Training in progress, epoch 0

config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "EleutherAI/pythia-70m",
+  "architectures": [
+    "GPTNeoXForCausalLM"
+  ],
+  "attention_bias": true,
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classifier_dropout": 0.1,
+  "eos_token_id": 0,
+  "hidden_act": "gelu",
+  "hidden_dropout": 0.0,
+  "hidden_size": 512,
+  "initializer_range": 0.02,
+  "intermediate_size": 2048,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neox",
+  "num_attention_heads": 8,
+  "num_hidden_layers": 6,
+  "rope_scaling": null,
+  "rotary_emb_base": 10000,
+  "rotary_pct": 0.25,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.2",
+  "use_cache": true,
+  "use_parallel_residual": true,
+  "vocab_size": 50304
+}

eval_job_output.txt

@@ -0,0 +1,210 @@
+slurm submission log: 2024-05-25 22:01:15.907170
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7651386
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-3177077
+#SBATCH --mem=60G
+#SBATCH --nodelist=sphinx1
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/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/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1,revision=main,dtype=float16,trust_remote_code=True --tasks piqa,arc_easy,xnli_en,xnli_fr,xnli_de,xnli_es,sciq,lambada --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/perf'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7651387
+
+
+
+###############################
+
+slurm submission log: 2024-05-25 22:02:25.155540
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7651417
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-3723367
+#SBATCH --mem=60G
+#SBATCH --nodelist=sphinx1
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/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/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1,revision=main,dtype=float16,trust_remote_code=True --tasks piqa,arc_easy,xnli_en,xnli_fr,xnli_de,xnli_es,sciq,lambada --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/perf'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7651418
+
+
+
+###############################
+
+slurm submission log: 2024-05-25 22:12:50.394186
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7651458
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-3832230
+#SBATCH --mem=60G
+#SBATCH --nodelist=sphinx1
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/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/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1,revision=main,dtype=float16,trust_remote_code=True --tasks piqa,arc_easy,xnli_en,xnli_fr,xnli_de,xnli_es,sciq,lambada --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/perf'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7651459
+
+
+
+###############################
+
+slurm submission log: 2024-05-25 22:15:55.997250
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7651486
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-2732481
+#SBATCH --mem=60G
+#SBATCH --nodelist=sphinx1
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/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/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1,revision=main,dtype=float16,trust_remote_code=True --tasks piqa,arc_easy,xnli_en,xnli_fr,xnli_de,xnli_es,sciq,lambada --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/perf'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7651487
+
+
+
+###############################
+
+slurm submission log: 2024-05-25 22:18:14.995526
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7651516
+#SBATCH --gres=gpu:1
+#SBATCH --job-name=tthrush-job-3308982
+#SBATCH --mem=60G
+#SBATCH --nodelist=sphinx1
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/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/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1,revision=main,dtype=float16,trust_remote_code=True --tasks piqa,arc_easy,xnli_en,xnli_fr,xnli_de,xnli_es,sciq,lambada --device cuda --output_path /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/perf'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7651517
+
+
+
+###############################
+

logs/events.out.tfevents.1716738249.sphinx2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11e15a6f6500f0062c4964dc38d6af748a94f95f7df5aeabc9657e2199256da5
+size 11781

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae13490cbf98ca8da07c69d1d78ddec321a71452b204a7e4e6c442f454038800
+size 281715176

train_job_output.txt

@@ -0,0 +1,287 @@
+slurm submission log: 2024-05-25 22:01:15.607519
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7651385
+#SBATCH --gres=gpu:2
+#SBATCH --job-name=tthrush-job-339895
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29508 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/data/xnli_es --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1 --output_hub_id pythia-70m_xnli_es --model_id EleutherAI/pythia-70m --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 14'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7651386
+
+
+
+###############################
+
+slurm submission log: 2024-05-25 22:02:24.879435
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7651416
+#SBATCH --gres=gpu:2
+#SBATCH --job-name=tthrush-job-1467582
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29508 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/data/xnli_es --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1 --output_hub_id pythia-70m_xnli_es --model_id EleutherAI/pythia-70m --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 14'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7651417
+
+
+
+###############################
+
+slurm submission log: 2024-05-25 22:12:50.153381
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7651457
+#SBATCH --gres=gpu:2
+#SBATCH --job-name=tthrush-job-1098424
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29508 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/data/xnli_es --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1 --output_hub_id pythia-70m_xnli_es --model_id EleutherAI/pythia-70m --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 14'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7651458
+
+
+
+###############################
+
+slurm submission log: 2024-05-25 22:15:55.753417
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7651485
+#SBATCH --gres=gpu:2
+#SBATCH --job-name=tthrush-job-3438810
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29508 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/data/xnli_es --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1 --output_hub_id pythia-70m_xnli_es --model_id EleutherAI/pythia-70m --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 14'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7651486
+
+
+
+###############################
+
+slurm submission log: 2024-05-25 22:18:14.763422
+created following sbatch script: 
+
+###############################
+
+#!/bin/bash
+
+#SBATCH --account=nlp
+#SBATCH --cpus-per-task=16
+#SBATCH --dependency=afterok:7651515
+#SBATCH --gres=gpu:2
+#SBATCH --job-name=tthrush-job-3850774
+#SBATCH --mem=100G
+#SBATCH --nodelist=sphinx2
+#SBATCH --open-mode=append
+#SBATCH --output=/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1/train_job_output.txt
+#SBATCH --partition=sphinx
+#SBATCH --time=14-0
+
+# activate your desired anaconda environment
+. /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh ; conda activate pretraining-coreset-selection
+
+# cd to working directory
+cd .
+
+# launch commands
+srun --unbuffered run_as_child_processes 'torchrun --master_port 29508 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/data/xnli_es --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1 --output_hub_id pythia-70m_xnli_es --model_id EleutherAI/pythia-70m --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 14'
+
+###############################
+
+submission to slurm complete!
+
+
+###############################
+slurm submission output
+
+Submitted batch job 7651516
+
+
+
+###############################
+
+/var/lib/slurm/slurmd/job7651516/slurm_script: line 16: /nlp/scr/tthrush/miniconda3/envs/pretraining-coreset-selection/etc/profile.d/conda.sh: No such file or directory
+
+CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'.
+To initialize your shell, run
+
+    $ conda init <SHELL_NAME>
+
+Currently supported shells are:
+  - bash
+  - fish
+  - tcsh
+  - xonsh
+  - zsh
+  - powershell
+
+See 'conda init --help' for more information and options.
+
+IMPORTANT: You may need to close and restart your shell after running 'conda init'.
+
+
+###############################
+start time: 2024-05-26 08:43:47.611801
+machine: sphinx2
+conda env: pretraining-coreset-selection
+###############################
+running following processes
+
+	torchrun --master_port 29508 --nproc_per_node=2 train_llm.py --dataset_id /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/data/xnli_es --output_dir /juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1 --output_hub_id pythia-70m_xnli_es --model_id EleutherAI/pythia-70m --learning_rate 1e-3 --warmup_ratio=0.1 --gradient_accumulation_steps 2 --per_device_train_batch_size 256 --seed 1 --num_train_epochs 14
+
+
+###############################
+command outputs: 
+
+
+[2024-05-26 08:43:51,176] torch.distributed.run: [WARNING] 
+[2024-05-26 08:43:51,176] torch.distributed.run: [WARNING] *****************************************
+[2024-05-26 08:43:51,176] torch.distributed.run: [WARNING] Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 
+[2024-05-26 08:43:51,176] torch.distributed.run: [WARNING] *****************************************
+05/26/2024 08:44:01 - INFO - __main__ - Script parameters ScriptArguments(seed=1, dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/data/xnli_es', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1', output_hub_id='pythia-70m_xnli_es', hf_hub_token=True, model_id='EleutherAI/pythia-70m', per_device_train_batch_size=256, num_train_epochs=14.0, learning_rate=0.001, gradient_accumulation_steps=2, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
+05/26/2024 08:44:06 - INFO - __main__ - Script parameters ScriptArguments(seed=1, dataset_id='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/data/xnli_es', output_dir='/juice5/scr5/tthrush/pretraining-coreset-selection/llm_pretraining/test_ordinal_constrained_big_upsample/llms/pythia-70m_xnli_es_1', output_hub_id='pythia-70m_xnli_es', hf_hub_token=True, model_id='EleutherAI/pythia-70m', per_device_train_batch_size=256, num_train_epochs=14.0, learning_rate=0.001, gradient_accumulation_steps=2, from_scratch=True, warmup_ratio=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, weight_decay=0.01, lr_scheduler_type='cosine', local_rank=0, resume_from_checkpoint=False, deepspeed=None, peft=False)
+
  0%|          | 0/11704 [00:00<?, ?it/s][rank1]:[W reducer.cpp:1360] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
+[rank0]:[W reducer.cpp:1360] Warning: find_unused_parameters=True was specified in DDP constructor, but did not find any unused parameters in the forward pass. This flag results in an extra traversal of the autograd graph every iteration,  which can adversely affect performance. If your model indeed never has any unused parameters in the forward pass, consider turning this flag off. Note that this warning may be a false positive if your model has flow control causing later iterations to have unused parameters. (function operator())
+
  0%|          | 1/11704 [00:11<36:01:50, 11.08s/it]
  0%|          | 2/11704 [00:15<22:29:09,  6.92s/it]
  0%|          | 3/11704 [00:18<17:13:49,  5.30s/it]
  0%|          | 4/11704 [00:21<13:50:30,  4.26s/it]
  0%|          | 5/11704 [00:22<10:32:47,  3.25s/it]
  0%|          | 6/11704 [00:24<9:19:43,  2.87s/it] 
  0%|          | 7/11704 [00:26<7:44:55,  2.38s/it]
  0%|          | 8/11704 [00:27<6:35:35,  2.03s/it]
  0%|          | 9/11704 [00:28<5:32:02,  1.70s/it]
  0%|          | 10/11704 [00:29<5:12:33,  1.60s/it]
  0%|          | 11/11704 [00:30<4:43:07,  1.45s/it]
  0%|          | 12/11704 [00:31<4:09:08,  1.28s/it]
  0%|          | 13/11704 [00:32<3:36:18,  1.11s/it]
  0%|          | 14/11704 [00:33<3:17:27,  1.01s/it]
  0%|          | 15/11704 [00:34<3:11:14,  1.02it/s]
  0%|          | 16/11704 [00:34<3:01:30,  1.07it/s]
  0%|          | 17/11704 [00:35<2:48:52,  1.15it/s]
  0%|          | 18/11704 [00:36<2:33:34,  1.27it/s]
  0%|          | 19/11704 [00:36<2:26:24,  1.33it/s]
  0%|          | 20/11704 [00:37<2:19:52,  1.39it/s]
  0%|          | 21/11704 [00:38<2:16:43,  1.42it/s]
  0%|          | 22/11704 [00:38<2:16:47,  1.42it/s]
  0%|          | 23/11704 [00:39<2:13:38,  1.46it/s]
  0%|          | 24/11704 [00:40<2:20:53,  1.38it/s]
  0%|          | 25/11704 [00:41<2:25:03,  1.34it/s]{'loss': 10.6948, 'grad_norm': 1.2803484201431274, 'learning_rate': 2.134927412467976e-05, 'epoch': 0.03}
                                                    
