second commit

Browse files

Files changed (7) hide show

all_results.json +7 -7
generated_predictions.jsonl +0 -0
llamaboard_config.yaml +12 -57
predict_results.json +9 -0
running_log.txt +70 -528
trainer_log.jsonl +15 -191
training_args.yaml +9 -21

all_results.json CHANGED Viewed

@@ -1,9 +1,9 @@
 {
-    "epoch": 4.887459807073955,
-    "num_input_tokens_seen": 1299392,
-    "total_flos": 5.151317702790349e+16,
-    "train_loss": 0.3433768034317166,
-    "train_runtime": 2162.0959,
-    "train_samples_per_second": 11.489,
-    "train_steps_per_second": 0.088
 }

 {
+    "predict_bleu-4": 87.56551722756411,
+    "predict_rouge-1": 95.1923076923077,
+    "predict_rouge-2": 0.0,
+    "predict_rouge-l": 95.1923076923077,
+    "predict_runtime": 11.1547,
+    "predict_samples_per_second": 111.433,
+    "predict_steps_per_second": 6.993
 }

generated_predictions.jsonl ADDED Viewed

The diff for this file is too large to render. See raw diff

llamaboard_config.yaml CHANGED Viewed

@@ -1,5 +1,16 @@
 top.booster: auto
-top.checkpoint_path: null
 top.finetuning_type: full
 top.model_name: LLaMA2-7B-Chat
 top.quantization_bit: none
@@ -7,59 +18,3 @@ top.quantization_method: bitsandbytes
 top.rope_scaling: none
 top.template: llama2
 top.visual_inputs: false
-train.additional_target: ''
-train.badam_mode: layer
-train.badam_switch_interval: 50
-train.badam_switch_mode: ascending
-train.badam_update_ratio: 0.05
-train.batch_size: 2
-train.compute_type: bf16
-train.create_new_adapter: false
-train.cutoff_len: 1024
-train.dataset:
-- truth_train_0716
-train.dataset_dir: data
-train.ds_offload: false
-train.ds_stage: '2'
-train.freeze_extra_modules: ''
-train.freeze_trainable_layers: 2
-train.freeze_trainable_modules: all
-train.galore_rank: 16
-train.galore_scale: 0.25
-train.galore_target: all
-train.galore_update_interval: 200
-train.gradient_accumulation_steps: 8
-train.learning_rate: 5e-6
-train.logging_steps: 1
-train.lora_alpha: 16
-train.lora_dropout: 0
-train.lora_rank: 8
-train.lora_target: ''
-train.loraplus_lr_ratio: 0
-train.lr_scheduler_type: cosine
-train.max_grad_norm: '1.0'
-train.max_samples: '100000'
-train.neat_packing: false
-train.neftune_alpha: 0
-train.num_train_epochs: '5.0'
-train.optim: adamw_torch
-train.packing: false
-train.ppo_score_norm: false
-train.ppo_whiten_rewards: false
-train.pref_beta: 0.1
-train.pref_ftx: 0
-train.pref_loss: sigmoid
-train.report_to: false
-train.resize_vocab: false
-train.reward_model: null
-train.save_steps: 1000
-train.shift_attn: false
-train.training_stage: Supervised Fine-Tuning
-train.use_badam: false
-train.use_dora: false
-train.use_galore: false
-train.use_llama_pro: false
-train.use_pissa: false
-train.use_rslora: false
-train.val_size: 0
-train.warmup_steps: 10

+eval.batch_size: 2
+eval.cutoff_len: 1024
+eval.dataset:
+- truth_dev_0716
+eval.dataset_dir: data
+eval.max_new_tokens: 512
+eval.max_samples: '100000'
+eval.output_dir: eval_2024-07-16-09-05-28
+eval.predict: true
+eval.temperature: 0.95
+eval.top_p: 0.7
 top.booster: auto
+top.checkpoint_path: train_2024-07-16-09-05-28_llama2
 top.finetuning_type: full
 top.model_name: LLaMA2-7B-Chat
 top.quantization_bit: none
 top.rope_scaling: none
 top.template: llama2
 top.visual_inputs: false

predict_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "predict_bleu-4": 87.56551722756411,
+    "predict_rouge-1": 95.1923076923077,
+    "predict_rouge-2": 0.0,
+    "predict_rouge-l": 95.1923076923077,
+    "predict_runtime": 11.1547,
+    "predict_samples_per_second": 111.433,
+    "predict_steps_per_second": 6.993
+}

