900/10682 [1:17:52<04:53,  2.66it/s]
                                                      
{'loss': 4.2049, 'grad_norm': 0.345893532037735, 'learning_rate': 1.6239414036870183e-05, 'epoch': 0.93}
-
 93%|█████████▎| 9900/10682 [1:17:52<04:53,  2.66it/s]
 93%|█████████▎| 9901/10682 [1:17:53<04:53,  2.66it/s]
 93%|█████████▎| 9902/10682 [1:17:53<04:52,  2.66it/s]
 93%|█████████▎| 9903/10682 [1:17:53<04:52,  2.67it/s]
 93%|█████████▎| 9904/10682 [1:17:54<04:51,  2.67it/s]
 93%|█████████▎| 9905/10682 [1:17:54<04:51,  2.67it/s]
 93%|█████████▎| 9906/10682 [1:17:55<04:50,  2.67it/s]
 93%|█████████▎| 9907/10682 [1:17:55<04:50,  2.67it/s]
 93%|█████████▎| 9908/10682 [1:17:55<04:49,  2.67it/s]
 93%|█████████▎| 9909/10682 [1:17:56<04:49,  2.67it/s]
 93%|█████████▎| 9910/10682 [1:17:56<04:49,  2.67it/s]
 93%|█████████▎| 9911/10682 [1:17:56<04:49,  2.66it/s]
 93%|█████████▎| 9912/10682 [1:17:57<04:49,  2.66it/s]
 93%|█████████▎| 9913/10682 [1:17:57<04:48,  2.66it/s]
 93%|█████████▎| 9914/10682 [1:17:58<04:48,  2.66it/s]
 93%|█████████▎| 9915/10682 [1:17:58<04:47,  2.67it/s]
 93%|█████████▎| 9916/10682 [1:17:58<04:47,  2.67it/s]
 93%|█████████▎| 9917/10682 [1:17:59<04:46,  2.67it/s]
 93%|█████████▎| 9918/10682 [1:17:59<04:46,  2.67it/s]
 93%|█████████▎| 9919/10682 [1:17:59<04:46,  2.67it/s]
 93%|█████████▎| 9920/10682 [1:18:00<04:45,  2.67it/s]
 93%|█████████▎| 9921/10682 [1:18:00<04:45,  2.67it/s]
 93%|█████████▎| 9922/10682 [1:18:01<04:44,  2.67it/s]
 93%|█████████▎| 9923/10682 [1:18:01<04:44,  2.67it/s]
 93%|█████████▎| 9924/10682 [1:18:01<04:44,  2.67it/s]
 93%|█████████▎| 9925/10682 [1:18:02<04:43,  2.67it/s]{'loss': 4.2012, 'grad_norm': 0.36059078574180603, 'learning_rate': 1.5222903086944684e-05, 'epoch': 0.93}
                                                      
-
 93%|█████████▎| 9925/10682 [1:18:02<04:43,  2.67it/s]
 93%|█████████▎| 9926/10682 [1:18:02<04:44,  2.66it/s]
 93%|█████████▎| 9927/10682 [1:18:02<04:43,  2.66it/s]
 93%|█████████▎| 9928/10682 [1:18:03<04:43,  2.66it/s]
 93%|█████████▎| 9929/10682 [1:18:03<04:42,  2.67it/s]
 93%|█████████▎| 9930/10682 [1:18:04<04:42,  2.67it/s]
 93%|█████████▎| 9931/10682 [1:18:04<04:41,  2.67it/s]
 93%|█████████▎| 9932/10682 [1:18:04<04:41,  2.67it/s]
 93%|█████████▎| 9933/10682 [1:18:05<04:40,  2.67it/s]
 93%|█████████▎| 9934/10682 [1:18:05<04:39,  2.67it/s]
 93%|█████████▎| 9935/10682 [1:18:05<04:39,  2.67it/s]
 93%|█████████▎| 9936/10682 [1:18:06<04:39,  2.67it/s]
 93%|█████████▎| 9937/10682 [1:18:06<04:38,  2.67it/s]
 93%|█████████▎| 9938/10682 [1:18:07<04:38,  2.67it/s]
 93%|█████████▎| 9939/10682 [1:18:07<04:38,  2.67it/s]
 93%|█████████▎| 9940/10682 [1:18:07<04:38,  2.67it/s]
 93%|█████████▎| 9941/10682 [1:18:08<04:37,  2.67it/s]
 93%|█████████▎| 9942/10682 [1:18:08<04:37,  2.67it/s]
 93%|█████████▎| 9943/10682 [1:18:08<04:37,  2.66it/s]
 93%|█████████▎| 9944/10682 [1:18:09<04:36,  2.67it/s]
 93%|█████████▎| 9945/10682 [1:18:09<04:36,  2.67it/s]
 93%|█████████▎| 9946/10682 [1:18:10<04:35,  2.67it/s]
 93%|█████████▎| 9947/10682 [1:18:10<04:35,  2.67it/s]
 93%|█████████▎| 9948/10682 [1:18:10<04:34,  2.67it/s]
 93%|█████████▎| 9949/10682 [1:18:11<04:34,  2.67it/s]
 93%|█████████▎| 9950/10682 [1:18:11<04:34,  2.67it/s]{'loss': 4.1963, 'grad_norm': 0.34588301181793213, 'learning_rate': 1.4238751618640577e-05, 'epoch': 0.93}
-
                                                      

 93%|█████████▎| 9950/10682 [1:18:11<04:34,  2.67it/s]
 93%|█████████▎| 9951/10682 [1:18:11<04:34,  2.66it/s]
 93%|█████████▎| 9952/10682 [1:18:12<04:33,  2.66it/s]
 93%|█████████▎| 9953/10682 [1:18:12<04:33,  2.66it/s]
 93%|█████████▎| 9954/10682 [1:18:13<04:33,  2.66it/s]
 93%|█████████▎| 9955/10682 [1:18:13<04:32,  2.67it/s]
 93%|█████████▎| 9956/10682 [1:18:13<04:31,  2.67it/s]
 93%|█████████▎| 9957/10682 [1:18:14<04:31,  2.67it/s]
 93%|█████████▎| 9958/10682 [1:18:14<04:31,  2.67it/s]
 93%|█████████▎| 9959/10682 [1:18:14<04:31,  2.66it/s]
 93%|█████████▎| 9960/10682 [1:18:15<04:30,  2.67it/s]
 93%|█████████▎| 9961/10682 [1:18:15<04:30,  2.67it/s]
 93%|█████████▎| 9962/10682 [1:18:16<04:29,  2.67it/s]
 93%|█████████▎| 9963/10682 [1:18:16<04:29,  2.67it/s]
 93%|█████████▎| 9964/10682 [1:18:16<04:29,  2.67it/s]
 93%|█████████▎| 9965/10682 [1:18:17<04:28,  2.67it/s]
 93%|█████████▎| 9966/10682 [1:18:17<04:28,  2.67it/s]
 93%|█████████▎| 9967/10682 [1:18:17<04:28,  2.66it/s]
 93%|█████████▎| 9968/10682 [1:18:18<04:28,  2.66it/s]
 93%|█████████▎| 9969/10682 [1:18:18<04:27,  2.67it/s]
 93%|█████████▎| 9970/10682 [1:18:19<04:26,  2.67it/s]
 93%|█████████▎| 9971/10682 [1:18:19<04:26,  2.67it/s]
 93%|█████████▎| 9972/10682 [1:18:19<04:25,  2.67it/s]
 93%|█████████▎| 9973/10682 [1:18:20<04:25,  2.67it/s]
 93%|█████████▎| 9974/10682 [1:18:20<04:25,  2.67it/s]
 93%|█████████▎| 9975/10682 [1:18:20<04:24,  2.67it/s]{'loss': 4.2001, 'grad_norm': 0.3558426797389984, 'learning_rate': 1.3287025325307511e-05, 'epoch': 0.93}
                                                      
-
 93%|█████████▎| 9975/10682 [1:18:20<04:24,  2.67it/s]
 93%|█████████▎| 9976/10682 [1:18:21<04:25,  2.66it/s]
 93%|█████████▎| 9977/10682 [1:18:21<04:24,  2.66it/s]
 93%|█████████▎| 9978/10682 [1:18:22<04:24,  2.66it/s]
 93%|█████████▎| 9979/10682 [1:18:22<04:23,  2.66it/s]
 93%|█████████▎| 9980/10682 [1:18:22<04:24,  2.66it/s]
 93%|█████████▎| 9981/10682 [1:18:23<04:23,  2.66it/s]
 93%|█████████▎| 9982/10682 [1:18:23<04:23,  2.66it/s]
 93%|█████████▎| 9983/10682 [1:18:23<04:22,  2.66it/s]
 93%|█████████▎| 9984/10682 [1:18:24<04:22,  2.66it/s]
 93%|█████████▎| 9985/10682 [1:18:24<04:21,  2.67it/s]
 93%|█████████▎| 9986/10682 [1:18:25<04:21,  2.66it/s]
 93%|█████████▎| 9987/10682 [1:18:25<04:20,  2.67it/s]
 94%|█████████▎| 9988/10682 [1:18:25<04:20,  2.67it/s]
 94%|█████████▎| 9989/10682 [1:18:26<04:19,  2.67it/s]
 94%|█████████▎| 9990/10682 [1:18:26<04:19,  2.67it/s]
 94%|█████████▎| 9991/10682 [1:18:26<04:19,  2.66it/s]
 94%|█████████▎| 9992/10682 [1:18:27<04:19,  2.66it/s]
 94%|█████████▎| 9993/10682 [1:18:27<04:19,  2.66it/s]
 94%|█████████▎| 9994/10682 [1:18:28<04:19,  2.66it/s]
 94%|█████████▎| 9995/10682 [1:18:28<04:18,  2.66it/s]
 94%|█████████▎| 9996/10682 [1:18:28<04:17,  2.66it/s]
 94%|█████████▎| 9997/10682 [1:18:29<04:17,  2.66it/s]
 94%|█████████▎| 9998/10682 [1:18:29<04:16,  2.66it/s]
 94%|█████████▎| 9999/10682 [1:18:30<04:16,  2.66it/s]
 94%|█████████▎| 10000/10682 [1:18:30<04:15,  2.67it/s]{'loss': 4.1938, 'grad_norm': 0.3428060710430145, 'learning_rate': 1.2367787735873993e-05, 'epoch': 0.94}
                                                       
-
 94%|█████████▎| 10000/10682 [1:18:30<04:15,  2.67it/s]
 94%|█████████▎| 10001/10682 [1:18:30<04:15,  2.66it/s]
 94%|█████████▎| 10002/10682 [1:18:31<04:15,  2.66it/s]
 94%|█████████▎| 10003/10682 [1:18:31<04:14,  2.67it/s]
 94%|█████████▎| 10004/10682 [1:18:31<04:14,  2.67it/s]
 94%|█████████▎| 10005/10682 [1:18:32<04:13,  2.67it/s]
 94%|█████████▎| 10006/10682 [1:18:32<04:13,  2.67it/s]
 94%|█████████▎| 10007/10682 [1:18:33<04:13,  2.67it/s]
 94%|█████████▎| 10008/10682 [1:18:33<04:12,  2.67it/s]
 94%|█████████▎| 10009/10682 [1:18:33<04:12,  2.67it/s]
 94%|█████████▎| 10010/10682 [1:18:34<04:11,  2.67it/s]
 94%|█████████▎| 10011/10682 [1:18:34<04:11,  2.67it/s]
 94%|█████████▎| 10012/10682 [1:18:34<04:11,  2.67it/s]
 94%|█████████▎| 10013/10682 [1:18:35<04:10,  2.67it/s]
 94%|█████████▎| 10014/10682 [1:18:35<04:10,  2.67it/s]
 94%|█████████▍| 10015/10682 [1:18:35<04:09,  2.67it/s]
 94%|█████████▍| 10016/10682 [1:18:36<04:09,  2.67it/s]
 94%|█████████▍| 10017/10682 [1:18:36<04:09,  2.66it/s]
 94%|█████████▍| 10018/10682 [1:18:37<04:09,  2.66it/s]
 94%|█████████▍| 10019/10682 [1:18:37<04:08,  2.67it/s]
 94%|█████████▍| 10020/10682 [1:18:37<04:08,  2.66it/s]
 94%|█████████▍| 10021/10682 [1:18:38<04:08,  2.66it/s]
 94%|█████████▍| 10022/10682 [1:18:38<04:07,  2.66it/s]
 94%|█████████▍| 10023/10682 [1:18:39<04:07,  2.66it/s]
 94%|█████████▍| 10024/10682 [1:18:39<04:07,  2.66it/s]
 94%|█████████▍| 10025/10682 [1:18:39<04:06,  2.66it/s]
                                                       
{'loss': 4.2, 'grad_norm': 0.3434779942035675, 'learning_rate': 1.1481100210606388e-05, 'epoch': 0.94}
-
 94%|█████████▍| 10025/10682 [1:18:39<04:06,  2.66it/s]
 94%|█████████▍| 10026/10682 [1:18:40<04:06,  2.66it/s]
 94%|█████████▍| 10027/10682 [1:18:40<04:06,  2.66it/s]
 94%|█████████▍| 10028/10682 [1:18:40<04:05,  2.66it/s]
 94%|█████████▍| 10029/10682 [1:18:41<04:05,  2.67it/s]
 94%|█████████▍| 10030/10682 [1:18:41<04:04,  2.67it/s]
 94%|█████████▍| 10031/10682 [1:18:42<04:04,  2.66it/s]
 94%|█████████▍| 10032/10682 [1:18:42<04:04,  2.66it/s]
 94%|█████████▍| 10033/10682 [1:18:42<04:04,  2.65it/s]
 94%|█████████▍| 10034/10682 [1:18:43<04:04,  2.65it/s]
 94%|█████████▍| 10035/10682 [1:18:43<04:03,  2.66it/s]
 94%|█████████▍| 10036/10682 [1:18:43<04:03,  2.66it/s]
 94%|█████████▍| 10037/10682 [1:18:44<04:02,  2.66it/s]
 94%|█████████▍| 10038/10682 [1:18:44<04:02,  2.66it/s]
 94%|█████████▍| 10039/10682 [1:18:45<04:01,  2.66it/s]
 94%|█████████▍| 10040/10682 [1:18:45<04:01,  2.66it/s]
 94%|█████████▍| 10041/10682 [1:18:45<04:00,  2.66it/s]
 94%|█████████▍| 10042/10682 [1:18:46<04:00,  2.66it/s]
 94%|█████████▍| 10043/10682 [1:18:46<03:59,  2.67it/s]
 94%|█████████▍| 10044/10682 [1:18:46<03:59,  2.67it/s]
 94%|█████████▍| 10045/10682 [1:18:47<03:58,  2.67it/s]
 94%|█████████▍| 10046/10682 [1:18:47<03:58,  2.66it/s]
 94%|█████████▍| 10047/10682 [1:18:48<03:58,  2.66it/s]
 94%|█████████▍| 10048/10682 [1:18:48<03:57,  2.67it/s]
 94%|█████████▍| 10049/10682 [1:18:48<03:57,  2.67it/s]
 94%|█████████▍| 10050/10682 [1:18:49<03:56,  2.67it/s]
                                                       
{'loss': 4.201, 'grad_norm': 0.34699127078056335, 'learning_rate': 1.0627021937013704e-05, 'epoch': 0.94}
-
 94%|█████████▍| 10050/10682 [1:18:49<03:56,  2.67it/s]
 94%|█████████▍| 10051/10682 [1:18:49<03:56,  2.67it/s]
 94%|█████████▍| 10052/10682 [1:18:49<03:56,  2.67it/s]
 94%|█████████▍| 10053/10682 [1:18:50<03:55,  2.67it/s]
 94%|█████████▍| 10054/10682 [1:18:50<03:55,  2.67it/s]
 94%|█████████▍| 10055/10682 [1:18:51<03:55,  2.67it/s]
 94%|█████████▍| 10056/10682 [1:18:51<03:54,  2.67it/s]
 94%|█████████▍| 10057/10682 [1:18:51<03:54,  2.67it/s]
 94%|█████████▍| 10058/10682 [1:18:52<03:54,  2.66it/s]
 94%|█████████▍| 10059/10682 [1:18:52<03:53,  2.67it/s]
 94%|█████████▍| 10060/10682 [1:18:52<03:53,  2.66it/s]
 94%|█████████▍| 10061/10682 [1:18:53<03:53,  2.66it/s]
 94%|█████████▍| 10062/10682 [1:18:53<03:52,  2.66it/s]
 94%|█████████▍| 10063/10682 [1:18:54<03:52,  2.66it/s]
 94%|█████████▍| 10064/10682 [1:18:54<03:51,  2.67it/s]
 94%|█████████▍| 10065/10682 [1:18:54<03:51,  2.67it/s]
 94%|█████████▍| 10066/10682 [1:18:55<03:50,  2.67it/s]
 94%|█████████▍| 10067/10682 [1:18:55<03:50,  2.67it/s]
 94%|█████████▍| 10068/10682 [1:18:55<03:50,  2.67it/s]
 94%|█████████▍| 10069/10682 [1:18:56<03:49,  2.67it/s]
 94%|█████████▍| 10070/10682 [1:18:56<03:49,  2.67it/s]
 94%|█████████▍| 10071/10682 [1:18:57<03:49,  2.66it/s]
 94%|█████████▍| 10072/10682 [1:18:57<03:49,  2.66it/s]
 94%|█████████▍| 10073/10682 [1:18:57<03:48,  2.66it/s]
 94%|█████████▍| 10074/10682 [1:18:58<03:48,  2.66it/s]
 94%|█████████▍| 10075/10682 [1:18:58<03:48,  2.66it/s]
                                                       {'loss': 4.1963, 'grad_norm': 0.3470246493816376, 'learning_rate': 9.805609925895964e-06, 'epoch': 0.94}
-
 94%|█████████▍| 10075/10682 [1:18:58<03:48,  2.66it/s]
 94%|█████████▍| 10076/10682 [1:18:58<03:47,  2.66it/s]
 94%|█████████▍| 10077/10682 [1:18:59<03:47,  2.66it/s]
 94%|█████████▍| 10078/10682 [1:18:59<03:46,  2.66it/s]
 94%|█████████▍| 10079/10682 [1:19:00<03:46,  2.66it/s]
 94%|█████████▍| 10080/10682 [1:19:00<03:45,  2.67it/s]
 94%|█████████▍| 10081/10682 [1:19:00<03:45,  2.67it/s]
 94%|█████████▍| 10082/10682 [1:19:01<03:45,  2.66it/s]
 94%|█████████▍| 10083/10682 [1:19:01<03:44,  2.67it/s]
 94%|█████████▍| 10084/10682 [1:19:01<03:44,  2.66it/s]
 94%|█████████▍| 10085/10682 [1:19:02<03:43,  2.67it/s]
 94%|█████████▍| 10086/10682 [1:19:02<03:43,  2.66it/s]
 94%|█████████▍| 10087/10682 [1:19:03<03:43,  2.66it/s]
 94%|█████████▍| 10088/10682 [1:19:03<03:43,  2.66it/s]
 94%|█████████▍| 10089/10682 [1:19:03<03:42,  2.66it/s]
 94%|█████████▍| 10090/10682 [1:19:04<03:42,  2.67it/s]
 94%|█████████▍| 10091/10682 [1:19:04<03:41,  2.67it/s]
 94%|█████████▍| 10092/10682 [1:19:04<03:41,  2.67it/s]
 94%|█████████▍| 10093/10682 [1:19:05<03:40,  2.67it/s]
 94%|█████████▍| 10094/10682 [1:19:05<03:40,  2.67it/s]
 95%|█████████▍| 10095/10682 [1:19:06<03:39,  2.67it/s]
 95%|█████████▍| 10096/10682 [1:19:06<03:39,  2.67it/s]
 95%|█████████▍| 10097/10682 [1:19:06<03:39,  2.67it/s]
 95%|█████████▍| 10098/10682 [1:19:07<03:38,  2.67it/s]
 95%|█████████▍| 10099/10682 [1:19:07<03:38,  2.67it/s]
 95%|█████████▍| 10100/10682 [1:19:07<03:38,  2.67it/s]
                                                       
{'loss': 4.1919, 'grad_norm': 0.33153823018074036, 'learning_rate': 9.01691900753926e-06, 'epoch': 0.95}
-
 95%|█████████▍| 10100/10682 [1:19:07<03:38,  2.67it/s]
 95%|█████████▍| 10101/10682 [1:19:08<03:37,  2.67it/s]
 95%|█████████▍| 10102/10682 [1:19:08<03:37,  2.67it/s]
 95%|█████████▍| 10103/10682 [1:19:09<03:37,  2.66it/s]
 95%|█████████▍| 10104/10682 [1:19:09<03:36,  2.67it/s]
 95%|█████████▍| 10105/10682 [1:19:09<03:36,  2.66it/s]
 95%|█████████▍| 10106/10682 [1:19:10<03:36,  2.66it/s]
 95%|█████████▍| 10107/10682 [1:19:10<03:35,  2.67it/s]
 95%|█████████▍| 10108/10682 [1:19:10<03:35,  2.67it/s]
 95%|█████████▍| 10109/10682 [1:19:11<03:34,  2.67it/s]
 95%|█████████▍| 10110/10682 [1:19:11<03:34,  2.67it/s]
 95%|█████████▍| 10111/10682 [1:19:12<03:34,  2.66it/s]
 95%|█████████▍| 10112/10682 [1:19:12<03:33,  2.67it/s]
 95%|█████████▍| 10113/10682 [1:19:12<03:33,  2.67it/s]
 95%|█████████▍| 10114/10682 [1:19:13<03:32,  2.67it/s]
 95%|█████████▍| 10115/10682 [1:19:13<03:32,  2.67it/s]
 95%|█████████▍| 10116/10682 [1:19:13<03:32,  2.67it/s]
 95%|█████████▍| 10117/10682 [1:19:14<03:31,  2.67it/s]
 95%|█████████▍| 10118/10682 [1:19:14<03:31,  2.67it/s]
 95%|█████████▍| 10119/10682 [1:19:15<03:31,  2.67it/s]
 95%|█████████▍| 10120/10682 [1:19:15<03:30,  2.67it/s]
 95%|█████████▍| 10121/10682 [1:19:15<03:30,  2.67it/s]
 95%|█████████▍| 10122/10682 [1:19:16<03:29,  2.67it/s]
 95%|█████████▍| 10123/10682 [1:19:16<03:29,  2.67it/s]
 95%|█████████▍| 10124/10682 [1:19:16<03:29,  2.66it/s]
 95%|█████████▍| 10125/10682 [1:19:17<03:28,  2.67it/s]
                                                       
{'loss': 4.1886, 'grad_norm': 0.3495642840862274, 'learning_rate': 8.261001828055447e-06, 'epoch': 0.95}
-
 95%|█████████▍| 10125/10682 [1:19:17<03:28,  2.67it/s]
 95%|█████████▍| 10126/10682 [1:19:17<03:29,  2.65it/s]
 95%|█████████▍| 10127/10682 [1:19:18<03:29,  2.65it/s]
 95%|█████████▍| 10128/10682 [1:19:18<03:29,  2.65it/s]
 95%|█████████▍| 10129/10682 [1:19:18<03:28,  2.66it/s]
 95%|█████████▍| 10130/10682 [1:19:19<03:27,  2.66it/s]
 95%|█████████▍| 10131/10682 [1:19:19<03:27,  2.66it/s]
 95%|█████████▍| 10132/10682 [1:19:19<03:26,  2.66it/s]
 95%|█████████▍| 10133/10682 [1:19:20<03:25,  2.67it/s]
 95%|█████████▍| 10134/10682 [1:19:20<03:25,  2.67it/s]
 95%|█████████▍| 10135/10682 [1:19:21<03:24,  2.67it/s]
 95%|█████████▍| 10136/10682 [1:19:21<03:24,  2.67it/s]
 95%|█████████▍| 10137/10682 [1:19:21<03:24,  2.67it/s]
 95%|█████████▍| 10138/10682 [1:19:22<03:24,  2.66it/s]
 95%|█████████▍| 10139/10682 [1:19:22<03:23,  2.66it/s]
 95%|█████████▍| 10140/10682 [1:19:22<03:23,  2.66it/s]
 95%|█████████▍| 10141/10682 [1:19:23<03:23,  2.66it/s]
 95%|█████████▍| 10142/10682 [1:19:23<03:22,  2.66it/s]
 95%|█████████▍| 10143/10682 [1:19:24<03:22,  2.66it/s]
 95%|█████████▍| 10144/10682 [1:19:24<03:21,  2.67it/s]
 95%|█████████▍| 10145/10682 [1:19:24<03:21,  2.67it/s]
 95%|█████████▍| 10146/10682 [1:19:25<03:21,  2.67it/s]
 95%|█████████▍| 10147/10682 [1:19:25<03:20,  2.67it/s]
 95%|█████████▌| 10148/10682 [1:19:25<03:20,  2.67it/s]
 95%|█████████▌| 10149/10682 [1:19:26<03:19,  2.67it/s]
 95%|█████████▌| 10150/10682 [1:19:26<03:19,  2.67it/s]{'loss': 4.1984, 'grad_norm': 0.35362622141838074, 'learning_rate': 7.537908845868024e-06, 'epoch': 0.95}
-
                                                       

 95%|█████████▌| 10150/10682 [1:19:26<03:19,  2.67it/s]
 95%|█████████▌| 10151/10682 [1:19:27<03:19,  2.66it/s]
 95%|█████████▌| 10152/10682 [1:19:27<03:18,  2.67it/s]
 95%|█████████▌| 10153/10682 [1:19:27<03:18,  2.66it/s]
 95%|█████████▌| 10154/10682 [1:19:28<03:18,  2.66it/s]
 95%|█████████▌| 10155/10682 [1:19:28<03:18,  2.66it/s]
 95%|█████████▌| 10156/10682 [1:19:28<03:17,  2.66it/s]
 95%|█████████▌| 10157/10682 [1:19:29<03:17,  2.66it/s]
 95%|█████████▌| 10158/10682 [1:19:29<03:16,  2.67it/s]
 95%|█████████▌| 10159/10682 [1:19:30<03:16,  2.67it/s]
 95%|█████████▌| 10160/10682 [1:19:30<03:15,  2.67it/s]
 95%|█████████▌| 10161/10682 [1:19:30<03:15,  2.67it/s]
 95%|█████████▌| 10162/10682 [1:19:31<03:14,  2.67it/s]
 95%|█████████▌| 10163/10682 [1:19:31<03:14,  2.67it/s]
 95%|█████████▌| 10164/10682 [1:19:31<03:14,  2.67it/s]
 95%|█████████▌| 10165/10682 [1:19:32<03:14,  2.66it/s]
 95%|█████████▌| 10166/10682 [1:19:32<03:13,  2.66it/s]
 95%|█████████▌| 10167/10682 [1:19:33<03:13,  2.67it/s]
 95%|█████████▌| 10168/10682 [1:19:33<03:12,  2.66it/s]
 95%|█████████▌| 10169/10682 [1:19:33<03:12,  2.67it/s]
 95%|█████████▌| 10170/10682 [1:19:34<03:12,  2.67it/s]
 95%|█████████▌| 10171/10682 [1:19:34<03:11,  2.66it/s]
 95%|█████████▌| 10172/10682 [1:19:34<03:11,  2.67it/s]
 95%|█████████▌| 10173/10682 [1:19:35<03:10,  2.67it/s]
 95%|█████████▌| 10174/10682 [1:19:35<03:10,  2.67it/s]
 95%|█████████▌| 10175/10682 [1:19:36<03:09,  2.67it/s]{'loss': 4.1965, 'grad_norm': 0.3380189538002014, 'learning_rate': 6.847688328344037e-06, 'epoch': 0.95}
                                                       
-
 95%|█████████▌| 10175/10682 [1:19:36<03:09,  2.67it/s]
 95%|█████████▌| 10176/10682 [1:19:36<03:09,  2.67it/s]
 95%|█████████▌| 10177/10682 [1:19:36<03:09,  2.67it/s]
 95%|█████████▌| 10178/10682 [1:19:37<03:08,  2.67it/s]
 95%|█████████▌| 10179/10682 [1:19:37<03:08,  2.67it/s]
 95%|█████████▌| 10180/10682 [1:19:37<03:08,  2.66it/s]
 95%|█████████▌| 10181/10682 [1:19:38<03:07,  2.67it/s]
 95%|█████████▌| 10182/10682 [1:19:38<03:07,  2.66it/s]
 95%|█████████▌| 10183/10682 [1:19:39<03:07,  2.66it/s]
 95%|█████████▌| 10184/10682 [1:19:39<03:07,  2.66it/s]
 95%|█████████▌| 10185/10682 [1:19:39<03:06,  2.67it/s]
 95%|█████████▌| 10186/10682 [1:19:40<03:05,  2.67it/s]
 95%|█████████▌| 10187/10682 [1:19:40<03:05,  2.67it/s]
 95%|█████████▌| 10188/10682 [1:19:40<03:04,  2.67it/s]
 95%|█████████▌| 10189/10682 [1:19:41<03:04,  2.67it/s]
 95%|█████████▌| 10190/10682 [1:19:41<03:04,  2.67it/s]
 95%|█████████▌| 10191/10682 [1:19:42<03:03,  2.67it/s]
 95%|█████████▌| 10192/10682 [1:19:42<03:03,  2.67it/s]
 95%|█████████▌| 10193/10682 [1:19:42<03:03,  2.67it/s]
 95%|█████████▌| 10194/10682 [1:19:43<03:03,  2.67it/s]
 95%|█████████▌| 10195/10682 [1:19:43<03:03,  2.66it/s]
 95%|█████████▌| 10196/10682 [1:19:43<03:02,  2.66it/s]
 95%|█████████▌| 10197/10682 [1:19:44<03:02,  2.66it/s]
 95%|█████████▌| 10198/10682 [1:19:44<03:01,  2.66it/s]
 95%|█████████▌| 10199/10682 [1:19:45<03:01,  2.66it/s]
 95%|█████████▌| 10200/10682 [1:19:45<03:00,  2.67it/s]{'loss': 4.1982, 'grad_norm': 0.33888542652130127, 'learning_rate': 6.190386348572108e-06, 'epoch': 0.95}
                                                       
-
 95%|█████████▌| 10200/10682 [1:19:45<03:00,  2.67it/s]
 95%|█████████▌| 10201/10682 [1:19:45<03:00,  2.66it/s]
 96%|█████████▌| 10202/10682 [1:19:46<03:00,  2.66it/s]
 96%|█████████▌| 10203/10682 [1:19:46<02:59,  2.67it/s]
 96%|█████████▌| 10204/10682 [1:19:46<02:59,  2.67it/s]
 96%|█████████▌| 10205/10682 [1:19:47<02:58,  2.67it/s]
 96%|█████████▌| 10206/10682 [1:19:47<02:58,  2.67it/s]
 96%|█████████▌| 10207/10682 [1:19:48<02:58,  2.66it/s]
 96%|█████████▌| 10208/10682 [1:19:48<02:57,  2.66it/s]
 96%|█████████▌| 10209/10682 [1:19:48<02:57,  2.67it/s]
 96%|█████████▌| 10210/10682 [1:19:49<02:57,  2.67it/s]
 96%|█████████▌| 10211/10682 [1:19:49<02:56,  2.67it/s]
 96%|█████████▌| 10212/10682 [1:19:49<02:56,  2.67it/s]
 96%|█████████▌| 10213/10682 [1:19:50<02:55,  2.67it/s]
 96%|█████████▌| 10214/10682 [1:19:50<02:55,  2.67it/s]
 96%|█████████▌| 10215/10682 [1:19:51<02:55,  2.66it/s]
 96%|█████████▌| 10216/10682 [1:19:51<02:54,  2.67it/s]
 96%|█████████▌| 10217/10682 [1:19:51<02:54,  2.66it/s]
 96%|█████████▌| 10218/10682 [1:19:52<02:54,  2.67it/s]
 96%|█████████▌| 10219/10682 [1:19:52<02:53,  2.66it/s]
 96%|█████████▌| 10220/10682 [1:19:52<02:53,  2.66it/s]
 96%|█████████▌| 10221/10682 [1:19:53<02:53,  2.66it/s]
 96%|█████████▌| 10222/10682 [1:19:53<02:52,  2.66it/s]
 96%|█████████▌| 10223/10682 [1:19:54<02:52,  2.66it/s]
 96%|█████████▌| 10224/10682 [1:19:54<02:52,  2.66it/s]
 96%|█████████▌| 10225/10682 [1:19:54<02:51,  2.67it/s]{'loss': 4.1952, 'grad_norm': 0.3332981467247009, 'learning_rate': 5.56604678228706e-06, 'epoch': 0.96}
                                                       
-
 96%|█████████▌| 10225/10682 [1:19:54<02:51,  2.67it/s]
 96%|█████████▌| 10226/10682 [1:19:55<02:51,  2.66it/s]
 96%|█████████▌| 10227/10682 [1:19:55<02:50,  2.66it/s]
 96%|█████████▌| 10228/10682 [1:19:55<02:50,  2.66it/s]
 96%|█████████▌| 10229/10682 [1:19:56<02:49,  2.67it/s]
 96%|█████████▌| 10230/10682 [1:19:56<02:49,  2.67it/s]
 96%|█████████▌| 10231/10682 [1:19:57<02:49,  2.66it/s]
 96%|█████████▌| 10232/10682 [1:19:57<02:48,  2.66it/s]
 96%|█████████▌| 10233/10682 [1:19:57<02:48,  2.66it/s]
 96%|█████████▌| 10234/10682 [1:19:58<02:48,  2.65it/s]
 96%|█████████▌| 10235/10682 [1:19:58<02:48,  2.66it/s]
 96%|█████████▌| 10236/10682 [1:19:58<02:47,  2.66it/s]
 96%|█████████▌| 10237/10682 [1:19:59<02:47,  2.66it/s]
 96%|█████████▌| 10238/10682 [1:19:59<02:46,  2.66it/s]
 96%|█████████▌| 10239/10682 [1:20:00<02:46,  2.66it/s]
 96%|█████████▌| 10240/10682 [1:20:00<02:46,  2.66it/s]
 96%|█████████▌| 10241/10682 [1:20:00<02:45,  2.66it/s]
 96%|█████████▌| 10242/10682 [1:20:01<02:45,  2.66it/s]
 96%|█████████▌| 10243/10682 [1:20:01<02:44,  2.66it/s]
 96%|█████████▌| 10244/10682 [1:20:01<02:44,  2.66it/s]
 96%|█████████▌| 10245/10682 [1:20:02<02:44,  2.66it/s]
 96%|█████████▌| 10246/10682 [1:20:02<02:43,  2.66it/s]
 96%|█████████▌| 10247/10682 [1:20:03<02:43,  2.66it/s]
 96%|█████████▌| 10248/10682 [1:20:03<02:43,  2.66it/s]
 96%|█████████▌| 10249/10682 [1:20:03<02:42,  2.66it/s]
 96%|█████████▌| 10250/10682 [1:20:04<02:42,  2.66it/s]
                                                       
{'loss': 4.2015, 'grad_norm': 0.33225977420806885, 'learning_rate': 4.974711304941093e-06, 'epoch': 0.96}
-
 96%|█████████▌| 10250/10682 [1:20:04<02:42,  2.66it/s]
 96%|█████████▌| 10251/10682 [1:20:04<02:42,  2.65it/s]
 96%|█████████▌| 10252/10682 [1:20:04<02:41,  2.66it/s]
 96%|█████████▌| 10253/10682 [1:20:05<02:41,  2.66it/s]
 96%|█████████▌| 10254/10682 [1:20:05<02:40,  2.66it/s]
 96%|█████████▌| 10255/10682 [1:20:06<02:40,  2.66it/s]
 96%|█████████▌| 10256/10682 [1:20:06<02:39,  2.67it/s]
 96%|█████████▌| 10257/10682 [1:20:06<02:39,  2.66it/s]
 96%|█████████▌| 10258/10682 [1:20:07<02:39,  2.66it/s]
 96%|█████████▌| 10259/10682 [1:20:07<02:38,  2.66it/s]
 96%|█████████▌| 10260/10682 [1:20:07<02:38,  2.66it/s]
 96%|█████████▌| 10261/10682 [1:20:08<02:38,  2.66it/s]
 96%|█████████▌| 10262/10682 [1:20:08<02:37,  2.67it/s]
 96%|█████████▌| 10263/10682 [1:20:09<02:37,  2.67it/s]
 96%|█████████▌| 10264/10682 [1:20:09<02:36,  2.66it/s]
 96%|█████████▌| 10265/10682 [1:20:09<02:36,  2.67it/s]
 96%|█████████▌| 10266/10682 [1:20:10<02:36,  2.67it/s]
 96%|█████████▌| 10267/10682 [1:20:10<02:35,  2.67it/s]
 96%|█████████▌| 10268/10682 [1:20:10<02:35,  2.67it/s]
 96%|█████████▌| 10269/10682 [1:20:11<02:34,  2.67it/s]
 96%|█████████▌| 10270/10682 [1:20:11<02:34,  2.67it/s]
 96%|█████████▌| 10271/10682 [1:20:12<02:34,  2.67it/s]
 96%|█████████▌| 10272/10682 [1:20:12<02:33,  2.67it/s]
 96%|█████████▌| 10273/10682 [1:20:12<02:33,  2.67it/s]
 96%|█████████▌| 10274/10682 [1:20:13<02:33,  2.66it/s]
 96%|█████████▌| 10275/10682 [1:20:13<02:32,  2.66it/s]{'loss': 4.2, 'grad_norm': 0.3397231698036194, 'learning_rate': 4.416419388921844e-06, 'epoch': 0.96}
                                                       
-
 96%|█████████▌| 10275/10682 [1:20:13<02:32,  2.66it/s]
 96%|█████████▌| 10276/10682 [1:20:13<02:32,  2.66it/s]
 96%|█████████▌| 10277/10682 [1:20:14<02:32,  2.66it/s]
 96%|█████████▌| 10278/10682 [1:20:14<02:31,  2.66it/s]
 96%|█████████▌| 10279/10682 [1:20:15<02:31,  2.67it/s]
 96%|█████████▌| 10280/10682 [1:20:15<02:30,  2.67it/s]
 96%|█████████▌| 10281/10682 [1:20:15<02:30,  2.67it/s]
 96%|█████████▋| 10282/10682 [1:20:16<02:29,  2.67it/s]
 96%|█████████▋| 10283/10682 [1:20:16<02:29,  2.67it/s]
 96%|█████████▋| 10284/10682 [1:20:16<02:29,  2.67it/s]
 96%|█████████▋| 10285/10682 [1:20:17<02:29,  2.66it/s]
 96%|█████████▋| 10286/10682 [1:20:17<02:28,  2.66it/s]
 96%|█████████▋| 10287/10682 [1:20:18<02:28,  2.67it/s]
 96%|█████████▋| 10288/10682 [1:20:18<02:27,  2.66it/s]
 96%|█████████▋| 10289/10682 [1:20:18<02:27,  2.66it/s]
 96%|█████████▋| 10290/10682 [1:20:19<02:27,  2.67it/s]
 96%|█████████▋| 10291/10682 [1:20:19<02:26,  2.66it/s]
 96%|█████████▋| 10292/10682 [1:20:19<02:26,  2.67it/s]
 96%|█████████▋| 10293/10682 [1:20:20<02:25,  2.67it/s]
 96%|█████████▋| 10294/10682 [1:20:20<02:25,  2.67it/s]
 96%|█████████▋| 10295/10682 [1:20:21<02:25,  2.67it/s]
 96%|█████████▋| 10296/10682 [1:20:21<02:24,  2.67it/s]
 96%|█████████▋| 10297/10682 [1:20:21<02:24,  2.67it/s]
 96%|█████████▋| 10298/10682 [1:20:22<02:24,  2.67it/s]
 96%|█████████▋| 10299/10682 [1:20:22<02:23,  2.67it/s]
 96%|█████████▋| 10300/10682 [1:20:22<02:23,  2.66it/s]
                                                       
{'loss': 4.1987, 'grad_norm': 0.33697494864463806, 'learning_rate': 3.891208300917604e-06, 'epoch': 0.96}
-
 96%|█████████▋| 10300/10682 [1:20:22<02:23,  2.66it/s]
 96%|█████████▋| 10301/10682 [1:20:23<02:23,  2.66it/s]
 96%|█████████▋| 10302/10682 [1:20:23<02:22,  2.66it/s]
 96%|█████████▋| 10303/10682 [1:20:24<02:22,  2.66it/s]
 96%|█████████▋| 10304/10682 [1:20:24<02:22,  2.66it/s]
 96%|█████████▋| 10305/10682 [1:20:24<02:21,  2.66it/s]
 96%|█████████▋| 10306/10682 [1:20:25<02:21,  2.66it/s]
 96%|█████████▋| 10307/10682 [1:20:25<02:20,  2.67it/s]
 96%|█████████▋| 10308/10682 [1:20:25<02:20,  2.66it/s]
 97%|█████████▋| 10309/10682 [1:20:26<02:20,  2.66it/s]
 97%|█████████▋| 10310/10682 [1:20:26<02:19,  2.66it/s]
 97%|█████████▋| 10311/10682 [1:20:27<02:19,  2.67it/s]
 97%|█████████▋| 10312/10682 [1:20:27<02:18,  2.67it/s]
 97%|█████████▋| 10313/10682 [1:20:27<02:18,  2.66it/s]
 97%|█████████▋| 10314/10682 [1:20:28<02:18,  2.66it/s]
 97%|█████████▋| 10315/10682 [1:20:28<02:17,  2.66it/s]
 97%|█████████▋| 10316/10682 [1:20:28<02:17,  2.67it/s]
 97%|█████████▋| 10317/10682 [1:20:29<02:16,  2.67it/s]
 97%|█████████▋| 10318/10682 [1:20:29<02:16,  2.67it/s]
 97%|█████████▋| 10319/10682 [1:20:30<02:15,  2.67it/s]
 97%|█████████▋| 10320/10682 [1:20:30<02:15,  2.67it/s]
 97%|█████████▋| 10321/10682 [1:20:30<02:15,  2.67it/s]
 97%|█████████▋| 10322/10682 [1:20:31<02:14,  2.67it/s]
 97%|█████████▋| 10323/10682 [1:20:31<02:14,  2.67it/s]
 97%|█████████▋| 10324/10682 [1:20:31<02:14,  2.67it/s]
 97%|█████████▋| 10325/10682 [1:20:32<02:13,  2.67it/s]
                                                       
{'loss': 4.2004, 'grad_norm': 0.33611249923706055, 'learning_rate': 3.3991130994299734e-06, 'epoch': 0.97}
-
 97%|█████████▋| 10325/10682 [1:20:32<02:13,  2.67it/s]
 97%|█████████▋| 10326/10682 [1:20:32<02:13,  2.66it/s]
 97%|█████████▋| 10327/10682 [1:20:33<02:13,  2.67it/s]
 97%|█████████▋| 10328/10682 [1:20:33<02:12,  2.67it/s]
 97%|█████████▋| 10329/10682 [1:20:33<02:12,  2.67it/s]
 97%|█████████▋| 10330/10682 [1:20:34<02:41,  2.19it/s]
 97%|█████████▋| 10331/10682 [1:20:34<02:32,  2.31it/s]
 97%|█████████▋| 10332/10682 [1:20:35<02:25,  2.40it/s]
 97%|█████████▋| 10333/10682 [1:20:35<02:20,  2.48it/s]
 97%|█████████▋| 10334/10682 [1:20:35<02:17,  2.53it/s]
 97%|█████████▋| 10335/10682 [1:20:36<02:14,  2.57it/s]
 97%|█████████▋| 10336/10682 [1:20:36<02:12,  2.60it/s]
 97%|█████████▋| 10337/10682 [1:20:37<02:11,  2.61it/s]
 97%|█████████▋| 10338/10682 [1:20:37<02:10,  2.63it/s]
 97%|█████████▋| 10339/10682 [1:20:37<02:10,  2.63it/s]
 97%|█████████▋| 10340/10682 [1:20:38<02:09,  2.64it/s]
 97%|█████████▋| 10341/10682 [1:20:38<02:08,  2.65it/s]
 97%|█████████▋| 10342/10682 [1:20:38<02:07,  2.66it/s]
 97%|█████████▋| 10343/10682 [1:20:39<02:07,  2.66it/s]
 97%|█████████▋| 10344/10682 [1:20:39<02:07,  2.66it/s]
 97%|█████████▋| 10345/10682 [1:20:40<02:06,  2.66it/s]
 97%|█████████▋| 10346/10682 [1:20:40<02:06,  2.66it/s]
 97%|█████████▋| 10347/10682 [1:20:40<02:05,  2.66it/s]
 97%|█████████▋| 10348/10682 [1:20:41<02:05,  2.66it/s]
 97%|█████████▋| 10349/10682 [1:20:41<02:05,  2.66it/s]
 97%|█████████▋| 10350/10682 [1:20:41<02:04,  2.66it/s]{'loss': 4.1946, 'grad_norm': 0.3460325002670288, 'learning_rate': 2.940166632433183e-06, 'epoch': 0.97}
                                                       
-
 97%|█████████▋| 10350/10682 [1:20:41<02:04,  2.66it/s]
 97%|█████████▋| 10351/10682 [1:20:42<02:04,  2.66it/s]
 97%|█████████▋| 10352/10682 [1:20:42<02:03,  2.66it/s]
 97%|█████████▋| 10353/10682 [1:20:43<02:03,  2.66it/s]
 97%|█████████▋| 10354/10682 [1:20:43<02:03,  2.66it/s]
 97%|█████████▋| 10355/10682 [1:20:43<02:02,  2.66it/s]
 97%|█████████▋| 10356/10682 [1:20:44<02:02,  2.66it/s]
 97%|█████████▋| 10357/10682 [1:20:44<02:01,  2.67it/s]
 97%|█████████▋| 10358/10682 [1:20:44<02:01,  2.67it/s]
 97%|█████████▋| 10359/10682 [1:20:45<02:01,  2.67it/s]
 97%|█████████▋| 10360/10682 [1:20:45<02:00,  2.67it/s]
 97%|█████████▋| 10361/10682 [1:20:46<02:00,  2.67it/s]
 97%|█████████▋| 10362/10682 [1:20:46<02:00,  2.67it/s]
 97%|█████████▋| 10363/10682 [1:20:46<01:59,  2.67it/s]
 97%|█████████▋| 10364/10682 [1:20:47<01:59,  2.67it/s]
 97%|█████████▋| 10365/10682 [1:20:47<01:58,  2.67it/s]
 97%|█████████▋| 10366/10682 [1:20:47<01:58,  2.67it/s]
 97%|█████████▋| 10367/10682 [1:20:48<01:58,  2.67it/s]
 97%|█████████▋| 10368/10682 [1:20:48<01:57,  2.67it/s]
 97%|█████████▋| 10369/10682 [1:20:49<01:57,  2.66it/s]
 97%|█████████▋| 10370/10682 [1:20:49<01:57,  2.66it/s]
 97%|█████████▋| 10371/10682 [1:20:49<01:56,  2.66it/s]
 97%|█████████▋| 10372/10682 [1:20:50<01:56,  2.66it/s]
 97%|█████████▋| 10373/10682 [1:20:50<02:21,  2.18it/s]
 97%|█████████▋| 10374/10682 [1:20:51<02:13,  2.31it/s]
 97%|█████████▋| 10375/10682 [1:20:51<02:07,  2.41it/s]{'loss': 4.2, 'grad_norm': 0.33119478821754456, 'learning_rate': 2.5143995351817882e-06, 'epoch': 0.97}
                                                       
-
 97%|█████████▋| 10375/10682 [1:20:51<02:07,  2.41it/s]
 97%|█████████▋| 10376/10682 [1:20:52<02:03,  2.47it/s]
 97%|█████████▋| 10377/10682 [1:20:52<02:00,  2.53it/s]
 97%|█████████▋| 10378/10682 [1:20:52<01:58,  2.57it/s]
 97%|█████████▋| 10379/10682 [1:20:53<01:56,  2.60it/s]
 97%|█████████▋| 10380/10682 [1:20:53<01:55,  2.62it/s]
 97%|█████████▋| 10381/10682 [1:20:53<01:54,  2.63it/s]
 97%|█████████▋| 10382/10682 [1:20:54<01:53,  2.64it/s]
 97%|█████████▋| 10383/10682 [1:20:54<01:52,  2.65it/s]
 97%|█████████▋| 10384/10682 [1:20:55<01:52,  2.65it/s]
 97%|█████████▋| 10385/10682 [1:20:55<01:51,  2.66it/s]
 97%|█████████▋| 10386/10682 [1:20:55<01:51,  2.66it/s]
 97%|█████████▋| 10387/10682 [1:20:56<01:50,  2.66it/s]
 97%|█████████▋| 10388/10682 [1:20:56<01:50,  2.66it/s]
 97%|█████████▋| 10389/10682 [1:20:56<01:50,  2.66it/s]
 97%|█████████▋| 10390/10682 [1:20:57<01:49,  2.66it/s]
 97%|█████████▋| 10391/10682 [1:20:57<01:49,  2.66it/s]
 97%|█████████▋| 10392/10682 [1:20:58<01:48,  2.66it/s]
 97%|█████████▋| 10393/10682 [1:20:58<01:48,  2.66it/s]
 97%|█████████▋| 10394/10682 [1:20:58<01:48,  2.66it/s]
 97%|█████████▋| 10395/10682 [1:20:59<01:47,  2.66it/s]
 97%|█████████▋| 10396/10682 [1:20:59<01:47,  2.66it/s]
 97%|█████████▋| 10397/10682 [1:20:59<01:46,  2.66it/s]
 97%|█████████▋| 10398/10682 [1:21:00<01:46,  2.67it/s]
 97%|█████████▋| 10399/10682 [1:21:00<01:46,  2.67it/s]
 97%|█████████▋| 10400/10682 [1:21:01<01:45,  2.67it/s]
                                                       
{'loss': 4.2054, 'grad_norm': 0.34603437781333923, 'learning_rate': 2.1218402281655835e-06, 'epoch': 0.97}
-
 97%|█████████▋| 10400/10682 [1:21:01<01:45,  2.67it/s]
 97%|█████████▋| 10401/10682 [1:21:01<01:45,  2.67it/s]
 97%|█████████▋| 10402/10682 [1:21:01<01:44,  2.67it/s]
 97%|█████████▋| 10403/10682 [1:21:02<01:44,  2.67it/s]
 97%|█████████▋| 10404/10682 [1:21:02<01:43,  2.67it/s]
 97%|█████████▋| 10405/10682 [1:21:02<01:43,  2.67it/s]
 97%|█████████▋| 10406/10682 [1:21:03<01:43,  2.67it/s]
 97%|█████████▋| 10407/10682 [1:21:03<01:43,  2.66it/s]
 97%|█████████▋| 10408/10682 [1:21:04<01:42,  2.66it/s]
 97%|█████████▋| 10409/10682 [1:21:04<01:42,  2.66it/s]
 97%|█████████▋| 10410/10682 [1:21:04<01:41,  2.67it/s]
 97%|█████████▋| 10411/10682 [1:21:05<01:41,  2.67it/s]
 97%|█████████▋| 10412/10682 [1:21:05<01:41,  2.66it/s]
 97%|█████████▋| 10413/10682 [1:21:05<01:40,  2.67it/s]
 97%|█████████▋| 10414/10682 [1:21:06<01:40,  2.67it/s]
 98%|█████████▊| 10415/10682 [1:21:06<01:40,  2.67it/s]
 98%|█████████▊| 10416/10682 [1:21:07<01:39,  2.67it/s]
 98%|█████████▊| 10417/10682 [1:21:07<01:39,  2.67it/s]
 98%|█████████▊| 10418/10682 [1:21:07<01:38,  2.67it/s]
 98%|█████████▊| 10419/10682 [1:21:08<01:38,  2.67it/s]
 98%|█████████▊| 10420/10682 [1:21:08<01:38,  2.67it/s]
 98%|█████████▊| 10421/10682 [1:21:08<01:37,  2.67it/s]
 98%|█████████▊| 10422/10682 [1:21:09<01:37,  2.67it/s]
 98%|█████████▊| 10423/10682 [1:21:09<01:37,  2.67it/s]
 98%|█████████▊| 10424/10682 [1:21:10<01:36,  2.67it/s]
 98%|█████████▊| 10425/10682 [1:21:10<01:36,  2.67it/s]{'loss': 4.2009, 'grad_norm': 0.3527858555316925, 'learning_rate': 1.7625149152127318e-06, 'epoch': 0.98}
                                                       
-
 98%|█████████▊| 10425/10682 [1:21:10<01:36,  2.67it/s]
 98%|█████████▊| 10426/10682 [1:21:10<01:36,  2.66it/s]
 98%|█████████▊| 10427/10682 [1:21:11<01:35,  2.67it/s]
 98%|█████████▊| 10428/10682 [1:21:11<01:35,  2.66it/s]
 98%|█████████▊| 10429/10682 [1:21:11<01:34,  2.67it/s]
 98%|█████████▊| 10430/10682 [1:21:12<01:34,  2.67it/s]
 98%|█████████▊| 10431/10682 [1:21:12<01:34,  2.67it/s]
 98%|█████████▊| 10432/10682 [1:21:13<01:33,  2.67it/s]
 98%|█████████▊| 10433/10682 [1:21:13<01:33,  2.67it/s]
 98%|█████████▊| 10434/10682 [1:21:13<01:32,  2.67it/s]
 98%|█████████▊| 10435/10682 [1:21:14<01:32,  2.67it/s]
 98%|█████████▊| 10436/10682 [1:21:14<01:32,  2.67it/s]
 98%|█████████▊| 10437/10682 [1:21:14<01:31,  2.67it/s]
 98%|█████████▊| 10438/10682 [1:21:15<01:31,  2.67it/s]
 98%|█████████▊| 10439/10682 [1:21:15<01:30,  2.67it/s]
 98%|█████████▊| 10440/10682 [1:21:16<01:30,  2.67it/s]
 98%|█████████▊| 10441/10682 [1:21:16<01:30,  2.67it/s]
 98%|█████████▊| 10442/10682 [1:21:16<01:29,  2.67it/s]
 98%|█████████▊| 10443/10682 [1:21:17<01:29,  2.67it/s]
 98%|█████████▊| 10444/10682 [1:21:17<01:29,  2.67it/s]
 98%|█████████▊| 10445/10682 [1:21:17<01:28,  2.67it/s]
 98%|█████████▊| 10446/10682 [1:21:18<01:28,  2.67it/s]
 98%|█████████▊| 10447/10682 [1:21:18<01:28,  2.66it/s]
 98%|█████████▊| 10448/10682 [1:21:19<01:27,  2.67it/s]
 98%|█████████▊| 10449/10682 [1:21:19<01:27,  2.66it/s]
 98%|█████████▊| 10450/10682 [1:21:19<01:26,  2.67it/s]
                                                       
{'loss': 4.195, 'grad_norm': 0.3437008857727051, 'learning_rate': 1.4364475817401635e-06, 'epoch': 0.98}
-
 98%|█████████▊| 10450/10682 [1:21:19<01:26,  2.67it/s]
 98%|█████████▊| 10451/10682 [1:21:20<01:26,  2.66it/s]
 98%|█████████▊| 10452/10682 [1:21:20<01:26,  2.66it/s]
 98%|█████████▊| 10453/10682 [1:21:20<01:25,  2.67it/s]
 98%|█████████▊| 10454/10682 [1:21:21<01:25,  2.66it/s]
 98%|█████████▊| 10455/10682 [1:21:21<01:25,  2.66it/s]
 98%|█████████▊| 10456/10682 [1:21:22<01:24,  2.66it/s]
 98%|█████████▊| 10457/10682 [1:21:22<01:24,  2.66it/s]
 98%|█████████▊| 10458/10682 [1:21:22<01:24,  2.66it/s]
 98%|█████████▊| 10459/10682 [1:21:23<01:23,  2.66it/s]
 98%|█████████▊| 10460/10682 [1:21:23<01:23,  2.66it/s]
 98%|█████████▊| 10461/10682 [1:21:23<01:23,  2.65it/s]
 98%|█████████▊| 10462/10682 [1:21:24<01:22,  2.65it/s]
 98%|█████████▊| 10463/10682 [1:21:24<01:22,  2.66it/s]
 98%|█████████▊| 10464/10682 [1:21:25<01:21,  2.66it/s]
 98%|█████████▊| 10465/10682 [1:21:25<01:21,  2.66it/s]
 98%|█████████▊| 10466/10682 [1:21:25<01:21,  2.66it/s]
 98%|█████████▊| 10467/10682 [1:21:26<01:20,  2.67it/s]
 98%|█████████▊| 10468/10682 [1:21:26<01:20,  2.67it/s]
 98%|█████████▊| 10469/10682 [1:21:26<01:19,  2.67it/s]
 98%|█████████▊| 10470/10682 [1:21:27<01:19,  2.67it/s]
 98%|█████████▊| 10471/10682 [1:21:27<01:19,  2.67it/s]
 98%|█████████▊| 10472/10682 [1:21:28<01:18,  2.67it/s]
 98%|█████████▊| 10473/10682 [1:21:28<01:18,  2.67it/s]
 98%|█████████▊| 10474/10682 [1:21:28<01:17,  2.67it/s]
 98%|█████████▊| 10475/10682 [1:21:29<01:17,  2.67it/s]
                                                       
{'loss': 4.2032, 'grad_norm': 0.33473604917526245, 'learning_rate': 1.143659993153079e-06, 'epoch': 0.98}
-
 98%|█████████▊| 10475/10682 [1:21:29<01:17,  2.67it/s]
 98%|█████████▊| 10476/10682 [1:21:29<01:17,  2.66it/s]
 98%|█████████▊| 10477/10682 [1:21:29<01:16,  2.66it/s]
 98%|█████████▊| 10478/10682 [1:21:30<01:16,  2.66it/s]
 98%|█████████▊| 10479/10682 [1:21:30<01:16,  2.67it/s]
 98%|█████████▊| 10480/10682 [1:21:31<01:15,  2.67it/s]
 98%|█████████▊| 10481/10682 [1:21:31<01:15,  2.67it/s]
 98%|█████████▊| 10482/10682 [1:21:31<01:14,  2.67it/s]
 98%|█████████▊| 10483/10682 [1:21:32<01:14,  2.67it/s]
 98%|█████████▊| 10484/10682 [1:21:32<01:14,  2.67it/s]
 98%|█████████▊| 10485/10682 [1:21:32<01:13,  2.67it/s]
 98%|█████████▊| 10486/10682 [1:21:33<01:13,  2.66it/s]
 98%|█████████▊| 10487/10682 [1:21:33<01:13,  2.66it/s]
 98%|█████████▊| 10488/10682 [1:21:34<01:12,  2.66it/s]
 98%|█████████▊| 10489/10682 [1:21:34<01:12,  2.66it/s]
 98%|█████████▊| 10490/10682 [1:21:34<01:12,  2.66it/s]
 98%|█████████▊| 10491/10682 [1:21:35<01:11,  2.67it/s]
 98%|█████████▊| 10492/10682 [1:21:35<01:11,  2.67it/s]
 98%|█████████▊| 10493/10682 [1:21:35<01:10,  2.67it/s]
 98%|█████████▊| 10494/10682 [1:21:36<01:10,  2.67it/s]
 98%|█████████▊| 10495/10682 [1:21:36<01:10,  2.67it/s]
 98%|█████████▊| 10496/10682 [1:21:37<01:09,  2.67it/s]
 98%|█████████▊| 10497/10682 [1:21:37<01:09,  2.67it/s]
 98%|█████████▊| 10498/10682 [1:21:37<01:08,  2.67it/s]
 98%|█████████▊| 10499/10682 [1:21:38<01:08,  2.67it/s]
 98%|█████████▊| 10500/10682 [1:21:38<01:08,  2.67it/s]
                                                       {'loss': 4.196, 'grad_norm': 0.3375387489795685, 'learning_rate': 8.841716933915555e-07, 'epoch': 0.98}
-
 98%|█████████▊| 10500/10682 [1:21:38<01:08,  2.67it/s]
 98%|█████████▊| 10501/10682 [1:21:38<01:07,  2.66it/s]
 98%|█████████▊| 10502/10682 [1:21:39<01:07,  2.67it/s]
 98%|█████████▊| 10503/10682 [1:21:39<01:07,  2.67it/s]
 98%|█████████▊| 10504/10682 [1:21:40<01:07,  2.64it/s]
 98%|█████████▊| 10505/10682 [1:21:40<01:07,  2.63it/s]
 98%|█████████▊| 10506/10682 [1:21:40<01:06,  2.64it/s]
 98%|█████████▊| 10507/10682 [1:21:41<01:06,  2.65it/s]
 98%|█████████▊| 10508/10682 [1:21:41<01:05,  2.65it/s]
 98%|█████████▊| 10509/10682 [1:21:41<01:05,  2.66it/s]
 98%|█████████▊| 10510/10682 [1:21:42<01:04,  2.66it/s]
 98%|█████████▊| 10511/10682 [1:21:42<01:04,  2.65it/s]
 98%|█████████▊| 10512/10682 [1:21:43<01:04,  2.64it/s]
 98%|█████████▊| 10513/10682 [1:21:43<01:03,  2.64it/s]
 98%|█████████▊| 10514/10682 [1:21:43<01:03,  2.65it/s]
 98%|█████████▊| 10515/10682 [1:21:44<01:02,  2.65it/s]
 98%|█████████▊| 10516/10682 [1:21:44<01:02,  2.66it/s]
 98%|█████████▊| 10517/10682 [1:21:44<01:02,  2.66it/s]
 98%|█████████▊| 10518/10682 [1:21:45<01:01,  2.66it/s]
 98%|█████████▊| 10519/10682 [1:21:45<01:01,  2.66it/s]
 98%|█████████▊| 10520/10682 [1:21:46<01:00,  2.67it/s]
 98%|█████████▊| 10521/10682 [1:21:46<01:00,  2.67it/s]
 99%|█████████▊| 10522/10682 [1:21:46<00:59,  2.67it/s]
 99%|█████████▊| 10523/10682 [1:21:47<00:59,  2.66it/s]
 99%|��████████▊| 10524/10682 [1:21:47<00:59,  2.67it/s]
 99%|█████████▊| 10525/10682 [1:21:47<00:58,  2.67it/s]{'loss': 4.1949, 'grad_norm': 0.3332563042640686, 'learning_rate': 6.580000036264244e-07, 'epoch': 0.99}
-
                                                       

 99%|█████████▊| 10525/10682 [1:21:47<00:58,  2.67it/s]
 99%|█████████▊| 10526/10682 [1:21:48<00:58,  2.66it/s]
 99%|█████████▊| 10527/10682 [1:21:48<00:58,  2.66it/s]
 99%|█████████▊| 10528/10682 [1:21:49<00:57,  2.66it/s]
 99%|█████████▊| 10529/10682 [1:21:49<00:57,  2.66it/s]
 99%|█████████▊| 10530/10682 [1:21:49<00:57,  2.66it/s]
 99%|█████████▊| 10531/10682 [1:21:50<00:56,  2.66it/s]
 99%|█████████▊| 10532/10682 [1:21:50<00:56,  2.67it/s]
 99%|█████████▊| 10533/10682 [1:21:50<00:55,  2.67it/s]
 99%|█████████▊| 10534/10682 [1:21:51<00:55,  2.67it/s]
 99%|█████████▊| 10535/10682 [1:21:51<00:55,  2.67it/s]
 99%|█████████▊| 10536/10682 [1:21:52<00:54,  2.67it/s]
 99%|█████████▊| 10537/10682 [1:21:52<00:54,  2.67it/s]
 99%|█████████▊| 10538/10682 [1:21:52<00:54,  2.67it/s]
 99%|█████████▊| 10539/10682 [1:21:53<00:53,  2.66it/s]
 99%|█████████▊| 10540/10682 [1:21:53<00:53,  2.66it/s]
 99%|█████████▊| 10541/10682 [1:21:53<00:52,  2.66it/s]
 99%|█████████▊| 10542/10682 [1:21:54<00:52,  2.66it/s]
 99%|█████████▊| 10543/10682 [1:21:54<00:52,  2.67it/s]
 99%|█████████▊| 10544/10682 [1:21:55<00:51,  2.67it/s]
 99%|█████████▊| 10545/10682 [1:21:55<00:51,  2.67it/s]
 99%|█████████▊| 10546/10682 [1:21:55<00:50,  2.67it/s]
 99%|█████████▊| 10547/10682 [1:21:56<00:50,  2.67it/s]
 99%|█████████▊| 10548/10682 [1:21:56<00:50,  2.67it/s]
 99%|█████████▉| 10549/10682 [1:21:56<00:49,  2.67it/s]
 99%|█████████▉| 10550/10682 [1:21:57<00:49,  2.67it/s]
                                                       
{'loss': 4.1947, 'grad_norm': 0.33778882026672363, 'learning_rate': 4.651600211027507e-07, 'epoch': 0.99}
-
 99%|█████████▉| 10550/10682 [1:21:57<00:49,  2.67it/s]
 99%|█████████▉| 10551/10682 [1:21:57<00:49,  2.67it/s]
 99%|█████████▉| 10552/10682 [1:21:58<00:48,  2.67it/s]
 99%|█████████▉| 10553/10682 [1:21:58<00:48,  2.67it/s]
 99%|█████████▉| 10554/10682 [1:21:58<00:47,  2.67it/s]
 99%|█████████▉| 10555/10682 [1:21:59<00:47,  2.67it/s]
 99%|█████████▉| 10556/10682 [1:21:59<00:47,  2.67it/s]
 99%|█████████▉| 10557/10682 [1:21:59<00:46,  2.67it/s]
 99%|█████████▉| 10558/10682 [1:22:00<00:46,  2.67it/s]
 99%|█████████▉| 10559/10682 [1:22:00<00:46,  2.67it/s]
 99%|█████████▉| 10560/10682 [1:22:01<00:45,  2.67it/s]
 99%|█████████▉| 10561/10682 [1:22:01<00:45,  2.67it/s]
 99%|█████████▉| 10562/10682 [1:22:01<00:44,  2.67it/s]
 99%|█████████▉| 10563/10682 [1:22:02<00:44,  2.67it/s]
 99%|█████████▉| 10564/10682 [1:22:02<00:44,  2.67it/s]
 99%|█████████▉| 10565/10682 [1:22:02<00:43,  2.67it/s]
 99%|█████████▉| 10566/10682 [1:22:03<00:43,  2.67it/s]
 99%|█████████▉| 10567/10682 [1:22:03<00:43,  2.66it/s]
 99%|█████████▉| 10568/10682 [1:22:04<00:42,  2.67it/s]
 99%|█████████▉| 10569/10682 [1:22:04<00:42,  2.67it/s]
 99%|█████████▉| 10570/10682 [1:22:04<00:41,  2.67it/s]
 99%|█████████▉| 10571/10682 [1:22:05<00:41,  2.67it/s]
 99%|█████████▉| 10572/10682 [1:22:05<00:41,  2.67it/s]
 99%|█████████▉| 10573/10682 [1:22:05<00:40,  2.67it/s]
 99%|█████████▉| 10574/10682 [1:22:06<00:40,  2.67it/s]
 99%|█████████▉| 10575/10682 [1:22:06<00:40,  2.67it/s]
                                                       {'loss': 4.1922, 'grad_norm': 0.3340144157409668, 'learning_rate': 3.0566461813213986e-07, 'epoch': 0.99}
-
 99%|█████████▉| 10575/10682 [1:22:06<00:40,  2.67it/s]
 99%|█████████▉| 10576/10682 [1:22:07<00:39,  2.67it/s]
 99%|█████████▉| 10577/10682 [1:22:07<00:39,  2.67it/s]
 99%|█████████▉| 10578/10682 [1:22:07<00:39,  2.66it/s]
 99%|█████████▉| 10579/10682 [1:22:08<00:38,  2.66it/s]
 99%|█████████▉| 10580/10682 [1:22:08<00:38,  2.66it/s]
 99%|█████████▉| 10581/10682 [1:22:08<00:37,  2.66it/s]
 99%|█████████▉| 10582/10682 [1:22:09<00:37,  2.66it/s]
 99%|█████████▉| 10583/10682 [1:22:09<00:37,  2.66it/s]
 99%|█████████▉| 10584/10682 [1:22:10<00:36,  2.67it/s]
 99%|█████████▉| 10585/10682 [1:22:10<00:36,  2.67it/s]
 99%|█████████▉| 10586/10682 [1:22:10<00:35,  2.67it/s]
 99%|█████████▉| 10587/10682 [1:22:11<00:35,  2.67it/s]
 99%|█████████▉| 10588/10682 [1:22:11<00:35,  2.67it/s]
 99%|█████████▉| 10589/10682 [1:22:11<00:34,  2.67it/s]
 99%|█████████▉| 10590/10682 [1:22:12<00:34,  2.66it/s]
 99%|█████████▉| 10591/10682 [1:22:12<00:34,  2.66it/s]
 99%|█████████▉| 10592/10682 [1:22:13<00:33,  2.66it/s]
 99%|█████████▉| 10593/10682 [1:22:13<00:33,  2.66it/s]
 99%|█████████▉| 10594/10682 [1:22:13<00:33,  2.67it/s]
 99%|█████████▉| 10595/10682 [1:22:14<00:32,  2.67it/s]
 99%|█████████▉| 10596/10682 [1:22:14<00:32,  2.67it/s]
 99%|█████████▉| 10597/10682 [1:22:14<00:31,  2.67it/s]
 99%|█████████▉| 10598/10682 [1:22:15<00:31,  2.67it/s]
 99%|█████████▉| 10599/10682 [1:22:15<00:31,  2.67it/s]
 99%|█████████▉| 10600/10682 [1:22:16<00:30,  2.67it/s]
                                                       
{'loss': 4.1949, 'grad_norm': 0.34571900963783264, 'learning_rate': 1.7952444123359167e-07, 'epoch': 0.99}
-
 99%|█████████▉| 10600/10682 [1:22:16<00:30,  2.67it/s]
 99%|█████████▉| 10601/10682 [1:22:16<00:30,  2.67it/s]
 99%|█████████▉| 10602/10682 [1:22:16<00:29,  2.67it/s]
 99%|█████████▉| 10603/10682 [1:22:17<00:29,  2.67it/s]
 99%|█████████▉| 10604/10682 [1:22:17<00:29,  2.67it/s]
 99%|█████████▉| 10605/10682 [1:22:17<00:28,  2.67it/s]
 99%|█████████▉| 10606/10682 [1:22:18<00:28,  2.67it/s]
 99%|█████████▉| 10607/10682 [1:22:18<00:28,  2.66it/s]
 99%|█████████▉| 10608/10682 [1:22:19<00:27,  2.67it/s]
 99%|█████████▉| 10609/10682 [1:22:19<00:27,  2.67it/s]
 99%|█████████▉| 10610/10682 [1:22:19<00:26,  2.67it/s]
 99%|█████████▉| 10611/10682 [1:22:20<00:26,  2.67it/s]
 99%|█████████▉| 10612/10682 [1:22:20<00:26,  2.67it/s]
 99%|█████████▉| 10613/10682 [1:22:20<00:25,  2.67it/s]
 99%|█████████▉| 10614/10682 [1:22:21<00:25,  2.67it/s]
 99%|█████████▉| 10615/10682 [1:22:21<00:25,  2.67it/s]
 99%|█████████▉| 10616/10682 [1:22:22<00:24,  2.67it/s]
 99%|█████████▉| 10617/10682 [1:22:22<00:24,  2.67it/s]
 99%|█████████▉| 10618/10682 [1:22:22<00:24,  2.66it/s]
 99%|█████████▉| 10619/10682 [1:22:23<00:23,  2.66it/s]
 99%|█████████▉| 10620/10682 [1:22:23<00:23,  2.66it/s]
 99%|█████████▉| 10621/10682 [1:22:23<00:22,  2.66it/s]
 99%|█████████▉| 10622/10682 [1:22:24<00:22,  2.66it/s]
 99%|█████████▉| 10623/10682 [1:22:24<00:22,  2.67it/s]
 99%|█████████▉| 10624/10682 [1:22:25<00:21,  2.66it/s]
 99%|█████████▉| 10625/10682 [1:22:25<00:21,  2.66it/s]
                                                       {'loss': 4.1946, 'grad_norm': 0.3399973213672638, 'learning_rate': 8.674791042273533e-08, 'epoch': 0.99}
-
 99%|█████████▉| 10625/10682 [1:22:25<00:21,  2.66it/s]
 99%|█████████▉| 10626/10682 [1:22:25<00:21,  2.66it/s]
 99%|█████████▉| 10627/10682 [1:22:26<00:20,  2.67it/s]
 99%|█████████▉| 10628/10682 [1:22:26<00:20,  2.67it/s]
100%|█████████▉| 10629/10682 [1:22:26<00:19,  2.67it/s]
100%|█████████▉| 10630/10682 [1:22:27<00:19,  2.67it/s]
100%|█████████▉| 10631/10682 [1:22:27<00:19,  2.66it/s]
100%|█████████▉| 10632/10682 [1:22:28<00:18,  2.66it/s]
100%|█████████▉| 10633/10682 [1:22:28<00:18,  2.66it/s]
100%|█████████▉| 10634/10682 [1:22:28<00:18,  2.66it/s]
100%|█████████▉| 10635/10682 [1:22:29<00:17,  2.66it/s]
100%|█████████▉| 10636/10682 [1:22:29<00:17,  2.67it/s]
100%|█████████▉| 10637/10682 [1:22:29<00:16,  2.67it/s]
100%|█████████▉| 10638/10682 [1:22:30<00:16,  2.67it/s]
100%|█████████▉| 10639/10682 [1:22:30<00:16,  2.67it/s]
100%|█████████▉| 10640/10682 [1:22:31<00:15,  2.65it/s]
100%|█████████▉| 10641/10682 [1:22:31<00:15,  2.65it/s]
100%|█████████▉| 10642/10682 [1:22:31<00:15,  2.66it/s]
100%|█████████▉| 10643/10682 [1:22:32<00:14,  2.66it/s]
100%|█████████▉| 10644/10682 [1:22:32<00:14,  2.66it/s]
100%|█████████▉| 10645/10682 [1:22:32<00:13,  2.66it/s]
100%|█████████▉| 10646/10682 [1:22:33<00:13,  2.67it/s]
100%|█████████▉| 10647/10682 [1:22:33<00:13,  2.67it/s]
100%|█████████▉| 10648/10682 [1:22:34<00:12,  2.66it/s]
100%|█████████▉| 10649/10682 [1:22:34<00:12,  2.67it/s]
100%|█████████▉| 10650/10682 [1:22:34<00:11,  2.67it/s]
                                                       
{'loss': 4.1963, 'grad_norm': 0.34263646602630615, 'learning_rate': 2.7341218649834522e-08, 'epoch': 1.0}
-
100%|█████████▉| 10650/10682 [1:22:34<00:11,  2.67it/s]
100%|█████████▉| 10651/10682 [1:22:35<00:11,  2.66it/s]
100%|█████████▉| 10652/10682 [1:22:35<00:11,  2.67it/s]
100%|█████████▉| 10653/10682 [1:22:35<00:10,  2.67it/s]
100%|█████████▉| 10654/10682 [1:22:36<00:10,  2.67it/s]
100%|█████████▉| 10655/10682 [1:22:36<00:10,  2.67it/s]
100%|█████████▉| 10656/10682 [1:22:37<00:09,  2.67it/s]
100%|█████████▉| 10657/10682 [1:22:37<00:09,  2.67it/s]
100%|█████████▉| 10658/10682 [1:22:37<00:09,  2.66it/s]
100%|█████████▉| 10659/10682 [1:22:38<00:08,  2.66it/s]
100%|█████████▉| 10660/10682 [1:22:38<00:08,  2.66it/s]
100%|█████████▉| 10661/10682 [1:22:38<00:07,  2.66it/s]
100%|█████████▉| 10662/10682 [1:22:39<00:07,  2.66it/s]
100%|█████████▉| 10663/10682 [1:22:39<00:07,  2.66it/s]
100%|█████████▉| 10664/10682 [1:22:40<00:06,  2.66it/s]
100%|█████████▉| 10665/10682 [1:22:40<00:06,  2.66it/s]
100%|█████████▉| 10666/10682 [1:22:40<00:05,  2.67it/s]
100%|█████████▉| 10667/10682 [1:22:41<00:05,  2.67it/s]
100%|█████████▉| 10668/10682 [1:22:41<00:05,  2.67it/s]
100%|█████████▉| 10669/10682 [1:22:41<00:04,  2.67it/s]
100%|█████████▉| 10670/10682 [1:22:42<00:04,  2.67it/s]
100%|█████████▉| 10671/10682 [1:22:42<00:04,  2.67it/s]
100%|█████████▉| 10672/10682 [1:22:43<00:03,  2.67it/s]
100%|█████████▉| 10673/10682 [1:22:43<00:03,  2.67it/s]
100%|█████████▉| 10674/10682 [1:22:43<00:03,  2.67it/s]
100%|█████████▉| 10675/10682 [1:22:44<00:02,  2.67it/s]
                                                       {'loss': 4.2034, 'grad_norm': 0.33637920022010803, 'learning_rate': 1.3083313863404555e-09, 'epoch': 1.0}
-
100%|█████████▉| 10675/10682 [1:22:44<00:02,  2.67it/s]
100%|█████████▉| 10676/10682 [1:22:44<00:02,  2.66it/s]
100%|█████████▉| 10677/10682 [1:22:44<00:01,  2.66it/s]
100%|█████████▉| 10678/10682 [1:22:45<00:01,  2.66it/s]
100%|█████████▉| 10679/10682 [1:22:45<00:01,  2.67it/s]
100%|█████████▉| 10680/10682 [1:22:46<00:00,  2.67it/s]
100%|█████████▉| 10681/10682 [1:22:46<00:00,  2.67it/s]
100%|██████████| 10682/10682 [1:22:47<00:00,  1.61it/s]
                                                       
{'train_runtime': 4970.0721, 'train_samples_per_second': 2200.823, 'train_steps_per_second': 2.149, 'train_loss': 4.595029320258859, 'epoch': 1.0}
-
100%|██████████| 10682/10682 [1:22:49<00:00,  1.61it/s]
100%|██████████| 10682/10682 [1:22:50<00:00,  2.15it/s]
-Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+05/19/2024 10:14:33 - INFO - __main__ - Script parameters ScriptArguments(seed=1, dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_more_data/piqa', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/14m_llm_seeds_more_data/pythia-14m_piqa_1', output_hub_id='pythia-14m_piqa', hf_hub_token=True, model_id='EleutherAI/pythia-14m', per_device_train_batch_size=320, num_train_epochs=0.2, learning_rate=0.006, gradient_accumulation_steps=1, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
+
  0%|          | 0/6837 [00:00<?, ?it/s][rank0]:[W reducer.cpp:1360] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
+
  0%|          | 1/6837 [01:08<129:23:49, 68.14s/it]
  0%|          | 2/6837 [01:39<87:46:13, 46.23s/it] 
  0%|          | 3/6837 [02:06<71:42:52, 37.78s/it]
  0%|          | 4/6837 [02:32<62:48:12, 33.09s/it]
  0%|          | 5/6837 [02:55<55:57:33, 29.49s/it]
  0%|          | 6/6837 [03:17<50:42:29, 26.72s/it]
  0%|          | 7/6837 [03:36<46:26:58, 24.48s/it]
  0%|          | 8/6837 [03:55<43:03:31, 22.70s/it]
  0%|          | 9/6837 [04:12<39:22:21, 20.76s/it]
  0%|          | 10/6837 [04:26<35:23:26, 18.66s/it]
  0%|          | 11/6837 [04:40<32:38:25, 17.21s/it]
  0%|          | 12/6837 [04:50<28:47:03, 15.18s/it]
  0%|          | 13/6837 [05:00<25:21:22, 13.38s/it]
  0%|          | 14/6837 [05:09<22:59:23, 12.13s/it]
  0%|          | 15/6837 [05:18<21:07:35, 11.15s/it]
  0%|          | 16/6837 [05:25<18:51:12,  9.95s/it]
  0%|          | 17/6837 [05:33<17:59:00,  9.49s/it]
  0%|          | 18/6837 [05:41<17:02:25,  9.00s/it]
  0%|          | 19/6837 [05:49<16:31:49,  8.73s/it]
  0%|          | 20/6837 [05:56<15:40:54,  8.28s/it]
  0%|          | 21/6837 [06:03<14:46:10,  7.80s/it]
  0%|          | 22/6837 [06:10<14:10:14,  7.49s/it]
  0%|          | 23/6837 [06:16<13:27:08,  7.11s/it]
  0%|          | 24/6837 [06:22<12:42:35,  6.72s/it]
  0%|          | 25/6837 [06:29<12:42:48,  6.72s/it]{'loss': 10.653, 'grad_norm': 0.7899429202079773, 'learning_rate': 0.0002192982456140351, 'epoch': 0.0}
                                                    
+
  0%|          | 25/6837 [06:29<12:42:48,  6.72s/it]
  0%|          | 26/6837 [06:35<12:16:43,  6.49s/it]
  0%|          | 27/6837 [06:40<11:48:16,  6.24s/it]
  0%|          | 28/6837 [06:46<11:29:36,  6.08s/it]
  0%|          | 29/6837 [06:51<11:09:11,  5.90s/it]
  0%|          | 30/6837 [06:58<11:17:46,  5.97s/it]
  0%|          | 31/6837 [07:03<10:44:30,  5.68s/it]
  0%|          | 32/6837 [07:08<10:23:08,  5.49s/it]
  0%|          | 33/6837 [07:13<10:22:21,  5.49s/it]
  0%|          | 34/6837 [07:19<10:29:36,  5.55s/it]
  1%|          | 35/6837 [07:24<10:05:12,  5.34s/it]
  1%|          | 36/6837 [07:28<9:30:12,  5.03s/it] 
  1%|          | 37/6837 [07:33<9:15:40,  4.90s/it]
  1%|          | 38/6837 [07:37<9:03:14,  4.79s/it]
  1%|          | 39/6837 [07:41<8:49:21,  4.67s/it]
  1%|          | 40/6837 [07:46<8:44:22,  4.63s/it]
  1%|          | 41/6837 [07:50<8:30:34,  4.51s/it]
  1%|          | 42/6837 [07:54<8:16:46,  4.39s/it]
  1%|          | 43/6837 [07:58<8:06:51,  4.30s/it]
  1%|          | 44/6837 [08:03<8:09:20,  4.32s/it]
  1%|          | 45/6837 [08:07<7:54:40,  4.19s/it]
  1%|          | 46/6837 [08:10<7:38:20,  4.05s/it]
  1%|          | 47/6837 [08:15<7:41:16,  4.08s/it]
  1%|          | 48/6837 [08:18<7:36:05,  4.03s/it]
  1%|          | 49/6837 [08:22<7:24:40,  3.93s/it]
  1%|          | 50/6837 [08:26<7:14:32,  3.84s/it]
                                                   
{'loss': 9.8835, 'grad_norm': 0.8061347603797913, 'learning_rate': 0.0004385964912280702, 'epoch': 0.0}
+
  1%|          | 50/6837 [08:26<7:14:32,  3.84s/it]
  1%|          | 51/6837 [08:30<7:22:57,  3.92s/it]
  1%|          | 52/6837 [08:34<7:17:33,  3.87s/it]
  1%|          | 53/6837 [08:37<6:53:09,  3.65s/it]
  1%|          | 54/6837 [08:40<6:47:14,  3.60s/it]
  1%|          | 55/6837 [08:43<6:20:21,  3.37s/it]
  1%|          | 56/6837 [08:46<6:19:22,  3.36s/it]
  1%|          | 57/6837 [08:50<6:16:07,  3.33s/it]
  1%|          | 58/6837 [08:53<6:10:39,  3.28s/it]
  1%|          | 59/6837 [08:56<6:05:20,  3.23s/it]
  1%|          | 60/6837 [08:59<6:14:56,  3.32s/it]
  1%|          | 61/6837 [09:02<6:03:10,  3.22s/it]
  1%|          | 62/6837 [09:05<5:52:00,  3.12s/it]
  1%|          | 63/6837 [09:08<5:45:40,  3.06s/it]
  1%|          | 64/6837 [09:11<5:38:51,  3.00s/it]
  1%|          | 65/6837 [09:14<5:28:02,  2.91s/it]
  1%|          | 66/6837 [09:17<5:32:27,  2.95s/it]
  1%|          | 67/6837 [09:20<5:26:00,  2.89s/it]
  1%|          | 68/6837 [09:22<5:20:18,  2.84s/it]
  1%|          | 69/6837 [09:25<5:02:51,  2.68s/it]
  1%|          | 70/6837 [09:27<4:58:18,  2.64s/it]
  1%|          | 71/6837 [09:30<4:50:51,  2.58s/it]
  1%|          | 72/6837 [09:32<4:40:16,  2.49s/it]
  1%|          | 73/6837 [09:34<4:34:43,  2.44s/it]
  1%|          | 74/6837 [09:37<4:38:44,  2.47s/it]
  1%|          | 75/6837 [09:39<4:36:38,  2.45s/it]
                                                   
{'loss': 8.7773, 'grad_norm': 0.7474189400672913, 'learning_rate': 0.0006578947368421052, 'epoch': 0.0}
+
  1%|          | 75/6837 [09:39<4:36:38,  2.45s/it]
  1%|          | 76/6837 [09:42<4:33:24,  2.43s/it]
  1%|          | 77/6837 [09:44<4:37:18,  2.46s/it]
  1%|          | 78/6837 [09:47<4:35:53,  2.45s/it]
  1%|          | 79/6837 [09:49<4:36:54,  2.46s/it]
  1%|          | 80/6837 [09:51<4:26:39,  2.37s/it]
  1%|          | 81/6837 [09:53<4:12:55,  2.25s/it]
  1%|          | 82/6837 [09:55<4:01:50,  2.15s/it]
  1%|          | 83/6837 [09:57<4:05:51,  2.18s/it]
  1%|          | 84/6837 [10:00<4:06:35,  2.19s/it]
  1%|          | 85/6837 [10:02<4:00:04,  2.13s/it]
  1%|▏         | 86/6837 [10:03<3:53:27,  2.07s/it]
  1%|▏         | 87/6837 [10:05<3:47:15,  2.02s/it]
  1%|▏         | 88/6837 [10:07<3:44:00,  1.99s/it]
  1%|▏         | 89/6837 [10:09<3:43:06,  1.98s/it]
  1%|▏         | 90/6837 [10:11<3:35:40,  1.92s/it]
  1%|▏         | 91/6837 [10:13<3:28:25,  1.85s/it]
  1%|▏         | 92/6837 [10:15<3:30:41,  1.87s/it]
  1%|▏         | 93/6837 [10:16<3:16:15,  1.75s/it]
  1%|▏         | 94/6837 [10:18<3:18:00,  1.76s/it]
  1%|▏         | 95/6837 [10:20<3:13:51,  1.73s/it]
  1%|▏         | 96/6837 [10:21<3:11:36,  1.71s/it]
  1%|▏         | 97/6837 [10:23<3:04:06,  1.64s/it]
  1%|▏         | 98/6837 [10:24<3:03:14,  1.63s/it]
  1%|▏         | 99/6837 [10:26<2:58:25,  1.59s/it]
  1%|▏         | 100/6837 [10:27<2:49:30,  1.51s/it]
                                                    
