diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..3ca2d1e2674dbfdfaad4c596e895f19213c47f12 100644 --- a/.gitattributes +++ b/.gitattributes @@ -33,3 +33,10 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.zip filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text +checkpoint-1207/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-2414/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-3621/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-4828/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-6035/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-7242/tokenizer.json filter=lfs diff=lfs merge=lfs -text +tokenizer.json filter=lfs diff=lfs merge=lfs -text diff --git a/README.md b/README.md new file mode 100644 index 0000000000000000000000000000000000000000..c6e26c277eecfdb22fe4190ccdd8d967900c381c --- /dev/null +++ b/README.md @@ -0,0 +1,143 @@ +--- +library_name: peft +license: llama3.1 +base_model: meta-llama/Llama-3.1-8B +tags: +- generated_from_trainer +datasets: +- ugaoo/detailed_response_dataset_doctor_patient +model-index: +- name: out/meta_llama_Llama_3.1_8B_ugaoo_detailed_response_dataset_doctor_patient + results: [] +--- + + + +[Built with Axolotl](https://github.com/axolotl-ai-cloud/axolotl) +
See axolotl config + +axolotl version: `0.8.0.dev0` +```yaml +base_model: meta-llama/Llama-3.1-8B +model_type: AutoModelForCausalLM +tokenizer_type: AutoTokenizer +trust_remote_code: true + +load_in_8bit: false +load_in_4bit: true +strict: false + +datasets: + - path: ugaoo/detailed_response_dataset_doctor_patient + type: alpaca +val_set_size: 0 +output_dir: ./out/meta_llama_Llama_3.1_8B_ugaoo_detailed_response_dataset_doctor_patient + +sequence_len: 4000 +sample_packing: true +pad_to_sequence_len: true + +adapter: qlora +lora_r: 256 +lora_alpha: 512 +lora_dropout: 0.05 +lora_target_linear: true +lora_target_modules: + - q_proj + - k_proj + - v_proj + - o_proj + - up_proj + - down_proj + - gate_proj +lora_modules_to_save: + - embed_tokens + - lm_head + +wandb_project: cosmosearch +wandb_entity: +wandb_watch: +wandb_name: meta_llama_Llama_3.1_8B_ugaoo_detailed_response_dataset_doctor_patient +wandb_log_model: + +gradient_accumulation_steps: 3 +micro_batch_size: 4 +num_epochs: 6 +optimizer: adamw_torch +lr_scheduler: cosine +learning_rate: 5e-6 + +train_on_inputs: false +group_by_length: false +bf16: auto +fp16: false +tf32: false + +gradient_checkpointing: true +early_stopping_patience: +resume_from_checkpoint: +logging_steps: 1 +xformers_attention: +flash_attention: true + +warmup_steps: 100 +evals_per_epoch: 6 +eval_table_size: +saves_per_epoch: 1 +debug: +deepspeed: +weight_decay: 0.0 +fsdp: +fsdp_config: +save_total_limit: 6 +special_tokens: + pad_token: <|end_of_text|> +``` + +

+ +# out/meta_llama_Llama_3.1_8B_ugaoo_detailed_response_dataset_doctor_patient + +This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on the ugaoo/detailed_response_dataset_doctor_patient dataset. + +## Model description + +More information needed + +## Intended uses & limitations + +More information needed + +## Training and evaluation data + +More information needed + +## Training procedure + +### Training hyperparameters + +The following hyperparameters were used during training: +- learning_rate: 5e-06 +- train_batch_size: 4 +- eval_batch_size: 4 +- seed: 42 +- distributed_type: multi-GPU +- gradient_accumulation_steps: 3 +- total_train_batch_size: 12 +- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments +- lr_scheduler_type: cosine +- lr_scheduler_warmup_steps: 100 +- num_epochs: 6.0 + +### Training results + + + +### Framework versions + +- PEFT 0.14.0 +- Transformers 4.49.0 +- Pytorch 2.5.1+cu124 +- Datasets 3.2.0 +- Tokenizers 0.21.0 \ No newline at end of file diff --git a/adapter_config.json b/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d6fca8a067dce19d599e7f1e24bea8019968fe0f --- /dev/null +++ b/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.1-8B", + "bias": "none", + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "up_proj", + "q_proj", + "k_proj", + "o_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/adapter_model.safetensors b/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8df017a42ac766c44c8af48fbe94759ea5b038ae --- /dev/null +++ b/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:07547a0137768f7add8b98278e833c88c14aa448fb45fe6b306fccdc96439647 +size 3443586272 diff --git a/checkpoint-1207/README.md b/checkpoint-1207/README.md new file mode 100644 index 0000000000000000000000000000000000000000..049d467664ca6172b7ffbe6ba60b3eac7479cac4 --- /dev/null +++ b/checkpoint-1207/README.md @@ -0,0 +1,202 @@ +--- +base_model: meta-llama/Llama-3.1-8B +library_name: peft +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.14.0 \ No newline at end of file diff --git a/checkpoint-1207/adapter_config.json b/checkpoint-1207/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d6fca8a067dce19d599e7f1e24bea8019968fe0f --- /dev/null +++ b/checkpoint-1207/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.1-8B", + "bias": "none", + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "up_proj", + "q_proj", + "k_proj", + "o_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-1207/adapter_model.safetensors b/checkpoint-1207/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5f975f5149bcad6f61fc0644c0147e4008545c13 --- /dev/null +++ b/checkpoint-1207/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3abf52bbc975a0e1ae0358a10b5a0d44d47e251bd769d6d610c3532dce1f792f +size 3443586272 diff --git a/checkpoint-1207/global_step1207/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-1207/global_step1207/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..d0b36c84fe29794db2469075c3fe11af640b0977 --- /dev/null +++ b/checkpoint-1207/global_step1207/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:64224c888ea1f25d4df3d34dd397745a90551147ba17013b4183dabf258dec83 +size 20661195036 diff --git a/checkpoint-1207/global_step1207/mp_rank_00_model_states.pt b/checkpoint-1207/global_step1207/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..737d7041120c7217844bf3396633c1470cee21f9 --- /dev/null +++ b/checkpoint-1207/global_step1207/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b170c6ae821f5a0908f4c52bc7675aabdfe8fe68510ba5215cfaff10c00034fd +size 3555326777 diff --git a/checkpoint-1207/latest b/checkpoint-1207/latest new file mode 100644 index 0000000000000000000000000000000000000000..589e8acf07000efa1c6870f875c4c07d84a6f5d9 --- /dev/null +++ b/checkpoint-1207/latest @@ -0,0 +1 @@ +global_step1207 \ No newline at end of file diff --git a/checkpoint-1207/rng_state.pth b/checkpoint-1207/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..5dd5025a8b04e9585f858bc561016da9b1a1738c --- /dev/null +++ b/checkpoint-1207/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b8d99e8b564fb438e9c28705f513b8a810d82cc0ad074caa2c4b42a69987818a +size 14244 diff --git a/checkpoint-1207/scheduler.pt b/checkpoint-1207/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..71a8b72b4bdadc0316189df99a17a51ac6c9b42c --- /dev/null +++ b/checkpoint-1207/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3da3ab357451ad3048ccc13d0c5679691b4af04c9091b4bc7c3120bf57dd8462 +size 1064 diff --git a/checkpoint-1207/special_tokens_map.json b/checkpoint-1207/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..e5b39b6305d89284b04934011c68dbb26bf588ca --- /dev/null +++ b/checkpoint-1207/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-1207/tokenizer.json b/checkpoint-1207/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-1207/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-1207/tokenizer_config.json b/checkpoint-1207/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..81dd14db6632ad5b35b9d447732e37ac074873a5 --- /dev/null +++ b/checkpoint-1207/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +} diff --git a/checkpoint-1207/trainer_state.json b/checkpoint-1207/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..efe60b21e494afab86ec883ee3f3358bdb088f81 --- /dev/null +++ b/checkpoint-1207/trainer_state.json @@ -0,0 +1,8482 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.999723909442297, + "eval_steps": 500, + "global_step": 1207, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0008282716731087797, + "grad_norm": 2.4795358180999756, + "learning_rate": 5.0000000000000004e-08, + "loss": 1.1215, + "step": 1 + }, + { + "epoch": 0.0016565433462175593, + "grad_norm": 2.6060028076171875, + "learning_rate": 1.0000000000000001e-07, + "loss": 1.1219, + "step": 2 + }, + { + "epoch": 0.002484815019326339, + "grad_norm": 2.750868082046509, + "learning_rate": 1.5000000000000002e-07, + "loss": 1.125, + "step": 3 + }, + { + "epoch": 0.0033130866924351186, + "grad_norm": 2.7321369647979736, + "learning_rate": 2.0000000000000002e-07, + "loss": 1.1164, + "step": 4 + }, + { + "epoch": 0.0041413583655438985, + "grad_norm": 2.8193929195404053, + "learning_rate": 2.5000000000000004e-07, + "loss": 1.1263, + "step": 5 + }, + { + "epoch": 0.004969630038652678, + "grad_norm": 2.648301839828491, + "learning_rate": 3.0000000000000004e-07, + "loss": 1.1092, + "step": 6 + }, + { + "epoch": 0.005797901711761457, + "grad_norm": 2.751993179321289, + "learning_rate": 3.5000000000000004e-07, + "loss": 1.0997, + "step": 7 + }, + { + "epoch": 0.006626173384870237, + "grad_norm": 2.592381000518799, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.112, + "step": 8 + }, + { + "epoch": 0.007454445057979017, + "grad_norm": 2.3135671615600586, + "learning_rate": 4.5000000000000003e-07, + "loss": 1.1045, + "step": 9 + }, + { + "epoch": 0.008282716731087797, + "grad_norm": 2.60491943359375, + "learning_rate": 5.000000000000001e-07, + "loss": 1.0956, + "step": 10 + }, + { + "epoch": 0.009110988404196576, + "grad_norm": 2.2435271739959717, + "learning_rate": 5.5e-07, + "loss": 1.1241, + "step": 11 + }, + { + "epoch": 0.009939260077305357, + "grad_norm": 2.4412319660186768, + "learning_rate": 6.000000000000001e-07, + "loss": 1.11, + "step": 12 + }, + { + "epoch": 0.010767531750414136, + "grad_norm": 2.2075185775756836, + "learning_rate": 6.5e-07, + "loss": 1.0878, + "step": 13 + }, + { + "epoch": 0.011595803423522915, + "grad_norm": 2.101902723312378, + "learning_rate": 7.000000000000001e-07, + "loss": 1.0863, + "step": 14 + }, + { + "epoch": 0.012424075096631695, + "grad_norm": 2.134535074234009, + "learning_rate": 7.5e-07, + "loss": 1.0735, + "step": 15 + }, + { + "epoch": 0.013252346769740474, + "grad_norm": 2.000504493713379, + "learning_rate": 8.000000000000001e-07, + "loss": 1.0875, + "step": 16 + }, + { + "epoch": 0.014080618442849255, + "grad_norm": 1.9191882610321045, + "learning_rate": 8.500000000000001e-07, + "loss": 1.1039, + "step": 17 + }, + { + "epoch": 0.014908890115958034, + "grad_norm": 1.7732031345367432, + "learning_rate": 9.000000000000001e-07, + "loss": 1.0819, + "step": 18 + }, + { + "epoch": 0.015737161789066815, + "grad_norm": 1.7775251865386963, + "learning_rate": 9.500000000000001e-07, + "loss": 1.0936, + "step": 19 + }, + { + "epoch": 0.016565433462175594, + "grad_norm": 1.6648598909378052, + "learning_rate": 1.0000000000000002e-06, + "loss": 1.0881, + "step": 20 + }, + { + "epoch": 0.017393705135284373, + "grad_norm": 1.5726720094680786, + "learning_rate": 1.0500000000000001e-06, + "loss": 1.07, + "step": 21 + }, + { + "epoch": 0.018221976808393152, + "grad_norm": 1.561417818069458, + "learning_rate": 1.1e-06, + "loss": 1.0668, + "step": 22 + }, + { + "epoch": 0.019050248481501934, + "grad_norm": 1.4953842163085938, + "learning_rate": 1.1500000000000002e-06, + "loss": 1.054, + "step": 23 + }, + { + "epoch": 0.019878520154610713, + "grad_norm": 1.4247829914093018, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.0363, + "step": 24 + }, + { + "epoch": 0.020706791827719492, + "grad_norm": 1.3483426570892334, + "learning_rate": 1.25e-06, + "loss": 1.0143, + "step": 25 + }, + { + "epoch": 0.02153506350082827, + "grad_norm": 1.3412944078445435, + "learning_rate": 1.3e-06, + "loss": 1.025, + "step": 26 + }, + { + "epoch": 0.02236333517393705, + "grad_norm": 1.2834237813949585, + "learning_rate": 1.3500000000000002e-06, + "loss": 1.0227, + "step": 27 + }, + { + "epoch": 0.02319160684704583, + "grad_norm": 1.2128241062164307, + "learning_rate": 1.4000000000000001e-06, + "loss": 1.0161, + "step": 28 + }, + { + "epoch": 0.024019878520154612, + "grad_norm": 1.1665573120117188, + "learning_rate": 1.45e-06, + "loss": 1.0034, + "step": 29 + }, + { + "epoch": 0.02484815019326339, + "grad_norm": 1.1575899124145508, + "learning_rate": 1.5e-06, + "loss": 0.9968, + "step": 30 + }, + { + "epoch": 0.02567642186637217, + "grad_norm": 1.195304274559021, + "learning_rate": 1.5500000000000002e-06, + "loss": 1.0176, + "step": 31 + }, + { + "epoch": 0.02650469353948095, + "grad_norm": 1.1321678161621094, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.0113, + "step": 32 + }, + { + "epoch": 0.027332965212589728, + "grad_norm": 1.1076338291168213, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.9971, + "step": 33 + }, + { + "epoch": 0.02816123688569851, + "grad_norm": 1.0770610570907593, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.9994, + "step": 34 + }, + { + "epoch": 0.02898950855880729, + "grad_norm": 1.0767767429351807, + "learning_rate": 1.75e-06, + "loss": 0.978, + "step": 35 + }, + { + "epoch": 0.02981778023191607, + "grad_norm": 1.054800033569336, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.9829, + "step": 36 + }, + { + "epoch": 0.030646051905024847, + "grad_norm": 1.031015157699585, + "learning_rate": 1.85e-06, + "loss": 0.97, + "step": 37 + }, + { + "epoch": 0.03147432357813363, + "grad_norm": 1.0182578563690186, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.9549, + "step": 38 + }, + { + "epoch": 0.032302595251242405, + "grad_norm": 0.9796578288078308, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.9762, + "step": 39 + }, + { + "epoch": 0.03313086692435119, + "grad_norm": 0.9956578016281128, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.9693, + "step": 40 + }, + { + "epoch": 0.03395913859745997, + "grad_norm": 0.9383260011672974, + "learning_rate": 2.05e-06, + "loss": 0.9556, + "step": 41 + }, + { + "epoch": 0.034787410270568746, + "grad_norm": 0.9372819066047668, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.9419, + "step": 42 + }, + { + "epoch": 0.03561568194367753, + "grad_norm": 0.9334004521369934, + "learning_rate": 2.15e-06, + "loss": 0.9587, + "step": 43 + }, + { + "epoch": 0.036443953616786304, + "grad_norm": 0.8916004300117493, + "learning_rate": 2.2e-06, + "loss": 0.9227, + "step": 44 + }, + { + "epoch": 0.037272225289895086, + "grad_norm": 0.8698756694793701, + "learning_rate": 2.25e-06, + "loss": 0.9275, + "step": 45 + }, + { + "epoch": 0.03810049696300387, + "grad_norm": 0.8775100708007812, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.941, + "step": 46 + }, + { + "epoch": 0.038928768636112644, + "grad_norm": 0.8763384222984314, + "learning_rate": 2.35e-06, + "loss": 0.9253, + "step": 47 + }, + { + "epoch": 0.03975704030922143, + "grad_norm": 0.8534179925918579, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.9327, + "step": 48 + }, + { + "epoch": 0.0405853119823302, + "grad_norm": 0.8432146906852722, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.9236, + "step": 49 + }, + { + "epoch": 0.041413583655438985, + "grad_norm": 0.9899975061416626, + "learning_rate": 2.5e-06, + "loss": 0.8839, + "step": 50 + }, + { + "epoch": 0.04224185532854776, + "grad_norm": 0.8370104432106018, + "learning_rate": 2.55e-06, + "loss": 0.906, + "step": 51 + }, + { + "epoch": 0.04307012700165654, + "grad_norm": 0.8207446336746216, + "learning_rate": 2.6e-06, + "loss": 0.8755, + "step": 52 + }, + { + "epoch": 0.043898398674765325, + "grad_norm": 0.805584728717804, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.9049, + "step": 53 + }, + { + "epoch": 0.0447266703478741, + "grad_norm": 0.8162860870361328, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.8993, + "step": 54 + }, + { + "epoch": 0.04555494202098288, + "grad_norm": 0.8151668906211853, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.9282, + "step": 55 + }, + { + "epoch": 0.04638321369409166, + "grad_norm": 0.7996147274971008, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.8946, + "step": 56 + }, + { + "epoch": 0.04721148536720044, + "grad_norm": 0.8223204016685486, + "learning_rate": 2.85e-06, + "loss": 0.897, + "step": 57 + }, + { + "epoch": 0.048039757040309224, + "grad_norm": 0.8117988109588623, + "learning_rate": 2.9e-06, + "loss": 0.8845, + "step": 58 + }, + { + "epoch": 0.048868028713418, + "grad_norm": 0.7902786731719971, + "learning_rate": 2.95e-06, + "loss": 0.8846, + "step": 59 + }, + { + "epoch": 0.04969630038652678, + "grad_norm": 0.804192304611206, + "learning_rate": 3e-06, + "loss": 0.8722, + "step": 60 + }, + { + "epoch": 0.05052457205963556, + "grad_norm": 0.7852040529251099, + "learning_rate": 3.05e-06, + "loss": 0.8821, + "step": 61 + }, + { + "epoch": 0.05135284373274434, + "grad_norm": 0.8045080304145813, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.8661, + "step": 62 + }, + { + "epoch": 0.05218111540585312, + "grad_norm": 0.7920243144035339, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.8712, + "step": 63 + }, + { + "epoch": 0.0530093870789619, + "grad_norm": 0.7601844072341919, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.8792, + "step": 64 + }, + { + "epoch": 0.05383765875207068, + "grad_norm": 0.7833033800125122, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.8811, + "step": 65 + }, + { + "epoch": 0.054665930425179456, + "grad_norm": 0.7691664099693298, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.8701, + "step": 66 + }, + { + "epoch": 0.05549420209828824, + "grad_norm": 0.7845340967178345, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.8929, + "step": 67 + }, + { + "epoch": 0.05632247377139702, + "grad_norm": 0.9076205492019653, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.8352, + "step": 68 + }, + { + "epoch": 0.057150745444505796, + "grad_norm": 0.7614927887916565, + "learning_rate": 3.45e-06, + "loss": 0.8766, + "step": 69 + }, + { + "epoch": 0.05797901711761458, + "grad_norm": 0.7722934484481812, + "learning_rate": 3.5e-06, + "loss": 0.8666, + "step": 70 + }, + { + "epoch": 0.058807288790723354, + "grad_norm": 0.7603748440742493, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.84, + "step": 71 + }, + { + "epoch": 0.05963556046383214, + "grad_norm": 0.7563591003417969, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.8784, + "step": 72 + }, + { + "epoch": 0.06046383213694092, + "grad_norm": 0.7835114002227783, + "learning_rate": 3.65e-06, + "loss": 0.8516, + "step": 73 + }, + { + "epoch": 0.061292103810049695, + "grad_norm": 0.7760982513427734, + "learning_rate": 3.7e-06, + "loss": 0.8265, + "step": 74 + }, + { + "epoch": 0.06212037548315848, + "grad_norm": 0.7978456020355225, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.8564, + "step": 75 + }, + { + "epoch": 0.06294864715626726, + "grad_norm": 0.7940747141838074, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.8635, + "step": 76 + }, + { + "epoch": 0.06377691882937604, + "grad_norm": 0.8055519461631775, + "learning_rate": 3.85e-06, + "loss": 0.8445, + "step": 77 + }, + { + "epoch": 0.06460519050248481, + "grad_norm": 0.7532406449317932, + "learning_rate": 3.900000000000001e-06, + "loss": 0.8772, + "step": 78 + }, + { + "epoch": 0.06543346217559359, + "grad_norm": 0.763933539390564, + "learning_rate": 3.95e-06, + "loss": 0.8574, + "step": 79 + }, + { + "epoch": 0.06626173384870238, + "grad_norm": 0.7833206057548523, + "learning_rate": 4.000000000000001e-06, + "loss": 0.8643, + "step": 80 + }, + { + "epoch": 0.06709000552181116, + "grad_norm": 0.7825527787208557, + "learning_rate": 4.05e-06, + "loss": 0.8518, + "step": 81 + }, + { + "epoch": 0.06791827719491994, + "grad_norm": 0.7749581336975098, + "learning_rate": 4.1e-06, + "loss": 0.8146, + "step": 82 + }, + { + "epoch": 0.06874654886802871, + "grad_norm": 0.8117144703865051, + "learning_rate": 4.15e-06, + "loss": 0.8719, + "step": 83 + }, + { + "epoch": 0.06957482054113749, + "grad_norm": 0.8440507054328918, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.8665, + "step": 84 + }, + { + "epoch": 0.07040309221424627, + "grad_norm": 0.7941984534263611, + "learning_rate": 4.25e-06, + "loss": 0.8492, + "step": 85 + }, + { + "epoch": 0.07123136388735506, + "grad_norm": 0.8301989436149597, + "learning_rate": 4.3e-06, + "loss": 0.8544, + "step": 86 + }, + { + "epoch": 0.07205963556046384, + "grad_norm": 0.757798969745636, + "learning_rate": 4.350000000000001e-06, + "loss": 0.8426, + "step": 87 + }, + { + "epoch": 0.07288790723357261, + "grad_norm": 0.7817673683166504, + "learning_rate": 4.4e-06, + "loss": 0.8554, + "step": 88 + }, + { + "epoch": 0.07371617890668139, + "grad_norm": 0.7609818577766418, + "learning_rate": 4.450000000000001e-06, + "loss": 0.8473, + "step": 89 + }, + { + "epoch": 0.07454445057979017, + "grad_norm": 0.8146091103553772, + "learning_rate": 4.5e-06, + "loss": 0.8317, + "step": 90 + }, + { + "epoch": 0.07537272225289895, + "grad_norm": 0.8058239817619324, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.8411, + "step": 91 + }, + { + "epoch": 0.07620099392600774, + "grad_norm": 0.7700346112251282, + "learning_rate": 4.600000000000001e-06, + "loss": 0.815, + "step": 92 + }, + { + "epoch": 0.0770292655991165, + "grad_norm": 0.7949067950248718, + "learning_rate": 4.65e-06, + "loss": 0.8512, + "step": 93 + }, + { + "epoch": 0.07785753727222529, + "grad_norm": 0.8112176060676575, + "learning_rate": 4.7e-06, + "loss": 0.8398, + "step": 94 + }, + { + "epoch": 0.07868580894533407, + "grad_norm": 0.7990434765815735, + "learning_rate": 4.75e-06, + "loss": 0.8276, + "step": 95 + }, + { + "epoch": 0.07951408061844285, + "grad_norm": 0.7942920923233032, + "learning_rate": 4.800000000000001e-06, + "loss": 0.8333, + "step": 96 + }, + { + "epoch": 0.08034235229155164, + "grad_norm": 0.8005534410476685, + "learning_rate": 4.85e-06, + "loss": 0.8453, + "step": 97 + }, + { + "epoch": 0.0811706239646604, + "grad_norm": 0.8103216290473938, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8303, + "step": 98 + }, + { + "epoch": 0.08199889563776919, + "grad_norm": 0.8217238187789917, + "learning_rate": 4.95e-06, + "loss": 0.8204, + "step": 99 + }, + { + "epoch": 0.08282716731087797, + "grad_norm": 0.7958654761314392, + "learning_rate": 5e-06, + "loss": 0.8237, + "step": 100 + }, + { + "epoch": 0.08365543898398675, + "grad_norm": 0.8631463050842285, + "learning_rate": 4.999999758136652e-06, + "loss": 0.821, + "step": 101 + }, + { + "epoch": 0.08448371065709552, + "grad_norm": 0.7666735053062439, + "learning_rate": 4.999999032546657e-06, + "loss": 0.8307, + "step": 102 + }, + { + "epoch": 0.0853119823302043, + "grad_norm": 0.7709226012229919, + "learning_rate": 4.999997823230153e-06, + "loss": 0.8299, + "step": 103 + }, + { + "epoch": 0.08614025400331309, + "grad_norm": 0.8020651340484619, + "learning_rate": 4.9999961301873736e-06, + "loss": 0.8075, + "step": 104 + }, + { + "epoch": 0.08696852567642187, + "grad_norm": 0.8019537329673767, + "learning_rate": 4.999993953418649e-06, + "loss": 0.8615, + "step": 105 + }, + { + "epoch": 0.08779679734953065, + "grad_norm": 0.7904282808303833, + "learning_rate": 4.999991292924399e-06, + "loss": 0.8353, + "step": 106 + }, + { + "epoch": 0.08862506902263942, + "grad_norm": 0.7854570150375366, + "learning_rate": 4.999988148705138e-06, + "loss": 0.8154, + "step": 107 + }, + { + "epoch": 0.0894533406957482, + "grad_norm": 0.8044222593307495, + "learning_rate": 4.999984520761475e-06, + "loss": 0.8117, + "step": 108 + }, + { + "epoch": 0.09028161236885698, + "grad_norm": 0.7986987829208374, + "learning_rate": 4.999980409094112e-06, + "loss": 0.8407, + "step": 109 + }, + { + "epoch": 0.09110988404196577, + "grad_norm": 0.7976786494255066, + "learning_rate": 4.999975813703844e-06, + "loss": 0.7871, + "step": 110 + }, + { + "epoch": 0.09193815571507455, + "grad_norm": 0.8330315351486206, + "learning_rate": 4.9999707345915605e-06, + "loss": 0.8129, + "step": 111 + }, + { + "epoch": 0.09276642738818332, + "grad_norm": 0.8350826501846313, + "learning_rate": 4.999965171758245e-06, + "loss": 0.8144, + "step": 112 + }, + { + "epoch": 0.0935946990612921, + "grad_norm": 0.8054768443107605, + "learning_rate": 4.999959125204973e-06, + "loss": 0.809, + "step": 113 + }, + { + "epoch": 0.09442297073440088, + "grad_norm": 0.8101645708084106, + "learning_rate": 4.999952594932914e-06, + "loss": 0.8081, + "step": 114 + }, + { + "epoch": 0.09525124240750966, + "grad_norm": 0.905028760433197, + "learning_rate": 4.999945580943332e-06, + "loss": 0.8251, + "step": 115 + }, + { + "epoch": 0.09607951408061845, + "grad_norm": 0.8529418706893921, + "learning_rate": 4.999938083237585e-06, + "loss": 0.7984, + "step": 116 + }, + { + "epoch": 0.09690778575372722, + "grad_norm": 0.8047451376914978, + "learning_rate": 4.9999301018171225e-06, + "loss": 0.8222, + "step": 117 + }, + { + "epoch": 0.097736057426836, + "grad_norm": 0.8073400259017944, + "learning_rate": 4.99992163668349e-06, + "loss": 0.8251, + "step": 118 + }, + { + "epoch": 0.09856432909994478, + "grad_norm": 0.8470693826675415, + "learning_rate": 4.999912687838324e-06, + "loss": 0.8291, + "step": 119 + }, + { + "epoch": 0.09939260077305356, + "grad_norm": 0.8261650800704956, + "learning_rate": 4.999903255283357e-06, + "loss": 0.808, + "step": 120 + }, + { + "epoch": 0.10022087244616235, + "grad_norm": 0.8421704769134521, + "learning_rate": 4.999893339020414e-06, + "loss": 0.8295, + "step": 121 + }, + { + "epoch": 0.10104914411927111, + "grad_norm": 0.8244299292564392, + "learning_rate": 4.9998829390514134e-06, + "loss": 0.8215, + "step": 122 + }, + { + "epoch": 0.1018774157923799, + "grad_norm": 0.8365658521652222, + "learning_rate": 4.999872055378368e-06, + "loss": 0.8245, + "step": 123 + }, + { + "epoch": 0.10270568746548868, + "grad_norm": 0.9089561104774475, + "learning_rate": 4.999860688003384e-06, + "loss": 0.8125, + "step": 124 + }, + { + "epoch": 0.10353395913859746, + "grad_norm": 0.8784728646278381, + "learning_rate": 4.9998488369286595e-06, + "loss": 0.8321, + "step": 125 + }, + { + "epoch": 0.10436223081170624, + "grad_norm": 0.8596073389053345, + "learning_rate": 4.999836502156487e-06, + "loss": 0.8333, + "step": 126 + }, + { + "epoch": 0.10519050248481501, + "grad_norm": 0.8472688794136047, + "learning_rate": 4.9998236836892566e-06, + "loss": 0.8068, + "step": 127 + }, + { + "epoch": 0.1060187741579238, + "grad_norm": 0.8716503381729126, + "learning_rate": 4.999810381529447e-06, + "loss": 0.8005, + "step": 128 + }, + { + "epoch": 0.10684704583103258, + "grad_norm": 0.8416318893432617, + "learning_rate": 4.999796595679629e-06, + "loss": 0.7914, + "step": 129 + }, + { + "epoch": 0.10767531750414136, + "grad_norm": 0.838101863861084, + "learning_rate": 4.999782326142474e-06, + "loss": 0.7957, + "step": 130 + }, + { + "epoch": 0.10850358917725014, + "grad_norm": 0.8640785217285156, + "learning_rate": 4.999767572920741e-06, + "loss": 0.8077, + "step": 131 + }, + { + "epoch": 0.10933186085035891, + "grad_norm": 0.8207422494888306, + "learning_rate": 4.999752336017286e-06, + "loss": 0.7973, + "step": 132 + }, + { + "epoch": 0.1101601325234677, + "grad_norm": 0.8640758395195007, + "learning_rate": 4.9997366154350566e-06, + "loss": 0.8165, + "step": 133 + }, + { + "epoch": 0.11098840419657648, + "grad_norm": 0.7965807914733887, + "learning_rate": 4.999720411177094e-06, + "loss": 0.8059, + "step": 134 + }, + { + "epoch": 0.11181667586968526, + "grad_norm": 0.8463795185089111, + "learning_rate": 4.999703723246533e-06, + "loss": 0.8106, + "step": 135 + }, + { + "epoch": 0.11264494754279404, + "grad_norm": 0.8118059635162354, + "learning_rate": 4.999686551646604e-06, + "loss": 0.8099, + "step": 136 + }, + { + "epoch": 0.11347321921590281, + "grad_norm": 0.8256771564483643, + "learning_rate": 4.999668896380629e-06, + "loss": 0.8043, + "step": 137 + }, + { + "epoch": 0.11430149088901159, + "grad_norm": 0.8203126192092896, + "learning_rate": 4.999650757452025e-06, + "loss": 0.8029, + "step": 138 + }, + { + "epoch": 0.11512976256212037, + "grad_norm": 0.8442854881286621, + "learning_rate": 4.9996321348643e-06, + "loss": 0.8004, + "step": 139 + }, + { + "epoch": 0.11595803423522916, + "grad_norm": 0.8523190021514893, + "learning_rate": 4.999613028621059e-06, + "loss": 0.8092, + "step": 140 + }, + { + "epoch": 0.11678630590833794, + "grad_norm": 0.821428656578064, + "learning_rate": 4.999593438725997e-06, + "loss": 0.8105, + "step": 141 + }, + { + "epoch": 0.11761457758144671, + "grad_norm": 0.8406105041503906, + "learning_rate": 4.999573365182906e-06, + "loss": 0.7757, + "step": 142 + }, + { + "epoch": 0.11844284925455549, + "grad_norm": 0.8620600700378418, + "learning_rate": 4.999552807995669e-06, + "loss": 0.781, + "step": 143 + }, + { + "epoch": 0.11927112092766427, + "grad_norm": 0.920437216758728, + "learning_rate": 4.9995317671682655e-06, + "loss": 0.7579, + "step": 144 + }, + { + "epoch": 0.12009939260077306, + "grad_norm": 0.8414936661720276, + "learning_rate": 4.999510242704765e-06, + "loss": 0.8031, + "step": 145 + }, + { + "epoch": 0.12092766427388184, + "grad_norm": 0.8372176289558411, + "learning_rate": 4.999488234609332e-06, + "loss": 0.7931, + "step": 146 + }, + { + "epoch": 0.1217559359469906, + "grad_norm": 0.8785244226455688, + "learning_rate": 4.999465742886226e-06, + "loss": 0.8189, + "step": 147 + }, + { + "epoch": 0.12258420762009939, + "grad_norm": 0.8404353260993958, + "learning_rate": 4.999442767539799e-06, + "loss": 0.7899, + "step": 148 + }, + { + "epoch": 0.12341247929320817, + "grad_norm": 0.8541820049285889, + "learning_rate": 4.999419308574494e-06, + "loss": 0.7764, + "step": 149 + }, + { + "epoch": 0.12424075096631695, + "grad_norm": 0.8220177292823792, + "learning_rate": 4.999395365994854e-06, + "loss": 0.7942, + "step": 150 + }, + { + "epoch": 0.12506902263942574, + "grad_norm": 0.8823731541633606, + "learning_rate": 4.999370939805509e-06, + "loss": 0.793, + "step": 151 + }, + { + "epoch": 0.12589729431253452, + "grad_norm": 0.838860034942627, + "learning_rate": 4.999346030011186e-06, + "loss": 0.7741, + "step": 152 + }, + { + "epoch": 0.1267255659856433, + "grad_norm": 0.8162702322006226, + "learning_rate": 4.999320636616705e-06, + "loss": 0.7989, + "step": 153 + }, + { + "epoch": 0.12755383765875208, + "grad_norm": 0.8328465819358826, + "learning_rate": 4.99929475962698e-06, + "loss": 0.8055, + "step": 154 + }, + { + "epoch": 0.12838210933186084, + "grad_norm": 0.8373605608940125, + "learning_rate": 4.999268399047016e-06, + "loss": 0.801, + "step": 155 + }, + { + "epoch": 0.12921038100496962, + "grad_norm": 0.8408295512199402, + "learning_rate": 4.999241554881915e-06, + "loss": 0.7749, + "step": 156 + }, + { + "epoch": 0.1300386526780784, + "grad_norm": 0.8326342701911926, + "learning_rate": 4.999214227136871e-06, + "loss": 0.7818, + "step": 157 + }, + { + "epoch": 0.13086692435118719, + "grad_norm": 0.8064060211181641, + "learning_rate": 4.99918641581717e-06, + "loss": 0.7755, + "step": 158 + }, + { + "epoch": 0.13169519602429597, + "grad_norm": 0.8274610042572021, + "learning_rate": 4.999158120928196e-06, + "loss": 0.8008, + "step": 159 + }, + { + "epoch": 0.13252346769740475, + "grad_norm": 0.8121576309204102, + "learning_rate": 4.999129342475422e-06, + "loss": 0.7713, + "step": 160 + }, + { + "epoch": 0.13335173937051353, + "grad_norm": 0.8227710723876953, + "learning_rate": 4.9991000804644176e-06, + "loss": 0.7902, + "step": 161 + }, + { + "epoch": 0.13418001104362232, + "grad_norm": 0.8437026739120483, + "learning_rate": 4.999070334900843e-06, + "loss": 0.8176, + "step": 162 + }, + { + "epoch": 0.1350082827167311, + "grad_norm": 0.8379645943641663, + "learning_rate": 4.999040105790455e-06, + "loss": 0.7973, + "step": 163 + }, + { + "epoch": 0.13583655438983988, + "grad_norm": 0.822542130947113, + "learning_rate": 4.999009393139103e-06, + "loss": 0.7955, + "step": 164 + }, + { + "epoch": 0.13666482606294864, + "grad_norm": 0.8141928315162659, + "learning_rate": 4.9989781969527274e-06, + "loss": 0.7766, + "step": 165 + }, + { + "epoch": 0.13749309773605742, + "grad_norm": 0.8325410485267639, + "learning_rate": 4.998946517237367e-06, + "loss": 0.7737, + "step": 166 + }, + { + "epoch": 0.1383213694091662, + "grad_norm": 0.8399263024330139, + "learning_rate": 4.99891435399915e-06, + "loss": 0.7746, + "step": 167 + }, + { + "epoch": 0.13914964108227498, + "grad_norm": 0.8694303035736084, + "learning_rate": 4.9988817072443e-06, + "loss": 0.776, + "step": 168 + }, + { + "epoch": 0.13997791275538377, + "grad_norm": 0.8650498390197754, + "learning_rate": 4.998848576979135e-06, + "loss": 0.7941, + "step": 169 + }, + { + "epoch": 0.14080618442849255, + "grad_norm": 0.8246620893478394, + "learning_rate": 4.998814963210062e-06, + "loss": 0.7604, + "step": 170 + }, + { + "epoch": 0.14163445610160133, + "grad_norm": 0.8137409090995789, + "learning_rate": 4.99878086594359e-06, + "loss": 0.7681, + "step": 171 + }, + { + "epoch": 0.1424627277747101, + "grad_norm": 0.9755353927612305, + "learning_rate": 4.998746285186312e-06, + "loss": 0.7887, + "step": 172 + }, + { + "epoch": 0.1432909994478189, + "grad_norm": 0.8673408031463623, + "learning_rate": 4.998711220944921e-06, + "loss": 0.7803, + "step": 173 + }, + { + "epoch": 0.14411927112092768, + "grad_norm": 0.8312270045280457, + "learning_rate": 4.998675673226202e-06, + "loss": 0.7709, + "step": 174 + }, + { + "epoch": 0.14494754279403643, + "grad_norm": 0.8126094341278076, + "learning_rate": 4.998639642037032e-06, + "loss": 0.7822, + "step": 175 + }, + { + "epoch": 0.14577581446714521, + "grad_norm": 0.8561585545539856, + "learning_rate": 4.9986031273843834e-06, + "loss": 0.7829, + "step": 176 + }, + { + "epoch": 0.146604086140254, + "grad_norm": 0.8787612915039062, + "learning_rate": 4.9985661292753214e-06, + "loss": 0.7913, + "step": 177 + }, + { + "epoch": 0.14743235781336278, + "grad_norm": 0.8798345327377319, + "learning_rate": 4.998528647717006e-06, + "loss": 0.797, + "step": 178 + }, + { + "epoch": 0.14826062948647156, + "grad_norm": 0.7956589460372925, + "learning_rate": 4.998490682716687e-06, + "loss": 0.7571, + "step": 179 + }, + { + "epoch": 0.14908890115958034, + "grad_norm": 0.8505003452301025, + "learning_rate": 4.998452234281712e-06, + "loss": 0.7736, + "step": 180 + }, + { + "epoch": 0.14991717283268913, + "grad_norm": 0.8857074975967407, + "learning_rate": 4.998413302419519e-06, + "loss": 0.7863, + "step": 181 + }, + { + "epoch": 0.1507454445057979, + "grad_norm": 0.8762485980987549, + "learning_rate": 4.998373887137642e-06, + "loss": 0.7588, + "step": 182 + }, + { + "epoch": 0.1515737161789067, + "grad_norm": 0.8497205972671509, + "learning_rate": 4.998333988443709e-06, + "loss": 0.7773, + "step": 183 + }, + { + "epoch": 0.15240198785201547, + "grad_norm": 0.8720596432685852, + "learning_rate": 4.998293606345437e-06, + "loss": 0.7834, + "step": 184 + }, + { + "epoch": 0.15323025952512423, + "grad_norm": 0.8567108511924744, + "learning_rate": 4.998252740850641e-06, + "loss": 0.7794, + "step": 185 + }, + { + "epoch": 0.154058531198233, + "grad_norm": 0.8227219581604004, + "learning_rate": 4.998211391967228e-06, + "loss": 0.7688, + "step": 186 + }, + { + "epoch": 0.1548868028713418, + "grad_norm": 0.8321112394332886, + "learning_rate": 4.998169559703199e-06, + "loss": 0.7747, + "step": 187 + }, + { + "epoch": 0.15571507454445058, + "grad_norm": 0.8556115627288818, + "learning_rate": 4.9981272440666486e-06, + "loss": 0.7916, + "step": 188 + }, + { + "epoch": 0.15654334621755936, + "grad_norm": 0.8473951816558838, + "learning_rate": 4.998084445065763e-06, + "loss": 0.7823, + "step": 189 + }, + { + "epoch": 0.15737161789066814, + "grad_norm": 0.8551077246665955, + "learning_rate": 4.998041162708823e-06, + "loss": 0.7816, + "step": 190 + }, + { + "epoch": 0.15819988956377692, + "grad_norm": 0.945099413394928, + "learning_rate": 4.997997397004206e-06, + "loss": 0.7828, + "step": 191 + }, + { + "epoch": 0.1590281612368857, + "grad_norm": 0.8747237920761108, + "learning_rate": 4.9979531479603785e-06, + "loss": 0.7559, + "step": 192 + }, + { + "epoch": 0.1598564329099945, + "grad_norm": 0.8479958176612854, + "learning_rate": 4.997908415585902e-06, + "loss": 0.7694, + "step": 193 + }, + { + "epoch": 0.16068470458310327, + "grad_norm": 0.8659047484397888, + "learning_rate": 4.997863199889433e-06, + "loss": 0.7534, + "step": 194 + }, + { + "epoch": 0.16151297625621203, + "grad_norm": 0.8744727969169617, + "learning_rate": 4.997817500879719e-06, + "loss": 0.7499, + "step": 195 + }, + { + "epoch": 0.1623412479293208, + "grad_norm": 0.896468997001648, + "learning_rate": 4.997771318565603e-06, + "loss": 0.7974, + "step": 196 + }, + { + "epoch": 0.1631695196024296, + "grad_norm": 0.9058924913406372, + "learning_rate": 4.997724652956021e-06, + "loss": 0.7753, + "step": 197 + }, + { + "epoch": 0.16399779127553837, + "grad_norm": 0.8904663920402527, + "learning_rate": 4.9976775040600014e-06, + "loss": 0.7569, + "step": 198 + }, + { + "epoch": 0.16482606294864716, + "grad_norm": 0.8709238767623901, + "learning_rate": 4.997629871886669e-06, + "loss": 0.763, + "step": 199 + }, + { + "epoch": 0.16565433462175594, + "grad_norm": 0.8400478959083557, + "learning_rate": 4.997581756445238e-06, + "loss": 0.7795, + "step": 200 + }, + { + "epoch": 0.16648260629486472, + "grad_norm": 0.8458966612815857, + "learning_rate": 4.9975331577450204e-06, + "loss": 0.7597, + "step": 201 + }, + { + "epoch": 0.1673108779679735, + "grad_norm": 0.8707828521728516, + "learning_rate": 4.997484075795417e-06, + "loss": 0.759, + "step": 202 + }, + { + "epoch": 0.16813914964108229, + "grad_norm": 0.8774807453155518, + "learning_rate": 4.997434510605927e-06, + "loss": 0.7706, + "step": 203 + }, + { + "epoch": 0.16896742131419104, + "grad_norm": 0.8987929821014404, + "learning_rate": 4.99738446218614e-06, + "loss": 0.7319, + "step": 204 + }, + { + "epoch": 0.16979569298729982, + "grad_norm": 0.8869186639785767, + "learning_rate": 4.99733393054574e-06, + "loss": 0.7812, + "step": 205 + }, + { + "epoch": 0.1706239646604086, + "grad_norm": 0.860407829284668, + "learning_rate": 4.997282915694505e-06, + "loss": 0.7524, + "step": 206 + }, + { + "epoch": 0.1714522363335174, + "grad_norm": 0.8719788789749146, + "learning_rate": 4.9972314176423035e-06, + "loss": 0.7796, + "step": 207 + }, + { + "epoch": 0.17228050800662617, + "grad_norm": 0.8596137762069702, + "learning_rate": 4.997179436399103e-06, + "loss": 0.7554, + "step": 208 + }, + { + "epoch": 0.17310877967973495, + "grad_norm": 0.8505175709724426, + "learning_rate": 4.997126971974959e-06, + "loss": 0.7607, + "step": 209 + }, + { + "epoch": 0.17393705135284374, + "grad_norm": 0.7858585119247437, + "learning_rate": 4.9970740243800245e-06, + "loss": 0.7622, + "step": 210 + }, + { + "epoch": 0.17476532302595252, + "grad_norm": 0.8634621500968933, + "learning_rate": 4.997020593624543e-06, + "loss": 0.7659, + "step": 211 + }, + { + "epoch": 0.1755935946990613, + "grad_norm": 0.8650951385498047, + "learning_rate": 4.9969666797188545e-06, + "loss": 0.7689, + "step": 212 + }, + { + "epoch": 0.17642186637217008, + "grad_norm": 0.8602325916290283, + "learning_rate": 4.996912282673389e-06, + "loss": 0.7695, + "step": 213 + }, + { + "epoch": 0.17725013804527884, + "grad_norm": 0.8453742265701294, + "learning_rate": 4.9968574024986735e-06, + "loss": 0.7434, + "step": 214 + }, + { + "epoch": 0.17807840971838762, + "grad_norm": 0.8542691469192505, + "learning_rate": 4.9968020392053255e-06, + "loss": 0.7298, + "step": 215 + }, + { + "epoch": 0.1789066813914964, + "grad_norm": 0.8806673288345337, + "learning_rate": 4.996746192804058e-06, + "loss": 0.7649, + "step": 216 + }, + { + "epoch": 0.17973495306460519, + "grad_norm": 0.8451183438301086, + "learning_rate": 4.9966898633056765e-06, + "loss": 0.7681, + "step": 217 + }, + { + "epoch": 0.18056322473771397, + "grad_norm": 0.8664626479148865, + "learning_rate": 4.99663305072108e-06, + "loss": 0.7753, + "step": 218 + }, + { + "epoch": 0.18139149641082275, + "grad_norm": 0.8828144669532776, + "learning_rate": 4.996575755061262e-06, + "loss": 0.7514, + "step": 219 + }, + { + "epoch": 0.18221976808393153, + "grad_norm": 0.8493948578834534, + "learning_rate": 4.996517976337308e-06, + "loss": 0.7684, + "step": 220 + }, + { + "epoch": 0.18304803975704032, + "grad_norm": 0.8561757802963257, + "learning_rate": 4.9964597145603975e-06, + "loss": 0.7723, + "step": 221 + }, + { + "epoch": 0.1838763114301491, + "grad_norm": 0.8863773941993713, + "learning_rate": 4.996400969741805e-06, + "loss": 0.7732, + "step": 222 + }, + { + "epoch": 0.18470458310325788, + "grad_norm": 0.8571644425392151, + "learning_rate": 4.996341741892895e-06, + "loss": 0.744, + "step": 223 + }, + { + "epoch": 0.18553285477636663, + "grad_norm": 0.8622035980224609, + "learning_rate": 4.9962820310251286e-06, + "loss": 0.7607, + "step": 224 + }, + { + "epoch": 0.18636112644947542, + "grad_norm": 0.8904410600662231, + "learning_rate": 4.9962218371500595e-06, + "loss": 0.7648, + "step": 225 + }, + { + "epoch": 0.1871893981225842, + "grad_norm": 0.8975846171379089, + "learning_rate": 4.996161160279335e-06, + "loss": 0.7469, + "step": 226 + }, + { + "epoch": 0.18801766979569298, + "grad_norm": 0.8788724541664124, + "learning_rate": 4.996100000424694e-06, + "loss": 0.7473, + "step": 227 + }, + { + "epoch": 0.18884594146880176, + "grad_norm": 0.9446991086006165, + "learning_rate": 4.996038357597971e-06, + "loss": 0.7635, + "step": 228 + }, + { + "epoch": 0.18967421314191055, + "grad_norm": 0.86910480260849, + "learning_rate": 4.995976231811094e-06, + "loss": 0.7539, + "step": 229 + }, + { + "epoch": 0.19050248481501933, + "grad_norm": 0.862197756767273, + "learning_rate": 4.995913623076084e-06, + "loss": 0.7546, + "step": 230 + }, + { + "epoch": 0.1913307564881281, + "grad_norm": 0.906878650188446, + "learning_rate": 4.995850531405054e-06, + "loss": 0.7737, + "step": 231 + }, + { + "epoch": 0.1921590281612369, + "grad_norm": 0.8557183742523193, + "learning_rate": 4.995786956810212e-06, + "loss": 0.7663, + "step": 232 + }, + { + "epoch": 0.19298729983434568, + "grad_norm": 0.8125383853912354, + "learning_rate": 4.995722899303859e-06, + "loss": 0.7647, + "step": 233 + }, + { + "epoch": 0.19381557150745443, + "grad_norm": 0.8610332012176514, + "learning_rate": 4.995658358898391e-06, + "loss": 0.7541, + "step": 234 + }, + { + "epoch": 0.19464384318056321, + "grad_norm": 0.8620829582214355, + "learning_rate": 4.995593335606294e-06, + "loss": 0.7625, + "step": 235 + }, + { + "epoch": 0.195472114853672, + "grad_norm": 0.8296957612037659, + "learning_rate": 4.9955278294401496e-06, + "loss": 0.7536, + "step": 236 + }, + { + "epoch": 0.19630038652678078, + "grad_norm": 0.8974814414978027, + "learning_rate": 4.995461840412634e-06, + "loss": 0.7651, + "step": 237 + }, + { + "epoch": 0.19712865819988956, + "grad_norm": 0.8701095581054688, + "learning_rate": 4.995395368536515e-06, + "loss": 0.788, + "step": 238 + }, + { + "epoch": 0.19795692987299834, + "grad_norm": 0.8556636571884155, + "learning_rate": 4.995328413824653e-06, + "loss": 0.7649, + "step": 239 + }, + { + "epoch": 0.19878520154610713, + "grad_norm": 0.836686909198761, + "learning_rate": 4.995260976290005e-06, + "loss": 0.7583, + "step": 240 + }, + { + "epoch": 0.1996134732192159, + "grad_norm": 0.8807066082954407, + "learning_rate": 4.995193055945618e-06, + "loss": 0.7766, + "step": 241 + }, + { + "epoch": 0.2004417448923247, + "grad_norm": 0.8363582491874695, + "learning_rate": 4.995124652804635e-06, + "loss": 0.7655, + "step": 242 + }, + { + "epoch": 0.20127001656543347, + "grad_norm": 0.8699461817741394, + "learning_rate": 4.9950557668802905e-06, + "loss": 0.743, + "step": 243 + }, + { + "epoch": 0.20209828823854223, + "grad_norm": 0.8571060299873352, + "learning_rate": 4.994986398185914e-06, + "loss": 0.7512, + "step": 244 + }, + { + "epoch": 0.202926559911651, + "grad_norm": 0.8429675102233887, + "learning_rate": 4.994916546734927e-06, + "loss": 0.7589, + "step": 245 + }, + { + "epoch": 0.2037548315847598, + "grad_norm": 0.874822199344635, + "learning_rate": 4.994846212540846e-06, + "loss": 0.7534, + "step": 246 + }, + { + "epoch": 0.20458310325786858, + "grad_norm": 0.878341555595398, + "learning_rate": 4.994775395617279e-06, + "loss": 0.7709, + "step": 247 + }, + { + "epoch": 0.20541137493097736, + "grad_norm": 0.897961437702179, + "learning_rate": 4.99470409597793e-06, + "loss": 0.7768, + "step": 248 + }, + { + "epoch": 0.20623964660408614, + "grad_norm": 0.8439584374427795, + "learning_rate": 4.994632313636593e-06, + "loss": 0.7558, + "step": 249 + }, + { + "epoch": 0.20706791827719492, + "grad_norm": 0.8394468426704407, + "learning_rate": 4.994560048607159e-06, + "loss": 0.7492, + "step": 250 + }, + { + "epoch": 0.2078961899503037, + "grad_norm": 0.8489627242088318, + "learning_rate": 4.9944873009036074e-06, + "loss": 0.7365, + "step": 251 + }, + { + "epoch": 0.2087244616234125, + "grad_norm": 0.8675844669342041, + "learning_rate": 4.994414070540018e-06, + "loss": 0.762, + "step": 252 + }, + { + "epoch": 0.20955273329652127, + "grad_norm": 0.8228297829627991, + "learning_rate": 4.994340357530558e-06, + "loss": 0.7475, + "step": 253 + }, + { + "epoch": 0.21038100496963003, + "grad_norm": 0.8794129490852356, + "learning_rate": 4.9942661618894915e-06, + "loss": 0.7631, + "step": 254 + }, + { + "epoch": 0.2112092766427388, + "grad_norm": 0.8616135716438293, + "learning_rate": 4.9941914836311735e-06, + "loss": 0.7749, + "step": 255 + }, + { + "epoch": 0.2120375483158476, + "grad_norm": 0.8470495939254761, + "learning_rate": 4.994116322770054e-06, + "loss": 0.7484, + "step": 256 + }, + { + "epoch": 0.21286581998895637, + "grad_norm": 0.9346716403961182, + "learning_rate": 4.9940406793206755e-06, + "loss": 0.7557, + "step": 257 + }, + { + "epoch": 0.21369409166206516, + "grad_norm": 0.8993160724639893, + "learning_rate": 4.993964553297676e-06, + "loss": 0.7422, + "step": 258 + }, + { + "epoch": 0.21452236333517394, + "grad_norm": 0.8257086277008057, + "learning_rate": 4.993887944715783e-06, + "loss": 0.7783, + "step": 259 + }, + { + "epoch": 0.21535063500828272, + "grad_norm": 0.8950703144073486, + "learning_rate": 4.993810853589819e-06, + "loss": 0.7311, + "step": 260 + }, + { + "epoch": 0.2161789066813915, + "grad_norm": 0.8733702898025513, + "learning_rate": 4.993733279934704e-06, + "loss": 0.7572, + "step": 261 + }, + { + "epoch": 0.21700717835450029, + "grad_norm": 0.8322939276695251, + "learning_rate": 4.993655223765444e-06, + "loss": 0.7632, + "step": 262 + }, + { + "epoch": 0.21783545002760907, + "grad_norm": 0.8666137456893921, + "learning_rate": 4.993576685097145e-06, + "loss": 0.7619, + "step": 263 + }, + { + "epoch": 0.21866372170071782, + "grad_norm": 0.8770918250083923, + "learning_rate": 4.993497663945002e-06, + "loss": 0.7538, + "step": 264 + }, + { + "epoch": 0.2194919933738266, + "grad_norm": 0.8940248489379883, + "learning_rate": 4.9934181603243045e-06, + "loss": 0.7512, + "step": 265 + }, + { + "epoch": 0.2203202650469354, + "grad_norm": 0.8979714512825012, + "learning_rate": 4.993338174250437e-06, + "loss": 0.7504, + "step": 266 + }, + { + "epoch": 0.22114853672004417, + "grad_norm": 0.8802001476287842, + "learning_rate": 4.993257705738876e-06, + "loss": 0.7501, + "step": 267 + }, + { + "epoch": 0.22197680839315295, + "grad_norm": 0.8734432458877563, + "learning_rate": 4.993176754805189e-06, + "loss": 0.7452, + "step": 268 + }, + { + "epoch": 0.22280508006626173, + "grad_norm": 0.9127711057662964, + "learning_rate": 4.993095321465042e-06, + "loss": 0.7404, + "step": 269 + }, + { + "epoch": 0.22363335173937052, + "grad_norm": 0.8533952832221985, + "learning_rate": 4.993013405734191e-06, + "loss": 0.7382, + "step": 270 + }, + { + "epoch": 0.2244616234124793, + "grad_norm": 0.8463877439498901, + "learning_rate": 4.9929310076284845e-06, + "loss": 0.7574, + "step": 271 + }, + { + "epoch": 0.22528989508558808, + "grad_norm": 0.8598008155822754, + "learning_rate": 4.992848127163868e-06, + "loss": 0.7624, + "step": 272 + }, + { + "epoch": 0.22611816675869686, + "grad_norm": 0.8759154081344604, + "learning_rate": 4.992764764356376e-06, + "loss": 0.7637, + "step": 273 + }, + { + "epoch": 0.22694643843180562, + "grad_norm": 0.8880444765090942, + "learning_rate": 4.992680919222141e-06, + "loss": 0.7514, + "step": 274 + }, + { + "epoch": 0.2277747101049144, + "grad_norm": 0.871366024017334, + "learning_rate": 4.9925965917773826e-06, + "loss": 0.7569, + "step": 275 + }, + { + "epoch": 0.22860298177802318, + "grad_norm": 0.862654983997345, + "learning_rate": 4.99251178203842e-06, + "loss": 0.7349, + "step": 276 + }, + { + "epoch": 0.22943125345113197, + "grad_norm": 0.8615298867225647, + "learning_rate": 4.992426490021662e-06, + "loss": 0.7438, + "step": 277 + }, + { + "epoch": 0.23025952512424075, + "grad_norm": 0.8735869526863098, + "learning_rate": 4.992340715743614e-06, + "loss": 0.7755, + "step": 278 + }, + { + "epoch": 0.23108779679734953, + "grad_norm": 0.9082722067832947, + "learning_rate": 4.9922544592208695e-06, + "loss": 0.7561, + "step": 279 + }, + { + "epoch": 0.23191606847045831, + "grad_norm": 0.8639047145843506, + "learning_rate": 4.992167720470119e-06, + "loss": 0.7493, + "step": 280 + }, + { + "epoch": 0.2327443401435671, + "grad_norm": 0.891640305519104, + "learning_rate": 4.9920804995081475e-06, + "loss": 0.75, + "step": 281 + }, + { + "epoch": 0.23357261181667588, + "grad_norm": 0.8880137205123901, + "learning_rate": 4.99199279635183e-06, + "loss": 0.7801, + "step": 282 + }, + { + "epoch": 0.23440088348978466, + "grad_norm": 0.8600473403930664, + "learning_rate": 4.991904611018137e-06, + "loss": 0.7477, + "step": 283 + }, + { + "epoch": 0.23522915516289342, + "grad_norm": 0.8685399889945984, + "learning_rate": 4.99181594352413e-06, + "loss": 0.741, + "step": 284 + }, + { + "epoch": 0.2360574268360022, + "grad_norm": 0.9006181955337524, + "learning_rate": 4.991726793886966e-06, + "loss": 0.7288, + "step": 285 + }, + { + "epoch": 0.23688569850911098, + "grad_norm": 0.863895833492279, + "learning_rate": 4.991637162123896e-06, + "loss": 0.7375, + "step": 286 + }, + { + "epoch": 0.23771397018221976, + "grad_norm": 0.8620381951332092, + "learning_rate": 4.9915470482522625e-06, + "loss": 0.7249, + "step": 287 + }, + { + "epoch": 0.23854224185532855, + "grad_norm": 0.9634964466094971, + "learning_rate": 4.991456452289499e-06, + "loss": 0.7313, + "step": 288 + }, + { + "epoch": 0.23937051352843733, + "grad_norm": 0.9154514670372009, + "learning_rate": 4.9913653742531385e-06, + "loss": 0.782, + "step": 289 + }, + { + "epoch": 0.2401987852015461, + "grad_norm": 0.9179158806800842, + "learning_rate": 4.991273814160802e-06, + "loss": 0.7388, + "step": 290 + }, + { + "epoch": 0.2410270568746549, + "grad_norm": 0.8609046936035156, + "learning_rate": 4.991181772030206e-06, + "loss": 0.7393, + "step": 291 + }, + { + "epoch": 0.24185532854776368, + "grad_norm": 0.8689826130867004, + "learning_rate": 4.991089247879159e-06, + "loss": 0.7302, + "step": 292 + }, + { + "epoch": 0.24268360022087246, + "grad_norm": 0.8414047956466675, + "learning_rate": 4.990996241725565e-06, + "loss": 0.7496, + "step": 293 + }, + { + "epoch": 0.2435118718939812, + "grad_norm": 0.8621289134025574, + "learning_rate": 4.990902753587418e-06, + "loss": 0.7317, + "step": 294 + }, + { + "epoch": 0.24434014356709, + "grad_norm": 0.856566309928894, + "learning_rate": 4.990808783482809e-06, + "loss": 0.7169, + "step": 295 + }, + { + "epoch": 0.24516841524019878, + "grad_norm": 0.8549672961235046, + "learning_rate": 4.990714331429919e-06, + "loss": 0.7357, + "step": 296 + }, + { + "epoch": 0.24599668691330756, + "grad_norm": 0.8813024759292603, + "learning_rate": 4.990619397447024e-06, + "loss": 0.7437, + "step": 297 + }, + { + "epoch": 0.24682495858641634, + "grad_norm": 0.9008268713951111, + "learning_rate": 4.990523981552493e-06, + "loss": 0.7652, + "step": 298 + }, + { + "epoch": 0.24765323025952513, + "grad_norm": 0.910277783870697, + "learning_rate": 4.990428083764788e-06, + "loss": 0.7598, + "step": 299 + }, + { + "epoch": 0.2484815019326339, + "grad_norm": 0.9261599183082581, + "learning_rate": 4.990331704102464e-06, + "loss": 0.7199, + "step": 300 + }, + { + "epoch": 0.2493097736057427, + "grad_norm": 0.9166674613952637, + "learning_rate": 4.99023484258417e-06, + "loss": 0.7414, + "step": 301 + }, + { + "epoch": 0.2501380452788515, + "grad_norm": 0.8701214790344238, + "learning_rate": 4.990137499228648e-06, + "loss": 0.7326, + "step": 302 + }, + { + "epoch": 0.25096631695196026, + "grad_norm": 0.9120466113090515, + "learning_rate": 4.990039674054732e-06, + "loss": 0.7574, + "step": 303 + }, + { + "epoch": 0.25179458862506904, + "grad_norm": 0.9449286460876465, + "learning_rate": 4.989941367081351e-06, + "loss": 0.725, + "step": 304 + }, + { + "epoch": 0.2526228602981778, + "grad_norm": 0.8989110589027405, + "learning_rate": 4.989842578327526e-06, + "loss": 0.7569, + "step": 305 + }, + { + "epoch": 0.2534511319712866, + "grad_norm": 0.891677975654602, + "learning_rate": 4.989743307812373e-06, + "loss": 0.7135, + "step": 306 + }, + { + "epoch": 0.2542794036443954, + "grad_norm": 0.8689535856246948, + "learning_rate": 4.989643555555099e-06, + "loss": 0.7439, + "step": 307 + }, + { + "epoch": 0.25510767531750417, + "grad_norm": 0.8779630064964294, + "learning_rate": 4.989543321575004e-06, + "loss": 0.7322, + "step": 308 + }, + { + "epoch": 0.2559359469906129, + "grad_norm": 0.8977622985839844, + "learning_rate": 4.9894426058914856e-06, + "loss": 0.7544, + "step": 309 + }, + { + "epoch": 0.2567642186637217, + "grad_norm": 0.8216757774353027, + "learning_rate": 4.989341408524028e-06, + "loss": 0.7344, + "step": 310 + }, + { + "epoch": 0.25759249033683046, + "grad_norm": 0.8494759798049927, + "learning_rate": 4.9892397294922126e-06, + "loss": 0.7389, + "step": 311 + }, + { + "epoch": 0.25842076200993924, + "grad_norm": 0.881244957447052, + "learning_rate": 4.989137568815715e-06, + "loss": 0.7121, + "step": 312 + }, + { + "epoch": 0.259249033683048, + "grad_norm": 0.8679410219192505, + "learning_rate": 4.9890349265143005e-06, + "loss": 0.7469, + "step": 313 + }, + { + "epoch": 0.2600773053561568, + "grad_norm": 0.9746971130371094, + "learning_rate": 4.988931802607831e-06, + "loss": 0.7585, + "step": 314 + }, + { + "epoch": 0.2609055770292656, + "grad_norm": 0.9222471117973328, + "learning_rate": 4.988828197116259e-06, + "loss": 0.761, + "step": 315 + }, + { + "epoch": 0.26173384870237437, + "grad_norm": 0.9104124903678894, + "learning_rate": 4.988724110059631e-06, + "loss": 0.7652, + "step": 316 + }, + { + "epoch": 0.26256212037548315, + "grad_norm": 0.8876475095748901, + "learning_rate": 4.988619541458088e-06, + "loss": 0.7529, + "step": 317 + }, + { + "epoch": 0.26339039204859194, + "grad_norm": 0.8705421686172485, + "learning_rate": 4.988514491331861e-06, + "loss": 0.7437, + "step": 318 + }, + { + "epoch": 0.2642186637217007, + "grad_norm": 0.8594399690628052, + "learning_rate": 4.9884089597012784e-06, + "loss": 0.7406, + "step": 319 + }, + { + "epoch": 0.2650469353948095, + "grad_norm": 0.8807039856910706, + "learning_rate": 4.988302946586759e-06, + "loss": 0.7429, + "step": 320 + }, + { + "epoch": 0.2658752070679183, + "grad_norm": 0.8881412148475647, + "learning_rate": 4.9881964520088144e-06, + "loss": 0.7234, + "step": 321 + }, + { + "epoch": 0.26670347874102707, + "grad_norm": 0.8764941096305847, + "learning_rate": 4.988089475988052e-06, + "loss": 0.7514, + "step": 322 + }, + { + "epoch": 0.26753175041413585, + "grad_norm": 0.8992927670478821, + "learning_rate": 4.987982018545169e-06, + "loss": 0.7143, + "step": 323 + }, + { + "epoch": 0.26836002208724463, + "grad_norm": 0.8961546421051025, + "learning_rate": 4.9878740797009585e-06, + "loss": 0.7396, + "step": 324 + }, + { + "epoch": 0.2691882937603534, + "grad_norm": 0.9031578302383423, + "learning_rate": 4.987765659476304e-06, + "loss": 0.7407, + "step": 325 + }, + { + "epoch": 0.2700165654334622, + "grad_norm": 0.8693655729293823, + "learning_rate": 4.987656757892186e-06, + "loss": 0.7215, + "step": 326 + }, + { + "epoch": 0.270844837106571, + "grad_norm": 0.8739856481552124, + "learning_rate": 4.987547374969676e-06, + "loss": 0.7471, + "step": 327 + }, + { + "epoch": 0.27167310877967976, + "grad_norm": 0.8916322588920593, + "learning_rate": 4.987437510729936e-06, + "loss": 0.7459, + "step": 328 + }, + { + "epoch": 0.2725013804527885, + "grad_norm": 0.890178382396698, + "learning_rate": 4.987327165194225e-06, + "loss": 0.7322, + "step": 329 + }, + { + "epoch": 0.27332965212589727, + "grad_norm": 0.8611598014831543, + "learning_rate": 4.987216338383896e-06, + "loss": 0.733, + "step": 330 + }, + { + "epoch": 0.27415792379900605, + "grad_norm": 0.87791907787323, + "learning_rate": 4.98710503032039e-06, + "loss": 0.7445, + "step": 331 + }, + { + "epoch": 0.27498619547211484, + "grad_norm": 0.963479220867157, + "learning_rate": 4.986993241025245e-06, + "loss": 0.7376, + "step": 332 + }, + { + "epoch": 0.2758144671452236, + "grad_norm": 0.9083762764930725, + "learning_rate": 4.986880970520092e-06, + "loss": 0.7258, + "step": 333 + }, + { + "epoch": 0.2766427388183324, + "grad_norm": 1.048991084098816, + "learning_rate": 4.9867682188266524e-06, + "loss": 0.7399, + "step": 334 + }, + { + "epoch": 0.2774710104914412, + "grad_norm": 0.9404491782188416, + "learning_rate": 4.986654985966745e-06, + "loss": 0.7133, + "step": 335 + }, + { + "epoch": 0.27829928216454997, + "grad_norm": 0.9168346524238586, + "learning_rate": 4.986541271962278e-06, + "loss": 0.74, + "step": 336 + }, + { + "epoch": 0.27912755383765875, + "grad_norm": 0.8716689944267273, + "learning_rate": 4.986427076835253e-06, + "loss": 0.7249, + "step": 337 + }, + { + "epoch": 0.27995582551076753, + "grad_norm": 0.9097291827201843, + "learning_rate": 4.986312400607767e-06, + "loss": 0.7151, + "step": 338 + }, + { + "epoch": 0.2807840971838763, + "grad_norm": 0.8621960282325745, + "learning_rate": 4.98619724330201e-06, + "loss": 0.7125, + "step": 339 + }, + { + "epoch": 0.2816123688569851, + "grad_norm": 0.9823791980743408, + "learning_rate": 4.986081604940261e-06, + "loss": 0.7191, + "step": 340 + }, + { + "epoch": 0.2824406405300939, + "grad_norm": 0.8999047875404358, + "learning_rate": 4.9859654855448966e-06, + "loss": 0.7175, + "step": 341 + }, + { + "epoch": 0.28326891220320266, + "grad_norm": 0.8690859079360962, + "learning_rate": 4.985848885138386e-06, + "loss": 0.7309, + "step": 342 + }, + { + "epoch": 0.28409718387631144, + "grad_norm": 0.889500617980957, + "learning_rate": 4.985731803743287e-06, + "loss": 0.7391, + "step": 343 + }, + { + "epoch": 0.2849254555494202, + "grad_norm": 0.8992016911506653, + "learning_rate": 4.985614241382257e-06, + "loss": 0.7437, + "step": 344 + }, + { + "epoch": 0.285753727222529, + "grad_norm": 0.9100068211555481, + "learning_rate": 4.985496198078041e-06, + "loss": 0.722, + "step": 345 + }, + { + "epoch": 0.2865819988956378, + "grad_norm": 0.9013107419013977, + "learning_rate": 4.985377673853481e-06, + "loss": 0.7434, + "step": 346 + }, + { + "epoch": 0.2874102705687466, + "grad_norm": 0.8794704675674438, + "learning_rate": 4.9852586687315094e-06, + "loss": 0.7387, + "step": 347 + }, + { + "epoch": 0.28823854224185536, + "grad_norm": 0.9288201928138733, + "learning_rate": 4.9851391827351526e-06, + "loss": 0.7165, + "step": 348 + }, + { + "epoch": 0.2890668139149641, + "grad_norm": 0.8735847473144531, + "learning_rate": 4.98501921588753e-06, + "loss": 0.7479, + "step": 349 + }, + { + "epoch": 0.28989508558807286, + "grad_norm": 0.890616238117218, + "learning_rate": 4.984898768211853e-06, + "loss": 0.76, + "step": 350 + }, + { + "epoch": 0.29072335726118165, + "grad_norm": 0.9181028604507446, + "learning_rate": 4.98477783973143e-06, + "loss": 0.7366, + "step": 351 + }, + { + "epoch": 0.29155162893429043, + "grad_norm": 0.9947429299354553, + "learning_rate": 4.984656430469657e-06, + "loss": 0.7199, + "step": 352 + }, + { + "epoch": 0.2923799006073992, + "grad_norm": 0.8730769157409668, + "learning_rate": 4.984534540450027e-06, + "loss": 0.7153, + "step": 353 + }, + { + "epoch": 0.293208172280508, + "grad_norm": 0.8675997853279114, + "learning_rate": 4.9844121696961235e-06, + "loss": 0.7196, + "step": 354 + }, + { + "epoch": 0.2940364439536168, + "grad_norm": 0.9558542966842651, + "learning_rate": 4.984289318231624e-06, + "loss": 0.7311, + "step": 355 + }, + { + "epoch": 0.29486471562672556, + "grad_norm": 0.8817659616470337, + "learning_rate": 4.9841659860803e-06, + "loss": 0.7367, + "step": 356 + }, + { + "epoch": 0.29569298729983434, + "grad_norm": 0.8679526448249817, + "learning_rate": 4.984042173266016e-06, + "loss": 0.719, + "step": 357 + }, + { + "epoch": 0.2965212589729431, + "grad_norm": 0.9102020859718323, + "learning_rate": 4.983917879812726e-06, + "loss": 0.7301, + "step": 358 + }, + { + "epoch": 0.2973495306460519, + "grad_norm": 0.8987215757369995, + "learning_rate": 4.983793105744482e-06, + "loss": 0.7557, + "step": 359 + }, + { + "epoch": 0.2981778023191607, + "grad_norm": 0.8779832720756531, + "learning_rate": 4.983667851085426e-06, + "loss": 0.7471, + "step": 360 + }, + { + "epoch": 0.29900607399226947, + "grad_norm": 0.8762708306312561, + "learning_rate": 4.983542115859792e-06, + "loss": 0.7676, + "step": 361 + }, + { + "epoch": 0.29983434566537825, + "grad_norm": 0.910353422164917, + "learning_rate": 4.98341590009191e-06, + "loss": 0.7285, + "step": 362 + }, + { + "epoch": 0.30066261733848704, + "grad_norm": 0.9385405778884888, + "learning_rate": 4.983289203806202e-06, + "loss": 0.7142, + "step": 363 + }, + { + "epoch": 0.3014908890115958, + "grad_norm": 1.016781210899353, + "learning_rate": 4.983162027027182e-06, + "loss": 0.7408, + "step": 364 + }, + { + "epoch": 0.3023191606847046, + "grad_norm": 0.8642508387565613, + "learning_rate": 4.983034369779457e-06, + "loss": 0.7397, + "step": 365 + }, + { + "epoch": 0.3031474323578134, + "grad_norm": 0.9469170570373535, + "learning_rate": 4.982906232087728e-06, + "loss": 0.7158, + "step": 366 + }, + { + "epoch": 0.30397570403092217, + "grad_norm": 0.8818212151527405, + "learning_rate": 4.982777613976789e-06, + "loss": 0.7408, + "step": 367 + }, + { + "epoch": 0.30480397570403095, + "grad_norm": 0.8639838099479675, + "learning_rate": 4.982648515471526e-06, + "loss": 0.7217, + "step": 368 + }, + { + "epoch": 0.3056322473771397, + "grad_norm": 0.923780083656311, + "learning_rate": 4.982518936596917e-06, + "loss": 0.7284, + "step": 369 + }, + { + "epoch": 0.30646051905024846, + "grad_norm": 0.8992834687232971, + "learning_rate": 4.982388877378037e-06, + "loss": 0.7395, + "step": 370 + }, + { + "epoch": 0.30728879072335724, + "grad_norm": 0.8671404123306274, + "learning_rate": 4.982258337840049e-06, + "loss": 0.7157, + "step": 371 + }, + { + "epoch": 0.308117062396466, + "grad_norm": 0.8689168691635132, + "learning_rate": 4.982127318008212e-06, + "loss": 0.7398, + "step": 372 + }, + { + "epoch": 0.3089453340695748, + "grad_norm": 0.8964621424674988, + "learning_rate": 4.9819958179078765e-06, + "loss": 0.7385, + "step": 373 + }, + { + "epoch": 0.3097736057426836, + "grad_norm": 0.8984093070030212, + "learning_rate": 4.981863837564488e-06, + "loss": 0.7253, + "step": 374 + }, + { + "epoch": 0.31060187741579237, + "grad_norm": 0.8776745200157166, + "learning_rate": 4.981731377003583e-06, + "loss": 0.7126, + "step": 375 + }, + { + "epoch": 0.31143014908890115, + "grad_norm": 0.8782477378845215, + "learning_rate": 4.98159843625079e-06, + "loss": 0.7164, + "step": 376 + }, + { + "epoch": 0.31225842076200994, + "grad_norm": 0.8993703722953796, + "learning_rate": 4.981465015331832e-06, + "loss": 0.7391, + "step": 377 + }, + { + "epoch": 0.3130866924351187, + "grad_norm": 0.8975916504859924, + "learning_rate": 4.981331114272527e-06, + "loss": 0.7385, + "step": 378 + }, + { + "epoch": 0.3139149641082275, + "grad_norm": 0.8726824522018433, + "learning_rate": 4.9811967330987816e-06, + "loss": 0.7426, + "step": 379 + }, + { + "epoch": 0.3147432357813363, + "grad_norm": 0.8741514682769775, + "learning_rate": 4.981061871836597e-06, + "loss": 0.7246, + "step": 380 + }, + { + "epoch": 0.31557150745444507, + "grad_norm": 0.8807122707366943, + "learning_rate": 4.98092653051207e-06, + "loss": 0.7364, + "step": 381 + }, + { + "epoch": 0.31639977912755385, + "grad_norm": 0.835006058216095, + "learning_rate": 4.980790709151385e-06, + "loss": 0.7248, + "step": 382 + }, + { + "epoch": 0.31722805080066263, + "grad_norm": 0.9473997354507446, + "learning_rate": 4.980654407780824e-06, + "loss": 0.7331, + "step": 383 + }, + { + "epoch": 0.3180563224737714, + "grad_norm": 0.8332815170288086, + "learning_rate": 4.980517626426758e-06, + "loss": 0.7299, + "step": 384 + }, + { + "epoch": 0.3188845941468802, + "grad_norm": 0.8536360859870911, + "learning_rate": 4.980380365115656e-06, + "loss": 0.7095, + "step": 385 + }, + { + "epoch": 0.319712865819989, + "grad_norm": 0.9488157629966736, + "learning_rate": 4.980242623874075e-06, + "loss": 0.7382, + "step": 386 + }, + { + "epoch": 0.32054113749309776, + "grad_norm": 0.8623067736625671, + "learning_rate": 4.980104402728666e-06, + "loss": 0.7187, + "step": 387 + }, + { + "epoch": 0.32136940916620654, + "grad_norm": 0.9110252857208252, + "learning_rate": 4.979965701706175e-06, + "loss": 0.7354, + "step": 388 + }, + { + "epoch": 0.32219768083931527, + "grad_norm": 0.8966642618179321, + "learning_rate": 4.979826520833439e-06, + "loss": 0.7251, + "step": 389 + }, + { + "epoch": 0.32302595251242405, + "grad_norm": 0.856674075126648, + "learning_rate": 4.979686860137387e-06, + "loss": 0.723, + "step": 390 + }, + { + "epoch": 0.32385422418553284, + "grad_norm": 0.8623953461647034, + "learning_rate": 4.979546719645043e-06, + "loss": 0.7224, + "step": 391 + }, + { + "epoch": 0.3246824958586416, + "grad_norm": 0.8685624599456787, + "learning_rate": 4.979406099383523e-06, + "loss": 0.7283, + "step": 392 + }, + { + "epoch": 0.3255107675317504, + "grad_norm": 0.8610525727272034, + "learning_rate": 4.979264999380035e-06, + "loss": 0.7385, + "step": 393 + }, + { + "epoch": 0.3263390392048592, + "grad_norm": 0.8860796093940735, + "learning_rate": 4.979123419661882e-06, + "loss": 0.743, + "step": 394 + }, + { + "epoch": 0.32716731087796797, + "grad_norm": 0.8748682737350464, + "learning_rate": 4.978981360256457e-06, + "loss": 0.7447, + "step": 395 + }, + { + "epoch": 0.32799558255107675, + "grad_norm": 0.8750934600830078, + "learning_rate": 4.978838821191247e-06, + "loss": 0.7259, + "step": 396 + }, + { + "epoch": 0.32882385422418553, + "grad_norm": 0.9335777163505554, + "learning_rate": 4.978695802493834e-06, + "loss": 0.7271, + "step": 397 + }, + { + "epoch": 0.3296521258972943, + "grad_norm": 0.8823036551475525, + "learning_rate": 4.978552304191888e-06, + "loss": 0.723, + "step": 398 + }, + { + "epoch": 0.3304803975704031, + "grad_norm": 0.9503542184829712, + "learning_rate": 4.9784083263131764e-06, + "loss": 0.752, + "step": 399 + }, + { + "epoch": 0.3313086692435119, + "grad_norm": 0.8779938220977783, + "learning_rate": 4.978263868885557e-06, + "loss": 0.7098, + "step": 400 + }, + { + "epoch": 0.33213694091662066, + "grad_norm": 0.873857855796814, + "learning_rate": 4.978118931936981e-06, + "loss": 0.7143, + "step": 401 + }, + { + "epoch": 0.33296521258972944, + "grad_norm": 0.8437919020652771, + "learning_rate": 4.9779735154954924e-06, + "loss": 0.7275, + "step": 402 + }, + { + "epoch": 0.3337934842628382, + "grad_norm": 0.8914555311203003, + "learning_rate": 4.977827619589229e-06, + "loss": 0.722, + "step": 403 + }, + { + "epoch": 0.334621755935947, + "grad_norm": 0.9062597751617432, + "learning_rate": 4.977681244246417e-06, + "loss": 0.7067, + "step": 404 + }, + { + "epoch": 0.3354500276090558, + "grad_norm": 0.8555092811584473, + "learning_rate": 4.977534389495382e-06, + "loss": 0.7318, + "step": 405 + }, + { + "epoch": 0.33627829928216457, + "grad_norm": 0.8717684745788574, + "learning_rate": 4.9773870553645386e-06, + "loss": 0.7398, + "step": 406 + }, + { + "epoch": 0.33710657095527335, + "grad_norm": 0.8746877908706665, + "learning_rate": 4.977239241882393e-06, + "loss": 0.7154, + "step": 407 + }, + { + "epoch": 0.3379348426283821, + "grad_norm": 0.8905707597732544, + "learning_rate": 4.977090949077547e-06, + "loss": 0.7345, + "step": 408 + }, + { + "epoch": 0.33876311430149086, + "grad_norm": 0.9037347435951233, + "learning_rate": 4.9769421769786935e-06, + "loss": 0.7271, + "step": 409 + }, + { + "epoch": 0.33959138597459965, + "grad_norm": 0.8588722944259644, + "learning_rate": 4.976792925614619e-06, + "loss": 0.7335, + "step": 410 + }, + { + "epoch": 0.34041965764770843, + "grad_norm": 0.8786173462867737, + "learning_rate": 4.9766431950142005e-06, + "loss": 0.7254, + "step": 411 + }, + { + "epoch": 0.3412479293208172, + "grad_norm": 0.9019079804420471, + "learning_rate": 4.976492985206412e-06, + "loss": 0.715, + "step": 412 + }, + { + "epoch": 0.342076200993926, + "grad_norm": 0.878132700920105, + "learning_rate": 4.976342296220316e-06, + "loss": 0.7063, + "step": 413 + }, + { + "epoch": 0.3429044726670348, + "grad_norm": 0.8722684979438782, + "learning_rate": 4.976191128085069e-06, + "loss": 0.7481, + "step": 414 + }, + { + "epoch": 0.34373274434014356, + "grad_norm": 0.9076220393180847, + "learning_rate": 4.976039480829922e-06, + "loss": 0.7197, + "step": 415 + }, + { + "epoch": 0.34456101601325234, + "grad_norm": 0.8674967288970947, + "learning_rate": 4.975887354484217e-06, + "loss": 0.7282, + "step": 416 + }, + { + "epoch": 0.3453892876863611, + "grad_norm": 0.9406816959381104, + "learning_rate": 