running_log.txt CHANGED Viewed

@@ -1,61 +1,45 @@
-07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
-07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
-07/16/2024 09:07:34 - INFO - llamafactory.hparams.parser - Process rank: 3, device: cuda:3, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
-[INFO|parser.py:325] 2024-07-16 09:07:34,077 >> Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
-07/16/2024 09:07:34 - INFO - llamafactory.hparams.parser - Process rank: 7, device: cuda:7, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
-07/16/2024 09:07:34 - INFO - llamafactory.hparams.parser - Process rank: 5, device: cuda:5, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
-07/16/2024 09:07:34 - INFO - llamafactory.hparams.parser - Process rank: 2, device: cuda:2, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
-[INFO|tokenization_utils_base.py:2161] 2024-07-16 09:07:34,347 >> loading file tokenizer.model from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/tokenizer.model
-[INFO|tokenization_utils_base.py:2161] 2024-07-16 09:07:34,347 >> loading file tokenizer.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/tokenizer.json
-[INFO|tokenization_utils_base.py:2161] 2024-07-16 09:07:34,348 >> loading file added_tokens.json from cache at None
-[INFO|tokenization_utils_base.py:2161] 2024-07-16 09:07:34,348 >> loading file special_tokens_map.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/special_tokens_map.json
-[INFO|tokenization_utils_base.py:2161] 2024-07-16 09:07:34,348 >> loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/tokenizer_config.json
-07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
-07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
-07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
-07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
-[INFO|template.py:372] 2024-07-16 09:07:34,452 >> Add pad token: </s>
-[INFO|loader.py:50] 2024-07-16 09:07:34,453 >> Loading dataset 0716_truthfulqa_benchmark_train.json...
-07/16/2024 09:07:34 - INFO - llamafactory.hparams.parser - Process rank: 6, device: cuda:6, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
-07/16/2024 09:07:34 - INFO - llamafactory.data.template - Add pad token: </s>
-07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
-07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
-07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
-07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
-07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
-07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
-07/16/2024 09:07:36 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train.json...
-[INFO|configuration_utils.py:733] 2024-07-16 09:07:37,470 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/config.json
-[INFO|configuration_utils.py:800] 2024-07-16 09:07:37,473 >> Model config LlamaConfig {
-  "_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
   "architectures": [
     "LlamaForCausalLM"
   ],
@@ -78,48 +62,48 @@
   "rope_scaling": null,
   "rope_theta": 10000.0,
   "tie_word_embeddings": false,
-  "torch_dtype": "float16",
   "transformers_version": "4.42.3",
-  "use_cache": true,
   "vocab_size": 32000
 }
-[INFO|modeling_utils.py:3556] 2024-07-16 09:07:37,523 >> loading weights file model.safetensors from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/model.safetensors.index.json
-[INFO|modeling_utils.py:1531] 2024-07-16 09:07:37,524 >> Instantiating LlamaForCausalLM model under default dtype torch.bfloat16.
-[INFO|configuration_utils.py:1000] 2024-07-16 09:07:37,526 >> Generate config GenerationConfig {
   "bos_token_id": 1,
   "eos_token_id": 2
 }
-[INFO|modeling_utils.py:4364] 2024-07-16 09:07:54,870 >> All model checkpoint weights were used when initializing LlamaForCausalLM.
-[INFO|modeling_utils.py:4372] 2024-07-16 09:07:54,870 >> All the weights of LlamaForCausalLM were initialized from the model checkpoint at meta-llama/Llama-2-7b-chat-hf.
-If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
-[INFO|configuration_utils.py:955] 2024-07-16 09:07:55,055 >> loading configuration file generation_config.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/generation_config.json
-[INFO|configuration_utils.py:1000] 2024-07-16 09:07:55,055 >> Generate config GenerationConfig {
   "bos_token_id": 1,
   "do_sample": true,
   "eos_token_id": 2,
@@ -130,502 +114,60 @@ If your task is similar to the task the model of the checkpoint was trained on,
 }
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
-[INFO|checkpointing.py:103] 2024-07-16 09:07:55,062 >> Gradient checkpointing enabled.
-[INFO|attention.py:80] 2024-07-16 09:07:55,062 >> Using torch SDPA for faster training and inference.
-[INFO|adapter.py:302] 2024-07-16 09:07:55,062 >> Upcasting trainable params to float32.
-[INFO|adapter.py:48] 2024-07-16 09:07:55,062 >> Fine-tuning method: Full
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
-07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
-07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
-07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
-[INFO|loader.py:196] 2024-07-16 09:07:55,174 >> trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
-07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
-07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
-07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
-[INFO|trainer.py:642] 2024-07-16 09:07:55,179 >> Using auto half precision backend
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
-07/16/2024 09:07:55 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
-07/16/2024 09:07:55 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
-07/16/2024 09:07:55 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
-[INFO|trainer.py:2128] 2024-07-16 09:08:14,231 >> ***** Running training *****
-[INFO|trainer.py:2129] 2024-07-16 09:08:14,231 >>   Num examples = 4,968
-[INFO|trainer.py:2130] 2024-07-16 09:08:14,231 >>   Num Epochs = 5
-[INFO|trainer.py:2131] 2024-07-16 09:08:14,231 >>   Instantaneous batch size per device = 2
-[INFO|trainer.py:2134] 2024-07-16 09:08:14,231 >>   Total train batch size (w. parallel, distributed & accumulation) = 128
-[INFO|trainer.py:2135] 2024-07-16 09:08:14,231 >>   Gradient Accumulation steps = 8
-[INFO|trainer.py:2136] 2024-07-16 09:08:14,231 >>   Total optimization steps = 190
-[INFO|trainer.py:2137] 2024-07-16 09:08:14,233 >>   Number of trainable parameters = 6,738,415,616
-[INFO|callbacks.py:310] 2024-07-16 09:08:27,214 >> {'loss': 8.3599, 'learning_rate': 5.0000e-07, 'epoch': 0.03, 'throughput': 548.54}
-[INFO|callbacks.py:310] 2024-07-16 09:08:38,345 >> {'loss': 8.1891, 'learning_rate': 1.0000e-06, 'epoch': 0.05, 'throughput': 575.99}
-[INFO|callbacks.py:310] 2024-07-16 09:08:49,459 >> {'loss': 8.0792, 'learning_rate': 1.5000e-06, 'epoch': 0.08, 'throughput': 586.40}
-[INFO|callbacks.py:310] 2024-07-16 09:09:00,554 >> {'loss': 7.9682, 'learning_rate': 2.0000e-06, 'epoch': 0.10, 'throughput': 586.87}
-[INFO|callbacks.py:310] 2024-07-16 09:09:11,650 >> {'loss': 6.9482, 'learning_rate': 2.5000e-06, 'epoch': 0.13, 'throughput': 599.42}
-[INFO|callbacks.py:310] 2024-07-16 09:09:22,743 >> {'loss': 5.1505, 'learning_rate': 3.0000e-06, 'epoch': 0.15, 'throughput': 599.28}
-[INFO|callbacks.py:310] 2024-07-16 09:09:33,861 >> {'loss': 4.7491, 'learning_rate': 3.5000e-06, 'epoch': 0.18, 'throughput': 596.99}
-[INFO|callbacks.py:310] 2024-07-16 09:09:44,975 >> {'loss': 3.2164, 'learning_rate': 4.0000e-06, 'epoch': 0.21, 'throughput': 600.21}
-[INFO|callbacks.py:310] 2024-07-16 09:09:56,098 >> {'loss': 2.7761, 'learning_rate': 4.5000e-06, 'epoch': 0.23, 'throughput': 603.94}
-[INFO|callbacks.py:310] 2024-07-16 09:10:07,219 >> {'loss': 0.6703, 'learning_rate': 5.0000e-06, 'epoch': 0.26, 'throughput': 605.96}
-[INFO|callbacks.py:310] 2024-07-16 09:10:18,354 >> {'loss': 0.3255, 'learning_rate': 4.9996e-06, 'epoch': 0.28, 'throughput': 605.48}
-[INFO|callbacks.py:310] 2024-07-16 09:10:29,443 >> {'loss': 0.3301, 'learning_rate': 4.9985e-06, 'epoch': 0.31, 'throughput': 605.64}
-[INFO|callbacks.py:310] 2024-07-16 09:10:40,548 >> {'loss': 0.2121, 'learning_rate': 4.9966e-06, 'epoch': 0.33, 'throughput': 606.15}
-[INFO|callbacks.py:310] 2024-07-16 09:10:51,649 >> {'loss': 1.1565, 'learning_rate': 4.9939e-06, 'epoch': 0.36, 'throughput': 607.41}
-[INFO|callbacks.py:310] 2024-07-16 09:11:02,736 >> {'loss': 0.8054, 'learning_rate': 4.9905e-06, 'epoch': 0.39, 'throughput': 606.19}
-[INFO|callbacks.py:310] 2024-07-16 09:11:13,851 >> {'loss': 0.2386, 'learning_rate': 4.9863e-06, 'epoch': 0.41, 'throughput': 607.52}
-[INFO|callbacks.py:310] 2024-07-16 09:11:24,984 >> {'loss': 0.3161, 'learning_rate': 4.9814e-06, 'epoch': 0.44, 'throughput': 606.78}
-[INFO|callbacks.py:310] 2024-07-16 09:11:36,116 >> {'loss': 0.2773, 'learning_rate': 4.9757e-06, 'epoch': 0.46, 'throughput': 607.80}
-[INFO|callbacks.py:310] 2024-07-16 09:11:47,247 >> {'loss': 0.2062, 'learning_rate': 4.9692e-06, 'epoch': 0.49, 'throughput': 608.19}
-[INFO|callbacks.py:310] 2024-07-16 09:11:58,362 >> {'loss': 0.1837, 'learning_rate': 4.9620e-06, 'epoch': 0.51, 'throughput': 609.22}
-[INFO|callbacks.py:310] 2024-07-16 09:12:09,457 >> {'loss': 0.1735, 'learning_rate': 4.9541e-06, 'epoch': 0.54, 'throughput': 610.01}
-[INFO|callbacks.py:310] 2024-07-16 09:12:20,540 >> {'loss': 0.1588, 'learning_rate': 4.9454e-06, 'epoch': 0.57, 'throughput': 609.91}
-[INFO|callbacks.py:310] 2024-07-16 09:12:31,647 >> {'loss': 0.1443, 'learning_rate': 4.9359e-06, 'epoch': 0.59, 'throughput': 610.82}
-[INFO|callbacks.py:310] 2024-07-16 09:12:42,733 >> {'loss': 0.1570, 'learning_rate': 4.9257e-06, 'epoch': 0.62, 'throughput': 609.97}
-[INFO|callbacks.py:310] 2024-07-16 09:12:53,861 >> {'loss': 0.1199, 'learning_rate': 4.9148e-06, 'epoch': 0.64, 'throughput': 609.21}
-[INFO|callbacks.py:310] 2024-07-16 09:13:04,974 >> {'loss': 0.1539, 'learning_rate': 4.9032e-06, 'epoch': 0.67, 'throughput': 609.20}
-[INFO|callbacks.py:310] 2024-07-16 09:13:16,096 >> {'loss': 0.1208, 'learning_rate': 4.8908e-06, 'epoch': 0.69, 'throughput': 609.87}
-[INFO|callbacks.py:310] 2024-07-16 09:13:27,217 >> {'loss': 0.0954, 'learning_rate': 4.8776e-06, 'epoch': 0.72, 'throughput': 610.39}
-[INFO|callbacks.py:310] 2024-07-16 09:13:38,328 >> {'loss': 0.1387, 'learning_rate': 4.8638e-06, 'epoch': 0.75, 'throughput': 611.13}
-[INFO|callbacks.py:310] 2024-07-16 09:13:49,415 >> {'loss': 0.1484, 'learning_rate': 4.8492e-06, 'epoch': 0.77, 'throughput': 612.02}
-[INFO|callbacks.py:310] 2024-07-16 09:14:00,513 >> {'loss': 0.0998, 'learning_rate': 4.8340e-06, 'epoch': 0.80, 'throughput': 612.22}
-[INFO|callbacks.py:310] 2024-07-16 09:14:11,593 >> {'loss': 0.1068, 'learning_rate': 4.8180e-06, 'epoch': 0.82, 'throughput': 612.05}
-[INFO|callbacks.py:310] 2024-07-16 09:14:22,685 >> {'loss': 0.0801, 'learning_rate': 4.8013e-06, 'epoch': 0.85, 'throughput': 612.99}
-[INFO|callbacks.py:310] 2024-07-16 09:14:33,813 >> {'loss': 0.1066, 'learning_rate': 4.7839e-06, 'epoch': 0.87, 'throughput': 612.89}
-[INFO|callbacks.py:310] 2024-07-16 09:14:44,935 >> {'loss': 0.1038, 'learning_rate': 4.7658e-06, 'epoch': 0.90, 'throughput': 613.01}
-[INFO|callbacks.py:310] 2024-07-16 09:14:56,047 >> {'loss': 0.1060, 'learning_rate': 4.7470e-06, 'epoch': 0.93, 'throughput': 612.94}
-[INFO|callbacks.py:310] 2024-07-16 09:15:07,172 >> {'loss': 0.1107, 'learning_rate': 4.7275e-06, 'epoch': 0.95, 'throughput': 613.01}
-[INFO|callbacks.py:310] 2024-07-16 09:15:18,265 >> {'loss': 0.1372, 'learning_rate': 4.7074e-06, 'epoch': 0.98, 'throughput': 613.54}
-[INFO|callbacks.py:310] 2024-07-16 09:15:29,366 >> {'loss': 0.0816, 'learning_rate': 4.6865e-06, 'epoch': 1.00, 'throughput': 613.88}
-[INFO|callbacks.py:310] 2024-07-16 09:15:40,449 >> {'loss': 0.0743, 'learning_rate': 4.6651e-06, 'epoch': 1.03, 'throughput': 614.30}
-[INFO|callbacks.py:310] 2024-07-16 09:15:51,540 >> {'loss': 0.0720, 'learning_rate': 4.6429e-06, 'epoch': 1.05, 'throughput': 614.77}
-[INFO|callbacks.py:310] 2024-07-16 09:16:02,629 >> {'loss': 0.0596, 'learning_rate': 4.6201e-06, 'epoch': 1.08, 'throughput': 614.97}
-[INFO|callbacks.py:310] 2024-07-16 09:16:13,746 >> {'loss': 0.0544, 'learning_rate': 4.5967e-06, 'epoch': 1.11, 'throughput': 615.46}
-[INFO|callbacks.py:310] 2024-07-16 09:16:24,855 >> {'loss': 0.0342, 'learning_rate': 4.5726e-06, 'epoch': 1.13, 'throughput': 615.55}
-[INFO|callbacks.py:310] 2024-07-16 09:16:35,985 >> {'loss': 0.0394, 'learning_rate': 4.5479e-06, 'epoch': 1.16, 'throughput': 615.19}
-[INFO|callbacks.py:310] 2024-07-16 09:16:47,103 >> {'loss': 0.0196, 'learning_rate': 4.5225e-06, 'epoch': 1.18, 'throughput': 615.36}
-[INFO|callbacks.py:310] 2024-07-16 09:16:58,199 >> {'loss': 0.0411, 'learning_rate': 4.4966e-06, 'epoch': 1.21, 'throughput': 615.43}
-[INFO|callbacks.py:310] 2024-07-16 09:17:09,282 >> {'loss': 0.0257, 'learning_rate': 4.4700e-06, 'epoch': 1.23, 'throughput': 614.94}
-[INFO|callbacks.py:310] 2024-07-16 09:17:20,373 >> {'loss': 0.0289, 'learning_rate': 4.4429e-06, 'epoch': 1.26, 'throughput': 615.29}
-[INFO|callbacks.py:310] 2024-07-16 09:17:31,470 >> {'loss': 0.1193, 'learning_rate': 4.4151e-06, 'epoch': 1.29, 'throughput': 615.01}
-[INFO|callbacks.py:310] 2024-07-16 09:17:42,559 >> {'loss': 0.0883, 'learning_rate': 4.3868e-06, 'epoch': 1.31, 'throughput': 614.92}
-[INFO|callbacks.py:310] 2024-07-16 09:17:53,670 >> {'loss': 0.0377, 'learning_rate': 4.3579e-06, 'epoch': 1.34, 'throughput': 614.86}
-[INFO|callbacks.py:310] 2024-07-16 09:18:04,800 >> {'loss': 0.0602, 'learning_rate': 4.3284e-06, 'epoch': 1.36, 'throughput': 614.73}
-[INFO|callbacks.py:310] 2024-07-16 09:18:15,923 >> {'loss': 0.0830, 'learning_rate': 4.2983e-06, 'epoch': 1.39, 'throughput': 614.38}
-[INFO|callbacks.py:310] 2024-07-16 09:18:27,039 >> {'loss': 0.0358, 'learning_rate': 4.2678e-06, 'epoch': 1.41, 'throughput': 614.72}
-[INFO|callbacks.py:310] 2024-07-16 09:18:38,136 >> {'loss': 0.0321, 'learning_rate': 4.2366e-06, 'epoch': 1.44, 'throughput': 614.84}
-[INFO|callbacks.py:310] 2024-07-16 09:18:49,231 >> {'loss': 0.0452, 'learning_rate': 4.2050e-06, 'epoch': 1.47, 'throughput': 615.11}
-[INFO|callbacks.py:310] 2024-07-16 09:19:00,331 >> {'loss': 0.0915, 'learning_rate': 4.1728e-06, 'epoch': 1.49, 'throughput': 615.02}
-[INFO|callbacks.py:310] 2024-07-16 09:19:11,424 >> {'loss': 0.0651, 'learning_rate': 4.1401e-06, 'epoch': 1.52, 'throughput': 614.81}
-[INFO|callbacks.py:310] 2024-07-16 09:19:22,545 >> {'loss': 0.0868, 'learning_rate': 4.1070e-06, 'epoch': 1.54, 'throughput': 614.92}
-[INFO|callbacks.py:310] 2024-07-16 09:19:33,666 >> {'loss': 0.0554, 'learning_rate': 4.0733e-06, 'epoch': 1.57, 'throughput': 615.06}
-[INFO|callbacks.py:310] 2024-07-16 09:19:44,774 >> {'loss': 0.0336, 'learning_rate': 4.0392e-06, 'epoch': 1.59, 'throughput': 615.29}
-[INFO|callbacks.py:310] 2024-07-16 09:19:55,885 >> {'loss': 0.0455, 'learning_rate': 4.0045e-06, 'epoch': 1.62, 'throughput': 615.69}
-[INFO|callbacks.py:310] 2024-07-16 09:20:07,002 >> {'loss': 0.0406, 'learning_rate': 3.9695e-06, 'epoch': 1.65, 'throughput': 615.45}
-[INFO|callbacks.py:310] 2024-07-16 09:20:18,095 >> {'loss': 0.0461, 'learning_rate': 3.9339e-06, 'epoch': 1.67, 'throughput': 615.37}
-[INFO|callbacks.py:310] 2024-07-16 09:20:29,180 >> {'loss': 0.0466, 'learning_rate': 3.8980e-06, 'epoch': 1.70, 'throughput': 615.10}
-[INFO|callbacks.py:310] 2024-07-16 09:20:40,282 >> {'loss': 0.0382, 'learning_rate': 3.8616e-06, 'epoch': 1.72, 'throughput': 615.23}
-[INFO|callbacks.py:310] 2024-07-16 09:20:51,381 >> {'loss': 0.0426, 'learning_rate': 3.8248e-06, 'epoch': 1.75, 'throughput': 614.90}
-[INFO|callbacks.py:310] 2024-07-16 09:21:02,489 >> {'loss': 0.0264, 'learning_rate': 3.7876e-06, 'epoch': 1.77, 'throughput': 615.03}
-[INFO|callbacks.py:310] 2024-07-16 09:21:13,594 >> {'loss': 0.0567, 'learning_rate': 3.7500e-06, 'epoch': 1.80, 'throughput': 615.11}
-[INFO|callbacks.py:310] 2024-07-16 09:21:24,706 >> {'loss': 0.0688, 'learning_rate': 3.7120e-06, 'epoch': 1.83, 'throughput': 615.35}
-[INFO|callbacks.py:310] 2024-07-16 09:21:35,842 >> {'loss': 0.0351, 'learning_rate': 3.6737e-06, 'epoch': 1.85, 'throughput': 614.88}
-[INFO|callbacks.py:310] 2024-07-16 09:21:46,947 >> {'loss': 0.0246, 'learning_rate': 3.6350e-06, 'epoch': 1.88, 'throughput': 614.93}
-[INFO|callbacks.py:310] 2024-07-16 09:21:58,021 >> {'loss': 0.0364, 'learning_rate': 3.5959e-06, 'epoch': 1.90, 'throughput': 615.23}
-[INFO|callbacks.py:310] 2024-07-16 09:22:09,127 >> {'loss': 0.0352, 'learning_rate': 3.5565e-06, 'epoch': 1.93, 'throughput': 615.23}
-[INFO|callbacks.py:310] 2024-07-16 09:22:20,219 >> {'loss': 0.0915, 'learning_rate': 3.5168e-06, 'epoch': 1.95, 'throughput': 615.24}
-[INFO|callbacks.py:310] 2024-07-16 09:22:31,310 >> {'loss': 0.0327, 'learning_rate': 3.4768e-06, 'epoch': 1.98, 'throughput': 614.95}
-[INFO|callbacks.py:310] 2024-07-16 09:22:42,417 >> {'loss': 0.0448, 'learning_rate': 3.4365e-06, 'epoch': 2.01, 'throughput': 615.21}
-[INFO|callbacks.py:310] 2024-07-16 09:22:53,536 >> {'loss': 0.0186, 'learning_rate': 3.3959e-06, 'epoch': 2.03, 'throughput': 615.29}
-[INFO|callbacks.py:310] 2024-07-16 09:23:04,675 >> {'loss': 0.0342, 'learning_rate': 3.3551e-06, 'epoch': 2.06, 'throughput': 615.30}
-[INFO|callbacks.py:310] 2024-07-16 09:23:15,801 >> {'loss': 0.0079, 'learning_rate': 3.3139e-06, 'epoch': 2.08, 'throughput': 615.14}