+
  0%|          | 25/11704 [00:41<2:25:03,  1.34it/s]
  0%|          | 26/11704 [00:41<2:14:46,  1.44it/s]
  0%|          | 27/11704 [00:42<2:08:18,  1.52it/s]
  0%|          | 28/11704 [00:42<2:06:17,  1.54it/s]
  0%|          | 29/11704 [00:43<2:03:23,  1.58it/s]
  0%|          | 30/11704 [00:44<1:58:17,  1.64it/s]
  0%|          | 31/11704 [00:44<1:53:07,  1.72it/s]
  0%|          | 32/11704 [00:45<1:50:58,  1.75it/s]
  0%|          | 33/11704 [00:45<1:50:04,  1.77it/s]
  0%|          | 34/11704 [00:46<1:51:19,  1.75it/s]
  0%|          | 35/11704 [00:46<1:47:30,  1.81it/s]
  0%|          | 36/11704 [00:47<1:47:57,  1.80it/s]
  0%|          | 37/11704 [00:47<1:46:41,  1.82it/s]
  0%|          | 38/11704 [00:48<1:45:56,  1.84it/s]
  0%|          | 39/11704 [00:49<1:47:28,  1.81it/s]
  0%|          | 40/11704 [00:49<1:46:05,  1.83it/s]
  0%|          | 41/11704 [00:50<1:45:06,  1.85it/s]
  0%|          | 42/11704 [00:50<1:45:14,  1.85it/s]
  0%|          | 43/11704 [00:51<1:46:16,  1.83it/s]
  0%|          | 44/11704 [00:51<1:44:48,  1.85it/s]
  0%|          | 45/11704 [00:52<1:42:40,  1.89it/s]
  0%|          | 46/11704 [00:52<1:43:30,  1.88it/s]
  0%|          | 47/11704 [00:53<1:43:45,  1.87it/s]
  0%|          | 48/11704 [00:53<1:42:51,  1.89it/s]
  0%|          | 49/11704 [00:54<1:42:14,  1.90it/s]
  0%|          | 50/11704 [00:54<1:41:41,  1.91it/s]{'loss': 9.9789, 'grad_norm': 1.1821939945220947, 'learning_rate': 4.269854824935952e-05, 'epoch': 0.06}
+
                                                    
  0%|          | 50/11704 [00:54<1:41:41,  1.91it/s]
  0%|          | 51/11704 [00:55<1:41:01,  1.92it/s]
  0%|          | 52/11704 [00:55<1:40:15,  1.94it/s]
  0%|          | 53/11704 [00:56<1:39:39,  1.95it/s]
  0%|          | 54/11704 [00:56<1:40:20,  1.94it/s]
  0%|          | 55/11704 [00:57<1:39:05,  1.96it/s]
  0%|          | 56/11704 [00:57<1:39:10,  1.96it/s]
  0%|          | 57/11704 [00:58<1:39:27,  1.95it/s]
  0%|          | 58/11704 [00:58<1:39:40,  1.95it/s]
  1%|          | 59/11704 [00:59<1:52:14,  1.73it/s]
  1%|          | 60/11704 [01:00<1:48:03,  1.80it/s]
  1%|          | 61/11704 [01:00<1:45:25,  1.84it/s]
  1%|          | 62/11704 [01:01<1:43:05,  1.88it/s]
  1%|          | 63/11704 [01:01<1:41:38,  1.91it/s]
  1%|          | 64/11704 [01:02<1:40:08,  1.94it/s]
  1%|          | 65/11704 [01:02<1:39:38,  1.95it/s]
  1%|          | 66/11704 [01:03<1:38:25,  1.97it/s]
  1%|          | 67/11704 [01:03<1:38:36,  1.97it/s]
  1%|          | 68/11704 [01:04<1:38:36,  1.97it/s]
  1%|          | 69/11704 [01:04<1:37:32,  1.99it/s]
  1%|          | 70/11704 [01:05<1:37:04,  2.00it/s]
  1%|          | 71/11704 [01:05<1:36:37,  2.01it/s]
  1%|          | 72/11704 [01:06<1:36:46,  2.00it/s]
  1%|          | 73/11704 [01:06<1:36:54,  2.00it/s]
  1%|          | 74/11704 [01:07<1:37:03,  2.00it/s]
  1%|          | 75/11704 [01:07<1:36:35,  2.01it/s]{'loss': 9.2071, 'grad_norm': 1.037238597869873, 'learning_rate': 6.404782237403927e-05, 'epoch': 0.09}
                                                    
+
  1%|          | 75/11704 [01:07<1:36:35,  2.01it/s]
  1%|          | 76/11704 [01:08<1:36:49,  2.00it/s]
  1%|          | 77/11704 [01:08<1:36:51,  2.00it/s]
  1%|          | 78/11704 [01:09<1:36:22,  2.01it/s]
  1%|          | 79/11704 [01:09<1:36:16,  2.01it/s]
  1%|          | 80/11704 [01:10<1:36:33,  2.01it/s]
  1%|          | 81/11704 [01:10<1:36:15,  2.01it/s]
  1%|          | 82/11704 [01:11<1:36:11,  2.01it/s]
  1%|          | 83/11704 [01:11<1:36:08,  2.01it/s]
  1%|          | 84/11704 [01:12<1:35:57,  2.02it/s]
  1%|          | 85/11704 [01:12<1:36:27,  2.01it/s]
  1%|          | 86/11704 [01:13<1:36:25,  2.01it/s]
  1%|          | 87/11704 [01:13<1:36:11,  2.01it/s]
  1%|          | 88/11704 [01:14<1:36:08,  2.01it/s]
  1%|          | 89/11704 [01:14<1:37:01,  2.00it/s]
  1%|          | 90/11704 [01:15<1:39:29,  1.95it/s]
  1%|          | 91/11704 [01:15<1:38:15,  1.97it/s]
  1%|          | 92/11704 [01:16<1:38:03,  1.97it/s]
  1%|          | 93/11704 [01:16<1:37:15,  1.99it/s]
  1%|          | 94/11704 [01:17<1:36:57,  2.00it/s]
  1%|          | 95/11704 [01:17<1:36:48,  2.00it/s]
  1%|          | 96/11704 [01:18<1:36:31,  2.00it/s]
  1%|          | 97/11704 [01:18<1:36:15,  2.01it/s]
  1%|          | 98/11704 [01:19<1:36:02,  2.01it/s]
  1%|          | 99/11704 [01:19<1:36:18,  2.01it/s]
  1%|          | 100/11704 [01:20<1:36:11,  2.01it/s]{'loss': 8.3139, 'grad_norm': 0.7935464978218079, 'learning_rate': 8.539709649871905e-05, 'epoch': 0.12}
+
                                                     
  1%|          | 100/11704 [01:20<1:36:11,  2.01it/s]
  1%|          | 101/11704 [01:20<1:36:06,  2.01it/s]
  1%|          | 102/11704 [01:21<1:35:47,  2.02it/s]
  1%|          | 103/11704 [01:21<1:36:08,  2.01it/s]
  1%|          | 104/11704 [01:22<1:35:49,  2.02it/s]
  1%|          | 105/11704 [01:22<1:35:43,  2.02it/s]
  1%|          | 106/11704 [01:23<1:35:41,  2.02it/s]
  1%|          | 107/11704 [01:23<1:35:33,  2.02it/s]
  1%|          | 108/11704 [01:24<1:35:22,  2.03it/s]
  1%|          | 109/11704 [01:24<1:35:39,  2.02it/s]
  1%|          | 110/11704 [01:25<1:35:33,  2.02it/s]
  1%|          | 111/11704 [01:25<1:35:49,  2.02it/s]
  1%|          | 112/11704 [01:26<1:35:39,  2.02it/s]
  1%|          | 113/11704 [01:26<1:35:34,  2.02it/s]
  1%|          | 114/11704 [01:27<1:35:24,  2.02it/s]
  1%|          | 115/11704 [01:27<1:35:20,  2.03it/s]
  1%|          | 116/11704 [01:28<1:35:31,  2.02it/s]
  1%|          | 117/11704 [01:28<1:35:27,  2.02it/s]
  1%|          | 118/11704 [01:29<1:35:24,  2.02it/s]
  1%|          | 119/11704 [01:29<1:35:14,  2.03it/s]
  1%|          | 120/11704 [01:30<1:35:08,  2.03it/s]
  1%|          | 121/11704 [01:30<1:35:10,  2.03it/s]
  1%|          | 122/11704 [01:31<1:35:03,  2.03it/s]
  1%|          | 123/11704 [01:31<1:34:57,  2.03it/s]
  1%|          | 124/11704 [01:32<1:35:01,  2.03it/s]
  1%|          | 125/11704 [01:32<1:35:08,  2.03it/s]{'loss': 7.526, 'grad_norm': 0.5917059183120728, 'learning_rate': 0.0001067463706233988, 'epoch': 0.15}
+
                                                     
  1%|          | 125/11704 [01:32<1:35:08,  2.03it/s]
  1%|          | 126/11704 [01:33<1:35:09,  2.03it/s]
  1%|          | 127/11704 [01:33<1:35:04,  2.03it/s]
  1%|          | 128/11704 [01:34<1:34:59,  2.03it/s]
  1%|          | 129/11704 [01:34<1:34:59,  2.03it/s]
  1%|          | 130/11704 [01:35<1:35:09,  2.03it/s]
  1%|          | 131/11704 [01:35<1:35:07,  2.03it/s]
  1%|          | 132/11704 [01:35<1:35:02,  2.03it/s]
  1%|          | 133/11704 [01:36<1:34:58,  2.03it/s]
  1%|          | 134/11704 [01:36<1:34:58,  2.03it/s]
  1%|          | 135/11704 [01:37<1:34:56,  2.03it/s]
  1%|          | 136/11704 [01:37<1:35:01,  2.03it/s]
  1%|          | 137/11704 [01:38<1:34:56,  2.03it/s]
  1%|          | 138/11704 [01:38<1:35:30,  2.02it/s]
  1%|          | 139/11704 [01:39<1:35:41,  2.01it/s]
  1%|          | 140/11704 [01:39<1:35:29,  2.02it/s]
  1%|          | 141/11704 [01:40<1:35:14,  2.02it/s]
  1%|          | 142/11704 [01:40<1:35:11,  2.02it/s]
  1%|          | 143/11704 [01:41<1:35:00,  2.03it/s]
  1%|          | 144/11704 [01:41<1:34:55,  2.03it/s]
  1%|          | 145/11704 [01:42<1:34:52,  2.03it/s]
  1%|          | 146/11704 [01:42<1:34:48,  2.03it/s]
  1%|▏         | 147/11704 [01:43<1:34:51,  2.03it/s]
  1%|▏         | 148/11704 [01:43<1:34:45,  2.03it/s]
  1%|▏         | 149/11704 [01:44<1:34:43,  2.03it/s]
  1%|▏         | 150/11704 [01:44<1:34:48,  2.03it/s]{'loss': 6.8951, 'grad_norm': 0.37877357006073, 'learning_rate': 0.00012809564474807855, 'epoch': 0.18}
+
                                                     
