Training in progress, epoch 1

eval_job_output.txt

@@ -1,4 +1,4 @@
-slurm submission log: 2024-05-09 15:03:34.490765
+slurm submission log: 2024-05-09 22:15:12.853075
 created following sbatch script: 
 
 ###############################
@@ -7,13 +7,13 @@ created following sbatch script:
 
 #SBATCH --account=nlp
 #SBATCH --cpus-per-task=16
-#SBATCH --dependency=afterok:7592325
+#SBATCH --dependency=afterok:7593078
 #SBATCH --gres=gpu:1
-#SBATCH --job-name=tthrush-job-707129
+#SBATCH --job-name=tthrush-job-4451503
 #SBATCH --mem=60G
 #SBATCH --nodelist=sphinx2
 #SBATCH --open-mode=append
-#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_2/pythia-70m_lambada/eval_job_output.txt
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada/eval_job_output.txt
 #SBATCH --partition=sphinx
 #SBATCH --time=14-0
 
@@ -24,7 +24,7 @@ created following sbatch script:
 cd .
 
 # launch commands
-srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_2/pythia-70m_lambada,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_2/pythia-70m_lambada/perf'
+srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada/perf'
 
 ###############################
 
@@ -34,7 +34,133 @@ submission to slurm complete!
 ###############################
 slurm submission output
 
-Submitted batch job 7592326
+Submitted batch job 7593079
+
+
+
+###############################
+
+slurm submission log: 2024-05-09 23:03:15.777721
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7593156
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-1291558
+#SBATCH --mem=60G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada/eval_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada/perf'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7593157
+
+
+
+###############################
+
+slurm submission log: 2024-05-10 08:21:54.596836
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7593613
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-4024076
+#SBATCH --mem=60G
+#SBATCH --nodelist=sphinx1
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada/eval_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada/perf'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7593614
+
+
+
+###############################
+
+slurm submission log: 2024-05-10 08:23:20.760002
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7593626
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-2274793
+#SBATCH --mem=60G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada/eval_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'lm_eval --model hf --model_args pretrained=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada,revision=main,dtype=float16,trust_remote_code=True --tasks xnli_en,xnli_fr,sciq,piqa,lambada,arc_easy --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada/perf'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7593627
 
 
 

logs/events.out.tfevents.1715379123.sphinx2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4698683d4f74bd9401eafd13ebbde448a7459409fe0fb8bc5de5f97584a07da0
+size 11521

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a06cdaa84294a3863ee36854edc2ef42c43d3e9d9952576d141ad2a327e355c2
+oid sha256:4382129d8fe1cb8ab29623ad0dd185faaff0891d845fd3019824795e3de47445
 size 281715176

train_job_o

@@ -0,0 +1,84 @@
+slurm submission log: 2024-05-09 22:15:12.576191
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7593077
+#SBATCH --gres=gpu:2
+#SBATCH --job-name=tthrush-job-1473981
+#SBATCH --mem=400G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada/train_job_o
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'utput.txt' 'torchrun --master_port 29504 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_3/lambada --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada --o    utput_hub_id pythia-70m_lambada --model_id EleutherAI/pythia-70m --num_train_epochs 1 --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7593078
+
+
+
+###############################
+
+slurm submission log: 2024-05-09 23:03:15.492375
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7593155
+#SBATCH --gres=gpu:2
+#SBATCH --job-name=tthrush-job-2751890
+#SBATCH --mem=400G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada/train_job_o
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'utput.txt' 'torchrun --master_port 29504 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/train_data_3/lambada --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/llms_3/pythia-70m_lambada --o    utput_hub_id pythia-70m_lambada --model_id EleutherAI/pythia-70m --num_train_epochs 1 --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7593156
+
+
+
+###############################
+

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec3c176db2130799a9567203f329d022f06795f8c51369fa4dd4c083e05ea6d8
+oid sha256:ed33f0523d3b8c8997f046572fbd844d1d7a5a95369e48dffa870aab9131ffc6
 size 5112