-[INFO|callbacks.py:310] 2024-07-16 09:23:26,896 >> {'loss': 0.0177, 'learning_rate': 3.2725e-06, 'epoch': 2.11, 'throughput': 615.01}
-[INFO|callbacks.py:310] 2024-07-16 09:23:37,991 >> {'loss': 0.0139, 'learning_rate': 3.2309e-06, 'epoch': 2.14, 'throughput': 614.74}
-[INFO|callbacks.py:310] 2024-07-16 09:23:49,080 >> {'loss': 0.0103, 'learning_rate': 3.1891e-06, 'epoch': 2.16, 'throughput': 615.15}
-[INFO|callbacks.py:310] 2024-07-16 09:24:00,169 >> {'loss': 0.0221, 'learning_rate': 3.1470e-06, 'epoch': 2.19, 'throughput': 615.35}
-[INFO|callbacks.py:310] 2024-07-16 09:24:11,255 >> {'loss': 0.0021, 'learning_rate': 3.1048e-06, 'epoch': 2.21, 'throughput': 615.26}
-[INFO|callbacks.py:310] 2024-07-16 09:24:22,375 >> {'loss': 0.0110, 'learning_rate': 3.0624e-06, 'epoch': 2.24, 'throughput': 615.65}
-[INFO|callbacks.py:310] 2024-07-16 09:24:33,470 >> {'loss': 0.0081, 'learning_rate': 3.0198e-06, 'epoch': 2.26, 'throughput': 615.45}
-[INFO|callbacks.py:310] 2024-07-16 09:24:44,602 >> {'loss': 0.0149, 'learning_rate': 2.9770e-06, 'epoch': 2.29, 'throughput': 615.35}
-[INFO|callbacks.py:310] 2024-07-16 09:24:55,725 >> {'loss': 0.0010, 'learning_rate': 2.9341e-06, 'epoch': 2.32, 'throughput': 615.53}
-[INFO|callbacks.py:310] 2024-07-16 09:25:06,826 >> {'loss': 0.0070, 'learning_rate': 2.8911e-06, 'epoch': 2.34, 'throughput': 615.54}
-[INFO|callbacks.py:310] 2024-07-16 09:25:17,934 >> {'loss': 0.0089, 'learning_rate': 2.8479e-06, 'epoch': 2.37, 'throughput': 615.48}
-[INFO|callbacks.py:310] 2024-07-16 09:25:29,026 >> {'loss': 0.0013, 'learning_rate': 2.8047e-06, 'epoch': 2.39, 'throughput': 615.67}
-[INFO|callbacks.py:310] 2024-07-16 09:25:40,116 >> {'loss': 0.0267, 'learning_rate': 2.7613e-06, 'epoch': 2.42, 'throughput': 615.82}
-[INFO|callbacks.py:310] 2024-07-16 09:25:51,214 >> {'loss': 0.0171, 'learning_rate': 2.7179e-06, 'epoch': 2.44, 'throughput': 615.76}
-[INFO|callbacks.py:310] 2024-07-16 09:26:02,342 >> {'loss': 0.0375, 'learning_rate': 2.6744e-06, 'epoch': 2.47, 'throughput': 615.50}
-[INFO|callbacks.py:310] 2024-07-16 09:26:13,469 >> {'loss': 0.0101, 'learning_rate': 2.6308e-06, 'epoch': 2.50, 'throughput': 615.37}
-[INFO|callbacks.py:310] 2024-07-16 09:26:24,600 >> {'loss': 0.0282, 'learning_rate': 2.5872e-06, 'epoch': 2.52, 'throughput': 615.50}
-[INFO|callbacks.py:310] 2024-07-16 09:26:35,708 >> {'loss': 0.0069, 'learning_rate': 2.5436e-06, 'epoch': 2.55, 'throughput': 615.47}
-[INFO|callbacks.py:310] 2024-07-16 09:26:46,803 >> {'loss': 0.0135, 'learning_rate': 2.5000e-06, 'epoch': 2.57, 'throughput': 615.66}
-[INFO|callbacks.py:310] 2024-07-16 09:26:57,903 >> {'loss': 0.0062, 'learning_rate': 2.4564e-06, 'epoch': 2.60, 'throughput': 615.71}
-[INFO|callbacks.py:310] 2024-07-16 09:27:08,991 >> {'loss': 0.0050, 'learning_rate': 2.4128e-06, 'epoch': 2.62, 'throughput': 615.56}
-[INFO|callbacks.py:310] 2024-07-16 09:27:20,085 >> {'loss': 0.0285, 'learning_rate': 2.3692e-06, 'epoch': 2.65, 'throughput': 615.65}
-[INFO|callbacks.py:310] 2024-07-16 09:27:31,191 >> {'loss': 0.0225, 'learning_rate': 2.3256e-06, 'epoch': 2.68, 'throughput': 615.86}
-[INFO|callbacks.py:310] 2024-07-16 09:27:42,299 >> {'loss': 0.0280, 'learning_rate': 2.2821e-06, 'epoch': 2.70, 'throughput': 615.69}
-[INFO|callbacks.py:310] 2024-07-16 09:27:53,416 >> {'loss': 0.0176, 'learning_rate': 2.2387e-06, 'epoch': 2.73, 'throughput': 615.60}
-[INFO|callbacks.py:310] 2024-07-16 09:28:04,554 >> {'loss': 0.0047, 'learning_rate': 2.1953e-06, 'epoch': 2.75, 'throughput': 615.36}
-[INFO|callbacks.py:310] 2024-07-16 09:28:15,674 >> {'loss': 0.0135, 'learning_rate': 2.1521e-06, 'epoch': 2.78, 'throughput': 615.25}
-[INFO|callbacks.py:310] 2024-07-16 09:28:26,766 >> {'loss': 0.0044, 'learning_rate': 2.1089e-06, 'epoch': 2.80, 'throughput': 615.51}
-[INFO|callbacks.py:310] 2024-07-16 09:28:37,852 >> {'loss': 0.0252, 'learning_rate': 2.0659e-06, 'epoch': 2.83, 'throughput': 615.50}
-[INFO|callbacks.py:310] 2024-07-16 09:28:48,945 >> {'loss': 0.0249, 'learning_rate': 2.0230e-06, 'epoch': 2.86, 'throughput': 615.61}
-[INFO|callbacks.py:310] 2024-07-16 09:29:00,043 >> {'loss': 0.0146, 'learning_rate': 1.9802e-06, 'epoch': 2.88, 'throughput': 615.75}
-[INFO|callbacks.py:310] 2024-07-16 09:29:11,162 >> {'loss': 0.0044, 'learning_rate': 1.9376e-06, 'epoch': 2.91, 'throughput': 615.69}
-[INFO|callbacks.py:310] 2024-07-16 09:29:22,253 >> {'loss': 0.0054, 'learning_rate': 1.8952e-06, 'epoch': 2.93, 'throughput': 615.71}
-[INFO|callbacks.py:310] 2024-07-16 09:29:33,390 >> {'loss': 0.0106, 'learning_rate': 1.8530e-06, 'epoch': 2.96, 'throughput': 615.58}
-[INFO|callbacks.py:310] 2024-07-16 09:29:44,539 >> {'loss': 0.0167, 'learning_rate': 1.8109e-06, 'epoch': 2.98, 'throughput': 615.53}
-[INFO|callbacks.py:310] 2024-07-16 09:29:55,648 >> {'loss': 0.0090, 'learning_rate': 1.7691e-06, 'epoch': 3.01, 'throughput': 615.55}
-[INFO|callbacks.py:310] 2024-07-16 09:30:06,727 >> {'loss': 0.0024, 'learning_rate': 1.7275e-06, 'epoch': 3.04, 'throughput': 615.66}
-[INFO|callbacks.py:310] 2024-07-16 09:30:17,830 >> {'loss': 0.0235, 'learning_rate': 1.6861e-06, 'epoch': 3.06, 'throughput': 615.60}
-[INFO|callbacks.py:310] 2024-07-16 09:30:28,918 >> {'loss': 0.0179, 'learning_rate': 1.6449e-06, 'epoch': 3.09, 'throughput': 615.53}
-[INFO|callbacks.py:310] 2024-07-16 09:30:40,012 >> {'loss': 0.0059, 'learning_rate': 1.6041e-06, 'epoch': 3.11, 'throughput': 615.35}
-[INFO|callbacks.py:310] 2024-07-16 09:30:51,146 >> {'loss': 0.0017, 'learning_rate': 1.5635e-06, 'epoch': 3.14, 'throughput': 615.08}
-[INFO|callbacks.py:310] 2024-07-16 09:31:02,257 >> {'loss': 0.0018, 'learning_rate': 1.5232e-06, 'epoch': 3.16, 'throughput': 615.02}
-[INFO|callbacks.py:310] 2024-07-16 09:31:13,381 >> {'loss': 0.0032, 'learning_rate': 1.4832e-06, 'epoch': 3.19, 'throughput': 615.23}
-[INFO|callbacks.py:310] 2024-07-16 09:31:24,512 >> {'loss': 0.0019, 'learning_rate': 1.4435e-06, 'epoch': 3.22, 'throughput': 615.29}
-[INFO|callbacks.py:310] 2024-07-16 09:31:35,616 >> {'loss': 0.0014, 'learning_rate': 1.4041e-06, 'epoch': 3.24, 'throughput': 615.27}
-[INFO|callbacks.py:310] 2024-07-16 09:31:46,705 >> {'loss': 0.0052, 'learning_rate': 1.3650e-06, 'epoch': 3.27, 'throughput': 615.45}
-[INFO|callbacks.py:310] 2024-07-16 09:31:57,796 >> {'loss': 0.0005, 'learning_rate': 1.3263e-06, 'epoch': 3.29, 'throughput': 615.55}
-[INFO|callbacks.py:310] 2024-07-16 09:32:08,900 >> {'loss': 0.0131, 'learning_rate': 1.2880e-06, 'epoch': 3.32, 'throughput': 615.52}
-[INFO|callbacks.py:310] 2024-07-16 09:32:19,991 >> {'loss': 0.0009, 'learning_rate': 1.2500e-06, 'epoch': 3.34, 'throughput': 615.54}
-[INFO|callbacks.py:310] 2024-07-16 09:32:31,110 >> {'loss': 0.0057, 'learning_rate': 1.2124e-06, 'epoch': 3.37, 'throughput': 615.63}
-[INFO|callbacks.py:310] 2024-07-16 09:32:42,232 >> {'loss': 0.0002, 'learning_rate': 1.1752e-06, 'epoch': 3.40, 'throughput': 615.53}
-[INFO|callbacks.py:310] 2024-07-16 09:32:53,354 >> {'loss': 0.0002, 'learning_rate': 1.1384e-06, 'epoch': 3.42, 'throughput': 615.37}
-[INFO|callbacks.py:310] 2024-07-16 09:33:04,458 >> {'loss': 0.0145, 'learning_rate': 1.1020e-06, 'epoch': 3.45, 'throughput': 615.56}
-[INFO|callbacks.py:310] 2024-07-16 09:33:15,548 >> {'loss': 0.0034, 'learning_rate': 1.0661e-06, 'epoch': 3.47, 'throughput': 615.59}
-[INFO|callbacks.py:310] 2024-07-16 09:33:26,645 >> {'loss': 0.0156, 'learning_rate': 1.0305e-06, 'epoch': 3.50, 'throughput': 615.43}
-[INFO|callbacks.py:310] 2024-07-16 09:33:37,738 >> {'loss': 0.0013, 'learning_rate': 9.9546e-07, 'epoch': 3.52, 'throughput': 615.59}
-[INFO|callbacks.py:310] 2024-07-16 09:33:48,828 >> {'loss': 0.0007, 'learning_rate': 9.6085e-07, 'epoch': 3.55, 'throughput': 615.56}
-[INFO|callbacks.py:310] 2024-07-16 09:33:59,926 >> {'loss': 0.0005, 'learning_rate': 9.2670e-07, 'epoch': 3.58, 'throughput': 615.58}
-[INFO|callbacks.py:310] 2024-07-16 09:34:11,043 >> {'loss': 0.0034, 'learning_rate': 8.9303e-07, 'epoch': 3.60, 'throughput': 615.52}
-[INFO|callbacks.py:310] 2024-07-16 09:34:22,149 >> {'loss': 0.0001, 'learning_rate': 8.5985e-07, 'epoch': 3.63, 'throughput': 615.41}
-[INFO|callbacks.py:310] 2024-07-16 09:34:33,264 >> {'loss': 0.0010, 'learning_rate': 8.2717e-07, 'epoch': 3.65, 'throughput': 615.49}
-[INFO|callbacks.py:310] 2024-07-16 09:34:44,385 >> {'loss': 0.0123, 'learning_rate': 7.9500e-07, 'epoch': 3.68, 'throughput': 615.44}
-[INFO|callbacks.py:310] 2024-07-16 09:34:55,486 >> {'loss': 0.0002, 'learning_rate': 7.6335e-07, 'epoch': 3.70, 'throughput': 615.35}
-[INFO|callbacks.py:310] 2024-07-16 09:35:06,571 >> {'loss': 0.0110, 'learning_rate': 7.3223e-07, 'epoch': 3.73, 'throughput': 615.39}
-[INFO|callbacks.py:310] 2024-07-16 09:35:17,657 >> {'loss': 0.0008, 'learning_rate': 7.0165e-07, 'epoch': 3.76, 'throughput': 615.17}
-[INFO|callbacks.py:310] 2024-07-16 09:35:28,737 >> {'loss': 0.0003, 'learning_rate': 6.7162e-07, 'epoch': 3.78, 'throughput': 615.50}
-[INFO|callbacks.py:310] 2024-07-16 09:35:39,839 >> {'loss': 0.0018, 'learning_rate': 6.4214e-07, 'epoch': 3.81, 'throughput': 615.55}
-[INFO|callbacks.py:310] 2024-07-16 09:35:50,950 >> {'loss': 0.0016, 'learning_rate': 6.1323e-07, 'epoch': 3.83, 'throughput': 615.59}
-[INFO|callbacks.py:310] 2024-07-16 09:36:02,073 >> {'loss': 0.0021, 'learning_rate': 5.8489e-07, 'epoch': 3.86, 'throughput': 615.56}
-[INFO|callbacks.py:310] 2024-07-16 09:36:13,188 >> {'loss': 0.0001, 'learning_rate': 5.5714e-07, 'epoch': 3.88, 'throughput': 615.68}
-[INFO|callbacks.py:310] 2024-07-16 09:36:24,312 >> {'loss': 0.0001, 'learning_rate': 5.2997e-07, 'epoch': 3.91, 'throughput': 615.50}
-[INFO|callbacks.py:310] 2024-07-16 09:36:35,410 >> {'loss': 0.0003, 'learning_rate': 5.0341e-07, 'epoch': 3.94, 'throughput': 615.45}
-[INFO|callbacks.py:310] 2024-07-16 09:36:46,505 >> {'loss': 0.0002, 'learning_rate': 4.7746e-07, 'epoch': 3.96, 'throughput': 615.52}
-[INFO|callbacks.py:310] 2024-07-16 09:36:57,590 >> {'loss': 0.0002, 'learning_rate': 4.5212e-07, 'epoch': 3.99, 'throughput': 615.41}
-[INFO|callbacks.py:310] 2024-07-16 09:37:08,665 >> {'loss': 0.0000, 'learning_rate': 4.2741e-07, 'epoch': 4.01, 'throughput': 615.56}
-[INFO|callbacks.py:310] 2024-07-16 09:37:19,763 >> {'loss': 0.0003, 'learning_rate': 4.0332e-07, 'epoch': 4.04, 'throughput': 615.58}
-[INFO|callbacks.py:310] 2024-07-16 09:37:30,878 >> {'loss': 0.0002, 'learning_rate': 3.7988e-07, 'epoch': 4.06, 'throughput': 615.55}
-[INFO|callbacks.py:310] 2024-07-16 09:37:41,999 >> {'loss': 0.0001, 'learning_rate': 3.5708e-07, 'epoch': 4.09, 'throughput': 615.40}
-[INFO|callbacks.py:310] 2024-07-16 09:37:53,113 >> {'loss': 0.0001, 'learning_rate': 3.3494e-07, 'epoch': 4.12, 'throughput': 615.53}
-[INFO|callbacks.py:310] 2024-07-16 09:38:04,234 >> {'loss': 0.0001, 'learning_rate': 3.1345e-07, 'epoch': 4.14, 'throughput': 615.52}
-[INFO|callbacks.py:310] 2024-07-16 09:38:15,321 >> {'loss': 0.0000, 'learning_rate': 2.9263e-07, 'epoch': 4.17, 'throughput': 615.59}
-[INFO|callbacks.py:310] 2024-07-16 09:38:26,408 >> {'loss': 0.0001, 'learning_rate': 2.7248e-07, 'epoch': 4.19, 'throughput': 615.69}
-[INFO|callbacks.py:310] 2024-07-16 09:38:37,489 >> {'loss': 0.0000, 'learning_rate': 2.5301e-07, 'epoch': 4.22, 'throughput': 615.71}
-[INFO|callbacks.py:310] 2024-07-16 09:38:48,575 >> {'loss': 0.0001, 'learning_rate': 2.3423e-07, 'epoch': 4.24, 'throughput': 615.55}
-[INFO|callbacks.py:310] 2024-07-16 09:38:59,677 >> {'loss': 0.0001, 'learning_rate': 2.1614e-07, 'epoch': 4.27, 'throughput': 615.61}
-[INFO|callbacks.py:310] 2024-07-16 09:39:10,799 >> {'loss': 0.0002, 'learning_rate': 1.9874e-07, 'epoch': 4.30, 'throughput': 615.64}
-[INFO|callbacks.py:310] 2024-07-16 09:39:21,928 >> {'loss': 0.0002, 'learning_rate': 1.8204e-07, 'epoch': 4.32, 'throughput': 615.58}
-[INFO|callbacks.py:310] 2024-07-16 09:39:33,061 >> {'loss': 0.0001, 'learning_rate': 1.6605e-07, 'epoch': 4.35, 'throughput': 615.46}
-[INFO|callbacks.py:310] 2024-07-16 09:39:44,166 >> {'loss': 0.0001, 'learning_rate': 1.5077e-07, 'epoch': 4.37, 'throughput': 615.43}
-[INFO|callbacks.py:310] 2024-07-16 09:39:55,251 >> {'loss': 0.0115, 'learning_rate': 1.3620e-07, 'epoch': 4.40, 'throughput': 615.48}
-[INFO|callbacks.py:310] 2024-07-16 09:40:06,330 >> {'loss': 0.0005, 'learning_rate': 1.2236e-07, 'epoch': 4.42, 'throughput': 615.47}
-[INFO|callbacks.py:310] 2024-07-16 09:40:17,431 >> {'loss': 0.0003, 'learning_rate': 1.0924e-07, 'epoch': 4.45, 'throughput': 615.57}
-[INFO|callbacks.py:310] 2024-07-16 09:40:28,525 >> {'loss': 0.0050, 'learning_rate': 9.6846e-08, 'epoch': 4.48, 'throughput': 615.49}
-[INFO|callbacks.py:310] 2024-07-16 09:40:39,640 >> {'loss': 0.0003, 'learning_rate': 8.5185e-08, 'epoch': 4.50, 'throughput': 615.38}
-[INFO|callbacks.py:310] 2024-07-16 09:40:50,749 >> {'loss': 0.0008, 'learning_rate': 7.4261e-08, 'epoch': 4.53, 'throughput': 615.27}
-[INFO|callbacks.py:310] 2024-07-16 09:41:01,862 >> {'loss': 0.0000, 'learning_rate': 6.4075e-08, 'epoch': 4.55, 'throughput': 615.37}
-[INFO|callbacks.py:310] 2024-07-16 09:41:12,986 >> {'loss': 0.0000, 'learning_rate': 5.4631e-08, 'epoch': 4.58, 'throughput': 615.39}
-[INFO|callbacks.py:310] 2024-07-16 09:41:24,079 >> {'loss': 0.0042, 'learning_rate': 4.5932e-08, 'epoch': 4.60, 'throughput': 615.48}
-[INFO|callbacks.py:310] 2024-07-16 09:41:35,169 >> {'loss': 0.0004, 'learning_rate': 3.7981e-08, 'epoch': 4.63, 'throughput': 615.54}
-[INFO|callbacks.py:310] 2024-07-16 09:41:46,249 >> {'loss': 0.0001, 'learning_rate': 3.0779e-08, 'epoch': 4.66, 'throughput': 615.42}
-[INFO|callbacks.py:310] 2024-07-16 09:41:57,352 >> {'loss': 0.0001, 'learning_rate': 2.4330e-08, 'epoch': 4.68, 'throughput': 615.30}
-[INFO|callbacks.py:310] 2024-07-16 09:42:08,449 >> {'loss': 0.0004, 'learning_rate': 1.8635e-08, 'epoch': 4.71, 'throughput': 615.12}
-[INFO|callbacks.py:310] 2024-07-16 09:42:19,548 >> {'loss': 0.0000, 'learning_rate': 1.3695e-08, 'epoch': 4.73, 'throughput': 615.05}
-[INFO|callbacks.py:310] 2024-07-16 09:42:30,662 >> {'loss': 0.0009, 'learning_rate': 9.5133e-09, 'epoch': 4.76, 'throughput': 615.11}
-[INFO|callbacks.py:310] 2024-07-16 09:42:41,790 >> {'loss': 0.0001, 'learning_rate': 6.0899e-09, 'epoch': 4.78, 'throughput': 615.12}
-[INFO|callbacks.py:310] 2024-07-16 09:42:52,921 >> {'loss': 0.0004, 'learning_rate': 3.4262e-09, 'epoch': 4.81, 'throughput': 615.30}
-[INFO|callbacks.py:310] 2024-07-16 09:43:04,012 >> {'loss': 0.0002, 'learning_rate': 1.5229e-09, 'epoch': 4.84, 'throughput': 615.28}
-[INFO|callbacks.py:310] 2024-07-16 09:43:15,108 >> {'loss': 0.0000, 'learning_rate': 3.8076e-10, 'epoch': 4.86, 'throughput': 615.26}
-[INFO|callbacks.py:310] 2024-07-16 09:43:26,201 >> {'loss': 0.0001, 'learning_rate': 0.0000e+00, 'epoch': 4.89, 'throughput': 615.25}
-[INFO|trainer.py:3478] 2024-07-16 09:43:32,570 >> Saving model checkpoint to saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/checkpoint-190
-[INFO|configuration_utils.py:472] 2024-07-16 09:43:32,573 >> Configuration saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/checkpoint-190/config.json
-[INFO|configuration_utils.py:769] 2024-07-16 09:43:32,573 >> Configuration saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/checkpoint-190/generation_config.json
-[INFO|modeling_utils.py:2698] 2024-07-16 09:43:46,233 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 3 checkpoint shards. You can find where each parameters has been saved in the index located at saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/checkpoint-190/model.safetensors.index.json.
-[INFO|tokenization_utils_base.py:2574] 2024-07-16 09:43:46,233 >> tokenizer config file saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/checkpoint-190/tokenizer_config.json
-[INFO|tokenization_utils_base.py:2583] 2024-07-16 09:43:46,234 >> Special tokens file saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/checkpoint-190/special_tokens_map.json
-[INFO|trainer.py:2383] 2024-07-16 09:44:16,328 >>
-Training completed. Do not forget to share your model on huggingface.co/models =)
-[INFO|trainer.py:3478] 2024-07-16 09:44:22,736 >> Saving model checkpoint to saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2
-[INFO|configuration_utils.py:472] 2024-07-16 09:44:22,738 >> Configuration saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/config.json
-[INFO|configuration_utils.py:769] 2024-07-16 09:44:22,739 >> Configuration saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/generation_config.json
-[INFO|modeling_utils.py:2698] 2024-07-16 09:44:36,499 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 3 checkpoint shards. You can find where each parameters has been saved in the index located at saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/model.safetensors.index.json.
-[INFO|tokenization_utils_base.py:2574] 2024-07-16 09:44:36,499 >> tokenizer config file saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/tokenizer_config.json
-[INFO|tokenization_utils_base.py:2583] 2024-07-16 09:44:36,499 >> Special tokens file saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/special_tokens_map.json
-[WARNING|ploting.py:89] 2024-07-16 09:44:37,565 >> No metric eval_loss to plot.
-[WARNING|ploting.py:89] 2024-07-16 09:44:37,565 >> No metric eval_accuracy to plot.
-[INFO|modelcard.py:449] 2024-07-16 09:44:37,565 >> Dropping the following result as it does not have all the necessary fields:
-{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}