4.9757347490773885e-06, + "loss": 0.7222, + "step": 417 + }, + { + "epoch": 0.3462175593594699, + "grad_norm": 0.8747051358222961, + "learning_rate": 4.975581664638965e-06, + "loss": 0.7283, + "step": 418 + }, + { + "epoch": 0.3470458310325787, + "grad_norm": 1.0587687492370605, + "learning_rate": 4.9754281011985666e-06, + "loss": 0.7272, + "step": 419 + }, + { + "epoch": 0.34787410270568747, + "grad_norm": 0.8596946597099304, + "learning_rate": 4.975274058785906e-06, + "loss": 0.7407, + "step": 420 + }, + { + "epoch": 0.34870237437879625, + "grad_norm": 0.9230688214302063, + "learning_rate": 4.975119537430789e-06, + "loss": 0.7303, + "step": 421 + }, + { + "epoch": 0.34953064605190504, + "grad_norm": 0.9276477694511414, + "learning_rate": 4.974964537163115e-06, + "loss": 0.7296, + "step": 422 + }, + { + "epoch": 0.3503589177250138, + "grad_norm": 0.8969418406486511, + "learning_rate": 4.974809058012874e-06, + "loss": 0.7275, + "step": 423 + }, + { + "epoch": 0.3511871893981226, + "grad_norm": 0.8454467058181763, + "learning_rate": 4.97465310001015e-06, + "loss": 0.7144, + "step": 424 + }, + { + "epoch": 0.3520154610712314, + "grad_norm": 0.8833656907081604, + "learning_rate": 4.97449666318512e-06, + "loss": 0.7073, + "step": 425 + }, + { + "epoch": 0.35284373274434017, + "grad_norm": 0.9251653552055359, + "learning_rate": 4.974339747568052e-06, + "loss": 0.7351, + "step": 426 + }, + { + "epoch": 0.35367200441744895, + "grad_norm": 0.8938575983047485, + "learning_rate": 4.974182353189309e-06, + "loss": 0.7031, + "step": 427 + }, + { + "epoch": 0.3545002760905577, + "grad_norm": 0.8487107753753662, + "learning_rate": 4.974024480079343e-06, + "loss": 0.704, + "step": 428 + }, + { + "epoch": 0.35532854776366646, + "grad_norm": 0.8515727519989014, + "learning_rate": 4.973866128268705e-06, + "loss": 0.7092, + "step": 429 + }, + { + "epoch": 0.35615681943677524, + "grad_norm": 0.8869542479515076, + "learning_rate": 4.97370729778803e-06, + "loss": 0.7099, + "step": 430 + }, + { + "epoch": 0.356985091109884, + "grad_norm": 0.8830757141113281, + "learning_rate": 4.973547988668055e-06, + "loss": 0.7294, + "step": 431 + }, + { + "epoch": 0.3578133627829928, + "grad_norm": 0.9321587681770325, + "learning_rate": 4.973388200939601e-06, + "loss": 0.7276, + "step": 432 + }, + { + "epoch": 0.3586416344561016, + "grad_norm": 0.9408143758773804, + "learning_rate": 4.9732279346335866e-06, + "loss": 0.7216, + "step": 433 + }, + { + "epoch": 0.35946990612921037, + "grad_norm": 0.9189308285713196, + "learning_rate": 4.973067189781022e-06, + "loss": 0.7079, + "step": 434 + }, + { + "epoch": 0.36029817780231915, + "grad_norm": 0.9198120832443237, + "learning_rate": 4.97290596641301e-06, + "loss": 0.7188, + "step": 435 + }, + { + "epoch": 0.36112644947542794, + "grad_norm": 0.9123285412788391, + "learning_rate": 4.972744264560746e-06, + "loss": 0.7328, + "step": 436 + }, + { + "epoch": 0.3619547211485367, + "grad_norm": 0.8722008466720581, + "learning_rate": 4.972582084255517e-06, + "loss": 0.7228, + "step": 437 + }, + { + "epoch": 0.3627829928216455, + "grad_norm": 1.035325288772583, + "learning_rate": 4.972419425528704e-06, + "loss": 0.72, + "step": 438 + }, + { + "epoch": 0.3636112644947543, + "grad_norm": 0.9577400088310242, + "learning_rate": 4.97225628841178e-06, + "loss": 0.7382, + "step": 439 + }, + { + "epoch": 0.36443953616786307, + "grad_norm": 0.9206854104995728, + "learning_rate": 4.972092672936309e-06, + "loss": 0.7294, + "step": 440 + }, + { + "epoch": 0.36526780784097185, + "grad_norm": 0.9281194806098938, + "learning_rate": 4.971928579133952e-06, + "loss": 0.7137, + "step": 441 + }, + { + "epoch": 0.36609607951408063, + "grad_norm": 0.951049268245697, + "learning_rate": 4.971764007036457e-06, + "loss": 0.7354, + "step": 442 + }, + { + "epoch": 0.3669243511871894, + "grad_norm": 0.8779566287994385, + "learning_rate": 4.971598956675668e-06, + "loss": 0.6858, + "step": 443 + }, + { + "epoch": 0.3677526228602982, + "grad_norm": 0.9221998453140259, + "learning_rate": 4.971433428083522e-06, + "loss": 0.7116, + "step": 444 + }, + { + "epoch": 0.368580894533407, + "grad_norm": 0.9608840346336365, + "learning_rate": 4.971267421292045e-06, + "loss": 0.7129, + "step": 445 + }, + { + "epoch": 0.36940916620651576, + "grad_norm": 0.8761363625526428, + "learning_rate": 4.971100936333359e-06, + "loss": 0.718, + "step": 446 + }, + { + "epoch": 0.37023743787962454, + "grad_norm": 0.8808798789978027, + "learning_rate": 4.970933973239677e-06, + "loss": 0.6986, + "step": 447 + }, + { + "epoch": 0.37106570955273327, + "grad_norm": 0.8818857669830322, + "learning_rate": 4.9707665320433055e-06, + "loss": 0.7465, + "step": 448 + }, + { + "epoch": 0.37189398122584205, + "grad_norm": 0.9144098162651062, + "learning_rate": 4.970598612776641e-06, + "loss": 0.7187, + "step": 449 + }, + { + "epoch": 0.37272225289895083, + "grad_norm": 0.8439158797264099, + "learning_rate": 4.9704302154721776e-06, + "loss": 0.7195, + "step": 450 + }, + { + "epoch": 0.3735505245720596, + "grad_norm": 0.8661074638366699, + "learning_rate": 4.970261340162495e-06, + "loss": 0.712, + "step": 451 + }, + { + "epoch": 0.3743787962451684, + "grad_norm": 0.8818789124488831, + "learning_rate": 4.970091986880271e-06, + "loss": 0.7199, + "step": 452 + }, + { + "epoch": 0.3752070679182772, + "grad_norm": 0.9560503363609314, + "learning_rate": 4.969922155658273e-06, + "loss": 0.7145, + "step": 453 + }, + { + "epoch": 0.37603533959138596, + "grad_norm": 0.8723775744438171, + "learning_rate": 4.969751846529363e-06, + "loss": 0.7152, + "step": 454 + }, + { + "epoch": 0.37686361126449475, + "grad_norm": 0.8612253069877625, + "learning_rate": 4.9695810595264935e-06, + "loss": 0.728, + "step": 455 + }, + { + "epoch": 0.37769188293760353, + "grad_norm": 0.8995444178581238, + "learning_rate": 4.969409794682709e-06, + "loss": 0.678, + "step": 456 + }, + { + "epoch": 0.3785201546107123, + "grad_norm": 0.8770595788955688, + "learning_rate": 4.96923805203115e-06, + "loss": 0.7354, + "step": 457 + }, + { + "epoch": 0.3793484262838211, + "grad_norm": 0.893329918384552, + "learning_rate": 4.969065831605045e-06, + "loss": 0.7102, + "step": 458 + }, + { + "epoch": 0.3801766979569299, + "grad_norm": 0.954058051109314, + "learning_rate": 4.968893133437718e-06, + "loss": 0.7144, + "step": 459 + }, + { + "epoch": 0.38100496963003866, + "grad_norm": 0.9711002707481384, + "learning_rate": 4.968719957562585e-06, + "loss": 0.7452, + "step": 460 + }, + { + "epoch": 0.38183324130314744, + "grad_norm": 0.9036396741867065, + "learning_rate": 4.968546304013153e-06, + "loss": 0.7535, + "step": 461 + }, + { + "epoch": 0.3826615129762562, + "grad_norm": 0.8948055505752563, + "learning_rate": 4.968372172823023e-06, + "loss": 0.7284, + "step": 462 + }, + { + "epoch": 0.383489784649365, + "grad_norm": 0.9134189486503601, + "learning_rate": 4.968197564025886e-06, + "loss": 0.723, + "step": 463 + }, + { + "epoch": 0.3843180563224738, + "grad_norm": 0.9191608428955078, + "learning_rate": 4.9680224776555305e-06, + "loss": 0.7078, + "step": 464 + }, + { + "epoch": 0.38514632799558257, + "grad_norm": 0.879683792591095, + "learning_rate": 4.9678469137458315e-06, + "loss": 0.6805, + "step": 465 + }, + { + "epoch": 0.38597459966869135, + "grad_norm": 0.8887293338775635, + "learning_rate": 4.96767087233076e-06, + "loss": 0.7296, + "step": 466 + }, + { + "epoch": 0.38680287134180014, + "grad_norm": 0.889991044998169, + "learning_rate": 4.967494353444377e-06, + "loss": 0.702, + "step": 467 + }, + { + "epoch": 0.38763114301490886, + "grad_norm": 0.9667506814002991, + "learning_rate": 4.967317357120839e-06, + "loss": 0.7305, + "step": 468 + }, + { + "epoch": 0.38845941468801765, + "grad_norm": 0.9297274351119995, + "learning_rate": 4.967139883394392e-06, + "loss": 0.7344, + "step": 469 + }, + { + "epoch": 0.38928768636112643, + "grad_norm": 0.9167365431785583, + "learning_rate": 4.966961932299377e-06, + "loss": 0.7016, + "step": 470 + }, + { + "epoch": 0.3901159580342352, + "grad_norm": 0.877378523349762, + "learning_rate": 4.966783503870224e-06, + "loss": 0.7313, + "step": 471 + }, + { + "epoch": 0.390944229707344, + "grad_norm": 0.9625933766365051, + "learning_rate": 4.966604598141458e-06, + "loss": 0.7245, + "step": 472 + }, + { + "epoch": 0.3917725013804528, + "grad_norm": 0.8859156966209412, + "learning_rate": 4.9664252151476955e-06, + "loss": 0.7094, + "step": 473 + }, + { + "epoch": 0.39260077305356156, + "grad_norm": 0.9172984957695007, + "learning_rate": 4.966245354923645e-06, + "loss": 0.6966, + "step": 474 + }, + { + "epoch": 0.39342904472667034, + "grad_norm": 0.9040864109992981, + "learning_rate": 4.96606501750411e-06, + "loss": 0.722, + "step": 475 + }, + { + "epoch": 0.3942573163997791, + "grad_norm": 0.8900951147079468, + "learning_rate": 4.965884202923982e-06, + "loss": 0.7263, + "step": 476 + }, + { + "epoch": 0.3950855880728879, + "grad_norm": 0.9096625447273254, + "learning_rate": 4.965702911218247e-06, + "loss": 0.702, + "step": 477 + }, + { + "epoch": 0.3959138597459967, + "grad_norm": 0.8801544904708862, + "learning_rate": 4.965521142421984e-06, + "loss": 0.7113, + "step": 478 + }, + { + "epoch": 0.39674213141910547, + "grad_norm": 0.9054175615310669, + "learning_rate": 4.965338896570363e-06, + "loss": 0.7135, + "step": 479 + }, + { + "epoch": 0.39757040309221425, + "grad_norm": 0.8734445571899414, + "learning_rate": 4.965156173698648e-06, + "loss": 0.7181, + "step": 480 + }, + { + "epoch": 0.39839867476532304, + "grad_norm": 0.9113215208053589, + "learning_rate": 4.964972973842191e-06, + "loss": 0.7008, + "step": 481 + }, + { + "epoch": 0.3992269464384318, + "grad_norm": 0.8941736221313477, + "learning_rate": 4.964789297036444e-06, + "loss": 0.7201, + "step": 482 + }, + { + "epoch": 0.4000552181115406, + "grad_norm": 0.8555464148521423, + "learning_rate": 4.964605143316944e-06, + "loss": 0.7173, + "step": 483 + }, + { + "epoch": 0.4008834897846494, + "grad_norm": 0.8719691634178162, + "learning_rate": 4.9644205127193225e-06, + "loss": 0.7211, + "step": 484 + }, + { + "epoch": 0.40171176145775817, + "grad_norm": 0.901626467704773, + "learning_rate": 4.964235405279306e-06, + "loss": 0.7223, + "step": 485 + }, + { + "epoch": 0.40254003313086695, + "grad_norm": 0.856769859790802, + "learning_rate": 4.9640498210327095e-06, + "loss": 0.7103, + "step": 486 + }, + { + "epoch": 0.40336830480397573, + "grad_norm": 0.9053338766098022, + "learning_rate": 4.963863760015442e-06, + "loss": 0.7301, + "step": 487 + }, + { + "epoch": 0.40419657647708446, + "grad_norm": 0.8989829421043396, + "learning_rate": 4.963677222263505e-06, + "loss": 0.7391, + "step": 488 + }, + { + "epoch": 0.40502484815019324, + "grad_norm": 0.9187886118888855, + "learning_rate": 4.963490207812991e-06, + "loss": 0.7179, + "step": 489 + }, + { + "epoch": 0.405853119823302, + "grad_norm": 0.8712778687477112, + "learning_rate": 4.963302716700087e-06, + "loss": 0.7088, + "step": 490 + }, + { + "epoch": 0.4066813914964108, + "grad_norm": 1.0485447645187378, + "learning_rate": 4.9631147489610695e-06, + "loss": 0.7068, + "step": 491 + }, + { + "epoch": 0.4075096631695196, + "grad_norm": 0.9332602024078369, + "learning_rate": 4.962926304632309e-06, + "loss": 0.7201, + "step": 492 + }, + { + "epoch": 0.40833793484262837, + "grad_norm": 0.9264400601387024, + "learning_rate": 4.962737383750268e-06, + "loss": 0.7136, + "step": 493 + }, + { + "epoch": 0.40916620651573715, + "grad_norm": 0.8812651634216309, + "learning_rate": 4.962547986351501e-06, + "loss": 0.7315, + "step": 494 + }, + { + "epoch": 0.40999447818884593, + "grad_norm": 0.8905262351036072, + "learning_rate": 4.962358112472653e-06, + "loss": 0.7127, + "step": 495 + }, + { + "epoch": 0.4108227498619547, + "grad_norm": 0.887514591217041, + "learning_rate": 4.962167762150465e-06, + "loss": 0.7602, + "step": 496 + }, + { + "epoch": 0.4116510215350635, + "grad_norm": 0.8683431148529053, + "learning_rate": 4.9619769354217665e-06, + "loss": 0.7091, + "step": 497 + }, + { + "epoch": 0.4124792932081723, + "grad_norm": 0.944595217704773, + "learning_rate": 4.961785632323481e-06, + "loss": 0.708, + "step": 498 + }, + { + "epoch": 0.41330756488128106, + "grad_norm": 0.8834611773490906, + "learning_rate": 4.961593852892625e-06, + "loss": 0.7308, + "step": 499 + }, + { + "epoch": 0.41413583655438985, + "grad_norm": 0.8479524850845337, + "learning_rate": 4.961401597166304e-06, + "loss": 0.7061, + "step": 500 + }, + { + "epoch": 0.41496410822749863, + "grad_norm": 0.9092981219291687, + "learning_rate": 4.96120886518172e-06, + "loss": 0.7176, + "step": 501 + }, + { + "epoch": 0.4157923799006074, + "grad_norm": 0.9024932384490967, + "learning_rate": 4.961015656976163e-06, + "loss": 0.6944, + "step": 502 + }, + { + "epoch": 0.4166206515737162, + "grad_norm": 0.8945576548576355, + "learning_rate": 4.960821972587018e-06, + "loss": 0.6958, + "step": 503 + }, + { + "epoch": 0.417448923246825, + "grad_norm": 0.8668336868286133, + "learning_rate": 4.9606278120517606e-06, + "loss": 0.7219, + "step": 504 + }, + { + "epoch": 0.41827719491993376, + "grad_norm": 0.8831227421760559, + "learning_rate": 4.96043317540796e-06, + "loss": 0.7124, + "step": 505 + }, + { + "epoch": 0.41910546659304254, + "grad_norm": 0.9637073874473572, + "learning_rate": 4.960238062693274e-06, + "loss": 0.7337, + "step": 506 + }, + { + "epoch": 0.4199337382661513, + "grad_norm": 0.8895899653434753, + "learning_rate": 4.960042473945459e-06, + "loss": 0.7459, + "step": 507 + }, + { + "epoch": 0.42076200993926005, + "grad_norm": 0.8762213587760925, + "learning_rate": 4.959846409202356e-06, + "loss": 0.7319, + "step": 508 + }, + { + "epoch": 0.42159028161236883, + "grad_norm": 0.9177393317222595, + "learning_rate": 4.959649868501903e-06, + "loss": 0.7114, + "step": 509 + }, + { + "epoch": 0.4224185532854776, + "grad_norm": 0.88541179895401, + "learning_rate": 4.9594528518821305e-06, + "loss": 0.716, + "step": 510 + }, + { + "epoch": 0.4232468249585864, + "grad_norm": 0.865566611289978, + "learning_rate": 4.959255359381158e-06, + "loss": 0.7098, + "step": 511 + }, + { + "epoch": 0.4240750966316952, + "grad_norm": 0.8629707098007202, + "learning_rate": 4.959057391037197e-06, + "loss": 0.7146, + "step": 512 + }, + { + "epoch": 0.42490336830480396, + "grad_norm": 0.8642273545265198, + "learning_rate": 4.958858946888554e-06, + "loss": 0.7345, + "step": 513 + }, + { + "epoch": 0.42573163997791275, + "grad_norm": 0.9157605171203613, + "learning_rate": 4.958660026973626e-06, + "loss": 0.7177, + "step": 514 + }, + { + "epoch": 0.42655991165102153, + "grad_norm": 0.9108191728591919, + "learning_rate": 4.958460631330903e-06, + "loss": 0.7317, + "step": 515 + }, + { + "epoch": 0.4273881833241303, + "grad_norm": 0.8524913191795349, + "learning_rate": 4.958260759998965e-06, + "loss": 0.7193, + "step": 516 + }, + { + "epoch": 0.4282164549972391, + "grad_norm": 0.9197820425033569, + "learning_rate": 4.958060413016485e-06, + "loss": 0.7127, + "step": 517 + }, + { + "epoch": 0.4290447266703479, + "grad_norm": 0.8744186758995056, + "learning_rate": 4.9578595904222296e-06, + "loss": 0.6968, + "step": 518 + }, + { + "epoch": 0.42987299834345666, + "grad_norm": 0.9071037769317627, + "learning_rate": 4.957658292255055e-06, + "loss": 0.6963, + "step": 519 + }, + { + "epoch": 0.43070127001656544, + "grad_norm": 0.937796950340271, + "learning_rate": 4.95745651855391e-06, + "loss": 0.7193, + "step": 520 + }, + { + "epoch": 0.4315295416896742, + "grad_norm": 0.883471667766571, + "learning_rate": 4.9572542693578375e-06, + "loss": 0.6886, + "step": 521 + }, + { + "epoch": 0.432357813362783, + "grad_norm": 0.8738589882850647, + "learning_rate": 4.9570515447059705e-06, + "loss": 0.7072, + "step": 522 + }, + { + "epoch": 0.4331860850358918, + "grad_norm": 0.9039497971534729, + "learning_rate": 4.9568483446375335e-06, + "loss": 0.7251, + "step": 523 + }, + { + "epoch": 0.43401435670900057, + "grad_norm": 0.896789014339447, + "learning_rate": 4.956644669191845e-06, + "loss": 0.7021, + "step": 524 + }, + { + "epoch": 0.43484262838210935, + "grad_norm": 0.8883570432662964, + "learning_rate": 4.9564405184083126e-06, + "loss": 0.7153, + "step": 525 + }, + { + "epoch": 0.43567090005521814, + "grad_norm": 0.8869197368621826, + "learning_rate": 4.956235892326438e-06, + "loss": 0.7061, + "step": 526 + }, + { + "epoch": 0.4364991717283269, + "grad_norm": 0.9237920641899109, + "learning_rate": 4.9560307909858165e-06, + "loss": 0.6988, + "step": 527 + }, + { + "epoch": 0.43732744340143564, + "grad_norm": 0.9359283447265625, + "learning_rate": 4.955825214426131e-06, + "loss": 0.738, + "step": 528 + }, + { + "epoch": 0.4381557150745444, + "grad_norm": 0.8602650761604309, + "learning_rate": 4.95561916268716e-06, + "loss": 0.709, + "step": 529 + }, + { + "epoch": 0.4389839867476532, + "grad_norm": 0.9013513922691345, + "learning_rate": 4.955412635808771e-06, + "loss": 0.6836, + "step": 530 + }, + { + "epoch": 0.439812258420762, + "grad_norm": 0.9136670827865601, + "learning_rate": 4.955205633830927e-06, + "loss": 0.7263, + "step": 531 + }, + { + "epoch": 0.4406405300938708, + "grad_norm": 0.9013793468475342, + "learning_rate": 4.95499815679368e-06, + "loss": 0.6937, + "step": 532 + }, + { + "epoch": 0.44146880176697956, + "grad_norm": 0.9017112851142883, + "learning_rate": 4.954790204737174e-06, + "loss": 0.6913, + "step": 533 + }, + { + "epoch": 0.44229707344008834, + "grad_norm": 0.8901436924934387, + "learning_rate": 4.954581777701647e-06, + "loss": 0.7241, + "step": 534 + }, + { + "epoch": 0.4431253451131971, + "grad_norm": 0.8882759809494019, + "learning_rate": 4.954372875727427e-06, + "loss": 0.7115, + "step": 535 + }, + { + "epoch": 0.4439536167863059, + "grad_norm": 0.8909885883331299, + "learning_rate": 4.954163498854936e-06, + "loss": 0.7146, + "step": 536 + }, + { + "epoch": 0.4447818884594147, + "grad_norm": 0.8951786756515503, + "learning_rate": 4.953953647124684e-06, + "loss": 0.699, + "step": 537 + }, + { + "epoch": 0.44561016013252347, + "grad_norm": 0.8890933990478516, + "learning_rate": 4.953743320577278e-06, + "loss": 0.7018, + "step": 538 + }, + { + "epoch": 0.44643843180563225, + "grad_norm": 0.910393238067627, + "learning_rate": 4.953532519253412e-06, + "loss": 0.7299, + "step": 539 + }, + { + "epoch": 0.44726670347874103, + "grad_norm": 0.9152857065200806, + "learning_rate": 4.953321243193876e-06, + "loss": 0.6984, + "step": 540 + }, + { + "epoch": 0.4480949751518498, + "grad_norm": 0.9001837372779846, + "learning_rate": 4.9531094924395485e-06, + "loss": 0.727, + "step": 541 + }, + { + "epoch": 0.4489232468249586, + "grad_norm": 0.882060170173645, + "learning_rate": 4.952897267031401e-06, + "loss": 0.7016, + "step": 542 + }, + { + "epoch": 0.4497515184980674, + "grad_norm": 0.8665159344673157, + "learning_rate": 4.9526845670104995e-06, + "loss": 0.7139, + "step": 543 + }, + { + "epoch": 0.45057979017117616, + "grad_norm": 0.9200844764709473, + "learning_rate": 4.952471392417997e-06, + "loss": 0.7016, + "step": 544 + }, + { + "epoch": 0.45140806184428495, + "grad_norm": 0.8852037191390991, + "learning_rate": 4.9522577432951416e-06, + "loss": 0.7048, + "step": 545 + }, + { + "epoch": 0.45223633351739373, + "grad_norm": 0.9003865122795105, + "learning_rate": 4.952043619683274e-06, + "loss": 0.7106, + "step": 546 + }, + { + "epoch": 0.4530646051905025, + "grad_norm": 0.8888446092605591, + "learning_rate": 4.951829021623822e-06, + "loss": 0.7202, + "step": 547 + }, + { + "epoch": 0.45389287686361124, + "grad_norm": 0.8940277099609375, + "learning_rate": 4.9516139491583094e-06, + "loss": 0.7156, + "step": 548 + }, + { + "epoch": 0.45472114853672, + "grad_norm": 0.9066211581230164, + "learning_rate": 4.951398402328353e-06, + "loss": 0.6865, + "step": 549 + }, + { + "epoch": 0.4555494202098288, + "grad_norm": 0.9352865815162659, + "learning_rate": 4.9511823811756565e-06, + "loss": 0.7117, + "step": 550 + }, + { + "epoch": 0.4563776918829376, + "grad_norm": 0.8672269582748413, + "learning_rate": 4.95096588574202e-06, + "loss": 0.7166, + "step": 551 + }, + { + "epoch": 0.45720596355604637, + "grad_norm": 0.8730685114860535, + "learning_rate": 4.95074891606933e-06, + "loss": 0.6998, + "step": 552 + }, + { + "epoch": 0.45803423522915515, + "grad_norm": 0.896821916103363, + "learning_rate": 4.950531472199572e-06, + "loss": 0.6839, + "step": 553 + }, + { + "epoch": 0.45886250690226393, + "grad_norm": 0.918572187423706, + "learning_rate": 4.950313554174817e-06, + "loss": 0.7155, + "step": 554 + }, + { + "epoch": 0.4596907785753727, + "grad_norm": 0.8670752048492432, + "learning_rate": 4.950095162037232e-06, + "loss": 0.708, + "step": 555 + }, + { + "epoch": 0.4605190502484815, + "grad_norm": 0.875618040561676, + "learning_rate": 4.9498762958290715e-06, + "loss": 0.6956, + "step": 556 + }, + { + "epoch": 0.4613473219215903, + "grad_norm": 0.9120676517486572, + "learning_rate": 4.949656955592685e-06, + "loss": 0.7327, + "step": 557 + }, + { + "epoch": 0.46217559359469906, + "grad_norm": 0.9670370817184448, + "learning_rate": 4.949437141370514e-06, + "loss": 0.705, + "step": 558 + }, + { + "epoch": 0.46300386526780785, + "grad_norm": 0.9155322313308716, + "learning_rate": 4.949216853205089e-06, + "loss": 0.7126, + "step": 559 + }, + { + "epoch": 0.46383213694091663, + "grad_norm": 0.8945392370223999, + "learning_rate": 4.948996091139034e-06, + "loss": 0.7083, + "step": 560 + }, + { + "epoch": 0.4646604086140254, + "grad_norm": 0.8922162055969238, + "learning_rate": 4.948774855215065e-06, + "loss": 0.7189, + "step": 561 + }, + { + "epoch": 0.4654886802871342, + "grad_norm": 0.8940657377243042, + "learning_rate": 4.948553145475989e-06, + "loss": 0.6934, + "step": 562 + }, + { + "epoch": 0.466316951960243, + "grad_norm": 0.8946855068206787, + "learning_rate": 4.948330961964703e-06, + "loss": 0.6835, + "step": 563 + }, + { + "epoch": 0.46714522363335176, + "grad_norm": 0.9002820253372192, + "learning_rate": 4.948108304724201e-06, + "loss": 0.712, + "step": 564 + }, + { + "epoch": 0.46797349530646054, + "grad_norm": 0.8715711832046509, + "learning_rate": 4.947885173797561e-06, + "loss": 0.7197, + "step": 565 + }, + { + "epoch": 0.4688017669795693, + "grad_norm": 0.9269874095916748, + "learning_rate": 4.94766156922796e-06, + "loss": 0.7311, + "step": 566 + }, + { + "epoch": 0.4696300386526781, + "grad_norm": 0.8999423980712891, + "learning_rate": 4.947437491058662e-06, + "loss": 0.7144, + "step": 567 + }, + { + "epoch": 0.47045831032578683, + "grad_norm": 0.8704255223274231, + "learning_rate": 4.947212939333024e-06, + "loss": 0.7156, + "step": 568 + }, + { + "epoch": 0.4712865819988956, + "grad_norm": 0.8770347237586975, + "learning_rate": 4.946987914094495e-06, + "loss": 0.6969, + "step": 569 + }, + { + "epoch": 0.4721148536720044, + "grad_norm": 0.8865682482719421, + "learning_rate": 4.946762415386615e-06, + "loss": 0.7089, + "step": 570 + }, + { + "epoch": 0.4729431253451132, + "grad_norm": 0.8756279349327087, + "learning_rate": 4.946536443253016e-06, + "loss": 0.7159, + "step": 571 + }, + { + "epoch": 0.47377139701822196, + "grad_norm": 0.8844839930534363, + "learning_rate": 4.946309997737422e-06, + "loss": 0.7234, + "step": 572 + }, + { + "epoch": 0.47459966869133074, + "grad_norm": 0.8549885153770447, + "learning_rate": 4.946083078883647e-06, + "loss": 0.6927, + "step": 573 + }, + { + "epoch": 0.4754279403644395, + "grad_norm": 0.8967456817626953, + "learning_rate": 4.945855686735599e-06, + "loss": 0.7082, + "step": 574 + }, + { + "epoch": 0.4762562120375483, + "grad_norm": 0.9120200276374817, + "learning_rate": 4.9456278213372755e-06, + "loss": 0.7135, + "step": 575 + }, + { + "epoch": 0.4770844837106571, + "grad_norm": 0.8780766725540161, + "learning_rate": 4.945399482732765e-06, + "loss": 0.7139, + "step": 576 + }, + { + "epoch": 0.4779127553837659, + "grad_norm": 0.8900999426841736, + "learning_rate": 4.945170670966253e-06, + "loss": 0.7132, + "step": 577 + }, + { + "epoch": 0.47874102705687466, + "grad_norm": 0.9402151107788086, + "learning_rate": 4.944941386082008e-06, + "loss": 0.7139, + "step": 578 + }, + { + "epoch": 0.47956929872998344, + "grad_norm": 0.8769465088844299, + "learning_rate": 4.944711628124396e-06, + "loss": 0.7236, + "step": 579 + }, + { + "epoch": 0.4803975704030922, + "grad_norm": 0.9185771346092224, + "learning_rate": 4.944481397137874e-06, + "loss": 0.6884, + "step": 580 + }, + { + "epoch": 0.481225842076201, + "grad_norm": 0.8775897026062012, + "learning_rate": 4.944250693166988e-06, + "loss": 0.731, + "step": 581 + }, + { + "epoch": 0.4820541137493098, + "grad_norm": 0.8751668930053711, + "learning_rate": 4.94401951625638e-06, + "loss": 0.7007, + "step": 582 + }, + { + "epoch": 0.48288238542241857, + "grad_norm": 0.9042515754699707, + "learning_rate": 4.943787866450776e-06, + "loss": 0.7064, + "step": 583 + }, + { + "epoch": 0.48371065709552735, + "grad_norm": 0.8845478892326355, + "learning_rate": 4.943555743795002e-06, + "loss": 0.7069, + "step": 584 + }, + { + "epoch": 0.48453892876863613, + "grad_norm": 0.8959096670150757, + "learning_rate": 4.943323148333969e-06, + "loss": 0.6871, + "step": 585 + }, + { + "epoch": 0.4853672004417449, + "grad_norm": 0.9021669626235962, + "learning_rate": 4.943090080112684e-06, + "loss": 0.7156, + "step": 586 + }, + { + "epoch": 0.4861954721148537, + "grad_norm": 0.8768396973609924, + "learning_rate": 4.942856539176243e-06, + "loss": 0.681, + "step": 587 + }, + { + "epoch": 0.4870237437879624, + "grad_norm": 0.8734706044197083, + "learning_rate": 4.942622525569832e-06, + "loss": 0.7371, + "step": 588 + }, + { + "epoch": 0.4878520154610712, + "grad_norm": 0.9155406355857849, + "learning_rate": 4.942388039338733e-06, + "loss": 0.7012, + "step": 589 + }, + { + "epoch": 0.48868028713418, + "grad_norm": 0.8778123259544373, + "learning_rate": 4.942153080528317e-06, + "loss": 0.7201, + "step": 590 + }, + { + "epoch": 0.4895085588072888, + "grad_norm": 0.9102092981338501, + "learning_rate": 4.941917649184045e-06, + "loss": 0.6951, + "step": 591 + }, + { + "epoch": 0.49033683048039756, + "grad_norm": 0.8855768442153931, + "learning_rate": 4.941681745351471e-06, + "loss": 0.6997, + "step": 592 + }, + { + "epoch": 0.49116510215350634, + "grad_norm": 0.8770707845687866, + "learning_rate": 4.94144536907624e-06, + "loss": 0.7092, + "step": 593 + }, + { + "epoch": 0.4919933738266151, + "grad_norm": 0.900458574295044, + "learning_rate": 4.9412085204040896e-06, + "loss": 0.6909, + "step": 594 + }, + { + "epoch": 0.4928216454997239, + "grad_norm": 0.903489887714386, + "learning_rate": 4.940971199380846e-06, + "loss": 0.7066, + "step": 595 + }, + { + "epoch": 0.4936499171728327, + "grad_norm": 0.9331871271133423, + "learning_rate": 4.940733406052431e-06, + "loss": 0.6924, + "step": 596 + }, + { + "epoch": 0.49447818884594147, + "grad_norm": 0.8855593800544739, + "learning_rate": 4.9404951404648535e-06, + "loss": 0.6991, + "step": 597 + }, + { + "epoch": 0.49530646051905025, + "grad_norm": 0.9075731039047241, + "learning_rate": 4.940256402664216e-06, + "loss": 0.6973, + "step": 598 + }, + { + "epoch": 0.49613473219215903, + "grad_norm": 0.8970743417739868, + "learning_rate": 4.940017192696713e-06, + "loss": 0.6767, + "step": 599 + }, + { + "epoch": 0.4969630038652678, + "grad_norm": 0.8644358515739441, + "learning_rate": 4.939777510608628e-06, + "loss": 0.7284, + "step": 600 + }, + { + "epoch": 0.4977912755383766, + "grad_norm": 0.8807653784751892, + "learning_rate": 4.939537356446339e-06, + "loss": 0.7143, + "step": 601 + }, + { + "epoch": 0.4986195472114854, + "grad_norm": 0.892946183681488, + "learning_rate": 4.939296730256312e-06, + "loss": 0.7021, + "step": 602 + }, + { + "epoch": 0.49944781888459416, + "grad_norm": 0.9661110639572144, + "learning_rate": 4.939055632085107e-06, + "loss": 0.7198, + "step": 603 + }, + { + "epoch": 0.500276090557703, + "grad_norm": 0.8638496398925781, + "learning_rate": 4.9388140619793735e-06, + "loss": 0.6943, + "step": 604 + }, + { + "epoch": 0.5011043622308117, + "grad_norm": 0.9025457501411438, + "learning_rate": 4.938572019985853e-06, + "loss": 0.7114, + "step": 605 + }, + { + "epoch": 0.5019326339039205, + "grad_norm": 0.9622734785079956, + "learning_rate": 4.93832950615138e-06, + "loss": 0.7086, + "step": 606 + }, + { + "epoch": 0.5027609055770292, + "grad_norm": 0.9565468430519104, + "learning_rate": 4.938086520522876e-06, + "loss": 0.7125, + "step": 607 + }, + { + "epoch": 0.5035891772501381, + "grad_norm": 0.8970094323158264, + "learning_rate": 4.937843063147359e-06, + "loss": 0.7023, + "step": 608 + }, + { + "epoch": 0.5044174489232468, + "grad_norm": 0.8920996785163879, + "learning_rate": 4.937599134071933e-06, + "loss": 0.7041, + "step": 609 + }, + { + "epoch": 0.5052457205963556, + "grad_norm": 0.9304755926132202, + "learning_rate": 4.937354733343799e-06, + "loss": 0.7165, + "step": 610 + }, + { + "epoch": 0.5060739922694644, + "grad_norm": 0.8995933532714844, + "learning_rate": 4.937109861010244e-06, + "loss": 0.6907, + "step": 611 + }, + { + "epoch": 0.5069022639425732, + "grad_norm": 0.8648850321769714, + "learning_rate": 4.936864517118649e-06, + "loss": 0.6994, + "step": 612 + }, + { + "epoch": 0.5077305356156819, + "grad_norm": 0.8414506316184998, + "learning_rate": 4.936618701716487e-06, + "loss": 0.7081, + "step": 613 + }, + { + "epoch": 0.5085588072887908, + "grad_norm": 0.9282650947570801, + "learning_rate": 4.936372414851321e-06, + "loss": 0.7303, + "step": 614 + }, + { + "epoch": 0.5093870789618995, + "grad_norm": 0.877757728099823, + "learning_rate": 4.936125656570802e-06, + "loss": 0.7132, + "step": 615 + }, + { + "epoch": 0.5102153506350083, + "grad_norm": 0.8645135164260864, + "learning_rate": 4.935878426922679e-06, + "loss": 0.723, + "step": 616 + }, + { + "epoch": 0.5110436223081171, + "grad_norm": 0.9014747738838196, + "learning_rate": 4.935630725954787e-06, + "loss": 0.7167, + "step": 617 + }, + { + "epoch": 0.5118718939812258, + "grad_norm": 0.9295386672019958, + "learning_rate": 4.935382553715055e-06, + "loss": 0.705, + "step": 618 + }, + { + "epoch": 0.5127001656543346, + "grad_norm": 0.8910616040229797, + "learning_rate": 4.935133910251499e-06, + "loss": 0.6781, + "step": 619 + }, + { + "epoch": 0.5135284373274434, + "grad_norm": 0.8804931640625, + "learning_rate": 4.934884795612234e-06, + "loss": 0.6984, + "step": 620 + }, + { + "epoch": 0.5143567090005522, + "grad_norm": 0.8223545551300049, + "learning_rate": 4.934635209845458e-06, + "loss": 0.6892, + "step": 621 + }, + { + "epoch": 0.5151849806736609, + "grad_norm": 0.8509450554847717, + "learning_rate": 4.934385152999463e-06, + "loss": 0.7139, + "step": 622 + }, + { + "epoch": 0.5160132523467698, + "grad_norm": 0.8944748044013977, + "learning_rate": 4.934134625122636e-06, + "loss": 0.6976, + "step": 623 + }, + { + "epoch": 0.5168415240198785, + "grad_norm": 0.9021086692810059, + "learning_rate": 4.933883626263448e-06, + "loss": 0.7189, + "step": 624 + }, + { + "epoch": 0.5176697956929873, + "grad_norm": 0.8939547538757324, + "learning_rate": 4.933632156470467e-06, + "loss": 0.704, + "step": 625 + }, + { + "epoch": 0.518498067366096, + "grad_norm": 0.8793498277664185, + "learning_rate": 4.93338021579235e-06, + "loss": 0.7046, + "step": 626 + }, + { + "epoch": 0.5193263390392049, + "grad_norm": 0.8728756308555603, + "learning_rate": 4.933127804277845e-06, + "loss": 0.6826, + "step": 627 + }, + { + "epoch": 0.5201546107123136, + "grad_norm": 0.8989412188529968, + "learning_rate": 4.932874921975792e-06, + "loss": 0.6842, + "step": 628 + }, + { + "epoch": 0.5209828823854225, + "grad_norm": 0.8836191296577454, + "learning_rate": 4.93262156893512e-06, + "loss": 0.7061, + "step": 629 + }, + { + "epoch": 0.5218111540585312, + "grad_norm": 0.9429040551185608, + "learning_rate": 4.9323677452048505e-06, + "loss": 0.7053, + "step": 630 + }, + { + "epoch": 0.52263942573164, + "grad_norm": 0.879463255405426, + "learning_rate": 4.932113450834097e-06, + "loss": 0.6926, + "step": 631 + }, + { + "epoch": 0.5234676974047487, + "grad_norm": 0.8566175699234009, + "learning_rate": 4.931858685872062e-06, + "loss": 0.7126, + "step": 632 + }, + { + "epoch": 0.5242959690778576, + "grad_norm": 0.8694538474082947, + "learning_rate": 4.931603450368042e-06, + "loss": 0.7153, + "step": 633 + }, + { + "epoch": 0.5251242407509663, + "grad_norm": 0.8754484057426453, + "learning_rate": 4.931347744371421e-06, + "loss": 0.711, + "step": 634 + }, + { + "epoch": 0.5259525124240751, + "grad_norm": 0.8982204794883728, + "learning_rate": 4.931091567931675e-06, + "loss": 0.7168, + "step": 635 + }, + { + "epoch": 0.5267807840971839, + "grad_norm": 0.9262725710868835, + "learning_rate": 4.930834921098374e-06, + "loss": 0.7315, + "step": 636 + }, + { + "epoch": 0.5276090557702926, + "grad_norm": 0.9271771311759949, + "learning_rate": 4.930577803921176e-06, + "loss": 0.6968, + "step": 637 + }, + { + "epoch": 0.5284373274434014, + "grad_norm": 0.8855197429656982, + "learning_rate": 4.93032021644983e-06, + "loss": 0.6986, + "step": 638 + }, + { + "epoch": 0.5292655991165102, + "grad_norm": 0.8615745902061462, + "learning_rate": 4.930062158734178e-06, + "loss": 0.7014, + "step": 639 + }, + { + "epoch": 0.530093870789619, + "grad_norm": 0.9004185795783997, + "learning_rate": 4.92980363082415e-06, + "loss": 0.7045, + "step": 640 + }, + { + "epoch": 0.5309221424627277, + "grad_norm": 0.8533529043197632, + "learning_rate": 4.92954463276977e-06, + "loss": 0.6911, + "step": 641 + }, + { + "epoch": 0.5317504141358366, + "grad_norm": 0.881901204586029, + "learning_rate": 4.929285164621152e-06, + "loss": 0.7148, + "step": 642 + }, + { + "epoch": 0.5325786858089453, + "grad_norm": 0.8778982162475586, + "learning_rate": 4.9290252264285e-06, + "loss": 0.6879, + "step": 643 + }, + { + "epoch": 0.5334069574820541, + "grad_norm": 0.9455735087394714, + "learning_rate": 4.928764818242109e-06, + "loss": 0.7088, + "step": 644 + }, + { + "epoch": 0.5342352291551629, + "grad_norm": 0.9255897998809814, + "learning_rate": 4.928503940112367e-06, + "loss": 0.6817, + "step": 645 + }, + { + "epoch": 0.5350635008282717, + "grad_norm": 0.8854040503501892, + "learning_rate": 4.928242592089752e-06, + "loss": 0.7074, + "step": 646 + }, + { + "epoch": 0.5358917725013804, + "grad_norm": 0.8824400901794434, + "learning_rate": 4.927980774224829e-06, + "loss": 0.7092, + "step": 647 + }, + { + "epoch": 0.5367200441744893, + "grad_norm": 0.8873158693313599, + "learning_rate": 4.927718486568261e-06, + "loss": 0.7214, + "step": 648 + }, + { + "epoch": 0.537548315847598, + "grad_norm": 0.9013757109642029, + "learning_rate": 4.927455729170796e-06, + "loss": 0.6955, + "step": 649 + }, + { + "epoch": 0.5383765875207068, + "grad_norm": 0.9100024104118347, + "learning_rate": 4.927192502083276e-06, + "loss": 0.7089, + "step": 650 + }, + { + "epoch": 0.5392048591938156, + "grad_norm": 0.8935851454734802, + "learning_rate": 4.9269288053566335e-06, + "loss": 0.7018, + "step": 651 + }, + { + "epoch": 0.5400331308669244, + "grad_norm": 0.8726267218589783, + "learning_rate": 4.92666463904189e-06, + "loss": 0.6829, + "step": 652 + }, + { + "epoch": 0.5408614025400331, + "grad_norm": 0.9320974349975586, + "learning_rate": 4.926400003190161e-06, + "loss": 0.6766, + "step": 653 + }, + { + "epoch": 0.541689674213142, + "grad_norm": 0.9257067441940308, + "learning_rate": 4.926134897852648e-06, + "loss": 0.6913, + "step": 654 + }, + { + "epoch": 0.5425179458862507, + "grad_norm": 0.8642577528953552, + "learning_rate": 4.92586932308065e-06, + "loss": 0.7155, + "step": 655 + }, + { + "epoch": 0.5433462175593595, + "grad_norm": 0.8861536383628845, + "learning_rate": 4.9256032789255505e-06, + "loss": 0.6695, + "step": 656 + }, + { + "epoch": 0.5441744892324683, + "grad_norm": 0.8901777267456055, + "learning_rate": 4.925336765438829e-06, + "loss": 0.6847, + "step": 657 + }, + { + "epoch": 0.545002760905577, + "grad_norm": 0.8810168504714966, + "learning_rate": 4.9250697826720515e-06, + "loss": 0.7038, + "step": 658 + }, + { + "epoch": 0.5458310325786858, + "grad_norm": 0.8655433654785156, + "learning_rate": 4.924802330676877e-06, + "loss": 0.6908, + "step": 659 + }, + { + "epoch": 0.5466593042517945, + "grad_norm": 0.8786777853965759, + "learning_rate": 4.924534409505055e-06, + "loss": 0.663, + "step": 660 + }, + { + "epoch": 0.5474875759249034, + "grad_norm": 0.8837904334068298, + "learning_rate": 4.924266019208427e-06, + "loss": 0.6843, + "step": 661 + }, + { + "epoch": 0.5483158475980121, + "grad_norm": 0.8715303540229797, + "learning_rate": 4.923997159838923e-06, + "loss": 0.6841, + "step": 662 + }, + { + "epoch": 0.549144119271121, + "grad_norm": 0.8825753927230835, + "learning_rate": 4.923727831448564e-06, + "loss": 0.7053, + "step": 663 + }, + { + "epoch": 0.5499723909442297, + "grad_norm": 0.877668559551239, + "learning_rate": 4.923458034089463e-06, + "loss": 0.6624, + "step": 664 + }, + { + "epoch": 0.5508006626173385, + "grad_norm": 0.8721411824226379, + "learning_rate": 4.923187767813825e-06, + "loss": 0.7339, + "step": 665 + }, + { + "epoch": 0.5516289342904472, + "grad_norm": 0.8607308268547058, + "learning_rate": 4.922917032673941e-06, + "loss": 0.7184, + "step": 666 + }, + { + "epoch": 0.5524572059635561, + "grad_norm": 0.8456798791885376, + "learning_rate": 4.922645828722199e-06, + "loss": 0.7056, + "step": 667 + }, + { + "epoch": 0.5532854776366648, + "grad_norm": 0.8557775020599365, + "learning_rate": 4.922374156011071e-06, + "loss": 0.6942, + "step": 668 + }, + { + "epoch": 0.5541137493097736, + "grad_norm": 0.8877674341201782, + "learning_rate": 4.922102014593126e-06, + "loss": 0.7173, + "step": 669 + }, + { + "epoch": 0.5549420209828824, + "grad_norm": 0.8662481904029846, + "learning_rate": 4.921829404521019e-06, + "loss": 0.7113, + "step": 670 + }, + { + "epoch": 0.5557702926559912, + "grad_norm": 0.8887649774551392, + "learning_rate": 4.921556325847499e-06, + "loss": 0.7028, + "step": 671 + }, + { + "epoch": 0.5565985643290999, + "grad_norm": 0.9837973713874817, + "learning_rate": 4.921282778625403e-06, + "loss": 0.6975, + "step": 672 + }, + { + "epoch": 0.5574268360022088, + "grad_norm": 0.8754701018333435, + "learning_rate": 4.921008762907661e-06, + "loss": 0.6983, + "step": 673 + }, + { + "epoch": 0.5582551076753175, + "grad_norm": 0.9125426411628723, + "learning_rate": 4.920734278747291e-06, + "loss": 0.6982, + "step": 674 + }, + { + "epoch": 0.5590833793484263, + "grad_norm": 0.9424439072608948, + "learning_rate": 4.920459326197403e-06, + "loss": 0.7304, + "step": 675 + }, + { + "epoch": 0.5599116510215351, + "grad_norm": 0.8718393445014954, + "learning_rate": 4.9201839053112e-06, + "loss": 0.6644, + "step": 676 + }, + { + "epoch": 0.5607399226946438, + "grad_norm": 0.9002220630645752, + "learning_rate": 4.919908016141972e-06, + "loss": 0.6966, + "step": 677 + }, + { + "epoch": 0.5615681943677526, + "grad_norm": 0.8925678133964539, + "learning_rate": 4.9196316587431e-06, + "loss": 0.713, + "step": 678 + }, + { + "epoch": 0.5623964660408614, + "grad_norm": 0.8854464292526245, + "learning_rate": 4.919354833168059e-06, + "loss": 0.7075, + "step": 679 + }, + { + "epoch": 0.5632247377139702, + "grad_norm": 0.8867958188056946, + "learning_rate": 4.919077539470409e-06, + "loss": 0.7166, + "step": 680 + }, + { + "epoch": 0.5640530093870789, + "grad_norm": 0.862531840801239, + "learning_rate": 4.918799777703806e-06, + "loss": 0.7119, + "step": 681 + }, + { + "epoch": 0.5648812810601878, + "grad_norm": 0.8623220920562744, + "learning_rate": 4.9185215479219944e-06, + "loss": 0.6758, + "step": 682 + }, + { + "epoch": 0.5657095527332965, + "grad_norm": 0.9171786904335022, + "learning_rate": 4.918242850178808e-06, + "loss": 0.7049, + "step": 683 + }, + { + "epoch": 0.5665378244064053, + "grad_norm": 0.8947017192840576, + "learning_rate": 4.9179636845281735e-06, + "loss": 0.7144, + "step": 684 + }, + { + "epoch": 0.567366096079514, + "grad_norm": 0.9343388080596924, + "learning_rate": 4.917684051024104e-06, + "loss": 0.7024, + "step": 685 + }, + { + "epoch": 0.5681943677526229, + "grad_norm": 0.8929025530815125, + "learning_rate": 4.917403949720709e-06, + "loss": 0.7162, + "step": 686 + }, + { + "epoch": 0.5690226394257316, + "grad_norm": 0.9000421762466431, + "learning_rate": 4.9171233806721854e-06, + "loss": 0.7085, + "step": 687 + }, + { + "epoch": 0.5698509110988405, + "grad_norm": 0.9404364228248596, + "learning_rate": 4.91684234393282e-06, + "loss": 0.6996, + "step": 688 + }, + { + "epoch": 0.5706791827719492, + "grad_norm": 0.9566411375999451, + "learning_rate": 4.916560839556989e-06, + "loss": 0.7241, + "step": 689 + }, + { + "epoch": 0.571507454445058, + "grad_norm": 0.9066181182861328, + "learning_rate": 4.916278867599163e-06, + "loss": 0.724, + "step": 690 + }, + { + "epoch": 0.5723357261181667, + "grad_norm": 0.8914163708686829, + "learning_rate": 4.915996428113901e-06, + "loss": 0.7046, + "step": 691 + }, + { + "epoch": 0.5731639977912756, + "grad_norm": 0.8977548480033875, + "learning_rate": 4.91571352115585e-06, + "loss": 0.6823, + "step": 692 + }, + { + "epoch": 0.5739922694643843, + "grad_norm": 0.8902036547660828, + "learning_rate": 4.915430146779754e-06, + "loss": 0.702, + "step": 693 + }, + { + "epoch": 0.5748205411374931, + "grad_norm": 0.8859509229660034, + "learning_rate": 4.915146305040439e-06, + "loss": 0.6996, + "step": 694 + }, + { + "epoch": 0.5756488128106019, + "grad_norm": 0.8916864991188049, + "learning_rate": 4.914861995992829e-06, + "loss": 0.6978, + "step": 695 + }, + { + "epoch": 0.5764770844837107, + "grad_norm": 0.8961358070373535, + "learning_rate": 4.9145772196919336e-06, + "loss": 0.6803, + "step": 696 + }, + { + "epoch": 0.5773053561568194, + "grad_norm": 0.8817684650421143, + "learning_rate": 4.914291976192854e-06, + "loss": 0.6891, + "step": 697 + }, + { + "epoch": 0.5781336278299282, + "grad_norm": 0.8821113705635071, + "learning_rate": 4.914006265550783e-06, + "loss": 0.7009, + "step": 698 + }, + { + "epoch": 0.578961899503037, + "grad_norm": 0.9553062319755554, + "learning_rate": 4.913720087821003e-06, + "loss": 0.7118, + "step": 699 + }, + { + "epoch": 0.5797901711761457, + "grad_norm": 0.8997507691383362, + "learning_rate": 4.913433443058887e-06, + "loss": 0.7036, + "step": 700 + }, + { + "epoch": 0.5806184428492546, + "grad_norm": 0.8659473657608032, + "learning_rate": 4.9131463313198964e-06, + "loss": 0.6766, + "step": 701 + }, + { + "epoch": 0.5814467145223633, + "grad_norm": 0.8916030526161194, + "learning_rate": 4.912858752659586e-06, + "loss": 0.6976, + "step": 702 + }, + { + "epoch": 0.5822749861954721, + "grad_norm": 0.8726498484611511, + "learning_rate": 4.912570707133599e-06, + "loss": 0.6989, + "step": 703 + }, + { + "epoch": 0.5831032578685809, + "grad_norm": 0.8725529909133911, + "learning_rate": 4.91228219479767e-06, + "loss": 0.6862, + "step": 704 + }, + { + "epoch": 0.5839315295416897, + "grad_norm": 0.8807168006896973, + "learning_rate": 4.911993215707624e-06, + "loss": 0.6964, + "step": 705 + }, + { + "epoch": 0.5847598012147984, + "grad_norm": 0.9244167804718018, + "learning_rate": 4.911703769919375e-06, + "loss": 0.6742, + "step": 706 + }, + { + "epoch": 0.5855880728879073, + "grad_norm": 0.9174355268478394, + "learning_rate": 4.9114138574889285e-06, + "loss": 0.6852, + "step": 707 + }, + { + "epoch": 0.586416344561016, + "grad_norm": 0.8894474506378174, + "learning_rate": 4.9111234784723785e-06, + "loss": 0.689, + "step": 708 + }, + { + "epoch": 0.5872446162341248, + "grad_norm": 0.8856050968170166, + "learning_rate": 4.910832632925912e-06, + "loss": 0.6742, + "step": 709 + }, + { + "epoch": 0.5880728879072336, + "grad_norm": 0.9120550751686096, + "learning_rate": 4.910541320905804e-06, + "loss": 0.7016, + "step": 710 + }, + { + "epoch": 0.5889011595803424, + "grad_norm": 0.8925642967224121, + "learning_rate": 4.910249542468422e-06, + "loss": 0.7165, + "step": 711 + }, + { + "epoch": 0.5897294312534511, + "grad_norm": 0.8786419630050659, + "learning_rate": 4.909957297670222e-06, + "loss": 0.7207, + "step": 712 + }, + { + "epoch": 0.59055770292656, + "grad_norm": 0.89600670337677, + "learning_rate": 4.909664586567749e-06, + "loss": 0.672, + "step": 713 + }, + { + "epoch": 0.5913859745996687, + "grad_norm": 0.889835000038147, + "learning_rate": 4.909371409217642e-06, + "loss": 0.7089, + "step": 714 + }, + { + "epoch": 0.5922142462727775, + "grad_norm": 0.9029288291931152, + "learning_rate": 4.909077765676627e-06, + "loss": 0.7065, + "step": 715 + }, + { + "epoch": 0.5930425179458862, + "grad_norm": 0.9212722778320312, + "learning_rate": 4.908783656001521e-06, + "loss": 0.6787, + "step": 716 + }, + { + "epoch": 0.593870789618995, + "grad_norm": 0.9079681038856506, + "learning_rate": 4.9084890802492325e-06, + "loss": 0.6996, + "step": 717 + }, + { + "epoch": 0.5946990612921038, + "grad_norm": 0.9110735654830933, + "learning_rate": 4.908194038476758e-06, + "loss": 0.6962, + "step": 718 + }, + { + "epoch": 0.5955273329652125, + "grad_norm": 0.8945547938346863, + "learning_rate": 4.9078985307411865e-06, + "loss": 0.6648, + "step": 719 + }, + { + "epoch": 0.5963556046383214, + "grad_norm": 0.8831511735916138, + "learning_rate": 4.907602557099696e-06, + "loss": 0.7077, + "step": 720 + }, + { + "epoch": 0.5971838763114301, + "grad_norm": 0.9393531680107117, + "learning_rate": 4.9073061176095525e-06, + "loss": 0.6722, + "step": 721 + }, + { + "epoch": 0.5980121479845389, + "grad_norm": 0.8807207942008972, + "learning_rate": 4.907009212328117e-06, + "loss": 0.69, + "step": 722 + }, + { + "epoch": 0.5988404196576477, + "grad_norm": 0.8811002969741821, + "learning_rate": 4.906711841312836e-06, + "loss": 0.6902, + "step": 723 + }, + { + "epoch": 0.5996686913307565, + "grad_norm": 0.8953952789306641, + "learning_rate": 4.906414004621249e-06, + "loss": 0.6747, + "step": 724 + }, + { + "epoch": 0.6004969630038652, + "grad_norm": 0.8661350607872009, + "learning_rate": 4.9061157023109855e-06, + "loss": 0.7047, + "step": 725 + }, + { + "epoch": 0.6013252346769741, + "grad_norm": 0.8722146153450012, + "learning_rate": 4.905816934439762e-06, + "loss": 0.7231, + "step": 726 + }, + { + "epoch": 0.6021535063500828, + "grad_norm": 0.8933189511299133, + "learning_rate": 4.905517701065389e-06, + "loss": 0.7029, + "step": 727 + }, + { + "epoch": 0.6029817780231916, + "grad_norm": 0.8863442540168762, + "learning_rate": 4.905218002245764e-06, + "loss": 0.6886, + "step": 728 + }, + { + "epoch": 0.6038100496963004, + "grad_norm": 0.8672969341278076, + "learning_rate": 4.904917838038877e-06, + "loss": 0.7048, + "step": 729 + }, + { + "epoch": 0.6046383213694092, + "grad_norm": 0.8892956972122192, + "learning_rate": 4.904617208502807e-06, + "loss": 0.7009, + "step": 730 + }, + { + "epoch": 0.6054665930425179, + "grad_norm": 0.9083442687988281, + "learning_rate": 4.904316113695723e-06, + "loss": 0.7169, + "step": 731 + }, + { + "epoch": 0.6062948647156268, + "grad_norm": 0.8885907530784607, + "learning_rate": 4.904014553675883e-06, + "loss": 0.6939, + "step": 732 + }, + { + "epoch": 0.6071231363887355, + "grad_norm": 0.8859131336212158, + "learning_rate": 4.903712528501637e-06, + "loss": 0.6881, + "step": 733 + }, + { + "epoch": 0.6079514080618443, + "grad_norm": 0.874146580696106, + "learning_rate": 4.903410038231425e-06, + "loss": 0.6857, + "step": 734 + }, + { + "epoch": 0.6087796797349531, + "grad_norm": 0.8719701170921326, + "learning_rate": 4.903107082923774e-06, + "loss": 0.6834, + "step": 735 + }, + { + "epoch": 0.6096079514080619, + "grad_norm": 0.9656700491905212, + "learning_rate": 4.902803662637304e-06, + "loss": 0.6933, + "step": 736 + }, + { + "epoch": 0.6104362230811706, + "grad_norm": 0.9022018313407898, + "learning_rate": 4.9024997774307235e-06, + "loss": 0.683, + "step": 737 + }, + { + "epoch": 0.6112644947542794, + "grad_norm": 0.8694130182266235, + "learning_rate": 4.902195427362832e-06, + "loss": 0.642, + "step": 738 + }, + { + "epoch": 0.6120927664273882, + "grad_norm": 0.9153549075126648, + "learning_rate": 4.901890612492519e-06, + "loss": 0.6674, + "step": 739 + }, + { + "epoch": 0.6129210381004969, + "grad_norm": 0.9431132674217224, + "learning_rate": 4.901585332878762e-06, + "loss": 0.6918, + "step": 740 + }, + { + "epoch": 0.6137493097736058, + "grad_norm": 0.881587028503418, + "learning_rate": 4.901279588580631e-06, + "loss": 0.687, + "step": 741 + }, + { + "epoch": 0.6145775814467145, + "grad_norm": 0.9289383888244629, + "learning_rate": 4.9009733796572834e-06, + "loss": 0.6789, + "step": 742 + }, + { + "epoch": 0.6154058531198233, + "grad_norm": 0.9058337211608887, + "learning_rate": 4.900666706167968e-06, + "loss": 0.6856, + "step": 743 + }, + { + "epoch": 0.616234124792932, + "grad_norm": 0.8999397158622742, + "learning_rate": 4.900359568172024e-06, + "loss": 0.7159, + "step": 744 + }, + { + "epoch": 0.6170623964660409, + "grad_norm": 0.895321249961853, + "learning_rate": 4.90005196572888e-06, + "loss": 0.7071, + "step": 745 + }, + { + "epoch": 0.6178906681391496, + "grad_norm": 0.9142102599143982, + "learning_rate": 4.899743898898054e-06, + "loss": 0.6918, + "step": 746 + }, + { + "epoch": 0.6187189398122585, + "grad_norm": 0.8867097496986389, + "learning_rate": 4.899435367739152e-06, + "loss": 0.6991, + "step": 747 + }, + { + "epoch": 0.6195472114853672, + "grad_norm": 0.8594791293144226, + "learning_rate": 4.899126372311876e-06, + "loss": 0.7153, + "step": 748 + }, + { + "epoch": 0.620375483158476, + "grad_norm": 0.8740563988685608, + "learning_rate": 4.8988169126760085e-06, + "loss": 0.6786, + "step": 749 + }, + { + "epoch": 0.6212037548315847, + "grad_norm": 0.8818353414535522, + "learning_rate": 4.898506988891432e-06, + "loss": 0.6965, + "step": 750 + }, + { + "epoch": 0.6220320265046936, + "grad_norm": 0.9459843635559082, + "learning_rate": 4.898196601018111e-06, + "loss": 0.6821, + "step": 751 + }, + { + "epoch": 0.6228602981778023, + "grad_norm": 0.923457682132721, + "learning_rate": 4.897885749116104e-06, + "loss": 0.6822, + "step": 752 + }, + { + "epoch": 0.6236885698509111, + "grad_norm": 0.9342297315597534, + "learning_rate": 4.897574433245557e-06, + "loss": 0.7001, + "step": 753 + }, + { + "epoch": 0.6245168415240199, + "grad_norm": 0.9327735900878906, + "learning_rate": 4.897262653466706e-06, + "loss": 0.6921, + "step": 754 + }, + { + "epoch": 0.6253451131971287, + "grad_norm": 0.8876991868019104, + "learning_rate": 4.8969504098398805e-06, + "loss": 0.7087, + "step": 755 + }, + { + "epoch": 0.6261733848702374, + "grad_norm": 0.8983528017997742, + "learning_rate": 4.896637702425493e-06, + "loss": 0.6824, + "step": 756 + }, + { + "epoch": 0.6270016565433462, + "grad_norm": 0.9074965715408325, + "learning_rate": 4.896324531284053e-06, + "loss": 0.6533, + "step": 757 + }, + { + "epoch": 0.627829928216455, + "grad_norm": 0.8938930630683899, + "learning_rate": 4.896010896476153e-06, + "loss": 0.6859, + "step": 758 + }, + { + "epoch": 0.6286581998895637, + "grad_norm": 0.9036867022514343, + "learning_rate": 4.8956967980624806e-06, + "loss": 0.6904, + "step": 759 + }, + { + "epoch": 0.6294864715626726, + "grad_norm": 0.9162903428077698, + "learning_rate": 4.895382236103809e-06, + "loss": 0.6995, + "step": 760 + }, + { + "epoch": 0.6303147432357813, + "grad_norm": 1.0000487565994263, + "learning_rate": 4.895067210661005e-06, + "loss": 0.671, + "step": 761 + }, + { + "epoch": 0.6311430149088901, + "grad_norm": 0.895163893699646, + "learning_rate": 4.894751721795022e-06, + "loss": 0.6871, + "step": 762 + }, + { + "epoch": 0.6319712865819989, + "grad_norm": 0.9401512742042542, + "learning_rate": 4.894435769566905e-06, + "loss": 0.6811, + "step": 763 + }, + { + "epoch": 0.6327995582551077, + "grad_norm": 0.924578845500946, + "learning_rate": 4.894119354037787e-06, + "loss": 0.692, + "step": 764 + }, + { + "epoch": 0.6336278299282164, + "grad_norm": 0.9114750623703003, + "learning_rate": 4.893802475268892e-06, + "loss": 0.6977, + "step": 765 + }, + { + "epoch": 0.6344561016013253, + "grad_norm": 1.1472550630569458, + "learning_rate": 4.893485133321533e-06, + "loss": 0.679, + "step": 766 + }, + { + "epoch": 0.635284373274434, + "grad_norm": 0.9460884928703308, + "learning_rate": 4.893167328257111e-06, + "loss": 0.6988, + "step": 767 + }, + { + "epoch": 0.6361126449475428, + "grad_norm": 0.9040783643722534, + "learning_rate": 4.8928490601371216e-06, + "loss": 0.6706, + "step": 768 + }, + { + "epoch": 0.6369409166206516, + "grad_norm": 0.9785428047180176, + "learning_rate": 4.892530329023144e-06, + "loss": 0.7185, + "step": 769 + }, + { + "epoch": 0.6377691882937604, + "grad_norm": 0.9795847535133362, + "learning_rate": 4.892211134976851e-06, + "loss": 0.703, + "step": 770 + }, + { + "epoch": 0.6385974599668691, + "grad_norm": 0.9539455771446228, + "learning_rate": 4.8918914780600025e-06, + "loss": 0.6985, + "step": 771 + }, + { + "epoch": 0.639425731639978, + "grad_norm": 0.9304025173187256, + "learning_rate": 4.8915713583344505e-06, + "loss": 0.6731, + "step": 772 + }, + { + "epoch": 0.6402540033130867, + "grad_norm": 0.9466667771339417, + "learning_rate": 4.891250775862134e-06, + "loss": 0.6972, + "step": 773 + }, + { + "epoch": 0.6410822749861955, + "grad_norm": 0.9522298574447632, + "learning_rate": 4.890929730705084e-06, + "loss": 0.6793, + "step": 774 + }, + { + "epoch": 0.6419105466593042, + "grad_norm": 0.9077293276786804, + "learning_rate": 4.890608222925419e-06, + "loss": 0.6892, + "step": 775 + }, + { + "epoch": 0.6427388183324131, + "grad_norm": 0.8763419389724731, + "learning_rate": 4.890286252585348e-06, + "loss": 0.6985, + "step": 776 + }, + { + "epoch": 0.6435670900055218, + "grad_norm": 0.9089516401290894, + "learning_rate": 4.8899638197471685e-06, + "loss": 0.6778, + "step": 777 + }, + { + "epoch": 0.6443953616786305, + "grad_norm": 0.8733325600624084, + "learning_rate": 4.88964092447327e-06, + "loss": 0.6915, + "step": 778 + }, + { + "epoch": 0.6452236333517394, + "grad_norm": 0.9472198486328125, + "learning_rate": 4.889317566826128e-06, + "loss": 0.6984, + "step": 779 + }, + { + "epoch": 0.6460519050248481, + "grad_norm": 0.8729666471481323, + "learning_rate": 4.888993746868309e-06, + "loss": 0.6724, + "step": 780 + }, + { + "epoch": 0.6468801766979569, + "grad_norm": 0.9395233988761902, + "learning_rate": 4.88866946466247e-06, + "loss": 0.6697, + "step": 781 + }, + { + "epoch": 0.6477084483710657, + "grad_norm": 0.9158744812011719, + "learning_rate": 4.888344720271357e-06, + "loss": 0.6838, + "step": 782 + }, + { + "epoch": 0.6485367200441745, + "grad_norm": 0.8961113095283508, + "learning_rate": 4.888019513757804e-06, + "loss": 0.6488, + "step": 783 + }, + { + "epoch": 0.6493649917172832, + "grad_norm": 0.8911786079406738, + "learning_rate": 4.8876938451847364e-06, + "loss": 0.7089, + "step": 784 + }, + { + "epoch": 0.6501932633903921, + "grad_norm": 0.9214697480201721, + "learning_rate": 4.887367714615167e-06, + "loss": 0.6929, + "step": 785 + }, + { + "epoch": 0.6510215350635008, + "grad_norm": 0.9161441326141357, + "learning_rate": 4.8870411221122e-06, + "loss": 0.6761, + "step": 786 + }, + { + "epoch": 0.6518498067366096, + "grad_norm": 0.9161947965621948, + "learning_rate": 4.886714067739028e-06, + "loss": 0.6804, + "step": 787 + }, + { + "epoch": 0.6526780784097184, + "grad_norm": 0.8706585168838501, + "learning_rate": 4.886386551558933e-06, + "loss": 0.6836, + "step": 788 + }, + { + "epoch": 0.6535063500828272, + "grad_norm": 0.9037661552429199, + "learning_rate": 4.886058573635284e-06, + "loss": 0.6786, + "step": 789 + }, + { + "epoch": 0.6543346217559359, + "grad_norm": 0.8977832794189453, + "learning_rate": 4.885730134031545e-06, + "loss": 0.6938, + "step": 790 + }, + { + "epoch": 0.6551628934290448, + "grad_norm": 0.9106955528259277, + "learning_rate": 4.8854012328112645e-06, + "loss": 0.678, + "step": 791 + }, + { + "epoch": 0.6559911651021535, + "grad_norm": 0.8769396543502808, + "learning_rate": 4.885071870038082e-06, + "loss": 0.6969, + "step": 792 + }, + { + "epoch": 0.6568194367752623, + "grad_norm": 0.9946926832199097, + "learning_rate": 4.884742045775725e-06, + "loss": 0.6799, + "step": 793 + }, + { + "epoch": 0.6576477084483711, + "grad_norm": 0.9399042725563049, + "learning_rate": 4.884411760088012e-06, + "loss": 0.7038, + "step": 794 + }, + { + "epoch": 0.6584759801214799, + "grad_norm": 0.8914628624916077, + "learning_rate": 4.8840810130388526e-06, + "loss": 0.6867, + "step": 795 + }, + { + "epoch": 0.6593042517945886, + "grad_norm": 0.8900558352470398, + "learning_rate": 4.88374980469224e-06, + "loss": 0.6823, + "step": 796 + }, + { + "epoch": 0.6601325234676974, + "grad_norm": 0.8608913421630859, + "learning_rate": 4.8834181351122615e-06, + "loss": 0.6705, + "step": 797 + }, + { + "epoch": 0.6609607951408062, + "grad_norm": 0.900460422039032, + "learning_rate": 4.883086004363092e-06, + "loss": 0.7029, + "step": 798 + }, + { + "epoch": 0.6617890668139149, + "grad_norm": 0.8932703733444214, + "learning_rate": 4.882753412508996e-06, + "loss": 0.6845, + "step": 799 + }, + { + "epoch": 0.6626173384870238, + "grad_norm": 0.8737874627113342, + "learning_rate": 4.882420359614325e-06, + "loss": 0.6979, + "step": 800 + }, + { + "epoch": 0.6634456101601325, + "grad_norm": 0.8805396556854248, + "learning_rate": 4.882086845743524e-06, + "loss": 0.6935, + "step": 801 + }, + { + "epoch": 0.6642738818332413, + "grad_norm": 0.8952041268348694, + "learning_rate": 4.881752870961123e-06, + "loss": 0.7137, + "step": 802 + }, + { + "epoch": 0.66510215350635, + "grad_norm": 0.889869213104248, + "learning_rate": 4.881418435331745e-06, + "loss": 0.6643, + "step": 803 + }, + { + "epoch": 0.6659304251794589, + "grad_norm": 0.8912904858589172, + "learning_rate": 4.881083538920098e-06, + "loss": 0.6712, + "step": 804 + }, + { + "epoch": 0.6667586968525676, + "grad_norm": 0.8943006992340088, + "learning_rate": 4.880748181790984e-06, + "loss": 0.7001, + "step": 805 + }, + { + "epoch": 0.6675869685256764, + "grad_norm": 0.9042482376098633, + "learning_rate": 4.880412364009289e-06, + "loss": 0.6744, + "step": 806 + }, + { + "epoch": 0.6684152401987852, + "grad_norm": 0.9052988886833191, + "learning_rate": 4.880076085639991e-06, + "loss": 0.72, + "step": 807 + }, + { + "epoch": 0.669243511871894, + "grad_norm": 0.9702905416488647, + "learning_rate": 4.879739346748158e-06, + "loss": 0.6871, + "step": 808 + }, + { + "epoch": 0.6700717835450027, + "grad_norm": 0.8741766810417175, + "learning_rate": 4.879402147398944e-06, + "loss": 0.7174, + "step": 809 + }, + { + "epoch": 0.6709000552181116, + "grad_norm": 0.9226665496826172, + "learning_rate": 4.879064487657596e-06, + "loss": 0.6991, + "step": 810 + }, + { + "epoch": 0.6717283268912203, + "grad_norm": 0.878304123878479, + "learning_rate": 4.878726367589446e-06, + "loss": 0.6925, + "step": 811 + }, + { + "epoch": 0.6725565985643291, + "grad_norm": 0.893649160861969, + "learning_rate": 4.878387787259919e-06, + "loss": 0.6689, + "step": 812 + }, + { + "epoch": 0.6733848702374379, + "grad_norm": 0.917861819267273, + "learning_rate": 4.878048746734526e-06, + "loss": 0.7171, + "step": 813 + }, + { + "epoch": 0.6742131419105467, + "grad_norm": 0.879375696182251, + "learning_rate": 4.877709246078869e-06, + "loss": 0.6662, + "step": 814 + }, + { + "epoch": 0.6750414135836554, + "grad_norm": 0.9098542928695679, + "learning_rate": 4.877369285358637e-06, + "loss": 0.6793, + "step": 815 + }, + { + "epoch": 0.6758696852567642, + "grad_norm": 0.9465262293815613, + "learning_rate": 4.877028864639609e-06, + "loss": 0.7065, + "step": 816 + }, + { + "epoch": 0.676697956929873, + "grad_norm": 0.9029957056045532, + "learning_rate": 4.876687983987654e-06, + "loss": 0.7008, + "step": 817 + }, + { + "epoch": 0.6775262286029817, + "grad_norm": 0.9073920845985413, + "learning_rate": 4.876346643468731e-06, + "loss": 0.6836, + "step": 818 + }, + { + "epoch": 0.6783545002760906, + "grad_norm": 0.9028745889663696, + "learning_rate": 4.876004843148884e-06, + "loss": 0.6965, + "step": 819 + }, + { + "epoch": 0.6791827719491993, + "grad_norm": 0.8745572566986084, + "learning_rate": 4.875662583094249e-06, + "loss": 0.6974, + "step": 820 + }, + { + "epoch": 0.6800110436223081, + "grad_norm": 0.9203384518623352, + "learning_rate": 4.875319863371048e-06, + "loss": 0.686, + "step": 821 + }, + { + "epoch": 0.6808393152954169, + "grad_norm": 0.9087005853652954, + "learning_rate": 4.874976684045597e-06, + "loss": 0.683, + "step": 822 + }, + { + "epoch": 0.6816675869685257, + "grad_norm": 0.9007741808891296, + "learning_rate": 4.874633045184297e-06, + "loss": 0.6938, + "step": 823 + }, + { + "epoch": 0.6824958586416344, + "grad_norm": 0.9854089617729187, + "learning_rate": 4.874288946853639e-06, + "loss": 0.6945, + "step": 824 + }, + { + "epoch": 0.6833241303147433, + "grad_norm": 0.9044802784919739, + "learning_rate": 4.8739443891202025e-06, + "loss": 0.714, + "step": 825 + }, + { + "epoch": 0.684152401987852, + "grad_norm": 0.9388754963874817, + "learning_rate": 4.873599372050656e-06, + "loss": 0.6838, + "step": 826 + }, + { + "epoch": 0.6849806736609608, + "grad_norm": 0.9215866327285767, + "learning_rate": 4.873253895711757e-06, + "loss": 0.6994, + "step": 827 + }, + { + "epoch": 0.6858089453340696, + "grad_norm": 0.8786854147911072, + "learning_rate": 4.872907960170353e-06, + "loss": 0.6883, + "step": 828 + }, + { + "epoch": 0.6866372170071784, + "grad_norm": 0.9032605290412903, + "learning_rate": 4.872561565493379e-06, + "loss": 0.6847, + "step": 829 + }, + { + "epoch": 0.6874654886802871, + "grad_norm": 0.8940309882164001, + "learning_rate": 4.872214711747858e-06, + "loss": 0.6877, + "step": 830 + }, + { + "epoch": 0.688293760353396, + "grad_norm": 0.8703469634056091, + "learning_rate": 4.8718673990009045e-06, + "loss": 0.6829, + "step": 831 + }, + { + "epoch": 0.6891220320265047, + "grad_norm": 0.8618217706680298, + "learning_rate": 4.871519627319719e-06, + "loss": 0.6987, + "step": 832 + }, + { + "epoch": 0.6899503036996135, + "grad_norm": 0.8871967196464539, + "learning_rate": 4.871171396771594e-06, + "loss": 0.6824, + "step": 833 + }, + { + "epoch": 0.6907785753727222, + "grad_norm": 0.9162507057189941, + "learning_rate": 4.8708227074239064e-06, + "loss": 0.7158, + "step": 834 + }, + { + "epoch": 0.6916068470458311, + "grad_norm": 0.8829584717750549, + "learning_rate": 4.8704735593441255e-06, + "loss": 0.6714, + "step": 835 + }, + { + "epoch": 0.6924351187189398, + "grad_norm": 0.9130700826644897, + "learning_rate": 4.8701239525998076e-06, + "loss": 0.6841, + "step": 836 + }, + { + "epoch": 0.6932633903920485, + "grad_norm": 0.8706396222114563, + "learning_rate": 4.8697738872586e-06, + "loss": 0.7162, + "step": 837 + }, + { + "epoch": 0.6940916620651574, + "grad_norm": 0.8906163573265076, + "learning_rate": 4.869423363388235e-06, + "loss": 0.6586, + "step": 838 + }, + { + "epoch": 0.6949199337382661, + "grad_norm": 0.9134309887886047, + "learning_rate": 4.869072381056537e-06, + "loss": 0.6842, + "step": 839 + }, + { + "epoch": 0.6957482054113749, + "grad_norm": 0.869674563407898, + "learning_rate": 4.8687209403314175e-06, + "loss": 0.7049, + "step": 840 + }, + { + "epoch": 0.6965764770844837, + "grad_norm": 0.8405192494392395, + "learning_rate": 4.8683690412808775e-06, + "loss": 0.6677, + "step": 841 + }, + { + "epoch": 0.6974047487575925, + "grad_norm": 0.9173605442047119, + "learning_rate": 4.868016683973005e-06, + "loss": 0.6784, + "step": 842 + }, + { + "epoch": 0.6982330204307012, + "grad_norm": 0.9220597743988037, + "learning_rate": 4.867663868475978e-06, + "loss": 0.6774, + "step": 843 + }, + { + "epoch": 0.6990612921038101, + "grad_norm": 0.8687174320220947, + "learning_rate": 4.867310594858065e-06, + "loss": 0.7094, + "step": 844 + }, + { + "epoch": 0.6998895637769188, + "grad_norm": 0.9161131381988525, + "learning_rate": 4.866956863187619e-06, + "loss": 0.6867, + "step": 845 + }, + { + "epoch": 0.7007178354500276, + "grad_norm": 0.9338458776473999, + "learning_rate": 4.866602673533084e-06, + "loss": 0.6783, + "step": 846 + }, + { + "epoch": 0.7015461071231364, + "grad_norm": 0.9268187284469604, + "learning_rate": 4.866248025962993e-06, + "loss": 0.6681, + "step": 847 + }, + { + "epoch": 0.7023743787962452, + "grad_norm": 0.8843579292297363, + "learning_rate": 4.865892920545967e-06, + "loss": 0.6854, + "step": 848 + }, + { + "epoch": 0.7032026504693539, + "grad_norm": 0.889299213886261, + "learning_rate": 4.865537357350716e-06, + "loss": 0.7055, + "step": 849 + }, + { + "epoch": 0.7040309221424628, + "grad_norm": 0.9062727093696594, + "learning_rate": 4.865181336446036e-06, + "loss": 0.6806, + "step": 850 + }, + { + "epoch": 0.7048591938155715, + "grad_norm": 0.9007996320724487, + "learning_rate": 4.864824857900817e-06, + "loss": 0.6937, + "step": 851 + }, + { + "epoch": 0.7056874654886803, + "grad_norm": 0.8883542418479919, + "learning_rate": 4.864467921784032e-06, + "loss": 0.6957, + "step": 852 + }, + { + "epoch": 0.7065157371617891, + "grad_norm": 0.9422078132629395, + "learning_rate": 4.864110528164746e-06, + "loss": 0.7025, + "step": 853 + }, + { + "epoch": 0.7073440088348979, + "grad_norm": 0.8961433172225952, + "learning_rate": 4.86375267711211e-06, + "loss": 0.6795, + "step": 854 + }, + { + "epoch": 0.7081722805080066, + "grad_norm": 0.925815761089325, + "learning_rate": 4.863394368695366e-06, + "loss": 0.6824, + "step": 855 + }, + { + "epoch": 0.7090005521811154, + "grad_norm": 0.9326804280281067, + "learning_rate": 4.863035602983843e-06, + "loss": 0.6829, + "step": 856 + }, + { + "epoch": 0.7098288238542242, + "grad_norm": 0.8784539699554443, + "learning_rate": 4.862676380046959e-06, + "loss": 0.6776, + "step": 857 + }, + { + "epoch": 0.7106570955273329, + "grad_norm": 0.8577290177345276, + "learning_rate": 4.86231669995422e-06, + "loss": 0.6573, + "step": 858 + }, + { + "epoch": 0.7114853672004418, + "grad_norm": 0.9261177778244019, + "learning_rate": 4.86195656277522e-06, + "loss": 0.6891, + "step": 859 + }, + { + "epoch": 0.7123136388735505, + "grad_norm": 0.9008646607398987, + "learning_rate": 4.861595968579643e-06, + "loss": 0.6714, + "step": 860 + }, + { + "epoch": 0.7131419105466593, + "grad_norm": 0.8729333877563477, + "learning_rate": 4.861234917437261e-06, + "loss": 0.6819, + "step": 861 + }, + { + "epoch": 0.713970182219768, + "grad_norm": 0.9006103873252869, + "learning_rate": 4.860873409417934e-06, + "loss": 0.6702, + "step": 862 + }, + { + "epoch": 0.7147984538928769, + "grad_norm": 0.8699921369552612, + "learning_rate": 4.860511444591609e-06, + "loss": 0.6681, + "step": 863 + }, + { + "epoch": 0.7156267255659856, + "grad_norm": 0.9359689950942993, + "learning_rate": 4.860149023028325e-06, + "loss": 0.6834, + "step": 864 + }, + { + "epoch": 0.7164549972390944, + "grad_norm": 0.874977707862854, + "learning_rate": 4.859786144798205e-06, + "loss": 0.6692, + "step": 865 + }, + { + "epoch": 0.7172832689122032, + "grad_norm": 0.8821915984153748, + "learning_rate": 4.859422809971464e-06, + "loss": 0.6865, + "step": 866 + }, + { + "epoch": 0.718111540585312, + "grad_norm": 0.905248761177063, + "learning_rate": 4.859059018618405e-06, + "loss": 0.6705, + "step": 867 + }, + { + "epoch": 0.7189398122584207, + "grad_norm": 0.9134542346000671, + "learning_rate": 4.858694770809415e-06, + "loss": 0.7039, + "step": 868 + }, + { + "epoch": 0.7197680839315296, + "grad_norm": 0.9048055410385132, + "learning_rate": 4.858330066614975e-06, + "loss": 0.6848, + "step": 869 + }, + { + "epoch": 0.7205963556046383, + "grad_norm": 0.8780005574226379, + "learning_rate": 4.857964906105651e-06, + "loss": 0.689, + "step": 870 + }, + { + "epoch": 0.7214246272777471, + "grad_norm": 0.8826956152915955, + "learning_rate": 4.857599289352098e-06, + "loss": 0.6799, + "step": 871 + }, + { + "epoch": 0.7222528989508559, + "grad_norm": 0.919905960559845, + "learning_rate": 4.857233216425061e-06, + "loss": 0.6587, + "step": 872 + }, + { + "epoch": 0.7230811706239647, + "grad_norm": 0.8711065053939819, + "learning_rate": 4.8568666873953696e-06, + "loss": 0.6754, + "step": 873 + }, + { + "epoch": 0.7239094422970734, + "grad_norm": 0.8740426898002625, + "learning_rate": 4.856499702333944e-06, + "loss": 0.6837, + "step": 874 + }, + { + "epoch": 0.7247377139701823, + "grad_norm": 0.921416699886322, + "learning_rate": 4.8561322613117945e-06, + "loss": 0.6709, + "step": 875 + }, + { + "epoch": 0.725565985643291, + "grad_norm": 0.8988221287727356, + "learning_rate": 4.855764364400016e-06, + "loss": 0.7111, + "step": 876 + }, + { + "epoch": 0.7263942573163997, + "grad_norm": 0.8838291168212891, + "learning_rate": 4.8553960116697915e-06, + "loss": 0.6821, + "step": 877 + }, + { + "epoch": 0.7272225289895086, + "grad_norm": 0.9051623940467834, + "learning_rate": 4.8550272031923965e-06, + "loss": 0.6848, + "step": 878 + }, + { + "epoch": 0.7280508006626173, + "grad_norm": 0.9308082461357117, + "learning_rate": 4.85465793903919e-06, + "loss": 0.7087, + "step": 879 + }, + { + "epoch": 0.7288790723357261, + "grad_norm": 0.873214066028595, + "learning_rate": 4.854288219281623e-06, + "loss": 0.6877, + "step": 880 + }, + { + "epoch": 0.7297073440088349, + "grad_norm": 0.8890054225921631, + "learning_rate": 4.853918043991232e-06, + "loss": 0.6719, + "step": 881 + }, + { + "epoch": 0.7305356156819437, + "grad_norm": 0.9580398201942444, + "learning_rate": 4.853547413239642e-06, + "loss": 0.6439, + "step": 882 + }, + { + "epoch": 0.7313638873550524, + "grad_norm": 0.875035285949707, + "learning_rate": 4.853176327098567e-06, + "loss": 0.6645, + "step": 883 + }, + { + "epoch": 0.7321921590281613, + "grad_norm": 0.9299217462539673, + "learning_rate": 4.85280478563981e-06, + "loss": 0.707, + "step": 884 + }, + { + "epoch": 0.73302043070127, + "grad_norm": 0.8840609192848206, + "learning_rate": 4.8524327889352585e-06, + "loss": 0.6789, + "step": 885 + }, + { + "epoch": 0.7338487023743788, + "grad_norm": 0.8983805179595947, + "learning_rate": 4.852060337056891e-06, + "loss": 0.6814, + "step": 