  1%|▏         | 150/11704 [01:44<1:34:48,  2.03it/s]
  1%|▏         | 151/11704 [01:45<1:34:56,  2.03it/s]
  1%|▏         | 152/11704 [01:45<1:34:58,  2.03it/s]
  1%|▏         | 153/11704 [01:46<1:34:45,  2.03it/s]
  1%|▏         | 154/11704 [01:46<1:34:38,  2.03it/s]
  1%|▏         | 155/11704 [01:47<1:34:42,  2.03it/s]
  1%|▏         | 156/11704 [01:47<1:34:49,  2.03it/s]
  1%|▏         | 157/11704 [01:48<1:34:49,  2.03it/s]
  1%|▏         | 158/11704 [01:48<1:34:47,  2.03it/s]
  1%|▏         | 159/11704 [01:49<1:34:42,  2.03it/s]
  1%|▏         | 160/11704 [01:49<1:34:40,  2.03it/s]
  1%|▏         | 161/11704 [01:50<1:34:38,  2.03it/s]
  1%|▏         | 162/11704 [01:50<1:34:39,  2.03it/s]
  1%|▏         | 163/11704 [01:51<1:34:47,  2.03it/s]
  1%|▏         | 164/11704 [01:51<1:34:47,  2.03it/s]
  1%|▏         | 165/11704 [01:52<1:34:54,  2.03it/s]
  1%|▏         | 166/11704 [01:52<1:34:47,  2.03it/s]
  1%|▏         | 167/11704 [01:53<1:34:47,  2.03it/s]
  1%|▏         | 168/11704 [01:53<1:34:38,  2.03it/s]
  1%|▏         | 169/11704 [01:54<1:34:42,  2.03it/s]
  1%|▏         | 170/11704 [01:54<1:34:41,  2.03it/s]
  1%|▏         | 171/11704 [01:55<1:34:32,  2.03it/s]
  1%|▏         | 172/11704 [01:55<1:34:38,  2.03it/s]
  1%|▏         | 173/11704 [01:56<1:34:32,  2.03it/s]
  1%|▏         | 174/11704 [01:56<1:34:27,  2.03it/s]
  1%|▏         | 175/11704 [01:57<1:34:29,  2.03it/s]{'loss': 6.3957, 'grad_norm': 0.3709251582622528, 'learning_rate': 0.00014944491887275833, 'epoch': 0.21}
+
                                                     
  1%|▏         | 175/11704 [01:57<1:34:29,  2.03it/s]
  2%|▏         | 176/11704 [01:57<1:34:26,  2.03it/s]
  2%|▏         | 177/11704 [01:58<1:34:35,  2.03it/s]
  2%|▏         | 178/11704 [01:58<1:34:30,  2.03it/s]
  2%|▏         | 179/11704 [01:59<1:34:27,  2.03it/s]
  2%|▏         | 180/11704 [01:59<1:34:35,  2.03it/s]
  2%|▏         | 181/11704 [02:00<1:34:38,  2.03it/s]
  2%|▏         | 182/11704 [02:00<1:34:39,  2.03it/s]
  2%|▏         | 183/11704 [02:01<1:34:32,  2.03it/s]
  2%|▏         | 184/11704 [02:01<1:34:33,  2.03it/s]
  2%|▏         | 185/11704 [02:02<1:34:35,  2.03it/s]
  2%|▏         | 186/11704 [02:02<1:34:32,  2.03it/s]
  2%|▏         | 187/11704 [02:03<1:34:35,  2.03it/s]
  2%|▏         | 188/11704 [02:03<1:34:30,  2.03it/s]
  2%|▏         | 189/11704 [02:04<1:34:32,  2.03it/s]
  2%|▏         | 190/11704 [02:04<1:34:33,  2.03it/s]
  2%|▏         | 191/11704 [02:05<1:34:28,  2.03it/s]
  2%|▏         | 192/11704 [02:05<1:34:33,  2.03it/s]
  2%|▏         | 193/11704 [02:06<1:34:30,  2.03it/s]
  2%|▏         | 194/11704 [02:06<1:34:35,  2.03it/s]
  2%|▏         | 195/11704 [02:07<1:34:27,  2.03it/s]
  2%|▏         | 196/11704 [02:07<1:34:29,  2.03it/s]
  2%|▏         | 197/11704 [02:08<1:34:33,  2.03it/s]
  2%|▏         | 198/11704 [02:08<1:34:28,  2.03it/s]
  2%|▏         | 199/11704 [02:09<1:34:28,  2.03it/s]
  2%|▏         | 200/11704 [02:09<1:34:30,  2.03it/s]{'loss': 5.9826, 'grad_norm': 0.6036785840988159, 'learning_rate': 0.0001707941929974381, 'epoch': 0.24}
+
                                                     
  2%|▏         | 200/11704 [02:09<1:34:30,  2.03it/s]
  2%|▏         | 201/11704 [02:09<1:34:24,  2.03it/s]
  2%|▏         | 202/11704 [02:10<1:34:31,  2.03it/s]
  2%|▏         | 203/11704 [02:10<1:34:23,  2.03it/s]
  2%|▏         | 204/11704 [02:11<1:34:21,  2.03it/s]
  2%|▏         | 205/11704 [02:11<1:34:27,  2.03it/s]
  2%|▏         | 206/11704 [02:12<1:34:31,  2.03it/s]
  2%|▏         | 207/11704 [02:12<1:34:36,  2.03it/s]
  2%|▏         | 208/11704 [02:13<1:34:28,  2.03it/s]
  2%|▏         | 209/11704 [02:13<1:34:31,  2.03it/s]
  2%|▏         | 210/11704 [02:14<1:34:28,  2.03it/s]
  2%|▏         | 211/11704 [02:14<1:34:26,  2.03it/s]
  2%|▏         | 212/11704 [02:15<1:34:28,  2.03it/s]
  2%|▏         | 213/11704 [02:15<1:34:23,  2.03it/s]
  2%|▏         | 214/11704 [02:16<1:34:28,  2.03it/s]
  2%|▏         | 215/11704 [02:16<1:34:30,  2.03it/s]
  2%|▏         | 216/11704 [02:17<1:34:29,  2.03it/s]
  2%|▏         | 217/11704 [02:17<1:34:34,  2.02it/s]
  2%|▏         | 218/11704 [02:18<1:34:38,  2.02it/s]
  2%|▏         | 219/11704 [02:18<1:34:37,  2.02it/s]
  2%|▏         | 220/11704 [02:19<1:34:45,  2.02it/s]
  2%|▏         | 221/11704 [02:19<1:34:34,  2.02it/s]
  2%|▏         | 222/11704 [02:20<1:34:43,  2.02it/s]
  2%|▏         | 223/11704 [02:20<1:34:35,  2.02it/s]
  2%|▏         | 224/11704 [02:21<1:34:33,  2.02it/s]
  2%|▏         | 225/11704 [02:21<1:34:27,  2.03it/s]{'loss': 5.659, 'grad_norm': 0.9925475716590881, 'learning_rate': 0.00019214346712211785, 'epoch': 0.27}
                                                     
+
  2%|▏         | 225/11704 [02:21<1:34:27,  2.03it/s]
  2%|▏         | 226/11704 [02:22<1:34:29,  2.02it/s]
  2%|▏         | 227/11704 [02:22<1:34:27,  2.03it/s]
  2%|▏         | 228/11704 [02:23<1:34:41,  2.02it/s]
  2%|▏         | 229/11704 [02:23<1:34:33,  2.02it/s]
  2%|▏         | 230/11704 [02:24<1:34:35,  2.02it/s]
  2%|▏         | 231/11704 [02:24<1:34:27,  2.02it/s]
  2%|▏         | 232/11704 [02:25<1:34:25,  2.03it/s]
  2%|▏         | 233/11704 [02:25<1:34:22,  2.03it/s]
  2%|▏         | 234/11704 [02:26<1:34:20,  2.03it/s]
  2%|▏         | 235/11704 [02:26<1:34:18,  2.03it/s]
  2%|▏         | 236/11704 [02:27<1:34:21,  2.03it/s]
  2%|▏         | 237/11704 [02:27<1:34:25,  2.02it/s]
  2%|▏         | 238/11704 [02:28<1:34:20,  2.03it/s]
  2%|▏         | 239/11704 [02:28<1:34:29,  2.02it/s]
  2%|▏         | 240/11704 [02:29<1:34:21,  2.02it/s]
  2%|▏         | 241/11704 [02:29<1:34:26,  2.02it/s]
  2%|▏         | 242/11704 [02:30<1:34:19,  2.03it/s]
  2%|▏         | 243/11704 [02:30<1:34:16,  2.03it/s]
  2%|▏         | 244/11704 [02:31<1:34:14,  2.03it/s]
  2%|▏         | 245/11704 [02:31<1:34:16,  2.03it/s]
  2%|▏         | 246/11704 [02:32<1:34:15,  2.03it/s]
  2%|▏         | 247/11704 [02:32<1:34:18,  2.02it/s]
  2%|▏         | 248/11704 [02:33<1:34:35,  2.02it/s]
  2%|▏         | 249/11704 [02:33<1:34:34,  2.02it/s]
  2%|▏         | 250/11704 [02:34<1:34:36,  2.02it/s]{'loss': 5.4253, 'grad_norm': 0.6297341585159302, 'learning_rate': 0.0002134927412467976, 'epoch': 0.3}
+
                                                     
  2%|▏         | 250/11704 [02:34<1:34:36,  2.02it/s]
  2%|▏         | 251/11704 [02:34<1:34:48,  2.01it/s]
  2%|▏         | 252/11704 [02:35<1:34:35,  2.02it/s]
  2%|▏         | 253/11704 [02:35<1:34:32,  2.02it/s]
  2%|▏         | 254/11704 [02:36<1:34:26,  2.02it/s]
  2%|▏         | 255/11704 [02:36<1:34:20,  2.02it/s]
  2%|▏         | 256/11704 [02:37<1:34:34,  2.02it/s]
  2%|▏         | 257/11704 [02:37<1:34:28,  2.02it/s]
  2%|▏         | 258/11704 [02:38<1:34:19,  2.02it/s]
  2%|▏         | 259/11704 [02:38<1:34:10,  2.03it/s]
  2%|▏         | 260/11704 [02:39<1:34:10,  2.03it/s]
  2%|▏         | 261/11704 [02:39<1:34:06,  2.03it/s]
  2%|▏         | 262/11704 [02:40<1:33:56,  2.03it/s]
  2%|▏         | 263/11704 [02:40<1:33:59,  2.03it/s]
  2%|▏         | 264/11704 [02:41<1:33:55,  2.03it/s]
  2%|▏         | 265/11704 [02:41<1:34:01,  2.03it/s]
  2%|▏         | 266/11704 [02:42<1:34:04,  2.03it/s]
  2%|▏         | 267/11704 [02:42<1:33:59,  2.03it/s]
  2%|▏         | 268/11704 [02:43<1:34:06,  2.03it/s]
  2%|▏         | 269/11704 [02:43<1:34:01,  2.03it/s]
  2%|▏         | 270/11704 [02:44<1:34:04,  2.03it/s]
  2%|▏         | 271/11704 [02:44<1:33:59,  2.03it/s]
  2%|▏         | 272/11704 [02:45<1:33:57,  2.03it/s]
  2%|▏         | 273/11704 [02:45<1:33:59,  2.03it/s]
  2%|▏         | 274/11704 [02:46<1:33:57,  2.03it/s]
  2%|▏         | 275/11704 [02:46<1:34:06,  2.02it/s]{'loss': 5.2293, 'grad_norm': 0.5968512892723083, 'learning_rate': 0.00023484201537147736, 'epoch': 0.33}
                                                     