+[INFO|tokenization_utils_base.py:2159] 2024-07-16 09:45:07,144 >> loading file tokenizer.json
+[INFO|tokenization_utils_base.py:2159] 2024-07-16 09:45:07,144 >> loading file added_tokens.json
+[INFO|tokenization_utils_base.py:2159] 2024-07-16 09:45:07,144 >> loading file special_tokens_map.json
+[INFO|tokenization_utils_base.py:2159] 2024-07-16 09:45:07,144 >> loading file tokenizer_config.json
+[INFO|loader.py:50] 2024-07-16 09:45:07,193 >> Loading dataset 0716_truthfulqa_benchmark_test.json...
+07/16/2024 09:45:07 - INFO - llamafactory.hparams.parser - Process rank: 4, device: cuda:4, n_gpu: 1, distributed training: True, compute dtype: None
+07/16/2024 09:45:07 - INFO - llamafactory.hparams.parser - Process rank: 2, device: cuda:2, n_gpu: 1, distributed training: True, compute dtype: None
+07/16/2024 09:45:07 - INFO - llamafactory.hparams.parser - Process rank: 7, device: cuda:7, n_gpu: 1, distributed training: True, compute dtype: None
+07/16/2024 09:45:07 - INFO - llamafactory.hparams.parser - Process rank: 3, device: cuda:3, n_gpu: 1, distributed training: True, compute dtype: None
+07/16/2024 09:45:07 - INFO - llamafactory.hparams.parser - Process rank: 1, device: cuda:1, n_gpu: 1, distributed training: True, compute dtype: None
+07/16/2024 09:45:07 - INFO - llamafactory.hparams.parser - Process rank: 6, device: cuda:6, n_gpu: 1, distributed training: True, compute dtype: None
+07/16/2024 09:45:07 - INFO - llamafactory.hparams.parser - Process rank: 5, device: cuda:5, n_gpu: 1, distributed training: True, compute dtype: None
+07/16/2024 09:45:08 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_test.json...
+07/16/2024 09:45:08 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_test.json...
+07/16/2024 09:45:08 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_test.json...
+07/16/2024 09:45:08 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_test.json...
+07/16/2024 09:45:08 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_test.json...
+07/16/2024 09:45:08 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_test.json...
+07/16/2024 09:45:08 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_test.json...
+[INFO|configuration_utils.py:731] 2024-07-16 09:45:10,115 >> loading configuration file saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/config.json
+[INFO|configuration_utils.py:800] 2024-07-16 09:45:10,118 >> Model config LlamaConfig {
+  "_name_or_path": "saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2",
   "architectures": [
     "LlamaForCausalLM"
   ],
   "rope_scaling": null,
   "rope_theta": 10000.0,
   "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
   "transformers_version": "4.42.3",
+  "use_cache": false,
   "vocab_size": 32000
 }
+[INFO|patcher.py:81] 2024-07-16 09:45:10,118 >> Using KV cache for faster generation.
+[INFO|modeling_utils.py:3553] 2024-07-16 09:45:10,162 >> loading weights file saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/model.safetensors.index.json
+[INFO|modeling_utils.py:1531] 2024-07-16 09:45:10,162 >> Instantiating LlamaForCausalLM model under default dtype torch.bfloat16.
+[INFO|configuration_utils.py:1000] 2024-07-16 09:45:10,163 >> Generate config GenerationConfig {
   "bos_token_id": 1,
   "eos_token_id": 2
 }
+07/16/2024 09:45:10 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
+07/16/2024 09:45:10 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
+07/16/2024 09:45:10 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
+07/16/2024 09:45:10 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
+07/16/2024 09:45:10 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
+07/16/2024 09:45:10 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
+07/16/2024 09:45:10 - INFO - llamafactory.model.patcher - Using KV cache for faster generation.
+[INFO|modeling_utils.py:4364] 2024-07-16 09:45:13,329 >> All model checkpoint weights were used when initializing LlamaForCausalLM.
+[INFO|modeling_utils.py:4372] 2024-07-16 09:45:13,329 >> All the weights of LlamaForCausalLM were initialized from the model checkpoint at saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2.
+If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
+[INFO|configuration_utils.py:953] 2024-07-16 09:45:13,333 >> loading configuration file saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2/generation_config.json
+[INFO|configuration_utils.py:1000] 2024-07-16 09:45:13,333 >> Generate config GenerationConfig {
   "bos_token_id": 1,
   "do_sample": true,
   "eos_token_id": 2,
 }
+[INFO|attention.py:80] 2024-07-16 09:45:13,339 >> Using torch SDPA for faster training and inference.
+[INFO|loader.py:196] 2024-07-16 09:45:13,344 >> all params: 6,738,415,616
+[INFO|trainer.py:3788] 2024-07-16 09:45:13,449 >>
+***** Running Prediction *****
+[INFO|trainer.py:3790] 2024-07-16 09:45:13,449 >>   Num examples = 1243
+[INFO|trainer.py:3793] 2024-07-16 09:45:13,449 >>   Batch size = 2
+[WARNING|logging.py:328] 2024-07-16 09:45:14,110 >> We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)
+07/16/2024 09:45:14 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+07/16/2024 09:45:14 - INFO - llamafactory.model.loader - all params: 6,738,415,616
+07/16/2024 09:45:14 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+07/16/2024 09:45:14 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+07/16/2024 09:45:14 - INFO - llamafactory.model.loader - all params: 6,738,415,616
+07/16/2024 09:45:14 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+07/16/2024 09:45:14 - INFO - llamafactory.model.loader - all params: 6,738,415,616
+07/16/2024 09:45:14 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+07/16/2024 09:45:14 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+07/16/2024 09:45:14 - INFO - llamafactory.model.loader - all params: 6,738,415,616
+07/16/2024 09:45:14 - INFO - llamafactory.model.loader - all params: 6,738,415,616
+07/16/2024 09:45:14 - INFO - llamafactory.model.loader - all params: 6,738,415,616
+07/16/2024 09:45:14 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+07/16/2024 09:45:14 - INFO - llamafactory.model.loader - all params: 6,738,415,616
+07/16/2024 09:45:15 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)
+07/16/2024 09:45:15 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)
+07/16/2024 09:45:15 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)
+07/16/2024 09:45:16 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)
+07/16/2024 09:45:16 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)
+07/16/2024 09:45:16 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)
+07/16/2024 09:45:16 - WARNING - transformers.models.llama.modeling_llama - We detected that you are passing `past_key_values` as a tuple and this is deprecated and will be removed in v4.43. Please use an appropriate `Cache` class (https://huggingface.co/docs/transformers/v4.41.3/en/internal/generation_utils#transformers.Cache)
+[INFO|trainer.py:127] 2024-07-16 09:45:24,567 >> Saving prediction results to saves/LLaMA2-7B-Chat/full/eval_2024-07-16-09-05-28/generated_predictions.jsonl