886 + }, + { + "epoch": 0.7346769740474876, + "grad_norm": 0.8730597496032715, + "learning_rate": 4.851687430076775e-06, + "loss": 0.7006, + "step": 887 + }, + { + "epoch": 0.7355052457205964, + "grad_norm": 0.8980515003204346, + "learning_rate": 4.851314068067063e-06, + "loss": 0.6729, + "step": 888 + }, + { + "epoch": 0.7363335173937051, + "grad_norm": 0.8919468522071838, + "learning_rate": 4.850940251099997e-06, + "loss": 0.6764, + "step": 889 + }, + { + "epoch": 0.737161789066814, + "grad_norm": 0.8987070918083191, + "learning_rate": 4.850565979247908e-06, + "loss": 0.6997, + "step": 890 + }, + { + "epoch": 0.7379900607399227, + "grad_norm": 0.9056184887886047, + "learning_rate": 4.850191252583214e-06, + "loss": 0.6887, + "step": 891 + }, + { + "epoch": 0.7388183324130315, + "grad_norm": 0.8714621067047119, + "learning_rate": 4.84981607117842e-06, + "loss": 0.6946, + "step": 892 + }, + { + "epoch": 0.7396466040861402, + "grad_norm": 0.8845617175102234, + "learning_rate": 4.849440435106122e-06, + "loss": 0.6845, + "step": 893 + }, + { + "epoch": 0.7404748757592491, + "grad_norm": 0.8926048278808594, + "learning_rate": 4.849064344439e-06, + "loss": 0.6922, + "step": 894 + }, + { + "epoch": 0.7413031474323578, + "grad_norm": 0.8731569647789001, + "learning_rate": 4.848687799249826e-06, + "loss": 0.6837, + "step": 895 + }, + { + "epoch": 0.7421314191054665, + "grad_norm": 0.8801183104515076, + "learning_rate": 4.848310799611457e-06, + "loss": 0.6932, + "step": 896 + }, + { + "epoch": 0.7429596907785754, + "grad_norm": 0.8818139433860779, + "learning_rate": 4.847933345596838e-06, + "loss": 0.6758, + "step": 897 + }, + { + "epoch": 0.7437879624516841, + "grad_norm": 0.9117923974990845, + "learning_rate": 4.847555437279004e-06, + "loss": 0.6628, + "step": 898 + }, + { + "epoch": 0.7446162341247929, + "grad_norm": 0.891360342502594, + "learning_rate": 4.847177074731078e-06, + "loss": 0.6876, + "step": 899 + }, + { + "epoch": 0.7454445057979017, + "grad_norm": 0.8905263543128967, + "learning_rate": 4.846798258026267e-06, + "loss": 0.6862, + "step": 900 + }, + { + "epoch": 0.7462727774710105, + "grad_norm": 0.8853707909584045, + "learning_rate": 4.846418987237868e-06, + "loss": 0.654, + "step": 901 + }, + { + "epoch": 0.7471010491441192, + "grad_norm": 0.8782089948654175, + "learning_rate": 4.846039262439271e-06, + "loss": 0.7005, + "step": 902 + }, + { + "epoch": 0.7479293208172281, + "grad_norm": 0.8774383068084717, + "learning_rate": 4.8456590837039445e-06, + "loss": 0.7044, + "step": 903 + }, + { + "epoch": 0.7487575924903368, + "grad_norm": 0.9385423064231873, + "learning_rate": 4.845278451105451e-06, + "loss": 0.676, + "step": 904 + }, + { + "epoch": 0.7495858641634456, + "grad_norm": 0.9211432337760925, + "learning_rate": 4.8448973647174405e-06, + "loss": 0.697, + "step": 905 + }, + { + "epoch": 0.7504141358365544, + "grad_norm": 0.887610912322998, + "learning_rate": 4.844515824613649e-06, + "loss": 0.6965, + "step": 906 + }, + { + "epoch": 0.7512424075096632, + "grad_norm": 0.885761559009552, + "learning_rate": 4.8441338308679e-06, + "loss": 0.6851, + "step": 907 + }, + { + "epoch": 0.7520706791827719, + "grad_norm": 0.9209930300712585, + "learning_rate": 4.843751383554106e-06, + "loss": 0.6718, + "step": 908 + }, + { + "epoch": 0.7528989508558808, + "grad_norm": 0.9308252334594727, + "learning_rate": 4.843368482746269e-06, + "loss": 0.6802, + "step": 909 + }, + { + "epoch": 0.7537272225289895, + "grad_norm": 0.8931897282600403, + "learning_rate": 4.842985128518473e-06, + "loss": 0.7076, + "step": 910 + }, + { + "epoch": 0.7545554942020983, + "grad_norm": 0.9102733731269836, + "learning_rate": 4.842601320944898e-06, + "loss": 0.6898, + "step": 911 + }, + { + "epoch": 0.7553837658752071, + "grad_norm": 0.9336893558502197, + "learning_rate": 4.842217060099804e-06, + "loss": 0.6709, + "step": 912 + }, + { + "epoch": 0.7562120375483159, + "grad_norm": 0.9366991519927979, + "learning_rate": 4.8418323460575425e-06, + "loss": 0.6665, + "step": 913 + }, + { + "epoch": 0.7570403092214246, + "grad_norm": 0.9371415376663208, + "learning_rate": 4.841447178892552e-06, + "loss": 0.7013, + "step": 914 + }, + { + "epoch": 0.7578685808945335, + "grad_norm": 1.031123399734497, + "learning_rate": 4.84106155867936e-06, + "loss": 0.687, + "step": 915 + }, + { + "epoch": 0.7586968525676422, + "grad_norm": 0.9319218397140503, + "learning_rate": 4.8406754854925785e-06, + "loss": 0.669, + "step": 916 + }, + { + "epoch": 0.7595251242407509, + "grad_norm": 0.8563560843467712, + "learning_rate": 4.840288959406911e-06, + "loss": 0.6905, + "step": 917 + }, + { + "epoch": 0.7603533959138598, + "grad_norm": 0.9424378275871277, + "learning_rate": 4.839901980497145e-06, + "loss": 0.6685, + "step": 918 + }, + { + "epoch": 0.7611816675869685, + "grad_norm": 0.9202269315719604, + "learning_rate": 4.839514548838158e-06, + "loss": 0.6728, + "step": 919 + }, + { + "epoch": 0.7620099392600773, + "grad_norm": 0.9152582287788391, + "learning_rate": 4.839126664504915e-06, + "loss": 0.6842, + "step": 920 + }, + { + "epoch": 0.762838210933186, + "grad_norm": 0.9457129836082458, + "learning_rate": 4.838738327572468e-06, + "loss": 0.6822, + "step": 921 + }, + { + "epoch": 0.7636664826062949, + "grad_norm": 0.93124920129776, + "learning_rate": 4.838349538115955e-06, + "loss": 0.691, + "step": 922 + }, + { + "epoch": 0.7644947542794036, + "grad_norm": 0.9249301552772522, + "learning_rate": 4.837960296210605e-06, + "loss": 0.6574, + "step": 923 + }, + { + "epoch": 0.7653230259525124, + "grad_norm": 1.0551042556762695, + "learning_rate": 4.837570601931731e-06, + "loss": 0.6782, + "step": 924 + }, + { + "epoch": 0.7661512976256212, + "grad_norm": 0.9404964447021484, + "learning_rate": 4.8371804553547365e-06, + "loss": 0.6703, + "step": 925 + }, + { + "epoch": 0.76697956929873, + "grad_norm": 0.909015417098999, + "learning_rate": 4.836789856555111e-06, + "loss": 0.6812, + "step": 926 + }, + { + "epoch": 0.7678078409718387, + "grad_norm": 0.8572261929512024, + "learning_rate": 4.836398805608431e-06, + "loss": 0.6831, + "step": 927 + }, + { + "epoch": 0.7686361126449476, + "grad_norm": 0.951753556728363, + "learning_rate": 4.836007302590362e-06, + "loss": 0.6672, + "step": 928 + }, + { + "epoch": 0.7694643843180563, + "grad_norm": 0.8875932693481445, + "learning_rate": 4.8356153475766564e-06, + "loss": 0.6819, + "step": 929 + }, + { + "epoch": 0.7702926559911651, + "grad_norm": 0.9405681490898132, + "learning_rate": 4.835222940643153e-06, + "loss": 0.6917, + "step": 930 + }, + { + "epoch": 0.7711209276642739, + "grad_norm": 0.942520022392273, + "learning_rate": 4.834830081865779e-06, + "loss": 0.6657, + "step": 931 + }, + { + "epoch": 0.7719491993373827, + "grad_norm": 0.9063259363174438, + "learning_rate": 4.83443677132055e-06, + "loss": 0.6857, + "step": 932 + }, + { + "epoch": 0.7727774710104914, + "grad_norm": 0.8778199553489685, + "learning_rate": 4.834043009083566e-06, + "loss": 0.7028, + "step": 933 + }, + { + "epoch": 0.7736057426836003, + "grad_norm": 0.9215186834335327, + "learning_rate": 4.833648795231018e-06, + "loss": 0.6712, + "step": 934 + }, + { + "epoch": 0.774434014356709, + "grad_norm": 0.8992031812667847, + "learning_rate": 4.8332541298391825e-06, + "loss": 0.6827, + "step": 935 + }, + { + "epoch": 0.7752622860298177, + "grad_norm": 0.9203833341598511, + "learning_rate": 4.8328590129844224e-06, + "loss": 0.6835, + "step": 936 + }, + { + "epoch": 0.7760905577029266, + "grad_norm": 0.9109310507774353, + "learning_rate": 4.832463444743191e-06, + "loss": 0.673, + "step": 937 + }, + { + "epoch": 0.7769188293760353, + "grad_norm": 0.978349506855011, + "learning_rate": 4.832067425192024e-06, + "loss": 0.6638, + "step": 938 + }, + { + "epoch": 0.7777471010491441, + "grad_norm": 0.9552261829376221, + "learning_rate": 4.831670954407551e-06, + "loss": 0.6919, + "step": 939 + }, + { + "epoch": 0.7785753727222529, + "grad_norm": 0.9106539487838745, + "learning_rate": 4.831274032466484e-06, + "loss": 0.6964, + "step": 940 + }, + { + "epoch": 0.7794036443953617, + "grad_norm": 0.9195299744606018, + "learning_rate": 4.830876659445623e-06, + "loss": 0.6793, + "step": 941 + }, + { + "epoch": 0.7802319160684704, + "grad_norm": 0.9075512290000916, + "learning_rate": 4.8304788354218564e-06, + "loss": 0.6649, + "step": 942 + }, + { + "epoch": 0.7810601877415793, + "grad_norm": 0.9142526984214783, + "learning_rate": 4.83008056047216e-06, + "loss": 0.6811, + "step": 943 + }, + { + "epoch": 0.781888459414688, + "grad_norm": 0.8898900151252747, + "learning_rate": 4.829681834673597e-06, + "loss": 0.6906, + "step": 944 + }, + { + "epoch": 0.7827167310877968, + "grad_norm": 0.8884983062744141, + "learning_rate": 4.8292826581033135e-06, + "loss": 0.6903, + "step": 945 + }, + { + "epoch": 0.7835450027609056, + "grad_norm": 0.9075689315795898, + "learning_rate": 4.828883030838551e-06, + "loss": 0.6794, + "step": 946 + }, + { + "epoch": 0.7843732744340144, + "grad_norm": 0.8970300555229187, + "learning_rate": 4.828482952956631e-06, + "loss": 0.6715, + "step": 947 + }, + { + "epoch": 0.7852015461071231, + "grad_norm": 0.8906447887420654, + "learning_rate": 4.828082424534965e-06, + "loss": 0.6878, + "step": 948 + }, + { + "epoch": 0.786029817780232, + "grad_norm": 0.9083748459815979, + "learning_rate": 4.827681445651052e-06, + "loss": 0.6787, + "step": 949 + }, + { + "epoch": 0.7868580894533407, + "grad_norm": 0.8975276947021484, + "learning_rate": 4.8272800163824784e-06, + "loss": 0.687, + "step": 950 + }, + { + "epoch": 0.7876863611264495, + "grad_norm": 0.9573839902877808, + "learning_rate": 4.8268781368069156e-06, + "loss": 0.6996, + "step": 951 + }, + { + "epoch": 0.7885146327995582, + "grad_norm": 0.913536787033081, + "learning_rate": 4.826475807002125e-06, + "loss": 0.6844, + "step": 952 + }, + { + "epoch": 0.7893429044726671, + "grad_norm": 0.9202503561973572, + "learning_rate": 4.826073027045952e-06, + "loss": 0.6702, + "step": 953 + }, + { + "epoch": 0.7901711761457758, + "grad_norm": 0.8800237774848938, + "learning_rate": 4.8256697970163315e-06, + "loss": 0.685, + "step": 954 + }, + { + "epoch": 0.7909994478188846, + "grad_norm": 0.8923048377037048, + "learning_rate": 4.825266116991285e-06, + "loss": 0.6839, + "step": 955 + }, + { + "epoch": 0.7918277194919934, + "grad_norm": 0.879743754863739, + "learning_rate": 4.824861987048922e-06, + "loss": 0.7112, + "step": 956 + }, + { + "epoch": 0.7926559911651021, + "grad_norm": 0.8575169444084167, + "learning_rate": 4.824457407267436e-06, + "loss": 0.671, + "step": 957 + }, + { + "epoch": 0.7934842628382109, + "grad_norm": 0.8851709961891174, + "learning_rate": 4.824052377725109e-06, + "loss": 0.669, + "step": 958 + }, + { + "epoch": 0.7943125345113197, + "grad_norm": 0.9084329605102539, + "learning_rate": 4.823646898500311e-06, + "loss": 0.6731, + "step": 959 + }, + { + "epoch": 0.7951408061844285, + "grad_norm": 0.908256471157074, + "learning_rate": 4.823240969671501e-06, + "loss": 0.688, + "step": 960 + }, + { + "epoch": 0.7959690778575372, + "grad_norm": 0.8993384838104248, + "learning_rate": 4.822834591317219e-06, + "loss": 0.6736, + "step": 961 + }, + { + "epoch": 0.7967973495306461, + "grad_norm": 0.893530011177063, + "learning_rate": 4.822427763516098e-06, + "loss": 0.6787, + "step": 962 + }, + { + "epoch": 0.7976256212037548, + "grad_norm": 0.9479735493659973, + "learning_rate": 4.822020486346852e-06, + "loss": 0.6842, + "step": 963 + }, + { + "epoch": 0.7984538928768636, + "grad_norm": 0.8716689348220825, + "learning_rate": 4.821612759888289e-06, + "loss": 0.6726, + "step": 964 + }, + { + "epoch": 0.7992821645499724, + "grad_norm": 0.8902738690376282, + "learning_rate": 4.821204584219299e-06, + "loss": 0.6672, + "step": 965 + }, + { + "epoch": 0.8001104362230812, + "grad_norm": 0.8713969588279724, + "learning_rate": 4.820795959418859e-06, + "loss": 0.6783, + "step": 966 + }, + { + "epoch": 0.8009387078961899, + "grad_norm": 0.891788899898529, + "learning_rate": 4.820386885566036e-06, + "loss": 0.6691, + "step": 967 + }, + { + "epoch": 0.8017669795692988, + "grad_norm": 0.8484700918197632, + "learning_rate": 4.81997736273998e-06, + "loss": 0.6766, + "step": 968 + }, + { + "epoch": 0.8025952512424075, + "grad_norm": 0.8931223154067993, + "learning_rate": 4.819567391019931e-06, + "loss": 0.6944, + "step": 969 + }, + { + "epoch": 0.8034235229155163, + "grad_norm": 0.9707916975021362, + "learning_rate": 4.819156970485216e-06, + "loss": 0.7076, + "step": 970 + }, + { + "epoch": 0.8042517945886251, + "grad_norm": 0.9172869920730591, + "learning_rate": 4.818746101215245e-06, + "loss": 0.6864, + "step": 971 + }, + { + "epoch": 0.8050800662617339, + "grad_norm": 0.8982006311416626, + "learning_rate": 4.818334783289518e-06, + "loss": 0.6685, + "step": 972 + }, + { + "epoch": 0.8059083379348426, + "grad_norm": 0.8960995078086853, + "learning_rate": 4.817923016787623e-06, + "loss": 0.6959, + "step": 973 + }, + { + "epoch": 0.8067366096079515, + "grad_norm": 0.8806949257850647, + "learning_rate": 4.81751080178923e-06, + "loss": 0.7028, + "step": 974 + }, + { + "epoch": 0.8075648812810602, + "grad_norm": 0.8798978924751282, + "learning_rate": 4.817098138374102e-06, + "loss": 0.6718, + "step": 975 + }, + { + "epoch": 0.8083931529541689, + "grad_norm": 0.9217897653579712, + "learning_rate": 4.8166850266220835e-06, + "loss": 0.6782, + "step": 976 + }, + { + "epoch": 0.8092214246272778, + "grad_norm": 0.8915667533874512, + "learning_rate": 4.8162714666131074e-06, + "loss": 0.6917, + "step": 977 + }, + { + "epoch": 0.8100496963003865, + "grad_norm": 0.8895093202590942, + "learning_rate": 4.815857458427195e-06, + "loss": 0.698, + "step": 978 + }, + { + "epoch": 0.8108779679734953, + "grad_norm": 0.9167448282241821, + "learning_rate": 4.815443002144453e-06, + "loss": 0.6796, + "step": 979 + }, + { + "epoch": 0.811706239646604, + "grad_norm": 0.913827657699585, + "learning_rate": 4.815028097845074e-06, + "loss": 0.6731, + "step": 980 + }, + { + "epoch": 0.8125345113197129, + "grad_norm": 0.9183399081230164, + "learning_rate": 4.814612745609338e-06, + "loss": 0.7076, + "step": 981 + }, + { + "epoch": 0.8133627829928216, + "grad_norm": 0.8812996745109558, + "learning_rate": 4.814196945517613e-06, + "loss": 0.6416, + "step": 982 + }, + { + "epoch": 0.8141910546659304, + "grad_norm": 0.9032955169677734, + "learning_rate": 4.813780697650351e-06, + "loss": 0.6799, + "step": 983 + }, + { + "epoch": 0.8150193263390392, + "grad_norm": 0.9380100965499878, + "learning_rate": 4.813364002088093e-06, + "loss": 0.6771, + "step": 984 + }, + { + "epoch": 0.815847598012148, + "grad_norm": 0.9470316171646118, + "learning_rate": 4.8129468589114655e-06, + "loss": 0.6851, + "step": 985 + }, + { + "epoch": 0.8166758696852567, + "grad_norm": 0.9401133060455322, + "learning_rate": 4.812529268201183e-06, + "loss": 0.676, + "step": 986 + }, + { + "epoch": 0.8175041413583656, + "grad_norm": 0.9379092454910278, + "learning_rate": 4.812111230038043e-06, + "loss": 0.672, + "step": 987 + }, + { + "epoch": 0.8183324130314743, + "grad_norm": 0.9591637253761292, + "learning_rate": 4.811692744502933e-06, + "loss": 0.6684, + "step": 988 + }, + { + "epoch": 0.8191606847045831, + "grad_norm": 0.8911624550819397, + "learning_rate": 4.8112738116768276e-06, + "loss": 0.6448, + "step": 989 + }, + { + "epoch": 0.8199889563776919, + "grad_norm": 0.9036282896995544, + "learning_rate": 4.810854431640784e-06, + "loss": 0.703, + "step": 990 + }, + { + "epoch": 0.8208172280508007, + "grad_norm": 0.9197635650634766, + "learning_rate": 4.81043460447595e-06, + "loss": 0.6655, + "step": 991 + }, + { + "epoch": 0.8216454997239094, + "grad_norm": 0.9548643827438354, + "learning_rate": 4.8100143302635575e-06, + "loss": 0.6892, + "step": 992 + }, + { + "epoch": 0.8224737713970183, + "grad_norm": 0.9139766097068787, + "learning_rate": 4.809593609084925e-06, + "loss": 0.6723, + "step": 993 + }, + { + "epoch": 0.823302043070127, + "grad_norm": 0.9067809581756592, + "learning_rate": 4.8091724410214595e-06, + "loss": 0.6649, + "step": 994 + }, + { + "epoch": 0.8241303147432358, + "grad_norm": 0.9705583453178406, + "learning_rate": 4.808750826154652e-06, + "loss": 0.6889, + "step": 995 + }, + { + "epoch": 0.8249585864163446, + "grad_norm": 0.9664806127548218, + "learning_rate": 4.808328764566082e-06, + "loss": 0.6758, + "step": 996 + }, + { + "epoch": 0.8257868580894533, + "grad_norm": 0.8840530514717102, + "learning_rate": 4.807906256337414e-06, + "loss": 0.6719, + "step": 997 + }, + { + "epoch": 0.8266151297625621, + "grad_norm": 0.8803462386131287, + "learning_rate": 4.8074833015504e-06, + "loss": 0.6903, + "step": 998 + }, + { + "epoch": 0.8274434014356709, + "grad_norm": 0.9404571652412415, + "learning_rate": 4.8070599002868756e-06, + "loss": 0.6917, + "step": 999 + }, + { + "epoch": 0.8282716731087797, + "grad_norm": 1.0206997394561768, + "learning_rate": 4.806636052628768e-06, + "loss": 0.6854, + "step": 1000 + }, + { + "epoch": 0.8290999447818884, + "grad_norm": 0.9294790029525757, + "learning_rate": 4.806211758658086e-06, + "loss": 0.6791, + "step": 1001 + }, + { + "epoch": 0.8299282164549973, + "grad_norm": 0.874011754989624, + "learning_rate": 4.805787018456927e-06, + "loss": 0.6915, + "step": 1002 + }, + { + "epoch": 0.830756488128106, + "grad_norm": 0.9098061323165894, + "learning_rate": 4.805361832107474e-06, + "loss": 0.6643, + "step": 1003 + }, + { + "epoch": 0.8315847598012148, + "grad_norm": 0.9783986210823059, + "learning_rate": 4.804936199691997e-06, + "loss": 0.6853, + "step": 1004 + }, + { + "epoch": 0.8324130314743236, + "grad_norm": 0.9316959381103516, + "learning_rate": 4.804510121292852e-06, + "loss": 0.6696, + "step": 1005 + }, + { + "epoch": 0.8332413031474324, + "grad_norm": 0.8930666446685791, + "learning_rate": 4.80408359699248e-06, + "loss": 0.6805, + "step": 1006 + }, + { + "epoch": 0.8340695748205411, + "grad_norm": 0.9417555332183838, + "learning_rate": 4.803656626873412e-06, + "loss": 0.66, + "step": 1007 + }, + { + "epoch": 0.83489784649365, + "grad_norm": 0.9435075521469116, + "learning_rate": 4.80322921101826e-06, + "loss": 0.6664, + "step": 1008 + }, + { + "epoch": 0.8357261181667587, + "grad_norm": 0.9090076684951782, + "learning_rate": 4.8028013495097255e-06, + "loss": 0.6595, + "step": 1009 + }, + { + "epoch": 0.8365543898398675, + "grad_norm": 0.9024788737297058, + "learning_rate": 4.802373042430597e-06, + "loss": 0.6672, + "step": 1010 + }, + { + "epoch": 0.8373826615129762, + "grad_norm": 0.899813711643219, + "learning_rate": 4.801944289863747e-06, + "loss": 0.6743, + "step": 1011 + }, + { + "epoch": 0.8382109331860851, + "grad_norm": 0.887106716632843, + "learning_rate": 4.801515091892136e-06, + "loss": 0.6823, + "step": 1012 + }, + { + "epoch": 0.8390392048591938, + "grad_norm": 0.8690978288650513, + "learning_rate": 4.801085448598809e-06, + "loss": 0.6642, + "step": 1013 + }, + { + "epoch": 0.8398674765323026, + "grad_norm": 0.9131479859352112, + "learning_rate": 4.800655360066898e-06, + "loss": 0.6841, + "step": 1014 + }, + { + "epoch": 0.8406957482054114, + "grad_norm": 0.9031343460083008, + "learning_rate": 4.800224826379622e-06, + "loss": 0.6879, + "step": 1015 + }, + { + "epoch": 0.8415240198785201, + "grad_norm": 0.8914244174957275, + "learning_rate": 4.799793847620283e-06, + "loss": 0.6562, + "step": 1016 + }, + { + "epoch": 0.8423522915516289, + "grad_norm": 0.8692408800125122, + "learning_rate": 4.799362423872274e-06, + "loss": 0.6759, + "step": 1017 + }, + { + "epoch": 0.8431805632247377, + "grad_norm": 0.9511806964874268, + "learning_rate": 4.798930555219069e-06, + "loss": 0.6637, + "step": 1018 + }, + { + "epoch": 0.8440088348978465, + "grad_norm": 0.9556595683097839, + "learning_rate": 4.7984982417442325e-06, + "loss": 0.6925, + "step": 1019 + }, + { + "epoch": 0.8448371065709552, + "grad_norm": 0.9107236862182617, + "learning_rate": 4.798065483531413e-06, + "loss": 0.67, + "step": 1020 + }, + { + "epoch": 0.8456653782440641, + "grad_norm": 0.875293493270874, + "learning_rate": 4.797632280664345e-06, + "loss": 0.6755, + "step": 1021 + }, + { + "epoch": 0.8464936499171728, + "grad_norm": 0.8972271084785461, + "learning_rate": 4.797198633226849e-06, + "loss": 0.6637, + "step": 1022 + }, + { + "epoch": 0.8473219215902816, + "grad_norm": 0.8903595805168152, + "learning_rate": 4.796764541302831e-06, + "loss": 0.6997, + "step": 1023 + }, + { + "epoch": 0.8481501932633904, + "grad_norm": 0.8948636651039124, + "learning_rate": 4.796330004976285e-06, + "loss": 0.6579, + "step": 1024 + }, + { + "epoch": 0.8489784649364992, + "grad_norm": 0.8590925335884094, + "learning_rate": 4.795895024331289e-06, + "loss": 0.6651, + "step": 1025 + }, + { + "epoch": 0.8498067366096079, + "grad_norm": 0.870811402797699, + "learning_rate": 4.795459599452007e-06, + "loss": 0.6707, + "step": 1026 + }, + { + "epoch": 0.8506350082827168, + "grad_norm": 0.8807471394538879, + "learning_rate": 4.795023730422692e-06, + "loss": 0.6787, + "step": 1027 + }, + { + "epoch": 0.8514632799558255, + "grad_norm": 0.8817671537399292, + "learning_rate": 4.794587417327678e-06, + "loss": 0.66, + "step": 1028 + }, + { + "epoch": 0.8522915516289343, + "grad_norm": 0.8924570679664612, + "learning_rate": 4.794150660251389e-06, + "loss": 0.6842, + "step": 1029 + }, + { + "epoch": 0.8531198233020431, + "grad_norm": 0.8746559619903564, + "learning_rate": 4.793713459278333e-06, + "loss": 0.6582, + "step": 1030 + }, + { + "epoch": 0.8539480949751519, + "grad_norm": 0.8885277509689331, + "learning_rate": 4.793275814493104e-06, + "loss": 0.6784, + "step": 1031 + }, + { + "epoch": 0.8547763666482606, + "grad_norm": 0.8797710537910461, + "learning_rate": 4.792837725980383e-06, + "loss": 0.6638, + "step": 1032 + }, + { + "epoch": 0.8556046383213695, + "grad_norm": 0.8932235240936279, + "learning_rate": 4.7923991938249355e-06, + "loss": 0.6918, + "step": 1033 + }, + { + "epoch": 0.8564329099944782, + "grad_norm": 0.9317487478256226, + "learning_rate": 4.791960218111613e-06, + "loss": 0.6905, + "step": 1034 + }, + { + "epoch": 0.8572611816675869, + "grad_norm": 0.8760019540786743, + "learning_rate": 4.791520798925354e-06, + "loss": 0.673, + "step": 1035 + }, + { + "epoch": 0.8580894533406958, + "grad_norm": 0.8644204139709473, + "learning_rate": 4.7910809363511825e-06, + "loss": 0.6755, + "step": 1036 + }, + { + "epoch": 0.8589177250138045, + "grad_norm": 0.8761067390441895, + "learning_rate": 4.790640630474206e-06, + "loss": 0.699, + "step": 1037 + }, + { + "epoch": 0.8597459966869133, + "grad_norm": 0.9135014414787292, + "learning_rate": 4.790199881379621e-06, + "loss": 0.6659, + "step": 1038 + }, + { + "epoch": 0.860574268360022, + "grad_norm": 0.9065434336662292, + "learning_rate": 4.789758689152708e-06, + "loss": 0.6664, + "step": 1039 + }, + { + "epoch": 0.8614025400331309, + "grad_norm": 0.8814099431037903, + "learning_rate": 4.789317053878833e-06, + "loss": 0.681, + "step": 1040 + }, + { + "epoch": 0.8622308117062396, + "grad_norm": 0.872502863407135, + "learning_rate": 4.788874975643449e-06, + "loss": 0.7044, + "step": 1041 + }, + { + "epoch": 0.8630590833793484, + "grad_norm": 0.9532367587089539, + "learning_rate": 4.788432454532093e-06, + "loss": 0.655, + "step": 1042 + }, + { + "epoch": 0.8638873550524572, + "grad_norm": 0.9009373784065247, + "learning_rate": 4.787989490630391e-06, + "loss": 0.6946, + "step": 1043 + }, + { + "epoch": 0.864715626725566, + "grad_norm": 0.889933168888092, + "learning_rate": 4.787546084024051e-06, + "loss": 0.6595, + "step": 1044 + }, + { + "epoch": 0.8655438983986747, + "grad_norm": 0.9179873466491699, + "learning_rate": 4.787102234798867e-06, + "loss": 0.658, + "step": 1045 + }, + { + "epoch": 0.8663721700717836, + "grad_norm": 0.8873292207717896, + "learning_rate": 4.786657943040721e-06, + "loss": 0.7023, + "step": 1046 + }, + { + "epoch": 0.8672004417448923, + "grad_norm": 0.8388610482215881, + "learning_rate": 4.78621320883558e-06, + "loss": 0.6644, + "step": 1047 + }, + { + "epoch": 0.8680287134180011, + "grad_norm": 0.8971349596977234, + "learning_rate": 4.785768032269493e-06, + "loss": 0.6552, + "step": 1048 + }, + { + "epoch": 0.8688569850911099, + "grad_norm": 0.8789131045341492, + "learning_rate": 4.785322413428601e-06, + "loss": 0.6681, + "step": 1049 + }, + { + "epoch": 0.8696852567642187, + "grad_norm": 0.8867958784103394, + "learning_rate": 4.784876352399125e-06, + "loss": 0.6859, + "step": 1050 + }, + { + "epoch": 0.8705135284373274, + "grad_norm": 0.8584237694740295, + "learning_rate": 4.784429849267373e-06, + "loss": 0.6655, + "step": 1051 + }, + { + "epoch": 0.8713418001104363, + "grad_norm": 0.9230811595916748, + "learning_rate": 4.783982904119743e-06, + "loss": 0.6648, + "step": 1052 + }, + { + "epoch": 0.872170071783545, + "grad_norm": 0.9217789769172668, + "learning_rate": 4.783535517042709e-06, + "loss": 0.6867, + "step": 1053 + }, + { + "epoch": 0.8729983434566538, + "grad_norm": 0.9209116697311401, + "learning_rate": 4.783087688122842e-06, + "loss": 0.6972, + "step": 1054 + }, + { + "epoch": 0.8738266151297626, + "grad_norm": 0.9493632316589355, + "learning_rate": 4.782639417446789e-06, + "loss": 0.6785, + "step": 1055 + }, + { + "epoch": 0.8746548868028713, + "grad_norm": 0.8790786862373352, + "learning_rate": 4.782190705101286e-06, + "loss": 0.6925, + "step": 1056 + }, + { + "epoch": 0.8754831584759801, + "grad_norm": 0.8662691712379456, + "learning_rate": 4.781741551173157e-06, + "loss": 0.6686, + "step": 1057 + }, + { + "epoch": 0.8763114301490889, + "grad_norm": 0.919658899307251, + "learning_rate": 4.781291955749307e-06, + "loss": 0.6814, + "step": 1058 + }, + { + "epoch": 0.8771397018221977, + "grad_norm": 0.9639708399772644, + "learning_rate": 4.780841918916731e-06, + "loss": 0.6629, + "step": 1059 + }, + { + "epoch": 0.8779679734953064, + "grad_norm": 0.8694306015968323, + "learning_rate": 4.780391440762505e-06, + "loss": 0.6893, + "step": 1060 + }, + { + "epoch": 0.8787962451684153, + "grad_norm": 0.9725603461265564, + "learning_rate": 4.779940521373793e-06, + "loss": 0.67, + "step": 1061 + }, + { + "epoch": 0.879624516841524, + "grad_norm": 0.93159419298172, + "learning_rate": 4.779489160837842e-06, + "loss": 0.686, + "step": 1062 + }, + { + "epoch": 0.8804527885146328, + "grad_norm": 0.9366991519927979, + "learning_rate": 4.779037359241989e-06, + "loss": 0.649, + "step": 1063 + }, + { + "epoch": 0.8812810601877415, + "grad_norm": 0.9052289724349976, + "learning_rate": 4.7785851166736505e-06, + "loss": 0.6677, + "step": 1064 + }, + { + "epoch": 0.8821093318608504, + "grad_norm": 0.882251501083374, + "learning_rate": 4.778132433220334e-06, + "loss": 0.6886, + "step": 1065 + }, + { + "epoch": 0.8829376035339591, + "grad_norm": 0.8867698907852173, + "learning_rate": 4.777679308969627e-06, + "loss": 0.6645, + "step": 1066 + }, + { + "epoch": 0.883765875207068, + "grad_norm": 0.890396773815155, + "learning_rate": 4.777225744009207e-06, + "loss": 0.6684, + "step": 1067 + }, + { + "epoch": 0.8845941468801767, + "grad_norm": 0.8827376365661621, + "learning_rate": 4.776771738426833e-06, + "loss": 0.6738, + "step": 1068 + }, + { + "epoch": 0.8854224185532855, + "grad_norm": 0.8953343033790588, + "learning_rate": 4.776317292310352e-06, + "loss": 0.684, + "step": 1069 + }, + { + "epoch": 0.8862506902263942, + "grad_norm": 0.8751050233840942, + "learning_rate": 4.775862405747694e-06, + "loss": 0.6779, + "step": 1070 + }, + { + "epoch": 0.8870789618995031, + "grad_norm": 0.9136618971824646, + "learning_rate": 4.775407078826876e-06, + "loss": 0.6939, + "step": 1071 + }, + { + "epoch": 0.8879072335726118, + "grad_norm": 0.8887941837310791, + "learning_rate": 4.774951311635999e-06, + "loss": 0.6547, + "step": 1072 + }, + { + "epoch": 0.8887355052457206, + "grad_norm": 0.8873274326324463, + "learning_rate": 4.77449510426325e-06, + "loss": 0.6769, + "step": 1073 + }, + { + "epoch": 0.8895637769188294, + "grad_norm": 0.9307074546813965, + "learning_rate": 4.774038456796901e-06, + "loss": 0.6984, + "step": 1074 + }, + { + "epoch": 0.8903920485919381, + "grad_norm": 0.9022747874259949, + "learning_rate": 4.77358136932531e-06, + "loss": 0.6733, + "step": 1075 + }, + { + "epoch": 0.8912203202650469, + "grad_norm": 0.8779094219207764, + "learning_rate": 4.773123841936916e-06, + "loss": 0.6434, + "step": 1076 + }, + { + "epoch": 0.8920485919381557, + "grad_norm": 0.9465747475624084, + "learning_rate": 4.7726658747202505e-06, + "loss": 0.6761, + "step": 1077 + }, + { + "epoch": 0.8928768636112645, + "grad_norm": 0.8694362044334412, + "learning_rate": 4.772207467763923e-06, + "loss": 0.6756, + "step": 1078 + }, + { + "epoch": 0.8937051352843732, + "grad_norm": 0.8839150071144104, + "learning_rate": 4.771748621156631e-06, + "loss": 0.6714, + "step": 1079 + }, + { + "epoch": 0.8945334069574821, + "grad_norm": 0.9220975041389465, + "learning_rate": 4.7712893349871585e-06, + "loss": 0.6724, + "step": 1080 + }, + { + "epoch": 0.8953616786305908, + "grad_norm": 0.9234760999679565, + "learning_rate": 4.770829609344372e-06, + "loss": 0.67, + "step": 1081 + }, + { + "epoch": 0.8961899503036996, + "grad_norm": 0.9262274503707886, + "learning_rate": 4.770369444317225e-06, + "loss": 0.6619, + "step": 1082 + }, + { + "epoch": 0.8970182219768084, + "grad_norm": 0.8889190554618835, + "learning_rate": 4.769908839994753e-06, + "loss": 0.688, + "step": 1083 + }, + { + "epoch": 0.8978464936499172, + "grad_norm": 0.9028982520103455, + "learning_rate": 4.769447796466082e-06, + "loss": 0.6633, + "step": 1084 + }, + { + "epoch": 0.8986747653230259, + "grad_norm": 0.9155063033103943, + "learning_rate": 4.7689863138204185e-06, + "loss": 0.6709, + "step": 1085 + }, + { + "epoch": 0.8995030369961348, + "grad_norm": 0.9066855311393738, + "learning_rate": 4.768524392147052e-06, + "loss": 0.6713, + "step": 1086 + }, + { + "epoch": 0.9003313086692435, + "grad_norm": 0.9365159273147583, + "learning_rate": 4.768062031535365e-06, + "loss": 0.6675, + "step": 1087 + }, + { + "epoch": 0.9011595803423523, + "grad_norm": 0.9106162190437317, + "learning_rate": 4.767599232074816e-06, + "loss": 0.6804, + "step": 1088 + }, + { + "epoch": 0.9019878520154611, + "grad_norm": 0.9183253645896912, + "learning_rate": 4.767135993854955e-06, + "loss": 0.658, + "step": 1089 + }, + { + "epoch": 0.9028161236885699, + "grad_norm": 0.9543409943580627, + "learning_rate": 4.766672316965413e-06, + "loss": 0.696, + "step": 1090 + }, + { + "epoch": 0.9036443953616786, + "grad_norm": 0.8894531726837158, + "learning_rate": 4.766208201495906e-06, + "loss": 0.6781, + "step": 1091 + }, + { + "epoch": 0.9044726670347875, + "grad_norm": 0.915542721748352, + "learning_rate": 4.765743647536238e-06, + "loss": 0.6804, + "step": 1092 + }, + { + "epoch": 0.9053009387078962, + "grad_norm": 0.8959763646125793, + "learning_rate": 4.765278655176296e-06, + "loss": 0.6873, + "step": 1093 + }, + { + "epoch": 0.906129210381005, + "grad_norm": 0.881188690662384, + "learning_rate": 4.76481322450605e-06, + "loss": 0.6582, + "step": 1094 + }, + { + "epoch": 0.9069574820541138, + "grad_norm": 0.9096754193305969, + "learning_rate": 4.764347355615558e-06, + "loss": 0.693, + "step": 1095 + }, + { + "epoch": 0.9077857537272225, + "grad_norm": 0.891132652759552, + "learning_rate": 4.76388104859496e-06, + "loss": 0.6781, + "step": 1096 + }, + { + "epoch": 0.9086140254003313, + "grad_norm": 0.893624484539032, + "learning_rate": 4.763414303534484e-06, + "loss": 0.6636, + "step": 1097 + }, + { + "epoch": 0.90944229707344, + "grad_norm": 0.8795678019523621, + "learning_rate": 4.762947120524438e-06, + "loss": 0.6604, + "step": 1098 + }, + { + "epoch": 0.9102705687465489, + "grad_norm": 0.9177265763282776, + "learning_rate": 4.76247949965522e-06, + "loss": 0.6557, + "step": 1099 + }, + { + "epoch": 0.9110988404196576, + "grad_norm": 0.9416985511779785, + "learning_rate": 4.762011441017309e-06, + "loss": 0.7008, + "step": 1100 + }, + { + "epoch": 0.9119271120927664, + "grad_norm": 0.9255479574203491, + "learning_rate": 4.7615429447012704e-06, + "loss": 0.6665, + "step": 1101 + }, + { + "epoch": 0.9127553837658752, + "grad_norm": 0.8948953151702881, + "learning_rate": 4.761074010797754e-06, + "loss": 0.6699, + "step": 1102 + }, + { + "epoch": 0.913583655438984, + "grad_norm": 0.9027442336082458, + "learning_rate": 4.760604639397493e-06, + "loss": 0.6649, + "step": 1103 + }, + { + "epoch": 0.9144119271120927, + "grad_norm": 0.9094238877296448, + "learning_rate": 4.7601348305913085e-06, + "loss": 0.6753, + "step": 1104 + }, + { + "epoch": 0.9152401987852016, + "grad_norm": 0.9098204374313354, + "learning_rate": 4.759664584470103e-06, + "loss": 0.6845, + "step": 1105 + }, + { + "epoch": 0.9160684704583103, + "grad_norm": 0.9339573383331299, + "learning_rate": 4.759193901124864e-06, + "loss": 0.6285, + "step": 1106 + }, + { + "epoch": 0.9168967421314191, + "grad_norm": 0.8812451958656311, + "learning_rate": 4.758722780646666e-06, + "loss": 0.6764, + "step": 1107 + }, + { + "epoch": 0.9177250138045279, + "grad_norm": 0.8832173347473145, + "learning_rate": 4.758251223126665e-06, + "loss": 0.6908, + "step": 1108 + }, + { + "epoch": 0.9185532854776367, + "grad_norm": 0.9124787449836731, + "learning_rate": 4.757779228656103e-06, + "loss": 0.6855, + "step": 1109 + }, + { + "epoch": 0.9193815571507454, + "grad_norm": 0.8778328895568848, + "learning_rate": 4.757306797326309e-06, + "loss": 0.6404, + "step": 1110 + }, + { + "epoch": 0.9202098288238543, + "grad_norm": 0.8831816911697388, + "learning_rate": 4.75683392922869e-06, + "loss": 0.6425, + "step": 1111 + }, + { + "epoch": 0.921038100496963, + "grad_norm": 0.9060006141662598, + "learning_rate": 4.756360624454744e-06, + "loss": 0.6738, + "step": 1112 + }, + { + "epoch": 0.9218663721700718, + "grad_norm": 0.8746777176856995, + "learning_rate": 4.755886883096051e-06, + "loss": 0.6665, + "step": 1113 + }, + { + "epoch": 0.9226946438431806, + "grad_norm": 0.9185945987701416, + "learning_rate": 4.755412705244274e-06, + "loss": 0.6564, + "step": 1114 + }, + { + "epoch": 0.9235229155162893, + "grad_norm": 0.9648581743240356, + "learning_rate": 4.754938090991165e-06, + "loss": 0.6597, + "step": 1115 + }, + { + "epoch": 0.9243511871893981, + "grad_norm": 0.8800092935562134, + "learning_rate": 4.754463040428555e-06, + "loss": 0.6884, + "step": 1116 + }, + { + "epoch": 0.9251794588625069, + "grad_norm": 0.8659892082214355, + "learning_rate": 4.753987553648363e-06, + "loss": 0.68, + "step": 1117 + }, + { + "epoch": 0.9260077305356157, + "grad_norm": 0.88991379737854, + "learning_rate": 4.75351163074259e-06, + "loss": 0.668, + "step": 1118 + }, + { + "epoch": 0.9268360022087244, + "grad_norm": 0.9683054089546204, + "learning_rate": 4.753035271803323e-06, + "loss": 0.6789, + "step": 1119 + }, + { + "epoch": 0.9276642738818333, + "grad_norm": 0.928001880645752, + "learning_rate": 4.752558476922735e-06, + "loss": 0.6786, + "step": 1120 + }, + { + "epoch": 0.928492545554942, + "grad_norm": 0.8760973215103149, + "learning_rate": 4.752081246193078e-06, + "loss": 0.6716, + "step": 1121 + }, + { + "epoch": 0.9293208172280508, + "grad_norm": 0.9171329140663147, + "learning_rate": 4.751603579706695e-06, + "loss": 0.662, + "step": 1122 + }, + { + "epoch": 0.9301490889011595, + "grad_norm": 0.9380231499671936, + "learning_rate": 4.751125477556008e-06, + "loss": 0.6836, + "step": 1123 + }, + { + "epoch": 0.9309773605742684, + "grad_norm": 0.9471802711486816, + "learning_rate": 4.750646939833526e-06, + "loss": 0.668, + "step": 1124 + }, + { + "epoch": 0.9318056322473771, + "grad_norm": 0.8860809206962585, + "learning_rate": 4.750167966631842e-06, + "loss": 0.6635, + "step": 1125 + }, + { + "epoch": 0.932633903920486, + "grad_norm": 0.8694644570350647, + "learning_rate": 4.7496885580436305e-06, + "loss": 0.6505, + "step": 1126 + }, + { + "epoch": 0.9334621755935947, + "grad_norm": 0.9602686166763306, + "learning_rate": 4.749208714161655e-06, + "loss": 0.6874, + "step": 1127 + }, + { + "epoch": 0.9342904472667035, + "grad_norm": 0.9198856353759766, + "learning_rate": 4.748728435078761e-06, + "loss": 0.697, + "step": 1128 + }, + { + "epoch": 0.9351187189398122, + "grad_norm": 0.9162499904632568, + "learning_rate": 4.748247720887876e-06, + "loss": 0.6521, + "step": 1129 + }, + { + "epoch": 0.9359469906129211, + "grad_norm": 0.9064576029777527, + "learning_rate": 4.747766571682016e-06, + "loss": 0.6525, + "step": 1130 + }, + { + "epoch": 0.9367752622860298, + "grad_norm": 0.9518057703971863, + "learning_rate": 4.747284987554277e-06, + "loss": 0.6745, + "step": 1131 + }, + { + "epoch": 0.9376035339591386, + "grad_norm": 0.8963000178337097, + "learning_rate": 4.746802968597842e-06, + "loss": 0.6999, + "step": 1132 + }, + { + "epoch": 0.9384318056322474, + "grad_norm": 0.8687212467193604, + "learning_rate": 4.746320514905977e-06, + "loss": 0.6353, + "step": 1133 + }, + { + "epoch": 0.9392600773053562, + "grad_norm": 0.8598489165306091, + "learning_rate": 4.745837626572033e-06, + "loss": 0.6897, + "step": 1134 + }, + { + "epoch": 0.9400883489784649, + "grad_norm": 0.9425498843193054, + "learning_rate": 4.745354303689444e-06, + "loss": 0.667, + "step": 1135 + }, + { + "epoch": 0.9409166206515737, + "grad_norm": 0.9026349782943726, + "learning_rate": 4.744870546351727e-06, + "loss": 0.676, + "step": 1136 + }, + { + "epoch": 0.9417448923246825, + "grad_norm": 0.8824150562286377, + "learning_rate": 4.744386354652486e-06, + "loss": 0.6941, + "step": 1137 + }, + { + "epoch": 0.9425731639977912, + "grad_norm": 0.8910468816757202, + "learning_rate": 4.743901728685406e-06, + "loss": 0.6701, + "step": 1138 + }, + { + "epoch": 0.9434014356709001, + "grad_norm": 0.9114013314247131, + "learning_rate": 4.74341666854426e-06, + "loss": 0.6735, + "step": 1139 + }, + { + "epoch": 0.9442297073440088, + "grad_norm": 0.9107307195663452, + "learning_rate": 4.742931174322902e-06, + "loss": 0.6505, + "step": 1140 + }, + { + "epoch": 0.9450579790171176, + "grad_norm": 0.8906923532485962, + "learning_rate": 4.742445246115271e-06, + "loss": 0.6648, + "step": 1141 + }, + { + "epoch": 0.9458862506902264, + "grad_norm": 0.902644157409668, + "learning_rate": 4.741958884015387e-06, + "loss": 0.6674, + "step": 1142 + }, + { + "epoch": 0.9467145223633352, + "grad_norm": 0.9522087574005127, + "learning_rate": 4.741472088117358e-06, + "loss": 0.6825, + "step": 1143 + }, + { + "epoch": 0.9475427940364439, + "grad_norm": 0.9152255654335022, + "learning_rate": 4.740984858515376e-06, + "loss": 0.6708, + "step": 1144 + }, + { + "epoch": 0.9483710657095528, + "grad_norm": 0.8779835104942322, + "learning_rate": 4.740497195303713e-06, + "loss": 0.6562, + "step": 1145 + }, + { + "epoch": 0.9491993373826615, + "grad_norm": 0.8842780590057373, + "learning_rate": 4.740009098576729e-06, + "loss": 0.6769, + "step": 1146 + }, + { + "epoch": 0.9500276090557703, + "grad_norm": 0.9137415289878845, + "learning_rate": 4.739520568428866e-06, + "loss": 0.6705, + "step": 1147 + }, + { + "epoch": 0.950855880728879, + "grad_norm": 0.8867765665054321, + "learning_rate": 4.739031604954648e-06, + "loss": 0.659, + "step": 1148 + }, + { + "epoch": 0.9516841524019879, + "grad_norm": 0.8965260982513428, + "learning_rate": 4.7385422082486874e-06, + "loss": 0.6744, + "step": 1149 + }, + { + "epoch": 0.9525124240750966, + "grad_norm": 0.8861697912216187, + "learning_rate": 4.738052378405677e-06, + "loss": 0.6575, + "step": 1150 + }, + { + "epoch": 0.9533406957482055, + "grad_norm": 0.9442310333251953, + "learning_rate": 4.737562115520394e-06, + "loss": 0.7164, + "step": 1151 + }, + { + "epoch": 0.9541689674213142, + "grad_norm": 0.9340727925300598, + "learning_rate": 4.7370714196877e-06, + "loss": 0.6577, + "step": 1152 + }, + { + "epoch": 0.954997239094423, + "grad_norm": 0.9337487816810608, + "learning_rate": 4.7365802910025405e-06, + "loss": 0.6664, + "step": 1153 + }, + { + "epoch": 0.9558255107675317, + "grad_norm": 0.8849001526832581, + "learning_rate": 4.736088729559943e-06, + "loss": 0.6721, + "step": 1154 + }, + { + "epoch": 0.9566537824406405, + "grad_norm": 0.917323648929596, + "learning_rate": 4.735596735455022e-06, + "loss": 0.6715, + "step": 1155 + }, + { + "epoch": 0.9574820541137493, + "grad_norm": 0.9480959177017212, + "learning_rate": 4.735104308782972e-06, + "loss": 0.6784, + "step": 1156 + }, + { + "epoch": 0.958310325786858, + "grad_norm": 0.9519251585006714, + "learning_rate": 4.734611449639073e-06, + "loss": 0.6617, + "step": 1157 + }, + { + "epoch": 0.9591385974599669, + "grad_norm": 0.8652079105377197, + "learning_rate": 4.734118158118689e-06, + "loss": 0.6846, + "step": 1158 + }, + { + "epoch": 0.9599668691330756, + "grad_norm": 0.8860392570495605, + "learning_rate": 4.733624434317269e-06, + "loss": 0.6729, + "step": 1159 + }, + { + "epoch": 0.9607951408061844, + "grad_norm": 0.9467882513999939, + "learning_rate": 4.7331302783303416e-06, + "loss": 0.6362, + "step": 1160 + }, + { + "epoch": 0.9616234124792932, + "grad_norm": 0.9138529896736145, + "learning_rate": 4.732635690253523e-06, + "loss": 0.6613, + "step": 1161 + }, + { + "epoch": 0.962451684152402, + "grad_norm": 0.9401054978370667, + "learning_rate": 4.7321406701825105e-06, + "loss": 0.6845, + "step": 1162 + }, + { + "epoch": 0.9632799558255107, + "grad_norm": 0.9049288630485535, + "learning_rate": 4.731645218213086e-06, + "loss": 0.6857, + "step": 1163 + }, + { + "epoch": 0.9641082274986196, + "grad_norm": 0.8972534537315369, + "learning_rate": 4.731149334441114e-06, + "loss": 0.6745, + "step": 1164 + }, + { + "epoch": 0.9649364991717283, + "grad_norm": 1.0799483060836792, + "learning_rate": 4.730653018962545e-06, + "loss": 0.6754, + "step": 1165 + }, + { + "epoch": 0.9657647708448371, + "grad_norm": 0.9841833710670471, + "learning_rate": 4.730156271873412e-06, + "loss": 0.6718, + "step": 1166 + }, + { + "epoch": 0.9665930425179459, + "grad_norm": 0.9388216733932495, + "learning_rate": 4.729659093269828e-06, + "loss": 0.6622, + "step": 1167 + }, + { + "epoch": 0.9674213141910547, + "grad_norm": 0.8879573345184326, + "learning_rate": 4.729161483247994e-06, + "loss": 0.6909, + "step": 1168 + }, + { + "epoch": 0.9682495858641634, + "grad_norm": 0.9713517427444458, + "learning_rate": 4.728663441904194e-06, + "loss": 0.6956, + "step": 1169 + }, + { + "epoch": 0.9690778575372723, + "grad_norm": 0.9200823903083801, + "learning_rate": 4.7281649693347925e-06, + "loss": 0.6766, + "step": 1170 + }, + { + "epoch": 0.969906129210381, + "grad_norm": 0.8983714580535889, + "learning_rate": 4.727666065636241e-06, + "loss": 0.6626, + "step": 1171 + }, + { + "epoch": 0.9707344008834898, + "grad_norm": 0.8792792558670044, + "learning_rate": 4.72716673090507e-06, + "loss": 0.666, + "step": 1172 + }, + { + "epoch": 0.9715626725565986, + "grad_norm": 1.0116846561431885, + "learning_rate": 4.7266669652379005e-06, + "loss": 0.6574, + "step": 1173 + }, + { + "epoch": 0.9723909442297074, + "grad_norm": 0.9433228373527527, + "learning_rate": 4.7261667687314285e-06, + "loss": 0.6972, + "step": 1174 + }, + { + "epoch": 0.9732192159028161, + "grad_norm": 0.8979770541191101, + "learning_rate": 4.725666141482439e-06, + "loss": 0.6523, + "step": 1175 + }, + { + "epoch": 0.9740474875759249, + "grad_norm": 0.8990604281425476, + "learning_rate": 4.725165083587798e-06, + "loss": 0.6535, + "step": 1176 + }, + { + "epoch": 0.9748757592490337, + "grad_norm": 0.9411956667900085, + "learning_rate": 4.724663595144457e-06, + "loss": 0.6947, + "step": 1177 + }, + { + "epoch": 0.9757040309221424, + "grad_norm": 1.0203242301940918, + "learning_rate": 4.724161676249448e-06, + "loss": 0.6894, + "step": 1178 + }, + { + "epoch": 0.9765323025952513, + "grad_norm": 0.8938170075416565, + "learning_rate": 4.723659326999888e-06, + "loss": 0.6648, + "step": 1179 + }, + { + "epoch": 0.97736057426836, + "grad_norm": 0.8810412883758545, + "learning_rate": 4.7231565474929765e-06, + "loss": 0.6761, + "step": 1180 + }, + { + "epoch": 0.9781888459414688, + "grad_norm": 0.8903076648712158, + "learning_rate": 4.722653337825998e-06, + "loss": 0.6738, + "step": 1181 + }, + { + "epoch": 0.9790171176145775, + "grad_norm": 0.905060350894928, + "learning_rate": 4.7221496980963175e-06, + "loss": 0.6625, + "step": 1182 + }, + { + "epoch": 0.9798453892876864, + "grad_norm": 0.8791100978851318, + "learning_rate": 4.721645628401385e-06, + "loss": 0.6884, + "step": 1183 + }, + { + "epoch": 0.9806736609607951, + "grad_norm": 0.897101104259491, + "learning_rate": 4.721141128838733e-06, + "loss": 0.6693, + "step": 1184 + }, + { + "epoch": 0.981501932633904, + "grad_norm": 0.9102833867073059, + "learning_rate": 4.7206361995059775e-06, + "loss": 0.6682, + "step": 1185 + }, + { + "epoch": 0.9823302043070127, + "grad_norm": 0.9492197632789612, + "learning_rate": 4.720130840500819e-06, + "loss": 0.6501, + "step": 1186 + }, + { + "epoch": 0.9831584759801215, + "grad_norm": 0.9783265590667725, + "learning_rate": 4.719625051921038e-06, + "loss": 0.6454, + "step": 1187 + }, + { + "epoch": 0.9839867476532302, + "grad_norm": 0.8938109874725342, + "learning_rate": 4.719118833864501e-06, + "loss": 0.6671, + "step": 1188 + }, + { + "epoch": 0.9848150193263391, + "grad_norm": 0.897484302520752, + "learning_rate": 4.718612186429156e-06, + "loss": 0.6664, + "step": 1189 + }, + { + "epoch": 0.9856432909994478, + "grad_norm": 0.9640376567840576, + "learning_rate": 4.718105109713034e-06, + "loss": 0.6845, + "step": 1190 + }, + { + "epoch": 0.9864715626725566, + "grad_norm": 0.9495493769645691, + "learning_rate": 4.7175976038142505e-06, + "loss": 0.6864, + "step": 1191 + }, + { + "epoch": 0.9872998343456654, + "grad_norm": 0.9390130639076233, + "learning_rate": 4.717089668831002e-06, + "loss": 0.6734, + "step": 1192 + }, + { + "epoch": 0.9881281060187742, + "grad_norm": 0.8854318261146545, + "learning_rate": 4.716581304861571e-06, + "loss": 0.6546, + "step": 1193 + }, + { + "epoch": 0.9889563776918829, + "grad_norm": 0.9242726564407349, + "learning_rate": 4.71607251200432e-06, + "loss": 0.6663, + "step": 1194 + }, + { + "epoch": 0.9897846493649917, + "grad_norm": 0.9302003383636475, + "learning_rate": 4.715563290357696e-06, + "loss": 0.6844, + "step": 1195 + }, + { + "epoch": 0.9906129210381005, + "grad_norm": 0.9279932975769043, + "learning_rate": 4.715053640020228e-06, + "loss": 0.6731, + "step": 1196 + }, + { + "epoch": 0.9914411927112092, + "grad_norm": 0.8753612041473389, + "learning_rate": 4.71454356109053e-06, + "loss": 0.6738, + "step": 1197 + }, + { + "epoch": 0.9922694643843181, + "grad_norm": 0.890133261680603, + "learning_rate": 4.714033053667296e-06, + "loss": 0.6638, + "step": 1198 + }, + { + "epoch": 0.9930977360574268, + "grad_norm": 0.9343729615211487, + "learning_rate": 4.713522117849305e-06, + "loss": 0.6826, + "step": 1199 + }, + { + "epoch": 0.9939260077305356, + "grad_norm": 0.9052755832672119, + "learning_rate": 4.7130107537354185e-06, + "loss": 0.6897, + "step": 1200 + }, + { + "epoch": 0.9947542794036444, + "grad_norm": 0.8683924674987793, + "learning_rate": 4.7124989614245805e-06, + "loss": 0.6687, + "step": 1201 + }, + { + "epoch": 0.9955825510767532, + "grad_norm": 0.8677132725715637, + "learning_rate": 4.711986741015818e-06, + "loss": 0.6917, + "step": 1202 + }, + { + "epoch": 0.9964108227498619, + "grad_norm": 0.8840690851211548, + "learning_rate": 4.711474092608243e-06, + "loss": 0.6841, + "step": 1203 + }, + { + "epoch": 0.9972390944229708, + "grad_norm": 0.9750391840934753, + "learning_rate": 4.7109610163010435e-06, + "loss": 0.6574, + "step": 1204 + }, + { + "epoch": 0.9980673660960795, + "grad_norm": 0.8752943873405457, + "learning_rate": 4.710447512193499e-06, + "loss": 0.636, + "step": 1205 + }, + { + "epoch": 0.9988956377691883, + "grad_norm": 0.9100313782691956, + "learning_rate": 4.709933580384965e-06, + "loss": 0.6712, + "step": 1206 + }, + { + "epoch": 0.999723909442297, + "grad_norm": 0.9202356338500977, + "learning_rate": 4.709419220974886e-06, + "loss": 0.6641, + "step": 1207 + } + ], + "logging_steps": 1, + "max_steps": 7242, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 1207, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 3.048926292989706e+18, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-1207/training_args.bin b/checkpoint-1207/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..7017eeafdf1f43b26668f0e0e5710ca52345a005 --- /dev/null +++ b/checkpoint-1207/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fdc878d2541c09353dd68f4f8a2464140b28fd84e1bc48b59914a6e84328e85e +size 8056 diff --git a/checkpoint-1207/zero_to_fp32.py b/checkpoint-1207/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-1207/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-2414/README.md b/checkpoint-2414/README.md new file mode 100644 index 0000000000000000000000000000000000000000..049d467664ca6172b7ffbe6ba60b3eac7479cac4 --- /dev/null +++ b/checkpoint-2414/README.md @@ -0,0 +1,202 @@ +--- +base_model: meta-llama/Llama-3.1-8B +library_name: peft +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.14.0 \ No newline at end of file diff --git a/checkpoint-2414/adapter_config.json b/checkpoint-2414/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d6fca8a067dce19d599e7f1e24bea8019968fe0f --- /dev/null +++ b/checkpoint-2414/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.1-8B", + "bias": "none", + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "up_proj", + "q_proj", + "k_proj", + "o_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-2414/adapter_model.safetensors b/checkpoint-2414/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..fdd000ad150afefd52669c8011e30895f98bfa2f --- /dev/null +++ b/checkpoint-2414/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c2cf8a6442a77a65156c31c7985bf47408a93f1acf526a2c27ad9a13cf81955d +size 3443586272 diff --git a/checkpoint-2414/global_step2413/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-2414/global_step2413/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..28348f7a5e0d6cb559cffa7fdf6e55007f0cddba --- /dev/null +++ b/checkpoint-2414/global_step2413/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3c5919c8a2d0422e7e060fbd89d6514bbfdfecb645edf107722283802315d203 +size 20661195036 diff --git a/checkpoint-2414/global_step2413/mp_rank_00_model_states.pt b/checkpoint-2414/global_step2413/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..617c79c92ac0c58f3c4ff170472dc7d92d4fdb7b --- /dev/null +++ b/checkpoint-2414/global_step2413/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2f8cd1e5392f6d8788fae2d60d078395fca40ea6de135d9ad66f61e82f4bcdc4 +size 3555326777 diff --git a/checkpoint-2414/latest b/checkpoint-2414/latest new file mode 100644 index 0000000000000000000000000000000000000000..a702fba0cdc4202ef0a860d32cfe432271b815cc --- /dev/null +++ b/checkpoint-2414/latest @@ -0,0 +1 @@ +global_step2413 \ No newline at end of file diff --git a/checkpoint-2414/rng_state.pth b/checkpoint-2414/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..e83f9fecb1334078298398f607f0c310687d19cb --- /dev/null +++ b/checkpoint-2414/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30696f16b946a5896425b687e986dd55e6f87bf06d8b59ba8a2d9c80c40f9cca +size 14244 diff --git a/checkpoint-2414/scheduler.pt b/checkpoint-2414/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..86cd77a44260d068bb734e315a796326a28f2de1 --- /dev/null +++ b/checkpoint-2414/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:11ad7cf24e2c2a19624735da1c22f0fb67444599c197d9d698fd75f5a5163d7b +size 1064 diff --git a/checkpoint-2414/special_tokens_map.json b/checkpoint-2414/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..e5b39b6305d89284b04934011c68dbb26bf588ca --- /dev/null +++ b/checkpoint-2414/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-2414/tokenizer.json b/checkpoint-2414/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-2414/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-2414/tokenizer_config.json b/checkpoint-2414/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..81dd14db6632ad5b35b9d447732e37ac074873a5 --- /dev/null +++ b/checkpoint-2414/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +} diff --git a/checkpoint-2414/trainer_state.json b/checkpoint-2414/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..553bacf6e3ea4809ed8f644c04ef775b7fe23152 --- /dev/null +++ b/checkpoint-2414/trainer_state.json @@ -0,0 +1,16931 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.9988956377691882, + "eval_steps": 500, + "global_step": 2414, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0008282716731087797, + "grad_norm": 2.4795358180999756, + "learning_rate": 5.0000000000000004e-08, + "loss": 1.1215, + "step": 1 + }, + { + "epoch": 0.0016565433462175593, + "grad_norm": 2.6060028076171875, + "learning_rate": 1.0000000000000001e-07, + "loss": 1.1219, + "step": 2 + }, + { + "epoch": 0.002484815019326339, + "grad_norm": 2.750868082046509, + "learning_rate": 1.5000000000000002e-07, + "loss": 1.125, + "step": 3 + }, + { + "epoch": 0.0033130866924351186, + "grad_norm": 2.7321369647979736, + "learning_rate": 2.0000000000000002e-07, + "loss": 1.1164, + "step": 4 + }, + { + "epoch": 0.0041413583655438985, + "grad_norm": 2.8193929195404053, + "learning_rate": 2.5000000000000004e-07, + "loss": 1.1263, + "step": 5 + }, + { + "epoch": 0.004969630038652678, + "grad_norm": 2.648301839828491, + "learning_rate": 3.0000000000000004e-07, + "loss": 1.1092, + "step": 6 + }, + { + "epoch": 0.005797901711761457, + "grad_norm": 2.751993179321289, + "learning_rate": 3.5000000000000004e-07, + "loss": 1.0997, + "step": 7 + }, + { + "epoch": 0.006626173384870237, + "grad_norm": 2.592381000518799, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.112, + "step": 8 + }, + { + "epoch": 0.007454445057979017, + "grad_norm": 2.3135671615600586, + "learning_rate": 4.5000000000000003e-07, + "loss": 1.1045, + "step": 9 + }, + { + "epoch": 0.008282716731087797, + "grad_norm": 2.60491943359375, + "learning_rate": 5.000000000000001e-07, + "loss": 1.0956, + "step": 10 + }, + { + "epoch": 0.009110988404196576, + "grad_norm": 2.2435271739959717, + "learning_rate": 5.5e-07, + "loss": 1.1241, + "step": 11 + }, + { + "epoch": 0.009939260077305357, + "grad_norm": 2.4412319660186768, + "learning_rate": 6.000000000000001e-07, + "loss": 1.11, + "step": 12 + }, + { + "epoch": 0.010767531750414136, + "grad_norm": 2.2075185775756836, + "learning_rate": 6.5e-07, + "loss": 1.0878, + "step": 13 + }, + { + "epoch": 0.011595803423522915, + "grad_norm": 2.101902723312378, + "learning_rate": 7.000000000000001e-07, + "loss": 1.0863, + "step": 14 + }, + { + "epoch": 0.012424075096631695, + "grad_norm": 2.134535074234009, + "learning_rate": 7.5e-07, + "loss": 1.0735, + "step": 15 + }, + { + "epoch": 0.013252346769740474, + "grad_norm": 2.000504493713379, + "learning_rate": 8.000000000000001e-07, + "loss": 1.0875, + "step": 16 + }, + { + "epoch": 0.014080618442849255, + "grad_norm": 1.9191882610321045, + "learning_rate": 8.500000000000001e-07, + "loss": 1.1039, + "step": 17 + }, + { + "epoch": 0.014908890115958034, + "grad_norm": 1.7732031345367432, + "learning_rate": 9.000000000000001e-07, + "loss": 1.0819, + "step": 18 + }, + { + "epoch": 0.015737161789066815, + "grad_norm": 1.7775251865386963, + "learning_rate": 9.500000000000001e-07, + "loss": 1.0936, + "step": 19 + }, + { + "epoch": 0.016565433462175594, + "grad_norm": 1.6648598909378052, + "learning_rate": 1.0000000000000002e-06, + "loss": 1.0881, + "step": 20 + }, + { + "epoch": 0.017393705135284373, + "grad_norm": 1.5726720094680786, + "learning_rate": 1.0500000000000001e-06, + "loss": 1.07, + "step": 21 + }, + { + "epoch": 0.018221976808393152, + "grad_norm": 1.561417818069458, + "learning_rate": 1.1e-06, + "loss": 1.0668, + "step": 22 + }, + { + "epoch": 0.019050248481501934, + "grad_norm": 1.4953842163085938, + "learning_rate": 1.1500000000000002e-06, + "loss": 1.054, + "step": 23 + }, + { + "epoch": 0.019878520154610713, + "grad_norm": 1.4247829914093018, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.0363, + "step": 24 + }, + { + "epoch": 0.020706791827719492, + "grad_norm": 1.3483426570892334, + "learning_rate": 1.25e-06, + "loss": 1.0143, + "step": 25 + }, + { + "epoch": 0.02153506350082827, + "grad_norm": 1.3412944078445435, + "learning_rate": 1.3e-06, + "loss": 1.025, + "step": 26 + }, + { + "epoch": 0.02236333517393705, + "grad_norm": 1.2834237813949585, + "learning_rate": 1.3500000000000002e-06, + "loss": 1.0227, + "step": 27 + }, + { + "epoch": 0.02319160684704583, + "grad_norm": 1.2128241062164307, + "learning_rate": 1.4000000000000001e-06, + "loss": 1.0161, + "step": 28 + }, + { + "epoch": 0.024019878520154612, + "grad_norm": 1.1665573120117188, + "learning_rate": 1.45e-06, + "loss": 1.0034, + "step": 29 + }, + { + "epoch": 0.02484815019326339, + "grad_norm": 1.1575899124145508, + "learning_rate": 1.5e-06, + "loss": 0.9968, + "step": 30 + }, + { + "epoch": 0.02567642186637217, + "grad_norm": 1.195304274559021, + "learning_rate": 1.5500000000000002e-06, + "loss": 1.0176, + "step": 31 + }, + { + "epoch": 0.02650469353948095, + "grad_norm": 1.1321678161621094, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.0113, + "step": 32 + }, + { + "epoch": 0.027332965212589728, + "grad_norm": 1.1076338291168213, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.9971, + "step": 33 + }, + { + "epoch": 0.02816123688569851, + "grad_norm": 1.0770610570907593, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.9994, + "step": 34 + }, + { + "epoch": 0.02898950855880729, + "grad_norm": 1.0767767429351807, + "learning_rate": 1.75e-06, + "loss": 0.978, + "step": 35 + }, + { + "epoch": 0.02981778023191607, + "grad_norm": 1.054800033569336, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.9829, + "step": 36 + }, + { + "epoch": 0.030646051905024847, + "grad_norm": 1.031015157699585, + "learning_rate": 1.85e-06, + "loss": 0.97, + "step": 37 + }, + { + "epoch": 0.03147432357813363, + "grad_norm": 1.0182578563690186, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.9549, + "step": 38 + }, + { + "epoch": 0.032302595251242405, + "grad_norm": 0.9796578288078308, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.9762, + "step": 39 + }, + { + "epoch": 0.03313086692435119, + "grad_norm": 0.9956578016281128, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.9693, + "step": 40 + }, + { + "epoch": 0.03395913859745997, + "grad_norm": 0.9383260011672974, + "learning_rate": 2.05e-06, + "loss": 0.9556, + "step": 41 + }, + { + "epoch": 0.034787410270568746, + "grad_norm": 0.9372819066047668, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.9419, + "step": 42 + }, + { + "epoch": 0.03561568194367753, + "grad_norm": 0.9334004521369934, + "learning_rate": 2.15e-06, + "loss": 0.9587, + "step": 43 + }, + { + "epoch": 0.036443953616786304, + "grad_norm": 0.8916004300117493, + "learning_rate": 2.2e-06, + "loss": 0.9227, + "step": 44 + }, + { + "epoch": 0.037272225289895086, + "grad_norm": 0.8698756694793701, + "learning_rate": 2.25e-06, + "loss": 0.9275, + "step": 45 + }, + { + "epoch": 0.03810049696300387, + "grad_norm": 0.8775100708007812, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.941, + "step": 46 + }, + { + "epoch": 0.038928768636112644, + "grad_norm": 0.8763384222984314, + "learning_rate": 2.35e-06, + "loss": 0.9253, + "step": 47 + }, + { + "epoch": 0.03975704030922143, + "grad_norm": 0.8534179925918579, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.9327, + "step": 48 + }, + { + "epoch": 0.0405853119823302, + "grad_norm": 0.8432146906852722, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.9236, + "step": 49 + }, + { + "epoch": 0.041413583655438985, + "grad_norm": 0.9899975061416626, + "learning_rate": 2.5e-06, + "loss": 0.8839, + "step": 50 + }, + { + "epoch": 0.04224185532854776, + "grad_norm": 0.8370104432106018, + "learning_rate": 2.55e-06, + "loss": 0.906, + "step": 51 + }, + { + "epoch": 0.04307012700165654, + "grad_norm": 0.8207446336746216, + "learning_rate": 2.6e-06, + "loss": 0.8755, + "step": 52 + }, + { + "epoch": 0.043898398674765325, + "grad_norm": 0.805584728717804, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.9049, + "step": 53 + }, + { + "epoch": 0.0447266703478741, + "grad_norm": 0.8162860870361328, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.8993, + "step": 54 + }, + { + "epoch": 0.04555494202098288, + "grad_norm": 0.8151668906211853, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.9282, + "step": 55 + }, + { + "epoch": 0.04638321369409166, + "grad_norm": 0.7996147274971008, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.8946, + "step": 56 + }, + { + "epoch": 0.04721148536720044, + "grad_norm": 0.8223204016685486, + "learning_rate": 2.85e-06, + "loss": 0.897, + "step": 57 + }, + { + "epoch": 0.048039757040309224, + "grad_norm": 0.8117988109588623, + "learning_rate": 2.9e-06, + "loss": 0.8845, + "step": 58 + }, + { + "epoch": 0.048868028713418, + "grad_norm": 0.7902786731719971, + "learning_rate": 2.95e-06, + "loss": 0.8846, + "step": 59 + }, + { + "epoch": 0.04969630038652678, + "grad_norm": 0.804192304611206, + "learning_rate": 3e-06, + "loss": 0.8722, + "step": 60 + }, + { + "epoch": 0.05052457205963556, + "grad_norm": 0.7852040529251099, + "learning_rate": 3.05e-06, + "loss": 0.8821, + "step": 61 + }, + { + "epoch": 0.05135284373274434, + "grad_norm": 0.8045080304145813, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.8661, + "step": 62 + }, + { + "epoch": 0.05218111540585312, + "grad_norm": 0.7920243144035339, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.8712, + "step": 63 + }, + { + "epoch": 0.0530093870789619, + "grad_norm": 0.7601844072341919, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.8792, + "step": 64 + }, + { + "epoch": 0.05383765875207068, + "grad_norm": 0.7833033800125122, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.8811, + "step": 65 + }, + { + "epoch": 0.054665930425179456, + "grad_norm": 0.7691664099693298, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.8701, + "step": 66 + }, + { + "epoch": 0.05549420209828824, + "grad_norm": 0.7845340967178345, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.8929, + "step": 67 + }, + { + "epoch": 0.05632247377139702, + "grad_norm": 0.9076205492019653, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.8352, + "step": 68 + }, + { + "epoch": 0.057150745444505796, + "grad_norm": 0.7614927887916565, + "learning_rate": 3.45e-06, + "loss": 0.8766, + "step": 69 + }, + { + "epoch": 0.05797901711761458, + "grad_norm": 0.7722934484481812, + "learning_rate": 3.5e-06, + "loss": 0.8666, + "step": 70 + }, + { + "epoch": 0.058807288790723354, + "grad_norm": 0.7603748440742493, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.84, + "step": 71 + }, + { + "epoch": 0.05963556046383214, + "grad_norm": 0.7563591003417969, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.8784, + "step": 72 + }, + { + "epoch": 0.06046383213694092, + "grad_norm": 0.7835114002227783, + "learning_rate": 3.65e-06, + "loss": 0.8516, + "step": 73 + }, + { + "epoch": 0.061292103810049695, + "grad_norm": 0.7760982513427734, + "learning_rate": 3.7e-06, + "loss": 0.8265, + "step": 74 + }, + { + "epoch": 0.06212037548315848, + "grad_norm": 0.7978456020355225, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.8564, + "step": 75 + }, + { + "epoch": 0.06294864715626726, + "grad_norm": 0.7940747141838074, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.8635, + "step": 76 + }, + { + "epoch": 0.06377691882937604, + "grad_norm": 0.8055519461631775, + "learning_rate": 3.85e-06, + "loss": 0.8445, + "step": 77 + }, + { + "epoch": 0.06460519050248481, + "grad_norm": 0.7532406449317932, + "learning_rate": 3.900000000000001e-06, + "loss": 0.8772, + "step": 78 + }, + { + "epoch": 0.06543346217559359, + "grad_norm": 0.763933539390564, + "learning_rate": 3.95e-06, + "loss": 0.8574, + "step": 79 + }, + { + "epoch": 0.06626173384870238, + "grad_norm": 0.7833206057548523, + "learning_rate": 4.000000000000001e-06, + "loss": 0.8643, + "step": 80 + }, + { + "epoch": 0.06709000552181116, + "grad_norm": 0.7825527787208557, + "learning_rate": 4.05e-06, + "loss": 0.8518, + "step": 81 + }, + { + "epoch": 0.06791827719491994, + "grad_norm": 0.7749581336975098, + "learning_rate": 4.1e-06, + "loss": 0.8146, + "step": 82 + }, + { + "epoch": 0.06874654886802871, + "grad_norm": 0.8117144703865051, + "learning_rate": 4.15e-06, + "loss": 0.8719, + "step": 83 + }, + { + "epoch": 0.06957482054113749, + "grad_norm": 0.8440507054328918, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.8665, + "step": 84 + }, + { + "epoch": 0.07040309221424627, + "grad_norm": 0.7941984534263611, + "learning_rate": 4.25e-06, + "loss": 0.8492, + "step": 85 + }, + { + "epoch": 0.07123136388735506, + "grad_norm": 0.8301989436149597, + "learning_rate": 4.3e-06, + "loss": 0.8544, + "step": 86 + }, + { + "epoch": 0.07205963556046384, + "grad_norm": 0.757798969745636, + "learning_rate": 4.350000000000001e-06, + "loss": 0.8426, + "step": 87 + }, + { + "epoch": 0.07288790723357261, + "grad_norm": 0.7817673683166504, + "learning_rate": 4.4e-06, + "loss": 0.8554, + "step": 88 + }, + { + "epoch": 0.07371617890668139, + "grad_norm": 0.7609818577766418, + "learning_rate": 4.450000000000001e-06, + "loss": 0.8473, + "step": 89 + }, + { + "epoch": 0.07454445057979017, + "grad_norm": 0.8146091103553772, + "learning_rate": 4.5e-06, + "loss": 0.8317, + "step": 90 + }, + { + "epoch": 0.07537272225289895, + "grad_norm": 0.8058239817619324, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.8411, + "step": 91 + }, + { + "epoch": 0.07620099392600774, + "grad_norm": 0.7700346112251282, + "learning_rate": 4.600000000000001e-06, + "loss": 0.815, + "step": 92 + }, + { + "epoch": 0.0770292655991165, + "grad_norm": 0.7949067950248718, + "learning_rate": 4.65e-06, + "loss": 0.8512, + "step": 93 + }, + { + "epoch": 0.07785753727222529, + "grad_norm": 0.8112176060676575, + "learning_rate": 4.7e-06, + "loss": 0.8398, + "step": 94 + }, + { + "epoch": 0.07868580894533407, + "grad_norm": 0.7990434765815735, + "learning_rate": 4.75e-06, + "loss": 0.8276, + "step": 95 + }, + { + "epoch": 0.07951408061844285, + "grad_norm": 0.7942920923233032, + "learning_rate": 4.800000000000001e-06, + "loss": 0.8333, + "step": 96 + }, + { + "epoch": 0.08034235229155164, + "grad_norm": 0.8005534410476685, + "learning_rate": 4.85e-06, + "loss": 0.8453, + "step": 97 + }, + { + "epoch": 0.0811706239646604, + "grad_norm": 0.8103216290473938, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8303, + "step": 98 + }, + { + "epoch": 0.08199889563776919, + "grad_norm": 0.8217238187789917, + "learning_rate": 4.95e-06, + "loss": 0.8204, + "step": 99 + }, + { + "epoch": 0.08282716731087797, + "grad_norm": 0.7958654761314392, + "learning_rate": 5e-06, + "loss": 0.8237, + "step": 100 + }, + { + "epoch": 0.08365543898398675, + "grad_norm": 0.8631463050842285, + "learning_rate": 4.999999758136652e-06, + "loss": 0.821, + "step": 101 + }, + { + "epoch": 0.08448371065709552, + "grad_norm": 0.7666735053062439, + "learning_rate": 4.999999032546657e-06, + "loss": 0.8307, + "step": 102 + }, + { + "epoch": 0.0853119823302043, + "grad_norm": 0.7709226012229919, + "learning_rate": 4.999997823230153e-06, + "loss": 0.8299, + "step": 103 + }, + { + "epoch": 0.08614025400331309, + "grad_norm": 0.8020651340484619, + "learning_rate": 4.9999961301873736e-06, + "loss": 0.8075, + "step": 104 + }, + { + "epoch": 0.08696852567642187, + "grad_norm": 0.8019537329673767, + "learning_rate": 4.999993953418649e-06, + "loss": 0.8615, + "step": 105 + }, + { + "epoch": 0.08779679734953065, + "grad_norm": 0.7904282808303833, + "learning_rate": 4.999991292924399e-06, + "loss": 0.8353, + "step": 106 + }, + { + "epoch": 0.08862506902263942, + "grad_norm": 0.7854570150375366, + "learning_rate": 4.999988148705138e-06, + "loss": 0.8154, + "step": 107 + }, + { + "epoch": 0.0894533406957482, + "grad_norm": 0.8044222593307495, + "learning_rate": 4.999984520761475e-06, + "loss": 0.8117, + "step": 108 + }, + { + "epoch": 0.09028161236885698, + "grad_norm": 0.7986987829208374, + "learning_rate": 4.999980409094112e-06, + "loss": 0.8407, + "step": 109 + }, + { + "epoch": 0.09110988404196577, + "grad_norm": 0.7976786494255066, + "learning_rate": 4.999975813703844e-06, + "loss": 0.7871, + "step": 110 + }, + { + "epoch": 0.09193815571507455, + "grad_norm": 0.8330315351486206, + "learning_rate": 4.9999707345915605e-06, + "loss": 0.8129, + "step": 111 + }, + { + "epoch": 0.09276642738818332, + "grad_norm": 0.8350826501846313, + "learning_rate": 4.999965171758245e-06, + "loss": 0.8144, + "step": 112 + }, + { + "epoch": 0.0935946990612921, + "grad_norm": 0.8054768443107605, + "learning_rate": 4.999959125204973e-06, + "loss": 0.809, + "step": 113 + }, + { + "epoch": 0.09442297073440088, + "grad_norm": 0.8101645708084106, + "learning_rate": 4.999952594932914e-06, + "loss": 0.8081, + "step": 114 + }, + { + "epoch": 0.09525124240750966, + "grad_norm": 0.905028760433197, + "learning_rate": 4.999945580943332e-06, + "loss": 0.8251, + "step": 115 + }, + { + "epoch": 0.09607951408061845, + "grad_norm": 0.8529418706893921, + "learning_rate": 4.999938083237585e-06, + "loss": 0.7984, + "step": 116 + }, + { + "epoch": 0.09690778575372722, + "grad_norm": 0.8047451376914978, + "learning_rate": 4.9999301018171225e-06, + "loss": 0.8222, + "step": 117 + }, + { + "epoch": 0.097736057426836, + "grad_norm": 0.8073400259017944, + "learning_rate": 4.99992163668349e-06, + "loss": 0.8251, + "step": 118 + }, + { + "epoch": 0.09856432909994478, + "grad_norm": 0.8470693826675415, + "learning_rate": 4.999912687838324e-06, + "loss": 0.8291, + "step": 119 + }, + { + "epoch": 0.09939260077305356, + "grad_norm": 0.8261650800704956, + "learning_rate": 4.999903255283357e-06, + "loss": 0.808, + "step": 120 + }, + { + "epoch": 0.10022087244616235, + "grad_norm": 0.8421704769134521, + "learning_rate": 