{'loss': 7.9608, 'grad_norm': 0.5506234169006348, 'learning_rate': 0.0008771929824561404, 'epoch': 0.0}
+
  1%|▏         | 100/6837 [10:27<2:49:30,  1.51s/it]
  1%|▏         | 101/6837 [10:29<2:52:51,  1.54s/it]
  1%|▏         | 102/6837 [10:30<2:55:52,  1.57s/it]
  2%|▏         | 103/6837 [10:32<2:52:17,  1.54s/it]
  2%|▏         | 104/6837 [10:33<2:48:01,  1.50s/it]
  2%|▏         | 105/6837 [10:35<2:54:19,  1.55s/it]
  2%|▏         | 106/6837 [10:36<2:55:29,  1.56s/it]
  2%|▏         | 107/6837 [10:38<2:51:47,  1.53s/it]
  2%|▏         | 108/6837 [10:40<2:56:07,  1.57s/it]
  2%|▏         | 109/6837 [10:41<3:01:39,  1.62s/it]
  2%|▏         | 110/6837 [10:43<2:57:56,  1.59s/it]
  2%|▏         | 111/6837 [10:44<2:54:21,  1.56s/it]
  2%|▏         | 112/6837 [10:46<2:49:28,  1.51s/it]
  2%|▏         | 113/6837 [10:47<2:40:42,  1.43s/it]
  2%|▏         | 114/6837 [10:48<2:40:02,  1.43s/it]
  2%|▏         | 115/6837 [10:50<2:46:40,  1.49s/it]
  2%|▏         | 116/6837 [10:52<2:49:48,  1.52s/it]
  2%|▏         | 117/6837 [10:53<2:48:51,  1.51s/it]
  2%|▏         | 118/6837 [10:55<2:46:19,  1.49s/it]
  2%|▏         | 119/6837 [10:56<2:38:12,  1.41s/it]
  2%|▏         | 120/6837 [10:57<2:34:52,  1.38s/it]
  2%|▏         | 121/6837 [10:58<2:32:32,  1.36s/it]
  2%|▏         | 122/6837 [11:00<2:28:58,  1.33s/it]
  2%|▏         | 123/6837 [11:01<2:25:32,  1.30s/it]
  2%|▏         | 124/6837 [11:02<2:24:42,  1.29s/it]
  2%|▏         | 125/6837 [11:03<2:25:22,  1.30s/it]{'loss': 7.3905, 'grad_norm': 0.2840506434440613, 'learning_rate': 0.0010964912280701754, 'epoch': 0.0}
                                                    
+
  2%|▏         | 125/6837 [11:03<2:25:22,  1.30s/it]
  2%|▏         | 126/6837 [11:05<2:31:07,  1.35s/it]
  2%|▏         | 127/6837 [11:06<2:32:05,  1.36s/it]
  2%|▏         | 128/6837 [11:08<2:26:04,  1.31s/it]
  2%|▏         | 129/6837 [11:09<2:28:26,  1.33s/it]
  2%|▏         | 130/6837 [11:10<2:27:47,  1.32s/it]
  2%|▏         | 131/6837 [11:11<2:23:00,  1.28s/it]
  2%|▏         | 132/6837 [11:13<2:18:15,  1.24s/it]
  2%|▏         | 133/6837 [11:14<2:19:03,  1.24s/it]
  2%|▏         | 134/6837 [11:15<2:21:09,  1.26s/it]
  2%|▏         | 135/6837 [11:16<2:19:05,  1.25s/it]
  2%|▏         | 136/6837 [11:17<2:11:42,  1.18s/it]
  2%|▏         | 137/6837 [11:19<2:16:20,  1.22s/it]
  2%|▏         | 138/6837 [11:20<2:14:16,  1.20s/it]
  2%|▏         | 139/6837 [11:21<2:11:58,  1.18s/it]
  2%|▏         | 140/6837 [11:22<2:08:47,  1.15s/it]
  2%|▏         | 141/6837 [11:23<2:04:33,  1.12s/it]
  2%|▏         | 142/6837 [11:24<2:08:31,  1.15s/it]
  2%|▏         | 143/6837 [11:25<2:00:58,  1.08s/it]
  2%|▏         | 144/6837 [11:26<1:57:55,  1.06s/it]
  2%|▏         | 145/6837 [11:27<1:56:31,  1.04s/it]
  2%|▏         | 146/6837 [11:28<2:01:49,  1.09s/it]
  2%|▏         | 147/6837 [11:29<1:58:11,  1.06s/it]
  2%|▏         | 148/6837 [11:30<1:52:53,  1.01s/it]
  2%|▏         | 149/6837 [11:31<1:54:27,  1.03s/it]
  2%|▏         | 150/6837 [11:32<1:54:00,  1.02s/it]
                                                    
{'loss': 6.9569, 'grad_norm': 0.2873937785625458, 'learning_rate': 0.0013157894736842105, 'epoch': 0.0}
+
  2%|▏         | 150/6837 [11:32<1:54:00,  1.02s/it]
  2%|▏         | 151/6837 [11:34<2:00:47,  1.08s/it]
  2%|▏         | 152/6837 [11:35<1:55:35,  1.04s/it]
  2%|▏         | 153/6837 [11:35<1:52:03,  1.01s/it]
  2%|▏         | 154/6837 [11:37<1:53:44,  1.02s/it]
  2%|▏         | 155/6837 [11:38<1:56:37,  1.05s/it]
  2%|▏         | 156/6837 [11:39<1:59:14,  1.07s/it]
  2%|▏         | 157/6837 [11:40<2:00:44,  1.08s/it]
  2%|▏         | 158/6837 [11:41<1:53:12,  1.02s/it]
  2%|▏         | 159/6837 [11:42<1:57:05,  1.05s/it]
  2%|▏         | 160/6837 [11:43<2:01:09,  1.09s/it]
  2%|▏         | 161/6837 [11:44<1:51:13,  1.00it/s]
  2%|▏         | 162/6837 [11:45<1:50:57,  1.00it/s]
  2%|▏         | 163/6837 [11:46<1:45:48,  1.05it/s]
  2%|▏         | 164/6837 [11:47<1:43:01,  1.08it/s]
  2%|▏         | 165/6837 [11:47<1:42:40,  1.08it/s]
  2%|▏         | 166/6837 [11:49<1:47:54,  1.03it/s]
  2%|▏         | 167/6837 [11:50<1:54:07,  1.03s/it]
  2%|▏         | 168/6837 [11:51<1:48:52,  1.02it/s]
  2%|▏         | 169/6837 [11:52<1:48:12,  1.03it/s]
  2%|▏         | 170/6837 [11:52<1:44:56,  1.06it/s]
  3%|▎         | 171/6837 [11:54<1:51:04,  1.00it/s]
  3%|▎         | 172/6837 [11:54<1:45:23,  1.05it/s]
  3%|▎         | 173/6837 [11:55<1:39:35,  1.12it/s]
  3%|▎         | 174/6837 [11:56<1:38:34,  1.13it/s]
  3%|▎         | 175/6837 [11:57<1:36:38,  1.15it/s]
                                                    
{'loss': 6.6535, 'grad_norm': 0.2848723530769348, 'learning_rate': 0.0015350877192982456, 'epoch': 0.01}
+
  3%|▎         | 175/6837 [11:57<1:36:38,  1.15it/s]
  3%|▎         | 176/6837 [11:58<1:41:15,  1.10it/s]
  3%|▎         | 177/6837 [11:59<1:48:50,  1.02it/s]
  3%|▎         | 178/6837 [12:00<1:52:11,  1.01s/it]
  3%|▎         | 179/6837 [12:01<1:51:57,  1.01s/it]
  3%|▎         | 180/6837 [12:02<1:47:22,  1.03it/s]
  3%|▎         | 181/6837 [12:03<1:46:22,  1.04it/s]
  3%|▎         | 182/6837 [12:04<1:44:47,  1.06it/s]
  3%|▎         | 183/6837 [12:05<1:43:48,  1.07it/s]
  3%|▎         | 184/6837 [12:06<1:39:46,  1.11it/s]
  3%|▎         | 185/6837 [12:06<1:41:20,  1.09it/s]
  3%|▎         | 186/6837 [12:07<1:38:07,  1.13it/s]
  3%|▎         | 187/6837 [12:08<1:38:11,  1.13it/s]
  3%|▎         | 188/6837 [12:09<1:38:51,  1.12it/s]
  3%|▎         | 189/6837 [12:10<1:37:16,  1.14it/s]
  3%|▎         | 190/6837 [12:11<1:34:52,  1.17it/s]
  3%|▎         | 191/6837 [12:12<1:34:59,  1.17it/s]
  3%|▎         | 192/6837 [12:12<1:35:03,  1.17it/s]
  3%|▎         | 193/6837 [12:13<1:36:32,  1.15it/s]
  3%|▎         | 194/6837 [12:14<1:37:00,  1.14it/s]
  3%|▎         | 195/6837 [12:15<1:39:11,  1.12it/s]
  3%|▎         | 196/6837 [12:16<1:34:54,  1.17it/s]
  3%|▎         | 197/6837 [12:17<1:32:18,  1.20it/s]
  3%|▎         | 198/6837 [12:18<1:34:41,  1.17it/s]
  3%|▎         | 199/6837 [12:19<1:35:05,  1.16it/s]
  3%|▎         | 200/6837 [12:19<1:32:26,  1.20it/s]
                                                    
{'loss': 6.4353, 'grad_norm': 0.38353198766708374, 'learning_rate': 0.0017543859649122807, 'epoch': 0.01}
+
  3%|▎         | 200/6837 [12:19<1:32:26,  1.20it/s]
  3%|▎         | 201/6837 [12:20<1:31:35,  1.21it/s]
  3%|▎         | 202/6837 [12:21<1:32:26,  1.20it/s]
  3%|▎         | 203/6837 [12:22<1:28:38,  1.25it/s]
  3%|▎         | 204/6837 [12:23<1:30:15,  1.22it/s]
  3%|▎         | 205/6837 [12:23<1:30:58,  1.21it/s]
  3%|▎         | 206/6837 [12:24<1:41:12,  1.09it/s]
  3%|▎         | 207/6837 [12:25<1:35:04,  1.16it/s]
  3%|▎         | 208/6837 [12:26<1:32:03,  1.20it/s]
  3%|▎         | 209/6837 [12:27<1:25:15,  1.30it/s]
  3%|▎         | 210/6837 [12:27<1:22:40,  1.34it/s]
  3%|▎         | 211/6837 [12:28<1:23:52,  1.32it/s]
  3%|▎         | 212/6837 [12:29<1:23:15,  1.33it/s]
  3%|▎         | 213/6837 [12:30<1:22:52,  1.33it/s]
  3%|▎         | 214/6837 [12:30<1:25:01,  1.30it/s]
  3%|▎         | 215/6837 [12:31<1:26:50,  1.27it/s]
  3%|▎         | 216/6837 [12:32<1:23:36,  1.32it/s]
  3%|▎         | 217/6837 [12:33<1:26:04,  1.28it/s]
  3%|▎         | 218/6837 [12:34<1:30:17,  1.22it/s]
  3%|▎         | 219/6837 [12:35<1:31:04,  1.21it/s]
  3%|▎         | 220/6837 [12:35<1:28:07,  1.25it/s]
  3%|▎         | 221/6837 [12:36<1:29:15,  1.24it/s]
  3%|▎         | 222/6837 [12:37<1:25:07,  1.30it/s]
  3%|▎         | 223/6837 [12:38<1:25:08,  1.29it/s]
  3%|▎         | 224/6837 [12:38<1:29:36,  1.23it/s]
  3%|▎         | 225/6837 [12:39<1:33:09,  1.18it/s]
                                                    
{'loss': 6.2598, 'grad_norm': 0.3620648980140686, 'learning_rate': 0.001973684210526316, 'epoch': 0.01}
+
  3%|▎         | 225/6837 [12:39<1:33:09,  1.18it/s]
  3%|▎         | 226/6837 [12:40<1:34:44,  1.16it/s]
  3%|▎         | 227/6837 [12:41<1:29:34,  1.23it/s]
  3%|▎         | 228/6837 [12:42<1:25:35,  1.29it/s]
  3%|▎         | 229/6837 [12:42<1:25:58,  1.28it/s]
  3%|▎         | 230/6837 [12:43<1:25:36,  1.29it/s]
  3%|▎         | 231/6837 [12:44<1:25:29,  1.29it/s]
  3%|▎         | 232/6837 [12:45<1:19:55,  1.38it/s]
  3%|▎         | 233/6837 [12:45<1:20:02,  1.37it/s]
  3%|▎         | 234/6837 [12:46<1:20:37,  1.36it/s]
  3%|▎         | 235/6837 [12:47<1:21:54,  1.34it/s]
  3%|▎         | 236/6837 [12:48<1:24:59,  1.29it/s]
  3%|▎         | 237/6837 [12:48<1:24:43,  1.30it/s]
  3%|▎         | 238/6837 [12:49<1:28:02,  1.25it/s]
  3%|▎         | 239/6837 [12:50<1:25:40,  1.28it/s]
  4%|▎         | 240/6837 [12:51<1:27:00,  1.26it/s]
  4%|▎         | 241/6837 [12:52<1:25:17,  1.29it/s]
  4%|▎         | 242/6837 [12:53<1:29:40,  1.23it/s]
  4%|▎         | 243/6837 [12:53<1:27:55,  1.25it/s]
  4%|▎         | 244/6837 [12:54<1:29:36,  1.23it/s]
  4%|▎         | 245/6837 [12:55<1:31:42,  1.20it/s]
  4%|▎         | 246/6837 [12:56<1:29:28,  1.23it/s]
  4%|▎         | 247/6837 [12:56<1:25:42,  1.28it/s]
  4%|▎         | 248/6837 [12:57<1:24:40,  1.30it/s]
  4%|▎         | 249/6837 [12:58<1:23:17,  1.32it/s]
  4%|▎         | 250/6837 [12:59<1:20:47,  1.36it/s]
                                                    
{'loss': 6.1052, 'grad_norm': 0.3460889160633087, 'learning_rate': 0.0021929824561403508, 'epoch': 0.01}
+
  4%|▎         | 250/6837 [12:59<1:20:47,  1.36it/s]
  4%|▎         | 251/6837 [12:59<1:19:09,  1.39it/s]
  4%|▎         | 252/6837 [13:00<1:17:31,  1.42it/s]
  4%|▎         | 253/6837 [13:01<1:14:37,  1.47it/s]
  4%|▎         | 254/6837 [13:01<1:16:09,  1.44it/s]
  4%|▎         | 255/6837 [13:02<1:11:16,  1.54it/s]
  4%|▎         | 256/6837 [13:03<1:17:55,  1.41it/s]
  4%|▍         | 257/6837 [13:04<1:23:00,  1.32it/s]
  4%|▍         | 258/6837 [13:04<1:22:36,  1.33it/s]
  4%|▍         | 259/6837 [13:05<1:22:30,  1.33it/s]
  4%|▍         | 260/6837 [13:06<1:18:59,  1.39it/s]
  4%|▍         | 261/6837 [13:06<1:14:47,  1.47it/s]
  4%|▍         | 262/6837 [13:07<1:15:12,  1.46it/s]
  4%|▍         | 263/6837 [13:08<1:16:33,  1.43it/s]
  4%|▍         | 264/6837 [13:08<1:15:06,  1.46it/s]
  4%|▍         | 265/6837 [13:09<1:13:50,  1.48it/s]
  4%|▍         | 266/6837 [13:10<1:14:28,  1.47it/s]
  4%|▍         | 267/6837 [13:10<1:13:34,  1.49it/s]
  4%|▍         | 268/6837 [13:11<1:17:13,  1.42it/s]
  4%|▍         | 269/6837 [13:12<1:14:17,  1.47it/s]
  4%|▍         | 270/6837 [13:12<1:11:27,  1.53it/s]
  4%|▍         | 271/6837 [13:13<1:10:32,  1.55it/s]
  4%|▍         | 272/6837 [13:14<1:10:30,  1.55it/s]
  4%|▍         | 273/6837 [13:14<1:10:09,  1.56it/s]
  4%|▍         | 274/6837 [13:15<1:11:25,  1.53it/s]
  4%|▍         | 275/6837 [13:16<1:11:01,  1.54it/s]
                                                    
{'loss': 5.9682, 'grad_norm': 0.3327798843383789, 'learning_rate': 0.002412280701754386, 'epoch': 0.01}
+
  4%|▍         | 275/6837 [13:16<1:11:01,  1.54it/s]
  4%|▍         | 276/6837 [13:16<1:11:30,  1.53it/s]
  4%|▍         | 277/6837 [13:17<1:08:39,  1.59it/s]
  4%|▍         | 278/6837 [13:18<1:09:54,  1.56it/s]
  4%|▍         | 279/6837 [13:18<1:09:17,  1.58it/s]
  4%|▍         | 280/6837 [13:19<1:08:04,  1.61it/s]
  4%|▍         | 281/6837 [13:19<1:08:59,  1.58it/s]
  4%|▍         | 282/6837 [13:20<1:08:13,  1.60it/s]
  4%|▍         | 283/6837 [13:21<1:08:35,  1.59it/s]
  4%|▍         | 284/6837 [13:21<1:05:05,  1.68it/s]
  4%|▍         | 285/6837 [13:22<1:03:44,  1.71it/s]
  4%|▍         | 286/6837 [13:22<1:01:00,  1.79it/s]
  4%|▍         | 287/6837 [13:23<1:03:49,  1.71it/s]
  4%|▍         | 288/6837 [13:23<1:02:23,  1.75it/s]
  4%|▍         | 289/6837 [13:24<1:05:05,  1.68it/s]
  4%|▍         | 290/6837 [13:25<1:06:00,  1.65it/s]
  4%|▍         | 291/6837 [13:25<1:03:32,  1.72it/s]
  4%|▍         | 292/6837 [13:26<1:03:20,  1.72it/s]
  4%|▍         | 293/6837 [13:26<1:00:53,  1.79it/s]
  4%|▍         | 294/6837 [13:27<1:03:13,  1.72it/s]
  4%|▍         | 295/6837 [13:27<1:01:43,  1.77it/s]
  4%|▍         | 296/6837 [13:28<1:04:20,  1.69it/s]
  4%|▍         | 297/6837 [13:29<1:02:45,  1.74it/s]
  4%|▍         | 298/6837 [13:29<1:00:10,  1.81it/s]
  4%|▍         | 299/6837 [13:30<1:00:46,  1.79it/s]
  4%|▍         | 300/6837 [13:30<1:01:33,  1.77it/s]
                                                    
{'loss': 5.8526, 'grad_norm': 0.4219636619091034, 'learning_rate': 0.002631578947368421, 'epoch': 0.01}
+
  4%|▍         | 300/6837 [13:30<1:01:33,  1.77it/s]
  4%|▍         | 301/6837 [13:31<59:13,  1.84it/s]  
  4%|▍         | 302/6837 [13:31<55:32,  1.96it/s]
  4%|▍         | 303/6837 [13:32<54:24,  2.00it/s]
  4%|▍         | 304/6837 [13:32<55:39,  1.96it/s]
  4%|▍         | 305/6837 [13:33<55:08,  1.97it/s]
  4%|▍         | 306/6837 [13:33<56:21,  1.93it/s]
  4%|▍         | 307/6837 [13:34<57:59,  1.88it/s]
  5%|▍         | 308/6837 [13:34<54:55,  1.98it/s]
  5%|▍         | 309/6837 [13:35<56:52,  1.91it/s]
  5%|▍         | 310/6837 [13:35<56:34,  1.92it/s]
  5%|▍         | 311/6837 [13:36<56:41,  1.92it/s]
  5%|▍         | 312/6837 [13:36<55:48,  1.95it/s]
  5%|▍         | 313/6837 [13:37<54:52,  1.98it/s]
  5%|▍         | 314/6837 [13:37<54:34,  1.99it/s]
  5%|▍         | 315/6837 [13:38<53:45,  2.02it/s]
  5%|▍         | 316/6837 [13:38<52:49,  2.06it/s]
  5%|▍         | 317/6837 [13:39<53:30,  2.03it/s]
  5%|▍         | 318/6837 [13:39<56:10,  1.93it/s]
  5%|▍         | 319/6837 [13:40<55:46,  1.95it/s]
  5%|▍         | 320/6837 [13:40<53:45,  2.02it/s]
  5%|▍         | 321/6837 [13:41<51:40,  2.10it/s]
  5%|▍         | 322/6837 [13:41<50:13,  2.16it/s]
  5%|▍         | 323/6837 [13:42<50:13,  2.16it/s]
  5%|▍         | 324/6837 [13:42<54:05,  2.01it/s]
  5%|▍         | 325/6837 [13:43<56:02,  1.94it/s]
                                                  
{'loss': 5.7558, 'grad_norm': 0.4793476164340973, 'learning_rate': 0.0028508771929824563, 'epoch': 0.01}
+
  5%|▍         | 325/6837 [13:43<56:02,  1.94it/s]
  5%|▍         | 326/6837 [13:43<54:39,  1.99it/s]
  5%|▍         | 327/6837 [13:44<57:09,  1.90it/s]
  5%|▍         | 328/6837 [13:44<54:19,  2.00it/s]
  5%|▍         | 329/6837 [13:45<54:18,  2.00it/s]
  5%|▍         | 330/6837 [13:45<53:20,  2.03it/s]
  5%|▍         | 331/6837 [13:46<51:31,  2.10it/s]
  5%|▍         | 332/6837 [13:46<50:06,  2.16it/s]
  5%|▍         | 333/6837 [13:47<52:25,  2.07it/s]
  5%|▍         | 334/6837 [13:47<51:03,  2.12it/s]
  5%|▍         | 335/6837 [13:48<55:19,  1.96it/s]
  5%|▍         | 336/6837 [13:48<55:41,  1.95it/s]
  5%|▍         | 337/6837 [13:49<54:26,  1.99it/s]
  5%|▍         | 338/6837 [13:49<53:55,  2.01it/s]
  5%|▍         | 339/6837 [13:50<55:00,  1.97it/s]
  5%|▍         | 340/6837 [13:50<53:12,  2.04it/s]
  5%|▍         | 341/6837 [13:51<50:17,  2.15it/s]
  5%|▌         | 342/6837 [13:51<48:52,  2.21it/s]
  5%|▌         | 343/6837 [13:51<49:20,  2.19it/s]
  5%|▌         | 344/6837 [13:52<49:02,  2.21it/s]
  5%|▌         | 345/6837 [13:52<48:08,  2.25it/s]
  5%|▌         | 346/6837 [13:53<51:59,  2.08it/s]
  5%|▌         | 347/6837 [13:53<52:33,  2.06it/s]
  5%|▌         | 348/6837 [13:54<55:03,  1.96it/s]
  5%|▌         | 349/6837 [13:54<54:50,  1.97it/s]
  5%|▌         | 350/6837 [13:55<54:16,  1.99it/s]
                                                  
{'loss': 5.6523, 'grad_norm': 0.38318943977355957, 'learning_rate': 0.0030701754385964912, 'epoch': 0.01}
+
  5%|▌         | 350/6837 [13:55<54:16,  1.99it/s]
  5%|▌         | 351/6837 [13:55<52:36,  2.05it/s]
  5%|▌         | 352/6837 [13:56<51:50,  2.08it/s]
  5%|▌         | 353/6837 [13:56<49:40,  2.18it/s]
  5%|▌         | 354/6837 [13:57<49:45,  2.17it/s]
  5%|▌         | 355/6837 [13:57<54:20,  1.99it/s]
  5%|▌         | 356/6837 [13:58<54:23,  1.99it/s]
  5%|▌         | 357/6837 [13:58<52:52,  2.04it/s]
  5%|▌         | 358/6837 [13:59<57:29,  1.88it/s]
  5%|▌         | 359/6837 [14:00<1:00:18,  1.79it/s]
  5%|▌         | 360/6837 [14:00<58:43,  1.84it/s]  
  5%|▌         | 361/6837 [14:01<54:45,  1.97it/s]
  5%|▌         | 362/6837 [14:01<54:32,  1.98it/s]
  5%|▌         | 363/6837 [14:02<54:16,  1.99it/s]
  5%|▌         | 364/6837 [14:02<50:48,  2.12it/s]
  5%|▌         | 365/6837 [14:02<49:49,  2.17it/s]
  5%|▌         | 366/6837 [14:03<50:14,  2.15it/s]
  5%|▌         | 367/6837 [14:03<50:24,  2.14it/s]
  5%|▌         | 368/6837 [14:04<52:24,  2.06it/s]
  5%|▌         | 369/6837 [14:04<50:18,  2.14it/s]
  5%|▌         | 370/6837 [14:05<47:45,  2.26it/s]
  5%|▌         | 371/6837 [14:05<45:28,  2.37it/s]
  5%|▌         | 372/6837 [14:05<44:52,  2.40it/s]
  5%|▌         | 373/6837 [14:06<44:27,  2.42it/s]
  5%|▌         | 374/6837 [14:06<42:40,  2.52it/s]
  5%|▌         | 375/6837 [14:07<41:40,  2.58it/s]
                                                  
{'loss': 5.5936, 'grad_norm': 0.3405887484550476, 'learning_rate': 0.0032894736842105266, 'epoch': 0.01}
+
  5%|▌         | 375/6837 [14:07<41:40,  2.58it/s]
  5%|▌         | 376/6837 [14:07<44:04,  2.44it/s]
  6%|▌         | 377/6837 [14:07<45:15,  2.38it/s]
  6%|▌         | 378/6837 [14:08<44:47,  2.40it/s]
  6%|▌         | 379/6837 [14:08<44:53,  2.40it/s]
  6%|▌         | 380/6837 [14:09<45:58,  2.34it/s]
  6%|▌         | 381/6837 [14:09<50:37,  2.13it/s]
  6%|▌         | 382/6837 [14:10<49:01,  2.19it/s]
  6%|▌         | 383/6837 [14:10<48:09,  2.23it/s]
  6%|▌         | 384/6837 [14:11<48:21,  2.22it/s]
  6%|▌         | 385/6837 [14:11<47:09,  2.28it/s]
  6%|▌         | 386/6837 [14:11<46:31,  2.31it/s]
  6%|▌         | 387/6837 [14:12<46:32,  2.31it/s]
  6%|▌         | 388/6837 [14:12<46:24,  2.32it/s]
  6%|▌         | 389/6837 [14:13<45:41,  2.35it/s]
  6%|▌         | 390/6837 [14:13<45:42,  2.35it/s]
  6%|▌         | 391/6837 [14:14<46:46,  2.30it/s]
  6%|▌         | 392/6837 [14:14<45:59,  2.34it/s]
  6%|▌         | 393/6837 [14:14<45:11,  2.38it/s]
  6%|▌         | 394/6837 [14:15<47:16,  2.27it/s]
  6%|▌         | 395/6837 [14:15<49:05,  2.19it/s]
  6%|▌         | 396/6837 [14:16<48:45,  2.20it/s]
  6%|▌         | 397/6837 [14:16<48:01,  2.24it/s]
  6%|▌         | 398/6837 [14:17<48:28,  2.21it/s]
  6%|▌         | 399/6837 [14:17<48:35,  2.21it/s]
  6%|▌         | 400/6837 [14:18<47:59,  2.24it/s]
                                                  
{'loss': 5.5289, 'grad_norm': 0.41496971249580383, 'learning_rate': 0.0035087719298245615, 'epoch': 0.01}
+
  6%|▌         | 400/6837 [14:18<47:59,  2.24it/s]
  6%|▌         | 401/6837 [14:18<46:50,  2.29it/s]
  6%|▌         | 402/6837 [14:18<46:57,  2.28it/s]
  6%|▌         | 403/6837 [14:19<44:48,  2.39it/s]
  6%|▌         | 404/6837 [14:19<46:58,  2.28it/s]
  6%|▌         | 405/6837 [14:20<47:27,  2.26it/s]
  6%|▌         | 406/6837 [14:20<46:07,  2.32it/s]
  6%|▌         | 407/6837 [14:21<43:53,  2.44it/s]
  6%|▌         | 408/6837 [14:21<41:44,  2.57it/s]
  6%|▌         | 409/6837 [14:21<41:13,  2.60it/s]
  6%|▌         | 410/6837 [14:22<45:08,  2.37it/s]
  6%|▌         | 411/6837 [14:22<49:04,  2.18it/s]
  6%|▌         | 412/6837 [14:23<48:23,  2.21it/s]
  6%|▌         | 413/6837 [14:23<47:46,  2.24it/s]
  6%|▌         | 414/6837 [14:24<47:49,  2.24it/s]
  6%|▌         | 415/6837 [14:24<47:37,  2.25it/s]
  6%|▌         | 416/6837 [14:24<46:39,  2.29it/s]
  6%|▌         | 417/6837 [14:25<45:49,  2.33it/s]
  6%|▌         | 418/6837 [14:25<45:48,  2.34it/s]
  6%|▌         | 419/6837 [14:26<44:26,  2.41it/s]
  6%|▌         | 420/6837 [14:26<42:31,  2.52it/s]
  6%|▌         | 421/6837 [14:26<41:46,  2.56it/s]
  6%|▌         | 422/6837 [14:27<39:57,  2.68it/s]
  6%|▌         | 423/6837 [14:27<40:57,  2.61it/s]
  6%|▌         | 424/6837 [14:28<41:13,  2.59it/s]
  6%|▌         | 425/6837 [14:28<41:48,  2.56it/s]
                                                  
{'loss': 5.4794, 'grad_norm': 0.3674440383911133, 'learning_rate': 0.003728070175438597, 'epoch': 0.01}
+
  6%|▌         | 425/6837 [14:28<41:48,  2.56it/s]
  6%|▌         | 426/6837 [14:28<41:20,  2.58it/s]
  6%|▌         | 427/6837 [14:29<42:43,  2.50it/s]
  6%|▋         | 428/6837 [14:29<43:12,  2.47it/s]
  6%|▋         | 429/6837 [14:30<44:16,  2.41it/s]
  6%|▋         | 430/6837 [14:30<43:27,  2.46it/s]
  6%|▋         | 431/6837 [14:30<41:45,  2.56it/s]
  6%|▋         | 432/6837 [14:31<42:51,  2.49it/s]
  6%|▋         | 433/6837 [14:31<42:55,  2.49it/s]
  6%|▋         | 434/6837 [14:32<42:18,  2.52it/s]
  6%|▋         | 435/6837 [14:32<43:39,  2.44it/s]
  6%|▋         | 436/6837 [14:32<42:18,  2.52it/s]
  6%|▋         | 437/6837 [14:33<41:14,  2.59it/s]
  6%|▋         | 438/6837 [14:33<42:11,  2.53it/s]
  6%|▋         | 439/6837 [14:34<41:58,  2.54it/s]
  6%|▋         | 440/6837 [14:34<40:19,  2.64it/s]
  6%|▋         | 441/6837 [14:34<43:45,  2.44it/s]
  6%|▋         | 442/6837 [14:35<43:27,  2.45it/s]
  6%|▋         | 443/6837 [14:35<43:35,  2.44it/s]
  6%|▋         | 444/6837 [14:36<42:32,  2.50it/s]
  7%|▋         | 445/6837 [14:36<42:24,  2.51it/s]
  7%|▋         | 446/6837 [14:36<41:20,  2.58it/s]
  7%|▋         | 447/6837 [14:37<40:18,  2.64it/s]
  7%|▋         | 448/6837 [14:37<41:36,  2.56it/s]
  7%|▋         | 449/6837 [14:37<41:24,  2.57it/s]
  7%|▋         | 450/6837 [14:38<40:55,  2.60it/s]
                                                  
{'loss': 5.4306, 'grad_norm': 0.3237208127975464, 'learning_rate': 0.003947368421052632, 'epoch': 0.01}
+
  7%|▋         | 450/6837 [14:38<40:55,  2.60it/s]
  7%|▋         | 451/6837 [14:38<41:53,  2.54it/s]
  7%|▋         | 452/6837 [14:39<42:05,  2.53it/s]
  7%|▋         | 453/6837 [14:39<41:49,  2.54it/s]
  7%|▋         | 454/6837 [14:39<40:26,  2.63it/s]
  7%|▋         | 455/6837 [14:40<40:02,  2.66it/s]
  7%|▋         | 456/6837 [14:40<39:40,  2.68it/s]
  7%|▋         | 457/6837 [14:41<41:12,  2.58it/s]
  7%|▋         | 458/6837 [14:41<42:05,  2.53it/s]
  7%|▋         | 459/6837 [14:41<41:00,  2.59it/s]
  7%|▋         | 460/6837 [14:42<41:16,  2.58it/s]
  7%|▋         | 461/6837 [14:42<42:32,  2.50it/s]
  7%|▋         | 462/6837 [14:43<41:49,  2.54it/s]
  7%|▋         | 463/6837 [14:43<45:17,  2.35it/s]
  7%|▋         | 464/6837 [14:43<44:13,  2.40it/s]
  7%|▋         | 465/6837 [14:44<45:05,  2.36it/s]
  7%|▋         | 466/6837 [14:44<45:43,  2.32it/s]
  7%|▋         | 467/6837 [14:45<42:09,  2.52it/s]
  7%|▋         | 468/6837 [14:45<41:20,  2.57it/s]
  7%|▋         | 469/6837 [14:45<40:27,  2.62it/s]
  7%|▋         | 470/6837 [14:46<40:01,  2.65it/s]
  7%|▋         | 471/6837 [14:46<39:10,  2.71it/s]
  7%|▋         | 472/6837 [14:46<37:18,  2.84it/s]
  7%|▋         | 473/6837 [14:47<38:06,  2.78it/s]
  7%|▋         | 474/6837 [14:47<38:20,  2.77it/s]
  7%|▋         | 475/6837 [14:48<39:26,  2.69it/s]
                                                  
{'loss': 5.3872, 'grad_norm': 0.3057781755924225, 'learning_rate': 0.004166666666666667, 'epoch': 0.01}
+
  7%|▋         | 475/6837 [14:48<39:26,  2.69it/s]
  7%|▋         | 476/6837 [14:48<39:03,  2.71it/s]
  7%|▋         | 477/6837 [14:48<39:35,  2.68it/s]
  7%|▋         | 478/6837 [14:49<39:37,  2.67it/s]
  7%|▋         | 479/6837 [14:49<39:39,  2.67it/s]
  7%|▋         | 480/6837 [14:49<38:07,  2.78it/s]
  7%|▋         | 481/6837 [14:50<40:24,  2.62it/s]
  7%|▋         | 482/6837 [14:50<42:54,  2.47it/s]
  7%|▋         | 483/6837 [14:51<40:30,  2.61it/s]
  7%|▋         | 484/6837 [14:51<40:34,  2.61it/s]
  7%|▋         | 485/6837 [14:51<38:28,  2.75it/s]
  7%|▋         | 486/6837 [14:52<37:44,  2.80it/s]
  7%|▋         | 487/6837 [14:52<38:05,  2.78it/s]
  7%|▋         | 488/6837 [14:52<37:08,  2.85it/s]
  7%|▋         | 489/6837 [14:53<37:35,  2.81it/s]
  7%|▋         | 490/6837 [14:53<38:10,  2.77it/s]
  7%|▋         | 491/6837 [14:54<40:12,  2.63it/s]
  7%|▋         | 492/6837 [14:54<40:14,  2.63it/s]
  7%|▋         | 493/6837 [14:54<38:56,  2.71it/s]
  7%|▋         | 494/6837 [14:55<38:46,  2.73it/s]
  7%|▋         | 495/6837 [14:55<39:43,  2.66it/s]
  7%|▋         | 496/6837 [14:55<37:53,  2.79it/s]
  7%|▋         | 497/6837 [14:56<38:37,  2.74it/s]
  7%|▋         | 498/6837 [14:56<38:46,  2.72it/s]
  7%|▋         | 499/6837 [14:56<38:31,  2.74it/s]
  7%|▋         | 500/6837 [14:57<38:04,  2.77it/s]{'loss': 5.3308, 'grad_norm': 0.25192293524742126, 'learning_rate': 0.0043859649122807015, 'epoch': 0.01}
                                                  
+
  7%|▋         | 500/6837 [14:57<38:04,  2.77it/s]
  7%|▋         | 501/6837 [14:57<37:34,  2.81it/s]
  7%|▋         | 502/6837 [14:57<38:22,  2.75it/s]
  7%|▋         | 503/6837 [14:58<36:11,  2.92it/s]
  7%|▋         | 504/6837 [14:58<38:07,  2.77it/s]
  7%|▋         | 505/6837 [14:59<37:40,  2.80it/s]
  7%|▋         | 506/6837 [14:59<38:50,  2.72it/s]
  7%|▋         | 507/6837 [14:59<39:42,  2.66it/s]
  7%|▋         | 508/6837 [15:00<37:41,  2.80it/s]
  7%|▋         | 509/6837 [15:00<36:32,  2.89it/s]
  7%|▋         | 510/6837 [15:00<35:48,  2.94it/s]
  7%|▋         | 511/6837 [15:01<34:42,  3.04it/s]
  7%|▋         | 512/6837 [15:01<35:58,  2.93it/s]
  8%|▊         | 513/6837 [15:01<36:08,  2.92it/s]
  8%|▊         | 514/6837 [15:02<35:43,  2.95it/s]
  8%|▊         | 515/6837 [15:02<36:07,  2.92it/s]
  8%|▊         | 516/6837 [15:02<36:29,  2.89it/s]
  8%|▊         | 517/6837 [15:03<36:19,  2.90it/s]
  8%|▊         | 518/6837 [15:03<36:05,  2.92it/s]
  8%|▊         | 519/6837 [15:03<35:43,  2.95it/s]
  8%|▊         | 520/6837 [15:04<36:45,  2.86it/s]
  8%|▊         | 521/6837 [15:04<35:55,  2.93it/s]
  8%|▊         | 522/6837 [15:04<36:57,  2.85it/s]
  8%|▊         | 523/6837 [15:05<36:29,  2.88it/s]
  8%|▊         | 524/6837 [15:05<35:20,  2.98it/s]
  8%|▊         | 525/6837 [15:05<35:19,  2.98it/s]
                                                  
{'loss': 5.2953, 'grad_norm': 0.26653507351875305, 'learning_rate': 0.004605263157894737, 'epoch': 0.02}
+
  8%|▊         | 525/6837 [15:05<35:19,  2.98it/s]
  8%|▊         | 526/6837 [15:06<36:42,  2.87it/s]
  8%|▊         | 527/6837 [15:06<35:59,  2.92it/s]
  8%|▊         | 528/6837 [15:06<35:34,  2.96it/s]
  8%|▊         | 529/6837 [15:07<34:35,  3.04it/s]
  8%|▊         | 530/6837 [15:07<33:07,  3.17it/s]
  8%|▊         | 531/6837 [15:07<34:03,  3.09it/s]
  8%|▊         | 532/6837 [15:08<35:48,  2.93it/s]
  8%|▊         | 533/6837 [15:08<35:59,  2.92it/s]
  8%|▊         | 534/6837 [15:08<36:00,  2.92it/s]
  8%|▊         | 535/6837 [15:09<34:21,  3.06it/s]
  8%|▊         | 536/6837 [15:09<34:49,  3.02it/s]
  8%|▊         | 537/6837 [15:09<34:52,  3.01it/s]
  8%|▊         | 538/6837 [15:10<34:29,  3.04it/s]
  8%|▊         | 539/6837 [15:10<35:01,  3.00it/s]
  8%|▊         | 540/6837 [15:10<34:10,  3.07it/s]
  8%|▊         | 541/6837 [15:11<34:05,  3.08it/s]
  8%|▊         | 542/6837 [15:11<33:28,  3.13it/s]
  8%|▊         | 543/6837 [15:11<34:03,  3.08it/s]
  8%|▊         | 544/6837 [15:12<34:14,  3.06it/s]
  8%|▊         | 545/6837 [15:12<35:13,  2.98it/s]
  8%|▊         | 546/6837 [15:12<36:27,  2.88it/s]
  8%|▊         | 547/6837 [15:13<36:29,  2.87it/s]
  8%|▊         | 548/6837 [15:13<36:46,  2.85it/s]
  8%|▊         | 549/6837 [15:14<38:00,  2.76it/s]
  8%|▊         | 550/6837 [15:14<37:11,  2.82it/s]
                                                  
{'loss': 5.2706, 'grad_norm': 0.2730398178100586, 'learning_rate': 0.004824561403508772, 'epoch': 0.02}
+
  8%|▊         | 550/6837 [15:14<37:11,  2.82it/s]
  8%|▊         | 551/6837 [15:14<38:51,  2.70it/s]
  8%|▊         | 552/6837 [15:15<37:51,  2.77it/s]
  8%|▊         | 553/6837 [15:15<37:50,  2.77it/s]
  8%|▊         | 554/6837 [15:15<39:08,  2.67it/s]
  8%|▊         | 555/6837 [15:16<37:25,  2.80it/s]
  8%|▊         | 556/6837 [15:16<38:26,  2.72it/s]
  8%|▊         | 557/6837 [15:17<41:28,  2.52it/s]
  8%|▊         | 558/6837 [15:17<40:35,  2.58it/s]
  8%|▊         | 559/6837 [15:17<39:58,  2.62it/s]
  8%|▊         | 560/6837 [15:18<39:22,  2.66it/s]
  8%|▊         | 561/6837 [15:18<39:29,  2.65it/s]
  8%|▊         | 562/6837 [15:18<40:19,  2.59it/s]
  8%|▊         | 563/6837 [15:19<38:45,  2.70it/s]
  8%|▊         | 564/6837 [15:19<38:01,  2.75it/s]
  8%|▊         | 565/6837 [15:19<38:05,  2.74it/s]
  8%|▊         | 566/6837 [15:20<36:48,  2.84it/s]
  8%|▊         | 567/6837 [15:20<35:08,  2.97it/s]
  8%|▊         | 568/6837 [15:20<35:27,  2.95it/s]
  8%|▊         | 569/6837 [15:21<33:44,  3.10it/s]
  8%|▊         | 570/6837 [15:21<33:20,  3.13it/s]
  8%|▊         | 571/6837 [15:21<32:49,  3.18it/s]
  8%|▊         | 572/6837 [15:22<32:55,  3.17it/s]
  8%|▊         | 573/6837 [15:22<32:22,  3.23it/s]
  8%|▊         | 574/6837 [15:22<32:28,  3.21it/s]
  8%|▊         | 575/6837 [15:23<33:05,  3.15it/s]{'loss': 5.2456, 'grad_norm': 0.26077064871788025, 'learning_rate': 0.005043859649122807, 'epoch': 0.02}
                                                  
+
  8%|▊         | 575/6837 [15:23<33:05,  3.15it/s]
  8%|▊         | 576/6837 [15:23<38:03,  2.74it/s]
  8%|▊         | 577/6837 [15:23<38:01,  2.74it/s]
  8%|▊         | 578/6837 [15:24<35:58,  2.90it/s]
  8%|▊         | 579/6837 [15:24<36:17,  2.87it/s]
  8%|▊         | 580/6837 [15:24<34:29,  3.02it/s]
  8%|▊         | 581/6837 [15:25<34:55,  2.98it/s]
  9%|▊         | 582/6837 [15:25<34:17,  3.04it/s]
  9%|▊         | 583/6837 [15:25<34:21,  3.03it/s]
  9%|▊         | 584/6837 [15:26<34:31,  3.02it/s]
  9%|▊         | 585/6837 [15:26<33:24,  3.12it/s]
  9%|▊         | 586/6837 [15:26<32:30,  3.21it/s]
  9%|▊         | 587/6837 [15:27<32:12,  3.23it/s]
  9%|▊         | 588/6837 [15:27<33:34,  3.10it/s]
  9%|▊         | 589/6837 [15:27<33:17,  3.13it/s]
  9%|▊         | 590/6837 [15:28<32:45,  3.18it/s]
  9%|▊         | 591/6837 [15:28<32:25,  3.21it/s]
  9%|▊         | 592/6837 [15:28<31:09,  3.34it/s]
  9%|▊         | 593/6837 [15:28<32:09,  3.24it/s]
  9%|▊         | 594/6837 [15:29<34:41,  3.00it/s]
  9%|▊         | 595/6837 [15:29<33:38,  3.09it/s]
  9%|▊         | 596/6837 [15:29<33:36,  3.09it/s]
  9%|▊         | 597/6837 [15:30<32:26,  3.21it/s]
  9%|▊         | 598/6837 [15:30<34:30,  3.01it/s]
  9%|▉         | 599/6837 [15:30<34:16,  3.03it/s]
  9%|▉         | 600/6837 [15:31<33:04,  3.14it/s]
                                                  
{'loss': 5.2218, 'grad_norm': 0.2542518675327301, 'learning_rate': 0.005263157894736842, 'epoch': 0.02}
+
  9%|▉         | 600/6837 [15:31<33:04,  3.14it/s]
  9%|▉         | 601/6837 [15:31<33:09,  3.14it/s]
  9%|▉         | 602/6837 [15:31<32:11,  3.23it/s]
  9%|▉         | 603/6837 [15:32<33:22,  3.11it/s]
  9%|▉         | 604/6837 [15:32<33:18,  3.12it/s]
  9%|▉         | 605/6837 [15:32<31:56,  3.25it/s]
  9%|▉         | 606/6837 [15:33<32:14,  3.22it/s]
  9%|▉         | 607/6837 [15:33<32:48,  3.16it/s]
  9%|▉         | 608/6837 [15:33<32:11,  3.22it/s]
  9%|▉         | 609/6837 [15:34<31:04,  3.34it/s]
  9%|▉         | 610/6837 [15:34<30:49,  3.37it/s]
  9%|▉         | 611/6837 [15:34<32:41,  3.17it/s]
  9%|▉         | 612/6837 [15:34<32:19,  3.21it/s]
  9%|▉         | 613/6837 [15:35<31:40,  3.27it/s]
  9%|▉         | 614/6837 [15:35<31:54,  3.25it/s]
  9%|▉         | 615/6837 [15:35<31:58,  3.24it/s]
  9%|▉         | 616/6837 [15:36<31:00,  3.34it/s]
  9%|▉         | 617/6837 [15:36<30:44,  3.37it/s]
  9%|▉         | 618/6837 [15:36<29:59,  3.46it/s]
  9%|▉         | 619/6837 [15:37<30:05,  3.44it/s]
  9%|▉         | 620/6837 [15:37<30:09,  3.44it/s]
  9%|▉         | 621/6837 [15:37<29:51,  3.47it/s]
  9%|▉         | 622/6837 [15:37<30:24,  3.41it/s]
  9%|▉         | 623/6837 [15:38<31:17,  3.31it/s]
  9%|▉         | 624/6837 [15:38<31:45,  3.26it/s]
  9%|▉         | 625/6837 [15:38<31:14,  3.31it/s]
                                                  
{'loss': 5.2173, 'grad_norm': 0.24375107884407043, 'learning_rate': 0.005482456140350877, 'epoch': 0.02}
+
  9%|▉         | 625/6837 [15:38<31:14,  3.31it/s]
  9%|▉         | 626/6837 [15:39<30:52,  3.35it/s]
  9%|▉         | 627/6837 [15:39<30:08,  3.43it/s]
  9%|▉         | 628/6837 [15:39<31:57,  3.24it/s]
  9%|▉         | 629/6837 [15:40<31:43,  3.26it/s]
  9%|▉         | 630/6837 [15:40<31:54,  3.24it/s]
  9%|▉         | 631/6837 [15:40<31:55,  3.24it/s]
  9%|▉         | 632/6837 [15:40<32:07,  3.22it/s]
  9%|▉         | 633/6837 [15:41<31:51,  3.24it/s]
  9%|▉         | 634/6837 [15:41<31:35,  3.27it/s]
  9%|▉         | 635/6837 [15:41<31:31,  3.28it/s]
  9%|▉         | 636/6837 [15:42<31:15,  3.31it/s]
  9%|▉         | 637/6837 [15:42<31:29,  3.28it/s]
  9%|▉         | 638/6837 [15:42<31:31,  3.28it/s]
  9%|▉         | 639/6837 [15:43<32:25,  3.19it/s]
  9%|▉         | 640/6837 [15:43<32:49,  3.15it/s]
  9%|▉         | 641/6837 [15:43<34:24,  3.00it/s]
  9%|▉         | 642/6837 [15:44<34:03,  3.03it/s]
  9%|▉         | 643/6837 [15:44<34:03,  3.03it/s]
  9%|▉         | 644/6837 [15:44<33:41,  3.06it/s]
  9%|▉         | 645/6837 [15:45<34:01,  3.03it/s]
  9%|▉         | 646/6837 [15:45<34:39,  2.98it/s]
  9%|▉         | 647/6837 [15:45<34:47,  2.97it/s]
  9%|▉         | 648/6837 [15:46<34:26,  3.00it/s]
  9%|▉         | 649/6837 [15:46<33:43,  3.06it/s]
 10%|▉         | 650/6837 [15:46<33:46,  3.05it/s]
                                                  
{'loss': 5.1708, 'grad_norm': 0.22548918426036835, 'learning_rate': 0.005701754385964913, 'epoch': 0.02}
+
 10%|▉         | 650/6837 [15:46<33:46,  3.05it/s]
 10%|▉         | 651/6837 [15:47<33:17,  3.10it/s]
 10%|▉         | 652/6837 [15:47<41:33,  2.48it/s]
 10%|▉         | 653/6837 [15:48<38:21,  2.69it/s]
 10%|▉         | 654/6837 [15:48<35:21,  2.91it/s]
 10%|▉         | 655/6837 [15:48<33:55,  3.04it/s]
 10%|▉         | 656/6837 [15:48<33:35,  3.07it/s]
 10%|▉         | 657/6837 [15:49<33:02,  3.12it/s]
 10%|▉         | 658/6837 [15:49<31:46,  3.24it/s]
 10%|▉         | 659/6837 [15:49<30:38,  3.36it/s]
 10%|▉         | 660/6837 [15:50<31:04,  3.31it/s]
 10%|▉         | 661/6837 [15:50<30:36,  3.36it/s]
 10%|▉         | 662/6837 [15:50<30:13,  3.40it/s]
 10%|▉         | 663/6837 [15:50<30:21,  3.39it/s]
 10%|▉         | 664/6837 [15:51<30:00,  3.43it/s]
 10%|▉         | 665/6837 [15:51<29:21,  3.50it/s]
 10%|▉         | 666/6837 [15:51<30:09,  3.41it/s]
 10%|▉         | 667/6837 [15:52<29:39,  3.47it/s]
 10%|▉         | 668/6837 [15:52<30:32,  3.37it/s]
 10%|▉         | 669/6837 [15:52<31:18,  3.28it/s]
 10%|▉         | 670/6837 [15:53<30:46,  3.34it/s]
 10%|▉         | 671/6837 [15:53<30:46,  3.34it/s]
 10%|▉         | 672/6837 [15:53<30:05,  3.41it/s]
 10%|▉         | 673/6837 [15:53<30:00,  3.42it/s]
 10%|▉         | 674/6837 [15:54<30:14,  3.40it/s]
 10%|▉         | 675/6837 [15:54<31:47,  3.23it/s]
                                                  
{'loss': 5.17, 'grad_norm': 0.21462322771549225, 'learning_rate': 0.0059210526315789476, 'epoch': 0.02}
+
 10%|▉         | 675/6837 [15:54<31:47,  3.23it/s]
 10%|▉         | 676/6837 [15:54<30:39,  3.35it/s]
 10%|▉         | 677/6837 [15:55<30:03,  3.42it/s]
 10%|▉         | 678/6837 [15:55<30:20,  3.38it/s]
 10%|▉         | 679/6837 [15:55<29:30,  3.48it/s]
 10%|▉         | 680/6837 [15:55<29:03,  3.53it/s]
 10%|▉         | 681/6837 [15:56<28:44,  3.57it/s]
 10%|▉         | 682/6837 [15:56<29:19,  3.50it/s]
 10%|▉         | 683/6837 [15:56<31:06,  3.30it/s]
 10%|█         | 684/6837 [15:57<31:17,  3.28it/s]
 10%|█         | 685/6837 [15:57<30:27,  3.37it/s]
 10%|█         | 686/6837 [15:57<29:55,  3.43it/s]
 10%|█         | 687/6837 [15:58<31:05,  3.30it/s]
 10%|█         | 688/6837 [15:58<30:59,  3.31it/s]
 10%|█         | 689/6837 [15:58<31:15,  3.28it/s]
 10%|█         | 690/6837 [15:58<31:38,  3.24it/s]
 10%|█         | 691/6837 [15:59<32:11,  3.18it/s]
 10%|█         | 692/6837 [15:59<31:01,  3.30it/s]
 10%|█         | 693/6837 [15:59<30:54,  3.31it/s]
 10%|█         | 694/6837 [16:00<30:11,  3.39it/s]
 10%|█         | 695/6837 [16:00<29:53,  3.42it/s]
 10%|█         | 696/6837 [16:00<29:57,  3.42it/s]
 10%|█         | 697/6837 [16:01<29:40,  3.45it/s]
 10%|█         | 698/6837 [16:01<29:37,  3.45it/s]
 10%|█         | 699/6837 [16:01<29:11,  3.50it/s]
 10%|█         | 700/6837 [16:01<30:32,  3.35it/s]
                                                  
{'loss': 5.1409, 'grad_norm': 0.21344247460365295, 'learning_rate': 0.005999899895241028, 'epoch': 0.02}
+
 10%|█         | 700/6837 [16:01<30:32,  3.35it/s]
 10%|█         | 701/6837 [16:02<30:00,  3.41it/s]
 10%|█         | 702/6837 [16:02<29:39,  3.45it/s]
 10%|█         | 703/6837 [16:02<29:17,  3.49it/s]
 10%|█         | 704/6837 [16:03<29:09,  3.51it/s]
 10%|█         | 705/6837 [16:03<29:18,  3.49it/s]
 10%|█         | 706/6837 [16:03<29:20,  3.48it/s]
 10%|█         | 707/6837 [16:03<28:39,  3.57it/s]
 10%|█         | 708/6837 [16:04<28:30,  3.58it/s]
 10%|█         | 709/6837 [16:04<28:08,  3.63it/s]
 10%|█         | 710/6837 [16:04<28:34,  3.57it/s]
 10%|█         | 711/6837 [16:05<30:40,  3.33it/s]
 10%|█         | 712/6837 [16:05<30:14,  3.37it/s]
 10%|█         | 713/6837 [16:05<30:27,  3.35it/s]
 10%|█         | 714/6837 [16:05<30:48,  3.31it/s]
 10%|█         | 715/6837 [16:06<30:03,  3.39it/s]
 10%|█         | 716/6837 [16:06<29:05,  3.51it/s]
 10%|█         | 717/6837 [16:06<28:30,  3.58it/s]
 11%|█         | 718/6837 [16:07<28:20,  3.60it/s]
 11%|█         | 719/6837 [16:07<28:43,  3.55it/s]
 11%|█         | 720/6837 [16:07<28:23,  3.59it/s]
 11%|█         | 721/6837 [16:07<27:52,  3.66it/s]
 11%|█         | 722/6837 [16:08<27:54,  3.65it/s]
 11%|█         | 723/6837 [16:08<28:11,  3.61it/s]
 11%|█         | 724/6837 [16:08<30:09,  3.38it/s]
 11%|█         | 725/6837 [16:09<29:32,  3.45it/s]
                                                  {'loss': 5.1216, 'grad_norm': 0.18738315999507904, 'learning_rate': 0.005999342691833758, 'epoch': 0.02}
+

 11%|█         | 725/6837 [16:09<29:32,  3.45it/s]
 11%|█         | 726/6837 [16:09<30:03,  3.39it/s]
 11%|█         | 727/6837 [16:09<29:29,  3.45it/s]
 11%|█         | 728/6837 [16:09<29:28,  3.45it/s]
 11%|█         | 729/6837 [16:10<29:20,  3.47it/s]
 11%|█         | 730/6837 [16:10<28:32,  3.57it/s]
 11%|█         | 731/6837 [16:10<29:10,  3.49it/s]
 11%|█         | 732/6837 [16:11<28:53,  3.52it/s]
 11%|█         | 733/6837 [16:11<28:22,  3.58it/s]
 11%|█         | 734/6837 [16:11<28:20,  3.59it/s]
 11%|█         | 735/6837 [16:11<28:10,  3.61it/s]
 11%|█         | 736/6837 [16:12<28:58,  3.51it/s]
 11%|█         | 737/6837 [16:12<28:30,  3.57it/s]
 11%|█         | 738/6837 [16:12<28:07,  3.61it/s]
 11%|█         | 739/6837 [16:12<28:31,  3.56it/s]
 11%|█         | 740/6837 [16:13<28:53,  3.52it/s]
 11%|█         | 741/6837 [16:13<29:14,  3.48it/s]
 11%|█         | 742/6837 [16:13<30:34,  3.32it/s]
 11%|█         | 743/6837 [16:14<32:11,  3.16it/s]
 11%|█         | 744/6837 [16:14<31:06,  3.26it/s]
 11%|█         | 745/6837 [16:14<31:12,  3.25it/s]
 11%|█         | 746/6837 [16:15<30:38,  3.31it/s]
 11%|█         | 747/6837 [16:15<29:44,  3.41it/s]
 11%|█         | 748/6837 [16:15<29:49,  3.40it/s]
 11%|█         | 749/6837 [16:16<29:54,  3.39it/s]
 11%|█         | 750/6837 [16:16<30:47,  3.30it/s]
                                                  
{'loss': 5.0885, 'grad_norm': 0.20718792080879211, 'learning_rate': 0.005998296806671316, 'epoch': 0.02}
+
 11%|█         | 750/6837 [16:16<30:47,  3.30it/s]
 11%|█         | 751/6837 [16:16<30:15,  3.35it/s]
 11%|█         | 752/6837 [16:16<29:10,  3.48it/s]
 11%|█         | 753/6837 [16:17<29:15,  3.47it/s]
 11%|█         | 754/6837 [16:17<28:30,  3.56it/s]
 11%|█         | 755/6837 [16:17<28:08,  3.60it/s]
 11%|█         | 756/6837 [16:17<27:22,  3.70it/s]
 11%|█         | 757/6837 [16:18<28:29,  3.56it/s]
 11%|█         | 758/6837 [16:18<28:12,  3.59it/s]
 11%|█         | 759/6837 [16:18<27:59,  3.62it/s]
 11%|█         | 760/6837 [16:19<28:00,  3.62it/s]
 11%|█         | 761/6837 [16:19<27:50,  3.64it/s]
 11%|█         | 762/6837 [16:19<28:23,  3.57it/s]
 11%|█         | 763/6837 [16:19<28:41,  3.53it/s]
 11%|█         | 764/6837 [16:20<28:20,  3.57it/s]
 11%|█         | 765/6837 [16:20<29:23,  3.44it/s]
 11%|█         | 766/6837 [16:20<30:01,  3.37it/s]
 11%|█         | 767/6837 [16:21<29:23,  3.44it/s]
 11%|█         | 768/6837 [16:21<29:52,  3.39it/s]
 11%|█         | 769/6837 [16:21<30:01,  3.37it/s]
 11%|█▏        | 770/6837 [16:21<29:07,  3.47it/s]
 11%|█▏        | 771/6837 [16:22<28:55,  3.50it/s]
 11%|█▏        | 772/6837 [16:22<29:39,  3.41it/s]
 11%|█▏        | 773/6837 [16:22<28:42,  3.52it/s]
 11%|█▏        | 774/6837 [16:23<28:29,  3.55it/s]
 11%|█▏        | 775/6837 [16:23<28:08,  3.59it/s]
                                                  
{'loss': 5.0715, 'grad_norm': 0.2014467865228653, 'learning_rate': 0.005996762410159371, 'epoch': 0.02}
+
 11%|█▏        | 775/6837 [16:23<28:08,  3.59it/s]
 11%|█▏        | 776/6837 [16:23<28:03,  3.60it/s]
 11%|█▏        | 777/6837 [16:23<27:31,  3.67it/s]
 11%|█▏        | 778/6837 [16:24<27:05,  3.73it/s]
 11%|█▏        | 779/6837 [16:24<26:59,  3.74it/s]
 11%|█▏        | 780/6837 [16:24<27:41,  3.65it/s]
 11%|█▏        | 781/6837 [16:25<27:23,  3.69it/s]
 11%|█▏        | 782/6837 [16:25<27:00,  3.74it/s]
 11%|█▏        | 783/6837 [16:25<27:45,  3.63it/s]
 11%|█▏        | 784/6837 [16:25<27:53,  3.62it/s]
 11%|█▏        | 785/6837 [16:26<28:25,  3.55it/s]
 11%|█▏        | 786/6837 [16:26<28:11,  3.58it/s]
 12%|█▏        | 787/6837 [16:26<27:51,  3.62it/s]
 12%|█▏        | 788/6837 [16:26<29:23,  3.43it/s]
 12%|█▏        | 789/6837 [16:27<28:48,  3.50it/s]
 12%|█▏        | 790/6837 [16:27<28:19,  3.56it/s]
 12%|█▏        | 791/6837 [16:27<28:28,  3.54it/s]
 12%|█▏        | 792/6837 [16:28<28:01,  3.60it/s]
 12%|█▏        | 793/6837 [16:28<27:43,  3.63it/s]
 12%|█▏        | 794/6837 [16:28<28:48,  3.50it/s]
 12%|█▏        | 795/6837 [16:28<28:23,  3.55it/s]
 12%|█▏        | 796/6837 [16:29<28:21,  3.55it/s]
 12%|█▏        | 797/6837 [16:29<29:27,  3.42it/s]
 12%|█▏        | 798/6837 [16:29<28:57,  3.48it/s]
 12%|█▏        | 799/6837 [16:30<28:13,  3.57it/s]
 12%|█▏        | 800/6837 [16:30<27:54,  3.60it/s]
                                                  
{'loss': 5.0601, 'grad_norm': 0.20796576142311096, 'learning_rate': 0.005994739752296558, 'epoch': 0.02}
+
 12%|█▏        | 800/6837 [16:30<27:54,  3.60it/s]
 12%|█▏        | 801/6837 [16:30<27:54,  3.60it/s]
 12%|█▏        | 802/6837 [16:30<28:03,  3.58it/s]
 12%|█▏        | 803/6837 [16:31<27:38,  3.64it/s]
 12%|█▏        | 804/6837 [16:31<27:09,  3.70it/s]
 12%|█▏        | 805/6837 [16:31<27:08,  3.70it/s]
 12%|█▏        | 806/6837 [16:31<26:53,  3.74it/s]
 12%|█▏        | 807/6837 [16:32<26:54,  3.73it/s]
 12%|█▏        | 808/6837 [16:32<26:43,  3.76it/s]
 12%|█▏        | 809/6837 [16:32<27:12,  3.69it/s]
 12%|█▏        | 810/6837 [16:33<28:03,  3.58it/s]
 12%|█▏        | 811/6837 [16:33<28:10,  3.56it/s]
 12%|█▏        | 812/6837 [16:33<29:56,  3.35it/s]
 12%|█▏        | 813/6837 [16:33<29:05,  3.45it/s]
 12%|█▏        | 814/6837 [16:34<28:20,  3.54it/s]
 12%|█▏        | 815/6837 [16:34<27:52,  3.60it/s]
 12%|█▏        | 816/6837 [16:34<27:34,  3.64it/s]
 12%|█▏        | 817/6837 [16:35<27:42,  3.62it/s]
 12%|█▏        | 818/6837 [16:35<27:57,  3.59it/s]
 12%|█▏        | 819/6837 [16:35<28:27,  3.52it/s]
 12%|█▏        | 820/6837 [16:35<27:31,  3.64it/s]
 12%|█▏        | 821/6837 [16:36<26:55,  3.72it/s]
 12%|█▏        | 822/6837 [16:36<26:22,  3.80it/s]
 12%|█▏        | 823/6837 [16:36<26:01,  3.85it/s]
 12%|█▏        | 824/6837 [16:36<26:46,  3.74it/s]
 12%|█▏        | 825/6837 [16:37<26:59,  3.71it/s]
                                                  
{'loss': 5.0282, 'grad_norm': 0.18012750148773193, 'learning_rate': 0.00599222916263375, 'epoch': 0.02}
+
 12%|█▏        | 825/6837 [16:37<26:59,  3.71it/s]
 12%|█▏        | 826/6837 [16:37<27:32,  3.64it/s]
 12%|█▏        | 827/6837 [16:37<28:14,  3.55it/s]
 12%|█▏        | 828/6837 [16:38<27:39,  3.62it/s]
 12%|█▏        | 829/6837 [16:38<27:17,  3.67it/s]
 12%|█▏        | 830/6837 [16:38<26:49,  3.73it/s]
 12%|█▏        | 831/6837 [16:38<27:08,  3.69it/s]
 12%|█▏        | 832/6837 [16:39<26:46,  3.74it/s]
 12%|█▏        | 833/6837 [16:39<27:14,  3.67it/s]
 12%|█▏        | 834/6837 [16:39<26:51,  3.72it/s]
 12%|█▏        | 835/6837 [16:39<27:32,  3.63it/s]
 12%|█▏        | 836/6837 [16:40<27:14,  3.67it/s]
 12%|█▏        | 837/6837 [16:40<28:50,  3.47it/s]
 12%|█▏        | 838/6837 [16:40<28:45,  3.48it/s]
 12%|█▏        | 839/6837 [16:41<29:29,  3.39it/s]
 12%|█▏        | 840/6837 [16:41<29:29,  3.39it/s]
 12%|█▏        | 841/6837 [16:41<29:58,  3.33it/s]
 12%|█▏        | 842/6837 [16:42<30:25,  3.28it/s]
 12%|█▏        | 843/6837 [16:42<30:20,  3.29it/s]
 12%|█▏        | 844/6837 [16:42<31:22,  3.18it/s]
 12%|█▏        | 845/6837 [16:42<29:41,  3.36it/s]
 12%|█▏        | 846/6837 [16:43<29:55,  3.34it/s]
 12%|█▏        | 847/6837 [16:43<31:12,  3.20it/s]
 12%|█▏        | 848/6837 [16:43<30:41,  3.25it/s]
 12%|█▏        | 849/6837 [16:44<30:23,  3.28it/s]
 12%|█▏        | 850/6837 [16:44<30:24,  3.28it/s]
                                                  
{'loss': 5.0196, 'grad_norm': 0.1821032166481018, 'learning_rate': 0.005989231050220356, 'epoch': 0.02}
+
 12%|█▏        | 850/6837 [16:44<30:24,  3.28it/s]
 12%|█▏        | 851/6837 [16:44<30:39,  3.25it/s]
 12%|█▏        | 852/6837 [16:45<30:26,  3.28it/s]
 12%|█▏        | 853/6837 [16:45<30:01,  3.32it/s]
 12%|█▏        | 854/6837 [16:45<31:07,  3.20it/s]
 13%|█▎        | 855/6837 [16:46<30:56,  3.22it/s]
 13%|█▎        | 856/6837 [16:46<29:24,  3.39it/s]
 13%|█▎        | 857/6837 [16:46<29:01,  3.43it/s]
 13%|█▎        | 858/6837 [16:46<28:06,  3.54it/s]
 13%|█▎        | 859/6837 [16:47<27:50,  3.58it/s]
 13%|█▎        | 860/6837 [16:47<27:27,  3.63it/s]
 13%|█▎        | 861/6837 [16:47<27:34,  3.61it/s]
 13%|█▎        | 862/6837 [16:47<27:36,  3.61it/s]
 13%|█▎        | 863/6837 [16:48<27:16,  3.65it/s]
 13%|█▎        | 864/6837 [16:48<27:35,  3.61it/s]
 13%|█▎        | 865/6837 [16:48<27:47,  3.58it/s]
 13%|█▎        | 866/6837 [16:49<27:50,  3.57it/s]
 13%|█▎        | 867/6837 [16:49<27:55,  3.56it/s]
 13%|█▎        | 868/6837 [16:49<28:01,  3.55it/s]
 13%|█▎        | 869/6837 [16:49<28:17,  3.52it/s]
 13%|█▎        | 870/6837 [16:50<27:58,  3.55it/s]
 13%|█▎        | 871/6837 [16:50<28:04,  3.54it/s]
 13%|█▎        | 872/6837 [16:50<27:29,  3.62it/s]
 13%|█▎        | 873/6837 [16:51<28:01,  3.55it/s]
 13%|█▎        | 874/6837 [16:51<27:26,  3.62it/s]
 13%|█▎        | 875/6837 [16:51<27:18,  3.64it/s]
                                                  
{'loss': 4.9863, 'grad_norm': 0.16888366639614105, 'learning_rate': 0.005985745903537683, 'epoch': 0.03}
+
 13%|█▎        | 875/6837 [16:51<27:18,  3.64it/s]
 13%|█▎        | 876/6837 [16:51<26:51,  3.70it/s]
 13%|█▎        | 877/6837 [16:52<26:30,  3.75it/s]
 13%|█▎        | 878/6837 [16:52<26:44,  3.71it/s]
 13%|█▎        | 879/6837 [16:52<26:51,  3.70it/s]
 13%|█▎        | 880/6837 [16:52<27:14,  3.65it/s]
 13%|█▎        | 881/6837 [16:53<26:59,  3.68it/s]
 13%|█▎        | 882/6837 [16:53<26:39,  3.72it/s]
 13%|█▎        | 883/6837 [16:53<26:52,  3.69it/s]
 13%|█▎        | 884/6837 [16:54<27:45,  3.57it/s]
 13%|█▎        | 885/6837 [16:54<27:22,  3.62it/s]
 13%|█▎        | 886/6837 [16:54<27:14,  3.64it/s]
 13%|█▎        | 887/6837 [16:54<26:50,  3.70it/s]
 13%|█▎        | 888/6837 [16:55<27:39,  3.58it/s]
 13%|█▎        | 889/6837 [16:55<27:39,  3.58it/s]
 13%|█▎        | 890/6837 [16:55<27:42,  3.58it/s]
 13%|█▎        | 891/6837 [16:55<27:28,  3.61it/s]
 13%|█▎        | 892/6837 [16:56<27:23,  3.62it/s]
 13%|█▎        | 893/6837 [16:56<26:50,  3.69it/s]
 13%|█▎        | 894/6837 [16:56<27:15,  3.63it/s]
 13%|█▎        | 895/6837 [16:57<27:04,  3.66it/s]
 13%|█▎        | 896/6837 [16:57<26:40,  3.71it/s]
 13%|█▎        | 897/6837 [16:57<26:54,  3.68it/s]
 13%|█▎        | 898/6837 [16:57<26:51,  3.69it/s]
 13%|█▎        | 899/6837 [16:58<26:48,  3.69it/s]
 13%|█▎        | 900/6837 [16:58<26:38,  3.71it/s]
                                                  
{'loss': 4.9737, 'grad_norm': 0.17831626534461975, 'learning_rate': 0.005981774290419344, 'epoch': 0.03}
+
 13%|█▎        | 900/6837 [16:58<26:38,  3.71it/s]
 13%|█▎        | 901/6837 [16:58<27:10,  3.64it/s]
 13%|█▎        | 902/6837 [16:58<26:56,  3.67it/s]
 13%|█▎        | 903/6837 [16:59<26:49,  3.69it/s]
 13%|█▎        | 904/6837 [16:59<26:53,  3.68it/s]
 13%|█▎        | 905/6837 [16:59<26:37,  3.71it/s]
 13%|█▎        | 906/6837 [17:00<26:55,  3.67it/s]
 13%|█▎        | 907/6837 [17:00<27:05,  3.65it/s]
 13%|█▎        | 908/6837 [17:00<27:06,  3.65it/s]
 13%|█▎        | 909/6837 [17:00<26:38,  3.71it/s]
 13%|█▎        | 910/6837 [17:01<26:58,  3.66it/s]
 13%|█▎        | 911/6837 [17:01<27:20,  3.61it/s]
 13%|█▎        | 912/6837 [17:01<27:19,  3.61it/s]
 13%|█▎        | 913/6837 [17:01<27:09,  3.64it/s]
 13%|█▎        | 914/6837 [17:02<26:31,  3.72it/s]
 13%|█▎        | 915/6837 [17:02<26:26,  3.73it/s]
 13%|█▎        | 916/6837 [17:02<26:20,  3.75it/s]
 13%|█▎        | 917/6837 [17:03<26:18,  3.75it/s]
 13%|█▎        | 918/6837 [17:03<26:40,  3.70it/s]
 13%|█▎        | 919/6837 [17:03<26:46,  3.68it/s]
 13%|█▎        | 920/6837 [17:03<26:35,  3.71it/s]
 13%|█▎        | 921/6837 [17:04<26:01,  3.79it/s]
 13%|█▎        | 922/6837 [17:04<26:45,  3.69it/s]
 14%|█▎        | 923/6837 [17:04<26:43,  3.69it/s]
 14%|█▎        | 924/6837 [17:04<27:20,  3.60it/s]
 14%|█▎        | 925/6837 [17:05<27:19,  3.61it/s]{'loss': 4.9537, 'grad_norm': 0.19819270074367523, 'learning_rate': 0.005977316857958743, 'epoch': 0.03}
                                                  
+
 14%|█▎        | 925/6837 [17:05<27:19,  3.61it/s]
 14%|█▎        | 926/6837 [17:05<27:04,  3.64it/s]
 14%|█▎        | 927/6837 [17:05<26:41,  3.69it/s]
 14%|█▎        | 928/6837 [17:06<27:31,  3.58it/s]
 14%|█▎        | 929/6837 [17:06<27:31,  3.58it/s]
 14%|█▎        | 930/6837 [17:06<27:10,  3.62it/s]
 14%|█▎        | 931/6837 [17:06<27:16,  3.61it/s]
 14%|█▎        | 932/6837 [17:07<26:37,  3.70it/s]
 14%|█▎        | 933/6837 [17:07<26:09,  3.76it/s]
 14%|█▎        | 934/6837 [17:07<26:29,  3.71it/s]
 14%|█▎        | 935/6837 [17:07<26:20,  3.73it/s]
 14%|█▎        | 936/6837 [17:08<26:56,  3.65it/s]
 14%|█▎        | 937/6837 [17:08<26:19,  3.74it/s]
 14%|█▎        | 938/6837 [17:08<26:52,  3.66it/s]
 14%|█▎        | 939/6837 [17:08<26:20,  3.73it/s]
 14%|█▎        | 940/6837 [17:09<26:26,  3.72it/s]
 14%|█▍        | 941/6837 [17:09<26:29,  3.71it/s]
 14%|█▍        | 942/6837 [17:09<27:15,  3.60it/s]
 14%|█▍        | 943/6837 [17:10<26:54,  3.65it/s]
 14%|█▍        | 944/6837 [17:10<27:29,  3.57it/s]
 14%|█▍        | 945/6837 [17:10<28:24,  3.46it/s]
 14%|█▍        | 946/6837 [17:10<28:26,  3.45it/s]
 14%|█▍        | 947/6837 [17:11<28:15,  3.47it/s]
 14%|█▍        | 948/6837 [17:11<28:29,  3.45it/s]
 14%|█▍        | 949/6837 [17:11<27:44,  3.54it/s]
 14%|█▍        | 950/6837 [17:12<27:33,  3.56it/s]
                                                  
{'loss': 4.9608, 'grad_norm': 0.17900893092155457, 'learning_rate': 0.005972374332403642, 'epoch': 0.03}
+
 14%|█▍        | 950/6837 [17:12<27:33,  3.56it/s]
 14%|█▍        | 951/6837 [17:12<27:09,  3.61it/s]
 14%|█▍        | 952/6837 [17:12<26:38,  3.68it/s]
 14%|█▍        | 953/6837 [17:12<26:27,  3.71it/s]
 14%|█▍        | 954/6837 [17:13<26:28,  3.70it/s]
 14%|█▍        | 955/6837 [17:13<26:48,  3.66it/s]
 14%|█▍        | 956/6837 [17:13<26:09,  3.75it/s]
 14%|█▍        | 957/6837 [17:13<25:40,  3.82it/s]
 14%|█▍        | 958/6837 [17:14<25:46,  3.80it/s]
 14%|█▍        | 959/6837 [17:14<25:49,  3.79it/s]
 14%|█▍        | 960/6837 [17:14<25:53,  3.78it/s]
 14%|█▍        | 961/6837 [17:15<26:14,  3.73it/s]
 14%|█▍        | 962/6837 [17:15<26:16,  3.73it/s]
 14%|█▍        | 963/6837 [17:15<26:01,  3.76it/s]
 14%|█▍        | 964/6837 [17:15<25:38,  3.82it/s]
 14%|█▍        | 965/6837 [17:16<26:49,  3.65it/s]
 14%|█▍        | 966/6837 [17:16<26:45,  3.66it/s]
 14%|█▍        | 967/6837 [17:16<26:46,  3.65it/s]
 14%|█▍        | 968/6837 [17:16<26:51,  3.64it/s]
 14%|█▍        | 969/6837 [17:17<26:58,  3.62it/s]
 14%|█▍        | 970/6837 [17:17<26:25,  3.70it/s]
 14%|█▍        | 971/6837 [17:17<26:01,  3.76it/s]
 14%|█▍        | 972/6837 [17:18<26:13,  3.73it/s]
 14%|█▍        | 973/6837 [17:18<26:00,  3.76it/s]
 14%|█▍        | 974/6837 [17:18<25:36,  3.81it/s]
 14%|█▍        | 975/6837 [17:18<25:44,  3.79it/s]
                                                  
{'loss': 4.9347, 'grad_norm': 0.1771428883075714, 'learning_rate': 0.005966947519037834, 'epoch': 0.03}
+
 14%|█▍        | 975/6837 [17:18<25:44,  3.79it/s]
 14%|█▍        | 976/6837 [17:19<25:47,  3.79it/s]
 14%|█▍        | 977/6837 [17:19<25:28,  3.83it/s]
 14%|█▍        | 978/6837 [17:19<25:15,  3.87it/s]
 14%|█▍        | 979/6837 [17:19<25:22,  3.85it/s]
 14%|█▍        | 980/6837 [17:20<25:25,  3.84it/s]
 14%|█▍        | 981/6837 [17:20<25:12,  3.87it/s]
 14%|█▍        | 982/6837 [17:20<25:13,  3.87it/s]
 14%|█▍        | 983/6837 [17:20<26:46,  3.64it/s]
 14%|█▍        | 984/6837 [17:21<26:45,  3.64it/s]
 14%|█▍        | 985/6837 [17:21<26:01,  3.75it/s]
 14%|█▍        | 986/6837 [17:21<25:28,  3.83it/s]
 14%|█▍        | 987/6837 [17:21<25:13,  3.87it/s]
 14%|█▍        | 988/6837 [17:22<24:52,  3.92it/s]
 14%|█▍        | 989/6837 [17:22<24:53,  3.92it/s]
 14%|█▍        | 990/6837 [17:22<24:48,  3.93it/s]
 14%|█▍        | 991/6837 [17:22<25:23,  3.84it/s]
 15%|█▍        | 992/6837 [17:23<25:12,  3.86it/s]
 15%|█▍        | 993/6837 [17:23<25:09,  3.87it/s]
 15%|█▍        | 994/6837 [17:23<25:09,  3.87it/s]
 15%|█▍        | 995/6837 [17:23<25:08,  3.87it/s]
 15%|█▍        | 996/6837 [17:24<24:48,  3.92it/s]
 15%|█▍        | 997/6837 [17:24<24:46,  3.93it/s]
 15%|█▍        | 998/6837 [17:24<24:41,  3.94it/s]
 15%|█▍        | 999/6837 [17:25<24:45,  3.93it/s]
 15%|█▍        | 1000/6837 [17:25<24:57,  3.90it/s]
                                                   
{'loss': 4.9352, 'grad_norm': 0.15494541823863983, 'learning_rate': 0.005961037302049941, 'epoch': 0.03}
+
 15%|█▍        | 1000/6837 [17:25<24:57,  3.90it/s]
 15%|█▍        | 1001/6837 [17:25<25:01,  3.89it/s]
 15%|█▍        | 1002/6837 [17:25<24:51,  3.91it/s]
 15%|█▍        | 1003/6837 [17:26<24:35,  3.95it/s]
 15%|█▍        | 1004/6837 [17:26<24:41,  3.94it/s]
 15%|█▍        | 1005/6837 [17:26<24:52,  3.91it/s]
 15%|█▍        | 1006/6837 [17:26<25:31,  3.81it/s]
 15%|█▍        | 1007/6837 [17:27<25:07,  3.87it/s]
 15%|█▍        | 1008/6837 [17:27<25:20,  3.83it/s]
 15%|█▍        | 1009/6837 [17:27<25:06,  3.87it/s]
 15%|█▍        | 1010/6837 [17:27<24:55,  3.90it/s]
 15%|█▍        | 1011/6837 [17:28<24:45,  3.92it/s]
 15%|█▍        | 1012/6837 [17:28<24:44,  3.92it/s]
 15%|█▍        | 1013/6837 [17:28<24:39,  3.94it/s]
 15%|█▍        | 1014/6837 [17:28<24:30,  3.96it/s]
 15%|█▍        | 1015/6837 [17:29<24:41,  3.93it/s]
 15%|█▍        | 1016/6837 [17:29<24:43,  3.92it/s]
 15%|█▍        | 1017/6837 [17:29<25:18,  3.83it/s]
 15%|█▍        | 1018/6837 [17:29<24:59,  3.88it/s]
 15%|█▍        | 1019/6837 [17:30<25:38,  3.78it/s]
 15%|█▍        | 1020/6837 [17:30<25:20,  3.83it/s]
 15%|█▍        | 1021/6837 [17:30<26:24,  3.67it/s]
 15%|█▍        | 1022/6837 [17:30<26:24,  3.67it/s]
 15%|█▍        | 1023/6837 [17:31<25:40,  3.77it/s]
 15%|█▍        | 1024/6837 [17:31<25:24,  3.81it/s]
 15%|█▍        | 1025/6837 [17:31<25:11,  3.84it/s]
                                                   
{'loss': 4.9, 'grad_norm': 0.15592190623283386, 'learning_rate': 0.005954644644389354, 'epoch': 0.03}
+
 15%|█▍        | 1025/6837 [17:31<25:11,  3.84it/s]
 15%|█▌        | 1026/6837 [17:32<25:02,  3.87it/s]
 15%|█▌        | 1027/6837 [17:32<25:11,  3.84it/s]
 15%|█▌        | 1028/6837 [17:32<25:13,  3.84it/s]
 15%|█▌        | 1029/6837 [17:32<24:52,  3.89it/s]
 15%|█▌        | 1030/6837 [17:33<24:55,  3.88it/s]
 15%|█▌        | 1031/6837 [17:33<24:48,  3.90it/s]
 15%|█▌        | 1032/6837 [17:33<24:42,  3.91it/s]
 15%|█▌        | 1033/6837 [17:33<24:28,  3.95it/s]
 15%|█▌        | 1034/6837 [17:34<25:05,  3.86it/s]
 15%|█▌        | 1035/6837 [17:34<24:45,  3.90it/s]
 15%|█▌        | 1036/6837 [17:34<24:37,  3.93it/s]
 15%|█▌        | 1037/6837 [17:34<24:29,  3.95it/s]
 15%|█▌        | 1038/6837 [17:35<24:35,  3.93it/s]
 15%|█▌        | 1039/6837 [17:35<24:47,  3.90it/s]
 15%|█▌        | 1040/6837 [17:35<24:45,  3.90it/s]
 15%|█▌        | 1041/6837 [17:35<24:45,  3.90it/s]
 15%|█▌        | 1042/6837 [17:36<24:43,  3.91it/s]
 15%|█▌        | 1043/6837 [17:36<24:42,  3.91it/s]
 15%|█▌        | 1044/6837 [17:36<24:53,  3.88it/s]
 15%|█▌        | 1045/6837 [17:36<24:54,  3.88it/s]
 15%|█▌        | 1046/6837 [17:37<24:39,  3.91it/s]
 15%|█▌        | 1047/6837 [17:37<24:48,  3.89it/s]
 15%|█▌        | 1048/6837 [17:37<24:44,  3.90it/s]
 15%|█▌        | 1049/6837 [17:37<24:35,  3.92it/s]
 15%|█▌        | 1050/6837 [17:38<24:32,  3.93it/s]{'loss': 4.8978, 'grad_norm': 0.15409070253372192, 'learning_rate': 0.0059477705876093325, 'epoch': 0.03}
+
                                                   

 15%|█▌        | 1050/6837 [17:38<24:32,  3.93it/s]
 15%|█▌        | 1051/6837 [17:38<25:27,  3.79it/s]
 15%|█▌        | 1052/6837 [17:38<25:24,  3.80it/s]
 15%|█▌        | 1053/6837 [17:38<24:55,  3.87it/s]
 15%|█▌        | 1054/6837 [17:39<25:12,  3.82it/s]
 15%|█▌        | 1055/6837 [17:39<24:53,  3.87it/s]
 15%|█▌        | 1056/6837 [17:39<24:38,  3.91it/s]
 15%|█▌        | 1057/6837 [17:39<24:28,  3.94it/s]
 15%|█▌        | 1058/6837 [17:40<24:35,  3.92it/s]
 15%|█▌        | 1059/6837 [17:40<24:30,  3.93it/s]
 16%|█▌        | 1060/6837 [17:40<24:19,  3.96it/s]
 16%|█▌        | 1061/6837 [17:40<24:14,  3.97it/s]
 16%|█▌        | 1062/6837 [17:41<24:08,  3.99it/s]
 16%|█▌        | 1063/6837 [17:41<24:06,  3.99it/s]
 16%|█▌        | 1064/6837 [17:41<24:07,  3.99it/s]
 16%|█▌        | 1065/6837 [17:41<23:59,  4.01it/s]
 16%|█���        | 1066/6837 [17:42<23:54,  4.02it/s]
 16%|█▌        | 1067/6837 [17:42<23:48,  4.04it/s]
 16%|█▌        | 1068/6837 [17:42<23:52,  4.03it/s]
 16%|█▌        | 1069/6837 [17:42<24:11,  3.97it/s]
 16%|█▌        | 1070/6837 [17:43<25:07,  3.83it/s]
 16%|█▌        | 1071/6837 [17:43<25:26,  3.78it/s]
 16%|█▌        | 1072/6837 [17:43<25:52,  3.71it/s]
 16%|█▌        | 1073/6837 [17:44<26:10,  3.67it/s]
 16%|█▌        | 1074/6837 [17:44<25:45,  3.73it/s]
 16%|█▌        | 1075/6837 [17:44<25:08,  3.82it/s]
                                                   
{'loss': 4.8736, 'grad_norm': 0.160922572016716, 'learning_rate': 0.0059404162516973145, 'epoch': 0.03}
+
 16%|█▌        | 1075/6837 [17:44<25:08,  3.82it/s]
 16%|█▌        | 1076/6837 [17:44<24:56,  3.85it/s]
 16%|█▌        | 1077/6837 [17:45<24:43,  3.88it/s]
 16%|█▌        | 1078/6837 [17:45<24:35,  3.90it/s]
 16%|█▌        | 1079/6837 [17:45<24:14,  3.96it/s]
 16%|█▌        | 1080/6837 [17:45<24:03,  3.99it/s]
 16%|█▌        | 1081/6837 [17:46<24:29,  3.92it/s]
 16%|█▌        | 1082/6837 [17:46<24:30,  3.91it/s]
 16%|█▌        | 1083/6837 [17:46<24:18,  3.94it/s]
 16%|█▌        | 1084/6837 [17:46<24:13,  3.96it/s]
 16%|█▌        | 1085/6837 [17:47<24:17,  3.95it/s]
 16%|█▌        | 1086/6837 [17:47<24:21,  3.94it/s]
 16%|█▌        | 1087/6837 [17:47<24:02,  3.99it/s]
 16%|█▌        | 1088/6837 [17:47<23:58,  4.00it/s]
 16%|█▌        | 1089/6837 [17:48<24:04,  3.98it/s]
 16%|█▌        | 1090/6837 [17:48<24:41,  3.88it/s]
 16%|█▌        | 1091/6837 [17:48<24:19,  3.94it/s]
 16%|█▌        | 1092/6837 [17:48<24:26,  3.92it/s]
 16%|█▌        | 1093/6837 [17:49<24:25,  3.92it/s]
 16%|█▌        | 1094/6837 [17:49<24:12,  3.95it/s]
 16%|█▌        | 1095/6837 [17:49<24:06,  3.97it/s]
 16%|█▌        | 1096/6837 [17:49<23:54,  4.00it/s]
 16%|█▌        | 1097/6837 [17:50<23:51,  4.01it/s]
 16%|█▌        | 1098/6837 [17:50<24:17,  3.94it/s]
 16%|█▌        | 1099/6837 [17:50<24:19,  3.93it/s]
 16%|█▌        | 1100/6837 [17:50<24:11,  3.95it/s]{'loss': 4.8651, 'grad_norm': 0.1596185714006424, 'learning_rate': 0.005932582834892431, 'epoch': 0.03}
                                                   