trainer_log.jsonl CHANGED Viewed

@@ -1,191 +1,15 @@
-{"current_steps": 1, "total_steps": 190, "loss": 8.3599, "learning_rate": 5.000000000000001e-07, "epoch": 0.02572347266881029, "percentage": 0.53, "elapsed_time": "0:00:12", "remaining_time": "0:40:53", "throughput": "548.54", "total_tokens": 7120}
-{"current_steps": 2, "total_steps": 190, "loss": 8.1891, "learning_rate": 1.0000000000000002e-06, "epoch": 0.05144694533762058, "percentage": 1.05, "elapsed_time": "0:00:24", "remaining_time": "0:37:46", "throughput": "575.99", "total_tokens": 13888}
-{"current_steps": 3, "total_steps": 190, "loss": 8.0792, "learning_rate": 1.5e-06, "epoch": 0.07717041800643087, "percentage": 1.58, "elapsed_time": "0:00:35", "remaining_time": "0:36:35", "throughput": "586.40", "total_tokens": 20656}
-{"current_steps": 4, "total_steps": 190, "loss": 7.9682, "learning_rate": 2.0000000000000003e-06, "epoch": 0.10289389067524116, "percentage": 2.11, "elapsed_time": "0:00:46", "remaining_time": "0:35:53", "throughput": "586.87", "total_tokens": 27184}
-{"current_steps": 5, "total_steps": 190, "loss": 6.9482, "learning_rate": 2.5e-06, "epoch": 0.12861736334405144, "percentage": 2.63, "elapsed_time": "0:00:57", "remaining_time": "0:35:24", "throughput": "599.42", "total_tokens": 34416}
-{"current_steps": 6, "total_steps": 190, "loss": 5.1505, "learning_rate": 3e-06, "epoch": 0.15434083601286175, "percentage": 3.16, "elapsed_time": "0:01:08", "remaining_time": "0:35:00", "throughput": "599.28", "total_tokens": 41056}
-{"current_steps": 7, "total_steps": 190, "loss": 4.7491, "learning_rate": 3.5e-06, "epoch": 0.18006430868167203, "percentage": 3.68, "elapsed_time": "0:01:19", "remaining_time": "0:34:41", "throughput": "596.99", "total_tokens": 47536}
-{"current_steps": 8, "total_steps": 190, "loss": 3.2164, "learning_rate": 4.000000000000001e-06, "epoch": 0.2057877813504823, "percentage": 4.21, "elapsed_time": "0:01:30", "remaining_time": "0:34:24", "throughput": "600.21", "total_tokens": 54464}
-{"current_steps": 9, "total_steps": 190, "loss": 2.7761, "learning_rate": 4.5e-06, "epoch": 0.2315112540192926, "percentage": 4.74, "elapsed_time": "0:01:41", "remaining_time": "0:34:08", "throughput": "603.94", "total_tokens": 61520}
-{"current_steps": 10, "total_steps": 190, "loss": 0.6703, "learning_rate": 5e-06, "epoch": 0.2572347266881029, "percentage": 5.26, "elapsed_time": "0:01:52", "remaining_time": "0:33:53", "throughput": "605.96", "total_tokens": 68464}
-{"current_steps": 11, "total_steps": 190, "loss": 0.3255, "learning_rate": 4.9996192378909785e-06, "epoch": 0.2829581993569132, "percentage": 5.79, "elapsed_time": "0:02:04", "remaining_time": "0:33:39", "throughput": "605.48", "total_tokens": 75152}
-{"current_steps": 12, "total_steps": 190, "loss": 0.3301, "learning_rate": 4.99847706754774e-06, "epoch": 0.3086816720257235, "percentage": 6.32, "elapsed_time": "0:02:15", "remaining_time": "0:33:25", "throughput": "605.64", "total_tokens": 81888}
-{"current_steps": 13, "total_steps": 190, "loss": 0.2121, "learning_rate": 4.9965738368864345e-06, "epoch": 0.33440514469453375, "percentage": 6.84, "elapsed_time": "0:02:26", "remaining_time": "0:33:12", "throughput": "606.15", "total_tokens": 88688}
-{"current_steps": 14, "total_steps": 190, "loss": 1.1565, "learning_rate": 4.993910125649561e-06, "epoch": 0.36012861736334406, "percentage": 7.37, "elapsed_time": "0:02:37", "remaining_time": "0:32:58", "throughput": "607.41", "total_tokens": 95616}
-{"current_steps": 15, "total_steps": 190, "loss": 0.8054, "learning_rate": 4.990486745229364e-06, "epoch": 0.3858520900321543, "percentage": 7.89, "elapsed_time": "0:02:48", "remaining_time": "0:32:45", "throughput": "606.19", "total_tokens": 102144}
-{"current_steps": 16, "total_steps": 190, "loss": 0.2386, "learning_rate": 4.986304738420684e-06, "epoch": 0.4115755627009646, "percentage": 8.42, "elapsed_time": "0:02:59", "remaining_time": "0:32:33", "throughput": "607.52", "total_tokens": 109120}
-{"current_steps": 17, "total_steps": 190, "loss": 0.3161, "learning_rate": 4.981365379103306e-06, "epoch": 0.43729903536977494, "percentage": 8.95, "elapsed_time": "0:03:10", "remaining_time": "0:32:21", "throughput": "606.78", "total_tokens": 115744}
-{"current_steps": 18, "total_steps": 190, "loss": 0.2773, "learning_rate": 4.975670171853926e-06, "epoch": 0.4630225080385852, "percentage": 9.47, "elapsed_time": "0:03:21", "remaining_time": "0:32:09", "throughput": "607.80", "total_tokens": 122704}
-{"current_steps": 19, "total_steps": 190, "loss": 0.2062, "learning_rate": 4.9692208514878445e-06, "epoch": 0.4887459807073955, "percentage": 10.0, "elapsed_time": "0:03:33", "remaining_time": "0:31:57", "throughput": "608.19", "total_tokens": 129552}
-{"current_steps": 20, "total_steps": 190, "loss": 0.1837, "learning_rate": 4.962019382530521e-06, "epoch": 0.5144694533762058, "percentage": 10.53, "elapsed_time": "0:03:44", "remaining_time": "0:31:45", "throughput": "609.22", "total_tokens": 136544}
-{"current_steps": 21, "total_steps": 190, "loss": 0.1735, "learning_rate": 4.9540679586191605e-06, "epoch": 0.5401929260450161, "percentage": 11.05, "elapsed_time": "0:03:55", "remaining_time": "0:31:32", "throughput": "610.01", "total_tokens": 143488}
-{"current_steps": 22, "total_steps": 190, "loss": 0.1588, "learning_rate": 4.9453690018345144e-06, "epoch": 0.5659163987138264, "percentage": 11.58, "elapsed_time": "0:04:06", "remaining_time": "0:31:20", "throughput": "609.91", "total_tokens": 150224}
-{"current_steps": 23, "total_steps": 190, "loss": 0.1443, "learning_rate": 4.935925161963089e-06, "epoch": 0.5916398713826366, "percentage": 12.11, "elapsed_time": "0:04:17", "remaining_time": "0:31:09", "throughput": "610.82", "total_tokens": 157232}
-{"current_steps": 24, "total_steps": 190, "loss": 0.157, "learning_rate": 4.925739315689991e-06, "epoch": 0.617363344051447, "percentage": 12.63, "elapsed_time": "0:04:28", "remaining_time": "0:30:57", "throughput": "609.97", "total_tokens": 163776}
-{"current_steps": 25, "total_steps": 190, "loss": 0.1199, "learning_rate": 4.914814565722671e-06, "epoch": 0.6430868167202572, "percentage": 13.16, "elapsed_time": "0:04:39", "remaining_time": "0:30:45", "throughput": "609.21", "total_tokens": 170352}
-{"current_steps": 26, "total_steps": 190, "loss": 0.1539, "learning_rate": 4.903154239845798e-06, "epoch": 0.6688102893890675, "percentage": 13.68, "elapsed_time": "0:04:50", "remaining_time": "0:30:33", "throughput": "609.20", "total_tokens": 177120}
-{"current_steps": 27, "total_steps": 190, "loss": 0.1208, "learning_rate": 4.890761889907589e-06, "epoch": 0.6945337620578779, "percentage": 14.21, "elapsed_time": "0:05:01", "remaining_time": "0:30:22", "throughput": "609.87", "total_tokens": 184096}
-{"current_steps": 28, "total_steps": 190, "loss": 0.0954, "learning_rate": 4.8776412907378845e-06, "epoch": 0.7202572347266881, "percentage": 14.74, "elapsed_time": "0:05:12", "remaining_time": "0:30:10", "throughput": "610.39", "total_tokens": 191040}
-{"current_steps": 29, "total_steps": 190, "loss": 0.1387, "learning_rate": 4.863796438998293e-06, "epoch": 0.7459807073954984, "percentage": 15.26, "elapsed_time": "0:05:24", "remaining_time": "0:29:59", "throughput": "611.13", "total_tokens": 198064}
-{"current_steps": 30, "total_steps": 190, "loss": 0.1484, "learning_rate": 4.849231551964771e-06, "epoch": 0.7717041800643086, "percentage": 15.79, "elapsed_time": "0:05:35", "remaining_time": "0:29:47", "throughput": "612.02", "total_tokens": 205136}
-{"current_steps": 31, "total_steps": 190, "loss": 0.0998, "learning_rate": 4.833951066243004e-06, "epoch": 0.797427652733119, "percentage": 16.32, "elapsed_time": "0:05:46", "remaining_time": "0:29:36", "throughput": "612.22", "total_tokens": 212000}
-{"current_steps": 32, "total_steps": 190, "loss": 0.1068, "learning_rate": 4.817959636416969e-06, "epoch": 0.8231511254019293, "percentage": 16.84, "elapsed_time": "0:05:57", "remaining_time": "0:29:24", "throughput": "612.05", "total_tokens": 218720}
-{"current_steps": 33, "total_steps": 190, "loss": 0.0801, "learning_rate": 4.801262133631101e-06, "epoch": 0.8488745980707395, "percentage": 17.37, "elapsed_time": "0:06:08", "remaining_time": "0:29:12", "throughput": "612.99", "total_tokens": 225856}
-{"current_steps": 34, "total_steps": 190, "loss": 0.1066, "learning_rate": 4.783863644106502e-06, "epoch": 0.8745980707395499, "percentage": 17.89, "elapsed_time": "0:06:19", "remaining_time": "0:29:01", "throughput": "612.89", "total_tokens": 232640}
-{"current_steps": 35, "total_steps": 190, "loss": 0.1038, "learning_rate": 4.765769467591626e-06, "epoch": 0.9003215434083601, "percentage": 18.42, "elapsed_time": "0:06:30", "remaining_time": "0:28:50", "throughput": "613.01", "total_tokens": 239504}
-{"current_steps": 36, "total_steps": 190, "loss": 0.106, "learning_rate": 4.746985115747918e-06, "epoch": 0.9260450160771704, "percentage": 18.95, "elapsed_time": "0:06:41", "remaining_time": "0:28:38", "throughput": "612.94", "total_tokens": 246288}
-{"current_steps": 37, "total_steps": 190, "loss": 0.1107, "learning_rate": 4.72751631047092e-06, "epoch": 0.9517684887459807, "percentage": 19.47, "elapsed_time": "0:06:52", "remaining_time": "0:28:27", "throughput": "613.01", "total_tokens": 253136}
-{"current_steps": 38, "total_steps": 190, "loss": 0.1372, "learning_rate": 4.707368982147318e-06, "epoch": 0.977491961414791, "percentage": 20.0, "elapsed_time": "0:07:04", "remaining_time": "0:28:16", "throughput": "613.54", "total_tokens": 260160}
-{"current_steps": 39, "total_steps": 190, "loss": 0.0816, "learning_rate": 4.68654926784849e-06, "epoch": 1.0032154340836013, "percentage": 20.53, "elapsed_time": "0:07:15", "remaining_time": "0:28:04", "throughput": "613.88", "total_tokens": 267120}
-{"current_steps": 40, "total_steps": 190, "loss": 0.0743, "learning_rate": 4.665063509461098e-06, "epoch": 1.0289389067524115, "percentage": 21.05, "elapsed_time": "0:07:26", "remaining_time": "0:27:53", "throughput": "614.30", "total_tokens": 274112}
-{"current_steps": 41, "total_steps": 190, "loss": 0.072, "learning_rate": 4.642918251755281e-06, "epoch": 1.0546623794212218, "percentage": 21.58, "elapsed_time": "0:07:37", "remaining_time": "0:27:41", "throughput": "614.77", "total_tokens": 281136}
-{"current_steps": 42, "total_steps": 190, "loss": 0.0596, "learning_rate": 4.620120240391065e-06, "epoch": 1.0803858520900322, "percentage": 22.11, "elapsed_time": "0:07:48", "remaining_time": "0:27:30", "throughput": "614.97", "total_tokens": 288048}
-{"current_steps": 43, "total_steps": 190, "loss": 0.0544, "learning_rate": 4.596676419863561e-06, "epoch": 1.1061093247588425, "percentage": 22.63, "elapsed_time": "0:07:59", "remaining_time": "0:27:19", "throughput": "615.46", "total_tokens": 295120}
-{"current_steps": 44, "total_steps": 190, "loss": 0.0342, "learning_rate": 4.572593931387604e-06, "epoch": 1.1318327974276527, "percentage": 23.16, "elapsed_time": "0:08:10", "remaining_time": "0:27:07", "throughput": "615.55", "total_tokens": 302000}
-{"current_steps": 45, "total_steps": 190, "loss": 0.0394, "learning_rate": 4.54788011072248e-06, "epoch": 1.157556270096463, "percentage": 23.68, "elapsed_time": "0:08:21", "remaining_time": "0:26:56", "throughput": "615.19", "total_tokens": 308672}
-{"current_steps": 46, "total_steps": 190, "loss": 0.0196, "learning_rate": 4.522542485937369e-06, "epoch": 1.1832797427652733, "percentage": 24.21, "elapsed_time": "0:08:32", "remaining_time": "0:26:45", "throughput": "615.36", "total_tokens": 315600}
-{"current_steps": 47, "total_steps": 190, "loss": 0.0411, "learning_rate": 4.496588775118232e-06, "epoch": 1.2090032154340835, "percentage": 24.74, "elapsed_time": "0:08:43", "remaining_time": "0:26:34", "throughput": "615.43", "total_tokens": 322464}
-{"current_steps": 48, "total_steps": 190, "loss": 0.0257, "learning_rate": 4.470026884016805e-06, "epoch": 1.234726688102894, "percentage": 25.26, "elapsed_time": "0:08:55", "remaining_time": "0:26:22", "throughput": "614.94", "total_tokens": 329024}
-{"current_steps": 49, "total_steps": 190, "loss": 0.0289, "learning_rate": 4.442864903642428e-06, "epoch": 1.2604501607717042, "percentage": 25.79, "elapsed_time": "0:09:06", "remaining_time": "0:26:11", "throughput": "615.29", "total_tokens": 336032}
-{"current_steps": 50, "total_steps": 190, "loss": 0.1193, "learning_rate": 4.415111107797445e-06, "epoch": 1.2861736334405145, "percentage": 26.32, "elapsed_time": "0:09:17", "remaining_time": "0:26:00", "throughput": "615.01", "total_tokens": 342704}
-{"current_steps": 51, "total_steps": 190, "loss": 0.0883, "learning_rate": 4.386773950556931e-06, "epoch": 1.3118971061093248, "percentage": 26.84, "elapsed_time": "0:09:28", "remaining_time": "0:25:48", "throughput": "614.92", "total_tokens": 349472}
-{"current_steps": 52, "total_steps": 190, "loss": 0.0377, "learning_rate": 4.357862063693486e-06, "epoch": 1.337620578778135, "percentage": 27.37, "elapsed_time": "0:09:39", "remaining_time": "0:25:37", "throughput": "614.86", "total_tokens": 356272}
-{"current_steps": 53, "total_steps": 190, "loss": 0.0602, "learning_rate": 4.328384254047927e-06, "epoch": 1.3633440514469453, "percentage": 27.89, "elapsed_time": "0:09:50", "remaining_time": "0:25:26", "throughput": "614.73", "total_tokens": 363040}
-{"current_steps": 54, "total_steps": 190, "loss": 0.083, "learning_rate": 4.2983495008466285e-06, "epoch": 1.3890675241157555, "percentage": 28.42, "elapsed_time": "0:10:01", "remaining_time": "0:25:15", "throughput": "614.38", "total_tokens": 369664}
-{"current_steps": 55, "total_steps": 190, "loss": 0.0358, "learning_rate": 4.267766952966369e-06, "epoch": 1.414790996784566, "percentage": 28.95, "elapsed_time": "0:10:12", "remaining_time": "0:25:04", "throughput": "614.72", "total_tokens": 376704}
-{"current_steps": 56, "total_steps": 190, "loss": 0.0321, "learning_rate": 4.236645926147493e-06, "epoch": 1.4405144694533762, "percentage": 29.47, "elapsed_time": "0:10:23", "remaining_time": "0:24:52", "throughput": "614.84", "total_tokens": 383600}
-{"current_steps": 57, "total_steps": 190, "loss": 0.0452, "learning_rate": 4.204995900156247e-06, "epoch": 1.4662379421221865, "percentage": 30.0, "elapsed_time": "0:10:34", "remaining_time": "0:24:41", "throughput": "615.11", "total_tokens": 390592}
-{"current_steps": 58, "total_steps": 190, "loss": 0.0915, "learning_rate": 4.172826515897146e-06, "epoch": 1.4919614147909968, "percentage": 30.53, "elapsed_time": "0:10:46", "remaining_time": "0:24:30", "throughput": "615.02", "total_tokens": 397360}
-{"current_steps": 59, "total_steps": 190, "loss": 0.0651, "learning_rate": 4.140147572476269e-06, "epoch": 1.517684887459807, "percentage": 31.05, "elapsed_time": "0:10:57", "remaining_time": "0:24:19", "throughput": "614.81", "total_tokens": 404048}
-{"current_steps": 60, "total_steps": 190, "loss": 0.0868, "learning_rate": 4.106969024216348e-06, "epoch": 1.5434083601286175, "percentage": 31.58, "elapsed_time": "0:11:08", "remaining_time": "0:24:08", "throughput": "614.92", "total_tokens": 410960}
-{"current_steps": 61, "total_steps": 190, "loss": 0.0554, "learning_rate": 4.073300977624594e-06, "epoch": 1.5691318327974275, "percentage": 32.11, "elapsed_time": "0:11:19", "remaining_time": "0:23:56", "throughput": "615.06", "total_tokens": 417888}
-{"current_steps": 62, "total_steps": 190, "loss": 0.0336, "learning_rate": 4.039153688314146e-06, "epoch": 1.594855305466238, "percentage": 32.63, "elapsed_time": "0:11:30", "remaining_time": "0:23:45", "throughput": "615.29", "total_tokens": 424880}
-{"current_steps": 63, "total_steps": 190, "loss": 0.0455, "learning_rate": 4.0045375578801216e-06, "epoch": 1.6205787781350482, "percentage": 33.16, "elapsed_time": "0:11:41", "remaining_time": "0:23:34", "throughput": "615.69", "total_tokens": 432000}
-{"current_steps": 64, "total_steps": 190, "loss": 0.0406, "learning_rate": 3.969463130731183e-06, "epoch": 1.6463022508038585, "percentage": 33.68, "elapsed_time": "0:11:52", "remaining_time": "0:23:23", "throughput": "615.45", "total_tokens": 438672}
-{"current_steps": 65, "total_steps": 190, "loss": 0.0461, "learning_rate": 3.933941090877615e-06, "epoch": 1.6720257234726688, "percentage": 34.21, "elapsed_time": "0:12:03", "remaining_time": "0:23:12", "throughput": "615.37", "total_tokens": 445440}
-{"current_steps": 66, "total_steps": 190, "loss": 0.0466, "learning_rate": 3.897982258676867e-06, "epoch": 1.697749196141479, "percentage": 34.74, "elapsed_time": "0:12:14", "remaining_time": "0:23:00", "throughput": "615.10", "total_tokens": 452064}
-{"current_steps": 67, "total_steps": 190, "loss": 0.0382, "learning_rate": 3.861597587537568e-06, "epoch": 1.7234726688102895, "percentage": 35.26, "elapsed_time": "0:12:26", "remaining_time": "0:22:49", "throughput": "615.23", "total_tokens": 458992}
-{"current_steps": 68, "total_steps": 190, "loss": 0.0426, "learning_rate": 3.824798160583012e-06, "epoch": 1.7491961414790995, "percentage": 35.79, "elapsed_time": "0:12:37", "remaining_time": "0:22:38", "throughput": "614.90", "total_tokens": 465568}
-{"current_steps": 69, "total_steps": 190, "loss": 0.0264, "learning_rate": 3.787595187275136e-06, "epoch": 1.77491961414791, "percentage": 36.32, "elapsed_time": "0:12:48", "remaining_time": "0:22:27", "throughput": "615.03", "total_tokens": 472496}
-{"current_steps": 70, "total_steps": 190, "loss": 0.0567, "learning_rate": 3.7500000000000005e-06, "epoch": 1.8006430868167203, "percentage": 36.84, "elapsed_time": "0:12:59", "remaining_time": "0:22:16", "throughput": "615.11", "total_tokens": 479392}
-{"current_steps": 71, "total_steps": 190, "loss": 0.0688, "learning_rate": 3.7120240506158433e-06, "epoch": 1.8263665594855305, "percentage": 37.37, "elapsed_time": "0:13:10", "remaining_time": "0:22:04", "throughput": "615.35", "total_tokens": 486416}
-{"current_steps": 72, "total_steps": 190, "loss": 0.0351, "learning_rate": 3.6736789069647273e-06, "epoch": 1.852090032154341, "percentage": 37.89, "elapsed_time": "0:13:21", "remaining_time": "0:21:53", "throughput": "614.88", "total_tokens": 492896}
-{"current_steps": 73, "total_steps": 190, "loss": 0.0246, "learning_rate": 3.634976249348867e-06, "epoch": 1.877813504823151, "percentage": 38.42, "elapsed_time": "0:13:32", "remaining_time": "0:21:42", "throughput": "614.93", "total_tokens": 499760}
-{"current_steps": 74, "total_steps": 190, "loss": 0.0364, "learning_rate": 3.595927866972694e-06, "epoch": 1.9035369774919615, "percentage": 38.95, "elapsed_time": "0:13:43", "remaining_time": "0:21:31", "throughput": "615.23", "total_tokens": 506816}
-{"current_steps": 75, "total_steps": 190, "loss": 0.0352, "learning_rate": 3.556545654351749e-06, "epoch": 1.9292604501607717, "percentage": 39.47, "elapsed_time": "0:13:54", "remaining_time": "0:21:20", "throughput": "615.23", "total_tokens": 513648}
-{"current_steps": 76, "total_steps": 190, "loss": 0.0915, "learning_rate": 3.516841607689501e-06, "epoch": 1.954983922829582, "percentage": 40.0, "elapsed_time": "0:14:05", "remaining_time": "0:21:08", "throughput": "615.24", "total_tokens": 520480}
-{"current_steps": 77, "total_steps": 190, "loss": 0.0327, "learning_rate": 3.476827821223184e-06, "epoch": 1.9807073954983923, "percentage": 40.53, "elapsed_time": "0:14:17", "remaining_time": "0:20:57", "throughput": "614.95", "total_tokens": 527056}
-{"current_steps": 78, "total_steps": 190, "loss": 0.0448, "learning_rate": 3.436516483539781e-06, "epoch": 2.0064308681672025, "percentage": 41.05, "elapsed_time": "0:14:28", "remaining_time": "0:20:46", "throughput": "615.21", "total_tokens": 534112}
-{"current_steps": 79, "total_steps": 190, "loss": 0.0186, "learning_rate": 3.39591987386325e-06, "epoch": 2.032154340836013, "percentage": 41.58, "elapsed_time": "0:14:39", "remaining_time": "0:20:35", "throughput": "615.29", "total_tokens": 541024}
-{"current_steps": 80, "total_steps": 190, "loss": 0.0342, "learning_rate": 3.3550503583141726e-06, "epoch": 2.057877813504823, "percentage": 42.11, "elapsed_time": "0:14:50", "remaining_time": "0:20:24", "throughput": "615.30", "total_tokens": 547888}
-{"current_steps": 81, "total_steps": 190, "loss": 0.0079, "learning_rate": 3.313920386142892e-06, "epoch": 2.0836012861736335, "percentage": 42.63, "elapsed_time": "0:15:01", "remaining_time": "0:20:13", "throughput": "615.14", "total_tokens": 554592}
-{"current_steps": 82, "total_steps": 190, "loss": 0.0177, "learning_rate": 3.272542485937369e-06, "epoch": 2.1093247588424435, "percentage": 43.16, "elapsed_time": "0:15:12", "remaining_time": "0:20:02", "throughput": "615.01", "total_tokens": 561296}
-{"current_steps": 83, "total_steps": 190, "loss": 0.0139, "learning_rate": 3.230929261806842e-06, "epoch": 2.135048231511254, "percentage": 43.68, "elapsed_time": "0:15:23", "remaining_time": "0:19:50", "throughput": "614.74", "total_tokens": 567872}
-{"current_steps": 84, "total_steps": 190, "loss": 0.0103, "learning_rate": 3.189093389542498e-06, "epoch": 2.1607717041800645, "percentage": 44.21, "elapsed_time": "0:15:34", "remaining_time": "0:19:39", "throughput": "615.15", "total_tokens": 575072}
-{"current_steps": 85, "total_steps": 190, "loss": 0.0221, "learning_rate": 3.147047612756302e-06, "epoch": 2.1864951768488745, "percentage": 44.74, "elapsed_time": "0:15:45", "remaining_time": "0:19:28", "throughput": "615.35", "total_tokens": 582080}
-{"current_steps": 86, "total_steps": 190, "loss": 0.0021, "learning_rate": 3.1048047389991693e-06, "epoch": 2.212218649517685, "percentage": 45.26, "elapsed_time": "0:15:57", "remaining_time": "0:19:17", "throughput": "615.26", "total_tokens": 588816}
-{"current_steps": 87, "total_steps": 190, "loss": 0.011, "learning_rate": 3.062377635859663e-06, "epoch": 2.237942122186495, "percentage": 45.79, "elapsed_time": "0:16:08", "remaining_time": "0:19:06", "throughput": "615.65", "total_tokens": 596032}
-{"current_steps": 88, "total_steps": 190, "loss": 0.0081, "learning_rate": 3.019779227044398e-06, "epoch": 2.2636655948553055, "percentage": 46.32, "elapsed_time": "0:16:19", "remaining_time": "0:18:55", "throughput": "615.45", "total_tokens": 602672}
-{"current_steps": 89, "total_steps": 190, "loss": 0.0149, "learning_rate": 2.9770224884413625e-06, "epoch": 2.289389067524116, "percentage": 46.84, "elapsed_time": "0:16:30", "remaining_time": "0:18:43", "throughput": "615.35", "total_tokens": 609424}
-{"current_steps": 90, "total_steps": 190, "loss": 0.001, "learning_rate": 2.9341204441673267e-06, "epoch": 2.315112540192926, "percentage": 47.37, "elapsed_time": "0:16:41", "remaining_time": "0:18:32", "throughput": "615.53", "total_tokens": 616448}
-{"current_steps": 91, "total_steps": 190, "loss": 0.007, "learning_rate": 2.8910861626005774e-06, "epoch": 2.3408360128617365, "percentage": 47.89, "elapsed_time": "0:16:52", "remaining_time": "0:18:21", "throughput": "615.54", "total_tokens": 623296}
-{"current_steps": 92, "total_steps": 190, "loss": 0.0089, "learning_rate": 2.847932752400164e-06, "epoch": 2.3665594855305465, "percentage": 48.42, "elapsed_time": "0:17:03", "remaining_time": "0:18:10", "throughput": "615.48", "total_tokens": 630064}
-{"current_steps": 93, "total_steps": 190, "loss": 0.0013, "learning_rate": 2.804673358512869e-06, "epoch": 2.392282958199357, "percentage": 48.95, "elapsed_time": "0:17:14", "remaining_time": "0:17:59", "throughput": "615.67", "total_tokens": 637088}
-{"current_steps": 94, "total_steps": 190, "loss": 0.0267, "learning_rate": 2.761321158169134e-06, "epoch": 2.418006430868167, "percentage": 49.47, "elapsed_time": "0:17:25", "remaining_time": "0:17:48", "throughput": "615.82", "total_tokens": 644080}
-{"current_steps": 95, "total_steps": 190, "loss": 0.0171, "learning_rate": 2.717889356869146e-06, "epoch": 2.4437299035369775, "percentage": 50.0, "elapsed_time": "0:17:36", "remaining_time": "0:17:36", "throughput": "615.76", "total_tokens": 650848}
-{"current_steps": 96, "total_steps": 190, "loss": 0.0375, "learning_rate": 2.6743911843603134e-06, "epoch": 2.469453376205788, "percentage": 50.53, "elapsed_time": "0:17:48", "remaining_time": "0:17:25", "throughput": "615.50", "total_tokens": 657424}