4.999893339020414e-06, + "loss": 0.8295, + "step": 121 + }, + { + "epoch": 0.10104914411927111, + "grad_norm": 0.8244299292564392, + "learning_rate": 4.9998829390514134e-06, + "loss": 0.8215, + "step": 122 + }, + { + "epoch": 0.1018774157923799, + "grad_norm": 0.8365658521652222, + "learning_rate": 4.999872055378368e-06, + "loss": 0.8245, + "step": 123 + }, + { + "epoch": 0.10270568746548868, + "grad_norm": 0.9089561104774475, + "learning_rate": 4.999860688003384e-06, + "loss": 0.8125, + "step": 124 + }, + { + "epoch": 0.10353395913859746, + "grad_norm": 0.8784728646278381, + "learning_rate": 4.9998488369286595e-06, + "loss": 0.8321, + "step": 125 + }, + { + "epoch": 0.10436223081170624, + "grad_norm": 0.8596073389053345, + "learning_rate": 4.999836502156487e-06, + "loss": 0.8333, + "step": 126 + }, + { + "epoch": 0.10519050248481501, + "grad_norm": 0.8472688794136047, + "learning_rate": 4.9998236836892566e-06, + "loss": 0.8068, + "step": 127 + }, + { + "epoch": 0.1060187741579238, + "grad_norm": 0.8716503381729126, + "learning_rate": 4.999810381529447e-06, + "loss": 0.8005, + "step": 128 + }, + { + "epoch": 0.10684704583103258, + "grad_norm": 0.8416318893432617, + "learning_rate": 4.999796595679629e-06, + "loss": 0.7914, + "step": 129 + }, + { + "epoch": 0.10767531750414136, + "grad_norm": 0.838101863861084, + "learning_rate": 4.999782326142474e-06, + "loss": 0.7957, + "step": 130 + }, + { + "epoch": 0.10850358917725014, + "grad_norm": 0.8640785217285156, + "learning_rate": 4.999767572920741e-06, + "loss": 0.8077, + "step": 131 + }, + { + "epoch": 0.10933186085035891, + "grad_norm": 0.8207422494888306, + "learning_rate": 4.999752336017286e-06, + "loss": 0.7973, + "step": 132 + }, + { + "epoch": 0.1101601325234677, + "grad_norm": 0.8640758395195007, + "learning_rate": 4.9997366154350566e-06, + "loss": 0.8165, + "step": 133 + }, + { + "epoch": 0.11098840419657648, + "grad_norm": 0.7965807914733887, + "learning_rate": 4.999720411177094e-06, + "loss": 0.8059, + "step": 134 + }, + { + "epoch": 0.11181667586968526, + "grad_norm": 0.8463795185089111, + "learning_rate": 4.999703723246533e-06, + "loss": 0.8106, + "step": 135 + }, + { + "epoch": 0.11264494754279404, + "grad_norm": 0.8118059635162354, + "learning_rate": 4.999686551646604e-06, + "loss": 0.8099, + "step": 136 + }, + { + "epoch": 0.11347321921590281, + "grad_norm": 0.8256771564483643, + "learning_rate": 4.999668896380629e-06, + "loss": 0.8043, + "step": 137 + }, + { + "epoch": 0.11430149088901159, + "grad_norm": 0.8203126192092896, + "learning_rate": 4.999650757452025e-06, + "loss": 0.8029, + "step": 138 + }, + { + "epoch": 0.11512976256212037, + "grad_norm": 0.8442854881286621, + "learning_rate": 4.9996321348643e-06, + "loss": 0.8004, + "step": 139 + }, + { + "epoch": 0.11595803423522916, + "grad_norm": 0.8523190021514893, + "learning_rate": 4.999613028621059e-06, + "loss": 0.8092, + "step": 140 + }, + { + "epoch": 0.11678630590833794, + "grad_norm": 0.821428656578064, + "learning_rate": 4.999593438725997e-06, + "loss": 0.8105, + "step": 141 + }, + { + "epoch": 0.11761457758144671, + "grad_norm": 0.8406105041503906, + "learning_rate": 4.999573365182906e-06, + "loss": 0.7757, + "step": 142 + }, + { + "epoch": 0.11844284925455549, + "grad_norm": 0.8620600700378418, + "learning_rate": 4.999552807995669e-06, + "loss": 0.781, + "step": 143 + }, + { + "epoch": 0.11927112092766427, + "grad_norm": 0.920437216758728, + "learning_rate": 4.9995317671682655e-06, + "loss": 0.7579, + "step": 144 + }, + { + "epoch": 0.12009939260077306, + "grad_norm": 0.8414936661720276, + "learning_rate": 4.999510242704765e-06, + "loss": 0.8031, + "step": 145 + }, + { + "epoch": 0.12092766427388184, + "grad_norm": 0.8372176289558411, + "learning_rate": 4.999488234609332e-06, + "loss": 0.7931, + "step": 146 + }, + { + "epoch": 0.1217559359469906, + "grad_norm": 0.8785244226455688, + "learning_rate": 4.999465742886226e-06, + "loss": 0.8189, + "step": 147 + }, + { + "epoch": 0.12258420762009939, + "grad_norm": 0.8404353260993958, + "learning_rate": 4.999442767539799e-06, + "loss": 0.7899, + "step": 148 + }, + { + "epoch": 0.12341247929320817, + "grad_norm": 0.8541820049285889, + "learning_rate": 4.999419308574494e-06, + "loss": 0.7764, + "step": 149 + }, + { + "epoch": 0.12424075096631695, + "grad_norm": 0.8220177292823792, + "learning_rate": 4.999395365994854e-06, + "loss": 0.7942, + "step": 150 + }, + { + "epoch": 0.12506902263942574, + "grad_norm": 0.8823731541633606, + "learning_rate": 4.999370939805509e-06, + "loss": 0.793, + "step": 151 + }, + { + "epoch": 0.12589729431253452, + "grad_norm": 0.838860034942627, + "learning_rate": 4.999346030011186e-06, + "loss": 0.7741, + "step": 152 + }, + { + "epoch": 0.1267255659856433, + "grad_norm": 0.8162702322006226, + "learning_rate": 4.999320636616705e-06, + "loss": 0.7989, + "step": 153 + }, + { + "epoch": 0.12755383765875208, + "grad_norm": 0.8328465819358826, + "learning_rate": 4.99929475962698e-06, + "loss": 0.8055, + "step": 154 + }, + { + "epoch": 0.12838210933186084, + "grad_norm": 0.8373605608940125, + "learning_rate": 4.999268399047016e-06, + "loss": 0.801, + "step": 155 + }, + { + "epoch": 0.12921038100496962, + "grad_norm": 0.8408295512199402, + "learning_rate": 4.999241554881915e-06, + "loss": 0.7749, + "step": 156 + }, + { + "epoch": 0.1300386526780784, + "grad_norm": 0.8326342701911926, + "learning_rate": 4.999214227136871e-06, + "loss": 0.7818, + "step": 157 + }, + { + "epoch": 0.13086692435118719, + "grad_norm": 0.8064060211181641, + "learning_rate": 4.99918641581717e-06, + "loss": 0.7755, + "step": 158 + }, + { + "epoch": 0.13169519602429597, + "grad_norm": 0.8274610042572021, + "learning_rate": 4.999158120928196e-06, + "loss": 0.8008, + "step": 159 + }, + { + "epoch": 0.13252346769740475, + "grad_norm": 0.8121576309204102, + "learning_rate": 4.999129342475422e-06, + "loss": 0.7713, + "step": 160 + }, + { + "epoch": 0.13335173937051353, + "grad_norm": 0.8227710723876953, + "learning_rate": 4.9991000804644176e-06, + "loss": 0.7902, + "step": 161 + }, + { + "epoch": 0.13418001104362232, + "grad_norm": 0.8437026739120483, + "learning_rate": 4.999070334900843e-06, + "loss": 0.8176, + "step": 162 + }, + { + "epoch": 0.1350082827167311, + "grad_norm": 0.8379645943641663, + "learning_rate": 4.999040105790455e-06, + "loss": 0.7973, + "step": 163 + }, + { + "epoch": 0.13583655438983988, + "grad_norm": 0.822542130947113, + "learning_rate": 4.999009393139103e-06, + "loss": 0.7955, + "step": 164 + }, + { + "epoch": 0.13666482606294864, + "grad_norm": 0.8141928315162659, + "learning_rate": 4.9989781969527274e-06, + "loss": 0.7766, + "step": 165 + }, + { + "epoch": 0.13749309773605742, + "grad_norm": 0.8325410485267639, + "learning_rate": 4.998946517237367e-06, + "loss": 0.7737, + "step": 166 + }, + { + "epoch": 0.1383213694091662, + "grad_norm": 0.8399263024330139, + "learning_rate": 4.99891435399915e-06, + "loss": 0.7746, + "step": 167 + }, + { + "epoch": 0.13914964108227498, + "grad_norm": 0.8694303035736084, + "learning_rate": 4.9988817072443e-06, + "loss": 0.776, + "step": 168 + }, + { + "epoch": 0.13997791275538377, + "grad_norm": 0.8650498390197754, + "learning_rate": 4.998848576979135e-06, + "loss": 0.7941, + "step": 169 + }, + { + "epoch": 0.14080618442849255, + "grad_norm": 0.8246620893478394, + "learning_rate": 4.998814963210062e-06, + "loss": 0.7604, + "step": 170 + }, + { + "epoch": 0.14163445610160133, + "grad_norm": 0.8137409090995789, + "learning_rate": 4.99878086594359e-06, + "loss": 0.7681, + "step": 171 + }, + { + "epoch": 0.1424627277747101, + "grad_norm": 0.9755353927612305, + "learning_rate": 4.998746285186312e-06, + "loss": 0.7887, + "step": 172 + }, + { + "epoch": 0.1432909994478189, + "grad_norm": 0.8673408031463623, + "learning_rate": 4.998711220944921e-06, + "loss": 0.7803, + "step": 173 + }, + { + "epoch": 0.14411927112092768, + "grad_norm": 0.8312270045280457, + "learning_rate": 4.998675673226202e-06, + "loss": 0.7709, + "step": 174 + }, + { + "epoch": 0.14494754279403643, + "grad_norm": 0.8126094341278076, + "learning_rate": 4.998639642037032e-06, + "loss": 0.7822, + "step": 175 + }, + { + "epoch": 0.14577581446714521, + "grad_norm": 0.8561585545539856, + "learning_rate": 4.9986031273843834e-06, + "loss": 0.7829, + "step": 176 + }, + { + "epoch": 0.146604086140254, + "grad_norm": 0.8787612915039062, + "learning_rate": 4.9985661292753214e-06, + "loss": 0.7913, + "step": 177 + }, + { + "epoch": 0.14743235781336278, + "grad_norm": 0.8798345327377319, + "learning_rate": 4.998528647717006e-06, + "loss": 0.797, + "step": 178 + }, + { + "epoch": 0.14826062948647156, + "grad_norm": 0.7956589460372925, + "learning_rate": 4.998490682716687e-06, + "loss": 0.7571, + "step": 179 + }, + { + "epoch": 0.14908890115958034, + "grad_norm": 0.8505003452301025, + "learning_rate": 4.998452234281712e-06, + "loss": 0.7736, + "step": 180 + }, + { + "epoch": 0.14991717283268913, + "grad_norm": 0.8857074975967407, + "learning_rate": 4.998413302419519e-06, + "loss": 0.7863, + "step": 181 + }, + { + "epoch": 0.1507454445057979, + "grad_norm": 0.8762485980987549, + "learning_rate": 4.998373887137642e-06, + "loss": 0.7588, + "step": 182 + }, + { + "epoch": 0.1515737161789067, + "grad_norm": 0.8497205972671509, + "learning_rate": 4.998333988443709e-06, + "loss": 0.7773, + "step": 183 + }, + { + "epoch": 0.15240198785201547, + "grad_norm": 0.8720596432685852, + "learning_rate": 4.998293606345437e-06, + "loss": 0.7834, + "step": 184 + }, + { + "epoch": 0.15323025952512423, + "grad_norm": 0.8567108511924744, + "learning_rate": 4.998252740850641e-06, + "loss": 0.7794, + "step": 185 + }, + { + "epoch": 0.154058531198233, + "grad_norm": 0.8227219581604004, + "learning_rate": 4.998211391967228e-06, + "loss": 0.7688, + "step": 186 + }, + { + "epoch": 0.1548868028713418, + "grad_norm": 0.8321112394332886, + "learning_rate": 4.998169559703199e-06, + "loss": 0.7747, + "step": 187 + }, + { + "epoch": 0.15571507454445058, + "grad_norm": 0.8556115627288818, + "learning_rate": 4.9981272440666486e-06, + "loss": 0.7916, + "step": 188 + }, + { + "epoch": 0.15654334621755936, + "grad_norm": 0.8473951816558838, + "learning_rate": 4.998084445065763e-06, + "loss": 0.7823, + "step": 189 + }, + { + "epoch": 0.15737161789066814, + "grad_norm": 0.8551077246665955, + "learning_rate": 4.998041162708823e-06, + "loss": 0.7816, + "step": 190 + }, + { + "epoch": 0.15819988956377692, + "grad_norm": 0.945099413394928, + "learning_rate": 4.997997397004206e-06, + "loss": 0.7828, + "step": 191 + }, + { + "epoch": 0.1590281612368857, + "grad_norm": 0.8747237920761108, + "learning_rate": 4.9979531479603785e-06, + "loss": 0.7559, + "step": 192 + }, + { + "epoch": 0.1598564329099945, + "grad_norm": 0.8479958176612854, + "learning_rate": 4.997908415585902e-06, + "loss": 0.7694, + "step": 193 + }, + { + "epoch": 0.16068470458310327, + "grad_norm": 0.8659047484397888, + "learning_rate": 4.997863199889433e-06, + "loss": 0.7534, + "step": 194 + }, + { + "epoch": 0.16151297625621203, + "grad_norm": 0.8744727969169617, + "learning_rate": 4.997817500879719e-06, + "loss": 0.7499, + "step": 195 + }, + { + "epoch": 0.1623412479293208, + "grad_norm": 0.896468997001648, + "learning_rate": 4.997771318565603e-06, + "loss": 0.7974, + "step": 196 + }, + { + "epoch": 0.1631695196024296, + "grad_norm": 0.9058924913406372, + "learning_rate": 4.997724652956021e-06, + "loss": 0.7753, + "step": 197 + }, + { + "epoch": 0.16399779127553837, + "grad_norm": 0.8904663920402527, + "learning_rate": 4.9976775040600014e-06, + "loss": 0.7569, + "step": 198 + }, + { + "epoch": 0.16482606294864716, + "grad_norm": 0.8709238767623901, + "learning_rate": 4.997629871886669e-06, + "loss": 0.763, + "step": 199 + }, + { + "epoch": 0.16565433462175594, + "grad_norm": 0.8400478959083557, + "learning_rate": 4.997581756445238e-06, + "loss": 0.7795, + "step": 200 + }, + { + "epoch": 0.16648260629486472, + "grad_norm": 0.8458966612815857, + "learning_rate": 4.9975331577450204e-06, + "loss": 0.7597, + "step": 201 + }, + { + "epoch": 0.1673108779679735, + "grad_norm": 0.8707828521728516, + "learning_rate": 4.997484075795417e-06, + "loss": 0.759, + "step": 202 + }, + { + "epoch": 0.16813914964108229, + "grad_norm": 0.8774807453155518, + "learning_rate": 4.997434510605927e-06, + "loss": 0.7706, + "step": 203 + }, + { + "epoch": 0.16896742131419104, + "grad_norm": 0.8987929821014404, + "learning_rate": 4.99738446218614e-06, + "loss": 0.7319, + "step": 204 + }, + { + "epoch": 0.16979569298729982, + "grad_norm": 0.8869186639785767, + "learning_rate": 4.99733393054574e-06, + "loss": 0.7812, + "step": 205 + }, + { + "epoch": 0.1706239646604086, + "grad_norm": 0.860407829284668, + "learning_rate": 4.997282915694505e-06, + "loss": 0.7524, + "step": 206 + }, + { + "epoch": 0.1714522363335174, + "grad_norm": 0.8719788789749146, + "learning_rate": 4.9972314176423035e-06, + "loss": 0.7796, + "step": 207 + }, + { + "epoch": 0.17228050800662617, + "grad_norm": 0.8596137762069702, + "learning_rate": 4.997179436399103e-06, + "loss": 0.7554, + "step": 208 + }, + { + "epoch": 0.17310877967973495, + "grad_norm": 0.8505175709724426, + "learning_rate": 4.997126971974959e-06, + "loss": 0.7607, + "step": 209 + }, + { + "epoch": 0.17393705135284374, + "grad_norm": 0.7858585119247437, + "learning_rate": 4.9970740243800245e-06, + "loss": 0.7622, + "step": 210 + }, + { + "epoch": 0.17476532302595252, + "grad_norm": 0.8634621500968933, + "learning_rate": 4.997020593624543e-06, + "loss": 0.7659, + "step": 211 + }, + { + "epoch": 0.1755935946990613, + "grad_norm": 0.8650951385498047, + "learning_rate": 4.9969666797188545e-06, + "loss": 0.7689, + "step": 212 + }, + { + "epoch": 0.17642186637217008, + "grad_norm": 0.8602325916290283, + "learning_rate": 4.996912282673389e-06, + "loss": 0.7695, + "step": 213 + }, + { + "epoch": 0.17725013804527884, + "grad_norm": 0.8453742265701294, + "learning_rate": 4.9968574024986735e-06, + "loss": 0.7434, + "step": 214 + }, + { + "epoch": 0.17807840971838762, + "grad_norm": 0.8542691469192505, + "learning_rate": 4.9968020392053255e-06, + "loss": 0.7298, + "step": 215 + }, + { + "epoch": 0.1789066813914964, + "grad_norm": 0.8806673288345337, + "learning_rate": 4.996746192804058e-06, + "loss": 0.7649, + "step": 216 + }, + { + "epoch": 0.17973495306460519, + "grad_norm": 0.8451183438301086, + "learning_rate": 4.9966898633056765e-06, + "loss": 0.7681, + "step": 217 + }, + { + "epoch": 0.18056322473771397, + "grad_norm": 0.8664626479148865, + "learning_rate": 4.99663305072108e-06, + "loss": 0.7753, + "step": 218 + }, + { + "epoch": 0.18139149641082275, + "grad_norm": 0.8828144669532776, + "learning_rate": 4.996575755061262e-06, + "loss": 0.7514, + "step": 219 + }, + { + "epoch": 0.18221976808393153, + "grad_norm": 0.8493948578834534, + "learning_rate": 4.996517976337308e-06, + "loss": 0.7684, + "step": 220 + }, + { + "epoch": 0.18304803975704032, + "grad_norm": 0.8561757802963257, + "learning_rate": 4.9964597145603975e-06, + "loss": 0.7723, + "step": 221 + }, + { + "epoch": 0.1838763114301491, + "grad_norm": 0.8863773941993713, + "learning_rate": 4.996400969741805e-06, + "loss": 0.7732, + "step": 222 + }, + { + "epoch": 0.18470458310325788, + "grad_norm": 0.8571644425392151, + "learning_rate": 4.996341741892895e-06, + "loss": 0.744, + "step": 223 + }, + { + "epoch": 0.18553285477636663, + "grad_norm": 0.8622035980224609, + "learning_rate": 4.9962820310251286e-06, + "loss": 0.7607, + "step": 224 + }, + { + "epoch": 0.18636112644947542, + "grad_norm": 0.8904410600662231, + "learning_rate": 4.9962218371500595e-06, + "loss": 0.7648, + "step": 225 + }, + { + "epoch": 0.1871893981225842, + "grad_norm": 0.8975846171379089, + "learning_rate": 4.996161160279335e-06, + "loss": 0.7469, + "step": 226 + }, + { + "epoch": 0.18801766979569298, + "grad_norm": 0.8788724541664124, + "learning_rate": 4.996100000424694e-06, + "loss": 0.7473, + "step": 227 + }, + { + "epoch": 0.18884594146880176, + "grad_norm": 0.9446991086006165, + "learning_rate": 4.996038357597971e-06, + "loss": 0.7635, + "step": 228 + }, + { + "epoch": 0.18967421314191055, + "grad_norm": 0.86910480260849, + "learning_rate": 4.995976231811094e-06, + "loss": 0.7539, + "step": 229 + }, + { + "epoch": 0.19050248481501933, + "grad_norm": 0.862197756767273, + "learning_rate": 4.995913623076084e-06, + "loss": 0.7546, + "step": 230 + }, + { + "epoch": 0.1913307564881281, + "grad_norm": 0.906878650188446, + "learning_rate": 4.995850531405054e-06, + "loss": 0.7737, + "step": 231 + }, + { + "epoch": 0.1921590281612369, + "grad_norm": 0.8557183742523193, + "learning_rate": 4.995786956810212e-06, + "loss": 0.7663, + "step": 232 + }, + { + "epoch": 0.19298729983434568, + "grad_norm": 0.8125383853912354, + "learning_rate": 4.995722899303859e-06, + "loss": 0.7647, + "step": 233 + }, + { + "epoch": 0.19381557150745443, + "grad_norm": 0.8610332012176514, + "learning_rate": 4.995658358898391e-06, + "loss": 0.7541, + "step": 234 + }, + { + "epoch": 0.19464384318056321, + "grad_norm": 0.8620829582214355, + "learning_rate": 4.995593335606294e-06, + "loss": 0.7625, + "step": 235 + }, + { + "epoch": 0.195472114853672, + "grad_norm": 0.8296957612037659, + "learning_rate": 4.9955278294401496e-06, + "loss": 0.7536, + "step": 236 + }, + { + "epoch": 0.19630038652678078, + "grad_norm": 0.8974814414978027, + "learning_rate": 4.995461840412634e-06, + "loss": 0.7651, + "step": 237 + }, + { + "epoch": 0.19712865819988956, + "grad_norm": 0.8701095581054688, + "learning_rate": 4.995395368536515e-06, + "loss": 0.788, + "step": 238 + }, + { + "epoch": 0.19795692987299834, + "grad_norm": 0.8556636571884155, + "learning_rate": 4.995328413824653e-06, + "loss": 0.7649, + "step": 239 + }, + { + "epoch": 0.19878520154610713, + "grad_norm": 0.836686909198761, + "learning_rate": 4.995260976290005e-06, + "loss": 0.7583, + "step": 240 + }, + { + "epoch": 0.1996134732192159, + "grad_norm": 0.8807066082954407, + "learning_rate": 4.995193055945618e-06, + "loss": 0.7766, + "step": 241 + }, + { + "epoch": 0.2004417448923247, + "grad_norm": 0.8363582491874695, + "learning_rate": 4.995124652804635e-06, + "loss": 0.7655, + "step": 242 + }, + { + "epoch": 0.20127001656543347, + "grad_norm": 0.8699461817741394, + "learning_rate": 4.9950557668802905e-06, + "loss": 0.743, + "step": 243 + }, + { + "epoch": 0.20209828823854223, + "grad_norm": 0.8571060299873352, + "learning_rate": 4.994986398185914e-06, + "loss": 0.7512, + "step": 244 + }, + { + "epoch": 0.202926559911651, + "grad_norm": 0.8429675102233887, + "learning_rate": 4.994916546734927e-06, + "loss": 0.7589, + "step": 245 + }, + { + "epoch": 0.2037548315847598, + "grad_norm": 0.874822199344635, + "learning_rate": 4.994846212540846e-06, + "loss": 0.7534, + "step": 246 + }, + { + "epoch": 0.20458310325786858, + "grad_norm": 0.878341555595398, + "learning_rate": 4.994775395617279e-06, + "loss": 0.7709, + "step": 247 + }, + { + "epoch": 0.20541137493097736, + "grad_norm": 0.897961437702179, + "learning_rate": 4.99470409597793e-06, + "loss": 0.7768, + "step": 248 + }, + { + "epoch": 0.20623964660408614, + "grad_norm": 0.8439584374427795, + "learning_rate": 4.994632313636593e-06, + "loss": 0.7558, + "step": 249 + }, + { + "epoch": 0.20706791827719492, + "grad_norm": 0.8394468426704407, + "learning_rate": 4.994560048607159e-06, + "loss": 0.7492, + "step": 250 + }, + { + "epoch": 0.2078961899503037, + "grad_norm": 0.8489627242088318, + "learning_rate": 4.9944873009036074e-06, + "loss": 0.7365, + "step": 251 + }, + { + "epoch": 0.2087244616234125, + "grad_norm": 0.8675844669342041, + "learning_rate": 4.994414070540018e-06, + "loss": 0.762, + "step": 252 + }, + { + "epoch": 0.20955273329652127, + "grad_norm": 0.8228297829627991, + "learning_rate": 4.994340357530558e-06, + "loss": 0.7475, + "step": 253 + }, + { + "epoch": 0.21038100496963003, + "grad_norm": 0.8794129490852356, + "learning_rate": 4.9942661618894915e-06, + "loss": 0.7631, + "step": 254 + }, + { + "epoch": 0.2112092766427388, + "grad_norm": 0.8616135716438293, + "learning_rate": 4.9941914836311735e-06, + "loss": 0.7749, + "step": 255 + }, + { + "epoch": 0.2120375483158476, + "grad_norm": 0.8470495939254761, + "learning_rate": 4.994116322770054e-06, + "loss": 0.7484, + "step": 256 + }, + { + "epoch": 0.21286581998895637, + "grad_norm": 0.9346716403961182, + "learning_rate": 4.9940406793206755e-06, + "loss": 0.7557, + "step": 257 + }, + { + "epoch": 0.21369409166206516, + "grad_norm": 0.8993160724639893, + "learning_rate": 4.993964553297676e-06, + "loss": 0.7422, + "step": 258 + }, + { + "epoch": 0.21452236333517394, + "grad_norm": 0.8257086277008057, + "learning_rate": 4.993887944715783e-06, + "loss": 0.7783, + "step": 259 + }, + { + "epoch": 0.21535063500828272, + "grad_norm": 0.8950703144073486, + "learning_rate": 4.993810853589819e-06, + "loss": 0.7311, + "step": 260 + }, + { + "epoch": 0.2161789066813915, + "grad_norm": 0.8733702898025513, + "learning_rate": 4.993733279934704e-06, + "loss": 0.7572, + "step": 261 + }, + { + "epoch": 0.21700717835450029, + "grad_norm": 0.8322939276695251, + "learning_rate": 4.993655223765444e-06, + "loss": 0.7632, + "step": 262 + }, + { + "epoch": 0.21783545002760907, + "grad_norm": 0.8666137456893921, + "learning_rate": 4.993576685097145e-06, + "loss": 0.7619, + "step": 263 + }, + { + "epoch": 0.21866372170071782, + "grad_norm": 0.8770918250083923, + "learning_rate": 4.993497663945002e-06, + "loss": 0.7538, + "step": 264 + }, + { + "epoch": 0.2194919933738266, + "grad_norm": 0.8940248489379883, + "learning_rate": 4.9934181603243045e-06, + "loss": 0.7512, + "step": 265 + }, + { + "epoch": 0.2203202650469354, + "grad_norm": 0.8979714512825012, + "learning_rate": 4.993338174250437e-06, + "loss": 0.7504, + "step": 266 + }, + { + "epoch": 0.22114853672004417, + "grad_norm": 0.8802001476287842, + "learning_rate": 4.993257705738876e-06, + "loss": 0.7501, + "step": 267 + }, + { + "epoch": 0.22197680839315295, + "grad_norm": 0.8734432458877563, + "learning_rate": 4.993176754805189e-06, + "loss": 0.7452, + "step": 268 + }, + { + "epoch": 0.22280508006626173, + "grad_norm": 0.9127711057662964, + "learning_rate": 4.993095321465042e-06, + "loss": 0.7404, + "step": 269 + }, + { + "epoch": 0.22363335173937052, + "grad_norm": 0.8533952832221985, + "learning_rate": 4.993013405734191e-06, + "loss": 0.7382, + "step": 270 + }, + { + "epoch": 0.2244616234124793, + "grad_norm": 0.8463877439498901, + "learning_rate": 4.9929310076284845e-06, + "loss": 0.7574, + "step": 271 + }, + { + "epoch": 0.22528989508558808, + "grad_norm": 0.8598008155822754, + "learning_rate": 4.992848127163868e-06, + "loss": 0.7624, + "step": 272 + }, + { + "epoch": 0.22611816675869686, + "grad_norm": 0.8759154081344604, + "learning_rate": 4.992764764356376e-06, + "loss": 0.7637, + "step": 273 + }, + { + "epoch": 0.22694643843180562, + "grad_norm": 0.8880444765090942, + "learning_rate": 4.992680919222141e-06, + "loss": 0.7514, + "step": 274 + }, + { + "epoch": 0.2277747101049144, + "grad_norm": 0.871366024017334, + "learning_rate": 4.9925965917773826e-06, + "loss": 0.7569, + "step": 275 + }, + { + "epoch": 0.22860298177802318, + "grad_norm": 0.862654983997345, + "learning_rate": 4.99251178203842e-06, + "loss": 0.7349, + "step": 276 + }, + { + "epoch": 0.22943125345113197, + "grad_norm": 0.8615298867225647, + "learning_rate": 4.992426490021662e-06, + "loss": 0.7438, + "step": 277 + }, + { + "epoch": 0.23025952512424075, + "grad_norm": 0.8735869526863098, + "learning_rate": 4.992340715743614e-06, + "loss": 0.7755, + "step": 278 + }, + { + "epoch": 0.23108779679734953, + "grad_norm": 0.9082722067832947, + "learning_rate": 4.9922544592208695e-06, + "loss": 0.7561, + "step": 279 + }, + { + "epoch": 0.23191606847045831, + "grad_norm": 0.8639047145843506, + "learning_rate": 4.992167720470119e-06, + "loss": 0.7493, + "step": 280 + }, + { + "epoch": 0.2327443401435671, + "grad_norm": 0.891640305519104, + "learning_rate": 4.9920804995081475e-06, + "loss": 0.75, + "step": 281 + }, + { + "epoch": 0.23357261181667588, + "grad_norm": 0.8880137205123901, + "learning_rate": 4.99199279635183e-06, + "loss": 0.7801, + "step": 282 + }, + { + "epoch": 0.23440088348978466, + "grad_norm": 0.8600473403930664, + "learning_rate": 4.991904611018137e-06, + "loss": 0.7477, + "step": 283 + }, + { + "epoch": 0.23522915516289342, + "grad_norm": 0.8685399889945984, + "learning_rate": 4.99181594352413e-06, + "loss": 0.741, + "step": 284 + }, + { + "epoch": 0.2360574268360022, + "grad_norm": 0.9006181955337524, + "learning_rate": 4.991726793886966e-06, + "loss": 0.7288, + "step": 285 + }, + { + "epoch": 0.23688569850911098, + "grad_norm": 0.863895833492279, + "learning_rate": 4.991637162123896e-06, + "loss": 0.7375, + "step": 286 + }, + { + "epoch": 0.23771397018221976, + "grad_norm": 0.8620381951332092, + "learning_rate": 4.9915470482522625e-06, + "loss": 0.7249, + "step": 287 + }, + { + "epoch": 0.23854224185532855, + "grad_norm": 0.9634964466094971, + "learning_rate": 4.991456452289499e-06, + "loss": 0.7313, + "step": 288 + }, + { + "epoch": 0.23937051352843733, + "grad_norm": 0.9154514670372009, + "learning_rate": 4.9913653742531385e-06, + "loss": 0.782, + "step": 289 + }, + { + "epoch": 0.2401987852015461, + "grad_norm": 0.9179158806800842, + "learning_rate": 4.991273814160802e-06, + "loss": 0.7388, + "step": 290 + }, + { + "epoch": 0.2410270568746549, + "grad_norm": 0.8609046936035156, + "learning_rate": 4.991181772030206e-06, + "loss": 0.7393, + "step": 291 + }, + { + "epoch": 0.24185532854776368, + "grad_norm": 0.8689826130867004, + "learning_rate": 4.991089247879159e-06, + "loss": 0.7302, + "step": 292 + }, + { + "epoch": 0.24268360022087246, + "grad_norm": 0.8414047956466675, + "learning_rate": 4.990996241725565e-06, + "loss": 0.7496, + "step": 293 + }, + { + "epoch": 0.2435118718939812, + "grad_norm": 0.8621289134025574, + "learning_rate": 4.990902753587418e-06, + "loss": 0.7317, + "step": 294 + }, + { + "epoch": 0.24434014356709, + "grad_norm": 0.856566309928894, + "learning_rate": 4.990808783482809e-06, + "loss": 0.7169, + "step": 295 + }, + { + "epoch": 0.24516841524019878, + "grad_norm": 0.8549672961235046, + "learning_rate": 4.990714331429919e-06, + "loss": 0.7357, + "step": 296 + }, + { + "epoch": 0.24599668691330756, + "grad_norm": 0.8813024759292603, + "learning_rate": 4.990619397447024e-06, + "loss": 0.7437, + "step": 297 + }, + { + "epoch": 0.24682495858641634, + "grad_norm": 0.9008268713951111, + "learning_rate": 4.990523981552493e-06, + "loss": 0.7652, + "step": 298 + }, + { + "epoch": 0.24765323025952513, + "grad_norm": 0.910277783870697, + "learning_rate": 4.990428083764788e-06, + "loss": 0.7598, + "step": 299 + }, + { + "epoch": 0.2484815019326339, + "grad_norm": 0.9261599183082581, + "learning_rate": 4.990331704102464e-06, + "loss": 0.7199, + "step": 300 + }, + { + "epoch": 0.2493097736057427, + "grad_norm": 0.9166674613952637, + "learning_rate": 4.99023484258417e-06, + "loss": 0.7414, + "step": 301 + }, + { + "epoch": 0.2501380452788515, + "grad_norm": 0.8701214790344238, + "learning_rate": 4.990137499228648e-06, + "loss": 0.7326, + "step": 302 + }, + { + "epoch": 0.25096631695196026, + "grad_norm": 0.9120466113090515, + "learning_rate": 4.990039674054732e-06, + "loss": 0.7574, + "step": 303 + }, + { + "epoch": 0.25179458862506904, + "grad_norm": 0.9449286460876465, + "learning_rate": 4.989941367081351e-06, + "loss": 0.725, + "step": 304 + }, + { + "epoch": 0.2526228602981778, + "grad_norm": 0.8989110589027405, + "learning_rate": 4.989842578327526e-06, + "loss": 0.7569, + "step": 305 + }, + { + "epoch": 0.2534511319712866, + "grad_norm": 0.891677975654602, + "learning_rate": 4.989743307812373e-06, + "loss": 0.7135, + "step": 306 + }, + { + "epoch": 0.2542794036443954, + "grad_norm": 0.8689535856246948, + "learning_rate": 4.989643555555099e-06, + "loss": 0.7439, + "step": 307 + }, + { + "epoch": 0.25510767531750417, + "grad_norm": 0.8779630064964294, + "learning_rate": 4.989543321575004e-06, + "loss": 0.7322, + "step": 308 + }, + { + "epoch": 0.2559359469906129, + "grad_norm": 0.8977622985839844, + "learning_rate": 4.9894426058914856e-06, + "loss": 0.7544, + "step": 309 + }, + { + "epoch": 0.2567642186637217, + "grad_norm": 0.8216757774353027, + "learning_rate": 4.989341408524028e-06, + "loss": 0.7344, + "step": 310 + }, + { + "epoch": 0.25759249033683046, + "grad_norm": 0.8494759798049927, + "learning_rate": 4.9892397294922126e-06, + "loss": 0.7389, + "step": 311 + }, + { + "epoch": 0.25842076200993924, + "grad_norm": 0.881244957447052, + "learning_rate": 4.989137568815715e-06, + "loss": 0.7121, + "step": 312 + }, + { + "epoch": 0.259249033683048, + "grad_norm": 0.8679410219192505, + "learning_rate": 4.9890349265143005e-06, + "loss": 0.7469, + "step": 313 + }, + { + "epoch": 0.2600773053561568, + "grad_norm": 0.9746971130371094, + "learning_rate": 4.988931802607831e-06, + "loss": 0.7585, + "step": 314 + }, + { + "epoch": 0.2609055770292656, + "grad_norm": 0.9222471117973328, + "learning_rate": 4.988828197116259e-06, + "loss": 0.761, + "step": 315 + }, + { + "epoch": 0.26173384870237437, + "grad_norm": 0.9104124903678894, + "learning_rate": 4.988724110059631e-06, + "loss": 0.7652, + "step": 316 + }, + { + "epoch": 0.26256212037548315, + "grad_norm": 0.8876475095748901, + "learning_rate": 4.988619541458088e-06, + "loss": 0.7529, + "step": 317 + }, + { + "epoch": 0.26339039204859194, + "grad_norm": 0.8705421686172485, + "learning_rate": 4.988514491331861e-06, + "loss": 0.7437, + "step": 318 + }, + { + "epoch": 0.2642186637217007, + "grad_norm": 0.8594399690628052, + "learning_rate": 4.9884089597012784e-06, + "loss": 0.7406, + "step": 319 + }, + { + "epoch": 0.2650469353948095, + "grad_norm": 0.8807039856910706, + "learning_rate": 4.988302946586759e-06, + "loss": 0.7429, + "step": 320 + }, + { + "epoch": 0.2658752070679183, + "grad_norm": 0.8881412148475647, + "learning_rate": 4.9881964520088144e-06, + "loss": 0.7234, + "step": 321 + }, + { + "epoch": 0.26670347874102707, + "grad_norm": 0.8764941096305847, + "learning_rate": 4.988089475988052e-06, + "loss": 0.7514, + "step": 322 + }, + { + "epoch": 0.26753175041413585, + "grad_norm": 0.8992927670478821, + "learning_rate": 4.987982018545169e-06, + "loss": 0.7143, + "step": 323 + }, + { + "epoch": 0.26836002208724463, + "grad_norm": 0.8961546421051025, + "learning_rate": 4.9878740797009585e-06, + "loss": 0.7396, + "step": 324 + }, + { + "epoch": 0.2691882937603534, + "grad_norm": 0.9031578302383423, + "learning_rate": 4.987765659476304e-06, + "loss": 0.7407, + "step": 325 + }, + { + "epoch": 0.2700165654334622, + "grad_norm": 0.8693655729293823, + "learning_rate": 4.987656757892186e-06, + "loss": 0.7215, + "step": 326 + }, + { + "epoch": 0.270844837106571, + "grad_norm": 0.8739856481552124, + "learning_rate": 4.987547374969676e-06, + "loss": 0.7471, + "step": 327 + }, + { + "epoch": 0.27167310877967976, + "grad_norm": 0.8916322588920593, + "learning_rate": 4.987437510729936e-06, + "loss": 0.7459, + "step": 328 + }, + { + "epoch": 0.2725013804527885, + "grad_norm": 0.890178382396698, + "learning_rate": 4.987327165194225e-06, + "loss": 0.7322, + "step": 329 + }, + { + "epoch": 0.27332965212589727, + "grad_norm": 0.8611598014831543, + "learning_rate": 4.987216338383896e-06, + "loss": 0.733, + "step": 330 + }, + { + "epoch": 0.27415792379900605, + "grad_norm": 0.87791907787323, + "learning_rate": 4.98710503032039e-06, + "loss": 0.7445, + "step": 331 + }, + { + "epoch": 0.27498619547211484, + "grad_norm": 0.963479220867157, + "learning_rate": 4.986993241025245e-06, + "loss": 0.7376, + "step": 332 + }, + { + "epoch": 0.2758144671452236, + "grad_norm": 0.9083762764930725, + "learning_rate": 4.986880970520092e-06, + "loss": 0.7258, + "step": 333 + }, + { + "epoch": 0.2766427388183324, + "grad_norm": 1.048991084098816, + "learning_rate": 4.9867682188266524e-06, + "loss": 0.7399, + "step": 334 + }, + { + "epoch": 0.2774710104914412, + "grad_norm": 0.9404491782188416, + "learning_rate": 4.986654985966745e-06, + "loss": 0.7133, + "step": 335 + }, + { + "epoch": 0.27829928216454997, + "grad_norm": 0.9168346524238586, + "learning_rate": 4.986541271962278e-06, + "loss": 0.74, + "step": 336 + }, + { + "epoch": 0.27912755383765875, + "grad_norm": 0.8716689944267273, + "learning_rate": 4.986427076835253e-06, + "loss": 0.7249, + "step": 337 + }, + { + "epoch": 0.27995582551076753, + "grad_norm": 0.9097291827201843, + "learning_rate": 4.986312400607767e-06, + "loss": 0.7151, + "step": 338 + }, + { + "epoch": 0.2807840971838763, + "grad_norm": 0.8621960282325745, + "learning_rate": 4.98619724330201e-06, + "loss": 0.7125, + "step": 339 + }, + { + "epoch": 0.2816123688569851, + "grad_norm": 0.9823791980743408, + "learning_rate": 4.986081604940261e-06, + "loss": 0.7191, + "step": 340 + }, + { + "epoch": 0.2824406405300939, + "grad_norm": 0.8999047875404358, + "learning_rate": 4.9859654855448966e-06, + "loss": 0.7175, + "step": 341 + }, + { + "epoch": 0.28326891220320266, + "grad_norm": 0.8690859079360962, + "learning_rate": 4.985848885138386e-06, + "loss": 0.7309, + "step": 342 + }, + { + "epoch": 0.28409718387631144, + "grad_norm": 0.889500617980957, + "learning_rate": 4.985731803743287e-06, + "loss": 0.7391, + "step": 343 + }, + { + "epoch": 0.2849254555494202, + "grad_norm": 0.8992016911506653, + "learning_rate": 4.985614241382257e-06, + "loss": 0.7437, + "step": 344 + }, + { + "epoch": 0.285753727222529, + "grad_norm": 0.9100068211555481, + "learning_rate": 4.985496198078041e-06, + "loss": 0.722, + "step": 345 + }, + { + "epoch": 0.2865819988956378, + "grad_norm": 0.9013107419013977, + "learning_rate": 4.985377673853481e-06, + "loss": 0.7434, + "step": 346 + }, + { + "epoch": 0.2874102705687466, + "grad_norm": 0.8794704675674438, + "learning_rate": 4.9852586687315094e-06, + "loss": 0.7387, + "step": 347 + }, + { + "epoch": 0.28823854224185536, + "grad_norm": 0.9288201928138733, + "learning_rate": 4.9851391827351526e-06, + "loss": 0.7165, + "step": 348 + }, + { + "epoch": 0.2890668139149641, + "grad_norm": 0.8735847473144531, + "learning_rate": 4.98501921588753e-06, + "loss": 0.7479, + "step": 349 + }, + { + "epoch": 0.28989508558807286, + "grad_norm": 0.890616238117218, + "learning_rate": 4.984898768211853e-06, + "loss": 0.76, + "step": 350 + }, + { + "epoch": 0.29072335726118165, + "grad_norm": 0.9181028604507446, + "learning_rate": 4.98477783973143e-06, + "loss": 0.7366, + "step": 351 + }, + { + "epoch": 0.29155162893429043, + "grad_norm": 0.9947429299354553, + "learning_rate": 4.984656430469657e-06, + "loss": 0.7199, + "step": 352 + }, + { + "epoch": 0.2923799006073992, + "grad_norm": 0.8730769157409668, + "learning_rate": 4.984534540450027e-06, + "loss": 0.7153, + "step": 353 + }, + { + "epoch": 0.293208172280508, + "grad_norm": 0.8675997853279114, + "learning_rate": 4.9844121696961235e-06, + "loss": 0.7196, + "step": 354 + }, + { + "epoch": 0.2940364439536168, + "grad_norm": 0.9558542966842651, + "learning_rate": 4.984289318231624e-06, + "loss": 0.7311, + "step": 355 + }, + { + "epoch": 0.29486471562672556, + "grad_norm": 0.8817659616470337, + "learning_rate": 4.9841659860803e-06, + "loss": 0.7367, + "step": 356 + }, + { + "epoch": 0.29569298729983434, + "grad_norm": 0.8679526448249817, + "learning_rate": 4.984042173266016e-06, + "loss": 0.719, + "step": 357 + }, + { + "epoch": 0.2965212589729431, + "grad_norm": 0.9102020859718323, + "learning_rate": 4.983917879812726e-06, + "loss": 0.7301, + "step": 358 + }, + { + "epoch": 0.2973495306460519, + "grad_norm": 0.8987215757369995, + "learning_rate": 4.983793105744482e-06, + "loss": 0.7557, + "step": 359 + }, + { + "epoch": 0.2981778023191607, + "grad_norm": 0.8779832720756531, + "learning_rate": 4.983667851085426e-06, + "loss": 0.7471, + "step": 360 + }, + { + "epoch": 0.29900607399226947, + "grad_norm": 0.8762708306312561, + "learning_rate": 4.983542115859792e-06, + "loss": 0.7676, + "step": 361 + }, + { + "epoch": 0.29983434566537825, + "grad_norm": 0.910353422164917, + "learning_rate": 4.98341590009191e-06, + "loss": 0.7285, + "step": 362 + }, + { + "epoch": 0.30066261733848704, + "grad_norm": 0.9385405778884888, + "learning_rate": 4.983289203806202e-06, + "loss": 0.7142, + "step": 363 + }, + { + "epoch": 0.3014908890115958, + "grad_norm": 1.016781210899353, + "learning_rate": 4.983162027027182e-06, + "loss": 0.7408, + "step": 364 + }, + { + "epoch": 0.3023191606847046, + "grad_norm": 0.8642508387565613, + "learning_rate": 4.983034369779457e-06, + "loss": 0.7397, + "step": 365 + }, + { + "epoch": 0.3031474323578134, + "grad_norm": 0.9469170570373535, + "learning_rate": 4.982906232087728e-06, + "loss": 0.7158, + "step": 366 + }, + { + "epoch": 0.30397570403092217, + "grad_norm": 0.8818212151527405, + "learning_rate": 4.982777613976789e-06, + "loss": 0.7408, + "step": 367 + }, + { + "epoch": 0.30480397570403095, + "grad_norm": 0.8639838099479675, + "learning_rate": 4.982648515471526e-06, + "loss": 0.7217, + "step": 368 + }, + { + "epoch": 0.3056322473771397, + "grad_norm": 0.923780083656311, + "learning_rate": 4.982518936596917e-06, + "loss": 0.7284, + "step": 369 + }, + { + "epoch": 0.30646051905024846, + "grad_norm": 0.8992834687232971, + "learning_rate": 4.982388877378037e-06, + "loss": 0.7395, + "step": 370 + }, + { + "epoch": 0.30728879072335724, + "grad_norm": 0.8671404123306274, + "learning_rate": 4.982258337840049e-06, + "loss": 0.7157, + "step": 371 + }, + { + "epoch": 0.308117062396466, + "grad_norm": 0.8689168691635132, + "learning_rate": 4.982127318008212e-06, + "loss": 0.7398, + "step": 372 + }, + { + "epoch": 0.3089453340695748, + "grad_norm": 0.8964621424674988, + "learning_rate": 4.9819958179078765e-06, + "loss": 0.7385, + "step": 373 + }, + { + "epoch": 0.3097736057426836, + "grad_norm": 0.8984093070030212, + "learning_rate": 4.981863837564488e-06, + "loss": 0.7253, + "step": 374 + }, + { + "epoch": 0.31060187741579237, + "grad_norm": 0.8776745200157166, + "learning_rate": 4.981731377003583e-06, + "loss": 0.7126, + "step": 375 + }, + { + "epoch": 0.31143014908890115, + "grad_norm": 0.8782477378845215, + "learning_rate": 4.98159843625079e-06, + "loss": 0.7164, + "step": 376 + }, + { + "epoch": 0.31225842076200994, + "grad_norm": 0.8993703722953796, + "learning_rate": 4.981465015331832e-06, + "loss": 0.7391, + "step": 377 + }, + { + "epoch": 0.3130866924351187, + "grad_norm": 0.8975916504859924, + "learning_rate": 4.981331114272527e-06, + "loss": 0.7385, + "step": 378 + }, + { + "epoch": 0.3139149641082275, + "grad_norm": 0.8726824522018433, + "learning_rate": 4.9811967330987816e-06, + "loss": 0.7426, + "step": 379 + }, + { + "epoch": 0.3147432357813363, + "grad_norm": 0.8741514682769775, + "learning_rate": 4.981061871836597e-06, + "loss": 0.7246, + "step": 380 + }, + { + "epoch": 0.31557150745444507, + "grad_norm": 0.8807122707366943, + "learning_rate": 4.98092653051207e-06, + "loss": 0.7364, + "step": 381 + }, + { + "epoch": 0.31639977912755385, + "grad_norm": 0.835006058216095, + "learning_rate": 4.980790709151385e-06, + "loss": 0.7248, + "step": 382 + }, + { + "epoch": 0.31722805080066263, + "grad_norm": 0.9473997354507446, + "learning_rate": 4.980654407780824e-06, + "loss": 0.7331, + "step": 383 + }, + { + "epoch": 0.3180563224737714, + "grad_norm": 0.8332815170288086, + "learning_rate": 4.980517626426758e-06, + "loss": 0.7299, + "step": 384 + }, + { + "epoch": 0.3188845941468802, + "grad_norm": 0.8536360859870911, + "learning_rate": 4.980380365115656e-06, + "loss": 0.7095, + "step": 385 + }, + { + "epoch": 0.319712865819989, + "grad_norm": 0.9488157629966736, + "learning_rate": 4.980242623874075e-06, + "loss": 0.7382, + "step": 386 + }, + { + "epoch": 0.32054113749309776, + "grad_norm": 0.8623067736625671, + "learning_rate": 4.980104402728666e-06, + "loss": 0.7187, + "step": 387 + }, + { + "epoch": 0.32136940916620654, + "grad_norm": 0.9110252857208252, + "learning_rate": 4.979965701706175e-06, + "loss": 0.7354, + "step": 388 + }, + { + "epoch": 0.32219768083931527, + "grad_norm": 0.8966642618179321, + "learning_rate": 4.979826520833439e-06, + "loss": 0.7251, + "step": 389 + }, + { + "epoch": 0.32302595251242405, + "grad_norm": 0.856674075126648, + "learning_rate": 4.979686860137387e-06, + "loss": 0.723, + "step": 390 + }, + { + "epoch": 0.32385422418553284, + "grad_norm": 0.8623953461647034, + "learning_rate": 4.979546719645043e-06, + "loss": 0.7224, + "step": 391 + }, + { + "epoch": 0.3246824958586416, + "grad_norm": 0.8685624599456787, + "learning_rate": 4.979406099383523e-06, + "loss": 0.7283, + "step": 392 + }, + { + "epoch": 0.3255107675317504, + "grad_norm": 0.8610525727272034, + "learning_rate": 4.979264999380035e-06, + "loss": 0.7385, + "step": 393 + }, + { + "epoch": 0.3263390392048592, + "grad_norm": 0.8860796093940735, + "learning_rate": 4.979123419661882e-06, + "loss": 0.743, + "step": 394 + }, + { + "epoch": 0.32716731087796797, + "grad_norm": 0.8748682737350464, + "learning_rate": 4.978981360256457e-06, + "loss": 0.7447, + "step": 395 + }, + { + "epoch": 0.32799558255107675, + "grad_norm": 0.8750934600830078, + "learning_rate": 4.978838821191247e-06, + "loss": 0.7259, + "step": 396 + }, + { + "epoch": 0.32882385422418553, + "grad_norm": 0.9335777163505554, + "learning_rate": 4.978695802493834e-06, + "loss": 0.7271, + "step": 397 + }, + { + "epoch": 0.3296521258972943, + "grad_norm": 0.8823036551475525, + "learning_rate": 4.978552304191888e-06, + "loss": 0.723, + "step": 398 + }, + { + "epoch": 0.3304803975704031, + "grad_norm": 0.9503542184829712, + "learning_rate": 4.9784083263131764e-06, + "loss": 0.752, + "step": 399 + }, + { + "epoch": 0.3313086692435119, + "grad_norm": 0.8779938220977783, + "learning_rate": 4.978263868885557e-06, + "loss": 0.7098, + "step": 400 + }, + { + "epoch": 0.33213694091662066, + "grad_norm": 0.873857855796814, + "learning_rate": 4.978118931936981e-06, + "loss": 0.7143, + "step": 401 + }, + { + "epoch": 0.33296521258972944, + "grad_norm": 0.8437919020652771, + "learning_rate": 4.9779735154954924e-06, + "loss": 0.7275, + "step": 402 + }, + { + "epoch": 0.3337934842628382, + "grad_norm": 0.8914555311203003, + "learning_rate": 4.977827619589229e-06, + "loss": 0.722, + "step": 403 + }, + { + "epoch": 0.334621755935947, + "grad_norm": 0.9062597751617432, + "learning_rate": 4.977681244246417e-06, + "loss": 0.7067, + "step": 404 + }, + { + "epoch": 0.3354500276090558, + "grad_norm": 0.8555092811584473, + "learning_rate": 4.977534389495382e-06, + "loss": 0.7318, + "step": 405 + }, + { + "epoch": 0.33627829928216457, + "grad_norm": 0.8717684745788574, + "learning_rate": 4.9773870553645386e-06, + "loss": 0.7398, + "step": 406 + }, + { + "epoch": 0.33710657095527335, + "grad_norm": 0.8746877908706665, + "learning_rate": 4.977239241882393e-06, + "loss": 0.7154, + "step": 407 + }, + { + "epoch": 0.3379348426283821, + "grad_norm": 0.8905707597732544, + "learning_rate": 4.977090949077547e-06, + "loss": 0.7345, + "step": 408 + }, + { + "epoch": 0.33876311430149086, + "grad_norm": 0.9037347435951233, + "learning_rate": 4.9769421769786935e-06, + "loss": 0.7271, + "step": 409 + }, + { + "epoch": 0.33959138597459965, + "grad_norm": 0.8588722944259644, + "learning_rate": 4.976792925614619e-06, + "loss": 0.7335, + "step": 410 + }, + { + "epoch": 0.34041965764770843, + "grad_norm": 0.8786173462867737, + "learning_rate": 4.9766431950142005e-06, + "loss": 0.7254, + "step": 411 + }, + { + "epoch": 0.3412479293208172, + "grad_norm": 0.9019079804420471, + "learning_rate": 4.976492985206412e-06, + "loss": 0.715, + "step": 412 + }, + { + "epoch": 0.342076200993926, + "grad_norm": 0.878132700920105, + "learning_rate": 4.976342296220316e-06, + "loss": 0.7063, + "step": 413 + }, + { + "epoch": 0.3429044726670348, + "grad_norm": 0.8722684979438782, + "learning_rate": 4.976191128085069e-06, + "loss": 0.7481, + "step": 414 + }, + { + "epoch": 0.34373274434014356, + "grad_norm": 0.9076220393180847, + "learning_rate": 4.976039480829922e-06, + "loss": 0.7197, + "step": 415 + }, + { + "epoch": 0.34456101601325234, + "grad_norm": 0.8674967288970947, + "learning_rate": 4.975887354484217e-06, + "loss": 0.7282, + "step": 416 + }, + { + "epoch": 0.3453892876863611, + "grad_norm": 0.9406816959381104, + "learning_rate": 4.9757347490773885e-06, + "loss": 0.7222, + "step": 417 + }, + { + "epoch": 0.3462175593594699, + "grad_norm": 0.8747051358222961, + "learning_rate": 4.975581664638965e-06, + "loss": 0.7283, + "step": 418 + }, + { + "epoch": 0.3470458310325787, + "grad_norm": 1.0587687492370605, + "learning_rate": 4.9754281011985666e-06, + "loss": 0.7272, + "step": 419 + }, + { + "epoch": 0.34787410270568747, + "grad_norm": 0.8596946597099304, + "learning_rate": 4.975274058785906e-06, + "loss": 0.7407, + "step": 420 + }, + { + "epoch": 0.34870237437879625, + "grad_norm": 0.9230688214302063, + "learning_rate": 4.975119537430789e-06, + "loss": 0.7303, + "step": 421 + }, + { + "epoch": 0.34953064605190504, + "grad_norm": 0.9276477694511414, + "learning_rate": 4.974964537163115e-06, + "loss": 0.7296, + "step": 422 + }, + { + "epoch": 0.3503589177250138, + "grad_norm": 0.8969418406486511, + "learning_rate": 4.974809058012874e-06, + "loss": 0.7275, + "step": 423 + }, + { + "epoch": 0.3511871893981226, + "grad_norm": 0.8454467058181763, + "learning_rate": 4.97465310001015e-06, + "loss": 0.7144, + "step": 424 + }, + { + "epoch": 0.3520154610712314, + "grad_norm": 0.8833656907081604, + "learning_rate": 4.97449666318512e-06, + "loss": 0.7073, + "step": 425 + }, + { + "epoch": 0.35284373274434017, + "grad_norm": 0.9251653552055359, + "learning_rate": 4.974339747568052e-06, + "loss": 0.7351, + "step": 426 + }, + { + "epoch": 0.35367200441744895, + "grad_norm": 0.8938575983047485, + "learning_rate": 4.974182353189309e-06, + "loss": 0.7031, + "step": 427 + }, + { + "epoch": 0.3545002760905577, + "grad_norm": 0.8487107753753662, + "learning_rate": 4.974024480079343e-06, + "loss": 0.704, + "step": 428 + }, + { + "epoch": 0.35532854776366646, + "grad_norm": 0.8515727519989014, + "learning_rate": 4.973866128268705e-06, + "loss": 0.7092, + "step": 429 + }, + { + "epoch": 0.35615681943677524, + "grad_norm": 0.8869542479515076, + "learning_rate": 4.97370729778803e-06, + "loss": 0.7099, + "step": 430 + }, + { + "epoch": 0.356985091109884, + "grad_norm": 0.8830757141113281, + "learning_rate": 4.973547988668055e-06, + "loss": 0.7294, + "step": 431 + }, + { + "epoch": 0.3578133627829928, + "grad_norm": 0.9321587681770325, + "learning_rate": 4.973388200939601e-06, + "loss": 0.7276, + "step": 432 + }, + { + "epoch": 0.3586416344561016, + "grad_norm": 0.9408143758773804, + "learning_rate": 4.9732279346335866e-06, + "loss": 0.7216, + "step": 433 + }, + { + "epoch": 0.35946990612921037, + "grad_norm": 0.9189308285713196, + "learning_rate": 4.973067189781022e-06, + "loss": 0.7079, + "step": 434 + }, + { + "epoch": 0.36029817780231915, + "grad_norm": 0.9198120832443237, + "learning_rate": 4.97290596641301e-06, + "loss": 0.7188, + "step": 435 + }, + { + "epoch": 0.36112644947542794, + "grad_norm": 0.9123285412788391, + "learning_rate": 4.972744264560746e-06, + "loss": 0.7328, + "step": 436 + }, + { + "epoch": 0.3619547211485367, + "grad_norm": 0.8722008466720581, + "learning_rate": 4.972582084255517e-06, + "loss": 0.7228, + "step": 437 + }, + { + "epoch": 0.3627829928216455, + "grad_norm": 1.035325288772583, + "learning_rate": 4.972419425528704e-06, + "loss": 0.72, + "step": 438 + }, + { + "epoch": 0.3636112644947543, + "grad_norm": 0.9577400088310242, + "learning_rate": 4.97225628841178e-06, + "loss": 0.7382, + "step": 439 + }, + { + "epoch": 0.36443953616786307, + "grad_norm": 0.9206854104995728, + "learning_rate": 4.972092672936309e-06, + "loss": 0.7294, + "step": 440 + }, + { + "epoch": 0.36526780784097185, + "grad_norm": 0.9281194806098938, + "learning_rate": 4.971928579133952e-06, + "loss": 0.7137, + "step": 441 + }, + { + "epoch": 0.36609607951408063, + "grad_norm": 0.951049268245697, + "learning_rate": 4.971764007036457e-06, + "loss": 0.7354, + "step": 442 + }, + { + "epoch": 0.3669243511871894, + "grad_norm": 0.8779566287994385, + "learning_rate": 4.971598956675668e-06, + "loss": 0.6858, + "step": 443 + }, + { + "epoch": 0.3677526228602982, + "grad_norm": 0.9221998453140259, + "learning_rate": 4.971433428083522e-06, + "loss": 0.7116, + "step": 444 + }, + { + "epoch": 0.368580894533407, + "grad_norm": 0.9608840346336365, + "learning_rate": 4.971267421292045e-06, + "loss": 0.7129, + "step": 445 + }, + { + "epoch": 0.36940916620651576, + "grad_norm": 0.8761363625526428, + "learning_rate": 4.971100936333359e-06, + "loss": 0.718, + "step": 446 + }, + { + "epoch": 0.37023743787962454, + "grad_norm": 0.8808798789978027, + "learning_rate": 4.970933973239677e-06, + "loss": 0.6986, + "step": 447 + }, + { + "epoch": 0.37106570955273327, + "grad_norm": 0.8818857669830322, + "learning_rate": 4.9707665320433055e-06, + "loss": 0.7465, + "step": 448 + }, + { + "epoch": 0.37189398122584205, + "grad_norm": 0.9144098162651062, + "learning_rate": 4.970598612776641e-06, + "loss": 0.7187, + "step": 449 + }, + { + "epoch": 0.37272225289895083, + "grad_norm": 0.8439158797264099, + "learning_rate": 4.9704302154721776e-06, + "loss": 0.7195, + "step": 450 + }, + { + "epoch": 0.3735505245720596, + "grad_norm": 0.8661074638366699, + "learning_rate": 4.970261340162495e-06, + "loss": 0.712, + "step": 451 + }, + { + "epoch": 0.3743787962451684, + "grad_norm": 0.8818789124488831, + "learning_rate": 4.970091986880271e-06, + "loss": 0.7199, + "step": 452 + }, + { + "epoch": 0.3752070679182772, + "grad_norm": 0.9560503363609314, + "learning_rate": 4.969922155658273e-06, + "loss": 0.7145, + "step": 453 + }, + { + "epoch": 0.37603533959138596, + "grad_norm": 0.8723775744438171, + "learning_rate": 4.969751846529363e-06, + "loss": 0.7152, + "step": 454 + }, + { + "epoch": 0.37686361126449475, + "grad_norm": 0.8612253069877625, + "learning_rate": 4.9695810595264935e-06, + "loss": 0.728, + "step": 455 + }, + { + "epoch": 0.37769188293760353, + "grad_norm": 0.8995444178581238, + "learning_rate": 4.969409794682709e-06, + "loss": 0.678, + "step": 456 + }, + { + "epoch": 0.3785201546107123, + "grad_norm": 0.8770595788955688, + "learning_rate": 4.96923805203115e-06, + "loss": 0.7354, + "step": 457 + }, + { + "epoch": 0.3793484262838211, + "grad_norm": 0.893329918384552, + "learning_rate": 4.969065831605045e-06, + "loss": 0.7102, + "step": 458 + }, + { + "epoch": 0.3801766979569299, + "grad_norm": 0.954058051109314, + "learning_rate": 4.968893133437718e-06, + "loss": 0.7144, + "step": 459 + }, + { + "epoch": 0.38100496963003866, + "grad_norm": 0.9711002707481384, + "learning_rate": 4.968719957562585e-06, + "loss": 0.7452, + "step": 460 + }, + { + "epoch": 0.38183324130314744, + "grad_norm": 0.9036396741867065, + "learning_rate": 4.968546304013153e-06, + "loss": 0.7535, + "step": 461 + }, + { + "epoch": 0.3826615129762562, + "grad_norm": 0.8948055505752563, + "learning_rate": 4.968372172823023e-06, + "loss": 0.7284, + "step": 462 + }, + { + "epoch": 0.383489784649365, + "grad_norm": 0.9134189486503601, + "learning_rate": 4.968197564025886e-06, + "loss": 0.723, + "step": 463 + }, + { + "epoch": 0.3843180563224738, + "grad_norm": 0.9191608428955078, + "learning_rate": 4.9680224776555305e-06, + "loss": 0.7078, + "step": 464 + }, + { + "epoch": 0.38514632799558257, + "grad_norm": 0.879683792591095, + "learning_rate": 4.9678469137458315e-06, + "loss": 0.6805, + "step": 465 + }, + { + "epoch": 0.38597459966869135, + "grad_norm": 0.8887293338775635, + "learning_rate": 4.96767087233076e-06, + "loss": 0.7296, + "step": 466 + }, + { + "epoch": 0.38680287134180014, + "grad_norm": 0.889991044998169, + "learning_rate": 4.967494353444377e-06, + "loss": 0.702, + "step": 467 + }, + { + "epoch": 0.38763114301490886, + "grad_norm": 0.9667506814002991, + "learning_rate": 4.967317357120839e-06, + "loss": 0.7305, + "step": 468 + }, + { + "epoch": 0.38845941468801765, + "grad_norm": 0.9297274351119995, + "learning_rate": 4.967139883394392e-06, + "loss": 0.7344, + "step": 469 + }, + { + "epoch": 0.38928768636112643, + "grad_norm": 0.9167365431785583, + "learning_rate": 4.966961932299377e-06, + "loss": 0.7016, + "step": 470 + }, + { + "epoch": 0.3901159580342352, + "grad_norm": 0.877378523349762, + "learning_rate": 4.966783503870224e-06, + "loss": 0.7313, + "step": 471 + }, + { + "epoch": 0.390944229707344, + "grad_norm": 0.9625933766365051, + "learning_rate": 4.966604598141458e-06, + "loss": 0.7245, + "step": 472 + }, + { + "epoch": 0.3917725013804528, + "grad_norm": 0.8859156966209412, + "learning_rate": 4.9664252151476955e-06, + "loss": 0.7094, + "step": 473 + }, + { + "epoch": 0.39260077305356156, + "grad_norm": 0.9172984957695007, + "learning_rate": 4.966245354923645e-06, + "loss": 0.6966, + "step": 474 + }, + { + "epoch": 0.39342904472667034, + "grad_norm": 0.9040864109992981, + "learning_rate": 4.96606501750411e-06, + "loss": 0.722, + "step": 475 + }, + { + "epoch": 0.3942573163997791, + "grad_norm": 0.8900951147079468, + "learning_rate": 4.965884202923982e-06, + "loss": 0.7263, + "step": 476 + }, + { + "epoch": 0.3950855880728879, + "grad_norm": 0.9096625447273254, + "learning_rate": 4.965702911218247e-06, + "loss": 0.702, + "step": 477 + }, + { + "epoch": 0.3959138597459967, + "grad_norm": 0.8801544904708862, + "learning_rate": 4.965521142421984e-06, + "loss": 0.7113, + "step": 478 + }, + { + "epoch": 0.39674213141910547, + "grad_norm": 0.9054175615310669, + "learning_rate": 4.965338896570363e-06, + "loss": 0.7135, + "step": 479 + }, + { + "epoch": 0.39757040309221425, + "grad_norm": 0.8734445571899414, + "learning_rate": 4.965156173698648e-06, + "loss": 0.7181, + "step": 480 + }, + { + "epoch": 0.39839867476532304, + "grad_norm": 0.9113215208053589, + "learning_rate": 4.964972973842191e-06, + "loss": 0.7008, + "step": 481 + }, + { + "epoch": 0.3992269464384318, + "grad_norm": 0.8941736221313477, + "learning_rate": 4.964789297036444e-06, + "loss": 0.7201, + "step": 482 + }, + { + "epoch": 0.4000552181115406, + "grad_norm": 0.8555464148521423, + "learning_rate": 4.964605143316944e-06, + "loss": 0.7173, + "step": 483 + }, + { + "epoch": 0.4008834897846494, + "grad_norm": 0.8719691634178162, + "learning_rate": 4.9644205127193225e-06, + "loss": 0.7211, + "step": 484 + }, + { + "epoch": 0.40171176145775817, + "grad_norm": 0.901626467704773, + "learning_rate": 4.964235405279306e-06, + "loss": 0.7223, + "step": 485 + }, + { + "epoch": 0.40254003313086695, + "grad_norm": 0.856769859790802, + "learning_rate": 4.9640498210327095e-06, + "loss": 0.7103, + "step": 486 + }, + { + "epoch": 0.40336830480397573, + "grad_norm": 0.9053338766098022, + "learning_rate": 4.963863760015442e-06, + "loss": 0.7301, + "step": 487 + }, + { + "epoch": 0.40419657647708446, + "grad_norm": 0.8989829421043396, + "learning_rate": 4.963677222263505e-06, + "loss": 0.7391, + "step": 488 + }, + { + "epoch": 0.40502484815019324, + "grad_norm": 0.9187886118888855, + "learning_rate": 4.963490207812991e-06, + "loss": 0.7179, + "step": 489 + }, + { + "epoch": 0.405853119823302, + "grad_norm": 0.8712778687477112, + "learning_rate": 4.963302716700087e-06, + "loss": 0.7088, + "step": 490 + }, + { + "epoch": 0.4066813914964108, + "grad_norm": 1.0485447645187378, + "learning_rate": 4.9631147489610695e-06, + "loss": 0.7068, + "step": 491 + }, + { + "epoch": 0.4075096631695196, + "grad_norm": 0.9332602024078369, + "learning_rate": 4.962926304632309e-06, + "loss": 0.7201, + "step": 492 + }, + { + "epoch": 0.40833793484262837, + "grad_norm": 0.9264400601387024, + "learning_rate": 4.962737383750268e-06, + "loss": 0.7136, + "step": 493 + }, + { + "epoch": 0.40916620651573715, + "grad_norm": 0.8812651634216309, + "learning_rate": 4.962547986351501e-06, + "loss": 0.7315, + "step": 494 + }, + { + "epoch": 0.40999447818884593, + "grad_norm": 0.8905262351036072, + "learning_rate": 4.962358112472653e-06, + "loss": 0.7127, + "step": 495 + }, + { + "epoch": 0.4108227498619547, + "grad_norm": 0.887514591217041, + "learning_rate": 4.962167762150465e-06, + "loss": 0.7602, + "step": 496 + }, + { + "epoch": 0.4116510215350635, + "grad_norm": 0.8683431148529053, + "learning_rate": 4.9619769354217665e-06, + "loss": 0.7091, + "step": 497 + }, + { + "epoch": 0.4124792932081723, + "grad_norm": 0.944595217704773, + "learning_rate": 4.961785632323481e-06, + "loss": 0.708, + "step": 498 + }, + { + "epoch": 0.41330756488128106, + "grad_norm": 0.8834611773490906, + "learning_rate": 4.961593852892625e-06, + "loss": 0.7308, + "step": 499 + }, + { + "epoch": 0.41413583655438985, + "grad_norm": 0.8479524850845337, + "learning_rate": 4.961401597166304e-06, + "loss": 0.7061, + "step": 500 + }, + { + "epoch": 0.41496410822749863, + "grad_norm": 0.9092981219291687, + "learning_rate": 4.96120886518172e-06, + "loss": 0.7176, + "step": 501 + }, + { + "epoch": 0.4157923799006074, + "grad_norm": 0.9024932384490967, + "learning_rate": 4.961015656976163e-06, + "loss": 0.6944, + "step": 502 + }, + { + "epoch": 0.4166206515737162, + "grad_norm": 0.8945576548576355, + "learning_rate": 4.960821972587018e-06, + "loss": 0.6958, + "step": 503 + }, + { + "epoch": 0.417448923246825, + "grad_norm": 0.8668336868286133, + "learning_rate": 4.9606278120517606e-06, + "loss": 0.7219, + "step": 504 + }, + { + "epoch": 0.41827719491993376, + "grad_norm": 0.8831227421760559, + "learning_rate": 4.96043317540796e-06, + "loss": 0.7124, + "step": 505 + }, + { + "epoch": 0.41910546659304254, + "grad_norm": 0.9637073874473572, + "learning_rate": 4.960238062693274e-06, + "loss": 0.7337, + "step": 506 + }, + { + "epoch": 0.4199337382661513, + "grad_norm": 0.8895899653434753, + "learning_rate": 4.960042473945459e-06, + "loss": 0.7459, + "step": 507 + }, + { + "epoch": 0.42076200993926005, + "grad_norm": 0.8762213587760925, + "learning_rate": 4.959846409202356e-06, + "loss": 0.7319, + "step": 508 + }, + { + "epoch": 0.42159028161236883, + "grad_norm": 0.9177393317222595, + "learning_rate": 4.959649868501903e-06, + "loss": 0.7114, + "step": 509 + }, + { + "epoch": 0.4224185532854776, + "grad_norm": 0.88541179895401, + "learning_rate": 4.9594528518821305e-06, + "loss": 0.716, + "step": 510 + }, + { + "epoch": 0.4232468249585864, + "grad_norm": 0.865566611289978, + "learning_rate": 4.959255359381158e-06, + "loss": 0.7098, + "step": 511 + }, + { + "epoch": 0.4240750966316952, + "grad_norm": 0.8629707098007202, + "learning_rate": 4.959057391037197e-06, + "loss": 0.7146, + "step": 512 + }, + { + "epoch": 0.42490336830480396, + "grad_norm": 0.8642273545265198, + "learning_rate": 4.958858946888554e-06, + "loss": 0.7345, + "step": 513 + }, + { + "epoch": 0.42573163997791275, + "grad_norm": 0.9157605171203613, + "learning_rate": 4.958660026973626e-06, + "loss": 0.7177, + "step": 514 + }, + { + "epoch": 0.42655991165102153, + "grad_norm": 0.9108191728591919, + "learning_rate": 4.958460631330903e-06, + "loss": 0.7317, + "step": 515 + }, + { + "epoch": 0.4273881833241303, + "grad_norm": 0.8524913191795349, + "learning_rate": 4.958260759998965e-06, + "loss": 0.7193, + "step": 516 + }, + { + "epoch": 0.4282164549972391, + "grad_norm": 0.9197820425033569, + "learning_rate": 4.958060413016485e-06, + "loss": 0.7127, + "step": 517 + }, + { + "epoch": 0.4290447266703479, + "grad_norm": 0.8744186758995056, + "learning_rate": 4.9578595904222296e-06, + "loss": 0.6968, + "step": 518 + }, + { + "epoch": 0.42987299834345666, + "grad_norm": 0.9071037769317627, + "learning_rate": 4.957658292255055e-06, + "loss": 0.6963, + "step": 519 + }, + { + "epoch": 0.43070127001656544, + "grad_norm": 0.937796950340271, + "learning_rate": 4.95745651855391e-06, + "loss": 0.7193, + "step": 520 + }, + { + "epoch": 0.4315295416896742, + "grad_norm": 0.883471667766571, + "learning_rate": 4.9572542693578375e-06, + "loss": 0.6886, + "step": 521 + }, + { + "epoch": 0.432357813362783, + "grad_norm": 0.8738589882850647, + "learning_rate": 4.9570515447059705e-06, + "loss": 0.7072, + "step": 522 + }, + { + "epoch": 0.4331860850358918, + "grad_norm": 0.9039497971534729, + "learning_rate": 4.9568483446375335e-06, + "loss": 0.7251, + "step": 523 + }, + { + "epoch": 0.43401435670900057, + "grad_norm": 0.896789014339447, + "learning_rate": 4.956644669191845e-06, + "loss": 0.7021, + "step": 524 + }, + { + "epoch": 0.43484262838210935, + "grad_norm": 0.8883570432662964, + "learning_rate": 4.9564405184083126e-06, + "loss": 0.7153, + "step": 525 + }, + { + "epoch": 0.43567090005521814, + "grad_norm": 0.8869197368621826, + "learning_rate": 4.956235892326438e-06, + "loss": 0.7061, + "step": 526 + }, + { + "epoch": 0.4364991717283269, + "grad_norm": 0.9237920641899109, + "learning_rate": 4.9560307909858165e-06, + "loss": 0.6988, + "step": 527 + }, + { + "epoch": 0.43732744340143564, + "grad_norm": 0.9359283447265625, + "learning_rate": 4.955825214426131e-06, + "loss": 0.738, + "step": 528 + }, + { + "epoch": 0.4381557150745444, + "grad_norm": 0.8602650761604309, + "learning_rate": 4.95561916268716e-06, + "loss": 0.709, + "step": 529 + }, + { + "epoch": 0.4389839867476532, + "grad_norm": 0.9013513922691345, + "learning_rate": 4.955412635808771e-06, + "loss": 0.6836, + "step": 530 + }, + { + "epoch": 0.439812258420762, + "grad_norm": 0.9136670827865601, + "learning_rate": 4.955205633830927e-06, + "loss": 0.7263, + "step": 531 + }, + { + "epoch": 0.4406405300938708, + "grad_norm": 0.9013793468475342, + "learning_rate": 4.95499815679368e-06, + "loss": 0.6937, + "step": 532 + }, + { + "epoch": 0.44146880176697956, + "grad_norm": 0.9017112851142883, + "learning_rate": 4.954790204737174e-06, + "loss": 0.6913, + "step": 533 + }, + { + "epoch": 0.44229707344008834, + "grad_norm": 0.8901436924934387, + "learning_rate": 4.954581777701647e-06, + "loss": 0.7241, + "step": 534 + }, + { + "epoch": 0.4431253451131971, + "grad_norm": 0.8882759809494019, + "learning_rate": 4.954372875727427e-06, + "loss": 0.7115, + "step": 535 + }, + { + "epoch": 0.4439536167863059, + "grad_norm": 0.8909885883331299, + "learning_rate": 4.954163498854936e-06, + "loss": 0.7146, + "step": 536 + }, + { + "epoch": 0.4447818884594147, + "grad_norm": 0.8951786756515503, + "learning_rate": 4.953953647124684e-06, + "loss": 0.699, + "step": 537 + }, + { + "epoch": 0.44561016013252347, + "grad_norm": 0.8890933990478516, + "learning_rate": 4.953743320577278e-06, + "loss": 0.7018, + "step": 538 + }, + { + "epoch": 0.44643843180563225, + "grad_norm": 0.910393238067627, + "learning_rate": 4.953532519253412e-06, + "loss": 0.7299, + "step": 539 + }, + { + "epoch": 0.44726670347874103, + "grad_norm": 0.9152857065200806, + "learning_rate": 4.953321243193876e-06, + "loss": 0.6984, + "step": 540 + }, + { + "epoch": 0.4480949751518498, + "grad_norm": 0.9001837372779846, + "learning_rate": 4.9531094924395485e-06, + "loss": 0.727, + "step": 541 + }, + { + "epoch": 0.4489232468249586, + "grad_norm": 0.882060170173645, + "learning_rate": 4.952897267031401e-06, + "loss": 0.7016, + "step": 542 + }, + { + "epoch": 0.4497515184980674, + "grad_norm": 0.8665159344673157, + "learning_rate": 4.9526845670104995e-06, + "loss": 0.7139, + "step": 543 + }, + { + "epoch": 0.45057979017117616, + "grad_norm": 0.9200844764709473, + "learning_rate": 4.952471392417997e-06, + "loss": 0.7016, + "step": 544 + }, + { + "epoch": 0.45140806184428495, + "grad_norm": 0.8852037191390991, + "learning_rate": 4.9522577432951416e-06, + "loss": 0.7048, + "step": 545 + }, + { + "epoch": 0.45223633351739373, + "grad_norm": 0.9003865122795105, + "learning_rate": 4.952043619683274e-06, + "loss": 0.7106, + "step": 546 + }, + { + "epoch": 0.4530646051905025, + "grad_norm": 0.8888446092605591, + "learning_rate": 4.951829021623822e-06, + "loss": 0.7202, + "step": 547 + }, + { + "epoch": 0.45389287686361124, + "grad_norm": 0.8940277099609375, + "learning_rate": 4.9516139491583094e-06, + "loss": 0.7156, + "step": 548 + }, + { + "epoch": 0.45472114853672, + "grad_norm": 0.9066211581230164, + "learning_rate": 4.951398402328353e-06, + "loss": 0.6865, + "step": 549 + }, + { + "epoch": 0.4555494202098288, + "grad_norm": 0.9352865815162659, + "learning_rate": 4.9511823811756565e-06, + "loss": 0.7117, + "step": 550 + }, + { + "epoch": 0.4563776918829376, + "grad_norm": 0.8672269582748413, + "learning_rate": 4.95096588574202e-06, + "loss": 0.7166, + "step": 551 + }, + { + "epoch": 0.45720596355604637, + "grad_norm": 0.8730685114860535, + "learning_rate": 4.95074891606933e-06, + "loss": 0.6998, + "step": 552 + }, + { + "epoch": 0.45803423522915515, + "grad_norm": 0.896821916103363, + "learning_rate": 4.950531472199572e-06, + "loss": 0.6839, + "step": 553 + }, + { + "epoch": 0.45886250690226393, + "grad_norm": 0.918572187423706, + "learning_rate": 4.950313554174817e-06, + "loss": 0.7155, + "step": 554 + }, + { + "epoch": 0.4596907785753727, + "grad_norm": 0.8670752048492432, + "learning_rate": 4.950095162037232e-06, + "loss": 0.708, + "step": 555 + }, + { + "epoch": 0.4605190502484815, + "grad_norm": 0.875618040561676, + "learning_rate": 4.9498762958290715e-06, + "loss": 0.6956, + "step": 556 + }, + { + "epoch": 0.4613473219215903, + "grad_norm": 0.9120676517486572, + "learning_rate": 4.949656955592685e-06, + "loss": 0.7327, + "step": 557 + }, + { + "epoch": 0.46217559359469906, + "grad_norm": 0.9670370817184448, + "learning_rate": 4.949437141370514e-06, + "loss": 0.705, + "step": 558 + }, + { + "epoch": 0.46300386526780785, + "grad_norm": 0.9155322313308716, + "learning_rate": 4.949216853205089e-06, + "loss": 0.7126, + "step": 559 + }, + { + "epoch": 0.46383213694091663, + "grad_norm": 0.8945392370223999, + "learning_rate": 4.948996091139034e-06, + "loss": 0.7083, + "step": 560 + }, + { + "epoch": 0.4646604086140254, + "grad_norm": 0.8922162055969238, + "learning_rate": 4.948774855215065e-06, + "loss": 0.7189, + "step": 561 + }, + { + "epoch": 0.4654886802871342, + "grad_norm": 0.8940657377243042, + "learning_rate": 4.948553145475989e-06, + "loss": 0.6934, + "step": 562 + }, + { + "epoch": 0.466316951960243, + "grad_norm": 0.8946855068206787, + "learning_rate": 4.948330961964703e-06, + "loss": 0.6835, + "step": 563 + }, + { + "epoch": 0.46714522363335176, + "grad_norm": 0.9002820253372192, + "learning_rate": 4.948108304724201e-06, + "loss": 0.712, + "step": 564 + }, + { + "epoch": 0.46797349530646054, + "grad_norm": 0.8715711832046509, + "learning_rate": 4.947885173797561e-06, + "loss": 0.7197, + "step": 565 + }, + { + "epoch": 0.4688017669795693, + "grad_norm": 0.9269874095916748, + "learning_rate": 4.94766156922796e-06, + "loss": 0.7311, + "step": 566 + }, + { + "epoch": 0.4696300386526781, + "grad_norm": 0.8999423980712891, + "learning_rate": 4.947437491058662e-06, + "loss": 0.7144, + "step": 567 + }, + { + "epoch": 0.47045831032578683, + "grad_norm": 0.8704255223274231, + "learning_rate": 4.947212939333024e-06, + "loss": 0.7156, + "step": 568 + }, + { + "epoch": 0.4712865819988956, + "grad_norm": 0.8770347237586975, + "learning_rate": 4.946987914094495e-06, + "loss": 0.6969, + "step": 569 + }, + { + "epoch": 0.4721148536720044, + "grad_norm": 0.8865682482719421, + "learning_rate": 4.946762415386615e-06, + "loss": 0.7089, + "step": 570 + }, + { + "epoch": 0.4729431253451132, + "grad_norm": 0.8756279349327087, + "learning_rate": 4.946536443253016e-06, + "loss": 0.7159, + "step": 571 + }, + { + "epoch": 0.47377139701822196, + "grad_norm": 0.8844839930534363, + "learning_rate": 4.946309997737422e-06, + "loss": 0.7234, + "step": 572 + }, + { + "epoch": 0.47459966869133074, + "grad_norm": 0.8549885153770447, + "learning_rate": 4.946083078883647e-06, + "loss": 0.6927, + "step": 573 + }, + { + "epoch": 0.4754279403644395, + "grad_norm": 0.8967456817626953, + "learning_rate": 4.945855686735599e-06, + "loss": 0.7082, + "step": 574 + }, + { + "epoch": 0.4762562120375483, + "grad_norm": 0.9120200276374817, + "learning_rate": 4.9456278213372755e-06, + "loss": 0.7135, + "step": 575 + }, + { + "epoch": 0.4770844837106571, + "grad_norm": 0.8780766725540161, + "learning_rate": 4.945399482732765e-06, + "loss": 0.7139, + "step": 576 + }, + { + "epoch": 0.4779127553837659, + "grad_norm": 0.8900999426841736, + "learning_rate": 4.945170670966253e-06, + "loss": 0.7132, + "step": 577 + }, + { + "epoch": 0.47874102705687466, + "grad_norm": 0.9402151107788086, + "learning_rate": 4.944941386082008e-06, + "loss": 0.7139, + "step": 578 + }, + { + "epoch": 0.47956929872998344, + "grad_norm": 0.8769465088844299, + "learning_rate": 4.944711628124396e-06, + "loss": 0.7236, + "step": 579 + }, + { + "epoch": 0.4803975704030922, + "grad_norm": 0.9185771346092224, + "learning_rate": 4.944481397137874e-06, + "loss": 0.6884, + "step": 580 + }, + { + "epoch": 0.481225842076201, + "grad_norm": 0.8775897026062012, + "learning_rate": 4.944250693166988e-06, + "loss": 0.731, + "step": 581 + }, + { + "epoch": 0.4820541137493098, + "grad_norm": 0.8751668930053711, + "learning_rate": 4.94401951625638e-06, + "loss": 0.7007, + "step": 582 + }, + { + "epoch": 0.48288238542241857, + "grad_norm": 0.9042515754699707, + "learning_rate": 4.943787866450776e-06, + "loss": 0.7064, + "step": 583 + }, + { + "epoch": 0.48371065709552735, + "grad_norm": 0.8845478892326355, + "learning_rate": 4.943555743795002e-06, + "loss": 0.7069, + "step": 584 + }, + { + "epoch": 0.48453892876863613, + "grad_norm": 0.8959096670150757, + "learning_rate": 4.943323148333969e-06, + "loss": 0.6871, + "step": 585 + }, + { + "epoch": 0.4853672004417449, + "grad_norm": 0.9021669626235962, + "learning_rate": 4.943090080112684e-06, + "loss": 0.7156, + "step": 586 + }, + { + "epoch": 0.4861954721148537, + "grad_norm": 0.8768396973609924, + "learning_rate": 4.942856539176243e-06, + "loss": 0.681, + "step": 587 + }, + { + "epoch": 0.4870237437879624, + "grad_norm": 0.8734706044197083, + "learning_rate": 4.942622525569832e-06, + "loss": 0.7371, + "step": 588 + }, + { + "epoch": 0.4878520154610712, + "grad_norm": 0.9155406355857849, + "learning_rate": 