+
  2%|▏         | 275/11704 [02:46<1:34:06,  2.02it/s]
  2%|▏         | 276/11704 [02:47<1:34:02,  2.03it/s]
  2%|▏         | 277/11704 [02:47<1:34:05,  2.02it/s]
  2%|▏         | 278/11704 [02:48<1:33:59,  2.03it/s]
  2%|▏         | 279/11704 [02:48<1:34:01,  2.02it/s]
  2%|▏         | 280/11704 [02:49<1:33:54,  2.03it/s]
  2%|▏         | 281/11704 [02:49<1:34:00,  2.03it/s]
  2%|▏         | 282/11704 [02:49<1:33:55,  2.03it/s]
  2%|▏         | 283/11704 [02:50<1:33:55,  2.03it/s]
  2%|▏         | 284/11704 [02:50<1:33:52,  2.03it/s]
  2%|▏         | 285/11704 [02:51<1:33:47,  2.03it/s]
  2%|▏         | 286/11704 [02:51<1:33:49,  2.03it/s]
  2%|▏         | 287/11704 [02:52<1:33:49,  2.03it/s]
  2%|▏         | 288/11704 [02:52<1:33:50,  2.03it/s]
  2%|▏         | 289/11704 [02:53<1:33:43,  2.03it/s]
  2%|▏         | 290/11704 [02:53<1:33:47,  2.03it/s]
  2%|▏         | 291/11704 [02:54<1:33:42,  2.03it/s]
  2%|▏         | 292/11704 [02:54<1:33:40,  2.03it/s]
  3%|▎         | 293/11704 [02:55<1:33:43,  2.03it/s]
  3%|▎         | 294/11704 [02:55<1:33:38,  2.03it/s]
  3%|▎         | 295/11704 [02:56<1:33:41,  2.03it/s]
  3%|▎         | 296/11704 [02:56<1:33:37,  2.03it/s]
  3%|▎         | 297/11704 [02:57<1:33:29,  2.03it/s]
  3%|▎         | 298/11704 [02:57<1:33:33,  2.03it/s]
  3%|▎         | 299/11704 [02:58<1:33:34,  2.03it/s]
  3%|▎         | 300/11704 [02:58<1:33:27,  2.03it/s]{'loss': 5.0541, 'grad_norm': 0.9146194458007812, 'learning_rate': 0.0002561912894961571, 'epoch': 0.36}
+
                                                     
  3%|▎         | 300/11704 [02:58<1:33:27,  2.03it/s]
  3%|▎         | 301/11704 [02:59<1:33:30,  2.03it/s]
  3%|▎         | 302/11704 [02:59<1:33:27,  2.03it/s]
  3%|▎         | 303/11704 [03:00<1:33:27,  2.03it/s]
  3%|▎         | 304/11704 [03:00<1:33:31,  2.03it/s]
  3%|▎         | 305/11704 [03:01<1:33:25,  2.03it/s]
  3%|▎         | 306/11704 [03:01<1:33:30,  2.03it/s]
  3%|▎         | 307/11704 [03:02<1:33:24,  2.03it/s]
  3%|▎         | 308/11704 [03:02<1:33:26,  2.03it/s]
  3%|▎         | 309/11704 [03:03<1:33:31,  2.03it/s]
  3%|▎         | 310/11704 [03:03<1:33:21,  2.03it/s]
  3%|▎         | 311/11704 [03:04<1:33:24,  2.03it/s]
  3%|▎         | 312/11704 [03:04<1:33:26,  2.03it/s]
  3%|▎         | 313/11704 [03:05<1:33:25,  2.03it/s]
  3%|▎         | 314/11704 [03:05<1:33:30,  2.03it/s]
  3%|▎         | 315/11704 [03:06<1:33:22,  2.03it/s]
  3%|▎         | 316/11704 [03:06<1:33:25,  2.03it/s]
  3%|▎         | 317/11704 [03:07<1:33:24,  2.03it/s]
  3%|▎         | 318/11704 [03:07<1:33:20,  2.03it/s]
  3%|▎         | 319/11704 [03:08<1:33:22,  2.03it/s]
  3%|▎         | 320/11704 [03:08<1:33:22,  2.03it/s]
  3%|▎         | 321/11704 [03:09<1:33:16,  2.03it/s]
  3%|▎         | 322/11704 [03:09<1:33:24,  2.03it/s]
  3%|▎         | 323/11704 [03:10<1:33:18,  2.03it/s]
  3%|▎         | 324/11704 [03:10<1:33:15,  2.03it/s]
  3%|▎         | 325/11704 [03:11<1:33:18,  2.03it/s]{'loss': 4.934, 'grad_norm': 0.9477291703224182, 'learning_rate': 0.0002775405636208369, 'epoch': 0.39}
+
                                                     
  3%|▎         | 325/11704 [03:11<1:33:18,  2.03it/s]
  3%|▎         | 326/11704 [03:11<1:33:15,  2.03it/s]
  3%|▎         | 327/11704 [03:12<1:33:14,  2.03it/s]
  3%|▎         | 328/11704 [03:12<1:33:18,  2.03it/s]
  3%|▎         | 329/11704 [03:13<1:33:10,  2.03it/s]
  3%|▎         | 330/11704 [03:13<1:33:10,  2.03it/s]
  3%|▎         | 331/11704 [03:14<1:33:11,  2.03it/s]
  3%|▎         | 332/11704 [03:14<1:33:11,  2.03it/s]
  3%|▎         | 333/11704 [03:15<1:33:12,  2.03it/s]
  3%|▎         | 334/11704 [03:15<1:33:15,  2.03it/s]
  3%|▎         | 335/11704 [03:16<1:33:09,  2.03it/s]
  3%|▎         | 336/11704 [03:16<1:33:09,  2.03it/s]
  3%|▎         | 337/11704 [03:17<1:33:16,  2.03it/s]
  3%|▎         | 338/11704 [03:17<1:33:49,  2.02it/s]
  3%|▎         | 339/11704 [03:18<1:33:39,  2.02it/s]
  3%|▎         | 340/11704 [03:18<1:33:32,  2.02it/s]
  3%|▎         | 341/11704 [03:19<1:33:29,  2.03it/s]
  3%|▎         | 342/11704 [03:19<1:33:20,  2.03it/s]
  3%|▎         | 343/11704 [03:20<1:33:18,  2.03it/s]
  3%|▎         | 344/11704 [03:20<1:33:18,  2.03it/s]
  3%|▎         | 345/11704 [03:21<1:33:12,  2.03it/s]
  3%|▎         | 346/11704 [03:21<1:33:12,  2.03it/s]
  3%|▎         | 347/11704 [03:21<1:33:08,  2.03it/s]
  3%|▎         | 348/11704 [03:22<1:33:03,  2.03it/s]
  3%|▎         | 349/11704 [03:22<1:33:10,  2.03it/s]
  3%|▎         | 350/11704 [03:23<1:33:03,  2.03it/s]{'loss': 4.8046, 'grad_norm': 0.824380099773407, 'learning_rate': 0.00029888983774551667, 'epoch': 0.42}
+
                                                     
  3%|▎         | 350/11704 [03:23<1:33:03,  2.03it/s]
  3%|▎         | 351/11704 [03:23<1:33:04,  2.03it/s]
  3%|▎         | 352/11704 [03:24<1:33:07,  2.03it/s]
  3%|▎         | 353/11704 [03:24<1:33:07,  2.03it/s]
  3%|▎         | 354/11704 [03:25<1:33:00,  2.03it/s]
  3%|▎         | 355/11704 [03:25<1:33:01,  2.03it/s]
  3%|▎         | 356/11704 [03:26<1:32:58,  2.03it/s]
  3%|▎         | 357/11704 [03:26<1:33:00,  2.03it/s]
  3%|▎         | 358/11704 [03:27<1:32:59,  2.03it/s]
  3%|▎         | 359/11704 [03:27<1:32:59,  2.03it/s]
  3%|▎         | 360/11704 [03:28<1:32:57,  2.03it/s]
  3%|▎         | 361/11704 [03:28<1:32:57,  2.03it/s]
  3%|▎         | 362/11704 [03:29<1:32:54,  2.03it/s]
  3%|▎         | 363/11704 [03:29<1:32:51,  2.04it/s]
  3%|▎         | 364/11704 [03:30<1:32:54,  2.03it/s]
  3%|▎         | 365/11704 [03:30<1:32:55,  2.03it/s]
  3%|▎         | 366/11704 [03:31<1:32:52,  2.03it/s]
  3%|▎         | 367/11704 [03:31<1:33:03,  2.03it/s]
  3%|▎         | 368/11704 [03:32<1:33:03,  2.03it/s]
  3%|▎         | 369/11704 [03:32<1:33:10,  2.03it/s]
  3%|▎         | 370/11704 [03:33<1:33:01,  2.03it/s]
  3%|▎         | 371/11704 [03:33<1:32:54,  2.03it/s]
  3%|▎         | 372/11704 [03:34<1:32:58,  2.03it/s]
  3%|▎         | 373/11704 [03:34<1:32:51,  2.03it/s]
  3%|▎         | 374/11704 [03:35<1:32:56,  2.03it/s]
  3%|▎         | 375/11704 [03:35<1:33:04,  2.03it/s]{'loss': 4.6929, 'grad_norm': 0.7358281016349792, 'learning_rate': 0.0003202391118701964, 'epoch': 0.45}
+
                                                     
  3%|▎         | 375/11704 [03:35<1:33:04,  2.03it/s]
  3%|▎         | 376/11704 [03:36<1:33:02,  2.03it/s]
  3%|▎         | 377/11704 [03:36<1:33:04,  2.03it/s]
  3%|▎         | 378/11704 [03:37<1:33:01,  2.03it/s]
  3%|▎         | 379/11704 [03:37<1:33:04,  2.03it/s]
  3%|▎         | 380/11704 [03:38<1:33:01,  2.03it/s]
  3%|▎         | 381/11704 [03:38<1:32:53,  2.03it/s]
  3%|▎         | 382/11704 [03:39<1:33:07,  2.03it/s]
  3%|▎         | 383/11704 [03:39<1:33:02,  2.03it/s]
  3%|▎         | 384/11704 [03:40<1:33:01,  2.03it/s]
  3%|▎         | 385/11704 [03:40<1:32:59,  2.03it/s]
  3%|▎         | 386/11704 [03:41<1:32:51,  2.03it/s]
  3%|▎         | 387/11704 [03:41<1:32:44,  2.03it/s]
  3%|▎         | 388/11704 [03:42<1:32:52,  2.03it/s]
  3%|▎         | 389/11704 [03:42<1:32:47,  2.03it/s]
  3%|▎         | 390/11704 [03:43<1:32:51,  2.03it/s]
  3%|▎         | 391/11704 [03:43<1:32:56,  2.03it/s]
  3%|▎         | 392/11704 [03:44<1:32:50,  2.03it/s]
  3%|▎         | 393/11704 [03:44<1:32:50,  2.03it/s]
  3%|▎         | 394/11704 [03:45<1:32:44,  2.03it/s]
  3%|▎         | 395/11704 [03:45<1:32:41,  2.03it/s]
  3%|▎         | 396/11704 [03:46<1:32:36,  2.04it/s]
  3%|▎         | 397/11704 [03:46<1:32:45,  2.03it/s]
  3%|▎         | 398/11704 [03:47<1:32:39,  2.03it/s]
  3%|▎         | 399/11704 [03:47<1:32:35,  2.03it/s]
  3%|▎         | 400/11704 [03:48<1:32:43,  2.03it/s]
                                                     