-{"current_steps": 97, "total_steps": 190, "loss": 0.0101, "learning_rate": 2.6308398906073603e-06, "epoch": 2.495176848874598, "percentage": 51.05, "elapsed_time": "0:17:59", "remaining_time": "0:17:14", "throughput": "615.37", "total_tokens": 664128}
-{"current_steps": 98, "total_steps": 190, "loss": 0.0282, "learning_rate": 2.587248741756253e-06, "epoch": 2.5209003215434085, "percentage": 51.58, "elapsed_time": "0:18:10", "remaining_time": "0:17:03", "throughput": "615.50", "total_tokens": 671120}
-{"current_steps": 99, "total_steps": 190, "loss": 0.0069, "learning_rate": 2.543631016093209e-06, "epoch": 2.5466237942122185, "percentage": 52.11, "elapsed_time": "0:18:21", "remaining_time": "0:16:52", "throughput": "615.47", "total_tokens": 677920}
-{"current_steps": 100, "total_steps": 190, "loss": 0.0135, "learning_rate": 2.5e-06, "epoch": 2.572347266881029, "percentage": 52.63, "elapsed_time": "0:18:32", "remaining_time": "0:16:41", "throughput": "615.66", "total_tokens": 684960}
-{"current_steps": 101, "total_steps": 190, "loss": 0.0062, "learning_rate": 2.4563689839067913e-06, "epoch": 2.598070739549839, "percentage": 53.16, "elapsed_time": "0:18:43", "remaining_time": "0:16:30", "throughput": "615.71", "total_tokens": 691856}
-{"current_steps": 102, "total_steps": 190, "loss": 0.005, "learning_rate": 2.4127512582437486e-06, "epoch": 2.6237942122186495, "percentage": 53.68, "elapsed_time": "0:18:54", "remaining_time": "0:16:19", "throughput": "615.56", "total_tokens": 698512}
-{"current_steps": 103, "total_steps": 190, "loss": 0.0285, "learning_rate": 2.3691601093926406e-06, "epoch": 2.64951768488746, "percentage": 54.21, "elapsed_time": "0:19:05", "remaining_time": "0:16:07", "throughput": "615.65", "total_tokens": 705440}
-{"current_steps": 104, "total_steps": 190, "loss": 0.0225, "learning_rate": 2.325608815639687e-06, "epoch": 2.67524115755627, "percentage": 54.74, "elapsed_time": "0:19:16", "remaining_time": "0:15:56", "throughput": "615.86", "total_tokens": 712528}
-{"current_steps": 105, "total_steps": 190, "loss": 0.028, "learning_rate": 2.2821106431308546e-06, "epoch": 2.7009646302250805, "percentage": 55.26, "elapsed_time": "0:19:28", "remaining_time": "0:15:45", "throughput": "615.69", "total_tokens": 719168}
-{"current_steps": 106, "total_steps": 190, "loss": 0.0176, "learning_rate": 2.238678841830867e-06, "epoch": 2.7266881028938905, "percentage": 55.79, "elapsed_time": "0:19:39", "remaining_time": "0:15:34", "throughput": "615.60", "total_tokens": 725904}
-{"current_steps": 107, "total_steps": 190, "loss": 0.0047, "learning_rate": 2.195326641487132e-06, "epoch": 2.752411575562701, "percentage": 56.32, "elapsed_time": "0:19:50", "remaining_time": "0:15:23", "throughput": "615.36", "total_tokens": 732480}
-{"current_steps": 108, "total_steps": 190, "loss": 0.0135, "learning_rate": 2.1520672475998374e-06, "epoch": 2.778135048231511, "percentage": 56.84, "elapsed_time": "0:20:01", "remaining_time": "0:15:12", "throughput": "615.25", "total_tokens": 739184}
-{"current_steps": 109, "total_steps": 190, "loss": 0.0044, "learning_rate": 2.1089138373994226e-06, "epoch": 2.8038585209003215, "percentage": 57.37, "elapsed_time": "0:20:12", "remaining_time": "0:15:01", "throughput": "615.51", "total_tokens": 746320}
-{"current_steps": 110, "total_steps": 190, "loss": 0.0252, "learning_rate": 2.0658795558326745e-06, "epoch": 2.829581993569132, "percentage": 57.89, "elapsed_time": "0:20:23", "remaining_time": "0:14:49", "throughput": "615.50", "total_tokens": 753136}
-{"current_steps": 111, "total_steps": 190, "loss": 0.0249, "learning_rate": 2.022977511558638e-06, "epoch": 2.855305466237942, "percentage": 58.42, "elapsed_time": "0:20:34", "remaining_time": "0:14:38", "throughput": "615.61", "total_tokens": 760096}
-{"current_steps": 112, "total_steps": 190, "loss": 0.0146, "learning_rate": 1.9802207729556023e-06, "epoch": 2.8810289389067525, "percentage": 58.95, "elapsed_time": "0:20:45", "remaining_time": "0:14:27", "throughput": "615.75", "total_tokens": 767104}
-{"current_steps": 113, "total_steps": 190, "loss": 0.0044, "learning_rate": 1.937622364140338e-06, "epoch": 2.906752411575563, "percentage": 59.47, "elapsed_time": "0:20:56", "remaining_time": "0:14:16", "throughput": "615.69", "total_tokens": 773872}
-{"current_steps": 114, "total_steps": 190, "loss": 0.0054, "learning_rate": 1.895195261000831e-06, "epoch": 2.932475884244373, "percentage": 60.0, "elapsed_time": "0:21:08", "remaining_time": "0:14:05", "throughput": "615.71", "total_tokens": 780736}
-{"current_steps": 115, "total_steps": 190, "loss": 0.0106, "learning_rate": 1.852952387243698e-06, "epoch": 2.958199356913183, "percentage": 60.53, "elapsed_time": "0:21:19", "remaining_time": "0:13:54", "throughput": "615.58", "total_tokens": 787424}
-{"current_steps": 116, "total_steps": 190, "loss": 0.0167, "learning_rate": 1.8109066104575023e-06, "epoch": 2.9839228295819935, "percentage": 61.05, "elapsed_time": "0:21:30", "remaining_time": "0:13:43", "throughput": "615.53", "total_tokens": 794224}
-{"current_steps": 117, "total_steps": 190, "loss": 0.009, "learning_rate": 1.7690707381931585e-06, "epoch": 3.009646302250804, "percentage": 61.58, "elapsed_time": "0:21:41", "remaining_time": "0:13:31", "throughput": "615.55", "total_tokens": 801088}
-{"current_steps": 118, "total_steps": 190, "loss": 0.0024, "learning_rate": 1.7274575140626318e-06, "epoch": 3.035369774919614, "percentage": 62.11, "elapsed_time": "0:21:52", "remaining_time": "0:13:20", "throughput": "615.66", "total_tokens": 808048}
-{"current_steps": 119, "total_steps": 190, "loss": 0.0235, "learning_rate": 1.686079613857109e-06, "epoch": 3.0610932475884245, "percentage": 62.63, "elapsed_time": "0:22:03", "remaining_time": "0:13:09", "throughput": "615.60", "total_tokens": 814800}
-{"current_steps": 120, "total_steps": 190, "loss": 0.0179, "learning_rate": 1.6449496416858285e-06, "epoch": 3.0868167202572345, "percentage": 63.16, "elapsed_time": "0:22:14", "remaining_time": "0:12:58", "throughput": "615.53", "total_tokens": 821536}
-{"current_steps": 121, "total_steps": 190, "loss": 0.0059, "learning_rate": 1.6040801261367494e-06, "epoch": 3.112540192926045, "percentage": 63.68, "elapsed_time": "0:22:25", "remaining_time": "0:12:47", "throughput": "615.35", "total_tokens": 828128}
-{"current_steps": 122, "total_steps": 190, "loss": 0.0017, "learning_rate": 1.56348351646022e-06, "epoch": 3.1382636655948555, "percentage": 64.21, "elapsed_time": "0:22:36", "remaining_time": "0:12:36", "throughput": "615.08", "total_tokens": 834608}
-{"current_steps": 123, "total_steps": 190, "loss": 0.0018, "learning_rate": 1.5231721787768162e-06, "epoch": 3.1639871382636655, "percentage": 64.74, "elapsed_time": "0:22:48", "remaining_time": "0:12:25", "throughput": "615.02", "total_tokens": 841360}
-{"current_steps": 124, "total_steps": 190, "loss": 0.0032, "learning_rate": 1.4831583923105e-06, "epoch": 3.189710610932476, "percentage": 65.26, "elapsed_time": "0:22:59", "remaining_time": "0:12:14", "throughput": "615.23", "total_tokens": 848496}
-{"current_steps": 125, "total_steps": 190, "loss": 0.0019, "learning_rate": 1.443454345648252e-06, "epoch": 3.215434083601286, "percentage": 65.79, "elapsed_time": "0:23:10", "remaining_time": "0:12:02", "throughput": "615.29", "total_tokens": 855424}
-{"current_steps": 126, "total_steps": 190, "loss": 0.0014, "learning_rate": 1.4040721330273063e-06, "epoch": 3.2411575562700965, "percentage": 66.32, "elapsed_time": "0:23:21", "remaining_time": "0:11:51", "throughput": "615.27", "total_tokens": 862224}
-{"current_steps": 127, "total_steps": 190, "loss": 0.0052, "learning_rate": 1.3650237506511333e-06, "epoch": 3.266881028938907, "percentage": 66.84, "elapsed_time": "0:23:32", "remaining_time": "0:11:40", "throughput": "615.45", "total_tokens": 869312}
-{"current_steps": 128, "total_steps": 190, "loss": 0.0005, "learning_rate": 1.3263210930352737e-06, "epoch": 3.292604501607717, "percentage": 67.37, "elapsed_time": "0:23:43", "remaining_time": "0:11:29", "throughput": "615.55", "total_tokens": 876272}
-{"current_steps": 129, "total_steps": 190, "loss": 0.0131, "learning_rate": 1.2879759493841577e-06, "epoch": 3.3183279742765275, "percentage": 67.89, "elapsed_time": "0:23:54", "remaining_time": "0:11:18", "throughput": "615.52", "total_tokens": 883072}
-{"current_steps": 130, "total_steps": 190, "loss": 0.0009, "learning_rate": 1.2500000000000007e-06, "epoch": 3.3440514469453375, "percentage": 68.42, "elapsed_time": "0:24:05", "remaining_time": "0:11:07", "throughput": "615.54", "total_tokens": 889920}
-{"current_steps": 131, "total_steps": 190, "loss": 0.0057, "learning_rate": 1.2124048127248644e-06, "epoch": 3.369774919614148, "percentage": 68.95, "elapsed_time": "0:24:16", "remaining_time": "0:10:56", "throughput": "615.63", "total_tokens": 896896}
-{"current_steps": 132, "total_steps": 190, "loss": 0.0002, "learning_rate": 1.1752018394169882e-06, "epoch": 3.395498392282958, "percentage": 69.47, "elapsed_time": "0:24:27", "remaining_time": "0:10:45", "throughput": "615.53", "total_tokens": 903600}
-{"current_steps": 133, "total_steps": 190, "loss": 0.0002, "learning_rate": 1.1384024124624324e-06, "epoch": 3.4212218649517685, "percentage": 70.0, "elapsed_time": "0:24:39", "remaining_time": "0:10:33", "throughput": "615.37", "total_tokens": 910208}
-{"current_steps": 134, "total_steps": 190, "loss": 0.0145, "learning_rate": 1.1020177413231334e-06, "epoch": 3.446945337620579, "percentage": 70.53, "elapsed_time": "0:24:50", "remaining_time": "0:10:22", "throughput": "615.56", "total_tokens": 917328}
-{"current_steps": 135, "total_steps": 190, "loss": 0.0034, "learning_rate": 1.0660589091223854e-06, "epoch": 3.472668810289389, "percentage": 71.05, "elapsed_time": "0:25:01", "remaining_time": "0:10:11", "throughput": "615.59", "total_tokens": 924192}
-{"current_steps": 136, "total_steps": 190, "loss": 0.0156, "learning_rate": 1.0305368692688175e-06, "epoch": 3.4983922829581995, "percentage": 71.58, "elapsed_time": "0:25:12", "remaining_time": "0:10:00", "throughput": "615.43", "total_tokens": 930784}
-{"current_steps": 137, "total_steps": 190, "loss": 0.0013, "learning_rate": 9.95462442119879e-07, "epoch": 3.5241157556270095, "percentage": 72.11, "elapsed_time": "0:25:23", "remaining_time": "0:09:49", "throughput": "615.59", "total_tokens": 937856}
-{"current_steps": 138, "total_steps": 190, "loss": 0.0007, "learning_rate": 9.608463116858544e-07, "epoch": 3.54983922829582, "percentage": 72.63, "elapsed_time": "0:25:34", "remaining_time": "0:09:38", "throughput": "615.56", "total_tokens": 944640}
-{"current_steps": 139, "total_steps": 190, "loss": 0.0005, "learning_rate": 9.266990223754069e-07, "epoch": 3.57556270096463, "percentage": 73.16, "elapsed_time": "0:25:45", "remaining_time": "0:09:27", "throughput": "615.58", "total_tokens": 951504}
-{"current_steps": 140, "total_steps": 190, "loss": 0.0034, "learning_rate": 8.930309757836517e-07, "epoch": 3.6012861736334405, "percentage": 73.68, "elapsed_time": "0:25:56", "remaining_time": "0:09:16", "throughput": "615.52", "total_tokens": 958240}
-{"current_steps": 141, "total_steps": 190, "loss": 0.0001, "learning_rate": 8.598524275237321e-07, "epoch": 3.627009646302251, "percentage": 74.21, "elapsed_time": "0:26:07", "remaining_time": "0:09:04", "throughput": "615.41", "total_tokens": 964912}
-{"current_steps": 142, "total_steps": 190, "loss": 0.001, "learning_rate": 8.271734841028553e-07, "epoch": 3.652733118971061, "percentage": 74.74, "elapsed_time": "0:26:19", "remaining_time": "0:08:53", "throughput": "615.49", "total_tokens": 971872}
-{"current_steps": 143, "total_steps": 190, "loss": 0.0123, "learning_rate": 7.950040998437541e-07, "epoch": 3.6784565916398715, "percentage": 75.26, "elapsed_time": "0:26:30", "remaining_time": "0:08:42", "throughput": "615.44", "total_tokens": 978640}
-{"current_steps": 144, "total_steps": 190, "loss": 0.0002, "learning_rate": 7.633540738525066e-07, "epoch": 3.7041800643086815, "percentage": 75.79, "elapsed_time": "0:26:41", "remaining_time": "0:08:31", "throughput": "615.35", "total_tokens": 985328}
-{"current_steps": 145, "total_steps": 190, "loss": 0.011, "learning_rate": 7.322330470336314e-07, "epoch": 3.729903536977492, "percentage": 76.32, "elapsed_time": "0:26:52", "remaining_time": "0:08:20", "throughput": "615.39", "total_tokens": 992224}
-{"current_steps": 146, "total_steps": 190, "loss": 0.0008, "learning_rate": 7.016504991533727e-07, "epoch": 3.755627009646302, "percentage": 76.84, "elapsed_time": "0:27:03", "remaining_time": "0:08:09", "throughput": "615.17", "total_tokens": 998688}
-{"current_steps": 147, "total_steps": 190, "loss": 0.0003, "learning_rate": 6.716157459520739e-07, "epoch": 3.7813504823151125, "percentage": 77.37, "elapsed_time": "0:27:14", "remaining_time": "0:07:58", "throughput": "615.50", "total_tokens": 1006032}
-{"current_steps": 148, "total_steps": 190, "loss": 0.0018, "learning_rate": 6.421379363065142e-07, "epoch": 3.807073954983923, "percentage": 77.89, "elapsed_time": "0:27:25", "remaining_time": "0:07:46", "throughput": "615.55", "total_tokens": 1012944}
-{"current_steps": 149, "total_steps": 190, "loss": 0.0016, "learning_rate": 6.1322604944307e-07, "epoch": 3.832797427652733, "percentage": 78.42, "elapsed_time": "0:27:36", "remaining_time": "0:07:35", "throughput": "615.59", "total_tokens": 1019856}
-{"current_steps": 150, "total_steps": 190, "loss": 0.0021, "learning_rate": 5.848888922025553e-07, "epoch": 3.8585209003215435, "percentage": 78.95, "elapsed_time": "0:27:47", "remaining_time": "0:07:24", "throughput": "615.56", "total_tokens": 1026656}
-{"current_steps": 151, "total_steps": 190, "loss": 0.0001, "learning_rate": 5.571350963575728e-07, "epoch": 3.884244372990354, "percentage": 79.47, "elapsed_time": "0:27:58", "remaining_time": "0:07:13", "throughput": "615.68", "total_tokens": 1033696}
-{"current_steps": 152, "total_steps": 190, "loss": 0.0001, "learning_rate": 5.299731159831953e-07, "epoch": 3.909967845659164, "percentage": 80.0, "elapsed_time": "0:28:10", "remaining_time": "0:07:02", "throughput": "615.50", "total_tokens": 1040240}
-{"current_steps": 153, "total_steps": 190, "loss": 0.0003, "learning_rate": 5.034112248817685e-07, "epoch": 3.935691318327974, "percentage": 80.53, "elapsed_time": "0:28:21", "remaining_time": "0:06:51", "throughput": "615.45", "total_tokens": 1046992}
-{"current_steps": 154, "total_steps": 190, "loss": 0.0002, "learning_rate": 4.774575140626317e-07, "epoch": 3.9614147909967845, "percentage": 81.05, "elapsed_time": "0:28:32", "remaining_time": "0:06:40", "throughput": "615.52", "total_tokens": 1053936}
-{"current_steps": 155, "total_steps": 190, "loss": 0.0002, "learning_rate": 4.5211988927752026e-07, "epoch": 3.987138263665595, "percentage": 81.58, "elapsed_time": "0:28:43", "remaining_time": "0:06:29", "throughput": "615.41", "total_tokens": 1060576}
-{"current_steps": 156, "total_steps": 190, "loss": 0.0, "learning_rate": 4.27406068612396e-07, "epoch": 4.012861736334405, "percentage": 82.11, "elapsed_time": "0:28:54", "remaining_time": "0:06:18", "throughput": "615.56", "total_tokens": 1067648}
-{"current_steps": 157, "total_steps": 190, "loss": 0.0003, "learning_rate": 4.033235801364402e-07, "epoch": 4.038585209003215, "percentage": 82.63, "elapsed_time": "0:29:05", "remaining_time": "0:06:06", "throughput": "615.58", "total_tokens": 1074512}
-{"current_steps": 158, "total_steps": 190, "loss": 0.0002, "learning_rate": 3.798797596089351e-07, "epoch": 4.064308681672026, "percentage": 83.16, "elapsed_time": "0:29:16", "remaining_time": "0:05:55", "throughput": "615.55", "total_tokens": 1081296}
-{"current_steps": 159, "total_steps": 190, "loss": 0.0001, "learning_rate": 3.5708174824471947e-07, "epoch": 4.090032154340836, "percentage": 83.68, "elapsed_time": "0:29:27", "remaining_time": "0:05:44", "throughput": "615.40", "total_tokens": 1087888}
-{"current_steps": 160, "total_steps": 190, "loss": 0.0001, "learning_rate": 3.3493649053890325e-07, "epoch": 4.115755627009646, "percentage": 84.21, "elapsed_time": "0:29:38", "remaining_time": "0:05:33", "throughput": "615.53", "total_tokens": 1094960}
-{"current_steps": 161, "total_steps": 190, "loss": 0.0001, "learning_rate": 3.134507321515107e-07, "epoch": 4.141479099678457, "percentage": 84.74, "elapsed_time": "0:29:50", "remaining_time": "0:05:22", "throughput": "615.52", "total_tokens": 1101776}
-{"current_steps": 162, "total_steps": 190, "loss": 0.0, "learning_rate": 2.9263101785268253e-07, "epoch": 4.167202572347267, "percentage": 85.26, "elapsed_time": "0:30:01", "remaining_time": "0:05:11", "throughput": "615.59", "total_tokens": 1108736}
-{"current_steps": 163, "total_steps": 190, "loss": 0.0001, "learning_rate": 2.7248368952908055e-07, "epoch": 4.192926045016077, "percentage": 85.79, "elapsed_time": "0:30:12", "remaining_time": "0:05:00", "throughput": "615.69", "total_tokens": 1115744}
-{"current_steps": 164, "total_steps": 190, "loss": 0.0, "learning_rate": 2.53014884252083e-07, "epoch": 4.218649517684887, "percentage": 86.32, "elapsed_time": "0:30:23", "remaining_time": "0:04:49", "throughput": "615.71", "total_tokens": 1122592}
-{"current_steps": 165, "total_steps": 190, "loss": 0.0001, "learning_rate": 2.3423053240837518e-07, "epoch": 4.244372990353698, "percentage": 86.84, "elapsed_time": "0:30:34", "remaining_time": "0:04:37", "throughput": "615.55", "total_tokens": 1129136}
-{"current_steps": 166, "total_steps": 190, "loss": 0.0001, "learning_rate": 2.1613635589349756e-07, "epoch": 4.270096463022508, "percentage": 87.37, "elapsed_time": "0:30:45", "remaining_time": "0:04:26", "throughput": "615.61", "total_tokens": 1136080}
-{"current_steps": 167, "total_steps": 190, "loss": 0.0002, "learning_rate": 1.9873786636889908e-07, "epoch": 4.295819935691318, "percentage": 87.89, "elapsed_time": "0:30:56", "remaining_time": "0:04:15", "throughput": "615.64", "total_tokens": 1142976}
-{"current_steps": 168, "total_steps": 190, "loss": 0.0002, "learning_rate": 1.8204036358303173e-07, "epoch": 4.321543408360129, "percentage": 88.42, "elapsed_time": "0:31:07", "remaining_time": "0:04:04", "throughput": "615.58", "total_tokens": 1149712}
-{"current_steps": 169, "total_steps": 190, "loss": 0.0001, "learning_rate": 1.6604893375699594e-07, "epoch": 4.347266881028939, "percentage": 88.95, "elapsed_time": "0:31:18", "remaining_time": "0:03:53", "throughput": "615.46", "total_tokens": 1156336}
-{"current_steps": 170, "total_steps": 190, "loss": 0.0001, "learning_rate": 1.507684480352292e-07, "epoch": 4.372990353697749, "percentage": 89.47, "elapsed_time": "0:31:29", "remaining_time": "0:03:42", "throughput": "615.43", "total_tokens": 1163120}
-{"current_steps": 171, "total_steps": 190, "loss": 0.0115, "learning_rate": 1.362035610017079e-07, "epoch": 4.39871382636656, "percentage": 90.0, "elapsed_time": "0:31:41", "remaining_time": "0:03:31", "throughput": "615.48", "total_tokens": 1170032}
-{"current_steps": 172, "total_steps": 190, "loss": 0.0005, "learning_rate": 1.223587092621162e-07, "epoch": 4.42443729903537, "percentage": 90.53, "elapsed_time": "0:31:52", "remaining_time": "0:03:20", "throughput": "615.47", "total_tokens": 1176832}
-{"current_steps": 173, "total_steps": 190, "loss": 0.0003, "learning_rate": 1.0923811009241142e-07, "epoch": 4.45016077170418, "percentage": 91.05, "elapsed_time": "0:32:03", "remaining_time": "0:03:08", "throughput": "615.57", "total_tokens": 1183856}
-{"current_steps": 174, "total_steps": 190, "loss": 0.005, "learning_rate": 9.684576015420277e-08, "epoch": 4.47588424437299, "percentage": 91.58, "elapsed_time": "0:32:14", "remaining_time": "0:02:57", "throughput": "615.49", "total_tokens": 1190544}
-{"current_steps": 175, "total_steps": 190, "loss": 0.0003, "learning_rate": 8.518543427732951e-08, "epoch": 4.501607717041801, "percentage": 92.11, "elapsed_time": "0:32:25", "remaining_time": "0:02:46", "throughput": "615.38", "total_tokens": 1197168}
-{"current_steps": 176, "total_steps": 190, "loss": 0.0008, "learning_rate": 7.426068431000883e-08, "epoch": 4.527331189710611, "percentage": 92.63, "elapsed_time": "0:32:36", "remaining_time": "0:02:35", "throughput": "615.27", "total_tokens": 1203776}
-{"current_steps": 177, "total_steps": 190, "loss": 0.0, "learning_rate": 6.407483803691216e-08, "epoch": 4.553054662379421, "percentage": 93.16, "elapsed_time": "0:32:47", "remaining_time": "0:02:24", "throughput": "615.37", "total_tokens": 1210816}
-{"current_steps": 178, "total_steps": 190, "loss": 0.0, "learning_rate": 5.463099816548578e-08, "epoch": 4.578778135048232, "percentage": 93.68, "elapsed_time": "0:32:58", "remaining_time": "0:02:13", "throughput": "615.39", "total_tokens": 1217696}
-{"current_steps": 179, "total_steps": 190, "loss": 0.0042, "learning_rate": 4.593204138084006e-08, "epoch": 4.604501607717042, "percentage": 94.21, "elapsed_time": "0:33:09", "remaining_time": "0:02:02", "throughput": "615.48", "total_tokens": 1224704}
-{"current_steps": 180, "total_steps": 190, "loss": 0.0004, "learning_rate": 3.798061746947995e-08, "epoch": 4.630225080385852, "percentage": 94.74, "elapsed_time": "0:33:20", "remaining_time": "0:01:51", "throughput": "615.54", "total_tokens": 1231664}
-{"current_steps": 181, "total_steps": 190, "loss": 0.0001, "learning_rate": 3.077914851215585e-08, "epoch": 4.655948553054662, "percentage": 95.26, "elapsed_time": "0:33:32", "remaining_time": "0:01:40", "throughput": "615.42", "total_tokens": 1238240}
-{"current_steps": 182, "total_steps": 190, "loss": 0.0001, "learning_rate": 2.4329828146074096e-08, "epoch": 4.681672025723473, "percentage": 95.79, "elapsed_time": "0:33:43", "remaining_time": "0:01:28", "throughput": "615.30", "total_tokens": 1244832}
-{"current_steps": 183, "total_steps": 190, "loss": 0.0004, "learning_rate": 1.8634620896695044e-08, "epoch": 4.707395498392283, "percentage": 96.32, "elapsed_time": "0:33:54", "remaining_time": "0:01:17", "throughput": "615.12", "total_tokens": 1251296}
-{"current_steps": 184, "total_steps": 190, "loss": 0.0, "learning_rate": 1.3695261579316776e-08, "epoch": 4.733118971061093, "percentage": 96.84, "elapsed_time": "0:34:05", "remaining_time": "0:01:06", "throughput": "615.05", "total_tokens": 1257968}
-{"current_steps": 185, "total_steps": 190, "loss": 0.0009, "learning_rate": 9.513254770636138e-09, "epoch": 4.758842443729904, "percentage": 97.37, "elapsed_time": "0:34:16", "remaining_time": "0:00:55", "throughput": "615.11", "total_tokens": 1264928}
-{"current_steps": 186, "total_steps": 190, "loss": 0.0001, "learning_rate": 6.089874350439507e-09, "epoch": 4.784565916398714, "percentage": 97.89, "elapsed_time": "0:34:27", "remaining_time": "0:00:44", "throughput": "615.12", "total_tokens": 1271792}
-{"current_steps": 187, "total_steps": 190, "loss": 0.0004, "learning_rate": 3.4261631135654174e-09, "epoch": 4.810289389067524, "percentage": 98.42, "elapsed_time": "0:34:38", "remaining_time": "0:00:33", "throughput": "615.30", "total_tokens": 1279024}
-{"current_steps": 188, "total_steps": 190, "loss": 0.0002, "learning_rate": 1.5229324522605949e-09, "epoch": 4.836012861736334, "percentage": 98.95, "elapsed_time": "0:34:49", "remaining_time": "0:00:22", "throughput": "615.28", "total_tokens": 1285792}
-{"current_steps": 189, "total_steps": 190, "loss": 0.0, "learning_rate": 3.8076210902182607e-10, "epoch": 4.861736334405145, "percentage": 99.47, "elapsed_time": "0:35:00", "remaining_time": "0:00:11", "throughput": "615.26", "total_tokens": 1292576}
-{"current_steps": 190, "total_steps": 190, "loss": 0.0001, "learning_rate": 0.0, "epoch": 4.887459807073955, "percentage": 100.0, "elapsed_time": "0:35:11", "remaining_time": "0:00:00", "throughput": "615.25", "total_tokens": 1299392}
-{"current_steps": 190, "total_steps": 190, "epoch": 4.887459807073955, "percentage": 100.0, "elapsed_time": "0:36:02", "remaining_time": "0:00:00", "throughput": "600.99", "total_tokens": 1299392}