4.942388039338733e-06, + "loss": 0.7012, + "step": 589 + }, + { + "epoch": 0.48868028713418, + "grad_norm": 0.8778123259544373, + "learning_rate": 4.942153080528317e-06, + "loss": 0.7201, + "step": 590 + }, + { + "epoch": 0.4895085588072888, + "grad_norm": 0.9102092981338501, + "learning_rate": 4.941917649184045e-06, + "loss": 0.6951, + "step": 591 + }, + { + "epoch": 0.49033683048039756, + "grad_norm": 0.8855768442153931, + "learning_rate": 4.941681745351471e-06, + "loss": 0.6997, + "step": 592 + }, + { + "epoch": 0.49116510215350634, + "grad_norm": 0.8770707845687866, + "learning_rate": 4.94144536907624e-06, + "loss": 0.7092, + "step": 593 + }, + { + "epoch": 0.4919933738266151, + "grad_norm": 0.900458574295044, + "learning_rate": 4.9412085204040896e-06, + "loss": 0.6909, + "step": 594 + }, + { + "epoch": 0.4928216454997239, + "grad_norm": 0.903489887714386, + "learning_rate": 4.940971199380846e-06, + "loss": 0.7066, + "step": 595 + }, + { + "epoch": 0.4936499171728327, + "grad_norm": 0.9331871271133423, + "learning_rate": 4.940733406052431e-06, + "loss": 0.6924, + "step": 596 + }, + { + "epoch": 0.49447818884594147, + "grad_norm": 0.8855593800544739, + "learning_rate": 4.9404951404648535e-06, + "loss": 0.6991, + "step": 597 + }, + { + "epoch": 0.49530646051905025, + "grad_norm": 0.9075731039047241, + "learning_rate": 4.940256402664216e-06, + "loss": 0.6973, + "step": 598 + }, + { + "epoch": 0.49613473219215903, + "grad_norm": 0.8970743417739868, + "learning_rate": 4.940017192696713e-06, + "loss": 0.6767, + "step": 599 + }, + { + "epoch": 0.4969630038652678, + "grad_norm": 0.8644358515739441, + "learning_rate": 4.939777510608628e-06, + "loss": 0.7284, + "step": 600 + }, + { + "epoch": 0.4977912755383766, + "grad_norm": 0.8807653784751892, + "learning_rate": 4.939537356446339e-06, + "loss": 0.7143, + "step": 601 + }, + { + "epoch": 0.4986195472114854, + "grad_norm": 0.892946183681488, + "learning_rate": 4.939296730256312e-06, + "loss": 0.7021, + "step": 602 + }, + { + "epoch": 0.49944781888459416, + "grad_norm": 0.9661110639572144, + "learning_rate": 4.939055632085107e-06, + "loss": 0.7198, + "step": 603 + }, + { + "epoch": 0.500276090557703, + "grad_norm": 0.8638496398925781, + "learning_rate": 4.9388140619793735e-06, + "loss": 0.6943, + "step": 604 + }, + { + "epoch": 0.5011043622308117, + "grad_norm": 0.9025457501411438, + "learning_rate": 4.938572019985853e-06, + "loss": 0.7114, + "step": 605 + }, + { + "epoch": 0.5019326339039205, + "grad_norm": 0.9622734785079956, + "learning_rate": 4.93832950615138e-06, + "loss": 0.7086, + "step": 606 + }, + { + "epoch": 0.5027609055770292, + "grad_norm": 0.9565468430519104, + "learning_rate": 4.938086520522876e-06, + "loss": 0.7125, + "step": 607 + }, + { + "epoch": 0.5035891772501381, + "grad_norm": 0.8970094323158264, + "learning_rate": 4.937843063147359e-06, + "loss": 0.7023, + "step": 608 + }, + { + "epoch": 0.5044174489232468, + "grad_norm": 0.8920996785163879, + "learning_rate": 4.937599134071933e-06, + "loss": 0.7041, + "step": 609 + }, + { + "epoch": 0.5052457205963556, + "grad_norm": 0.9304755926132202, + "learning_rate": 4.937354733343799e-06, + "loss": 0.7165, + "step": 610 + }, + { + "epoch": 0.5060739922694644, + "grad_norm": 0.8995933532714844, + "learning_rate": 4.937109861010244e-06, + "loss": 0.6907, + "step": 611 + }, + { + "epoch": 0.5069022639425732, + "grad_norm": 0.8648850321769714, + "learning_rate": 4.936864517118649e-06, + "loss": 0.6994, + "step": 612 + }, + { + "epoch": 0.5077305356156819, + "grad_norm": 0.8414506316184998, + "learning_rate": 4.936618701716487e-06, + "loss": 0.7081, + "step": 613 + }, + { + "epoch": 0.5085588072887908, + "grad_norm": 0.9282650947570801, + "learning_rate": 4.936372414851321e-06, + "loss": 0.7303, + "step": 614 + }, + { + "epoch": 0.5093870789618995, + "grad_norm": 0.877757728099823, + "learning_rate": 4.936125656570802e-06, + "loss": 0.7132, + "step": 615 + }, + { + "epoch": 0.5102153506350083, + "grad_norm": 0.8645135164260864, + "learning_rate": 4.935878426922679e-06, + "loss": 0.723, + "step": 616 + }, + { + "epoch": 0.5110436223081171, + "grad_norm": 0.9014747738838196, + "learning_rate": 4.935630725954787e-06, + "loss": 0.7167, + "step": 617 + }, + { + "epoch": 0.5118718939812258, + "grad_norm": 0.9295386672019958, + "learning_rate": 4.935382553715055e-06, + "loss": 0.705, + "step": 618 + }, + { + "epoch": 0.5127001656543346, + "grad_norm": 0.8910616040229797, + "learning_rate": 4.935133910251499e-06, + "loss": 0.6781, + "step": 619 + }, + { + "epoch": 0.5135284373274434, + "grad_norm": 0.8804931640625, + "learning_rate": 4.934884795612234e-06, + "loss": 0.6984, + "step": 620 + }, + { + "epoch": 0.5143567090005522, + "grad_norm": 0.8223545551300049, + "learning_rate": 4.934635209845458e-06, + "loss": 0.6892, + "step": 621 + }, + { + "epoch": 0.5151849806736609, + "grad_norm": 0.8509450554847717, + "learning_rate": 4.934385152999463e-06, + "loss": 0.7139, + "step": 622 + }, + { + "epoch": 0.5160132523467698, + "grad_norm": 0.8944748044013977, + "learning_rate": 4.934134625122636e-06, + "loss": 0.6976, + "step": 623 + }, + { + "epoch": 0.5168415240198785, + "grad_norm": 0.9021086692810059, + "learning_rate": 4.933883626263448e-06, + "loss": 0.7189, + "step": 624 + }, + { + "epoch": 0.5176697956929873, + "grad_norm": 0.8939547538757324, + "learning_rate": 4.933632156470467e-06, + "loss": 0.704, + "step": 625 + }, + { + "epoch": 0.518498067366096, + "grad_norm": 0.8793498277664185, + "learning_rate": 4.93338021579235e-06, + "loss": 0.7046, + "step": 626 + }, + { + "epoch": 0.5193263390392049, + "grad_norm": 0.8728756308555603, + "learning_rate": 4.933127804277845e-06, + "loss": 0.6826, + "step": 627 + }, + { + "epoch": 0.5201546107123136, + "grad_norm": 0.8989412188529968, + "learning_rate": 4.932874921975792e-06, + "loss": 0.6842, + "step": 628 + }, + { + "epoch": 0.5209828823854225, + "grad_norm": 0.8836191296577454, + "learning_rate": 4.93262156893512e-06, + "loss": 0.7061, + "step": 629 + }, + { + "epoch": 0.5218111540585312, + "grad_norm": 0.9429040551185608, + "learning_rate": 4.9323677452048505e-06, + "loss": 0.7053, + "step": 630 + }, + { + "epoch": 0.52263942573164, + "grad_norm": 0.879463255405426, + "learning_rate": 4.932113450834097e-06, + "loss": 0.6926, + "step": 631 + }, + { + "epoch": 0.5234676974047487, + "grad_norm": 0.8566175699234009, + "learning_rate": 4.931858685872062e-06, + "loss": 0.7126, + "step": 632 + }, + { + "epoch": 0.5242959690778576, + "grad_norm": 0.8694538474082947, + "learning_rate": 4.931603450368042e-06, + "loss": 0.7153, + "step": 633 + }, + { + "epoch": 0.5251242407509663, + "grad_norm": 0.8754484057426453, + "learning_rate": 4.931347744371421e-06, + "loss": 0.711, + "step": 634 + }, + { + "epoch": 0.5259525124240751, + "grad_norm": 0.8982204794883728, + "learning_rate": 4.931091567931675e-06, + "loss": 0.7168, + "step": 635 + }, + { + "epoch": 0.5267807840971839, + "grad_norm": 0.9262725710868835, + "learning_rate": 4.930834921098374e-06, + "loss": 0.7315, + "step": 636 + }, + { + "epoch": 0.5276090557702926, + "grad_norm": 0.9271771311759949, + "learning_rate": 4.930577803921176e-06, + "loss": 0.6968, + "step": 637 + }, + { + "epoch": 0.5284373274434014, + "grad_norm": 0.8855197429656982, + "learning_rate": 4.93032021644983e-06, + "loss": 0.6986, + "step": 638 + }, + { + "epoch": 0.5292655991165102, + "grad_norm": 0.8615745902061462, + "learning_rate": 4.930062158734178e-06, + "loss": 0.7014, + "step": 639 + }, + { + "epoch": 0.530093870789619, + "grad_norm": 0.9004185795783997, + "learning_rate": 4.92980363082415e-06, + "loss": 0.7045, + "step": 640 + }, + { + "epoch": 0.5309221424627277, + "grad_norm": 0.8533529043197632, + "learning_rate": 4.92954463276977e-06, + "loss": 0.6911, + "step": 641 + }, + { + "epoch": 0.5317504141358366, + "grad_norm": 0.881901204586029, + "learning_rate": 4.929285164621152e-06, + "loss": 0.7148, + "step": 642 + }, + { + "epoch": 0.5325786858089453, + "grad_norm": 0.8778982162475586, + "learning_rate": 4.9290252264285e-06, + "loss": 0.6879, + "step": 643 + }, + { + "epoch": 0.5334069574820541, + "grad_norm": 0.9455735087394714, + "learning_rate": 4.928764818242109e-06, + "loss": 0.7088, + "step": 644 + }, + { + "epoch": 0.5342352291551629, + "grad_norm": 0.9255897998809814, + "learning_rate": 4.928503940112367e-06, + "loss": 0.6817, + "step": 645 + }, + { + "epoch": 0.5350635008282717, + "grad_norm": 0.8854040503501892, + "learning_rate": 4.928242592089752e-06, + "loss": 0.7074, + "step": 646 + }, + { + "epoch": 0.5358917725013804, + "grad_norm": 0.8824400901794434, + "learning_rate": 4.927980774224829e-06, + "loss": 0.7092, + "step": 647 + }, + { + "epoch": 0.5367200441744893, + "grad_norm": 0.8873158693313599, + "learning_rate": 4.927718486568261e-06, + "loss": 0.7214, + "step": 648 + }, + { + "epoch": 0.537548315847598, + "grad_norm": 0.9013757109642029, + "learning_rate": 4.927455729170796e-06, + "loss": 0.6955, + "step": 649 + }, + { + "epoch": 0.5383765875207068, + "grad_norm": 0.9100024104118347, + "learning_rate": 4.927192502083276e-06, + "loss": 0.7089, + "step": 650 + }, + { + "epoch": 0.5392048591938156, + "grad_norm": 0.8935851454734802, + "learning_rate": 4.9269288053566335e-06, + "loss": 0.7018, + "step": 651 + }, + { + "epoch": 0.5400331308669244, + "grad_norm": 0.8726267218589783, + "learning_rate": 4.92666463904189e-06, + "loss": 0.6829, + "step": 652 + }, + { + "epoch": 0.5408614025400331, + "grad_norm": 0.9320974349975586, + "learning_rate": 4.926400003190161e-06, + "loss": 0.6766, + "step": 653 + }, + { + "epoch": 0.541689674213142, + "grad_norm": 0.9257067441940308, + "learning_rate": 4.926134897852648e-06, + "loss": 0.6913, + "step": 654 + }, + { + "epoch": 0.5425179458862507, + "grad_norm": 0.8642577528953552, + "learning_rate": 4.92586932308065e-06, + "loss": 0.7155, + "step": 655 + }, + { + "epoch": 0.5433462175593595, + "grad_norm": 0.8861536383628845, + "learning_rate": 4.9256032789255505e-06, + "loss": 0.6695, + "step": 656 + }, + { + "epoch": 0.5441744892324683, + "grad_norm": 0.8901777267456055, + "learning_rate": 4.925336765438829e-06, + "loss": 0.6847, + "step": 657 + }, + { + "epoch": 0.545002760905577, + "grad_norm": 0.8810168504714966, + "learning_rate": 4.9250697826720515e-06, + "loss": 0.7038, + "step": 658 + }, + { + "epoch": 0.5458310325786858, + "grad_norm": 0.8655433654785156, + "learning_rate": 4.924802330676877e-06, + "loss": 0.6908, + "step": 659 + }, + { + "epoch": 0.5466593042517945, + "grad_norm": 0.8786777853965759, + "learning_rate": 4.924534409505055e-06, + "loss": 0.663, + "step": 660 + }, + { + "epoch": 0.5474875759249034, + "grad_norm": 0.8837904334068298, + "learning_rate": 4.924266019208427e-06, + "loss": 0.6843, + "step": 661 + }, + { + "epoch": 0.5483158475980121, + "grad_norm": 0.8715303540229797, + "learning_rate": 4.923997159838923e-06, + "loss": 0.6841, + "step": 662 + }, + { + "epoch": 0.549144119271121, + "grad_norm": 0.8825753927230835, + "learning_rate": 4.923727831448564e-06, + "loss": 0.7053, + "step": 663 + }, + { + "epoch": 0.5499723909442297, + "grad_norm": 0.877668559551239, + "learning_rate": 4.923458034089463e-06, + "loss": 0.6624, + "step": 664 + }, + { + "epoch": 0.5508006626173385, + "grad_norm": 0.8721411824226379, + "learning_rate": 4.923187767813825e-06, + "loss": 0.7339, + "step": 665 + }, + { + "epoch": 0.5516289342904472, + "grad_norm": 0.8607308268547058, + "learning_rate": 4.922917032673941e-06, + "loss": 0.7184, + "step": 666 + }, + { + "epoch": 0.5524572059635561, + "grad_norm": 0.8456798791885376, + "learning_rate": 4.922645828722199e-06, + "loss": 0.7056, + "step": 667 + }, + { + "epoch": 0.5532854776366648, + "grad_norm": 0.8557775020599365, + "learning_rate": 4.922374156011071e-06, + "loss": 0.6942, + "step": 668 + }, + { + "epoch": 0.5541137493097736, + "grad_norm": 0.8877674341201782, + "learning_rate": 4.922102014593126e-06, + "loss": 0.7173, + "step": 669 + }, + { + "epoch": 0.5549420209828824, + "grad_norm": 0.8662481904029846, + "learning_rate": 4.921829404521019e-06, + "loss": 0.7113, + "step": 670 + }, + { + "epoch": 0.5557702926559912, + "grad_norm": 0.8887649774551392, + "learning_rate": 4.921556325847499e-06, + "loss": 0.7028, + "step": 671 + }, + { + "epoch": 0.5565985643290999, + "grad_norm": 0.9837973713874817, + "learning_rate": 4.921282778625403e-06, + "loss": 0.6975, + "step": 672 + }, + { + "epoch": 0.5574268360022088, + "grad_norm": 0.8754701018333435, + "learning_rate": 4.921008762907661e-06, + "loss": 0.6983, + "step": 673 + }, + { + "epoch": 0.5582551076753175, + "grad_norm": 0.9125426411628723, + "learning_rate": 4.920734278747291e-06, + "loss": 0.6982, + "step": 674 + }, + { + "epoch": 0.5590833793484263, + "grad_norm": 0.9424439072608948, + "learning_rate": 4.920459326197403e-06, + "loss": 0.7304, + "step": 675 + }, + { + "epoch": 0.5599116510215351, + "grad_norm": 0.8718393445014954, + "learning_rate": 4.9201839053112e-06, + "loss": 0.6644, + "step": 676 + }, + { + "epoch": 0.5607399226946438, + "grad_norm": 0.9002220630645752, + "learning_rate": 4.919908016141972e-06, + "loss": 0.6966, + "step": 677 + }, + { + "epoch": 0.5615681943677526, + "grad_norm": 0.8925678133964539, + "learning_rate": 4.9196316587431e-06, + "loss": 0.713, + "step": 678 + }, + { + "epoch": 0.5623964660408614, + "grad_norm": 0.8854464292526245, + "learning_rate": 4.919354833168059e-06, + "loss": 0.7075, + "step": 679 + }, + { + "epoch": 0.5632247377139702, + "grad_norm": 0.8867958188056946, + "learning_rate": 4.919077539470409e-06, + "loss": 0.7166, + "step": 680 + }, + { + "epoch": 0.5640530093870789, + "grad_norm": 0.862531840801239, + "learning_rate": 4.918799777703806e-06, + "loss": 0.7119, + "step": 681 + }, + { + "epoch": 0.5648812810601878, + "grad_norm": 0.8623220920562744, + "learning_rate": 4.9185215479219944e-06, + "loss": 0.6758, + "step": 682 + }, + { + "epoch": 0.5657095527332965, + "grad_norm": 0.9171786904335022, + "learning_rate": 4.918242850178808e-06, + "loss": 0.7049, + "step": 683 + }, + { + "epoch": 0.5665378244064053, + "grad_norm": 0.8947017192840576, + "learning_rate": 4.9179636845281735e-06, + "loss": 0.7144, + "step": 684 + }, + { + "epoch": 0.567366096079514, + "grad_norm": 0.9343388080596924, + "learning_rate": 4.917684051024104e-06, + "loss": 0.7024, + "step": 685 + }, + { + "epoch": 0.5681943677526229, + "grad_norm": 0.8929025530815125, + "learning_rate": 4.917403949720709e-06, + "loss": 0.7162, + "step": 686 + }, + { + "epoch": 0.5690226394257316, + "grad_norm": 0.9000421762466431, + "learning_rate": 4.9171233806721854e-06, + "loss": 0.7085, + "step": 687 + }, + { + "epoch": 0.5698509110988405, + "grad_norm": 0.9404364228248596, + "learning_rate": 4.91684234393282e-06, + "loss": 0.6996, + "step": 688 + }, + { + "epoch": 0.5706791827719492, + "grad_norm": 0.9566411375999451, + "learning_rate": 4.916560839556989e-06, + "loss": 0.7241, + "step": 689 + }, + { + "epoch": 0.571507454445058, + "grad_norm": 0.9066181182861328, + "learning_rate": 4.916278867599163e-06, + "loss": 0.724, + "step": 690 + }, + { + "epoch": 0.5723357261181667, + "grad_norm": 0.8914163708686829, + "learning_rate": 4.915996428113901e-06, + "loss": 0.7046, + "step": 691 + }, + { + "epoch": 0.5731639977912756, + "grad_norm": 0.8977548480033875, + "learning_rate": 4.91571352115585e-06, + "loss": 0.6823, + "step": 692 + }, + { + "epoch": 0.5739922694643843, + "grad_norm": 0.8902036547660828, + "learning_rate": 4.915430146779754e-06, + "loss": 0.702, + "step": 693 + }, + { + "epoch": 0.5748205411374931, + "grad_norm": 0.8859509229660034, + "learning_rate": 4.915146305040439e-06, + "loss": 0.6996, + "step": 694 + }, + { + "epoch": 0.5756488128106019, + "grad_norm": 0.8916864991188049, + "learning_rate": 4.914861995992829e-06, + "loss": 0.6978, + "step": 695 + }, + { + "epoch": 0.5764770844837107, + "grad_norm": 0.8961358070373535, + "learning_rate": 4.9145772196919336e-06, + "loss": 0.6803, + "step": 696 + }, + { + "epoch": 0.5773053561568194, + "grad_norm": 0.8817684650421143, + "learning_rate": 4.914291976192854e-06, + "loss": 0.6891, + "step": 697 + }, + { + "epoch": 0.5781336278299282, + "grad_norm": 0.8821113705635071, + "learning_rate": 4.914006265550783e-06, + "loss": 0.7009, + "step": 698 + }, + { + "epoch": 0.578961899503037, + "grad_norm": 0.9553062319755554, + "learning_rate": 4.913720087821003e-06, + "loss": 0.7118, + "step": 699 + }, + { + "epoch": 0.5797901711761457, + "grad_norm": 0.8997507691383362, + "learning_rate": 4.913433443058887e-06, + "loss": 0.7036, + "step": 700 + }, + { + "epoch": 0.5806184428492546, + "grad_norm": 0.8659473657608032, + "learning_rate": 4.9131463313198964e-06, + "loss": 0.6766, + "step": 701 + }, + { + "epoch": 0.5814467145223633, + "grad_norm": 0.8916030526161194, + "learning_rate": 4.912858752659586e-06, + "loss": 0.6976, + "step": 702 + }, + { + "epoch": 0.5822749861954721, + "grad_norm": 0.8726498484611511, + "learning_rate": 4.912570707133599e-06, + "loss": 0.6989, + "step": 703 + }, + { + "epoch": 0.5831032578685809, + "grad_norm": 0.8725529909133911, + "learning_rate": 4.91228219479767e-06, + "loss": 0.6862, + "step": 704 + }, + { + "epoch": 0.5839315295416897, + "grad_norm": 0.8807168006896973, + "learning_rate": 4.911993215707624e-06, + "loss": 0.6964, + "step": 705 + }, + { + "epoch": 0.5847598012147984, + "grad_norm": 0.9244167804718018, + "learning_rate": 4.911703769919375e-06, + "loss": 0.6742, + "step": 706 + }, + { + "epoch": 0.5855880728879073, + "grad_norm": 0.9174355268478394, + "learning_rate": 4.9114138574889285e-06, + "loss": 0.6852, + "step": 707 + }, + { + "epoch": 0.586416344561016, + "grad_norm": 0.8894474506378174, + "learning_rate": 4.9111234784723785e-06, + "loss": 0.689, + "step": 708 + }, + { + "epoch": 0.5872446162341248, + "grad_norm": 0.8856050968170166, + "learning_rate": 4.910832632925912e-06, + "loss": 0.6742, + "step": 709 + }, + { + "epoch": 0.5880728879072336, + "grad_norm": 0.9120550751686096, + "learning_rate": 4.910541320905804e-06, + "loss": 0.7016, + "step": 710 + }, + { + "epoch": 0.5889011595803424, + "grad_norm": 0.8925642967224121, + "learning_rate": 4.910249542468422e-06, + "loss": 0.7165, + "step": 711 + }, + { + "epoch": 0.5897294312534511, + "grad_norm": 0.8786419630050659, + "learning_rate": 4.909957297670222e-06, + "loss": 0.7207, + "step": 712 + }, + { + "epoch": 0.59055770292656, + "grad_norm": 0.89600670337677, + "learning_rate": 4.909664586567749e-06, + "loss": 0.672, + "step": 713 + }, + { + "epoch": 0.5913859745996687, + "grad_norm": 0.889835000038147, + "learning_rate": 4.909371409217642e-06, + "loss": 0.7089, + "step": 714 + }, + { + "epoch": 0.5922142462727775, + "grad_norm": 0.9029288291931152, + "learning_rate": 4.909077765676627e-06, + "loss": 0.7065, + "step": 715 + }, + { + "epoch": 0.5930425179458862, + "grad_norm": 0.9212722778320312, + "learning_rate": 4.908783656001521e-06, + "loss": 0.6787, + "step": 716 + }, + { + "epoch": 0.593870789618995, + "grad_norm": 0.9079681038856506, + "learning_rate": 4.9084890802492325e-06, + "loss": 0.6996, + "step": 717 + }, + { + "epoch": 0.5946990612921038, + "grad_norm": 0.9110735654830933, + "learning_rate": 4.908194038476758e-06, + "loss": 0.6962, + "step": 718 + }, + { + "epoch": 0.5955273329652125, + "grad_norm": 0.8945547938346863, + "learning_rate": 4.9078985307411865e-06, + "loss": 0.6648, + "step": 719 + }, + { + "epoch": 0.5963556046383214, + "grad_norm": 0.8831511735916138, + "learning_rate": 4.907602557099696e-06, + "loss": 0.7077, + "step": 720 + }, + { + "epoch": 0.5971838763114301, + "grad_norm": 0.9393531680107117, + "learning_rate": 4.9073061176095525e-06, + "loss": 0.6722, + "step": 721 + }, + { + "epoch": 0.5980121479845389, + "grad_norm": 0.8807207942008972, + "learning_rate": 4.907009212328117e-06, + "loss": 0.69, + "step": 722 + }, + { + "epoch": 0.5988404196576477, + "grad_norm": 0.8811002969741821, + "learning_rate": 4.906711841312836e-06, + "loss": 0.6902, + "step": 723 + }, + { + "epoch": 0.5996686913307565, + "grad_norm": 0.8953952789306641, + "learning_rate": 4.906414004621249e-06, + "loss": 0.6747, + "step": 724 + }, + { + "epoch": 0.6004969630038652, + "grad_norm": 0.8661350607872009, + "learning_rate": 4.9061157023109855e-06, + "loss": 0.7047, + "step": 725 + }, + { + "epoch": 0.6013252346769741, + "grad_norm": 0.8722146153450012, + "learning_rate": 4.905816934439762e-06, + "loss": 0.7231, + "step": 726 + }, + { + "epoch": 0.6021535063500828, + "grad_norm": 0.8933189511299133, + "learning_rate": 4.905517701065389e-06, + "loss": 0.7029, + "step": 727 + }, + { + "epoch": 0.6029817780231916, + "grad_norm": 0.8863442540168762, + "learning_rate": 4.905218002245764e-06, + "loss": 0.6886, + "step": 728 + }, + { + "epoch": 0.6038100496963004, + "grad_norm": 0.8672969341278076, + "learning_rate": 4.904917838038877e-06, + "loss": 0.7048, + "step": 729 + }, + { + "epoch": 0.6046383213694092, + "grad_norm": 0.8892956972122192, + "learning_rate": 4.904617208502807e-06, + "loss": 0.7009, + "step": 730 + }, + { + "epoch": 0.6054665930425179, + "grad_norm": 0.9083442687988281, + "learning_rate": 4.904316113695723e-06, + "loss": 0.7169, + "step": 731 + }, + { + "epoch": 0.6062948647156268, + "grad_norm": 0.8885907530784607, + "learning_rate": 4.904014553675883e-06, + "loss": 0.6939, + "step": 732 + }, + { + "epoch": 0.6071231363887355, + "grad_norm": 0.8859131336212158, + "learning_rate": 4.903712528501637e-06, + "loss": 0.6881, + "step": 733 + }, + { + "epoch": 0.6079514080618443, + "grad_norm": 0.874146580696106, + "learning_rate": 4.903410038231425e-06, + "loss": 0.6857, + "step": 734 + }, + { + "epoch": 0.6087796797349531, + "grad_norm": 0.8719701170921326, + "learning_rate": 4.903107082923774e-06, + "loss": 0.6834, + "step": 735 + }, + { + "epoch": 0.6096079514080619, + "grad_norm": 0.9656700491905212, + "learning_rate": 4.902803662637304e-06, + "loss": 0.6933, + "step": 736 + }, + { + "epoch": 0.6104362230811706, + "grad_norm": 0.9022018313407898, + "learning_rate": 4.9024997774307235e-06, + "loss": 0.683, + "step": 737 + }, + { + "epoch": 0.6112644947542794, + "grad_norm": 0.8694130182266235, + "learning_rate": 4.902195427362832e-06, + "loss": 0.642, + "step": 738 + }, + { + "epoch": 0.6120927664273882, + "grad_norm": 0.9153549075126648, + "learning_rate": 4.901890612492519e-06, + "loss": 0.6674, + "step": 739 + }, + { + "epoch": 0.6129210381004969, + "grad_norm": 0.9431132674217224, + "learning_rate": 4.901585332878762e-06, + "loss": 0.6918, + "step": 740 + }, + { + "epoch": 0.6137493097736058, + "grad_norm": 0.881587028503418, + "learning_rate": 4.901279588580631e-06, + "loss": 0.687, + "step": 741 + }, + { + "epoch": 0.6145775814467145, + "grad_norm": 0.9289383888244629, + "learning_rate": 4.9009733796572834e-06, + "loss": 0.6789, + "step": 742 + }, + { + "epoch": 0.6154058531198233, + "grad_norm": 0.9058337211608887, + "learning_rate": 4.900666706167968e-06, + "loss": 0.6856, + "step": 743 + }, + { + "epoch": 0.616234124792932, + "grad_norm": 0.8999397158622742, + "learning_rate": 4.900359568172024e-06, + "loss": 0.7159, + "step": 744 + }, + { + "epoch": 0.6170623964660409, + "grad_norm": 0.895321249961853, + "learning_rate": 4.90005196572888e-06, + "loss": 0.7071, + "step": 745 + }, + { + "epoch": 0.6178906681391496, + "grad_norm": 0.9142102599143982, + "learning_rate": 4.899743898898054e-06, + "loss": 0.6918, + "step": 746 + }, + { + "epoch": 0.6187189398122585, + "grad_norm": 0.8867097496986389, + "learning_rate": 4.899435367739152e-06, + "loss": 0.6991, + "step": 747 + }, + { + "epoch": 0.6195472114853672, + "grad_norm": 0.8594791293144226, + "learning_rate": 4.899126372311876e-06, + "loss": 0.7153, + "step": 748 + }, + { + "epoch": 0.620375483158476, + "grad_norm": 0.8740563988685608, + "learning_rate": 4.8988169126760085e-06, + "loss": 0.6786, + "step": 749 + }, + { + "epoch": 0.6212037548315847, + "grad_norm": 0.8818353414535522, + "learning_rate": 4.898506988891432e-06, + "loss": 0.6965, + "step": 750 + }, + { + "epoch": 0.6220320265046936, + "grad_norm": 0.9459843635559082, + "learning_rate": 4.898196601018111e-06, + "loss": 0.6821, + "step": 751 + }, + { + "epoch": 0.6228602981778023, + "grad_norm": 0.923457682132721, + "learning_rate": 4.897885749116104e-06, + "loss": 0.6822, + "step": 752 + }, + { + "epoch": 0.6236885698509111, + "grad_norm": 0.9342297315597534, + "learning_rate": 4.897574433245557e-06, + "loss": 0.7001, + "step": 753 + }, + { + "epoch": 0.6245168415240199, + "grad_norm": 0.9327735900878906, + "learning_rate": 4.897262653466706e-06, + "loss": 0.6921, + "step": 754 + }, + { + "epoch": 0.6253451131971287, + "grad_norm": 0.8876991868019104, + "learning_rate": 4.8969504098398805e-06, + "loss": 0.7087, + "step": 755 + }, + { + "epoch": 0.6261733848702374, + "grad_norm": 0.8983528017997742, + "learning_rate": 4.896637702425493e-06, + "loss": 0.6824, + "step": 756 + }, + { + "epoch": 0.6270016565433462, + "grad_norm": 0.9074965715408325, + "learning_rate": 4.896324531284053e-06, + "loss": 0.6533, + "step": 757 + }, + { + "epoch": 0.627829928216455, + "grad_norm": 0.8938930630683899, + "learning_rate": 4.896010896476153e-06, + "loss": 0.6859, + "step": 758 + }, + { + "epoch": 0.6286581998895637, + "grad_norm": 0.9036867022514343, + "learning_rate": 4.8956967980624806e-06, + "loss": 0.6904, + "step": 759 + }, + { + "epoch": 0.6294864715626726, + "grad_norm": 0.9162903428077698, + "learning_rate": 4.895382236103809e-06, + "loss": 0.6995, + "step": 760 + }, + { + "epoch": 0.6303147432357813, + "grad_norm": 1.0000487565994263, + "learning_rate": 4.895067210661005e-06, + "loss": 0.671, + "step": 761 + }, + { + "epoch": 0.6311430149088901, + "grad_norm": 0.895163893699646, + "learning_rate": 4.894751721795022e-06, + "loss": 0.6871, + "step": 762 + }, + { + "epoch": 0.6319712865819989, + "grad_norm": 0.9401512742042542, + "learning_rate": 4.894435769566905e-06, + "loss": 0.6811, + "step": 763 + }, + { + "epoch": 0.6327995582551077, + "grad_norm": 0.924578845500946, + "learning_rate": 4.894119354037787e-06, + "loss": 0.692, + "step": 764 + }, + { + "epoch": 0.6336278299282164, + "grad_norm": 0.9114750623703003, + "learning_rate": 4.893802475268892e-06, + "loss": 0.6977, + "step": 765 + }, + { + "epoch": 0.6344561016013253, + "grad_norm": 1.1472550630569458, + "learning_rate": 4.893485133321533e-06, + "loss": 0.679, + "step": 766 + }, + { + "epoch": 0.635284373274434, + "grad_norm": 0.9460884928703308, + "learning_rate": 4.893167328257111e-06, + "loss": 0.6988, + "step": 767 + }, + { + "epoch": 0.6361126449475428, + "grad_norm": 0.9040783643722534, + "learning_rate": 4.8928490601371216e-06, + "loss": 0.6706, + "step": 768 + }, + { + "epoch": 0.6369409166206516, + "grad_norm": 0.9785428047180176, + "learning_rate": 4.892530329023144e-06, + "loss": 0.7185, + "step": 769 + }, + { + "epoch": 0.6377691882937604, + "grad_norm": 0.9795847535133362, + "learning_rate": 4.892211134976851e-06, + "loss": 0.703, + "step": 770 + }, + { + "epoch": 0.6385974599668691, + "grad_norm": 0.9539455771446228, + "learning_rate": 4.8918914780600025e-06, + "loss": 0.6985, + "step": 771 + }, + { + "epoch": 0.639425731639978, + "grad_norm": 0.9304025173187256, + "learning_rate": 4.8915713583344505e-06, + "loss": 0.6731, + "step": 772 + }, + { + "epoch": 0.6402540033130867, + "grad_norm": 0.9466667771339417, + "learning_rate": 4.891250775862134e-06, + "loss": 0.6972, + "step": 773 + }, + { + "epoch": 0.6410822749861955, + "grad_norm": 0.9522298574447632, + "learning_rate": 4.890929730705084e-06, + "loss": 0.6793, + "step": 774 + }, + { + "epoch": 0.6419105466593042, + "grad_norm": 0.9077293276786804, + "learning_rate": 4.890608222925419e-06, + "loss": 0.6892, + "step": 775 + }, + { + "epoch": 0.6427388183324131, + "grad_norm": 0.8763419389724731, + "learning_rate": 4.890286252585348e-06, + "loss": 0.6985, + "step": 776 + }, + { + "epoch": 0.6435670900055218, + "grad_norm": 0.9089516401290894, + "learning_rate": 4.8899638197471685e-06, + "loss": 0.6778, + "step": 777 + }, + { + "epoch": 0.6443953616786305, + "grad_norm": 0.8733325600624084, + "learning_rate": 4.88964092447327e-06, + "loss": 0.6915, + "step": 778 + }, + { + "epoch": 0.6452236333517394, + "grad_norm": 0.9472198486328125, + "learning_rate": 4.889317566826128e-06, + "loss": 0.6984, + "step": 779 + }, + { + "epoch": 0.6460519050248481, + "grad_norm": 0.8729666471481323, + "learning_rate": 4.888993746868309e-06, + "loss": 0.6724, + "step": 780 + }, + { + "epoch": 0.6468801766979569, + "grad_norm": 0.9395233988761902, + "learning_rate": 4.88866946466247e-06, + "loss": 0.6697, + "step": 781 + }, + { + "epoch": 0.6477084483710657, + "grad_norm": 0.9158744812011719, + "learning_rate": 4.888344720271357e-06, + "loss": 0.6838, + "step": 782 + }, + { + "epoch": 0.6485367200441745, + "grad_norm": 0.8961113095283508, + "learning_rate": 4.888019513757804e-06, + "loss": 0.6488, + "step": 783 + }, + { + "epoch": 0.6493649917172832, + "grad_norm": 0.8911786079406738, + "learning_rate": 4.8876938451847364e-06, + "loss": 0.7089, + "step": 784 + }, + { + "epoch": 0.6501932633903921, + "grad_norm": 0.9214697480201721, + "learning_rate": 4.887367714615167e-06, + "loss": 0.6929, + "step": 785 + }, + { + "epoch": 0.6510215350635008, + "grad_norm": 0.9161441326141357, + "learning_rate": 4.8870411221122e-06, + "loss": 0.6761, + "step": 786 + }, + { + "epoch": 0.6518498067366096, + "grad_norm": 0.9161947965621948, + "learning_rate": 4.886714067739028e-06, + "loss": 0.6804, + "step": 787 + }, + { + "epoch": 0.6526780784097184, + "grad_norm": 0.8706585168838501, + "learning_rate": 4.886386551558933e-06, + "loss": 0.6836, + "step": 788 + }, + { + "epoch": 0.6535063500828272, + "grad_norm": 0.9037661552429199, + "learning_rate": 4.886058573635284e-06, + "loss": 0.6786, + "step": 789 + }, + { + "epoch": 0.6543346217559359, + "grad_norm": 0.8977832794189453, + "learning_rate": 4.885730134031545e-06, + "loss": 0.6938, + "step": 790 + }, + { + "epoch": 0.6551628934290448, + "grad_norm": 0.9106955528259277, + "learning_rate": 4.8854012328112645e-06, + "loss": 0.678, + "step": 791 + }, + { + "epoch": 0.6559911651021535, + "grad_norm": 0.8769396543502808, + "learning_rate": 4.885071870038082e-06, + "loss": 0.6969, + "step": 792 + }, + { + "epoch": 0.6568194367752623, + "grad_norm": 0.9946926832199097, + "learning_rate": 4.884742045775725e-06, + "loss": 0.6799, + "step": 793 + }, + { + "epoch": 0.6576477084483711, + "grad_norm": 0.9399042725563049, + "learning_rate": 4.884411760088012e-06, + "loss": 0.7038, + "step": 794 + }, + { + "epoch": 0.6584759801214799, + "grad_norm": 0.8914628624916077, + "learning_rate": 4.8840810130388526e-06, + "loss": 0.6867, + "step": 795 + }, + { + "epoch": 0.6593042517945886, + "grad_norm": 0.8900558352470398, + "learning_rate": 4.88374980469224e-06, + "loss": 0.6823, + "step": 796 + }, + { + "epoch": 0.6601325234676974, + "grad_norm": 0.8608913421630859, + "learning_rate": 4.8834181351122615e-06, + "loss": 0.6705, + "step": 797 + }, + { + "epoch": 0.6609607951408062, + "grad_norm": 0.900460422039032, + "learning_rate": 4.883086004363092e-06, + "loss": 0.7029, + "step": 798 + }, + { + "epoch": 0.6617890668139149, + "grad_norm": 0.8932703733444214, + "learning_rate": 4.882753412508996e-06, + "loss": 0.6845, + "step": 799 + }, + { + "epoch": 0.6626173384870238, + "grad_norm": 0.8737874627113342, + "learning_rate": 4.882420359614325e-06, + "loss": 0.6979, + "step": 800 + }, + { + "epoch": 0.6634456101601325, + "grad_norm": 0.8805396556854248, + "learning_rate": 4.882086845743524e-06, + "loss": 0.6935, + "step": 801 + }, + { + "epoch": 0.6642738818332413, + "grad_norm": 0.8952041268348694, + "learning_rate": 4.881752870961123e-06, + "loss": 0.7137, + "step": 802 + }, + { + "epoch": 0.66510215350635, + "grad_norm": 0.889869213104248, + "learning_rate": 4.881418435331745e-06, + "loss": 0.6643, + "step": 803 + }, + { + "epoch": 0.6659304251794589, + "grad_norm": 0.8912904858589172, + "learning_rate": 4.881083538920098e-06, + "loss": 0.6712, + "step": 804 + }, + { + "epoch": 0.6667586968525676, + "grad_norm": 0.8943006992340088, + "learning_rate": 4.880748181790984e-06, + "loss": 0.7001, + "step": 805 + }, + { + "epoch": 0.6675869685256764, + "grad_norm": 0.9042482376098633, + "learning_rate": 4.880412364009289e-06, + "loss": 0.6744, + "step": 806 + }, + { + "epoch": 0.6684152401987852, + "grad_norm": 0.9052988886833191, + "learning_rate": 4.880076085639991e-06, + "loss": 0.72, + "step": 807 + }, + { + "epoch": 0.669243511871894, + "grad_norm": 0.9702905416488647, + "learning_rate": 4.879739346748158e-06, + "loss": 0.6871, + "step": 808 + }, + { + "epoch": 0.6700717835450027, + "grad_norm": 0.8741766810417175, + "learning_rate": 4.879402147398944e-06, + "loss": 0.7174, + "step": 809 + }, + { + "epoch": 0.6709000552181116, + "grad_norm": 0.9226665496826172, + "learning_rate": 4.879064487657596e-06, + "loss": 0.6991, + "step": 810 + }, + { + "epoch": 0.6717283268912203, + "grad_norm": 0.878304123878479, + "learning_rate": 4.878726367589446e-06, + "loss": 0.6925, + "step": 811 + }, + { + "epoch": 0.6725565985643291, + "grad_norm": 0.893649160861969, + "learning_rate": 4.878387787259919e-06, + "loss": 0.6689, + "step": 812 + }, + { + "epoch": 0.6733848702374379, + "grad_norm": 0.917861819267273, + "learning_rate": 4.878048746734526e-06, + "loss": 0.7171, + "step": 813 + }, + { + "epoch": 0.6742131419105467, + "grad_norm": 0.879375696182251, + "learning_rate": 4.877709246078869e-06, + "loss": 0.6662, + "step": 814 + }, + { + "epoch": 0.6750414135836554, + "grad_norm": 0.9098542928695679, + "learning_rate": 4.877369285358637e-06, + "loss": 0.6793, + "step": 815 + }, + { + "epoch": 0.6758696852567642, + "grad_norm": 0.9465262293815613, + "learning_rate": 4.877028864639609e-06, + "loss": 0.7065, + "step": 816 + }, + { + "epoch": 0.676697956929873, + "grad_norm": 0.9029957056045532, + "learning_rate": 4.876687983987654e-06, + "loss": 0.7008, + "step": 817 + }, + { + "epoch": 0.6775262286029817, + "grad_norm": 0.9073920845985413, + "learning_rate": 4.876346643468731e-06, + "loss": 0.6836, + "step": 818 + }, + { + "epoch": 0.6783545002760906, + "grad_norm": 0.9028745889663696, + "learning_rate": 4.876004843148884e-06, + "loss": 0.6965, + "step": 819 + }, + { + "epoch": 0.6791827719491993, + "grad_norm": 0.8745572566986084, + "learning_rate": 4.875662583094249e-06, + "loss": 0.6974, + "step": 820 + }, + { + "epoch": 0.6800110436223081, + "grad_norm": 0.9203384518623352, + "learning_rate": 4.875319863371048e-06, + "loss": 0.686, + "step": 821 + }, + { + "epoch": 0.6808393152954169, + "grad_norm": 0.9087005853652954, + "learning_rate": 4.874976684045597e-06, + "loss": 0.683, + "step": 822 + }, + { + "epoch": 0.6816675869685257, + "grad_norm": 0.9007741808891296, + "learning_rate": 4.874633045184297e-06, + "loss": 0.6938, + "step": 823 + }, + { + "epoch": 0.6824958586416344, + "grad_norm": 0.9854089617729187, + "learning_rate": 4.874288946853639e-06, + "loss": 0.6945, + "step": 824 + }, + { + "epoch": 0.6833241303147433, + "grad_norm": 0.9044802784919739, + "learning_rate": 4.8739443891202025e-06, + "loss": 0.714, + "step": 825 + }, + { + "epoch": 0.684152401987852, + "grad_norm": 0.9388754963874817, + "learning_rate": 4.873599372050656e-06, + "loss": 0.6838, + "step": 826 + }, + { + "epoch": 0.6849806736609608, + "grad_norm": 0.9215866327285767, + "learning_rate": 4.873253895711757e-06, + "loss": 0.6994, + "step": 827 + }, + { + "epoch": 0.6858089453340696, + "grad_norm": 0.8786854147911072, + "learning_rate": 4.872907960170353e-06, + "loss": 0.6883, + "step": 828 + }, + { + "epoch": 0.6866372170071784, + "grad_norm": 0.9032605290412903, + "learning_rate": 4.872561565493379e-06, + "loss": 0.6847, + "step": 829 + }, + { + "epoch": 0.6874654886802871, + "grad_norm": 0.8940309882164001, + "learning_rate": 4.872214711747858e-06, + "loss": 0.6877, + "step": 830 + }, + { + "epoch": 0.688293760353396, + "grad_norm": 0.8703469634056091, + "learning_rate": 4.8718673990009045e-06, + "loss": 0.6829, + "step": 831 + }, + { + "epoch": 0.6891220320265047, + "grad_norm": 0.8618217706680298, + "learning_rate": 4.871519627319719e-06, + "loss": 0.6987, + "step": 832 + }, + { + "epoch": 0.6899503036996135, + "grad_norm": 0.8871967196464539, + "learning_rate": 4.871171396771594e-06, + "loss": 0.6824, + "step": 833 + }, + { + "epoch": 0.6907785753727222, + "grad_norm": 0.9162507057189941, + "learning_rate": 4.8708227074239064e-06, + "loss": 0.7158, + "step": 834 + }, + { + "epoch": 0.6916068470458311, + "grad_norm": 0.8829584717750549, + "learning_rate": 4.8704735593441255e-06, + "loss": 0.6714, + "step": 835 + }, + { + "epoch": 0.6924351187189398, + "grad_norm": 0.9130700826644897, + "learning_rate": 4.8701239525998076e-06, + "loss": 0.6841, + "step": 836 + }, + { + "epoch": 0.6932633903920485, + "grad_norm": 0.8706396222114563, + "learning_rate": 4.8697738872586e-06, + "loss": 0.7162, + "step": 837 + }, + { + "epoch": 0.6940916620651574, + "grad_norm": 0.8906163573265076, + "learning_rate": 4.869423363388235e-06, + "loss": 0.6586, + "step": 838 + }, + { + "epoch": 0.6949199337382661, + "grad_norm": 0.9134309887886047, + "learning_rate": 4.869072381056537e-06, + "loss": 0.6842, + "step": 839 + }, + { + "epoch": 0.6957482054113749, + "grad_norm": 0.869674563407898, + "learning_rate": 4.8687209403314175e-06, + "loss": 0.7049, + "step": 840 + }, + { + "epoch": 0.6965764770844837, + "grad_norm": 0.8405192494392395, + "learning_rate": 4.8683690412808775e-06, + "loss": 0.6677, + "step": 841 + }, + { + "epoch": 0.6974047487575925, + "grad_norm": 0.9173605442047119, + "learning_rate": 4.868016683973005e-06, + "loss": 0.6784, + "step": 842 + }, + { + "epoch": 0.6982330204307012, + "grad_norm": 0.9220597743988037, + "learning_rate": 4.867663868475978e-06, + "loss": 0.6774, + "step": 843 + }, + { + "epoch": 0.6990612921038101, + "grad_norm": 0.8687174320220947, + "learning_rate": 4.867310594858065e-06, + "loss": 0.7094, + "step": 844 + }, + { + "epoch": 0.6998895637769188, + "grad_norm": 0.9161131381988525, + "learning_rate": 4.866956863187619e-06, + "loss": 0.6867, + "step": 845 + }, + { + "epoch": 0.7007178354500276, + "grad_norm": 0.9338458776473999, + "learning_rate": 4.866602673533084e-06, + "loss": 0.6783, + "step": 846 + }, + { + "epoch": 0.7015461071231364, + "grad_norm": 0.9268187284469604, + "learning_rate": 4.866248025962993e-06, + "loss": 0.6681, + "step": 847 + }, + { + "epoch": 0.7023743787962452, + "grad_norm": 0.8843579292297363, + "learning_rate": 4.865892920545967e-06, + "loss": 0.6854, + "step": 848 + }, + { + "epoch": 0.7032026504693539, + "grad_norm": 0.889299213886261, + "learning_rate": 4.865537357350716e-06, + "loss": 0.7055, + "step": 849 + }, + { + "epoch": 0.7040309221424628, + "grad_norm": 0.9062727093696594, + "learning_rate": 4.865181336446036e-06, + "loss": 0.6806, + "step": 850 + }, + { + "epoch": 0.7048591938155715, + "grad_norm": 0.9007996320724487, + "learning_rate": 4.864824857900817e-06, + "loss": 0.6937, + "step": 851 + }, + { + "epoch": 0.7056874654886803, + "grad_norm": 0.8883542418479919, + "learning_rate": 4.864467921784032e-06, + "loss": 0.6957, + "step": 852 + }, + { + "epoch": 0.7065157371617891, + "grad_norm": 0.9422078132629395, + "learning_rate": 4.864110528164746e-06, + "loss": 0.7025, + "step": 853 + }, + { + "epoch": 0.7073440088348979, + "grad_norm": 0.8961433172225952, + "learning_rate": 4.86375267711211e-06, + "loss": 0.6795, + "step": 854 + }, + { + "epoch": 0.7081722805080066, + "grad_norm": 0.925815761089325, + "learning_rate": 4.863394368695366e-06, + "loss": 0.6824, + "step": 855 + }, + { + "epoch": 0.7090005521811154, + "grad_norm": 0.9326804280281067, + "learning_rate": 4.863035602983843e-06, + "loss": 0.6829, + "step": 856 + }, + { + "epoch": 0.7098288238542242, + "grad_norm": 0.8784539699554443, + "learning_rate": 4.862676380046959e-06, + "loss": 0.6776, + "step": 857 + }, + { + "epoch": 0.7106570955273329, + "grad_norm": 0.8577290177345276, + "learning_rate": 4.86231669995422e-06, + "loss": 0.6573, + "step": 858 + }, + { + "epoch": 0.7114853672004418, + "grad_norm": 0.9261177778244019, + "learning_rate": 4.86195656277522e-06, + "loss": 0.6891, + "step": 859 + }, + { + "epoch": 0.7123136388735505, + "grad_norm": 0.9008646607398987, + "learning_rate": 4.861595968579643e-06, + "loss": 0.6714, + "step": 860 + }, + { + "epoch": 0.7131419105466593, + "grad_norm": 0.8729333877563477, + "learning_rate": 4.861234917437261e-06, + "loss": 0.6819, + "step": 861 + }, + { + "epoch": 0.713970182219768, + "grad_norm": 0.9006103873252869, + "learning_rate": 4.860873409417934e-06, + "loss": 0.6702, + "step": 862 + }, + { + "epoch": 0.7147984538928769, + "grad_norm": 0.8699921369552612, + "learning_rate": 4.860511444591609e-06, + "loss": 0.6681, + "step": 863 + }, + { + "epoch": 0.7156267255659856, + "grad_norm": 0.9359689950942993, + "learning_rate": 4.860149023028325e-06, + "loss": 0.6834, + "step": 864 + }, + { + "epoch": 0.7164549972390944, + "grad_norm": 0.874977707862854, + "learning_rate": 4.859786144798205e-06, + "loss": 0.6692, + "step": 865 + }, + { + "epoch": 0.7172832689122032, + "grad_norm": 0.8821915984153748, + "learning_rate": 4.859422809971464e-06, + "loss": 0.6865, + "step": 866 + }, + { + "epoch": 0.718111540585312, + "grad_norm": 0.905248761177063, + "learning_rate": 4.859059018618405e-06, + "loss": 0.6705, + "step": 867 + }, + { + "epoch": 0.7189398122584207, + "grad_norm": 0.9134542346000671, + "learning_rate": 4.858694770809415e-06, + "loss": 0.7039, + "step": 868 + }, + { + "epoch": 0.7197680839315296, + "grad_norm": 0.9048055410385132, + "learning_rate": 4.858330066614975e-06, + "loss": 0.6848, + "step": 869 + }, + { + "epoch": 0.7205963556046383, + "grad_norm": 0.8780005574226379, + "learning_rate": 4.857964906105651e-06, + "loss": 0.689, + "step": 870 + }, + { + "epoch": 0.7214246272777471, + "grad_norm": 0.8826956152915955, + "learning_rate": 4.857599289352098e-06, + "loss": 0.6799, + "step": 871 + }, + { + "epoch": 0.7222528989508559, + "grad_norm": 0.919905960559845, + "learning_rate": 4.857233216425061e-06, + "loss": 0.6587, + "step": 872 + }, + { + "epoch": 0.7230811706239647, + "grad_norm": 0.8711065053939819, + "learning_rate": 4.8568666873953696e-06, + "loss": 0.6754, + "step": 873 + }, + { + "epoch": 0.7239094422970734, + "grad_norm": 0.8740426898002625, + "learning_rate": 4.856499702333944e-06, + "loss": 0.6837, + "step": 874 + }, + { + "epoch": 0.7247377139701823, + "grad_norm": 0.921416699886322, + "learning_rate": 4.8561322613117945e-06, + "loss": 0.6709, + "step": 875 + }, + { + "epoch": 0.725565985643291, + "grad_norm": 0.8988221287727356, + "learning_rate": 4.855764364400016e-06, + "loss": 0.7111, + "step": 876 + }, + { + "epoch": 0.7263942573163997, + "grad_norm": 0.8838291168212891, + "learning_rate": 4.8553960116697915e-06, + "loss": 0.6821, + "step": 877 + }, + { + "epoch": 0.7272225289895086, + "grad_norm": 0.9051623940467834, + "learning_rate": 4.8550272031923965e-06, + "loss": 0.6848, + "step": 878 + }, + { + "epoch": 0.7280508006626173, + "grad_norm": 0.9308082461357117, + "learning_rate": 4.85465793903919e-06, + "loss": 0.7087, + "step": 879 + }, + { + "epoch": 0.7288790723357261, + "grad_norm": 0.873214066028595, + "learning_rate": 4.854288219281623e-06, + "loss": 0.6877, + "step": 880 + }, + { + "epoch": 0.7297073440088349, + "grad_norm": 0.8890054225921631, + "learning_rate": 4.853918043991232e-06, + "loss": 0.6719, + "step": 881 + }, + { + "epoch": 0.7305356156819437, + "grad_norm": 0.9580398201942444, + "learning_rate": 4.853547413239642e-06, + "loss": 0.6439, + "step": 882 + }, + { + "epoch": 0.7313638873550524, + "grad_norm": 0.875035285949707, + "learning_rate": 4.853176327098567e-06, + "loss": 0.6645, + "step": 883 + }, + { + "epoch": 0.7321921590281613, + "grad_norm": 0.9299217462539673, + "learning_rate": 4.85280478563981e-06, + "loss": 0.707, + "step": 884 + }, + { + "epoch": 0.73302043070127, + "grad_norm": 0.8840609192848206, + "learning_rate": 4.8524327889352585e-06, + "loss": 0.6789, + "step": 885 + }, + { + "epoch": 0.7338487023743788, + "grad_norm": 0.8983805179595947, + "learning_rate": 4.852060337056891e-06, + "loss": 0.6814, + "step": 886 + }, + { + "epoch": 0.7346769740474876, + "grad_norm": 0.8730597496032715, + "learning_rate": 4.851687430076775e-06, + "loss": 0.7006, + "step": 887 + }, + { + "epoch": 0.7355052457205964, + "grad_norm": 0.8980515003204346, + "learning_rate": 4.851314068067063e-06, + "loss": 0.6729, + "step": 888 + }, + { + "epoch": 0.7363335173937051, + "grad_norm": 0.8919468522071838, + "learning_rate": 4.850940251099997e-06, + "loss": 0.6764, + "step": 889 + }, + { + "epoch": 0.737161789066814, + "grad_norm": 0.8987070918083191, + "learning_rate": 4.850565979247908e-06, + "loss": 0.6997, + "step": 890 + }, + { + "epoch": 0.7379900607399227, + "grad_norm": 0.9056184887886047, + "learning_rate": 4.850191252583214e-06, + "loss": 0.6887, + "step": 891 + }, + { + "epoch": 0.7388183324130315, + "grad_norm": 0.8714621067047119, + "learning_rate": 4.84981607117842e-06, + "loss": 0.6946, + "step": 892 + }, + { + "epoch": 0.7396466040861402, + "grad_norm": 0.8845617175102234, + "learning_rate": 4.849440435106122e-06, + "loss": 0.6845, + "step": 893 + }, + { + "epoch": 0.7404748757592491, + "grad_norm": 0.8926048278808594, + "learning_rate": 4.849064344439e-06, + "loss": 0.6922, + "step": 894 + }, + { + "epoch": 0.7413031474323578, + "grad_norm": 0.8731569647789001, + "learning_rate": 4.848687799249826e-06, + "loss": 0.6837, + "step": 895 + }, + { + "epoch": 0.7421314191054665, + "grad_norm": 0.8801183104515076, + "learning_rate": 4.848310799611457e-06, + "loss": 0.6932, + "step": 896 + }, + { + "epoch": 0.7429596907785754, + "grad_norm": 0.8818139433860779, + "learning_rate": 4.847933345596838e-06, + "loss": 0.6758, + "step": 897 + }, + { + "epoch": 0.7437879624516841, + "grad_norm": 0.9117923974990845, + "learning_rate": 4.847555437279004e-06, + "loss": 0.6628, + "step": 898 + }, + { + "epoch": 0.7446162341247929, + "grad_norm": 0.891360342502594, + "learning_rate": 4.847177074731078e-06, + "loss": 0.6876, + "step": 899 + }, + { + "epoch": 0.7454445057979017, + "grad_norm": 0.8905263543128967, + "learning_rate": 4.846798258026267e-06, + "loss": 0.6862, + "step": 900 + }, + { + "epoch": 0.7462727774710105, + "grad_norm": 0.8853707909584045, + "learning_rate": 4.846418987237868e-06, + "loss": 0.654, + "step": 901 + }, + { + "epoch": 0.7471010491441192, + "grad_norm": 0.8782089948654175, + "learning_rate": 4.846039262439271e-06, + "loss": 0.7005, + "step": 902 + }, + { + "epoch": 0.7479293208172281, + "grad_norm": 0.8774383068084717, + "learning_rate": 4.8456590837039445e-06, + "loss": 0.7044, + "step": 903 + }, + { + "epoch": 0.7487575924903368, + "grad_norm": 0.9385423064231873, + "learning_rate": 4.845278451105451e-06, + "loss": 0.676, + "step": 904 + }, + { + "epoch": 0.7495858641634456, + "grad_norm": 0.9211432337760925, + "learning_rate": 4.8448973647174405e-06, + "loss": 0.697, + "step": 905 + }, + { + "epoch": 0.7504141358365544, + "grad_norm": 0.887610912322998, + "learning_rate": 4.844515824613649e-06, + "loss": 0.6965, + "step": 906 + }, + { + "epoch": 0.7512424075096632, + "grad_norm": 0.885761559009552, + "learning_rate": 4.8441338308679e-06, + "loss": 0.6851, + "step": 907 + }, + { + "epoch": 0.7520706791827719, + "grad_norm": 0.9209930300712585, + "learning_rate": 4.843751383554106e-06, + "loss": 0.6718, + "step": 908 + }, + { + "epoch": 0.7528989508558808, + "grad_norm": 0.9308252334594727, + "learning_rate": 4.843368482746269e-06, + "loss": 0.6802, + "step": 909 + }, + { + "epoch": 0.7537272225289895, + "grad_norm": 0.8931897282600403, + "learning_rate": 4.842985128518473e-06, + "loss": 0.7076, + "step": 910 + }, + { + "epoch": 0.7545554942020983, + "grad_norm": 0.9102733731269836, + "learning_rate": 4.842601320944898e-06, + "loss": 0.6898, + "step": 911 + }, + { + "epoch": 0.7553837658752071, + "grad_norm": 0.9336893558502197, + "learning_rate": 4.842217060099804e-06, + "loss": 0.6709, + "step": 912 + }, + { + "epoch": 0.7562120375483159, + "grad_norm": 0.9366991519927979, + "learning_rate": 4.8418323460575425e-06, + "loss": 0.6665, + "step": 913 + }, + { + "epoch": 0.7570403092214246, + "grad_norm": 0.9371415376663208, + "learning_rate": 4.841447178892552e-06, + "loss": 0.7013, + "step": 914 + }, + { + "epoch": 0.7578685808945335, + "grad_norm": 1.031123399734497, + "learning_rate": 4.84106155867936e-06, + "loss": 0.687, + "step": 915 + }, + { + "epoch": 0.7586968525676422, + "grad_norm": 0.9319218397140503, + "learning_rate": 4.8406754854925785e-06, + "loss": 0.669, + "step": 916 + }, + { + "epoch": 0.7595251242407509, + "grad_norm": 0.8563560843467712, + "learning_rate": 4.840288959406911e-06, + "loss": 0.6905, + "step": 917 + }, + { + "epoch": 0.7603533959138598, + "grad_norm": 0.9424378275871277, + "learning_rate": 4.839901980497145e-06, + "loss": 0.6685, + "step": 918 + }, + { + "epoch": 0.7611816675869685, + "grad_norm": 0.9202269315719604, + "learning_rate": 4.839514548838158e-06, + "loss": 0.6728, + "step": 919 + }, + { + "epoch": 0.7620099392600773, + "grad_norm": 0.9152582287788391, + "learning_rate": 4.839126664504915e-06, + "loss": 0.6842, + "step": 920 + }, + { + "epoch": 0.762838210933186, + "grad_norm": 0.9457129836082458, + "learning_rate": 4.838738327572468e-06, + "loss": 0.6822, + "step": 921 + }, + { + "epoch": 0.7636664826062949, + "grad_norm": 0.93124920129776, + "learning_rate": 4.838349538115955e-06, + "loss": 0.691, + "step": 922 + }, + { + "epoch": 0.7644947542794036, + "grad_norm": 0.9249301552772522, + "learning_rate": 4.837960296210605e-06, + "loss": 0.6574, + "step": 923 + }, + { + "epoch": 0.7653230259525124, + "grad_norm": 1.0551042556762695, + "learning_rate": 4.837570601931731e-06, + "loss": 0.6782, + "step": 924 + }, + { + "epoch": 0.7661512976256212, + "grad_norm": 0.9404964447021484, + "learning_rate": 4.8371804553547365e-06, + "loss": 0.6703, + "step": 925 + }, + { + "epoch": 0.76697956929873, + "grad_norm": 0.909015417098999, + "learning_rate": 4.836789856555111e-06, + "loss": 0.6812, + "step": 926 + }, + { + "epoch": 0.7678078409718387, + "grad_norm": 0.8572261929512024, + "learning_rate": 4.836398805608431e-06, + "loss": 0.6831, + "step": 927 + }, + { + "epoch": 0.7686361126449476, + "grad_norm": 0.951753556728363, + "learning_rate": 4.836007302590362e-06, + "loss": 0.6672, + "step": 928 + }, + { + "epoch": 0.7694643843180563, + "grad_norm": 0.8875932693481445, + "learning_rate": 4.8356153475766564e-06, + "loss": 0.6819, + "step": 929 + }, + { + "epoch": 0.7702926559911651, + "grad_norm": 0.9405681490898132, + "learning_rate": 4.835222940643153e-06, + "loss": 0.6917, + "step": 930 + }, + { + "epoch": 0.7711209276642739, + "grad_norm": 0.942520022392273, + "learning_rate": 4.834830081865779e-06, + "loss": 0.6657, + "step": 931 + }, + { + "epoch": 0.7719491993373827, + "grad_norm": 0.9063259363174438, + "learning_rate": 4.83443677132055e-06, + "loss": 0.6857, + "step": 932 + }, + { + "epoch": 0.7727774710104914, + "grad_norm": 0.8778199553489685, + "learning_rate": 4.834043009083566e-06, + "loss": 0.7028, + "step": 933 + }, + { + "epoch": 0.7736057426836003, + "grad_norm": 0.9215186834335327, + "learning_rate": 4.833648795231018e-06, + "loss": 0.6712, + "step": 934 + }, + { + "epoch": 0.774434014356709, + "grad_norm": 0.8992031812667847, + "learning_rate": 4.8332541298391825e-06, + "loss": 0.6827, + "step": 935 + }, + { + "epoch": 0.7752622860298177, + "grad_norm": 0.9203833341598511, + "learning_rate": 4.8328590129844224e-06, + "loss": 0.6835, + "step": 936 + }, + { + "epoch": 0.7760905577029266, + "grad_norm": 0.9109310507774353, + "learning_rate": 4.832463444743191e-06, + "loss": 0.673, + "step": 937 + }, + { + "epoch": 0.7769188293760353, + "grad_norm": 0.978349506855011, + "learning_rate": 4.832067425192024e-06, + "loss": 0.6638, + "step": 938 + }, + { + "epoch": 0.7777471010491441, + "grad_norm": 0.9552261829376221, + "learning_rate": 4.831670954407551e-06, + "loss": 0.6919, + "step": 939 + }, + { + "epoch": 0.7785753727222529, + "grad_norm": 0.9106539487838745, + "learning_rate": 4.831274032466484e-06, + "loss": 0.6964, + "step": 940 + }, + { + "epoch": 0.7794036443953617, + "grad_norm": 0.9195299744606018, + "learning_rate": 4.830876659445623e-06, + "loss": 0.6793, + "step": 941 + }, + { + "epoch": 0.7802319160684704, + "grad_norm": 0.9075512290000916, + "learning_rate": 4.8304788354218564e-06, + "loss": 0.6649, + "step": 942 + }, + { + "epoch": 0.7810601877415793, + "grad_norm": 0.9142526984214783, + "learning_rate": 4.83008056047216e-06, + "loss": 0.6811, + "step": 943 + }, + { + "epoch": 0.781888459414688, + "grad_norm": 0.8898900151252747, + "learning_rate": 4.829681834673597e-06, + "loss": 0.6906, + "step": 944 + }, + { + "epoch": 0.7827167310877968, + "grad_norm": 0.8884983062744141, + "learning_rate": 4.8292826581033135e-06, + "loss": 0.6903, + "step": 945 + }, + { + "epoch": 0.7835450027609056, + "grad_norm": 0.9075689315795898, + "learning_rate": 4.828883030838551e-06, + "loss": 0.6794, + "step": 946 + }, + { + "epoch": 0.7843732744340144, + "grad_norm": 0.8970300555229187, + "learning_rate": 4.828482952956631e-06, + "loss": 0.6715, + "step": 947 + }, + { + "epoch": 0.7852015461071231, + "grad_norm": 0.8906447887420654, + "learning_rate": 4.828082424534965e-06, + "loss": 0.6878, + "step": 948 + }, + { + "epoch": 0.786029817780232, + "grad_norm": 0.9083748459815979, + "learning_rate": 4.827681445651052e-06, + "loss": 0.6787, + "step": 949 + }, + { + "epoch": 0.7868580894533407, + "grad_norm": 0.8975276947021484, + "learning_rate": 4.8272800163824784e-06, + "loss": 0.687, + "step": 950 + }, + { + "epoch": 0.7876863611264495, + "grad_norm": 0.9573839902877808, + "learning_rate": 4.8268781368069156e-06, + "loss": 0.6996, + "step": 951 + }, + { + "epoch": 0.7885146327995582, + "grad_norm": 0.913536787033081, + "learning_rate": 4.826475807002125e-06, + "loss": 0.6844, + "step": 952 + }, + { + "epoch": 0.7893429044726671, + "grad_norm": 0.9202503561973572, + "learning_rate": 4.826073027045952e-06, + "loss": 0.6702, + "step": 953 + }, + { + "epoch": 0.7901711761457758, + "grad_norm": 0.8800237774848938, + "learning_rate": 4.8256697970163315e-06, + "loss": 0.685, + "step": 954 + }, + { + "epoch": 0.7909994478188846, + "grad_norm": 0.8923048377037048, + "learning_rate": 4.825266116991285e-06, + "loss": 0.6839, + "step": 955 + }, + { + "epoch": 0.7918277194919934, + "grad_norm": 0.879743754863739, + "learning_rate": 4.824861987048922e-06, + "loss": 0.7112, + "step": 956 + }, + { + "epoch": 0.7926559911651021, + "grad_norm": 0.8575169444084167, + "learning_rate": 4.824457407267436e-06, + "loss": 0.671, + "step": 957 + }, + { + "epoch": 0.7934842628382109, + "grad_norm": 0.8851709961891174, + "learning_rate": 4.824052377725109e-06, + "loss": 0.669, + "step": 958 + }, + { + "epoch": 0.7943125345113197, + "grad_norm": 0.9084329605102539, + "learning_rate": 4.823646898500311e-06, + "loss": 0.6731, + "step": 959 + }, + { + "epoch": 0.7951408061844285, + "grad_norm": 0.908256471157074, + "learning_rate": 4.823240969671501e-06, + "loss": 0.688, + "step": 960 + }, + { + "epoch": 0.7959690778575372, + "grad_norm": 0.8993384838104248, + "learning_rate": 4.822834591317219e-06, + "loss": 0.6736, + "step": 961 + }, + { + "epoch": 0.7967973495306461, + "grad_norm": 0.893530011177063, + "learning_rate": 4.822427763516098e-06, + "loss": 0.6787, + "step": 962 + }, + { + "epoch": 0.7976256212037548, + "grad_norm": 0.9479735493659973, + "learning_rate": 4.822020486346852e-06, + "loss": 0.6842, + "step": 963 + }, + { + "epoch": 0.7984538928768636, + "grad_norm": 0.8716689348220825, + "learning_rate": 4.821612759888289e-06, + "loss": 0.6726, + "step": 964 + }, + { + "epoch": 0.7992821645499724, + "grad_norm": 0.8902738690376282, + "learning_rate": 4.821204584219299e-06, + "loss": 0.6672, + "step": 965 + }, + { + "epoch": 0.8001104362230812, + "grad_norm": 0.8713969588279724, + "learning_rate": 4.820795959418859e-06, + "loss": 0.6783, + "step": 966 + }, + { + "epoch": 0.8009387078961899, + "grad_norm": 0.891788899898529, + "learning_rate": 4.820386885566036e-06, + "loss": 0.6691, + "step": 967 + }, + { + "epoch": 0.8017669795692988, + "grad_norm": 0.8484700918197632, + "learning_rate": 4.81997736273998e-06, + "loss": 0.6766, + "step": 968 + }, + { + "epoch": 0.8025952512424075, + "grad_norm": 0.8931223154067993, + "learning_rate": 4.819567391019931e-06, + "loss": 0.6944, + "step": 969 + }, + { + "epoch": 0.8034235229155163, + "grad_norm": 0.9707916975021362, + "learning_rate": 4.819156970485216e-06, + "loss": 0.7076, + "step": 970 + }, + { + "epoch": 0.8042517945886251, + "grad_norm": 0.9172869920730591, + "learning_rate": 4.818746101215245e-06, + "loss": 0.6864, + "step": 971 + }, + { + "epoch": 0.8050800662617339, + "grad_norm": 0.8982006311416626, + "learning_rate": 4.818334783289518e-06, + "loss": 0.6685, + "step": 972 + }, + { + "epoch": 0.8059083379348426, + "grad_norm": 0.8960995078086853, + "learning_rate": 4.817923016787623e-06, + "loss": 0.6959, + "step": 973 + }, + { + "epoch": 0.8067366096079515, + "grad_norm": 0.8806949257850647, + "learning_rate": 4.81751080178923e-06, + "loss": 0.7028, + "step": 974 + }, + { + "epoch": 0.8075648812810602, + "grad_norm": 0.8798978924751282, + "learning_rate": 4.817098138374102e-06, + "loss": 0.6718, + "step": 975 + }, + { + "epoch": 0.8083931529541689, + "grad_norm": 0.9217897653579712, + "learning_rate": 4.8166850266220835e-06, + "loss": 0.6782, + "step": 976 + }, + { + "epoch": 0.8092214246272778, + "grad_norm": 0.8915667533874512, + "learning_rate": 4.8162714666131074e-06, + "loss": 0.6917, + "step": 977 + }, + { + "epoch": 0.8100496963003865, + "grad_norm": 0.8895093202590942, + "learning_rate": 4.815857458427195e-06, + "loss": 0.698, + "step": 978 + }, + { + "epoch": 0.8108779679734953, + "grad_norm": 0.9167448282241821, + "learning_rate": 4.815443002144453e-06, + "loss": 0.6796, + "step": 979 + }, + { + "epoch": 0.811706239646604, + "grad_norm": 0.913827657699585, + "learning_rate": 4.815028097845074e-06, + "loss": 0.6731, + "step": 980 + }, + { + "epoch": 0.8125345113197129, + "grad_norm": 0.9183399081230164, + "learning_rate": 4.814612745609338e-06, + "loss": 0.7076, + "step": 981 + }, + { + "epoch": 0.8133627829928216, + "grad_norm": 0.8812996745109558, + "learning_rate": 4.814196945517613e-06, + "loss": 0.6416, + "step": 982 + }, + { + "epoch": 0.8141910546659304, + "grad_norm": 0.9032955169677734, + "learning_rate": 4.813780697650351e-06, + "loss": 0.6799, + "step": 983 + }, + { + "epoch": 0.8150193263390392, + "grad_norm": 0.9380100965499878, + "learning_rate": 4.813364002088093e-06, + "loss": 0.6771, + "step": 984 + }, + { + "epoch": 0.815847598012148, + "grad_norm": 0.9470316171646118, + "learning_rate": 4.8129468589114655e-06, + "loss": 0.6851, + "step": 985 + }, + { + "epoch": 0.8166758696852567, + "grad_norm": 0.9401133060455322, + "learning_rate": 4.812529268201183e-06, + "loss": 0.676, + "step": 986 + }, + { + "epoch": 0.8175041413583656, + "grad_norm": 0.9379092454910278, + "learning_rate": 4.812111230038043e-06, + "loss": 0.672, + "step": 987 + }, + { + "epoch": 0.8183324130314743, + "grad_norm": 0.9591637253761292, + "learning_rate": 4.811692744502933e-06, + "loss": 0.6684, + "step": 988 + }, + { + "epoch": 0.8191606847045831, + "grad_norm": 0.8911624550819397, + "learning_rate": 4.8112738116768276e-06, + "loss": 0.6448, + "step": 989 + }, + { + "epoch": 0.8199889563776919, + "grad_norm": 0.9036282896995544, + "learning_rate": 4.810854431640784e-06, + "loss": 0.703, + "step": 990 + }, + { + "epoch": 0.8208172280508007, + "grad_norm": 0.9197635650634766, + "learning_rate": 4.81043460447595e-06, + "loss": 0.6655, + "step": 991 + }, + { + "epoch": 0.8216454997239094, + "grad_norm": 0.9548643827438354, + "learning_rate": 4.8100143302635575e-06, + "loss": 0.6892, + "step": 992 + }, + { + "epoch": 0.8224737713970183, + "grad_norm": 0.9139766097068787, + "learning_rate": 4.809593609084925e-06, + "loss": 0.6723, + "step": 993 + }, + { + "epoch": 0.823302043070127, + "grad_norm": 0.9067809581756592, + "learning_rate": 4.8091724410214595e-06, + "loss": 0.6649, + "step": 994 + }, + { + "epoch": 0.8241303147432358, + "grad_norm": 0.9705583453178406, + "learning_rate": 4.808750826154652e-06, + "loss": 0.6889, + "step": 995 + }, + { + "epoch": 0.8249585864163446, + "grad_norm": 0.9664806127548218, + "learning_rate": 4.808328764566082e-06, + "loss": 0.6758, + "step": 996 + }, + { + "epoch": 0.8257868580894533, + "grad_norm": 0.8840530514717102, + "learning_rate": 4.807906256337414e-06, + "loss": 0.6719, + "step": 997 + }, + { + "epoch": 0.8266151297625621, + "grad_norm": 0.8803462386131287, + "learning_rate": 4.8074833015504e-06, + "loss": 0.6903, + "step": 998 + }, + { + "epoch": 0.8274434014356709, + "grad_norm": 0.9404571652412415, + "learning_rate": 4.8070599002868756e-06, + "loss": 0.6917, + "step": 999 + }, + { + "epoch": 0.8282716731087797, + "grad_norm": 1.0206997394561768, + "learning_rate": 4.806636052628768e-06, + "loss": 0.6854, + "step": 1000 + }, + { + "epoch": 0.8290999447818884, + "grad_norm": 0.9294790029525757, + "learning_rate": 4.806211758658086e-06, + "loss": 0.6791, + "step": 1001 + }, + { + "epoch": 0.8299282164549973, + "grad_norm": 0.874011754989624, + "learning_rate": 4.805787018456927e-06, + "loss": 0.6915, + "step": 1002 + }, + { + "epoch": 0.830756488128106, + "grad_norm": 0.9098061323165894, + "learning_rate": 4.805361832107474e-06, + "loss": 0.6643, + "step": 1003 + }, + { + "epoch": 0.8315847598012148, + "grad_norm": 0.9783986210823059, + "learning_rate": 4.804936199691997e-06, + "loss": 0.6853, + "step": 1004 + }, + { + "epoch": 0.8324130314743236, + "grad_norm": 0.9316959381103516, + "learning_rate": 4.804510121292852e-06, + "loss": 0.6696, + "step": 1005 + }, + { + "epoch": 0.8332413031474324, + "grad_norm": 0.8930666446685791, + "learning_rate": 4.80408359699248e-06, + "loss": 0.6805, + "step": 1006 + }, + { + "epoch": 0.8340695748205411, + "grad_norm": 0.9417555332183838, + "learning_rate": 4.803656626873412e-06, + "loss": 0.66, + "step": 1007 + }, + { + "epoch": 0.83489784649365, + "grad_norm": 0.9435075521469116, + "learning_rate": 4.80322921101826e-06, + "loss": 0.6664, + "step": 1008 + }, + { + "epoch": 0.8357261181667587, + "grad_norm": 0.9090076684951782, + "learning_rate": 4.8028013495097255e-06, + "loss": 0.6595, + "step": 1009 + }, + { + "epoch": 0.8365543898398675, + "grad_norm": 0.9024788737297058, + "learning_rate": 4.802373042430597e-06, + "loss": 0.6672, + "step": 1010 + }, + { + "epoch": 0.8373826615129762, + "grad_norm": 0.899813711643219, + "learning_rate": 4.801944289863747e-06, + "loss": 0.6743, + "step": 1011 + }, + { + "epoch": 0.8382109331860851, + "grad_norm": 0.887106716632843, + "learning_rate": 4.801515091892136e-06, + "loss": 0.6823, + "step": 1012 + }, + { + "epoch": 0.8390392048591938, + "grad_norm": 0.8690978288650513, + "learning_rate": 4.801085448598809e-06, + "loss": 0.6642, + "step": 1013 + }, + { + "epoch": 0.8398674765323026, + "grad_norm": 0.9131479859352112, + "learning_rate": 4.800655360066898e-06, + "loss": 0.6841, + "step": 1014 + }, + { + "epoch": 0.8406957482054114, + "grad_norm": 0.9031343460083008, + "learning_rate": 4.800224826379622e-06, + "loss": 0.6879, + "step": 1015 + }, + { + "epoch": 0.8415240198785201, + "grad_norm": 0.8914244174957275, + "learning_rate": 4.799793847620283e-06, + "loss": 0.6562, + "step": 1016 + }, + { + "epoch": 0.8423522915516289, + "grad_norm": 0.8692408800125122, + "learning_rate": 4.799362423872274e-06, + "loss": 0.6759, + "step": 1017 + }, + { + "epoch": 0.8431805632247377, + "grad_norm": 0.9511806964874268, + "learning_rate": 4.798930555219069e-06, + "loss": 0.6637, + "step": 1018 + }, + { + "epoch": 0.8440088348978465, + "grad_norm": 0.9556595683097839, + "learning_rate": 4.7984982417442325e-06, + "loss": 0.6925, + "step": 1019 + }, + { + "epoch": 0.8448371065709552, + "grad_norm": 0.9107236862182617, + "learning_rate": 4.798065483531413e-06, + "loss": 0.67, + "step": 1020 + }, + { + "epoch": 0.8456653782440641, + "grad_norm": 0.875293493270874, + "learning_rate": 4.797632280664345e-06, + "loss": 0.6755, + "step": 1021 + }, + { + "epoch": 0.8464936499171728, + "grad_norm": 0.8972271084785461, + "learning_rate": 4.797198633226849e-06, + "loss": 0.6637, + "step": 1022 + }, + { + "epoch": 0.8473219215902816, + "grad_norm": 0.8903595805168152, + "learning_rate": 4.796764541302831e-06, + "loss": 0.6997, + "step": 1023 + }, + { + "epoch": 0.8481501932633904, + "grad_norm": 0.8948636651039124, + "learning_rate": 4.796330004976285e-06, + "loss": 0.6579, + "step": 1024 + }, + { + "epoch": 0.8489784649364992, + "grad_norm": 0.8590925335884094, + "learning_rate": 4.795895024331289e-06, + "loss": 0.6651, + "step": 1025 + }, + { + "epoch": 0.8498067366096079, + "grad_norm": 0.870811402797699, + "learning_rate": 4.795459599452007e-06, + "loss": 0.6707, + "step": 1026 + }, + { + "epoch": 0.8506350082827168, + "grad_norm": 0.8807471394538879, + "learning_rate": 4.795023730422692e-06, + "loss": 0.6787, + "step": 1027 + }, + { + "epoch": 0.8514632799558255, + "grad_norm": 0.8817671537399292, + "learning_rate": 4.794587417327678e-06, + "loss": 0.66, + "step": 1028 + }, + { + "epoch": 0.8522915516289343, + "grad_norm": 0.8924570679664612, + "learning_rate": 4.794150660251389e-06, + "loss": 0.6842, + "step": 1029 + }, + { + "epoch": 0.8531198233020431, + "grad_norm": 0.8746559619903564, + "learning_rate": 4.793713459278333e-06, + "loss": 0.6582, + "step": 1030 + }, + { + "epoch": 0.8539480949751519, + "grad_norm": 0.8885277509689331, + "learning_rate": 4.793275814493104e-06, + "loss": 0.6784, + "step": 1031 + }, + { + "epoch": 0.8547763666482606, + "grad_norm": 0.8797710537910461, + "learning_rate": 4.792837725980383e-06, + "loss": 0.6638, + "step": 1032 + }, + { + "epoch": 0.8556046383213695, + "grad_norm": 0.8932235240936279, + "learning_rate": 4.7923991938249355e-06, + "loss": 0.6918, + "step": 1033 + }, + { + "epoch": 0.8564329099944782, + "grad_norm": 0.9317487478256226, + "learning_rate": 4.791960218111613e-06, + "loss": 0.6905, + "step": 1034 + }, + { + "epoch": 0.8572611816675869, + "grad_norm": 0.8760019540786743, + "learning_rate": 4.791520798925354e-06, + "loss": 0.673, + "step": 1035 + }, + { + "epoch": 0.8580894533406958, + "grad_norm": 0.8644204139709473, + "learning_rate": 4.7910809363511825e-06, + "loss": 0.6755, + "step": 1036 + }, + { + "epoch": 0.8589177250138045, + "grad_norm": 0.8761067390441895, + "learning_rate": 4.790640630474206e-06, + "loss": 0.699, + "step": 1037 + }, + { + "epoch": 0.8597459966869133, + "grad_norm": 0.9135014414787292, + "learning_rate": 4.790199881379621e-06, + "loss": 0.6659, + "step": 1038 + }, + { + "epoch": 0.860574268360022, + "grad_norm": 0.9065434336662292, + "learning_rate": 4.789758689152708e-06, + "loss": 0.6664, + "step": 1039 + }, + { + "epoch": 0.8614025400331309, + "grad_norm": 0.8814099431037903, + "learning_rate": 4.789317053878833e-06, + "loss": 0.681, + "step": 1040 + }, + { + "epoch": 0.8622308117062396, + "grad_norm": 0.872502863407135, + "learning_rate": 4.788874975643449e-06, + "loss": 0.7044, + "step": 1041 + }, + { + "epoch": 0.8630590833793484, + "grad_norm": 0.9532367587089539, + "learning_rate": 4.788432454532093e-06, + "loss": 0.655, + "step": 1042 + }, + { + "epoch": 0.8638873550524572, + "grad_norm": 0.9009373784065247, + "learning_rate": 4.787989490630391e-06, + "loss": 0.6946, + "step": 1043 + }, + { + "epoch": 0.864715626725566, + "grad_norm": 0.889933168888092, + "learning_rate": 4.787546084024051e-06, + "loss": 0.6595, + "step": 1044 + }, + { + "epoch": 0.8655438983986747, + "grad_norm": 0.9179873466491699, + "learning_rate": 4.787102234798867e-06, + "loss": 0.658, + "step": 1045 + }, + { + "epoch": 0.8663721700717836, + "grad_norm": 0.8873292207717896, + "learning_rate": 4.786657943040721e-06, + "loss": 0.7023, + "step": 1046 + }, + { + "epoch": 0.8672004417448923, + "grad_norm": 0.8388610482215881, + "learning_rate": 4.78621320883558e-06, + "loss": 0.6644, + "step": 1047 + }, + { + "epoch": 0.8680287134180011, + "grad_norm": 0.8971349596977234, + "learning_rate": 4.785768032269493e-06, + "loss": 0.6552, + "step": 1048 + }, + { + "epoch": 0.8688569850911099, + "grad_norm": 0.8789131045341492, + "learning_rate": 4.785322413428601e-06, + "loss": 0.6681, + "step": 1049 + }, + { + "epoch": 0.8696852567642187, + "grad_norm": 0.8867958784103394, + "learning_rate": 4.784876352399125e-06, + "loss": 0.6859, + "step": 1050 + }, + { + "epoch": 0.8705135284373274, + "grad_norm": 0.8584237694740295, + "learning_rate": 4.784429849267373e-06, + "loss": 0.6655, + "step": 1051 + }, + { + "epoch": 0.8713418001104363, + "grad_norm": 0.9230811595916748, + "learning_rate": 4.783982904119743e-06, + "loss": 0.6648, + "step": 1052 + }, + { + "epoch": 0.872170071783545, + "grad_norm": 0.9217789769172668, + "learning_rate": 4.783535517042709e-06, + "loss": 0.6867, + "step": 1053 + }, + { + "epoch": 0.8729983434566538, + "grad_norm": 0.9209116697311401, + "learning_rate": 4.783087688122842e-06, + "loss": 0.6972, + "step": 1054 + }, + { + "epoch": 0.8738266151297626, + "grad_norm": 0.9493632316589355, + "learning_rate": 4.782639417446789e-06, + "loss": 0.6785, + "step": 1055 + }, + { + "epoch": 0.8746548868028713, + "grad_norm": 0.8790786862373352, + "learning_rate": 4.782190705101286e-06, + "loss": 0.6925, + "step": 1056 + }, + { + "epoch": 0.8754831584759801, + "grad_norm": 0.8662691712379456, + "learning_rate": 4.781741551173157e-06, + "loss": 0.6686, + "step": 1057 + }, + { + "epoch": 0.8763114301490889, + "grad_norm": 0.919658899307251, + "learning_rate": 4.781291955749307e-06, + "loss": 0.6814, + "step": 1058 + }, + { + "epoch": 0.8771397018221977, + "grad_norm": 