+
 16%|█▌        | 1100/6837 [17:50<24:11,  3.95it/s]
 16%|█▌        | 1101/6837 [17:51<24:04,  3.97it/s]
 16%|█▌        | 1102/6837 [17:51<23:50,  4.01it/s]
 16%|█▌        | 1103/6837 [17:51<23:46,  4.02it/s]
 16%|█▌        | 1104/6837 [17:51<23:54,  4.00it/s]
 16%|█▌        | 1105/6837 [17:52<24:00,  3.98it/s]
 16%|█▌        | 1106/6837 [17:52<23:50,  4.01it/s]
 16%|█▌        | 1107/6837 [17:52<23:59,  3.98it/s]
 16%|█▌        | 1108/6837 [17:52<24:09,  3.95it/s]
 16%|█▌        | 1109/6837 [17:53<24:00,  3.98it/s]
 16%|█▌        | 1110/6837 [17:53<23:57,  3.98it/s]
 16%|█▌        | 1111/6837 [17:53<23:52,  4.00it/s]
 16%|█▋        | 1112/6837 [17:53<24:30,  3.89it/s]
 16%|█▋        | 1113/6837 [17:54<24:22,  3.91it/s]
 16%|█▋        | 1114/6837 [17:54<24:02,  3.97it/s]
 16%|█▋        | 1115/6837 [17:54<23:53,  3.99it/s]
 16%|█▋        | 1116/6837 [17:54<23:52,  3.99it/s]
 16%|█▋        | 1117/6837 [17:55<23:49,  4.00it/s]
 16%|█▋        | 1118/6837 [17:55<23:50,  4.00it/s]
 16%|█▋        | 1119/6837 [17:55<23:59,  3.97it/s]
 16%|█▋        | 1120/6837 [17:55<23:54,  3.98it/s]
 16%|█▋        | 1121/6837 [17:56<23:44,  4.01it/s]
 16%|█▋        | 1122/6837 [17:56<23:49,  4.00it/s]
 16%|█▋        | 1123/6837 [17:56<24:03,  3.96it/s]
 16%|█▋        | 1124/6837 [17:56<24:10,  3.94it/s]
 16%|█▋        | 1125/6837 [17:57<24:02,  3.96it/s]
                                                   
{'loss': 4.8762, 'grad_norm': 0.15350407361984253, 'learning_rate': 0.0059242716134902795, 'epoch': 0.03}
+
 16%|█▋        | 1125/6837 [17:57<24:02,  3.96it/s]
 16%|█▋        | 1126/6837 [17:57<24:06,  3.95it/s]
 16%|█▋        | 1127/6837 [17:57<23:56,  3.98it/s]
 16%|█▋        | 1128/6837 [17:57<23:54,  3.98it/s]
 17%|█▋        | 1129/6837 [17:58<23:59,  3.96it/s]
 17%|█▋        | 1130/6837 [17:58<23:45,  4.00it/s]
 17%|█▋        | 1131/6837 [17:58<23:49,  3.99it/s]
 17%|█▋        | 1132/6837 [17:58<23:48,  3.99it/s]
 17%|█▋        | 1133/6837 [17:59<23:50,  3.99it/s]
 17%|█▋        | 1134/6837 [17:59<23:48,  3.99it/s]
 17%|█▋        | 1135/6837 [17:59<24:23,  3.90it/s]
 17%|█▋        | 1136/6837 [17:59<24:09,  3.93it/s]
 17%|█▋        | 1137/6837 [18:00<24:06,  3.94it/s]
 17%|█▋        | 1138/6837 [18:00<24:07,  3.94it/s]
 17%|█▋        | 1139/6837 [18:00<24:55,  3.81it/s]
 17%|█▋        | 1140/6837 [18:01<25:27,  3.73it/s]
 17%|█▋        | 1141/6837 [18:01<25:47,  3.68it/s]
 17%|█▋        | 1142/6837 [18:01<25:36,  3.71it/s]
 17%|█▋        | 1143/6837 [18:01<24:53,  3.81it/s]
 17%|█▋        | 1144/6837 [18:02<24:35,  3.86it/s]
 17%|█▋        | 1145/6837 [18:02<24:19,  3.90it/s]
 17%|█▋        | 1146/6837 [18:02<24:03,  3.94it/s]
 17%|█▋        | 1147/6837 [18:02<23:47,  3.99it/s]
 17%|█▋        | 1148/6837 [18:03<23:50,  3.98it/s]
 17%|█▋        | 1149/6837 [18:03<23:38,  4.01it/s]
 17%|█▋        | 1150/6837 [18:03<23:41,  4.00it/s]
                                                   
{'loss': 4.853, 'grad_norm': 0.14462624490261078, 'learning_rate': 0.0059154839416349775, 'epoch': 0.03}
+
 17%|█▋        | 1150/6837 [18:03<23:41,  4.00it/s]
 17%|█▋        | 1151/6837 [18:03<23:42,  4.00it/s]
 17%|█▋        | 1152/6837 [18:04<23:38,  4.01it/s]
 17%|█▋        | 1153/6837 [18:04<23:39,  4.00it/s]
 17%|█▋        | 1154/6837 [18:04<23:28,  4.04it/s]
 17%|█▋        | 1155/6837 [18:04<24:06,  3.93it/s]
 17%|█▋        | 1156/6837 [18:05<24:04,  3.93it/s]
 17%|█▋        | 1157/6837 [18:05<23:58,  3.95it/s]
 17%|█▋        | 1158/6837 [18:05<23:44,  3.99it/s]
 17%|█▋        | 1159/6837 [18:05<23:39,  4.00it/s]
 17%|█▋        | 1160/6837 [18:06<23:34,  4.01it/s]
 17%|█▋        | 1161/6837 [18:06<23:37,  4.00it/s]
 17%|█▋        | 1162/6837 [18:06<23:37,  4.00it/s]
 17%|█▋        | 1163/6837 [18:06<24:07,  3.92it/s]
 17%|█▋        | 1164/6837 [18:07<23:49,  3.97it/s]
 17%|█▋        | 1165/6837 [18:07<23:47,  3.97it/s]
 17%|█▋        | 1166/6837 [18:07<23:39,  4.00it/s]
 17%|█▋        | 1167/6837 [18:07<23:41,  3.99it/s]
 17%|█▋        | 1168/6837 [18:08<23:37,  4.00it/s]
 17%|█▋        | 1169/6837 [18:08<23:39,  3.99it/s]
 17%|█▋        | 1170/6837 [18:08<23:36,  4.00it/s]
 17%|█▋        | 1171/6837 [18:08<23:46,  3.97it/s]
 17%|█▋        | 1172/6837 [18:09<23:41,  3.99it/s]
 17%|█▋        | 1173/6837 [18:09<24:20,  3.88it/s]
 17%|█▋        | 1174/6837 [18:09<23:59,  3.93it/s]
 17%|█▋        | 1175/6837 [18:09<24:27,  3.86it/s]
                                                   
{'loss': 4.8587, 'grad_norm': 0.14773821830749512, 'learning_rate': 0.005906221251098531, 'epoch': 0.03}
+
 17%|█▋        | 1175/6837 [18:09<24:27,  3.86it/s]
 17%|█▋        | 1176/6837 [18:10<24:09,  3.90it/s]
 17%|█▋        | 1177/6837 [18:10<24:08,  3.91it/s]
 17%|█▋        | 1178/6837 [18:10<23:59,  3.93it/s]
 17%|█▋        | 1179/6837 [18:10<23:48,  3.96it/s]
 17%|█▋        | 1180/6837 [18:11<23:42,  3.98it/s]
 17%|█▋        | 1181/6837 [18:11<23:49,  3.96it/s]
 17%|█▋        | 1182/6837 [18:11<23:33,  4.00it/s]
 17%|█▋        | 1183/6837 [18:11<23:22,  4.03it/s]
 17%|█▋        | 1184/6837 [18:12<23:19,  4.04it/s]
 17%|█▋        | 1185/6837 [18:12<23:15,  4.05it/s]
 17%|█▋        | 1186/6837 [18:12<23:11,  4.06it/s]
 17%|█▋        | 1187/6837 [18:12<23:11,  4.06it/s]
 17%|█▋        | 1188/6837 [18:13<23:05,  4.08it/s]
 17%|█▋        | 1189/6837 [18:13<23:51,  3.95it/s]
 17%|█▋        | 1190/6837 [18:13<23:43,  3.97it/s]
 17%|█▋        | 1191/6837 [18:13<23:38,  3.98it/s]
 17%|█▋        | 1192/6837 [18:14<23:40,  3.97it/s]
 17%|█▋        | 1193/6837 [18:14<23:35,  3.99it/s]
 17%|█▋        | 1194/6837 [18:14<23:27,  4.01it/s]
 17%|█▋        | 1195/6837 [18:14<23:32,  3.99it/s]
 17%|█▋        | 1196/6837 [18:15<23:20,  4.03it/s]
 18%|█▊        | 1197/6837 [18:15<23:17,  4.04it/s]
 18%|█▊        | 1198/6837 [18:15<23:27,  4.01it/s]
 18%|█▊        | 1199/6837 [18:15<23:31,  4.00it/s]
 18%|█▊        | 1200/6837 [18:16<23:28,  4.00it/s]
                                                   
{'loss': 4.8266, 'grad_norm': 0.15739105641841888, 'learning_rate': 0.005896485051047562, 'epoch': 0.04}
+
 18%|█▊        | 1200/6837 [18:16<23:28,  4.00it/s]
 18%|█▊        | 1201/6837 [18:16<23:26,  4.01it/s]
 18%|█▊        | 1202/6837 [18:16<23:24,  4.01it/s]
 18%|█▊        | 1203/6837 [18:16<23:27,  4.00it/s]
 18%|█▊        | 1204/6837 [18:17<23:28,  4.00it/s]
 18%|█▊        | 1205/6837 [18:17<23:20,  4.02it/s]
 18%|█▊        | 1206/6837 [18:17<23:26,  4.00it/s]
 18%|█▊        | 1207/6837 [18:17<23:30,  3.99it/s]
 18%|█▊        | 1208/6837 [18:18<23:25,  4.00it/s]
 18%|█▊        | 1209/6837 [18:18<23:26,  4.00it/s]
 18%|█▊        | 1210/6837 [18:18<23:24,  4.01it/s]
 18%|█▊        | 1211/6837 [18:18<23:20,  4.02it/s]
 18%|█▊        | 1212/6837 [18:19<23:15,  4.03it/s]
 18%|█▊        | 1213/6837 [18:19<23:09,  4.05it/s]
 18%|█▊        | 1214/6837 [18:19<23:50,  3.93it/s]
 18%|█▊        | 1215/6837 [18:19<23:37,  3.97it/s]
 18%|█▊        | 1216/6837 [18:20<23:26,  4.00it/s]
 18%|█▊        | 1217/6837 [18:20<23:25,  4.00it/s]
 18%|█▊        | 1218/6837 [18:20<23:17,  4.02it/s]
 18%|█▊        | 1219/6837 [18:20<23:12,  4.03it/s]
 18%|█▊        | 1220/6837 [18:21<23:13,  4.03it/s]
 18%|█▊        | 1221/6837 [18:21<23:02,  4.06it/s]
 18%|█▊        | 1222/6837 [18:21<23:01,  4.07it/s]
 18%|█▊        | 1223/6837 [18:21<23:10,  4.04it/s]
 18%|█▊        | 1224/6837 [18:22<23:03,  4.06it/s]
 18%|█▊        | 1225/6837 [18:22<23:01,  4.06it/s]
                                                   {'loss': 4.829, 'grad_norm': 0.14097338914871216, 'learning_rate': 0.0058862769277974085, 'epoch': 0.04}
+
 18%|█▊        | 1225/6837 [18:22<23:01,  4.06it/s]
 18%|█▊        | 1226/6837 [18:22<23:15,  4.02it/s]
 18%|█▊        | 1227/6837 [18:22<23:09,  4.04it/s]
 18%|█▊        | 1228/6837 [18:23<22:58,  4.07it/s]
 18%|█▊        | 1229/6837 [18:23<23:01,  4.06it/s]
 18%|█▊        | 1230/6837 [18:23<22:57,  4.07it/s]
 18%|█▊        | 1231/6837 [18:23<23:08,  4.04it/s]
 18%|█▊        | 1232/6837 [18:24<23:02,  4.05it/s]
 18%|█▊        | 1233/6837 [18:24<23:02,  4.05it/s]
 18%|█▊        | 1234/6837 [18:24<22:59,  4.06it/s]
 18%|█▊        | 1235/6837 [18:24<23:06,  4.04it/s]
 18%|█▊        | 1236/6837 [18:25<23:08,  4.04it/s]
 18%|█▊        | 1237/6837 [18:25<23:01,  4.05it/s]
 18%|█▊        | 1238/6837 [18:25<22:58,  4.06it/s]
 18%|█▊        | 1239/6837 [18:25<22:58,  4.06it/s]
 18%|█▊        | 1240/6837 [18:26<23:28,  3.97it/s]
 18%|█▊        | 1241/6837 [18:26<23:16,  4.01it/s]
 18%|█▊        | 1242/6837 [18:26<23:20,  3.99it/s]
 18%|█▊        | 1243/6837 [18:26<23:10,  4.02it/s]
 18%|█▊        | 1244/6837 [18:27<23:10,  4.02it/s]
 18%|█▊        | 1245/6837 [18:27<23:09,  4.02it/s]
 18%|█▊        | 1246/6837 [18:27<23:06,  4.03it/s]
 18%|█▊        | 1247/6837 [18:27<23:28,  3.97it/s]
 18%|█▊        | 1248/6837 [18:28<23:51,  3.90it/s]
 18%|█▊        | 1249/6837 [18:28<24:05,  3.87it/s]
 18%|█▊        | 1250/6837 [18:28<23:45,  3.92it/s]
                                                   
{'loss': 4.8191, 'grad_norm': 0.14332738518714905, 'learning_rate': 0.0058755985445536835, 'epoch': 0.04}
+
 18%|█▊        | 1250/6837 [18:28<23:45,  3.92it/s]
 18%|█▊        | 1251/6837 [18:28<23:41,  3.93it/s]
 18%|█▊        | 1252/6837 [18:29<23:42,  3.93it/s]
 18%|█▊        | 1253/6837 [18:29<23:42,  3.92it/s]
 18%|█▊        | 1254/6837 [18:29<24:04,  3.86it/s]
 18%|█▊        | 1255/6837 [18:29<24:02,  3.87it/s]
 18%|█▊        | 1256/6837 [18:30<24:02,  3.87it/s]
 18%|█▊        | 1257/6837 [18:30<24:19,  3.82it/s]
 18%|█▊        | 1258/6837 [18:30<24:20,  3.82it/s]
 18%|█▊        | 1259/6837 [18:30<24:03,  3.86it/s]
 18%|█▊        | 1260/6837 [18:31<23:46,  3.91it/s]
 18%|█▊        | 1261/6837 [18:31<23:24,  3.97it/s]
 18%|█▊        | 1262/6837 [18:31<24:42,  3.76it/s]
 18%|█▊        | 1263/6837 [18:31<24:09,  3.85it/s]
 18%|█▊        | 1264/6837 [18:32<23:42,  3.92it/s]
 19%|█▊        | 1265/6837 [18:32<24:56,  3.72it/s]
 19%|█▊        | 1266/6837 [18:32<24:18,  3.82it/s]
 19%|█▊        | 1267/6837 [18:32<23:54,  3.88it/s]
 19%|█▊        | 1268/6837 [18:33<24:36,  3.77it/s]
 19%|█▊        | 1269/6837 [18:33<24:05,  3.85it/s]
 19%|█▊        | 1270/6837 [18:33<23:46,  3.90it/s]
 19%|█▊        | 1271/6837 [18:34<23:24,  3.96it/s]
 19%|█▊        | 1272/6837 [18:34<23:12,  4.00it/s]
 19%|█▊        | 1273/6837 [18:34<23:18,  3.98it/s]
 19%|█▊        | 1274/6837 [18:34<23:05,  4.01it/s]
 19%|█▊        | 1275/6837 [18:34<22:58,  4.04it/s]
                                                   
{'loss': 4.8108, 'grad_norm': 0.18900059163570404, 'learning_rate': 0.005864451641141273, 'epoch': 0.04}
+
 19%|█▊        | 1275/6837 [18:34<22:58,  4.04it/s]
 19%|█▊        | 1276/6837 [18:35<23:00,  4.03it/s]
 19%|█▊        | 1277/6837 [18:35<22:53,  4.05it/s]
 19%|█▊        | 1278/6837 [18:35<22:48,  4.06it/s]
 19%|█▊        | 1279/6837 [18:35<22:53,  4.05it/s]
 19%|█▊        | 1280/6837 [18:36<23:02,  4.02it/s]
 19%|█▊        | 1281/6837 [18:36<23:01,  4.02it/s]
 19%|█▉        | 1282/6837 [18:36<23:00,  4.03it/s]
 19%|█▉        | 1283/6837 [18:36<22:54,  4.04it/s]
 19%|█▉        | 1284/6837 [18:37<22:58,  4.03it/s]
 19%|█▉        | 1285/6837 [18:37<22:57,  4.03it/s]
 19%|█▉        | 1286/6837 [18:37<23:05,  4.01it/s]
 19%|█▉        | 1287/6837 [18:37<22:54,  4.04it/s]
 19%|█▉        | 1288/6837 [18:38<23:04,  4.01it/s]
 19%|█▉        | 1289/6837 [18:38<22:53,  4.04it/s]
 19%|█▉        | 1290/6837 [18:38<22:48,  4.05it/s]
 19%|█▉        | 1291/6837 [18:38<22:43,  4.07it/s]
 19%|█▉        | 1292/6837 [18:39<22:45,  4.06it/s]
 19%|█▉        | 1293/6837 [18:39<22:43,  4.07it/s]
 19%|█▉        | 1294/6837 [18:39<22:43,  4.07it/s]
 19%|█▉        | 1295/6837 [18:39<22:58,  4.02it/s]
 19%|█▉        | 1296/6837 [18:40<22:55,  4.03it/s]
 19%|█▉        | 1297/6837 [18:40<22:48,  4.05it/s]
 19%|█▉        | 1298/6837 [18:40<22:45,  4.06it/s]
 19%|█▉        | 1299/6837 [18:40<22:42,  4.07it/s]
 19%|█▉        | 1300/6837 [18:41<22:39,  4.07it/s]
                                                   
{'loss': 4.7889, 'grad_norm': 0.15760479867458344, 'learning_rate': 0.0058528380337208785, 'epoch': 0.04}
+
 19%|█▉        | 1300/6837 [18:41<22:39,  4.07it/s]
 19%|█▉        | 1301/6837 [18:41<22:49,  4.04it/s]
 19%|█▉        | 1302/6837 [18:41<22:41,  4.07it/s]
 19%|█▉        | 1303/6837 [18:41<22:39,  4.07it/s]
 19%|█▉        | 1304/6837 [18:42<22:39,  4.07it/s]
 19%|█▉        | 1305/6837 [18:42<22:39,  4.07it/s]
 19%|█▉        | 1306/6837 [18:42<22:34,  4.08it/s]
 19%|█▉        | 1307/6837 [18:42<22:38,  4.07it/s]
 19%|█▉        | 1308/6837 [18:43<22:36,  4.08it/s]
 19%|█▉        | 1309/6837 [18:43<23:15,  3.96it/s]
 19%|█▉        | 1310/6837 [18:43<23:42,  3.89it/s]
 19%|█▉        | 1311/6837 [18:43<23:17,  3.95it/s]
 19%|█▉        | 1312/6837 [18:44<23:04,  3.99it/s]
 19%|█▉        | 1313/6837 [18:44<22:54,  4.02it/s]
 19%|█▉        | 1314/6837 [18:44<22:45,  4.05it/s]
 19%|█▉        | 1315/6837 [18:44<22:38,  4.06it/s]
 19%|█▉        | 1316/6837 [18:45<22:32,  4.08it/s]
 19%|█▉        | 1317/6837 [18:45<22:27,  4.10it/s]
 19%|█▉        | 1318/6837 [18:45<22:29,  4.09it/s]
 19%|█▉        | 1319/6837 [18:45<22:35,  4.07it/s]
 19%|█▉        | 1320/6837 [18:46<22:50,  4.03it/s]
 19%|█▉        | 1321/6837 [18:46<22:43,  4.05it/s]
 19%|█▉        | 1322/6837 [18:46<22:32,  4.08it/s]
 19%|█▉        | 1323/6837 [18:46<22:27,  4.09it/s]
 19%|█▉        | 1324/6837 [18:47<22:33,  4.07it/s]
 19%|█▉        | 1325/6837 [18:47<22:34,  4.07it/s]
                                                   {'loss': 4.7944, 'grad_norm': 0.1523684561252594, 'learning_rate': 0.005840759614493106, 'epoch': 0.04}
+
 19%|█▉        | 1325/6837 [18:47<22:34,  4.07it/s]
 19%|█▉        | 1326/6837 [18:47<22:35,  4.07it/s]
 19%|█▉        | 1327/6837 [18:47<22:41,  4.05it/s]
 19%|█▉        | 1328/6837 [18:48<22:33,  4.07it/s]
 19%|█▉        | 1329/6837 [18:48<22:27,  4.09it/s]
 19%|█▉        | 1330/6837 [18:48<22:21,  4.10it/s]
 19%|█▉        | 1331/6837 [18:48<22:18,  4.11it/s]
 19%|█▉        | 1332/6837 [18:49<22:38,  4.05it/s]
 19%|█▉        | 1333/6837 [18:49<22:36,  4.06it/s]
 20%|█▉        | 1334/6837 [18:49<22:26,  4.09it/s]
 20%|█▉        | 1335/6837 [18:49<22:23,  4.09it/s]
 20%|█▉        | 1336/6837 [18:50<22:26,  4.09it/s]
 20%|█▉        | 1337/6837 [18:50<22:25,  4.09it/s]
 20%|█▉        | 1338/6837 [18:50<22:18,  4.11it/s]
 20%|█▉        | 1339/6837 [18:50<22:28,  4.08it/s]
 20%|█▉        | 1340/6837 [18:51<22:29,  4.07it/s]
 20%|█▉        | 1341/6837 [18:51<22:23,  4.09it/s]
 20%|█▉        | 1342/6837 [18:51<22:32,  4.06it/s]
 20%|█▉        | 1343/6837 [18:51<22:39,  4.04it/s]
 20%|█▉        | 1344/6837 [18:52<22:33,  4.06it/s]
 20%|█▉        | 1345/6837 [18:52<22:30,  4.07it/s]
 20%|█▉        | 1346/6837 [18:52<22:53,  4.00it/s]
 20%|█▉        | 1347/6837 [18:52<22:42,  4.03it/s]
 20%|█▉        | 1348/6837 [18:53<22:36,  4.05it/s]
 20%|█▉        | 1349/6837 [18:53<22:40,  4.03it/s]
 20%|█▉        | 1350/6837 [18:53<22:35,  4.05it/s]
                                                   {'loss': 4.7685, 'grad_norm': 0.15037256479263306, 'learning_rate': 0.005828218351390171, 'epoch': 0.04}
+

 20%|█▉        | 1350/6837 [18:53<22:35,  4.05it/s]
 20%|█▉        | 1351/6837 [18:53<22:41,  4.03it/s]
 20%|█▉        | 1352/6837 [18:53<22:33,  4.05it/s]
 20%|█▉        | 1353/6837 [18:54<22:28,  4.07it/s]
 20%|█▉        | 1354/6837 [18:54<22:20,  4.09it/s]
 20%|█▉        | 1355/6837 [18:54<22:15,  4.10it/s]
 20%|█▉        | 1356/6837 [18:54<22:18,  4.09it/s]
 20%|█▉        | 1357/6837 [18:55<22:18,  4.09it/s]
 20%|█▉        | 1358/6837 [18:55<22:18,  4.09it/s]
 20%|█▉        | 1359/6837 [18:55<22:14,  4.11it/s]
 20%|█▉        | 1360/6837 [18:55<22:11,  4.11it/s]
 20%|█▉        | 1361/6837 [18:56<22:13,  4.11it/s]
 20%|█▉        | 1362/6837 [18:56<22:09,  4.12it/s]
 20%|█▉        | 1363/6837 [18:56<22:06,  4.13it/s]
 20%|█▉        | 1364/6837 [18:56<22:09,  4.12it/s]
 20%|█▉        | 1365/6837 [18:57<22:09,  4.12it/s]
 20%|█▉        | 1366/6837 [18:57<22:12,  4.11it/s]
 20%|█▉        | 1367/6837 [18:57<22:24,  4.07it/s]
 20%|██        | 1368/6837 [18:57<22:23,  4.07it/s]
 20%|██        | 1369/6837 [18:58<22:24,  4.07it/s]
 20%|██        | 1370/6837 [18:58<22:20,  4.08it/s]
 20%|██        | 1371/6837 [18:58<22:19,  4.08it/s]
 20%|██        | 1372/6837 [18:58<22:18,  4.08it/s]
 20%|██        | 1373/6837 [18:59<22:12,  4.10it/s]
 20%|██        | 1374/6837 [18:59<22:10,  4.11it/s]
 20%|██        | 1375/6837 [18:59<22:10,  4.11it/s]{'loss': 4.7698, 'grad_norm': 0.1460862159729004, 'learning_rate': 0.005815216287755262, 'epoch': 0.04}
                                                   
+
 20%|██        | 1375/6837 [18:59<22:10,  4.11it/s]
 20%|██        | 1376/6837 [18:59<22:12,  4.10it/s]
 20%|██        | 1377/6837 [19:00<22:12,  4.10it/s]
 20%|██        | 1378/6837 [19:00<22:44,  4.00it/s]
 20%|██        | 1379/6837 [19:00<23:09,  3.93it/s]
 20%|██        | 1380/6837 [19:00<22:44,  4.00it/s]
 20%|██        | 1381/6837 [19:01<22:27,  4.05it/s]
 20%|██        | 1382/6837 [19:01<22:21,  4.07it/s]
 20%|██        | 1383/6837 [19:01<22:12,  4.09it/s]
 20%|██        | 1384/6837 [19:01<22:10,  4.10it/s]
 20%|██        | 1385/6837 [19:02<22:12,  4.09it/s]
 20%|██        | 1386/6837 [19:02<22:08,  4.10it/s]
 20%|██        | 1387/6837 [19:02<22:08,  4.10it/s]
 20%|██        | 1388/6837 [19:02<22:04,  4.11it/s]
 20%|██        | 1389/6837 [19:03<22:05,  4.11it/s]
 20%|██        | 1390/6837 [19:03<22:08,  4.10it/s]
 20%|██        | 1391/6837 [19:03<22:13,  4.08it/s]
 20%|██        | 1392/6837 [19:03<22:14,  4.08it/s]
 20%|██        | 1393/6837 [19:04<22:13,  4.08it/s]
 20%|██        | 1394/6837 [19:04<22:07,  4.10it/s]
 20%|██        | 1395/6837 [19:04<22:04,  4.11it/s]
 20%|██        | 1396/6837 [19:04<22:06,  4.10it/s]
 20%|██        | 1397/6837 [19:05<22:37,  4.01it/s]
 20%|██        | 1398/6837 [19:05<22:42,  3.99it/s]
 20%|██        | 1399/6837 [19:05<22:30,  4.03it/s]
 20%|██        | 1400/6837 [19:05<22:52,  3.96it/s]
                                                   
{'loss': 4.7646, 'grad_norm': 0.149286687374115, 'learning_rate': 0.005801755542009626, 'epoch': 0.04}
+
 20%|██        | 1400/6837 [19:05<22:52,  3.96it/s]
 20%|██        | 1401/6837 [19:06<23:11,  3.91it/s]
 21%|██        | 1402/6837 [19:06<23:04,  3.93it/s]
 21%|██        | 1403/6837 [19:06<23:21,  3.88it/s]
 21%|██        | 1404/6837 [19:06<23:15,  3.89it/s]
 21%|██        | 1405/6837 [19:07<23:02,  3.93it/s]
 21%|██        | 1406/6837 [19:07<23:13,  3.90it/s]
 21%|██        | 1407/6837 [19:08<39:16,  2.30it/s]
 21%|██        | 1408/6837 [19:08<34:19,  2.64it/s]
 21%|██        | 1409/6837 [19:08<30:38,  2.95it/s]
 21%|██        | 1410/6837 [19:08<28:03,  3.22it/s]
 21%|██        | 1411/6837 [19:09<27:01,  3.35it/s]
 21%|██        | 1412/6837 [19:09<25:30,  3.54it/s]
 21%|██        | 1413/6837 [19:09<25:08,  3.60it/s]
 21%|██        | 1414/6837 [19:09<24:42,  3.66it/s]
 21%|██        | 1415/6837 [19:10<24:38,  3.67it/s]
 21%|██        | 1416/6837 [19:10<23:51,  3.79it/s]
 21%|██        | 1417/6837 [19:10<23:17,  3.88it/s]
 21%|██        | 1418/6837 [19:10<22:51,  3.95it/s]
 21%|██        | 1419/6837 [19:11<22:39,  3.99it/s]
 21%|██        | 1420/6837 [19:11<22:28,  4.02it/s]
 21%|██        | 1421/6837 [19:11<22:19,  4.04it/s]
 21%|██        | 1422/6837 [19:11<22:23,  4.03it/s]
 21%|██        | 1423/6837 [19:12<22:13,  4.06it/s]
 21%|██        | 1424/6837 [19:12<22:04,  4.09it/s]
 21%|██        | 1425/6837 [19:12<22:03,  4.09it/s]
                                                   
{'loss': 4.7572, 'grad_norm': 0.1465584635734558, 'learning_rate': 0.005787838307307407, 'epoch': 0.04}
+
 21%|██        | 1425/6837 [19:12<22:03,  4.09it/s]
 21%|██        | 1426/6837 [19:12<22:01,  4.09it/s]
 21%|██        | 1427/6837 [19:13<22:01,  4.09it/s]
 21%|██        | 1428/6837 [19:13<22:41,  3.97it/s]
 21%|██        | 1429/6837 [19:13<23:01,  3.91it/s]
 21%|██        | 1430/6837 [19:13<23:40,  3.81it/s]
 21%|██        | 1431/6837 [19:14<23:52,  3.77it/s]
 21%|██        | 1432/6837 [19:14<24:00,  3.75it/s]
 21%|██        | 1433/6837 [19:14<24:24,  3.69it/s]
 21%|██        | 1434/6837 [19:15<24:24,  3.69it/s]
 21%|██        | 1435/6837 [19:15<24:41,  3.65it/s]
 21%|██        | 1436/6837 [19:15<24:49,  3.63it/s]
 21%|██        | 1437/6837 [19:15<24:44,  3.64it/s]
 21%|██        | 1438/6837 [19:16<24:38,  3.65it/s]
 21%|██        | 1439/6837 [19:16<24:38,  3.65it/s]
 21%|██        | 1440/6837 [19:16<24:35,  3.66it/s]
 21%|██        | 1441/6837 [19:16<24:40,  3.64it/s]
 21%|██        | 1442/6837 [19:17<24:31,  3.67it/s]
 21%|██        | 1443/6837 [19:17<24:37,  3.65it/s]
 21%|██        | 1444/6837 [19:17<24:22,  3.69it/s]
 21%|██        | 1445/6837 [19:18<24:37,  3.65it/s]
 21%|██        | 1446/6837 [19:18<24:44,  3.63it/s]
 21%|██        | 1447/6837 [19:18<24:44,  3.63it/s]
 21%|██        | 1448/6837 [19:18<24:35,  3.65it/s]
 21%|██        | 1449/6837 [19:19<23:57,  3.75it/s]
 21%|██        | 1450/6837 [19:19<23:15,  3.86it/s]{'loss': 4.7592, 'grad_norm': 0.1481638252735138, 'learning_rate': 0.00577346685117832, 'epoch': 0.04}
+
                                                   

 21%|██        | 1450/6837 [19:19<23:15,  3.86it/s]
 21%|██        | 1451/6837 [19:19<23:01,  3.90it/s]
 21%|██        | 1452/6837 [19:19<22:41,  3.96it/s]
 21%|██▏       | 1453/6837 [19:20<22:29,  3.99it/s]
 21%|██▏       | 1454/6837 [19:20<22:21,  4.01it/s]
 21%|██▏       | 1455/6837 [19:20<22:12,  4.04it/s]
 21%|██▏       | 1456/6837 [19:20<22:27,  3.99it/s]
 21%|██▏       | 1457/6837 [19:21<22:13,  4.03it/s]
 21%|██▏       | 1458/6837 [19:21<22:02,  4.07it/s]
 21%|██▏       | 1459/6837 [19:21<21:57,  4.08it/s]
 21%|██▏       | 1460/6837 [19:21<21:54,  4.09it/s]
 21%|██▏       | 1461/6837 [19:22<21:52,  4.09it/s]
 21%|██▏       | 1462/6837 [19:22<21:54,  4.09it/s]
 21%|██▏       | 1463/6837 [19:22<21:59,  4.07it/s]
 21%|██▏       | 1464/6837 [19:22<22:05,  4.05it/s]
 21%|██▏       | 1465/6837 [19:23<22:02,  4.06it/s]
 21%|██▏       | 1466/6837 [19:23<22:46,  3.93it/s]
 21%|██▏       | 1467/6837 [19:23<22:32,  3.97it/s]
 21%|██▏       | 1468/6837 [19:23<22:19,  4.01it/s]
 21%|██▏       | 1469/6837 [19:24<22:04,  4.05it/s]
 22%|██▏       | 1470/6837 [19:24<22:00,  4.07it/s]
 22%|██▏       | 1471/6837 [19:24<21:50,  4.09it/s]
 22%|██▏       | 1472/6837 [19:24<21:45,  4.11it/s]
 22%|██▏       | 1473/6837 [19:25<21:45,  4.11it/s]
 22%|██▏       | 1474/6837 [19:25<21:41,  4.12it/s]
 22%|██▏       | 1475/6837 [19:25<21:41,  4.12it/s]{'loss': 4.7474, 'grad_norm': 0.13737046718597412, 'learning_rate': 0.005758643515158205, 'epoch': 0.04}
                                                   
+
 22%|██▏       | 1475/6837 [19:25<21:41,  4.12it/s]
 22%|██▏       | 1476/6837 [19:25<21:39,  4.12it/s]
 22%|██▏       | 1477/6837 [19:26<21:38,  4.13it/s]
 22%|██▏       | 1478/6837 [19:26<21:36,  4.13it/s]
 22%|██▏       | 1479/6837 [19:26<21:35,  4.13it/s]
 22%|██▏       | 1480/6837 [19:26<21:34,  4.14it/s]
 22%|██▏       | 1481/6837 [19:26<21:40,  4.12it/s]
 22%|██▏       | 1482/6837 [19:27<21:43,  4.11it/s]
 22%|██▏       | 1483/6837 [19:27<21:43,  4.11it/s]
 22%|██▏       | 1484/6837 [19:27<21:43,  4.11it/s]
 22%|██▏       | 1485/6837 [19:27<21:41,  4.11it/s]
 22%|██▏       | 1486/6837 [19:28<21:42,  4.11it/s]
 22%|██▏       | 1487/6837 [19:28<21:47,  4.09it/s]
 22%|██▏       | 1488/6837 [19:28<21:48,  4.09it/s]
 22%|██▏       | 1489/6837 [19:28<21:41,  4.11it/s]
 22%|██▏       | 1490/6837 [19:29<21:40,  4.11it/s]
 22%|██▏       | 1491/6837 [19:29<21:39,  4.11it/s]
 22%|██▏       | 1492/6837 [19:29<21:36,  4.12it/s]
 22%|██▏       | 1493/6837 [19:29<21:37,  4.12it/s]
 22%|██▏       | 1494/6837 [19:30<21:34,  4.13it/s]
 22%|██▏       | 1495/6837 [19:30<21:35,  4.12it/s]
 22%|██▏       | 1496/6837 [19:30<21:31,  4.14it/s]
 22%|██▏       | 1497/6837 [19:30<21:29,  4.14it/s]
 22%|██▏       | 1498/6837 [19:31<21:26,  4.15it/s]
 22%|██▏       | 1499/6837 [19:31<21:39,  4.11it/s]
 22%|██▏       | 1500/6837 [19:31<21:39,  4.11it/s]
                                                   
{'loss': 4.7354, 'grad_norm': 0.13927464187145233, 'learning_rate': 0.005743370714407518, 'epoch': 0.04}
+
 22%|██▏       | 1500/6837 [19:31<21:39,  4.11it/s]
 22%|██▏       | 1501/6837 [19:31<21:41,  4.10it/s]
 22%|██▏       | 1502/6837 [19:32<21:58,  4.05it/s]
 22%|██▏       | 1503/6837 [19:32<21:46,  4.08it/s]
 22%|██▏       | 1504/6837 [19:32<21:36,  4.11it/s]
 22%|██▏       | 1505/6837 [19:32<21:37,  4.11it/s]
 22%|██▏       | 1506/6837 [19:33<21:35,  4.11it/s]
 22%|██▏       | 1507/6837 [19:33<21:33,  4.12it/s]
 22%|██▏       | 1508/6837 [19:33<21:28,  4.14it/s]
 22%|██▏       | 1509/6837 [19:33<21:23,  4.15it/s]
 22%|██▏       | 1510/6837 [19:34<21:24,  4.15it/s]
 22%|██▏       | 1511/6837 [19:34<21:21,  4.15it/s]
 22%|██▏       | 1512/6837 [19:34<21:20,  4.16it/s]
 22%|██▏       | 1513/6837 [19:34<21:20,  4.16it/s]
 22%|██▏       | 1514/6837 [19:35<21:18,  4.16it/s]
 22%|██▏       | 1515/6837 [19:35<21:25,  4.14it/s]
 22%|██▏       | 1516/6837 [19:35<21:24,  4.14it/s]
 22%|██▏       | 1517/6837 [19:35<21:24,  4.14it/s]
 22%|██▏       | 1518/6837 [19:35<21:24,  4.14it/s]
 22%|██▏       | 1519/6837 [19:36<21:21,  4.15it/s]
 22%|██▏       | 1520/6837 [19:36<21:21,  4.15it/s]
 22%|██▏       | 1521/6837 [19:36<21:22,  4.15it/s]
 22%|██▏       | 1522/6837 [19:36<21:19,  4.15it/s]
 22%|██▏       | 1523/6837 [19:37<21:20,  4.15it/s]
 22%|██▏       | 1524/6837 [19:37<21:20,  4.15it/s]
 22%|██▏       | 1525/6837 [19:37<21:21,  4.14it/s]{'loss': 4.7286, 'grad_norm': 0.1442137211561203, 'learning_rate': 0.00572765093731783, 'epoch': 0.04}
                                                   
+
 22%|██▏       | 1525/6837 [19:37<21:21,  4.14it/s]
 22%|██▏       | 1526/6837 [19:37<21:30,  4.12it/s]
 22%|██▏       | 1527/6837 [19:38<21:34,  4.10it/s]
 22%|██▏       | 1528/6837 [19:38<21:29,  4.12it/s]
 22%|██▏       | 1529/6837 [19:38<21:29,  4.12it/s]
 22%|██▏       | 1530/6837 [19:38<21:27,  4.12it/s]
 22%|██▏       | 1531/6837 [19:39<21:25,  4.13it/s]
 22%|██▏       | 1532/6837 [19:39<21:23,  4.13it/s]
 22%|██▏       | 1533/6837 [19:39<21:24,  4.13it/s]
 22%|██▏       | 1534/6837 [19:39<21:22,  4.14it/s]
 22%|██▏       | 1535/6837 [19:40<21:21,  4.14it/s]
 22%|██▏       | 1536/6837 [19:40<21:23,  4.13it/s]
 22%|██▏       | 1537/6837 [19:40<21:21,  4.14it/s]
 22%|██▏       | 1538/6837 [19:40<21:22,  4.13it/s]
 23%|██▎       | 1539/6837 [19:41<21:20,  4.14it/s]
 23%|██▎       | 1540/6837 [19:41<21:24,  4.12it/s]
 23%|██▎       | 1541/6837 [19:41<21:20,  4.14it/s]
 23%|██▎       | 1542/6837 [19:41<21:18,  4.14it/s]
 23%|██▎       | 1543/6837 [19:42<21:16,  4.15it/s]
 23%|██▎       | 1544/6837 [19:42<21:18,  4.14it/s]
 23%|██▎       | 1545/6837 [19:42<21:19,  4.14it/s]
 23%|██▎       | 1546/6837 [19:42<21:20,  4.13it/s]
 23%|██▎       | 1547/6837 [19:42<21:13,  4.15it/s]
 23%|██▎       | 1548/6837 [19:43<21:55,  4.02it/s]
 23%|██▎       | 1549/6837 [19:43<22:24,  3.93it/s]
 23%|██▎       | 1550/6837 [19:43<21:59,  4.01it/s]
                                                   
{'loss': 4.7346, 'grad_norm': 0.1326439082622528, 'learning_rate': 0.005711486745106398, 'epoch': 0.05}
+
 23%|██▎       | 1550/6837 [19:43<21:59,  4.01it/s]
 23%|██▎       | 1551/6837 [19:43<21:48,  4.04it/s]
 23%|██▎       | 1552/6837 [19:44<21:35,  4.08it/s]
 23%|██▎       | 1553/6837 [19:44<21:25,  4.11it/s]
 23%|██▎       | 1554/6837 [19:44<21:18,  4.13it/s]
 23%|██▎       | 1555/6837 [19:44<21:11,  4.15it/s]
 23%|██▎       | 1556/6837 [19:45<21:13,  4.15it/s]
 23%|██▎       | 1557/6837 [19:45<21:15,  4.14it/s]
 23%|██▎       | 1558/6837 [19:45<21:15,  4.14it/s]
 23%|██▎       | 1559/6837 [19:45<21:15,  4.14it/s]
 23%|██▎       | 1560/6837 [19:46<21:13,  4.14it/s]
 23%|██▎       | 1561/6837 [19:46<21:11,  4.15it/s]
 23%|██▎       | 1562/6837 [19:46<21:08,  4.16it/s]
 23%|██▎       | 1563/6837 [19:46<21:08,  4.16it/s]
 23%|██▎       | 1564/6837 [19:47<21:08,  4.16it/s]
 23%|██▎       | 1565/6837 [19:47<21:05,  4.17it/s]
 23%|██▎       | 1566/6837 [19:47<21:03,  4.17it/s]
 23%|██▎       | 1567/6837 [19:47<21:10,  4.15it/s]
 23%|██▎       | 1568/6837 [19:48<21:08,  4.15it/s]
 23%|██▎       | 1569/6837 [19:48<21:13,  4.14it/s]
 23%|██▎       | 1570/6837 [19:48<21:13,  4.14it/s]
 23%|██▎       | 1571/6837 [19:48<21:09,  4.15it/s]
 23%|██▎       | 1572/6837 [19:49<21:10,  4.14it/s]
 23%|██▎       | 1573/6837 [19:49<21:07,  4.15it/s]
 23%|██▎       | 1574/6837 [19:49<21:05,  4.16it/s]
 23%|██▎       | 1575/6837 [19:49<21:05,  4.16it/s]
                                                   
{'loss': 4.7234, 'grad_norm': 0.1355282962322235, 'learning_rate': 0.0056948807713988634, 'epoch': 0.05}
+
 23%|██▎       | 1575/6837 [19:49<21:05,  4.16it/s]
 23%|██▎       | 1576/6837 [19:50<21:07,  4.15it/s]
 23%|██▎       | 1577/6837 [19:50<21:08,  4.15it/s]
 23%|██▎       | 1578/6837 [19:50<21:06,  4.15it/s]
 23%|██▎       | 1579/6837 [19:50<21:04,  4.16it/s]
 23%|██▎       | 1580/6837 [19:50<21:01,  4.17it/s]
 23%|██▎       | 1581/6837 [19:51<21:05,  4.15it/s]
 23%|██▎       | 1582/6837 [19:51<21:01,  4.17it/s]
 23%|██▎       | 1583/6837 [19:51<21:04,  4.16it/s]
 23%|██▎       | 1584/6837 [19:51<21:04,  4.15it/s]
 23%|██▎       | 1585/6837 [19:52<21:10,  4.13it/s]
 23%|██▎       | 1586/6837 [19:52<21:06,  4.15it/s]
 23%|██▎       | 1587/6837 [19:52<21:07,  4.14it/s]
 23%|██▎       | 1588/6837 [19:52<21:06,  4.14it/s]
 23%|██▎       | 1589/6837 [19:53<21:02,  4.16it/s]
 23%|██▎       | 1590/6837 [19:53<21:07,  4.14it/s]
 23%|██▎       | 1591/6837 [19:53<21:07,  4.14it/s]
 23%|██▎       | 1592/6837 [19:53<21:05,  4.15it/s]
 23%|██▎       | 1593/6837 [19:54<21:00,  4.16it/s]
 23%|██▎       | 1594/6837 [19:54<21:01,  4.16it/s]
 23%|██▎       | 1595/6837 [19:54<20:59,  4.16it/s]
 23%|██▎       | 1596/6837 [19:54<20:55,  4.18it/s]
 23%|██▎       | 1597/6837 [19:55<20:56,  4.17it/s]
 23%|██▎       | 1598/6837 [19:55<20:55,  4.17it/s]
 23%|██▎       | 1599/6837 [19:55<20:55,  4.17it/s]
 23%|██▎       | 1600/6837 [19:55<20:52,  4.18it/s]{'loss': 4.7179, 'grad_norm': 0.13755303621292114, 'learning_rate': 0.005677835721800159, 'epoch': 0.05}
                                                   
+
 23%|██▎       | 1600/6837 [19:55<20:52,  4.18it/s]
 23%|██▎       | 1601/6837 [19:56<21:01,  4.15it/s]
 23%|██▎       | 1602/6837 [19:56<21:03,  4.14it/s]
 23%|██▎       | 1603/6837 [19:56<20:58,  4.16it/s]
 23%|██▎       | 1604/6837 [19:56<20:57,  4.16it/s]
 23%|██▎       | 1605/6837 [19:56<20:55,  4.17it/s]
 23%|██▎       | 1606/6837 [19:57<20:54,  4.17it/s]
 24%|██▎       | 1607/6837 [19:57<20:53,  4.17it/s]
 24%|██▎       | 1608/6837 [19:57<20:53,  4.17it/s]
 24%|██▎       | 1609/6837 [19:57<20:53,  4.17it/s]
 24%|██▎       | 1610/6837 [19:58<20:53,  4.17it/s]
 24%|██▎       | 1611/6837 [19:58<20:53,  4.17it/s]
 24%|██▎       | 1612/6837 [19:58<20:50,  4.18it/s]
 24%|██▎       | 1613/6837 [19:58<20:50,  4.18it/s]
 24%|██▎       | 1614/6837 [19:59<20:48,  4.18it/s]
 24%|██▎       | 1615/6837 [19:59<20:48,  4.18it/s]
 24%|██▎       | 1616/6837 [19:59<20:45,  4.19it/s]
 24%|██▎       | 1617/6837 [19:59<20:45,  4.19it/s]
 24%|██▎       | 1618/6837 [20:00<20:46,  4.19it/s]
 24%|██▎       | 1619/6837 [20:00<20:46,  4.19it/s]
 24%|██▎       | 1620/6837 [20:00<20:44,  4.19it/s]
 24%|██▎       | 1621/6837 [20:00<20:48,  4.18it/s]
 24%|██▎       | 1622/6837 [20:01<20:49,  4.17it/s]
 24%|██▎       | 1623/6837 [20:01<20:47,  4.18it/s]
 24%|██▍       | 1624/6837 [20:01<20:45,  4.19it/s]
 24%|██▍       | 1625/6837 [20:01<20:45,  4.19it/s]
                                                   
{'loss': 4.6983, 'grad_norm': 0.14186429977416992, 'learning_rate': 0.00566035437345368, 'epoch': 0.05}
+
 24%|██▍       | 1625/6837 [20:01<20:45,  4.19it/s]
 24%|██▍       | 1626/6837 [20:02<20:53,  4.16it/s]
 24%|██▍       | 1627/6837 [20:02<20:56,  4.14it/s]
 24%|██▍       | 1628/6837 [20:02<21:02,  4.13it/s]
 24%|██▍       | 1629/6837 [20:02<21:04,  4.12it/s]
 24%|██▍       | 1630/6837 [20:02<20:59,  4.13it/s]
 24%|██▍       | 1631/6837 [20:03<20:54,  4.15it/s]
 24%|██▍       | 1632/6837 [20:03<20:50,  4.16it/s]
 24%|██▍       | 1633/6837 [20:03<20:48,  4.17it/s]
 24%|██▍       | 1634/6837 [20:03<20:45,  4.18it/s]
 24%|██▍       | 1635/6837 [20:04<20:43,  4.18it/s]
 24%|██▍       | 1636/6837 [20:04<20:46,  4.17it/s]
 24%|██▍       | 1637/6837 [20:04<20:42,  4.19it/s]
 24%|██▍       | 1638/6837 [20:04<20:43,  4.18it/s]
 24%|██▍       | 1639/6837 [20:05<21:24,  4.05it/s]
 24%|██▍       | 1640/6837 [20:05<21:55,  3.95it/s]
 24%|██▍       | 1641/6837 [20:05<22:17,  3.89it/s]
 24%|██▍       | 1642/6837 [20:05<22:36,  3.83it/s]
 24%|██▍       | 1643/6837 [20:06<22:06,  3.92it/s]
 24%|██▍       | 1644/6837 [20:06<21:39,  4.00it/s]
 24%|██▍       | 1645/6837 [20:06<21:23,  4.05it/s]
 24%|██▍       | 1646/6837 [20:06<21:09,  4.09it/s]
 24%|██▍       | 1647/6837 [20:07<21:00,  4.12it/s]
 24%|██▍       | 1648/6837 [20:07<20:56,  4.13it/s]
 24%|██▍       | 1649/6837 [20:07<20:54,  4.14it/s]
 24%|██▍       | 1650/6837 [20:07<20:51,  4.14it/s]
                                                   
{'loss': 4.6805, 'grad_norm': 0.18044713139533997, 'learning_rate': 0.005642439574588813, 'epoch': 0.05}
+
 24%|██▍       | 1650/6837 [20:07<20:51,  4.14it/s]
 24%|██▍       | 1651/6837 [20:08<20:51,  4.14it/s]
 24%|██▍       | 1652/6837 [20:08<20:46,  4.16it/s]
 24%|██▍       | 1653/6837 [20:08<20:45,  4.16it/s]
 24%|██▍       | 1654/6837 [20:08<20:47,  4.15it/s]
 24%|██▍       | 1655/6837 [20:09<20:44,  4.16it/s]
 24%|██▍       | 1656/6837 [20:09<20:44,  4.16it/s]
 24%|██▍       | 1657/6837 [20:09<20:43,  4.17it/s]
 24%|██▍       | 1658/6837 [20:09<20:42,  4.17it/s]
 24%|██▍       | 1659/6837 [20:10<20:45,  4.16it/s]
 24%|██▍       | 1660/6837 [20:10<20:43,  4.16it/s]
 24%|██▍       | 1661/6837 [20:10<20:39,  4.18it/s]
 24%|██▍       | 1662/6837 [20:10<20:38,  4.18it/s]
 24%|██▍       | 1663/6837 [20:11<20:37,  4.18it/s]
 24%|██▍       | 1664/6837 [20:11<20:37,  4.18it/s]
 24%|██▍       | 1665/6837 [20:11<20:41,  4.17it/s]
 24%|██▍       | 1666/6837 [20:11<20:38,  4.17it/s]
 24%|██▍       | 1667/6837 [20:11<20:42,  4.16it/s]
 24%|██▍       | 1668/6837 [20:12<20:43,  4.16it/s]
 24%|██▍       | 1669/6837 [20:12<20:45,  4.15it/s]
 24%|██▍       | 1670/6837 [20:12<20:40,  4.16it/s]
 24%|██▍       | 1671/6837 [20:12<20:37,  4.17it/s]
 24%|██▍       | 1672/6837 [20:13<20:37,  4.17it/s]
 24%|██▍       | 1673/6837 [20:13<20:40,  4.16it/s]
 24%|██▍       | 1674/6837 [20:13<20:47,  4.14it/s]
 24%|██▍       | 1675/6837 [20:13<20:42,  4.16it/s]
                                                   
{'loss': 4.702, 'grad_norm': 0.13621020317077637, 'learning_rate': 0.005624094244056873, 'epoch': 0.05}
+
 24%|██▍       | 1675/6837 [20:13<20:42,  4.16it/s]
 25%|██▍       | 1676/6837 [20:14<20:46,  4.14it/s]
 25%|██▍       | 1677/6837 [20:14<20:46,  4.14it/s]
 25%|██▍       | 1678/6837 [20:14<20:43,  4.15it/s]
 25%|██▍       | 1679/6837 [20:14<20:43,  4.15it/s]
 25%|██▍       | 1680/6837 [20:15<20:40,  4.16it/s]
 25%|██▍       | 1681/6837 [20:15<20:38,  4.16it/s]
 25%|██▍       | 1682/6837 [20:15<20:38,  4.16it/s]
 25%|██▍       | 1683/6837 [20:15<20:34,  4.17it/s]
 25%|██▍       | 1684/6837 [20:16<20:35,  4.17it/s]
 25%|██▍       | 1685/6837 [20:16<20:33,  4.18it/s]
 25%|██▍       | 1686/6837 [20:16<20:33,  4.17it/s]
 25%|██▍       | 1687/6837 [20:16<20:33,  4.18it/s]
 25%|██▍       | 1688/6837 [20:17<20:35,  4.17it/s]
 25%|██▍       | 1689/6837 [20:17<20:36,  4.16it/s]
 25%|██▍       | 1690/6837 [20:17<20:35,  4.17it/s]
 25%|██▍       | 1691/6837 [20:17<20:37,  4.16it/s]
 25%|██▍       | 1692/6837 [20:17<20:33,  4.17it/s]
 25%|██▍       | 1693/6837 [20:18<20:33,  4.17it/s]
 25%|██▍       | 1694/6837 [20:18<20:32,  4.17it/s]
 25%|██▍       | 1695/6837 [20:18<20:29,  4.18it/s]
 25%|██▍       | 1696/6837 [20:18<20:28,  4.18it/s]
 25%|██▍       | 1697/6837 [20:19<20:28,  4.19it/s]
 25%|██▍       | 1698/6837 [20:19<20:28,  4.18it/s]
 25%|██▍       | 1699/6837 [20:19<20:27,  4.19it/s]
 25%|██▍       | 1700/6837 [20:19<20:29,  4.18it/s]{'loss': 4.6829, 'grad_norm': 0.15837280452251434, 'learning_rate': 0.005605321370855529, 'epoch': 0.05}
                                                   
+
 25%|██▍       | 1700/6837 [20:19<20:29,  4.18it/s]
 25%|██▍       | 1701/6837 [20:20<20:35,  4.16it/s]
 25%|██▍       | 1702/6837 [20:20<20:31,  4.17it/s]
 25%|██▍       | 1703/6837 [20:20<20:26,  4.18it/s]
 25%|██▍       | 1704/6837 [20:20<20:24,  4.19it/s]
 25%|██▍       | 1705/6837 [20:21<20:26,  4.19it/s]
 25%|██▍       | 1706/6837 [20:21<20:27,  4.18it/s]
 25%|██▍       | 1707/6837 [20:21<20:36,  4.15it/s]
 25%|██▍       | 1708/6837 [20:21<20:39,  4.14it/s]
 25%|██▍       | 1709/6837 [20:22<20:39,  4.14it/s]
 25%|██▌       | 1710/6837 [20:22<20:34,  4.15it/s]
 25%|██▌       | 1711/6837 [20:22<20:30,  4.16it/s]
 25%|██▌       | 1712/6837 [20:22<20:30,  4.16it/s]
 25%|██▌       | 1713/6837 [20:23<20:31,  4.16it/s]
 25%|██▌       | 1714/6837 [20:23<20:29,  4.17it/s]
 25%|██▌       | 1715/6837 [20:23<20:27,  4.17it/s]
 25%|██▌       | 1716/6837 [20:23<20:31,  4.16it/s]
 25%|██▌       | 1717/6837 [20:23<20:29,  4.16it/s]
 25%|██▌       | 1718/6837 [20:24<20:28,  4.17it/s]
 25%|██▌       | 1719/6837 [20:24<20:27,  4.17it/s]
 25%|██▌       | 1720/6837 [20:24<20:27,  4.17it/s]
 25%|██▌       | 1721/6837 [20:24<20:27,  4.17it/s]
 25%|██▌       | 1722/6837 [20:25<20:25,  4.17it/s]
 25%|██▌       | 1723/6837 [20:25<20:23,  4.18it/s]
 25%|██▌       | 1724/6837 [20:25<20:27,  4.16it/s]
 25%|██▌       | 1725/6837 [20:25<20:25,  4.17it/s]
                                                   
{'loss': 4.6889, 'grad_norm': 0.16804726421833038, 'learning_rate': 0.005586124013641817, 'epoch': 0.05}
+
 25%|██▌       | 1725/6837 [20:25<20:25,  4.17it/s]
 25%|██▌       | 1726/6837 [20:26<20:29,  4.16it/s]
 25%|██▌       | 1727/6837 [20:26<20:25,  4.17it/s]
 25%|██▌       | 1728/6837 [20:26<20:26,  4.16it/s]
 25%|██▌       | 1729/6837 [20:26<20:24,  4.17it/s]
 25%|██▌       | 1730/6837 [20:27<20:24,  4.17it/s]
 25%|██▌       | 1731/6837 [20:27<20:24,  4.17it/s]
 25%|██▌       | 1732/6837 [20:27<20:24,  4.17it/s]
 25%|██▌       | 1733/6837 [20:27<20:22,  4.18it/s]
 25%|██▌       | 1734/6837 [20:28<20:16,  4.19it/s]
 25%|██▌       | 1735/6837 [20:28<20:17,  4.19it/s]
 25%|██▌       | 1736/6837 [20:28<20:18,  4.19it/s]
 25%|██▌       | 1737/6837 [20:28<20:21,  4.18it/s]
 25%|██▌       | 1738/6837 [20:28<20:18,  4.19it/s]
 25%|██▌       | 1739/6837 [20:29<20:18,  4.18it/s]
 25%|██▌       | 1740/6837 [20:29<20:18,  4.18it/s]
 25%|██▌       | 1741/6837 [20:29<20:14,  4.20it/s]
 25%|██▌       | 1742/6837 [20:29<20:14,  4.20it/s]
 25%|██▌       | 1743/6837 [20:30<20:15,  4.19it/s]
 26%|██▌       | 1744/6837 [20:30<20:20,  4.17it/s]
 26%|██▌       | 1745/6837 [20:30<20:20,  4.17it/s]
 26%|██▌       | 1746/6837 [20:30<20:20,  4.17it/s]
 26%|██▌       | 1747/6837 [20:31<20:19,  4.18it/s]
 26%|██▌       | 1748/6837 [20:31<20:16,  4.18it/s]
 26%|██▌       | 1749/6837 [20:31<20:19,  4.17it/s]
 26%|██▌       | 1750/6837 [20:31<20:18,  4.17it/s]
                                                   
{'loss': 4.6683, 'grad_norm': 0.14777465164661407, 'learning_rate': 0.005566505300233791, 'epoch': 0.05}
+
 26%|██▌       | 1750/6837 [20:31<20:18,  4.17it/s]
 26%|██▌       | 1751/6837 [20:32<20:20,  4.17it/s]
 26%|██▌       | 1752/6837 [20:32<20:16,  4.18it/s]
 26%|██▌       | 1753/6837 [20:32<20:15,  4.18it/s]
 26%|██▌       | 1754/6837 [20:32<20:15,  4.18it/s]
 26%|██▌       | 1755/6837 [20:33<20:16,  4.18it/s]
 26%|██▌       | 1756/6837 [20:33<20:15,  4.18it/s]
 26%|██▌       | 1757/6837 [20:33<20:16,  4.18it/s]
 26%|██▌       | 1758/6837 [20:33<20:20,  4.16it/s]
 26%|██▌       | 1759/6837 [20:34<20:23,  4.15it/s]
 26%|██▌       | 1760/6837 [20:34<20:18,  4.17it/s]
 26%|██▌       | 1761/6837 [20:34<20:16,  4.17it/s]
 26%|██▌       | 1762/6837 [20:34<20:15,  4.18it/s]
 26%|██▌       | 1763/6837 [20:34<20:09,  4.19it/s]
 26%|██▌       | 1764/6837 [20:35<20:11,  4.19it/s]
 26%|██▌       | 1765/6837 [20:35<20:10,  4.19it/s]
 26%|██▌       | 1766/6837 [20:35<20:10,  4.19it/s]
 26%|██▌       | 1767/6837 [20:35<20:10,  4.19it/s]
 26%|██▌       | 1768/6837 [20:36<20:09,  4.19it/s]
 26%|██▌       | 1769/6837 [20:36<20:07,  4.20it/s]
 26%|██▌       | 1770/6837 [20:36<20:08,  4.19it/s]
 26%|██▌       | 1771/6837 [20:36<20:06,  4.20it/s]
 26%|██▌       | 1772/6837 [20:37<20:07,  4.19it/s]
 26%|██▌       | 1773/6837 [20:37<20:08,  4.19it/s]
 26%|██▌       | 1774/6837 [20:37<20:14,  4.17it/s]
 26%|██▌       | 1775/6837 [20:37<20:12,  4.18it/s]
                                                   
{'loss': 4.6645, 'grad_norm': 0.13642410933971405, 'learning_rate': 0.0055464684271009064, 'epoch': 0.05}
+
 26%|██▌       | 1775/6837 [20:37<20:12,  4.18it/s]
 26%|██▌       | 1776/6837 [20:38<20:14,  4.17it/s]
 26%|██▌       | 1777/6837 [20:38<20:16,  4.16it/s]
 26%|██▌       | 1778/6837 [20:38<20:16,  4.16it/s]
 26%|██▌       | 1779/6837 [20:38<20:12,  4.17it/s]
 26%|██▌       | 1780/6837 [20:39<20:14,  4.16it/s]
 26%|██▌       | 1781/6837 [20:39<20:13,  4.17it/s]
 26%|██▌       | 1782/6837 [20:39<20:13,  4.17it/s]
 26%|██▌       | 1783/6837 [20:39<20:07,  4.19it/s]
 26%|██▌       | 1784/6837 [20:40<20:08,  4.18it/s]
 26%|██▌       | 1785/6837 [20:40<20:08,  4.18it/s]
 26%|██▌       | 1786/6837 [20:40<20:07,  4.18it/s]
 26%|██▌       | 1787/6837 [20:40<20:03,  4.20it/s]
 26%|██▌       | 1788/6837 [20:40<20:04,  4.19it/s]
 26%|██▌       | 1789/6837 [20:41<20:01,  4.20it/s]
 26%|██▌       | 1790/6837 [20:41<20:02,  4.20it/s]
 26%|██▌       | 1791/6837 [20:41<20:00,  4.20it/s]
 26%|██▌       | 1792/6837 [20:41<20:01,  4.20it/s]
 26%|██▌       | 1793/6837 [20:42<20:14,  4.15it/s]
 26%|██▌       | 1794/6837 [20:42<20:12,  4.16it/s]
 26%|██▋       | 1795/6837 [20:42<20:09,  4.17it/s]
 26%|██▋       | 1796/6837 [20:42<20:09,  4.17it/s]
 26%|██▋       | 1797/6837 [20:43<20:06,  4.18it/s]
 26%|██▋       | 1798/6837 [20:43<20:07,  4.17it/s]
 26%|██▋       | 1799/6837 [20:43<20:12,  4.15it/s]
 26%|██▋       | 1800/6837 [20:43<20:11,  4.16it/s]
                                                   
{'loss': 4.6599, 'grad_norm': 0.14138217270374298, 'learning_rate': 0.005526016658843222, 'epoch': 0.05}
+
 26%|██▋       | 1800/6837 [20:43<20:11,  4.16it/s]
 26%|██▋       | 1801/6837 [20:44<20:12,  4.15it/s]
 26%|██▋       | 1802/6837 [20:44<20:06,  4.17it/s]
 26%|██▋       | 1803/6837 [20:44<20:05,  4.18it/s]
 26%|██▋       | 1804/6837 [20:44<20:08,  4.17it/s]
 26%|██▋       | 1805/6837 [20:45<20:08,  4.16it/s]
 26%|██▋       | 1806/6837 [20:45<20:09,  4.16it/s]
 26%|██▋       | 1807/6837 [20:45<20:07,  4.17it/s]
 26%|██▋       | 1808/6837 [20:45<20:03,  4.18it/s]
 26%|██▋       | 1809/6837 [20:45<20:02,  4.18it/s]
 26%|██▋       | 1810/6837 [20:46<20:02,  4.18it/s]
 26%|██▋       | 1811/6837 [20:46<20:00,  4.19it/s]
 27%|██▋       | 1812/6837 [20:46<20:00,  4.19it/s]
 27%|██▋       | 1813/6837 [20:46<20:13,  4.14it/s]
 27%|██▋       | 1814/6837 [20:47<20:07,  4.16it/s]
 27%|██▋       | 1815/6837 [20:47<20:10,  4.15it/s]
 27%|██▋       | 1816/6837 [20:47<20:06,  4.16it/s]
 27%|██▋       | 1817/6837 [20:47<20:00,  4.18it/s]
 27%|██▋       | 1818/6837 [20:48<20:01,  4.18it/s]
 27%|██▋       | 1819/6837 [20:48<20:01,  4.18it/s]
 27%|██▋       | 1820/6837 [20:48<19:58,  4.19it/s]
 27%|██▋       | 1821/6837 [20:48<19:57,  4.19it/s]
 27%|██▋       | 1822/6837 [20:49<19:58,  4.18it/s]
 27%|██▋       | 1823/6837 [20:49<19:58,  4.18it/s]
 27%|██▋       | 1824/6837 [20:49<19:56,  4.19it/s]
 27%|██▋       | 1825/6837 [20:49<19:54,  4.20it/s]{'loss': 4.6691, 'grad_norm': 0.13456492125988007, 'learning_rate': 0.0055051533276595024, 'epoch': 0.05}
+
                                                   

 27%|██▋       | 1825/6837 [20:49<19:54,  4.20it/s]
 27%|██▋       | 1826/6837 [20:50<20:00,  4.17it/s]
 27%|██▋       | 1827/6837 [20:50<19:59,  4.18it/s]
 27%|██▋       | 1828/6837 [20:50<19:56,  4.19it/s]
 27%|██▋       | 1829/6837 [20:50<19:56,  4.19it/s]
 27%|██▋       | 1830/6837 [20:51<19:57,  4.18it/s]
 27%|██▋       | 1831/6837 [20:51<19:58,  4.18it/s]
 27%|██▋       | 1832/6837 [20:51<19:56,  4.18it/s]
 27%|██▋       | 1833/6837 [20:51<19:58,  4.18it/s]
 27%|██▋       | 1834/6837 [20:51<19:56,  4.18it/s]
 27%|██▋       | 1835/6837 [20:52<19:55,  4.19it/s]
 27%|██▋       | 1836/6837 [20:52<19:52,  4.19it/s]
 27%|██▋       | 1837/6837 [20:52<19:52,  4.19it/s]
 27%|██▋       | 1838/6837 [20:52<19:54,  4.18it/s]
 27%|██▋       | 1839/6837 [20:53<19:51,  4.19it/s]
 27%|██▋       | 1840/6837 [20:53<19:49,  4.20it/s]
 27%|██▋       | 1841/6837 [20:53<19:46,  4.21it/s]
 27%|██▋       | 1842/6837 [20:53<19:51,  4.19it/s]
 27%|██▋       | 1843/6837 [20:54<19:48,  4.20it/s]
 27%|██▋       | 1844/6837 [20:54<19:46,  4.21it/s]
 27%|██▋       | 1845/6837 [20:54<19:47,  4.20it/s]
 27%|██▋       | 1846/6837 [20:54<19:49,  4.19it/s]
 27%|██▋       | 1847/6837 [20:55<19:49,  4.20it/s]
 27%|██▋       | 1848/6837 [20:55<19:48,  4.20it/s]
 27%|██▋       | 1849/6837 [20:55<19:46,  4.20it/s]
 27%|██▋       | 1850/6837 [20:55<19:49,  4.19it/s]
                                                   
{'loss': 4.6531, 'grad_norm': 0.14033429324626923, 'learning_rate': 0.005483881832804299, 'epoch': 0.05}
+
 27%|██▋       | 1850/6837 [20:55<19:49,  4.19it/s]
 27%|██▋       | 1851/6837 [20:56<19:50,  4.19it/s]
 27%|██▋       | 1852/6837 [20:56<19:48,  4.19it/s]
 27%|██▋       | 1853/6837 [20:56<19:47,  4.20it/s]
 27%|██▋       | 1854/6837 [20:56<19:47,  4.20it/s]
 27%|██▋       | 1855/6837 [20:56<19:44,  4.21it/s]
 27%|██▋       | 1856/6837 [20:57<19:42,  4.21it/s]
 27%|██▋       | 1857/6837 [20:57<19:44,  4.20it/s]
 27%|██▋       | 1858/6837 [20:57<19:43,  4.21it/s]
 27%|██▋       | 1859/6837 [20:57<19:41,  4.21it/s]
 27%|██▋       | 1860/6837 [20:58<19:41,  4.21it/s]
 27%|██▋       | 1861/6837 [20:58<19:41,  4.21it/s]
 27%|██▋       | 1862/6837 [20:58<19:41,  4.21it/s]
 27%|██▋       | 1863/6837 [20:58<19:41,  4.21it/s]
 27%|██▋       | 1864/6837 [20:59<19:42,  4.21it/s]
 27%|██▋       | 1865/6837 [20:59<19:43,  4.20it/s]
 27%|██▋       | 1866/6837 [20:59<19:46,  4.19it/s]
 27%|██▋       | 1867/6837 [20:59<19:42,  4.20it/s]
 27%|██▋       | 1868/6837 [21:00<19:42,  4.20it/s]
 27%|██▋       | 1869/6837 [21:00<19:41,  4.21it/s]
 27%|██▋       | 1870/6837 [21:00<19:39,  4.21it/s]
 27%|██▋       | 1871/6837 [21:00<19:39,  4.21it/s]
 27%|██▋       | 1872/6837 [21:01<19:42,  4.20it/s]
 27%|██▋       | 1873/6837 [21:01<19:42,  4.20it/s]
 27%|██▋       | 1874/6837 [21:01<19:40,  4.20it/s]
 27%|██▋       | 1875/6837 [21:01<19:40,  4.20it/s]
                                                   
{'loss': 4.6486, 'grad_norm': 0.1333823949098587, 'learning_rate': 0.005462205640034116, 'epoch': 0.05}
+
 27%|██▋       | 1875/6837 [21:01<19:40,  4.20it/s]
 27%|██▋       | 1876/6837 [21:01<19:42,  4.20it/s]
 27%|██▋       | 1877/6837 [21:02<19:41,  4.20it/s]
 27%|██▋       | 1878/6837 [21:02<19:39,  4.20it/s]
 27%|██▋       | 1879/6837 [21:02<19:40,  4.20it/s]
 27%|██▋       | 1880/6837 [21:02<19:41,  4.19it/s]
 28%|██▊       | 1881/6837 [21:03<19:41,  4.20it/s]
 28%|██▊       | 1882/6837 [21:03<19:41,  4.19it/s]
 28%|██▊       | 1883/6837 [21:03<19:39,  4.20it/s]
 28%|██▊       | 1884/6837 [21:03<19:38,  4.20it/s]
 28%|██▊       | 1885/6837 [21:04<19:37,  4.21it/s]
 28%|██▊       | 1886/6837 [21:04<19:36,  4.21it/s]
 28%|██▊       | 1887/6837 [21:04<19:38,  4.20it/s]
 28%|██▊       | 1888/6837 [21:04<19:37,  4.20it/s]
 28%|██▊       | 1889/6837 [21:05<19:36,  4.20it/s]
 28%|██▊       | 1890/6837 [21:05<19:35,  4.21it/s]
 28%|██▊       | 1891/6837 [21:05<19:38,  4.20it/s]
 28%|██▊       | 1892/6837 [21:05<19:37,  4.20it/s]
 28%|██▊       | 1893/6837 [21:06<19:37,  4.20it/s]
 28%|██▊       | 1894/6837 [21:06<19:34,  4.21it/s]
 28%|██▊       | 1895/6837 [21:06<19:34,  4.21it/s]
 28%|██▊       | 1896/6837 [21:06<19:36,  4.20it/s]
 28%|██▊       | 1897/6837 [21:06<19:33,  4.21it/s]
 28%|██▊       | 1898/6837 [21:07<19:34,  4.21it/s]
 28%|██▊       | 1899/6837 [21:07<19:32,  4.21it/s]
 28%|██▊       | 1900/6837 [21:07<19:30,  4.22it/s]
                                                   
{'loss': 4.6442, 'grad_norm': 0.13534818589687347, 'learning_rate': 0.005440128281042736, 'epoch': 0.06}
+
 28%|██▊       | 1900/6837 [21:07<19:30,  4.22it/s]
 28%|██▊       | 1901/6837 [21:07<19:34,  4.20it/s]
 28%|██▊       | 1902/6837 [21:08<19:36,  4.20it/s]
 28%|██▊       | 1903/6837 [21:08<19:34,  4.20it/s]
 28%|██▊       | 1904/6837 [21:08<19:33,  4.20it/s]
 28%|██▊       | 1905/6837 [21:08<19:33,  4.20it/s]
 28%|██▊       | 1906/6837 [21:09<19:33,  4.20it/s]
 28%|██▊       | 1907/6837 [21:09<19:32,  4.20it/s]
 28%|██▊       | 1908/6837 [21:09<19:31,  4.21it/s]
 28%|██▊       | 1909/6837 [21:09<19:28,  4.22it/s]
 28%|██▊       | 1910/6837 [21:10<19:29,  4.21it/s]
 28%|██▊       | 1911/6837 [21:10<19:30,  4.21it/s]
 28%|██▊       | 1912/6837 [21:10<19:30,  4.21it/s]
 28%|██▊       | 1913/6837 [21:10<19:29,  4.21it/s]
 28%|██▊       | 1914/6837 [21:11<19:30,  4.21it/s]
 28%|██▊       | 1915/6837 [21:11<19:28,  4.21it/s]
 28%|██▊       | 1916/6837 [21:11<19:27,  4.21it/s]
 28%|██▊       | 1917/6837 [21:11<19:30,  4.20it/s]
 28%|██▊       | 1918/6837 [21:11<19:31,  4.20it/s]
 28%|██▊       | 1919/6837 [21:12<19:31,  4.20it/s]
 28%|██▊       | 1920/6837 [21:12<19:30,  4.20it/s]
 28%|██▊       | 1921/6837 [21:12<19:30,  4.20it/s]
 28%|██▊       | 1922/6837 [21:12<19:29,  4.20it/s]
 28%|██▊       | 1923/6837 [21:13<19:27,  4.21it/s]
 28%|██▊       | 1924/6837 [21:13<20:11,  4.06it/s]
 28%|██▊       | 1925/6837 [21:13<20:02,  4.08it/s]
                                                   
{'loss': 4.6472, 'grad_norm': 0.13941456377506256, 'learning_rate': 0.005417653352885793, 'epoch': 0.06}
+
 28%|██▊       | 1925/6837 [21:13<20:02,  4.08it/s]
 28%|██▊       | 1926/6837 [21:13<19:54,  4.11it/s]
 28%|██▊       | 1927/6837 [21:14<19:46,  4.14it/s]
 28%|██▊       | 1928/6837 [21:14<19:43,  4.15it/s]
 28%|██▊       | 1929/6837 [21:14<19:39,  4.16it/s]
 28%|██▊       | 1930/6837 [21:14<19:36,  4.17it/s]
 28%|██▊       | 1931/6837 [21:15<19:31,  4.19it/s]
 28%|██▊       | 1932/6837 [21:15<19:31,  4.19it/s]
 28%|██▊       | 1933/6837 [21:15<19:31,  4.19it/s]
 28%|██▊       | 1934/6837 [21:15<19:31,  4.19it/s]
 28%|██▊       | 1935/6837 [21:16<19:30,  4.19it/s]
 28%|██▊       | 1936/6837 [21:16<19:30,  4.19it/s]
 28%|██▊       | 1937/6837 [21:16<19:29,  4.19it/s]
 28%|██▊       | 1938/6837 [21:16<19:28,  4.19it/s]
 28%|██▊       | 1939/6837 [21:16<19:28,  4.19it/s]
 28%|██▊       | 1940/6837 [21:17<19:29,  4.19it/s]
 28%|██▊       | 1941/6837 [21:17<19:29,  4.19it/s]
 28%|██▊       | 1942/6837 [21:17<19:25,  4.20it/s]
 28%|██▊       | 1943/6837 [21:17<19:25,  4.20it/s]
 28%|██▊       | 1944/6837 [21:18<19:25,  4.20it/s]
 28%|██▊       | 1945/6837 [21:18<19:29,  4.18it/s]
 28%|██▊       | 1946/6837 [21:18<19:26,  4.19it/s]
 28%|██▊       | 1947/6837 [21:18<19:28,  4.19it/s]
 28%|██▊       | 1948/6837 [21:19<19:28,  4.19it/s]
 29%|██▊       | 1949/6837 [21:19<19:24,  4.20it/s]
 29%|██▊       | 1950/6837 [21:19<19:26,  4.19it/s]{'loss': 4.6264, 'grad_norm': 0.14234456419944763, 'learning_rate': 0.0053947845173947205, 'epoch': 0.06}
                                                   
+
 29%|██▊       | 1950/6837 [21:19<19:26,  4.19it/s]
 29%|██▊       | 1951/6837 [21:19<19:29,  4.18it/s]
 29%|██▊       | 1952/6837 [21:20<19:26,  4.19it/s]
 29%|██▊       | 1953/6837 [21:20<19:22,  4.20it/s]
 29%|██▊       | 1954/6837 [21:20<19:22,  4.20it/s]
 29%|██▊       | 1955/6837 [21:20<19:21,  4.20it/s]
 29%|██▊       | 1956/6837 [21:21<19:22,  4.20it/s]
 29%|██▊       | 1957/6837 [21:21<19:22,  4.20it/s]
 29%|██▊       | 1958/6837 [21:21<19:23,  4.19it/s]
 29%|██▊       | 1959/6837 [21:21<19:21,  4.20it/s]
 29%|██▊       | 1960/6837 [21:22<19:20,  4.20it/s]
 29%|██▊       | 1961/6837 [21:22<19:20,  4.20it/s]
 29%|██▊       | 1962/6837 [21:22<19:21,  4.20it/s]
 29%|██▊       | 1963/6837 [21:22<19:18,  4.21it/s]
 29%|██▊       | 1964/6837 [21:22<19:16,  4.21it/s]
 29%|██▊       | 1965/6837 [21:23<19:15,  4.22it/s]
 29%|██▉       | 1966/6837 [21:23<19:13,  4.22it/s]
 29%|██▉       | 1967/6837 [21:23<19:19,  4.20it/s]
 29%|██▉       | 1968/6837 [21:23<19:16,  4.21it/s]
 29%|██▉       | 1969/6837 [21:24<19:16,  4.21it/s]
 29%|██▉       | 1970/6837 [21:24<19:16,  4.21it/s]
 29%|██▉       | 1971/6837 [21:24<19:14,  4.21it/s]
 29%|██▉       | 1972/6837 [21:24<19:11,  4.22it/s]
 29%|██▉       | 1973/6837 [21:25<19:12,  4.22it/s]
 29%|██▉       | 1974/6837 [21:25<19:10,  4.23it/s]
 29%|██▉       | 1975/6837 [21:25<19:10,  4.23it/s]{'loss': 4.631, 'grad_norm': 0.13691064715385437, 'learning_rate': 0.005371525500580118, 'epoch': 0.06}
                                                   
+
 29%|██▉       | 1975/6837 [21:25<19:10,  4.23it/s]
 29%|██▉       | 1976/6837 [21:25<19:14,  4.21it/s]
 29%|██▉       | 1977/6837 [21:26<19:14,  4.21it/s]
 29%|██▉       | 1978/6837 [21:26<19:12,  4.22it/s]
 29%|██▉       | 1979/6837 [21:26<19:11,  4.22it/s]
 29%|██▉       | 1980/6837 [21:26<19:12,  4.21it/s]
 29%|██▉       | 1981/6837 [21:26<19:13,  4.21it/s]
 29%|██▉       | 1982/6837 [21:27<19:11,  4.22it/s]
 29%|██▉       | 1983/6837 [21:27<19:10,  4.22it/s]
 29%|██▉       | 1984/6837 [21:27<19:11,  4.22it/s]
 29%|██▉       | 1985/6837 [21:27<19:10,  4.22it/s]
 29%|██▉       | 1986/6837 [21:28<19:09,  4.22it/s]
 29%|██▉       | 1987/6837 [21:28<19:09,  4.22it/s]
 29%|██▉       | 1988/6837 [21:28<19:10,  4.22it/s]
 29%|██▉       | 1989/6837 [21:28<19:09,  4.22it/s]
 29%|██▉       | 1990/6837 [21:29<19:09,  4.21it/s]
 29%|██▉       | 1991/6837 [21:29<19:10,  4.21it/s]
 29%|██▉       | 1992/6837 [21:29<19:09,  4.22it/s]
 29%|██▉       | 1993/6837 [21:29<19:08,  4.22it/s]
 29%|██▉       | 1994/6837 [21:30<19:06,  4.22it/s]
 29%|██▉       | 1995/6837 [21:30<19:07,  4.22it/s]
 29%|██▉       | 1996/6837 [21:30<19:09,  4.21it/s]
 29%|██▉       | 1997/6837 [21:30<19:07,  4.22it/s]
 29%|██▉       | 1998/6837 [21:31<19:08,  4.21it/s]
 29%|██▉       | 1999/6837 [21:31<19:54,  4.05it/s]
 29%|██▉       | 2000/6837 [21:31<20:21,  3.96it/s]
                                                   
{'loss': 4.626, 'grad_norm': 0.14199486374855042, 'learning_rate': 0.0053478800920246814, 'epoch': 0.06}
+
 29%|██▉       | 2000/6837 [21:31<20:21,  3.96it/s]
 29%|██▉       | 2001/6837 [21:31<20:43,  3.89it/s]
 29%|██▉       | 2002/6837 [21:32<20:50,  3.87it/s]
 29%|██▉       | 2003/6837 [21:32<21:01,  3.83it/s]
 29%|██▉       | 2004/6837 [21:32<21:10,  3.80it/s]
 29%|██▉       | 2005/6837 [21:32<21:16,  3.79it/s]
 29%|██▉       | 2006/6837 [21:33<21:18,  3.78it/s]
 29%|██▉       | 2007/6837 [21:33<21:21,  3.77it/s]
 29%|██▉       | 2008/6837 [21:33<21:27,  3.75it/s]
 29%|██▉       | 2009/6837 [21:33<21:14,  3.79it/s]
 29%|██▉       | 2010/6837 [21:34<21:15,  3.78it/s]
 29%|██▉       | 2011/6837 [21:34<21:17,  3.78it/s]
 29%|██▉       | 2012/6837 [21:34<21:21,  3.77it/s]
 29%|██▉       | 2013/6837 [21:35<21:19,  3.77it/s]
 29%|██▉       | 2014/6837 [21:35<21:14,  3.78it/s]
 29%|██▉       | 2015/6837 [21:35<20:43,  3.88it/s]
 29%|██▉       | 2016/6837 [21:35<20:14,  3.97it/s]
 30%|██▉       | 2017/6837 [21:35<19:51,  4.04it/s]
 30%|██▉       | 2018/6837 [21:36<19:35,  4.10it/s]
 30%|██▉       | 2019/6837 [21:36<19:22,  4.14it/s]
 30%|██▉       | 2020/6837 [21:36<19:17,  4.16it/s]
 30%|██▉       | 2021/6837 [21:36<19:11,  4.18it/s]
 30%|██▉       | 2022/6837 [21:37<19:06,  4.20it/s]
 30%|██▉       | 2023/6837 [21:37<19:04,  4.21it/s]
 30%|██▉       | 2024/6837 [21:37<19:01,  4.22it/s]
 30%|██▉       | 2025/6837 [21:37<19:02,  4.21it/s]
                                                   
{'loss': 4.6118, 'grad_norm': 0.13648205995559692, 'learning_rate': 0.005323852144265764, 'epoch': 0.06}
+
 30%|██▉       | 2025/6837 [21:37<19:02,  4.21it/s]
 30%|██▉       | 2026/6837 [21:38<19:00,  4.22it/s]
 30%|██▉       | 2027/6837 [21:38<18:59,  4.22it/s]
 30%|██▉       | 2028/6837 [21:38<19:00,  4.22it/s]
 30%|██▉       | 2029/6837 [21:38<18:59,  4.22it/s]
 30%|██▉       | 2030/6837 [21:39<19:00,  4.21it/s]
 30%|██▉       | 2031/6837 [21:39<18:57,  4.23it/s]
 30%|██▉       | 2032/6837 [21:39<18:57,  4.22it/s]
 30%|██▉       | 2033/6837 [21:39<18:54,  4.24it/s]
 30%|██▉       | 2034/6837 [21:40<18:55,  4.23it/s]
 30%|██▉       | 2035/6837 [21:40<18:55,  4.23it/s]
 30%|██▉       | 2036/6837 [21:40<18:55,  4.23it/s]
 30%|██▉       | 2037/6837 [21:40<19:01,  4.21it/s]
 30%|██▉       | 2038/6837 [21:40<19:01,  4.20it/s]
 30%|██▉       | 2039/6837 [21:41<18:57,  4.22it/s]
 30%|██▉       | 2040/6837 [21:41<18:58,  4.21it/s]
 30%|██▉       | 2041/6837 [21:41<18:55,  4.22it/s]
 30%|██▉       | 2042/6837 [21:41<18:54,  4.23it/s]
 30%|██▉       | 2043/6837 [21:42<18:55,  4.22it/s]
 30%|██▉       | 2044/6837 [21:42<18:57,  4.22it/s]
 30%|██▉       | 2045/6837 [21:42<18:54,  4.22it/s]
 30%|██▉       | 2046/6837 [21:42<18:53,  4.23it/s]
 30%|██▉       | 2047/6837 [21:43<18:53,  4.22it/s]
 30%|██▉       | 2048/6837 [21:43<19:33,  4.08it/s]
 30%|██▉       | 2049/6837 [21:43<19:28,  4.10it/s]
 30%|██▉       | 2050/6837 [21:43<19:20,  4.13it/s]{'loss': 4.6142, 'grad_norm': 0.12774842977523804, 'learning_rate': 0.005299445572167684, 'epoch': 0.06}
                                                   
+
 30%|██▉       | 2050/6837 [21:43<19:20,  4.13it/s]
 30%|██▉       | 2051/6837 [21:44<19:14,  4.14it/s]
 30%|███       | 2052/6837 [21:44<19:08,  4.17it/s]
 30%|███       | 2053/6837 [21:44<19:03,  4.18it/s]
 30%|███       | 2054/6837 [21:44<18:59,  4.20it/s]
 30%|███       | 2055/6837 [21:45<18:57,  4.20it/s]
 30%|███       | 2056/6837 [21:45<18:56,  4.21it/s]
 30%|███       | 2057/6837 [21:45<18:56,  4.21it/s]
 30%|███       | 2058/6837 [21:45<18:56,  4.21it/s]
 30%|███       | 2059/6837 [21:45<18:54,  4.21it/s]
 30%|███       | 2060/6837 [21:46<18:52,  4.22it/s]
 30%|███       | 2061/6837 [21:46<18:52,  4.22it/s]
 30%|███       | 2062/6837 [21:46<18:50,  4.22it/s]
 30%|███       | 2063/6837 [21:46<18:48,  4.23it/s]
 30%|███       | 2064/6837 [21:47<18:47,  4.23it/s]
 30%|███       | 2065/6837 [21:47<18:47,  4.23it/s]
 30%|███       | 2066/6837 [21:47<18:50,  4.22it/s]
 30%|███       | 2067/6837 [21:47<18:51,  4.21it/s]
 30%|███       | 2068/6837 [21:48<18:51,  4.22it/s]
 30%|███       | 2069/6837 [21:48<18:51,  4.21it/s]
 30%|███       | 2070/6837 [21:48<18:49,  4.22it/s]
 30%|███       | 2071/6837 [21:48<18:48,  4.22it/s]
 30%|███       | 2072/6837 [21:49<18:46,  4.23it/s]
 30%|███       | 2073/6837 [21:49<18:46,  4.23it/s]
 30%|███       | 2074/6837 [21:49<18:45,  4.23it/s]
 30%|███       | 2075/6837 [21:49<18:44,  4.23it/s]
                                                   