+
  3%|▎         | 400/11704 [03:48<1:32:43,  2.03it/s]
  3%|▎         | 401/11704 [03:48<1:32:46,  2.03it/s]
  3%|▎         | 402/11704 [03:49<1:32:45,  2.03it/s]
  3%|▎         | 403/11704 [03:49<1:32:48,  2.03it/s]
  3%|▎         | 404/11704 [03:50<1:32:43,  2.03it/s]
  3%|▎         | 405/11704 [03:50<1:32:41,  2.03it/s]
  3%|▎         | 406/11704 [03:51<1:32:39,  2.03it/s]
  3%|▎         | 407/11704 [03:51<1:32:34,  2.03it/s]
  3%|▎         | 408/11704 [03:52<1:32:39,  2.03it/s]
  3%|▎         | 409/11704 [03:52<1:32:39,  2.03it/s]
  4%|▎         | 410/11704 [03:53<1:32:35,  2.03it/s]
  4%|▎         | 411/11704 [03:53<1:32:35,  2.03it/s]
  4%|▎         | 412/11704 [03:53<1:32:32,  2.03it/s]
  4%|▎         | 413/11704 [03:54<1:32:25,  2.04it/s]
  4%|▎         | 414/11704 [03:54<1:32:29,  2.03it/s]
  4%|▎         | 415/11704 [03:55<1:32:26,  2.04it/s]
  4%|▎         | 416/11704 [03:55<1:32:17,  2.04it/s]
  4%|▎         | 417/11704 [03:56<1:32:23,  2.04it/s]
  4%|▎         | 418/11704 [03:56<1:32:26,  2.03it/s]
  4%|▎         | 419/11704 [03:57<1:32:22,  2.04it/s]
  4%|▎         | 420/11704 [03:57<1:32:24,  2.04it/s]
  4%|▎         | 421/11704 [03:58<1:32:22,  2.04it/s]
  4%|▎         | 422/11704 [03:58<1:32:19,  2.04it/s]
  4%|▎         | 423/11704 [03:59<1:32:20,  2.04it/s]
  4%|▎         | 424/11704 [03:59<1:32:19,  2.04it/s]
  4%|▎         | 425/11704 [04:00<1:32:14,  2.04it/s]
                                                     
+
  4%|▎         | 425/11704 [04:00<1:32:14,  2.04it/s]
  4%|▎         | 426/11704 [04:00<1:32:21,  2.04it/s]
  4%|▎         | 427/11704 [04:01<1:32:20,  2.04it/s]
  4%|▎         | 428/11704 [04:01<1:32:16,  2.04it/s]
  4%|▎         | 429/11704 [04:02<1:32:26,  2.03it/s]
  4%|▎         | 430/11704 [04:02<1:32:20,  2.03it/s]
  4%|▎         | 431/11704 [04:03<1:32:17,  2.04it/s]
  4%|▎         | 432/11704 [04:03<1:32:23,  2.03it/s]
  4%|▎         | 433/11704 [04:04<1:32:20,  2.03it/s]
  4%|▎         | 434/11704 [04:04<1:32:17,  2.04it/s]
  4%|▎         | 435/11704 [04:05<1:32:23,  2.03it/s]
  4%|▎         | 436/11704 [04:05<1:32:19,  2.03it/s]
  4%|▎         | 437/11704 [04:06<1:32:21,  2.03it/s]
  4%|▎         | 438/11704 [04:06<1:32:25,  2.03it/s]
  4%|▍         | 439/11704 [04:07<1:32:13,  2.04it/s]
  4%|▍         | 440/11704 [04:07<1:32:22,  2.03it/s]
  4%|▍         | 441/11704 [04:08<1:32:18,  2.03it/s]
  4%|▍         | 442/11704 [04:08<1:32:09,  2.04it/s]
  4%|▍         | 443/11704 [04:09<1:32:17,  2.03it/s]
  4%|▍         | 444/11704 [04:09<1:32:16,  2.03it/s]
  4%|▍         | 445/11704 [04:10<1:32:08,  2.04it/s]
  4%|▍         | 446/11704 [04:10<1:32:13,  2.03it/s]
  4%|▍         | 447/11704 [04:11<1:32:10,  2.04it/s]
  4%|▍         | 448/11704 [04:11<1:32:08,  2.04it/s]
  4%|▍         | 449/11704 [04:12<1:32:12,  2.03it/s]
  4%|▍         | 450/11704 [04:12<1:32:11,  2.03it/s]{'loss': 4.4307, 'grad_norm': 0.6776960492134094, 'learning_rate': 0.0003842869342442357, 'epoch': 0.54}
                                                     
+
  4%|▍         | 450/11704 [04:12<1:32:11,  2.03it/s]
  4%|▍         | 451/11704 [04:13<1:32:13,  2.03it/s]
  4%|▍         | 452/11704 [04:13<1:32:17,  2.03it/s]
  4%|▍         | 453/11704 [04:14<1:32:13,  2.03it/s]
  4%|▍         | 454/11704 [04:14<1:32:16,  2.03it/s]
  4%|▍         | 455/11704 [04:15<1:32:11,  2.03it/s]
  4%|▍         | 456/11704 [04:15<1:32:06,  2.04it/s]
  4%|▍         | 457/11704 [04:16<1:32:11,  2.03it/s]
  4%|▍         | 458/11704 [04:16<1:32:09,  2.03it/s]
  4%|▍         | 459/11704 [04:17<1:32:04,  2.04it/s]
  4%|▍         | 460/11704 [04:17<1:32:09,  2.03it/s]
  4%|▍         | 461/11704 [04:18<1:32:10,  2.03it/s]
  4%|▍         | 462/11704 [04:18<1:32:04,  2.03it/s]
  4%|▍         | 463/11704 [04:19<1:32:10,  2.03it/s]
  4%|▍         | 464/11704 [04:19<1:32:03,  2.04it/s]
  4%|▍         | 465/11704 [04:20<1:31:56,  2.04it/s]
  4%|▍         | 466/11704 [04:20<1:32:00,  2.04it/s]
  4%|▍         | 467/11704 [04:21<1:32:05,  2.03it/s]
  4%|▍         | 468/11704 [04:21<1:31:58,  2.04it/s]
  4%|▍         | 469/11704 [04:22<1:32:03,  2.03it/s]
  4%|▍         | 470/11704 [04:22<1:31:58,  2.04it/s]
  4%|▍         | 471/11704 [04:22<1:31:56,  2.04it/s]
  4%|▍         | 472/11704 [04:23<1:31:58,  2.04it/s]
  4%|▍         | 473/11704 [04:23<1:32:04,  2.03it/s]
  4%|▍         | 474/11704 [04:24<1:31:58,  2.04it/s]
  4%|▍         | 475/11704 [04:24<1:32:03,  2.03it/s]{'loss': 4.3608, 'grad_norm': 0.7181493043899536, 'learning_rate': 0.00040563620836891546, 'epoch': 0.57}
+
                                                     
  4%|▍         | 475/11704 [04:24<1:32:03,  2.03it/s]
  4%|▍         | 476/11704 [04:25<1:32:02,  2.03it/s]
  4%|▍         | 477/11704 [04:25<1:32:02,  2.03it/s]
  4%|▍         | 478/11704 [04:26<1:32:07,  2.03it/s]
  4%|▍         | 479/11704 [04:26<1:31:59,  2.03it/s]
  4%|▍         | 480/11704 [04:27<1:32:01,  2.03it/s]
  4%|▍         | 481/11704 [04:27<1:32:03,  2.03it/s]
  4%|▍         | 482/11704 [04:28<1:31:58,  2.03it/s]
  4%|▍         | 483/11704 [04:28<1:32:07,  2.03it/s]
  4%|▍         | 484/11704 [04:29<1:32:02,  2.03it/s]
  4%|▍         | 485/11704 [04:29<1:31:57,  2.03it/s]
  4%|▍         | 486/11704 [04:30<1:31:53,  2.03it/s]
  4%|▍         | 487/11704 [04:30<1:31:57,  2.03it/s]
  4%|▍         | 488/11704 [04:31<1:31:56,  2.03it/s]
  4%|▍         | 489/11704 [04:31<1:31:56,  2.03it/s]
  4%|▍         | 490/11704 [04:32<1:31:57,  2.03it/s]
  4%|▍         | 491/11704 [04:32<1:31:54,  2.03it/s]
  4%|▍         | 492/11704 [04:33<1:31:57,  2.03it/s]
  4%|▍         | 493/11704 [04:33<1:31:54,  2.03it/s]
  4%|▍         | 494/11704 [04:34<1:31:55,  2.03it/s]
  4%|▍         | 495/11704 [04:34<1:31:55,  2.03it/s]
  4%|▍         | 496/11704 [04:35<1:31:53,  2.03it/s]
  4%|▍         | 497/11704 [04:35<1:31:52,  2.03it/s]
  4%|▍         | 498/11704 [04:36<1:31:53,  2.03it/s]
  4%|▍         | 499/11704 [04:36<1:31:47,  2.03it/s]
  4%|▍         | 500/11704 [04:37<1:31:42,  2.04it/s]{'loss': 4.2807, 'grad_norm': 0.6647967100143433, 'learning_rate': 0.0004269854824935952, 'epoch': 0.6}
+
                                                     
  4%|▍         | 500/11704 [04:37<1:31:42,  2.04it/s]
  4%|▍         | 501/11704 [04:37<1:31:46,  2.03it/s]
  4%|▍         | 502/11704 [04:38<1:31:41,  2.04it/s]
  4%|▍         | 503/11704 [04:38<1:31:44,  2.03it/s]
  4%|▍         | 504/11704 [04:39<1:31:46,  2.03it/s]
  4%|▍         | 505/11704 [04:39<1:31:44,  2.03it/s]
  4%|▍         | 506/11704 [04:40<1:31:47,  2.03it/s]
  4%|▍         | 507/11704 [04:40<1:31:46,  2.03it/s]
  4%|▍         | 508/11704 [04:41<1:31:43,  2.03it/s]
  4%|▍         | 509/11704 [04:41<1:31:50,  2.03it/s]
  4%|▍         | 510/11704 [04:42<1:31:44,  2.03it/s]
  4%|▍         | 511/11704 [04:42<1:31:40,  2.04it/s]
  4%|▍         | 512/11704 [04:43<1:31:45,  2.03it/s]
  4%|▍         | 513/11704 [04:43<1:31:42,  2.03it/s]
  4%|▍         | 514/11704 [04:44<1:31:42,  2.03it/s]
  4%|▍         | 515/11704 [04:44<1:31:45,  2.03it/s]
  4%|▍         | 516/11704 [04:45<1:31:43,  2.03it/s]
  4%|▍         | 517/11704 [04:45<1:31:48,  2.03it/s]
  4%|▍         | 518/11704 [04:46<1:32:34,  2.01it/s]
  4%|▍         | 519/11704 [04:46<1:32:16,  2.02it/s]
  4%|▍         | 520/11704 [04:47<1:31:57,  2.03it/s]
  4%|▍         | 521/11704 [04:47<1:31:50,  2.03it/s]
  4%|▍         | 522/11704 [04:48<1:32:03,  2.02it/s]
  4%|▍         | 523/11704 [04:48<1:31:50,  2.03it/s]
  4%|▍         | 524/11704 [04:49<1:31:45,  2.03it/s]
  4%|▍         | 525/11704 [04:49<1:31:44,  2.03it/s]{'loss': 4.221, 'grad_norm': 0.5851206183433533, 'learning_rate': 0.00044833475661827497, 'epoch': 0.63}
+
                                                     