0.9639708399772644, + "learning_rate": 4.780841918916731e-06, + "loss": 0.6629, + "step": 1059 + }, + { + "epoch": 0.8779679734953064, + "grad_norm": 0.8694306015968323, + "learning_rate": 4.780391440762505e-06, + "loss": 0.6893, + "step": 1060 + }, + { + "epoch": 0.8787962451684153, + "grad_norm": 0.9725603461265564, + "learning_rate": 4.779940521373793e-06, + "loss": 0.67, + "step": 1061 + }, + { + "epoch": 0.879624516841524, + "grad_norm": 0.93159419298172, + "learning_rate": 4.779489160837842e-06, + "loss": 0.686, + "step": 1062 + }, + { + "epoch": 0.8804527885146328, + "grad_norm": 0.9366991519927979, + "learning_rate": 4.779037359241989e-06, + "loss": 0.649, + "step": 1063 + }, + { + "epoch": 0.8812810601877415, + "grad_norm": 0.9052289724349976, + "learning_rate": 4.7785851166736505e-06, + "loss": 0.6677, + "step": 1064 + }, + { + "epoch": 0.8821093318608504, + "grad_norm": 0.882251501083374, + "learning_rate": 4.778132433220334e-06, + "loss": 0.6886, + "step": 1065 + }, + { + "epoch": 0.8829376035339591, + "grad_norm": 0.8867698907852173, + "learning_rate": 4.777679308969627e-06, + "loss": 0.6645, + "step": 1066 + }, + { + "epoch": 0.883765875207068, + "grad_norm": 0.890396773815155, + "learning_rate": 4.777225744009207e-06, + "loss": 0.6684, + "step": 1067 + }, + { + "epoch": 0.8845941468801767, + "grad_norm": 0.8827376365661621, + "learning_rate": 4.776771738426833e-06, + "loss": 0.6738, + "step": 1068 + }, + { + "epoch": 0.8854224185532855, + "grad_norm": 0.8953343033790588, + "learning_rate": 4.776317292310352e-06, + "loss": 0.684, + "step": 1069 + }, + { + "epoch": 0.8862506902263942, + "grad_norm": 0.8751050233840942, + "learning_rate": 4.775862405747694e-06, + "loss": 0.6779, + "step": 1070 + }, + { + "epoch": 0.8870789618995031, + "grad_norm": 0.9136618971824646, + "learning_rate": 4.775407078826876e-06, + "loss": 0.6939, + "step": 1071 + }, + { + "epoch": 0.8879072335726118, + "grad_norm": 0.8887941837310791, + "learning_rate": 4.774951311635999e-06, + "loss": 0.6547, + "step": 1072 + }, + { + "epoch": 0.8887355052457206, + "grad_norm": 0.8873274326324463, + "learning_rate": 4.77449510426325e-06, + "loss": 0.6769, + "step": 1073 + }, + { + "epoch": 0.8895637769188294, + "grad_norm": 0.9307074546813965, + "learning_rate": 4.774038456796901e-06, + "loss": 0.6984, + "step": 1074 + }, + { + "epoch": 0.8903920485919381, + "grad_norm": 0.9022747874259949, + "learning_rate": 4.77358136932531e-06, + "loss": 0.6733, + "step": 1075 + }, + { + "epoch": 0.8912203202650469, + "grad_norm": 0.8779094219207764, + "learning_rate": 4.773123841936916e-06, + "loss": 0.6434, + "step": 1076 + }, + { + "epoch": 0.8920485919381557, + "grad_norm": 0.9465747475624084, + "learning_rate": 4.7726658747202505e-06, + "loss": 0.6761, + "step": 1077 + }, + { + "epoch": 0.8928768636112645, + "grad_norm": 0.8694362044334412, + "learning_rate": 4.772207467763923e-06, + "loss": 0.6756, + "step": 1078 + }, + { + "epoch": 0.8937051352843732, + "grad_norm": 0.8839150071144104, + "learning_rate": 4.771748621156631e-06, + "loss": 0.6714, + "step": 1079 + }, + { + "epoch": 0.8945334069574821, + "grad_norm": 0.9220975041389465, + "learning_rate": 4.7712893349871585e-06, + "loss": 0.6724, + "step": 1080 + }, + { + "epoch": 0.8953616786305908, + "grad_norm": 0.9234760999679565, + "learning_rate": 4.770829609344372e-06, + "loss": 0.67, + "step": 1081 + }, + { + "epoch": 0.8961899503036996, + "grad_norm": 0.9262274503707886, + "learning_rate": 4.770369444317225e-06, + "loss": 0.6619, + "step": 1082 + }, + { + "epoch": 0.8970182219768084, + "grad_norm": 0.8889190554618835, + "learning_rate": 4.769908839994753e-06, + "loss": 0.688, + "step": 1083 + }, + { + "epoch": 0.8978464936499172, + "grad_norm": 0.9028982520103455, + "learning_rate": 4.769447796466082e-06, + "loss": 0.6633, + "step": 1084 + }, + { + "epoch": 0.8986747653230259, + "grad_norm": 0.9155063033103943, + "learning_rate": 4.7689863138204185e-06, + "loss": 0.6709, + "step": 1085 + }, + { + "epoch": 0.8995030369961348, + "grad_norm": 0.9066855311393738, + "learning_rate": 4.768524392147052e-06, + "loss": 0.6713, + "step": 1086 + }, + { + "epoch": 0.9003313086692435, + "grad_norm": 0.9365159273147583, + "learning_rate": 4.768062031535365e-06, + "loss": 0.6675, + "step": 1087 + }, + { + "epoch": 0.9011595803423523, + "grad_norm": 0.9106162190437317, + "learning_rate": 4.767599232074816e-06, + "loss": 0.6804, + "step": 1088 + }, + { + "epoch": 0.9019878520154611, + "grad_norm": 0.9183253645896912, + "learning_rate": 4.767135993854955e-06, + "loss": 0.658, + "step": 1089 + }, + { + "epoch": 0.9028161236885699, + "grad_norm": 0.9543409943580627, + "learning_rate": 4.766672316965413e-06, + "loss": 0.696, + "step": 1090 + }, + { + "epoch": 0.9036443953616786, + "grad_norm": 0.8894531726837158, + "learning_rate": 4.766208201495906e-06, + "loss": 0.6781, + "step": 1091 + }, + { + "epoch": 0.9044726670347875, + "grad_norm": 0.915542721748352, + "learning_rate": 4.765743647536238e-06, + "loss": 0.6804, + "step": 1092 + }, + { + "epoch": 0.9053009387078962, + "grad_norm": 0.8959763646125793, + "learning_rate": 4.765278655176296e-06, + "loss": 0.6873, + "step": 1093 + }, + { + "epoch": 0.906129210381005, + "grad_norm": 0.881188690662384, + "learning_rate": 4.76481322450605e-06, + "loss": 0.6582, + "step": 1094 + }, + { + "epoch": 0.9069574820541138, + "grad_norm": 0.9096754193305969, + "learning_rate": 4.764347355615558e-06, + "loss": 0.693, + "step": 1095 + }, + { + "epoch": 0.9077857537272225, + "grad_norm": 0.891132652759552, + "learning_rate": 4.76388104859496e-06, + "loss": 0.6781, + "step": 1096 + }, + { + "epoch": 0.9086140254003313, + "grad_norm": 0.893624484539032, + "learning_rate": 4.763414303534484e-06, + "loss": 0.6636, + "step": 1097 + }, + { + "epoch": 0.90944229707344, + "grad_norm": 0.8795678019523621, + "learning_rate": 4.762947120524438e-06, + "loss": 0.6604, + "step": 1098 + }, + { + "epoch": 0.9102705687465489, + "grad_norm": 0.9177265763282776, + "learning_rate": 4.76247949965522e-06, + "loss": 0.6557, + "step": 1099 + }, + { + "epoch": 0.9110988404196576, + "grad_norm": 0.9416985511779785, + "learning_rate": 4.762011441017309e-06, + "loss": 0.7008, + "step": 1100 + }, + { + "epoch": 0.9119271120927664, + "grad_norm": 0.9255479574203491, + "learning_rate": 4.7615429447012704e-06, + "loss": 0.6665, + "step": 1101 + }, + { + "epoch": 0.9127553837658752, + "grad_norm": 0.8948953151702881, + "learning_rate": 4.761074010797754e-06, + "loss": 0.6699, + "step": 1102 + }, + { + "epoch": 0.913583655438984, + "grad_norm": 0.9027442336082458, + "learning_rate": 4.760604639397493e-06, + "loss": 0.6649, + "step": 1103 + }, + { + "epoch": 0.9144119271120927, + "grad_norm": 0.9094238877296448, + "learning_rate": 4.7601348305913085e-06, + "loss": 0.6753, + "step": 1104 + }, + { + "epoch": 0.9152401987852016, + "grad_norm": 0.9098204374313354, + "learning_rate": 4.759664584470103e-06, + "loss": 0.6845, + "step": 1105 + }, + { + "epoch": 0.9160684704583103, + "grad_norm": 0.9339573383331299, + "learning_rate": 4.759193901124864e-06, + "loss": 0.6285, + "step": 1106 + }, + { + "epoch": 0.9168967421314191, + "grad_norm": 0.8812451958656311, + "learning_rate": 4.758722780646666e-06, + "loss": 0.6764, + "step": 1107 + }, + { + "epoch": 0.9177250138045279, + "grad_norm": 0.8832173347473145, + "learning_rate": 4.758251223126665e-06, + "loss": 0.6908, + "step": 1108 + }, + { + "epoch": 0.9185532854776367, + "grad_norm": 0.9124787449836731, + "learning_rate": 4.757779228656103e-06, + "loss": 0.6855, + "step": 1109 + }, + { + "epoch": 0.9193815571507454, + "grad_norm": 0.8778328895568848, + "learning_rate": 4.757306797326309e-06, + "loss": 0.6404, + "step": 1110 + }, + { + "epoch": 0.9202098288238543, + "grad_norm": 0.8831816911697388, + "learning_rate": 4.75683392922869e-06, + "loss": 0.6425, + "step": 1111 + }, + { + "epoch": 0.921038100496963, + "grad_norm": 0.9060006141662598, + "learning_rate": 4.756360624454744e-06, + "loss": 0.6738, + "step": 1112 + }, + { + "epoch": 0.9218663721700718, + "grad_norm": 0.8746777176856995, + "learning_rate": 4.755886883096051e-06, + "loss": 0.6665, + "step": 1113 + }, + { + "epoch": 0.9226946438431806, + "grad_norm": 0.9185945987701416, + "learning_rate": 4.755412705244274e-06, + "loss": 0.6564, + "step": 1114 + }, + { + "epoch": 0.9235229155162893, + "grad_norm": 0.9648581743240356, + "learning_rate": 4.754938090991165e-06, + "loss": 0.6597, + "step": 1115 + }, + { + "epoch": 0.9243511871893981, + "grad_norm": 0.8800092935562134, + "learning_rate": 4.754463040428555e-06, + "loss": 0.6884, + "step": 1116 + }, + { + "epoch": 0.9251794588625069, + "grad_norm": 0.8659892082214355, + "learning_rate": 4.753987553648363e-06, + "loss": 0.68, + "step": 1117 + }, + { + "epoch": 0.9260077305356157, + "grad_norm": 0.88991379737854, + "learning_rate": 4.75351163074259e-06, + "loss": 0.668, + "step": 1118 + }, + { + "epoch": 0.9268360022087244, + "grad_norm": 0.9683054089546204, + "learning_rate": 4.753035271803323e-06, + "loss": 0.6789, + "step": 1119 + }, + { + "epoch": 0.9276642738818333, + "grad_norm": 0.928001880645752, + "learning_rate": 4.752558476922735e-06, + "loss": 0.6786, + "step": 1120 + }, + { + "epoch": 0.928492545554942, + "grad_norm": 0.8760973215103149, + "learning_rate": 4.752081246193078e-06, + "loss": 0.6716, + "step": 1121 + }, + { + "epoch": 0.9293208172280508, + "grad_norm": 0.9171329140663147, + "learning_rate": 4.751603579706695e-06, + "loss": 0.662, + "step": 1122 + }, + { + "epoch": 0.9301490889011595, + "grad_norm": 0.9380231499671936, + "learning_rate": 4.751125477556008e-06, + "loss": 0.6836, + "step": 1123 + }, + { + "epoch": 0.9309773605742684, + "grad_norm": 0.9471802711486816, + "learning_rate": 4.750646939833526e-06, + "loss": 0.668, + "step": 1124 + }, + { + "epoch": 0.9318056322473771, + "grad_norm": 0.8860809206962585, + "learning_rate": 4.750167966631842e-06, + "loss": 0.6635, + "step": 1125 + }, + { + "epoch": 0.932633903920486, + "grad_norm": 0.8694644570350647, + "learning_rate": 4.7496885580436305e-06, + "loss": 0.6505, + "step": 1126 + }, + { + "epoch": 0.9334621755935947, + "grad_norm": 0.9602686166763306, + "learning_rate": 4.749208714161655e-06, + "loss": 0.6874, + "step": 1127 + }, + { + "epoch": 0.9342904472667035, + "grad_norm": 0.9198856353759766, + "learning_rate": 4.748728435078761e-06, + "loss": 0.697, + "step": 1128 + }, + { + "epoch": 0.9351187189398122, + "grad_norm": 0.9162499904632568, + "learning_rate": 4.748247720887876e-06, + "loss": 0.6521, + "step": 1129 + }, + { + "epoch": 0.9359469906129211, + "grad_norm": 0.9064576029777527, + "learning_rate": 4.747766571682016e-06, + "loss": 0.6525, + "step": 1130 + }, + { + "epoch": 0.9367752622860298, + "grad_norm": 0.9518057703971863, + "learning_rate": 4.747284987554277e-06, + "loss": 0.6745, + "step": 1131 + }, + { + "epoch": 0.9376035339591386, + "grad_norm": 0.8963000178337097, + "learning_rate": 4.746802968597842e-06, + "loss": 0.6999, + "step": 1132 + }, + { + "epoch": 0.9384318056322474, + "grad_norm": 0.8687212467193604, + "learning_rate": 4.746320514905977e-06, + "loss": 0.6353, + "step": 1133 + }, + { + "epoch": 0.9392600773053562, + "grad_norm": 0.8598489165306091, + "learning_rate": 4.745837626572033e-06, + "loss": 0.6897, + "step": 1134 + }, + { + "epoch": 0.9400883489784649, + "grad_norm": 0.9425498843193054, + "learning_rate": 4.745354303689444e-06, + "loss": 0.667, + "step": 1135 + }, + { + "epoch": 0.9409166206515737, + "grad_norm": 0.9026349782943726, + "learning_rate": 4.744870546351727e-06, + "loss": 0.676, + "step": 1136 + }, + { + "epoch": 0.9417448923246825, + "grad_norm": 0.8824150562286377, + "learning_rate": 4.744386354652486e-06, + "loss": 0.6941, + "step": 1137 + }, + { + "epoch": 0.9425731639977912, + "grad_norm": 0.8910468816757202, + "learning_rate": 4.743901728685406e-06, + "loss": 0.6701, + "step": 1138 + }, + { + "epoch": 0.9434014356709001, + "grad_norm": 0.9114013314247131, + "learning_rate": 4.74341666854426e-06, + "loss": 0.6735, + "step": 1139 + }, + { + "epoch": 0.9442297073440088, + "grad_norm": 0.9107307195663452, + "learning_rate": 4.742931174322902e-06, + "loss": 0.6505, + "step": 1140 + }, + { + "epoch": 0.9450579790171176, + "grad_norm": 0.8906923532485962, + "learning_rate": 4.742445246115271e-06, + "loss": 0.6648, + "step": 1141 + }, + { + "epoch": 0.9458862506902264, + "grad_norm": 0.902644157409668, + "learning_rate": 4.741958884015387e-06, + "loss": 0.6674, + "step": 1142 + }, + { + "epoch": 0.9467145223633352, + "grad_norm": 0.9522087574005127, + "learning_rate": 4.741472088117358e-06, + "loss": 0.6825, + "step": 1143 + }, + { + "epoch": 0.9475427940364439, + "grad_norm": 0.9152255654335022, + "learning_rate": 4.740984858515376e-06, + "loss": 0.6708, + "step": 1144 + }, + { + "epoch": 0.9483710657095528, + "grad_norm": 0.8779835104942322, + "learning_rate": 4.740497195303713e-06, + "loss": 0.6562, + "step": 1145 + }, + { + "epoch": 0.9491993373826615, + "grad_norm": 0.8842780590057373, + "learning_rate": 4.740009098576729e-06, + "loss": 0.6769, + "step": 1146 + }, + { + "epoch": 0.9500276090557703, + "grad_norm": 0.9137415289878845, + "learning_rate": 4.739520568428866e-06, + "loss": 0.6705, + "step": 1147 + }, + { + "epoch": 0.950855880728879, + "grad_norm": 0.8867765665054321, + "learning_rate": 4.739031604954648e-06, + "loss": 0.659, + "step": 1148 + }, + { + "epoch": 0.9516841524019879, + "grad_norm": 0.8965260982513428, + "learning_rate": 4.7385422082486874e-06, + "loss": 0.6744, + "step": 1149 + }, + { + "epoch": 0.9525124240750966, + "grad_norm": 0.8861697912216187, + "learning_rate": 4.738052378405677e-06, + "loss": 0.6575, + "step": 1150 + }, + { + "epoch": 0.9533406957482055, + "grad_norm": 0.9442310333251953, + "learning_rate": 4.737562115520394e-06, + "loss": 0.7164, + "step": 1151 + }, + { + "epoch": 0.9541689674213142, + "grad_norm": 0.9340727925300598, + "learning_rate": 4.7370714196877e-06, + "loss": 0.6577, + "step": 1152 + }, + { + "epoch": 0.954997239094423, + "grad_norm": 0.9337487816810608, + "learning_rate": 4.7365802910025405e-06, + "loss": 0.6664, + "step": 1153 + }, + { + "epoch": 0.9558255107675317, + "grad_norm": 0.8849001526832581, + "learning_rate": 4.736088729559943e-06, + "loss": 0.6721, + "step": 1154 + }, + { + "epoch": 0.9566537824406405, + "grad_norm": 0.917323648929596, + "learning_rate": 4.735596735455022e-06, + "loss": 0.6715, + "step": 1155 + }, + { + "epoch": 0.9574820541137493, + "grad_norm": 0.9480959177017212, + "learning_rate": 4.735104308782972e-06, + "loss": 0.6784, + "step": 1156 + }, + { + "epoch": 0.958310325786858, + "grad_norm": 0.9519251585006714, + "learning_rate": 4.734611449639073e-06, + "loss": 0.6617, + "step": 1157 + }, + { + "epoch": 0.9591385974599669, + "grad_norm": 0.8652079105377197, + "learning_rate": 4.734118158118689e-06, + "loss": 0.6846, + "step": 1158 + }, + { + "epoch": 0.9599668691330756, + "grad_norm": 0.8860392570495605, + "learning_rate": 4.733624434317269e-06, + "loss": 0.6729, + "step": 1159 + }, + { + "epoch": 0.9607951408061844, + "grad_norm": 0.9467882513999939, + "learning_rate": 4.7331302783303416e-06, + "loss": 0.6362, + "step": 1160 + }, + { + "epoch": 0.9616234124792932, + "grad_norm": 0.9138529896736145, + "learning_rate": 4.732635690253523e-06, + "loss": 0.6613, + "step": 1161 + }, + { + "epoch": 0.962451684152402, + "grad_norm": 0.9401054978370667, + "learning_rate": 4.7321406701825105e-06, + "loss": 0.6845, + "step": 1162 + }, + { + "epoch": 0.9632799558255107, + "grad_norm": 0.9049288630485535, + "learning_rate": 4.731645218213086e-06, + "loss": 0.6857, + "step": 1163 + }, + { + "epoch": 0.9641082274986196, + "grad_norm": 0.8972534537315369, + "learning_rate": 4.731149334441114e-06, + "loss": 0.6745, + "step": 1164 + }, + { + "epoch": 0.9649364991717283, + "grad_norm": 1.0799483060836792, + "learning_rate": 4.730653018962545e-06, + "loss": 0.6754, + "step": 1165 + }, + { + "epoch": 0.9657647708448371, + "grad_norm": 0.9841833710670471, + "learning_rate": 4.730156271873412e-06, + "loss": 0.6718, + "step": 1166 + }, + { + "epoch": 0.9665930425179459, + "grad_norm": 0.9388216733932495, + "learning_rate": 4.729659093269828e-06, + "loss": 0.6622, + "step": 1167 + }, + { + "epoch": 0.9674213141910547, + "grad_norm": 0.8879573345184326, + "learning_rate": 4.729161483247994e-06, + "loss": 0.6909, + "step": 1168 + }, + { + "epoch": 0.9682495858641634, + "grad_norm": 0.9713517427444458, + "learning_rate": 4.728663441904194e-06, + "loss": 0.6956, + "step": 1169 + }, + { + "epoch": 0.9690778575372723, + "grad_norm": 0.9200823903083801, + "learning_rate": 4.7281649693347925e-06, + "loss": 0.6766, + "step": 1170 + }, + { + "epoch": 0.969906129210381, + "grad_norm": 0.8983714580535889, + "learning_rate": 4.727666065636241e-06, + "loss": 0.6626, + "step": 1171 + }, + { + "epoch": 0.9707344008834898, + "grad_norm": 0.8792792558670044, + "learning_rate": 4.72716673090507e-06, + "loss": 0.666, + "step": 1172 + }, + { + "epoch": 0.9715626725565986, + "grad_norm": 1.0116846561431885, + "learning_rate": 4.7266669652379005e-06, + "loss": 0.6574, + "step": 1173 + }, + { + "epoch": 0.9723909442297074, + "grad_norm": 0.9433228373527527, + "learning_rate": 4.7261667687314285e-06, + "loss": 0.6972, + "step": 1174 + }, + { + "epoch": 0.9732192159028161, + "grad_norm": 0.8979770541191101, + "learning_rate": 4.725666141482439e-06, + "loss": 0.6523, + "step": 1175 + }, + { + "epoch": 0.9740474875759249, + "grad_norm": 0.8990604281425476, + "learning_rate": 4.725165083587798e-06, + "loss": 0.6535, + "step": 1176 + }, + { + "epoch": 0.9748757592490337, + "grad_norm": 0.9411956667900085, + "learning_rate": 4.724663595144457e-06, + "loss": 0.6947, + "step": 1177 + }, + { + "epoch": 0.9757040309221424, + "grad_norm": 1.0203242301940918, + "learning_rate": 4.724161676249448e-06, + "loss": 0.6894, + "step": 1178 + }, + { + "epoch": 0.9765323025952513, + "grad_norm": 0.8938170075416565, + "learning_rate": 4.723659326999888e-06, + "loss": 0.6648, + "step": 1179 + }, + { + "epoch": 0.97736057426836, + "grad_norm": 0.8810412883758545, + "learning_rate": 4.7231565474929765e-06, + "loss": 0.6761, + "step": 1180 + }, + { + "epoch": 0.9781888459414688, + "grad_norm": 0.8903076648712158, + "learning_rate": 4.722653337825998e-06, + "loss": 0.6738, + "step": 1181 + }, + { + "epoch": 0.9790171176145775, + "grad_norm": 0.905060350894928, + "learning_rate": 4.7221496980963175e-06, + "loss": 0.6625, + "step": 1182 + }, + { + "epoch": 0.9798453892876864, + "grad_norm": 0.8791100978851318, + "learning_rate": 4.721645628401385e-06, + "loss": 0.6884, + "step": 1183 + }, + { + "epoch": 0.9806736609607951, + "grad_norm": 0.897101104259491, + "learning_rate": 4.721141128838733e-06, + "loss": 0.6693, + "step": 1184 + }, + { + "epoch": 0.981501932633904, + "grad_norm": 0.9102833867073059, + "learning_rate": 4.7206361995059775e-06, + "loss": 0.6682, + "step": 1185 + }, + { + "epoch": 0.9823302043070127, + "grad_norm": 0.9492197632789612, + "learning_rate": 4.720130840500819e-06, + "loss": 0.6501, + "step": 1186 + }, + { + "epoch": 0.9831584759801215, + "grad_norm": 0.9783265590667725, + "learning_rate": 4.719625051921038e-06, + "loss": 0.6454, + "step": 1187 + }, + { + "epoch": 0.9839867476532302, + "grad_norm": 0.8938109874725342, + "learning_rate": 4.719118833864501e-06, + "loss": 0.6671, + "step": 1188 + }, + { + "epoch": 0.9848150193263391, + "grad_norm": 0.897484302520752, + "learning_rate": 4.718612186429156e-06, + "loss": 0.6664, + "step": 1189 + }, + { + "epoch": 0.9856432909994478, + "grad_norm": 0.9640376567840576, + "learning_rate": 4.718105109713034e-06, + "loss": 0.6845, + "step": 1190 + }, + { + "epoch": 0.9864715626725566, + "grad_norm": 0.9495493769645691, + "learning_rate": 4.7175976038142505e-06, + "loss": 0.6864, + "step": 1191 + }, + { + "epoch": 0.9872998343456654, + "grad_norm": 0.9390130639076233, + "learning_rate": 4.717089668831002e-06, + "loss": 0.6734, + "step": 1192 + }, + { + "epoch": 0.9881281060187742, + "grad_norm": 0.8854318261146545, + "learning_rate": 4.716581304861571e-06, + "loss": 0.6546, + "step": 1193 + }, + { + "epoch": 0.9889563776918829, + "grad_norm": 0.9242726564407349, + "learning_rate": 4.71607251200432e-06, + "loss": 0.6663, + "step": 1194 + }, + { + "epoch": 0.9897846493649917, + "grad_norm": 0.9302003383636475, + "learning_rate": 4.715563290357696e-06, + "loss": 0.6844, + "step": 1195 + }, + { + "epoch": 0.9906129210381005, + "grad_norm": 0.9279932975769043, + "learning_rate": 4.715053640020228e-06, + "loss": 0.6731, + "step": 1196 + }, + { + "epoch": 0.9914411927112092, + "grad_norm": 0.8753612041473389, + "learning_rate": 4.71454356109053e-06, + "loss": 0.6738, + "step": 1197 + }, + { + "epoch": 0.9922694643843181, + "grad_norm": 0.890133261680603, + "learning_rate": 4.714033053667296e-06, + "loss": 0.6638, + "step": 1198 + }, + { + "epoch": 0.9930977360574268, + "grad_norm": 0.9343729615211487, + "learning_rate": 4.713522117849305e-06, + "loss": 0.6826, + "step": 1199 + }, + { + "epoch": 0.9939260077305356, + "grad_norm": 0.9052755832672119, + "learning_rate": 4.7130107537354185e-06, + "loss": 0.6897, + "step": 1200 + }, + { + "epoch": 0.9947542794036444, + "grad_norm": 0.8683924674987793, + "learning_rate": 4.7124989614245805e-06, + "loss": 0.6687, + "step": 1201 + }, + { + "epoch": 0.9955825510767532, + "grad_norm": 0.8677132725715637, + "learning_rate": 4.711986741015818e-06, + "loss": 0.6917, + "step": 1202 + }, + { + "epoch": 0.9964108227498619, + "grad_norm": 0.8840690851211548, + "learning_rate": 4.711474092608243e-06, + "loss": 0.6841, + "step": 1203 + }, + { + "epoch": 0.9972390944229708, + "grad_norm": 0.9750391840934753, + "learning_rate": 4.7109610163010435e-06, + "loss": 0.6574, + "step": 1204 + }, + { + "epoch": 0.9980673660960795, + "grad_norm": 0.8752943873405457, + "learning_rate": 4.710447512193499e-06, + "loss": 0.636, + "step": 1205 + }, + { + "epoch": 0.9988956377691883, + "grad_norm": 0.9100313782691956, + "learning_rate": 4.709933580384965e-06, + "loss": 0.6712, + "step": 1206 + }, + { + "epoch": 0.999723909442297, + "grad_norm": 0.9202356338500977, + "learning_rate": 4.709419220974886e-06, + "loss": 0.6641, + "step": 1207 + }, + { + "epoch": 1.0, + "grad_norm": 0.9202356338500977, + "learning_rate": 4.708904434062783e-06, + "loss": 0.6199, + "step": 1208 + }, + { + "epoch": 1.0008282716731087, + "grad_norm": 1.6839245557785034, + "learning_rate": 4.708389219748263e-06, + "loss": 0.643, + "step": 1209 + }, + { + "epoch": 1.0016565433462175, + "grad_norm": 0.8636482357978821, + "learning_rate": 4.707873578131015e-06, + "loss": 0.6467, + "step": 1210 + }, + { + "epoch": 1.0024848150193264, + "grad_norm": 0.8548988103866577, + "learning_rate": 4.707357509310811e-06, + "loss": 0.6574, + "step": 1211 + }, + { + "epoch": 1.0033130866924351, + "grad_norm": 0.8961352705955505, + "learning_rate": 4.7068410133875066e-06, + "loss": 0.6367, + "step": 1212 + }, + { + "epoch": 1.0041413583655439, + "grad_norm": 0.8914269208908081, + "learning_rate": 4.706324090461037e-06, + "loss": 0.6294, + "step": 1213 + }, + { + "epoch": 1.0049696300386526, + "grad_norm": 0.879915177822113, + "learning_rate": 4.7058067406314235e-06, + "loss": 0.6346, + "step": 1214 + }, + { + "epoch": 1.0057979017117615, + "grad_norm": 0.8798143863677979, + "learning_rate": 4.7052889639987674e-06, + "loss": 0.6559, + "step": 1215 + }, + { + "epoch": 1.0066261733848703, + "grad_norm": 0.8888561129570007, + "learning_rate": 4.704770760663255e-06, + "loss": 0.6471, + "step": 1216 + }, + { + "epoch": 1.007454445057979, + "grad_norm": 0.8707575798034668, + "learning_rate": 4.704252130725152e-06, + "loss": 0.6485, + "step": 1217 + }, + { + "epoch": 1.0082827167310877, + "grad_norm": 0.8609346151351929, + "learning_rate": 4.7037330742848105e-06, + "loss": 0.6544, + "step": 1218 + }, + { + "epoch": 1.0091109884041967, + "grad_norm": 0.8588053584098816, + "learning_rate": 4.703213591442663e-06, + "loss": 0.6421, + "step": 1219 + }, + { + "epoch": 1.0099392600773054, + "grad_norm": 0.8932695388793945, + "learning_rate": 4.702693682299222e-06, + "loss": 0.6333, + "step": 1220 + }, + { + "epoch": 1.0107675317504141, + "grad_norm": 0.8615652322769165, + "learning_rate": 4.7021733469550865e-06, + "loss": 0.6371, + "step": 1221 + }, + { + "epoch": 1.0115958034235228, + "grad_norm": 0.8876634240150452, + "learning_rate": 4.7016525855109376e-06, + "loss": 0.6368, + "step": 1222 + }, + { + "epoch": 1.0124240750966318, + "grad_norm": 0.895575225353241, + "learning_rate": 4.701131398067537e-06, + "loss": 0.6226, + "step": 1223 + }, + { + "epoch": 1.0132523467697405, + "grad_norm": 0.8836749196052551, + "learning_rate": 4.70060978472573e-06, + "loss": 0.6516, + "step": 1224 + }, + { + "epoch": 1.0140806184428492, + "grad_norm": 0.9471280574798584, + "learning_rate": 4.700087745586442e-06, + "loss": 0.633, + "step": 1225 + }, + { + "epoch": 1.014908890115958, + "grad_norm": 0.8721561431884766, + "learning_rate": 4.699565280750685e-06, + "loss": 0.6568, + "step": 1226 + }, + { + "epoch": 1.015737161789067, + "grad_norm": 0.8946447372436523, + "learning_rate": 4.69904239031955e-06, + "loss": 0.6503, + "step": 1227 + }, + { + "epoch": 1.0165654334621756, + "grad_norm": 0.8854631185531616, + "learning_rate": 4.698519074394212e-06, + "loss": 0.6341, + "step": 1228 + }, + { + "epoch": 1.0173937051352844, + "grad_norm": 0.9348397850990295, + "learning_rate": 4.697995333075927e-06, + "loss": 0.6317, + "step": 1229 + }, + { + "epoch": 1.018221976808393, + "grad_norm": 0.9037336111068726, + "learning_rate": 4.697471166466035e-06, + "loss": 0.6618, + "step": 1230 + }, + { + "epoch": 1.0190502484815018, + "grad_norm": 0.8866400718688965, + "learning_rate": 4.696946574665957e-06, + "loss": 0.6421, + "step": 1231 + }, + { + "epoch": 1.0198785201546108, + "grad_norm": 0.8897701501846313, + "learning_rate": 4.696421557777196e-06, + "loss": 0.6608, + "step": 1232 + }, + { + "epoch": 1.0207067918277195, + "grad_norm": 0.9046173095703125, + "learning_rate": 4.6958961159013386e-06, + "loss": 0.6464, + "step": 1233 + }, + { + "epoch": 1.0215350635008282, + "grad_norm": 0.9492906332015991, + "learning_rate": 4.695370249140052e-06, + "loss": 0.6578, + "step": 1234 + }, + { + "epoch": 1.022363335173937, + "grad_norm": 0.8906951546669006, + "learning_rate": 4.6948439575950885e-06, + "loss": 0.6276, + "step": 1235 + }, + { + "epoch": 1.023191606847046, + "grad_norm": 0.9218266606330872, + "learning_rate": 4.694317241368278e-06, + "loss": 0.627, + "step": 1236 + }, + { + "epoch": 1.0240198785201546, + "grad_norm": 0.901157557964325, + "learning_rate": 4.693790100561537e-06, + "loss": 0.6379, + "step": 1237 + }, + { + "epoch": 1.0248481501932634, + "grad_norm": 0.8992452025413513, + "learning_rate": 4.693262535276863e-06, + "loss": 0.6585, + "step": 1238 + }, + { + "epoch": 1.025676421866372, + "grad_norm": 0.9525597095489502, + "learning_rate": 4.692734545616331e-06, + "loss": 0.6625, + "step": 1239 + }, + { + "epoch": 1.026504693539481, + "grad_norm": 0.9229065775871277, + "learning_rate": 4.692206131682106e-06, + "loss": 0.6454, + "step": 1240 + }, + { + "epoch": 1.0273329652125898, + "grad_norm": 1.012373685836792, + "learning_rate": 4.691677293576431e-06, + "loss": 0.6492, + "step": 1241 + }, + { + "epoch": 1.0281612368856985, + "grad_norm": 0.8650060892105103, + "learning_rate": 4.69114803140163e-06, + "loss": 0.6739, + "step": 1242 + }, + { + "epoch": 1.0289895085588072, + "grad_norm": 0.9396429657936096, + "learning_rate": 4.69061834526011e-06, + "loss": 0.6729, + "step": 1243 + }, + { + "epoch": 1.0298177802319162, + "grad_norm": 0.9584783911705017, + "learning_rate": 4.6900882352543614e-06, + "loss": 0.6678, + "step": 1244 + }, + { + "epoch": 1.030646051905025, + "grad_norm": 0.985394299030304, + "learning_rate": 4.689557701486954e-06, + "loss": 0.6478, + "step": 1245 + }, + { + "epoch": 1.0314743235781336, + "grad_norm": 0.9014914631843567, + "learning_rate": 4.689026744060543e-06, + "loss": 0.6393, + "step": 1246 + }, + { + "epoch": 1.0323025952512423, + "grad_norm": 0.9904847145080566, + "learning_rate": 4.688495363077863e-06, + "loss": 0.6752, + "step": 1247 + }, + { + "epoch": 1.0331308669243513, + "grad_norm": 0.9439241886138916, + "learning_rate": 4.687963558641731e-06, + "loss": 0.6417, + "step": 1248 + }, + { + "epoch": 1.03395913859746, + "grad_norm": 0.9277704954147339, + "learning_rate": 4.687431330855046e-06, + "loss": 0.6493, + "step": 1249 + }, + { + "epoch": 1.0347874102705688, + "grad_norm": 0.9456571340560913, + "learning_rate": 4.686898679820789e-06, + "loss": 0.6774, + "step": 1250 + }, + { + "epoch": 1.0356156819436775, + "grad_norm": 0.9043509364128113, + "learning_rate": 4.686365605642025e-06, + "loss": 0.6198, + "step": 1251 + }, + { + "epoch": 1.0364439536167862, + "grad_norm": 0.9246267676353455, + "learning_rate": 4.685832108421896e-06, + "loss": 0.6513, + "step": 1252 + }, + { + "epoch": 1.0372722252898952, + "grad_norm": 0.9227612614631653, + "learning_rate": 4.685298188263631e-06, + "loss": 0.6597, + "step": 1253 + }, + { + "epoch": 1.0381004969630039, + "grad_norm": 0.8927509188652039, + "learning_rate": 4.684763845270537e-06, + "loss": 0.641, + "step": 1254 + }, + { + "epoch": 1.0389287686361126, + "grad_norm": 0.8812937140464783, + "learning_rate": 4.684229079546005e-06, + "loss": 0.6292, + "step": 1255 + }, + { + "epoch": 1.0397570403092213, + "grad_norm": 0.9186881184577942, + "learning_rate": 4.683693891193508e-06, + "loss": 0.6485, + "step": 1256 + }, + { + "epoch": 1.0405853119823303, + "grad_norm": 0.8928134441375732, + "learning_rate": 4.683158280316599e-06, + "loss": 0.6435, + "step": 1257 + }, + { + "epoch": 1.041413583655439, + "grad_norm": 0.8919920325279236, + "learning_rate": 4.682622247018913e-06, + "loss": 0.622, + "step": 1258 + }, + { + "epoch": 1.0422418553285477, + "grad_norm": 0.885259747505188, + "learning_rate": 4.682085791404169e-06, + "loss": 0.6521, + "step": 1259 + }, + { + "epoch": 1.0430701270016565, + "grad_norm": 0.9034169316291809, + "learning_rate": 4.681548913576165e-06, + "loss": 0.6437, + "step": 1260 + }, + { + "epoch": 1.0438983986747654, + "grad_norm": 0.884848952293396, + "learning_rate": 4.6810116136387825e-06, + "loss": 0.6733, + "step": 1261 + }, + { + "epoch": 1.0447266703478741, + "grad_norm": 0.9137836694717407, + "learning_rate": 4.680473891695985e-06, + "loss": 0.6415, + "step": 1262 + }, + { + "epoch": 1.0455549420209829, + "grad_norm": 0.9044780135154724, + "learning_rate": 4.6799357478518145e-06, + "loss": 0.6453, + "step": 1263 + }, + { + "epoch": 1.0463832136940916, + "grad_norm": 0.9111152291297913, + "learning_rate": 4.6793971822103985e-06, + "loss": 0.6644, + "step": 1264 + }, + { + "epoch": 1.0472114853672005, + "grad_norm": 0.9158084988594055, + "learning_rate": 4.678858194875944e-06, + "loss": 0.6434, + "step": 1265 + }, + { + "epoch": 1.0480397570403093, + "grad_norm": 0.8970723152160645, + "learning_rate": 4.678318785952739e-06, + "loss": 0.6391, + "step": 1266 + }, + { + "epoch": 1.048868028713418, + "grad_norm": 0.8958215117454529, + "learning_rate": 4.677778955545155e-06, + "loss": 0.6562, + "step": 1267 + }, + { + "epoch": 1.0496963003865267, + "grad_norm": 0.9229645729064941, + "learning_rate": 4.677238703757644e-06, + "loss": 0.646, + "step": 1268 + }, + { + "epoch": 1.0505245720596355, + "grad_norm": 0.8986689448356628, + "learning_rate": 4.676698030694741e-06, + "loss": 0.6241, + "step": 1269 + }, + { + "epoch": 1.0513528437327444, + "grad_norm": 0.8943328261375427, + "learning_rate": 4.6761569364610585e-06, + "loss": 0.6455, + "step": 1270 + }, + { + "epoch": 1.0521811154058531, + "grad_norm": 0.9147390127182007, + "learning_rate": 4.675615421161295e-06, + "loss": 0.6299, + "step": 1271 + }, + { + "epoch": 1.0530093870789619, + "grad_norm": 0.9124075174331665, + "learning_rate": 4.675073484900229e-06, + "loss": 0.6379, + "step": 1272 + }, + { + "epoch": 1.0538376587520706, + "grad_norm": 0.9267525672912598, + "learning_rate": 4.674531127782718e-06, + "loss": 0.6478, + "step": 1273 + }, + { + "epoch": 1.0546659304251795, + "grad_norm": 0.8710078001022339, + "learning_rate": 4.673988349913707e-06, + "loss": 0.6314, + "step": 1274 + }, + { + "epoch": 1.0554942020982883, + "grad_norm": 0.8905670642852783, + "learning_rate": 4.673445151398214e-06, + "loss": 0.6456, + "step": 1275 + }, + { + "epoch": 1.056322473771397, + "grad_norm": 0.9125292897224426, + "learning_rate": 4.672901532341346e-06, + "loss": 0.6439, + "step": 1276 + }, + { + "epoch": 1.0571507454445057, + "grad_norm": 0.9003625512123108, + "learning_rate": 4.6723574928482865e-06, + "loss": 0.6561, + "step": 1277 + }, + { + "epoch": 1.0579790171176147, + "grad_norm": 0.96099454164505, + "learning_rate": 4.6718130330243014e-06, + "loss": 0.6515, + "step": 1278 + }, + { + "epoch": 1.0588072887907234, + "grad_norm": 0.8841069340705872, + "learning_rate": 4.671268152974742e-06, + "loss": 0.6389, + "step": 1279 + }, + { + "epoch": 1.0596355604638321, + "grad_norm": 0.8906645178794861, + "learning_rate": 4.670722852805033e-06, + "loss": 0.6448, + "step": 1280 + }, + { + "epoch": 1.0604638321369408, + "grad_norm": 0.9538317918777466, + "learning_rate": 4.670177132620689e-06, + "loss": 0.6544, + "step": 1281 + }, + { + "epoch": 1.0612921038100498, + "grad_norm": 0.9158053994178772, + "learning_rate": 4.669630992527299e-06, + "loss": 0.6494, + "step": 1282 + }, + { + "epoch": 1.0621203754831585, + "grad_norm": 0.9756888151168823, + "learning_rate": 4.6690844326305374e-06, + "loss": 0.6419, + "step": 1283 + }, + { + "epoch": 1.0629486471562672, + "grad_norm": 0.905055820941925, + "learning_rate": 4.668537453036159e-06, + "loss": 0.6463, + "step": 1284 + }, + { + "epoch": 1.063776918829376, + "grad_norm": 0.9011061191558838, + "learning_rate": 4.667990053849997e-06, + "loss": 0.6353, + "step": 1285 + }, + { + "epoch": 1.0646051905024847, + "grad_norm": 0.955047070980072, + "learning_rate": 4.667442235177969e-06, + "loss": 0.6485, + "step": 1286 + }, + { + "epoch": 1.0654334621755936, + "grad_norm": 0.9253432750701904, + "learning_rate": 4.666893997126074e-06, + "loss": 0.6391, + "step": 1287 + }, + { + "epoch": 1.0662617338487024, + "grad_norm": 0.8840277194976807, + "learning_rate": 4.66634533980039e-06, + "loss": 0.6372, + "step": 1288 + }, + { + "epoch": 1.067090005521811, + "grad_norm": 1.0881837606430054, + "learning_rate": 4.6657962633070765e-06, + "loss": 0.6419, + "step": 1289 + }, + { + "epoch": 1.06791827719492, + "grad_norm": 0.9263116717338562, + "learning_rate": 4.665246767752376e-06, + "loss": 0.6226, + "step": 1290 + }, + { + "epoch": 1.0687465488680288, + "grad_norm": 0.9110594987869263, + "learning_rate": 4.66469685324261e-06, + "loss": 0.6495, + "step": 1291 + }, + { + "epoch": 1.0695748205411375, + "grad_norm": 0.9086888432502747, + "learning_rate": 4.664146519884182e-06, + "loss": 0.6408, + "step": 1292 + }, + { + "epoch": 1.0704030922142462, + "grad_norm": 0.8991205096244812, + "learning_rate": 4.663595767783575e-06, + "loss": 0.6424, + "step": 1293 + }, + { + "epoch": 1.071231363887355, + "grad_norm": 0.926933228969574, + "learning_rate": 4.6630445970473576e-06, + "loss": 0.6379, + "step": 1294 + }, + { + "epoch": 1.072059635560464, + "grad_norm": 0.8793080449104309, + "learning_rate": 4.662493007782174e-06, + "loss": 0.6254, + "step": 1295 + }, + { + "epoch": 1.0728879072335726, + "grad_norm": 0.9588155150413513, + "learning_rate": 4.661941000094751e-06, + "loss": 0.6307, + "step": 1296 + }, + { + "epoch": 1.0737161789066814, + "grad_norm": 0.9468860030174255, + "learning_rate": 4.661388574091898e-06, + "loss": 0.6308, + "step": 1297 + }, + { + "epoch": 1.07454445057979, + "grad_norm": 0.8930124640464783, + "learning_rate": 4.660835729880505e-06, + "loss": 0.6371, + "step": 1298 + }, + { + "epoch": 1.075372722252899, + "grad_norm": 0.9233878254890442, + "learning_rate": 4.660282467567541e-06, + "loss": 0.6327, + "step": 1299 + }, + { + "epoch": 1.0762009939260078, + "grad_norm": 0.9714951515197754, + "learning_rate": 4.659728787260057e-06, + "loss": 0.6218, + "step": 1300 + }, + { + "epoch": 1.0770292655991165, + "grad_norm": 0.9464428424835205, + "learning_rate": 4.659174689065185e-06, + "loss": 0.6406, + "step": 1301 + }, + { + "epoch": 1.0778575372722252, + "grad_norm": 0.8984559774398804, + "learning_rate": 4.6586201730901395e-06, + "loss": 0.6483, + "step": 1302 + }, + { + "epoch": 1.0786858089453342, + "grad_norm": 0.9007774591445923, + "learning_rate": 4.658065239442212e-06, + "loss": 0.6288, + "step": 1303 + }, + { + "epoch": 1.079514080618443, + "grad_norm": 0.907171368598938, + "learning_rate": 4.657509888228779e-06, + "loss": 0.6445, + "step": 1304 + }, + { + "epoch": 1.0803423522915516, + "grad_norm": 0.9207282066345215, + "learning_rate": 4.656954119557293e-06, + "loss": 0.6148, + "step": 1305 + }, + { + "epoch": 1.0811706239646603, + "grad_norm": 0.9467616677284241, + "learning_rate": 4.656397933535293e-06, + "loss": 0.6374, + "step": 1306 + }, + { + "epoch": 1.0819988956377693, + "grad_norm": 0.9310930967330933, + "learning_rate": 4.655841330270393e-06, + "loss": 0.6362, + "step": 1307 + }, + { + "epoch": 1.082827167310878, + "grad_norm": 0.890816330909729, + "learning_rate": 4.655284309870294e-06, + "loss": 0.6372, + "step": 1308 + }, + { + "epoch": 1.0836554389839868, + "grad_norm": 0.9206095933914185, + "learning_rate": 4.654726872442771e-06, + "loss": 0.6297, + "step": 1309 + }, + { + "epoch": 1.0844837106570955, + "grad_norm": 0.9015856385231018, + "learning_rate": 4.654169018095685e-06, + "loss": 0.658, + "step": 1310 + }, + { + "epoch": 1.0853119823302042, + "grad_norm": 0.8909441828727722, + "learning_rate": 4.653610746936975e-06, + "loss": 0.6382, + "step": 1311 + }, + { + "epoch": 1.0861402540033132, + "grad_norm": 0.9035969972610474, + "learning_rate": 4.653052059074661e-06, + "loss": 0.6441, + "step": 1312 + }, + { + "epoch": 1.0869685256764219, + "grad_norm": 0.8988869786262512, + "learning_rate": 4.652492954616844e-06, + "loss": 0.6359, + "step": 1313 + }, + { + "epoch": 1.0877967973495306, + "grad_norm": 0.881885826587677, + "learning_rate": 4.6519334336717055e-06, + "loss": 0.6397, + "step": 1314 + }, + { + "epoch": 1.0886250690226393, + "grad_norm": 0.980633020401001, + "learning_rate": 4.651373496347508e-06, + "loss": 0.6223, + "step": 1315 + }, + { + "epoch": 1.0894533406957483, + "grad_norm": 0.9027420878410339, + "learning_rate": 4.650813142752593e-06, + "loss": 0.6211, + "step": 1316 + }, + { + "epoch": 1.090281612368857, + "grad_norm": 0.9011778831481934, + "learning_rate": 4.6502523729953855e-06, + "loss": 0.6442, + "step": 1317 + }, + { + "epoch": 1.0911098840419657, + "grad_norm": 0.926692545413971, + "learning_rate": 4.649691187184387e-06, + "loss": 0.6661, + "step": 1318 + }, + { + "epoch": 1.0919381557150745, + "grad_norm": 0.9758741855621338, + "learning_rate": 4.649129585428184e-06, + "loss": 0.6461, + "step": 1319 + }, + { + "epoch": 1.0927664273881834, + "grad_norm": 0.967698335647583, + "learning_rate": 4.64856756783544e-06, + "loss": 0.6461, + "step": 1320 + }, + { + "epoch": 1.0935946990612921, + "grad_norm": 0.9372200965881348, + "learning_rate": 4.6480051345149e-06, + "loss": 0.6348, + "step": 1321 + }, + { + "epoch": 1.0944229707344009, + "grad_norm": 0.8976414799690247, + "learning_rate": 4.64744228557539e-06, + "loss": 0.6297, + "step": 1322 + }, + { + "epoch": 1.0952512424075096, + "grad_norm": 0.9163550734519958, + "learning_rate": 4.646879021125816e-06, + "loss": 0.6746, + "step": 1323 + }, + { + "epoch": 1.0960795140806185, + "grad_norm": 0.9064576625823975, + "learning_rate": 4.646315341275164e-06, + "loss": 0.6242, + "step": 1324 + }, + { + "epoch": 1.0969077857537273, + "grad_norm": 0.9352027773857117, + "learning_rate": 4.645751246132501e-06, + "loss": 0.6568, + "step": 1325 + }, + { + "epoch": 1.097736057426836, + "grad_norm": 0.8680023550987244, + "learning_rate": 4.645186735806976e-06, + "loss": 0.6484, + "step": 1326 + }, + { + "epoch": 1.0985643290999447, + "grad_norm": 0.8782095909118652, + "learning_rate": 4.644621810407813e-06, + "loss": 0.646, + "step": 1327 + }, + { + "epoch": 1.0993926007730535, + "grad_norm": 0.8931819796562195, + "learning_rate": 4.644056470044323e-06, + "loss": 0.6708, + "step": 1328 + }, + { + "epoch": 1.1002208724461624, + "grad_norm": 0.9228624105453491, + "learning_rate": 4.6434907148258915e-06, + "loss": 0.6255, + "step": 1329 + }, + { + "epoch": 1.1010491441192711, + "grad_norm": 0.8960729837417603, + "learning_rate": 4.642924544861988e-06, + "loss": 0.6523, + "step": 1330 + }, + { + "epoch": 1.1018774157923799, + "grad_norm": 0.9092081785202026, + "learning_rate": 4.642357960262162e-06, + "loss": 0.6534, + "step": 1331 + }, + { + "epoch": 1.1027056874654886, + "grad_norm": 0.9024838209152222, + "learning_rate": 4.641790961136041e-06, + "loss": 0.6238, + "step": 1332 + }, + { + "epoch": 1.1035339591385975, + "grad_norm": 0.9016388058662415, + "learning_rate": 4.641223547593334e-06, + "loss": 0.6643, + "step": 1333 + }, + { + "epoch": 1.1043622308117063, + "grad_norm": 0.8877143859863281, + "learning_rate": 4.6406557197438324e-06, + "loss": 0.6222, + "step": 1334 + }, + { + "epoch": 1.105190502484815, + "grad_norm": 0.9066653251647949, + "learning_rate": 4.640087477697403e-06, + "loss": 0.6205, + "step": 1335 + }, + { + "epoch": 1.1060187741579237, + "grad_norm": 0.903465747833252, + "learning_rate": 4.639518821563997e-06, + "loss": 0.6299, + "step": 1336 + }, + { + "epoch": 1.1068470458310327, + "grad_norm": 0.8908408880233765, + "learning_rate": 4.6389497514536415e-06, + "loss": 0.6247, + "step": 1337 + }, + { + "epoch": 1.1076753175041414, + "grad_norm": 0.8957250714302063, + "learning_rate": 4.63838026747645e-06, + "loss": 0.632, + "step": 1338 + }, + { + "epoch": 1.1085035891772501, + "grad_norm": 0.938494086265564, + "learning_rate": 4.637810369742609e-06, + "loss": 0.6309, + "step": 1339 + }, + { + "epoch": 1.1093318608503588, + "grad_norm": 0.9913409948348999, + "learning_rate": 4.637240058362391e-06, + "loss": 0.6571, + "step": 1340 + }, + { + "epoch": 1.1101601325234678, + "grad_norm": 0.9037449359893799, + "learning_rate": 4.636669333446145e-06, + "loss": 0.6467, + "step": 1341 + }, + { + "epoch": 1.1109884041965765, + "grad_norm": 0.911696195602417, + "learning_rate": 4.636098195104299e-06, + "loss": 0.6361, + "step": 1342 + }, + { + "epoch": 1.1118166758696852, + "grad_norm": 0.9105570316314697, + "learning_rate": 4.6355266434473665e-06, + "loss": 0.6418, + "step": 1343 + }, + { + "epoch": 1.112644947542794, + "grad_norm": 0.9101701378822327, + "learning_rate": 4.634954678585935e-06, + "loss": 0.6454, + "step": 1344 + }, + { + "epoch": 1.1134732192159027, + "grad_norm": 0.8980433344841003, + "learning_rate": 4.634382300630675e-06, + "loss": 0.6159, + "step": 1345 + }, + { + "epoch": 1.1143014908890116, + "grad_norm": 0.9253267049789429, + "learning_rate": 4.633809509692336e-06, + "loss": 0.6465, + "step": 1346 + }, + { + "epoch": 1.1151297625621204, + "grad_norm": 0.9150062203407288, + "learning_rate": 4.6332363058817484e-06, + "loss": 0.639, + "step": 1347 + }, + { + "epoch": 1.115958034235229, + "grad_norm": 0.9257630705833435, + "learning_rate": 4.632662689309821e-06, + "loss": 0.6571, + "step": 1348 + }, + { + "epoch": 1.116786305908338, + "grad_norm": 0.9183084964752197, + "learning_rate": 4.632088660087545e-06, + "loss": 0.6291, + "step": 1349 + }, + { + "epoch": 1.1176145775814468, + "grad_norm": 0.9229403138160706, + "learning_rate": 4.631514218325987e-06, + "loss": 0.6487, + "step": 1350 + }, + { + "epoch": 1.1184428492545555, + "grad_norm": 0.9379578828811646, + "learning_rate": 4.630939364136298e-06, + "loss": 0.6564, + "step": 1351 + }, + { + "epoch": 1.1192711209276642, + "grad_norm": 0.9094021320343018, + "learning_rate": 4.630364097629706e-06, + "loss": 0.6354, + "step": 1352 + }, + { + "epoch": 1.120099392600773, + "grad_norm": 0.9528741836547852, + "learning_rate": 4.62978841891752e-06, + "loss": 0.6259, + "step": 1353 + }, + { + "epoch": 1.120927664273882, + "grad_norm": 0.8859900832176208, + "learning_rate": 4.629212328111129e-06, + "loss": 0.6783, + "step": 1354 + }, + { + "epoch": 1.1217559359469906, + "grad_norm": 0.8739688396453857, + "learning_rate": 4.628635825322e-06, + "loss": 0.6518, + "step": 1355 + }, + { + "epoch": 1.1225842076200994, + "grad_norm": 0.8773418068885803, + "learning_rate": 4.628058910661683e-06, + "loss": 0.6477, + "step": 1356 + }, + { + "epoch": 1.123412479293208, + "grad_norm": 0.9025301933288574, + "learning_rate": 4.627481584241803e-06, + "loss": 0.6471, + "step": 1357 + }, + { + "epoch": 1.124240750966317, + "grad_norm": 0.89899080991745, + "learning_rate": 4.62690384617407e-06, + "loss": 0.6196, + "step": 1358 + }, + { + "epoch": 1.1250690226394258, + "grad_norm": 0.9514591693878174, + "learning_rate": 4.6263256965702684e-06, + "loss": 0.6495, + "step": 1359 + }, + { + "epoch": 1.1258972943125345, + "grad_norm": 0.908789336681366, + "learning_rate": 4.6257471355422654e-06, + "loss": 0.6606, + "step": 1360 + }, + { + "epoch": 1.1267255659856432, + "grad_norm": 0.9262552857398987, + "learning_rate": 4.625168163202009e-06, + "loss": 0.6308, + "step": 1361 + }, + { + "epoch": 1.1275538376587522, + "grad_norm": 0.9395532608032227, + "learning_rate": 4.624588779661523e-06, + "loss": 0.6196, + "step": 1362 + }, + { + "epoch": 1.128382109331861, + "grad_norm": 0.9126531481742859, + "learning_rate": 4.624008985032913e-06, + "loss": 0.6474, + "step": 1363 + }, + { + "epoch": 1.1292103810049696, + "grad_norm": 0.9105198383331299, + "learning_rate": 4.623428779428365e-06, + "loss": 0.6455, + "step": 1364 + }, + { + "epoch": 1.1300386526780783, + "grad_norm": 0.9048190712928772, + "learning_rate": 4.6228481629601415e-06, + "loss": 0.6647, + "step": 1365 + }, + { + "epoch": 1.1308669243511873, + "grad_norm": 0.9415149092674255, + "learning_rate": 4.622267135740588e-06, + "loss": 0.6403, + "step": 1366 + }, + { + "epoch": 1.131695196024296, + "grad_norm": 0.954728901386261, + "learning_rate": 4.6216856978821275e-06, + "loss": 0.6351, + "step": 1367 + }, + { + "epoch": 1.1325234676974048, + "grad_norm": 0.9392480850219727, + "learning_rate": 4.621103849497261e-06, + "loss": 0.6481, + "step": 1368 + }, + { + "epoch": 1.1333517393705135, + "grad_norm": 0.9199283719062805, + "learning_rate": 4.620521590698574e-06, + "loss": 0.6578, + "step": 1369 + }, + { + "epoch": 1.1341800110436222, + "grad_norm": 0.9175401926040649, + "learning_rate": 4.6199389215987266e-06, + "loss": 0.6508, + "step": 1370 + }, + { + "epoch": 1.1350082827167312, + "grad_norm": 0.9140470027923584, + "learning_rate": 4.619355842310459e-06, + "loss": 0.6536, + "step": 1371 + }, + { + "epoch": 1.1358365543898399, + "grad_norm": 0.8863652348518372, + "learning_rate": 4.618772352946593e-06, + "loss": 0.6376, + "step": 1372 + }, + { + "epoch": 1.1366648260629486, + "grad_norm": 0.9074656963348389, + "learning_rate": 4.618188453620026e-06, + "loss": 0.6253, + "step": 1373 + }, + { + "epoch": 1.1374930977360573, + "grad_norm": 0.891055703163147, + "learning_rate": 4.61760414444374e-06, + "loss": 0.6507, + "step": 1374 + }, + { + "epoch": 1.1383213694091663, + "grad_norm": 0.9153634905815125, + "learning_rate": 4.617019425530791e-06, + "loss": 0.6699, + "step": 1375 + }, + { + "epoch": 1.139149641082275, + "grad_norm": 0.9172940850257874, + "learning_rate": 4.6164342969943196e-06, + "loss": 0.6385, + "step": 1376 + }, + { + "epoch": 1.1399779127553837, + "grad_norm": 0.8978968262672424, + "learning_rate": 4.615848758947539e-06, + "loss": 0.6394, + "step": 1377 + }, + { + "epoch": 1.1408061844284925, + "grad_norm": 0.9372277855873108, + "learning_rate": 4.615262811503749e-06, + "loss": 0.6418, + "step": 1378 + }, + { + "epoch": 1.1416344561016014, + "grad_norm": 0.9596112370491028, + "learning_rate": 4.614676454776321e-06, + "loss": 0.6573, + "step": 1379 + }, + { + "epoch": 1.1424627277747101, + "grad_norm": 0.9243675470352173, + "learning_rate": 4.614089688878713e-06, + "loss": 0.6546, + "step": 1380 + }, + { + "epoch": 1.1432909994478189, + "grad_norm": 0.8839207291603088, + "learning_rate": 4.613502513924459e-06, + "loss": 0.654, + "step": 1381 + }, + { + "epoch": 1.1441192711209276, + "grad_norm": 0.8940262198448181, + "learning_rate": 4.612914930027168e-06, + "loss": 0.6223, + "step": 1382 + }, + { + "epoch": 1.1449475427940365, + "grad_norm": 0.9091162085533142, + "learning_rate": 4.612326937300535e-06, + "loss": 0.6512, + "step": 1383 + }, + { + "epoch": 1.1457758144671453, + "grad_norm": 0.9332976937294006, + "learning_rate": 4.6117385358583315e-06, + "loss": 0.6373, + "step": 1384 + }, + { + "epoch": 1.146604086140254, + "grad_norm": 0.9103875756263733, + "learning_rate": 4.611149725814406e-06, + "loss": 0.6502, + "step": 1385 + }, + { + "epoch": 1.1474323578133627, + "grad_norm": 0.9004862308502197, + "learning_rate": 4.610560507282688e-06, + "loss": 0.6361, + "step": 1386 + }, + { + "epoch": 1.1482606294864715, + "grad_norm": 0.880293071269989, + "learning_rate": 4.609970880377187e-06, + "loss": 0.655, + "step": 1387 + }, + { + "epoch": 1.1490889011595804, + "grad_norm": 0.9099947214126587, + "learning_rate": 4.60938084521199e-06, + "loss": 0.6517, + "step": 1388 + }, + { + "epoch": 1.1499171728326891, + "grad_norm": 0.8913806080818176, + "learning_rate": 4.608790401901262e-06, + "loss": 0.6429, + "step": 1389 + }, + { + "epoch": 1.1507454445057979, + "grad_norm": 0.9198290109634399, + "learning_rate": 4.608199550559249e-06, + "loss": 0.6537, + "step": 1390 + }, + { + "epoch": 1.1515737161789068, + "grad_norm": 0.9192278981208801, + "learning_rate": 4.607608291300275e-06, + "loss": 0.6531, + "step": 1391 + }, + { + "epoch": 1.1524019878520155, + "grad_norm": 0.8977665305137634, + "learning_rate": 4.607016624238744e-06, + "loss": 0.6358, + "step": 1392 + }, + { + "epoch": 1.1532302595251243, + "grad_norm": 0.9576854705810547, + "learning_rate": 4.606424549489138e-06, + "loss": 0.6493, + "step": 1393 + }, + { + "epoch": 1.154058531198233, + "grad_norm": 0.905611515045166, + "learning_rate": 4.605832067166017e-06, + "loss": 0.63, + "step": 1394 + }, + { + "epoch": 1.1548868028713417, + "grad_norm": 0.8840015530586243, + "learning_rate": 4.605239177384021e-06, + "loss": 0.6504, + "step": 1395 + }, + { + "epoch": 1.1557150745444507, + "grad_norm": 0.9247309565544128, + "learning_rate": 4.604645880257869e-06, + "loss": 0.6477, + "step": 1396 + }, + { + "epoch": 1.1565433462175594, + "grad_norm": 0.9473649263381958, + "learning_rate": 4.604052175902359e-06, + "loss": 0.6374, + "step": 1397 + }, + { + "epoch": 1.1573716178906681, + "grad_norm": 0.9354954361915588, + "learning_rate": 4.603458064432367e-06, + "loss": 0.6491, + "step": 1398 + }, + { + "epoch": 1.1581998895637768, + "grad_norm": 0.8788221478462219, + "learning_rate": 4.6028635459628476e-06, + "loss": 0.6187, + "step": 1399 + }, + { + "epoch": 1.1590281612368858, + "grad_norm": 0.8836075067520142, + "learning_rate": 4.602268620608834e-06, + "loss": 0.665, + "step": 1400 + }, + { + "epoch": 1.1598564329099945, + "grad_norm": 0.9195554256439209, + "learning_rate": 4.60167328848544e-06, + "loss": 0.6291, + "step": 1401 + }, + { + "epoch": 1.1606847045831032, + "grad_norm": 0.8984520435333252, + "learning_rate": 4.601077549707856e-06, + "loss": 0.6361, + "step": 1402 + }, + { + "epoch": 1.161512976256212, + "grad_norm": 0.8667109608650208, + "learning_rate": 4.600481404391352e-06, + "loss": 0.6345, + "step": 1403 + }, + { + "epoch": 1.1623412479293207, + "grad_norm": 0.9312148094177246, + "learning_rate": 4.599884852651277e-06, + "loss": 0.6388, + "step": 1404 + }, + { + "epoch": 1.1631695196024296, + "grad_norm": 0.9096866846084595, + "learning_rate": 4.5992878946030575e-06, + "loss": 0.6383, + "step": 1405 + }, + { + "epoch": 1.1639977912755384, + "grad_norm": 0.9010055661201477, + "learning_rate": 4.5986905303622e-06, + "loss": 0.6524, + "step": 1406 + }, + { + "epoch": 1.164826062948647, + "grad_norm": 0.9414717555046082, + "learning_rate": 4.598092760044288e-06, + "loss": 0.6825, + "step": 1407 + }, + { + "epoch": 1.165654334621756, + "grad_norm": 0.9162404537200928, + "learning_rate": 4.597494583764987e-06, + "loss": 0.6467, + "step": 1408 + }, + { + "epoch": 1.1664826062948648, + "grad_norm": 0.8935617208480835, + "learning_rate": 4.596896001640034e-06, + "loss": 0.6375, + "step": 1409 + }, + { + "epoch": 1.1673108779679735, + "grad_norm": 0.8976510167121887, + "learning_rate": 4.596297013785253e-06, + "loss": 0.6135, + "step": 1410 + }, + { + "epoch": 1.1681391496410822, + "grad_norm": 0.9503909349441528, + "learning_rate": 4.5956976203165414e-06, + "loss": 0.6305, + "step": 1411 + }, + { + "epoch": 1.168967421314191, + "grad_norm": 0.9147273302078247, + "learning_rate": 4.595097821349876e-06, + "loss": 0.6548, + "step": 1412 + }, + { + "epoch": 1.1697956929873, + "grad_norm": 0.9171034693717957, + "learning_rate": 4.594497617001312e-06, + "loss": 0.6253, + "step": 1413 + }, + { + "epoch": 1.1706239646604086, + "grad_norm": 0.8970120549201965, + "learning_rate": 4.593897007386985e-06, + "loss": 0.6414, + "step": 1414 + }, + { + "epoch": 1.1714522363335174, + "grad_norm": 0.9032952189445496, + "learning_rate": 4.593295992623105e-06, + "loss": 0.6595, + "step": 1415 + }, + { + "epoch": 1.172280508006626, + "grad_norm": 0.9066448211669922, + "learning_rate": 4.592694572825964e-06, + "loss": 0.6654, + "step": 1416 + }, + { + "epoch": 1.173108779679735, + "grad_norm": 0.8747309446334839, + "learning_rate": 4.592092748111931e-06, + "loss": 0.6619, + "step": 1417 + }, + { + "epoch": 1.1739370513528438, + "grad_norm": 0.8836600184440613, + "learning_rate": 4.5914905185974535e-06, + "loss": 0.6443, + "step": 1418 + }, + { + "epoch": 1.1747653230259525, + "grad_norm": 0.9340765476226807, + "learning_rate": 4.590887884399058e-06, + "loss": 0.6633, + "step": 1419 + }, + { + "epoch": 1.1755935946990612, + "grad_norm": 0.8889044523239136, + "learning_rate": 4.590284845633347e-06, + "loss": 0.6489, + "step": 1420 + }, + { + "epoch": 1.1764218663721702, + "grad_norm": 0.9043524265289307, + "learning_rate": 4.589681402417005e-06, + "loss": 0.6392, + "step": 1421 + }, + { + "epoch": 1.177250138045279, + "grad_norm": 0.9210046529769897, + "learning_rate": 4.5890775548667895e-06, + "loss": 0.6105, + "step": 1422 + }, + { + "epoch": 1.1780784097183876, + "grad_norm": 0.8954651951789856, + "learning_rate": 4.588473303099542e-06, + "loss": 0.6425, + "step": 1423 + }, + { + "epoch": 1.1789066813914963, + "grad_norm": 0.945517897605896, + "learning_rate": 4.587868647232179e-06, + "loss": 0.6404, + "step": 1424 + }, + { + "epoch": 1.1797349530646053, + "grad_norm": 0.9059463739395142, + "learning_rate": 4.587263587381696e-06, + "loss": 0.6521, + "step": 1425 + }, + { + "epoch": 1.180563224737714, + "grad_norm": 0.8876296281814575, + "learning_rate": 4.586658123665165e-06, + "loss": 0.6447, + "step": 1426 + }, + { + "epoch": 1.1813914964108228, + "grad_norm": 0.8862487077713013, + "learning_rate": 4.58605225619974e-06, + "loss": 0.6445, + "step": 1427 + }, + { + "epoch": 1.1822197680839315, + "grad_norm": 0.9066351056098938, + "learning_rate": 4.585445985102649e-06, + "loss": 0.6563, + "step": 1428 + }, + { + "epoch": 1.1830480397570402, + "grad_norm": 0.913724958896637, + "learning_rate": 4.584839310491199e-06, + "loss": 0.6507, + "step": 1429 + }, + { + "epoch": 1.1838763114301492, + "grad_norm": 0.8942010402679443, + "learning_rate": 4.584232232482778e-06, + "loss": 0.6424, + "step": 1430 + }, + { + "epoch": 1.1847045831032579, + "grad_norm": 0.8986000418663025, + "learning_rate": 4.58362475119485e-06, + "loss": 0.6491, + "step": 1431 + }, + { + "epoch": 1.1855328547763666, + "grad_norm": 0.8879220485687256, + "learning_rate": 4.5830168667449544e-06, + "loss": 0.6412, + "step": 1432 + }, + { + "epoch": 1.1863611264494753, + "grad_norm": 0.9497302770614624, + "learning_rate": 4.5824085792507135e-06, + "loss": 0.6453, + "step": 1433 + }, + { + "epoch": 1.1871893981225843, + "grad_norm": 0.8933898210525513, + "learning_rate": 4.581799888829824e-06, + "loss": 0.6381, + "step": 1434 + }, + { + "epoch": 1.188017669795693, + "grad_norm": 0.9034610986709595, + "learning_rate": 4.581190795600062e-06, + "loss": 0.6551, + "step": 1435 + }, + { + "epoch": 1.1888459414688017, + "grad_norm": 0.8968634605407715, + "learning_rate": 4.580581299679282e-06, + "loss": 0.637, + "step": 1436 + }, + { + "epoch": 1.1896742131419105, + "grad_norm": 0.8974384069442749, + "learning_rate": 4.579971401185416e-06, + "loss": 0.6301, + "step": 1437 + }, + { + "epoch": 1.1905024848150194, + "grad_norm": 0.8726125359535217, + "learning_rate": 4.579361100236473e-06, + "loss": 0.6487, + "step": 1438 + }, + { + "epoch": 1.1913307564881281, + "grad_norm": 0.9131139516830444, + "learning_rate": 4.5787503969505405e-06, + "loss": 0.6211, + "step": 1439 + }, + { + "epoch": 1.1921590281612369, + "grad_norm": 0.8924953937530518, + "learning_rate": 4.578139291445783e-06, + "loss": 0.6339, + "step": 1440 + }, + { + "epoch": 1.1929872998343456, + "grad_norm": 0.890883207321167, + "learning_rate": 4.577527783840447e-06, + "loss": 0.66, + "step": 1441 + }, + { + "epoch": 1.1938155715074545, + "grad_norm": 0.9326318502426147, + "learning_rate": 4.5769158742528494e-06, + "loss": 0.6611, + "step": 1442 + }, + { + "epoch": 1.1946438431805633, + "grad_norm": 0.9354713559150696, + "learning_rate": 4.576303562801392e-06, + "loss": 0.6445, + "step": 1443 + }, + { + "epoch": 1.195472114853672, + "grad_norm": 0.8868383169174194, + "learning_rate": 4.57569084960455e-06, + "loss": 0.6344, + "step": 1444 + }, + { + "epoch": 1.1963003865267807, + "grad_norm": 0.893924355506897, + "learning_rate": 4.5750777347808775e-06, + "loss": 0.602, + "step": 1445 + }, + { + "epoch": 1.1971286581998895, + "grad_norm": 0.9299339652061462, + "learning_rate": 4.574464218449006e-06, + "loss": 0.6206, + "step": 1446 + }, + { + "epoch": 1.1979569298729984, + "grad_norm": 0.9058142304420471, + "learning_rate": 4.573850300727648e-06, + "loss": 0.6073, + "step": 1447 + }, + { + "epoch": 1.1987852015461071, + "grad_norm": 0.8893733620643616, + "learning_rate": 4.5732359817355885e-06, + "loss": 0.6336, + "step": 1448 + }, + { + "epoch": 1.1996134732192159, + "grad_norm": 0.9018000960350037, + "learning_rate": 4.572621261591692e-06, + "loss": 0.6268, + "step": 1449 + }, + { + "epoch": 1.2004417448923248, + "grad_norm": 0.9087574481964111, + "learning_rate": 4.572006140414903e-06, + "loss": 0.6511, + "step": 1450 + }, + { + "epoch": 1.2012700165654335, + "grad_norm": 0.9092984795570374, + "learning_rate": 4.5713906183242404e-06, + "loss": 0.6424, + "step": 1451 + }, + { + "epoch": 1.2020982882385423, + "grad_norm": 0.9040535092353821, + "learning_rate": 4.5707746954388034e-06, + "loss": 0.6242, + "step": 1452 + }, + { + "epoch": 1.202926559911651, + "grad_norm": 0.891659677028656, + "learning_rate": 4.570158371877766e-06, + "loss": 0.643, + "step": 1453 + }, + { + "epoch": 1.2037548315847597, + "grad_norm": 0.9254544973373413, + "learning_rate": 4.569541647760381e-06, + "loss": 0.6191, + "step": 1454 + }, + { + "epoch": 1.2045831032578687, + "grad_norm": 0.8978536128997803, + "learning_rate": 4.568924523205979e-06, + "loss": 0.6632, + "step": 1455 + }, + { + "epoch": 1.2054113749309774, + "grad_norm": 0.9179201126098633, + "learning_rate": 4.568306998333968e-06, + "loss": 0.6485, + "step": 1456 + }, + { + "epoch": 1.2062396466040861, + "grad_norm": 0.9061882495880127, + "learning_rate": 4.567689073263834e-06, + "loss": 0.6704, + "step": 1457 + }, + { + "epoch": 1.2070679182771948, + "grad_norm": 0.9143427610397339, + "learning_rate": 4.567070748115139e-06, + "loss": 0.6453, + "step": 1458 + }, + { + "epoch": 1.2078961899503038, + "grad_norm": 0.8740119338035583, + "learning_rate": 4.566452023007523e-06, + "loss": 0.6453, + "step": 1459 + }, + { + "epoch": 1.2087244616234125, + "grad_norm": 0.8977463841438293, + "learning_rate": 4.565832898060703e-06, + "loss": 0.6451, + "step": 1460 + }, + { + "epoch": 1.2095527332965212, + "grad_norm": 0.9416611194610596, + "learning_rate": 4.565213373394476e-06, + "loss": 0.6265, + "step": 1461 + }, + { + "epoch": 1.21038100496963, + "grad_norm": 0.9132000207901001, + "learning_rate": 4.5645934491287115e-06, + "loss": 0.6412, + "step": 1462 + }, + { + "epoch": 1.2112092766427387, + "grad_norm": 0.9480498433113098, + "learning_rate": 4.563973125383362e-06, + "loss": 0.6363, + "step": 1463 + }, + { + "epoch": 1.2120375483158476, + "grad_norm": 0.9170709848403931, + "learning_rate": 4.563352402278451e-06, + "loss": 0.6201, + "step": 1464 + }, + { + "epoch": 1.2128658199889564, + "grad_norm": 0.9655918478965759, + "learning_rate": 4.562731279934086e-06, + "loss": 0.6497, + "step": 1465 + }, + { + "epoch": 1.213694091662065, + "grad_norm": 0.8885562419891357, + "learning_rate": 4.562109758470446e-06, + "loss": 0.6265, + "step": 1466 + }, + { + "epoch": 1.214522363335174, + "grad_norm": 0.9072791337966919, + "learning_rate": 4.561487838007792e-06, + "loss": 0.6257, + "step": 1467 + }, + { + "epoch": 1.2153506350082828, + "grad_norm": 0.8992362022399902, + "learning_rate": 4.560865518666458e-06, + "loss": 0.6425, + "step": 1468 + }, + { + "epoch": 1.2161789066813915, + "grad_norm": 0.8934991955757141, + "learning_rate": 4.560242800566855e-06, + "loss": 0.6583, + "step": 1469 + }, + { + "epoch": 1.2170071783545002, + "grad_norm": 0.9292325973510742, + "learning_rate": 4.559619683829477e-06, + "loss": 0.6403, + "step": 1470 + }, + { + "epoch": 1.217835450027609, + "grad_norm": 0.8756814002990723, + "learning_rate": 4.55899616857489e-06, + "loss": 0.6261, + "step": 1471 + }, + { + "epoch": 1.218663721700718, + "grad_norm": 0.9050565361976624, + "learning_rate": 4.558372254923738e-06, + "loss": 0.6325, + "step": 1472 + }, + { + "epoch": 1.2194919933738266, + "grad_norm": 0.9230931401252747, + "learning_rate": 4.557747942996742e-06, + "loss": 0.6275, + "step": 1473 + }, + { + "epoch": 1.2203202650469354, + "grad_norm": 0.894026517868042, + "learning_rate": 4.557123232914702e-06, + "loss": 0.6348, + "step": 1474 + }, + { + "epoch": 1.221148536720044, + "grad_norm": 0.8882637023925781, + "learning_rate": 4.556498124798492e-06, + "loss": 0.6123, + "step": 1475 + }, + { + "epoch": 1.221976808393153, + "grad_norm": 0.9191063642501831, + "learning_rate": 4.555872618769065e-06, + "loss": 0.6531, + "step": 1476 + }, + { + "epoch": 1.2228050800662618, + "grad_norm": 0.9398623704910278, + "learning_rate": 4.555246714947452e-06, + "loss": 0.6553, + "step": 1477 + }, + { + "epoch": 1.2236333517393705, + "grad_norm": 0.9399452805519104, + "learning_rate": 4.554620413454757e-06, + "loss": 0.6485, + "step": 1478 + }, + { + "epoch": 1.2244616234124792, + "grad_norm": 0.9001572728157043, + "learning_rate": 4.553993714412166e-06, + "loss": 0.6467, + "step": 1479 + }, + { + "epoch": 1.2252898950855882, + "grad_norm": 0.9266572594642639, + "learning_rate": 4.553366617940938e-06, + "loss": 0.6325, + "step": 1480 + }, + { + "epoch": 1.226118166758697, + "grad_norm": 0.928301215171814, + "learning_rate": 4.552739124162411e-06, + "loss": 0.6398, + "step": 1481 + }, + { + "epoch": 1.2269464384318056, + "grad_norm": 0.880954921245575, + "learning_rate": 4.552111233197999e-06, + "loss": 0.669, + "step": 1482 + }, + { + "epoch": 1.2277747101049143, + "grad_norm": 0.9089685082435608, + "learning_rate": 4.551482945169191e-06, + "loss": 0.62, + "step": 1483 + }, + { + "epoch": 1.2286029817780233, + "grad_norm": 0.9187189936637878, + "learning_rate": 4.550854260197559e-06, + "loss": 0.6381, + "step": 1484 + }, + { + "epoch": 1.229431253451132, + "grad_norm": 0.8782045245170593, + "learning_rate": 4.550225178404744e-06, + "loss": 0.6397, + "step": 1485 + }, + { + "epoch": 1.2302595251242407, + "grad_norm": 0.8798948526382446, + "learning_rate": 4.54959569991247e-06, + "loss": 0.6522, + "step": 1486 + }, + { + "epoch": 1.2310877967973495, + "grad_norm": 0.9257029294967651, + "learning_rate": 4.548965824842534e-06, + "loss": 0.6522, + "step": 1487 + }, + { + "epoch": 1.2319160684704582, + "grad_norm": 0.8923602104187012, + "learning_rate": 4.548335553316811e-06, + "loss": 0.6538, + "step": 1488 + }, + { + "epoch": 1.2327443401435672, + "grad_norm": 0.9071294069290161, + "learning_rate": 4.5477048854572524e-06, + "loss": 0.6305, + "step": 1489 + }, + { + "epoch": 1.2335726118166759, + "grad_norm": 0.9495804905891418, + "learning_rate": 4.547073821385888e-06, + "loss": 0.6398, + "step": 1490 + }, + { + "epoch": 1.2344008834897846, + "grad_norm": 0.9163299202919006, + "learning_rate": 4.54644236122482e-06, + "loss": 0.6454, + "step": 1491 + }, + { + "epoch": 1.2352291551628933, + "grad_norm": 0.9286475777626038, + "learning_rate": 4.545810505096233e-06, + "loss": 0.6345, + "step": 1492 + }, + { + "epoch": 1.2360574268360023, + "grad_norm": 0.9116879105567932, + "learning_rate": 4.545178253122384e-06, + "loss": 0.6451, + "step": 1493 + }, + { + "epoch": 1.236885698509111, + "grad_norm": 0.917826771736145, + "learning_rate": 4.544545605425607e-06, + "loss": 0.6374, + "step": 1494 + }, + { + "epoch": 1.2377139701822197, + "grad_norm": 0.9433108568191528, + "learning_rate": 4.543912562128316e-06, + "loss": 0.6321, + "step": 1495 + }, + { + "epoch": 1.2385422418553285, + "grad_norm": 0.9017132520675659, + "learning_rate": 4.543279123352996e-06, + "loss": 0.6341, + "step": 1496 + }, + { + "epoch": 1.2393705135284374, + "grad_norm": 0.9557158350944519, + "learning_rate": 4.542645289222214e-06, + "loss": 0.6364, + "step": 1497 + }, + { + "epoch": 1.2401987852015461, + "grad_norm": 0.9258891344070435, + "learning_rate": 4.542011059858609e-06, + "loss": 0.6598, + "step": 1498 + }, + { + "epoch": 1.2410270568746549, + "grad_norm": 0.8988100290298462, + "learning_rate": 4.541376435384899e-06, + "loss": 0.6496, + "step": 1499 + }, + { + "epoch": 1.2418553285477636, + "grad_norm": 0.898154079914093, + "learning_rate": 4.540741415923878e-06, + "loss": 0.6486, + "step": 1500 + }, + { + "epoch": 1.2426836002208725, + "grad_norm": 0.9216037392616272, + "learning_rate": 4.540106001598418e-06, + "loss": 0.6436, + "step": 1501 + }, + { + "epoch": 1.2435118718939813, + "grad_norm": 0.9086095690727234, + "learning_rate": 4.539470192531463e-06, + "loss": 0.6287, + "step": 1502 + }, + { + "epoch": 1.24434014356709, + "grad_norm": 0.921291172504425, + "learning_rate": 4.538833988846039e-06, + "loss": 0.629, + "step": 1503 + }, + { + "epoch": 1.2451684152401987, + "grad_norm": 0.8621019721031189, + "learning_rate": 4.538197390665242e-06, + "loss": 0.6543, + "step": 1504 + }, + { + "epoch": 1.2459966869133074, + "grad_norm": 0.9523105025291443, + "learning_rate": 4.537560398112251e-06, + "loss": 0.6482, + "step": 1505 + }, + { + "epoch": 1.2468249585864164, + "grad_norm": 0.971931517124176, + "learning_rate": 4.536923011310317e-06, + "loss": 0.6463, + "step": 1506 + }, + { + "epoch": 1.2476532302595251, + "grad_norm": 0.927219033241272, + "learning_rate": 4.536285230382767e-06, + "loss": 0.6443, + "step": 1507 + }, + { + "epoch": 1.2484815019326339, + "grad_norm": 0.9433056116104126, + "learning_rate": 4.535647055453007e-06, + "loss": 0.6301, + "step": 1508 + }, + { + "epoch": 1.2493097736057428, + "grad_norm": 0.8992047905921936, + "learning_rate": 4.5350084866445195e-06, + "loss": 0.6177, + "step": 1509 + }, + { + "epoch": 1.2501380452788515, + "grad_norm": 0.8946895003318787, + "learning_rate": 4.534369524080858e-06, + "loss": 0.6275, + "step": 1510 + }, + { + "epoch": 1.2509663169519603, + "grad_norm": 0.8732156157493591, + "learning_rate": 4.5337301678856595e-06, + "loss": 0.6572, + "step": 1511 + }, + { + "epoch": 1.251794588625069, + "grad_norm": 0.9505333304405212, + "learning_rate": 4.533090418182631e-06, + "loss": 0.6352, + "step": 1512 + }, + { + "epoch": 1.2526228602981777, + "grad_norm": 0.8931122422218323, + "learning_rate": 4.532450275095558e-06, + "loss": 0.649, + "step": 1513 + }, + { + "epoch": 1.2534511319712867, + "grad_norm": 0.8781880140304565, + "learning_rate": 4.531809738748304e-06, + "loss": 0.651, + "step": 1514 + }, + { + "epoch": 1.2542794036443954, + "grad_norm": 0.9201871156692505, + "learning_rate": 4.531168809264805e-06, + "loss": 0.6451, + "step": 1515 + }, + { + "epoch": 1.2551076753175041, + "grad_norm": 0.9318148493766785, + "learning_rate": 4.530527486769077e-06, + "loss": 0.6458, + "step": 1516 + }, + { + "epoch": 1.2559359469906128, + "grad_norm": 0.9131407141685486, + "learning_rate": 4.529885771385208e-06, + "loss": 0.6679, + "step": 1517 + }, + { + "epoch": 1.2567642186637218, + "grad_norm": 0.8942316770553589, + "learning_rate": 4.529243663237365e-06, + "loss": 0.6262, + "step": 1518 + }, + { + "epoch": 1.2575924903368305, + "grad_norm": 0.9058274030685425, + "learning_rate": 4.528601162449788e-06, + "loss": 0.635, + "step": 1519 + }, + { + "epoch": 1.2584207620099392, + "grad_norm": 0.9129453897476196, + "learning_rate": 4.527958269146798e-06, + "loss": 0.6375, + "step": 1520 + }, + { + "epoch": 1.259249033683048, + "grad_norm": 0.8737504482269287, + "learning_rate": 4.527314983452787e-06, + "loss": 0.6437, + "step": 1521 + }, + { + "epoch": 1.2600773053561567, + "grad_norm": 0.9680130481719971, + "learning_rate": 4.526671305492225e-06, + "loss": 0.6546, + "step": 1522 + }, + { + "epoch": 1.2609055770292656, + "grad_norm": 0.9186084270477295, + "learning_rate": 4.526027235389658e-06, + "loss": 0.6505, + "step": 1523 + }, + { + "epoch": 1.2617338487023744, + "grad_norm": 0.9164871573448181, + "learning_rate": 4.525382773269706e-06, + "loss": 0.6541, + "step": 1524 + }, + { + "epoch": 1.262562120375483, + "grad_norm": 0.8994123935699463, + "learning_rate": 4.5247379192570695e-06, + "loss": 0.6313, + "step": 1525 + }, + { + "epoch": 1.263390392048592, + "grad_norm": 