{'loss': 4.6038, 'grad_norm': 0.13706155121326447, 'learning_rate': 0.005274664352283877, 'epoch': 0.06}
+
 30%|███       | 2075/6837 [21:49<18:44,  4.23it/s]
 30%|███       | 2076/6837 [21:49<18:48,  4.22it/s]
 30%|███       | 2077/6837 [21:50<18:49,  4.22it/s]
 30%|███       | 2078/6837 [21:50<18:47,  4.22it/s]
 30%|███       | 2079/6837 [21:50<18:46,  4.22it/s]
 30%|███       | 2080/6837 [21:50<18:49,  4.21it/s]
 30%|███       | 2081/6837 [21:51<25:31,  3.10it/s]
 30%|███       | 2082/6837 [21:51<23:30,  3.37it/s]
 30%|███       | 2083/6837 [21:51<22:07,  3.58it/s]
 30%|███       | 2084/6837 [21:52<21:05,  3.76it/s]
 30%|███       | 2085/6837 [21:52<20:24,  3.88it/s]
 31%|███       | 2086/6837 [21:52<19:59,  3.96it/s]
 31%|███       | 2087/6837 [21:52<19:35,  4.04it/s]
 31%|███       | 2088/6837 [21:53<19:18,  4.10it/s]
 31%|███       | 2089/6837 [21:53<19:08,  4.13it/s]
 31%|███       | 2090/6837 [21:53<19:02,  4.16it/s]
 31%|███       | 2091/6837 [21:53<18:56,  4.17it/s]
 31%|███       | 2092/6837 [21:54<18:52,  4.19it/s]
 31%|███       | 2093/6837 [21:54<18:51,  4.19it/s]
 31%|███       | 2094/6837 [21:54<18:48,  4.20it/s]
 31%|███       | 2095/6837 [21:54<18:48,  4.20it/s]
 31%|███       | 2096/6837 [21:55<18:46,  4.21it/s]
 31%|███       | 2097/6837 [21:55<18:44,  4.21it/s]
 31%|███       | 2098/6837 [21:55<18:49,  4.20it/s]
 31%|███       | 2099/6837 [21:55<18:46,  4.20it/s]
 31%|███       | 2100/6837 [21:55<18:46,  4.21it/s]
                                                   
{'loss': 4.6014, 'grad_norm': 0.13223719596862793, 'learning_rate': 0.005249512522209001, 'epoch': 0.06}
+
 31%|███       | 2100/6837 [21:55<18:46,  4.21it/s]
 31%|███       | 2101/6837 [21:56<18:46,  4.21it/s]
 31%|███       | 2102/6837 [21:56<18:43,  4.22it/s]
 31%|███       | 2103/6837 [21:56<18:40,  4.22it/s]
 31%|███       | 2104/6837 [21:56<18:39,  4.23it/s]
 31%|███       | 2105/6837 [21:57<18:38,  4.23it/s]
 31%|███       | 2106/6837 [21:57<18:39,  4.23it/s]
 31%|███       | 2107/6837 [21:57<18:38,  4.23it/s]
 31%|███       | 2108/6837 [21:57<18:37,  4.23it/s]
 31%|███       | 2109/6837 [21:58<18:38,  4.23it/s]
 31%|███       | 2110/6837 [21:58<18:36,  4.23it/s]
 31%|███       | 2111/6837 [21:58<18:36,  4.23it/s]
 31%|███       | 2112/6837 [21:58<18:34,  4.24it/s]
 31%|███       | 2113/6837 [21:59<18:36,  4.23it/s]
 31%|███       | 2114/6837 [21:59<18:35,  4.23it/s]
 31%|███       | 2115/6837 [21:59<18:34,  4.24it/s]
 31%|███       | 2116/6837 [21:59<18:36,  4.23it/s]
 31%|███       | 2117/6837 [21:59<18:37,  4.23it/s]
 31%|███       | 2118/6837 [22:00<18:34,  4.23it/s]
 31%|███       | 2119/6837 [22:00<18:35,  4.23it/s]
 31%|███       | 2120/6837 [22:00<18:37,  4.22it/s]
 31%|███       | 2121/6837 [22:00<18:39,  4.21it/s]
 31%|███       | 2122/6837 [22:01<18:36,  4.22it/s]
 31%|███       | 2123/6837 [22:01<18:36,  4.22it/s]
 31%|███       | 2124/6837 [22:01<18:37,  4.22it/s]
 31%|███       | 2125/6837 [22:01<18:37,  4.22it/s]
                                                   
{'loss': 4.6248, 'grad_norm': 0.13747967779636383, 'learning_rate': 0.005223994179921092, 'epoch': 0.06}
+
 31%|███       | 2125/6837 [22:01<18:37,  4.22it/s]
 31%|███       | 2126/6837 [22:02<18:37,  4.22it/s]
 31%|███       | 2127/6837 [22:02<18:35,  4.22it/s]
 31%|███       | 2128/6837 [22:02<18:37,  4.22it/s]
 31%|███       | 2129/6837 [22:02<18:35,  4.22it/s]
 31%|███       | 2130/6837 [22:03<18:36,  4.21it/s]
 31%|███       | 2131/6837 [22:03<18:33,  4.23it/s]
 31%|███       | 2132/6837 [22:03<18:32,  4.23it/s]
 31%|███       | 2133/6837 [22:03<19:37,  4.00it/s]
 31%|███       | 2134/6837 [22:04<20:07,  3.89it/s]
 31%|███       | 2135/6837 [22:04<20:22,  3.85it/s]
 31%|███       | 2136/6837 [22:04<20:26,  3.83it/s]
 31%|███▏      | 2137/6837 [22:04<20:08,  3.89it/s]
 31%|███▏      | 2138/6837 [22:05<19:39,  3.99it/s]
 31%|███▏      | 2139/6837 [22:05<19:15,  4.07it/s]
 31%|███▏      | 2140/6837 [22:05<19:01,  4.11it/s]
 31%|███▏      | 2141/6837 [22:05<18:52,  4.15it/s]
 31%|███▏      | 2142/6837 [22:06<18:45,  4.17it/s]
 31%|███▏      | 2143/6837 [22:06<19:19,  4.05it/s]
 31%|███▏      | 2144/6837 [22:06<19:43,  3.97it/s]
 31%|███▏      | 2145/6837 [22:06<19:19,  4.05it/s]
 31%|███▏      | 2146/6837 [22:07<19:04,  4.10it/s]
 31%|███▏      | 2147/6837 [22:07<18:53,  4.14it/s]
 31%|███▏      | 2148/6837 [22:07<18:46,  4.16it/s]
 31%|███▏      | 2149/6837 [22:07<18:42,  4.18it/s]
 31%|███▏      | 2150/6837 [22:08<18:37,  4.19it/s]
                                                   
{'loss': 4.6057, 'grad_norm': 0.16631564497947693, 'learning_rate': 0.005198113483113877, 'epoch': 0.06}
+
 31%|███▏      | 2150/6837 [22:08<18:37,  4.19it/s]
 31%|███▏      | 2151/6837 [22:08<18:36,  4.20it/s]
 31%|███▏      | 2152/6837 [22:08<18:33,  4.21it/s]
 31%|███▏      | 2153/6837 [22:08<18:31,  4.21it/s]
 32%|███▏      | 2154/6837 [22:08<18:29,  4.22it/s]
 32%|███▏      | 2155/6837 [22:09<18:29,  4.22it/s]
 32%|███▏      | 2156/6837 [22:09<18:27,  4.23it/s]
 32%|███▏      | 2157/6837 [22:09<18:25,  4.23it/s]
 32%|███▏      | 2158/6837 [22:09<18:23,  4.24it/s]
 32%|███▏      | 2159/6837 [22:10<18:25,  4.23it/s]
 32%|███▏      | 2160/6837 [22:10<18:28,  4.22it/s]
 32%|███▏      | 2161/6837 [22:10<18:28,  4.22it/s]
 32%|███▏      | 2162/6837 [22:10<18:26,  4.22it/s]
 32%|███▏      | 2163/6837 [22:11<18:25,  4.23it/s]
 32%|███▏      | 2164/6837 [22:11<18:23,  4.23it/s]
 32%|███▏      | 2165/6837 [22:11<18:24,  4.23it/s]
 32%|███▏      | 2166/6837 [22:11<18:23,  4.23it/s]
 32%|███▏      | 2167/6837 [22:12<18:22,  4.24it/s]
 32%|███▏      | 2168/6837 [22:12<18:21,  4.24it/s]
 32%|███▏      | 2169/6837 [22:12<18:20,  4.24it/s]
 32%|███▏      | 2170/6837 [22:12<18:21,  4.24it/s]
 32%|███▏      | 2171/6837 [22:12<18:23,  4.23it/s]
 32%|███▏      | 2172/6837 [22:13<19:02,  4.08it/s]
 32%|███▏      | 2173/6837 [22:13<19:38,  3.96it/s]
 32%|███▏      | 2174/6837 [22:13<20:04,  3.87it/s]
 32%|███▏      | 2175/6837 [22:14<20:09,  3.86it/s]
                                                   
{'loss': 4.6092, 'grad_norm': 0.14016544818878174, 'learning_rate': 0.005171874648519365, 'epoch': 0.06}
+
 32%|███▏      | 2175/6837 [22:14<20:09,  3.86it/s]
 32%|███▏      | 2176/6837 [22:14<20:15,  3.83it/s]
 32%|███▏      | 2177/6837 [22:14<20:22,  3.81it/s]
 32%|███▏      | 2178/6837 [22:14<20:27,  3.80it/s]
 32%|███▏      | 2179/6837 [22:15<20:27,  3.79it/s]
 32%|███▏      | 2180/6837 [22:15<20:28,  3.79it/s]
 32%|███▏      | 2181/6837 [22:15<20:24,  3.80it/s]
 32%|███▏      | 2182/6837 [22:15<19:47,  3.92it/s]
 32%|███▏      | 2183/6837 [22:16<19:19,  4.02it/s]
 32%|███▏      | 2184/6837 [22:16<19:00,  4.08it/s]
 32%|███▏      | 2185/6837 [22:16<18:50,  4.11it/s]
 32%|███▏      | 2186/6837 [22:16<18:43,  4.14it/s]
 32%|███▏      | 2187/6837 [22:17<18:35,  4.17it/s]
 32%|███▏      | 2188/6837 [22:17<18:29,  4.19it/s]
 32%|███▏      | 2189/6837 [22:17<18:25,  4.21it/s]
 32%|███▏      | 2190/6837 [22:17<18:24,  4.21it/s]
 32%|███▏      | 2191/6837 [22:17<18:24,  4.21it/s]
 32%|███▏      | 2192/6837 [22:18<18:25,  4.20it/s]
 32%|███▏      | 2193/6837 [22:18<18:23,  4.21it/s]
 32%|███▏      | 2194/6837 [22:18<18:22,  4.21it/s]
 32%|███▏      | 2195/6837 [22:18<18:20,  4.22it/s]
 32%|███▏      | 2196/6837 [22:19<18:23,  4.21it/s]
 32%|███▏      | 2197/6837 [22:19<18:21,  4.21it/s]
 32%|███▏      | 2198/6837 [22:19<18:21,  4.21it/s]
 32%|███▏      | 2199/6837 [22:19<18:19,  4.22it/s]
 32%|███▏      | 2200/6837 [22:20<18:20,  4.21it/s]
                                                   
{'loss': 4.5938, 'grad_norm': 0.14113828539848328, 'learning_rate': 0.005145281951220822, 'epoch': 0.06}
+
 32%|███▏      | 2200/6837 [22:20<18:20,  4.21it/s]
 32%|███▏      | 2201/6837 [22:20<18:22,  4.21it/s]
 32%|███▏      | 2202/6837 [22:20<18:21,  4.21it/s]
 32%|███▏      | 2203/6837 [22:20<18:22,  4.20it/s]
 32%|███▏      | 2204/6837 [22:21<18:21,  4.21it/s]
 32%|███▏      | 2205/6837 [22:21<18:20,  4.21it/s]
 32%|███▏      | 2206/6837 [22:21<18:19,  4.21it/s]
 32%|███▏      | 2207/6837 [22:21<18:20,  4.21it/s]
 32%|███▏      | 2208/6837 [22:22<18:18,  4.21it/s]
 32%|███▏      | 2209/6837 [22:22<18:18,  4.21it/s]
 32%|███▏      | 2210/6837 [22:22<18:17,  4.22it/s]
 32%|███▏      | 2211/6837 [22:22<18:17,  4.21it/s]
 32%|███▏      | 2212/6837 [22:22<18:15,  4.22it/s]
 32%|███▏      | 2213/6837 [22:23<18:13,  4.23it/s]
 32%|███▏      | 2214/6837 [22:23<18:13,  4.23it/s]
 32%|███▏      | 2215/6837 [22:23<18:13,  4.23it/s]
 32%|███▏      | 2216/6837 [22:23<18:12,  4.23it/s]
 32%|███▏      | 2217/6837 [22:24<18:10,  4.24it/s]
 32%|███▏      | 2218/6837 [22:24<18:10,  4.23it/s]
 32%|███▏      | 2219/6837 [22:24<18:10,  4.24it/s]
 32%|███▏      | 2220/6837 [22:24<18:07,  4.25it/s]
 32%|███▏      | 2221/6837 [22:25<18:06,  4.25it/s]
 32%|███▏      | 2222/6837 [22:25<18:07,  4.24it/s]
 33%|███▎      | 2223/6837 [22:25<18:05,  4.25it/s]
 33%|███▎      | 2224/6837 [22:25<18:05,  4.25it/s]
 33%|███▎      | 2225/6837 [22:26<18:08,  4.24it/s]
                                                   
{'loss': 4.6016, 'grad_norm': 0.13770692050457, 'learning_rate': 0.005118339723956226, 'epoch': 0.07}
+
 33%|███▎      | 2225/6837 [22:26<18:08,  4.24it/s]
 33%|███▎      | 2226/6837 [22:26<18:11,  4.23it/s]
 33%|███▎      | 2227/6837 [22:26<18:11,  4.22it/s]
 33%|███▎      | 2228/6837 [22:26<18:11,  4.22it/s]
 33%|███▎      | 2229/6837 [22:26<18:12,  4.22it/s]
 33%|███▎      | 2230/6837 [22:27<18:13,  4.21it/s]
 33%|███▎      | 2231/6837 [22:27<18:11,  4.22it/s]
 33%|███▎      | 2232/6837 [22:27<18:09,  4.23it/s]
 33%|███▎      | 2233/6837 [22:27<18:08,  4.23it/s]
 33%|███▎      | 2234/6837 [22:28<18:06,  4.24it/s]
 33%|███▎      | 2235/6837 [22:28<18:06,  4.24it/s]
 33%|███▎      | 2236/6837 [22:28<18:05,  4.24it/s]
 33%|███▎      | 2237/6837 [22:28<18:04,  4.24it/s]
 33%|███▎      | 2238/6837 [22:29<18:02,  4.25it/s]
 33%|███▎      | 2239/6837 [22:29<18:02,  4.25it/s]
 33%|███▎      | 2240/6837 [22:29<18:01,  4.25it/s]
 33%|███▎      | 2241/6837 [22:29<18:02,  4.25it/s]
 33%|███▎      | 2242/6837 [22:30<18:00,  4.25it/s]
 33%|███▎      | 2243/6837 [22:30<18:01,  4.25it/s]
 33%|███▎      | 2244/6837 [22:30<18:02,  4.24it/s]
 33%|███▎      | 2245/6837 [22:30<17:59,  4.25it/s]
 33%|███▎      | 2246/6837 [22:30<18:01,  4.25it/s]
 33%|███▎      | 2247/6837 [22:31<18:01,  4.25it/s]
 33%|███▎      | 2248/6837 [22:31<18:01,  4.24it/s]
 33%|███▎      | 2249/6837 [22:31<18:00,  4.25it/s]
 33%|███▎      | 2250/6837 [22:31<18:01,  4.24it/s]
                                                   {'loss': 4.5901, 'grad_norm': 0.13336360454559326, 'learning_rate': 0.005091052356412338, 'epoch': 0.07}
+

 33%|███▎      | 2250/6837 [22:31<18:01,  4.24it/s]
 33%|███▎      | 2251/6837 [22:32<18:08,  4.21it/s]
 33%|███▎      | 2252/6837 [22:32<18:04,  4.23it/s]
 33%|███▎      | 2253/6837 [22:32<18:04,  4.23it/s]
 33%|███▎      | 2254/6837 [22:32<18:04,  4.23it/s]
 33%|███▎      | 2255/6837 [22:33<18:03,  4.23it/s]
 33%|███▎      | 2256/6837 [22:33<18:01,  4.24it/s]
 33%|███▎      | 2257/6837 [22:33<18:00,  4.24it/s]
 33%|███▎      | 2258/6837 [22:33<17:59,  4.24it/s]
 33%|███▎      | 2259/6837 [22:34<17:59,  4.24it/s]
 33%|███▎      | 2260/6837 [22:34<18:01,  4.23it/s]
 33%|███▎      | 2261/6837 [22:34<18:00,  4.23it/s]
 33%|███▎      | 2262/6837 [22:34<18:02,  4.23it/s]
 33%|███▎      | 2263/6837 [22:35<18:01,  4.23it/s]
 33%|███▎      | 2264/6837 [22:35<18:01,  4.23it/s]
 33%|███▎      | 2265/6837 [22:35<18:01,  4.23it/s]
 33%|███▎      | 2266/6837 [22:35<18:01,  4.23it/s]
 33%|███▎      | 2267/6837 [22:35<18:00,  4.23it/s]
 33%|███▎      | 2268/6837 [22:36<18:00,  4.23it/s]
 33%|███▎      | 2269/6837 [22:36<17:59,  4.23it/s]
 33%|███▎      | 2270/6837 [22:36<18:00,  4.23it/s]
 33%|███▎      | 2271/6837 [22:36<17:59,  4.23it/s]
 33%|███▎      | 2272/6837 [22:37<18:01,  4.22it/s]
 33%|███▎      | 2273/6837 [22:37<18:07,  4.20it/s]
 33%|███▎      | 2274/6837 [22:37<18:10,  4.18it/s]
 33%|███▎      | 2275/6837 [22:37<18:11,  4.18it/s]
                                                   
{'loss': 4.5819, 'grad_norm': 0.14527961611747742, 'learning_rate': 0.005063424294509492, 'epoch': 0.07}
+
 33%|███▎      | 2275/6837 [22:37<18:11,  4.18it/s]
 33%|███▎      | 2276/6837 [22:38<18:10,  4.18it/s]
 33%|███▎      | 2277/6837 [22:38<18:07,  4.19it/s]
 33%|███▎      | 2278/6837 [22:38<18:04,  4.20it/s]
 33%|███▎      | 2279/6837 [22:38<18:02,  4.21it/s]
 33%|███▎      | 2280/6837 [22:39<18:01,  4.22it/s]
 33%|███▎      | 2281/6837 [22:39<17:59,  4.22it/s]
 33%|███▎      | 2282/6837 [22:39<17:58,  4.22it/s]
 33%|███▎      | 2283/6837 [22:39<17:56,  4.23it/s]
 33%|███▎      | 2284/6837 [22:39<17:56,  4.23it/s]
 33%|███▎      | 2285/6837 [22:40<17:56,  4.23it/s]
 33%|███▎      | 2286/6837 [22:40<17:54,  4.23it/s]
 33%|███▎      | 2287/6837 [22:40<17:54,  4.23it/s]
 33%|███▎      | 2288/6837 [22:40<17:57,  4.22it/s]
 33%|███▎      | 2289/6837 [22:41<17:56,  4.23it/s]
 33%|███▎      | 2290/6837 [22:41<17:56,  4.22it/s]
 34%|███▎      | 2291/6837 [22:41<17:57,  4.22it/s]
 34%|███▎      | 2292/6837 [22:41<17:56,  4.22it/s]
 34%|███▎      | 2293/6837 [22:42<17:55,  4.23it/s]
 34%|███▎      | 2294/6837 [22:42<17:55,  4.23it/s]
 34%|███▎      | 2295/6837 [22:42<17:57,  4.21it/s]
 34%|███▎      | 2296/6837 [22:42<17:55,  4.22it/s]
 34%|███▎      | 2297/6837 [22:43<17:54,  4.22it/s]
 34%|███▎      | 2298/6837 [22:43<17:53,  4.23it/s]
 34%|███▎      | 2299/6837 [22:43<18:30,  4.09it/s]
 34%|███▎      | 2300/6837 [22:43<18:18,  4.13it/s]
                                                   
{'loss': 4.5948, 'grad_norm': 0.1316913664340973, 'learning_rate': 0.005035460039677225, 'epoch': 0.07}
+
 34%|███▎      | 2300/6837 [22:43<18:18,  4.13it/s]
 34%|███▎      | 2301/6837 [22:44<18:14,  4.15it/s]
 34%|███▎      | 2302/6837 [22:44<18:06,  4.17it/s]
 34%|███▎      | 2303/6837 [22:44<18:02,  4.19it/s]
 34%|███▎      | 2304/6837 [22:44<17:59,  4.20it/s]
 34%|███▎      | 2305/6837 [22:44<17:55,  4.21it/s]
 34%|███▎      | 2306/6837 [22:45<17:55,  4.21it/s]
 34%|███▎      | 2307/6837 [22:45<17:55,  4.21it/s]
 34%|███▍      | 2308/6837 [22:45<17:55,  4.21it/s]
 34%|███▍      | 2309/6837 [22:45<17:55,  4.21it/s]
 34%|███▍      | 2310/6837 [22:46<17:54,  4.21it/s]
 34%|███▍      | 2311/6837 [22:46<17:52,  4.22it/s]
 34%|███▍      | 2312/6837 [22:46<17:51,  4.22it/s]
 34%|███▍      | 2313/6837 [22:46<17:51,  4.22it/s]
 34%|███▍      | 2314/6837 [22:47<18:16,  4.12it/s]
 34%|███▍      | 2315/6837 [22:47<18:08,  4.15it/s]
 34%|███▍      | 2316/6837 [22:47<18:03,  4.17it/s]
 34%|███▍      | 2317/6837 [22:47<18:02,  4.18it/s]
 34%|███▍      | 2318/6837 [22:48<17:59,  4.19it/s]
 34%|███▍      | 2319/6837 [22:48<17:56,  4.20it/s]
 34%|███▍      | 2320/6837 [22:48<17:55,  4.20it/s]
 34%|███▍      | 2321/6837 [22:48<17:54,  4.20it/s]
 34%|███▍      | 2322/6837 [22:49<17:52,  4.21it/s]
 34%|███▍      | 2323/6837 [22:49<17:51,  4.21it/s]
 34%|███▍      | 2324/6837 [22:49<17:54,  4.20it/s]
 34%|███▍      | 2325/6837 [22:49<17:58,  4.18it/s]
                                                   
{'loss': 4.5796, 'grad_norm': 0.14166171848773956, 'learning_rate': 0.00500716414812086, 'epoch': 0.07}
+
 34%|███▍      | 2325/6837 [22:49<17:58,  4.18it/s]
 34%|███▍      | 2326/6837 [22:49<17:57,  4.18it/s]
 34%|███▍      | 2327/6837 [22:50<17:54,  4.20it/s]
 34%|███▍      | 2328/6837 [22:50<17:51,  4.21it/s]
 34%|███▍      | 2329/6837 [22:50<17:49,  4.22it/s]
 34%|███▍      | 2330/6837 [22:50<17:46,  4.23it/s]
 34%|███▍      | 2331/6837 [22:51<17:47,  4.22it/s]
 34%|███▍      | 2332/6837 [22:51<18:26,  4.07it/s]
 34%|███▍      | 2333/6837 [22:51<18:53,  3.97it/s]
 34%|███▍      | 2334/6837 [22:51<19:13,  3.90it/s]
 34%|███▍      | 2335/6837 [22:52<18:46,  4.00it/s]
 34%|███▍      | 2336/6837 [22:52<18:26,  4.07it/s]
 34%|███▍      | 2337/6837 [22:52<18:11,  4.12it/s]
 34%|███▍      | 2338/6837 [22:52<18:01,  4.16it/s]
 34%|███▍      | 2339/6837 [22:53<17:54,  4.18it/s]
 34%|███▍      | 2340/6837 [22:53<17:51,  4.20it/s]
 34%|███▍      | 2341/6837 [22:53<17:48,  4.21it/s]
 34%|███▍      | 2342/6837 [22:53<17:47,  4.21it/s]
 34%|███▍      | 2343/6837 [22:54<17:45,  4.22it/s]
 34%|███▍      | 2344/6837 [22:54<17:44,  4.22it/s]
 34%|███▍      | 2345/6837 [22:54<17:42,  4.23it/s]
 34%|███▍      | 2346/6837 [22:54<17:42,  4.23it/s]
 34%|███▍      | 2347/6837 [22:55<17:42,  4.23it/s]
 34%|███▍      | 2348/6837 [22:55<17:40,  4.23it/s]
 34%|███▍      | 2349/6837 [22:55<17:41,  4.23it/s]
 34%|███▍      | 2350/6837 [22:55<17:42,  4.22it/s]
                                                   
{'loss': 4.5792, 'grad_norm': 0.12408694624900818, 'learning_rate': 0.004978541230079161, 'epoch': 0.07}
+
 34%|███▍      | 2350/6837 [22:55<17:42,  4.22it/s]
 34%|███▍      | 2351/6837 [22:55<17:46,  4.20it/s]
 34%|███▍      | 2352/6837 [22:56<17:42,  4.22it/s]
 34%|███▍      | 2353/6837 [22:56<17:46,  4.20it/s]
 34%|███▍      | 2354/6837 [22:56<17:42,  4.22it/s]
 34%|███▍      | 2355/6837 [22:56<17:41,  4.22it/s]
 34%|███▍      | 2356/6837 [22:57<17:40,  4.23it/s]
 34%|███▍      | 2357/6837 [22:57<17:41,  4.22it/s]
 34%|███▍      | 2358/6837 [22:57<17:39,  4.23it/s]
 35%|███▍      | 2359/6837 [22:57<17:38,  4.23it/s]
 35%|███▍      | 2360/6837 [22:58<17:39,  4.23it/s]
 35%|███▍      | 2361/6837 [22:58<17:40,  4.22it/s]
 35%|███▍      | 2362/6837 [22:58<17:38,  4.23it/s]
 35%|███▍      | 2363/6837 [22:58<17:36,  4.23it/s]
 35%|███▍      | 2364/6837 [22:59<17:37,  4.23it/s]
 35%|███▍      | 2365/6837 [22:59<17:36,  4.23it/s]
 35%|███▍      | 2366/6837 [22:59<17:35,  4.23it/s]
 35%|███▍      | 2367/6837 [22:59<17:35,  4.24it/s]
 35%|███▍      | 2368/6837 [23:00<17:35,  4.23it/s]
 35%|███▍      | 2369/6837 [23:00<17:35,  4.23it/s]
 35%|███▍      | 2370/6837 [23:00<17:35,  4.23it/s]
 35%|███▍      | 2371/6837 [23:00<17:34,  4.23it/s]
 35%|███▍      | 2372/6837 [23:00<17:35,  4.23it/s]
 35%|███▍      | 2373/6837 [23:01<17:34,  4.23it/s]
 35%|███▍      | 2374/6837 [23:01<17:33,  4.24it/s]
 35%|███▍      | 2375/6837 [23:01<17:34,  4.23it/s]
                                                   
{'loss': 4.5897, 'grad_norm': 0.16600622236728668, 'learning_rate': 0.004949595949073194, 'epoch': 0.07}
+
 35%|███▍      | 2375/6837 [23:01<17:34,  4.23it/s]
 35%|███▍      | 2376/6837 [23:01<18:19,  4.06it/s]
 35%|███▍      | 2377/6837 [23:02<18:35,  4.00it/s]
 35%|███▍      | 2378/6837 [23:02<18:56,  3.92it/s]
 35%|███▍      | 2379/6837 [23:02<19:12,  3.87it/s]
 35%|███▍      | 2380/6837 [23:02<19:17,  3.85it/s]
 35%|███▍      | 2381/6837 [23:03<19:20,  3.84it/s]
 35%|███▍      | 2382/6837 [23:03<19:21,  3.83it/s]
 35%|███▍      | 2383/6837 [23:03<18:48,  3.95it/s]
 35%|███▍      | 2384/6837 [23:03<18:27,  4.02it/s]
 35%|███▍      | 2385/6837 [23:04<18:08,  4.09it/s]
 35%|███▍      | 2386/6837 [23:04<17:55,  4.14it/s]
 35%|███▍      | 2387/6837 [23:04<17:49,  4.16it/s]
 35%|███▍      | 2388/6837 [23:04<17:42,  4.19it/s]
 35%|███▍      | 2389/6837 [23:05<17:38,  4.20it/s]
 35%|███▍      | 2390/6837 [23:05<17:36,  4.21it/s]
 35%|███▍      | 2391/6837 [23:05<17:33,  4.22it/s]
 35%|███▍      | 2392/6837 [23:05<17:31,  4.23it/s]
 35%|███▌      | 2393/6837 [23:06<17:31,  4.22it/s]
 35%|███▌      | 2394/6837 [23:06<17:29,  4.23it/s]
 35%|███▌      | 2395/6837 [23:06<18:06,  4.09it/s]
 35%|███▌      | 2396/6837 [23:06<18:26,  4.01it/s]
 35%|███▌      | 2397/6837 [23:07<18:48,  3.93it/s]
 35%|███▌      | 2398/6837 [23:07<18:58,  3.90it/s]
 35%|███▌      | 2399/6837 [23:07<19:06,  3.87it/s]
 35%|███▌      | 2400/6837 [23:07<19:13,  3.85it/s]
                                                   
{'loss': 4.5745, 'grad_norm': 0.16214387118816376, 'learning_rate': 0.004920333021146501, 'epoch': 0.07}
+
 35%|███▌      | 2400/6837 [23:07<19:13,  3.85it/s]
 35%|███▌      | 2401/6837 [23:08<19:15,  3.84it/s]
 35%|███▌      | 2402/6837 [23:08<19:19,  3.82it/s]
 35%|███▌      | 2403/6837 [23:08<19:20,  3.82it/s]
 35%|███▌      | 2404/6837 [23:08<19:22,  3.81it/s]
 35%|███▌      | 2405/6837 [23:09<19:16,  3.83it/s]
 35%|███▌      | 2406/6837 [23:09<19:21,  3.82it/s]
 35%|███▌      | 2407/6837 [23:09<19:23,  3.81it/s]
 35%|███▌      | 2408/6837 [23:10<19:25,  3.80it/s]
 35%|███▌      | 2409/6837 [23:10<19:21,  3.81it/s]
 35%|███▌      | 2410/6837 [23:10<19:20,  3.82it/s]
 35%|███▌      | 2411/6837 [23:10<19:19,  3.82it/s]
 35%|███▌      | 2412/6837 [23:11<19:23,  3.80it/s]
 35%|███▌      | 2413/6837 [23:11<19:28,  3.79it/s]
 35%|███▌      | 2414/6837 [23:11<19:27,  3.79it/s]
 35%|███▌      | 2415/6837 [23:11<19:28,  3.78it/s]
 35%|███▌      | 2416/6837 [23:12<19:37,  3.75it/s]
 35%|███▌      | 2417/6837 [23:12<19:34,  3.76it/s]
 35%|███▌      | 2418/6837 [23:12<19:26,  3.79it/s]
 35%|███▌      | 2419/6837 [23:12<19:24,  3.79it/s]
 35%|███▌      | 2420/6837 [23:13<19:28,  3.78it/s]
 35%|███▌      | 2421/6837 [23:13<19:28,  3.78it/s]
 35%|███▌      | 2422/6837 [23:13<19:22,  3.80it/s]
 35%|███▌      | 2423/6837 [23:13<19:25,  3.79it/s]
 35%|███▌      | 2424/6837 [23:14<19:27,  3.78it/s]
 35%|███▌      | 2425/6837 [23:14<19:29,  3.77it/s]
                                                   
{'loss': 4.5609, 'grad_norm': 0.1423948109149933, 'learning_rate': 0.00489075721409671, 'epoch': 0.07}
+
 35%|███▌      | 2425/6837 [23:14<19:29,  3.77it/s]
 35%|███▌      | 2426/6837 [23:14<19:43,  3.73it/s]
 35%|███▌      | 2427/6837 [23:15<19:43,  3.73it/s]
 36%|███▌      | 2428/6837 [23:15<19:42,  3.73it/s]
 36%|███▌      | 2429/6837 [23:15<19:38,  3.74it/s]
 36%|███▌      | 2430/6837 [23:15<19:36,  3.75it/s]
 36%|███▌      | 2431/6837 [23:16<19:09,  3.83it/s]
 36%|███▌      | 2432/6837 [23:16<18:36,  3.95it/s]
 36%|███▌      | 2433/6837 [23:16<18:13,  4.03it/s]
 36%|███▌      | 2434/6837 [23:16<17:56,  4.09it/s]
 36%|███▌      | 2435/6837 [23:17<17:44,  4.14it/s]
 36%|███▌      | 2436/6837 [23:17<17:36,  4.17it/s]
 36%|███▌      | 2437/6837 [23:17<17:31,  4.18it/s]
 36%|███▌      | 2438/6837 [23:17<17:25,  4.21it/s]
 36%|███▌      | 2439/6837 [23:18<17:23,  4.21it/s]
 36%|███▌      | 2440/6837 [23:18<17:23,  4.22it/s]
 36%|███▌      | 2441/6837 [23:18<17:19,  4.23it/s]
 36%|███▌      | 2442/6837 [23:18<17:17,  4.23it/s]
 36%|███▌      | 2443/6837 [23:18<17:17,  4.23it/s]
 36%|███▌      | 2444/6837 [23:19<17:17,  4.24it/s]
 36%|███▌      | 2445/6837 [23:19<17:17,  4.23it/s]
 36%|███▌      | 2446/6837 [23:19<17:16,  4.24it/s]
 36%|███▌      | 2447/6837 [23:19<17:16,  4.23it/s]
 36%|███▌      | 2448/6837 [23:20<17:15,  4.24it/s]
 36%|███▌      | 2449/6837 [23:20<17:16,  4.23it/s]
 36%|███▌      | 2450/6837 [23:20<17:16,  4.23it/s]
                                                   
{'loss': 4.571, 'grad_norm': 0.1584578901529312, 'learning_rate': 0.004860873346698729, 'epoch': 0.07}
+
 36%|███▌      | 2450/6837 [23:20<17:16,  4.23it/s]
 36%|███▌      | 2451/6837 [23:20<17:20,  4.22it/s]
 36%|███▌      | 2452/6837 [23:21<17:17,  4.23it/s]
 36%|███▌      | 2453/6837 [23:21<17:15,  4.23it/s]
 36%|███▌      | 2454/6837 [23:21<17:16,  4.23it/s]
 36%|███▌      | 2455/6837 [23:21<17:14,  4.23it/s]
 36%|███▌      | 2456/6837 [23:22<17:14,  4.23it/s]
 36%|███▌      | 2457/6837 [23:22<17:15,  4.23it/s]
 36%|███▌      | 2458/6837 [23:22<17:15,  4.23it/s]
 36%|███▌      | 2459/6837 [23:22<17:15,  4.23it/s]
 36%|███▌      | 2460/6837 [23:22<17:13,  4.24it/s]
 36%|███▌      | 2461/6837 [23:23<17:12,  4.24it/s]
 36%|███▌      | 2462/6837 [23:23<17:13,  4.23it/s]
 36%|███▌      | 2463/6837 [23:23<17:13,  4.23it/s]
 36%|███▌      | 2464/6837 [23:23<17:13,  4.23it/s]
 36%|███▌      | 2465/6837 [23:24<17:16,  4.22it/s]
 36%|███▌      | 2466/6837 [23:24<17:14,  4.22it/s]
 36%|███▌      | 2467/6837 [23:24<17:14,  4.22it/s]
 36%|███▌      | 2468/6837 [23:24<17:13,  4.23it/s]
 36%|███▌      | 2469/6837 [23:25<17:48,  4.09it/s]
 36%|███▌      | 2470/6837 [23:25<18:12,  4.00it/s]
 36%|███▌      | 2471/6837 [23:25<18:31,  3.93it/s]
 36%|███▌      | 2472/6837 [23:25<18:40,  3.90it/s]
 36%|███▌      | 2473/6837 [23:26<18:46,  3.87it/s]
 36%|███▌      | 2474/6837 [23:26<18:54,  3.85it/s]
 36%|███▌      | 2475/6837 [23:26<19:02,  3.82it/s]
                                                   
{'loss': 4.5662, 'grad_norm': 0.1679723709821701, 'learning_rate': 0.004830686287919619, 'epoch': 0.07}
+
 36%|███▌      | 2475/6837 [23:26<19:02,  3.82it/s]
 36%|███▌      | 2476/6837 [23:26<19:09,  3.79it/s]
 36%|███▌      | 2477/6837 [23:27<19:12,  3.78it/s]
 36%|███▌      | 2478/6837 [23:27<19:19,  3.76it/s]
 36%|███▋      | 2479/6837 [23:27<19:24,  3.74it/s]
 36%|███▋      | 2480/6837 [23:28<19:30,  3.72it/s]
 36%|███▋      | 2481/6837 [23:28<19:37,  3.70it/s]
 36%|███▋      | 2482/6837 [23:28<19:31,  3.72it/s]
 36%|███▋      | 2483/6837 [23:28<19:27,  3.73it/s]
 36%|███▋      | 2484/6837 [23:29<19:30,  3.72it/s]
 36%|███▋      | 2485/6837 [23:29<19:28,  3.72it/s]
 36%|███▋      | 2486/6837 [23:29<19:24,  3.74it/s]
 36%|███▋      | 2487/6837 [23:29<19:21,  3.74it/s]
 36%|███▋      | 2488/6837 [23:30<18:50,  3.85it/s]
 36%|███▋      | 2489/6837 [23:30<18:18,  3.96it/s]
 36%|███▋      | 2490/6837 [23:30<17:56,  4.04it/s]
 36%|███▋      | 2491/6837 [23:30<17:40,  4.10it/s]
 36%|███▋      | 2492/6837 [23:31<18:06,  4.00it/s]
 36%|███▋      | 2493/6837 [23:31<18:27,  3.92it/s]
 36%|███▋      | 2494/6837 [23:31<18:40,  3.87it/s]
 36%|███▋      | 2495/6837 [23:31<18:52,  3.83it/s]
 37%|███▋      | 2496/6837 [23:32<18:55,  3.82it/s]
 37%|███▋      | 2497/6837 [23:32<19:03,  3.79it/s]
 37%|███▋      | 2498/6837 [23:32<19:07,  3.78it/s]
 37%|███▋      | 2499/6837 [23:33<19:12,  3.77it/s]
 37%|███▋      | 2500/6837 [23:33<19:07,  3.78it/s]{'loss': 4.5668, 'grad_norm': 0.14782841503620148, 'learning_rate': 0.004800200956125296, 'epoch': 0.07}
+
                                                   

 37%|███▋      | 2500/6837 [23:33<19:07,  3.78it/s]
 37%|███▋      | 2501/6837 [23:33<19:15,  3.75it/s]
 37%|███▋      | 2502/6837 [23:33<19:14,  3.76it/s]
 37%|███▋      | 2503/6837 [23:34<19:13,  3.76it/s]
 37%|███▋      | 2504/6837 [23:34<19:14,  3.75it/s]
 37%|███▋      | 2505/6837 [23:34<19:11,  3.76it/s]
 37%|███▋      | 2506/6837 [23:34<19:11,  3.76it/s]
 37%|███▋      | 2507/6837 [23:35<19:11,  3.76it/s]
 37%|███▋      | 2508/6837 [23:35<19:14,  3.75it/s]
 37%|███▋      | 2509/6837 [23:35<19:13,  3.75it/s]
 37%|███▋      | 2510/6837 [23:35<19:11,  3.76it/s]
 37%|███▋      | 2511/6837 [23:36<19:11,  3.76it/s]
 37%|███▋      | 2512/6837 [23:36<19:10,  3.76it/s]
 37%|███▋      | 2513/6837 [23:36<19:10,  3.76it/s]
 37%|███▋      | 2514/6837 [23:36<19:16,  3.74it/s]
 37%|███▋      | 2515/6837 [23:37<19:10,  3.76it/s]
 37%|███▋      | 2516/6837 [23:37<19:09,  3.76it/s]
 37%|███▋      | 2517/6837 [23:37<19:09,  3.76it/s]
 37%|███▋      | 2518/6837 [23:38<19:10,  3.75it/s]
 37%|███▋      | 2519/6837 [23:38<19:03,  3.78it/s]
 37%|███▋      | 2520/6837 [23:38<19:08,  3.76it/s]
 37%|███▋      | 2521/6837 [23:38<19:09,  3.75it/s]
 37%|███▋      | 2522/6837 [23:39<19:09,  3.75it/s]
 37%|███▋      | 2523/6837 [23:39<19:12,  3.74it/s]
 37%|███▋      | 2524/6837 [23:39<19:11,  3.74it/s]
 37%|███▋      | 2525/6837 [23:39<19:04,  3.77it/s]
                                                   
{'loss': 4.5425, 'grad_norm': 0.14905227720737457, 'learning_rate': 0.0047694223182791875, 'epoch': 0.07}
+
 37%|███▋      | 2525/6837 [23:39<19:04,  3.77it/s]
 37%|███▋      | 2526/6837 [23:40<19:07,  3.76it/s]
 37%|███▋      | 2527/6837 [23:40<19:02,  3.77it/s]
 37%|███▋      | 2528/6837 [23:40<19:06,  3.76it/s]
 37%|███▋      | 2529/6837 [23:40<19:06,  3.76it/s]
 37%|███▋      | 2530/6837 [23:41<19:09,  3.75it/s]
 37%|███▋      | 2531/6837 [23:41<19:08,  3.75it/s]
 37%|███▋      | 2532/6837 [23:41<19:09,  3.75it/s]
 37%|███▋      | 2533/6837 [23:42<19:10,  3.74it/s]
 37%|███▋      | 2534/6837 [23:42<19:09,  3.74it/s]
 37%|███▋      | 2535/6837 [23:42<19:05,  3.76it/s]
 37%|███▋      | 2536/6837 [23:42<19:07,  3.75it/s]
 37%|███▋      | 2537/6837 [23:43<19:11,  3.74it/s]
 37%|███▋      | 2538/6837 [23:43<19:12,  3.73it/s]
 37%|███▋      | 2539/6837 [23:43<19:13,  3.73it/s]
 37%|███▋      | 2540/6837 [23:43<19:12,  3.73it/s]
 37%|███▋      | 2541/6837 [23:44<19:03,  3.76it/s]
 37%|███▋      | 2542/6837 [23:44<19:05,  3.75it/s]
 37%|███▋      | 2543/6837 [23:44<19:04,  3.75it/s]
 37%|███▋      | 2544/6837 [23:44<19:09,  3.73it/s]
 37%|███▋      | 2545/6837 [23:45<19:08,  3.74it/s]
 37%|███▋      | 2546/6837 [23:45<18:54,  3.78it/s]
 37%|███▋      | 2547/6837 [23:45<18:57,  3.77it/s]
 37%|███▋      | 2548/6837 [23:46<18:55,  3.78it/s]
 37%|███▋      | 2549/6837 [23:46<18:54,  3.78it/s]
 37%|███▋      | 2550/6837 [23:46<19:01,  3.76it/s]
                                                   
{'loss': 4.5424, 'grad_norm': 0.143193319439888, 'learning_rate': 0.004738355389132956, 'epoch': 0.07}
+
 37%|███▋      | 2550/6837 [23:46<19:01,  3.76it/s]
 37%|███▋      | 2551/6837 [23:46<19:08,  3.73it/s]
 37%|███▋      | 2552/6837 [23:47<19:00,  3.76it/s]
 37%|███▋      | 2553/6837 [23:47<18:54,  3.78it/s]
 37%|███▋      | 2554/6837 [23:47<18:53,  3.78it/s]
 37%|███▋      | 2555/6837 [23:47<18:42,  3.82it/s]
 37%|███▋      | 2556/6837 [23:48<18:09,  3.93it/s]
 37%|███▋      | 2557/6837 [23:48<17:46,  4.01it/s]
 37%|███▋      | 2558/6837 [23:48<17:29,  4.08it/s]
 37%|███▋      | 2559/6837 [23:48<17:18,  4.12it/s]
 37%|███▋      | 2560/6837 [23:49<17:11,  4.15it/s]
 37%|███▋      | 2561/6837 [23:49<17:06,  4.17it/s]
 37%|███▋      | 2562/6837 [23:49<17:03,  4.18it/s]
 37%|███▋      | 2563/6837 [23:49<16:59,  4.19it/s]
 38%|███▊      | 2564/6837 [23:50<17:00,  4.19it/s]
 38%|███▊      | 2565/6837 [23:50<16:58,  4.19it/s]
 38%|███▊      | 2566/6837 [23:50<16:55,  4.20it/s]
 38%|███▊      | 2567/6837 [23:50<16:54,  4.21it/s]
 38%|███▊      | 2568/6837 [23:50<16:53,  4.21it/s]
 38%|███▊      | 2569/6837 [23:51<16:52,  4.22it/s]
 38%|███▊      | 2570/6837 [23:51<16:52,  4.22it/s]
 38%|███▊      | 2571/6837 [23:51<16:50,  4.22it/s]
 38%|███▊      | 2572/6837 [23:51<16:47,  4.23it/s]
 38%|███▊      | 2573/6837 [23:52<16:46,  4.24it/s]
 38%|███▊      | 2574/6837 [23:52<16:46,  4.24it/s]
 38%|███▊      | 2575/6837 [23:52<16:47,  4.23it/s]
                                                   
{'loss': 4.5479, 'grad_norm': 0.15022441744804382, 'learning_rate': 0.004707005230409463, 'epoch': 0.08}
+
 38%|███▊      | 2575/6837 [23:52<16:47,  4.23it/s]
 38%|███▊      | 2576/6837 [23:52<16:50,  4.22it/s]
 38%|███▊      | 2577/6837 [23:53<16:47,  4.23it/s]
 38%|███▊      | 2578/6837 [23:53<16:47,  4.23it/s]
 38%|███▊      | 2579/6837 [23:53<16:46,  4.23it/s]
 38%|███▊      | 2580/6837 [23:53<16:44,  4.24it/s]
 38%|███▊      | 2581/6837 [23:54<16:45,  4.23it/s]
 38%|███▊      | 2582/6837 [23:54<16:53,  4.20it/s]
 38%|███▊      | 2583/6837 [23:54<17:29,  4.05it/s]
 38%|███▊      | 2584/6837 [23:54<17:53,  3.96it/s]
 38%|███▊      | 2585/6837 [23:55<18:11,  3.90it/s]
 38%|███▊      | 2586/6837 [23:55<18:30,  3.83it/s]
 38%|███▊      | 2587/6837 [23:55<18:36,  3.81it/s]
 38%|███▊      | 2588/6837 [23:55<18:35,  3.81it/s]
 38%|███▊      | 2589/6837 [23:56<18:39,  3.79it/s]
 38%|███▊      | 2590/6837 [23:56<18:39,  3.79it/s]
 38%|███▊      | 2591/6837 [23:56<18:50,  3.76it/s]
 38%|███▊      | 2592/6837 [23:56<18:55,  3.74it/s]
 38%|███▊      | 2593/6837 [23:57<18:57,  3.73it/s]
 38%|███▊      | 2594/6837 [23:57<18:54,  3.74it/s]
 38%|███▊      | 2595/6837 [23:57<18:52,  3.75it/s]
 38%|███▊      | 2596/6837 [23:58<18:51,  3.75it/s]
 38%|███▊      | 2597/6837 [23:58<18:45,  3.77it/s]
 38%|███▊      | 2598/6837 [23:58<18:21,  3.85it/s]
 38%|███▊      | 2599/6837 [23:58<17:50,  3.96it/s]
 38%|███▊      | 2600/6837 [23:59<17:27,  4.04it/s]
                                                   
{'loss': 4.5438, 'grad_norm': 0.14342527091503143, 'learning_rate': 0.004675376949978053, 'epoch': 0.08}
+
 38%|███▊      | 2600/6837 [23:59<17:27,  4.04it/s]
 38%|███▊      | 2601/6837 [23:59<17:23,  4.06it/s]
 38%|███▊      | 2602/6837 [23:59<17:09,  4.11it/s]
 38%|███▊      | 2603/6837 [23:59<17:07,  4.12it/s]
 38%|███▊      | 2604/6837 [23:59<16:59,  4.15it/s]
 38%|███▊      | 2605/6837 [24:00<16:53,  4.17it/s]
 38%|███▊      | 2606/6837 [24:00<16:51,  4.18it/s]
 38%|███▊      | 2607/6837 [24:00<16:47,  4.20it/s]
 38%|███▊      | 2608/6837 [24:00<16:44,  4.21it/s]
 38%|███▊      | 2609/6837 [24:01<16:41,  4.22it/s]
 38%|███▊      | 2610/6837 [24:01<16:40,  4.22it/s]
 38%|███▊      | 2611/6837 [24:01<16:38,  4.23it/s]
 38%|███▊      | 2612/6837 [24:01<16:37,  4.24it/s]
 38%|███▊      | 2613/6837 [24:02<16:37,  4.23it/s]
 38%|███▊      | 2614/6837 [24:02<16:37,  4.24it/s]
 38%|███▊      | 2615/6837 [24:02<16:37,  4.23it/s]
 38%|███▊      | 2616/6837 [24:02<16:37,  4.23it/s]
 38%|███▊      | 2617/6837 [24:03<16:36,  4.23it/s]
 38%|███▊      | 2618/6837 [24:03<16:35,  4.24it/s]
 38%|███▊      | 2619/6837 [24:03<16:36,  4.23it/s]
 38%|███▊      | 2620/6837 [24:03<16:36,  4.23it/s]
 38%|███▊      | 2621/6837 [24:03<16:36,  4.23it/s]
 38%|███▊      | 2622/6837 [24:04<16:35,  4.24it/s]
 38%|███▊      | 2623/6837 [24:04<16:35,  4.23it/s]
 38%|███▊      | 2624/6837 [24:04<16:37,  4.22it/s]
 38%|███▊      | 2625/6837 [24:04<16:36,  4.23it/s]
                                                   
{'loss': 4.5556, 'grad_norm': 0.18744206428527832, 'learning_rate': 0.004643475701022332, 'epoch': 0.08}
+
 38%|███▊      | 2625/6837 [24:04<16:36,  4.23it/s]
 38%|███▊      | 2626/6837 [24:05<16:37,  4.22it/s]
 38%|███▊      | 2627/6837 [24:05<16:35,  4.23it/s]
 38%|███▊      | 2628/6837 [24:05<16:36,  4.22it/s]
 38%|███▊      | 2629/6837 [24:05<17:19,  4.05it/s]
 38%|███▊      | 2630/6837 [24:06<17:37,  3.98it/s]
 38%|███▊      | 2631/6837 [24:06<17:55,  3.91it/s]
 38%|███▊      | 2632/6837 [24:06<18:03,  3.88it/s]
 39%|███▊      | 2633/6837 [24:06<18:11,  3.85it/s]
 39%|███▊      | 2634/6837 [24:07<18:16,  3.83it/s]
 39%|███▊      | 2635/6837 [24:07<18:17,  3.83it/s]
 39%|███▊      | 2636/6837 [24:07<18:18,  3.83it/s]
 39%|███▊      | 2637/6837 [24:08<18:21,  3.81it/s]
 39%|███▊      | 2638/6837 [24:08<18:23,  3.81it/s]
 39%|███▊      | 2639/6837 [24:08<18:24,  3.80it/s]
 39%|███▊      | 2640/6837 [24:08<18:25,  3.80it/s]
 39%|███▊      | 2641/6837 [24:09<18:20,  3.81it/s]
 39%|███▊      | 2642/6837 [24:09<18:24,  3.80it/s]
 39%|███▊      | 2643/6837 [24:09<18:28,  3.79it/s]
 39%|███▊      | 2644/6837 [24:09<18:28,  3.78it/s]
 39%|███▊      | 2645/6837 [24:10<18:26,  3.79it/s]
 39%|███▊      | 2646/6837 [24:10<18:23,  3.80it/s]
 39%|███▊      | 2647/6837 [24:10<18:23,  3.80it/s]
 39%|███▊      | 2648/6837 [24:10<18:27,  3.78it/s]
 39%|███▊      | 2649/6837 [24:11<18:29,  3.78it/s]
 39%|███▉      | 2650/6837 [24:11<18:25,  3.79it/s]
                                                   
{'loss': 4.5428, 'grad_norm': 0.13695579767227173, 'learning_rate': 0.004611306681200567, 'epoch': 0.08}
+
 39%|███▉      | 2650/6837 [24:11<18:25,  3.79it/s]
 39%|███▉      | 2651/6837 [24:11<18:25,  3.79it/s]
 39%|███▉      | 2652/6837 [24:11<18:24,  3.79it/s]
 39%|███▉      | 2653/6837 [24:12<18:25,  3.79it/s]
 39%|███▉      | 2654/6837 [24:12<18:26,  3.78it/s]
 39%|███▉      | 2655/6837 [24:12<18:23,  3.79it/s]
 39%|███▉      | 2656/6837 [24:13<18:23,  3.79it/s]
 39%|███▉      | 2657/6837 [24:13<18:20,  3.80it/s]
 39%|███▉      | 2658/6837 [24:13<18:18,  3.80it/s]
 39%|███▉      | 2659/6837 [24:13<18:14,  3.82it/s]
 39%|███▉      | 2660/6837 [24:14<18:17,  3.81it/s]
 39%|███▉      | 2661/6837 [24:14<18:20,  3.79it/s]
 39%|███▉      | 2662/6837 [24:14<18:23,  3.78it/s]
 39%|███▉      | 2663/6837 [24:14<18:25,  3.77it/s]
 39%|███▉      | 2664/6837 [24:15<18:19,  3.80it/s]
 39%|███▉      | 2665/6837 [24:15<18:16,  3.80it/s]
 39%|███▉      | 2666/6837 [24:15<18:15,  3.81it/s]
 39%|███▉      | 2667/6837 [24:15<18:11,  3.82it/s]
 39%|███▉      | 2668/6837 [24:16<18:17,  3.80it/s]
 39%|███▉      | 2669/6837 [24:16<18:19,  3.79it/s]
 39%|███▉      | 2670/6837 [24:16<18:23,  3.78it/s]
 39%|███▉      | 2671/6837 [24:17<18:33,  3.74it/s]
 39%|███▉      | 2672/6837 [24:17<18:33,  3.74it/s]
 39%|███▉      | 2673/6837 [24:17<18:29,  3.75it/s]
 39%|███▉      | 2674/6837 [24:17<18:23,  3.77it/s]
 39%|███▉      | 2675/6837 [24:18<18:21,  3.78it/s]
                                                   
{'loss': 4.5324, 'grad_norm': 0.1539468914270401, 'learning_rate': 0.004578875131798827, 'epoch': 0.08}
+
 39%|███▉      | 2675/6837 [24:18<18:21,  3.78it/s]
 39%|███▉      | 2676/6837 [24:18<18:28,  3.75it/s]
 39%|███▉      | 2677/6837 [24:18<18:27,  3.76it/s]
 39%|███▉      | 2678/6837 [24:18<18:16,  3.79it/s]
 39%|███▉      | 2679/6837 [24:19<18:14,  3.80it/s]
 39%|███▉      | 2680/6837 [24:19<18:15,  3.79it/s]
 39%|███▉      | 2681/6837 [24:19<17:48,  3.89it/s]
 39%|███▉      | 2682/6837 [24:19<17:21,  3.99it/s]
 39%|███▉      | 2683/6837 [24:20<17:03,  4.06it/s]
 39%|███▉      | 2684/6837 [24:20<16:50,  4.11it/s]
 39%|███▉      | 2685/6837 [24:20<16:41,  4.15it/s]
 39%|███▉      | 2686/6837 [24:20<16:34,  4.18it/s]
 39%|███▉      | 2687/6837 [24:21<16:30,  4.19it/s]
 39%|███▉      | 2688/6837 [24:21<16:26,  4.20it/s]
 39%|███▉      | 2689/6837 [24:21<16:24,  4.21it/s]
 39%|███▉      | 2690/6837 [24:21<16:22,  4.22it/s]
 39%|███▉      | 2691/6837 [24:21<16:20,  4.23it/s]
 39%|███▉      | 2692/6837 [24:22<16:19,  4.23it/s]
 39%|███▉      | 2693/6837 [24:22<16:17,  4.24it/s]
 39%|███▉      | 2694/6837 [24:22<16:16,  4.24it/s]
 39%|███▉      | 2695/6837 [24:22<16:16,  4.24it/s]
 39%|███▉      | 2696/6837 [24:23<16:18,  4.23it/s]
 39%|███▉      | 2697/6837 [24:23<16:17,  4.23it/s]
 39%|███▉      | 2698/6837 [24:23<16:19,  4.23it/s]
 39%|███▉      | 2699/6837 [24:23<16:18,  4.23it/s]
 39%|███▉      | 2700/6837 [24:24<16:16,  4.24it/s]
                                                   
{'loss': 4.5364, 'grad_norm': 0.13192936778068542, 'learning_rate': 0.004546186336877026, 'epoch': 0.08}
+
 39%|███▉      | 2700/6837 [24:24<16:16,  4.24it/s]
 40%|███▉      | 2701/6837 [24:24<16:17,  4.23it/s]
 40%|███▉      | 2702/6837 [24:24<16:16,  4.23it/s]
 40%|███▉      | 2703/6837 [24:24<16:15,  4.24it/s]
 40%|███▉      | 2704/6837 [24:25<16:14,  4.24it/s]
 40%|███▉      | 2705/6837 [24:25<16:13,  4.24it/s]
 40%|███▉      | 2706/6837 [24:25<16:14,  4.24it/s]
 40%|███▉      | 2707/6837 [24:25<16:13,  4.24it/s]
 40%|███▉      | 2708/6837 [24:25<16:14,  4.24it/s]
 40%|███▉      | 2709/6837 [24:26<16:16,  4.23it/s]
 40%|███▉      | 2710/6837 [24:26<16:14,  4.23it/s]
 40%|███▉      | 2711/6837 [24:26<16:13,  4.24it/s]
 40%|███▉      | 2712/6837 [24:26<16:13,  4.24it/s]
 40%|███▉      | 2713/6837 [24:27<16:13,  4.24it/s]
 40%|███▉      | 2714/6837 [24:27<16:13,  4.24it/s]
 40%|███▉      | 2715/6837 [24:27<16:12,  4.24it/s]
 40%|███▉      | 2716/6837 [24:27<16:13,  4.24it/s]
 40%|███▉      | 2717/6837 [24:28<16:12,  4.24it/s]
 40%|███▉      | 2718/6837 [24:28<16:12,  4.23it/s]
 40%|███▉      | 2719/6837 [24:28<16:13,  4.23it/s]
 40%|███▉      | 2720/6837 [24:28<16:12,  4.23it/s]
 40%|███▉      | 2721/6837 [24:29<16:11,  4.24it/s]
 40%|███▉      | 2722/6837 [24:29<16:10,  4.24it/s]
 40%|███▉      | 2723/6837 [24:29<16:11,  4.23it/s]
 40%|███▉      | 2724/6837 [24:29<16:12,  4.23it/s]
 40%|███▉      | 2725/6837 [24:30<16:12,  4.23it/s]
                                                   
{'loss': 4.524, 'grad_norm': 0.13700827956199646, 'learning_rate': 0.00451324562240799, 'epoch': 0.08}
+
 40%|███▉      | 2725/6837 [24:30<16:12,  4.23it/s]
 40%|███▉      | 2726/6837 [24:30<16:15,  4.21it/s]
 40%|███▉      | 2727/6837 [24:30<16:15,  4.21it/s]
 40%|███▉      | 2728/6837 [24:30<16:13,  4.22it/s]
 40%|███▉      | 2729/6837 [24:30<16:11,  4.23it/s]
 40%|███▉      | 2730/6837 [24:31<16:11,  4.23it/s]
 40%|███▉      | 2731/6837 [24:31<16:11,  4.23it/s]
 40%|███▉      | 2732/6837 [24:31<16:09,  4.23it/s]
 40%|███▉      | 2733/6837 [24:31<16:08,  4.24it/s]
 40%|███▉      | 2734/6837 [24:32<16:08,  4.23it/s]
 40%|████      | 2735/6837 [24:32<16:08,  4.24it/s]
 40%|████      | 2736/6837 [24:32<16:07,  4.24it/s]
 40%|████      | 2737/6837 [24:32<16:07,  4.24it/s]
 40%|████      | 2738/6837 [24:33<16:08,  4.23it/s]
 40%|████      | 2739/6837 [24:33<16:08,  4.23it/s]
 40%|████      | 2740/6837 [24:33<16:07,  4.24it/s]
 40%|████      | 2741/6837 [24:33<16:05,  4.24it/s]
 40%|████      | 2742/6837 [24:34<16:07,  4.23it/s]
 40%|████      | 2743/6837 [24:34<16:06,  4.24it/s]
 40%|████      | 2744/6837 [24:34<16:06,  4.23it/s]
 40%|████      | 2745/6837 [24:34<16:06,  4.24it/s]
 40%|████      | 2746/6837 [24:34<16:05,  4.24it/s]
 40%|████      | 2747/6837 [24:35<16:05,  4.24it/s]
 40%|████      | 2748/6837 [24:35<16:04,  4.24it/s]
 40%|████      | 2749/6837 [24:35<16:08,  4.22it/s]
 40%|████      | 2750/6837 [24:35<16:07,  4.22it/s]
                                                   
{'loss': 4.5337, 'grad_norm': 0.14694513380527496, 'learning_rate': 0.004480058355409703, 'epoch': 0.08}
+
 40%|████      | 2750/6837 [24:35<16:07,  4.22it/s]
 40%|████      | 2751/6837 [24:36<16:08,  4.22it/s]
 40%|████      | 2752/6837 [24:36<16:06,  4.22it/s]
 40%|████      | 2753/6837 [24:36<16:07,  4.22it/s]
 40%|████      | 2754/6837 [24:36<16:05,  4.23it/s]
 40%|████      | 2755/6837 [24:37<16:06,  4.23it/s]
 40%|████      | 2756/6837 [24:37<16:05,  4.23it/s]
 40%|████      | 2757/6837 [24:37<16:05,  4.23it/s]
 40%|████      | 2758/6837 [24:37<16:03,  4.23it/s]
 40%|████      | 2759/6837 [24:38<16:03,  4.23it/s]
 40%|████      | 2760/6837 [24:38<16:03,  4.23it/s]
 40%|████      | 2761/6837 [24:38<16:03,  4.23it/s]
 40%|████      | 2762/6837 [24:38<16:04,  4.23it/s]
 40%|████      | 2763/6837 [24:38<16:02,  4.23it/s]
 40%|████      | 2764/6837 [24:39<16:01,  4.24it/s]
 40%|████      | 2765/6837 [24:39<16:01,  4.24it/s]
 40%|████      | 2766/6837 [24:39<16:01,  4.24it/s]
 40%|████      | 2767/6837 [24:39<16:01,  4.23it/s]
 40%|████      | 2768/6837 [24:40<16:01,  4.23it/s]
 41%|████      | 2769/6837 [24:40<16:02,  4.23it/s]
 41%|████      | 2770/6837 [24:40<16:00,  4.23it/s]
 41%|████      | 2771/6837 [24:40<16:04,  4.22it/s]
 41%|████      | 2772/6837 [24:41<16:02,  4.22it/s]
 41%|████      | 2773/6837 [24:41<16:03,  4.22it/s]
 41%|████      | 2774/6837 [24:41<16:03,  4.22it/s]
 41%|████      | 2775/6837 [24:41<16:02,  4.22it/s]
                                                   
{'loss': 4.5223, 'grad_norm': 0.13326427340507507, 'learning_rate': 0.0044466299430708555, 'epoch': 0.08}
+
 41%|████      | 2775/6837 [24:41<16:02,  4.22it/s]
 41%|████      | 2776/6837 [24:42<16:03,  4.21it/s]
 41%|████      | 2777/6837 [24:42<16:02,  4.22it/s]
 41%|████      | 2778/6837 [24:42<16:00,  4.23it/s]
 41%|████      | 2779/6837 [24:42<15:59,  4.23it/s]
 41%|████      | 2780/6837 [24:43<15:59,  4.23it/s]
 41%|████      | 2781/6837 [24:43<16:31,  4.09it/s]
 41%|████      | 2782/6837 [24:43<16:49,  4.02it/s]
 41%|████      | 2783/6837 [24:43<16:34,  4.08it/s]
 41%|████      | 2784/6837 [24:44<16:22,  4.13it/s]
 41%|████      | 2785/6837 [24:44<16:13,  4.16it/s]
 41%|████      | 2786/6837 [24:44<16:06,  4.19it/s]
 41%|████      | 2787/6837 [24:44<16:01,  4.21it/s]
 41%|████      | 2788/6837 [24:44<15:59,  4.22it/s]
 41%|████      | 2789/6837 [24:45<15:58,  4.22it/s]
 41%|████      | 2790/6837 [24:45<15:57,  4.23it/s]
 41%|████      | 2791/6837 [24:45<15:55,  4.23it/s]
 41%|████      | 2792/6837 [24:45<15:56,  4.23it/s]
 41%|████      | 2793/6837 [24:46<15:54,  4.24it/s]
 41%|████      | 2794/6837 [24:46<15:53,  4.24it/s]
 41%|████      | 2795/6837 [24:46<15:51,  4.25it/s]
 41%|████      | 2796/6837 [24:46<15:51,  4.25it/s]
 41%|████      | 2797/6837 [24:47<15:51,  4.25it/s]
 41%|████      | 2798/6837 [24:47<15:51,  4.24it/s]
 41%|████      | 2799/6837 [24:47<15:51,  4.24it/s]
 41%|████      | 2800/6837 [24:47<15:52,  4.24it/s]
                                                   
{'loss': 4.5217, 'grad_norm': 0.14217203855514526, 'learning_rate': 0.004412965831869858, 'epoch': 0.08}
+
 41%|████      | 2800/6837 [24:47<15:52,  4.24it/s]
 41%|████      | 2801/6837 [24:48<15:54,  4.23it/s]
 41%|████      | 2802/6837 [24:48<15:54,  4.23it/s]
 41%|████      | 2803/6837 [24:48<15:54,  4.23it/s]
 41%|████      | 2804/6837 [24:48<15:53,  4.23it/s]
 41%|████      | 2805/6837 [24:48<15:51,  4.24it/s]
 41%|████      | 2806/6837 [24:49<15:51,  4.24it/s]
 41%|████      | 2807/6837 [24:49<15:50,  4.24it/s]
 41%|████      | 2808/6837 [24:49<15:50,  4.24it/s]
 41%|████      | 2809/6837 [24:49<15:49,  4.24it/s]
 41%|████      | 2810/6837 [24:50<15:49,  4.24it/s]
 41%|████      | 2811/6837 [24:50<15:50,  4.24it/s]
 41%|████      | 2812/6837 [24:50<15:51,  4.23it/s]
 41%|████      | 2813/6837 [24:50<15:50,  4.23it/s]
 41%|████      | 2814/6837 [24:51<15:50,  4.23it/s]
 41%|████      | 2815/6837 [24:51<15:50,  4.23it/s]
 41%|████      | 2816/6837 [24:51<15:51,  4.23it/s]
 41%|████      | 2817/6837 [24:51<15:49,  4.24it/s]
 41%|████      | 2818/6837 [24:52<15:49,  4.23it/s]
 41%|████      | 2819/6837 [24:52<15:50,  4.23it/s]
 41%|████      | 2820/6837 [24:52<15:51,  4.22it/s]
 41%|████▏     | 2821/6837 [24:52<15:49,  4.23it/s]
 41%|████▏     | 2822/6837 [24:52<15:49,  4.23it/s]
 41%|████▏     | 2823/6837 [24:53<15:49,  4.23it/s]
 41%|█��██▏     | 2824/6837 [24:53<15:49,  4.23it/s]
 41%|████▏     | 2825/6837 [24:53<15:48,  4.23it/s]
                                                   
{'loss': 4.5218, 'grad_norm': 0.13789354264736176, 'learning_rate': 0.004379071506687446, 'epoch': 0.08}
+
 41%|████▏     | 2825/6837 [24:53<15:48,  4.23it/s]
 41%|████▏     | 2826/6837 [24:53<15:49,  4.22it/s]
 41%|████▏     | 2827/6837 [24:54<15:48,  4.23it/s]
 41%|████▏     | 2828/6837 [24:54<15:48,  4.23it/s]
 41%|████▏     | 2829/6837 [24:54<15:48,  4.23it/s]
 41%|████▏     | 2830/6837 [24:54<15:48,  4.23it/s]
 41%|████▏     | 2831/6837 [24:55<15:47,  4.23it/s]
 41%|████▏     | 2832/6837 [24:55<15:46,  4.23it/s]
 41%|████▏     | 2833/6837 [24:55<15:45,  4.23it/s]
 41%|████▏     | 2834/6837 [24:55<15:44,  4.24it/s]
 41%|████▏     | 2835/6837 [24:56<15:44,  4.24it/s]
 41%|████▏     | 2836/6837 [24:56<15:45,  4.23it/s]
 41%|████▏     | 2837/6837 [24:56<15:44,  4.24it/s]
 42%|████▏     | 2838/6837 [24:56<15:42,  4.24it/s]
 42%|████▏     | 2839/6837 [24:56<15:42,  4.24it/s]
 42%|████▏     | 2840/6837 [24:57<15:42,  4.24it/s]
 42%|████▏     | 2841/6837 [24:57<15:42,  4.24it/s]
 42%|████▏     | 2842/6837 [24:57<15:42,  4.24it/s]
 42%|████▏     | 2843/6837 [24:57<15:40,  4.24it/s]
 42%|████▏     | 2844/6837 [24:58<15:40,  4.24it/s]
 42%|████▏     | 2845/6837 [24:58<15:39,  4.25it/s]
 42%|████▏     | 2846/6837 [24:58<15:40,  4.24it/s]
 42%|████▏     | 2847/6837 [24:58<15:40,  4.24it/s]
 42%|████▏     | 2848/6837 [24:59<15:40,  4.24it/s]
 42%|████▏     | 2849/6837 [24:59<15:39,  4.25it/s]
 42%|████▏     | 2850/6837 [24:59<15:38,  4.25it/s]
                                                   
{'loss': 4.521, 'grad_norm': 0.14034835994243622, 'learning_rate': 0.00434495248991303, 'epoch': 0.08}
+
 42%|████▏     | 2850/6837 [24:59<15:38,  4.25it/s]
 42%|████▏     | 2851/6837 [24:59<15:40,  4.24it/s]
 42%|████▏     | 2852/6837 [25:00<15:39,  4.24it/s]
 42%|████▏     | 2853/6837 [25:00<16:11,  4.10it/s]
 42%|████▏     | 2854/6837 [25:00<16:36,  4.00it/s]
 42%|████▏     | 2855/6837 [25:00<16:54,  3.92it/s]
 42%|████▏     | 2856/6837 [25:01<17:06,  3.88it/s]
 42%|████▏     | 2857/6837 [25:01<16:56,  3.92it/s]
 42%|████▏     | 2858/6837 [25:01<16:37,  3.99it/s]
 42%|████▏     | 2859/6837 [25:01<16:22,  4.05it/s]
 42%|████▏     | 2860/6837 [25:02<16:44,  3.96it/s]
 42%|████▏     | 2861/6837 [25:02<17:02,  3.89it/s]
 42%|████▏     | 2862/6837 [25:02<17:12,  3.85it/s]
 42%|████▏     | 2863/6837 [25:02<17:17,  3.83it/s]
 42%|████▏     | 2864/6837 [25:03<17:20,  3.82it/s]
 42%|████▏     | 2865/6837 [25:03<17:25,  3.80it/s]
 42%|████▏     | 2866/6837 [25:03<17:25,  3.80it/s]
 42%|████▏     | 2867/6837 [25:03<17:31,  3.78it/s]
 42%|████▏     | 2868/6837 [25:04<17:42,  3.74it/s]
 42%|████▏     | 2869/6837 [25:04<17:40,  3.74it/s]
 42%|████▏     | 2870/6837 [25:04<17:37,  3.75it/s]
 42%|████▏     | 2871/6837 [25:05<17:29,  3.78it/s]
 42%|████▏     | 2872/6837 [25:05<17:26,  3.79it/s]
 42%|████▏     | 2873/6837 [25:05<17:35,  3.76it/s]
 42%|████▏     | 2874/6837 [25:05<17:37,  3.75it/s]
 42%|████▏     | 2875/6837 [25:06<17:31,  3.77it/s]
                                                   {'loss': 4.4993, 'grad_norm': 0.13763543963432312, 'learning_rate': 0.004310614340544934, 'epoch': 0.08}
+
 42%|████▏     | 2875/6837 [25:06<17:31,  3.77it/s]
 42%|████▏     | 2876/6837 [25:06<17:01,  3.88it/s]
 42%|████▏     | 2877/6837 [25:06<16:36,  3.98it/s]
 42%|████▏     | 2878/6837 [25:06<16:27,  4.01it/s]
 42%|████▏     | 2879/6837 [25:07<16:12,  4.07it/s]
 42%|████▏     | 2880/6837 [25:07<16:02,  4.11it/s]
 42%|████▏     | 2881/6837 [25:07<15:56,  4.14it/s]
 42%|████▏     | 2882/6837 [25:07<15:52,  4.15it/s]
 42%|████▏     | 2883/6837 [25:08<15:48,  4.17it/s]
 42%|████▏     | 2884/6837 [25:08<15:45,  4.18it/s]
 42%|████▏     | 2885/6837 [25:08<15:43,  4.19it/s]
 42%|████▏     | 2886/6837 [25:08<15:41,  4.20it/s]
 42%|████▏     | 2887/6837 [25:08<15:40,  4.20it/s]
 42%|████▏     | 2888/6837 [25:09<15:39,  4.21it/s]
 42%|████▏     | 2889/6837 [25:09<15:39,  4.20it/s]
 42%|████▏     | 2890/6837 [25:09<15:38,  4.20it/s]
 42%|████▏     | 2891/6837 [25:09<15:37,  4.21it/s]
 42%|████▏     | 2892/6837 [25:10<15:36,  4.21it/s]
 42%|████▏     | 2893/6837 [25:10<15:37,  4.21it/s]
 42%|████▏     | 2894/6837 [25:10<15:36,  4.21it/s]
 42%|████▏     | 2895/6837 [25:10<15:36,  4.21it/s]
 42%|████▏     | 2896/6837 [25:11<15:36,  4.21it/s]
 42%|████▏     | 2897/6837 [25:11<15:35,  4.21it/s]
 42%|████▏     | 2898/6837 [25:11<15:34,  4.21it/s]
 42%|████▏     | 2899/6837 [25:11<15:35,  4.21it/s]
 42%|████▏     | 2900/6837 [25:12<15:35,  4.21it/s]{'loss': 4.5119, 'grad_norm': 0.13792113959789276, 'learning_rate': 0.004276062653284676, 'epoch': 0.08}
+
                                                   

 42%|████▏     | 2900/6837 [25:12<15:35,  4.21it/s]
 42%|████▏     | 2901/6837 [25:12<15:36,  4.21it/s]
 42%|████▏     | 2902/6837 [25:12<15:34,  4.21it/s]
 42%|████▏     | 2903/6837 [25:12<15:34,  4.21it/s]
 42%|████▏     | 2904/6837 [25:13<15:33,  4.21it/s]
 42%|████▏     | 2905/6837 [25:13<16:07,  4.06it/s]
 43%|████▎     | 2906/6837 [25:13<16:29,  3.97it/s]
 43%|████▎     | 2907/6837 [25:13<16:11,  4.05it/s]
 43%|████▎     | 2908/6837 [25:14<15:57,  4.10it/s]
 43%|████▎     | 2909/6837 [25:14<15:47,  4.15it/s]
 43%|████▎     | 2910/6837 [25:14<15:40,  4.18it/s]
 43%|████▎     | 2911/6837 [25:14<15:36,  4.19it/s]
 43%|████▎     | 2912/6837 [25:14<15:35,  4.20it/s]
 43%|████▎     | 2913/6837 [25:15<15:33,  4.21it/s]
 43%|████▎     | 2914/6837 [25:15<15:31,  4.21it/s]
 43%|████▎     | 2915/6837 [25:15<15:29,  4.22it/s]
 43%|████▎     | 2916/6837 [25:15<15:28,  4.22it/s]
 43%|████▎     | 2917/6837 [25:16<15:27,  4.23it/s]
 43%|████▎     | 2918/6837 [25:16<15:27,  4.23it/s]
 43%|████▎     | 2919/6837 [25:16<15:27,  4.23it/s]
 43%|████▎     | 2920/6837 [25:16<15:26,  4.23it/s]
 43%|████▎     | 2921/6837 [25:17<15:27,  4.22it/s]
 43%|████▎     | 2922/6837 [25:17<15:27,  4.22it/s]
 43%|████▎     | 2923/6837 [25:17<15:26,  4.23it/s]
 43%|████▎     | 2924/6837 [25:17<15:25,  4.23it/s]
 43%|████▎     | 2925/6837 [25:18<15:26,  4.22it/s]
                                                   
{'loss': 4.516, 'grad_norm': 0.16946350038051605, 'learning_rate': 0.004241303057625416, 'epoch': 0.09}
+
 43%|████▎     | 2925/6837 [25:18<15:26,  4.22it/s]
 43%|████▎     | 2926/6837 [25:18<15:34,  4.19it/s]
 43%|████▎     | 2927/6837 [25:18<16:07,  4.04it/s]
 43%|████▎     | 2928/6837 [25:18<16:26,  3.96it/s]
 43%|████▎     | 2929/6837 [25:19<16:39,  3.91it/s]
 43%|████▎     | 2930/6837 [25:19<16:55,  3.85it/s]
 43%|████▎     | 2931/6837 [25:19<16:58,  3.83it/s]
 43%|████▎     | 2932/6837 [25:19<17:11,  3.79it/s]
 43%|████▎     | 2933/6837 [25:20<17:10,  3.79it/s]
 43%|████▎     | 2934/6837 [25:20<17:10,  3.79it/s]
 43%|████▎     | 2935/6837 [25:20<17:05,  3.81it/s]
 43%|████▎     | 2936/6837 [25:20<17:04,  3.81it/s]
 43%|████▎     | 2937/6837 [25:21<17:08,  3.79it/s]
 43%|████▎     | 2938/6837 [25:21<17:13,  3.77it/s]
 43%|████▎     | 2939/6837 [25:21<17:13,  3.77it/s]
 43%|████▎     | 2940/6837 [25:21<17:12,  3.78it/s]
 43%|████▎     | 2941/6837 [25:22<17:18,  3.75it/s]
 43%|████▎     | 2942/6837 [25:22<17:14,  3.77it/s]
 43%|████▎     | 2943/6837 [25:22<17:14,  3.76it/s]
 43%|████▎     | 2944/6837 [25:23<17:11,  3.78it/s]
 43%|████▎     | 2945/6837 [25:23<17:13,  3.77it/s]
 43%|████▎     | 2946/6837 [25:23<17:14,  3.76it/s]
 43%|████▎     | 2947/6837 [25:23<17:11,  3.77it/s]
 43%|████▎     | 2948/6837 [25:24<17:13,  3.76it/s]
 43%|████▎     | 2949/6837 [25:24<17:12,  3.76it/s]
 43%|████▎     | 2950/6837 [25:24<17:05,  3.79it/s]
                                                   
{'loss': 4.5024, 'grad_norm': 0.1543588638305664, 'learning_rate': 0.004206341216934752, 'epoch': 0.09}
+
 43%|████▎     | 2950/6837 [25:24<17:05,  3.79it/s]
 43%|████▎     | 2951/6837 [25:24<17:07,  3.78it/s]
 43%|████▎     | 2952/6837 [25:25<17:06,  3.78it/s]
 43%|████▎     | 2953/6837 [25:25<17:04,  3.79it/s]
 43%|████▎     | 2954/6837 [25:25<17:01,  3.80it/s]
 43%|████▎     | 2955/6837 [25:25<17:07,  3.78it/s]
 43%|████▎     | 2956/6837 [25:26<16:42,  3.87it/s]
 43%|████▎     | 2957/6837 [25:26<16:17,  3.97it/s]
 43%|████▎     | 2958/6837 [25:26<16:00,  4.04it/s]
 43%|████▎     | 2959/6837 [25:26<15:48,  4.09it/s]
 43%|████▎     | 2960/6837 [25:27<15:38,  4.13it/s]
 43%|████▎     | 2961/6837 [25:27<15:35,  4.14it/s]
 43%|████▎     | 2962/6837 [25:27<15:28,  4.17it/s]
 43%|████▎     | 2963/6837 [25:27<15:25,  4.18it/s]
 43%|████▎     | 2964/6837 [25:28<15:23,  4.19it/s]
 43%|████▎     | 2965/6837 [25:28<15:23,  4.19it/s]
 43%|████▎     | 2966/6837 [25:28<15:21,  4.20it/s]
 43%|████▎     | 2967/6837 [25:28<15:20,  4.21it/s]
 43%|████▎     | 2968/6837 [25:29<15:18,  4.21it/s]
 43%|████▎     | 2969/6837 [25:29<15:28,  4.17it/s]
 43%|████▎     | 2970/6837 [25:29<15:24,  4.18it/s]
 43%|████▎     | 2971/6837 [25:29<15:21,  4.20it/s]
 43%|████▎     | 2972/6837 [25:30<15:19,  4.20it/s]
 43%|████▎     | 2973/6837 [25:30<15:17,  4.21it/s]
 43%|████▎     | 2974/6837 [25:30<15:15,  4.22it/s]
 44%|████▎     | 2975/6837 [25:30<15:14,  4.22it/s]
                                                   
{'loss': 4.4904, 'grad_norm': 0.170481875538826, 'learning_rate': 0.004171182827531988, 'epoch': 0.09}
+
 44%|████▎     | 2975/6837 [25:30<15:14,  4.22it/s]
 44%|████▎     | 2976/6837 [25:30<15:17,  4.21it/s]
 44%|████▎     | 2977/6837 [25:31<15:15,  4.22it/s]
 44%|████▎     | 2978/6837 [25:31<15:13,  4.22it/s]
 44%|████▎     | 2979/6837 [25:31<15:14,  4.22it/s]
 44%|████▎     | 2980/6837 [25:31<15:14,  4.22it/s]
 44%|████▎     | 2981/6837 [25:32<15:12,  4.23it/s]
 44%|████▎     | 2982/6837 [25:32<15:11,  4.23it/s]
 44%|████▎     | 2983/6837 [25:32<15:11,  4.23it/s]
 44%|████▎     | 2984/6837 [25:32<15:11,  4.23it/s]
 44%|████▎     | 2985/6837 [25:33<15:12,  4.22it/s]
 44%|████▎     | 2986/6837 [25:33<15:11,  4.22it/s]
 44%|████▎     | 2987/6837 [25:33<15:12,  4.22it/s]
 44%|████▎     | 2988/6837 [25:33<15:12,  4.22it/s]
 44%|████▎     | 2989/6837 [25:34<15:11,  4.22it/s]
 44%|████▎     | 2990/6837 [25:34<15:11,  4.22it/s]
 44%|████▎     | 2991/6837 [25:34<15:11,  4.22it/s]
 44%|████▍     | 2992/6837 [25:34<15:11,  4.22it/s]
 44%|████▍     | 2993/6837 [25:34<15:11,  4.22it/s]
 44%|████▍     | 2994/6837 [25:35<15:11,  4.22it/s]
 44%|████▍     | 2995/6837 [25:35<15:09,  4.22it/s]
 44%|████▍     | 2996/6837 [25:35<15:08,  4.23it/s]
 44%|████▍     | 2997/6837 [25:35<15:10,  4.22it/s]
 44%|████▍     | 2998/6837 [25:36<15:08,  4.22it/s]
 44%|████▍     | 2999/6837 [25:36<15:12,  4.20it/s]
 44%|████▍     | 3000/6837 [25:36<15:11,  4.21it/s]
                                                   
{'loss': 4.4967, 'grad_norm': 0.14753445982933044, 'learning_rate': 0.0041358336177600334, 'epoch': 0.09}
+
 44%|████▍     | 3000/6837 [25:36<15:11,  4.21it/s]
 44%|████▍     | 3001/6837 [25:36<15:13,  4.20it/s]
 44%|████▍     | 3002/6837 [25:37<15:09,  4.22it/s]
 44%|████▍     | 3003/6837 [25:37<15:07,  4.22it/s]
 44%|████▍     | 3004/6837 [25:37<15:06,  4.23it/s]
 44%|████▍     | 3005/6837 [25:37<15:05,  4.23it/s]
 44%|████▍     | 3006/6837 [25:38<15:04,  4.23it/s]
 44%|████▍     | 3007/6837 [25:38<15:03,  4.24it/s]
 44%|████▍     | 3008/6837 [25:38<15:06,  4.23it/s]
 44%|████▍     | 3009/6837 [25:38<15:05,  4.23it/s]
 44%|████▍     | 3010/6837 [25:39<15:03,  4.23it/s]
 44%|████▍     | 3011/6837 [25:39<15:02,  4.24it/s]
 44%|████▍     | 3012/6837 [25:39<15:02,  4.24it/s]
 44%|████▍     | 3013/6837 [25:39<15:02,  4.24it/s]
 44%|████▍     | 3014/6837 [25:39<15:01,  4.24it/s]
 44%|████▍     | 3015/6837 [25:40<15:02,  4.24it/s]
 44%|████▍     | 3016/6837 [25:40<15:02,  4.23it/s]
 44%|████▍     | 3017/6837 [25:40<15:02,  4.23it/s]
 44%|████▍     | 3018/6837 [25:40<15:01,  4.24it/s]
 44%|████▍     | 3019/6837 [25:41<15:00,  4.24it/s]
 44%|████▍     | 3020/6837 [25:41<15:00,  4.24it/s]
 44%|████▍     | 3021/6837 [25:41<15:00,  4.24it/s]
 44%|████▍     | 3022/6837 [25:41<14:59,  4.24it/s]
 44%|████▍     | 3023/6837 [25:42<14:59,  4.24it/s]
 44%|████▍     | 3024/6837 [25:42<14:58,  4.24it/s]
 44%|████▍     | 3025/6837 [25:42<14:57,  4.25it/s]
                                                   
{'loss': 4.5004, 'grad_norm': 0.12795390188694, 'learning_rate': 0.0041002993470520875, 'epoch': 0.09}
+
 44%|████▍     | 3025/6837 [25:42<14:57,  4.25it/s]
 44%|████▍     | 3026/6837 [25:42<15:03,  4.22it/s]
 44%|████▍     | 3027/6837 [25:43<15:08,  4.19it/s]
 44%|████▍     | 3028/6837 [25:43<15:39,  4.05it/s]
 44%|████▍     | 3029/6837 [25:43<15:59,  3.97it/s]
 44%|████▍     | 3030/6837 [25:43<15:41,  4.05it/s]
 44%|████▍     | 3031/6837 [25:44<15:28,  4.10it/s]
 44%|████▍     | 3032/6837 [25:44<15:19,  4.14it/s]
 44%|████▍     | 3033/6837 [25:44<15:12,  4.17it/s]
 44%|████▍     | 3034/6837 [25:44<15:07,  4.19it/s]
 44%|████▍     | 3035/6837 [25:44<15:04,  4.20it/s]
 44%|████▍     | 3036/6837 [25:45<15:03,  4.21it/s]
 44%|████▍     | 3037/6837 [25:45<15:00,  4.22it/s]
 44%|████▍     | 3038/6837 [25:45<14:59,  4.23it/s]
 44%|████▍     | 3039/6837 [25:45<14:57,  4.23it/s]
 44%|████▍     | 3040/6837 [25:46<14:56,  4.23it/s]
 44%|████▍     | 3041/6837 [25:46<14:55,  4.24it/s]
 44%|████▍     | 3042/6837 [25:46<14:57,  4.23it/s]
 45%|████▍     | 3043/6837 [25:46<15:33,  4.06it/s]
 45%|████▍     | 3044/6837 [25:47<15:25,  4.10it/s]
 45%|████▍     | 3045/6837 [25:47<15:16,  4.14it/s]
 45%|████▍     | 3046/6837 [25:47<15:10,  4.16it/s]
 45%|████▍     | 3047/6837 [25:47<15:08,  4.17it/s]
 45%|████▍     | 3048/6837 [25:48<15:05,  4.18it/s]
 45%|████▍     | 3049/6837 [25:48<15:04,  4.19it/s]
 45%|████▍     | 3050/6837 [25:48<15:01,  4.20it/s]
                                                   