  4%|▍         | 525/11704 [04:49<1:31:44,  2.03it/s]
  4%|▍         | 526/11704 [04:50<1:31:38,  2.03it/s]
  5%|▍         | 527/11704 [04:50<1:31:31,  2.04it/s]
  5%|▍         | 528/11704 [04:51<1:31:36,  2.03it/s]
  5%|▍         | 529/11704 [04:51<1:31:34,  2.03it/s]
  5%|▍         | 530/11704 [04:52<1:31:33,  2.03it/s]
  5%|▍         | 531/11704 [04:52<1:31:38,  2.03it/s]
  5%|▍         | 532/11704 [04:53<1:31:36,  2.03it/s]
  5%|▍         | 533/11704 [04:53<1:31:28,  2.04it/s]
  5%|▍         | 534/11704 [04:53<1:31:30,  2.03it/s]
  5%|▍         | 535/11704 [04:54<1:31:34,  2.03it/s]
  5%|▍         | 536/11704 [04:54<1:31:33,  2.03it/s]
  5%|▍         | 537/11704 [04:55<1:31:29,  2.03it/s]
  5%|▍         | 538/11704 [04:55<1:31:32,  2.03it/s]
  5%|▍         | 539/11704 [04:56<1:31:29,  2.03it/s]
  5%|▍         | 540/11704 [04:56<1:31:23,  2.04it/s]
  5%|▍         | 541/11704 [04:57<1:31:26,  2.03it/s]
  5%|▍         | 542/11704 [04:57<1:31:21,  2.04it/s]
  5%|▍         | 543/11704 [04:58<1:31:19,  2.04it/s]
  5%|▍         | 544/11704 [04:58<1:31:27,  2.03it/s]
  5%|▍         | 545/11704 [04:59<1:31:24,  2.03it/s]
  5%|▍         | 546/11704 [04:59<1:31:18,  2.04it/s]
  5%|▍         | 547/11704 [05:00<1:31:20,  2.04it/s]
  5%|▍         | 548/11704 [05:00<1:31:17,  2.04it/s]
  5%|▍         | 549/11704 [05:01<1:31:09,  2.04it/s]
  5%|▍         | 550/11704 [05:01<1:31:18,  2.04it/s]{'loss': 4.1699, 'grad_norm': 0.8667837381362915, 'learning_rate': 0.00046968403074295473, 'epoch': 0.66}
+
                                                     
  5%|▍         | 550/11704 [05:01<1:31:18,  2.04it/s]
  5%|▍         | 551/11704 [05:02<1:31:19,  2.04it/s]
  5%|▍         | 552/11704 [05:02<1:31:17,  2.04it/s]
  5%|▍         | 553/11704 [05:03<1:31:25,  2.03it/s]
  5%|▍         | 554/11704 [05:03<1:31:22,  2.03it/s]
  5%|▍         | 555/11704 [05:04<1:31:22,  2.03it/s]
  5%|▍         | 556/11704 [05:04<1:31:25,  2.03it/s]
  5%|▍         | 557/11704 [05:05<1:31:21,  2.03it/s]
  5%|▍         | 558/11704 [05:05<1:31:20,  2.03it/s]
  5%|▍         | 559/11704 [05:06<1:31:21,  2.03it/s]
  5%|▍         | 560/11704 [05:06<1:31:22,  2.03it/s]
  5%|▍         | 561/11704 [05:07<1:31:22,  2.03it/s]
  5%|▍         | 562/11704 [05:07<1:31:17,  2.03it/s]
  5%|▍         | 563/11704 [05:08<1:31:13,  2.04it/s]
  5%|▍         | 564/11704 [05:08<1:31:18,  2.03it/s]
  5%|▍         | 565/11704 [05:09<1:31:13,  2.03it/s]
  5%|▍         | 566/11704 [05:09<1:31:07,  2.04it/s]
  5%|▍         | 567/11704 [05:10<1:31:11,  2.04it/s]
  5%|▍         | 568/11704 [05:10<1:31:09,  2.04it/s]
  5%|▍         | 569/11704 [05:11<1:31:03,  2.04it/s]
  5%|▍         | 570/11704 [05:11<1:31:11,  2.03it/s]
  5%|▍         | 571/11704 [05:12<1:31:07,  2.04it/s]
  5%|▍         | 572/11704 [05:12<1:31:04,  2.04it/s]
  5%|▍         | 573/11704 [05:13<1:31:09,  2.04it/s]
  5%|▍         | 574/11704 [05:13<1:31:05,  2.04it/s]
  5%|▍         | 575/11704 [05:14<1:31:05,  2.04it/s]{'loss': 4.1233, 'grad_norm': 0.720784604549408, 'learning_rate': 0.0004910333048676345, 'epoch': 0.69}
+
                                                     
  5%|▍         | 575/11704 [05:14<1:31:05,  2.04it/s]
  5%|▍         | 576/11704 [05:14<1:31:14,  2.03it/s]
  5%|▍         | 577/11704 [05:15<1:31:06,  2.04it/s]
  5%|▍         | 578/11704 [05:15<1:31:06,  2.04it/s]
  5%|▍         | 579/11704 [05:16<1:31:10,  2.03it/s]
  5%|▍         | 580/11704 [05:16<1:31:03,  2.04it/s]
  5%|▍         | 581/11704 [05:17<1:31:00,  2.04it/s]
  5%|▍         | 582/11704 [05:17<1:31:08,  2.03it/s]
  5%|▍         | 583/11704 [05:18<1:31:06,  2.03it/s]
  5%|▍         | 584/11704 [05:18<1:31:09,  2.03it/s]
  5%|▍         | 585/11704 [05:19<1:31:07,  2.03it/s]
  5%|▌         | 586/11704 [05:19<1:31:03,  2.04it/s]
  5%|▌         | 587/11704 [05:20<1:31:06,  2.03it/s]
  5%|▌         | 588/11704 [05:20<1:31:03,  2.03it/s]
  5%|▌         | 589/11704 [05:21<1:30:56,  2.04it/s]
  5%|▌         | 590/11704 [05:21<1:31:02,  2.03it/s]
  5%|▌         | 591/11704 [05:21<1:31:02,  2.03it/s]
  5%|▌         | 592/11704 [05:22<1:30:59,  2.04it/s]
  5%|▌         | 593/11704 [05:22<1:31:05,  2.03it/s]
  5%|▌         | 594/11704 [05:23<1:31:00,  2.03it/s]
  5%|▌         | 595/11704 [05:23<1:30:57,  2.04it/s]
  5%|▌         | 596/11704 [05:24<1:30:54,  2.04it/s]
  5%|▌         | 597/11704 [05:24<1:30:55,  2.04it/s]
  5%|▌         | 598/11704 [05:25<1:30:59,  2.03it/s]
  5%|▌         | 599/11704 [05:25<1:31:07,  2.03it/s]
  5%|▌         | 600/11704 [05:26<1:31:00,  2.03it/s]{'loss': 4.0671, 'grad_norm': 0.5922385454177856, 'learning_rate': 0.0005123825789923142, 'epoch': 0.72}
+
                                                     
  5%|▌         | 600/11704 [05:26<1:31:00,  2.03it/s]
  5%|▌         | 601/11704 [05:26<1:31:07,  2.03it/s]
  5%|▌         | 602/11704 [05:27<1:31:07,  2.03it/s]
  5%|▌         | 603/11704 [05:27<1:30:58,  2.03it/s]
  5%|▌         | 604/11704 [05:28<1:31:02,  2.03it/s]
  5%|▌         | 605/11704 [05:28<1:30:57,  2.03it/s]
  5%|▌         | 606/11704 [05:29<1:30:54,  2.03it/s]
  5%|▌         | 607/11704 [05:29<1:31:00,  2.03it/s]
  5%|▌         | 608/11704 [05:30<1:30:53,  2.03it/s]
  5%|▌         | 609/11704 [05:30<1:30:51,  2.04it/s]
  5%|▌         | 610/11704 [05:31<1:30:58,  2.03it/s]
  5%|▌         | 611/11704 [05:31<1:30:50,  2.04it/s]
  5%|▌         | 612/11704 [05:32<1:30:44,  2.04it/s]
  5%|▌         | 613/11704 [05:32<1:30:56,  2.03it/s]
  5%|▌         | 614/11704 [05:33<1:30:48,  2.04it/s]
  5%|▌         | 615/11704 [05:33<1:39:03,  1.87it/s]
  5%|▌         | 616/11704 [05:34<1:44:26,  1.77it/s]
  5%|▌         | 617/11704 [05:35<1:40:15,  1.84it/s]
  5%|▌         | 618/11704 [05:35<1:37:27,  1.90it/s]
  5%|▌         | 619/11704 [05:36<1:35:28,  1.93it/s]
  5%|▌         | 620/11704 [05:36<1:33:54,  1.97it/s]
  5%|▌         | 621/11704 [05:37<1:33:01,  1.99it/s]
  5%|▌         | 622/11704 [05:37<1:32:24,  2.00it/s]
  5%|▌         | 623/11704 [05:38<1:31:51,  2.01it/s]
  5%|▌         | 624/11704 [05:38<1:31:28,  2.02it/s]
  5%|▌         | 625/11704 [05:38<1:31:20,  2.02it/s]{'loss': 4.0331, 'grad_norm': 0.5989447236061096, 'learning_rate': 0.000533731853116994, 'epoch': 0.75}
+
                                                     