0.8780331015586853, + "learning_rate": 4.524092673476519e-06, + "loss": 0.6496, + "step": 1526 + }, + { + "epoch": 1.2642186637217008, + "grad_norm": 0.8670293092727661, + "learning_rate": 4.5234470360529045e-06, + "loss": 0.6422, + "step": 1527 + }, + { + "epoch": 1.2650469353948095, + "grad_norm": 0.9305918216705322, + "learning_rate": 4.522801007111152e-06, + "loss": 0.6352, + "step": 1528 + }, + { + "epoch": 1.2658752070679182, + "grad_norm": 0.9444066286087036, + "learning_rate": 4.52215458677626e-06, + "loss": 0.6357, + "step": 1529 + }, + { + "epoch": 1.266703478741027, + "grad_norm": 0.8987369537353516, + "learning_rate": 4.521507775173305e-06, + "loss": 0.6407, + "step": 1530 + }, + { + "epoch": 1.267531750414136, + "grad_norm": 0.9200116395950317, + "learning_rate": 4.5208605724274415e-06, + "loss": 0.6366, + "step": 1531 + }, + { + "epoch": 1.2683600220872446, + "grad_norm": 0.8680561780929565, + "learning_rate": 4.520212978663894e-06, + "loss": 0.6389, + "step": 1532 + }, + { + "epoch": 1.2691882937603534, + "grad_norm": 0.8928949236869812, + "learning_rate": 4.519564994007968e-06, + "loss": 0.6552, + "step": 1533 + }, + { + "epoch": 1.2700165654334623, + "grad_norm": 0.8986184000968933, + "learning_rate": 4.518916618585042e-06, + "loss": 0.6554, + "step": 1534 + }, + { + "epoch": 1.270844837106571, + "grad_norm": 0.9239423274993896, + "learning_rate": 4.51826785252057e-06, + "loss": 0.6448, + "step": 1535 + }, + { + "epoch": 1.2716731087796798, + "grad_norm": 0.9534674286842346, + "learning_rate": 4.517618695940082e-06, + "loss": 0.6303, + "step": 1536 + }, + { + "epoch": 1.2725013804527885, + "grad_norm": 0.9341598749160767, + "learning_rate": 4.516969148969186e-06, + "loss": 0.6335, + "step": 1537 + }, + { + "epoch": 1.2733296521258972, + "grad_norm": 0.9264217615127563, + "learning_rate": 4.51631921173356e-06, + "loss": 0.6543, + "step": 1538 + }, + { + "epoch": 1.274157923799006, + "grad_norm": 0.9460370540618896, + "learning_rate": 4.515668884358963e-06, + "loss": 0.6531, + "step": 1539 + }, + { + "epoch": 1.274986195472115, + "grad_norm": 0.9690954089164734, + "learning_rate": 4.515018166971227e-06, + "loss": 0.6495, + "step": 1540 + }, + { + "epoch": 1.2758144671452236, + "grad_norm": 0.8746740818023682, + "learning_rate": 4.514367059696259e-06, + "loss": 0.6169, + "step": 1541 + }, + { + "epoch": 1.2766427388183323, + "grad_norm": 0.9288576245307922, + "learning_rate": 4.513715562660043e-06, + "loss": 0.6459, + "step": 1542 + }, + { + "epoch": 1.2774710104914413, + "grad_norm": 0.9935420751571655, + "learning_rate": 4.513063675988637e-06, + "loss": 0.6471, + "step": 1543 + }, + { + "epoch": 1.27829928216455, + "grad_norm": 0.9584968686103821, + "learning_rate": 4.512411399808175e-06, + "loss": 0.6473, + "step": 1544 + }, + { + "epoch": 1.2791275538376587, + "grad_norm": 0.9049419164657593, + "learning_rate": 4.511758734244867e-06, + "loss": 0.6667, + "step": 1545 + }, + { + "epoch": 1.2799558255107675, + "grad_norm": 0.9456910490989685, + "learning_rate": 4.511105679424998e-06, + "loss": 0.625, + "step": 1546 + }, + { + "epoch": 1.2807840971838762, + "grad_norm": 0.9267060160636902, + "learning_rate": 4.510452235474926e-06, + "loss": 0.6239, + "step": 1547 + }, + { + "epoch": 1.2816123688569852, + "grad_norm": 0.9363065958023071, + "learning_rate": 4.509798402521088e-06, + "loss": 0.6416, + "step": 1548 + }, + { + "epoch": 1.2824406405300939, + "grad_norm": 0.85664302110672, + "learning_rate": 4.509144180689995e-06, + "loss": 0.6297, + "step": 1549 + }, + { + "epoch": 1.2832689122032026, + "grad_norm": 0.9298888444900513, + "learning_rate": 4.508489570108231e-06, + "loss": 0.6305, + "step": 1550 + }, + { + "epoch": 1.2840971838763116, + "grad_norm": 0.9485051035881042, + "learning_rate": 4.507834570902459e-06, + "loss": 0.6409, + "step": 1551 + }, + { + "epoch": 1.2849254555494203, + "grad_norm": 0.9129849076271057, + "learning_rate": 4.507179183199414e-06, + "loss": 0.6215, + "step": 1552 + }, + { + "epoch": 1.285753727222529, + "grad_norm": 0.9630280137062073, + "learning_rate": 4.506523407125907e-06, + "loss": 0.6229, + "step": 1553 + }, + { + "epoch": 1.2865819988956377, + "grad_norm": 0.9047848582267761, + "learning_rate": 4.505867242808826e-06, + "loss": 0.6434, + "step": 1554 + }, + { + "epoch": 1.2874102705687465, + "grad_norm": 0.9219448566436768, + "learning_rate": 4.5052106903751315e-06, + "loss": 0.6449, + "step": 1555 + }, + { + "epoch": 1.2882385422418554, + "grad_norm": 0.9347684383392334, + "learning_rate": 4.504553749951861e-06, + "loss": 0.6272, + "step": 1556 + }, + { + "epoch": 1.2890668139149641, + "grad_norm": 0.8930262923240662, + "learning_rate": 4.503896421666126e-06, + "loss": 0.6364, + "step": 1557 + }, + { + "epoch": 1.2898950855880729, + "grad_norm": 0.9067763090133667, + "learning_rate": 4.503238705645114e-06, + "loss": 0.6317, + "step": 1558 + }, + { + "epoch": 1.2907233572611816, + "grad_norm": 0.9524715542793274, + "learning_rate": 4.5025806020160865e-06, + "loss": 0.6525, + "step": 1559 + }, + { + "epoch": 1.2915516289342905, + "grad_norm": 0.927255392074585, + "learning_rate": 4.5019221109063795e-06, + "loss": 0.6343, + "step": 1560 + }, + { + "epoch": 1.2923799006073993, + "grad_norm": 0.904295802116394, + "learning_rate": 4.5012632324434065e-06, + "loss": 0.6484, + "step": 1561 + }, + { + "epoch": 1.293208172280508, + "grad_norm": 0.9138978123664856, + "learning_rate": 4.500603966754653e-06, + "loss": 0.6374, + "step": 1562 + }, + { + "epoch": 1.2940364439536167, + "grad_norm": 0.9038779139518738, + "learning_rate": 4.499944313967681e-06, + "loss": 0.6395, + "step": 1563 + }, + { + "epoch": 1.2948647156267254, + "grad_norm": 0.9218693971633911, + "learning_rate": 4.499284274210128e-06, + "loss": 0.6325, + "step": 1564 + }, + { + "epoch": 1.2956929872998344, + "grad_norm": 0.9210799932479858, + "learning_rate": 4.4986238476097055e-06, + "loss": 0.6476, + "step": 1565 + }, + { + "epoch": 1.2965212589729431, + "grad_norm": 0.8977149724960327, + "learning_rate": 4.4979630342941994e-06, + "loss": 0.6489, + "step": 1566 + }, + { + "epoch": 1.2973495306460519, + "grad_norm": 0.9423681497573853, + "learning_rate": 4.49730183439147e-06, + "loss": 0.6391, + "step": 1567 + }, + { + "epoch": 1.2981778023191608, + "grad_norm": 0.9404032826423645, + "learning_rate": 4.4966402480294545e-06, + "loss": 0.6233, + "step": 1568 + }, + { + "epoch": 1.2990060739922695, + "grad_norm": 0.9104741215705872, + "learning_rate": 4.495978275336164e-06, + "loss": 0.6369, + "step": 1569 + }, + { + "epoch": 1.2998343456653783, + "grad_norm": 0.8869790434837341, + "learning_rate": 4.4953159164396835e-06, + "loss": 0.6331, + "step": 1570 + }, + { + "epoch": 1.300662617338487, + "grad_norm": 0.9269986152648926, + "learning_rate": 4.494653171468173e-06, + "loss": 0.6361, + "step": 1571 + }, + { + "epoch": 1.3014908890115957, + "grad_norm": 0.9210803508758545, + "learning_rate": 4.4939900405498675e-06, + "loss": 0.6314, + "step": 1572 + }, + { + "epoch": 1.3023191606847047, + "grad_norm": 0.9426820874214172, + "learning_rate": 4.4933265238130775e-06, + "loss": 0.6625, + "step": 1573 + }, + { + "epoch": 1.3031474323578134, + "grad_norm": 0.9198150634765625, + "learning_rate": 4.492662621386186e-06, + "loss": 0.627, + "step": 1574 + }, + { + "epoch": 1.3039757040309221, + "grad_norm": 0.9312534928321838, + "learning_rate": 4.4919983333976525e-06, + "loss": 0.6386, + "step": 1575 + }, + { + "epoch": 1.304803975704031, + "grad_norm": 0.8926329016685486, + "learning_rate": 4.491333659976011e-06, + "loss": 0.6559, + "step": 1576 + }, + { + "epoch": 1.3056322473771398, + "grad_norm": 0.9172425270080566, + "learning_rate": 4.4906686012498705e-06, + "loss": 0.6254, + "step": 1577 + }, + { + "epoch": 1.3064605190502485, + "grad_norm": 0.9896116256713867, + "learning_rate": 4.490003157347911e-06, + "loss": 0.6331, + "step": 1578 + }, + { + "epoch": 1.3072887907233572, + "grad_norm": 0.9267039895057678, + "learning_rate": 4.489337328398891e-06, + "loss": 0.6296, + "step": 1579 + }, + { + "epoch": 1.308117062396466, + "grad_norm": 0.9033195376396179, + "learning_rate": 4.488671114531643e-06, + "loss": 0.6354, + "step": 1580 + }, + { + "epoch": 1.3089453340695747, + "grad_norm": 0.9047731161117554, + "learning_rate": 4.488004515875072e-06, + "loss": 0.6351, + "step": 1581 + }, + { + "epoch": 1.3097736057426836, + "grad_norm": 0.9096142649650574, + "learning_rate": 4.4873375325581596e-06, + "loss": 0.6504, + "step": 1582 + }, + { + "epoch": 1.3106018774157924, + "grad_norm": 0.8848153352737427, + "learning_rate": 4.48667016470996e-06, + "loss": 0.6354, + "step": 1583 + }, + { + "epoch": 1.311430149088901, + "grad_norm": 0.9036188721656799, + "learning_rate": 4.486002412459603e-06, + "loss": 0.6431, + "step": 1584 + }, + { + "epoch": 1.31225842076201, + "grad_norm": 0.8881173133850098, + "learning_rate": 4.4853342759362924e-06, + "loss": 0.6509, + "step": 1585 + }, + { + "epoch": 1.3130866924351188, + "grad_norm": 0.9195591807365417, + "learning_rate": 4.484665755269307e-06, + "loss": 0.6522, + "step": 1586 + }, + { + "epoch": 1.3139149641082275, + "grad_norm": 0.9021509885787964, + "learning_rate": 4.483996850587998e-06, + "loss": 0.6532, + "step": 1587 + }, + { + "epoch": 1.3147432357813362, + "grad_norm": 0.8902403116226196, + "learning_rate": 4.483327562021794e-06, + "loss": 0.5942, + "step": 1588 + }, + { + "epoch": 1.315571507454445, + "grad_norm": 0.8933504819869995, + "learning_rate": 4.482657889700194e-06, + "loss": 0.6305, + "step": 1589 + }, + { + "epoch": 1.316399779127554, + "grad_norm": 0.9233856201171875, + "learning_rate": 4.481987833752776e-06, + "loss": 0.6275, + "step": 1590 + }, + { + "epoch": 1.3172280508006626, + "grad_norm": 0.9482141733169556, + "learning_rate": 4.4813173943091865e-06, + "loss": 0.6232, + "step": 1591 + }, + { + "epoch": 1.3180563224737714, + "grad_norm": 0.94732266664505, + "learning_rate": 4.480646571499151e-06, + "loss": 0.6297, + "step": 1592 + }, + { + "epoch": 1.3188845941468803, + "grad_norm": 0.9066787958145142, + "learning_rate": 4.479975365452468e-06, + "loss": 0.6334, + "step": 1593 + }, + { + "epoch": 1.319712865819989, + "grad_norm": 0.9063938856124878, + "learning_rate": 4.479303776299008e-06, + "loss": 0.6415, + "step": 1594 + }, + { + "epoch": 1.3205411374930978, + "grad_norm": 0.9286819696426392, + "learning_rate": 4.478631804168719e-06, + "loss": 0.6534, + "step": 1595 + }, + { + "epoch": 1.3213694091662065, + "grad_norm": 0.9161240458488464, + "learning_rate": 4.47795944919162e-06, + "loss": 0.6401, + "step": 1596 + }, + { + "epoch": 1.3221976808393152, + "grad_norm": 0.956538200378418, + "learning_rate": 4.4772867114978056e-06, + "loss": 0.6649, + "step": 1597 + }, + { + "epoch": 1.323025952512424, + "grad_norm": 0.910504162311554, + "learning_rate": 4.4766135912174455e-06, + "loss": 0.6456, + "step": 1598 + }, + { + "epoch": 1.323854224185533, + "grad_norm": 0.9054026007652283, + "learning_rate": 4.4759400884807805e-06, + "loss": 0.6458, + "step": 1599 + }, + { + "epoch": 1.3246824958586416, + "grad_norm": 0.933506429195404, + "learning_rate": 4.475266203418127e-06, + "loss": 0.626, + "step": 1600 + }, + { + "epoch": 1.3255107675317503, + "grad_norm": 0.9042014479637146, + "learning_rate": 4.474591936159878e-06, + "loss": 0.6376, + "step": 1601 + }, + { + "epoch": 1.3263390392048593, + "grad_norm": 0.9005138278007507, + "learning_rate": 4.473917286836496e-06, + "loss": 0.6093, + "step": 1602 + }, + { + "epoch": 1.327167310877968, + "grad_norm": 0.9096264839172363, + "learning_rate": 4.4732422555785196e-06, + "loss": 0.6344, + "step": 1603 + }, + { + "epoch": 1.3279955825510767, + "grad_norm": 0.9171631336212158, + "learning_rate": 4.4725668425165605e-06, + "loss": 0.6454, + "step": 1604 + }, + { + "epoch": 1.3288238542241855, + "grad_norm": 0.8821735382080078, + "learning_rate": 4.4718910477813055e-06, + "loss": 0.6357, + "step": 1605 + }, + { + "epoch": 1.3296521258972942, + "grad_norm": 0.918499231338501, + "learning_rate": 4.471214871503514e-06, + "loss": 0.6361, + "step": 1606 + }, + { + "epoch": 1.3304803975704032, + "grad_norm": 0.9187904596328735, + "learning_rate": 4.470538313814021e-06, + "loss": 0.6362, + "step": 1607 + }, + { + "epoch": 1.3313086692435119, + "grad_norm": 0.9272632598876953, + "learning_rate": 4.469861374843734e-06, + "loss": 0.6659, + "step": 1608 + }, + { + "epoch": 1.3321369409166206, + "grad_norm": 0.8950392603874207, + "learning_rate": 4.469184054723632e-06, + "loss": 0.6551, + "step": 1609 + }, + { + "epoch": 1.3329652125897296, + "grad_norm": 0.9604961276054382, + "learning_rate": 4.468506353584773e-06, + "loss": 0.643, + "step": 1610 + }, + { + "epoch": 1.3337934842628383, + "grad_norm": 0.933073103427887, + "learning_rate": 4.467828271558286e-06, + "loss": 0.6193, + "step": 1611 + }, + { + "epoch": 1.334621755935947, + "grad_norm": 0.890964686870575, + "learning_rate": 4.467149808775371e-06, + "loss": 0.6423, + "step": 1612 + }, + { + "epoch": 1.3354500276090557, + "grad_norm": 0.8916820883750916, + "learning_rate": 4.466470965367306e-06, + "loss": 0.6399, + "step": 1613 + }, + { + "epoch": 1.3362782992821645, + "grad_norm": 0.9151459336280823, + "learning_rate": 4.465791741465439e-06, + "loss": 0.6439, + "step": 1614 + }, + { + "epoch": 1.3371065709552734, + "grad_norm": 0.9155884385108948, + "learning_rate": 4.465112137201197e-06, + "loss": 0.6516, + "step": 1615 + }, + { + "epoch": 1.3379348426283821, + "grad_norm": 0.9187855124473572, + "learning_rate": 4.464432152706073e-06, + "loss": 0.6263, + "step": 1616 + }, + { + "epoch": 1.3387631143014909, + "grad_norm": 0.9506946206092834, + "learning_rate": 4.46375178811164e-06, + "loss": 0.6586, + "step": 1617 + }, + { + "epoch": 1.3395913859745996, + "grad_norm": 0.9056413769721985, + "learning_rate": 4.463071043549543e-06, + "loss": 0.6301, + "step": 1618 + }, + { + "epoch": 1.3404196576477085, + "grad_norm": 0.9056457877159119, + "learning_rate": 4.462389919151498e-06, + "loss": 0.6352, + "step": 1619 + }, + { + "epoch": 1.3412479293208173, + "grad_norm": 0.9204022288322449, + "learning_rate": 4.461708415049297e-06, + "loss": 0.6446, + "step": 1620 + }, + { + "epoch": 1.342076200993926, + "grad_norm": 0.9411336183547974, + "learning_rate": 4.461026531374804e-06, + "loss": 0.6594, + "step": 1621 + }, + { + "epoch": 1.3429044726670347, + "grad_norm": 0.8956605792045593, + "learning_rate": 4.460344268259958e-06, + "loss": 0.6492, + "step": 1622 + }, + { + "epoch": 1.3437327443401434, + "grad_norm": 0.9010135531425476, + "learning_rate": 4.45966162583677e-06, + "loss": 0.6233, + "step": 1623 + }, + { + "epoch": 1.3445610160132524, + "grad_norm": 0.8948563933372498, + "learning_rate": 4.458978604237325e-06, + "loss": 0.6447, + "step": 1624 + }, + { + "epoch": 1.3453892876863611, + "grad_norm": 0.9227688312530518, + "learning_rate": 4.458295203593782e-06, + "loss": 0.6459, + "step": 1625 + }, + { + "epoch": 1.3462175593594699, + "grad_norm": 0.9532939195632935, + "learning_rate": 4.4576114240383725e-06, + "loss": 0.6724, + "step": 1626 + }, + { + "epoch": 1.3470458310325788, + "grad_norm": 0.9089633226394653, + "learning_rate": 4.4569272657034005e-06, + "loss": 0.6303, + "step": 1627 + }, + { + "epoch": 1.3478741027056875, + "grad_norm": 0.8903496861457825, + "learning_rate": 4.456242728721244e-06, + "loss": 0.6655, + "step": 1628 + }, + { + "epoch": 1.3487023743787963, + "grad_norm": 0.9009859561920166, + "learning_rate": 4.455557813224356e-06, + "loss": 0.6268, + "step": 1629 + }, + { + "epoch": 1.349530646051905, + "grad_norm": 0.9265439510345459, + "learning_rate": 4.454872519345261e-06, + "loss": 0.6586, + "step": 1630 + }, + { + "epoch": 1.3503589177250137, + "grad_norm": 0.8945439457893372, + "learning_rate": 4.454186847216556e-06, + "loss": 0.62, + "step": 1631 + }, + { + "epoch": 1.3511871893981227, + "grad_norm": 0.9313004016876221, + "learning_rate": 4.453500796970913e-06, + "loss": 0.6452, + "step": 1632 + }, + { + "epoch": 1.3520154610712314, + "grad_norm": 0.9005258083343506, + "learning_rate": 4.452814368741076e-06, + "loss": 0.6436, + "step": 1633 + }, + { + "epoch": 1.35284373274434, + "grad_norm": 0.8860014081001282, + "learning_rate": 4.452127562659864e-06, + "loss": 0.6423, + "step": 1634 + }, + { + "epoch": 1.353672004417449, + "grad_norm": 0.9049930572509766, + "learning_rate": 4.4514403788601645e-06, + "loss": 0.6343, + "step": 1635 + }, + { + "epoch": 1.3545002760905578, + "grad_norm": 0.8849244117736816, + "learning_rate": 4.4507528174749425e-06, + "loss": 0.6481, + "step": 1636 + }, + { + "epoch": 1.3553285477636665, + "grad_norm": 0.9214544892311096, + "learning_rate": 4.450064878637235e-06, + "loss": 0.6166, + "step": 1637 + }, + { + "epoch": 1.3561568194367752, + "grad_norm": 0.8948760628700256, + "learning_rate": 4.449376562480153e-06, + "loss": 0.6195, + "step": 1638 + }, + { + "epoch": 1.356985091109884, + "grad_norm": 0.9000245928764343, + "learning_rate": 4.448687869136878e-06, + "loss": 0.6378, + "step": 1639 + }, + { + "epoch": 1.3578133627829927, + "grad_norm": 0.8963801264762878, + "learning_rate": 4.447998798740665e-06, + "loss": 0.6118, + "step": 1640 + }, + { + "epoch": 1.3586416344561016, + "grad_norm": 0.8774341344833374, + "learning_rate": 4.447309351424843e-06, + "loss": 0.6184, + "step": 1641 + }, + { + "epoch": 1.3594699061292104, + "grad_norm": 0.8901870846748352, + "learning_rate": 4.446619527322815e-06, + "loss": 0.6499, + "step": 1642 + }, + { + "epoch": 1.360298177802319, + "grad_norm": 0.9108647108078003, + "learning_rate": 4.445929326568055e-06, + "loss": 0.6243, + "step": 1643 + }, + { + "epoch": 1.361126449475428, + "grad_norm": 0.9178787469863892, + "learning_rate": 4.445238749294109e-06, + "loss": 0.6458, + "step": 1644 + }, + { + "epoch": 1.3619547211485368, + "grad_norm": 0.9278320670127869, + "learning_rate": 4.444547795634598e-06, + "loss": 0.6281, + "step": 1645 + }, + { + "epoch": 1.3627829928216455, + "grad_norm": 0.8769298195838928, + "learning_rate": 4.443856465723216e-06, + "loss": 0.6387, + "step": 1646 + }, + { + "epoch": 1.3636112644947542, + "grad_norm": 0.9313247203826904, + "learning_rate": 4.443164759693729e-06, + "loss": 0.6304, + "step": 1647 + }, + { + "epoch": 1.364439536167863, + "grad_norm": 1.0037778615951538, + "learning_rate": 4.4424726776799745e-06, + "loss": 0.6378, + "step": 1648 + }, + { + "epoch": 1.365267807840972, + "grad_norm": 0.9090721011161804, + "learning_rate": 4.441780219815863e-06, + "loss": 0.6456, + "step": 1649 + }, + { + "epoch": 1.3660960795140806, + "grad_norm": 0.8968444466590881, + "learning_rate": 4.441087386235382e-06, + "loss": 0.6377, + "step": 1650 + }, + { + "epoch": 1.3669243511871894, + "grad_norm": 0.9638763070106506, + "learning_rate": 4.440394177072586e-06, + "loss": 0.651, + "step": 1651 + }, + { + "epoch": 1.3677526228602983, + "grad_norm": 0.9224078059196472, + "learning_rate": 4.439700592461604e-06, + "loss": 0.6492, + "step": 1652 + }, + { + "epoch": 1.368580894533407, + "grad_norm": 0.891180157661438, + "learning_rate": 4.439006632536639e-06, + "loss": 0.6402, + "step": 1653 + }, + { + "epoch": 1.3694091662065158, + "grad_norm": 0.8817652463912964, + "learning_rate": 4.438312297431967e-06, + "loss": 0.6095, + "step": 1654 + }, + { + "epoch": 1.3702374378796245, + "grad_norm": 0.8814468383789062, + "learning_rate": 4.437617587281932e-06, + "loss": 0.6334, + "step": 1655 + }, + { + "epoch": 1.3710657095527332, + "grad_norm": 0.8895065784454346, + "learning_rate": 4.436922502220958e-06, + "loss": 0.6163, + "step": 1656 + }, + { + "epoch": 1.371893981225842, + "grad_norm": 0.9043355584144592, + "learning_rate": 4.436227042383535e-06, + "loss": 0.6625, + "step": 1657 + }, + { + "epoch": 1.372722252898951, + "grad_norm": 0.942014753818512, + "learning_rate": 4.435531207904229e-06, + "loss": 0.6298, + "step": 1658 + }, + { + "epoch": 1.3735505245720596, + "grad_norm": 0.8830228447914124, + "learning_rate": 4.434834998917677e-06, + "loss": 0.6215, + "step": 1659 + }, + { + "epoch": 1.3743787962451683, + "grad_norm": 0.9138136506080627, + "learning_rate": 4.434138415558588e-06, + "loss": 0.63, + "step": 1660 + }, + { + "epoch": 1.3752070679182773, + "grad_norm": 0.8880951404571533, + "learning_rate": 4.433441457961746e-06, + "loss": 0.6411, + "step": 1661 + }, + { + "epoch": 1.376035339591386, + "grad_norm": 0.9435725808143616, + "learning_rate": 4.432744126262005e-06, + "loss": 0.6279, + "step": 1662 + }, + { + "epoch": 1.3768636112644947, + "grad_norm": 0.894296407699585, + "learning_rate": 4.432046420594292e-06, + "loss": 0.6298, + "step": 1663 + }, + { + "epoch": 1.3776918829376035, + "grad_norm": 0.8850218653678894, + "learning_rate": 4.431348341093608e-06, + "loss": 0.627, + "step": 1664 + }, + { + "epoch": 1.3785201546107122, + "grad_norm": 0.8956788182258606, + "learning_rate": 4.430649887895025e-06, + "loss": 0.6412, + "step": 1665 + }, + { + "epoch": 1.3793484262838211, + "grad_norm": 0.9128378033638, + "learning_rate": 4.429951061133683e-06, + "loss": 0.6208, + "step": 1666 + }, + { + "epoch": 1.3801766979569299, + "grad_norm": 0.8763598203659058, + "learning_rate": 4.429251860944803e-06, + "loss": 0.6317, + "step": 1667 + }, + { + "epoch": 1.3810049696300386, + "grad_norm": 0.9210475087165833, + "learning_rate": 4.428552287463672e-06, + "loss": 0.6321, + "step": 1668 + }, + { + "epoch": 1.3818332413031476, + "grad_norm": 0.9203745126724243, + "learning_rate": 4.427852340825651e-06, + "loss": 0.6442, + "step": 1669 + }, + { + "epoch": 1.3826615129762563, + "grad_norm": 0.9229859709739685, + "learning_rate": 4.4271520211661735e-06, + "loss": 0.6402, + "step": 1670 + }, + { + "epoch": 1.383489784649365, + "grad_norm": 0.9455426335334778, + "learning_rate": 4.426451328620744e-06, + "loss": 0.6402, + "step": 1671 + }, + { + "epoch": 1.3843180563224737, + "grad_norm": 0.8894345760345459, + "learning_rate": 4.425750263324942e-06, + "loss": 0.6429, + "step": 1672 + }, + { + "epoch": 1.3851463279955825, + "grad_norm": 0.871449887752533, + "learning_rate": 4.425048825414414e-06, + "loss": 0.6449, + "step": 1673 + }, + { + "epoch": 1.3859745996686914, + "grad_norm": 0.9102624654769897, + "learning_rate": 4.424347015024884e-06, + "loss": 0.634, + "step": 1674 + }, + { + "epoch": 1.3868028713418001, + "grad_norm": 0.881172776222229, + "learning_rate": 4.4236448322921445e-06, + "loss": 0.621, + "step": 1675 + }, + { + "epoch": 1.3876311430149089, + "grad_norm": 0.9143398404121399, + "learning_rate": 4.422942277352063e-06, + "loss": 0.6379, + "step": 1676 + }, + { + "epoch": 1.3884594146880176, + "grad_norm": 0.9155046343803406, + "learning_rate": 4.4222393503405754e-06, + "loss": 0.6168, + "step": 1677 + }, + { + "epoch": 1.3892876863611265, + "grad_norm": 0.9501062631607056, + "learning_rate": 4.421536051393694e-06, + "loss": 0.6238, + "step": 1678 + }, + { + "epoch": 1.3901159580342353, + "grad_norm": 0.9055907130241394, + "learning_rate": 4.420832380647498e-06, + "loss": 0.634, + "step": 1679 + }, + { + "epoch": 1.390944229707344, + "grad_norm": 0.9199142456054688, + "learning_rate": 4.420128338238141e-06, + "loss": 0.6283, + "step": 1680 + }, + { + "epoch": 1.3917725013804527, + "grad_norm": 0.8932124376296997, + "learning_rate": 4.419423924301851e-06, + "loss": 0.6146, + "step": 1681 + }, + { + "epoch": 1.3926007730535614, + "grad_norm": 0.9010130167007446, + "learning_rate": 4.418719138974924e-06, + "loss": 0.6409, + "step": 1682 + }, + { + "epoch": 1.3934290447266704, + "grad_norm": 0.9154212474822998, + "learning_rate": 4.41801398239373e-06, + "loss": 0.6115, + "step": 1683 + }, + { + "epoch": 1.3942573163997791, + "grad_norm": 0.9208158850669861, + "learning_rate": 4.417308454694709e-06, + "loss": 0.6242, + "step": 1684 + }, + { + "epoch": 1.3950855880728878, + "grad_norm": 0.932996928691864, + "learning_rate": 4.4166025560143756e-06, + "loss": 0.6277, + "step": 1685 + }, + { + "epoch": 1.3959138597459968, + "grad_norm": 0.8990349769592285, + "learning_rate": 4.415896286489314e-06, + "loss": 0.6128, + "step": 1686 + }, + { + "epoch": 1.3967421314191055, + "grad_norm": 0.926906168460846, + "learning_rate": 4.4151896462561805e-06, + "loss": 0.6442, + "step": 1687 + }, + { + "epoch": 1.3975704030922143, + "grad_norm": 0.9330859184265137, + "learning_rate": 4.414482635451704e-06, + "loss": 0.5995, + "step": 1688 + }, + { + "epoch": 1.398398674765323, + "grad_norm": 0.9039576649665833, + "learning_rate": 4.413775254212683e-06, + "loss": 0.6363, + "step": 1689 + }, + { + "epoch": 1.3992269464384317, + "grad_norm": 0.9228506684303284, + "learning_rate": 4.413067502675992e-06, + "loss": 0.6294, + "step": 1690 + }, + { + "epoch": 1.4000552181115407, + "grad_norm": 0.921620786190033, + "learning_rate": 4.412359380978571e-06, + "loss": 0.64, + "step": 1691 + }, + { + "epoch": 1.4008834897846494, + "grad_norm": 0.9379149079322815, + "learning_rate": 4.4116508892574375e-06, + "loss": 0.6207, + "step": 1692 + }, + { + "epoch": 1.401711761457758, + "grad_norm": 1.0528016090393066, + "learning_rate": 4.410942027649677e-06, + "loss": 0.6204, + "step": 1693 + }, + { + "epoch": 1.402540033130867, + "grad_norm": 0.9510579705238342, + "learning_rate": 4.410232796292447e-06, + "loss": 0.6314, + "step": 1694 + }, + { + "epoch": 1.4033683048039758, + "grad_norm": 0.8922870755195618, + "learning_rate": 4.409523195322979e-06, + "loss": 0.6056, + "step": 1695 + }, + { + "epoch": 1.4041965764770845, + "grad_norm": 0.9239864349365234, + "learning_rate": 4.408813224878573e-06, + "loss": 0.6404, + "step": 1696 + }, + { + "epoch": 1.4050248481501932, + "grad_norm": 0.8696756362915039, + "learning_rate": 4.408102885096601e-06, + "loss": 0.6553, + "step": 1697 + }, + { + "epoch": 1.405853119823302, + "grad_norm": 0.9176343679428101, + "learning_rate": 4.407392176114509e-06, + "loss": 0.6286, + "step": 1698 + }, + { + "epoch": 1.4066813914964107, + "grad_norm": 0.9825496077537537, + "learning_rate": 4.40668109806981e-06, + "loss": 0.6421, + "step": 1699 + }, + { + "epoch": 1.4075096631695196, + "grad_norm": 0.9334873557090759, + "learning_rate": 4.405969651100093e-06, + "loss": 0.6347, + "step": 1700 + }, + { + "epoch": 1.4083379348426284, + "grad_norm": 0.8881192803382874, + "learning_rate": 4.405257835343015e-06, + "loss": 0.6377, + "step": 1701 + }, + { + "epoch": 1.409166206515737, + "grad_norm": 0.8896775245666504, + "learning_rate": 4.404545650936307e-06, + "loss": 0.624, + "step": 1702 + }, + { + "epoch": 1.409994478188846, + "grad_norm": 0.9361454844474792, + "learning_rate": 4.40383309801777e-06, + "loss": 0.6561, + "step": 1703 + }, + { + "epoch": 1.4108227498619548, + "grad_norm": 0.9082345962524414, + "learning_rate": 4.403120176725276e-06, + "loss": 0.6415, + "step": 1704 + }, + { + "epoch": 1.4116510215350635, + "grad_norm": 0.9453962445259094, + "learning_rate": 4.402406887196767e-06, + "loss": 0.6443, + "step": 1705 + }, + { + "epoch": 1.4124792932081722, + "grad_norm": 0.9420861005783081, + "learning_rate": 4.401693229570261e-06, + "loss": 0.6544, + "step": 1706 + }, + { + "epoch": 1.413307564881281, + "grad_norm": 0.869742214679718, + "learning_rate": 4.400979203983841e-06, + "loss": 0.6072, + "step": 1707 + }, + { + "epoch": 1.41413583655439, + "grad_norm": 0.915322482585907, + "learning_rate": 4.4002648105756665e-06, + "loss": 0.6657, + "step": 1708 + }, + { + "epoch": 1.4149641082274986, + "grad_norm": 0.9029238224029541, + "learning_rate": 4.3995500494839656e-06, + "loss": 0.6157, + "step": 1709 + }, + { + "epoch": 1.4157923799006074, + "grad_norm": 0.8992076516151428, + "learning_rate": 4.398834920847037e-06, + "loss": 0.6321, + "step": 1710 + }, + { + "epoch": 1.4166206515737163, + "grad_norm": 0.9099612236022949, + "learning_rate": 4.398119424803252e-06, + "loss": 0.6123, + "step": 1711 + }, + { + "epoch": 1.417448923246825, + "grad_norm": 0.9081803560256958, + "learning_rate": 4.397403561491052e-06, + "loss": 0.6369, + "step": 1712 + }, + { + "epoch": 1.4182771949199338, + "grad_norm": 0.8873599767684937, + "learning_rate": 4.3966873310489505e-06, + "loss": 0.6503, + "step": 1713 + }, + { + "epoch": 1.4191054665930425, + "grad_norm": 0.8841976523399353, + "learning_rate": 4.395970733615531e-06, + "loss": 0.6303, + "step": 1714 + }, + { + "epoch": 1.4199337382661512, + "grad_norm": 0.8922518491744995, + "learning_rate": 4.395253769329449e-06, + "loss": 0.6105, + "step": 1715 + }, + { + "epoch": 1.42076200993926, + "grad_norm": 0.9023404717445374, + "learning_rate": 4.39453643832943e-06, + "loss": 0.6292, + "step": 1716 + }, + { + "epoch": 1.421590281612369, + "grad_norm": 0.9042630791664124, + "learning_rate": 4.39381874075427e-06, + "loss": 0.6675, + "step": 1717 + }, + { + "epoch": 1.4224185532854776, + "grad_norm": 0.8933060169219971, + "learning_rate": 4.393100676742838e-06, + "loss": 0.6125, + "step": 1718 + }, + { + "epoch": 1.4232468249585863, + "grad_norm": 0.8821312189102173, + "learning_rate": 4.392382246434073e-06, + "loss": 0.6266, + "step": 1719 + }, + { + "epoch": 1.4240750966316953, + "grad_norm": 0.904460608959198, + "learning_rate": 4.391663449966983e-06, + "loss": 0.5984, + "step": 1720 + }, + { + "epoch": 1.424903368304804, + "grad_norm": 0.8999350666999817, + "learning_rate": 4.39094428748065e-06, + "loss": 0.6653, + "step": 1721 + }, + { + "epoch": 1.4257316399779127, + "grad_norm": 0.9061364531517029, + "learning_rate": 4.390224759114224e-06, + "loss": 0.6306, + "step": 1722 + }, + { + "epoch": 1.4265599116510215, + "grad_norm": 0.8995294570922852, + "learning_rate": 4.389504865006927e-06, + "loss": 0.6656, + "step": 1723 + }, + { + "epoch": 1.4273881833241302, + "grad_norm": 0.904737114906311, + "learning_rate": 4.388784605298053e-06, + "loss": 0.6234, + "step": 1724 + }, + { + "epoch": 1.4282164549972391, + "grad_norm": 0.8827239871025085, + "learning_rate": 4.388063980126966e-06, + "loss": 0.6418, + "step": 1725 + }, + { + "epoch": 1.4290447266703479, + "grad_norm": 0.8791747689247131, + "learning_rate": 4.387342989633097e-06, + "loss": 0.6174, + "step": 1726 + }, + { + "epoch": 1.4298729983434566, + "grad_norm": 0.8831204175949097, + "learning_rate": 4.386621633955955e-06, + "loss": 0.6222, + "step": 1727 + }, + { + "epoch": 1.4307012700165656, + "grad_norm": 0.908511221408844, + "learning_rate": 4.3858999132351124e-06, + "loss": 0.6373, + "step": 1728 + }, + { + "epoch": 1.4315295416896743, + "grad_norm": 0.900497555732727, + "learning_rate": 4.385177827610218e-06, + "loss": 0.6291, + "step": 1729 + }, + { + "epoch": 1.432357813362783, + "grad_norm": 0.930695652961731, + "learning_rate": 4.384455377220986e-06, + "loss": 0.6408, + "step": 1730 + }, + { + "epoch": 1.4331860850358917, + "grad_norm": 0.9583060145378113, + "learning_rate": 4.383732562207206e-06, + "loss": 0.6452, + "step": 1731 + }, + { + "epoch": 1.4340143567090005, + "grad_norm": 0.9173291921615601, + "learning_rate": 4.383009382708734e-06, + "loss": 0.6491, + "step": 1732 + }, + { + "epoch": 1.4348426283821094, + "grad_norm": 0.9054985642433167, + "learning_rate": 4.382285838865501e-06, + "loss": 0.6179, + "step": 1733 + }, + { + "epoch": 1.4356709000552181, + "grad_norm": 0.9162827730178833, + "learning_rate": 4.381561930817504e-06, + "loss": 0.6428, + "step": 1734 + }, + { + "epoch": 1.4364991717283269, + "grad_norm": 0.9013632535934448, + "learning_rate": 4.380837658704813e-06, + "loss": 0.6385, + "step": 1735 + }, + { + "epoch": 1.4373274434014356, + "grad_norm": 0.9274044632911682, + "learning_rate": 4.3801130226675685e-06, + "loss": 0.6508, + "step": 1736 + }, + { + "epoch": 1.4381557150745445, + "grad_norm": 0.8884710669517517, + "learning_rate": 4.37938802284598e-06, + "loss": 0.6675, + "step": 1737 + }, + { + "epoch": 1.4389839867476533, + "grad_norm": 0.9156724214553833, + "learning_rate": 4.378662659380329e-06, + "loss": 0.6424, + "step": 1738 + }, + { + "epoch": 1.439812258420762, + "grad_norm": 0.9267123341560364, + "learning_rate": 4.377936932410966e-06, + "loss": 0.6333, + "step": 1739 + }, + { + "epoch": 1.4406405300938707, + "grad_norm": 0.8859686851501465, + "learning_rate": 4.377210842078311e-06, + "loss": 0.6343, + "step": 1740 + }, + { + "epoch": 1.4414688017669794, + "grad_norm": 0.92928147315979, + "learning_rate": 4.376484388522858e-06, + "loss": 0.6124, + "step": 1741 + }, + { + "epoch": 1.4422970734400884, + "grad_norm": 0.8908365964889526, + "learning_rate": 4.375757571885169e-06, + "loss": 0.631, + "step": 1742 + }, + { + "epoch": 1.4431253451131971, + "grad_norm": 0.8956666588783264, + "learning_rate": 4.375030392305875e-06, + "loss": 0.6314, + "step": 1743 + }, + { + "epoch": 1.4439536167863058, + "grad_norm": 0.8836198449134827, + "learning_rate": 4.374302849925679e-06, + "loss": 0.6179, + "step": 1744 + }, + { + "epoch": 1.4447818884594148, + "grad_norm": 0.87999027967453, + "learning_rate": 4.373574944885354e-06, + "loss": 0.6226, + "step": 1745 + }, + { + "epoch": 1.4456101601325235, + "grad_norm": 0.9118010997772217, + "learning_rate": 4.3728466773257414e-06, + "loss": 0.6319, + "step": 1746 + }, + { + "epoch": 1.4464384318056323, + "grad_norm": 0.8989576101303101, + "learning_rate": 4.3721180473877555e-06, + "loss": 0.6049, + "step": 1747 + }, + { + "epoch": 1.447266703478741, + "grad_norm": 0.8896577954292297, + "learning_rate": 4.371389055212379e-06, + "loss": 0.6242, + "step": 1748 + }, + { + "epoch": 1.4480949751518497, + "grad_norm": 0.8993588089942932, + "learning_rate": 4.370659700940665e-06, + "loss": 0.6255, + "step": 1749 + }, + { + "epoch": 1.4489232468249587, + "grad_norm": 0.896486759185791, + "learning_rate": 4.369929984713738e-06, + "loss": 0.6354, + "step": 1750 + }, + { + "epoch": 1.4497515184980674, + "grad_norm": 0.9003145694732666, + "learning_rate": 4.369199906672789e-06, + "loss": 0.6464, + "step": 1751 + }, + { + "epoch": 1.450579790171176, + "grad_norm": 0.8885106444358826, + "learning_rate": 4.368469466959083e-06, + "loss": 0.6567, + "step": 1752 + }, + { + "epoch": 1.451408061844285, + "grad_norm": 0.8914943933486938, + "learning_rate": 4.367738665713953e-06, + "loss": 0.6784, + "step": 1753 + }, + { + "epoch": 1.4522363335173938, + "grad_norm": 0.933029055595398, + "learning_rate": 4.367007503078802e-06, + "loss": 0.6495, + "step": 1754 + }, + { + "epoch": 1.4530646051905025, + "grad_norm": 0.8788006901741028, + "learning_rate": 4.366275979195102e-06, + "loss": 0.6352, + "step": 1755 + }, + { + "epoch": 1.4538928768636112, + "grad_norm": 0.8768455982208252, + "learning_rate": 4.365544094204399e-06, + "loss": 0.6329, + "step": 1756 + }, + { + "epoch": 1.45472114853672, + "grad_norm": 0.9021796584129333, + "learning_rate": 4.364811848248303e-06, + "loss": 0.6616, + "step": 1757 + }, + { + "epoch": 1.4555494202098287, + "grad_norm": 0.9197986125946045, + "learning_rate": 4.364079241468499e-06, + "loss": 0.6248, + "step": 1758 + }, + { + "epoch": 1.4563776918829376, + "grad_norm": 0.9095743894577026, + "learning_rate": 4.3633462740067375e-06, + "loss": 0.6722, + "step": 1759 + }, + { + "epoch": 1.4572059635560464, + "grad_norm": 0.9114678502082825, + "learning_rate": 4.362612946004843e-06, + "loss": 0.6278, + "step": 1760 + }, + { + "epoch": 1.458034235229155, + "grad_norm": 0.8978061676025391, + "learning_rate": 4.3618792576047054e-06, + "loss": 0.633, + "step": 1761 + }, + { + "epoch": 1.458862506902264, + "grad_norm": 0.8578571677207947, + "learning_rate": 4.361145208948287e-06, + "loss": 0.6263, + "step": 1762 + }, + { + "epoch": 1.4596907785753728, + "grad_norm": 0.9333001375198364, + "learning_rate": 4.360410800177622e-06, + "loss": 0.6383, + "step": 1763 + }, + { + "epoch": 1.4605190502484815, + "grad_norm": 0.9448907971382141, + "learning_rate": 4.359676031434808e-06, + "loss": 0.6324, + "step": 1764 + }, + { + "epoch": 1.4613473219215902, + "grad_norm": 0.8806003928184509, + "learning_rate": 4.358940902862019e-06, + "loss": 0.6365, + "step": 1765 + }, + { + "epoch": 1.462175593594699, + "grad_norm": 0.9975590109825134, + "learning_rate": 4.358205414601493e-06, + "loss": 0.6436, + "step": 1766 + }, + { + "epoch": 1.463003865267808, + "grad_norm": 1.0076019763946533, + "learning_rate": 4.357469566795542e-06, + "loss": 0.6234, + "step": 1767 + }, + { + "epoch": 1.4638321369409166, + "grad_norm": 0.9254529476165771, + "learning_rate": 4.356733359586545e-06, + "loss": 0.6244, + "step": 1768 + }, + { + "epoch": 1.4646604086140254, + "grad_norm": 0.9129828810691833, + "learning_rate": 4.35599679311695e-06, + "loss": 0.6519, + "step": 1769 + }, + { + "epoch": 1.4654886802871343, + "grad_norm": 0.9141498804092407, + "learning_rate": 4.355259867529278e-06, + "loss": 0.6343, + "step": 1770 + }, + { + "epoch": 1.466316951960243, + "grad_norm": 0.920790433883667, + "learning_rate": 4.354522582966116e-06, + "loss": 0.6505, + "step": 1771 + }, + { + "epoch": 1.4671452236333518, + "grad_norm": 0.9683033227920532, + "learning_rate": 4.353784939570121e-06, + "loss": 0.6238, + "step": 1772 + }, + { + "epoch": 1.4679734953064605, + "grad_norm": 0.882957935333252, + "learning_rate": 4.3530469374840226e-06, + "loss": 0.6358, + "step": 1773 + }, + { + "epoch": 1.4688017669795692, + "grad_norm": 0.943648099899292, + "learning_rate": 4.352308576850615e-06, + "loss": 0.6274, + "step": 1774 + }, + { + "epoch": 1.4696300386526782, + "grad_norm": 0.9609524607658386, + "learning_rate": 4.351569857812764e-06, + "loss": 0.6368, + "step": 1775 + }, + { + "epoch": 1.4704583103257869, + "grad_norm": 0.9170458912849426, + "learning_rate": 4.3508307805134055e-06, + "loss": 0.6515, + "step": 1776 + }, + { + "epoch": 1.4712865819988956, + "grad_norm": 0.8956430554389954, + "learning_rate": 4.350091345095545e-06, + "loss": 0.6394, + "step": 1777 + }, + { + "epoch": 1.4721148536720043, + "grad_norm": 0.9145321846008301, + "learning_rate": 4.349351551702256e-06, + "loss": 0.6572, + "step": 1778 + }, + { + "epoch": 1.4729431253451133, + "grad_norm": 0.8807089328765869, + "learning_rate": 4.34861140047668e-06, + "loss": 0.6391, + "step": 1779 + }, + { + "epoch": 1.473771397018222, + "grad_norm": 0.8828510642051697, + "learning_rate": 4.347870891562032e-06, + "loss": 0.6583, + "step": 1780 + }, + { + "epoch": 1.4745996686913307, + "grad_norm": 0.9017962217330933, + "learning_rate": 4.34713002510159e-06, + "loss": 0.6337, + "step": 1781 + }, + { + "epoch": 1.4754279403644395, + "grad_norm": 0.8738356232643127, + "learning_rate": 4.346388801238708e-06, + "loss": 0.637, + "step": 1782 + }, + { + "epoch": 1.4762562120375482, + "grad_norm": 0.8805137872695923, + "learning_rate": 4.345647220116805e-06, + "loss": 0.6306, + "step": 1783 + }, + { + "epoch": 1.4770844837106571, + "grad_norm": 0.923599362373352, + "learning_rate": 4.34490528187937e-06, + "loss": 0.6254, + "step": 1784 + }, + { + "epoch": 1.4779127553837659, + "grad_norm": 0.9181063175201416, + "learning_rate": 4.344162986669961e-06, + "loss": 0.6257, + "step": 1785 + }, + { + "epoch": 1.4787410270568746, + "grad_norm": 0.9060764312744141, + "learning_rate": 4.343420334632204e-06, + "loss": 0.6326, + "step": 1786 + }, + { + "epoch": 1.4795692987299836, + "grad_norm": 0.9101755023002625, + "learning_rate": 4.342677325909798e-06, + "loss": 0.6319, + "step": 1787 + }, + { + "epoch": 1.4803975704030923, + "grad_norm": 0.9417787790298462, + "learning_rate": 4.341933960646505e-06, + "loss": 0.6147, + "step": 1788 + }, + { + "epoch": 1.481225842076201, + "grad_norm": 0.9138630628585815, + "learning_rate": 4.341190238986162e-06, + "loss": 0.6256, + "step": 1789 + }, + { + "epoch": 1.4820541137493097, + "grad_norm": 0.9004842638969421, + "learning_rate": 4.340446161072671e-06, + "loss": 0.6443, + "step": 1790 + }, + { + "epoch": 1.4828823854224185, + "grad_norm": 0.9181780219078064, + "learning_rate": 4.339701727050004e-06, + "loss": 0.6408, + "step": 1791 + }, + { + "epoch": 1.4837106570955274, + "grad_norm": 0.880736768245697, + "learning_rate": 4.338956937062203e-06, + "loss": 0.6436, + "step": 1792 + }, + { + "epoch": 1.4845389287686361, + "grad_norm": 0.9168913960456848, + "learning_rate": 4.338211791253376e-06, + "loss": 0.6391, + "step": 1793 + }, + { + "epoch": 1.4853672004417449, + "grad_norm": 0.942925751209259, + "learning_rate": 4.337466289767704e-06, + "loss": 0.6303, + "step": 1794 + }, + { + "epoch": 1.4861954721148538, + "grad_norm": 0.8768817186355591, + "learning_rate": 4.336720432749433e-06, + "loss": 0.6239, + "step": 1795 + }, + { + "epoch": 1.4870237437879625, + "grad_norm": 0.9388988614082336, + "learning_rate": 4.33597422034288e-06, + "loss": 0.6441, + "step": 1796 + }, + { + "epoch": 1.4878520154610713, + "grad_norm": 0.9257978200912476, + "learning_rate": 4.3352276526924295e-06, + "loss": 0.6373, + "step": 1797 + }, + { + "epoch": 1.48868028713418, + "grad_norm": 0.9225389957427979, + "learning_rate": 4.334480729942537e-06, + "loss": 0.6555, + "step": 1798 + }, + { + "epoch": 1.4895085588072887, + "grad_norm": 0.8992138504981995, + "learning_rate": 4.3337334522377235e-06, + "loss": 0.6375, + "step": 1799 + }, + { + "epoch": 1.4903368304803974, + "grad_norm": 0.9207738041877747, + "learning_rate": 4.332985819722582e-06, + "loss": 0.6439, + "step": 1800 + }, + { + "epoch": 1.4911651021535064, + "grad_norm": 0.8866941332817078, + "learning_rate": 4.3322378325417705e-06, + "loss": 0.6326, + "step": 1801 + }, + { + "epoch": 1.4919933738266151, + "grad_norm": 0.9069377779960632, + "learning_rate": 4.3314894908400175e-06, + "loss": 0.6193, + "step": 1802 + }, + { + "epoch": 1.4928216454997238, + "grad_norm": 0.8918569087982178, + "learning_rate": 4.3307407947621216e-06, + "loss": 0.6374, + "step": 1803 + }, + { + "epoch": 1.4936499171728328, + "grad_norm": 0.8993483781814575, + "learning_rate": 4.329991744452948e-06, + "loss": 0.6251, + "step": 1804 + }, + { + "epoch": 1.4944781888459415, + "grad_norm": 0.918445885181427, + "learning_rate": 4.32924234005743e-06, + "loss": 0.6494, + "step": 1805 + }, + { + "epoch": 1.4953064605190503, + "grad_norm": 0.9245757460594177, + "learning_rate": 4.328492581720573e-06, + "loss": 0.6172, + "step": 1806 + }, + { + "epoch": 1.496134732192159, + "grad_norm": 0.9166370034217834, + "learning_rate": 4.327742469587445e-06, + "loss": 0.622, + "step": 1807 + }, + { + "epoch": 1.4969630038652677, + "grad_norm": 0.9083957076072693, + "learning_rate": 4.326992003803187e-06, + "loss": 0.6205, + "step": 1808 + }, + { + "epoch": 1.4977912755383767, + "grad_norm": 0.8895595073699951, + "learning_rate": 4.326241184513008e-06, + "loss": 0.6245, + "step": 1809 + }, + { + "epoch": 1.4986195472114854, + "grad_norm": 0.9845629334449768, + "learning_rate": 4.325490011862184e-06, + "loss": 0.6163, + "step": 1810 + }, + { + "epoch": 1.499447818884594, + "grad_norm": 0.9501281976699829, + "learning_rate": 4.324738485996058e-06, + "loss": 0.6218, + "step": 1811 + }, + { + "epoch": 1.500276090557703, + "grad_norm": 0.9685792922973633, + "learning_rate": 4.323986607060047e-06, + "loss": 0.634, + "step": 1812 + }, + { + "epoch": 1.5011043622308118, + "grad_norm": 0.8959722518920898, + "learning_rate": 4.32323437519963e-06, + "loss": 0.6316, + "step": 1813 + }, + { + "epoch": 1.5019326339039205, + "grad_norm": 0.914463222026825, + "learning_rate": 4.322481790560356e-06, + "loss": 0.6258, + "step": 1814 + }, + { + "epoch": 1.5027609055770292, + "grad_norm": 0.9052320718765259, + "learning_rate": 4.321728853287846e-06, + "loss": 0.6345, + "step": 1815 + }, + { + "epoch": 1.503589177250138, + "grad_norm": 0.9306650161743164, + "learning_rate": 4.320975563527785e-06, + "loss": 0.5976, + "step": 1816 + }, + { + "epoch": 1.5044174489232467, + "grad_norm": 0.8930769562721252, + "learning_rate": 4.320221921425926e-06, + "loss": 0.6427, + "step": 1817 + }, + { + "epoch": 1.5052457205963556, + "grad_norm": 0.9205043911933899, + "learning_rate": 4.3194679271280935e-06, + "loss": 0.617, + "step": 1818 + }, + { + "epoch": 1.5060739922694644, + "grad_norm": 1.0371367931365967, + "learning_rate": 4.318713580780179e-06, + "loss": 0.6301, + "step": 1819 + }, + { + "epoch": 1.5069022639425733, + "grad_norm": 0.9442588686943054, + "learning_rate": 4.317958882528139e-06, + "loss": 0.6252, + "step": 1820 + }, + { + "epoch": 1.507730535615682, + "grad_norm": 0.9214493036270142, + "learning_rate": 4.317203832518003e-06, + "loss": 0.6326, + "step": 1821 + }, + { + "epoch": 1.5085588072887908, + "grad_norm": 0.9270440340042114, + "learning_rate": 4.316448430895866e-06, + "loss": 0.6377, + "step": 1822 + }, + { + "epoch": 1.5093870789618995, + "grad_norm": 0.8822277188301086, + "learning_rate": 4.315692677807889e-06, + "loss": 0.6277, + "step": 1823 + }, + { + "epoch": 1.5102153506350082, + "grad_norm": 0.8926153779029846, + "learning_rate": 4.314936573400305e-06, + "loss": 0.6439, + "step": 1824 + }, + { + "epoch": 1.511043622308117, + "grad_norm": 0.9214072227478027, + "learning_rate": 4.314180117819412e-06, + "loss": 0.6519, + "step": 1825 + }, + { + "epoch": 1.5118718939812257, + "grad_norm": 0.881522536277771, + "learning_rate": 4.313423311211578e-06, + "loss": 0.6213, + "step": 1826 + }, + { + "epoch": 1.5127001656543346, + "grad_norm": 0.8960263729095459, + "learning_rate": 4.312666153723238e-06, + "loss": 0.6592, + "step": 1827 + }, + { + "epoch": 1.5135284373274434, + "grad_norm": 0.9176173210144043, + "learning_rate": 4.3119086455008956e-06, + "loss": 0.6569, + "step": 1828 + }, + { + "epoch": 1.5143567090005523, + "grad_norm": 0.9312153458595276, + "learning_rate": 4.311150786691119e-06, + "loss": 0.6324, + "step": 1829 + }, + { + "epoch": 1.515184980673661, + "grad_norm": 0.9202299118041992, + "learning_rate": 4.310392577440549e-06, + "loss": 0.6381, + "step": 1830 + }, + { + "epoch": 1.5160132523467698, + "grad_norm": 0.8895224928855896, + "learning_rate": 4.309634017895893e-06, + "loss": 0.6491, + "step": 1831 + }, + { + "epoch": 1.5168415240198785, + "grad_norm": 0.9424769878387451, + "learning_rate": 4.308875108203922e-06, + "loss": 0.6342, + "step": 1832 + }, + { + "epoch": 1.5176697956929872, + "grad_norm": 0.8676947951316833, + "learning_rate": 4.30811584851148e-06, + "loss": 0.6287, + "step": 1833 + }, + { + "epoch": 1.518498067366096, + "grad_norm": 0.8764673471450806, + "learning_rate": 4.3073562389654764e-06, + "loss": 0.6123, + "step": 1834 + }, + { + "epoch": 1.5193263390392049, + "grad_norm": 0.9222663640975952, + "learning_rate": 4.306596279712889e-06, + "loss": 0.6433, + "step": 1835 + }, + { + "epoch": 1.5201546107123136, + "grad_norm": 0.8839907050132751, + "learning_rate": 4.305835970900762e-06, + "loss": 0.6416, + "step": 1836 + }, + { + "epoch": 1.5209828823854226, + "grad_norm": 0.9095149636268616, + "learning_rate": 4.305075312676209e-06, + "loss": 0.6417, + "step": 1837 + }, + { + "epoch": 1.5218111540585313, + "grad_norm": 0.8718581199645996, + "learning_rate": 4.304314305186409e-06, + "loss": 0.6174, + "step": 1838 + }, + { + "epoch": 1.52263942573164, + "grad_norm": 0.9424081444740295, + "learning_rate": 4.303552948578611e-06, + "loss": 0.6064, + "step": 1839 + }, + { + "epoch": 1.5234676974047487, + "grad_norm": 0.9170297384262085, + "learning_rate": 4.30279124300013e-06, + "loss": 0.6078, + "step": 1840 + }, + { + "epoch": 1.5242959690778575, + "grad_norm": 0.8814254999160767, + "learning_rate": 4.302029188598348e-06, + "loss": 0.6194, + "step": 1841 + }, + { + "epoch": 1.5251242407509662, + "grad_norm": 0.9248694181442261, + "learning_rate": 4.301266785520718e-06, + "loss": 0.6452, + "step": 1842 + }, + { + "epoch": 1.5259525124240751, + "grad_norm": 0.8757627010345459, + "learning_rate": 4.300504033914754e-06, + "loss": 0.6486, + "step": 1843 + }, + { + "epoch": 1.5267807840971839, + "grad_norm": 0.9229605793952942, + "learning_rate": 4.299740933928045e-06, + "loss": 0.6363, + "step": 1844 + }, + { + "epoch": 1.5276090557702926, + "grad_norm": 0.9192973971366882, + "learning_rate": 4.298977485708243e-06, + "loss": 0.6383, + "step": 1845 + }, + { + "epoch": 1.5284373274434015, + "grad_norm": 0.9071361422538757, + "learning_rate": 4.2982136894030655e-06, + "loss": 0.6541, + "step": 1846 + }, + { + "epoch": 1.5292655991165103, + "grad_norm": 0.9067169427871704, + "learning_rate": 4.297449545160303e-06, + "loss": 0.6317, + "step": 1847 + }, + { + "epoch": 1.530093870789619, + "grad_norm": 0.9047722816467285, + "learning_rate": 4.2966850531278086e-06, + "loss": 0.6365, + "step": 1848 + }, + { + "epoch": 1.5309221424627277, + "grad_norm": 0.8948541879653931, + "learning_rate": 4.295920213453504e-06, + "loss": 0.6268, + "step": 1849 + }, + { + "epoch": 1.5317504141358365, + "grad_norm": 0.9061904549598694, + "learning_rate": 4.295155026285382e-06, + "loss": 0.6271, + "step": 1850 + }, + { + "epoch": 1.5325786858089452, + "grad_norm": 0.9092041850090027, + "learning_rate": 4.2943894917714946e-06, + "loss": 0.6148, + "step": 1851 + }, + { + "epoch": 1.5334069574820541, + "grad_norm": 0.932331919670105, + "learning_rate": 4.293623610059968e-06, + "loss": 0.647, + "step": 1852 + }, + { + "epoch": 1.5342352291551629, + "grad_norm": 0.9136486649513245, + "learning_rate": 4.292857381298993e-06, + "loss": 0.6326, + "step": 1853 + }, + { + "epoch": 1.5350635008282718, + "grad_norm": 0.9760140180587769, + "learning_rate": 4.292090805636827e-06, + "loss": 0.6324, + "step": 1854 + }, + { + "epoch": 1.5358917725013805, + "grad_norm": 0.9075965881347656, + "learning_rate": 4.291323883221797e-06, + "loss": 0.6194, + "step": 1855 + }, + { + "epoch": 1.5367200441744893, + "grad_norm": 0.8804707527160645, + "learning_rate": 4.290556614202292e-06, + "loss": 0.6246, + "step": 1856 + }, + { + "epoch": 1.537548315847598, + "grad_norm": 0.9123021960258484, + "learning_rate": 4.2897889987267755e-06, + "loss": 0.6156, + "step": 1857 + }, + { + "epoch": 1.5383765875207067, + "grad_norm": 0.9396944642066956, + "learning_rate": 4.289021036943771e-06, + "loss": 0.6268, + "step": 1858 + }, + { + "epoch": 1.5392048591938154, + "grad_norm": 0.858566164970398, + "learning_rate": 4.288252729001874e-06, + "loss": 0.6237, + "step": 1859 + }, + { + "epoch": 1.5400331308669244, + "grad_norm": 0.899341344833374, + "learning_rate": 4.287484075049743e-06, + "loss": 0.5973, + "step": 1860 + }, + { + "epoch": 1.5408614025400331, + "grad_norm": 0.9021979570388794, + "learning_rate": 4.286715075236106e-06, + "loss": 0.6164, + "step": 1861 + }, + { + "epoch": 1.541689674213142, + "grad_norm": 0.9693799018859863, + "learning_rate": 4.285945729709757e-06, + "loss": 0.6409, + "step": 1862 + }, + { + "epoch": 1.5425179458862508, + "grad_norm": 0.9177312254905701, + "learning_rate": 4.285176038619559e-06, + "loss": 0.6276, + "step": 1863 + }, + { + "epoch": 1.5433462175593595, + "grad_norm": 0.9382041692733765, + "learning_rate": 4.284406002114438e-06, + "loss": 0.643, + "step": 1864 + }, + { + "epoch": 1.5441744892324683, + "grad_norm": 0.8778985142707825, + "learning_rate": 4.28363562034339e-06, + "loss": 0.6119, + "step": 1865 + }, + { + "epoch": 1.545002760905577, + "grad_norm": 0.8890271782875061, + "learning_rate": 4.282864893455476e-06, + "loss": 0.6388, + "step": 1866 + }, + { + "epoch": 1.5458310325786857, + "grad_norm": 0.9081838130950928, + "learning_rate": 4.282093821599824e-06, + "loss": 0.6464, + "step": 1867 + }, + { + "epoch": 1.5466593042517944, + "grad_norm": 0.8805630803108215, + "learning_rate": 4.281322404925631e-06, + "loss": 0.6533, + "step": 1868 + }, + { + "epoch": 1.5474875759249034, + "grad_norm": 0.8817034363746643, + "learning_rate": 4.280550643582158e-06, + "loss": 0.643, + "step": 1869 + }, + { + "epoch": 1.548315847598012, + "grad_norm": 0.9003176689147949, + "learning_rate": 4.279778537718733e-06, + "loss": 0.6322, + "step": 1870 + }, + { + "epoch": 1.549144119271121, + "grad_norm": 0.8982899785041809, + "learning_rate": 4.2790060874847525e-06, + "loss": 0.6456, + "step": 1871 + }, + { + "epoch": 1.5499723909442298, + "grad_norm": 0.8730716109275818, + "learning_rate": 4.278233293029677e-06, + "loss": 0.6428, + "step": 1872 + }, + { + "epoch": 1.5508006626173385, + "grad_norm": 0.9135062098503113, + "learning_rate": 4.277460154503036e-06, + "loss": 0.645, + "step": 1873 + }, + { + "epoch": 1.5516289342904472, + "grad_norm": 0.9091652631759644, + "learning_rate": 4.276686672054424e-06, + "loss": 0.6256, + "step": 1874 + }, + { + "epoch": 1.552457205963556, + "grad_norm": 0.9090970754623413, + "learning_rate": 4.275912845833504e-06, + "loss": 0.6404, + "step": 1875 + }, + { + "epoch": 1.5532854776366647, + "grad_norm": 0.91478031873703, + "learning_rate": 4.275138675990003e-06, + "loss": 0.6498, + "step": 1876 + }, + { + "epoch": 1.5541137493097736, + "grad_norm": 0.8979254961013794, + "learning_rate": 4.274364162673715e-06, + "loss": 0.6275, + "step": 1877 + }, + { + "epoch": 1.5549420209828824, + "grad_norm": 0.92198646068573, + "learning_rate": 4.273589306034503e-06, + "loss": 0.6585, + "step": 1878 + }, + { + "epoch": 1.5557702926559913, + "grad_norm": 0.9540958404541016, + "learning_rate": 4.272814106222293e-06, + "loss": 0.6111, + "step": 1879 + }, + { + "epoch": 1.5565985643291, + "grad_norm": 0.8989667892456055, + "learning_rate": 4.272038563387079e-06, + "loss": 0.6255, + "step": 1880 + }, + { + "epoch": 1.5574268360022088, + "grad_norm": 0.9551357626914978, + "learning_rate": 4.2712626776789224e-06, + "loss": 0.6164, + "step": 1881 + }, + { + "epoch": 1.5582551076753175, + "grad_norm": 0.9278994798660278, + "learning_rate": 4.270486449247949e-06, + "loss": 0.6376, + "step": 1882 + }, + { + "epoch": 1.5590833793484262, + "grad_norm": 0.927180826663971, + "learning_rate": 4.269709878244352e-06, + "loss": 0.6602, + "step": 1883 + }, + { + "epoch": 1.559911651021535, + "grad_norm": 0.932837724685669, + "learning_rate": 4.268932964818389e-06, + "loss": 0.6229, + "step": 1884 + }, + { + "epoch": 1.5607399226946437, + "grad_norm": 0.895219087600708, + "learning_rate": 4.26815570912039e-06, + "loss": 0.6192, + "step": 1885 + }, + { + "epoch": 1.5615681943677526, + "grad_norm": 0.9075414538383484, + "learning_rate": 4.2673781113007415e-06, + "loss": 0.6418, + "step": 1886 + }, + { + "epoch": 1.5623964660408614, + "grad_norm": 0.9130772948265076, + "learning_rate": 4.266600171509905e-06, + "loss": 0.6402, + "step": 1887 + }, + { + "epoch": 1.5632247377139703, + "grad_norm": 0.8892185688018799, + "learning_rate": 4.265821889898402e-06, + "loss": 0.6213, + "step": 1888 + }, + { + "epoch": 1.564053009387079, + "grad_norm": 0.8853991031646729, + "learning_rate": 4.265043266616825e-06, + "loss": 0.6357, + "step": 1889 + }, + { + "epoch": 1.5648812810601878, + "grad_norm": 0.9273191690444946, + "learning_rate": 4.26426430181583e-06, + "loss": 0.6311, + "step": 1890 + }, + { + "epoch": 1.5657095527332965, + "grad_norm": 0.8890771865844727, + "learning_rate": 4.263484995646138e-06, + "loss": 0.6053, + "step": 1891 + }, + { + "epoch": 1.5665378244064052, + "grad_norm": 0.9022220373153687, + "learning_rate": 4.262705348258538e-06, + "loss": 0.6256, + "step": 1892 + }, + { + "epoch": 1.567366096079514, + "grad_norm": 0.9201255440711975, + "learning_rate": 4.261925359803886e-06, + "loss": 0.6302, + "step": 1893 + }, + { + "epoch": 1.5681943677526229, + "grad_norm": 0.9675379991531372, + "learning_rate": 4.261145030433101e-06, + "loss": 0.602, + "step": 1894 + }, + { + "epoch": 1.5690226394257316, + "grad_norm": 0.9039328098297119, + "learning_rate": 4.26036436029717e-06, + "loss": 0.6477, + "step": 1895 + }, + { + "epoch": 1.5698509110988406, + "grad_norm": 0.911361813545227, + "learning_rate": 4.259583349547144e-06, + "loss": 0.6211, + "step": 1896 + }, + { + "epoch": 1.5706791827719493, + "grad_norm": 0.8901238441467285, + "learning_rate": 4.258801998334143e-06, + "loss": 0.6442, + "step": 1897 + }, + { + "epoch": 1.571507454445058, + "grad_norm": 0.9290407299995422, + "learning_rate": 4.258020306809351e-06, + "loss": 0.6382, + "step": 1898 + }, + { + "epoch": 1.5723357261181667, + "grad_norm": 0.9179307222366333, + "learning_rate": 4.257238275124017e-06, + "loss": 0.6513, + "step": 1899 + }, + { + "epoch": 1.5731639977912755, + "grad_norm": 0.9212455153465271, + "learning_rate": 4.256455903429459e-06, + "loss": 0.6257, + "step": 1900 + }, + { + "epoch": 1.5739922694643842, + "grad_norm": 0.9126132130622864, + "learning_rate": 4.2556731918770565e-06, + "loss": 0.6384, + "step": 1901 + }, + { + "epoch": 1.5748205411374931, + "grad_norm": 0.8735355734825134, + "learning_rate": 4.254890140618257e-06, + "loss": 0.6031, + "step": 1902 + }, + { + "epoch": 1.5756488128106019, + "grad_norm": 0.9375724196434021, + "learning_rate": 4.254106749804575e-06, + "loss": 0.6333, + "step": 1903 + }, + { + "epoch": 1.5764770844837108, + "grad_norm": 0.9451910853385925, + "learning_rate": 4.253323019587588e-06, + "loss": 0.6186, + "step": 1904 + }, + { + "epoch": 1.5773053561568195, + "grad_norm": 0.9468234777450562, + "learning_rate": 4.25253895011894e-06, + "loss": 0.6023, + "step": 1905 + }, + { + "epoch": 1.5781336278299283, + "grad_norm": 0.8941681385040283, + "learning_rate": 4.251754541550343e-06, + "loss": 0.634, + "step": 1906 + }, + { + "epoch": 1.578961899503037, + "grad_norm": 0.9508136510848999, + "learning_rate": 4.250969794033573e-06, + "loss": 0.6346, + "step": 1907 + }, + { + "epoch": 1.5797901711761457, + "grad_norm": 0.8964453935623169, + "learning_rate": 4.2501847077204685e-06, + "loss": 0.6298, + "step": 1908 + }, + { + "epoch": 1.5806184428492545, + "grad_norm": 0.8885358572006226, + "learning_rate": 4.249399282762938e-06, + "loss": 0.6319, + "step": 1909 + }, + { + "epoch": 1.5814467145223632, + "grad_norm": 0.8929858207702637, + "learning_rate": 4.248613519312955e-06, + "loss": 0.6272, + "step": 1910 + }, + { + "epoch": 1.5822749861954721, + "grad_norm": 0.9664214849472046, + "learning_rate": 4.247827417522557e-06, + "loss": 0.6336, + "step": 1911 + }, + { + "epoch": 1.5831032578685809, + "grad_norm": 0.890846312046051, + "learning_rate": 4.247040977543845e-06, + "loss": 0.6242, + "step": 1912 + }, + { + "epoch": 1.5839315295416898, + "grad_norm": 0.8973182439804077, + "learning_rate": 4.246254199528991e-06, + "loss": 0.6336, + "step": 1913 + }, + { + "epoch": 1.5847598012147985, + "grad_norm": 0.9789074659347534, + "learning_rate": 4.245467083630227e-06, + "loss": 0.6191, + "step": 1914 + }, + { + "epoch": 1.5855880728879073, + "grad_norm": 0.9195513129234314, + "learning_rate": 4.244679629999853e-06, + "loss": 0.6333, + "step": 1915 + }, + { + "epoch": 1.586416344561016, + "grad_norm": 0.950316846370697, + "learning_rate": 4.243891838790235e-06, + "loss": 0.635, + "step": 1916 + }, + { + "epoch": 1.5872446162341247, + "grad_norm": 0.8971677422523499, + "learning_rate": 4.243103710153802e-06, + "loss": 0.6352, + "step": 1917 + }, + { + "epoch": 1.5880728879072334, + "grad_norm": 0.9198834896087646, + "learning_rate": 4.24231524424305e-06, + "loss": 0.6252, + "step": 1918 + }, + { + "epoch": 1.5889011595803424, + "grad_norm": 0.9535729885101318, + "learning_rate": 4.241526441210541e-06, + "loss": 0.6177, + "step": 1919 + }, + { + "epoch": 1.5897294312534511, + "grad_norm": 0.8910471200942993, + "learning_rate": 4.240737301208899e-06, + "loss": 0.6408, + "step": 1920 + }, + { + "epoch": 1.59055770292656, + "grad_norm": 0.862991452217102, + "learning_rate": 4.2399478243908164e-06, + "loss": 0.6221, + "step": 1921 + }, + { + "epoch": 1.5913859745996688, + "grad_norm": 0.8935141563415527, + "learning_rate": 4.239158010909049e-06, + "loss": 0.608, + "step": 1922 + }, + { + "epoch": 1.5922142462727775, + "grad_norm": 0.9177968502044678, + "learning_rate": 4.23836786091642e-06, + "loss": 0.6442, + "step": 1923 + }, + { + "epoch": 1.5930425179458862, + "grad_norm": 0.9157758951187134, + "learning_rate": 4.237577374565813e-06, + "loss": 0.6419, + "step": 1924 + }, + { + "epoch": 1.593870789618995, + "grad_norm": 0.9484630227088928, + "learning_rate": 4.236786552010182e-06, + "loss": 0.6733, + "step": 1925 + }, + { + "epoch": 1.5946990612921037, + "grad_norm": 0.9401251077651978, + "learning_rate": 4.235995393402544e-06, + "loss": 0.6325, + "step": 1926 + }, + { + "epoch": 1.5955273329652124, + "grad_norm": 0.9104607105255127, + "learning_rate": 4.23520389889598e-06, + "loss": 0.6313, + "step": 1927 + }, + { + "epoch": 1.5963556046383214, + "grad_norm": 0.9173511862754822, + "learning_rate": 4.234412068643637e-06, + "loss": 0.6044, + "step": 1928 + }, + { + "epoch": 1.59718387631143, + "grad_norm": 0.9189788699150085, + "learning_rate": 4.233619902798727e-06, + "loss": 0.6182, + "step": 1929 + }, + { + "epoch": 1.598012147984539, + "grad_norm": 0.910804808139801, + "learning_rate": 4.232827401514525e-06, + "loss": 0.6435, + "step": 1930 + }, + { + "epoch": 1.5988404196576478, + "grad_norm": 0.8986726403236389, + "learning_rate": 4.232034564944375e-06, + "loss": 0.6268, + "step": 1931 + }, + { + "epoch": 1.5996686913307565, + "grad_norm": 0.8782910108566284, + "learning_rate": 4.231241393241682e-06, + "loss": 0.6381, + "step": 1932 + }, + { + "epoch": 1.6004969630038652, + "grad_norm": 0.9176427721977234, + "learning_rate": 4.230447886559919e-06, + "loss": 0.6131, + "step": 1933 + }, + { + "epoch": 1.601325234676974, + "grad_norm": 0.8755510449409485, + "learning_rate": 4.22965404505262e-06, + "loss": 0.654, + "step": 1934 + }, + { + "epoch": 1.6021535063500827, + "grad_norm": 0.8644832968711853, + "learning_rate": 4.228859868873387e-06, + "loss": 0.627, + "step": 1935 + }, + { + "epoch": 1.6029817780231916, + "grad_norm": 0.9181885719299316, + "learning_rate": 4.228065358175886e-06, + "loss": 0.6062, + "step": 1936 + }, + { + "epoch": 1.6038100496963004, + "grad_norm": 0.9515942335128784, + "learning_rate": 4.227270513113846e-06, + "loss": 0.6262, + "step": 1937 + }, + { + "epoch": 1.6046383213694093, + "grad_norm": 0.9062270522117615, + "learning_rate": 4.226475333841063e-06, + "loss": 0.6139, + "step": 1938 + }, + { + "epoch": 1.605466593042518, + "grad_norm": 0.9040980339050293, + "learning_rate": 4.225679820511398e-06, + "loss": 0.64, + "step": 1939 + }, + { + "epoch": 1.6062948647156268, + "grad_norm": 0.8910935521125793, + "learning_rate": 4.224883973278775e-06, + "loss": 0.6239, + "step": 1940 + }, + { + "epoch": 1.6071231363887355, + "grad_norm": 0.9150440692901611, + "learning_rate": 4.224087792297181e-06, + "loss": 0.6283, + "step": 1941 + }, + { + "epoch": 1.6079514080618442, + "grad_norm": 0.9175113439559937, + "learning_rate": 4.223291277720671e-06, + "loss": 0.6498, + "step": 1942 + }, + { + "epoch": 1.608779679734953, + "grad_norm": 0.8720189332962036, + "learning_rate": 4.222494429703363e-06, + "loss": 0.623, + "step": 1943 + }, + { + "epoch": 1.609607951408062, + "grad_norm": 0.8868989944458008, + "learning_rate": 4.22169724839944e-06, + "loss": 0.6357, + "step": 1944 + }, + { + "epoch": 1.6104362230811706, + "grad_norm": 0.9084488153457642, + "learning_rate": 4.22089973396315e-06, + "loss": 0.6607, + "step": 1945 + }, + { + "epoch": 1.6112644947542794, + "grad_norm": 0.9061319231987, + "learning_rate": 4.220101886548802e-06, + "loss": 0.6309, + "step": 1946 + }, + { + "epoch": 1.6120927664273883, + "grad_norm": 0.895646870136261, + "learning_rate": 4.219303706310775e-06, + "loss": 0.6276, + "step": 1947 + }, + { + "epoch": 1.612921038100497, + "grad_norm": 0.8863298892974854, + "learning_rate": 4.218505193403506e-06, + "loss": 0.6444, + "step": 1948 + }, + { + "epoch": 1.6137493097736058, + "grad_norm": 0.9005767107009888, + "learning_rate": 4.2177063479815025e-06, + "loss": 0.6356, + "step": 1949 + }, + { + "epoch": 1.6145775814467145, + "grad_norm": 0.9094089865684509, + "learning_rate": 4.216907170199334e-06, + "loss": 0.6053, + "step": 1950 + }, + { + "epoch": 1.6154058531198232, + "grad_norm": 0.8869677782058716, + "learning_rate": 4.216107660211632e-06, + "loss": 0.6308, + "step": 1951 + }, + { + "epoch": 1.616234124792932, + "grad_norm": 0.9285000562667847, + "learning_rate": 4.215307818173095e-06, + "loss": 0.6188, + "step": 1952 + }, + { + "epoch": 1.6170623964660409, + "grad_norm": 0.9039749503135681, + "learning_rate": 4.214507644238485e-06, + "loss": 0.63, + "step": 1953 + }, + { + "epoch": 1.6178906681391496, + "grad_norm": 0.9202860593795776, + "learning_rate": 4.213707138562628e-06, + "loss": 0.632, + "step": 1954 + }, + { + "epoch": 1.6187189398122586, + "grad_norm": 0.8759778141975403, + "learning_rate": 4.212906301300415e-06, + "loss": 0.6253, + "step": 1955 + }, + { + "epoch": 1.6195472114853673, + "grad_norm": 0.8918818831443787, + "learning_rate": 4.212105132606801e-06, + "loss": 0.625, + "step": 1956 + }, + { + "epoch": 1.620375483158476, + "grad_norm": 0.9184549450874329, + "learning_rate": 4.211303632636804e-06, + "loss": 0.6293, + "step": 1957 + }, + { + "epoch": 1.6212037548315847, + "grad_norm": 0.8887206315994263, + "learning_rate": 4.210501801545506e-06, + "loss": 0.6118, + "step": 1958 + }, + { + "epoch": 1.6220320265046935, + "grad_norm": 0.9010420441627502, + "learning_rate": 4.209699639488054e-06, + "loss": 0.6209, + "step": 1959 + }, + { + "epoch": 1.6228602981778022, + "grad_norm": 0.8888309001922607, + "learning_rate": 4.20889714661966e-06, + "loss": 0.6301, + "step": 1960 + }, + { + "epoch": 1.6236885698509111, + "grad_norm": 0.9068322777748108, + "learning_rate": 4.208094323095597e-06, + "loss": 0.6234, + "step": 1961 + }, + { + "epoch": 1.6245168415240199, + "grad_norm": 0.8940340876579285, + "learning_rate": 4.207291169071207e-06, + "loss": 0.6076, + "step": 1962 + }, + { + "epoch": 1.6253451131971288, + "grad_norm": 0.907363772392273, + "learning_rate": 4.20648768470189e-06, + "loss": 0.6522, + "step": 1963 + }, + { + "epoch": 1.6261733848702375, + "grad_norm": 0.9104205965995789, + "learning_rate": 4.205683870143114e-06, + "loss": 0.6177, + "step": 1964 + }, + { + "epoch": 1.6270016565433463, + "grad_norm": 0.9051296710968018, + "learning_rate": 4.204879725550409e-06, + "loss": 0.6069, + "step": 1965 + }, + { + "epoch": 1.627829928216455, + "grad_norm": 0.9034865498542786, + "learning_rate": 4.20407525107937e-06, + "loss": 0.6099, + "step": 1966 + }, + { + "epoch": 1.6286581998895637, + "grad_norm": 0.9055590033531189, + "learning_rate": 4.203270446885655e-06, + "loss": 0.6226, + "step": 1967 + }, + { + "epoch": 1.6294864715626725, + "grad_norm": 0.897927463054657, + "learning_rate": 4.202465313124986e-06, + "loss": 0.6342, + "step": 1968 + }, + { + "epoch": 1.6303147432357812, + "grad_norm": 0.8995800018310547, + "learning_rate": 4.2016598499531495e-06, + "loss": 0.6408, + "step": 1969 + }, + { + "epoch": 1.6311430149088901, + "grad_norm": 0.8727264404296875, + "learning_rate": 4.200854057525995e-06, + "loss": 0.6263, + "step": 1970 + }, + { + "epoch": 1.6319712865819989, + "grad_norm": 0.9289891719818115, + "learning_rate": 4.200047935999435e-06, + "loss": 0.6351, + "step": 1971 + }, + { + "epoch": 1.6327995582551078, + "grad_norm": 0.8936645984649658, + "learning_rate": 4.199241485529446e-06, + "loss": 0.6205, + "step": 1972 + }, + { + "epoch": 1.6336278299282165, + "grad_norm": 0.9054183959960938, + "learning_rate": 4.198434706272072e-06, + "loss": 0.6101, + "step": 1973 + }, + { + "epoch": 1.6344561016013253, + "grad_norm": 0.9181169271469116, + "learning_rate": 4.1976275983834135e-06, + "loss": 0.6437, + "step": 1974 + }, + { + "epoch": 1.635284373274434, + "grad_norm": 0.9325777888298035, + "learning_rate": 4.1968201620196404e-06, + "loss": 0.6343, + "step": 1975 + }, + { + "epoch": 1.6361126449475427, + "grad_norm": 0.878094494342804, + "learning_rate": 4.196012397336983e-06, + "loss": 0.6298, + "step": 1976 + }, + { + "epoch": 1.6369409166206514, + "grad_norm": 0.9336429238319397, + "learning_rate": 4.195204304491737e-06, + "loss": 0.6232, + "step": 1977 + }, + { + "epoch": 1.6377691882937604, + "grad_norm": 0.9159499406814575, + "learning_rate": 4.19439588364026e-06, + "loss": 0.621, + "step": 1978 + }, + { + "epoch": 1.6385974599668691, + "grad_norm": 0.9075011610984802, + "learning_rate": 4.193587134938975e-06, + "loss": 0.6375, + "step": 1979 + }, + { + "epoch": 1.639425731639978, + "grad_norm": 0.8912638425827026, + "learning_rate": 4.192778058544367e-06, + "loss": 0.6494, + "step": 1980 + }, + { + "epoch": 1.6402540033130868, + "grad_norm": 0.9125227928161621, + "learning_rate": 4.191968654612984e-06, + "loss": 0.6238, + "step": 1981 + }, + { + "epoch": 1.6410822749861955, + "grad_norm": 0.906326174736023, + "learning_rate": 4.191158923301439e-06, + "loss": 0.5904, + "step": 1982 + }, + { + "epoch": 1.6419105466593042, + "grad_norm": 0.8858581781387329, + "learning_rate": 4.190348864766407e-06, + "loss": 0.6222, + "step": 1983 + }, + { + "epoch": 1.642738818332413, + "grad_norm": 0.864389955997467, + "learning_rate": 4.189538479164627e-06, + "loss": 0.6218, + "step": 1984 + }, + { + "epoch": 1.6435670900055217, + "grad_norm": 0.8983911871910095, + "learning_rate": 4.188727766652901e-06, + "loss": 0.6511, + "step": 1985 + }, + { + "epoch": 1.6443953616786304, + "grad_norm": 0.9633772373199463, + "learning_rate": 4.187916727388094e-06, + "loss": 0.623, + "step": 1986 + }, + { + "epoch": 1.6452236333517394, + "grad_norm": 0.9191408753395081, + "learning_rate": 4.187105361527135e-06, + "loss": 0.6245, + "step": 1987 + }, + { + "epoch": 1.646051905024848, + "grad_norm": 0.907792866230011, + "learning_rate": 4.186293669227016e-06, + "loss": 0.5994, + "step": 1988 + }, + { + "epoch": 1.646880176697957, + "grad_norm": 0.906218409538269, + "learning_rate": 4.18548165064479e-06, + "loss": 0.653, + "step": 1989 + }, + { + "epoch": 1.6477084483710658, + "grad_norm": 0.9408275485038757, + "learning_rate": 4.184669305937577e-06, + "loss": 0.5978, + "step": 1990 + }, + { + "epoch": 1.6485367200441745, + "grad_norm": 0.9278435111045837, + "learning_rate": 4.183856635262558e-06, + "loss": 0.6167, + "step": 1991 + }, + { + "epoch": 1.6493649917172832, + "grad_norm": 0.91252601146698, + "learning_rate": 4.183043638776978e-06, + "loss": 0.6229, + "step": 1992 + }, + { + "epoch": 1.650193263390392, + "grad_norm": 0.9178313612937927, + "learning_rate": 4.182230316638142e-06, + "loss": 