+{"current_steps": 5, "total_steps": 78, "percentage": 6.41, "elapsed_time": "0:00:00", "remaining_time": "0:00:05"}
+{"current_steps": 10, "total_steps": 78, "percentage": 12.82, "elapsed_time": "0:00:00", "remaining_time": "0:00:05"}
+{"current_steps": 15, "total_steps": 78, "percentage": 19.23, "elapsed_time": "0:00:01", "remaining_time": "0:00:05"}
+{"current_steps": 20, "total_steps": 78, "percentage": 25.64, "elapsed_time": "0:00:01", "remaining_time": "0:00:05"}
+{"current_steps": 25, "total_steps": 78, "percentage": 32.05, "elapsed_time": "0:00:02", "remaining_time": "0:00:04"}
+{"current_steps": 30, "total_steps": 78, "percentage": 38.46, "elapsed_time": "0:00:02", "remaining_time": "0:00:04"}
+{"current_steps": 35, "total_steps": 78, "percentage": 44.87, "elapsed_time": "0:00:03", "remaining_time": "0:00:03"}
+{"current_steps": 40, "total_steps": 78, "percentage": 51.28, "elapsed_time": "0:00:03", "remaining_time": "0:00:03"}
+{"current_steps": 45, "total_steps": 78, "percentage": 57.69, "elapsed_time": "0:00:04", "remaining_time": "0:00:03"}
+{"current_steps": 50, "total_steps": 78, "percentage": 64.1, "elapsed_time": "0:00:04", "remaining_time": "0:00:02"}
+{"current_steps": 55, "total_steps": 78, "percentage": 70.51, "elapsed_time": "0:00:05", "remaining_time": "0:00:02"}
+{"current_steps": 60, "total_steps": 78, "percentage": 76.92, "elapsed_time": "0:00:05", "remaining_time": "0:00:01"}
+{"current_steps": 65, "total_steps": 78, "percentage": 83.33, "elapsed_time": "0:00:06", "remaining_time": "0:00:01"}
+{"current_steps": 70, "total_steps": 78, "percentage": 89.74, "elapsed_time": "0:00:06", "remaining_time": "0:00:00"}
+{"current_steps": 75, "total_steps": 78, "percentage": 96.15, "elapsed_time": "0:00:06", "remaining_time": "0:00:00"}