  5%|▌         | 625/11704 [05:39<1:31:20,  2.02it/s]
  5%|▌         | 626/11704 [05:39<1:31:10,  2.03it/s]
  5%|▌         | 627/11704 [05:39<1:30:59,  2.03it/s]
  5%|▌         | 628/11704 [05:40<1:31:00,  2.03it/s]
  5%|▌         | 629/11704 [05:40<1:30:50,  2.03it/s]
  5%|▌         | 630/11704 [05:41<1:30:48,  2.03it/s]
  5%|▌         | 631/11704 [05:41<1:30:49,  2.03it/s]
  5%|▌         | 632/11704 [05:42<1:30:45,  2.03it/s]
  5%|▌         | 633/11704 [05:42<1:30:51,  2.03it/s]
  5%|▌         | 634/11704 [05:43<1:30:44,  2.03it/s]
  5%|▌         | 635/11704 [05:43<1:30:39,  2.04it/s]
  5%|▌         | 636/11704 [05:44<1:30:46,  2.03it/s]
  5%|▌         | 637/11704 [05:44<1:30:45,  2.03it/s]
  5%|▌         | 638/11704 [05:45<1:30:43,  2.03it/s]
  5%|▌         | 639/11704 [05:45<1:30:43,  2.03it/s]
  5%|▌         | 640/11704 [05:46<1:30:38,  2.03it/s]
  5%|▌         | 641/11704 [05:46<1:30:40,  2.03it/s]
  5%|▌         | 642/11704 [05:47<1:30:42,  2.03it/s]
  5%|▌         | 643/11704 [05:47<1:30:39,  2.03it/s]
  6%|▌         | 644/11704 [05:48<1:30:44,  2.03it/s]
  6%|▌         | 645/11704 [05:48<1:30:35,  2.03it/s]
  6%|▌         | 646/11704 [05:49<1:30:40,  2.03it/s]
  6%|▌         | 647/11704 [05:49<1:30:42,  2.03it/s]
  6%|▌         | 648/11704 [05:50<1:30:35,  2.03it/s]
  6%|▌         | 649/11704 [05:50<1:30:36,  2.03it/s]
  6%|▌         | 650/11704 [05:51<1:30:42,  2.03it/s]{'loss': 3.9926, 'grad_norm': 0.630885899066925, 'learning_rate': 0.0005550811272416738, 'epoch': 0.78}
+
                                                     
  6%|▌         | 650/11704 [05:51<1:30:42,  2.03it/s]
  6%|▌         | 651/11704 [05:51<1:30:40,  2.03it/s]
  6%|▌         | 652/11704 [05:52<1:30:43,  2.03it/s]
  6%|▌         | 653/11704 [05:52<1:30:36,  2.03it/s]
  6%|▌         | 654/11704 [05:53<1:30:33,  2.03it/s]
  6%|▌         | 655/11704 [05:53<1:30:43,  2.03it/s]
  6%|▌         | 656/11704 [05:54<1:30:37,  2.03it/s]
  6%|▌         | 657/11704 [05:54<1:30:37,  2.03it/s]
  6%|▌         | 658/11704 [05:55<1:30:42,  2.03it/s]
  6%|▌         | 659/11704 [05:55<1:30:42,  2.03it/s]
  6%|▌         | 660/11704 [05:56<1:30:41,  2.03it/s]
  6%|▌         | 661/11704 [05:56<1:30:43,  2.03it/s]
  6%|▌         | 662/11704 [05:57<1:30:33,  2.03it/s]
  6%|▌         | 663/11704 [05:57<1:30:33,  2.03it/s]
  6%|▌         | 664/11704 [05:58<1:30:35,  2.03it/s]
  6%|▌         | 665/11704 [05:58<1:30:35,  2.03it/s]
  6%|▌         | 666/11704 [05:59<1:30:29,  2.03it/s]
  6%|▌         | 667/11704 [05:59<1:30:36,  2.03it/s]
  6%|▌         | 668/11704 [06:00<1:30:36,  2.03it/s]
  6%|▌         | 669/11704 [06:00<1:30:32,  2.03it/s]
  6%|▌         | 670/11704 [06:01<1:30:35,  2.03it/s]
  6%|▌         | 671/11704 [06:01<1:30:30,  2.03it/s]
  6%|▌         | 672/11704 [06:02<1:30:28,  2.03it/s]
  6%|▌         | 673/11704 [06:02<1:30:26,  2.03it/s]
  6%|▌         | 674/11704 [06:03<1:30:24,  2.03it/s]
  6%|▌         | 675/11704 [06:03<1:30:22,  2.03it/s]{'loss': 3.9532, 'grad_norm': 0.6419950723648071, 'learning_rate': 0.0005764304013663536, 'epoch': 0.81}
+
                                                     
  6%|▌         | 675/11704 [06:03<1:30:22,  2.03it/s]
  6%|▌         | 676/11704 [06:04<1:30:31,  2.03it/s]
  6%|▌         | 677/11704 [06:04<1:30:28,  2.03it/s]
  6%|▌         | 678/11704 [06:05<1:30:33,  2.03it/s]
  6%|▌         | 679/11704 [06:05<1:30:34,  2.03it/s]
  6%|▌         | 680/11704 [06:06<1:30:24,  2.03it/s]
  6%|▌         | 681/11704 [06:06<1:30:30,  2.03it/s]
  6%|▌         | 682/11704 [06:07<1:30:29,  2.03it/s]
  6%|▌         | 683/11704 [06:07<1:30:31,  2.03it/s]
  6%|▌         | 684/11704 [06:08<1:30:31,  2.03it/s]
  6%|▌         | 685/11704 [06:08<1:30:25,  2.03it/s]
  6%|▌         | 686/11704 [06:09<1:30:28,  2.03it/s]
  6%|▌         | 687/11704 [06:09<1:30:25,  2.03it/s]
  6%|▌         | 688/11704 [06:10<1:30:24,  2.03it/s]
  6%|▌         | 689/11704 [06:10<1:30:25,  2.03it/s]
  6%|▌         | 690/11704 [06:10<1:30:24,  2.03it/s]
  6%|▌         | 691/11704 [06:11<1:30:24,  2.03it/s]
  6%|▌         | 692/11704 [06:11<1:30:19,  2.03it/s]
  6%|▌         | 693/11704 [06:12<1:30:15,  2.03it/s]
  6%|▌         | 694/11704 [06:12<1:30:20,  2.03it/s]
  6%|▌         | 695/11704 [06:13<1:30:17,  2.03it/s]
  6%|▌         | 696/11704 [06:13<1:30:23,  2.03it/s]
  6%|▌         | 697/11704 [06:14<1:30:18,  2.03it/s]
  6%|▌         | 698/11704 [06:14<1:30:13,  2.03it/s]
  6%|▌         | 699/11704 [06:15<1:30:15,  2.03it/s]
  6%|▌         | 700/11704 [06:15<1:30:13,  2.03it/s]{'loss': 3.9036, 'grad_norm': 0.5995571613311768, 'learning_rate': 0.0005977796754910333, 'epoch': 0.84}
+
                                                     
  6%|▌         | 700/11704 [06:15<1:30:13,  2.03it/s]
  6%|▌         | 701/11704 [06:16<1:30:21,  2.03it/s]
  6%|▌         | 702/11704 [06:16<1:30:22,  2.03it/s]
  6%|▌         | 703/11704 [06:17<1:30:16,  2.03it/s]
  6%|▌         | 704/11704 [06:17<1:30:19,  2.03it/s]
  6%|▌         | 705/11704 [06:18<1:30:16,  2.03it/s]
  6%|▌         | 706/11704 [06:18<1:30:18,  2.03it/s]
  6%|▌         | 707/11704 [06:19<1:30:13,  2.03it/s]
  6%|▌         | 708/11704 [06:19<1:30:10,  2.03it/s]
  6%|▌         | 709/11704 [06:20<1:30:16,  2.03it/s]
  6%|▌         | 710/11704 [06:20<1:30:10,  2.03it/s]
  6%|▌         | 711/11704 [06:21<1:30:22,  2.03it/s]
  6%|▌         | 712/11704 [06:21<1:30:14,  2.03it/s]
  6%|▌         | 713/11704 [06:22<1:30:10,  2.03it/s]
  6%|▌         | 714/11704 [06:22<1:30:16,  2.03it/s]
  6%|▌         | 715/11704 [06:23<1:30:10,  2.03it/s]
  6%|▌         | 716/11704 [06:23<1:30:13,  2.03it/s]
  6%|▌         | 717/11704 [06:24<1:30:17,  2.03it/s]
  6%|▌         | 718/11704 [06:24<1:30:09,  2.03it/s]
  6%|▌         | 719/11704 [06:25<1:30:13,  2.03it/s]
  6%|▌         | 720/11704 [06:25<1:30:05,  2.03it/s]
  6%|▌         | 721/11704 [06:26<1:30:09,  2.03it/s]
  6%|▌         | 722/11704 [06:26<1:30:09,  2.03it/s]
  6%|▌         | 723/11704 [06:27<1:30:08,  2.03it/s]
  6%|▌         | 724/11704 [06:27<1:30:10,  2.03it/s]
  6%|▌         | 725/11704 [06:28<1:30:07,  2.03it/s]{'loss': 3.8761, 'grad_norm': 0.5167131423950195, 'learning_rate': 0.000619128949615713, 'epoch': 0.87}
+
                                                     
  6%|▌         | 725/11704 [06:28<1:30:07,  2.03it/s]
  6%|▌         | 726/11704 [06:28<1:30:10,  2.03it/s]
  6%|▌         | 727/11704 [06:29<1:30:07,  2.03it/s]
  6%|▌         | 728/11704 [06:29<1:30:06,  2.03it/s]
  6%|▌         | 729/11704 [06:30<1:30:10,  2.03it/s]
  6%|▌         | 730/11704 [06:30<1:30:08,  2.03it/s]
  6%|▌         | 731/11704 [06:31<1:30:10,  2.03it/s]
  6%|▋         | 732/11704 [06:31<1:30:04,  2.03it/s]
  6%|▋         | 733/11704 [06:32<1:30:01,  2.03it/s]
  6%|▋         | 734/11704 [06:32<1:30:04,  2.03it/s]
  6%|▋         | 735/11704 [06:33<1:29:58,  2.03it/s]
  6%|▋         | 736/11704 [06:33<1:30:04,  2.03it/s]
  6%|▋         | 737/11704 [06:34<1:30:03,  2.03it/s]
  6%|▋         | 738/11704 [06:34<1:30:02,  2.03it/s]
  6%|▋         | 739/11704 [06:35<1:30:03,  2.03it/s]
  6%|▋         | 740/11704 [06:35<1:29:57,  2.03it/s]
  6%|▋         | 741/11704 [06:36<1:30:03,  2.03it/s]
  6%|▋         | 742/11704 [06:36<1:30:01,  2.03it/s]
  6%|▋         | 743/11704 [06:37<1:29:59,  2.03it/s]
  6%|▋         | 744/11704 [06:37<1:30:00,  2.03it/s]
  6%|▋         | 745/11704 [06:38<1:29:54,  2.03it/s]
  6%|▋         | 746/11704 [06:38<1:30:01,  2.03it/s]
  6%|▋         | 747/11704 [06:39<1:30:00,  2.03it/s]
  6%|▋         | 748/11704 [06:39<1:29:55,  2.03it/s]
  6%|▋         | 749/11704 [06:40<1:30:01,  2.03it/s]
  6%|▋         | 750/11704 [06:40<1:29:54,  2.03it/s]{'loss': 3.851, 'grad_norm': 0.5299991369247437, 'learning_rate': 0.0006404782237403928, 'epoch': 0.9}
+
                                                     