{'loss': 4.4861, 'grad_norm': 0.1477264165878296, 'learning_rate': 0.004064585804993258, 'epoch': 0.09}
+
 45%|████▍     | 3050/6837 [25:48<15:01,  4.20it/s]
 45%|████▍     | 3051/6837 [25:48<15:03,  4.19it/s]
 45%|████▍     | 3052/6837 [25:49<15:01,  4.20it/s]
 45%|████▍     | 3053/6837 [25:49<14:59,  4.21it/s]
 45%|████▍     | 3054/6837 [25:49<14:58,  4.21it/s]
 45%|████▍     | 3055/6837 [25:49<14:57,  4.22it/s]
 45%|████▍     | 3056/6837 [25:49<14:55,  4.22it/s]
 45%|████▍     | 3057/6837 [25:50<14:54,  4.22it/s]
 45%|████▍     | 3058/6837 [25:50<14:55,  4.22it/s]
 45%|████▍     | 3059/6837 [25:50<14:54,  4.22it/s]
 45%|████▍     | 3060/6837 [25:50<14:53,  4.23it/s]
 45%|████▍     | 3061/6837 [25:51<14:52,  4.23it/s]
 45%|████▍     | 3062/6837 [25:51<14:53,  4.23it/s]
 45%|████▍     | 3063/6837 [25:51<14:51,  4.23it/s]
 45%|████▍     | 3064/6837 [25:51<14:51,  4.23it/s]
 45%|████▍     | 3065/6837 [25:52<14:51,  4.23it/s]
 45%|████▍     | 3066/6837 [25:52<14:51,  4.23it/s]
 45%|████▍     | 3067/6837 [25:52<14:51,  4.23it/s]
 45%|████▍     | 3068/6837 [25:52<14:51,  4.23it/s]
 45%|████▍     | 3069/6837 [25:53<14:50,  4.23it/s]
 45%|████▍     | 3070/6837 [25:53<14:49,  4.23it/s]
 45%|████▍     | 3071/6837 [25:53<14:50,  4.23it/s]
 45%|████▍     | 3072/6837 [25:53<14:50,  4.23it/s]
 45%|████▍     | 3073/6837 [25:53<14:49,  4.23it/s]
 45%|████▍     | 3074/6837 [25:54<14:49,  4.23it/s]
 45%|████▍     | 3075/6837 [25:54<14:50,  4.23it/s]
                                                   
{'loss': 4.4949, 'grad_norm': 0.17127594351768494, 'learning_rate': 0.004028698810377262, 'epoch': 0.09}
+
 45%|████▍     | 3075/6837 [25:54<14:50,  4.23it/s]
 45%|████▍     | 3076/6837 [25:54<14:52,  4.21it/s]
 45%|████▌     | 3077/6837 [25:54<14:51,  4.22it/s]
 45%|████▌     | 3078/6837 [25:55<14:49,  4.23it/s]
 45%|████▌     | 3079/6837 [25:55<14:48,  4.23it/s]
 45%|████▌     | 3080/6837 [25:55<14:48,  4.23it/s]
 45%|████▌     | 3081/6837 [25:55<14:48,  4.23it/s]
 45%|████▌     | 3082/6837 [25:56<14:48,  4.23it/s]
 45%|████▌     | 3083/6837 [25:56<14:47,  4.23it/s]
 45%|████▌     | 3084/6837 [25:56<14:47,  4.23it/s]
 45%|████▌     | 3085/6837 [25:56<14:47,  4.23it/s]
 45%|████▌     | 3086/6837 [25:57<14:47,  4.23it/s]
 45%|████▌     | 3087/6837 [25:57<14:48,  4.22it/s]
 45%|████▌     | 3088/6837 [25:57<14:47,  4.22it/s]
 45%|████▌     | 3089/6837 [25:57<14:46,  4.23it/s]
 45%|████▌     | 3090/6837 [25:58<14:46,  4.23it/s]
 45%|████▌     | 3091/6837 [25:58<14:46,  4.23it/s]
 45%|████▌     | 3092/6837 [25:58<14:44,  4.24it/s]
 45%|████▌     | 3093/6837 [25:58<14:43,  4.24it/s]
 45%|████▌     | 3094/6837 [25:58<14:43,  4.24it/s]
 45%|████▌     | 3095/6837 [25:59<14:42,  4.24it/s]
 45%|████▌     | 3096/6837 [25:59<14:41,  4.24it/s]
 45%|████▌     | 3097/6837 [25:59<14:42,  4.24it/s]
 45%|████▌     | 3098/6837 [25:59<14:43,  4.23it/s]
 45%|████▌     | 3099/6837 [26:00<14:41,  4.24it/s]
 45%|████▌     | 3100/6837 [26:00<14:41,  4.24it/s]
                                                   
{'loss': 4.4896, 'grad_norm': 0.17606590688228607, 'learning_rate': 0.003992644210258377, 'epoch': 0.09}
+
 45%|████▌     | 3100/6837 [26:00<14:41,  4.24it/s]
 45%|████▌     | 3101/6837 [26:00<14:43,  4.23it/s]
 45%|████▌     | 3102/6837 [26:00<14:43,  4.23it/s]
 45%|████▌     | 3103/6837 [26:01<14:42,  4.23it/s]
 45%|████▌     | 3104/6837 [26:01<14:43,  4.23it/s]
 45%|████▌     | 3105/6837 [26:01<14:44,  4.22it/s]
 45%|████▌     | 3106/6837 [26:01<14:43,  4.22it/s]
 45%|████▌     | 3107/6837 [26:02<14:41,  4.23it/s]
 45%|████▌     | 3108/6837 [26:02<14:42,  4.23it/s]
 45%|████▌     | 3109/6837 [26:02<14:42,  4.23it/s]
 45%|████▌     | 3110/6837 [26:02<14:41,  4.23it/s]
 46%|████▌     | 3111/6837 [26:02<14:39,  4.24it/s]
 46%|████▌     | 3112/6837 [26:03<14:41,  4.23it/s]
 46%|████▌     | 3113/6837 [26:03<14:41,  4.22it/s]
 46%|████▌     | 3114/6837 [26:03<14:40,  4.23it/s]
 46%|████▌     | 3115/6837 [26:03<14:41,  4.22it/s]
 46%|████▌     | 3116/6837 [26:04<14:41,  4.22it/s]
 46%|████▌     | 3117/6837 [26:04<14:39,  4.23it/s]
 46%|████▌     | 3118/6837 [26:04<14:41,  4.22it/s]
 46%|████▌     | 3119/6837 [26:04<14:38,  4.23it/s]
 46%|████▌     | 3120/6837 [26:05<14:38,  4.23it/s]
 46%|████▌     | 3121/6837 [26:05<14:36,  4.24it/s]
 46%|████▌     | 3122/6837 [26:05<14:37,  4.23it/s]
 46%|████▌     | 3123/6837 [26:05<14:37,  4.23it/s]
 46%|████▌     | 3124/6837 [26:06<14:37,  4.23it/s]
 46%|████▌     | 3125/6837 [26:06<14:37,  4.23it/s]
                                                   
{'loss': 4.4878, 'grad_norm': 0.15565237402915955, 'learning_rate': 0.003956427878998784, 'epoch': 0.09}
+
 46%|████▌     | 3125/6837 [26:06<14:37,  4.23it/s]
 46%|████▌     | 3126/6837 [26:06<14:39,  4.22it/s]
 46%|████▌     | 3127/6837 [26:06<14:39,  4.22it/s]
 46%|████▌     | 3128/6837 [26:07<14:41,  4.21it/s]
 46%|████▌     | 3129/6837 [26:07<14:38,  4.22it/s]
 46%|████▌     | 3130/6837 [26:07<14:39,  4.22it/s]
 46%|████▌     | 3131/6837 [26:07<14:38,  4.22it/s]
 46%|████▌     | 3132/6837 [26:07<14:36,  4.23it/s]
 46%|████▌     | 3133/6837 [26:08<14:36,  4.23it/s]
 46%|████▌     | 3134/6837 [26:08<14:35,  4.23it/s]
 46%|████▌     | 3135/6837 [26:08<14:35,  4.23it/s]
 46%|████▌     | 3136/6837 [26:08<14:35,  4.23it/s]
 46%|████▌     | 3137/6837 [26:09<14:36,  4.22it/s]
 46%|████▌     | 3138/6837 [26:09<14:37,  4.22it/s]
 46%|████▌     | 3139/6837 [26:09<14:36,  4.22it/s]
 46%|████▌     | 3140/6837 [26:09<14:36,  4.22it/s]
 46%|████▌     | 3141/6837 [26:10<14:35,  4.22it/s]
 46%|████▌     | 3142/6837 [26:10<14:33,  4.23it/s]
 46%|████▌     | 3143/6837 [26:10<14:35,  4.22it/s]
 46%|████▌     | 3144/6837 [26:10<14:34,  4.22it/s]
 46%|████▌     | 3145/6837 [26:11<14:35,  4.22it/s]
 46%|████▌     | 3146/6837 [26:11<14:34,  4.22it/s]
 46%|████▌     | 3147/6837 [26:11<14:34,  4.22it/s]
 46%|████▌     | 3148/6837 [26:11<14:34,  4.22it/s]
 46%|████▌     | 3149/6837 [26:11<14:35,  4.21it/s]
 46%|████▌     | 3150/6837 [26:12<14:34,  4.22it/s]
                                                   
{'loss': 4.4914, 'grad_norm': 0.1586657464504242, 'learning_rate': 0.003920055717311451, 'epoch': 0.09}
+
 46%|████▌     | 3150/6837 [26:12<14:34,  4.22it/s]
 46%|████▌     | 3151/6837 [26:12<14:36,  4.21it/s]
 46%|████▌     | 3152/6837 [26:12<14:36,  4.20it/s]
 46%|████▌     | 3153/6837 [26:12<14:35,  4.21it/s]
 46%|████▌     | 3154/6837 [26:13<14:33,  4.22it/s]
 46%|████▌     | 3155/6837 [26:13<15:01,  4.09it/s]
 46%|████▌     | 3156/6837 [26:13<14:53,  4.12it/s]
 46%|████▌     | 3157/6837 [26:13<14:46,  4.15it/s]
 46%|████▌     | 3158/6837 [26:14<14:40,  4.18it/s]
 46%|████▌     | 3159/6837 [26:14<14:38,  4.18it/s]
 46%|████▌     | 3160/6837 [26:14<14:37,  4.19it/s]
 46%|████▌     | 3161/6837 [26:14<14:34,  4.20it/s]
 46%|████▌     | 3162/6837 [26:15<14:32,  4.21it/s]
 46%|████▋     | 3163/6837 [26:15<14:31,  4.22it/s]
 46%|████▋     | 3164/6837 [26:15<14:29,  4.22it/s]
 46%|████▋     | 3165/6837 [26:15<14:28,  4.23it/s]
 46%|████▋     | 3166/6837 [26:16<14:29,  4.22it/s]
 46%|████▋     | 3167/6837 [26:16<14:27,  4.23it/s]
 46%|████▋     | 3168/6837 [26:16<14:27,  4.23it/s]
 46%|████▋     | 3169/6837 [26:16<14:27,  4.23it/s]
 46%|████▋     | 3170/6837 [26:16<14:27,  4.23it/s]
 46%|████▋     | 3171/6837 [26:17<14:26,  4.23it/s]
 46%|████▋     | 3172/6837 [26:17<14:25,  4.24it/s]
 46%|████▋     | 3173/6837 [26:17<14:25,  4.23it/s]
 46%|████▋     | 3174/6837 [26:17<14:25,  4.23it/s]
 46%|████▋     | 3175/6837 [26:18<14:24,  4.24it/s]
                                                   
{'loss': 4.4865, 'grad_norm': 0.14921404421329498, 'learning_rate': 0.0038835336512987417, 'epoch': 0.09}
+
 46%|████▋     | 3175/6837 [26:18<14:24,  4.24it/s]
 46%|████▋     | 3176/6837 [26:18<14:26,  4.23it/s]
 46%|████▋     | 3177/6837 [26:18<14:25,  4.23it/s]
 46%|████▋     | 3178/6837 [26:18<14:24,  4.23it/s]
 46%|████▋     | 3179/6837 [26:19<14:24,  4.23it/s]
 47%|████▋     | 3180/6837 [26:19<14:24,  4.23it/s]
 47%|████▋     | 3181/6837 [26:19<14:25,  4.23it/s]
 47%|████▋     | 3182/6837 [26:19<14:23,  4.23it/s]
 47%|████▋     | 3183/6837 [26:20<14:22,  4.24it/s]
 47%|████▋     | 3184/6837 [26:20<14:23,  4.23it/s]
 47%|████▋     | 3185/6837 [26:20<14:21,  4.24it/s]
 47%|████▋     | 3186/6837 [26:20<14:22,  4.24it/s]
 47%|████▋     | 3187/6837 [26:20<14:23,  4.23it/s]
 47%|████▋     | 3188/6837 [26:21<14:23,  4.23it/s]
 47%|████▋     | 3189/6837 [26:21<14:23,  4.22it/s]
 47%|████▋     | 3190/6837 [26:21<14:22,  4.23it/s]
 47%|████▋     | 3191/6837 [26:21<14:23,  4.22it/s]
 47%|████▋     | 3192/6837 [26:22<14:22,  4.22it/s]
 47%|████▋     | 3193/6837 [26:22<14:21,  4.23it/s]
 47%|████▋     | 3194/6837 [26:22<14:20,  4.23it/s]
 47%|████▋     | 3195/6837 [26:22<14:21,  4.23it/s]
 47%|████▋     | 3196/6837 [26:23<14:22,  4.22it/s]
 47%|████▋     | 3197/6837 [26:23<14:20,  4.23it/s]
 47%|████▋     | 3198/6837 [26:23<14:20,  4.23it/s]
 47%|████▋     | 3199/6837 [26:23<14:20,  4.23it/s]
 47%|████▋     | 3200/6837 [26:24<14:20,  4.23it/s]
                                                   
{'loss': 4.4808, 'grad_norm': 0.14431630074977875, 'learning_rate': 0.0038468676314868736, 'epoch': 0.09}
+
 47%|████▋     | 3200/6837 [26:24<14:20,  4.23it/s]
 47%|████▋     | 3201/6837 [26:24<14:22,  4.22it/s]
 47%|████▋     | 3202/6837 [26:24<14:23,  4.21it/s]
 47%|████▋     | 3203/6837 [26:24<14:21,  4.22it/s]
 47%|████▋     | 3204/6837 [26:25<14:21,  4.22it/s]
 47%|████▋     | 3205/6837 [26:25<14:20,  4.22it/s]
 47%|████▋     | 3206/6837 [26:25<14:20,  4.22it/s]
 47%|████▋     | 3207/6837 [26:25<14:20,  4.22it/s]
 47%|████▋     | 3208/6837 [26:25<14:19,  4.22it/s]
 47%|████▋     | 3209/6837 [26:26<14:18,  4.23it/s]
 47%|████▋     | 3210/6837 [26:26<14:18,  4.22it/s]
 47%|████▋     | 3211/6837 [26:26<14:17,  4.23it/s]
 47%|████▋     | 3212/6837 [26:26<14:16,  4.23it/s]
 47%|████▋     | 3213/6837 [26:27<14:17,  4.23it/s]
 47%|████▋     | 3214/6837 [26:27<14:17,  4.22it/s]
 47%|████▋     | 3215/6837 [26:27<14:16,  4.23it/s]
 47%|████▋     | 3216/6837 [26:27<14:15,  4.23it/s]
 47%|████▋     | 3217/6837 [26:28<14:15,  4.23it/s]
 47%|████▋     | 3218/6837 [26:28<14:14,  4.23it/s]
 47%|████▋     | 3219/6837 [26:28<14:14,  4.23it/s]
 47%|████▋     | 3220/6837 [26:28<14:16,  4.22it/s]
 47%|████▋     | 3221/6837 [26:29<14:16,  4.22it/s]
 47%|████▋     | 3222/6837 [26:29<14:14,  4.23it/s]
 47%|████▋     | 3223/6837 [26:29<14:13,  4.24it/s]
 47%|████▋     | 3224/6837 [26:29<14:13,  4.23it/s]
 47%|████▋     | 3225/6837 [26:29<14:14,  4.23it/s]
                                                   
{'loss': 4.484, 'grad_norm': 0.14424818754196167, 'learning_rate': 0.003810063631856402, 'epoch': 0.09}
+
 47%|████▋     | 3225/6837 [26:29<14:14,  4.23it/s]
 47%|████▋     | 3226/6837 [26:30<14:14,  4.22it/s]
 47%|████▋     | 3227/6837 [26:30<14:13,  4.23it/s]
 47%|████▋     | 3228/6837 [26:30<14:12,  4.23it/s]
 47%|████▋     | 3229/6837 [26:30<14:11,  4.24it/s]
 47%|████▋     | 3230/6837 [26:31<14:08,  4.25it/s]
 47%|████▋     | 3231/6837 [26:31<14:10,  4.24it/s]
 47%|████▋     | 3232/6837 [26:31<14:10,  4.24it/s]
 47%|████▋     | 3233/6837 [26:31<14:11,  4.23it/s]
 47%|████▋     | 3234/6837 [26:32<14:10,  4.23it/s]
 47%|████▋     | 3235/6837 [26:32<14:09,  4.24it/s]
 47%|████▋     | 3236/6837 [26:32<14:09,  4.24it/s]
 47%|████▋     | 3237/6837 [26:32<14:08,  4.24it/s]
 47%|████▋     | 3238/6837 [26:33<14:08,  4.24it/s]
 47%|████▋     | 3239/6837 [26:33<14:09,  4.24it/s]
 47%|████▋     | 3240/6837 [26:33<14:09,  4.23it/s]
 47%|████▋     | 3241/6837 [26:33<14:08,  4.24it/s]
 47%|████▋     | 3242/6837 [26:34<14:10,  4.23it/s]
 47%|████▋     | 3243/6837 [26:34<14:09,  4.23it/s]
 47%|████▋     | 3244/6837 [26:34<14:09,  4.23it/s]
 47%|████▋     | 3245/6837 [26:34<14:09,  4.23it/s]
 47%|████▋     | 3246/6837 [26:34<14:10,  4.22it/s]
 47%|████▋     | 3247/6837 [26:35<14:10,  4.22it/s]
 48%|████▊     | 3248/6837 [26:35<14:09,  4.22it/s]
 48%|████▊     | 3249/6837 [26:35<14:09,  4.22it/s]
 48%|████▊     | 3250/6837 [26:35<14:08,  4.23it/s]
                                                   
{'loss': 4.4903, 'grad_norm': 0.13489964604377747, 'learning_rate': 0.0037731276488688773, 'epoch': 0.1}
+
 48%|████▊     | 3250/6837 [26:35<14:08,  4.23it/s]
 48%|████▊     | 3251/6837 [26:36<14:11,  4.21it/s]
 48%|████▊     | 3252/6837 [26:36<14:09,  4.22it/s]
 48%|████▊     | 3253/6837 [26:36<14:09,  4.22it/s]
 48%|████▊     | 3254/6837 [26:36<14:08,  4.22it/s]
 48%|████▊     | 3255/6837 [26:37<14:08,  4.22it/s]
 48%|████▊     | 3256/6837 [26:37<14:09,  4.21it/s]
 48%|████▊     | 3257/6837 [26:37<14:09,  4.21it/s]
 48%|████▊     | 3258/6837 [26:37<14:06,  4.23it/s]
 48%|████▊     | 3259/6837 [26:38<14:05,  4.23it/s]
 48%|████▊     | 3260/6837 [26:38<14:06,  4.22it/s]
 48%|████▊     | 3261/6837 [26:38<14:04,  4.23it/s]
 48%|████▊     | 3262/6837 [26:38<14:03,  4.24it/s]
 48%|████▊     | 3263/6837 [26:38<14:03,  4.24it/s]
 48%|████▊     | 3264/6837 [26:39<14:04,  4.23it/s]
 48%|████▊     | 3265/6837 [26:39<14:04,  4.23it/s]
 48%|████▊     | 3266/6837 [26:39<14:04,  4.23it/s]
 48%|████▊     | 3267/6837 [26:39<14:03,  4.23it/s]
 48%|████▊     | 3268/6837 [26:40<14:04,  4.23it/s]
 48%|████▊     | 3269/6837 [26:40<14:04,  4.23it/s]
 48%|████▊     | 3270/6837 [26:40<14:04,  4.22it/s]
 48%|████▊     | 3271/6837 [26:40<14:04,  4.22it/s]
 48%|████▊     | 3272/6837 [26:41<14:03,  4.22it/s]
 48%|████▊     | 3273/6837 [26:41<14:03,  4.23it/s]
 48%|████▊     | 3274/6837 [26:41<14:03,  4.22it/s]
 48%|████▊     | 3275/6837 [26:41<14:02,  4.23it/s]
                                                   
{'loss': 4.4782, 'grad_norm': 0.17107246816158295, 'learning_rate': 0.003736065700489853, 'epoch': 0.1}
+
 48%|████▊     | 3275/6837 [26:41<14:02,  4.23it/s]
 48%|████▊     | 3276/6837 [26:42<14:02,  4.22it/s]
 48%|████▊     | 3277/6837 [26:42<14:00,  4.23it/s]
 48%|████▊     | 3278/6837 [26:42<14:01,  4.23it/s]
 48%|████▊     | 3279/6837 [26:42<13:59,  4.24it/s]
 48%|████▊     | 3280/6837 [26:42<13:58,  4.24it/s]
 48%|████▊     | 3281/6837 [26:43<13:58,  4.24it/s]
 48%|████▊     | 3282/6837 [26:43<14:31,  4.08it/s]
 48%|████▊     | 3283/6837 [26:43<14:50,  3.99it/s]
 48%|████▊     | 3284/6837 [26:44<15:02,  3.94it/s]
 48%|████▊     | 3285/6837 [26:44<15:14,  3.88it/s]
 48%|████▊     | 3286/6837 [26:44<14:54,  3.97it/s]
 48%|████▊     | 3287/6837 [26:44<14:39,  4.03it/s]
 48%|████▊     | 3288/6837 [26:45<15:00,  3.94it/s]
 48%|████▊     | 3289/6837 [26:45<15:15,  3.87it/s]
 48%|████▊     | 3290/6837 [26:45<15:22,  3.84it/s]
 48%|████▊     | 3291/6837 [26:45<15:29,  3.82it/s]
 48%|████▊     | 3292/6837 [26:46<15:30,  3.81it/s]
 48%|████▊     | 3293/6837 [26:46<15:27,  3.82it/s]
 48%|████▊     | 3294/6837 [26:46<15:30,  3.81it/s]
 48%|████▊     | 3295/6837 [26:46<15:38,  3.78it/s]
 48%|████▊     | 3296/6837 [26:47<15:42,  3.76it/s]
 48%|████▊     | 3297/6837 [26:47<15:38,  3.77it/s]
 48%|████▊     | 3298/6837 [26:47<15:39,  3.77it/s]
 48%|████▊     | 3299/6837 [26:47<15:07,  3.90it/s]
 48%|████▊     | 3300/6837 [26:48<14:44,  4.00it/s]
                                                   
{'loss': 4.4687, 'grad_norm': 0.1348651647567749, 'learning_rate': 0.003698883825208372, 'epoch': 0.1}
+
 48%|████▊     | 3300/6837 [26:48<14:44,  4.00it/s]
 48%|████▊     | 3301/6837 [26:48<14:33,  4.05it/s]
 48%|████▊     | 3302/6837 [26:48<14:23,  4.10it/s]
 48%|████▊     | 3303/6837 [26:48<14:14,  4.13it/s]
 48%|████▊     | 3304/6837 [26:49<14:08,  4.16it/s]
 48%|████▊     | 3305/6837 [26:49<14:05,  4.18it/s]
 48%|████▊     | 3306/6837 [26:49<14:02,  4.19it/s]
 48%|████▊     | 3307/6837 [26:49<13:58,  4.21it/s]
 48%|████▊     | 3308/6837 [26:50<13:57,  4.22it/s]
 48%|████▊     | 3309/6837 [26:50<13:57,  4.21it/s]
 48%|████▊     | 3310/6837 [26:50<13:55,  4.22it/s]
 48%|████▊     | 3311/6837 [26:50<13:54,  4.22it/s]
 48%|████▊     | 3312/6837 [26:51<13:54,  4.22it/s]
 48%|████▊     | 3313/6837 [26:51<13:53,  4.23it/s]
 48%|████▊     | 3314/6837 [26:51<13:53,  4.23it/s]
 48%|████▊     | 3315/6837 [26:51<13:51,  4.24it/s]
 49%|████▊     | 3316/6837 [26:51<13:52,  4.23it/s]
 49%|████▊     | 3317/6837 [26:52<13:52,  4.23it/s]
 49%|████▊     | 3318/6837 [26:52<13:50,  4.24it/s]
 49%|████▊     | 3319/6837 [26:52<13:51,  4.23it/s]
 49%|████▊     | 3320/6837 [26:52<13:51,  4.23it/s]
 49%|████▊     | 3321/6837 [26:53<13:50,  4.23it/s]
 49%|████▊     | 3322/6837 [26:53<13:50,  4.23it/s]
 49%|████▊     | 3323/6837 [26:53<13:51,  4.23it/s]
 49%|████▊     | 3324/6837 [26:53<13:50,  4.23it/s]
 49%|████▊     | 3325/6837 [26:54<13:50,  4.23it/s]
                                                   
{'loss': 4.4671, 'grad_norm': 0.13594302535057068, 'learning_rate': 0.003661588081053123, 'epoch': 0.1}
+
 49%|████▊     | 3325/6837 [26:54<13:50,  4.23it/s]
 49%|████▊     | 3326/6837 [26:54<13:52,  4.22it/s]
 49%|████▊     | 3327/6837 [26:54<13:51,  4.22it/s]
 49%|████▊     | 3328/6837 [26:54<13:50,  4.23it/s]
 49%|████▊     | 3329/6837 [26:55<13:49,  4.23it/s]
 49%|████▊     | 3330/6837 [26:55<13:49,  4.23it/s]
 49%|████▊     | 3331/6837 [26:55<13:50,  4.22it/s]
 49%|████▊     | 3332/6837 [26:55<13:49,  4.23it/s]
 49%|████▊     | 3333/6837 [26:55<13:48,  4.23it/s]
 49%|████▉     | 3334/6837 [26:56<13:48,  4.23it/s]
 49%|████▉     | 3335/6837 [26:56<13:48,  4.23it/s]
 49%|████▉     | 3336/6837 [26:56<13:47,  4.23it/s]
 49%|████▉     | 3337/6837 [26:56<13:49,  4.22it/s]
 49%|████▉     | 3338/6837 [26:57<13:49,  4.22it/s]
 49%|████▉     | 3339/6837 [26:57<13:48,  4.22it/s]
 49%|████▉     | 3340/6837 [26:57<13:48,  4.22it/s]
 49%|████▉     | 3341/6837 [26:57<13:48,  4.22it/s]
 49%|████▉     | 3342/6837 [26:58<13:48,  4.22it/s]
 49%|████▉     | 3343/6837 [26:58<13:47,  4.22it/s]
 49%|████▉     | 3344/6837 [26:58<13:47,  4.22it/s]
 49%|████▉     | 3345/6837 [26:58<13:47,  4.22it/s]
 49%|████▉     | 3346/6837 [26:59<13:47,  4.22it/s]
 49%|████▉     | 3347/6837 [26:59<13:47,  4.22it/s]
 49%|████▉     | 3348/6837 [26:59<13:47,  4.22it/s]
 49%|████▉     | 3349/6837 [26:59<13:46,  4.22it/s]
 49%|████▉     | 3350/6837 [26:59<13:45,  4.22it/s]
                                                   
{'loss': 4.4844, 'grad_norm': 0.14646238088607788, 'learning_rate': 0.003624184544605406, 'epoch': 0.1}
+
 49%|████▉     | 3350/6837 [26:59<13:45,  4.22it/s]
 49%|████▉     | 3351/6837 [27:00<13:47,  4.21it/s]
 49%|████▉     | 3352/6837 [27:00<13:47,  4.21it/s]
 49%|████▉     | 3353/6837 [27:00<13:45,  4.22it/s]
 49%|████▉     | 3354/6837 [27:00<13:44,  4.22it/s]
 49%|████▉     | 3355/6837 [27:01<13:42,  4.23it/s]
 49%|████▉     | 3356/6837 [27:01<13:42,  4.23it/s]
 49%|████▉     | 3357/6837 [27:01<13:41,  4.24it/s]
 49%|████▉     | 3358/6837 [27:01<13:41,  4.24it/s]
 49%|████▉     | 3359/6837 [27:02<13:42,  4.23it/s]
 49%|████▉     | 3360/6837 [27:02<13:42,  4.22it/s]
 49%|████▉     | 3361/6837 [27:02<13:42,  4.23it/s]
 49%|████▉     | 3362/6837 [27:02<13:41,  4.23it/s]
 49%|████▉     | 3363/6837 [27:03<13:41,  4.23it/s]
 49%|████▉     | 3364/6837 [27:03<13:40,  4.23it/s]
 49%|████▉     | 3365/6837 [27:03<13:40,  4.23it/s]
 49%|████▉     | 3366/6837 [27:03<13:39,  4.23it/s]
 49%|████▉     | 3367/6837 [27:04<13:41,  4.23it/s]
 49%|████▉     | 3368/6837 [27:04<13:39,  4.23it/s]
 49%|████▉     | 3369/6837 [27:04<13:39,  4.23it/s]
 49%|████▉     | 3370/6837 [27:04<13:41,  4.22it/s]
 49%|████▉     | 3371/6837 [27:04<14:15,  4.05it/s]
 49%|████▉     | 3372/6837 [27:05<14:30,  3.98it/s]
 49%|████▉     | 3373/6837 [27:05<14:44,  3.92it/s]
 49%|████▉     | 3374/6837 [27:05<14:52,  3.88it/s]
 49%|████▉     | 3375/6837 [27:06<15:02,  3.83it/s]
                                                   
{'loss': 4.4763, 'grad_norm': 0.16114091873168945, 'learning_rate': 0.003586679310009079, 'epoch': 0.1}
+
 49%|████▉     | 3375/6837 [27:06<15:02,  3.83it/s]
 49%|████▉     | 3376/6837 [27:06<15:10,  3.80it/s]
 49%|████▉     | 3377/6837 [27:06<15:15,  3.78it/s]
 49%|████▉     | 3378/6837 [27:06<14:49,  3.89it/s]
 49%|████▉     | 3379/6837 [27:07<14:31,  3.97it/s]
 49%|████▉     | 3380/6837 [27:07<14:19,  4.02it/s]
 49%|████▉     | 3381/6837 [27:07<14:06,  4.08it/s]
 49%|████▉     | 3382/6837 [27:07<13:56,  4.13it/s]
 49%|████▉     | 3383/6837 [27:08<13:49,  4.16it/s]
 49%|████▉     | 3384/6837 [27:08<13:45,  4.18it/s]
 50%|████▉     | 3385/6837 [27:08<13:43,  4.19it/s]
 50%|████▉     | 3386/6837 [27:08<13:40,  4.20it/s]
 50%|████▉     | 3387/6837 [27:08<13:43,  4.19it/s]
 50%|████▉     | 3388/6837 [27:09<13:41,  4.20it/s]
 50%|████▉     | 3389/6837 [27:09<13:39,  4.21it/s]
 50%|████▉     | 3390/6837 [27:09<13:37,  4.21it/s]
 50%|████▉     | 3391/6837 [27:09<13:37,  4.22it/s]
 50%|████▉     | 3392/6837 [27:10<13:35,  4.22it/s]
 50%|████▉     | 3393/6837 [27:10<13:34,  4.23it/s]
 50%|████▉     | 3394/6837 [27:10<13:34,  4.23it/s]
 50%|████▉     | 3395/6837 [27:10<13:35,  4.22it/s]
 50%|████▉     | 3396/6837 [27:11<13:34,  4.22it/s]
 50%|████▉     | 3397/6837 [27:11<13:34,  4.22it/s]
 50%|████▉     | 3398/6837 [27:11<13:34,  4.22it/s]
 50%|████▉     | 3399/6837 [27:11<13:33,  4.23it/s]
 50%|████▉     | 3400/6837 [27:12<13:33,  4.23it/s]
                                                   
{'loss': 4.465, 'grad_norm': 0.15125402808189392, 'learning_rate': 0.003549078487977638, 'epoch': 0.1}
+
 50%|████▉     | 3400/6837 [27:12<13:33,  4.23it/s]
 50%|████▉     | 3401/6837 [27:12<13:34,  4.22it/s]
 50%|████▉     | 3402/6837 [27:12<13:34,  4.22it/s]
 50%|████▉     | 3403/6837 [27:12<13:33,  4.22it/s]
 50%|████▉     | 3404/6837 [27:12<13:31,  4.23it/s]
 50%|████▉     | 3405/6837 [27:13<14:03,  4.07it/s]
 50%|████▉     | 3406/6837 [27:13<14:22,  3.98it/s]
 50%|████▉     | 3407/6837 [27:13<14:05,  4.06it/s]
 50%|████▉     | 3408/6837 [27:13<13:53,  4.11it/s]
 50%|████▉     | 3409/6837 [27:14<13:46,  4.15it/s]
 50%|████▉     | 3410/6837 [27:14<13:40,  4.18it/s]
 50%|████▉     | 3411/6837 [27:14<13:36,  4.20it/s]
 50%|████▉     | 3412/6837 [27:14<13:36,  4.20it/s]
 50%|████▉     | 3413/6837 [27:15<13:34,  4.20it/s]
 50%|████▉     | 3414/6837 [27:15<13:31,  4.22it/s]
 50%|████▉     | 3415/6837 [27:15<13:31,  4.22it/s]
 50%|████▉     | 3416/6837 [27:15<13:31,  4.22it/s]
 50%|████▉     | 3417/6837 [27:16<13:30,  4.22it/s]
 50%|████▉     | 3418/6837 [27:16<13:29,  4.22it/s]
 50%|█████     | 3419/6837 [27:16<13:29,  4.22it/s]
 50%|█████     | 3420/6837 [27:16<13:28,  4.23it/s]
 50%|█████     | 3421/6837 [27:17<13:27,  4.23it/s]
 50%|█████     | 3422/6837 [27:17<13:26,  4.24it/s]
 50%|█████     | 3423/6837 [27:17<13:26,  4.23it/s]
 50%|█████     | 3424/6837 [27:17<13:26,  4.23it/s]
 50%|█████     | 3425/6837 [27:18<13:26,  4.23it/s]
                                                   
{'loss': 4.4552, 'grad_norm': 0.14624564349651337, 'learning_rate': 0.0035113882047986054, 'epoch': 0.1}
+
 50%|█████     | 3425/6837 [27:18<13:26,  4.23it/s]
 50%|█████     | 3426/6837 [27:18<13:26,  4.23it/s]
 50%|█████     | 3427/6837 [27:18<13:26,  4.23it/s]
 50%|█████     | 3428/6837 [27:18<13:25,  4.23it/s]
 50%|█████     | 3429/6837 [27:18<13:23,  4.24it/s]
 50%|█████     | 3430/6837 [27:19<13:25,  4.23it/s]
 50%|█████     | 3431/6837 [27:19<13:25,  4.23it/s]
 50%|█████     | 3432/6837 [27:19<13:24,  4.23it/s]
 50%|█████     | 3433/6837 [27:19<13:23,  4.24it/s]
 50%|█████     | 3434/6837 [27:20<13:23,  4.24it/s]
 50%|█████     | 3435/6837 [27:20<13:24,  4.23it/s]
 50%|█████     | 3436/6837 [27:20<13:24,  4.23it/s]
 50%|█████     | 3437/6837 [27:20<13:24,  4.23it/s]
 50%|█████     | 3438/6837 [27:21<13:25,  4.22it/s]
 50%|█████     | 3439/6837 [27:21<13:25,  4.22it/s]
 50%|█████     | 3440/6837 [27:21<13:25,  4.22it/s]
 50%|█████     | 3441/6837 [27:21<13:25,  4.22it/s]
 50%|█████     | 3442/6837 [27:22<13:23,  4.22it/s]
 50%|█████     | 3443/6837 [27:22<13:22,  4.23it/s]
 50%|█████     | 3444/6837 [27:22<13:22,  4.23it/s]
 50%|█████     | 3445/6837 [27:22<13:21,  4.23it/s]
 50%|█████     | 3446/6837 [27:22<13:20,  4.24it/s]
 50%|█████     | 3447/6837 [27:23<13:20,  4.23it/s]
 50%|█████     | 3448/6837 [27:23<13:22,  4.22it/s]
 50%|█████     | 3449/6837 [27:23<13:22,  4.22it/s]
 50%|█████     | 3450/6837 [27:23<13:21,  4.23it/s]
                                                   
{'loss': 4.4562, 'grad_norm': 0.15270282328128815, 'learning_rate': 0.003473614601335374, 'epoch': 0.1}
+
 50%|█████     | 3450/6837 [27:23<13:21,  4.23it/s]
 50%|█████     | 3451/6837 [27:24<13:23,  4.21it/s]
 50%|█████     | 3452/6837 [27:24<13:23,  4.21it/s]
 51%|█████     | 3453/6837 [27:24<13:23,  4.21it/s]
 51%|█████     | 3454/6837 [27:24<13:23,  4.21it/s]
 51%|█████     | 3455/6837 [27:25<13:21,  4.22it/s]
 51%|█████     | 3456/6837 [27:25<13:20,  4.22it/s]
 51%|█████     | 3457/6837 [27:25<13:19,  4.23it/s]
 51%|█████     | 3458/6837 [27:25<13:18,  4.23it/s]
 51%|█████     | 3459/6837 [27:26<13:21,  4.22it/s]
 51%|█████     | 3460/6837 [27:26<13:20,  4.22it/s]
 51%|█████     | 3461/6837 [27:26<13:19,  4.22it/s]
 51%|█████     | 3462/6837 [27:26<13:19,  4.22it/s]
 51%|█████     | 3463/6837 [27:27<13:18,  4.22it/s]
 51%|█████     | 3464/6837 [27:27<13:18,  4.23it/s]
 51%|█████     | 3465/6837 [27:27<13:17,  4.23it/s]
 51%|█████     | 3466/6837 [27:27<13:17,  4.23it/s]
 51%|█████     | 3467/6837 [27:27<13:18,  4.22it/s]
 51%|█████     | 3468/6837 [27:28<13:18,  4.22it/s]
 51%|█████     | 3469/6837 [27:28<13:17,  4.22it/s]
 51%|█████     | 3470/6837 [27:28<13:16,  4.23it/s]
 51%|█████     | 3471/6837 [27:28<13:16,  4.23it/s]
 51%|█████     | 3472/6837 [27:29<13:15,  4.23it/s]
 51%|█████     | 3473/6837 [27:29<13:16,  4.23it/s]
 51%|█████     | 3474/6837 [27:29<13:16,  4.22it/s]
 51%|█████     | 3475/6837 [27:29<13:16,  4.22it/s]
                                                   
{'loss': 4.4428, 'grad_norm': 0.14836426079273224, 'learning_rate': 0.0034357638320266773, 'epoch': 0.1}
+
 51%|█████     | 3475/6837 [27:29<13:16,  4.22it/s]
 51%|█████     | 3476/6837 [27:30<13:18,  4.21it/s]
 51%|█████     | 3477/6837 [27:30<13:18,  4.21it/s]
 51%|█████     | 3478/6837 [27:30<13:16,  4.22it/s]
 51%|█████     | 3479/6837 [27:30<13:15,  4.22it/s]
 51%|█████     | 3480/6837 [27:31<13:14,  4.22it/s]
 51%|█████     | 3481/6837 [27:31<13:15,  4.22it/s]
 51%|█████     | 3482/6837 [27:31<13:14,  4.22it/s]
 51%|█████     | 3483/6837 [27:31<13:13,  4.23it/s]
 51%|█████     | 3484/6837 [27:31<13:13,  4.23it/s]
 51%|█████     | 3485/6837 [27:32<13:13,  4.23it/s]
 51%|█████     | 3486/6837 [27:32<13:13,  4.22it/s]
 51%|█████     | 3487/6837 [27:32<13:12,  4.22it/s]
 51%|█████     | 3488/6837 [27:32<13:14,  4.22it/s]
 51%|█████     | 3489/6837 [27:33<13:12,  4.22it/s]
 51%|█████     | 3490/6837 [27:33<13:12,  4.22it/s]
 51%|█████     | 3491/6837 [27:33<13:11,  4.23it/s]
 51%|█████     | 3492/6837 [27:33<13:11,  4.23it/s]
 51%|█████     | 3493/6837 [27:34<13:11,  4.22it/s]
 51%|█████     | 3494/6837 [27:34<13:11,  4.22it/s]
 51%|█████     | 3495/6837 [27:34<13:13,  4.21it/s]
 51%|█████     | 3496/6837 [27:34<13:13,  4.21it/s]
 51%|█████     | 3497/6837 [27:35<13:12,  4.21it/s]
 51%|█████     | 3498/6837 [27:35<13:11,  4.22it/s]
 51%|█████     | 3499/6837 [27:35<13:09,  4.23it/s]
 51%|█████     | 3500/6837 [27:35<13:09,  4.23it/s]
                                                   
{'loss': 4.453, 'grad_norm': 0.1503460854291916, 'learning_rate': 0.0033978420638838503, 'epoch': 0.1}
+
 51%|█████     | 3500/6837 [27:35<13:09,  4.23it/s]
 51%|█████     | 3501/6837 [27:36<13:11,  4.22it/s]
 51%|█████     | 3502/6837 [27:36<13:13,  4.20it/s]
 51%|█████     | 3503/6837 [27:36<13:13,  4.20it/s]
 51%|█████▏    | 3504/6837 [27:36<13:11,  4.21it/s]
 51%|█████▏    | 3505/6837 [27:36<13:10,  4.21it/s]
 51%|█████▏    | 3506/6837 [27:37<13:09,  4.22it/s]
 51%|█████▏    | 3507/6837 [27:37<13:07,  4.23it/s]
 51%|█████▏    | 3508/6837 [27:37<13:06,  4.23it/s]
 51%|█████▏    | 3509/6837 [27:37<13:07,  4.23it/s]
 51%|█████▏    | 3510/6837 [27:38<17:44,  3.12it/s]
 51%|█████▏    | 3511/6837 [27:38<16:20,  3.39it/s]
 51%|█████▏    | 3512/6837 [27:38<15:25,  3.59it/s]
 51%|█████▏    | 3513/6837 [27:39<14:42,  3.77it/s]
 51%|█████▏    | 3514/6837 [27:39<14:13,  3.89it/s]
 51%|█████▏    | 3515/6837 [27:39<13:53,  3.99it/s]
 51%|█████▏    | 3516/6837 [27:39<13:37,  4.06it/s]
 51%|█████▏    | 3517/6837 [27:40<13:27,  4.11it/s]
 51%|█████▏    | 3518/6837 [27:40<13:19,  4.15it/s]
 51%|█████▏    | 3519/6837 [27:40<13:16,  4.16it/s]
 51%|█████▏    | 3520/6837 [27:40<13:13,  4.18it/s]
 51%|█████▏    | 3521/6837 [27:41<13:09,  4.20it/s]
 52%|█████▏    | 3522/6837 [27:41<13:07,  4.21it/s]
 52%|█████▏    | 3523/6837 [27:41<13:07,  4.21it/s]
 52%|█████▏    | 3524/6837 [27:41<13:05,  4.22it/s]
 52%|█████▏    | 3525/6837 [27:41<13:02,  4.23it/s]
                                                   
{'loss': 4.4278, 'grad_norm': 0.14883722364902496, 'learning_rate': 0.0033598554754860435, 'epoch': 0.1}
+
 52%|█████▏    | 3525/6837 [27:41<13:02,  4.23it/s]
 52%|█████▏    | 3526/6837 [27:42<13:04,  4.22it/s]
 52%|█████▏    | 3527/6837 [27:42<13:03,  4.23it/s]
 52%|█████▏    | 3528/6837 [27:42<13:01,  4.23it/s]
 52%|█████▏    | 3529/6837 [27:42<13:01,  4.23it/s]
 52%|█████▏    | 3530/6837 [27:43<13:00,  4.24it/s]
 52%|█████▏    | 3531/6837 [27:43<13:29,  4.08it/s]
 52%|█████▏    | 3532/6837 [27:43<13:50,  3.98it/s]
 52%|█████▏    | 3533/6837 [27:43<14:08,  3.89it/s]
 52%|█████▏    | 3534/6837 [27:44<14:15,  3.86it/s]
 52%|█████▏    | 3535/6837 [27:44<14:16,  3.86it/s]
 52%|█████▏    | 3536/6837 [27:44<13:53,  3.96it/s]
 52%|█████▏    | 3537/6837 [27:44<13:37,  4.04it/s]
 52%|█████▏    | 3538/6837 [27:45<13:25,  4.09it/s]
 52%|█████▏    | 3539/6837 [27:45<13:17,  4.13it/s]
 52%|█████▏    | 3540/6837 [27:45<13:13,  4.16it/s]
 52%|█████▏    | 3541/6837 [27:45<13:09,  4.17it/s]
 52%|█████▏    | 3542/6837 [27:46<13:06,  4.19it/s]
 52%|█████▏    | 3543/6837 [27:46<13:04,  4.20it/s]
 52%|█████▏    | 3544/6837 [27:46<13:04,  4.20it/s]
 52%|█████▏    | 3545/6837 [27:46<13:03,  4.20it/s]
 52%|█████▏    | 3546/6837 [27:47<13:01,  4.21it/s]
 52%|█████▏    | 3547/6837 [27:47<13:00,  4.22it/s]
 52%|█████▏    | 3548/6837 [27:47<13:02,  4.21it/s]
 52%|█████▏    | 3549/6837 [27:47<13:01,  4.21it/s]
 52%|█████▏    | 3550/6837 [27:48<13:00,  4.21it/s]{'loss': 4.4452, 'grad_norm': 0.1521454006433487, 'learning_rate': 0.003321810255973543, 'epoch': 0.1}
                                                   
+
 52%|█████▏    | 3550/6837 [27:48<13:00,  4.21it/s]
 52%|█████▏    | 3551/6837 [27:48<13:01,  4.20it/s]
 52%|█████▏    | 3552/6837 [27:48<13:01,  4.21it/s]
 52%|█████▏    | 3553/6837 [27:48<12:59,  4.21it/s]
 52%|█████▏    | 3554/6837 [27:48<12:58,  4.22it/s]
 52%|█████▏    | 3555/6837 [27:49<12:57,  4.22it/s]
 52%|█████▏    | 3556/6837 [27:49<12:58,  4.22it/s]
 52%|█████▏    | 3557/6837 [27:49<12:56,  4.23it/s]
 52%|█████▏    | 3558/6837 [27:49<12:54,  4.23it/s]
 52%|█████▏    | 3559/6837 [27:50<12:55,  4.23it/s]
 52%|█████▏    | 3560/6837 [27:50<12:55,  4.23it/s]
 52%|█████▏    | 3561/6837 [27:50<12:54,  4.23it/s]
 52%|█████▏    | 3562/6837 [27:50<12:54,  4.23it/s]
 52%|█████▏    | 3563/6837 [27:51<12:54,  4.23it/s]
 52%|█████▏    | 3564/6837 [27:51<12:54,  4.23it/s]
 52%|█████▏    | 3565/6837 [27:51<12:53,  4.23it/s]
 52%|█████▏    | 3566/6837 [27:51<12:56,  4.22it/s]
 52%|█████▏    | 3567/6837 [27:52<12:54,  4.22it/s]
 52%|█████▏    | 3568/6837 [27:52<12:53,  4.22it/s]
 52%|█████▏    | 3569/6837 [27:52<12:52,  4.23it/s]
 52%|█████▏    | 3570/6837 [27:52<12:52,  4.23it/s]
 52%|█████▏    | 3571/6837 [27:53<12:52,  4.23it/s]
 52%|█████▏    | 3572/6837 [27:53<12:51,  4.23it/s]
 52%|█████▏    | 3573/6837 [27:53<12:51,  4.23it/s]
 52%|█████▏    | 3574/6837 [27:53<12:50,  4.23it/s]
 52%|█████▏    | 3575/6837 [27:53<12:50,  4.23it/s]
                                                   
{'loss': 4.4529, 'grad_norm': 0.1537385731935501, 'learning_rate': 0.003283712604039382, 'epoch': 0.1}
+
 52%|█████▏    | 3575/6837 [27:53<12:50,  4.23it/s]
 52%|█████▏    | 3576/6837 [27:54<12:52,  4.22it/s]
 52%|█████▏    | 3577/6837 [27:54<12:51,  4.23it/s]
 52%|█████▏    | 3578/6837 [27:54<12:50,  4.23it/s]
 52%|█████▏    | 3579/6837 [27:54<12:49,  4.24it/s]
 52%|█████▏    | 3580/6837 [27:55<12:48,  4.24it/s]
 52%|█████▏    | 3581/6837 [27:55<12:48,  4.23it/s]
 52%|█████▏    | 3582/6837 [27:55<12:48,  4.24it/s]
 52%|█████▏    | 3583/6837 [27:55<12:47,  4.24it/s]
 52%|█████▏    | 3584/6837 [27:56<12:49,  4.23it/s]
 52%|█████▏    | 3585/6837 [27:56<12:49,  4.22it/s]
 52%|█████▏    | 3586/6837 [27:56<12:48,  4.23it/s]
 52%|█████▏    | 3587/6837 [27:56<12:48,  4.23it/s]
 52%|█████▏    | 3588/6837 [27:57<12:48,  4.23it/s]
 52%|█████▏    | 3589/6837 [27:57<12:48,  4.22it/s]
 53%|█████▎    | 3590/6837 [27:57<12:47,  4.23it/s]
 53%|█████▎    | 3591/6837 [27:57<12:47,  4.23it/s]
 53%|█████▎    | 3592/6837 [27:57<12:47,  4.23it/s]
 53%|█████▎    | 3593/6837 [27:58<12:46,  4.23it/s]
 53%|█████▎    | 3594/6837 [27:58<12:47,  4.23it/s]
 53%|█████▎    | 3595/6837 [27:58<12:47,  4.22it/s]
 53%|█████▎    | 3596/6837 [27:58<12:46,  4.23it/s]
 53%|█████▎    | 3597/6837 [27:59<12:46,  4.23it/s]
 53%|█████▎    | 3598/6837 [27:59<12:45,  4.23it/s]
 53%|█████▎    | 3599/6837 [27:59<12:45,  4.23it/s]
 53%|█████▎    | 3600/6837 [27:59<12:45,  4.23it/s]
                                                   
{'loss': 4.4419, 'grad_norm': 0.14868900179862976, 'learning_rate': 0.00324556872691939, 'epoch': 0.11}
+
 53%|█████▎    | 3600/6837 [27:59<12:45,  4.23it/s]
 53%|█████▎    | 3601/6837 [28:00<12:47,  4.22it/s]
 53%|█████▎    | 3602/6837 [28:00<12:46,  4.22it/s]
 53%|█████▎    | 3603/6837 [28:00<12:45,  4.22it/s]
 53%|█████▎    | 3604/6837 [28:00<12:44,  4.23it/s]
 53%|█████▎    | 3605/6837 [28:01<12:46,  4.22it/s]
 53%|█████▎    | 3606/6837 [28:01<12:45,  4.22it/s]
 53%|█████▎    | 3607/6837 [28:01<12:45,  4.22it/s]
 53%|█████▎    | 3608/6837 [28:01<12:44,  4.23it/s]
 53%|█████▎    | 3609/6837 [28:01<12:43,  4.23it/s]
 53%|█████▎    | 3610/6837 [28:02<12:44,  4.22it/s]
 53%|█████▎    | 3611/6837 [28:02<12:42,  4.23it/s]
 53%|█████▎    | 3612/6837 [28:02<12:42,  4.23it/s]
 53%|█████▎    | 3613/6837 [28:02<12:44,  4.22it/s]
 53%|█████▎    | 3614/6837 [28:03<12:43,  4.22it/s]
 53%|█��███▎    | 3615/6837 [28:03<12:42,  4.23it/s]
 53%|█████▎    | 3616/6837 [28:03<12:42,  4.22it/s]
 53%|█████▎    | 3617/6837 [28:03<12:41,  4.23it/s]
 53%|█████▎    | 3618/6837 [28:04<12:41,  4.23it/s]
 53%|█████▎    | 3619/6837 [28:04<12:40,  4.23it/s]
 53%|█████▎    | 3620/6837 [28:04<12:40,  4.23it/s]
 53%|█████▎    | 3621/6837 [28:04<12:40,  4.23it/s]
 53%|█████▎    | 3622/6837 [28:05<12:40,  4.23it/s]
 53%|█████▎    | 3623/6837 [28:05<12:40,  4.23it/s]
 53%|█████▎    | 3624/6837 [28:05<12:42,  4.22it/s]
 53%|█████▎    | 3625/6837 [28:05<12:40,  4.22it/s]{'loss': 4.4441, 'grad_norm': 0.15140162408351898, 'learning_rate': 0.0032073848393808424, 'epoch': 0.11}
+
                                                   

 53%|█████▎    | 3625/6837 [28:05<12:40,  4.22it/s]
 53%|█████▎    | 3626/6837 [28:06<12:43,  4.20it/s]
 53%|█████▎    | 3627/6837 [28:06<12:45,  4.19it/s]
 53%|█████▎    | 3628/6837 [28:06<12:47,  4.18it/s]
 53%|█████▎    | 3629/6837 [28:06<12:46,  4.18it/s]
 53%|█████▎    | 3630/6837 [28:06<12:43,  4.20it/s]
 53%|█████▎    | 3631/6837 [28:07<12:42,  4.20it/s]
 53%|█████▎    | 3632/6837 [28:07<12:40,  4.22it/s]
 53%|█████▎    | 3633/6837 [28:07<12:38,  4.23it/s]
 53%|█████▎    | 3634/6837 [28:07<12:37,  4.23it/s]
 53%|█████▎    | 3635/6837 [28:08<12:38,  4.22it/s]
 53%|█████▎    | 3636/6837 [28:08<12:38,  4.22it/s]
 53%|█████▎    | 3637/6837 [28:08<12:41,  4.20it/s]
 53%|█████▎    | 3638/6837 [28:08<12:43,  4.19it/s]
 53%|█████▎    | 3639/6837 [28:09<12:43,  4.19it/s]
 53%|█████▎    | 3640/6837 [28:09<12:40,  4.21it/s]
 53%|█████▎    | 3641/6837 [28:09<12:38,  4.22it/s]
 53%|█████▎    | 3642/6837 [28:09<12:37,  4.22it/s]
 53%|█████▎    | 3643/6837 [28:10<12:36,  4.22it/s]
 53%|█████▎    | 3644/6837 [28:10<12:36,  4.22it/s]
 53%|█████▎    | 3645/6837 [28:10<12:37,  4.21it/s]
 53%|█████▎    | 3646/6837 [28:10<12:38,  4.21it/s]
 53%|█████▎    | 3647/6837 [28:11<12:40,  4.20it/s]
 53%|█████▎    | 3648/6837 [28:11<12:42,  4.18it/s]
 53%|█████▎    | 3649/6837 [28:11<12:43,  4.18it/s]
 53%|█████▎    | 3650/6837 [28:11<12:40,  4.19it/s]
                                                   
{'loss': 4.439, 'grad_norm': 0.13878877460956573, 'learning_rate': 0.003169167162709903, 'epoch': 0.11}
+
 53%|█████▎    | 3650/6837 [28:11<12:40,  4.19it/s]
 53%|█████▎    | 3651/6837 [28:11<12:39,  4.19it/s]
 53%|█████▎    | 3652/6837 [28:12<12:38,  4.20it/s]
 53%|█████▎    | 3653/6837 [28:12<12:36,  4.21it/s]
 53%|█████▎    | 3654/6837 [28:12<12:35,  4.21it/s]
 53%|█████▎    | 3655/6837 [28:12<12:35,  4.21it/s]
 53%|█████▎    | 3656/6837 [28:13<12:35,  4.21it/s]
 53%|█████▎    | 3657/6837 [28:13<12:56,  4.09it/s]
 54%|█████▎    | 3658/6837 [28:13<13:13,  4.01it/s]
 54%|█████▎    | 3659/6837 [28:13<13:28,  3.93it/s]
 54%|█████▎    | 3660/6837 [28:14<13:39,  3.88it/s]
 54%|█████▎    | 3661/6837 [28:14<13:50,  3.82it/s]
 54%|█████▎    | 3662/6837 [28:14<13:52,  3.81it/s]
 54%|█████▎    | 3663/6837 [28:15<13:56,  3.80it/s]
 54%|█████▎    | 3664/6837 [28:15<13:58,  3.79it/s]
 54%|█████▎    | 3665/6837 [28:15<14:00,  3.77it/s]
 54%|█████▎    | 3666/6837 [28:15<13:58,  3.78it/s]
 54%|█████▎    | 3667/6837 [28:16<13:55,  3.79it/s]
 54%|█████▎    | 3668/6837 [28:16<13:55,  3.79it/s]
 54%|█████▎    | 3669/6837 [28:16<13:58,  3.78it/s]
 54%|█████▎    | 3670/6837 [28:16<13:36,  3.88it/s]
 54%|█████▎    | 3671/6837 [28:17<13:15,  3.98it/s]
 54%|█████▎    | 3672/6837 [28:17<13:03,  4.04it/s]
 54%|█████▎    | 3673/6837 [28:17<12:51,  4.10it/s]
 54%|█████▎    | 3674/6837 [28:17<12:44,  4.14it/s]
 54%|█████▍    | 3675/6837 [28:18<12:39,  4.16it/s]
                                                   
{'loss': 4.4463, 'grad_norm': 0.15668663382530212, 'learning_rate': 0.003130921923697983, 'epoch': 0.11}
+
 54%|█████▍    | 3675/6837 [28:18<12:39,  4.16it/s]
 54%|█████▍    | 3676/6837 [28:18<12:39,  4.16it/s]
 54%|█████▍    | 3677/6837 [28:18<12:34,  4.19it/s]
 54%|█████▍    | 3678/6837 [28:18<12:32,  4.20it/s]
 54%|█████▍    | 3679/6837 [28:18<12:29,  4.22it/s]
 54%|█████▍    | 3680/6837 [28:19<12:26,  4.23it/s]
 54%|█████▍    | 3681/6837 [28:19<12:25,  4.23it/s]
 54%|█████▍    | 3682/6837 [28:19<12:25,  4.23it/s]
 54%|█████▍    | 3683/6837 [28:19<12:24,  4.24it/s]
 54%|█████▍    | 3684/6837 [28:20<12:23,  4.24it/s]
 54%|█████▍    | 3685/6837 [28:20<12:22,  4.24it/s]
 54%|█████▍    | 3686/6837 [28:20<12:21,  4.25it/s]
 54%|█████▍    | 3687/6837 [28:20<12:21,  4.25it/s]
 54%|█████▍    | 3688/6837 [28:21<12:22,  4.24it/s]
 54%|█████▍    | 3689/6837 [28:21<12:21,  4.25it/s]
 54%|█████▍    | 3690/6837 [28:21<12:21,  4.24it/s]
 54%|█████▍    | 3691/6837 [28:21<12:22,  4.24it/s]
 54%|█████▍    | 3692/6837 [28:22<12:22,  4.24it/s]
 54%|█████▍    | 3693/6837 [28:22<12:21,  4.24it/s]
 54%|█████▍    | 3694/6837 [28:22<12:21,  4.24it/s]
 54%|█████▍    | 3695/6837 [28:22<12:20,  4.24it/s]
 54%|█████▍    | 3696/6837 [28:22<12:19,  4.25it/s]
 54%|█████▍    | 3697/6837 [28:23<12:18,  4.25it/s]
 54%|█████▍    | 3698/6837 [28:23<12:49,  4.08it/s]
 54%|█████▍    | 3699/6837 [28:23<13:06,  3.99it/s]
 54%|█████▍    | 3700/6837 [28:24<13:20,  3.92it/s]
                                                   
{'loss': 4.4368, 'grad_norm': 0.1446448713541031, 'learning_rate': 0.003092655353627216, 'epoch': 0.11}
+
 54%|█████▍    | 3700/6837 [28:24<13:20,  3.92it/s]
 54%|█████▍    | 3701/6837 [28:24<13:32,  3.86it/s]
 54%|█████▍    | 3702/6837 [28:24<13:10,  3.97it/s]
 54%|█████▍    | 3703/6837 [28:24<12:55,  4.04it/s]
 54%|█████▍    | 3704/6837 [28:24<12:46,  4.09it/s]
 54%|█████▍    | 3705/6837 [28:25<12:38,  4.13it/s]
 54%|█████▍    | 3706/6837 [28:25<12:34,  4.15it/s]
 54%|█████▍    | 3707/6837 [28:25<12:29,  4.18it/s]
 54%|█████▍    | 3708/6837 [28:25<12:26,  4.19it/s]
 54%|█████▍    | 3709/6837 [28:26<12:25,  4.19it/s]
 54%|█████▍    | 3710/6837 [28:26<12:25,  4.20it/s]
 54%|█████▍    | 3711/6837 [28:26<12:24,  4.20it/s]
 54%|█████▍    | 3712/6837 [28:26<12:22,  4.21it/s]
 54%|█████▍    | 3713/6837 [28:27<12:21,  4.21it/s]
 54%|█████▍    | 3714/6837 [28:27<12:20,  4.22it/s]
 54%|█████▍    | 3715/6837 [28:27<12:19,  4.22it/s]
 54%|█████▍    | 3716/6837 [28:27<12:19,  4.22it/s]
 54%|█████▍    | 3717/6837 [28:28<12:17,  4.23it/s]
 54%|█████▍    | 3718/6837 [28:28<12:17,  4.23it/s]
 54%|█████▍    | 3719/6837 [28:28<12:18,  4.22it/s]
 54%|█████▍    | 3720/6837 [28:28<12:18,  4.22it/s]
 54%|█████▍    | 3721/6837 [28:29<12:17,  4.23it/s]
 54%|█████▍    | 3722/6837 [28:29<12:16,  4.23it/s]
 54%|█████▍    | 3723/6837 [28:29<12:16,  4.23it/s]
 54%|█████▍    | 3724/6837 [28:29<12:14,  4.24it/s]
 54%|█████▍    | 3725/6837 [28:29<12:14,  4.23it/s]
                                                   
{'loss': 4.4277, 'grad_norm': 0.15262754261493683, 'learning_rate': 0.003054373687255196, 'epoch': 0.11}
+
 54%|█████▍    | 3725/6837 [28:29<12:14,  4.23it/s]
 54%|█████▍    | 3726/6837 [28:30<12:17,  4.22it/s]
 55%|█████▍    | 3727/6837 [28:30<12:17,  4.22it/s]
 55%|█████▍    | 3728/6837 [28:30<12:15,  4.23it/s]
 55%|█████▍    | 3729/6837 [28:30<12:15,  4.23it/s]
 55%|█████▍    | 3730/6837 [28:31<12:16,  4.22it/s]
 55%|█████▍    | 3731/6837 [28:31<12:14,  4.23it/s]
 55%|█████▍    | 3732/6837 [28:31<12:13,  4.23it/s]
 55%|█████▍    | 3733/6837 [28:31<12:14,  4.23it/s]
 55%|█████▍    | 3734/6837 [28:32<12:13,  4.23it/s]
 55%|█████▍    | 3735/6837 [28:32<12:12,  4.23it/s]
 55%|█████▍    | 3736/6837 [28:32<12:12,  4.23it/s]
 55%|█████▍    | 3737/6837 [28:32<12:12,  4.23it/s]
 55%|█████▍    | 3738/6837 [28:33<12:11,  4.24it/s]
 55%|█████▍    | 3739/6837 [28:33<12:11,  4.23it/s]
 55%|█████▍    | 3740/6837 [28:33<12:12,  4.23it/s]
 55%|█████▍    | 3741/6837 [28:33<12:12,  4.23it/s]
 55%|█████▍    | 3742/6837 [28:33<12:11,  4.23it/s]
 55%|█████▍    | 3743/6837 [28:34<12:11,  4.23it/s]
 55%|█████▍    | 3744/6837 [28:34<12:11,  4.23it/s]
 55%|█████▍    | 3745/6837 [28:34<12:11,  4.23it/s]
 55%|█████▍    | 3746/6837 [28:34<12:11,  4.23it/s]
 55%|█████▍    | 3747/6837 [28:35<12:11,  4.22it/s]
 55%|█████▍    | 3748/6837 [28:35<12:11,  4.22it/s]
 55%|█████▍    | 3749/6837 [28:35<12:11,  4.22it/s]
 55%|█████▍    | 3750/6837 [28:35<12:11,  4.22it/s]
                                                   
{'loss': 4.4425, 'grad_norm': 0.15154211223125458, 'learning_rate': 0.003016083161799152, 'epoch': 0.11}
+
 55%|█████▍    | 3750/6837 [28:35<12:11,  4.22it/s]
 55%|█████▍    | 3751/6837 [28:36<12:13,  4.21it/s]
 55%|█████▍    | 3752/6837 [28:36<12:11,  4.22it/s]
 55%|█████▍    | 3753/6837 [28:36<12:09,  4.23it/s]
 55%|█████▍    | 3754/6837 [28:36<12:09,  4.23it/s]
 55%|█████▍    | 3755/6837 [28:37<12:08,  4.23it/s]
 55%|█████▍    | 3756/6837 [28:37<12:07,  4.23it/s]
 55%|█████▍    | 3757/6837 [28:37<12:08,  4.23it/s]
 55%|█████▍    | 3758/6837 [28:37<12:14,  4.19it/s]
 55%|█████▍    | 3759/6837 [28:38<12:12,  4.20it/s]
 55%|█████▍    | 3760/6837 [28:38<12:10,  4.21it/s]
 55%|█████▌    | 3761/6837 [28:38<12:09,  4.22it/s]
 55%|█████▌    | 3762/6837 [28:38<12:09,  4.22it/s]
 55%|█████▌    | 3763/6837 [28:38<12:07,  4.22it/s]
 55%|█████▌    | 3764/6837 [28:39<12:06,  4.23it/s]
 55%|█████▌    | 3765/6837 [28:39<12:06,  4.23it/s]
 55%|█████▌    | 3766/6837 [28:39<12:05,  4.23it/s]
 55%|█████▌    | 3767/6837 [28:39<12:05,  4.23it/s]
 55%|█████▌    | 3768/6837 [28:40<12:05,  4.23it/s]
 55%|█████▌    | 3769/6837 [28:40<12:06,  4.22it/s]
 55%|█████▌    | 3770/6837 [28:40<12:04,  4.23it/s]
 55%|█████▌    | 3771/6837 [28:40<12:05,  4.23it/s]
 55%|█████▌    | 3772/6837 [28:41<12:04,  4.23it/s]
 55%|█████▌    | 3773/6837 [28:41<12:04,  4.23it/s]
 55%|█████▌    | 3774/6837 [28:41<12:04,  4.22it/s]
 55%|█████▌    | 3775/6837 [28:41<12:03,  4.23it/s]
                                                   
{'loss': 4.4304, 'grad_norm': 0.17801694571971893, 'learning_rate': 0.0029777900159197205, 'epoch': 0.11}
+
 55%|█████▌    | 3775/6837 [28:41<12:03,  4.23it/s]
 55%|█████▌    | 3776/6837 [28:42<12:05,  4.22it/s]
 55%|█████▌    | 3777/6837 [28:42<12:05,  4.22it/s]
 55%|█████▌    | 3778/6837 [28:42<12:04,  4.22it/s]
 55%|█████▌    | 3779/6837 [28:42<12:04,  4.22it/s]
 55%|█████▌    | 3780/6837 [28:42<12:05,  4.21it/s]
 55%|█████▌    | 3781/6837 [28:43<12:29,  4.08it/s]
 55%|█████▌    | 3782/6837 [28:43<12:45,  3.99it/s]
 55%|█████▌    | 3783/6837 [28:43<12:34,  4.05it/s]
 55%|█████▌    | 3784/6837 [28:44<12:52,  3.95it/s]
 55%|█████▌    | 3785/6837 [28:44<12:58,  3.92it/s]
 55%|█████▌    | 3786/6837 [28:44<12:56,  3.93it/s]
 55%|█████▌    | 3787/6837 [28:44<12:40,  4.01it/s]
 55%|█████▌    | 3788/6837 [28:44<12:29,  4.07it/s]
 55%|█████▌    | 3789/6837 [28:45<12:20,  4.12it/s]
 55%|█████▌    | 3790/6837 [28:45<12:15,  4.14it/s]
 55%|█████▌    | 3791/6837 [28:45<12:11,  4.16it/s]
 55%|█████▌    | 3792/6837 [28:45<12:08,  4.18it/s]
 55%|█████▌    | 3793/6837 [28:46<12:06,  4.19it/s]
 55%|█████▌    | 3794/6837 [28:46<12:04,  4.20it/s]
 56%|█████▌    | 3795/6837 [28:46<12:02,  4.21it/s]
 56%|█████▌    | 3796/6837 [28:46<12:00,  4.22it/s]
 56%|█████▌    | 3797/6837 [28:47<12:01,  4.21it/s]
 56%|█████▌    | 3798/6837 [28:47<12:00,  4.22it/s]
 56%|█████▌    | 3799/6837 [28:47<11:59,  4.22it/s]
 56%|█████▌    | 3800/6837 [28:47<11:59,  4.22it/s]
                                                   
{'loss': 4.4191, 'grad_norm': 0.15948429703712463, 'learning_rate': 0.002939500488704482, 'epoch': 0.11}
+
 56%|█████▌    | 3800/6837 [28:47<11:59,  4.22it/s]
 56%|█████▌    | 3801/6837 [28:48<12:00,  4.22it/s]
 56%|█████▌    | 3802/6837 [28:48<11:59,  4.22it/s]
 56%|█████▌    | 3803/6837 [28:48<11:58,  4.22it/s]
 56%|█████▌    | 3804/6837 [28:48<11:58,  4.22it/s]
 56%|█████▌    | 3805/6837 [28:49<11:59,  4.22it/s]
 56%|█████▌    | 3806/6837 [28:49<11:58,  4.22it/s]
 56%|█████▌    | 3807/6837 [28:49<11:58,  4.21it/s]
 56%|█████▌    | 3808/6837 [28:49<11:58,  4.21it/s]
 56%|█████▌    | 3809/6837 [28:49<11:58,  4.21it/s]
 56%|█████▌    | 3810/6837 [28:50<11:57,  4.22it/s]
 56%|█████▌    | 3811/6837 [28:50<11:56,  4.22it/s]
 56%|█████▌    | 3812/6837 [28:50<11:56,  4.22it/s]
 56%|█████▌    | 3813/6837 [28:50<11:55,  4.22it/s]
 56%|█████▌    | 3814/6837 [28:51<11:54,  4.23it/s]
 56%|█████▌    | 3815/6837 [28:51<11:54,  4.23it/s]
 56%|█████▌    | 3816/6837 [28:51<11:54,  4.23it/s]
 56%|█████▌    | 3817/6837 [28:51<11:53,  4.23it/s]
 56%|█████▌    | 3818/6837 [28:52<11:53,  4.23it/s]
 56%|█████▌    | 3819/6837 [28:52<11:53,  4.23it/s]
 56%|█████▌    | 3820/6837 [28:52<11:52,  4.23it/s]
 56%|█████▌    | 3821/6837 [28:52<11:52,  4.23it/s]
 56%|█████▌    | 3822/6837 [28:53<11:53,  4.22it/s]
 56%|█████▌    | 3823/6837 [28:53<11:52,  4.23it/s]
 56%|█████▌    | 3824/6837 [28:53<11:52,  4.23it/s]
 56%|█████▌    | 3825/6837 [28:53<11:52,  4.23it/s]
                                                   
{'loss': 4.4127, 'grad_norm': 0.14102542400360107, 'learning_rate': 0.00290122081865143, 'epoch': 0.11}
+
 56%|█████▌    | 3825/6837 [28:53<11:52,  4.23it/s]
 56%|█████▌    | 3826/6837 [28:53<11:54,  4.21it/s]
 56%|█████▌    | 3827/6837 [28:54<11:53,  4.22it/s]
 56%|█████▌    | 3828/6837 [28:54<11:53,  4.22it/s]
 56%|█████▌    | 3829/6837 [28:54<11:53,  4.22it/s]
 56%|█████▌    | 3830/6837 [28:54<11:53,  4.22it/s]
 56%|█████▌    | 3831/6837 [28:55<11:52,  4.22it/s]
 56%|█████▌    | 3832/6837 [28:55<11:51,  4.22it/s]
 56%|█████▌    | 3833/6837 [28:55<11:52,  4.22it/s]
 56%|█████▌    | 3834/6837 [28:55<11:50,  4.22it/s]
 56%|█████▌    | 3835/6837 [28:56<11:49,  4.23it/s]
 56%|█████▌    | 3836/6837 [28:56<11:49,  4.23it/s]
 56%|█████▌    | 3837/6837 [28:56<11:50,  4.22it/s]
 56%|█████▌    | 3838/6837 [28:56<11:50,  4.22it/s]
 56%|█████▌    | 3839/6837 [28:57<11:51,  4.21it/s]
 56%|█████▌    | 3840/6837 [28:57<11:50,  4.22it/s]
 56%|█████▌    | 3841/6837 [28:57<11:50,  4.22it/s]
 56%|█████▌    | 3842/6837 [28:57<11:50,  4.22it/s]
 56%|█████▌    | 3843/6837 [28:58<11:49,  4.22it/s]
 56%|█████▌    | 3844/6837 [28:58<11:50,  4.21it/s]
 56%|█████▌    | 3845/6837 [28:58<11:50,  4.21it/s]
 56%|█████▋    | 3846/6837 [28:58<11:48,  4.22it/s]
 56%|█████▋    | 3847/6837 [28:58<11:50,  4.21it/s]
 56%|█████▋    | 3848/6837 [28:59<11:49,  4.21it/s]
 56%|█████▋    | 3849/6837 [28:59<11:48,  4.22it/s]
 56%|█████▋    | 3850/6837 [28:59<11:48,  4.22it/s]
                                                   
{'loss': 4.4236, 'grad_norm': 0.1711367964744568, 'learning_rate': 0.0028629572426525354, 'epoch': 0.11}
+
 56%|█████▋    | 3850/6837 [28:59<11:48,  4.22it/s]
 56%|█████▋    | 3851/6837 [28:59<12:16,  4.06it/s]
 56%|█████▋    | 3852/6837 [29:00<12:17,  4.05it/s]
 56%|█████▋    | 3853/6837 [29:00<12:07,  4.10it/s]
 56%|█████▋    | 3854/6837 [29:00<12:00,  4.14it/s]
 56%|█████▋    | 3855/6837 [29:00<11:55,  4.17it/s]
 56%|█████▋    | 3856/6837 [29:01<11:51,  4.19it/s]
 56%|█████▋    | 3857/6837 [29:01<11:49,  4.20it/s]
 56%|█████▋    | 3858/6837 [29:01<11:48,  4.21it/s]
 56%|█████▋    | 3859/6837 [29:01<11:46,  4.21it/s]
 56%|█████▋    | 3860/6837 [29:02<11:44,  4.22it/s]
 56%|█████▋    | 3861/6837 [29:02<11:44,  4.22it/s]
 56%|█████▋    | 3862/6837 [29:02<11:44,  4.23it/s]
 57%|█████▋    | 3863/6837 [29:02<11:42,  4.23it/s]
 57%|█████▋    | 3864/6837 [29:03<11:42,  4.23it/s]
 57%|█████▋    | 3865/6837 [29:03<11:41,  4.24it/s]
 57%|█████▋    | 3866/6837 [29:03<11:40,  4.24it/s]
 57%|█████▋    | 3867/6837 [29:03<11:39,  4.24it/s]
 57%|█████▋    | 3868/6837 [29:03<11:39,  4.24it/s]
 57%|█████▋    | 3869/6837 [29:04<11:40,  4.24it/s]
 57%|█████▋    | 3870/6837 [29:04<11:40,  4.23it/s]
 57%|█████▋    | 3871/6837 [29:04<11:39,  4.24it/s]
 57%|█████▋    | 3872/6837 [29:04<11:39,  4.24it/s]
 57%|█████▋    | 3873/6837 [29:05<11:39,  4.24it/s]
 57%|█████▋    | 3874/6837 [29:05<11:39,  4.24it/s]
 57%|█████▋    | 3875/6837 [29:05<11:38,  4.24it/s]
                                                   
{'loss': 4.4324, 'grad_norm': 0.1498245745897293, 'learning_rate': 0.0028247159949775684, 'epoch': 0.11}
+
 57%|█████▋    | 3875/6837 [29:05<11:38,  4.24it/s]
 57%|█████▋    | 3876/6837 [29:05<11:39,  4.23it/s]
 57%|█████▋    | 3877/6837 [29:06<11:38,  4.24it/s]
 57%|█████▋    | 3878/6837 [29:06<11:39,  4.23it/s]
 57%|█████▋    | 3879/6837 [29:06<11:40,  4.22it/s]
 57%|█████▋    | 3880/6837 [29:06<11:39,  4.23it/s]
 57%|█████▋    | 3881/6837 [29:07<11:38,  4.23it/s]
 57%|█████▋    | 3882/6837 [29:07<11:37,  4.24it/s]
 57%|█████▋    | 3883/6837 [29:07<11:38,  4.23it/s]
 57%|█████▋    | 3884/6837 [29:07<11:38,  4.23it/s]
 57%|█████▋    | 3885/6837 [29:07<11:37,  4.23it/s]
 57%|█████▋    | 3886/6837 [29:08<11:39,  4.22it/s]
 57%|█████▋    | 3887/6837 [29:08<11:37,  4.23it/s]
 57%|█████▋    | 3888/6837 [29:08<11:37,  4.23it/s]
 57%|█████▋    | 3889/6837 [29:08<11:35,  4.24it/s]
 57%|█████▋    | 3890/6837 [29:09<11:36,  4.23it/s]
 57%|█████▋    | 3891/6837 [29:09<11:34,  4.24it/s]
 57%|█████▋    | 3892/6837 [29:09<11:35,  4.24it/s]
 57%|█████▋    | 3893/6837 [29:09<11:34,  4.24it/s]
 57%|█████▋    | 3894/6837 [29:10<11:34,  4.24it/s]
 57%|█████▋    | 3895/6837 [29:10<11:34,  4.24it/s]
 57%|█████▋    | 3896/6837 [29:10<11:34,  4.24it/s]
 57%|█████▋    | 3897/6837 [29:10<11:34,  4.23it/s]
 57%|█████▋    | 3898/6837 [29:11<11:33,  4.24it/s]
 57%|█████▋    | 3899/6837 [29:11<11:34,  4.23it/s]
 57%|█████▋    | 3900/6837 [29:11<11:33,  4.23it/s]
                                                   
{'loss': 4.4087, 'grad_norm': 0.15671294927597046, 'learning_rate': 0.0027865033062583556, 'epoch': 0.11}
+
 57%|█████▋    | 3900/6837 [29:11<11:33,  4.23it/s]
 57%|█████▋    | 3901/6837 [29:11<11:35,  4.22it/s]
 57%|█████▋    | 3902/6837 [29:12<11:34,  4.23it/s]
 57%|█████▋    | 3903/6837 [29:12<11:33,  4.23it/s]
 57%|█████▋    | 3904/6837 [29:12<11:34,  4.22it/s]
 57%|█████▋    | 3905/6837 [29:12<11:33,  4.23it/s]
 57%|█████▋    | 3906/6837 [29:12<11:32,  4.23it/s]
 57%|█████▋    | 3907/6837 [29:13<11:30,  4.24it/s]
 57%|█████▋    | 3908/6837 [29:13<11:55,  4.09it/s]
 57%|█████▋    | 3909/6837 [29:13<11:47,  4.14it/s]
 57%|█████▋    | 3910/6837 [29:13<11:41,  4.17it/s]
 57%|█████▋    | 3911/6837 [29:14<11:39,  4.19it/s]
 57%|█████▋    | 3912/6837 [29:14<11:36,  4.20it/s]
 57%|█████▋    | 3913/6837 [29:14<11:35,  4.21it/s]
 57%|█████▋    | 3914/6837 [29:14<11:33,  4.22it/s]
 57%|█████▋    | 3915/6837 [29:15<11:31,  4.23it/s]
 57%|█████▋    | 3916/6837 [29:15<11:30,  4.23it/s]
 57%|█████▋    | 3917/6837 [29:15<11:29,  4.23it/s]
 57%|█████▋    | 3918/6837 [29:15<11:29,  4.23it/s]
 57%|█████▋    | 3919/6837 [29:16<11:27,  4.24it/s]
 57%|█████▋    | 3920/6837 [29:16<11:27,  4.25it/s]
 57%|█████▋    | 3921/6837 [29:16<11:27,  4.24it/s]
 57%|█████▋    | 3922/6837 [29:16<11:27,  4.24it/s]
 57%|█████▋    | 3923/6837 [29:16<11:26,  4.24it/s]
 57%|█████▋    | 3924/6837 [29:17<11:26,  4.24it/s]
 57%|█████▋    | 3925/6837 [29:17<11:26,  4.24it/s]
                                                   
{'loss': 4.4142, 'grad_norm': 0.14212249219417572, 'learning_rate': 0.0027483254024736233, 'epoch': 0.11}
+
 57%|█████▋    | 3925/6837 [29:17<11:26,  4.24it/s]
 57%|█████▋    | 3926/6837 [29:17<11:28,  4.23it/s]
 57%|█████▋    | 3927/6837 [29:17<11:27,  4.23it/s]
 57%|█████▋    | 3928/6837 [29:18<11:26,  4.24it/s]
 57%|█████▋    | 3929/6837 [29:18<11:27,  4.23it/s]
 57%|█████▋    | 3930/6837 [29:18<11:26,  4.24it/s]
 57%|█████▋    | 3931/6837 [29:18<11:25,  4.24it/s]
 58%|█████▊    | 3932/6837 [29:19<11:24,  4.24it/s]
 58%|█████▊    | 3933/6837 [29:19<11:24,  4.24it/s]
 58%|█████▊    | 3934/6837 [29:19<11:25,  4.24it/s]
 58%|█████▊    | 3935/6837 [29:19<11:24,  4.24it/s]
 58%|█████▊    | 3936/6837 [29:20<11:24,  4.24it/s]
 58%|█████▊    | 3937/6837 [29:20<11:24,  4.24it/s]
 58%|█████▊    | 3938/6837 [29:20<11:25,  4.23it/s]
 58%|█████▊    | 3939/6837 [29:20<11:24,  4.23it/s]
 58%|█████▊    | 3940/6837 [29:21<11:23,  4.24it/s]
 58%|█████▊    | 3941/6837 [29:21<11:23,  4.24it/s]
 58%|█████▊    | 3942/6837 [29:21<11:21,  4.25it/s]
 58%|█████▊    | 3943/6837 [29:21<11:21,  4.25it/s]
 58%|█████▊    | 3944/6837 [29:21<11:21,  4.25it/s]
 58%|█████▊    | 3945/6837 [29:22<11:21,  4.24it/s]
 58%|█████▊    | 3946/6837 [29:22<11:21,  4.24it/s]
 58%|█████▊    | 3947/6837 [29:22<11:21,  4.24it/s]
 58%|█████▊    | 3948/6837 [29:22<11:19,  4.25it/s]
 58%|█████▊    | 3949/6837 [29:23<11:19,  4.25it/s]
 58%|█████▊    | 3950/6837 [29:23<11:19,  4.25it/s]
                                                   
{'loss': 4.4155, 'grad_norm': 0.17642661929130554, 'learning_rate': 0.002710188503934603, 'epoch': 0.12}
+
 58%|█████▊    | 3950/6837 [29:23<11:19,  4.25it/s]
 58%|█████▊    | 3951/6837 [29:23<11:22,  4.23it/s]
 58%|█████▊    | 3952/6837 [29:23<11:21,  4.23it/s]
 58%|█████▊    | 3953/6837 [29:24<11:21,  4.23it/s]
 58%|█████▊    | 3954/6837 [29:24<11:23,  4.22it/s]
 58%|█████▊    | 3955/6837 [29:24<11:23,  4.22it/s]
 58%|█████▊    | 3956/6837 [29:24<11:21,  4.22it/s]
 58%|█████▊    | 3957/6837 [29:25<11:21,  4.23it/s]
 58%|█████▊    | 3958/6837 [29:25<11:21,  4.23it/s]
 58%|█████▊    | 3959/6837 [29:25<11:20,  4.23it/s]
 58%|█████▊    | 3960/6837 [29:25<11:20,  4.23it/s]
 58%|█████▊    | 3961/6837 [29:25<11:20,  4.23it/s]
 58%|█████▊    | 3962/6837 [29:26<11:20,  4.23it/s]
 58%|█████▊    | 3963/6837 [29:26<11:19,  4.23it/s]
 58%|█████▊    | 3964/6837 [29:26<11:19,  4.23it/s]
 58%|█████▊    | 3965/6837 [29:26<11:19,  4.23it/s]
 58%|█████▊    | 3966/6837 [29:27<11:18,  4.23it/s]
 58%|█████▊    | 3967/6837 [29:27<11:17,  4.23it/s]
 58%|█████▊    | 3968/6837 [29:27<11:17,  4.23it/s]
 58%|█████▊    | 3969/6837 [29:27<11:18,  4.23it/s]
 58%|█████▊    | 3970/6837 [29:28<11:18,  4.23it/s]
 58%|█████▊    | 3971/6837 [29:28<11:17,  4.23it/s]
 58%|█████▊    | 3972/6837 [29:28<11:18,  4.22it/s]
 58%|█████▊    | 3973/6837 [29:28<11:17,  4.23it/s]
 58%|█████▊    | 3974/6837 [29:29<11:17,  4.23it/s]
 58%|█████▊    | 3975/6837 [29:29<11:16,  4.23it/s]
                                                   {'loss': 4.4186, 'grad_norm': 0.16185303032398224, 'learning_rate': 0.002672098824271554, 'epoch': 0.12}
+
 58%|█████▊    | 3975/6837 [29:29<11:16,  4.23it/s]
 58%|█████▊    | 3976/6837 [29:29<11:18,  4.22it/s]
 58%|█████▊    | 3977/6837 [29:29<11:16,  4.23it/s]
 58%|█████▊    | 3978/6837 [29:29<11:15,  4.23it/s]
 58%|█████▊    | 3979/6837 [29:30<11:15,  4.23it/s]
 58%|█████▊    | 3980/6837 [29:30<11:15,  4.23it/s]
 58%|█████▊    | 3981/6837 [29:30<11:14,  4.23it/s]
 58%|█████▊    | 3982/6837 [29:30<11:15,  4.23it/s]
 58%|█████▊    | 3983/6837 [29:31<11:16,  4.22it/s]
 58%|█████▊    | 3984/6837 [29:31<11:15,  4.23it/s]
 58%|█████▊    | 3985/6837 [29:31<11:15,  4.23it/s]
 58%|█████▊    | 3986/6837 [29:31<11:14,  4.23it/s]
 58%|█████▊    | 3987/6837 [29:32<11:14,  4.23it/s]
 58%|█████▊    | 3988/6837 [29:32<11:13,  4.23it/s]
 58%|█████▊    | 3989/6837 [29:32<11:13,  4.23it/s]
 58%|█████▊    | 3990/6837 [29:32<11:12,  4.23it/s]
 58%|█████▊    | 3991/6837 [29:33<11:12,  4.23it/s]
 58%|█████▊    | 3992/6837 [29:33<11:11,  4.23it/s]
 58%|█████▊    | 3993/6837 [29:33<11:13,  4.23it/s]
 58%|█████▊    | 3994/6837 [29:33<11:12,  4.23it/s]
 58%|█████▊    | 3995/6837 [29:34<11:11,  4.23it/s]
 58%|█████▊    | 3996/6837 [29:34<11:24,  4.15it/s]
 58%|█████▊    | 3997/6837 [29:34<11:20,  4.17it/s]
 58%|█████▊    | 3998/6837 [29:34<11:17,  4.19it/s]
 58%|█████▊    | 3999/6837 [29:34<11:14,  4.21it/s]
 59%|█████▊    | 4000/6837 [29:35<11:11,  4.22it/s]
                                                   