training_args.yaml CHANGED Viewed

@@ -1,30 +1,18 @@
-bf16: true
 cutoff_len: 1024
-dataset: truth_train_0716
 dataset_dir: data
-ddp_timeout: 180000000
-deepspeed: cache/ds_z2_config.json
-do_train: true
 finetuning_type: full
 flash_attn: auto
-gradient_accumulation_steps: 8
-include_num_input_tokens_seen: true
-learning_rate: 5.0e-06
-logging_steps: 1
-lr_scheduler_type: cosine
-max_grad_norm: 1.0
 max_samples: 100000
-model_name_or_path: meta-llama/Llama-2-7b-chat-hf
-num_train_epochs: 5.0
-optim: adamw_torch
-output_dir: saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2
-packing: false
-per_device_train_batch_size: 2
-plot_loss: true
 preprocessing_num_workers: 16
 quantization_method: bitsandbytes
-report_to: none
-save_steps: 1000
 stage: sft
 template: llama2
-warmup_steps: 10

 cutoff_len: 1024
+dataset: truth_dev_0716
 dataset_dir: data
+do_predict: true
 finetuning_type: full
 flash_attn: auto
+max_new_tokens: 512
 max_samples: 100000
+model_name_or_path: saves/LLaMA2-7B-Chat/full/train_2024-07-16-09-05-28_llama2
+output_dir: saves/LLaMA2-7B-Chat/full/eval_2024-07-16-09-05-28
+per_device_eval_batch_size: 2
+predict_with_generate: true
 preprocessing_num_workers: 16
 quantization_method: bitsandbytes
 stage: sft
+temperature: 0.95
 template: llama2
+top_p: 0.7