  6%|▋         | 750/11704 [06:40<1:29:54,  2.03it/s]
  6%|▋         | 751/11704 [06:41<1:29:57,  2.03it/s]
  6%|▋         | 752/11704 [06:41<1:30:00,  2.03it/s]
  6%|▋         | 753/11704 [06:42<1:29:52,  2.03it/s]
  6%|▋         | 754/11704 [06:42<1:30:01,  2.03it/s]
  6%|▋         | 755/11704 [06:43<1:29:53,  2.03it/s]
  6%|▋         | 756/11704 [06:43<1:29:51,  2.03it/s]
  6%|▋         | 757/11704 [06:43<1:29:52,  2.03it/s]
  6%|▋         | 758/11704 [06:44<1:29:51,  2.03it/s]
  6%|▋         | 759/11704 [06:44<1:29:56,  2.03it/s]
  6%|▋         | 760/11704 [06:45<1:29:50,  2.03it/s]
  7%|▋         | 761/11704 [06:45<1:29:48,  2.03it/s]
  7%|▋         | 762/11704 [06:46<1:29:55,  2.03it/s]
  7%|▋         | 763/11704 [06:46<1:29:48,  2.03it/s]
  7%|▋         | 764/11704 [06:47<1:29:48,  2.03it/s]
  7%|▋         | 765/11704 [06:47<1:29:54,  2.03it/s]
  7%|▋         | 766/11704 [06:48<1:29:47,  2.03it/s]
  7%|▋         | 767/11704 [06:48<1:29:56,  2.03it/s]
  7%|▋         | 768/11704 [06:49<1:29:52,  2.03it/s]
  7%|▋         | 769/11704 [06:49<1:29:52,  2.03it/s]
  7%|▋         | 770/11704 [06:50<1:29:50,  2.03it/s]
  7%|▋         | 771/11704 [06:50<1:29:44,  2.03it/s]
  7%|▋         | 772/11704 [06:51<1:29:44,  2.03it/s]
  7%|▋         | 773/11704 [06:51<1:29:36,  2.03it/s]
  7%|▋         | 774/11704 [06:52<1:29:42,  2.03it/s]
  7%|▋         | 775/11704 [06:52<1:29:42,  2.03it/s]{'loss': 3.7964, 'grad_norm': 0.5229047536849976, 'learning_rate': 0.0006618274978650726, 'epoch': 0.93}
+
                                                     
  7%|▋         | 775/11704 [06:52<1:29:42,  2.03it/s]
  7%|▋         | 776/11704 [06:53<1:29:43,  2.03it/s]
  7%|▋         | 777/11704 [06:53<1:29:50,  2.03it/s]
  7%|▋         | 778/11704 [06:54<1:29:41,  2.03it/s]
  7%|▋         | 779/11704 [06:54<1:29:43,  2.03it/s]
  7%|▋         | 780/11704 [06:55<1:29:41,  2.03it/s]
  7%|▋         | 781/11704 [06:55<1:29:40,  2.03it/s]
  7%|▋         | 782/11704 [06:56<1:29:39,  2.03it/s]
  7%|▋         | 783/11704 [06:56<1:29:35,  2.03it/s]
  7%|▋         | 784/11704 [06:57<1:29:34,  2.03it/s]
  7%|▋         | 785/11704 [06:57<1:29:40,  2.03it/s]
  7%|▋         | 786/11704 [06:58<1:29:40,  2.03it/s]
  7%|▋         | 787/11704 [06:58<1:29:44,  2.03it/s]
  7%|▋         | 788/11704 [06:59<1:29:40,  2.03it/s]
  7%|▋         | 789/11704 [06:59<1:29:42,  2.03it/s]
  7%|▋         | 790/11704 [07:00<1:29:42,  2.03it/s]
  7%|▋         | 791/11704 [07:00<1:29:41,  2.03it/s]
  7%|▋         | 792/11704 [07:01<1:29:39,  2.03it/s]
  7%|▋         | 793/11704 [07:01<1:29:34,  2.03it/s]
  7%|▋         | 794/11704 [07:02<1:29:26,  2.03it/s]
  7%|▋         | 795/11704 [07:02<1:29:29,  2.03it/s]
  7%|▋         | 796/11704 [07:03<1:29:29,  2.03it/s]
  7%|▋         | 797/11704 [07:03<1:29:29,  2.03it/s]
  7%|▋         | 798/11704 [07:04<1:29:32,  2.03it/s]
  7%|▋         | 799/11704 [07:04<1:29:33,  2.03it/s]
  7%|▋         | 800/11704 [07:05<1:29:32,  2.03it/s]{'loss': 3.7888, 'grad_norm': 0.48627805709838867, 'learning_rate': 0.0006831767719897524, 'epoch': 0.96}
+
                                                     
  7%|▋         | 800/11704 [07:05<1:29:32,  2.03it/s]
  7%|▋         | 801/11704 [07:05<1:29:38,  2.03it/s]
  7%|▋         | 802/11704 [07:06<1:29:34,  2.03it/s]
  7%|▋         | 803/11704 [07:06<1:29:34,  2.03it/s]
  7%|▋         | 804/11704 [07:07<1:29:28,  2.03it/s]
  7%|▋         | 805/11704 [07:07<1:29:23,  2.03it/s]
  7%|▋         | 806/11704 [07:08<1:29:21,  2.03it/s]
  7%|▋         | 807/11704 [07:08<1:29:29,  2.03it/s]
  7%|▋         | 808/11704 [07:09<1:29:27,  2.03it/s]
  7%|▋         | 809/11704 [07:09<1:29:26,  2.03it/s]
  7%|▋         | 810/11704 [07:10<1:29:27,  2.03it/s]
  7%|▋         | 811/11704 [07:10<1:29:23,  2.03it/s]
  7%|▋         | 812/11704 [07:11<1:29:25,  2.03it/s]
  7%|▋         | 813/11704 [07:11<1:29:24,  2.03it/s]
  7%|▋         | 814/11704 [07:12<1:29:16,  2.03it/s]
  7%|▋         | 815/11704 [07:12<1:29:08,  2.04it/s]
  7%|▋         | 816/11704 [07:13<1:29:13,  2.03it/s]
  7%|▋         | 817/11704 [07:13<1:29:13,  2.03it/s]
  7%|▋         | 818/11704 [07:14<1:29:15,  2.03it/s]
  7%|▋         | 819/11704 [07:14<1:29:19,  2.03it/s]
  7%|▋         | 820/11704 [07:15<1:29:12,  2.03it/s]
  7%|▋         | 821/11704 [07:15<1:29:15,  2.03it/s]
  7%|▋         | 822/11704 [07:16<1:29:14,  2.03it/s]
  7%|▋         | 823/11704 [07:16<1:29:12,  2.03it/s]
  7%|▋         | 824/11704 [07:16<1:29:17,  2.03it/s]
  7%|▋         | 825/11704 [07:17<1:29:15,  2.03it/s]{'loss': 3.7506, 'grad_norm': 0.5719705820083618, 'learning_rate': 0.0007045260461144322, 'epoch': 0.99}
+
                                                     
  7%|▋         | 825/11704 [07:17<1:29:15,  2.03it/s]
  7%|▋         | 826/11704 [07:17<1:29:22,  2.03it/s]
  7%|▋         | 827/11704 [07:18<1:29:19,  2.03it/s]
  7%|▋         | 828/11704 [07:18<1:29:13,  2.03it/s]
  7%|▋         | 829/11704 [07:19<1:29:19,  2.03it/s]
  7%|▋         | 830/11704 [07:19<1:29:18,  2.03it/s]
  7%|▋         | 831/11704 [07:20<1:29:20,  2.03it/s]
  7%|▋         | 832/11704 [07:20<1:29:17,  2.03it/s]
  7%|▋         | 833/11704 [07:21<1:29:13,  2.03it/s]
  7%|▋         | 834/11704 [07:21<1:29:20,  2.03it/s]
  7%|▋         | 835/11704 [07:22<1:29:11,  2.03it/s]
  7%|▋         | 836/11704 [07:22<1:31:45,  1.97it/s]
  7%|▋         | 837/11704 [07:34<11:50:27,  3.92s/it]
  7%|▋         | 838/11704 [07:35<8:44:46,  2.90s/it] 
  7%|▋         | 839/11704 [07:35<6:33:59,  2.18s/it]
  7%|▋         | 840/11704 [07:36<5:02:24,  1.67s/it]
  7%|▋         | 841/11704 [07:36<3:58:30,  1.32s/it]
  7%|▋         | 842/11704 [07:37<3:13:48,  1.07s/it]
  7%|▋         | 843/11704 [07:37<2:42:26,  1.11it/s]
  7%|▋         | 844/11704 [07:38<2:20:22,  1.29it/s]
  7%|▋         | 845/11704 [07:38<2:05:46,  1.44it/s]
  7%|▋         | 846/11704 [07:39<1:54:47,  1.58it/s]
  7%|▋         | 847/11704 [07:39<1:47:02,  1.69it/s]
  7%|▋         | 848/11704 [07:40<1:41:37,  1.78it/s]
  7%|▋         | 849/11704 [07:40<1:37:49,  1.85it/s]
  7%|▋         | 850/11704 [07:41<1:35:06,  1.90it/s]{'loss': 3.7125, 'grad_norm': 0.48182985186576843, 'learning_rate': 0.0007258753202391119, 'epoch': 1.02}
+
                                                     
  7%|▋         | 850/11704 [07:41<1:35:06,  1.90it/s]
  7%|▋         | 851/11704 [07:41<1:33:29,  1.93it/s]
  7%|▋         | 852/11704 [07:42<1:32:09,  1.96it/s]
  7%|▋         | 853/11704 [07:42<1:31:05,  1.99it/s]
  7%|▋         | 854/11704 [07:43<1:30:25,  2.00it/s]
  7%|▋         | 855/11704 [07:43<1:29:58,  2.01it/s]
  7%|▋         | 856/11704 [07:44<1:29:45,  2.01it/s]
  7%|▋         | 857/11704 [07:44<1:29:20,  2.02it/s]
  7%|▋         | 858/11704 [07:45<1:29:19,  2.02it/s]
  7%|▋         | 859/11704 [07:45<1:29:09,  2.03it/s]
  7%|▋         | 860/11704 [07:46<1:29:11,  2.03it/s]
  7%|▋         | 861/11704 [07:46<1:29:02,  2.03it/s]
  7%|▋         | 862/11704 [07:47<1:28:59,  2.03it/s]
  7%|▋         | 863/11704 [07:47<1:28:54,  2.03it/s]
  7%|▋         | 864/11704 [07:48<1:28:51,  2.03it/s]
  7%|▋         | 865/11704 [07:48<1:28:48,  2.03it/s]
  7%|▋         | 866/11704 [07:49<1:28:57,  2.03it/s]
  7%|▋         | 867/11704 [07:49<1:28:59,  2.03it/s]
  7%|▋         | 868/11704 [07:50<1:28:51,  2.03it/s]
  7%|▋         | 869/11704 [07:50<1:28:52,  2.03it/s]
  7%|▋         | 870/11704 [07:51<1:28:50,  2.03it/s]
  7%|▋         | 871/11704 [07:51<1:28:46,  2.03it/s]
  7%|▋         | 872/11704 [07:52<1:28:53,  2.03it/s]

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9509fc878acd0defe5adfeb20bab4c79a7a8244d939dbf9aedc248e6079ebae0
+size 5240