{'loss': 4.4102, 'grad_norm': 0.15099814534187317, 'learning_rate': 0.0026340625694213876, 'epoch': 0.12}
+
 59%|█████▊    | 4000/6837 [29:35<11:11,  4.22it/s]
 59%|█████▊    | 4001/6837 [29:35<11:12,  4.22it/s]
 59%|█████▊    | 4002/6837 [29:35<11:11,  4.22it/s]
 59%|█████▊    | 4003/6837 [29:35<11:10,  4.23it/s]
 59%|█████▊    | 4004/6837 [29:36<11:10,  4.22it/s]
 59%|█████▊    | 4005/6837 [29:36<11:09,  4.23it/s]
 59%|█████▊    | 4006/6837 [29:36<11:08,  4.23it/s]
 59%|█████▊    | 4007/6837 [29:36<11:08,  4.24it/s]
 59%|█████▊    | 4008/6837 [29:37<11:07,  4.24it/s]
 59%|█████▊    | 4009/6837 [29:37<11:07,  4.24it/s]
 59%|█████▊    | 4010/6837 [29:37<11:07,  4.23it/s]
 59%|█████▊    | 4011/6837 [29:37<11:08,  4.23it/s]
 59%|█████▊    | 4012/6837 [29:38<11:07,  4.23it/s]
 59%|█████▊    | 4013/6837 [29:38<11:05,  4.24it/s]
 59%|█████▊    | 4014/6837 [29:38<11:06,  4.24it/s]
 59%|█████▊    | 4015/6837 [29:38<11:06,  4.24it/s]
 59%|█████▊    | 4016/6837 [29:38<11:05,  4.24it/s]
 59%|█████▉    | 4017/6837 [29:39<11:04,  4.24it/s]
 59%|█████▉    | 4018/6837 [29:39<11:06,  4.23it/s]
 59%|█████▉    | 4019/6837 [29:39<11:05,  4.23it/s]
 59%|█████▉    | 4020/6837 [29:39<11:05,  4.23it/s]
 59%|█████▉    | 4021/6837 [29:40<11:05,  4.23it/s]
 59%|█████▉    | 4022/6837 [29:40<11:04,  4.24it/s]
 59%|█████▉    | 4023/6837 [29:40<11:04,  4.24it/s]
 59%|█████▉    | 4024/6837 [29:40<11:29,  4.08it/s]
 59%|█████▉    | 4025/6837 [29:41<11:47,  3.98it/s]
                                                   
{'loss': 4.4071, 'grad_norm': 0.15153741836547852, 'learning_rate': 0.0025960859366165275, 'epoch': 0.12}
+
 59%|█████▉    | 4025/6837 [29:41<11:47,  3.98it/s]
 59%|█████▉    | 4026/6837 [29:41<11:57,  3.92it/s]
 59%|█████▉    | 4027/6837 [29:41<12:04,  3.88it/s]
 59%|█████▉    | 4028/6837 [29:41<12:07,  3.86it/s]
 59%|█████▉    | 4029/6837 [29:42<11:53,  3.94it/s]
 59%|█████▉    | 4030/6837 [29:42<11:38,  4.02it/s]
 59%|█████▉    | 4031/6837 [29:42<11:28,  4.07it/s]
 59%|█████▉    | 4032/6837 [29:42<11:22,  4.11it/s]
 59%|█████▉    | 4033/6837 [29:43<11:16,  4.15it/s]
 59%|█████▉    | 4034/6837 [29:43<11:32,  4.05it/s]
 59%|█████▉    | 4035/6837 [29:43<11:49,  3.95it/s]
 59%|█████▉    | 4036/6837 [29:43<12:00,  3.89it/s]
 59%|█████▉    | 4037/6837 [29:44<12:03,  3.87it/s]
 59%|█████▉    | 4038/6837 [29:44<12:07,  3.85it/s]
 59%|█████▉    | 4039/6837 [29:44<12:10,  3.83it/s]
 59%|█████▉    | 4040/6837 [29:44<11:49,  3.94it/s]
 59%|█████▉    | 4041/6837 [29:45<11:33,  4.03it/s]
 59%|█████▉    | 4042/6837 [29:45<11:23,  4.09it/s]
 59%|█████▉    | 4043/6837 [29:45<11:15,  4.13it/s]
 59%|█████▉    | 4044/6837 [29:45<11:10,  4.17it/s]
 59%|█████▉    | 4045/6837 [29:46<11:07,  4.18it/s]
 59%|█████▉    | 4046/6837 [29:46<11:04,  4.20it/s]
 59%|█████▉    | 4047/6837 [29:46<11:02,  4.21it/s]
 59%|█████▉    | 4048/6837 [29:46<11:01,  4.21it/s]
 59%|█████▉    | 4049/6837 [29:47<11:02,  4.21it/s]
 59%|█████▉    | 4050/6837 [29:47<11:00,  4.22it/s]
                                                   
{'loss': 4.4103, 'grad_norm': 0.1720927506685257, 'learning_rate': 0.002558175113375197, 'epoch': 0.12}
+
 59%|█████▉    | 4050/6837 [29:47<11:00,  4.22it/s]
 59%|█████▉    | 4051/6837 [29:47<11:02,  4.21it/s]
 59%|█████▉    | 4052/6837 [29:47<11:01,  4.21it/s]
 59%|█████▉    | 4053/6837 [29:48<11:00,  4.21it/s]
 59%|█████▉    | 4054/6837 [29:48<10:59,  4.22it/s]
 59%|█████▉    | 4055/6837 [29:48<10:59,  4.22it/s]
 59%|█████▉    | 4056/6837 [29:48<10:59,  4.21it/s]
 59%|█████▉    | 4057/6837 [29:48<10:59,  4.21it/s]
 59%|█████▉    | 4058/6837 [29:49<10:57,  4.22it/s]
 59%|█████▉    | 4059/6837 [29:49<10:58,  4.22it/s]
 59%|█████▉    | 4060/6837 [29:49<10:58,  4.22it/s]
 59%|█████▉    | 4061/6837 [29:49<10:58,  4.22it/s]
 59%|█████▉    | 4062/6837 [29:50<10:56,  4.22it/s]
 59%|█████▉    | 4063/6837 [29:50<10:55,  4.23it/s]
 59%|█████▉    | 4064/6837 [29:50<10:54,  4.23it/s]
 59%|█████▉    | 4065/6837 [29:50<10:54,  4.24it/s]
 59%|█████▉    | 4066/6837 [29:51<10:54,  4.24it/s]
 59%|█████▉    | 4067/6837 [29:51<10:53,  4.24it/s]
 59%|█████▉    | 4068/6837 [29:51<10:52,  4.24it/s]
 60%|█████▉    | 4069/6837 [29:51<10:51,  4.25it/s]
 60%|█████▉    | 4070/6837 [29:52<10:52,  4.24it/s]
 60%|█████▉    | 4071/6837 [29:52<10:52,  4.24it/s]
 60%|█████▉    | 4072/6837 [29:52<10:52,  4.24it/s]
 60%|█████▉    | 4073/6837 [29:52<10:52,  4.24it/s]
 60%|█████▉    | 4074/6837 [29:53<10:52,  4.23it/s]
 60%|█████▉    | 4075/6837 [29:53<10:51,  4.24it/s]
                                                   
{'loss': 4.403, 'grad_norm': 0.1552221029996872, 'learning_rate': 0.0025203362764932994, 'epoch': 0.12}
+
 60%|█████▉    | 4075/6837 [29:53<10:51,  4.24it/s]
 60%|█████▉    | 4076/6837 [29:53<10:53,  4.22it/s]
 60%|█████▉    | 4077/6837 [29:53<10:52,  4.23it/s]
 60%|█████▉    | 4078/6837 [29:53<10:52,  4.23it/s]
 60%|█████▉    | 4079/6837 [29:54<10:51,  4.23it/s]
 60%|█████▉    | 4080/6837 [29:54<10:50,  4.24it/s]
 60%|█████▉    | 4081/6837 [29:54<10:51,  4.23it/s]
 60%|█████▉    | 4082/6837 [29:54<10:51,  4.23it/s]
 60%|█████▉    | 4083/6837 [29:55<10:50,  4.24it/s]
 60%|█████▉    | 4084/6837 [29:55<10:49,  4.24it/s]
 60%|█████▉    | 4085/6837 [29:55<10:49,  4.24it/s]
 60%|█████▉    | 4086/6837 [29:55<10:48,  4.24it/s]
 60%|█████▉    | 4087/6837 [29:56<10:47,  4.25it/s]
 60%|█████▉    | 4088/6837 [29:56<10:48,  4.24it/s]
 60%|█████▉    | 4089/6837 [29:56<10:47,  4.24it/s]
 60%|█████▉    | 4090/6837 [29:56<10:47,  4.24it/s]
 60%|█████▉    | 4091/6837 [29:57<10:47,  4.24it/s]
 60%|█████▉    | 4092/6837 [29:57<10:48,  4.24it/s]
 60%|█████▉    | 4093/6837 [29:57<10:47,  4.24it/s]
 60%|█████▉    | 4094/6837 [29:57<10:47,  4.24it/s]
 60%|█████▉    | 4095/6837 [29:57<10:46,  4.24it/s]
 60%|█████▉    | 4096/6837 [29:58<10:46,  4.24it/s]
 60%|█████▉    | 4097/6837 [29:58<10:46,  4.24it/s]
 60%|█████▉    | 4098/6837 [29:58<10:46,  4.24it/s]
 60%|█████▉    | 4099/6837 [29:58<10:47,  4.23it/s]
 60%|█████▉    | 4100/6837 [29:59<10:47,  4.23it/s]
                                                   
{'loss': 4.4132, 'grad_norm': 0.14017847180366516, 'learning_rate': 0.002482575591038024, 'epoch': 0.12}
+
 60%|█████▉    | 4100/6837 [29:59<10:47,  4.23it/s]
 60%|█████▉    | 4101/6837 [29:59<10:48,  4.22it/s]
 60%|█████▉    | 4102/6837 [29:59<10:48,  4.22it/s]
 60%|██████    | 4103/6837 [29:59<10:47,  4.22it/s]
 60%|██████    | 4104/6837 [30:00<10:47,  4.22it/s]
 60%|██████    | 4105/6837 [30:00<10:47,  4.22it/s]
 60%|██████    | 4106/6837 [30:00<10:46,  4.23it/s]
 60%|██████    | 4107/6837 [30:00<10:46,  4.23it/s]
 60%|██████    | 4108/6837 [30:01<10:46,  4.22it/s]
 60%|██████    | 4109/6837 [30:01<10:46,  4.22it/s]
 60%|██████    | 4110/6837 [30:01<10:46,  4.22it/s]
 60%|██████    | 4111/6837 [30:01<10:43,  4.23it/s]
 60%|██████    | 4112/6837 [30:01<10:43,  4.23it/s]
 60%|██████    | 4113/6837 [30:02<10:42,  4.24it/s]
 60%|██████    | 4114/6837 [30:02<10:42,  4.24it/s]
 60%|██████    | 4115/6837 [30:02<10:41,  4.24it/s]
 60%|██████    | 4116/6837 [30:02<10:41,  4.24it/s]
 60%|██████    | 4117/6837 [30:03<10:42,  4.24it/s]
 60%|██████    | 4118/6837 [30:03<10:42,  4.23it/s]
 60%|██████    | 4119/6837 [30:03<10:41,  4.24it/s]
 60%|██████    | 4120/6837 [30:03<10:41,  4.23it/s]
 60%|██████    | 4121/6837 [30:04<10:40,  4.24it/s]
 60%|██████    | 4122/6837 [30:04<11:03,  4.09it/s]
 60%|██████    | 4123/6837 [30:04<11:19,  3.99it/s]
 60%|██████    | 4124/6837 [30:04<11:27,  3.95it/s]
 60%|██████    | 4125/6837 [30:05<11:22,  3.97it/s]
                                                   
{'loss': 4.3988, 'grad_norm': 0.15832413733005524, 'learning_rate': 0.002444899209343379, 'epoch': 0.12}
+
 60%|██████    | 4125/6837 [30:05<11:22,  3.97it/s]
 60%|██████    | 4126/6837 [30:05<11:11,  4.04it/s]
 60%|██████    | 4127/6837 [30:05<11:01,  4.10it/s]
 60%|██████    | 4128/6837 [30:05<10:55,  4.13it/s]
 60%|██████    | 4129/6837 [30:06<10:50,  4.16it/s]
 60%|██████    | 4130/6837 [30:06<10:48,  4.17it/s]
 60%|██████    | 4131/6837 [30:06<10:45,  4.19it/s]
 60%|██████    | 4132/6837 [30:06<10:42,  4.21it/s]
 60%|██████    | 4133/6837 [30:07<10:40,  4.22it/s]
 60%|██████    | 4134/6837 [30:07<10:39,  4.23it/s]
 60%|██████    | 4135/6837 [30:07<10:39,  4.23it/s]
 60%|██████    | 4136/6837 [30:07<10:38,  4.23it/s]
 61%|██████    | 4137/6837 [30:07<10:37,  4.23it/s]
 61%|██████    | 4138/6837 [30:08<10:38,  4.23it/s]
 61%|██████    | 4139/6837 [30:08<10:37,  4.23it/s]
 61%|██████    | 4140/6837 [30:08<10:36,  4.24it/s]
 61%|██████    | 4141/6837 [30:08<10:36,  4.24it/s]
 61%|██████    | 4142/6837 [30:09<10:35,  4.24it/s]
 61%|██████    | 4143/6837 [30:09<10:35,  4.24it/s]
 61%|██████    | 4144/6837 [30:09<10:35,  4.24it/s]
 61%|██████    | 4145/6837 [30:09<10:35,  4.23it/s]
 61%|██████    | 4146/6837 [30:10<10:35,  4.23it/s]
 61%|██████    | 4147/6837 [30:10<10:35,  4.23it/s]
 61%|██████    | 4148/6837 [30:10<10:35,  4.23it/s]
 61%|██████    | 4149/6837 [30:10<10:34,  4.24it/s]
 61%|██████    | 4150/6837 [30:11<10:33,  4.24it/s]
                                                   
{'loss': 4.3935, 'grad_norm': 0.1576928049325943, 'learning_rate': 0.0024073132700077916, 'epoch': 0.12}
+
 61%|██████    | 4150/6837 [30:11<10:33,  4.24it/s]
 61%|██████    | 4151/6837 [30:11<10:34,  4.23it/s]
 61%|██████    | 4152/6837 [30:11<10:33,  4.24it/s]
 61%|██████    | 4153/6837 [30:11<10:33,  4.24it/s]
 61%|██████    | 4154/6837 [30:11<10:33,  4.24it/s]
 61%|██████    | 4155/6837 [30:12<10:32,  4.24it/s]
 61%|██████    | 4156/6837 [30:12<10:33,  4.23it/s]
 61%|██████    | 4157/6837 [30:12<10:34,  4.23it/s]
 61%|██████    | 4158/6837 [30:12<10:34,  4.22it/s]
 61%|██████    | 4159/6837 [30:13<10:33,  4.23it/s]
 61%|██████    | 4160/6837 [30:13<10:55,  4.08it/s]
 61%|██████    | 4161/6837 [30:13<10:48,  4.12it/s]
 61%|██████    | 4162/6837 [30:13<10:43,  4.16it/s]
 61%|██████    | 4163/6837 [30:14<10:40,  4.18it/s]
 61%|██████    | 4164/6837 [30:14<10:36,  4.20it/s]
 61%|██████    | 4165/6837 [30:14<10:34,  4.21it/s]
 61%|██████    | 4166/6837 [30:14<10:32,  4.22it/s]
 61%|██████    | 4167/6837 [30:15<10:32,  4.22it/s]
 61%|██████    | 4168/6837 [30:15<10:31,  4.23it/s]
 61%|██████    | 4169/6837 [30:15<10:30,  4.23it/s]
 61%|██████    | 4170/6837 [30:15<10:29,  4.23it/s]
 61%|██████    | 4171/6837 [30:16<10:29,  4.24it/s]
 61%|██████    | 4172/6837 [30:16<10:29,  4.24it/s]
 61%|██████    | 4173/6837 [30:16<10:28,  4.24it/s]
 61%|██████    | 4174/6837 [30:16<10:29,  4.23it/s]
 61%|██████    | 4175/6837 [30:16<10:29,  4.23it/s]
                                                   
{'loss': 4.4187, 'grad_norm': 0.18703840672969818, 'learning_rate': 0.002369823896893955, 'epoch': 0.12}
+
 61%|██████    | 4175/6837 [30:16<10:29,  4.23it/s]
 61%|██████    | 4176/6837 [30:17<10:30,  4.22it/s]
 61%|██████    | 4177/6837 [30:17<10:29,  4.23it/s]
 61%|██████    | 4178/6837 [30:17<10:28,  4.23it/s]
 61%|██████    | 4179/6837 [30:17<10:27,  4.23it/s]
 61%|██████    | 4180/6837 [30:18<10:28,  4.23it/s]
 61%|██████    | 4181/6837 [30:18<10:27,  4.23it/s]
 61%|██████    | 4182/6837 [30:18<10:27,  4.23it/s]
 61%|██████    | 4183/6837 [30:18<10:27,  4.23it/s]
 61%|██████    | 4184/6837 [30:19<10:28,  4.22it/s]
 61%|██████    | 4185/6837 [30:19<10:27,  4.23it/s]
 61%|██████    | 4186/6837 [30:19<10:26,  4.23it/s]
 61%|██████    | 4187/6837 [30:19<10:26,  4.23it/s]
 61%|██████▏   | 4188/6837 [30:20<10:26,  4.23it/s]
 61%|██████▏   | 4189/6837 [30:20<10:24,  4.24it/s]
 61%|██████▏   | 4190/6837 [30:20<10:24,  4.24it/s]
 61%|██████▏   | 4191/6837 [30:20<10:24,  4.24it/s]
 61%|██████▏   | 4192/6837 [30:20<10:25,  4.23it/s]
 61%|██████▏   | 4193/6837 [30:21<10:24,  4.23it/s]
 61%|██████▏   | 4194/6837 [30:21<10:24,  4.23it/s]
 61%|██████▏   | 4195/6837 [30:21<10:24,  4.23it/s]
 61%|██████▏   | 4196/6837 [30:21<10:24,  4.23it/s]
 61%|██████▏   | 4197/6837 [30:22<10:24,  4.23it/s]
 61%|██████▏   | 4198/6837 [30:22<10:24,  4.23it/s]
 61%|██████▏   | 4199/6837 [30:22<10:24,  4.22it/s]
 61%|██████▏   | 4200/6837 [30:22<10:24,  4.22it/s]
                                                   
{'loss': 4.4002, 'grad_norm': 0.15781816840171814, 'learning_rate': 0.002332437198131057, 'epoch': 0.12}
+
 61%|██████▏   | 4200/6837 [30:22<10:24,  4.22it/s]
 61%|██████▏   | 4201/6837 [30:23<10:25,  4.22it/s]
 61%|██████▏   | 4202/6837 [30:23<10:25,  4.21it/s]
 61%|██████▏   | 4203/6837 [30:23<10:24,  4.22it/s]
 61%|██████▏   | 4204/6837 [30:23<10:24,  4.22it/s]
 62%|██████▏   | 4205/6837 [30:24<10:24,  4.22it/s]
 62%|██████▏   | 4206/6837 [30:24<10:24,  4.21it/s]
 62%|██████▏   | 4207/6837 [30:24<10:23,  4.22it/s]
 62%|██████▏   | 4208/6837 [30:24<10:23,  4.22it/s]
 62%|██████▏   | 4209/6837 [30:25<10:23,  4.21it/s]
 62%|██████▏   | 4210/6837 [30:25<10:22,  4.22it/s]
 62%|██████▏   | 4211/6837 [30:25<10:21,  4.22it/s]
 62%|██████▏   | 4212/6837 [30:25<10:21,  4.23it/s]
 62%|██████▏   | 4213/6837 [30:25<10:20,  4.23it/s]
 62%|██████▏   | 4214/6837 [30:26<10:19,  4.24it/s]
 62%|██████▏   | 4215/6837 [30:26<10:18,  4.24it/s]
 62%|██████▏   | 4216/6837 [30:26<10:18,  4.24it/s]
 62%|██████▏   | 4217/6837 [30:26<10:19,  4.23it/s]
 62%|██████▏   | 4218/6837 [30:27<10:19,  4.23it/s]
 62%|██████▏   | 4219/6837 [30:27<10:18,  4.23it/s]
 62%|██████▏   | 4220/6837 [30:27<10:18,  4.23it/s]
 62%|██████▏   | 4221/6837 [30:27<10:17,  4.23it/s]
 62%|██████▏   | 4222/6837 [30:28<10:17,  4.23it/s]
 62%|██████▏   | 4223/6837 [30:28<10:17,  4.23it/s]
 62%|██████▏   | 4224/6837 [30:28<10:17,  4.23it/s]
 62%|██████▏   | 4225/6837 [30:28<10:16,  4.23it/s]
                                                   
{'loss': 4.3962, 'grad_norm': 0.15633727610111237, 'learning_rate': 0.0022951592651196, 'epoch': 0.12}
+
 62%|██████▏   | 4225/6837 [30:28<10:16,  4.23it/s]
 62%|██████▏   | 4226/6837 [30:29<10:19,  4.22it/s]
 62%|██████▏   | 4227/6837 [30:29<10:18,  4.22it/s]
 62%|██████▏   | 4228/6837 [30:29<10:17,  4.23it/s]
 62%|██████▏   | 4229/6837 [30:29<10:16,  4.23it/s]
 62%|██████▏   | 4230/6837 [30:29<10:16,  4.23it/s]
 62%|██████▏   | 4231/6837 [30:30<10:16,  4.23it/s]
 62%|██████▏   | 4232/6837 [30:30<10:14,  4.24it/s]
 62%|██████▏   | 4233/6837 [30:30<10:14,  4.24it/s]
 62%|██████▏   | 4234/6837 [30:30<10:14,  4.23it/s]
 62%|██████▏   | 4235/6837 [30:31<10:14,  4.23it/s]
 62%|██████▏   | 4236/6837 [30:31<10:14,  4.24it/s]
 62%|██████▏   | 4237/6837 [30:31<10:13,  4.23it/s]
 62%|██████▏   | 4238/6837 [30:31<10:13,  4.23it/s]
 62%|██████▏   | 4239/6837 [30:32<10:13,  4.23it/s]
 62%|██████▏   | 4240/6837 [30:32<10:13,  4.23it/s]
 62%|██████▏   | 4241/6837 [30:32<10:13,  4.23it/s]
 62%|██████▏   | 4242/6837 [30:32<10:12,  4.24it/s]
 62%|██████▏   | 4243/6837 [30:33<10:11,  4.24it/s]
 62%|██████▏   | 4244/6837 [30:33<10:11,  4.24it/s]
 62%|██████▏   | 4245/6837 [30:33<10:12,  4.23it/s]
 62%|██���███▏   | 4246/6837 [30:33<10:11,  4.23it/s]
 62%|██████▏   | 4247/6837 [30:34<10:11,  4.24it/s]
 62%|██████▏   | 4248/6837 [30:34<10:12,  4.23it/s]
 62%|██████▏   | 4249/6837 [30:34<10:11,  4.23it/s]
 62%|██████▏   | 4250/6837 [30:34<10:11,  4.23it/s]
                                                   
{'loss': 4.3871, 'grad_norm': 0.1562398374080658, 'learning_rate': 0.002257996171538921, 'epoch': 0.12}
+
 62%|██████▏   | 4250/6837 [30:34<10:11,  4.23it/s]
 62%|██████▏   | 4251/6837 [30:34<10:13,  4.22it/s]
 62%|██████▏   | 4252/6837 [30:35<10:12,  4.22it/s]
 62%|██████▏   | 4253/6837 [30:35<10:11,  4.22it/s]
 62%|██████▏   | 4254/6837 [30:35<10:11,  4.23it/s]
 62%|██████▏   | 4255/6837 [30:35<10:10,  4.23it/s]
 62%|██████▏   | 4256/6837 [30:36<10:09,  4.24it/s]
 62%|██████▏   | 4257/6837 [30:36<10:08,  4.24it/s]
 62%|██████▏   | 4258/6837 [30:36<10:10,  4.22it/s]
 62%|██████▏   | 4259/6837 [30:36<10:13,  4.20it/s]
 62%|██████▏   | 4260/6837 [30:37<10:14,  4.19it/s]
 62%|██████▏   | 4261/6837 [30:37<10:13,  4.20it/s]
 62%|██████▏   | 4262/6837 [30:37<10:11,  4.21it/s]
 62%|██████▏   | 4263/6837 [30:37<10:10,  4.22it/s]
 62%|██████▏   | 4264/6837 [30:38<10:08,  4.23it/s]
 62%|██████▏   | 4265/6837 [30:38<10:08,  4.23it/s]
 62%|██████▏   | 4266/6837 [30:38<10:07,  4.23it/s]
 62%|██████▏   | 4267/6837 [30:38<10:07,  4.23it/s]
 62%|██████▏   | 4268/6837 [30:38<10:06,  4.24it/s]
 62%|██████▏   | 4269/6837 [30:39<10:06,  4.24it/s]
 62%|██████▏   | 4270/6837 [30:39<10:05,  4.24it/s]
 62%|██████▏   | 4271/6837 [30:39<10:05,  4.24it/s]
 62%|██████▏   | 4272/6837 [30:39<10:04,  4.24it/s]
 62%|██████▏   | 4273/6837 [30:40<10:05,  4.23it/s]
 63%|██████▎   | 4274/6837 [30:40<10:04,  4.24it/s]
 63%|██████▎   | 4275/6837 [30:40<10:04,  4.24it/s]
                                                   
{'loss': 4.393, 'grad_norm': 0.14862550795078278, 'learning_rate': 0.0022209539723576123, 'epoch': 0.13}
+
 63%|██████▎   | 4275/6837 [30:40<10:04,  4.24it/s]
 63%|██████▎   | 4276/6837 [30:40<10:05,  4.23it/s]
 63%|██████▎   | 4277/6837 [30:41<10:04,  4.23it/s]
 63%|██████▎   | 4278/6837 [30:41<10:04,  4.24it/s]
 63%|██████▎   | 4279/6837 [30:41<10:03,  4.24it/s]
 63%|██████▎   | 4280/6837 [30:41<10:03,  4.24it/s]
 63%|██████▎   | 4281/6837 [30:42<10:02,  4.24it/s]
 63%|██████▎   | 4282/6837 [30:42<10:01,  4.24it/s]
 63%|██████▎   | 4283/6837 [30:42<10:01,  4.25it/s]
 63%|██████▎   | 4284/6837 [30:42<10:02,  4.24it/s]
 63%|██████▎   | 4285/6837 [30:43<10:26,  4.07it/s]
 63%|██████▎   | 4286/6837 [30:43<10:41,  3.98it/s]
 63%|██████▎   | 4287/6837 [30:43<10:53,  3.90it/s]
 63%|██████▎   | 4288/6837 [30:43<10:56,  3.88it/s]
 63%|██████▎   | 4289/6837 [30:44<10:49,  3.92it/s]
 63%|██████▎   | 4290/6837 [30:44<10:37,  4.00it/s]
 63%|██████▎   | 4291/6837 [30:44<10:27,  4.06it/s]
 63%|██████▎   | 4292/6837 [30:44<10:25,  4.07it/s]
 63%|██████▎   | 4293/6837 [30:45<10:18,  4.11it/s]
 63%|██████▎   | 4294/6837 [30:45<10:14,  4.14it/s]
 63%|██████▎   | 4295/6837 [30:45<10:10,  4.16it/s]
 63%|██████▎   | 4296/6837 [30:45<10:07,  4.18it/s]
 63%|██████▎   | 4297/6837 [30:45<10:05,  4.19it/s]
 63%|██████▎   | 4298/6837 [30:46<10:04,  4.20it/s]
 63%|██████▎   | 4299/6837 [30:46<10:03,  4.20it/s]
 63%|██████▎   | 4300/6837 [30:46<10:02,  4.21it/s]{'loss': 4.3904, 'grad_norm': 0.14203400909900665, 'learning_rate': 0.0021840387028469886, 'epoch': 0.13}
                                                   
+
 63%|██████▎   | 4300/6837 [30:46<10:02,  4.21it/s]
 63%|██████▎   | 4301/6837 [30:46<10:04,  4.20it/s]
 63%|██████▎   | 4302/6837 [30:47<10:03,  4.20it/s]
 63%|██████▎   | 4303/6837 [30:47<10:02,  4.21it/s]
 63%|██████▎   | 4304/6837 [30:47<10:01,  4.21it/s]
 63%|██████▎   | 4305/6837 [30:47<10:00,  4.22it/s]
 63%|██████▎   | 4306/6837 [30:48<09:58,  4.23it/s]
 63%|██████▎   | 4307/6837 [30:48<09:58,  4.23it/s]
 63%|██████▎   | 4308/6837 [30:48<09:59,  4.22it/s]
 63%|██████▎   | 4309/6837 [30:48<09:58,  4.23it/s]
 63%|██████▎   | 4310/6837 [30:49<09:57,  4.23it/s]
 63%|██████▎   | 4311/6837 [30:49<09:57,  4.23it/s]
 63%|██████▎   | 4312/6837 [30:49<09:57,  4.22it/s]
 63%|██████▎   | 4313/6837 [30:49<09:57,  4.23it/s]
 63%|██████▎   | 4314/6837 [30:49<09:56,  4.23it/s]
 63%|██████▎   | 4315/6837 [30:50<09:56,  4.23it/s]
 63%|██████▎   | 4316/6837 [30:50<09:55,  4.23it/s]
 63%|██████▎   | 4317/6837 [30:50<09:56,  4.23it/s]
 63%|██████▎   | 4318/6837 [30:50<09:55,  4.23it/s]
 63%|██████▎   | 4319/6837 [30:51<09:55,  4.23it/s]
 63%|██████▎   | 4320/6837 [30:51<09:54,  4.23it/s]
 63%|██████▎   | 4321/6837 [30:51<09:54,  4.23it/s]
 63%|██████▎   | 4322/6837 [30:51<09:54,  4.23it/s]
 63%|██████▎   | 4323/6837 [30:52<09:54,  4.23it/s]
 63%|██████▎   | 4324/6837 [30:52<09:54,  4.23it/s]
 63%|██████▎   | 4325/6837 [30:52<09:54,  4.23it/s]
                                                   {'loss': 4.3732, 'grad_norm': 0.1431817263364792, 'learning_rate': 0.002147256377597765, 'epoch': 0.13}
+
 63%|██████▎   | 4325/6837 [30:52<09:54,  4.23it/s]
 63%|██████▎   | 4326/6837 [30:52<09:56,  4.21it/s]
 63%|██████▎   | 4327/6837 [30:53<09:55,  4.21it/s]
 63%|██████▎   | 4328/6837 [30:53<09:55,  4.21it/s]
 63%|██████▎   | 4329/6837 [30:53<09:54,  4.22it/s]
 63%|██████▎   | 4330/6837 [30:53<09:54,  4.22it/s]
 63%|██████▎   | 4331/6837 [30:54<09:53,  4.22it/s]
 63%|██████▎   | 4332/6837 [30:54<09:53,  4.22it/s]
 63%|██████▎   | 4333/6837 [30:54<09:55,  4.21it/s]
 63%|██████▎   | 4334/6837 [30:54<09:54,  4.21it/s]
 63%|██████▎   | 4335/6837 [30:54<09:53,  4.22it/s]
 63%|██████▎   | 4336/6837 [30:55<09:53,  4.22it/s]
 63%|██████▎   | 4337/6837 [30:55<09:52,  4.22it/s]
 63%|██████▎   | 4338/6837 [30:55<09:52,  4.22it/s]
 63%|██████▎   | 4339/6837 [30:55<09:51,  4.23it/s]
 63%|██████▎   | 4340/6837 [30:56<09:50,  4.23it/s]
 63%|██████▎   | 4341/6837 [30:56<09:49,  4.23it/s]
 64%|██████▎   | 4342/6837 [30:56<09:54,  4.19it/s]
 64%|██████▎   | 4343/6837 [30:56<09:53,  4.20it/s]
 64%|██████▎   | 4344/6837 [30:57<09:52,  4.21it/s]
 64%|██████▎   | 4345/6837 [30:57<09:51,  4.21it/s]
 64%|██████▎   | 4346/6837 [30:57<09:50,  4.22it/s]
 64%|██████▎   | 4347/6837 [30:57<09:49,  4.23it/s]
 64%|██████▎   | 4348/6837 [30:58<09:48,  4.23it/s]
 64%|██████▎   | 4349/6837 [30:58<09:48,  4.23it/s]
 64%|██████▎   | 4350/6837 [30:58<09:48,  4.22it/s]
                                                   
{'loss': 4.3774, 'grad_norm': 0.14435188472270966, 'learning_rate': 0.002110612989540092, 'epoch': 0.13}
+
 64%|██████▎   | 4350/6837 [30:58<09:48,  4.22it/s]
 64%|██████▎   | 4351/6837 [30:58<09:52,  4.19it/s]
 64%|██████▎   | 4352/6837 [30:59<09:52,  4.19it/s]
 64%|██████▎   | 4353/6837 [30:59<09:51,  4.20it/s]
 64%|██████▎   | 4354/6837 [30:59<09:50,  4.20it/s]
 64%|██████▎   | 4355/6837 [30:59<09:49,  4.21it/s]
 64%|██████▎   | 4356/6837 [30:59<09:48,  4.22it/s]
 64%|██████▎   | 4357/6837 [31:00<09:47,  4.22it/s]
 64%|██████▎   | 4358/6837 [31:00<09:47,  4.22it/s]
 64%|██████▍   | 4359/6837 [31:00<09:47,  4.22it/s]
 64%|██████▍   | 4360/6837 [31:00<09:46,  4.22it/s]
 64%|██████▍   | 4361/6837 [31:01<10:05,  4.09it/s]
 64%|██████▍   | 4362/6837 [31:01<10:19,  3.99it/s]
 64%|██████▍   | 4363/6837 [31:01<10:30,  3.92it/s]
 64%|██████▍   | 4364/6837 [31:01<10:38,  3.88it/s]
 64%|██████▍   | 4365/6837 [31:02<10:26,  3.95it/s]
 64%|██████▍   | 4366/6837 [31:02<10:12,  4.03it/s]
 64%|██████▍   | 4367/6837 [31:02<10:03,  4.09it/s]
 64%|██████▍   | 4368/6837 [31:02<09:57,  4.14it/s]
 64%|██████▍   | 4369/6837 [31:03<09:52,  4.17it/s]
 64%|██████▍   | 4370/6837 [31:03<09:48,  4.19it/s]
 64%|██████▍   | 4371/6837 [31:03<09:46,  4.20it/s]
 64%|██████▍   | 4372/6837 [31:03<09:45,  4.21it/s]
 64%|██████▍   | 4373/6837 [31:04<09:44,  4.21it/s]
 64%|██████▍   | 4374/6837 [31:04<09:43,  4.22it/s]
 64%|██████▍   | 4375/6837 [31:04<09:45,  4.21it/s]
                                                   
{'loss': 4.3858, 'grad_norm': 0.1456858515739441, 'learning_rate': 0.002074114508967142, 'epoch': 0.13}
+
 64%|██████▍   | 4375/6837 [31:04<09:45,  4.21it/s]
 64%|██████▍   | 4376/6837 [31:04<09:45,  4.20it/s]
 64%|██████▍   | 4377/6837 [31:05<09:44,  4.21it/s]
 64%|██████▍   | 4378/6837 [31:05<09:44,  4.21it/s]
 64%|██████▍   | 4379/6837 [31:05<09:43,  4.21it/s]
 64%|██████▍   | 4380/6837 [31:05<09:42,  4.22it/s]
 64%|██████▍   | 4381/6837 [31:05<09:42,  4.21it/s]
 64%|██████▍   | 4382/6837 [31:06<09:44,  4.20it/s]
 64%|██████▍   | 4383/6837 [31:06<09:43,  4.20it/s]
 64%|██████▍   | 4384/6837 [31:06<09:42,  4.21it/s]
 64%|██████▍   | 4385/6837 [31:06<09:42,  4.21it/s]
 64%|██████▍   | 4386/6837 [31:07<09:42,  4.21it/s]
 64%|██████▍   | 4387/6837 [31:07<09:41,  4.22it/s]
 64%|██████▍   | 4388/6837 [31:07<09:40,  4.22it/s]
 64%|██████▍   | 4389/6837 [31:07<09:40,  4.21it/s]
 64%|██████▍   | 4390/6837 [31:08<09:40,  4.21it/s]
 64%|██████▍   | 4391/6837 [31:08<09:39,  4.22it/s]
 64%|██████▍   | 4392/6837 [31:08<09:39,  4.22it/s]
 64%|██████▍   | 4393/6837 [31:08<09:38,  4.22it/s]
 64%|██████▍   | 4394/6837 [31:09<09:37,  4.23it/s]
 64%|██████▍   | 4395/6837 [31:09<09:36,  4.23it/s]
 64%|██████▍   | 4396/6837 [31:09<09:37,  4.23it/s]
 64%|██████▍   | 4397/6837 [31:09<09:36,  4.23it/s]
 64%|██████▍   | 4398/6837 [31:10<09:36,  4.23it/s]
 64%|██████▍   | 4399/6837 [31:10<09:35,  4.24it/s]
 64%|██████▍   | 4400/6837 [31:10<09:35,  4.24it/s]
                                                   {'loss': 4.3873, 'grad_norm': 0.21040885150432587, 'learning_rate': 0.0020377668825623673, 'epoch': 0.13}
+

 64%|██████▍   | 4400/6837 [31:10<09:35,  4.24it/s]
 64%|██████▍   | 4401/6837 [31:10<09:36,  4.23it/s]
 64%|██████▍   | 4402/6837 [31:10<09:35,  4.23it/s]
 64%|██████▍   | 4403/6837 [31:11<09:34,  4.24it/s]
 64%|██████▍   | 4404/6837 [31:11<09:34,  4.24it/s]
 64%|██████▍   | 4405/6837 [31:11<09:33,  4.24it/s]
 64%|██████▍   | 4406/6837 [31:11<09:33,  4.24it/s]
 64%|██████▍   | 4407/6837 [31:12<09:33,  4.23it/s]
 64%|██████▍   | 4408/6837 [31:12<09:33,  4.24it/s]
 64%|██████▍   | 4409/6837 [31:12<09:32,  4.24it/s]
 65%|██████▍   | 4410/6837 [31:12<09:31,  4.24it/s]
 65%|██████▍   | 4411/6837 [31:13<09:31,  4.24it/s]
 65%|██████▍   | 4412/6837 [31:13<09:52,  4.10it/s]
 65%|██████▍   | 4413/6837 [31:13<10:06,  4.00it/s]
 65%|██████▍   | 4414/6837 [31:13<10:22,  3.90it/s]
 65%|██████▍   | 4415/6837 [31:14<10:25,  3.87it/s]
 65%|██████▍   | 4416/6837 [31:14<10:29,  3.85it/s]
 65%|██████▍   | 4417/6837 [31:14<10:33,  3.82it/s]
 65%|██████▍   | 4418/6837 [31:14<10:36,  3.80it/s]
 65%|██████▍   | 4419/6837 [31:15<10:35,  3.81it/s]
 65%|██████▍   | 4420/6837 [31:15<10:35,  3.81it/s]
 65%|██████▍   | 4421/6837 [31:15<10:39,  3.78it/s]
 65%|██████▍   | 4422/6837 [31:15<10:37,  3.79it/s]
 65%|██████▍   | 4423/6837 [31:16<10:37,  3.79it/s]
 65%|██████▍   | 4424/6837 [31:16<10:27,  3.85it/s]
 65%|██████▍   | 4425/6837 [31:16<10:10,  3.95it/s]
                                                   
{'loss': 4.3753, 'grad_norm': 0.15731613337993622, 'learning_rate': 0.0020015760324306048, 'epoch': 0.13}
+
 65%|██████▍   | 4425/6837 [31:16<10:10,  3.95it/s]
 65%|██████▍   | 4426/6837 [31:16<10:00,  4.02it/s]
 65%|██████▍   | 4427/6837 [31:17<09:51,  4.07it/s]
 65%|██████▍   | 4428/6837 [31:17<09:44,  4.12it/s]
 65%|██████▍   | 4429/6837 [31:17<09:39,  4.16it/s]
 65%|██████▍   | 4430/6837 [31:17<09:35,  4.18it/s]
 65%|██████▍   | 4431/6837 [31:18<09:32,  4.20it/s]
 65%|██████▍   | 4432/6837 [31:18<09:30,  4.21it/s]
 65%|██████▍   | 4433/6837 [31:18<09:28,  4.23it/s]
 65%|██████▍   | 4434/6837 [31:18<09:28,  4.23it/s]
 65%|██████▍   | 4435/6837 [31:19<09:27,  4.23it/s]
 65%|██████▍   | 4436/6837 [31:19<09:26,  4.24it/s]
 65%|██████▍   | 4437/6837 [31:19<09:26,  4.24it/s]
 65%|██████▍   | 4438/6837 [31:19<09:26,  4.24it/s]
 65%|██████▍   | 4439/6837 [31:20<09:25,  4.24it/s]
 65%|██████▍   | 4440/6837 [31:20<09:25,  4.24it/s]
 65%|██████▍   | 4441/6837 [31:20<09:24,  4.24it/s]
 65%|██████▍   | 4442/6837 [31:20<09:24,  4.24it/s]
 65%|██████▍   | 4443/6837 [31:20<09:24,  4.24it/s]
 65%|██████▍   | 4444/6837 [31:21<09:24,  4.24it/s]
 65%|██████▌   | 4445/6837 [31:21<09:23,  4.24it/s]
 65%|██████▌   | 4446/6837 [31:21<09:23,  4.24it/s]
 65%|██████▌   | 4447/6837 [31:21<09:22,  4.25it/s]
 65%|██████▌   | 4448/6837 [31:22<09:22,  4.25it/s]
 65%|██████▌   | 4449/6837 [31:22<09:22,  4.24it/s]
 65%|██████▌   | 4450/6837 [31:22<09:22,  4.24it/s]
                                                   
{'loss': 4.3839, 'grad_norm': 0.15175366401672363, 'learning_rate': 0.0019655478551331954, 'epoch': 0.13}
+
 65%|██████▌   | 4450/6837 [31:22<09:22,  4.24it/s]
 65%|██████▌   | 4451/6837 [31:22<09:24,  4.23it/s]
 65%|██████▌   | 4452/6837 [31:23<09:24,  4.23it/s]
 65%|██████▌   | 4453/6837 [31:23<09:23,  4.23it/s]
 65%|██████▌   | 4454/6837 [31:23<09:22,  4.24it/s]
 65%|██████▌   | 4455/6837 [31:23<09:21,  4.24it/s]
 65%|██████▌   | 4456/6837 [31:24<09:21,  4.24it/s]
 65%|██████▌   | 4457/6837 [31:24<09:21,  4.24it/s]
 65%|██████▌   | 4458/6837 [31:24<09:20,  4.24it/s]
 65%|██████▌   | 4459/6837 [31:24<09:21,  4.23it/s]
 65%|██████▌   | 4460/6837 [31:25<09:20,  4.24it/s]
 65%|██████▌   | 4461/6837 [31:25<09:19,  4.24it/s]
 65%|██████▌   | 4462/6837 [31:25<09:19,  4.24it/s]
 65%|██████▌   | 4463/6837 [31:25<09:19,  4.24it/s]
 65%|██████▌   | 4464/6837 [31:25<09:19,  4.24it/s]
 65%|██████▌   | 4465/6837 [31:26<09:19,  4.24it/s]
 65%|██████▌   | 4466/6837 [31:26<09:19,  4.24it/s]
 65%|██████▌   | 4467/6837 [31:26<09:18,  4.24it/s]
 65%|██████▌   | 4468/6837 [31:26<09:18,  4.24it/s]
 65%|██████▌   | 4469/6837 [31:27<09:18,  4.24it/s]
 65%|██████▌   | 4470/6837 [31:27<09:17,  4.24it/s]
 65%|██████▌   | 4471/6837 [31:27<09:16,  4.25it/s]
 65%|██████▌   | 4472/6837 [31:27<09:16,  4.25it/s]
 65%|██████▌   | 4473/6837 [31:28<09:15,  4.26it/s]
 65%|██████▌   | 4474/6837 [31:28<09:15,  4.25it/s]
 65%|██████▌   | 4475/6837 [31:28<09:14,  4.26it/s]
                                                   {'loss': 4.3856, 'grad_norm': 0.15692757070064545, 'learning_rate': 0.0019296882207272582, 'epoch': 0.13}
+
 65%|██████▌   | 4475/6837 [31:28<09:14,  4.26it/s]
 65%|██████▌   | 4476/6837 [31:28<09:15,  4.25it/s]
 65%|██████▌   | 4477/6837 [31:29<09:15,  4.25it/s]
 65%|██████▌   | 4478/6837 [31:29<09:14,  4.25it/s]
 66%|██████▌   | 4479/6837 [31:29<09:13,  4.26it/s]
 66%|██████▌   | 4480/6837 [31:29<09:14,  4.25it/s]
 66%|██████▌   | 4481/6837 [31:29<09:14,  4.25it/s]
 66%|██████▌   | 4482/6837 [31:30<09:13,  4.25it/s]
 66%|██████▌   | 4483/6837 [31:30<09:13,  4.25it/s]
 66%|██████▌   | 4484/6837 [31:30<09:14,  4.25it/s]
 66%|██████▌   | 4485/6837 [31:30<09:13,  4.25it/s]
 66%|██████▌   | 4486/6837 [31:31<09:14,  4.24it/s]
 66%|██████▌   | 4487/6837 [31:31<09:14,  4.24it/s]
 66%|██████▌   | 4488/6837 [31:31<09:13,  4.24it/s]
 66%|██████▌   | 4489/6837 [31:31<09:12,  4.25it/s]
 66%|██████▌   | 4490/6837 [31:32<09:12,  4.25it/s]
 66%|██████▌   | 4491/6837 [31:32<09:13,  4.24it/s]
 66%|██████▌   | 4492/6837 [31:32<09:13,  4.24it/s]
 66%|██████▌   | 4493/6837 [31:32<09:13,  4.24it/s]
 66%|██████▌   | 4494/6837 [31:33<09:13,  4.24it/s]
 66%|██████▌   | 4495/6837 [31:33<09:13,  4.23it/s]
 66%|██████▌   | 4496/6837 [31:33<09:13,  4.23it/s]
 66%|██████▌   | 4497/6837 [31:33<09:12,  4.24it/s]
 66%|██████▌   | 4498/6837 [31:33<09:12,  4.23it/s]
 66%|██████▌   | 4499/6837 [31:34<09:11,  4.24it/s]
 66%|██████▌   | 4500/6837 [31:34<09:11,  4.24it/s]
                                                   
{'loss': 4.363, 'grad_norm': 0.14754816889762878, 'learning_rate': 0.0018940029718092798, 'epoch': 0.13}
+
 66%|██████▌   | 4500/6837 [31:34<09:11,  4.24it/s]
 66%|██████▌   | 4501/6837 [31:34<09:13,  4.22it/s]
 66%|██████▌   | 4502/6837 [31:34<09:12,  4.22it/s]
 66%|██████▌   | 4503/6837 [31:35<09:12,  4.23it/s]
 66%|██████▌   | 4504/6837 [31:35<09:11,  4.23it/s]
 66%|██████▌   | 4505/6837 [31:35<09:10,  4.24it/s]
 66%|██████▌   | 4506/6837 [31:35<09:10,  4.24it/s]
 66%|██████▌   | 4507/6837 [31:36<09:10,  4.23it/s]
 66%|██████▌   | 4508/6837 [31:36<09:13,  4.20it/s]
 66%|██████▌   | 4509/6837 [31:36<09:13,  4.20it/s]
 66%|██████▌   | 4510/6837 [31:36<09:11,  4.22it/s]
 66%|██████▌   | 4511/6837 [31:37<09:10,  4.22it/s]
 66%|██████▌   | 4512/6837 [31:37<09:10,  4.22it/s]
 66%|██████▌   | 4513/6837 [31:37<09:09,  4.23it/s]
 66%|██████▌   | 4514/6837 [31:37<09:08,  4.23it/s]
 66%|██████▌   | 4515/6837 [31:37<09:08,  4.24it/s]
 66%|██████▌   | 4516/6837 [31:38<09:07,  4.24it/s]
 66%|██████▌   | 4517/6837 [31:38<09:06,  4.25it/s]
 66%|██████▌   | 4518/6837 [31:38<09:05,  4.25it/s]
 66%|██████▌   | 4519/6837 [31:38<09:05,  4.25it/s]
 66%|██████▌   | 4520/6837 [31:39<09:05,  4.25it/s]
 66%|██████▌   | 4521/6837 [31:39<09:05,  4.25it/s]
 66%|██████▌   | 4522/6837 [31:39<09:05,  4.24it/s]
 66%|██████▌   | 4523/6837 [31:39<09:06,  4.23it/s]
 66%|██████▌   | 4524/6837 [31:40<09:06,  4.23it/s]
 66%|██████▌   | 4525/6837 [31:40<09:05,  4.24it/s]
                                                   
{'loss': 4.3624, 'grad_norm': 0.14798496663570404, 'learning_rate': 0.0018584979225631905, 'epoch': 0.13}
+
 66%|██████▌   | 4525/6837 [31:40<09:05,  4.24it/s]
 66%|██████▌   | 4526/6837 [31:40<09:05,  4.24it/s]
 66%|██████▌   | 4527/6837 [31:40<09:05,  4.24it/s]
 66%|██████▌   | 4528/6837 [31:41<09:04,  4.24it/s]
 66%|██████▌   | 4529/6837 [31:41<09:03,  4.25it/s]
 66%|██████▋   | 4530/6837 [31:41<09:03,  4.25it/s]
 66%|██████▋   | 4531/6837 [31:41<09:03,  4.24it/s]
 66%|██████▋   | 4532/6837 [31:41<09:02,  4.25it/s]
 66%|██████▋   | 4533/6837 [31:42<09:01,  4.25it/s]
 66%|██████▋   | 4534/6837 [31:42<09:02,  4.25it/s]
 66%|██████▋   | 4535/6837 [31:42<09:01,  4.25it/s]
 66%|██████▋   | 4536/6837 [31:42<09:00,  4.25it/s]
 66%|██████▋   | 4537/6837 [31:43<09:00,  4.26it/s]
 66%|██████▋   | 4538/6837 [31:43<09:21,  4.10it/s]
 66%|██████▋   | 4539/6837 [31:43<09:39,  3.97it/s]
 66%|██████▋   | 4540/6837 [31:43<09:46,  3.91it/s]
 66%|██████▋   | 4541/6837 [31:44<09:53,  3.87it/s]
 66%|██████▋   | 4542/6837 [31:44<09:57,  3.84it/s]
 66%|██████▋   | 4543/6837 [31:44<10:01,  3.81it/s]
 66%|██████▋   | 4544/6837 [31:45<10:06,  3.78it/s]
 66%|██████▋   | 4545/6837 [31:45<10:06,  3.78it/s]
 66%|██████▋   | 4546/6837 [31:45<10:08,  3.76it/s]
 67%|██████▋   | 4547/6837 [31:45<10:03,  3.79it/s]
 67%|██████▋   | 4548/6837 [31:46<10:03,  3.79it/s]
 67%|██████▋   | 4549/6837 [31:46<10:03,  3.79it/s]
 67%|██████▋   | 4550/6837 [31:46<10:06,  3.77it/s]
                                                   
{'loss': 4.357, 'grad_norm': 0.14812426269054413, 'learning_rate': 0.0018231788578130553, 'epoch': 0.13}
+
 67%|██████▋   | 4550/6837 [31:46<10:06,  3.77it/s]
 67%|██████▋   | 4551/6837 [31:46<09:54,  3.85it/s]
 67%|██████▋   | 4552/6837 [31:47<09:39,  3.94it/s]
 67%|██████▋   | 4553/6837 [31:47<09:28,  4.02it/s]
 67%|██████▋   | 4554/6837 [31:47<09:19,  4.08it/s]
 67%|██████▋   | 4555/6837 [31:47<09:13,  4.12it/s]
 67%|██████▋   | 4556/6837 [31:48<09:09,  4.15it/s]
 67%|██████▋   | 4557/6837 [31:48<09:06,  4.17it/s]
 67%|██████▋   | 4558/6837 [31:48<09:04,  4.19it/s]
 67%|██████▋   | 4559/6837 [31:48<09:03,  4.19it/s]
 67%|██████▋   | 4560/6837 [31:48<09:01,  4.20it/s]
 67%|██████▋   | 4561/6837 [31:49<09:01,  4.20it/s]
 67%|██████▋   | 4562/6837 [31:49<09:00,  4.21it/s]
 67%|██████▋   | 4563/6837 [31:49<09:00,  4.21it/s]
 67%|██████▋   | 4564/6837 [31:49<09:19,  4.07it/s]
 67%|██████▋   | 4565/6837 [31:50<09:31,  3.97it/s]
 67%|██████▋   | 4566/6837 [31:50<09:40,  3.92it/s]
 67%|██████▋   | 4567/6837 [31:50<09:39,  3.92it/s]
 67%|██████▋   | 4568/6837 [31:50<09:26,  4.01it/s]
 67%|██████▋   | 4569/6837 [31:51<09:16,  4.07it/s]
 67%|██████▋   | 4570/6837 [31:51<09:11,  4.11it/s]
 67%|██████▋   | 4571/6837 [31:51<09:06,  4.15it/s]
 67%|██████▋   | 4572/6837 [31:51<09:02,  4.17it/s]
 67%|██████▋   | 4573/6837 [31:52<09:00,  4.19it/s]
 67%|██████▋   | 4574/6837 [31:52<08:58,  4.20it/s]
 67%|██████▋   | 4575/6837 [31:52<08:57,  4.21it/s]
                                                   
{'loss': 4.3635, 'grad_norm': 0.16064761579036713, 'learning_rate': 0.0017880515320805564, 'epoch': 0.13}
+
 67%|██████▋   | 4575/6837 [31:52<08:57,  4.21it/s]
 67%|██████▋   | 4576/6837 [31:52<08:58,  4.20it/s]
 67%|██████▋   | 4577/6837 [31:53<08:56,  4.21it/s]
 67%|██████▋   | 4578/6837 [31:53<08:55,  4.22it/s]
 67%|██████▋   | 4579/6837 [31:53<08:55,  4.22it/s]
 67%|██████▋   | 4580/6837 [31:53<08:53,  4.23it/s]
 67%|██████▋   | 4581/6837 [31:54<08:52,  4.23it/s]
 67%|██████▋   | 4582/6837 [31:54<08:53,  4.23it/s]
 67%|██████▋   | 4583/6837 [31:54<08:52,  4.23it/s]
 67%|██████▋   | 4584/6837 [31:54<08:52,  4.23it/s]
 67%|██████▋   | 4585/6837 [31:55<08:52,  4.23it/s]
 67%|████���█▋   | 4586/6837 [31:55<08:52,  4.23it/s]
 67%|██████▋   | 4587/6837 [31:55<08:51,  4.23it/s]
 67%|██████▋   | 4588/6837 [31:55<08:51,  4.23it/s]
 67%|██████▋   | 4589/6837 [31:55<08:52,  4.22it/s]
 67%|██████▋   | 4590/6837 [31:56<08:52,  4.22it/s]
 67%|██████▋   | 4591/6837 [31:56<08:52,  4.22it/s]
 67%|██████▋   | 4592/6837 [31:56<09:10,  4.08it/s]
 67%|██████▋   | 4593/6837 [31:56<09:25,  3.97it/s]
 67%|██████▋   | 4594/6837 [31:57<09:31,  3.92it/s]
 67%|██████▋   | 4595/6837 [31:57<09:37,  3.88it/s]
 67%|██████▋   | 4596/6837 [31:57<09:42,  3.84it/s]
 67%|██████▋   | 4597/6837 [31:58<09:46,  3.82it/s]
 67%|██████▋   | 4598/6837 [31:58<09:51,  3.78it/s]
 67%|██████▋   | 4599/6837 [31:58<09:47,  3.81it/s]
 67%|██████▋   | 4600/6837 [31:58<09:49,  3.80it/s]
                                                   
{'loss': 4.3666, 'grad_norm': 0.15804423391819, 'learning_rate': 0.0017531216686474112, 'epoch': 0.13}
+
 67%|██████▋   | 4600/6837 [31:58<09:49,  3.80it/s]
 67%|██████▋   | 4601/6837 [31:59<09:51,  3.78it/s]
 67%|██████▋   | 4602/6837 [31:59<09:51,  3.78it/s]
 67%|██████▋   | 4603/6837 [31:59<09:49,  3.79it/s]
 67%|██████▋   | 4604/6837 [31:59<09:49,  3.79it/s]
 67%|██████▋   | 4605/6837 [32:00<09:35,  3.88it/s]
 67%|██████▋   | 4606/6837 [32:00<09:21,  3.97it/s]
 67%|██████▋   | 4607/6837 [32:00<09:10,  4.05it/s]
 67%|██████▋   | 4608/6837 [32:00<09:03,  4.10it/s]
 67%|██████▋   | 4609/6837 [32:01<08:58,  4.14it/s]
 67%|██████▋   | 4610/6837 [32:01<08:54,  4.16it/s]
 67%|██████▋   | 4611/6837 [32:01<08:51,  4.19it/s]
 67%|██████▋   | 4612/6837 [32:01<08:49,  4.20it/s]
 67%|██████▋   | 4613/6837 [32:02<08:47,  4.22it/s]
 67%|██████▋   | 4614/6837 [32:02<08:47,  4.22it/s]
 68%|██████▊   | 4615/6837 [32:02<08:46,  4.22it/s]
 68%|██████▊   | 4616/6837 [32:02<08:45,  4.23it/s]
 68%|██████▊   | 4617/6837 [32:02<08:44,  4.23it/s]
 68%|██████▊   | 4618/6837 [32:03<09:03,  4.08it/s]
 68%|██████▊   | 4619/6837 [32:03<09:17,  3.98it/s]
 68%|██████▊   | 4620/6837 [32:03<09:24,  3.93it/s]
 68%|██████▊   | 4621/6837 [32:04<09:31,  3.88it/s]
 68%|██████▊   | 4622/6837 [32:04<09:36,  3.84it/s]
 68%|██████▊   | 4623/6837 [32:04<09:37,  3.83it/s]
 68%|██████▊   | 4624/6837 [32:04<09:36,  3.84it/s]
 68%|██████▊   | 4625/6837 [32:05<09:33,  3.86it/s]
                                                   
{'loss': 4.3635, 'grad_norm': 0.16471876204013824, 'learning_rate': 0.0017183949586228808, 'epoch': 0.14}
+
 68%|██████▊   | 4625/6837 [32:05<09:33,  3.86it/s]
 68%|██████▊   | 4626/6837 [32:05<09:37,  3.83it/s]
 68%|██████▊   | 4627/6837 [32:05<09:19,  3.95it/s]
 68%|██████▊   | 4628/6837 [32:05<09:07,  4.03it/s]
 68%|██████▊   | 4629/6837 [32:06<08:59,  4.09it/s]
 68%|██████▊   | 4630/6837 [32:06<08:53,  4.14it/s]
 68%|██████▊   | 4631/6837 [32:06<08:48,  4.17it/s]
 68%|██████▊   | 4632/6837 [32:06<08:45,  4.19it/s]
 68%|██████▊   | 4633/6837 [32:06<08:43,  4.21it/s]
 68%|██████▊   | 4634/6837 [32:07<08:41,  4.22it/s]
 68%|██████▊   | 4635/6837 [32:07<08:40,  4.23it/s]
 68%|██████▊   | 4636/6837 [32:07<08:39,  4.24it/s]
 68%|██████▊   | 4637/6837 [32:07<08:38,  4.25it/s]
 68%|██████▊   | 4638/6837 [32:08<08:36,  4.26it/s]
 68%|██████▊   | 4639/6837 [32:08<08:36,  4.25it/s]
 68%|██████▊   | 4640/6837 [32:08<08:36,  4.25it/s]
 68%|██████▊   | 4641/6837 [32:08<08:36,  4.25it/s]
 68%|██████▊   | 4642/6837 [32:09<08:35,  4.26it/s]
 68%|██████▊   | 4643/6837 [32:09<08:35,  4.26it/s]
 68%|██████▊   | 4644/6837 [32:09<08:35,  4.26it/s]
 68%|██████▊   | 4645/6837 [32:09<08:34,  4.26it/s]
 68%|██████▊   | 4646/6837 [32:10<08:34,  4.26it/s]
 68%|██████▊   | 4647/6837 [32:10<08:34,  4.26it/s]
 68%|██████▊   | 4648/6837 [32:10<08:33,  4.26it/s]
 68%|██████▊   | 4649/6837 [32:10<08:33,  4.26it/s]
 68%|██████▊   | 4650/6837 [32:10<08:33,  4.26it/s]
                                                   
{'loss': 4.367, 'grad_norm': 0.15519706904888153, 'learning_rate': 0.00168387706001652, 'epoch': 0.14}
+
 68%|██████▊   | 4650/6837 [32:10<08:33,  4.26it/s]
 68%|██████▊   | 4651/6837 [32:11<08:35,  4.24it/s]
 68%|██████▊   | 4652/6837 [32:11<08:34,  4.24it/s]
 68%|██████▊   | 4653/6837 [32:11<08:58,  4.05it/s]
 68%|██████▊   | 4654/6837 [32:11<09:11,  3.96it/s]
 68%|██████▊   | 4655/6837 [32:12<09:19,  3.90it/s]
 68%|██████▊   | 4656/6837 [32:12<09:22,  3.87it/s]
 68%|██████▊   | 4657/6837 [32:12<09:28,  3.84it/s]
 68%|██████▊   | 4658/6837 [32:13<09:35,  3.79it/s]
 68%|██████▊   | 4659/6837 [32:13<09:34,  3.79it/s]
 68%|██████▊   | 4660/6837 [32:13<09:33,  3.80it/s]
 68%|██████▊   | 4661/6837 [32:13<09:25,  3.85it/s]
 68%|██████▊   | 4662/6837 [32:14<09:09,  3.96it/s]
 68%|██████▊   | 4663/6837 [32:14<08:58,  4.04it/s]
 68%|██████▊   | 4664/6837 [32:14<08:50,  4.09it/s]
 68%|██████▊   | 4665/6837 [32:14<08:44,  4.14it/s]
 68%|██████▊   | 4666/6837 [32:15<08:40,  4.17it/s]
 68%|██████▊   | 4667/6837 [32:15<08:38,  4.19it/s]
 68%|██████▊   | 4668/6837 [32:15<08:36,  4.20it/s]
 68%|██████▊   | 4669/6837 [32:15<08:34,  4.21it/s]
 68%|██████▊   | 4670/6837 [32:15<08:33,  4.22it/s]
 68%|██████▊   | 4671/6837 [32:16<08:32,  4.22it/s]
 68%|██████▊   | 4672/6837 [32:16<08:31,  4.23it/s]
 68%|██████▊   | 4673/6837 [32:16<08:29,  4.24it/s]
 68%|██████▊   | 4674/6837 [32:16<08:29,  4.24it/s]
 68%|██████▊   | 4675/6837 [32:17<08:29,  4.25it/s]
                                                   
{'loss': 4.3535, 'grad_norm': 0.20246660709381104, 'learning_rate': 0.0016495735968163174, 'epoch': 0.14}
+
 68%|██████▊   | 4675/6837 [32:17<08:29,  4.25it/s]
 68%|██████▊   | 4676/6837 [32:17<08:29,  4.24it/s]
 68%|██████▊   | 4677/6837 [32:17<08:28,  4.24it/s]
 68%|██████▊   | 4678/6837 [32:17<08:29,  4.24it/s]
 68%|██████▊   | 4679/6837 [32:18<08:29,  4.23it/s]
 68%|██████▊   | 4680/6837 [32:18<08:28,  4.24it/s]
 68%|██████▊   | 4681/6837 [32:18<08:28,  4.24it/s]
 68%|██████▊   | 4682/6837 [32:18<08:28,  4.24it/s]
 68%|██████▊   | 4683/6837 [32:19<08:27,  4.24it/s]
 69%|██████▊   | 4684/6837 [32:19<08:27,  4.25it/s]
 69%|██████▊   | 4685/6837 [32:19<08:27,  4.24it/s]
 69%|██████▊   | 4686/6837 [32:19<08:27,  4.24it/s]
 69%|██████▊   | 4687/6837 [32:19<08:27,  4.24it/s]
 69%|██████▊   | 4688/6837 [32:20<08:27,  4.23it/s]
 69%|██████▊   | 4689/6837 [32:20<08:27,  4.24it/s]
 69%|██████▊   | 4690/6837 [32:20<08:26,  4.24it/s]
 69%|██████▊   | 4691/6837 [32:20<08:26,  4.23it/s]
 69%|██████▊   | 4692/6837 [32:21<08:27,  4.23it/s]
 69%|██████▊   | 4693/6837 [32:21<08:28,  4.22it/s]
 69%|██████▊   | 4694/6837 [32:21<08:27,  4.22it/s]
 69%|██████▊   | 4695/6837 [32:21<08:26,  4.23it/s]
 69%|██████▊   | 4696/6837 [32:22<08:26,  4.23it/s]
 69%|██████▊   | 4697/6837 [32:22<08:26,  4.23it/s]
 69%|██████▊   | 4698/6837 [32:22<08:25,  4.23it/s]
 69%|██████▊   | 4699/6837 [32:22<08:24,  4.24it/s]
 69%|██████▊   | 4700/6837 [32:23<08:24,  4.24it/s]
                                                   
{'loss': 4.372, 'grad_norm': 0.15515254437923431, 'learning_rate': 0.001615490158072385, 'epoch': 0.14}
+
 69%|██████▊   | 4700/6837 [32:23<08:24,  4.24it/s]
 69%|██████▉   | 4701/6837 [32:23<08:25,  4.22it/s]
 69%|██████▉   | 4702/6837 [32:23<08:25,  4.23it/s]
 69%|██████▉   | 4703/6837 [32:23<08:24,  4.23it/s]
 69%|██████▉   | 4704/6837 [32:23<08:24,  4.23it/s]
 69%|██████▉   | 4705/6837 [32:24<08:23,  4.23it/s]
 69%|██████▉   | 4706/6837 [32:24<08:23,  4.23it/s]
 69%|██████▉   | 4707/6837 [32:24<08:24,  4.22it/s]
 69%|██████▉   | 4708/6837 [32:24<08:23,  4.23it/s]
 69%|██████▉   | 4709/6837 [32:25<08:23,  4.23it/s]
 69%|██████▉   | 4710/6837 [32:25<08:24,  4.22it/s]
 69%|██████▉   | 4711/6837 [32:25<08:22,  4.23it/s]
 69%|██████▉   | 4712/6837 [32:25<08:22,  4.23it/s]
 69%|██████▉   | 4713/6837 [32:26<08:22,  4.23it/s]
 69%|██████▉   | 4714/6837 [32:26<08:21,  4.23it/s]
 69%|██████▉   | 4715/6837 [32:26<08:21,  4.23it/s]
 69%|██████▉   | 4716/6837 [32:26<08:21,  4.23it/s]
 69%|██████▉   | 4717/6837 [32:27<08:20,  4.23it/s]
 69%|██████▉   | 4718/6837 [32:27<08:20,  4.23it/s]
 69%|██████▉   | 4719/6837 [32:27<08:20,  4.24it/s]
 69%|██████▉   | 4720/6837 [32:27<08:19,  4.24it/s]
 69%|██████▉   | 4721/6837 [32:27<08:19,  4.24it/s]
 69%|██████▉   | 4722/6837 [32:28<08:19,  4.24it/s]
 69%|██████▉   | 4723/6837 [32:28<08:18,  4.24it/s]
 69%|██████▉   | 4724/6837 [32:28<08:18,  4.24it/s]
 69%|██████▉   | 4725/6837 [32:28<08:18,  4.23it/s]
                                                   
{'loss': 4.3559, 'grad_norm': 0.161404550075531, 'learning_rate': 0.001581632296986333, 'epoch': 0.14}
+
 69%|██████▉   | 4725/6837 [32:28<08:18,  4.23it/s]
 69%|██████▉   | 4726/6837 [32:29<08:20,  4.22it/s]
 69%|██████▉   | 4727/6837 [32:29<08:19,  4.22it/s]
 69%|██████▉   | 4728/6837 [32:29<08:19,  4.22it/s]
 69%|██████▉   | 4729/6837 [32:29<08:18,  4.23it/s]
 69%|██████▉   | 4730/6837 [32:30<08:18,  4.23it/s]
 69%|██████▉   | 4731/6837 [32:30<08:17,  4.23it/s]
 69%|██████▉   | 4732/6837 [32:30<08:17,  4.23it/s]
 69%|██████▉   | 4733/6837 [32:30<08:16,  4.23it/s]
 69%|██████▉   | 4734/6837 [32:31<08:15,  4.24it/s]
 69%|██████▉   | 4735/6837 [32:31<08:17,  4.23it/s]
 69%|██████▉   | 4736/6837 [32:31<08:17,  4.23it/s]
 69%|██████▉   | 4737/6837 [32:31<08:16,  4.23it/s]
 69%|██████▉   | 4738/6837 [32:32<08:15,  4.23it/s]
 69%|██████▉   | 4739/6837 [32:32<08:15,  4.23it/s]
 69%|██████▉   | 4740/6837 [32:32<08:14,  4.24it/s]
 69%|██████▉   | 4741/6837 [32:32<08:13,  4.25it/s]
 69%|██████▉   | 4742/6837 [32:32<08:13,  4.24it/s]
 69%|██████▉   | 4743/6837 [32:33<08:13,  4.24it/s]
 69%|██████▉   | 4744/6837 [32:33<08:13,  4.25it/s]
 69%|██████▉   | 4745/6837 [32:33<08:12,  4.25it/s]
 69%|██████▉   | 4746/6837 [32:33<08:13,  4.24it/s]
 69%|██████▉   | 4747/6837 [32:34<08:11,  4.25it/s]
 69%|██████▉   | 4748/6837 [32:34<08:11,  4.25it/s]
 69%|██████▉   | 4749/6837 [32:34<08:29,  4.10it/s]
 69%|██████▉   | 4750/6837 [32:34<08:41,  4.00it/s]
                                                   
{'loss': 4.3545, 'grad_norm': 0.1516709327697754, 'learning_rate': 0.0015480055300064908, 'epoch': 0.14}
+
 69%|██████▉   | 4750/6837 [32:34<08:41,  4.00it/s]
 69%|██████▉   | 4751/6837 [32:35<08:50,  3.94it/s]
 70%|██████▉   | 4752/6837 [32:35<08:56,  3.89it/s]
 70%|██████▉   | 4753/6837 [32:35<08:44,  3.97it/s]
 70%|██████▉   | 4754/6837 [32:35<08:34,  4.05it/s]
 70%|██████▉   | 4755/6837 [32:36<08:27,  4.10it/s]
 70%|██████▉   | 4756/6837 [32:36<08:23,  4.14it/s]
 70%|██████▉   | 4757/6837 [32:36<08:19,  4.17it/s]
 70%|██████▉   | 4758/6837 [32:36<08:16,  4.19it/s]
 70%|██████▉   | 4759/6837 [32:37<08:14,  4.20it/s]
 70%|██████▉   | 4760/6837 [32:37<08:12,  4.21it/s]
 70%|██████▉   | 4761/6837 [32:37<08:11,  4.23it/s]
 70%|██████▉   | 4762/6837 [32:37<08:10,  4.23it/s]
 70%|██████▉   | 4763/6837 [32:38<08:10,  4.23it/s]
 70%|██████▉   | 4764/6837 [32:38<08:23,  4.11it/s]
 70%|██████▉   | 4765/6837 [32:38<08:18,  4.16it/s]
 70%|██████▉   | 4766/6837 [32:38<08:14,  4.19it/s]
 70%|██████▉   | 4767/6837 [32:38<08:12,  4.20it/s]
 70%|██████▉   | 4768/6837 [32:39<08:12,  4.20it/s]
 70%|██████▉   | 4769/6837 [32:39<08:10,  4.22it/s]
 70%|██████▉   | 4770/6837 [32:39<08:09,  4.22it/s]
 70%|██████▉   | 4771/6837 [32:39<08:09,  4.22it/s]
 70%|██████▉   | 4772/6837 [32:40<08:08,  4.23it/s]
 70%|██████▉   | 4773/6837 [32:40<08:07,  4.23it/s]
 70%|██████▉   | 4774/6837 [32:40<08:06,  4.24it/s]
 70%|██████▉   | 4775/6837 [32:40<08:06,  4.24it/s]{'loss': 4.3464, 'grad_norm': 0.17767181992530823, 'learning_rate': 0.0015146153359291144, 'epoch': 0.14}
                                                   
+
 70%|██████▉   | 4775/6837 [32:40<08:06,  4.24it/s]
 70%|██████▉   | 4776/6837 [32:41<08:07,  4.23it/s]
 70%|██████▉   | 4777/6837 [32:41<08:05,  4.24it/s]
 70%|██████▉   | 4778/6837 [32:41<08:05,  4.24it/s]
 70%|██████▉   | 4779/6837 [32:41<08:04,  4.24it/s]
 70%|██████▉   | 4780/6837 [32:42<08:04,  4.24it/s]
 70%|██████▉   | 4781/6837 [32:42<08:04,  4.25it/s]
 70%|██████▉   | 4782/6837 [32:42<08:03,  4.25it/s]
 70%|██████▉   | 4783/6837 [32:42<08:03,  4.25it/s]
 70%|██████▉   | 4784/6837 [32:42<08:03,  4.25it/s]
 70%|██████▉   | 4785/6837 [32:43<08:21,  4.09it/s]
 70%|███████   | 4786/6837 [32:43<08:33,  3.99it/s]
 70%|███████   | 4787/6837 [32:43<08:43,  3.92it/s]
 70%|███████   | 4788/6837 [32:44<08:48,  3.88it/s]
 70%|███████   | 4789/6837 [32:44<08:50,  3.86it/s]
 70%|███████   | 4790/6837 [32:44<08:35,  3.97it/s]
 70%|███████   | 4791/6837 [32:44<08:25,  4.05it/s]
 70%|███████   | 4792/6837 [32:45<08:18,  4.10it/s]
 70%|███████   | 4793/6837 [32:45<08:13,  4.14it/s]
 70%|███████   | 4794/6837 [32:45<08:10,  4.17it/s]
 70%|███████   | 4795/6837 [32:45<08:07,  4.19it/s]
 70%|███████   | 4796/6837 [32:45<08:05,  4.20it/s]
 70%|███████   | 4797/6837 [32:46<08:03,  4.22it/s]
 70%|███████   | 4798/6837 [32:46<08:02,  4.23it/s]
 70%|███████   | 4799/6837 [32:46<08:02,  4.22it/s]
 70%|███████   | 4800/6837 [32:46<08:02,  4.22it/s]
                                                   
{'loss': 4.3384, 'grad_norm': 0.1568547934293747, 'learning_rate': 0.0014814671550057261, 'epoch': 0.14}
+
 70%|███████   | 4800/6837 [32:46<08:02,  4.22it/s]
 70%|███████   | 4801/6837 [32:47<08:03,  4.21it/s]
 70%|███████   | 4802/6837 [32:47<08:01,  4.22it/s]
 70%|███████   | 4803/6837 [32:47<08:02,  4.22it/s]
 70%|███████   | 4804/6837 [32:47<08:01,  4.23it/s]
 70%|███████   | 4805/6837 [32:48<08:00,  4.23it/s]
 70%|███████   | 4806/6837 [32:48<07:59,  4.24it/s]
 70%|███████   | 4807/6837 [32:48<07:58,  4.24it/s]
 70%|███████   | 4808/6837 [32:48<07:57,  4.25it/s]
 70%|███████   | 4809/6837 [32:49<07:58,  4.24it/s]
 70%|███████   | 4810/6837 [32:49<07:57,  4.24it/s]
 70%|███████   | 4811/6837 [32:49<07:56,  4.25it/s]
 70%|███████   | 4812/6837 [32:49<07:56,  4.25it/s]
 70%|███████   | 4813/6837 [32:49<07:56,  4.25it/s]
 70%|███████   | 4814/6837 [32:50<07:55,  4.25it/s]
 70%|███████   | 4815/6837 [32:50<07:55,  4.26it/s]
 70%|███████   | 4816/6837 [32:50<07:57,  4.24it/s]
 70%|███████   | 4817/6837 [32:50<08:10,  4.12it/s]
 70%|███████   | 4818/6837 [32:51<08:25,  3.99it/s]
 70%|███████   | 4819/6837 [32:51<08:33,  3.93it/s]
 70%|███████   | 4820/6837 [32:51<08:38,  3.89it/s]
 71%|███████   | 4821/6837 [32:52<08:43,  3.85it/s]
 71%|███████   | 4822/6837 [32:52<08:33,  3.92it/s]
 71%|███████   | 4823/6837 [32:52<08:22,  4.01it/s]
 71%|███████   | 4824/6837 [32:52<08:14,  4.07it/s]
 71%|███████   | 4825/6837 [32:52<08:08,  4.12it/s]
                                                   
{'loss': 4.3436, 'grad_norm': 0.1425759494304657, 'learning_rate': 0.0014485663880567387, 'epoch': 0.14}
+
 71%|███████   | 4825/6837 [32:52<08:08,  4.12it/s]
 71%|███████   | 4826/6837 [32:53<08:05,  4.14it/s]
 71%|███████   | 4827/6837 [32:53<08:01,  4.17it/s]
 71%|███████   | 4828/6837 [32:53<07:59,  4.19it/s]
 71%|███████   | 4829/6837 [32:53<07:57,  4.21it/s]
 71%|███████   | 4830/6837 [32:54<07:58,  4.19it/s]
 71%|███████   | 4831/6837 [32:54<07:58,  4.19it/s]
 71%|███████   | 4832/6837 [32:54<07:56,  4.21it/s]
 71%|███████   | 4833/6837 [32:54<07:54,  4.22it/s]
 71%|███████   | 4834/6837 [32:55<07:53,  4.23it/s]
 71%|███████   | 4835/6837 [32:55<07:53,  4.23it/s]
 71%|███████   | 4836/6837 [32:55<07:53,  4.23it/s]
 71%|███████   | 4837/6837 [32:55<07:54,  4.21it/s]
 71%|███████   | 4838/6837 [32:56<08:11,  4.07it/s]
 71%|███████   | 4839/6837 [32:56<08:21,  3.98it/s]
 71%|███████   | 4840/6837 [32:56<08:29,  3.92it/s]
 71%|███████   | 4841/6837 [32:56<08:34,  3.88it/s]
 71%|███████   | 4842/6837 [32:57<08:38,  3.84it/s]
 71%|███████   | 4843/6837 [32:57<08:41,  3.83it/s]
 71%|███████   | 4844/6837 [32:57<08:43,  3.81it/s]
 71%|███████   | 4845/6837 [32:57<08:45,  3.79it/s]
 71%|███████   | 4846/6837 [32:58<08:46,  3.78it/s]
 71%|███████   | 4847/6837 [32:58<08:44,  3.79it/s]
 71%|███████   | 4848/6837 [32:58<08:45,  3.78it/s]
 71%|███████   | 4849/6837 [32:58<08:46,  3.78it/s]
 71%|███████   | 4850/6837 [32:59<08:44,  3.79it/s]
                                                   
{'loss': 4.3434, 'grad_norm': 0.16639594733715057, 'learning_rate': 0.0014159183955915011, 'epoch': 0.14}
+
 71%|███████   | 4850/6837 [32:59<08:44,  3.79it/s]
 71%|███████   | 4851/6837 [32:59<08:46,  3.77it/s]
 71%|███████   | 4852/6837 [32:59<08:46,  3.77it/s]
 71%|███████   | 4853/6837 [33:00<08:46,  3.77it/s]
 71%|███████   | 4854/6837 [33:00<08:46,  3.77it/s]
 71%|███████   | 4855/6837 [33:00<08:44,  3.78it/s]
 71%|███████   | 4856/6837 [33:00<08:44,  3.78it/s]
 71%|███████   | 4857/6837 [33:01<08:30,  3.88it/s]
 71%|███████   | 4858/6837 [33:01<08:17,  3.98it/s]
 71%|███████   | 4859/6837 [33:01<08:08,  4.05it/s]
 71%|███████   | 4860/6837 [33:01<08:02,  4.09it/s]
 71%|███████   | 4861/6837 [33:02<07:58,  4.13it/s]
 71%|███████   | 4862/6837 [33:02<07:54,  4.16it/s]
 71%|███████   | 4863/6837 [33:02<07:52,  4.18it/s]
 71%|███████   | 4864/6837 [33:02<07:50,  4.20it/s]
 71%|███████   | 4865/6837 [33:02<07:48,  4.21it/s]
 71%|███████   | 4866/6837 [33:03<07:47,  4.22it/s]
 71%|███████   | 4867/6837 [33:03<07:47,  4.22it/s]
 71%|███████   | 4868/6837 [33:03<07:46,  4.22it/s]
 71%|███████   | 4869/6837 [33:03<07:46,  4.22it/s]
 71%|███████   | 4870/6837 [33:04<07:45,  4.23it/s]
 71%|███████   | 4871/6837 [33:04<07:44,  4.23it/s]
 71%|███████▏  | 4872/6837 [33:04<07:44,  4.23it/s]
 71%|███████▏  | 4873/6837 [33:04<07:43,  4.23it/s]
 71%|███████▏  | 4874/6837 [33:05<07:43,  4.24it/s]
 71%|███████▏  | 4875/6837 [33:05<07:43,  4.23it/s]
                                                   
{'loss': 4.3468, 'grad_norm': 0.15968827903270721, 'learning_rate': 0.0013835284969349116, 'epoch': 0.14}
+
 71%|███████▏  | 4875/6837 [33:05<07:43,  4.23it/s]
 71%|███████▏  | 4876/6837 [33:05<07:49,  4.18it/s]
 71%|███████▏  | 4877/6837 [33:05<07:47,  4.19it/s]
 71%|███████▏  | 4878/6837 [33:06<07:45,  4.21it/s]
 71%|███████▏  | 4879/6837 [33:06<07:44,  4.22it/s]
 71%|███████▏  | 4880/6837 [33:06<07:44,  4.22it/s]
 71%|███████▏  | 4881/6837 [33:06<07:42,  4.23it/s]
 71%|███████▏  | 4882/6837 [33:06<07:43,  4.22it/s]
 71%|███████▏  | 4883/6837 [33:07<07:44,  4.20it/s]
 71%|███████▏  | 4884/6837 [33:07<07:43,  4.21it/s]
 71%|███████▏  | 4885/6837 [33:07<07:42,  4.22it/s]
 71%|███████▏  | 4886/6837 [33:07<07:43,  4.21it/s]
 71%|███████▏  | 4887/6837 [33:08<07:42,  4.21it/s]
 71%|███████▏  | 4888/6837 [33:08<07:41,  4.22it/s]
 72%|███████▏  | 4889/6837 [33:08<07:40,  4.23it/s]
 72%|███████▏  | 4890/6837 [33:08<07:40,  4.23it/s]
 72%|███████▏  | 4891/6837 [33:09<07:39,  4.24it/s]
 72%|███████▏  | 4892/6837 [33:09<07:38,  4.24it/s]
 72%|███████▏  | 4893/6837 [33:09<07:38,  4.24it/s]
 72%|███████▏  | 4894/6837 [33:09<07:38,  4.23it/s]
 72%|███████▏  | 4895/6837 [33:10<07:38,  4.24it/s]
 72%|███████▏  | 4896/6837 [33:10<07:38,  4.23it/s]
 72%|███████▏  | 4897/6837 [33:10<07:38,  4.23it/s]
 72%|███████▏  | 4898/6837 [33:10<07:38,  4.23it/s]
 72%|███████▏  | 4899/6837 [33:11<07:41,  4.20it/s]
 72%|███████▏  | 4900/6837 [33:11<07:40,  4.21it/s]
                                                   
{'loss': 4.3507, 'grad_norm': 0.1706847995519638, 'learning_rate': 0.0013514019693607441, 'epoch': 0.14}
+
 72%|███████▏  | 4900/6837 [33:11<07:40,  4.21it/s]
 72%|███████▏  | 4901/6837 [33:11<07:41,  4.19it/s]
 72%|███████▏  | 4902/6837 [33:11<07:40,  4.20it/s]
 72%|███████▏  | 4903/6837 [33:11<07:39,  4.21it/s]
 72%|███████▏  | 4904/6837 [33:12<07:37,  4.22it/s]
 72%|███████▏  | 4905/6837 [33:12<07:37,  4.22it/s]
 72%|███████▏  | 4906/6837 [33:12<07:37,  4.22it/s]
 72%|███████▏  | 4907/6837 [33:12<07:36,  4.23it/s]
 72%|███████▏  | 4908/6837 [33:13<07:36,  4.23it/s]
 72%|███████▏  | 4909/6837 [33:13<07:37,  4.22it/s]
 72%|███████▏  | 4910/6837 [33:13<07:36,  4.22it/s]
 72%|███████▏  | 4911/6837 [33:13<07:35,  4.23it/s]
 72%|███████▏  | 4912/6837 [33:14<07:35,  4.23it/s]
 72%|███████▏  | 4913/6837 [33:14<07:34,  4.24it/s]
 72%|███████▏  | 4914/6837 [33:14<07:33,  4.24it/s]
 72%|███████▏  | 4915/6837 [33:14<07:34,  4.23it/s]
 72%|███████▏  | 4916/6837 [33:15<07:34,  4.23it/s]
 72%|███████▏  | 4917/6837 [33:15<07:33,  4.23it/s]
 72%|███████▏  | 4918/6837 [33:15<07:33,  4.24it/s]
 72%|███████▏  | 4919/6837 [33:15<07:32,  4.23it/s]
 72%|███████▏  | 4920/6837 [33:15<07:31,  4.25it/s]
 72%|███████▏  | 4921/6837 [33:16<07:30,  4.25it/s]
 72%|███████▏  | 4922/6837 [33:16<07:32,  4.24it/s]
 72%|███████▏  | 4923/6837 [33:16<07:32,  4.23it/s]
 72%|███████▏  | 4924/6837 [33:16<07:31,  4.23it/s]
 72%|███████▏  | 4925/6837 [33:17<07:31,  4.23it/s]
                                                   
{'loss': 4.3296, 'grad_norm': 0.14632856845855713, 'learning_rate': 0.0013195440472318226, 'epoch': 0.14}
+
 72%|███████▏  | 4925/6837 [33:17<07:31,  4.23it/s]
 72%|███████▏  | 4926/6837 [33:17<07:32,  4.22it/s]
 72%|███████▏  | 4927/6837 [33:17<07:31,  4.23it/s]
 72%|███████▏  | 4928/6837 [33:17<07:30,  4.23it/s]
 72%|███████▏  | 4929/6837 [33:18<07:31,  4.23it/s]
 72%|███████▏  | 4930/6837 [33:18<07:30,  4.23it/s]
 72%|███████▏  | 4931/6837 [33:18<07:30,  4.23it/s]
 72%|███████▏  | 4932/6837 [33:18<07:30,  4.23it/s]
 72%|███████▏  | 4933/6837 [33:19<07:29,  4.23it/s]
 72%|███████▏  | 4934/6837 [33:19<07:29,  4.24it/s]
 72%|███████▏  | 4935/6837 [33:19<07:28,  4.24it/s]
 72%|███████▏  | 4936/6837 [33:19<07:29,  4.23it/s]
 72%|███████▏  | 4937/6837 [33:20<07:28,  4.24it/s]
 72%|███████▏  | 4938/6837 [33:20<07:27,  4.24it/s]
 72%|███████▏  | 4939/6837 [33:20<10:18,  3.07it/s]
 72%|███████▏  | 4940/6837 [33:21<09:26,  3.35it/s]
 72%|███████▏  | 4941/6837 [33:21<08:50,  3.57it/s]
 72%|███████▏  | 4942/6837 [33:21<08:26,  3.74it/s]
 72%|███████▏  | 4943/6837 [33:21<08:08,  3.88it/s]
 72%|███████▏  | 4944/6837 [33:21<07:55,  3.98it/s]
 72%|███████▏  | 4945/6837 [33:22<07:46,  4.05it/s]
 72%|███████▏  | 4946/6837 [33:22<07:40,  4.11it/s]
 72%|███████▏  | 4947/6837 [33:22<07:35,  4.15it/s]
 72%|███████▏  | 4948/6837 [33:22<07:32,  4.18it/s]
 72%|███████▏  | 4949/6837 [33:23<07:29,  4.20it/s]
 72%|███████▏  | 4950/6837 [33:23<07:28,  4.21it/s]
                                                   
{'loss': 4.3515, 'grad_norm': 0.1598782241344452, 'learning_rate': 0.0012879599211471895, 'epoch': 0.14}
+
 72%|███████▏  | 4950/6837 [33:23<07:28,  4.21it/s]
 72%|███████▏  | 4951/6837 [33:23<07:28,  4.21it/s]
 72%|███████▏  | 4952/6837 [33:23<07:26,  4.22it/s]
 72%|███████▏  | 4953/6837 [33:24<07:26,  4.22it/s]
 72%|███████▏  | 4954/6837 [33:24<07:25,  4.23it/s]
 72%|███████▏  | 4955/6837 [33:24<07:24,  4.24it/s]
 72%|███████▏  | 4956/6837 [33:24<07:24,  4.23it/s]
 73%|███████▎  | 4957/6837 [33:25<07:24,  4.23it/s]
 73%|███████▎  | 4958/6837 [33:25<07:23,  4.23it/s]
 73%|███████▎  | 4959/6837 [33:25<07:23,  4.24it/s]
 73%|███████▎  | 4960/6837 [33:25<07:22,  4.24it/s]
 73%|███████▎  | 4961/6837 [33:25<07:22,  4.24it/s]
 73%|███████▎  | 4962/6837 [33:26<07:21,  4.24it/s]
 73%|███████▎  | 4963/6837 [33:26<07:24,  4.21it/s]
 73%|███████▎  | 4964/6837 [33:26<07:23,  4.22it/s]
 73%|███████▎  | 4965/6837 [33:26<07:22,  4.23it/s]
 73%|███████▎  | 4966/6837 [33:27<07:22,  4.23it/s]
 73%|███████▎  | 4967/6837 [33:27<07:22,  4.23it/s]
 73%|███████▎  | 4968/6837 [33:27<07:21,  4.23it/s]
 73%|███████▎  | 4969/6837 [33:27<07:20,  4.24it/s]
 73%|███████▎  | 4970/6837 [33:28<07:20,  4.24it/s]
 73%|███████▎  | 4971/6837 [33:28<07:20,  4.24it/s]
 73%|███████▎  | 4972/6837 [33:28<07:19,  4.24it/s]
 73%|███████▎  | 4973/6837 [33:28<07:19,  4.24it/s]
 73%|███████▎  | 4974/6837 [33:29<07:19,  4.24it/s]
 73%|███████▎  | 4975/6837 [33:29<07:19,  4.24it/s]
                                                   {'loss': 4.3423, 'grad_norm': 0.167702779173851, 'learning_rate': 0.001256654737096399, 'epoch': 0.15}
+

 73%|███████▎  | 4975/6837 [33:29<07:19,  4.24it/s]
 73%|███████▎  | 4976/6837 [33:29<07:21,  4.22it/s]
 73%|███████▎  | 4977/6837 [33:29<07:21,  4.21it/s]
 73%|███████▎  | 4978/6837 [33:29<07:22,  4.20it/s]
 73%|███████▎  | 4979/6837 [33:30<07:21,  4.21it/s]
 73%|███████▎  | 4980/6837 [33:30<07:20,  4.22it/s]
 73%|███████▎  | 4981/6837 [33:30<07:20,  4.22it/s]
 73%|███████▎  | 4982/6837 [33:30<07:19,  4.22it/s]
 73%|███████▎  | 4983/6837 [33:31<07:19,  4.22it/s]
 73%|███████▎  | 4984/6837 [33:31<07:18,  4.22it/s]
 73%|███████▎  | 4985/6837 [33:31<07:18,  4.23it/s]
 73%|███████▎  | 4986/6837 [33:31<07:18,  4.22it/s]
 73%|███████▎  | 4987/6837 [33:32<07:17,  4.23it/s]
 73%|███████▎  | 4988/6837 [33:32<07:16,  4.23it/s]
 73%|███████▎  | 4989/6837 [33:32<07:16,  4.23it/s]
 73%|███████▎  | 4990/6837 [33:32<07:16,  4.23it/s]
 73%|███████▎  | 4991/6837 [33:33<07:16,  4.23it/s]
 73%|███████▎  | 4992/6837 [33:33<07:16,  4.23it/s]
 73%|███████���  | 4993/6837 [33:33<07:15,  4.24it/s]
 73%|███████▎  | 4994/6837 [33:33<07:13,  4.25it/s]
 73%|███████▎  | 4995/6837 [33:34<07:13,  4.25it/s]
 73%|███████▎  | 4996/6837 [33:34<07:13,  4.25it/s]
 73%|███████▎  | 4997/6837 [33:34<07:13,  4.24it/s]
 73%|███████▎  | 4998/6837 [33:34<07:13,  4.24it/s]
 73%|███████▎  | 4999/6837 [33:34<07:14,  4.23it/s]
 73%|███████▎  | 5000/6837 [33:35<07:13,  4.24it/s]
                                                   
{'loss': 4.3393, 'grad_norm': 0.1614031344652176, 'learning_rate': 0.0012256335956210927, 'epoch': 0.15}
+
 73%|███████▎  | 5000/6837 [33:35<07:13,  4.24it/s]
 73%|███████▎  | 5001/6837 [33:35<07:14,  4.22it/s]
 73%|███████▎  | 5002/6837 [33:35<07:13,  4.23it/s]
 73%|███████▎  | 5003/6837 [33:35<07:29,  4.08it/s]
 73%|███████▎  | 5004/6837 [33:36<07:38,  4.00it/s]
 73%|███████▎  | 5005/6837 [33:36<07:30,  4.06it/s]
 73%|███████▎  | 5006/6837 [33:36<07:25,  4.11it/s]
 73%|███████▎  | 5007/6837 [33:36<07:21,  4.14it/s]
 73%|███████▎  | 5008/6837 [33:37<07:19,  4.17it/s]
 73%|███████▎  | 5009/6837 [33:37<07:17,  4.18it/s]
 73%|███████▎  | 5010/6837 [33:37<07:15,  4.19it/s]
 73%|███████▎  | 5011/6837 [33:37<07:14,  4.20it/s]
 73%|███████▎  | 5012/6837 [33:38<07:12,  4.21it/s]
 73%|███████▎  | 5013/6837 [33:38<07:12,  4.22it/s]
 73%|███████▎  | 5014/6837 [33:38<07:11,  4.22it/s]
 73%|███████▎  | 5015/6837 [33:38<07:10,  4.23it/s]
 73%|███████▎  | 5016/6837 [33:39<07:11,  4.22it/s]
 73%|███████▎  | 5017/6837 [33:39<07:10,  4.23it/s]
 73%|███████▎  | 5018/6837 [33:39<07:09,  4.23it/s]
 73%|███████▎  | 5019/6837 [33:39<07:09,  4.23it/s]
 73%|███████▎  | 5020/6837 [33:39<07:23,  4.10it/s]
 73%|███████▎  | 5021/6837 [33:40<07:23,  4.10it/s]
 73%|███████▎  | 5022/6837 [33:40<07:18,  4.14it/s]
 73%|███████▎  | 5023/6837 [33:40<07:15,  4.16it/s]
 73%|███████▎  | 5024/6837 [33:40<07:13,  4.18it/s]
 73%|███████▎  | 5025/6837 [33:41<07:10,  4.21it/s]
                                                   
{'loss': 4.3302, 'grad_norm': 0.15441203117370605, 'learning_rate': 0.0011949015509839602, 'epoch': 0.15}
+
 73%|███████▎  | 5025/6837 [33:41<07:10,  4.21it/s]
 74%|███████▎  | 5026/6837 [33:41<07:10,  4.20it/s]
 74%|███████▎  | 5027/6837 [33:41<07:09,  4.21it/s]
 74%|███████▎  | 5028/6837 [33:41<07:08,  4.22it/s]
 74%|███████▎  | 5029/6837 [33:42<07:07,  4.23it/s]
 74%|███████▎  | 5030/6837 [33:42<07:07,  4.23it/s]
 74%|███████▎  | 5031/6837 [33:42<07:07,  4.23it/s]
 74%|███████▎  | 5032/6837 [33:42<07:06,  4.23it/s]
 74%|███████▎  | 5033/6837 [33:43<07:06,  4.23it/s]
 74%|███████▎  | 5034/6837 [33:43<07:06,  4.22it/s]
 74%|███████▎  | 5035/6837 [33:43<07:23,  4.06it/s]
 74%|███████▎  | 5036/6837 [33:43<07:33,  3.97it/s]
 74%|███████▎  | 5037/6837 [33:44<07:38,  3.92it/s]
 74%|███████▎  | 5038/6837 [33:44<07:42,  3.89it/s]
 74%|███████▎  | 5039/6837 [33:44<07:46,  3.85it/s]
 74%|███████▎  | 5040/6837 [33:44<07:35,  3.94it/s]
 74%|███████▎  | 5041/6837 [33:45<07:25,  4.03it/s]
 74%|███████▎  | 5042/6837 [33:45<07:19,  4.09it/s]
 74%|███████▍  | 5043/6837 [33:45<07:13,  4.14it/s]
 74%|███████▍  | 5044/6837 [33:45<07:10,  4.16it/s]
 74%|███████▍  | 5045/6837 [33:46<07:08,  4.18it/s]
 74%|███████▍  | 5046/6837 [33:46<07:06,  4.20it/s]
 74%|███████▍  | 5047/6837 [33:46<07:06,  4.20it/s]
 74%|███████▍  | 5048/6837 [33:46<07:05,  4.21it/s]
 74%|███████▍  | 5049/6837 [33:47<07:04,  4.22it/s]
 74%|███████▍  | 5050/6837 [33:47<07:06,  4.19it/s]
                                                   
{'loss': 4.3347, 'grad_norm': 0.14135445654392242, 'learning_rate': 0.0011644636103452552, 'epoch': 0.15}
+
 74%|███████▍  | 5050/6837 [33:47<07:06,  4.19it/s]
 74%|███████▍  | 5051/6837 [33:47<07:05,  4.20it/s]
 74%|███████▍  | 5052/6837 [33:47<07:03,  4.21it/s]
 74%|███████▍  | 5053/6837 [33:47<07:03,  4.22it/s]
 74%|███████▍  | 5054/6837 [33:48<07:02,  4.22it/s]
 74%|███████▍  | 5055/6837 [33:48<07:01,  4.22it/s]
 74%|███████▍  | 5056/6837 [33:48<07:01,  4.23it/s]
 74%|███████▍  | 5057/6837 [33:48<07:01,  4.23it/s]
 74%|███████▍  | 5058/6837 [33:49<07:01,  4.22it/s]
 74%|███████▍  | 5059/6837 [33:49<07:02,  4.21it/s]
 74%|███████▍  | 5060/6837 [33:49<07:01,  4.21it/s]
 74%|███████▍  | 5061/6837 [33:49<07:01,  4.21it/s]
 74%|███████▍  | 5062/6837 [33:50<07:20,  4.03it/s]
 74%|███████▍  | 5063/6837 [33:50<07:32,  3.92it/s]
 74%|███████▍  | 5064/6837 [33:50<07:38,  3.87it/s]
 74%|███████▍  | 5065/6837 [33:50<07:42,  3.83it/s]
 74%|███████▍  | 5066/6837 [33:51<07:45,  3.81it/s]
 74%|███████▍  | 5067/6837 [33:51<07:46,  3.79it/s]
 74%|███████▍  | 5068/6837 [33:51<07:45,  3.80it/s]
 74%|███████▍  | 5069/6837 [33:51<07:44,  3.81it/s]
 74%|███████▍  | 5070/6837 [33:52<07:41,  3.83it/s]
 74%|███████▍  | 5071/6837 [33:52<07:47,  3.78it/s]
 74%|███████▍  | 5072/6837 [33:52<07:34,  3.88it/s]
 74%|███████▍  | 5073/6837 [33:52<07:23,  3.98it/s]
 74%|███████▍  | 5074/6837 [33:53<07:18,  4.02it/s]
 74%|███████▍  | 5075/6837 [33:53<07:12,  4.08it/s]
                                                   
{'loss': 4.3182, 'grad_norm': 0.1706584244966507, 'learning_rate': 0.0011343247329469795, 'epoch': 0.15}
+
 74%|███████▍  | 5075/6837 [33:53<07:12,  4.08it/s]
 74%|███████▍  | 5076/6837 [33:53<07:09,  4.10it/s]
 74%|███████▍  | 5077/6837 [33:53<07:07,  4.11it/s]
 74%|███████▍  | 5078/6837 [33:54<07:04,  4.15it/s]
 74%|███████▍  | 5079/6837 [33:54<07:01,  4.17it/s]
 74%|███████▍  | 5080/6837 [33:54<06:59,  4.19it/s]
 74%|███████▍  | 5081/6837 [33:54<06:58,  4.19it/s]
 74%|███████▍  | 5082/6837 [33:55<06:56,  4.21it/s]
 74%|███████▍  | 5083/6837 [33:55<06:55,  4.22it/s]
 74%|███████▍  | 5084/6837 [33:55<06:55,  4.22it/s]
 74%|███████▍  | 5085/6837 [33:55<06:55,  4.22it/s]
 74%|███████▍  | 5086/6837 [33:56<06:55,  4.22it/s]
 74%|███████▍  | 5087/6837 [33:56<06:55,  4.21it/s]
 74%|███████▍  | 5088/6837 [33:56<06:54,  4.22it/s]
 74%|███████▍  | 5089/6837 [33:56<06:54,  4.22it/s]
 74%|███████▍  | 5090/6837 [33:57<06:53,  4.23it/s]
 74%|███████▍  | 5091/6837 [33:57<06:53,  4.23it/s]
 74%|███████▍  | 5092/6837 [33:57<06:53,  4.22it/s]
 74%|███████▍  | 5093/6837 [33:57<06:52,  4.23it/s]
 75%|███████▍  | 5094/6837 [33:57<06:52,  4.23it/s]
 75%|███████▍  | 5095/6837 [33:58<06:52,  4.23it/s]
 75%|███████▍  | 5096/6837 [33:58<06:51,  4.23it/s]
 75%|███████▍  | 5097/6837 [33:58<06:51,  4.23it/s]
 75%|███████▍  | 5098/6837 [33:58<06:50,  4.23it/s]
 75%|███████▍  | 5099/6837 [33:59<06:51,  4.22it/s]
 75%|███████▍  | 5100/6837 [33:59<06:50,  4.23it/s]
                                                   
{'loss': 4.3179, 'grad_norm': 0.17500291764736176, 'learning_rate': 0.0011044898293048765, 'epoch': 0.15}
+
 75%|███████▍  | 5100/6837 [33:59<06:50,  4.23it/s]
 75%|███████▍  | 5101/6837 [33:59<06:52,  4.21it/s]
 75%|███████▍  | 5102/6837 [33:59<06:51,  4.21it/s]
 75%|███████▍  | 5103/6837 [34:00<06:51,  4.21it/s]
 75%|███████▍  | 5104/6837 [34:00<06:50,  4.22it/s]
 75%|███████▍  | 5105/6837 [34:00<06:50,  4.22it/s]
 75%|███████▍  | 5106/6837 [34:00<06:50,  4.22it/s]
 75%|███████▍  | 5107/6837 [34:01<06:50,  4.22it/s]
 75%|███████▍  | 5108/6837 [34:01<06:49,  4.23it/s]
 75%|███████▍  | 5109/6837 [34:01<06:48,  4.23it/s]
 75%|███████▍  | 5110/6837 [34:01<06:48,  4.23it/s]
 75%|███████▍  | 5111/6837 [34:02<06:47,  4.23it/s]
 75%|███████▍  | 5112/6837 [34:02<06:47,  4.24it/s]
 75%|███████▍  | 5113/6837 [34:02<06:48,  4.22it/s]
 75%|███████▍  | 5114/6837 [34:02<06:48,  4.22it/s]
 75%|███████▍  | 5115/6837 [34:02<06:48,  4.22it/s]
 75%|███████▍  | 5116/6837 [34:03<06:47,  4.22it/s]
 75%|███████▍  | 5117/6837 [34:03<06:47,  4.22it/s]
 75%|███████▍  | 5118/6837 [34:03<06:46,  4.23it/s]
 75%|███████▍  | 5119/6837 [34:03<06:46,  4.23it/s]
 75%|███████▍  | 5120/6837 [34:04<06:46,  4.22it/s]
 75%|███████▍  | 5121/6837 [34:04<06:46,  4.22it/s]
 75%|███████▍  | 5122/6837 [34:04<06:46,  4.22it/s]
 75%|███████▍  | 5123/6837 [34:04<06:46,  4.22it/s]
 75%|███████▍  | 5124/6837 [34:05<06:46,  4.22it/s]
 75%|███████▍  | 5125/6837 [34:05<06:45,  4.22it/s]
                                                   
{'loss': 4.3352, 'grad_norm': 0.14886602759361267, 'learning_rate': 0.0010749637604083582, 'epoch': 0.15}
+
 75%|███████▍  | 5125/6837 [34:05<06:45,  4.22it/s]
 75%|███████▍  | 5126/6837 [34:05<06:47,  4.20it/s]
 75%|███████▍  | 5127/6837 [34:05<06:46,  4.20it/s]
 75%|███████▌  | 5128/6837 [34:06<06:46,  4.20it/s]
 75%|███████▌  | 5129/6837 [34:06<06:46,  4.20it/s]
 75%|███████▌  | 5130/6837 [34:06<06:47,  4.19it/s]
 75%|███████▌  | 5131/6837 [34:06<06:46,  4.19it/s]
 75%|███████▌  | 5132/6837 [34:06<06:45,  4.20it/s]
 75%|███████▌  | 5133/6837 [34:07<06:45,  4.20it/s]
 75%|███████▌  | 5134/6837 [34:07<06:45,  4.20it/s]
 75%|███████▌  | 5135/6837 [34:07<06:57,  4.07it/s]
 75%|███████▌  | 5136/6837 [34:07<07:08,  3.97it/s]
 75%|███████▌  | 5137/6837 [34:08<07:16,  3.89it/s]
 75%|███████▌  | 5138/6837 [34:08<07:19,  3.87it/s]
 75%|███████▌  | 5139/6837 [34:08<07:22,  3.84it/s]
 75%|███████▌  | 5140/6837 [34:09<07:23,  3.83it/s]
 75%|███████▌  | 5141/6837 [34:09<07:25,  3.81it/s]
 75%|███████▌  | 5142/6837 [34:09<07:25,  3.80it/s]
 75%|███████▌  | 5143/6837 [34:09<07:26,  3.79it/s]
 75%|███████▌  | 5144/6837 [34:10<07:21,  3.84it/s]
 75%|███████▌  | 5145/6837 [34:10<07:09,  3.94it/s]
 75%|███████▌  | 5146/6837 [34:10<06:59,  4.03it/s]
 75%|███████▌  | 5147/6837 [34:10<06:53,  4.08it/s]
 75%|███████▌  | 5148/6837 [34:11<07:06,  3.96it/s]
 75%|███████▌  | 5149/6837 [34:11<07:12,  3.90it/s]
 75%|███████▌  | 5150/6837 [34:11<07:18,  3.85it/s]
                                                   
{'loss': 4.3402, 'grad_norm': 0.16718977689743042, 'learning_rate': 0.0010457513369285163, 'epoch': 0.15}
+
 75%|███████▌  | 5150/6837 [34:11<07:18,  3.85it/s]
 75%|███████▌  | 5151/6837 [34:11<07:22,  3.81it/s]
 75%|███████▌  | 5152/6837 [34:12<07:24,  3.79it/s]
 75%|███████▌  | 5153/6837 [34:12<07:27,  3.76it/s]
 75%|███████▌  | 5154/6837 [34:12<07:27,  3.76it/s]
 75%|███████▌  | 5155/6837 [34:12<07:20,  3.82it/s]
 75%|███████▌  | 5156/6837 [34:13<07:07,  3.93it/s]
 75%|███████▌  | 5157/6837 [34:13<07:12,  3.89it/s]
 75%|███████▌  | 5158/6837 [34:13<07:01,  3.98it/s]
 75%|███████▌  | 5159/6837 [34:13<06:53,  4.06it/s]
 75%|███████▌  | 5160/6837 [34:14<06:48,  4.11it/s]
 75%|███████▌  | 5161/6837 [34:14<06:44,  4.14it/s]
 76%|███████▌  | 5162/6837 [34:14<06:41,  4.17it/s]
 76%|███████▌  | 5163/6837 [34:14<06:39,  4.19it/s]
 76%|███████▌  | 5164/6837 [34:15<06:38,  4.20it/s]
 76%|███████▌  | 5165/6837 [34:15<06:36,  4.22it/s]
 76%|███████▌  | 5166/6837 [34:15<06:35,  4.22it/s]
 76%|███████▌  | 5167/6837 [34:15<06:34,  4.23it/s]
 76%|███████▌  | 5168/6837 [34:16<06:34,  4.23it/s]
 76%|███████▌  | 5169/6837 [34:16<06:33,  4.23it/s]
 76%|███████▌  | 5170/6837 [34:16<06:33,  4.23it/s]
 76%|███████▌  | 5171/6837 [34:16<06:34,  4.23it/s]
 76%|███████▌  | 5172/6837 [34:16<06:34,  4.22it/s]
 76%|███████▌  | 5173/6837 [34:17<06:33,  4.22it/s]
 76%|███████▌  | 5174/6837 [34:17<06:33,  4.23it/s]
 76%|███████▌  | 5175/6837 [34:17<06:33,  4.22it/s]
                                                   
{'loss': 4.3259, 'grad_norm': 0.17196804285049438, 'learning_rate': 0.0010168573184343105, 'epoch': 0.15}
+
 76%|███████▌  | 5175/6837 [34:17<06:33,  4.22it/s]
 76%|███████▌  | 5176/6837 [34:17<06:34,  4.21it/s]
 76%|███████▌  | 5177/6837 [34:18<06:33,  4.21it/s]
 76%|███████▌  | 5178/6837 [34:18<06:33,  4.22it/s]
 76%|███████▌  | 5179/6837 [34:18<06:32,  4.22it/s]
 76%|███████▌  | 5180/6837 [34:18<06:32,  4.22it/s]
 76%|███████▌  | 5181/6837 [34:19<06:32,  4.22it/s]
 76%|███████▌  | 5182/6837 [34:19<06:32,  4.21it/s]
 76%|███████▌  | 5183/6837 [34:19<06:32,  4.21it/s]
 76%|███████▌  | 5184/6837 [34:19<06:32,  4.21it/s]
 76%|███████▌  | 5185/6837 [34:20<06:31,  4.22it/s]
 76%|███████▌  | 5186/6837 [34:20<06:30,  4.22it/s]
 76%|███████▌  | 5187/6837 [34:20<06:30,  4.22it/s]
 76%|███████▌  | 5188/6837 [34:20<06:30,  4.22it/s]
 76%|███████▌  | 5189/6837 [34:21<06:30,  4.21it/s]
 76%|███████▌  | 5190/6837 [34:21<06:30,  4.22it/s]
 76%|███████▌  | 5191/6837 [34:21<06:30,  4.22it/s]
 76%|███████▌  | 5192/6837 [34:21<06:29,  4.22it/s]
 76%|███████▌  | 5193/6837 [34:21<06:29,  4.22it/s]
 76%|███████▌  | 5194/6837 [34:22<06:28,  4.23it/s]
 76%|███████▌  | 5195/6837 [34:22<06:27,  4.23it/s]
 76%|███████▌  | 5196/6837 [34:22<06:28,  4.22it/s]
 76%|███████▌  | 5197/6837 [34:22<06:28,  4.22it/s]
 76%|███████▌  | 5198/6837 [34:23<06:28,  4.22it/s]
 76%|███████▌  | 5199/6837 [34:23<06:28,  4.22it/s]
 76%|███████▌  | 5200/6837 [34:23<06:28,  4.22it/s]
                                                   
{'loss': 4.3245, 'grad_norm': 0.1394873857498169, 'learning_rate': 0.0009882864126170992, 'epoch': 0.15}
+
 76%|███████▌  | 5200/6837 [34:23<06:28,  4.22it/s]
 76%|███████▌  | 5201/6837 [34:23<06:29,  4.20it/s]
 76%|███████▌  | 5202/6837 [34:24<06:28,  4.21it/s]
 76%|███████▌  | 5203/6837 [34:24<06:28,  4.21it/s]
 76%|███████▌  | 5204/6837 [34:24<06:27,  4.21it/s]
 76%|███████▌  | 5205/6837 [34:24<06:27,  4.21it/s]
 76%|███████▌  | 5206/6837 [34:25<06:27,  4.21it/s]
 76%|███████▌  | 5207/6837 [34:25<06:26,  4.22it/s]
 76%|███████▌  | 5208/6837 [34:25<06:26,  4.21it/s]
 76%|███████▌  | 5209/6837 [34:25<06:26,  4.21it/s]
 76%|███████▌  | 5210/6837 [34:26<06:26,  4.21it/s]
 76%|███████▌  | 5211/6837 [34:26<06:26,  4.21it/s]
 76%|███████▌  | 5212/6837 [34:26<06:25,  4.21it/s]
 76%|███████▌  | 5213/6837 [34:26<06:37,  4.09it/s]
 76%|███████▋  | 5214/6837 [34:27<06:49,  3.97it/s]
 76%|███████▋  | 5215/6837 [34:27<06:58,  3.88it/s]
 76%|███████▋  | 5216/6837 [34:27<07:02,  3.84it/s]
 76%|███████▋  | 5217/6837 [34:27<07:08,  3.78it/s]
 76%|███████▋  | 5218/6837 [34:28<07:08,  3.77it/s]
 76%|███████▋  | 5219/6837 [34:28<07:08,  3.78it/s]
 76%|███████▋  | 5220/6837 [34:28<07:07,  3.78it/s]
 76%|███████▋  | 5221/6837 [34:28<07:07,  3.78it/s]
 76%|███████▋  | 5222/6837 [34:29<07:07,  3.78it/s]
 76%|███████▋  | 5223/6837 [34:29<07:06,  3.79it/s]
 76%|███████▋  | 5224/6837 [34:29<07:07,  3.77it/s]
 76%|███████▋  | 5225/6837 [34:29<07:08,  3.76it/s]
                                                   
{'loss': 4.3233, 'grad_norm': 0.14485645294189453, 'learning_rate': 0.0009600432745236163, 'epoch': 0.15}
+
 76%|███████▋  | 5225/6837 [34:29<07:08,  3.76it/s]
 76%|███████▋  | 5226/6837 [34:30<07:08,  3.76it/s]
 76%|███████▋  | 5227/6837 [34:30<07:07,  3.77it/s]
 76%|███████▋  | 5228/6837 [34:30<07:10,  3.74it/s]
 76%|███████▋  | 5229/6837 [34:31<07:08,  3.75it/s]
 76%|███████▋  | 5230/6837 [34:31<07:01,  3.81it/s]
 77%|███████▋  | 5231/6837 [34:31<06:48,  3.93it/s]
 77%|███████▋  | 5232/6837 [34:31<06:39,  4.01it/s]
 77%|███████▋  | 5233/6837 [34:31<06:34,  4.07it/s]
 77%|███████▋  | 5234/6837 [34:32<06:29,  4.11it/s]
 77%|███████▋  | 5235/6837 [34:32<06:26,  4.14it/s]
 77%|███████▋  | 5236/6837 [34:32<06:24,  4.17it/s]
 77%|███████▋  | 5237/6837 [34:32<06:22,  4.18it/s]
 77%|███████▋  | 5238/6837 [34:33<06:21,  4.19it/s]
 77%|███████▋  | 5239/6837 [34:33<06:20,  4.20it/s]
 77%|███████▋  | 5240/6837 [34:33<06:18,  4.21it/s]
 77%|███████▋  | 5241/6837 [34:33<06:17,  4.22it/s]
 77%|███████▋  | 5242/6837 [34:34<06:17,  4.23it/s]
 77%|███████▋  | 5243/6837 [34:34<06:16,  4.23it/s]
 77%|███████▋  | 5244/6837 [34:34<06:16,  4.23it/s]
 77%|███████▋  | 5245/6837 [34:34<06:16,  4.23it/s]
 77%|███████▋  | 5246/6837 [34:35<06:15,  4.24it/s]
 77%|███████▋  | 5247/6837 [34:35<06:29,  4.09it/s]
 77%|███████▋  | 5248/6837 [34:35<06:38,  3.99it/s]
 77%|███████▋  | 5249/6837 [34:35<06:46,  3.91it/s]
 77%|███████▋  | 5250/6837 [34:36<06:49,  3.87it/s]
                                                   
{'loss': 4.319, 'grad_norm': 0.16479887068271637, 'learning_rate': 0.0009321325057975276, 'epoch': 0.15}
+
 77%|███████▋  | 5250/6837 [34:36<06:49,  3.87it/s]
 77%|███████▋  | 5251/6837 [34:36<06:41,  3.95it/s]
 77%|███████▋  | 5252/6837 [34:36<06:48,  3.88it/s]
 77%|███████▋  | 5253/6837 [34:36<06:37,  3.98it/s]
 77%|███████▋  | 5254/6837 [34:37<06:30,  4.05it/s]
 77%|███████▋  | 5255/6837 [34:37<06:25,  4.10it/s]
 77%|███████▋  | 5256/6837 [34:37<06:21,  4.14it/s]
 77%|███████▋  | 5257/6837 [34:37<06:18,  4.17it/s]
 77%|███████▋  | 5258/6837 [34:38<06:16,  4.19it/s]
 77%|███████▋  | 5259/6837 [34:38<06:15,  4.20it/s]
 77%|███████▋  | 5260/6837 [34:38<06:14,  4.22it/s]
 77%|███████▋  | 5261/6837 [34:38<06:13,  4.22it/s]
 77%|███████▋  | 5262/6837 [34:38<06:12,  4.23it/s]
 77%|███████▋  | 5263/6837 [34:39<06:12,  4.23it/s]
 77%|███████▋  | 5264/6837 [34:39<06:11,  4.23it/s]
 77%|███████▋  | 5265/6837 [34:39<06:12,  4.22it/s]
 77%|███████▋  | 5266/6837 [34:39<06:13,  4.21it/s]
 77%|███████▋  | 5267/6837 [34:40<06:13,  4.20it/s]
 77%|███████▋  | 5268/6837 [34:40<06:12,  4.21it/s]
 77%|███████▋  | 5269/6837 [34:40<06:12,  4.21it/s]
 77%|███████▋  | 5270/6837 [34:40<06:11,  4.22it/s]
 77%|███████▋  | 5271/6837 [34:41<06:10,  4.22it/s]
 77%|███████▋  | 5272/6837 [34:41<06:11,  4.22it/s]
 77%|███████▋  | 5273/6837 [34:41<06:10,  4.22it/s]
 77%|███████▋  | 5274/6837 [34:41<06:10,  4.22it/s]
 77%|███████▋  | 5275/6837 [34:42<06:10,  4.22it/s]
                                                   
{'loss': 4.3029, 'grad_norm': 0.15019620954990387, 'learning_rate': 0.0009045586539296815, 'epoch': 0.15}
+
 77%|███████▋  | 5275/6837 [34:42<06:10,  4.22it/s]
 77%|███████▋  | 5276/6837 [34:42<06:16,  4.14it/s]
 77%|███████▋  | 5277/6837 [34:42<06:14,  4.17it/s]
 77%|███████▋  | 5278/6837 [34:42<06:12,  4.19it/s]
 77%|███████▋  | 5279/6837 [34:43<06:12,  4.18it/s]
 77%|███████▋  | 5280/6837 [34:43<06:25,  4.04it/s]
 77%|███████▋  | 5281/6837 [34:43<06:34,  3.95it/s]
 77%|███████▋  | 5282/6837 [34:43<06:25,  4.03it/s]
 77%|███████▋  | 5283/6837 [34:44<06:19,  4.10it/s]
 77%|███████▋  | 5284/6837 [34:44<06:15,  4.14it/s]
 77%|███████▋  | 5285/6837 [34:44<06:12,  4.17it/s]
 77%|███████▋  | 5286/6837 [34:44<06:11,  4.18it/s]
 77%|███████▋  | 5287/6837 [34:44<06:09,  4.19it/s]
 77%|███████▋  | 5288/6837 [34:45<06:09,  4.19it/s]
 77%|███████▋  | 5289/6837 [34:45<06:08,  4.20it/s]
 77%|███████▋  | 5290/6837 [34:45<06:08,  4.20it/s]
 77%|███████▋  | 5291/6837 [34:45<06:07,  4.21it/s]
 77%|███████▋  | 5292/6837 [34:46<06:07,  4.21it/s]
 77%|███████▋  | 5293/6837 [34:46<06:06,  4.21it/s]
 77%|███████▋  | 5294/6837 [34:46<06:06,  4.21it/s]
 77%|███████▋  | 5295/6837 [34:46<06:05,  4.22it/s]
 77%|███████▋  | 5296/6837 [34:47<06:05,  4.21it/s]
 77%|███████▋  | 5297/6837 [34:47<06:06,  4.21it/s]
 77%|███████▋  | 5298/6837 [34:47<06:05,  4.21it/s]
 78%|███████▊  | 5299/6837 [34:47<06:05,  4.21it/s]
 78%|███████▊  | 5300/6837 [34:48<06:05,  4.20it/s]
                                                   
{'loss': 4.3355, 'grad_norm': 0.15404148399829865, 'learning_rate': 0.0008773262115171941, 'epoch': 0.16}
+
 78%|███████▊  | 5300/6837 [34:48<06:05,  4.20it/s]
 78%|███████▊  | 5301/6837 [34:48<06:06,  4.19it/s]
 78%|███████▊  | 5302/6837 [34:48<06:05,  4.20it/s]
 78%|███████▊  | 5303/6837 [34:48<06:04,  4.21it/s]
 78%|███████▊  | 5304/6837 [34:49<06:03,  4.22it/s]
 78%|███████▊  | 5305/6837 [34:49<06:03,  4.22it/s]
 78%|███████▊  | 5306/6837 [34:49<06:02,  4.22it/s]
 78%|███████▊  | 5307/6837 [34:49<06:03,  4.21it/s]
 78%|███████▊  | 5308/6837 [34:49<06:02,  4.21it/s]
 78%|███████▊  | 5309/6837 [34:50<06:02,  4.22it/s]
 78%|███████▊  | 5310/6837 [34:50<06:01,  4.22it/s]
 78%|███████▊  | 5311/6837 [34:50<06:01,  4.22it/s]
 78%|███████▊  | 5312/6837 [34:50<06:01,  4.22it/s]
 78%|███████▊  | 5313/6837 [34:51<06:01,  4.22it/s]
 78%|███████▊  | 5314/6837 [34:51<06:01,  4.22it/s]
 78%|███████▊  | 5315/6837 [34:51<06:00,  4.22it/s]
 78%|███████▊  | 5316/6837 [34:51<06:00,  4.22it/s]
 78%|███████▊  | 5317/6837 [34:52<06:00,  4.22it/s]
 78%|███████▊  | 5318/6837 [34:52<06:00,  4.22it/s]
 78%|███████▊  | 5319/6837 [34:52<05:59,  4.22it/s]
 78%|███████▊  | 5320/6837 [34:52<05:59,  4.22it/s]
 78%|███████▊  | 5321/6837 [34:53<05:59,  4.21it/s]
 78%|███████▊  | 5322/6837 [34:53<06:00,  4.21it/s]
 78%|███████▊  | 5323/6837 [34:53<05:59,  4.21it/s]
 78%|███████▊  | 5324/6837 [34:53<05:59,  4.21it/s]
 78%|███████▊  | 5325/6837 [34:53<05:58,  4.21it/s]{'loss': 4.3336, 'grad_norm': 0.15235549211502075, 'learning_rate': 0.0008504396155314719, 'epoch': 0.16}
+
                                                   

 78%|███████▊  | 5325/6837 [34:53<05:58,  4.21it/s]
 78%|███████▊  | 5326/6837 [34:54<05:59,  4.20it/s]
 78%|███████▊  | 5327/6837 [34:54<05:58,  4.21it/s]
 78%|███████▊  | 5328/6837 [34:54<05:58,  4.21it/s]
 78%|███████▊  | 5329/6837 [34:54<06:14,  4.02it/s]
 78%|███████▊  | 5330/6837 [34:55<06:23,  3.93it/s]
 78%|███████▊  | 5331/6837 [34:55<06:21,  3.95it/s]
 78%|███████▊  | 5332/6837 [34:55<06:13,  4.03it/s]
 78%|███████▊  | 5333/6837 [34:55<06:07,  4.09it/s]
 78%|███████▊  | 5334/6837 [34:56<06:03,  4.14it/s]
 78%|███████▊  | 5335/6837 [34:56<06:00,  4.17it/s]
 78%|███████▊  | 5336/6837 [34:56<05:58,  4.19it/s]
 78%|███████▊  | 5337/6837 [34:56<05:56,  4.21it/s]
 78%|███████▊  | 5338/6837 [34:57<05:55,  4.22it/s]
 78%|███████▊  | 5339/6837 [34:57<05:54,  4.23it/s]
 78%|███████▊  | 5340/6837 [34:57<05:53,  4.24it/s]
 78%|███████▊  | 5341/6837 [34:57<05:52,  4.24it/s]
 78%|███████▊  | 5342/6837 [34:58<05:52,  4.24it/s]
 78%|███████▊  | 5343/6837 [34:58<05:52,  4.24it/s]
 78%|███████▊  | 5344/6837 [34:58<05:52,  4.24it/s]
 78%|███████▊  | 5345/6837 [34:58<05:52,  4.23it/s]
 78%|███████▊  | 5346/6837 [34:59<05:52,  4.23it/s]
 78%|███████▊  | 5347/6837 [34:59<05:52,  4.23it/s]
 78%|███████▊  | 5348/6837 [34:59<05:51,  4.23it/s]
 78%|███████▊  | 5349/6837 [34:59<05:52,  4.22it/s]
 78%|███████▊  | 5350/6837 [34:59<05:51,  4.23it/s]
                                                   
{'loss': 4.3149, 'grad_norm': 0.1456640362739563, 'learning_rate': 0.0008239032465952924, 'epoch': 0.16}
+
 78%|███████▊  | 5350/6837 [34:59<05:51,  4.23it/s]
 78%|███████▊  | 5351/6837 [35:00<05:52,  4.22it/s]
 78%|███████▊  | 5352/6837 [35:00<05:52,  4.22it/s]
 78%|███████▊  | 5353/6837 [35:00<05:51,  4.22it/s]
 78%|███████▊  | 5354/6837 [35:00<05:51,  4.22it/s]
 78%|███████▊  | 5355/6837 [35:01<05:51,  4.22it/s]
 78%|███████▊  | 5356/6837 [35:01<05:50,  4.22it/s]
 78%|███████▊  | 5357/6837 [35:01<05:51,  4.21it/s]
 78%|███████▊  | 5358/6837 [35:01<05:50,  4.22it/s]
 78%|███████▊  | 5359/6837 [35:02<05:49,  4.22it/s]
 78%|███████▊  | 5360/6837 [35:02<05:50,  4.22it/s]
 78%|███████▊  | 5361/6837 [35:02<05:49,  4.22it/s]
 78%|███████▊  | 5362/6837 [35:02<05:49,  4.23it/s]
 78%|███████▊  | 5363/6837 [35:03<05:48,  4.22it/s]
 78%|███████▊  | 5364/6837 [35:03<05:49,  4.22it/s]
 78%|███████▊  | 5365/6837 [35:03<05:48,  4.22it/s]
 78%|███████▊  | 5366/6837 [35:03<05:48,  4.22it/s]
 78%|███████▊  | 5367/6837 [35:04<05:48,  4.22it/s]
 79%|███████▊  | 5368/6837 [35:04<05:47,  4.22it/s]
 79%|███████▊  | 5369/6837 [35:04<05:47,  4.23it/s]
 79%|███████▊  | 5370/6837 [35:04<05:47,  4.22it/s]
 79%|███████▊  | 5371/6837 [35:04<05:47,  4.22it/s]
 79%|███████▊  | 5372/6837 [35:05<05:47,  4.22it/s]
 79%|███████▊  | 5373/6837 [35:05<05:47,  4.22it/s]
 79%|███████▊  | 5374/6837 [35:05<05:47,  4.21it/s]
 79%|███████▊  | 5375/6837 [35:05<05:47,  4.21it/s]
                                                   
{'loss': 4.312, 'grad_norm': 0.14772243797779083, 'learning_rate': 0.0007977214282690755, 'epoch': 0.16}
+
 79%|███████▊  | 5375/6837 [35:05<05:47,  4.21it/s]
 79%|███████▊  | 5376/6837 [35:06<05:47,  4.20it/s]
 79%|███████▊  | 5377/6837 [35:06<05:46,  4.21it/s]
 79%|███████▊  | 5378/6837 [35:06<05:46,  4.21it/s]
 79%|███████▊  | 5379/6837 [35:06<05:46,  4.21it/s]
 79%|███████▊  | 5380/6837 [35:07<05:45,  4.22it/s]
 79%|███████▊  | 5381/6837 [35:07<05:45,  4.22it/s]
 79%|███████▊  | 5382/6837 [35:07<05:44,  4.22it/s]
 79%|███████▊  | 5383/6837 [35:07<05:44,  4.22it/s]
 79%|███████▊  | 5384/6837 [35:08<05:43,  4.23it/s]
 79%|███████▉  | 5385/6837 [35:08<05:44,  4.22it/s]
 79%|███████▉  | 5386/6837 [35:08<05:43,  4.22it/s]
 79%|███████▉  | 5387/6837 [35:08<05:43,  4.22it/s]
 79%|███████▉  | 5388/6837 [35:08<05:43,  4.21it/s]
 79%|███████▉  | 5389/6837 [35:09<05:43,  4.22it/s]
 79%|███████▉  | 5390/6837 [35:09<05:43,  4.22it/s]
 79%|███████▉  | 5391/6837 [35:09<05:42,  4.22it/s]
 79%|███████▉  | 5392/6837 [35:09<05:43,  4.21it/s]
 79%|███████▉  | 5393/6837 [35:10<05:42,  4.21it/s]
 79%|███████▉  | 5394/6837 [35:10<05:42,  4.21it/s]
 79%|███████▉  | 5395/6837 [35:10<05:41,  4.22it/s]
 79%|███████▉  | 5396/6837 [35:10<05:41,  4.22it/s]
 79%|███████▉  | 5397/6837 [35:11<05:40,  4.23it/s]
 79%|███████▉  | 5398/6837 [35:11<05:40,  4.23it/s]
 79%|███████▉  | 5399/6837 [35:11<05:39,  4.23it/s]
 79%|███████▉  | 5400/6837 [35:11<05:40,  4.22it/s]
                                                   
{'loss': 4.3111, 'grad_norm': 0.14781531691551208, 'learning_rate': 0.0007718984263464471, 'epoch': 0.16}
+
 79%|███████▉  | 5400/6837 [35:11<05:40,  4.22it/s]
 79%|███████▉  | 5401/6837 [35:12<05:41,  4.21it/s]
 79%|███████▉  | 5402/6837 [35:12<05:40,  4.21it/s]
 79%|███████▉  | 5403/6837 [35:12<05:40,  4.21it/s]
 79%|███████▉  | 5404/6837 [35:12<05:39,  4.22it/s]
 79%|███████▉  | 5405/6837 [35:13<05:38,  4.22it/s]
 79%|███████▉  | 5406/6837 [35:13<05:51,  4.07it/s]
 79%|███████▉  | 5407/6837 [35:13<05:59,  3.97it/s]
 79%|███████▉  | 5408/6837 [35:13<05:53,  4.05it/s]
 79%|███████▉  | 5409/6837 [35:14<05:48,  4.10it/s]
 79%|███████▉  | 5410/6837 [35:14<05:45,  4.13it/s]
 79%|███████▉  | 5411/6837 [35:14<05:42,  4.16it/s]
 79%|███████▉  | 5412/6837 [35:14<05:40,  4.18it/s]
 79%|███████▉  | 5413/6837 [35:14<05:39,  4.19it/s]
 79%|███████▉  | 5414/6837 [35:15<05:38,  4.21it/s]
 79%|███████▉  | 5415/6837 [35:15<05:37,  4.21it/s]
 79%|███████▉  | 5416/6837 [35:15<05:36,  4.22it/s]
 79%|███████▉  | 5417/6837 [35:15<05:36,  4.22it/s]
 79%|███████▉  | 5418/6837 [35:16<05:35,  4.23it/s]
 79%|███████▉  | 5419/6837 [35:16<05:35,  4.23it/s]
 79%|███████▉  | 5420/6837 [35:16<05:35,  4.23it/s]
 79%|███████▉  | 5421/6837 [35:16<05:35,  4.23it/s]
 79%|███████▉  | 5422/6837 [35:17<05:34,  4.23it/s]
 79%|███████▉  | 5423/6837 [35:17<05:34,  4.23it/s]
 79%|███████▉  | 5424/6837 [35:17<05:35,  4.22it/s]
 79%|███████▉  | 5425/6837 [35:17<05:34,  4.22it/s]
                                                   
{'loss': 4.299, 'grad_norm': 0.15480075776576996, 'learning_rate': 0.0007464384481592105, 'epoch': 0.16}
+
 79%|███████▉  | 5425/6837 [35:17<05:34,  4.22it/s]
 79%|███████▉  | 5426/6837 [35:18<05:34,  4.21it/s]
 79%|███████▉  | 5427/6837 [35:18<05:34,  4.22it/s]
 79%|███████▉  | 5428/6837 [35:18<05:34,  4.21it/s]
 79%|███████▉  | 5429/6837 [35:18<05:34,  4.21it/s]
 79%|███████▉  | 5430/6837 [35:18<05:33,  4.22it/s]
 79%|███████▉  | 5431/6837 [35:19<05:33,  4.22it/s]
 79%|███████▉  | 5432/6837 [35:19<05:32,  4.22it/s]
 79%|███████▉  | 5433/6837 [35:19<05:44,  4.07it/s]
 79%|███████▉  | 5434/6837 [35:20<05:52,  3.98it/s]
 79%|███████▉  | 5435/6837 [35:20<05:57,  3.92it/s]
 80%|███████▉  | 5436/6837 [35:20<06:18,  3.71it/s]
 80%|███████▉  | 5437/6837 [35:20<06:13,  3.75it/s]
 80%|███████▉  | 5438/6837 [35:21<06:13,  3.75it/s]
 80%|███████▉  | 5439/6837 [35:21<06:01,  3.87it/s]
 80%|███████▉  | 5440/6837 [35:21<05:51,  3.98it/s]
 80%|███████▉  | 5441/6837 [35:21<05:44,  4.05it/s]
 80%|███████▉  | 5442/6837 [35:22<05:39,  4.11it/s]
 80%|███████▉  | 5443/6837 [35:22<05:36,  4.15it/s]
 80%|███████▉  | 5444/6837 [35:22<05:33,  4.17it/s]
 80%|███████▉  | 5445/6837 [35:22<05:31,  4.20it/s]
 80%|███████▉  | 5446/6837 [35:22<05:30,  4.21it/s]
 80%|███████▉  | 5447/6837 [35:23<05:29,  4.22it/s]
 80%|███████▉  | 5448/6837 [35:23<05:29,  4.22it/s]
 80%|███████▉  | 5449/6837 [35:23<05:28,  4.23it/s]
 80%|███████▉  | 5450/6837 [35:23<05:27,  4.23it/s]
                                                   
{'loss': 4.3252, 'grad_norm': 0.14403195679187775, 'learning_rate': 0.0007213456418918534, 'epoch': 0.16}
+
 80%|███████▉  | 5450/6837 [35:23<05:27,  4.23it/s]
 80%|███████▉  | 5451/6837 [35:24<05:30,  4.19it/s]
 80%|███████▉  | 5452/6837 [35:24<05:29,  4.20it/s]
 80%|███████▉  | 5453/6837 [35:24<05:28,  4.21it/s]
 80%|███████▉  | 5454/6837 [35:24<05:40,  4.06it/s]
 80%|███████▉  | 5455/6837 [35:25<05:42,  4.04it/s]
 80%|███████▉  | 5456/6837 [35:25<05:36,  4.10it/s]
 80%|███████▉  | 5457/6837 [35:25<05:33,  4.14it/s]
 80%|███████▉  | 5458/6837 [35:25<05:30,  4.17it/s]
 80%|███████▉  | 5459/6837 [35:26<05:28,  4.20it/s]
 80%|███████▉  | 5460/6837 [35:26<05:27,  4.21it/s]
 80%|███████▉  | 5461/6837 [35:26<05:26,  4.22it/s]
 80%|███████▉  | 5462/6837 [35:26<05:25,  4.23it/s]
 80%|███████▉  | 5463/6837 [35:27<05:25,  4.22it/s]
 80%|███████▉  | 5464/6837 [35:27<05:39,  4.05it/s]
 80%|███████▉  | 5465/6837 [35:27<05:44,  3.98it/s]
 80%|███████▉  | 5466/6837 [35:27<05:51,  3.90it/s]
 80%|███████▉  | 5467/6837 [35:28<05:55,  3.86it/s]
 80%|███████▉  | 5468/6837 [35:28<05:59,  3.81it/s]
 80%|███████▉  | 5469/6837 [35:28<06:01,  3.78it/s]
 80%|████████  | 5470/6837 [35:28<05:51,  3.89it/s]
 80%|████████  | 5471/6837 [35:29<05:42,  3.99it/s]
 80%|████████  | 5472/6837 [35:29<05:35,  4.07it/s]
 80%|████████  | 5473/6837 [35:29<05:31,  4.12it/s]
 80%|████████  | 5474/6837 [35:29<05:27,  4.16it/s]
 80%|████████  | 5475/6837 [35:30<05:25,  4.18it/s]{'loss': 4.3246, 'grad_norm': 0.15620769560337067, 'learning_rate': 0.0006966240959056853, 'epoch': 0.16}
+
                                                   

 80%|████████  | 5475/6837 [35:30<05:25,  4.18it/s]
 80%|████████  | 5476/6837 [35:30<05:24,  4.19it/s]
 80%|████████  | 5477/6837 [35:30<05:23,  4.20it/s]
 80%|████████  | 5478/6837 [35:30<05:22,  4.21it/s]
 80%|████████  | 5479/6837 [35:31<05:21,  4.22it/s]
 80%|████████  | 5480/6837 [35:31<05:20,  4.23it/s]
 80%|████████  | 5481/6837 [35:31<05:20,  4.24it/s]
 80%|████████  | 5482/6837 [35:31<05:20,  4.23it/s]
 80%|████████  | 5483/6837 [35:31<05:19,  4.23it/s]
 80%|████████  | 5484/6837 [35:32<05:19,  4.24it/s]
 80%|████████  | 5485/6837 [35:32<05:18,  4.24it/s]
 80%|████████  | 5486/6837 [35:32<05:18,  4.24it/s]
 80%|████████  | 5487/6837 [35:32<05:18,  4.24it/s]
 80%|████████  | 5488/6837 [35:33<05:17,  4.24it/s]
 80%|████████  | 5489/6837 [35:33<05:16,  4.25it/s]
 80%|████████  | 5490/6837 [35:33<05:17,  4.24it/s]
 80%|████████  | 5491/6837 [35:33<05:17,  4.24it/s]
 80%|████████  | 5492/6837 [35:34<05:16,  4.25it/s]
 80%|████████  | 5493/6837 [35:34<05:16,  4.25it/s]
 80%|████████  | 5494/6837 [35:34<05:16,  4.25it/s]
 80%|████████  | 5495/6837 [35:34<05:16,  4.25it/s]
 80%|████████  | 5496/6837 [35:35<05:16,  4.24it/s]
 80%|████████  | 5497/6837 [35:35<05:27,  4.09it/s]
 80%|████████  | 5498/6837 [35:35<05:35,  3.99it/s]
 80%|████████  | 5499/6837 [35:35<05:39,  3.94it/s]
 80%|████████  | 5500/6837 [35:36<05:44,  3.88it/s]
                                                   
{'loss': 4.3129, 'grad_norm': 0.14926698803901672, 'learning_rate': 0.0006722778380727169, 'epoch': 0.16}
+
 80%|████████  | 5500/6837 [35:36<05:44,  3.88it/s]
 80%|████████  | 5501/6837 [35:36<05:48,  3.83it/s]
 80%|████████  | 5502/6837 [35:36<05:49,  3.82it/s]
 80%|████████  | 5503/6837 [35:36<05:41,  3.91it/s]
 81%|████████  | 5504/6837 [35:37<05:33,  4.00it/s]
 81%|████████  | 5505/6837 [35:37<05:27,  4.07it/s]
 81%|████████  | 5506/6837 [35:37<05:22,  4.12it/s]
 81%|████████  | 5507/6837 [35:37<05:20,  4.15it/s]
 81%|████████  | 5508/6837 [35:38<05:17,  4.18it/s]
 81%|████████  | 5509/6837 [35:38<05:15,  4.20it/s]
 81%|████████  | 5510/6837 [35:38<05:14,  4.22it/s]
 81%|████████  | 5511/6837 [35:38<05:13,  4.22it/s]
 81%|████████  | 5512/6837 [35:38<05:13,  4.23it/s]
 81%|████████  | 5513/6837 [35:39<05:12,  4.23it/s]
 81%|████████  | 5514/6837 [35:39<05:12,  4.23it/s]
 81%|████████  | 5515/6837 [35:39<05:12,  4.23it/s]
 81%|████████  | 5516/6837 [35:39<05:13,  4.22it/s]
 81%|████████  | 5517/6837 [35:40<05:14,  4.20it/s]
 81%|████████  | 5518/6837 [35:40<05:13,  4.21it/s]
 81%|████████  | 5519/6837 [35:40<05:12,  4.22it/s]
 81%|████████  | 5520/6837 [35:40<05:11,  4.23it/s]
 81%|████████  | 5521/6837 [35:41<05:10,  4.23it/s]
 81%|████████  | 5522/6837 [35:41<05:10,  4.23it/s]
 81%|████████  | 5523/6837 [35:41<05:09,  4.24it/s]
 81%|████████  | 5524/6837 [35:41<05:09,  4.24it/s]
 81%|████████  | 5525/6837 [35:42<05:09,  4.24it/s]
                                                   
{'loss': 4.3128, 'grad_norm': 0.15168799459934235, 'learning_rate': 0.000648310835119406, 'epoch': 0.16}
+
 81%|████████  | 5525/6837 [35:42<05:09,  4.24it/s]
 81%|████████  | 5526/6837 [35:42<05:10,  4.22it/s]
 81%|████████  | 5527/6837 [35:42<05:09,  4.23it/s]
 81%|████████  | 5528/6837 [35:42<05:08,  4.24it/s]
 81%|████████  | 5529/6837 [35:42<05:08,  4.24it/s]
 81%|████████  | 5530/6837 [35:43<05:19,  4.09it/s]
 81%|████████  | 5531/6837 [35:43<05:24,  4.02it/s]
 81%|████████  | 5532/6837 [35:43<05:25,  4.00it/s]
 81%|████████  | 5533/6837 [35:44<05:19,  4.08it/s]
 81%|████████  | 5534/6837 [35:44<05:15,  4.13it/s]
 81%|████████  | 5535/6837 [35:44<05:12,  4.17it/s]
 81%|████████  | 5536/6837 [35:44<05:10,  4.20it/s]
 81%|████████  | 5537/6837 [35:44<05:08,  4.21it/s]
 81%|████████  | 5538/6837 [35:45<05:07,  4.23it/s]
 81%|████████  | 5539/6837 [35:45<05:06,  4.23it/s]
 81%|████████  | 5540/6837 [35:45<05:06,  4.23it/s]
 81%|████████  | 5541/6837 [35:45<05:05,  4.24it/s]
 81%|████████  | 5542/6837 [35:46<05:05,  4.24it/s]
 81%|████████  | 5543/6837 [35:46<05:05,  4.23it/s]
 81%|████████  | 5544/6837 [35:46<05:05,  4.24it/s]
 81%|████████  | 5545/6837 [35:46<05:04,  4.24it/s]
 81%|████████  | 5546/6837 [35:47<05:04,  4.24it/s]
 81%|████████  | 5547/6837 [35:47<05:04,  4.23it/s]
 81%|████████  | 5548/6837 [35:47<05:04,  4.24it/s]
 81%|████████  | 5549/6837 [35:47<05:04,  4.24it/s]
 81%|████████  | 5550/6837 [35:48<05:04,  4.23it/s]
                                                   
{'loss': 4.3123, 'grad_norm': 0.14819887280464172, 'learning_rate': 0.0006247269919803577, 'epoch': 0.16}
+
 81%|████████  | 5550/6837 [35:48<05:04,  4.23it/s]
 81%|████████  | 5551/6837 [35:48<05:04,  4.22it/s]
 81%|████████  | 5552/6837 [35:48<05:03,  4.23it/s]
 81%|████████  | 5553/6837 [35:48<05:03,  4.23it/s]
 81%|████████  | 5554/6837 [35:48<05:03,  4.23it/s]
 81%|████████  | 5555/6837 [35:49<05:02,  4.23it/s]
 81%|████████▏ | 5556/6837 [35:49<05:02,  4.24it/s]
 81%|████████▏ | 5557/6837 [35:49<05:02,  4.24it/s]
 81%|████████▏ | 5558/6837 [35:49<05:02,  4.23it/s]
 81%|████████▏ | 5559/6837 [35:50<05:01,  4.24it/s]
 81%|████████▏ | 5560/6837 [35:50<05:01,  4.24it/s]
 81%|████████▏ | 5561/6837 [35:50<05:01,  4.23it/s]
 81%|████████▏ | 5562/6837 [35:50<05:01,  4.23it/s]
 81%|████████▏ | 5563/6837 [35:51<05:01,  4.23it/s]
 81%|████████▏ | 5564/6837 [35:51<05:00,  4.23it/s]
 81%|████████▏ | 5565/6837 [35:51<05:00,  4.24it/s]
 81%|████████▏ | 5566/6837 [35:51<04:59,  4.24it/s]
 81%|████████▏ | 5567/6837 [35:52<04:59,  4.24it/s]
 81%|████████▏ | 5568/6837 [35:52<04:59,  4.24it/s]
 81%|████████▏ | 5569/6837 [35:52<04:59,  4.24it/s]
 81%|████████▏ | 5570/6837 [35:52<04:58,  4.24it/s]
 81%|████████▏ | 5571/6837 [35:52<04:59,  4.23it/s]
 81%|████████▏ | 5572/6837 [35:53<04:58,  4.23it/s]
 82%|████████▏ | 5573/6837 [35:53<04:58,  4.24it/s]
 82%|████████▏ | 5574/6837 [35:53<04:57,  4.24it/s]
 82%|████████▏ | 5575/6837 [35:53<04:58,  4.23it/s]
                                                   
{'loss': 4.3167, 'grad_norm': 0.14942596852779388, 'learning_rate': 0.0006015301511620938, 'epoch': 0.16}
+
 82%|████████▏ | 5575/6837 [35:53<04:58,  4.23it/s]
 82%|████████▏ | 5576/6837 [35:54<04:58,  4.23it/s]
 82%|████████▏ | 5577/6837 [35:54<04:58,  4.22it/s]
 82%|████████▏ | 5578/6837 [35:54<04:58,  4.22it/s]
 82%|████████▏ | 5579/6837 [35:54<04:57,  4.23it/s]
 82%|████████▏ | 5580/6837 [35:55<04:56,  4.23it/s]
 82%|████████▏ | 5581/6837 [35:55<04:56,  4.23it/s]
 82%|████████▏ | 5582/6837 [35:55<04:56,  4.23it/s]
 82%|████████▏ | 5583/6837 [35:55<04:56,  4.24it/s]
 82%|████████▏ | 5584/6837 [35:56<04:56,  4.23it/s]
 82%|████████▏ | 5585/6837 [35:56<04:55,  4.24it/s]
 82%|████████▏ | 5586/6837 [35:56<04:55,  4.24it/s]
 82%|████████▏ | 5587/6837 [35:56<04:54,  4.24it/s]
 82%|████████▏ | 5588/6837 [35:56<04:54,  4.24it/s]
 82%|████████▏ | 5589/6837 [35:57<04:54,  4.24it/s]
 82%|████████▏ | 5590/6837 [35:57<04:54,  4.24it/s]
 82%|████████▏ | 5591/6837 [35:57<04:53,  4.24it/s]
 82%|████████▏ | 5592/6837 [35:57<04:54,  4.23it/s]
 82%|████████▏ | 5593/6837 [35:58<04:54,  4.23it/s]
 82%|████████▏ | 5594/6837 [35:58<04:53,  4.23it/s]
 82%|████████▏ | 5595/6837 [35:58<04:53,  4.23it/s]
 82%|████████▏ | 5596/6837 [35:58<04:53,  4.22it/s]
 82%|████████▏ | 5597/6837 [35:59<04:53,  4.23it/s]
 82%|████████▏ | 5598/6837 [35:59<04:52,  4.23it/s]
 82%|████████▏ | 5599/6837 [35:59<04:52,  4.24it/s]
 82%|████████▏ | 5600/6837 [35:59<04:52,  4.24it/s]
                                                   
{'loss': 4.2944, 'grad_norm': 0.14451776444911957, 'learning_rate': 0.0005787240921169986, 'epoch': 0.16}
+
 82%|████████▏ | 5600/6837 [35:59<04:52,  4.24it/s]
 82%|████████▏ | 5601/6837 [36:00<05:03,  4.07it/s]
 82%|████████▏ | 5602/6837 [36:00<05:09,  3.99it/s]
 82%|████████▏ | 5603/6837 [36:00<05:14,  3.93it/s]
 82%|████████▏ | 5604/6837 [36:00<05:17,  3.89it/s]
 82%|████████▏ | 5605/6837 [36:01<05:21,  3.83it/s]
 82%|████████▏ | 5606/6837 [36:01<05:18,  3.87it/s]
 82%|████████▏ | 5607/6837 [36:01<05:10,  3.97it/s]
 82%|████████▏ | 5608/6837 [36:01<05:04,  4.04it/s]
 82%|████████▏ | 5609/6837 [36:02<05:00,  4.09it/s]
 82%|████████▏ | 5610/6837 [36:02<04:56,  4.13it/s]
 82%|████████▏ | 5611/6837 [36:02<04:54,  4.16it/s]
 82%|████████▏ | 5612/6837 [36:02<04:52,  4.19it/s]
 82%|████████▏ | 5613/6837 [36:03<04:52,  4.19it/s]
 82%|████████▏ | 5614/6837 [36:03<04:51,  4.20it/s]
 82%|████████▏ | 5615/6837 [36:03<04:49,  4.21it/s]
 82%|████████▏ | 5616/6837 [36:03<04:49,  4.22it/s]
 82%|████████▏ | 5617/6837 [36:04<04:49,  4.21it/s]
 82%|████████▏ | 5618/6837 [36:04<04:58,  4.08it/s]
 82%|████████▏ | 5619/6837 [36:04<04:55,  4.13it/s]
 82%|████████▏ | 5620/6837 [36:04<04:52,  4.16it/s]
 82%|████████▏ | 5621/6837 [36:04<04:51,  4.18it/s]
 82%|████████▏ | 5622/6837 [36:05<04:49,  4.20it/s]
 82%|████████▏ | 5623/6837 [36:05<04:48,  4.21it/s]
 82%|████████▏ | 5624/6837 [36:05<04:48,  4.21it/s]
 82%|████████▏ | 5625/6837 [36:05<04:47,  4.21it/s]
                                                   
{'loss': 4.3041, 'grad_norm': 0.14448697865009308, 'learning_rate': 0.0005563125306275306, 'epoch': 0.16}
+
 82%|████████▏ | 5625/6837 [36:05<04:47,  4.21it/s]
 82%|████████▏ | 5626/6837 [36:06<04:47,  4.21it/s]
 82%|████████▏ | 5627/6837 [36:06<04:47,  4.22it/s]
 82%|████████▏ | 5628/6837 [36:06<04:46,  4.21it/s]
 82%|████████▏ | 5629/6837 [36:06<04:46,  4.22it/s]
 82%|████████▏ | 5630/6837 [36:07<04:46,  4.22it/s]
 82%|████████▏ | 5631/6837 [36:07<04:46,  4.22it/s]
 82%|████████▏ | 5632/6837 [36:07<04:45,  4.22it/s]
 82%|████████▏ | 5633/6837 [36:07<04:45,  4.22it/s]
 82%|████████▏ | 5634/6837 [36:08<04:44,  4.22it/s]
 82%|████████▏ | 5635/6837 [36:08<04:44,  4.22it/s]
 82%|████████▏ | 5636/6837 [36:08<04:44,  4.22it/s]
 82%|████████▏ | 5637/6837 [36:08<04:43,  4.23it/s]
 82%|████████▏ | 5638/6837 [36:09<04:44,  4.21it/s]
 82%|████████▏ | 5639/6837 [36:09<04:44,  4.21it/s]
 82%|████████▏ | 5640/6837 [36:09<04:43,  4.22it/s]
 83%|████████▎ | 5641/6837 [36:09<04:43,  4.22it/s]
 83%|████████▎ | 5642/6837 [36:09<04:44,  4.20it/s]
 83%|████████▎ | 5643/6837 [36:10<04:42,  4.22it/s]
 83%|████████▎ | 5644/6837 [36:10<04:42,  4.23it/s]
 83%|████████▎ | 5645/6837 [36:10<04:41,  4.23it/s]
 83%|████████▎ | 5646/6837 [36:10<04:40,  4.24it/s]
 83%|████████▎ | 5647/6837 [36:11<04:40,  4.24it/s]
 83%|████████▎ | 5648/6837 [36:11<04:39,  4.25it/s]
 83%|████████▎ | 5649/6837 [36:11<04:39,  4.25it/s]
 83%|████████▎ | 5650/6837 [36:11<04:39,  4.25it/s]
                                                   
{'loss': 4.3027, 'grad_norm': 0.15426599979400635, 'learning_rate': 0.0005342991182008145, 'epoch': 0.17}
+
 83%|████████▎ | 5650/6837 [36:11<04:39,  4.25it/s]
 83%|████████▎ | 5651/6837 [36:12<04:39,  4.24it/s]
 83%|████████▎ | 5652/6837 [36:12<04:38,  4.25it/s]
 83%|████████▎ | 5653/6837 [36:12<04:38,  4.25it/s]
 83%|████████▎ | 5654/6837 [36:12<04:38,  4.25it/s]
 83%|████████▎ | 5655/6837 [36:13<04:37,  4.25it/s]
 83%|████████▎ | 5656/6837 [36:13<04:37,  4.25it/s]
 83%|████████▎ | 5657/6837 [36:13<04:47,  4.11it/s]
 83%|████████▎ | 5658/6837 [36:13<04:43,  4.15it/s]
 83%|████████▎ | 5659/6837 [36:13<04:41,  4.19it/s]
 83%|████████▎ | 5660/6837 [36:14<04:39,  4.20it/s]
 83%|████████▎ | 5661/6837 [36:14<04:38,  4.22it/s]
 83%|████████▎ | 5662/6837 [36:14<04:38,  4.22it/s]
 83%|████████▎ | 5663/6837 [36:14<04:37,  4.23it/s]
 83%|████████▎ | 5664/6837 [36:15<04:37,  4.23it/s]
 83%|████████▎ | 5665/6837 [36:15<04:36,  4.24it/s]
 83%|████████▎ | 5666/6837 [36:15<04:36,  4.24it/s]
 83%|████████▎ | 5667/6837 [36:15<04:35,  4.24it/s]
 83%|████████▎ | 5668/6837 [36:16<04:35,  4.25it/s]
 83%|████████▎ | 5669/6837 [36:16<04:35,  4.24it/s]
 83%|████████▎ | 5670/6837 [36:16<04:34,  4.24it/s]
 83%|████████▎ | 5671/6837 [36:16<04:34,  4.25it/s]
 83%|████████▎ | 5672/6837 [36:17<04:34,  4.24it/s]
 83%|████████▎ | 5673/6837 [36:17<04:34,  4.25it/s]
 83%|████████▎ | 5674/6837 [36:17<04:33,  4.25it/s]
 83%|████████▎ | 5675/6837 [36:17<04:33,  4.25it/s]
                                                   
{'loss': 4.295, 'grad_norm': 0.15341651439666748, 'learning_rate': 0.0005126874414736986, 'epoch': 0.17}
+
 83%|████████▎ | 5675/6837 [36:17<04:33,  4.25it/s]
 83%|████████▎ | 5676/6837 [36:17<04:34,  4.24it/s]
 83%|████████▎ | 5677/6837 [36:18<04:34,  4.23it/s]
 83%|████████▎ | 5678/6837 [36:18<04:33,  4.24it/s]
 83%|████████▎ | 5679/6837 [36:18<04:33,  4.23it/s]
 83%|████████▎ | 5680/6837 [36:18<04:33,  4.23it/s]
 83%|████████▎ | 5681/6837 [36:19<04:33,  4.23it/s]
 83%|████████▎ | 5682/6837 [36:19<04:32,  4.23it/s]
 83%|████████▎ | 5683/6837 [36:19<04:32,  4.23it/s]
 83%|████████▎ | 5684/6837 [36:19<04:32,  4.23it/s]
 83%|████████▎ | 5685/6837 [36:20<04:32,  4.23it/s]
 83%|████████▎ | 5686/6837 [36:20<04:32,  4.23it/s]
 83%|████████▎ | 5687/6837 [36:20<04:31,  4.23it/s]
 83%|████████▎ | 5688/6837 [36:20<04:31,  4.23it/s]
 83%|████████▎ | 5689/6837 [36:21<04:31,  4.23it/s]
 83%|████████▎ | 5690/6837 [36:21<04:30,  4.23it/s]
 83%|████████▎ | 5691/6837 [36:21<04:30,  4.24it/s]
 83%|████████▎ | 5692/6837 [36:21<04:29,  4.24it/s]
 83%|████████▎ | 5693/6837 [36:22<04:29,  4.24it/s]
 83%|████████▎ | 5694/6837 [36:22<04:29,  4.24it/s]
 83%|████████▎ | 5695/6837 [36:22<04:29,  4.23it/s]
 83%|████████▎ | 5696/6837 [36:22<04:29,  4.23it/s]
 83%|████████▎ | 5697/6837 [36:22<04:29,  4.23it/s]
 83%|████████▎ | 5698/6837 [36:23<04:29,  4.23it/s]
 83%|████████▎ | 5699/6837 [36:23<04:28,  4.23it/s]
 83%|████████▎ | 5700/6837 [36:23<04:28,  4.24it/s]
                                                   
{'loss': 4.2785, 'grad_norm': 0.15265800058841705, 'learning_rate': 0.0004914810216283896, 'epoch': 0.17}
+
 83%|████████▎ | 5700/6837 [36:23<04:28,  4.24it/s]
 83%|████████▎ | 5701/6837 [36:23<04:28,  4.23it/s]
 83%|████████▎ | 5702/6837 [36:24<04:28,  4.22it/s]
 83%|████████▎ | 5703/6837 [36:24<04:28,  4.23it/s]
 83%|████████▎ | 5704/6837 [36:24<04:27,  4.23it/s]
 83%|████████▎ | 5705/6837 [36:24<04:27,  4.23it/s]
 83%|████████▎ | 5706/6837 [36:25<04:27,  4.23it/s]
 83%|████████▎ | 5707/6837 [36:25<04:26,  4.23it/s]
 83%|████████▎ | 5708/6837 [36:25<04:26,  4.23it/s]
 84%|████████▎ | 5709/6837 [36:25<04:26,  4.23it/s]
 84%|████████▎ | 5710/6837 [36:26<04:25,  4.24it/s]
 84%|████████▎ | 5711/6837 [36:26<04:25,  4.24it/s]
 84%|████████▎ | 5712/6837 [36:26<04:25,  4.24it/s]
 84%|████████▎ | 5713/6837 [36:26<04:24,  4.24it/s]
 84%|████████▎ | 5714/6837 [36:26<04:24,  4.24it/s]
 84%|████████▎ | 5715/6837 [36:27<04:24,  4.25it/s]
 84%|████████▎ | 5716/6837 [36:27<04:24,  4.24it/s]
 84%|████████▎ | 5717/6837 [36:27<04:24,  4.23it/s]
 84%|████████▎ | 5718/6837 [36:27<04:24,  4.23it/s]
 84%|████████▎ | 5719/6837 [36:28<04:24,  4.23it/s]
 84%|████████▎ | 5720/6837 [36:28<04:24,  4.22it/s]
 84%|████████▎ | 5721/6837 [36:28<04:24,  4.23it/s]
 84%|████████▎ | 5722/6837 [36:28<04:23,  4.23it/s]
 84%|████████▎ | 5723/6837 [36:29<04:23,  4.23it/s]
 84%|████████▎ | 5724/6837 [36:29<04:23,  4.23it/s]
 84%|████████▎ | 5725/6837 [36:29<04:23,  4.22it/s]
                                                   {'loss': 4.306, 'grad_norm': 0.1513630896806717, 'learning_rate': 0.0004706833138187431, 'epoch': 0.17}
+
 84%|████████▎ | 5725/6837 [36:29<04:23,  4.22it/s]
 84%|████████▍ | 5726/6837 [36:29<04:23,  4.21it/s]
 84%|████████▍ | 5727/6837 [36:30<04:23,  4.22it/s]
 84%|████████▍ | 5728/6837 [36:30<04:22,  4.23it/s]
 84%|████████▍ | 5729/6837 [36:30<04:21,  4.23it/s]
 84%|████████▍ | 5730/6837 [36:30<04:21,  4.24it/s]
 84%|████████▍ | 5731/6837 [36:30<04:20,  4.24it/s]
 84%|████████▍ | 5732/6837 [36:31<04:20,  4.24it/s]
 84%|████████▍ | 5733/6837 [36:31<04:21,  4.22it/s]
 84%|████████▍ | 5734/6837 [36:31<04:22,  4.20it/s]
 84%|████████▍ | 5735/6837 [36:31<04:21,  4.22it/s]
 84%|████████▍ | 5736/6837 [36:32<04:20,  4.23it/s]
 84%|████████▍ | 5737/6837 [36:32<04:19,  4.24it/s]
 84%|████████▍ | 5738/6837 [36:32<04:19,  4.24it/s]
 84%|████████▍ | 5739/6837 [36:32<04:18,  4.24it/s]
 84%|████████▍ | 5740/6837 [36:33<04:18,  4.24it/s]
 84%|████████▍ | 5741/6837 [36:33<04:18,  4.23it/s]
 84%|████████▍ | 5742/6837 [36:33<04:18,  4.23it/s]
 84%|████████▍ | 5743/6837 [36:33<04:18,  4.24it/s]
 84%|████████▍ | 5744/6837 [36:34<04:17,  4.24it/s]
 84%|████████▍ | 5745/6837 [36:34<04:17,  4.24it/s]
 84%|████████▍ | 5746/6837 [36:34<04:17,  4.24it/s]
 84%|████████▍ | 5747/6837 [36:34<04:17,  4.24it/s]
 84%|████████▍ | 5748/6837 [36:34<04:16,  4.24it/s]
 84%|████████▍ | 5749/6837 [36:35<04:16,  4.24it/s]
 84%|████████▍ | 5750/6837 [36:35<04:16,  4.24it/s]
                                                   
{'loss': 4.3013, 'grad_norm': 0.16069135069847107, 'learning_rate': 0.0004502977066073216, 'epoch': 0.17}
+
 84%|████████▍ | 5750/6837 [36:35<04:16,  4.24it/s]
 84%|████████▍ | 5751/6837 [36:35<04:16,  4.23it/s]
 84%|████████▍ | 5752/6837 [36:35<04:16,  4.23it/s]
 84%|████████▍ | 5753/6837 [36:36<04:15,  4.24it/s]
 84%|████████▍ | 5754/6837 [36:36<04:15,  4.24it/s]
 84%|████████▍ | 5755/6837 [36:36<04:15,  4.24it/s]
 84%|████████▍ | 5756/6837 [36:36<04:15,  4.24it/s]
 84%|████████▍ | 5757/6837 [36:37<04:14,  4.24it/s]
 84%|████████▍ | 5758/6837 [36:37<04:14,  4.24it/s]
 84%|████████▍ | 5759/6837 [36:37<04:14,  4.24it/s]
 84%|████████▍ | 5760/6837 [36:37<04:13,  4.24it/s]
 84%|████████▍ | 5761/6837 [36:38<04:13,  4.24it/s]
 84%|████████▍ | 5762/6837 [36:38<04:13,  4.24it/s]
 84%|████████▍ | 5763/6837 [36:38<04:13,  4.24it/s]
 84%|████████▍ | 5764/6837 [36:38<04:12,  4.24it/s]
 84%|████████▍ | 5765/6837 [36:39<04:12,  4.24it/s]
 84%|████████▍ | 5766/6837 [36:39<04:12,  4.23it/s]
 84%|████████▍ | 5767/6837 [36:39<04:12,  4.24it/s]
 84%|████████▍ | 5768/6837 [36:39<04:12,  4.23it/s]
 84%|████████▍ | 5769/6837 [36:39<04:13,  4.22it/s]
 84%|████████▍ | 5770/6837 [36:40<04:13,  4.21it/s]
 84%|████████▍ | 5771/6837 [36:40<04:13,  4.21it/s]
 84%|████████▍ | 5772/6837 [36:40<04:12,  4.22it/s]
 84%|████████▍ | 5773/6837 [36:40<04:12,  4.22it/s]
 84%|████████▍ | 5774/6837 [36:41<04:11,  4.22it/s]
 84%|████████▍ | 5775/6837 [36:41<04:11,  4.23it/s]{'loss': 4.3016, 'grad_norm': 0.1558927446603775, 'learning_rate': 0.00043032752141329426, 'epoch': 0.17}
                                                   
+
 84%|████████▍ | 5775/6837 [36:41<04:11,  4.23it/s]
 84%|████████▍ | 5776/6837 [36:41<04:11,  4.21it/s]
 84%|████████▍ | 5777/6837 [36:41<04:11,  4.21it/s]
 85%|████████▍ | 5778/6837 [36:42<04:10,  4.22it/s]
 85%|████████▍ | 5779/6837 [36:42<04:10,  4.23it/s]
 85%|████████▍ | 5780/6837 [36:42<04:10,  4.22it/s]
 85%|████████▍ | 5781/6837 [36:42<04:10,  4.22it/s]
 85%|████████▍ | 5782/6837 [36:43<04:09,  4.22it/s]
 85%|████████▍ | 5783/6837 [36:43<04:16,  4.11it/s]
 85%|████████▍ | 5784/6837 [36:43<04:22,  4.01it/s]
 85%|████████▍ | 5785/6837 [36:43<04:25,  3.96it/s]
 85%|████████▍ | 5786/6837 [36:44<04:29,  3.90it/s]
 85%|████████▍ | 5787/6837 [36:44<04:31,  3.87it/s]
 85%|████████▍ | 5788/6837 [36:44<04:33,  3.84it/s]
 85%|████████▍ | 5789/6837 [36:44<04:34,  3.82it/s]
 85%|████████▍ | 5790/6837 [36:45<04:34,  3.81it/s]
 85%|████████▍ | 5791/6837 [36:45<04:33,  3.83it/s]
 85%|████████▍ | 5792/6837 [36:45<04:34,  3.81it/s]
 85%|████████▍ | 5793/6837 [36:45<04:35,  3.79it/s]
 85%|████████▍ | 5794/6837 [36:46<04:36,  3.78it/s]
 85%|████████▍ | 5795/6837 [36:46<04:34,  3.80it/s]
 85%|████████▍ | 5796/6837 [36:46<04:34,  3.79it/s]
 85%|████████▍ | 5797/6837 [36:46<04:33,  3.80it/s]
 85%|████████▍ | 5798/6837 [36:47<04:32,  3.81it/s]
 85%|████████▍ | 5799/6837 [36:47<04:35,  3.76it/s]
 85%|████████▍ | 5800/6837 [36:47<04:35,  3.76it/s]
                                                   
{'loss': 4.282, 'grad_norm': 0.14679650962352753, 'learning_rate': 0.00041077601197128167, 'epoch': 0.17}
+
 85%|████████▍ | 5800/6837 [36:47<04:35,  3.76it/s]
 85%|████████▍ | 5801/6837 [36:48<04:35,  3.75it/s]
 85%|████████▍ | 5802/6837 [36:48<04:34,  3.76it/s]
 85%|████████▍ | 5803/6837 [36:48<04:33,  3.78it/s]
 85%|████████▍ | 5804/6837 [36:48<04:24,  3.91it/s]
 85%|████████▍ | 5805/6837 [36:49<04:18,  4.00it/s]
 85%|████████▍ | 5806/6837 [36:49<04:13,  4.06it/s]
 85%|████████▍ | 5807/6837 [36:49<04:10,  4.11it/s]
 85%|████████▍ | 5808/6837 [36:49<04:08,  4.14it/s]
 85%|████████▍ | 5809/6837 [36:49<04:06,  4.17it/s]
 85%|████████▍ | 5810/6837 [36:50<04:05,  4.18it/s]
 85%|████████▍ | 5811/6837 [36:50<04:04,  4.20it/s]
 85%|████████▌ | 5812/6837 [36:50<04:03,  4.21it/s]
 85%|████████▌ | 5813/6837 [36:50<04:03,  4.21it/s]
 85%|████████▌ | 5814/6837 [36:51<04:02,  4.21it/s]
 85%|████████▌ | 5815/6837 [36:51<04:02,  4.22it/s]
 85%|████████▌ | 5816/6837 [36:51<04:01,  4.22it/s]
 85%|████████▌ | 5817/6837 [36:51<04:01,  4.22it/s]
 85%|████████▌ | 5818/6837 [36:52<04:01,  4.22it/s]
 85%|████████▌ | 5819/6837 [36:52<04:00,  4.23it/s]
 85%|████████▌ | 5820/6837 [36:52<04:00,  4.23it/s]
 85%|████████▌ | 5821/6837 [36:52<04:00,  4.23it/s]
 85%|████████▌ | 5822/6837 [36:53<04:00,  4.22it/s]
 85%|████████▌ | 5823/6837 [36:53<04:00,  4.22it/s]
 85%|████████▌ | 5824/6837 [36:53<04:00,  4.21it/s]
 85%|████████▌ | 5825/6837 [36:53<04:00,  4.22it/s]
                                                   
{'loss': 4.3036, 'grad_norm': 0.18000821769237518, 'learning_rate': 0.00039164636380122276, 'epoch': 0.17}
+
 85%|████████▌ | 5825/6837 [36:53<04:00,  4.22it/s]
 85%|████████▌ | 5826/6837 [36:54<04:00,  4.21it/s]
 85%|████████▌ | 5827/6837 [36:54<04:00,  4.21it/s]
 85%|████████▌ | 5828/6837 [36:54<03:59,  4.21it/s]
 85%|████████▌ | 5829/6837 [36:54<03:58,  4.22it/s]
 85%|████████▌ | 5830/6837 [36:54<03:58,  4.22it/s]
 85%|████████▌ | 5831/6837 [36:55<03:58,  4.22it/s]
 85%|████████▌ | 5832/6837 [36:55<03:58,  4.22it/s]
 85%|████████▌ | 5833/6837 [36:55<03:57,  4.22it/s]
 85%|████████▌ | 5834/6837 [36:55<03:57,  4.22it/s]
 85%|████████▌ | 5835/6837 [36:56<03:57,  4.22it/s]
 85%|████████▌ | 5836/6837 [36:56<03:57,  4.22it/s]
 85%|████████▌ | 5837/6837 [36:56<03:57,  4.22it/s]
 85%|████████▌ | 5838/6837 [36:56<03:56,  4.22it/s]
 85%|████████▌ | 5839/6837 [36:57<03:56,  4.22it/s]
 85%|████████▌ | 5840/6837 [36:57<03:55,  4.23it/s]
 85%|████████▌ | 5841/6837 [36:57<03:55,  4.23it/s]
 85%|████████▌ | 5842/6837 [36:57<03:55,  4.23it/s]
 85%|████████▌ | 5843/6837 [36:58<03:55,  4.23it/s]
 85%|████████▌ | 5844/6837 [36:58<03:54,  4.23it/s]
 85%|████████▌ | 5845/6837 [36:58<03:54,  4.23it/s]
 86%|████████▌ | 5846/6837 [36:58<03:54,  4.23it/s]
 86%|████████▌ | 5847/6837 [36:58<03:53,  4.23it/s]
 86%|████████▌ | 5848/6837 [36:59<03:54,  4.23it/s]
 86%|████████▌ | 5849/6837 [36:59<03:53,  4.23it/s]
 86%|████████▌ | 5850/6837 [36:59<03:53,  4.23it/s]
                                                   
{'loss': 4.2954, 'grad_norm': 0.14284411072731018, 'learning_rate': 0.0003729416936893619, 'epoch': 0.17}
+
 86%|████████▌ | 5850/6837 [36:59<03:53,  4.23it/s]
 86%|████████▌ | 5851/6837 [36:59<03:53,  4.22it/s]
 86%|████████▌ | 5852/6837 [37:00<03:53,  4.23it/s]
 86%|████████▌ | 5853/6837 [37:00<03:52,  4.23it/s]
 86%|████████▌ | 5854/6837 [37:00<03:52,  4.23it/s]
 86%|████████▌ | 5855/6837 [37:00<03:51,  4.23it/s]
 86%|████████▌ | 5856/6837 [37:01<03:51,  4.23it/s]
 86%|████████▌ | 5857/6837 [37:01<03:51,  4.24it/s]
 86%|████████▌ | 5858/6837 [37:01<03:51,  4.24it/s]
 86%|████████▌ | 5859/6837 [37:01<03:50,  4.23it/s]
 86%|████████▌ | 5860/6837 [37:02<03:50,  4.24it/s]
 86%|████████▌ | 5861/6837 [37:02<03:50,  4.24it/s]
 86%|████████▌ | 5862/6837 [37:02<03:49,  4.24it/s]
 86%|████████▌ | 5863/6837 [37:02<03:49,  4.24it/s]
 86%|████████▌ | 5864/6837 [37:03<03:49,  4.23it/s]
 86%|████████▌ | 5865/6837 [37:03<03:49,  4.24it/s]
 86%|████████▌ | 5866/6837 [37:03<03:49,  4.24it/s]
 86%|████████▌ | 5867/6837 [37:03<03:48,  4.24it/s]
 86%|████████▌ | 5868/6837 [37:03<03:48,  4.24it/s]
 86%|████████▌ | 5869/6837 [37:04<03:48,  4.24it/s]
 86%|████████▌ | 5870/6837 [37:04<03:48,  4.24it/s]
 86%|████████▌ | 5871/6837 [37:04<03:48,  4.23it/s]
 86%|████████▌ | 5872/6837 [37:04<03:48,  4.23it/s]
 86%|████████▌ | 5873/6837 [37:05<03:47,  4.24it/s]
 86%|████████▌ | 5874/6837 [37:05<03:47,  4.24it/s]
 86%|████████▌ | 5875/6837 [37:05<03:46,  4.24it/s]
                                                   
{'loss': 4.2906, 'grad_norm': 0.1507183313369751, 'learning_rate': 0.0003546650491804304, 'epoch': 0.17}
+
 86%|████████▌ | 5875/6837 [37:05<03:46,  4.24it/s]
 86%|████████▌ | 5876/6837 [37:05<03:46,  4.24it/s]
 86%|████████▌ | 5877/6837 [37:06<03:46,  4.24it/s]
 86%|████████▌ | 5878/6837 [37:06<03:46,  4.24it/s]
 86%|████████▌ | 5879/6837 [37:06<03:46,  4.23it/s]
 86%|████████▌ | 5880/6837 [37:06<03:45,  4.23it/s]
 86%|████████▌ | 5881/6837 [37:07<03:45,  4.23it/s]
 86%|████████▌ | 5882/6837 [37:07<03:45,  4.23it/s]
 86%|████████▌ | 5883/6837 [37:07<03:44,  4.24it/s]
 86%|████████▌ | 5884/6837 [37:07<03:44,  4.24it/s]
 86%|████████▌ | 5885/6837 [37:07<03:44,  4.24it/s]
 86%|████████▌ | 5886/6837 [37:08<03:44,  4.24it/s]
 86%|████████▌ | 5887/6837 [37:08<03:44,  4.24it/s]
 86%|████████▌ | 5888/6837 [37:08<03:44,  4.22it/s]
 86%|████████▌ | 5889/6837 [37:08<03:44,  4.23it/s]
 86%|████████▌ | 5890/6837 [37:09<03:44,  4.23it/s]
 86%|████████▌ | 5891/6837 [37:09<03:43,  4.23it/s]
 86%|████████▌ | 5892/6837 [37:09<03:43,  4.23it/s]
 86%|████████▌ | 5893/6837 [37:09<03:43,  4.23it/s]
 86%|████████▌ | 5894/6837 [37:10<03:43,  4.22it/s]
 86%|████████▌ | 5895/6837 [37:10<03:44,  4.19it/s]
 86%|████████▌ | 5896/6837 [37:10<03:45,  4.16it/s]
 86%|████████▋ | 5897/6837 [37:10<03:44,  4.18it/s]
 86%|████████▋ | 5898/6837 [37:11<03:43,  4.20it/s]
 86%|████████▋ | 5899/6837 [37:11<03:42,  4.21it/s]
 86%|████████▋ | 5900/6837 [37:11<03:42,  4.21it/s]
                                                   
{'loss': 4.2864, 'grad_norm': 0.1474820375442505, 'learning_rate': 0.0003368194080811134, 'epoch': 0.17}
+
 86%|████████▋ | 5900/6837 [37:11<03:42,  4.21it/s]
 86%|████████▋ | 5901/6837 [37:11<03:42,  4.21it/s]
 86%|████████▋ | 5902/6837 [37:12<03:41,  4.21it/s]
 86%|████████▋ | 5903/6837 [37:12<03:41,  4.22it/s]
 86%|████████▋ | 5904/6837 [37:12<03:41,  4.21it/s]
 86%|████████▋ | 5905/6837 [37:12<03:41,  4.20it/s]
 86%|████████▋ | 5906/6837 [37:12<03:42,  4.19it/s]
 86%|████████▋ | 5907/6837 [37:13<03:50,  4.04it/s]
 86%|████████▋ | 5908/6837 [37:13<03:56,  3.93it/s]
 86%|████████▋ | 5909/6837 [37:13<04:02,  3.82it/s]
 86%|████████▋ | 5910/6837 [37:14<03:56,  3.93it/s]
 86%|████████▋ | 5911/6837 [37:14<03:51,  4.01it/s]
 86%|████████▋ | 5912/6837 [37:14<03:47,  4.06it/s]
 86%|████████▋ | 5913/6837 [37:14<03:45,  4.10it/s]
 86%|████████▋ | 5914/6837 [37:14<03:43,  4.13it/s]
 87%|████████▋ | 5915/6837 [37:15<03:42,  4.15it/s]
 87%|████████▋ | 5916/6837 [37:15<03:41,  4.16it/s]
 87%|████████▋ | 5917/6837 [37:15<03:40,  4.18it/s]
 87%|████████▋ | 5918/6837 [37:15<03:39,  4.19it/s]
 87%|████████▋ | 5919/6837 [37:16<03:39,  4.19it/s]
 87%|████████▋ | 5920/6837 [37:16<03:38,  4.20it/s]
 87%|████████▋ | 5921/6837 [37:16<03:37,  4.20it/s]
 87%|████████▋ | 5922/6837 [37:16<03:37,  4.21it/s]
 87%|████████▋ | 5923/6837 [37:17<03:36,  4.22it/s]
 87%|████████▋ | 5924/6837 [37:17<03:36,  4.22it/s]
 87%|████████▋ | 5925/6837 [37:17<03:35,  4.22it/s]
                                                   
{'loss': 4.2767, 'grad_norm': 0.1477581262588501, 'learning_rate': 0.00031940767797487504, 'epoch': 0.17}
+
 87%|████████▋ | 5925/6837 [37:17<03:35,  4.22it/s]
 87%|████████▋ | 5926/6837 [37:17<03:36,  4.21it/s]
 87%|████████▋ | 5927/6837 [37:18<03:36,  4.21it/s]
 87%|████████▋ | 5928/6837 [37:18<03:35,  4.22it/s]
 87%|████████▋ | 5929/6837 [37:18<03:35,  4.22it/s]
 87%|████████▋ | 5930/6837 [37:18<03:34,  4.22it/s]
 87%|████████▋ | 5931/6837 [37:19<03:34,  4.22it/s]
 87%|████████▋ | 5932/6837 [37:19<03:34,  4.22it/s]
 87%|████████▋ | 5933/6837 [37:19<03:34,  4.21it/s]
 87%|████████▋ | 5934/6837 [37:19<03:34,  4.21it/s]
 87%|████████▋ | 5935/6837 [37:19<03:34,  4.21it/s]
 87%|████████▋ | 5936/6837 [37:20<03:33,  4.21it/s]
 87%|████████▋ | 5937/6837 [37:20<03:33,  4.21it/s]
 87%|████████▋ | 5938/6837 [37:20<03:33,  4.21it/s]
 87%|████████▋ | 5939/6837 [37:20<03:33,  4.22it/s]
 87%|████████▋ | 5940/6837 [37:21<03:32,  4.22it/s]
 87%|████████▋ | 5941/6837 [37:21<03:32,  4.22it/s]
 87%|████████▋ | 5942/6837 [37:21<03:32,  4.22it/s]
 87%|████████▋ | 5943/6837 [37:21<03:31,  4.22it/s]
 87%|████████▋ | 5944/6837 [37:22<03:32,  4.21it/s]
 87%|████████▋ | 5945/6837 [37:22<03:32,  4.19it/s]
 87%|████████▋ | 5946/6837 [37:22<03:32,  4.19it/s]
 87%|████████▋ | 5947/6837 [37:22<03:31,  4.20it/s]
 87%|████████▋ | 5948/6837 [37:23<03:31,  4.20it/s]
 87%|████████▋ | 5949/6837 [37:23<03:30,  4.21it/s]
 87%|████████▋ | 5950/6837 [37:23<03:30,  4.21it/s]{'loss': 4.2901, 'grad_norm': 0.15054039657115936, 'learning_rate': 0.0003024326957482272, 'epoch': 0.17}
+
                                                   

 87%|████████▋ | 5950/6837 [37:23<03:30,  4.21it/s]
 87%|████████▋ | 5951/6837 [37:23<03:31,  4.19it/s]
 87%|████████▋ | 5952/6837 [37:23<03:30,  4.20it/s]
 87%|████████▋ | 5953/6837 [37:24<03:30,  4.20it/s]
 87%|████████▋ | 5954/6837 [37:24<03:30,  4.20it/s]
 87%|████████▋ | 5955/6837 [37:24<03:29,  4.20it/s]
 87%|████████▋ | 5956/6837 [37:24<03:29,  4.21it/s]
 87%|████████▋ | 5957/6837 [37:25<03:29,  4.21it/s]
 87%|████████▋ | 5958/6837 [37:25<03:28,  4.21it/s]
 87%|████████▋ | 5959/6837 [37:25<03:28,  4.21it/s]
 87%|████████▋ | 5960/6837 [37:25<03:27,  4.22it/s]
 87%|████████▋ | 5961/6837 [37:26<03:27,  4.22it/s]
 87%|████████▋ | 5962/6837 [37:26<03:27,  4.21it/s]
 87%|████████▋ | 5963/6837 [37:26<03:27,  4.22it/s]
 87%|████████▋ | 5964/6837 [37:26<03:26,  4.22it/s]
 87%|████████▋ | 5965/6837 [37:27<03:27,  4.21it/s]
 87%|████████▋ | 5966/6837 [37:27<03:26,  4.22it/s]
 87%|████████▋ | 5967/6837 [37:27<03:26,  4.22it/s]
 87%|████████▋ | 5968/6837 [37:27<03:26,  4.22it/s]
 87%|████████▋ | 5969/6837 [37:28<03:25,  4.22it/s]
 87%|████████▋ | 5970/6837 [37:28<03:25,  4.22it/s]
 87%|████████▋ | 5971/6837 [37:28<03:25,  4.22it/s]
 87%|████████▋ | 5972/6837 [37:28<03:25,  4.21it/s]
 87%|████████▋ | 5973/6837 [37:28<03:25,  4.21it/s]
 87%|████████▋ | 5974/6837 [37:29<03:24,  4.22it/s]
 87%|████████▋ | 5975/6837 [37:29<03:24,  4.22it/s]
                                                   
{'loss': 4.2801, 'grad_norm': 0.15281051397323608, 'learning_rate': 0.0002858972271285194, 'epoch': 0.17}
+
 87%|████████▋ | 5975/6837 [37:29<03:24,  4.22it/s]
 87%|████████▋ | 5976/6837 [37:29<03:30,  4.09it/s]
 87%|████████▋ | 5977/6837 [37:29<03:35,  3.99it/s]
 87%|████████▋ | 5978/6837 [37:30<03:38,  3.93it/s]
 87%|████████▋ | 5979/6837 [37:30<03:40,  3.89it/s]
 87%|████████▋ | 5980/6837 [37:30<03:41,  3.87it/s]
 87%|████████▋ | 5981/6837 [37:31<03:42,  3.85it/s]
 87%|████████▋ | 5982/6837 [37:31<03:41,  3.86it/s]
 88%|████████▊ | 5983/6837 [37:31<03:42,  3.84it/s]
 88%|████████▊ | 5984/6837 [37:31<03:43,  3.81it/s]
 88%|████████▊ | 5985/6837 [37:32<03:43,  3.80it/s]
 88%|████████▊ | 5986/6837 [37:32<03:43,  3.80it/s]
 88%|████████▊ | 5987/6837 [37:32<03:43,  3.80it/s]
 88%|████████▊ | 5988/6837 [37:32<03:43,  3.79it/s]
 88%|████████▊ | 5989/6837 [37:33<03:44,  3.78it/s]
 88%|████████▊ | 5990/6837 [37:33<03:43,  3.79it/s]
 88%|████████▊ | 5991/6837 [37:33<03:43,  3.79it/s]
 88%|████████▊ | 5992/6837 [37:33<03:42,  3.79it/s]
 88%|████████▊ | 5993/6837 [37:34<03:40,  3.82it/s]
 88%|████████▊ | 5994/6837 [37:34<03:41,  3.80it/s]
 88%|████████▊ | 5995/6837 [37:34<03:42,  3.78it/s]
 88%|████████▊ | 5996/6837 [37:34<03:42,  3.78it/s]
 88%|████████▊ | 5997/6837 [37:35<03:41,  3.79it/s]
 88%|████████▊ | 5998/6837 [37:35<03:37,  3.85it/s]
 88%|████████▊ | 5999/6837 [37:35<03:31,  3.96it/s]
 88%|████████▊ | 6000/6837 [37:35<03:27,  4.04it/s]
                                                   
{'loss': 4.3079, 'grad_norm': 0.14913491904735565, 'learning_rate': 0.00026980396623331503, 'epoch': 0.18}
+
 88%|████████▊ | 6000/6837 [37:35<03:27,  4.04it/s]
 88%|████████▊ | 6001/6837 [37:36<03:24,  4.08it/s]
 88%|████████▊ | 6002/6837 [37:36<03:22,  4.12it/s]
 88%|████████▊ | 6003/6837 [37:36<03:20,  4.16it/s]
 88%|████████▊ | 6004/6837 [37:36<03:18,  4.19it/s]
 88%|████████▊ | 6005/6837 [37:37<03:17,  4.20it/s]
 88%|████████▊ | 6006/6837 [37:37<03:17,  4.21it/s]
 88%|████████▊ | 6007/6837 [37:37<03:16,  4.22it/s]
 88%|████████▊ | 6008/6837 [37:37<03:16,  4.22it/s]
 88%|████████▊ | 6009/6837 [37:38<03:16,  4.22it/s]
 88%|████████▊ | 6010/6837 [37:38<03:15,  4.23it/s]
 88%|████████▊ | 6011/6837 [37:38<03:15,  4.23it/s]
 88%|████████▊ | 6012/6837 [37:38<03:15,  4.23it/s]
 88%|████████▊ | 6013/6837 [37:39<03:14,  4.23it/s]
 88%|████████▊ | 6014/6837 [37:39<03:14,  4.24it/s]
 88%|████████▊ | 6015/6837 [37:39<03:14,  4.23it/s]
 88%|████████▊ | 6016/6837 [37:39<03:14,  4.23it/s]
 88%|████████▊ | 6017/6837 [37:39<03:13,  4.23it/s]
 88%|████████▊ | 6018/6837 [37:40<03:13,  4.23it/s]
 88%|████████▊ | 6019/6837 [37:40<03:14,  4.21it/s]
 88%|████████▊ | 6020/6837 [37:40<03:14,  4.20it/s]
 88%|████████▊ | 6021/6837 [37:40<03:14,  4.20it/s]
 88%|████████▊ | 6022/6837 [37:41<03:13,  4.21it/s]
 88%|████████▊ | 6023/6837 [37:41<03:13,  4.22it/s]
 88%|████████▊ | 6024/6837 [37:41<03:12,  4.22it/s]
 88%|████████▊ | 6025/6837 [37:41<03:11,  4.23it/s]
                                                   
{'loss': 4.2881, 'grad_norm': 0.15348763763904572, 'learning_rate': 0.0002541555351314442, 'epoch': 0.18}
+
 88%|████████▊ | 6025/6837 [37:41<03:11,  4.23it/s]
 88%|████████▊ | 6026/6837 [37:42<03:11,  4.23it/s]
 88%|████████▊ | 6027/6837 [37:42<03:11,  4.23it/s]
 88%|████████▊ | 6028/6837 [37:42<03:11,  4.23it/s]
 88%|████████▊ | 6029/6837 [37:42<03:12,  4.21it/s]
 88%|████████▊ | 6030/6837 [37:43<03:12,  4.20it/s]
 88%|████████▊ | 6031/6837 [37:43<03:19,  4.05it/s]
 88%|████████▊ | 6032/6837 [37:43<03:24,  3.94it/s]
 88%|████████▊ | 6033/6837 [37:43<03:21,  3.99it/s]
 88%|████████▊ | 6034/6837 [37:44<03:18,  4.06it/s]
 88%|████████▊ | 6035/6837 [37:44<03:15,  4.11it/s]
 88%|████████▊ | 6036/6837 [37:44<03:13,  4.14it/s]
 88%|████████▊ | 6037/6837 [37:44<03:11,  4.17it/s]
 88%|████████▊ | 6038/6837 [37:45<03:10,  4.19it/s]
 88%|████████▊ | 6039/6837 [37:45<03:09,  4.20it/s]
 88%|████████▊ | 6040/6837 [37:45<03:09,  4.21it/s]
 88%|████████▊ | 6041/6837 [37:45<03:09,  4.21it/s]
 88%|████████▊ | 6042/6837 [37:45<03:09,  4.20it/s]
 88%|████████▊ | 6043/6837 [37:46<03:09,  4.19it/s]
 88%|████████▊ | 6044/6837 [37:46<03:08,  4.20it/s]
 88%|████████▊ | 6045/6837 [37:46<03:07,  4.21it/s]
 88%|████████▊ | 6046/6837 [37:46<03:07,  4.22it/s]
 88%|████████▊ | 6047/6837 [37:47<03:06,  4.23it/s]
 88%|████████▊ | 6048/6837 [37:47<03:06,  4.23it/s]
 88%|████████▊ | 6049/6837 [37:47<03:06,  4.23it/s]
 88%|████████▊ | 6050/6837 [37:47<03:05,  4.24it/s]
                                                   
{'loss': 4.2975, 'grad_norm': 0.14375270903110504, 'learning_rate': 0.0002389544834157873, 'epoch': 0.18}
+
 88%|████████▊ | 6050/6837 [37:47<03:05,  4.24it/s]
 89%|████████▊ | 6051/6837 [37:48<03:06,  4.22it/s]
 89%|████████▊ | 6052/6837 [37:48<03:05,  4.23it/s]
 89%|████████▊ | 6053/6837 [37:48<03:05,  4.22it/s]
 89%|████████▊ | 6054/6837 [37:48<03:05,  4.23it/s]
 89%|████████▊ | 6055/6837 [37:49<03:11,  4.08it/s]
 89%|████████▊ | 6056/6837 [37:49<03:15,  4.00it/s]
 89%|████████▊ | 6057/6837 [37:49<03:17,  3.94it/s]
 89%|████████▊ | 6058/6837 [37:49<03:18,  3.91it/s]
 89%|████████▊ | 6059/6837 [37:50<03:20,  3.88it/s]
 89%|████████▊ | 6060/6837 [37:50<03:20,  3.87it/s]
 89%|████████▊ | 6061/6837 [37:50<03:22,  3.83it/s]
 89%|████████▊ | 6062/6837 [37:50<03:22,  3.82it/s]
 89%|████████▊ | 6063/6837 [37:51<03:24,  3.78it/s]
 89%|████████▊ | 6064/6837 [37:51<03:24,  3.78it/s]
 89%|████████▊ | 6065/6837 [37:51<03:17,  3.90it/s]
 89%|████████▊ | 6066/6837 [37:51<03:12,  4.00it/s]
 89%|████████▊ | 6067/6837 [37:52<03:09,  4.07it/s]
 89%|████████▉ | 6068/6837 [37:52<03:06,  4.12it/s]
 89%|████████▉ | 6069/6837 [37:52<03:04,  4.15it/s]
 89%|████████▉ | 6070/6837 [37:52<03:03,  4.18it/s]
 89%|████████▉ | 6071/6837 [37:53<03:02,  4.20it/s]
 89%|████████▉ | 6072/6837 [37:53<03:01,  4.22it/s]
 89%|████████▉ | 6073/6837 [37:53<03:01,  4.22it/s]
 89%|████████▉ | 6074/6837 [37:53<03:00,  4.23it/s]
 89%|████████▉ | 6075/6837 [37:54<02:59,  4.24it/s]
                                                   
{'loss': 4.2935, 'grad_norm': 0.14443939924240112, 'learning_rate': 0.00022420328778787426, 'epoch': 0.18}
+
 89%|████████▉ | 6075/6837 [37:54<02:59,  4.24it/s]
 89%|████████▉ | 6076/6837 [37:54<02:59,  4.23it/s]
 89%|████████▉ | 6077/6837 [37:54<02:59,  4.23it/s]
 89%|████████▉ | 6078/6837 [37:54<02:58,  4.24it/s]
 89%|████████▉ | 6079/6837 [37:54<02:58,  4.25it/s]
 89%|████████▉ | 6080/6837 [37:55<02:57,  4.25it/s]
 89%|████████▉ | 6081/6837 [37:55<02:57,  4.25it/s]
 89%|████████▉ | 6082/6837 [37:55<02:57,  4.25it/s]
 89%|████████▉ | 6083/6837 [37:55<02:57,  4.25it/s]
 89%|████████▉ | 6084/6837 [37:56<02:57,  4.25it/s]
 89%|████████▉ | 6085/6837 [37:56<02:56,  4.25it/s]
 89%|████████▉ | 6086/6837 [37:56<02:56,  4.26it/s]
 89%|████████▉ | 6087/6837 [37:56<02:56,  4.25it/s]
 89%|████████▉ | 6088/6837 [37:57<02:56,  4.25it/s]
 89%|████████▉ | 6089/6837 [37:57<02:55,  4.26it/s]
 89%|████████▉ | 6090/6837 [37:57<02:55,  4.26it/s]
 89%|████████▉ | 6091/6837 [37:57<02:55,  4.26it/s]
 89%|████████▉ | 6092/6837 [37:58<02:54,  4.26it/s]
 89%|████████▉ | 6093/6837 [37:58<02:54,  4.26it/s]
 89%|████████▉ | 6094/6837 [37:58<02:54,  4.25it/s]
 89%|████████▉ | 6095/6837 [37:58<02:54,  4.25it/s]
 89%|████████▉ | 6096/6837 [37:58<02:54,  4.25it/s]
 89%|████████▉ | 6097/6837 [37:59<02:54,  4.25it/s]
 89%|████████▉ | 6098/6837 [37:59<02:54,  4.25it/s]
 89%|████████▉ | 6099/6837 [37:59<02:53,  4.25it/s]
 89%|████████▉ | 6100/6837 [37:59<02:53,  4.24it/s]
                                                   
{'loss': 4.2807, 'grad_norm': 0.14695385098457336, 'learning_rate': 0.00020990435165435474, 'epoch': 0.18}
+
 89%|████████▉ | 6100/6837 [37:59<02:53,  4.24it/s]
 89%|████████▉ | 6101/6837 [38:00<02:53,  4.23it/s]
 89%|████████▉ | 6102/6837 [38:00<02:53,  4.23it/s]
 89%|████████▉ | 6103/6837 [38:00<02:53,  4.24it/s]
 89%|████████▉ | 6104/6837 [38:00<02:52,  4.24it/s]
 89%|████████▉ | 6105/6837 [38:01<02:52,  4.24it/s]
 89%|████████▉ | 6106/6837 [38:01<02:52,  4.25it/s]
 89%|████████▉ | 6107/6837 [38:01<02:52,  4.24it/s]
 89%|████████▉ | 6108/6837 [38:01<02:52,  4.24it/s]
 89%|████████▉ | 6109/6837 [38:02<02:52,  4.23it/s]
 89%|████████▉ | 6110/6837 [38:02<02:51,  4.23it/s]
 89%|████████▉ | 6111/6837 [38:02<02:51,  4.23it/s]
 89%|████████▉ | 6112/6837 [38:02<02:51,  4.22it/s]
 89%|████████▉ | 6113/6837 [38:03<02:51,  4.23it/s]
 89%|████████▉ | 6114/6837 [38:03<02:50,  4.23it/s]
 89%|████████▉ | 6115/6837 [38:03<02:50,  4.23it/s]
 89%|████████▉ | 6116/6837 [38:03<02:50,  4.23it/s]
 89%|████████▉ | 6117/6837 [38:03<02:50,  4.23it/s]
 89%|████████▉ | 6118/6837 [38:04<02:49,  4.24it/s]
 89%|████████▉ | 6119/6837 [38:04<02:49,  4.23it/s]
 90%|████████▉ | 6120/6837 [38:04<02:49,  4.23it/s]
 90%|████████▉ | 6121/6837 [38:04<02:48,  4.24it/s]
 90%|████████▉ | 6122/6837 [38:05<02:48,  4.24it/s]
 90%|████████▉ | 6123/6837 [38:05<02:48,  4.24it/s]
 90%|████████▉ | 6124/6837 [38:05<02:48,  4.24it/s]
 90%|████████▉ | 6125/6837 [38:05<02:47,  4.25it/s]
                                                   
{'loss': 4.3013, 'grad_norm': 0.14527039229869843, 'learning_rate': 0.0001960600047354132, 'epoch': 0.18}
+
 90%|████████▉ | 6125/6837 [38:05<02:47,  4.25it/s]
 90%|████████▉ | 6126/6837 [38:06<02:48,  4.23it/s]
 90%|████████▉ | 6127/6837 [38:06<02:47,  4.23it/s]
 90%|████████▉ | 6128/6837 [38:06<02:47,  4.24it/s]
 90%|████████▉ | 6129/6837 [38:06<02:53,  4.09it/s]
 90%|████████▉ | 6130/6837 [38:07<02:57,  3.98it/s]
 90%|████████▉ | 6131/6837 [38:07<03:00,  3.91it/s]
 90%|████████▉ | 6132/6837 [38:07<03:02,  3.87it/s]
 90%|████████▉ | 6133/6837 [38:07<03:04,  3.82it/s]
 90%|████████▉ | 6134/6837 [38:08<03:04,  3.80it/s]
 90%|████████▉ | 6135/6837 [38:08<03:04,  3.80it/s]
 90%|████████▉ | 6136/6837 [38:08<03:06,  3.77it/s]
 90%|████████▉ | 6137/6837 [38:08<03:04,  3.79it/s]
 90%|████████▉ | 6138/6837 [38:09<03:04,  3.79it/s]
 90%|████████▉ | 6139/6837 [38:09<03:04,  3.78it/s]
 90%|████████▉ | 6140/6837 [38:09<03:04,  3.78it/s]
 90%|████████▉ | 6141/6837 [38:09<02:59,  3.87it/s]
 90%|████████▉ | 6142/6837 [38:10<02:55,  3.95it/s]
 90%|████████▉ | 6143/6837 [38:10<02:52,  4.03it/s]
 90%|████████▉ | 6144/6837 [38:10<02:49,  4.09it/s]
 90%|████████▉ | 6145/6837 [38:10<02:47,  4.12it/s]
 90%|████████▉ | 6146/6837 [38:11<02:46,  4.15it/s]
 90%|████████▉ | 6147/6837 [38:11<02:45,  4.18it/s]
 90%|████████▉ | 6148/6837 [38:11<02:44,  4.19it/s]
 90%|████████▉ | 6149/6837 [38:11<02:43,  4.20it/s]
 90%|████████▉ | 6150/6837 [38:12<02:43,  4.21it/s]
                                                   
{'loss': 4.2981, 'grad_norm': 0.14275389909744263, 'learning_rate': 0.00018267250268518643, 'epoch': 0.18}
+
 90%|████████▉ | 6150/6837 [38:12<02:43,  4.21it/s]
 90%|████████▉ | 6151/6837 [38:12<02:43,  4.21it/s]
 90%|████████▉ | 6152/6837 [38:12<02:42,  4.21it/s]
 90%|████████▉ | 6153/6837 [38:12<02:42,  4.22it/s]
 90%|█████████ | 6154/6837 [38:13<02:41,  4.22it/s]
 90%|█████████ | 6155/6837 [38:13<02:46,  4.09it/s]
 90%|█████████ | 6156/6837 [38:13<02:51,  3.97it/s]
 90%|█████████ | 6157/6837 [38:13<02:52,  3.94it/s]
 90%|█████████ | 6158/6837 [38:14<02:54,  3.89it/s]
 90%|█████████ | 6159/6837 [38:14<02:56,  3.85it/s]
 90%|█████████ | 6160/6837 [38:14<02:55,  3.85it/s]
 90%|█████████ | 6161/6837 [38:14<02:56,  3.82it/s]
 90%|█████████ | 6162/6837 [38:15<02:57,  3.81it/s]
 90%|█████████ | 6163/6837 [38:15<02:58,  3.78it/s]
 90%|█████████ | 6164/6837 [38:15<02:57,  3.79it/s]
 90%|█████████ | 6165/6837 [38:15<02:59,  3.75it/s]
 90%|█████████ | 6166/6837 [38:16<03:00,  3.72it/s]
 90%|█████████ | 6167/6837 [38:16<02:58,  3.76it/s]
 90%|█████████ | 6168/6837 [38:16<02:56,  3.78it/s]
 90%|█████████ | 6169/6837 [38:17<02:57,  3.76it/s]
 90%|█████████ | 6170/6837 [38:17<02:58,  3.74it/s]
 90%|█████████ | 6171/6837 [38:17<02:53,  3.84it/s]
 90%|███████���█ | 6172/6837 [38:17<02:48,  3.95it/s]
 90%|█████████ | 6173/6837 [38:18<02:44,  4.03it/s]
 90%|█████████ | 6174/6837 [38:18<02:42,  4.09it/s]
 90%|█████████ | 6175/6837 [38:18<02:40,  4.13it/s]
                                                   
{'loss': 4.2863, 'grad_norm': 0.15762583911418915, 'learning_rate': 0.00016974402672425225, 'epoch': 0.18}
+
 90%|█████████ | 6175/6837 [38:18<02:40,  4.13it/s]
 90%|█████████ | 6176/6837 [38:18<02:39,  4.15it/s]
 90%|█████████ | 6177/6837 [38:18<02:38,  4.18it/s]
 90%|█████████ | 6178/6837 [38:19<02:37,  4.19it/s]
 90%|█████████ | 6179/6837 [38:19<02:36,  4.19it/s]
 90%|█████████ | 6180/6837 [38:19<02:36,  4.20it/s]
 90%|█████████ | 6181/6837 [38:19<02:35,  4.21it/s]
 90%|█████████ | 6182/6837 [38:20<02:35,  4.22it/s]
 90%|█████████ | 6183/6837 [38:20<02:34,  4.22it/s]
 90%|█████████ | 6184/6837 [38:20<02:34,  4.23it/s]
 90%|█████████ | 6185/6837 [38:20<02:34,  4.23it/s]
 90%|█████████ | 6186/6837 [38:21<02:34,  4.23it/s]
 90%|█████████ | 6187/6837 [38:21<02:33,  4.23it/s]
 91%|█████████ | 6188/6837 [38:21<02:33,  4.23it/s]
 91%|█████████ | 6189/6837 [38:21<02:32,  4.24it/s]
 91%|█████████ | 6190/6837 [38:22<02:32,  4.24it/s]
 91%|█████████ | 6191/6837 [38:22<02:32,  4.24it/s]
 91%|█████████ | 6192/6837 [38:22<02:32,  4.24it/s]
 91%|█████████ | 6193/6837 [38:22<02:32,  4.23it/s]
 91%|█████████ | 6194/6837 [38:22<02:31,  4.23it/s]
 91%|█████████ | 6195/6837 [38:23<02:31,  4.24it/s]
 91%|█████████ | 6196/6837 [38:23<02:31,  4.24it/s]
 91%|█████████ | 6197/6837 [38:23<02:31,  4.24it/s]
 91%|█████████ | 6198/6837 [38:23<02:30,  4.24it/s]
 91%|█████████ | 6199/6837 [38:24<02:30,  4.24it/s]
 91%|█████████ | 6200/6837 [38:24<02:30,  4.23it/s]
                                                   
{'loss': 4.2882, 'grad_norm': 0.15112021565437317, 'learning_rate': 0.00015727668328424315, 'epoch': 0.18}
+
 91%|█████████ | 6200/6837 [38:24<02:30,  4.23it/s]
 91%|█████████ | 6201/6837 [38:24<02:30,  4.21it/s]
 91%|█████████ | 6202/6837 [38:24<02:30,  4.22it/s]
 91%|█████████ | 6203/6837 [38:25<02:30,  4.22it/s]
 91%|█████████ | 6204/6837 [38:25<02:29,  4.23it/s]
 91%|█████████ | 6205/6837 [38:25<02:29,  4.23it/s]
 91%|█████████ | 6206/6837 [38:25<02:28,  4.24it/s]
 91%|█████████ | 6207/6837 [38:26<02:28,  4.23it/s]
 91%|█████████ | 6208/6837 [38:26<02:28,  4.23it/s]
 91%|█████████ | 6209/6837 [38:26<02:28,  4.23it/s]
 91%|█████████ | 6210/6837 [38:26<02:28,  4.24it/s]
 91%|█████████ | 6211/6837 [38:27<02:27,  4.24it/s]
 91%|█████████ | 6212/6837 [38:27<02:27,  4.23it/s]
 91%|█████████ | 6213/6837 [38:27<02:27,  4.23it/s]
 91%|█████████ | 6214/6837 [38:27<02:27,  4.23it/s]
 91%|█████████ | 6215/6837 [38:27<02:27,  4.23it/s]
 91%|█████████ | 6216/6837 [38:28<02:26,  4.23it/s]
 91%|█████████ | 6217/6837 [38:28<02:26,  4.23it/s]
 91%|█████████ | 6218/6837 [38:28<02:26,  4.23it/s]
 91%|█████████ | 6219/6837 [38:28<02:26,  4.23it/s]
 91%|█████████ | 6220/6837 [38:29<02:26,  4.23it/s]
 91%|█████████ | 6221/6837 [38:29<02:26,  4.20it/s]
 91%|█████████ | 6222/6837 [38:29<02:26,  4.20it/s]
 91%|█████████ | 6223/6837 [38:29<02:25,  4.21it/s]
 91%|█████████ | 6224/6837 [38:30<02:25,  4.21it/s]
 91%|█████████ | 6225/6837 [38:30<02:24,  4.22it/s]
                                                   
{'loss': 4.2794, 'grad_norm': 0.14710699021816254, 'learning_rate': 0.0001452725036646474, 'epoch': 0.18}
+
 91%|█████████ | 6225/6837 [38:30<02:24,  4.22it/s]
 91%|█████████ | 6226/6837 [38:30<02:25,  4.21it/s]
 91%|█████████ | 6227/6837 [38:30<02:24,  4.22it/s]
 91%|█████████ | 6228/6837 [38:31<02:24,  4.21it/s]
 91%|█████████ | 6229/6837 [38:31<02:24,  4.22it/s]
 91%|█████████ | 6230/6837 [38:31<02:23,  4.23it/s]
 91%|█████████ | 6231/6837 [38:31<02:23,  4.23it/s]
 91%|█████████ | 6232/6837 [38:31<02:23,  4.23it/s]
 91%|█████████ | 6233/6837 [38:32<02:22,  4.23it/s]
 91%|█████████ | 6234/6837 [38:32<02:22,  4.23it/s]
 91%|█████████ | 6235/6837 [38:32<02:22,  4.23it/s]
 91%|█████████ | 6236/6837 [38:32<02:22,  4.23it/s]
 91%|█████████ | 6237/6837 [38:33<02:21,  4.23it/s]
 91%|█████████ | 6238/6837 [38:33<02:21,  4.23it/s]
 91%|█████████▏| 6239/6837 [38:33<02:21,  4.23it/s]
 91%|█████████▏| 6240/6837 [38:33<02:21,  4.23it/s]
 91%|█████████▏| 6241/6837 [38:34<02:20,  4.24it/s]
 91%|█████████▏| 6242/6837 [38:34<02:20,  4.24it/s]
 91%|█████████▏| 6243/6837 [38:34<02:20,  4.24it/s]
 91%|█████████▏| 6244/6837 [38:34<02:19,  4.24it/s]
 91%|█████████▏| 6245/6837 [38:35<02:19,  4.24it/s]
 91%|█████████▏| 6246/6837 [38:35<02:19,  4.24it/s]
 91%|█████████▏| 6247/6837 [38:35<02:19,  4.24it/s]
 91%|█████████▏| 6248/6837 [38:35<02:18,  4.24it/s]
 91%|█████████▏| 6249/6837 [38:35<02:18,  4.24it/s]
 91%|█████████▏| 6250/6837 [38:36<02:18,  4.24it/s]
                                                   
{'loss': 4.2744, 'grad_norm': 0.14244778454303741, 'learning_rate': 0.0001337334437018498, 'epoch': 0.18}
+
 91%|█████████▏| 6250/6837 [38:36<02:18,  4.24it/s]
 91%|█████████▏| 6251/6837 [38:36<02:18,  4.23it/s]
 91%|█████████▏| 6252/6837 [38:36<02:18,  4.23it/s]
 91%|█████████▏| 6253/6837 [38:36<02:18,  4.23it/s]
 91%|█████████▏| 6254/6837 [38:37<02:17,  4.23it/s]
 91%|█████████▏| 6255/6837 [38:37<02:17,  4.24it/s]
 92%|█████████▏| 6256/6837 [38:37<02:17,  4.24it/s]
 92%|█████████▏| 6257/6837 [38:37<02:16,  4.24it/s]
 92%|█████████▏| 6258/6837 [38:38<02:16,  4.24it/s]
 92%|█████████▏| 6259/6837 [38:38<02:16,  4.24it/s]
 92%|█████████▏| 6260/6837 [38:38<02:16,  4.24it/s]
 92%|█████████▏| 6261/6837 [38:38<02:15,  4.24it/s]
 92%|█████████▏| 6262/6837 [38:39<02:15,  4.24it/s]
 92%|█████████▏| 6263/6837 [38:39<02:15,  4.24it/s]
 92%|█████████▏| 6264/6837 [38:39<02:20,  4.09it/s]
 92%|█████████▏| 6265/6837 [38:39<02:22,  4.01it/s]
 92%|█████████▏| 6266/6837 [38:40<02:24,  3.96it/s]
 92%|█████████▏| 6267/6837 [38:40<02:21,  4.04it/s]
 92%|█████████▏| 6268/6837 [38:40<02:18,  4.10it/s]
 92%|█████████▏| 6269/6837 [38:40<02:16,  4.15it/s]
 92%|█████████▏| 6270/6837 [38:41<02:15,  4.18it/s]
 92%|█████████▏| 6271/6837 [38:41<02:15,  4.19it/s]
 92%|█████████▏| 6272/6837 [38:41<02:14,  4.20it/s]
 92%|█████████▏| 6273/6837 [38:41<02:13,  4.21it/s]
 92%|█████████▏| 6274/6837 [38:41<02:13,  4.22it/s]
 92%|█████████▏| 6275/6837 [38:42<02:12,  4.23it/s]
                                                   {'loss': 4.2938, 'grad_norm': 0.15002533793449402, 'learning_rate': 0.0001226613834504664, 'epoch': 0.18}
+
 92%|█████████▏| 6275/6837 [38:42<02:12,  4.23it/s]
 92%|█████████▏| 6276/6837 [38:42<02:13,  4.21it/s]
 92%|█████████▏| 6277/6837 [38:42<02:12,  4.21it/s]
 92%|█████████▏| 6278/6837 [38:42<02:12,  4.22it/s]
 92%|█████████▏| 6279/6837 [38:43<02:12,  4.22it/s]
 92%|█████████▏| 6280/6837 [38:43<02:16,  4.07it/s]
 92%|█████████▏| 6281/6837 [38:43<02:15,  4.11it/s]
 92%|█████████▏| 6282/6837 [38:43<02:13,  4.15it/s]
 92%|█████████▏| 6283/6837 [38:44<02:12,  4.17it/s]
 92%|█████████▏| 6284/6837 [38:44<02:12,  4.18it/s]
 92%|█████████▏| 6285/6837 [38:44<02:11,  4.20it/s]
 92%|█████████▏| 6286/6837 [38:44<02:10,  4.21it/s]
 92%|█████████▏| 6287/6837 [38:45<02:10,  4.22it/s]
 92%|█████████▏| 6288/6837 [38:45<02:09,  4.23it/s]
 92%|█████████▏| 6289/6837 [38:45<02:09,  4.22it/s]
 92%|█████████▏| 6290/6837 [38:45<02:09,  4.23it/s]
 92%|█████████▏| 6291/6837 [38:46<02:08,  4.23it/s]
 92%|█████████▏| 6292/6837 [38:46<02:08,  4.24it/s]
 92%|█████████▏| 6293/6837 [38:46<02:08,  4.24it/s]
 92%|█████████▏| 6294/6837 [38:46<02:07,  4.25it/s]
 92%|█████████▏| 6295/6837 [38:46<02:07,  4.25it/s]
 92%|█████████▏| 6296/6837 [38:47<02:07,  4.25it/s]
 92%|█████████▏| 6297/6837 [38:47<02:07,  4.25it/s]
 92%|█████████▏| 6298/6837 [38:47<02:07,  4.24it/s]
 92%|█████████▏| 6299/6837 [38:47<02:06,  4.24it/s]
 92%|█████████▏| 6300/6837 [38:48<02:06,  4.25it/s]
                                                   
{'loss': 4.2938, 'grad_norm': 0.14287318289279938, 'learning_rate': 0.00011205812687703132, 'epoch': 0.18}
+
 92%|█████████▏| 6300/6837 [38:48<02:06,  4.25it/s]
 92%|█████████▏| 6301/6837 [38:48<02:06,  4.23it/s]
 92%|█████████▏| 6302/6837 [38:48<02:06,  4.24it/s]
 92%|█████████▏| 6303/6837 [38:48<02:05,  4.25it/s]
 92%|█████████▏| 6304/6837 [38:49<02:05,  4.25it/s]
 92%|█████████▏| 6305/6837 [38:49<02:05,  4.24it/s]
 92%|█████████▏| 6306/6837 [38:49<02:05,  4.24it/s]
 92%|█████████▏| 6307/6837 [38:49<02:05,  4.24it/s]
 92%|█████████▏| 6308/6837 [38:50<02:04,  4.25it/s]
 92%|█████████▏| 6309/6837 [38:50<02:04,  4.25it/s]
 92%|█████████▏| 6310/6837 [38:50<02:03,  4.25it/s]
 92%|█████████▏| 6311/6837 [38:50<02:03,  4.26it/s]
 92%|█████████▏| 6312/6837 [38:50<02:03,  4.26it/s]
 92%|█████████▏| 6313/6837 [38:51<02:03,  4.26it/s]
 92%|█████████▏| 6314/6837 [38:51<02:02,  4.26it/s]
 92%|█████████▏| 6315/6837 [38:51<02:02,  4.26it/s]
 92%|█████████▏| 6316/6837 [38:51<02:02,  4.26it/s]
 92%|█████████▏| 6317/6837 [38:52<02:02,  4.26it/s]
 92%|█████████▏| 6318/6837 [38:52<02:01,  4.25it/s]
 92%|█████████▏| 6319/6837 [38:52<02:01,  4.25it/s]
 92%|█████████▏| 6320/6837 [38:52<02:01,  4.24it/s]
 92%|█████████▏| 6321/6837 [38:53<02:01,  4.25it/s]
 92%|█████████▏| 6322/6837 [38:53<02:01,  4.24it/s]
 92%|█████████▏| 6323/6837 [38:53<02:01,  4.23it/s]
 92%|█████████▏| 6324/6837 [38:53<02:01,  4.23it/s]
 93%|█████████▎| 6325/6837 [38:54<02:00,  4.23it/s]
                                                   
{'loss': 4.2871, 'grad_norm': 0.146336629986763, 'learning_rate': 0.000101925401566074, 'epoch': 0.19}
+
 93%|█████████▎| 6325/6837 [38:54<02:00,  4.23it/s]
 93%|█████████▎| 6326/6837 [38:54<02:01,  4.22it/s]
 93%|█████████▎| 6327/6837 [38:54<02:00,  4.22it/s]
 93%|█████████▎| 6328/6837 [38:54<02:00,  4.23it/s]
 93%|█████████▎| 6329/6837 [38:54<02:00,  4.23it/s]
 93%|█████████▎| 6330/6837 [38:55<01:59,  4.23it/s]
 93%|█████████▎| 6331/6837 [38:55<01:59,  4.23it/s]
 93%|█████████▎| 6332/6837 [38:55<01:59,  4.23it/s]
 93%|█████████▎| 6333/6837 [38:55<01:59,  4.23it/s]
 93%|█████████▎| 6334/6837 [38:56<01:58,  4.23it/s]
 93%|█████████▎| 6335/6837 [38:56<01:58,  4.23it/s]
 93%|█████████▎| 6336/6837 [38:56<01:58,  4.23it/s]
 93%|█████████▎| 6337/6837 [38:56<01:58,  4.23it/s]
 93%|█████████▎| 6338/6837 [38:57<01:57,  4.23it/s]
 93%|█████████▎| 6339/6837 [38:57<01:57,  4.23it/s]
 93%|█████████▎| 6340/6837 [38:57<01:57,  4.22it/s]
 93%|█████████▎| 6341/6837 [38:57<01:57,  4.23it/s]
 93%|█████████▎| 6342/6837 [38:58<01:56,  4.23it/s]
 93%|█████████▎| 6343/6837 [38:58<01:56,  4.23it/s]
 93%|█████████▎| 6344/6837 [38:58<01:56,  4.22it/s]
 93%|█████████▎| 6345/6837 [38:58<01:56,  4.23it/s]
 93%|█████████▎| 6346/6837 [38:58<01:55,  4.23it/s]
 93%|█████████▎| 6347/6837 [38:59<02:00,  4.08it/s]
 93%|█████████▎| 6348/6837 [38:59<02:02,  4.00it/s]
 93%|█████████▎| 6349/6837 [38:59<02:04,  3.94it/s]
 93%|█████████▎| 6350/6837 [39:00<02:01,  3.99it/s]
                                                   
{'loss': 4.2968, 'grad_norm': 0.14501486718654633, 'learning_rate': 9.226485843864463e-05, 'epoch': 0.19}
+
 93%|█████████▎| 6350/6837 [39:00<02:01,  3.99it/s]
 93%|█████████▎| 6351/6837 [39:00<01:59,  4.05it/s]
 93%|█████████▎| 6352/6837 [39:00<01:57,  4.11it/s]
 93%|█████████▎| 6353/6837 [39:00<02:00,  4.01it/s]
 93%|█████████▎| 6354/6837 [39:01<02:03,  3.91it/s]
 93%|█████████▎| 6355/6837 [39:01<02:05,  3.85it/s]
 93%|█████████▎| 6356/6837 [39:01<02:04,  3.85it/s]
 93%|█████████▎| 6357/6837 [39:01<02:05,  3.83it/s]
 93%|█████████▎| 6358/6837 [39:02<02:05,  3.81it/s]
 93%|█████████▎| 6359/6837 [39:02<02:05,  3.82it/s]
 93%|█████████▎| 6360/6837 [39:02<02:01,  3.92it/s]
 93%|█████████▎| 6361/6837 [39:02<01:58,  4.02it/s]
 93%|█████████▎| 6362/6837 [39:03<01:56,  4.09it/s]
 93%|█████████▎| 6363/6837 [39:03<01:54,  4.14it/s]
 93%|█████████▎| 6364/6837 [39:03<01:53,  4.17it/s]
 93%|█████████▎| 6365/6837 [39:03<01:52,  4.20it/s]
 93%|█████████▎| 6366/6837 [39:04<01:51,  4.21it/s]
 93%|█████████▎| 6367/6837 [39:04<01:51,  4.22it/s]
 93%|█████████▎| 6368/6837 [39:04<02:30,  3.12it/s]
 93%|█████████▎| 6369/6837 [39:04<02:18,  3.39it/s]
 93%|█████████▎| 6370/6837 [39:05<02:09,  3.61it/s]
 93%|█████████▎| 6371/6837 [39:05<02:03,  3.78it/s]
 93%|█████████▎| 6372/6837 [39:05<01:59,  3.91it/s]
 93%|█████████▎| 6373/6837 [39:05<02:00,  3.87it/s]
 93%|█████████▎| 6374/6837 [39:06<02:00,  3.85it/s]
 93%|█████████▎| 6375/6837 [39:06<02:00,  3.84it/s]
                                                   
{'loss': 4.2869, 'grad_norm': 0.1443915218114853, 'learning_rate': 8.307807148333269e-05, 'epoch': 0.19}
+
 93%|█████████▎| 6375/6837 [39:06<02:00,  3.84it/s]
 93%|█████████▎| 6376/6837 [39:06<02:00,  3.83it/s]
 93%|█████████▎| 6377/6837 [39:07<02:00,  3.83it/s]
 93%|█████████▎| 6378/6837 [39:07<02:00,  3.82it/s]
 93%|█████████▎| 6379/6837 [39:07<02:00,  3.82it/s]
 93%|█████████▎| 6380/6837 [39:07<01:59,  3.81it/s]
 93%|█████████▎| 6381/6837 [39:08<02:00,  3.80it/s]
 93%|█████████▎| 6382/6837 [39:08<01:59,  3.80it/s]
 93%|█████████▎| 6383/6837 [39:08<01:59,  3.79it/s]
 93%|█████████▎| 6384/6837 [39:08<01:58,  3.81it/s]
 93%|█████████▎| 6385/6837 [39:09<01:59,  3.79it/s]
 93%|█████████▎| 6386/6837 [39:09<01:59,  3.79it/s]
 93%|█████████▎| 6387/6837 [39:09<01:59,  3.78it/s]
 93%|█████████▎| 6388/6837 [39:09<01:58,  3.78it/s]
 93%|█████████▎| 6389/6837 [39:10<01:58,  3.77it/s]
 93%|█████████▎| 6390/6837 [39:10<01:58,  3.78it/s]
 93%|█████████▎| 6391/6837 [39:10<01:58,  3.76it/s]
 93%|█████████▎| 6392/6837 [39:10<01:57,  3.77it/s]
 94%|█████████▎| 6393/6837 [39:11<01:56,  3.80it/s]
 94%|█████████▎| 6394/6837 [39:11<01:55,  3.83it/s]
 94%|█████████▎| 6395/6837 [39:11<01:55,  3.81it/s]
 94%|█████████▎| 6396/6837 [39:12<01:55,  3.81it/s]
 94%|█████████▎| 6397/6837 [39:12<01:55,  3.81it/s]
 94%|█████████▎| 6398/6837 [39:12<01:55,  3.80it/s]
 94%|█████████▎| 6399/6837 [39:12<01:55,  3.78it/s]
 94%|█████████▎| 6400/6837 [39:13<01:55,  3.79it/s]
                                                   
{'loss': 4.2862, 'grad_norm': 0.14435304701328278, 'learning_rate': 7.43665374998157e-05, 'epoch': 0.19}
+
 94%|█████████▎| 6400/6837 [39:13<01:55,  3.79it/s]
 94%|█████████▎| 6401/6837 [39:13<01:55,  3.78it/s]
 94%|█████████▎| 6402/6837 [39:13<01:55,  3.78it/s]
 94%|█████████▎| 6403/6837 [39:13<01:54,  3.79it/s]
 94%|█████████▎| 6404/6837 [39:14<01:54,  3.79it/s]
 94%|█████████▎| 6405/6837 [39:14<01:52,  3.84it/s]
 94%|█████████▎| 6406/6837 [39:14<01:49,  3.95it/s]
 94%|█████████▎| 6407/6837 [39:14<01:46,  4.03it/s]
 94%|█████████▎| 6408/6837 [39:15<01:45,  4.08it/s]
 94%|█████████▎| 6409/6837 [39:15<01:43,  4.13it/s]
 94%|█████████▍| 6410/6837 [39:15<01:42,  4.16it/s]
 94%|█████████▍| 6411/6837 [39:15<01:41,  4.18it/s]
 94%|█████████▍| 6412/6837 [39:16<01:41,  4.20it/s]
 94%|█████████▍| 6413/6837 [39:16<01:40,  4.21it/s]
 94%|█████████▍| 6414/6837 [39:16<01:40,  4.21it/s]
 94%|█████████▍| 6415/6837 [39:16<01:40,  4.21it/s]
 94%|█████████▍| 6416/6837 [39:16<01:39,  4.22it/s]
 94%|█████████▍| 6417/6837 [39:17<01:39,  4.23it/s]
 94%|█████████▍| 6418/6837 [39:17<01:39,  4.23it/s]
 94%|█████████▍| 6419/6837 [39:17<01:38,  4.23it/s]
 94%|█████████▍| 6420/6837 [39:17<01:38,  4.23it/s]
 94%|█████████▍| 6421/6837 [39:18<01:38,  4.24it/s]
 94%|█████████▍| 6422/6837 [39:18<01:38,  4.23it/s]
 94%|█████████▍| 6423/6837 [39:18<01:37,  4.23it/s]
 94%|█████████▍| 6424/6837 [39:18<01:37,  4.23it/s]
 94%|█████████▍| 6425/6837 [39:19<01:37,  4.23it/s]
                                                   
{'loss': 4.2897, 'grad_norm': 0.14084896445274353, 'learning_rate': 6.613167585498636e-05, 'epoch': 0.19}
+
 94%|█████████▍| 6425/6837 [39:19<01:37,  4.23it/s]
 94%|█████████▍| 6426/6837 [39:19<01:37,  4.22it/s]
 94%|█████████▍| 6427/6837 [39:19<01:36,  4.23it/s]
 94%|█████████▍| 6428/6837 [39:19<01:36,  4.23it/s]
 94%|█████████▍| 6429/6837 [39:20<01:36,  4.23it/s]
 94%|█████████▍| 6430/6837 [39:20<01:36,  4.23it/s]
 94%|█████████▍| 6431/6837 [39:20<01:35,  4.23it/s]
 94%|█████████▍| 6432/6837 [39:20<01:35,  4.23it/s]
 94%|█████████▍| 6433/6837 [39:21<01:35,  4.23it/s]
 94%|█████████▍| 6434/6837 [39:21<01:36,  4.17it/s]
 94%|█████████▍| 6435/6837 [39:21<01:35,  4.19it/s]
 94%|█████████▍| 6436/6837 [39:21<01:35,  4.20it/s]
 94%|█████████▍| 6437/6837 [39:21<01:34,  4.22it/s]
 94%|█████████▍| 6438/6837 [39:22<01:34,  4.23it/s]
 94%|█████████▍| 6439/6837 [39:22<01:34,  4.23it/s]
 94%|█████████▍| 6440/6837 [39:22<01:33,  4.23it/s]
 94%|█████████▍| 6441/6837 [39:22<01:33,  4.24it/s]
 94%|█████████▍| 6442/6837 [39:23<01:33,  4.24it/s]
 94%|█████████▍| 6443/6837 [39:23<01:32,  4.24it/s]
 94%|█████████▍| 6444/6837 [39:23<01:32,  4.24it/s]
 94%|█████████▍| 6445/6837 [39:23<01:32,  4.24it/s]
 94%|█████████▍| 6446/6837 [39:24<01:32,  4.24it/s]
 94%|█████████▍| 6447/6837 [39:24<01:32,  4.24it/s]
 94%|█████████▍| 6448/6837 [39:24<01:31,  4.24it/s]
 94%|█████████▍| 6449/6837 [39:24<01:31,  4.24it/s]
 94%|█████████▍| 6450/6837 [39:25<01:31,  4.23it/s]
                                                   
{'loss': 4.287, 'grad_norm': 0.14081266522407532, 'learning_rate': 5.837482825169616e-05, 'epoch': 0.19}
+
 94%|█████████▍| 6450/6837 [39:25<01:31,  4.23it/s]
 94%|█████████▍| 6451/6837 [39:25<01:31,  4.22it/s]
 94%|█████████▍| 6452/6837 [39:25<01:31,  4.22it/s]
 94%|█████████▍| 6453/6837 [39:25<01:30,  4.23it/s]
 94%|█████████▍| 6454/6837 [39:25<01:30,  4.23it/s]
 94%|█████████▍| 6455/6837 [39:26<01:30,  4.23it/s]
 94%|█████████▍| 6456/6837 [39:26<01:29,  4.24it/s]
 94%|█████████▍| 6457/6837 [39:26<01:29,  4.23it/s]
 94%|█████████▍| 6458/6837 [39:26<01:29,  4.24it/s]
 94%|█████████▍| 6459/6837 [39:27<01:29,  4.24it/s]
 94%|█████████▍| 6460/6837 [39:27<01:28,  4.24it/s]
 95%|█████████▍| 6461/6837 [39:27<01:28,  4.24it/s]
 95%|█████████▍| 6462/6837 [39:27<01:28,  4.25it/s]
 95%|█████████▍| 6463/6837 [39:28<01:31,  4.10it/s]
 95%|█████████▍| 6464/6837 [39:28<01:33,  3.99it/s]
 95%|█████████▍| 6465/6837 [39:28<01:34,  3.92it/s]
 95%|█████████▍| 6466/6837 [39:28<01:32,  4.01it/s]
 95%|█████████▍| 6467/6837 [39:29<01:31,  4.06it/s]
 95%|█████████▍| 6468/6837 [39:29<01:29,  4.10it/s]
 95%|█████████▍| 6469/6837 [39:29<01:29,  4.13it/s]
 95%|█████████▍| 6470/6837 [39:29<01:28,  4.16it/s]
 95%|█████████▍| 6471/6837 [39:30<01:27,  4.18it/s]
 95%|█████████▍| 6472/6837 [39:30<01:26,  4.20it/s]
 95%|█████████▍| 6473/6837 [39:30<01:26,  4.21it/s]
 95%|█████████▍| 6474/6837 [39:30<01:26,  4.22it/s]
 95%|█████████▍| 6475/6837 [39:31<01:25,  4.23it/s]
                                                   
{'loss': 4.2769, 'grad_norm': 0.14077018201351166, 'learning_rate': 5.109725851015256e-05, 'epoch': 0.19}
+
 95%|█████████▍| 6475/6837 [39:31<01:25,  4.23it/s]
 95%|█████████▍| 6476/6837 [39:31<01:25,  4.22it/s]
 95%|█████████▍| 6477/6837 [39:31<01:25,  4.23it/s]
 95%|█████████▍| 6478/6837 [39:31<01:24,  4.23it/s]
 95%|█████████▍| 6479/6837 [39:31<01:24,  4.24it/s]
 95%|█████████▍| 6480/6837 [39:32<01:24,  4.24it/s]
 95%|█████████▍| 6481/6837 [39:32<01:23,  4.25it/s]
 95%|█████████▍| 6482/6837 [39:32<01:23,  4.25it/s]
 95%|█████████▍| 6483/6837 [39:32<01:23,  4.23it/s]
 95%|█████████▍| 6484/6837 [39:33<01:23,  4.23it/s]
 95%|█████████▍| 6485/6837 [39:33<01:23,  4.23it/s]
 95%|█████████▍| 6486/6837 [39:33<01:22,  4.24it/s]
 95%|█████████▍| 6487/6837 [39:33<01:22,  4.24it/s]
 95%|█████████▍| 6488/6837 [39:34<01:22,  4.24it/s]
 95%|█████████▍| 6489/6837 [39:34<01:22,  4.24it/s]
 95%|█████████▍| 6490/6837 [39:34<01:21,  4.24it/s]
 95%|█████████▍| 6491/6837 [39:34<01:21,  4.24it/s]
 95%|█████████▍| 6492/6837 [39:35<01:21,  4.24it/s]
 95%|█████████▍| 6493/6837 [39:35<01:21,  4.24it/s]
 95%|█████████▍| 6494/6837 [39:35<01:20,  4.24it/s]
 95%|█████████▍| 6495/6837 [39:35<01:20,  4.25it/s]
 95%|█████████▌| 6496/6837 [39:35<01:20,  4.25it/s]
 95%|█████████▌| 6497/6837 [39:36<01:20,  4.25it/s]
 95%|█████████▌| 6498/6837 [39:36<01:19,  4.25it/s]
 95%|█████████▌| 6499/6837 [39:36<01:19,  4.25it/s]
 95%|█████████▌| 6500/6837 [39:36<01:19,  4.24it/s]
                                                   
{'loss': 4.2817, 'grad_norm': 0.14405563473701477, 'learning_rate': 4.4300152362003466e-05, 'epoch': 0.19}
+
 95%|█████████▌| 6500/6837 [39:36<01:19,  4.24it/s]
 95%|█████████▌| 6501/6837 [39:37<01:19,  4.23it/s]
 95%|█████████▌| 6502/6837 [39:37<01:19,  4.23it/s]
 95%|█████████▌| 6503/6837 [39:37<01:18,  4.23it/s]
 95%|█████████▌| 6504/6837 [39:37<01:18,  4.23it/s]
 95%|█████████▌| 6505/6837 [39:38<01:18,  4.23it/s]
 95%|█████████▌| 6506/6837 [39:38<01:18,  4.22it/s]
 95%|█████████▌| 6507/6837 [39:38<01:18,  4.22it/s]
 95%|█████████▌| 6508/6837 [39:38<01:17,  4.23it/s]
 95%|█████████▌| 6509/6837 [39:39<01:17,  4.23it/s]
 95%|█████████▌| 6510/6837 [39:39<01:17,  4.22it/s]
 95%|█████████▌| 6511/6837 [39:39<01:17,  4.23it/s]
 95%|█████████▌| 6512/6837 [39:39<01:16,  4.23it/s]
 95%|█████████▌| 6513/6837 [39:39<01:16,  4.23it/s]
 95%|█████████▌| 6514/6837 [39:40<01:16,  4.24it/s]
 95%|█████████▌| 6515/6837 [39:40<01:15,  4.24it/s]
 95%|█████████▌| 6516/6837 [39:40<01:15,  4.24it/s]
 95%|█████████▌| 6517/6837 [39:40<01:15,  4.24it/s]
 95%|█████████▌| 6518/6837 [39:41<01:15,  4.24it/s]
 95%|█████████▌| 6519/6837 [39:41<01:14,  4.24it/s]
 95%|█████████▌| 6520/6837 [39:41<01:14,  4.24it/s]
 95%|█████████▌| 6521/6837 [39:41<01:14,  4.24it/s]
 95%|█████████▌| 6522/6837 [39:42<01:14,  4.24it/s]
 95%|█████████▌| 6523/6837 [39:42<01:14,  4.24it/s]
 95%|█████████▌| 6524/6837 [39:42<01:13,  4.24it/s]
 95%|█████████▌| 6525/6837 [39:42<01:13,  4.24it/s]
                                                   
{'loss': 4.2885, 'grad_norm': 0.14118966460227966, 'learning_rate': 3.7984617257148634e-05, 'epoch': 0.19}
+
 95%|█████████▌| 6525/6837 [39:42<01:13,  4.24it/s]
 95%|█████████▌| 6526/6837 [39:43<01:13,  4.23it/s]
 95%|█████████▌| 6527/6837 [39:43<01:16,  4.06it/s]
 95%|█████████▌| 6528/6837 [39:43<01:18,  3.95it/s]
 95%|█████████▌| 6529/6837 [39:43<01:18,  3.93it/s]
 96%|█████████▌| 6530/6837 [39:44<01:17,  3.98it/s]
 96%|█████████▌| 6531/6837 [39:44<01:15,  4.06it/s]
 96%|█████████▌| 6532/6837 [39:44<01:14,  4.11it/s]
 96%|█████████▌| 6533/6837 [39:44<01:13,  4.14it/s]
 96%|█████████▌| 6534/6837 [39:45<01:12,  4.16it/s]
 96%|█████████▌| 6535/6837 [39:45<01:12,  4.18it/s]
 96%|█████████▌| 6536/6837 [39:45<01:11,  4.19it/s]
 96%|█████████▌| 6537/6837 [39:45<01:11,  4.20it/s]
 96%|█████████▌| 6538/6837 [39:45<01:11,  4.20it/s]
 96%|█████████▌| 6539/6837 [39:46<01:10,  4.21it/s]
 96%|█████████▌| 6540/6837 [39:46<01:10,  4.21it/s]
 96%|█████████▌| 6541/6837 [39:46<01:10,  4.21it/s]
 96%|█████████▌| 6542/6837 [39:46<01:10,  4.21it/s]
 96%|█████████▌| 6543/6837 [39:47<01:09,  4.22it/s]
 96%|█████████▌| 6544/6837 [39:47<01:09,  4.22it/s]
 96%|█████████▌| 6545/6837 [39:47<01:09,  4.22it/s]
 96%|█████████▌| 6546/6837 [39:47<01:08,  4.22it/s]
 96%|█████████▌| 6547/6837 [39:48<01:08,  4.22it/s]
 96%|█████████▌| 6548/6837 [39:48<01:08,  4.22it/s]
 96%|█████████▌| 6549/6837 [39:48<01:08,  4.22it/s]
 96%|█████████▌| 6550/6837 [39:48<01:07,  4.23it/s]
                                                   
{'loss': 4.2816, 'grad_norm': 0.1524515599012375, 'learning_rate': 3.2151682183301775e-05, 'epoch': 0.19}
+
 96%|█████████▌| 6550/6837 [39:48<01:07,  4.23it/s]
 96%|█████████▌| 6551/6837 [39:49<01:07,  4.22it/s]
 96%|█████████▌| 6552/6837 [39:49<01:07,  4.22it/s]
 96%|█████████▌| 6553/6837 [39:49<01:07,  4.22it/s]
 96%|█████████▌| 6554/6837 [39:49<01:07,  4.22it/s]
 96%|█████████▌| 6555/6837 [39:50<01:06,  4.22it/s]
 96%|█████████▌| 6556/6837 [39:50<01:06,  4.22it/s]
 96%|█████████▌| 6557/6837 [39:50<01:06,  4.23it/s]
 96%|█████████▌| 6558/6837 [39:50<01:06,  4.23it/s]
 96%|█████████▌| 6559/6837 [39:50<01:05,  4.23it/s]
 96%|█████████▌| 6560/6837 [39:51<01:05,  4.23it/s]
 96%|█████████▌| 6561/6837 [39:51<01:05,  4.24it/s]
 96%|█████████▌| 6562/6837 [39:51<01:04,  4.23it/s]
 96%|█████████▌| 6563/6837 [39:51<01:04,  4.23it/s]
 96%|█████████▌| 6564/6837 [39:52<01:04,  4.23it/s]
 96%|█████████▌| 6565/6837 [39:52<01:04,  4.24it/s]
 96%|█████████▌| 6566/6837 [39:52<01:04,  4.23it/s]
 96%|█████████▌| 6567/6837 [39:52<01:03,  4.24it/s]
 96%|█████████▌| 6568/6837 [39:53<01:03,  4.25it/s]
 96%|█████████▌| 6569/6837 [39:53<01:03,  4.25it/s]
 96%|█████████▌| 6570/6837 [39:53<01:02,  4.25it/s]
 96%|█████████▌| 6571/6837 [39:53<01:02,  4.25it/s]
 96%|█████████▌| 6572/6837 [39:54<01:02,  4.25it/s]
 96%|█████████▌| 6573/6837 [39:54<01:02,  4.25it/s]
 96%|█████████▌| 6574/6837 [39:54<01:01,  4.25it/s]
 96%|█████████▌| 6575/6837 [39:54<01:01,  4.25it/s]
                                                   
{'loss': 4.2794, 'grad_norm': 0.14205096662044525, 'learning_rate': 2.680229749833929e-05, 'epoch': 0.19}
+
 96%|█████████▌| 6575/6837 [39:54<01:01,  4.25it/s]
 96%|█████████▌| 6576/6837 [39:54<01:01,  4.23it/s]
 96%|█████████▌| 6577/6837 [39:55<01:01,  4.24it/s]
 96%|█████████▌| 6578/6837 [39:55<01:01,  4.24it/s]
 96%|█████████▌| 6579/6837 [39:55<01:00,  4.24it/s]
 96%|█████████▌| 6580/6837 [39:55<01:02,  4.08it/s]
 96%|█████████▋| 6581/6837 [39:56<01:04,  3.98it/s]
 96%|█████████▋| 6582/6837 [39:56<01:04,  3.93it/s]
 96%|█████████▋| 6583/6837 [39:56<01:03,  3.99it/s]
 96%|█████████▋| 6584/6837 [39:56<01:02,  4.05it/s]
 96%|█████████▋| 6585/6837 [39:57<01:01,  4.11it/s]
 96%|█████████▋| 6586/6837 [39:57<01:00,  4.14it/s]
 96%|█████████▋| 6587/6837 [39:57<01:00,  4.16it/s]
 96%|█████████▋| 6588/6837 [39:57<00:59,  4.18it/s]
 96%|█████████▋| 6589/6837 [39:58<00:59,  4.20it/s]
 96%|█████████▋| 6590/6837 [39:58<00:58,  4.21it/s]
 96%|█████████▋| 6591/6837 [39:58<00:58,  4.22it/s]
 96%|█████████▋| 6592/6837 [39:58<00:57,  4.23it/s]
 96%|█████████▋| 6593/6837 [39:59<00:57,  4.21it/s]
 96%|█████████▋| 6594/6837 [39:59<00:57,  4.20it/s]
 96%|█████████▋| 6595/6837 [39:59<00:57,  4.20it/s]
 96%|█████████▋| 6596/6837 [39:59<00:57,  4.20it/s]
 96%|█████████▋| 6597/6837 [40:00<00:57,  4.21it/s]
 97%|█████████▋| 6598/6837 [40:00<00:56,  4.22it/s]
 97%|█████████▋| 6599/6837 [40:00<00:56,  4.22it/s]
 97%|█████████▋| 6600/6837 [40:00<00:56,  4.22it/s]
                                                   
{'loss': 4.2741, 'grad_norm': 0.14347386360168457, 'learning_rate': 2.193733477545712e-05, 'epoch': 0.19}
+
 97%|█████████▋| 6600/6837 [40:00<00:56,  4.22it/s]
 97%|█████████▋| 6601/6837 [40:00<00:56,  4.21it/s]
 97%|█████████▋| 6602/6837 [40:01<00:55,  4.22it/s]
 97%|█████████▋| 6603/6837 [40:01<00:55,  4.22it/s]
 97%|█████████▋| 6604/6837 [40:01<00:55,  4.21it/s]
 97%|█████████▋| 6605/6837 [40:01<00:55,  4.22it/s]
 97%|█████████▋| 6606/6837 [40:02<00:54,  4.22it/s]
 97%|█████████▋| 6607/6837 [40:02<00:54,  4.22it/s]
 97%|█████████▋| 6608/6837 [40:02<00:54,  4.22it/s]
 97%|█████████▋| 6609/6837 [40:02<00:53,  4.23it/s]
 97%|█████████▋| 6610/6837 [40:03<00:53,  4.23it/s]
 97%|█████████▋| 6611/6837 [40:03<00:53,  4.23it/s]
 97%|█████████▋| 6612/6837 [40:03<00:53,  4.23it/s]
 97%|█████████▋| 6613/6837 [40:03<00:52,  4.23it/s]
 97%|█████████▋| 6614/6837 [40:04<00:52,  4.21it/s]
 97%|█████████▋| 6615/6837 [40:04<00:52,  4.22it/s]
 97%|█████████▋| 6616/6837 [40:04<00:52,  4.22it/s]
 97%|█████████▋| 6617/6837 [40:04<00:52,  4.22it/s]
 97%|█████████▋| 6618/6837 [40:05<00:51,  4.22it/s]
 97%|█████████▋| 6619/6837 [40:05<00:51,  4.22it/s]
 97%|█████████▋| 6620/6837 [40:05<00:51,  4.23it/s]
 97%|█████████▋| 6621/6837 [40:05<00:51,  4.23it/s]
 97%|█████████▋| 6622/6837 [40:05<00:50,  4.23it/s]
 97%|█████████▋| 6623/6837 [40:06<00:50,  4.23it/s]
 97%|█████████▋| 6624/6837 [40:06<00:50,  4.23it/s]
 97%|█████████▋| 6625/6837 [40:06<00:50,  4.23it/s]
                                                   
{'loss': 4.2872, 'grad_norm': 0.1419345885515213, 'learning_rate': 1.7557586661167357e-05, 'epoch': 0.19}
+
 97%|█████████▋| 6625/6837 [40:06<00:50,  4.23it/s]
 97%|█████████▋| 6626/6837 [40:06<00:49,  4.22it/s]
 97%|█████████▋| 6627/6837 [40:07<00:49,  4.23it/s]
 97%|█████████▋| 6628/6837 [40:07<00:49,  4.22it/s]
 97%|█████████▋| 6629/6837 [40:07<00:49,  4.23it/s]
 97%|█████████▋| 6630/6837 [40:07<00:48,  4.23it/s]
 97%|█████████▋| 6631/6837 [40:08<00:48,  4.23it/s]
 97%|█████████▋| 6632/6837 [40:08<00:48,  4.23it/s]
 97%|█████████▋| 6633/6837 [40:08<00:48,  4.24it/s]
 97%|█████████▋| 6634/6837 [40:08<00:47,  4.24it/s]
 97%|█████████▋| 6635/6837 [40:09<00:47,  4.23it/s]
 97%|█████████▋| 6636/6837 [40:09<00:47,  4.24it/s]
 97%|█████████▋| 6637/6837 [40:09<00:47,  4.24it/s]
 97%|█████████▋| 6638/6837 [40:09<00:46,  4.24it/s]
 97%|█████████▋| 6639/6837 [40:09<00:46,  4.24it/s]
 97%|█████████▋| 6640/6837 [40:10<00:46,  4.24it/s]
 97%|█████████▋| 6641/6837 [40:10<00:46,  4.24it/s]
 97%|█████████▋| 6642/6837 [40:10<00:45,  4.25it/s]
 97%|█████████▋| 6643/6837 [40:10<00:45,  4.25it/s]
 97%|█████████▋| 6644/6837 [40:11<00:45,  4.25it/s]
 97%|█████████▋| 6645/6837 [40:11<00:45,  4.25it/s]
 97%|█████████▋| 6646/6837 [40:11<00:44,  4.25it/s]
 97%|█████████▋| 6647/6837 [40:11<00:44,  4.24it/s]
 97%|█████████▋| 6648/6837 [40:12<00:44,  4.25it/s]
 97%|█████████▋| 6649/6837 [40:12<00:44,  4.25it/s]
 97%|█████████▋| 6650/6837 [40:12<00:44,  4.25it/s]
                                                   
{'loss': 4.2689, 'grad_norm': 0.14699147641658783, 'learning_rate': 1.3663766746151218e-05, 'epoch': 0.19}
+
 97%|█████████▋| 6650/6837 [40:12<00:44,  4.25it/s]
 97%|█████████▋| 6651/6837 [40:12<00:43,  4.24it/s]
 97%|█████████▋| 6652/6837 [40:13<00:43,  4.24it/s]
 97%|█████████▋| 6653/6837 [40:13<00:45,  4.06it/s]
 97%|█████████▋| 6654/6837 [40:13<00:46,  3.95it/s]
 97%|█████████▋| 6655/6837 [40:13<00:45,  4.03it/s]
 97%|█████████▋| 6656/6837 [40:14<00:44,  4.08it/s]
 97%|█████████▋| 6657/6837 [40:14<00:43,  4.13it/s]
 97%|█████████▋| 6658/6837 [40:14<00:42,  4.16it/s]
 97%|█████████▋| 6659/6837 [40:14<00:42,  4.19it/s]
 97%|█████████▋| 6660/6837 [40:14<00:42,  4.20it/s]
 97%|█████████▋| 6661/6837 [40:15<00:41,  4.21it/s]
 97%|█████████▋| 6662/6837 [40:15<00:41,  4.22it/s]
 97%|█████████▋| 6663/6837 [40:15<00:41,  4.22it/s]
 97%|█████████▋| 6664/6837 [40:15<00:40,  4.23it/s]
 97%|█████████▋| 6665/6837 [40:16<00:40,  4.23it/s]
 97%|█████████▋| 6666/6837 [40:16<00:40,  4.23it/s]
 98%|█████████▊| 6667/6837 [40:16<00:40,  4.23it/s]
 98%|█████████▊| 6668/6837 [40:16<00:39,  4.23it/s]
 98%|█████████▊| 6669/6837 [40:17<00:39,  4.23it/s]
 98%|█████████▊| 6670/6837 [40:17<00:39,  4.24it/s]
 98%|█████████▊| 6671/6837 [40:17<00:39,  4.24it/s]
 98%|█████████▊| 6672/6837 [40:17<00:38,  4.24it/s]
 98%|█████████▊| 6673/6837 [40:18<00:38,  4.24it/s]
 98%|█████████▊| 6674/6837 [40:18<00:38,  4.24it/s]
 98%|█████████▊| 6675/6837 [40:18<00:38,  4.25it/s]{'loss': 4.2927, 'grad_norm': 0.145946204662323, 'learning_rate': 1.0256509448995366e-05, 'epoch': 0.2}
                                                   
+
 98%|█████████▊| 6675/6837 [40:18<00:38,  4.25it/s]
 98%|█████████▊| 6676/6837 [40:18<00:38,  4.23it/s]
 98%|█████████▊| 6677/6837 [40:19<00:37,  4.22it/s]
 98%|█████████▊| 6678/6837 [40:19<00:37,  4.22it/s]
 98%|█████████▊| 6679/6837 [40:19<00:37,  4.23it/s]
 98%|█████████▊| 6680/6837 [40:19<00:37,  4.23it/s]
 98%|█████████▊| 6681/6837 [40:19<00:36,  4.23it/s]
 98%|█████████▊| 6682/6837 [40:20<00:36,  4.23it/s]
 98%|█████████▊| 6683/6837 [40:20<00:36,  4.23it/s]
 98%|█████████▊| 6684/6837 [40:20<00:36,  4.23it/s]
 98%|█████████▊| 6685/6837 [40:20<00:35,  4.23it/s]
 98%|█████████▊| 6686/6837 [40:21<00:35,  4.23it/s]
 98%|█████████▊| 6687/6837 [40:21<00:35,  4.23it/s]
 98%|█████████▊| 6688/6837 [40:21<00:35,  4.23it/s]
 98%|█████████▊| 6689/6837 [40:21<00:34,  4.23it/s]
 98%|█████████▊| 6690/6837 [40:22<00:34,  4.24it/s]
 98%|█████████▊| 6691/6837 [40:22<00:34,  4.23it/s]
 98%|█████████▊| 6692/6837 [40:22<00:34,  4.24it/s]
 98%|█████████▊| 6693/6837 [40:22<00:33,  4.24it/s]
 98%|█████████▊| 6694/6837 [40:23<00:33,  4.24it/s]
 98%|█████████▊| 6695/6837 [40:23<00:33,  4.24it/s]
 98%|█████████▊| 6696/6837 [40:23<00:33,  4.24it/s]
 98%|█████████▊| 6697/6837 [40:23<00:33,  4.24it/s]
 98%|█████████▊| 6698/6837 [40:23<00:32,  4.23it/s]
 98%|█████████▊| 6699/6837 [40:24<00:32,  4.23it/s]
 98%|█████████▊| 6700/6837 [40:24<00:32,  4.23it/s]
                                                   
{'loss': 4.2651, 'grad_norm': 0.14143434166908264, 'learning_rate': 7.336369912824936e-06, 'epoch': 0.2}
+
 98%|█████████▊| 6700/6837 [40:24<00:32,  4.23it/s]
 98%|█████████▊| 6701/6837 [40:24<00:32,  4.22it/s]
 98%|█████████▊| 6702/6837 [40:24<00:32,  4.22it/s]
 98%|█████████▊| 6703/6837 [40:25<00:31,  4.22it/s]
 98%|█████████▊| 6704/6837 [40:25<00:31,  4.23it/s]
 98%|█████████▊| 6705/6837 [40:25<00:31,  4.23it/s]
 98%|█████████▊| 6706/6837 [40:25<00:30,  4.23it/s]
 98%|█████████▊| 6707/6837 [40:26<00:30,  4.23it/s]
 98%|█████████▊| 6708/6837 [40:26<00:30,  4.23it/s]
 98%|█████████▊| 6709/6837 [40:26<00:30,  4.23it/s]
 98%|█████████▊| 6710/6837 [40:26<00:30,  4.23it/s]
 98%|█████████▊| 6711/6837 [40:27<00:29,  4.24it/s]
 98%|█████████▊| 6712/6837 [40:27<00:29,  4.23it/s]
 98%|█████████▊| 6713/6837 [40:27<00:29,  4.23it/s]
 98%|█████████▊| 6714/6837 [40:27<00:29,  4.23it/s]
 98%|█████████▊| 6715/6837 [40:27<00:28,  4.24it/s]
 98%|█████████▊| 6716/6837 [40:28<00:28,  4.24it/s]
 98%|█████████▊| 6717/6837 [40:28<00:28,  4.23it/s]
 98%|█████████▊| 6718/6837 [40:28<00:28,  4.24it/s]
 98%|█████████▊| 6719/6837 [40:28<00:27,  4.23it/s]
 98%|█████████▊| 6720/6837 [40:29<00:27,  4.22it/s]
 98%|█████████▊| 6721/6837 [40:29<00:27,  4.21it/s]
 98%|█████████▊| 6722/6837 [40:29<00:27,  4.22it/s]
 98%|█████████▊| 6723/6837 [40:29<00:27,  4.22it/s]
 98%|█████████▊| 6724/6837 [40:30<00:26,  4.23it/s]
 98%|█████████▊| 6725/6837 [40:30<00:26,  4.23it/s]
                                                   
{'loss': 4.2814, 'grad_norm': 0.14146795868873596, 'learning_rate': 4.903823914855554e-06, 'epoch': 0.2}
+
 98%|█████████▊| 6725/6837 [40:30<00:26,  4.23it/s]
 98%|█████████▊| 6726/6837 [40:30<00:26,  4.22it/s]
 98%|█████████▊| 6727/6837 [40:30<00:26,  4.23it/s]
 98%|█████████▊| 6728/6837 [40:31<00:25,  4.23it/s]
 98%|█████████▊| 6729/6837 [40:31<00:25,  4.23it/s]
 98%|█████████▊| 6730/6837 [40:31<00:25,  4.23it/s]
 98%|█████████▊| 6731/6837 [40:31<00:25,  4.24it/s]
 98%|█████████▊| 6732/6837 [40:32<00:24,  4.25it/s]
 98%|█████████▊| 6733/6837 [40:32<00:24,  4.25it/s]
 98%|█████████▊| 6734/6837 [40:32<00:24,  4.25it/s]
 99%|█████████▊| 6735/6837 [40:32<00:23,  4.25it/s]
 99%|█████████▊| 6736/6837 [40:32<00:23,  4.25it/s]
 99%|█████████▊| 6737/6837 [40:33<00:23,  4.24it/s]
 99%|█████████▊| 6738/6837 [40:33<00:23,  4.25it/s]
 99%|█████████▊| 6739/6837 [40:33<00:23,  4.25it/s]
 99%|████��████▊| 6740/6837 [40:33<00:22,  4.24it/s]
 99%|█████████▊| 6741/6837 [40:34<00:22,  4.25it/s]
 99%|█████████▊| 6742/6837 [40:34<00:22,  4.24it/s]
 99%|█████████▊| 6743/6837 [40:34<00:22,  4.24it/s]
 99%|█████████▊| 6744/6837 [40:34<00:21,  4.24it/s]
 99%|█████████▊| 6745/6837 [40:35<00:21,  4.23it/s]
 99%|█████████▊| 6746/6837 [40:35<00:21,  4.24it/s]
 99%|█████████▊| 6747/6837 [40:35<00:21,  4.24it/s]
 99%|█████████▊| 6748/6837 [40:35<00:20,  4.24it/s]
 99%|█████████▊| 6749/6837 [40:36<00:20,  4.25it/s]
 99%|█████████▊| 6750/6837 [40:36<00:20,  4.24it/s]
                                                   
{'loss': 4.2669, 'grad_norm': 0.14176252484321594, 'learning_rate': 2.9592677888740095e-06, 'epoch': 0.2}
+
 99%|█████████▊| 6750/6837 [40:36<00:20,  4.24it/s]
 99%|█████████▊| 6751/6837 [40:36<00:20,  4.22it/s]
 99%|█████████▉| 6752/6837 [40:36<00:20,  4.22it/s]
 99%|█████████▉| 6753/6837 [40:36<00:19,  4.23it/s]
 99%|█████████▉| 6754/6837 [40:37<00:19,  4.23it/s]
 99%|█████████▉| 6755/6837 [40:37<00:19,  4.23it/s]
 99%|█████████▉| 6756/6837 [40:37<00:19,  4.24it/s]
 99%|█████████▉| 6757/6837 [40:37<00:18,  4.24it/s]
 99%|█████████▉| 6758/6837 [40:38<00:18,  4.24it/s]
 99%|█████████▉| 6759/6837 [40:38<00:18,  4.24it/s]
 99%|█████████▉| 6760/6837 [40:38<00:18,  4.24it/s]
 99%|█████████▉| 6761/6837 [40:38<00:17,  4.24it/s]
 99%|█████████▉| 6762/6837 [40:39<00:18,  4.05it/s]
 99%|█████████▉| 6763/6837 [40:39<00:18,  3.96it/s]
 99%|█████████▉| 6764/6837 [40:39<00:18,  3.89it/s]
 99%|█████████▉| 6765/6837 [40:39<00:18,  3.82it/s]
 99%|█████████▉| 6766/6837 [40:40<00:18,  3.80it/s]
 99%|█████████▉| 6767/6837 [40:40<00:18,  3.80it/s]
 99%|█████████▉| 6768/6837 [40:40<00:18,  3.80it/s]
 99%|█████████▉| 6769/6837 [40:40<00:17,  3.92it/s]
 99%|█████████▉| 6770/6837 [40:41<00:16,  4.01it/s]
 99%|█████████▉| 6771/6837 [40:41<00:16,  4.08it/s]
 99%|█████████▉| 6772/6837 [40:41<00:15,  4.12it/s]
 99%|█████████▉| 6773/6837 [40:41<00:15,  4.16it/s]
 99%|█████████▉| 6774/6837 [40:42<00:15,  4.18it/s]
 99%|█████████▉| 6775/6837 [40:42<00:14,  4.20it/s]
                                                   
{'loss': 4.2874, 'grad_norm': 0.1411001980304718, 'learning_rate': 1.5030183606649096e-06, 'epoch': 0.2}
+
 99%|█████████▉| 6775/6837 [40:42<00:14,  4.20it/s]
 99%|█████████▉| 6776/6837 [40:42<00:14,  4.20it/s]
 99%|█████████▉| 6777/6837 [40:42<00:14,  4.21it/s]
 99%|█████████▉| 6778/6837 [40:43<00:14,  4.21it/s]
 99%|█████████▉| 6779/6837 [40:43<00:14,  4.05it/s]
 99%|█████████▉| 6780/6837 [40:43<00:14,  3.96it/s]
 99%|█████████▉| 6781/6837 [40:43<00:14,  3.89it/s]
 99%|█████████▉| 6782/6837 [40:44<00:14,  3.87it/s]
 99%|█████████▉| 6783/6837 [40:44<00:14,  3.83it/s]
 99%|█████████▉| 6784/6837 [40:44<00:13,  3.81it/s]
 99%|█████████▉| 6785/6837 [40:44<00:13,  3.80it/s]
 99%|█████████▉| 6786/6837 [40:45<00:13,  3.79it/s]
 99%|█████████▉| 6787/6837 [40:45<00:13,  3.81it/s]
 99%|█████████▉| 6788/6837 [40:45<00:12,  3.80it/s]
 99%|█████████▉| 6789/6837 [40:45<00:12,  3.80it/s]
 99%|█████████▉| 6790/6837 [40:46<00:12,  3.79it/s]
 99%|█████████▉| 6791/6837 [40:46<00:12,  3.82it/s]
 99%|█████████▉| 6792/6837 [40:46<00:11,  3.81it/s]
 99%|█████████▉| 6793/6837 [40:47<00:11,  3.79it/s]
 99%|█████████▉| 6794/6837 [40:47<00:11,  3.80it/s]
 99%|█████████▉| 6795/6837 [40:47<00:11,  3.80it/s]
 99%|█████████▉| 6796/6837 [40:47<00:10,  3.79it/s]
 99%|█████████▉| 6797/6837 [40:48<00:10,  3.80it/s]
 99%|█████████▉| 6798/6837 [40:48<00:10,  3.80it/s]
 99%|█████████▉| 6799/6837 [40:48<00:10,  3.79it/s]
 99%|█████████▉| 6800/6837 [40:48<00:09,  3.78it/s]
                                                   
{'loss': 4.2792, 'grad_norm': 0.14825549721717834, 'learning_rate': 5.353128963893062e-07, 'epoch': 0.2}
+
 99%|█████████▉| 6800/6837 [40:48<00:09,  3.78it/s]
 99%|█████████▉| 6801/6837 [40:49<00:09,  3.88it/s]
 99%|█████████▉| 6802/6837 [40:49<00:08,  3.98it/s]
100%|█████████▉| 6803/6837 [40:49<00:08,  4.05it/s]
100%|█████████▉| 6804/6837 [40:49<00:08,  4.10it/s]
100%|█████████▉| 6805/6837 [40:50<00:07,  4.14it/s]
100%|█████████▉| 6806/6837 [40:50<00:07,  4.16it/s]
100%|█████████▉| 6807/6837 [40:50<00:07,  4.18it/s]
100%|█████████▉| 6808/6837 [40:50<00:06,  4.19it/s]
100%|█████████▉| 6809/6837 [40:51<00:06,  4.21it/s]
100%|█████████▉| 6810/6837 [40:51<00:06,  4.22it/s]
100%|█████████▉| 6811/6837 [40:51<00:06,  4.22it/s]
100%|█████████▉| 6812/6837 [40:51<00:05,  4.22it/s]
100%|█████████▉| 6813/6837 [40:51<00:05,  4.23it/s]
100%|█████████▉| 6814/6837 [40:52<00:05,  4.23it/s]
100%|█████████▉| 6815/6837 [40:52<00:05,  4.23it/s]
100%|█████████▉| 6816/6837 [40:52<00:04,  4.23it/s]
100%|█████████▉| 6817/6837 [40:52<00:04,  4.24it/s]
100%|█████████▉| 6818/6837 [40:53<00:04,  4.23it/s]
100%|█████████▉| 6819/6837 [40:53<00:04,  4.24it/s]
100%|█████████▉| 6820/6837 [40:53<00:04,  4.24it/s]
100%|█████████▉| 6821/6837 [40:53<00:03,  4.23it/s]
100%|█████████▉| 6822/6837 [40:54<00:03,  4.23it/s]
100%|█████████▉| 6823/6837 [40:54<00:03,  4.23it/s]
100%|█████████▉| 6824/6837 [40:54<00:03,  4.24it/s]
100%|█████████▉| 6825/6837 [40:54<00:02,  4.23it/s]
                                                   
{'loss': 4.289, 'grad_norm': 0.14030522108078003, 'learning_rate': 5.6309063927617586e-08, 'epoch': 0.2}
+
100%|█████████▉| 6825/6837 [40:54<00:02,  4.23it/s]
100%|█████████▉| 6826/6837 [40:55<00:02,  4.22it/s]
100%|█████████▉| 6827/6837 [40:55<00:02,  4.23it/s]
100%|█████████▉| 6828/6837 [40:55<00:02,  4.23it/s]
100%|█████████▉| 6829/6837 [40:55<00:01,  4.23it/s]
100%|█████████▉| 6830/6837 [40:55<00:01,  4.24it/s]
100%|█████████▉| 6831/6837 [40:56<00:01,  4.24it/s]
100%|█████████▉| 6832/6837 [40:56<00:01,  4.24it/s]
100%|█████████▉| 6833/6837 [40:56<00:00,  4.24it/s]
100%|█████████▉| 6834/6837 [40:56<00:00,  4.24it/s]
100%|█████████▉| 6835/6837 [40:57<00:00,  4.23it/s]
100%|█████████▉| 6836/6837 [40:57<00:00,  4.24it/s]
100%|██████████| 6837/6837 [40:57<00:00,  4.24it/s]
                                                   {'train_runtime': 2489.6388, 'train_samples_per_second': 878.703, 'train_steps_per_second': 2.746, 'train_loss': 4.665441373295301, 'epoch': 0.2}
+
100%|██████████| 6837/6837 [41:29<00:00,  4.24it/s]
100%|██████████| 6837/6837 [41:29<00:00,  2.75it/s]