0.6318, + "step": 1993 + }, + { + "epoch": 1.6510215350635007, + "grad_norm": 0.9056826233863831, + "learning_rate": 4.181416669003421e-06, + "loss": 0.5932, + "step": 1994 + }, + { + "epoch": 1.6518498067366096, + "grad_norm": 0.9139112830162048, + "learning_rate": 4.180602696030248e-06, + "loss": 0.6407, + "step": 1995 + }, + { + "epoch": 1.6526780784097184, + "grad_norm": 0.9195094108581543, + "learning_rate": 4.179788397876121e-06, + "loss": 0.6371, + "step": 1996 + }, + { + "epoch": 1.6535063500828273, + "grad_norm": 0.9221140146255493, + "learning_rate": 4.178973774698597e-06, + "loss": 0.611, + "step": 1997 + }, + { + "epoch": 1.654334621755936, + "grad_norm": 0.9180639386177063, + "learning_rate": 4.178158826655299e-06, + "loss": 0.6138, + "step": 1998 + }, + { + "epoch": 1.6551628934290448, + "grad_norm": 0.9255500435829163, + "learning_rate": 4.177343553903912e-06, + "loss": 0.6425, + "step": 1999 + }, + { + "epoch": 1.6559911651021535, + "grad_norm": 0.9378511905670166, + "learning_rate": 4.176527956602182e-06, + "loss": 0.6357, + "step": 2000 + }, + { + "epoch": 1.6568194367752622, + "grad_norm": 0.9212007522583008, + "learning_rate": 4.175712034907922e-06, + "loss": 0.6307, + "step": 2001 + }, + { + "epoch": 1.657647708448371, + "grad_norm": 0.9185839891433716, + "learning_rate": 4.174895788979004e-06, + "loss": 0.6173, + "step": 2002 + }, + { + "epoch": 1.65847598012148, + "grad_norm": 0.9658777117729187, + "learning_rate": 4.174079218973364e-06, + "loss": 0.622, + "step": 2003 + }, + { + "epoch": 1.6593042517945886, + "grad_norm": 0.9310017824172974, + "learning_rate": 4.173262325049e-06, + "loss": 0.6407, + "step": 2004 + }, + { + "epoch": 1.6601325234676974, + "grad_norm": 0.9048784375190735, + "learning_rate": 4.172445107363974e-06, + "loss": 0.6557, + "step": 2005 + }, + { + "epoch": 1.6609607951408063, + "grad_norm": 0.8820594549179077, + "learning_rate": 4.17162756607641e-06, + "loss": 0.6077, + "step": 2006 + }, + { + "epoch": 1.661789066813915, + "grad_norm": 0.8961437344551086, + "learning_rate": 4.170809701344496e-06, + "loss": 0.6151, + "step": 2007 + }, + { + "epoch": 1.6626173384870238, + "grad_norm": 0.9773615598678589, + "learning_rate": 4.169991513326479e-06, + "loss": 0.6339, + "step": 2008 + }, + { + "epoch": 1.6634456101601325, + "grad_norm": 0.9309290647506714, + "learning_rate": 4.169173002180672e-06, + "loss": 0.6235, + "step": 2009 + }, + { + "epoch": 1.6642738818332412, + "grad_norm": 1.0016266107559204, + "learning_rate": 4.168354168065448e-06, + "loss": 0.6355, + "step": 2010 + }, + { + "epoch": 1.66510215350635, + "grad_norm": 0.8820549249649048, + "learning_rate": 4.167535011139246e-06, + "loss": 0.6258, + "step": 2011 + }, + { + "epoch": 1.6659304251794589, + "grad_norm": 0.9030908942222595, + "learning_rate": 4.166715531560564e-06, + "loss": 0.6217, + "step": 2012 + }, + { + "epoch": 1.6667586968525676, + "grad_norm": 0.983755350112915, + "learning_rate": 4.165895729487962e-06, + "loss": 0.6281, + "step": 2013 + }, + { + "epoch": 1.6675869685256766, + "grad_norm": 0.948001503944397, + "learning_rate": 4.165075605080066e-06, + "loss": 0.6242, + "step": 2014 + }, + { + "epoch": 1.6684152401987853, + "grad_norm": 0.9468982219696045, + "learning_rate": 4.164255158495563e-06, + "loss": 0.6563, + "step": 2015 + }, + { + "epoch": 1.669243511871894, + "grad_norm": 0.9012095332145691, + "learning_rate": 4.1634343898932e-06, + "loss": 0.618, + "step": 2016 + }, + { + "epoch": 1.6700717835450027, + "grad_norm": 0.9554924964904785, + "learning_rate": 4.162613299431789e-06, + "loss": 0.6306, + "step": 2017 + }, + { + "epoch": 1.6709000552181115, + "grad_norm": 0.9587515592575073, + "learning_rate": 4.161791887270203e-06, + "loss": 0.6305, + "step": 2018 + }, + { + "epoch": 1.6717283268912202, + "grad_norm": 0.8792864084243774, + "learning_rate": 4.160970153567379e-06, + "loss": 0.6181, + "step": 2019 + }, + { + "epoch": 1.6725565985643291, + "grad_norm": 0.8958392143249512, + "learning_rate": 4.160148098482313e-06, + "loss": 0.623, + "step": 2020 + }, + { + "epoch": 1.6733848702374379, + "grad_norm": 0.9501842260360718, + "learning_rate": 4.159325722174065e-06, + "loss": 0.6454, + "step": 2021 + }, + { + "epoch": 1.6742131419105468, + "grad_norm": 0.9107666015625, + "learning_rate": 4.158503024801759e-06, + "loss": 0.6237, + "step": 2022 + }, + { + "epoch": 1.6750414135836555, + "grad_norm": 0.9405720233917236, + "learning_rate": 4.157680006524577e-06, + "loss": 0.6423, + "step": 2023 + }, + { + "epoch": 1.6758696852567643, + "grad_norm": 0.9008631706237793, + "learning_rate": 4.1568566675017675e-06, + "loss": 0.6385, + "step": 2024 + }, + { + "epoch": 1.676697956929873, + "grad_norm": 0.9056677222251892, + "learning_rate": 4.156033007892637e-06, + "loss": 0.6225, + "step": 2025 + }, + { + "epoch": 1.6775262286029817, + "grad_norm": 0.899982750415802, + "learning_rate": 4.155209027856557e-06, + "loss": 0.6268, + "step": 2026 + }, + { + "epoch": 1.6783545002760905, + "grad_norm": 0.928506076335907, + "learning_rate": 4.15438472755296e-06, + "loss": 0.6343, + "step": 2027 + }, + { + "epoch": 1.6791827719491992, + "grad_norm": 0.8831804394721985, + "learning_rate": 4.15356010714134e-06, + "loss": 0.6404, + "step": 2028 + }, + { + "epoch": 1.6800110436223081, + "grad_norm": 0.8637033700942993, + "learning_rate": 4.152735166781253e-06, + "loss": 0.6305, + "step": 2029 + }, + { + "epoch": 1.6808393152954169, + "grad_norm": 0.9164902567863464, + "learning_rate": 4.151909906632319e-06, + "loss": 0.6171, + "step": 2030 + }, + { + "epoch": 1.6816675869685258, + "grad_norm": 0.9124466776847839, + "learning_rate": 4.151084326854216e-06, + "loss": 0.6446, + "step": 2031 + }, + { + "epoch": 1.6824958586416345, + "grad_norm": 0.9212199449539185, + "learning_rate": 4.150258427606687e-06, + "loss": 0.6074, + "step": 2032 + }, + { + "epoch": 1.6833241303147433, + "grad_norm": 0.9536449313163757, + "learning_rate": 4.149432209049536e-06, + "loss": 0.6149, + "step": 2033 + }, + { + "epoch": 1.684152401987852, + "grad_norm": 0.9232591390609741, + "learning_rate": 4.148605671342629e-06, + "loss": 0.6238, + "step": 2034 + }, + { + "epoch": 1.6849806736609607, + "grad_norm": 0.8844353556632996, + "learning_rate": 4.147778814645892e-06, + "loss": 0.6434, + "step": 2035 + }, + { + "epoch": 1.6858089453340694, + "grad_norm": 0.8500795960426331, + "learning_rate": 4.1469516391193155e-06, + "loss": 0.6304, + "step": 2036 + }, + { + "epoch": 1.6866372170071784, + "grad_norm": 0.8676083087921143, + "learning_rate": 4.146124144922949e-06, + "loss": 0.6158, + "step": 2037 + }, + { + "epoch": 1.6874654886802871, + "grad_norm": 0.8914940357208252, + "learning_rate": 4.1452963322169046e-06, + "loss": 0.6336, + "step": 2038 + }, + { + "epoch": 1.688293760353396, + "grad_norm": 0.9245527386665344, + "learning_rate": 4.144468201161358e-06, + "loss": 0.6385, + "step": 2039 + }, + { + "epoch": 1.6891220320265048, + "grad_norm": 0.8946509957313538, + "learning_rate": 4.143639751916544e-06, + "loss": 0.6243, + "step": 2040 + }, + { + "epoch": 1.6899503036996135, + "grad_norm": 0.882095217704773, + "learning_rate": 4.14281098464276e-06, + "loss": 0.6171, + "step": 2041 + }, + { + "epoch": 1.6907785753727222, + "grad_norm": 0.8981113433837891, + "learning_rate": 4.141981899500364e-06, + "loss": 0.6389, + "step": 2042 + }, + { + "epoch": 1.691606847045831, + "grad_norm": 0.928701639175415, + "learning_rate": 4.1411524966497775e-06, + "loss": 0.6488, + "step": 2043 + }, + { + "epoch": 1.6924351187189397, + "grad_norm": 0.8937084674835205, + "learning_rate": 4.140322776251481e-06, + "loss": 0.6415, + "step": 2044 + }, + { + "epoch": 1.6932633903920484, + "grad_norm": 0.8940730094909668, + "learning_rate": 4.1394927384660185e-06, + "loss": 0.6419, + "step": 2045 + }, + { + "epoch": 1.6940916620651574, + "grad_norm": 0.9067767262458801, + "learning_rate": 4.138662383453994e-06, + "loss": 0.608, + "step": 2046 + }, + { + "epoch": 1.694919933738266, + "grad_norm": 0.9123063087463379, + "learning_rate": 4.137831711376074e-06, + "loss": 0.633, + "step": 2047 + }, + { + "epoch": 1.695748205411375, + "grad_norm": 0.89399653673172, + "learning_rate": 4.137000722392985e-06, + "loss": 0.6338, + "step": 2048 + }, + { + "epoch": 1.6965764770844838, + "grad_norm": 0.9146568775177002, + "learning_rate": 4.136169416665517e-06, + "loss": 0.6214, + "step": 2049 + }, + { + "epoch": 1.6974047487575925, + "grad_norm": 0.9402440786361694, + "learning_rate": 4.135337794354518e-06, + "loss": 0.6266, + "step": 2050 + }, + { + "epoch": 1.6982330204307012, + "grad_norm": 0.8790574669837952, + "learning_rate": 4.134505855620902e-06, + "loss": 0.5989, + "step": 2051 + }, + { + "epoch": 1.69906129210381, + "grad_norm": 0.9421536326408386, + "learning_rate": 4.133673600625637e-06, + "loss": 0.611, + "step": 2052 + }, + { + "epoch": 1.6998895637769187, + "grad_norm": 0.9365546703338623, + "learning_rate": 4.132841029529762e-06, + "loss": 0.6201, + "step": 2053 + }, + { + "epoch": 1.7007178354500276, + "grad_norm": 0.9213638305664062, + "learning_rate": 4.132008142494367e-06, + "loss": 0.632, + "step": 2054 + }, + { + "epoch": 1.7015461071231364, + "grad_norm": 0.9204041957855225, + "learning_rate": 4.13117493968061e-06, + "loss": 0.6086, + "step": 2055 + }, + { + "epoch": 1.7023743787962453, + "grad_norm": 0.8972902297973633, + "learning_rate": 4.130341421249708e-06, + "loss": 0.656, + "step": 2056 + }, + { + "epoch": 1.703202650469354, + "grad_norm": 0.9163963794708252, + "learning_rate": 4.129507587362939e-06, + "loss": 0.623, + "step": 2057 + }, + { + "epoch": 1.7040309221424628, + "grad_norm": 0.9178527593612671, + "learning_rate": 4.128673438181642e-06, + "loss": 0.6327, + "step": 2058 + }, + { + "epoch": 1.7048591938155715, + "grad_norm": 0.8982755541801453, + "learning_rate": 4.127838973867217e-06, + "loss": 0.6165, + "step": 2059 + }, + { + "epoch": 1.7056874654886802, + "grad_norm": 0.8995897769927979, + "learning_rate": 4.127004194581125e-06, + "loss": 0.6204, + "step": 2060 + }, + { + "epoch": 1.706515737161789, + "grad_norm": 0.9148748517036438, + "learning_rate": 4.126169100484888e-06, + "loss": 0.6177, + "step": 2061 + }, + { + "epoch": 1.707344008834898, + "grad_norm": 0.9249931573867798, + "learning_rate": 4.12533369174009e-06, + "loss": 0.6113, + "step": 2062 + }, + { + "epoch": 1.7081722805080066, + "grad_norm": 0.9162423610687256, + "learning_rate": 4.124497968508372e-06, + "loss": 0.6119, + "step": 2063 + }, + { + "epoch": 1.7090005521811154, + "grad_norm": 0.8919503688812256, + "learning_rate": 4.123661930951441e-06, + "loss": 0.6194, + "step": 2064 + }, + { + "epoch": 1.7098288238542243, + "grad_norm": 0.8884986639022827, + "learning_rate": 4.122825579231063e-06, + "loss": 0.656, + "step": 2065 + }, + { + "epoch": 1.710657095527333, + "grad_norm": 0.9072293043136597, + "learning_rate": 4.121988913509062e-06, + "loss": 0.6351, + "step": 2066 + }, + { + "epoch": 1.7114853672004418, + "grad_norm": 0.9255547523498535, + "learning_rate": 4.121151933947327e-06, + "loss": 0.601, + "step": 2067 + }, + { + "epoch": 1.7123136388735505, + "grad_norm": 0.9141303300857544, + "learning_rate": 4.120314640707804e-06, + "loss": 0.6257, + "step": 2068 + }, + { + "epoch": 1.7131419105466592, + "grad_norm": 0.8791571855545044, + "learning_rate": 4.119477033952504e-06, + "loss": 0.6355, + "step": 2069 + }, + { + "epoch": 1.713970182219768, + "grad_norm": 0.8910380005836487, + "learning_rate": 4.118639113843493e-06, + "loss": 0.6405, + "step": 2070 + }, + { + "epoch": 1.7147984538928769, + "grad_norm": 0.9583408236503601, + "learning_rate": 4.117800880542904e-06, + "loss": 0.6408, + "step": 2071 + }, + { + "epoch": 1.7156267255659856, + "grad_norm": 0.9684842824935913, + "learning_rate": 4.116962334212924e-06, + "loss": 0.6155, + "step": 2072 + }, + { + "epoch": 1.7164549972390946, + "grad_norm": 0.8882633447647095, + "learning_rate": 4.116123475015806e-06, + "loss": 0.6332, + "step": 2073 + }, + { + "epoch": 1.7172832689122033, + "grad_norm": 0.8599907755851746, + "learning_rate": 4.1152843031138604e-06, + "loss": 0.6416, + "step": 2074 + }, + { + "epoch": 1.718111540585312, + "grad_norm": 0.9216133952140808, + "learning_rate": 4.114444818669461e-06, + "loss": 0.61, + "step": 2075 + }, + { + "epoch": 1.7189398122584207, + "grad_norm": 0.943729817867279, + "learning_rate": 4.113605021845037e-06, + "loss": 0.643, + "step": 2076 + }, + { + "epoch": 1.7197680839315295, + "grad_norm": 0.9158332943916321, + "learning_rate": 4.112764912803084e-06, + "loss": 0.6353, + "step": 2077 + }, + { + "epoch": 1.7205963556046382, + "grad_norm": 0.9036465287208557, + "learning_rate": 4.111924491706154e-06, + "loss": 0.6143, + "step": 2078 + }, + { + "epoch": 1.7214246272777471, + "grad_norm": 0.9264538884162903, + "learning_rate": 4.111083758716862e-06, + "loss": 0.6302, + "step": 2079 + }, + { + "epoch": 1.7222528989508559, + "grad_norm": 0.9309650659561157, + "learning_rate": 4.11024271399788e-06, + "loss": 0.6394, + "step": 2080 + }, + { + "epoch": 1.7230811706239648, + "grad_norm": 0.9261552095413208, + "learning_rate": 4.109401357711944e-06, + "loss": 0.6338, + "step": 2081 + }, + { + "epoch": 1.7239094422970735, + "grad_norm": 0.9027671217918396, + "learning_rate": 4.1085596900218476e-06, + "loss": 0.633, + "step": 2082 + }, + { + "epoch": 1.7247377139701823, + "grad_norm": 0.8899479508399963, + "learning_rate": 4.107717711090446e-06, + "loss": 0.6372, + "step": 2083 + }, + { + "epoch": 1.725565985643291, + "grad_norm": 0.9513740539550781, + "learning_rate": 4.106875421080654e-06, + "loss": 0.6057, + "step": 2084 + }, + { + "epoch": 1.7263942573163997, + "grad_norm": 0.8926359415054321, + "learning_rate": 4.106032820155447e-06, + "loss": 0.6219, + "step": 2085 + }, + { + "epoch": 1.7272225289895085, + "grad_norm": 0.8961042165756226, + "learning_rate": 4.105189908477862e-06, + "loss": 0.5926, + "step": 2086 + }, + { + "epoch": 1.7280508006626172, + "grad_norm": 0.8870651125907898, + "learning_rate": 4.104346686210992e-06, + "loss": 0.6298, + "step": 2087 + }, + { + "epoch": 1.7288790723357261, + "grad_norm": 0.9134538173675537, + "learning_rate": 4.103503153517993e-06, + "loss": 0.6544, + "step": 2088 + }, + { + "epoch": 1.7297073440088349, + "grad_norm": 0.900499165058136, + "learning_rate": 4.102659310562082e-06, + "loss": 0.6182, + "step": 2089 + }, + { + "epoch": 1.7305356156819438, + "grad_norm": 0.9146167039871216, + "learning_rate": 4.101815157506534e-06, + "loss": 0.6297, + "step": 2090 + }, + { + "epoch": 1.7313638873550525, + "grad_norm": 0.909670352935791, + "learning_rate": 4.100970694514687e-06, + "loss": 0.6272, + "step": 2091 + }, + { + "epoch": 1.7321921590281613, + "grad_norm": 1.127607822418213, + "learning_rate": 4.100125921749932e-06, + "loss": 0.6416, + "step": 2092 + }, + { + "epoch": 1.73302043070127, + "grad_norm": 0.8950872421264648, + "learning_rate": 4.099280839375729e-06, + "loss": 0.6411, + "step": 2093 + }, + { + "epoch": 1.7338487023743787, + "grad_norm": 0.9612913727760315, + "learning_rate": 4.098435447555591e-06, + "loss": 0.626, + "step": 2094 + }, + { + "epoch": 1.7346769740474874, + "grad_norm": 0.927638053894043, + "learning_rate": 4.097589746453094e-06, + "loss": 0.6197, + "step": 2095 + }, + { + "epoch": 1.7355052457205964, + "grad_norm": 0.9428266286849976, + "learning_rate": 4.096743736231875e-06, + "loss": 0.6407, + "step": 2096 + }, + { + "epoch": 1.7363335173937051, + "grad_norm": 0.8619174361228943, + "learning_rate": 4.095897417055627e-06, + "loss": 0.6386, + "step": 2097 + }, + { + "epoch": 1.737161789066814, + "grad_norm": 0.8943668007850647, + "learning_rate": 4.095050789088106e-06, + "loss": 0.6206, + "step": 2098 + }, + { + "epoch": 1.7379900607399228, + "grad_norm": 0.9190096259117126, + "learning_rate": 4.0942038524931245e-06, + "loss": 0.6395, + "step": 2099 + }, + { + "epoch": 1.7388183324130315, + "grad_norm": 0.9055671095848083, + "learning_rate": 4.09335660743456e-06, + "loss": 0.6384, + "step": 2100 + }, + { + "epoch": 1.7396466040861402, + "grad_norm": 0.9204139709472656, + "learning_rate": 4.092509054076345e-06, + "loss": 0.6438, + "step": 2101 + }, + { + "epoch": 1.740474875759249, + "grad_norm": 0.9572716355323792, + "learning_rate": 4.091661192582473e-06, + "loss": 0.6233, + "step": 2102 + }, + { + "epoch": 1.7413031474323577, + "grad_norm": 0.9479691982269287, + "learning_rate": 4.0908130231169975e-06, + "loss": 0.6284, + "step": 2103 + }, + { + "epoch": 1.7421314191054664, + "grad_norm": 0.9081270694732666, + "learning_rate": 4.089964545844032e-06, + "loss": 0.6284, + "step": 2104 + }, + { + "epoch": 1.7429596907785754, + "grad_norm": 0.9227046966552734, + "learning_rate": 4.0891157609277475e-06, + "loss": 0.6611, + "step": 2105 + }, + { + "epoch": 1.743787962451684, + "grad_norm": 0.9033496975898743, + "learning_rate": 4.088266668532377e-06, + "loss": 0.6122, + "step": 2106 + }, + { + "epoch": 1.744616234124793, + "grad_norm": 0.933539628982544, + "learning_rate": 4.087417268822211e-06, + "loss": 0.6167, + "step": 2107 + }, + { + "epoch": 1.7454445057979018, + "grad_norm": 0.9211486577987671, + "learning_rate": 4.086567561961602e-06, + "loss": 0.6364, + "step": 2108 + }, + { + "epoch": 1.7462727774710105, + "grad_norm": 0.8782284259796143, + "learning_rate": 4.08571754811496e-06, + "loss": 0.6164, + "step": 2109 + }, + { + "epoch": 1.7471010491441192, + "grad_norm": 0.8936499357223511, + "learning_rate": 4.084867227446753e-06, + "loss": 0.6155, + "step": 2110 + }, + { + "epoch": 1.747929320817228, + "grad_norm": 0.8931904435157776, + "learning_rate": 4.084016600121513e-06, + "loss": 0.6512, + "step": 2111 + }, + { + "epoch": 1.7487575924903367, + "grad_norm": 0.9070253372192383, + "learning_rate": 4.083165666303826e-06, + "loss": 0.6197, + "step": 2112 + }, + { + "epoch": 1.7495858641634456, + "grad_norm": 0.9232584238052368, + "learning_rate": 4.08231442615834e-06, + "loss": 0.6332, + "step": 2113 + }, + { + "epoch": 1.7504141358365544, + "grad_norm": 0.92015540599823, + "learning_rate": 4.081462879849765e-06, + "loss": 0.6351, + "step": 2114 + }, + { + "epoch": 1.7512424075096633, + "grad_norm": 0.9353263974189758, + "learning_rate": 4.080611027542863e-06, + "loss": 0.6479, + "step": 2115 + }, + { + "epoch": 1.752070679182772, + "grad_norm": 0.8943057060241699, + "learning_rate": 4.079758869402462e-06, + "loss": 0.6179, + "step": 2116 + }, + { + "epoch": 1.7528989508558808, + "grad_norm": 0.9062668681144714, + "learning_rate": 4.078906405593446e-06, + "loss": 0.6166, + "step": 2117 + }, + { + "epoch": 1.7537272225289895, + "grad_norm": 0.9368826150894165, + "learning_rate": 4.078053636280759e-06, + "loss": 0.6206, + "step": 2118 + }, + { + "epoch": 1.7545554942020982, + "grad_norm": 0.9012559652328491, + "learning_rate": 4.0772005616294034e-06, + "loss": 0.6445, + "step": 2119 + }, + { + "epoch": 1.755383765875207, + "grad_norm": 0.9061569571495056, + "learning_rate": 4.076347181804442e-06, + "loss": 0.6312, + "step": 2120 + }, + { + "epoch": 1.756212037548316, + "grad_norm": 0.892900288105011, + "learning_rate": 4.075493496970996e-06, + "loss": 0.64, + "step": 2121 + }, + { + "epoch": 1.7570403092214246, + "grad_norm": 0.9200263023376465, + "learning_rate": 4.074639507294243e-06, + "loss": 0.6435, + "step": 2122 + }, + { + "epoch": 1.7578685808945336, + "grad_norm": 0.921215295791626, + "learning_rate": 4.0737852129394265e-06, + "loss": 0.6135, + "step": 2123 + }, + { + "epoch": 1.7586968525676423, + "grad_norm": 0.8980262279510498, + "learning_rate": 4.07293061407184e-06, + "loss": 0.635, + "step": 2124 + }, + { + "epoch": 1.759525124240751, + "grad_norm": 0.9192932844161987, + "learning_rate": 4.072075710856843e-06, + "loss": 0.6317, + "step": 2125 + }, + { + "epoch": 1.7603533959138598, + "grad_norm": 0.9387120008468628, + "learning_rate": 4.07122050345985e-06, + "loss": 0.6324, + "step": 2126 + }, + { + "epoch": 1.7611816675869685, + "grad_norm": 0.8787287473678589, + "learning_rate": 4.070364992046338e-06, + "loss": 0.6177, + "step": 2127 + }, + { + "epoch": 1.7620099392600772, + "grad_norm": 0.8963474631309509, + "learning_rate": 4.069509176781838e-06, + "loss": 0.6158, + "step": 2128 + }, + { + "epoch": 1.762838210933186, + "grad_norm": 0.8926964402198792, + "learning_rate": 4.068653057831942e-06, + "loss": 0.6528, + "step": 2129 + }, + { + "epoch": 1.7636664826062949, + "grad_norm": 0.9290505647659302, + "learning_rate": 4.067796635362303e-06, + "loss": 0.6274, + "step": 2130 + }, + { + "epoch": 1.7644947542794036, + "grad_norm": 0.8848921060562134, + "learning_rate": 4.06693990953863e-06, + "loss": 0.6507, + "step": 2131 + }, + { + "epoch": 1.7653230259525126, + "grad_norm": 0.8977338075637817, + "learning_rate": 4.0660828805266915e-06, + "loss": 0.6437, + "step": 2132 + }, + { + "epoch": 1.7661512976256213, + "grad_norm": 0.9125422239303589, + "learning_rate": 4.065225548492314e-06, + "loss": 0.6221, + "step": 2133 + }, + { + "epoch": 1.76697956929873, + "grad_norm": 0.9186370968818665, + "learning_rate": 4.064367913601384e-06, + "loss": 0.6174, + "step": 2134 + }, + { + "epoch": 1.7678078409718387, + "grad_norm": 0.9084877967834473, + "learning_rate": 4.063509976019845e-06, + "loss": 0.6228, + "step": 2135 + }, + { + "epoch": 1.7686361126449475, + "grad_norm": 0.9180804491043091, + "learning_rate": 4.0626517359137e-06, + "loss": 0.6126, + "step": 2136 + }, + { + "epoch": 1.7694643843180562, + "grad_norm": 0.9099741578102112, + "learning_rate": 4.061793193449013e-06, + "loss": 0.6027, + "step": 2137 + }, + { + "epoch": 1.7702926559911651, + "grad_norm": 0.9060433506965637, + "learning_rate": 4.0609343487919005e-06, + "loss": 0.632, + "step": 2138 + }, + { + "epoch": 1.7711209276642739, + "grad_norm": 0.9404996037483215, + "learning_rate": 4.060075202108542e-06, + "loss": 0.6405, + "step": 2139 + }, + { + "epoch": 1.7719491993373828, + "grad_norm": 0.9031956791877747, + "learning_rate": 4.059215753565176e-06, + "loss": 0.6137, + "step": 2140 + }, + { + "epoch": 1.7727774710104915, + "grad_norm": 0.8913756012916565, + "learning_rate": 4.058356003328095e-06, + "loss": 0.6387, + "step": 2141 + }, + { + "epoch": 1.7736057426836003, + "grad_norm": 0.8986163139343262, + "learning_rate": 4.057495951563655e-06, + "loss": 0.6151, + "step": 2142 + }, + { + "epoch": 1.774434014356709, + "grad_norm": 0.9017037153244019, + "learning_rate": 4.0566355984382676e-06, + "loss": 0.6466, + "step": 2143 + }, + { + "epoch": 1.7752622860298177, + "grad_norm": 0.9211136698722839, + "learning_rate": 4.0557749441184024e-06, + "loss": 0.638, + "step": 2144 + }, + { + "epoch": 1.7760905577029265, + "grad_norm": 0.9097506999969482, + "learning_rate": 4.054913988770588e-06, + "loss": 0.6529, + "step": 2145 + }, + { + "epoch": 1.7769188293760352, + "grad_norm": 0.8985908627510071, + "learning_rate": 4.054052732561411e-06, + "loss": 0.6388, + "step": 2146 + }, + { + "epoch": 1.7777471010491441, + "grad_norm": 0.9066522121429443, + "learning_rate": 4.053191175657518e-06, + "loss": 0.6429, + "step": 2147 + }, + { + "epoch": 1.7785753727222529, + "grad_norm": 0.9080630540847778, + "learning_rate": 4.052329318225611e-06, + "loss": 0.619, + "step": 2148 + }, + { + "epoch": 1.7794036443953618, + "grad_norm": 0.8701719641685486, + "learning_rate": 4.0514671604324505e-06, + "loss": 0.6245, + "step": 2149 + }, + { + "epoch": 1.7802319160684705, + "grad_norm": 0.9059606194496155, + "learning_rate": 4.0506047024448585e-06, + "loss": 0.6124, + "step": 2150 + }, + { + "epoch": 1.7810601877415793, + "grad_norm": 0.8912079334259033, + "learning_rate": 4.04974194442971e-06, + "loss": 0.6295, + "step": 2151 + }, + { + "epoch": 1.781888459414688, + "grad_norm": 0.9063137769699097, + "learning_rate": 4.048878886553942e-06, + "loss": 0.6299, + "step": 2152 + }, + { + "epoch": 1.7827167310877967, + "grad_norm": 0.8850398659706116, + "learning_rate": 4.048015528984548e-06, + "loss": 0.6251, + "step": 2153 + }, + { + "epoch": 1.7835450027609054, + "grad_norm": 0.9059123992919922, + "learning_rate": 4.04715187188858e-06, + "loss": 0.6278, + "step": 2154 + }, + { + "epoch": 1.7843732744340144, + "grad_norm": 0.8777562379837036, + "learning_rate": 4.046287915433147e-06, + "loss": 0.6003, + "step": 2155 + }, + { + "epoch": 1.7852015461071231, + "grad_norm": 0.9048421382904053, + "learning_rate": 4.045423659785417e-06, + "loss": 0.6143, + "step": 2156 + }, + { + "epoch": 1.786029817780232, + "grad_norm": 0.9388517737388611, + "learning_rate": 4.044559105112614e-06, + "loss": 0.6314, + "step": 2157 + }, + { + "epoch": 1.7868580894533408, + "grad_norm": 0.9405019283294678, + "learning_rate": 4.043694251582023e-06, + "loss": 0.6381, + "step": 2158 + }, + { + "epoch": 1.7876863611264495, + "grad_norm": 0.8942162990570068, + "learning_rate": 4.0428290993609845e-06, + "loss": 0.6318, + "step": 2159 + }, + { + "epoch": 1.7885146327995582, + "grad_norm": 0.9065560698509216, + "learning_rate": 4.041963648616897e-06, + "loss": 0.6258, + "step": 2160 + }, + { + "epoch": 1.789342904472667, + "grad_norm": 0.8766914010047913, + "learning_rate": 4.041097899517218e-06, + "loss": 0.6292, + "step": 2161 + }, + { + "epoch": 1.7901711761457757, + "grad_norm": 0.8942935466766357, + "learning_rate": 4.040231852229462e-06, + "loss": 0.662, + "step": 2162 + }, + { + "epoch": 1.7909994478188846, + "grad_norm": 0.8892539143562317, + "learning_rate": 4.039365506921198e-06, + "loss": 0.6251, + "step": 2163 + }, + { + "epoch": 1.7918277194919934, + "grad_norm": 0.9359427690505981, + "learning_rate": 4.038498863760061e-06, + "loss": 0.6222, + "step": 2164 + }, + { + "epoch": 1.792655991165102, + "grad_norm": 0.8899942636489868, + "learning_rate": 4.037631922913733e-06, + "loss": 0.6318, + "step": 2165 + }, + { + "epoch": 1.793484262838211, + "grad_norm": 0.9233553409576416, + "learning_rate": 4.0367646845499635e-06, + "loss": 0.6188, + "step": 2166 + }, + { + "epoch": 1.7943125345113198, + "grad_norm": 0.8946877121925354, + "learning_rate": 4.035897148836553e-06, + "loss": 0.6247, + "step": 2167 + }, + { + "epoch": 1.7951408061844285, + "grad_norm": 0.9200116991996765, + "learning_rate": 4.03502931594136e-06, + "loss": 0.5981, + "step": 2168 + }, + { + "epoch": 1.7959690778575372, + "grad_norm": 0.9193991422653198, + "learning_rate": 4.034161186032306e-06, + "loss": 0.6144, + "step": 2169 + }, + { + "epoch": 1.796797349530646, + "grad_norm": 0.9637252688407898, + "learning_rate": 4.0332927592773615e-06, + "loss": 0.6094, + "step": 2170 + }, + { + "epoch": 1.7976256212037547, + "grad_norm": 0.8958486318588257, + "learning_rate": 4.032424035844564e-06, + "loss": 0.6304, + "step": 2171 + }, + { + "epoch": 1.7984538928768636, + "grad_norm": 0.9411646127700806, + "learning_rate": 4.031555015901998e-06, + "loss": 0.5984, + "step": 2172 + }, + { + "epoch": 1.7992821645499724, + "grad_norm": 0.8594862222671509, + "learning_rate": 4.030685699617815e-06, + "loss": 0.6176, + "step": 2173 + }, + { + "epoch": 1.8001104362230813, + "grad_norm": 0.8969242572784424, + "learning_rate": 4.029816087160217e-06, + "loss": 0.6294, + "step": 2174 + }, + { + "epoch": 1.80093870789619, + "grad_norm": 0.8950173854827881, + "learning_rate": 4.028946178697467e-06, + "loss": 0.6441, + "step": 2175 + }, + { + "epoch": 1.8017669795692988, + "grad_norm": 0.9084720611572266, + "learning_rate": 4.0280759743978835e-06, + "loss": 0.6245, + "step": 2176 + }, + { + "epoch": 1.8025952512424075, + "grad_norm": 0.9174644947052002, + "learning_rate": 4.027205474429845e-06, + "loss": 0.6136, + "step": 2177 + }, + { + "epoch": 1.8034235229155162, + "grad_norm": 0.9103249907493591, + "learning_rate": 4.026334678961783e-06, + "loss": 0.6341, + "step": 2178 + }, + { + "epoch": 1.804251794588625, + "grad_norm": 0.8916360139846802, + "learning_rate": 4.025463588162189e-06, + "loss": 0.6424, + "step": 2179 + }, + { + "epoch": 1.805080066261734, + "grad_norm": 0.9347325563430786, + "learning_rate": 4.024592202199611e-06, + "loss": 0.6425, + "step": 2180 + }, + { + "epoch": 1.8059083379348426, + "grad_norm": 0.9135183691978455, + "learning_rate": 4.023720521242653e-06, + "loss": 0.6265, + "step": 2181 + }, + { + "epoch": 1.8067366096079516, + "grad_norm": 0.8996950387954712, + "learning_rate": 4.022848545459979e-06, + "loss": 0.6083, + "step": 2182 + }, + { + "epoch": 1.8075648812810603, + "grad_norm": 0.9104750752449036, + "learning_rate": 4.021976275020307e-06, + "loss": 0.6087, + "step": 2183 + }, + { + "epoch": 1.808393152954169, + "grad_norm": 0.9272193908691406, + "learning_rate": 4.0211037100924135e-06, + "loss": 0.6254, + "step": 2184 + }, + { + "epoch": 1.8092214246272778, + "grad_norm": 0.8605275750160217, + "learning_rate": 4.020230850845132e-06, + "loss": 0.6244, + "step": 2185 + }, + { + "epoch": 1.8100496963003865, + "grad_norm": 0.8817747831344604, + "learning_rate": 4.019357697447352e-06, + "loss": 0.6226, + "step": 2186 + }, + { + "epoch": 1.8108779679734952, + "grad_norm": 0.8978784084320068, + "learning_rate": 4.018484250068021e-06, + "loss": 0.615, + "step": 2187 + }, + { + "epoch": 1.811706239646604, + "grad_norm": 0.8996117115020752, + "learning_rate": 4.017610508876143e-06, + "loss": 0.6248, + "step": 2188 + }, + { + "epoch": 1.8125345113197129, + "grad_norm": 0.9133120179176331, + "learning_rate": 4.016736474040778e-06, + "loss": 0.6364, + "step": 2189 + }, + { + "epoch": 1.8133627829928216, + "grad_norm": 0.903982400894165, + "learning_rate": 4.015862145731044e-06, + "loss": 0.6078, + "step": 2190 + }, + { + "epoch": 1.8141910546659306, + "grad_norm": 0.9343582987785339, + "learning_rate": 4.014987524116115e-06, + "loss": 0.6254, + "step": 2191 + }, + { + "epoch": 1.8150193263390393, + "grad_norm": 0.883305549621582, + "learning_rate": 4.014112609365224e-06, + "loss": 0.6306, + "step": 2192 + }, + { + "epoch": 1.815847598012148, + "grad_norm": 0.9044662714004517, + "learning_rate": 4.013237401647656e-06, + "loss": 0.6288, + "step": 2193 + }, + { + "epoch": 1.8166758696852567, + "grad_norm": 0.9016374349594116, + "learning_rate": 4.012361901132758e-06, + "loss": 0.6254, + "step": 2194 + }, + { + "epoch": 1.8175041413583655, + "grad_norm": 0.9198315739631653, + "learning_rate": 4.01148610798993e-06, + "loss": 0.621, + "step": 2195 + }, + { + "epoch": 1.8183324130314742, + "grad_norm": 0.8884013891220093, + "learning_rate": 4.01061002238863e-06, + "loss": 0.6195, + "step": 2196 + }, + { + "epoch": 1.8191606847045831, + "grad_norm": 0.901606023311615, + "learning_rate": 4.009733644498372e-06, + "loss": 0.646, + "step": 2197 + }, + { + "epoch": 1.8199889563776919, + "grad_norm": 0.9003238677978516, + "learning_rate": 4.008856974488728e-06, + "loss": 0.6276, + "step": 2198 + }, + { + "epoch": 1.8208172280508008, + "grad_norm": 0.8835832476615906, + "learning_rate": 4.007980012529324e-06, + "loss": 0.6232, + "step": 2199 + }, + { + "epoch": 1.8216454997239095, + "grad_norm": 0.8751473426818848, + "learning_rate": 4.007102758789846e-06, + "loss": 0.635, + "step": 2200 + }, + { + "epoch": 1.8224737713970183, + "grad_norm": 0.8990851640701294, + "learning_rate": 4.006225213440032e-06, + "loss": 0.605, + "step": 2201 + }, + { + "epoch": 1.823302043070127, + "grad_norm": 0.8766496181488037, + "learning_rate": 4.0053473766496805e-06, + "loss": 0.6396, + "step": 2202 + }, + { + "epoch": 1.8241303147432357, + "grad_norm": 0.914273202419281, + "learning_rate": 4.004469248588644e-06, + "loss": 0.6435, + "step": 2203 + }, + { + "epoch": 1.8249585864163445, + "grad_norm": 0.9063234925270081, + "learning_rate": 4.003590829426832e-06, + "loss": 0.6144, + "step": 2204 + }, + { + "epoch": 1.8257868580894532, + "grad_norm": 0.9479252696037292, + "learning_rate": 4.002712119334212e-06, + "loss": 0.6584, + "step": 2205 + }, + { + "epoch": 1.8266151297625621, + "grad_norm": 0.9253218770027161, + "learning_rate": 4.001833118480804e-06, + "loss": 0.6203, + "step": 2206 + }, + { + "epoch": 1.8274434014356709, + "grad_norm": 0.930874764919281, + "learning_rate": 4.000953827036689e-06, + "loss": 0.6254, + "step": 2207 + }, + { + "epoch": 1.8282716731087798, + "grad_norm": 0.9202263951301575, + "learning_rate": 4.000074245171999e-06, + "loss": 0.6351, + "step": 2208 + }, + { + "epoch": 1.8290999447818885, + "grad_norm": 0.8993806838989258, + "learning_rate": 3.999194373056926e-06, + "loss": 0.6439, + "step": 2209 + }, + { + "epoch": 1.8299282164549973, + "grad_norm": 0.9138032793998718, + "learning_rate": 3.998314210861718e-06, + "loss": 0.6011, + "step": 2210 + }, + { + "epoch": 1.830756488128106, + "grad_norm": 0.8995500206947327, + "learning_rate": 3.9974337587566765e-06, + "loss": 0.6285, + "step": 2211 + }, + { + "epoch": 1.8315847598012147, + "grad_norm": 0.9092652797698975, + "learning_rate": 3.996553016912161e-06, + "loss": 0.6364, + "step": 2212 + }, + { + "epoch": 1.8324130314743234, + "grad_norm": 0.9085352420806885, + "learning_rate": 3.995671985498589e-06, + "loss": 0.6245, + "step": 2213 + }, + { + "epoch": 1.8332413031474324, + "grad_norm": 0.9546533226966858, + "learning_rate": 3.994790664686429e-06, + "loss": 0.6512, + "step": 2214 + }, + { + "epoch": 1.8340695748205411, + "grad_norm": 0.8606089353561401, + "learning_rate": 3.99390905464621e-06, + "loss": 0.6135, + "step": 2215 + }, + { + "epoch": 1.83489784649365, + "grad_norm": 0.9450663328170776, + "learning_rate": 3.993027155548515e-06, + "loss": 0.6193, + "step": 2216 + }, + { + "epoch": 1.8357261181667588, + "grad_norm": 0.9496508836746216, + "learning_rate": 3.992144967563984e-06, + "loss": 0.5978, + "step": 2217 + }, + { + "epoch": 1.8365543898398675, + "grad_norm": 0.8837412595748901, + "learning_rate": 3.99126249086331e-06, + "loss": 0.6104, + "step": 2218 + }, + { + "epoch": 1.8373826615129762, + "grad_norm": 0.8774031400680542, + "learning_rate": 3.9903797256172464e-06, + "loss": 0.6302, + "step": 2219 + }, + { + "epoch": 1.838210933186085, + "grad_norm": 0.8978205919265747, + "learning_rate": 3.989496671996599e-06, + "loss": 0.6411, + "step": 2220 + }, + { + "epoch": 1.8390392048591937, + "grad_norm": 0.9087855815887451, + "learning_rate": 3.98861333017223e-06, + "loss": 0.592, + "step": 2221 + }, + { + "epoch": 1.8398674765323026, + "grad_norm": 0.9040684700012207, + "learning_rate": 3.987729700315059e-06, + "loss": 0.6121, + "step": 2222 + }, + { + "epoch": 1.8406957482054114, + "grad_norm": 0.8980532288551331, + "learning_rate": 3.986845782596058e-06, + "loss": 0.6231, + "step": 2223 + }, + { + "epoch": 1.84152401987852, + "grad_norm": 0.9104854464530945, + "learning_rate": 3.9859615771862595e-06, + "loss": 0.6228, + "step": 2224 + }, + { + "epoch": 1.842352291551629, + "grad_norm": 0.9314625263214111, + "learning_rate": 3.985077084256747e-06, + "loss": 0.6136, + "step": 2225 + }, + { + "epoch": 1.8431805632247378, + "grad_norm": 0.9033634662628174, + "learning_rate": 3.984192303978663e-06, + "loss": 0.6408, + "step": 2226 + }, + { + "epoch": 1.8440088348978465, + "grad_norm": 0.9022733569145203, + "learning_rate": 3.983307236523203e-06, + "loss": 0.5954, + "step": 2227 + }, + { + "epoch": 1.8448371065709552, + "grad_norm": 0.8900256156921387, + "learning_rate": 3.982421882061621e-06, + "loss": 0.631, + "step": 2228 + }, + { + "epoch": 1.845665378244064, + "grad_norm": 0.8882872462272644, + "learning_rate": 3.981536240765223e-06, + "loss": 0.6204, + "step": 2229 + }, + { + "epoch": 1.8464936499171727, + "grad_norm": 0.894923985004425, + "learning_rate": 3.980650312805373e-06, + "loss": 0.6013, + "step": 2230 + }, + { + "epoch": 1.8473219215902816, + "grad_norm": 0.8910638093948364, + "learning_rate": 3.979764098353491e-06, + "loss": 0.6373, + "step": 2231 + }, + { + "epoch": 1.8481501932633904, + "grad_norm": 0.8952053189277649, + "learning_rate": 3.978877597581049e-06, + "loss": 0.6374, + "step": 2232 + }, + { + "epoch": 1.8489784649364993, + "grad_norm": 0.9100720882415771, + "learning_rate": 3.977990810659579e-06, + "loss": 0.6417, + "step": 2233 + }, + { + "epoch": 1.849806736609608, + "grad_norm": 0.8954150676727295, + "learning_rate": 3.9771037377606634e-06, + "loss": 0.6027, + "step": 2234 + }, + { + "epoch": 1.8506350082827168, + "grad_norm": 0.8989905118942261, + "learning_rate": 3.976216379055944e-06, + "loss": 0.6228, + "step": 2235 + }, + { + "epoch": 1.8514632799558255, + "grad_norm": 0.9060820937156677, + "learning_rate": 3.975328734717117e-06, + "loss": 0.6291, + "step": 2236 + }, + { + "epoch": 1.8522915516289342, + "grad_norm": 0.9164367318153381, + "learning_rate": 3.974440804915932e-06, + "loss": 0.6249, + "step": 2237 + }, + { + "epoch": 1.853119823302043, + "grad_norm": 0.9227840304374695, + "learning_rate": 3.973552589824196e-06, + "loss": 0.6328, + "step": 2238 + }, + { + "epoch": 1.853948094975152, + "grad_norm": 0.8612478971481323, + "learning_rate": 3.972664089613772e-06, + "loss": 0.5794, + "step": 2239 + }, + { + "epoch": 1.8547763666482606, + "grad_norm": 0.8819253444671631, + "learning_rate": 3.971775304456573e-06, + "loss": 0.6145, + "step": 2240 + }, + { + "epoch": 1.8556046383213696, + "grad_norm": 0.9216543436050415, + "learning_rate": 3.970886234524573e-06, + "loss": 0.609, + "step": 2241 + }, + { + "epoch": 1.8564329099944783, + "grad_norm": 0.9114699959754944, + "learning_rate": 3.969996879989797e-06, + "loss": 0.6105, + "step": 2242 + }, + { + "epoch": 1.857261181667587, + "grad_norm": 0.9464704990386963, + "learning_rate": 3.969107241024328e-06, + "loss": 0.6389, + "step": 2243 + }, + { + "epoch": 1.8580894533406958, + "grad_norm": 0.8921800255775452, + "learning_rate": 3.968217317800303e-06, + "loss": 0.6338, + "step": 2244 + }, + { + "epoch": 1.8589177250138045, + "grad_norm": 0.897452175617218, + "learning_rate": 3.967327110489913e-06, + "loss": 0.6248, + "step": 2245 + }, + { + "epoch": 1.8597459966869132, + "grad_norm": 0.9093567132949829, + "learning_rate": 3.9664366192654066e-06, + "loss": 0.6258, + "step": 2246 + }, + { + "epoch": 1.860574268360022, + "grad_norm": 0.894188642501831, + "learning_rate": 3.965545844299084e-06, + "loss": 0.6095, + "step": 2247 + }, + { + "epoch": 1.8614025400331309, + "grad_norm": 0.9238452911376953, + "learning_rate": 3.964654785763301e-06, + "loss": 0.6237, + "step": 2248 + }, + { + "epoch": 1.8622308117062396, + "grad_norm": 0.8715214133262634, + "learning_rate": 3.9637634438304715e-06, + "loss": 0.6153, + "step": 2249 + }, + { + "epoch": 1.8630590833793486, + "grad_norm": 0.8790848255157471, + "learning_rate": 3.962871818673061e-06, + "loss": 0.6389, + "step": 2250 + }, + { + "epoch": 1.8638873550524573, + "grad_norm": 0.9859224557876587, + "learning_rate": 3.961979910463589e-06, + "loss": 0.6075, + "step": 2251 + }, + { + "epoch": 1.864715626725566, + "grad_norm": 0.9127727150917053, + "learning_rate": 3.9610877193746335e-06, + "loss": 0.6332, + "step": 2252 + }, + { + "epoch": 1.8655438983986747, + "grad_norm": 0.9248382449150085, + "learning_rate": 3.960195245578825e-06, + "loss": 0.622, + "step": 2253 + }, + { + "epoch": 1.8663721700717835, + "grad_norm": 0.9030913710594177, + "learning_rate": 3.959302489248848e-06, + "loss": 0.6326, + "step": 2254 + }, + { + "epoch": 1.8672004417448922, + "grad_norm": 0.8922805190086365, + "learning_rate": 3.958409450557441e-06, + "loss": 0.6099, + "step": 2255 + }, + { + "epoch": 1.8680287134180011, + "grad_norm": 0.9165307879447937, + "learning_rate": 3.957516129677402e-06, + "loss": 0.6245, + "step": 2256 + }, + { + "epoch": 1.8688569850911099, + "grad_norm": 0.9445623159408569, + "learning_rate": 3.956622526781579e-06, + "loss": 0.6317, + "step": 2257 + }, + { + "epoch": 1.8696852567642188, + "grad_norm": 0.8833973407745361, + "learning_rate": 3.955728642042874e-06, + "loss": 0.6264, + "step": 2258 + }, + { + "epoch": 1.8705135284373275, + "grad_norm": 0.9512470364570618, + "learning_rate": 3.9548344756342486e-06, + "loss": 0.6137, + "step": 2259 + }, + { + "epoch": 1.8713418001104363, + "grad_norm": 0.8834348320960999, + "learning_rate": 3.9539400277287134e-06, + "loss": 0.6597, + "step": 2260 + }, + { + "epoch": 1.872170071783545, + "grad_norm": 0.8881180882453918, + "learning_rate": 3.953045298499335e-06, + "loss": 0.6502, + "step": 2261 + }, + { + "epoch": 1.8729983434566537, + "grad_norm": 0.8921699523925781, + "learning_rate": 3.952150288119239e-06, + "loss": 0.6216, + "step": 2262 + }, + { + "epoch": 1.8738266151297625, + "grad_norm": 0.91098552942276, + "learning_rate": 3.9512549967615965e-06, + "loss": 0.6286, + "step": 2263 + }, + { + "epoch": 1.8746548868028712, + "grad_norm": 0.9343264698982239, + "learning_rate": 3.950359424599642e-06, + "loss": 0.6368, + "step": 2264 + }, + { + "epoch": 1.8754831584759801, + "grad_norm": 0.9168746471405029, + "learning_rate": 3.949463571806658e-06, + "loss": 0.6355, + "step": 2265 + }, + { + "epoch": 1.8763114301490889, + "grad_norm": 0.9206068515777588, + "learning_rate": 3.948567438555986e-06, + "loss": 0.607, + "step": 2266 + }, + { + "epoch": 1.8771397018221978, + "grad_norm": 0.9262584447860718, + "learning_rate": 3.947671025021017e-06, + "loss": 0.6197, + "step": 2267 + }, + { + "epoch": 1.8779679734953065, + "grad_norm": 0.8699168562889099, + "learning_rate": 3.946774331375199e-06, + "loss": 0.6181, + "step": 2268 + }, + { + "epoch": 1.8787962451684153, + "grad_norm": 0.8892862200737, + "learning_rate": 3.9458773577920365e-06, + "loss": 0.6245, + "step": 2269 + }, + { + "epoch": 1.879624516841524, + "grad_norm": 0.9038919806480408, + "learning_rate": 3.944980104445082e-06, + "loss": 0.6209, + "step": 2270 + }, + { + "epoch": 1.8804527885146327, + "grad_norm": 0.8780872821807861, + "learning_rate": 3.9440825715079475e-06, + "loss": 0.6355, + "step": 2271 + }, + { + "epoch": 1.8812810601877414, + "grad_norm": 0.9379895925521851, + "learning_rate": 3.943184759154297e-06, + "loss": 0.6381, + "step": 2272 + }, + { + "epoch": 1.8821093318608504, + "grad_norm": 0.9112582802772522, + "learning_rate": 3.942286667557849e-06, + "loss": 0.6273, + "step": 2273 + }, + { + "epoch": 1.8829376035339591, + "grad_norm": 0.9367797374725342, + "learning_rate": 3.9413882968923754e-06, + "loss": 0.6111, + "step": 2274 + }, + { + "epoch": 1.883765875207068, + "grad_norm": 0.8923424482345581, + "learning_rate": 3.940489647331703e-06, + "loss": 0.6187, + "step": 2275 + }, + { + "epoch": 1.8845941468801768, + "grad_norm": 0.8797968029975891, + "learning_rate": 3.939590719049712e-06, + "loss": 0.603, + "step": 2276 + }, + { + "epoch": 1.8854224185532855, + "grad_norm": 0.8977354168891907, + "learning_rate": 3.938691512220336e-06, + "loss": 0.614, + "step": 2277 + }, + { + "epoch": 1.8862506902263942, + "grad_norm": 0.9207090139389038, + "learning_rate": 3.937792027017564e-06, + "loss": 0.6162, + "step": 2278 + }, + { + "epoch": 1.887078961899503, + "grad_norm": 0.8815562725067139, + "learning_rate": 3.936892263615438e-06, + "loss": 0.6315, + "step": 2279 + }, + { + "epoch": 1.8879072335726117, + "grad_norm": 0.9983104467391968, + "learning_rate": 3.935992222188054e-06, + "loss": 0.6337, + "step": 2280 + }, + { + "epoch": 1.8887355052457206, + "grad_norm": 0.8987579941749573, + "learning_rate": 3.93509190290956e-06, + "loss": 0.644, + "step": 2281 + }, + { + "epoch": 1.8895637769188294, + "grad_norm": 0.9132242798805237, + "learning_rate": 3.934191305954161e-06, + "loss": 0.6063, + "step": 2282 + }, + { + "epoch": 1.890392048591938, + "grad_norm": 0.8940858840942383, + "learning_rate": 3.933290431496114e-06, + "loss": 0.6204, + "step": 2283 + }, + { + "epoch": 1.891220320265047, + "grad_norm": 0.8862229585647583, + "learning_rate": 3.932389279709729e-06, + "loss": 0.6257, + "step": 2284 + }, + { + "epoch": 1.8920485919381558, + "grad_norm": 0.8841557502746582, + "learning_rate": 3.931487850769371e-06, + "loss": 0.6329, + "step": 2285 + }, + { + "epoch": 1.8928768636112645, + "grad_norm": 0.9075404405593872, + "learning_rate": 3.930586144849459e-06, + "loss": 0.6195, + "step": 2286 + }, + { + "epoch": 1.8937051352843732, + "grad_norm": 0.9349889755249023, + "learning_rate": 3.929684162124462e-06, + "loss": 0.6159, + "step": 2287 + }, + { + "epoch": 1.894533406957482, + "grad_norm": 0.9378470182418823, + "learning_rate": 3.928781902768908e-06, + "loss": 0.6221, + "step": 2288 + }, + { + "epoch": 1.8953616786305907, + "grad_norm": 0.8877297043800354, + "learning_rate": 3.927879366957374e-06, + "loss": 0.6359, + "step": 2289 + }, + { + "epoch": 1.8961899503036996, + "grad_norm": 0.9335882067680359, + "learning_rate": 3.926976554864495e-06, + "loss": 0.6397, + "step": 2290 + }, + { + "epoch": 1.8970182219768084, + "grad_norm": 0.8982004523277283, + "learning_rate": 3.926073466664953e-06, + "loss": 0.6303, + "step": 2291 + }, + { + "epoch": 1.8978464936499173, + "grad_norm": 0.8909676671028137, + "learning_rate": 3.92517010253349e-06, + "loss": 0.6476, + "step": 2292 + }, + { + "epoch": 1.898674765323026, + "grad_norm": 0.9127813577651978, + "learning_rate": 3.924266462644896e-06, + "loss": 0.6078, + "step": 2293 + }, + { + "epoch": 1.8995030369961348, + "grad_norm": 0.8923580050468445, + "learning_rate": 3.923362547174019e-06, + "loss": 0.6371, + "step": 2294 + }, + { + "epoch": 1.9003313086692435, + "grad_norm": 0.912458062171936, + "learning_rate": 3.922458356295757e-06, + "loss": 0.6162, + "step": 2295 + }, + { + "epoch": 1.9011595803423522, + "grad_norm": 0.8866751790046692, + "learning_rate": 3.921553890185064e-06, + "loss": 0.6011, + "step": 2296 + }, + { + "epoch": 1.901987852015461, + "grad_norm": 0.9114490151405334, + "learning_rate": 3.920649149016944e-06, + "loss": 0.6322, + "step": 2297 + }, + { + "epoch": 1.90281612368857, + "grad_norm": 0.912030816078186, + "learning_rate": 3.919744132966457e-06, + "loss": 0.6293, + "step": 2298 + }, + { + "epoch": 1.9036443953616786, + "grad_norm": 0.887301504611969, + "learning_rate": 3.918838842208715e-06, + "loss": 0.6288, + "step": 2299 + }, + { + "epoch": 1.9044726670347876, + "grad_norm": 0.9039818048477173, + "learning_rate": 3.917933276918884e-06, + "loss": 0.6154, + "step": 2300 + }, + { + "epoch": 1.9053009387078963, + "grad_norm": 0.9260784983634949, + "learning_rate": 3.917027437272182e-06, + "loss": 0.6111, + "step": 2301 + }, + { + "epoch": 1.906129210381005, + "grad_norm": 0.8966069221496582, + "learning_rate": 3.916121323443879e-06, + "loss": 0.6334, + "step": 2302 + }, + { + "epoch": 1.9069574820541138, + "grad_norm": 0.8983698487281799, + "learning_rate": 3.9152149356093e-06, + "loss": 0.6235, + "step": 2303 + }, + { + "epoch": 1.9077857537272225, + "grad_norm": 0.9150595664978027, + "learning_rate": 3.914308273943826e-06, + "loss": 0.6118, + "step": 2304 + }, + { + "epoch": 1.9086140254003312, + "grad_norm": 0.8768419027328491, + "learning_rate": 3.913401338622883e-06, + "loss": 0.5935, + "step": 2305 + }, + { + "epoch": 1.90944229707344, + "grad_norm": 0.887562096118927, + "learning_rate": 3.912494129821958e-06, + "loss": 0.6232, + "step": 2306 + }, + { + "epoch": 1.9102705687465489, + "grad_norm": 0.8916515111923218, + "learning_rate": 3.911586647716585e-06, + "loss": 0.6409, + "step": 2307 + }, + { + "epoch": 1.9110988404196576, + "grad_norm": 0.9207882881164551, + "learning_rate": 3.910678892482354e-06, + "loss": 0.6155, + "step": 2308 + }, + { + "epoch": 1.9119271120927666, + "grad_norm": 0.9300670027732849, + "learning_rate": 3.9097708642949085e-06, + "loss": 0.6389, + "step": 2309 + }, + { + "epoch": 1.9127553837658753, + "grad_norm": 0.8685980439186096, + "learning_rate": 3.908862563329941e-06, + "loss": 0.6392, + "step": 2310 + }, + { + "epoch": 1.913583655438984, + "grad_norm": 0.9150967001914978, + "learning_rate": 3.907953989763202e-06, + "loss": 0.6387, + "step": 2311 + }, + { + "epoch": 1.9144119271120927, + "grad_norm": 0.9094190001487732, + "learning_rate": 3.90704514377049e-06, + "loss": 0.6055, + "step": 2312 + }, + { + "epoch": 1.9152401987852015, + "grad_norm": 0.8863343596458435, + "learning_rate": 3.90613602552766e-06, + "loss": 0.6073, + "step": 2313 + }, + { + "epoch": 1.9160684704583102, + "grad_norm": 0.9878994822502136, + "learning_rate": 3.905226635210616e-06, + "loss": 0.613, + "step": 2314 + }, + { + "epoch": 1.9168967421314191, + "grad_norm": 0.8863592147827148, + "learning_rate": 3.904316972995318e-06, + "loss": 0.6102, + "step": 2315 + }, + { + "epoch": 1.9177250138045279, + "grad_norm": 0.8859980702400208, + "learning_rate": 3.9034070390577765e-06, + "loss": 0.6182, + "step": 2316 + }, + { + "epoch": 1.9185532854776368, + "grad_norm": 0.8953589200973511, + "learning_rate": 3.9024968335740556e-06, + "loss": 0.6212, + "step": 2317 + }, + { + "epoch": 1.9193815571507455, + "grad_norm": 0.873466968536377, + "learning_rate": 3.90158635672027e-06, + "loss": 0.616, + "step": 2318 + }, + { + "epoch": 1.9202098288238543, + "grad_norm": 0.8589543700218201, + "learning_rate": 3.900675608672592e-06, + "loss": 0.6141, + "step": 2319 + }, + { + "epoch": 1.921038100496963, + "grad_norm": 0.8717133402824402, + "learning_rate": 3.899764589607239e-06, + "loss": 0.6281, + "step": 2320 + }, + { + "epoch": 1.9218663721700717, + "grad_norm": 0.893227219581604, + "learning_rate": 3.898853299700486e-06, + "loss": 0.6139, + "step": 2321 + }, + { + "epoch": 1.9226946438431805, + "grad_norm": 0.8736181259155273, + "learning_rate": 3.89794173912866e-06, + "loss": 0.5987, + "step": 2322 + }, + { + "epoch": 1.9235229155162892, + "grad_norm": 0.8975626826286316, + "learning_rate": 3.897029908068139e-06, + "loss": 0.6417, + "step": 2323 + }, + { + "epoch": 1.9243511871893981, + "grad_norm": 0.8916004300117493, + "learning_rate": 3.896117806695355e-06, + "loss": 0.6259, + "step": 2324 + }, + { + "epoch": 1.9251794588625069, + "grad_norm": 0.8739155530929565, + "learning_rate": 3.8952054351867885e-06, + "loss": 0.6252, + "step": 2325 + }, + { + "epoch": 1.9260077305356158, + "grad_norm": 0.8667280673980713, + "learning_rate": 3.894292793718976e-06, + "loss": 0.6083, + "step": 2326 + }, + { + "epoch": 1.9268360022087245, + "grad_norm": 0.8749732375144958, + "learning_rate": 3.893379882468506e-06, + "loss": 0.6294, + "step": 2327 + }, + { + "epoch": 1.9276642738818333, + "grad_norm": 0.9085330963134766, + "learning_rate": 3.8924667016120175e-06, + "loss": 0.6101, + "step": 2328 + }, + { + "epoch": 1.928492545554942, + "grad_norm": 0.9394292235374451, + "learning_rate": 3.891553251326203e-06, + "loss": 0.6182, + "step": 2329 + }, + { + "epoch": 1.9293208172280507, + "grad_norm": 0.8756951093673706, + "learning_rate": 3.890639531787805e-06, + "loss": 0.6159, + "step": 2330 + }, + { + "epoch": 1.9301490889011594, + "grad_norm": 0.8765501379966736, + "learning_rate": 3.8897255431736225e-06, + "loss": 0.6182, + "step": 2331 + }, + { + "epoch": 1.9309773605742684, + "grad_norm": 0.907835841178894, + "learning_rate": 3.888811285660501e-06, + "loss": 0.6336, + "step": 2332 + }, + { + "epoch": 1.9318056322473771, + "grad_norm": 1.0617926120758057, + "learning_rate": 3.887896759425342e-06, + "loss": 0.6336, + "step": 2333 + }, + { + "epoch": 1.932633903920486, + "grad_norm": 0.8745515942573547, + "learning_rate": 3.886981964645099e-06, + "loss": 0.6244, + "step": 2334 + }, + { + "epoch": 1.9334621755935948, + "grad_norm": 0.8918715119361877, + "learning_rate": 3.886066901496775e-06, + "loss": 0.6289, + "step": 2335 + }, + { + "epoch": 1.9342904472667035, + "grad_norm": 0.8779492378234863, + "learning_rate": 3.885151570157426e-06, + "loss": 0.6335, + "step": 2336 + }, + { + "epoch": 1.9351187189398122, + "grad_norm": 0.8955790400505066, + "learning_rate": 3.88423597080416e-06, + "loss": 0.6125, + "step": 2337 + }, + { + "epoch": 1.935946990612921, + "grad_norm": 0.8905174136161804, + "learning_rate": 3.883320103614137e-06, + "loss": 0.5913, + "step": 2338 + }, + { + "epoch": 1.9367752622860297, + "grad_norm": 0.9078717231750488, + "learning_rate": 3.88240396876457e-06, + "loss": 0.6356, + "step": 2339 + }, + { + "epoch": 1.9376035339591386, + "grad_norm": 0.9211397171020508, + "learning_rate": 3.881487566432722e-06, + "loss": 0.6305, + "step": 2340 + }, + { + "epoch": 1.9384318056322474, + "grad_norm": 0.9131335020065308, + "learning_rate": 3.880570896795908e-06, + "loss": 0.6184, + "step": 2341 + }, + { + "epoch": 1.9392600773053563, + "grad_norm": 0.8874300718307495, + "learning_rate": 3.879653960031494e-06, + "loss": 0.6191, + "step": 2342 + }, + { + "epoch": 1.940088348978465, + "grad_norm": 0.9244474768638611, + "learning_rate": 3.8787367563169e-06, + "loss": 0.5994, + "step": 2343 + }, + { + "epoch": 1.9409166206515738, + "grad_norm": 0.8986138701438904, + "learning_rate": 3.877819285829596e-06, + "loss": 0.6544, + "step": 2344 + }, + { + "epoch": 1.9417448923246825, + "grad_norm": 0.9084672927856445, + "learning_rate": 3.876901548747104e-06, + "loss": 0.6248, + "step": 2345 + }, + { + "epoch": 1.9425731639977912, + "grad_norm": 0.9052600264549255, + "learning_rate": 3.875983545246997e-06, + "loss": 0.6112, + "step": 2346 + }, + { + "epoch": 1.9434014356709, + "grad_norm": 0.9029438495635986, + "learning_rate": 3.8750652755069025e-06, + "loss": 0.6157, + "step": 2347 + }, + { + "epoch": 1.9442297073440087, + "grad_norm": 0.882449746131897, + "learning_rate": 3.874146739704494e-06, + "loss": 0.5886, + "step": 2348 + }, + { + "epoch": 1.9450579790171176, + "grad_norm": 0.8628272414207458, + "learning_rate": 3.873227938017502e-06, + "loss": 0.608, + "step": 2349 + }, + { + "epoch": 1.9458862506902264, + "grad_norm": 0.8977845907211304, + "learning_rate": 3.872308870623704e-06, + "loss": 0.6475, + "step": 2350 + }, + { + "epoch": 1.9467145223633353, + "grad_norm": 0.8801501393318176, + "learning_rate": 3.871389537700933e-06, + "loss": 0.6246, + "step": 2351 + }, + { + "epoch": 1.947542794036444, + "grad_norm": 0.9245358109474182, + "learning_rate": 3.87046993942707e-06, + "loss": 0.6317, + "step": 2352 + }, + { + "epoch": 1.9483710657095528, + "grad_norm": 0.8959249258041382, + "learning_rate": 3.869550075980049e-06, + "loss": 0.6157, + "step": 2353 + }, + { + "epoch": 1.9491993373826615, + "grad_norm": 0.861315906047821, + "learning_rate": 3.868629947537855e-06, + "loss": 0.6251, + "step": 2354 + }, + { + "epoch": 1.9500276090557702, + "grad_norm": 0.8839966654777527, + "learning_rate": 3.867709554278524e-06, + "loss": 0.6289, + "step": 2355 + }, + { + "epoch": 1.950855880728879, + "grad_norm": 0.8878759741783142, + "learning_rate": 3.8667888963801445e-06, + "loss": 0.6113, + "step": 2356 + }, + { + "epoch": 1.951684152401988, + "grad_norm": 0.8725037574768066, + "learning_rate": 3.865867974020854e-06, + "loss": 0.6324, + "step": 2357 + }, + { + "epoch": 1.9525124240750966, + "grad_norm": 0.9059397578239441, + "learning_rate": 3.8649467873788436e-06, + "loss": 0.5981, + "step": 2358 + }, + { + "epoch": 1.9533406957482056, + "grad_norm": 0.8883296847343445, + "learning_rate": 3.8640253366323535e-06, + "loss": 0.598, + "step": 2359 + }, + { + "epoch": 1.9541689674213143, + "grad_norm": 0.8987225890159607, + "learning_rate": 3.863103621959676e-06, + "loss": 0.6253, + "step": 2360 + }, + { + "epoch": 1.954997239094423, + "grad_norm": 0.8977320194244385, + "learning_rate": 3.862181643539155e-06, + "loss": 0.6424, + "step": 2361 + }, + { + "epoch": 1.9558255107675317, + "grad_norm": 0.9267589449882507, + "learning_rate": 3.861259401549183e-06, + "loss": 0.647, + "step": 2362 + }, + { + "epoch": 1.9566537824406405, + "grad_norm": 0.9297959804534912, + "learning_rate": 3.860336896168207e-06, + "loss": 0.6448, + "step": 2363 + }, + { + "epoch": 1.9574820541137492, + "grad_norm": 0.9627453088760376, + "learning_rate": 3.859414127574722e-06, + "loss": 0.6154, + "step": 2364 + }, + { + "epoch": 1.958310325786858, + "grad_norm": 0.8945866227149963, + "learning_rate": 3.858491095947275e-06, + "loss": 0.6201, + "step": 2365 + }, + { + "epoch": 1.9591385974599669, + "grad_norm": 0.8859817981719971, + "learning_rate": 3.857567801464466e-06, + "loss": 0.6171, + "step": 2366 + }, + { + "epoch": 1.9599668691330756, + "grad_norm": 0.9104642271995544, + "learning_rate": 3.8566442443049425e-06, + "loss": 0.6273, + "step": 2367 + }, + { + "epoch": 1.9607951408061846, + "grad_norm": 0.8821743726730347, + "learning_rate": 3.855720424647403e-06, + "loss": 0.5993, + "step": 2368 + }, + { + "epoch": 1.9616234124792933, + "grad_norm": 0.903322160243988, + "learning_rate": 3.854796342670601e-06, + "loss": 0.6471, + "step": 2369 + }, + { + "epoch": 1.962451684152402, + "grad_norm": 0.8744645714759827, + "learning_rate": 3.853871998553334e-06, + "loss": 0.6444, + "step": 2370 + }, + { + "epoch": 1.9632799558255107, + "grad_norm": 0.9538072347640991, + "learning_rate": 3.852947392474457e-06, + "loss": 0.656, + "step": 2371 + }, + { + "epoch": 1.9641082274986195, + "grad_norm": 0.875407874584198, + "learning_rate": 3.852022524612872e-06, + "loss": 0.6479, + "step": 2372 + }, + { + "epoch": 1.9649364991717282, + "grad_norm": 0.8908282518386841, + "learning_rate": 3.8510973951475305e-06, + "loss": 0.6262, + "step": 2373 + }, + { + "epoch": 1.9657647708448371, + "grad_norm": 0.8698679208755493, + "learning_rate": 3.850172004257438e-06, + "loss": 0.6431, + "step": 2374 + }, + { + "epoch": 1.9665930425179459, + "grad_norm": 0.8997004628181458, + "learning_rate": 3.84924635212165e-06, + "loss": 0.6349, + "step": 2375 + }, + { + "epoch": 1.9674213141910548, + "grad_norm": 0.8815768957138062, + "learning_rate": 3.8483204389192695e-06, + "loss": 0.6449, + "step": 2376 + }, + { + "epoch": 1.9682495858641635, + "grad_norm": 0.9148927927017212, + "learning_rate": 3.847394264829453e-06, + "loss": 0.6106, + "step": 2377 + }, + { + "epoch": 1.9690778575372723, + "grad_norm": 0.8647779226303101, + "learning_rate": 3.846467830031407e-06, + "loss": 0.5954, + "step": 2378 + }, + { + "epoch": 1.969906129210381, + "grad_norm": 0.8877570033073425, + "learning_rate": 3.845541134704387e-06, + "loss": 0.6006, + "step": 2379 + }, + { + "epoch": 1.9707344008834897, + "grad_norm": 0.899742603302002, + "learning_rate": 3.844614179027701e-06, + "loss": 0.6069, + "step": 2380 + }, + { + "epoch": 1.9715626725565985, + "grad_norm": 0.9085001349449158, + "learning_rate": 3.843686963180705e-06, + "loss": 0.6332, + "step": 2381 + }, + { + "epoch": 1.9723909442297074, + "grad_norm": 0.8954899311065674, + "learning_rate": 3.842759487342808e-06, + "loss": 0.6186, + "step": 2382 + }, + { + "epoch": 1.9732192159028161, + "grad_norm": 0.9011021256446838, + "learning_rate": 3.841831751693466e-06, + "loss": 0.6177, + "step": 2383 + }, + { + "epoch": 1.9740474875759249, + "grad_norm": 0.8824598789215088, + "learning_rate": 3.8409037564121895e-06, + "loss": 0.6392, + "step": 2384 + }, + { + "epoch": 1.9748757592490338, + "grad_norm": 0.8936701416969299, + "learning_rate": 3.839975501678536e-06, + "loss": 0.6508, + "step": 2385 + }, + { + "epoch": 1.9757040309221425, + "grad_norm": 0.9009436368942261, + "learning_rate": 3.839046987672113e-06, + "loss": 0.6169, + "step": 2386 + }, + { + "epoch": 1.9765323025952513, + "grad_norm": 0.8839378356933594, + "learning_rate": 3.8381182145725815e-06, + "loss": 0.6107, + "step": 2387 + }, + { + "epoch": 1.97736057426836, + "grad_norm": 0.9240637421607971, + "learning_rate": 3.837189182559649e-06, + "loss": 0.6083, + "step": 2388 + }, + { + "epoch": 1.9781888459414687, + "grad_norm": 0.881908118724823, + "learning_rate": 3.836259891813075e-06, + "loss": 0.5995, + "step": 2389 + }, + { + "epoch": 1.9790171176145774, + "grad_norm": 0.8754116296768188, + "learning_rate": 3.835330342512667e-06, + "loss": 0.6111, + "step": 2390 + }, + { + "epoch": 1.9798453892876864, + "grad_norm": 0.9713384509086609, + "learning_rate": 3.834400534838287e-06, + "loss": 0.641, + "step": 2391 + }, + { + "epoch": 1.9806736609607951, + "grad_norm": 0.9263403415679932, + "learning_rate": 3.833470468969842e-06, + "loss": 0.6254, + "step": 2392 + }, + { + "epoch": 1.981501932633904, + "grad_norm": 0.9033388495445251, + "learning_rate": 3.832540145087291e-06, + "loss": 0.6049, + "step": 2393 + }, + { + "epoch": 1.9823302043070128, + "grad_norm": 0.8710628151893616, + "learning_rate": 3.8316095633706455e-06, + "loss": 0.5999, + "step": 2394 + }, + { + "epoch": 1.9831584759801215, + "grad_norm": 0.9053885340690613, + "learning_rate": 3.830678723999961e-06, + "loss": 0.6393, + "step": 2395 + }, + { + "epoch": 1.9839867476532302, + "grad_norm": 0.9075314998626709, + "learning_rate": 3.829747627155348e-06, + "loss": 0.6188, + "step": 2396 + }, + { + "epoch": 1.984815019326339, + "grad_norm": 0.9151870012283325, + "learning_rate": 3.828816273016966e-06, + "loss": 0.625, + "step": 2397 + }, + { + "epoch": 1.9856432909994477, + "grad_norm": 0.903843104839325, + "learning_rate": 3.827884661765022e-06, + "loss": 0.618, + "step": 2398 + }, + { + "epoch": 1.9864715626725566, + "grad_norm": 0.875468373298645, + "learning_rate": 3.826952793579774e-06, + "loss": 0.6121, + "step": 2399 + }, + { + "epoch": 1.9872998343456654, + "grad_norm": 0.8908163905143738, + "learning_rate": 3.8260206686415295e-06, + "loss": 0.6292, + "step": 2400 + }, + { + "epoch": 1.9881281060187743, + "grad_norm": 0.8985518217086792, + "learning_rate": 3.825088287130647e-06, + "loss": 0.6356, + "step": 2401 + }, + { + "epoch": 1.988956377691883, + "grad_norm": 0.9277573227882385, + "learning_rate": 3.8241556492275335e-06, + "loss": 0.6424, + "step": 2402 + }, + { + "epoch": 1.9897846493649918, + "grad_norm": 0.9075362682342529, + "learning_rate": 3.823222755112646e-06, + "loss": 0.623, + "step": 2403 + }, + { + "epoch": 1.9906129210381005, + "grad_norm": 0.9038227796554565, + "learning_rate": 3.822289604966489e-06, + "loss": 0.6195, + "step": 2404 + }, + { + "epoch": 1.9914411927112092, + "grad_norm": 0.8869004845619202, + "learning_rate": 3.82135619896962e-06, + "loss": 0.6034, + "step": 2405 + }, + { + "epoch": 1.992269464384318, + "grad_norm": 0.9017077684402466, + "learning_rate": 3.820422537302645e-06, + "loss": 0.6469, + "step": 2406 + }, + { + "epoch": 1.9930977360574267, + "grad_norm": 0.8785225749015808, + "learning_rate": 3.819488620146218e-06, + "loss": 0.6327, + "step": 2407 + }, + { + "epoch": 1.9939260077305356, + "grad_norm": 1.0367780923843384, + "learning_rate": 3.818554447681041e-06, + "loss": 0.6259, + "step": 2408 + }, + { + "epoch": 1.9947542794036444, + "grad_norm": 0.9392277002334595, + "learning_rate": 3.817620020087872e-06, + "loss": 0.6393, + "step": 2409 + }, + { + "epoch": 1.9955825510767533, + "grad_norm": 0.8935515880584717, + "learning_rate": 3.816685337547511e-06, + "loss": 0.6153, + "step": 2410 + }, + { + "epoch": 1.996410822749862, + "grad_norm": 0.8895331025123596, + "learning_rate": 3.815750400240811e-06, + "loss": 0.5978, + "step": 2411 + }, + { + "epoch": 1.9972390944229708, + "grad_norm": 0.8733546733856201, + "learning_rate": 3.814815208348673e-06, + "loss": 0.6254, + "step": 2412 + }, + { + "epoch": 1.9980673660960795, + "grad_norm": 0.9177356958389282, + "learning_rate": 3.8138797620520507e-06, + "loss": 0.631, + "step": 2413 + }, + { + "epoch": 1.9988956377691882, + "grad_norm": 0.9259604811668396, + "learning_rate": 3.8129440615319414e-06, + "loss": 0.6138, + "step": 2414 + } + ], + "logging_steps": 1, + "max_steps": 7242, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 1207, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 6.096168561514906e+18, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-2414/training_args.bin b/checkpoint-2414/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..7017eeafdf1f43b26668f0e0e5710ca52345a005 --- /dev/null +++ b/checkpoint-2414/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fdc878d2541c09353dd68f4f8a2464140b28fd84e1bc48b59914a6e84328e85e +size 8056 diff --git a/checkpoint-2414/zero_to_fp32.py b/checkpoint-2414/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-2414/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-3621/README.md b/checkpoint-3621/README.md new file mode 100644 index 0000000000000000000000000000000000000000..049d467664ca6172b7ffbe6ba60b3eac7479cac4 --- /dev/null +++ b/checkpoint-3621/README.md @@ -0,0 +1,202 @@ +--- +base_model: meta-llama/Llama-3.1-8B +library_name: peft +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.14.0 \ No newline at end of file diff --git a/checkpoint-3621/adapter_config.json b/checkpoint-3621/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d6fca8a067dce19d599e7f1e24bea8019968fe0f --- /dev/null +++ b/checkpoint-3621/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.1-8B", + "bias": "none", + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "up_proj", + "q_proj", + "k_proj", + "o_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-3621/adapter_model.safetensors b/checkpoint-3621/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..57799cbbc3074f3f349b0b75e84483c03bf2c8aa --- /dev/null +++ b/checkpoint-3621/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9bdc5f0833fbce623850fbcfbfa1c4fcaabf3967308a279bdd43b99ad8ecda9e +size 3443586272 diff --git a/checkpoint-3621/global_step3619/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-3621/global_step3619/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..310f085c4ae4633dd11dbf5f6f125ce5093ef632 --- /dev/null +++ b/checkpoint-3621/global_step3619/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7f6ef52da6b29be92221f6b7cb22d8c8513a078190e8b08fc63e2ed4b8d70509 +size 20661195036 diff --git a/checkpoint-3621/global_step3619/mp_rank_00_model_states.pt b/checkpoint-3621/global_step3619/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..6711a91e97df8eef115285e6339b5a1d7f262789 --- /dev/null +++ b/checkpoint-3621/global_step3619/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ef4d7b32389656d2d26dce5a43bae5d31ecaba2ac67ca40a9e42326cb6d41a88 +size 3555326777 diff --git a/checkpoint-3621/latest b/checkpoint-3621/latest new file mode 100644 index 0000000000000000000000000000000000000000..e19502e7f3e77d1c91d2782dd4329f839c0f1053 --- /dev/null +++ b/checkpoint-3621/latest @@ -0,0 +1 @@ +global_step3619 \ No newline at end of file diff --git a/checkpoint-3621/rng_state.pth b/checkpoint-3621/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..de07ec49cf8ab449b14d17252b6b90804c8ecafd --- /dev/null +++ b/checkpoint-3621/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cfe32979334cdb5ee18dbe27fcf1df087869d13a7dadebeb2b733f20e4382c17 +size 14244 diff --git a/checkpoint-3621/scheduler.pt b/checkpoint-3621/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..d5413bc35b22af21c9d6b013a5ea47f72a733743 --- /dev/null +++ b/checkpoint-3621/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:37a46319320818039d5604ab94bc908f651f3f274c96df7747a1935f836fef1b +size 1064 diff --git a/checkpoint-3621/special_tokens_map.json b/checkpoint-3621/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..e5b39b6305d89284b04934011c68dbb26bf588ca --- /dev/null +++ b/checkpoint-3621/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-3621/tokenizer.json b/checkpoint-3621/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-3621/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-3621/tokenizer_config.json b/checkpoint-3621/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..81dd14db6632ad5b35b9d447732e37ac074873a5 --- /dev/null +++ b/checkpoint-3621/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +} diff --git a/checkpoint-3621/trainer_state.json b/checkpoint-3621/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..8049b256cda91e764b551c804cebc581e924f1e0 --- /dev/null +++ b/checkpoint-3621/trainer_state.json @@ -0,0 +1,25380 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.9980673660960795, + "eval_steps": 500, + "global_step": 3621, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0008282716731087797, + "grad_norm": 2.4795358180999756, + "learning_rate": 5.0000000000000004e-08, + "loss": 1.1215, + "step": 1 + }, + { + "epoch": 0.0016565433462175593, + "grad_norm": 2.6060028076171875, + "learning_rate": 1.0000000000000001e-07, + "loss": 1.1219, + "step": 2 + }, + { + "epoch": 0.002484815019326339, + "grad_norm": 2.750868082046509, + "learning_rate": 1.5000000000000002e-07, + "loss": 1.125, + "step": 3 + }, + { + "epoch": 0.0033130866924351186, + "grad_norm": 2.7321369647979736, + "learning_rate": 2.0000000000000002e-07, + "loss": 1.1164, + "step": 4 + }, + { + "epoch": 0.0041413583655438985, + "grad_norm": 2.8193929195404053, + "learning_rate": 2.5000000000000004e-07, + "loss": 1.1263, + "step": 5 + }, + { + "epoch": 0.004969630038652678, + "grad_norm": 2.648301839828491, + "learning_rate": 3.0000000000000004e-07, + "loss": 1.1092, + "step": 6 + }, + { + "epoch": 0.005797901711761457, + "grad_norm": 2.751993179321289, + "learning_rate": 3.5000000000000004e-07, + "loss": 1.0997, + "step": 7 + }, + { + "epoch": 0.006626173384870237, + "grad_norm": 2.592381000518799, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.112, + "step": 8 + }, + { + "epoch": 0.007454445057979017, + "grad_norm": 2.3135671615600586, + "learning_rate": 4.5000000000000003e-07, + "loss": 1.1045, + "step": 9 + }, + { + "epoch": 0.008282716731087797, + "grad_norm": 2.60491943359375, + "learning_rate": 5.000000000000001e-07, + "loss": 1.0956, + "step": 10 + }, + { + "epoch": 0.009110988404196576, + "grad_norm": 2.2435271739959717, + "learning_rate": 5.5e-07, + "loss": 1.1241, + "step": 11 + }, + { + "epoch": 0.009939260077305357, + "grad_norm": 2.4412319660186768, + "learning_rate": 6.000000000000001e-07, + "loss": 1.11, + "step": 12 + }, + { + "epoch": 0.010767531750414136, + "grad_norm": 2.2075185775756836, + "learning_rate": 6.5e-07, + "loss": 1.0878, + "step": 13 + }, + { + "epoch": 0.011595803423522915, + "grad_norm": 2.101902723312378, + "learning_rate": 7.000000000000001e-07, + "loss": 1.0863, + "step": 14 + }, + { + "epoch": 0.012424075096631695, + "grad_norm": 2.134535074234009, + "learning_rate": 7.5e-07, + "loss": 1.0735, + "step": 15 + }, + { + "epoch": 0.013252346769740474, + "grad_norm": 2.000504493713379, + "learning_rate": 8.000000000000001e-07, + "loss": 1.0875, + "step": 16 + }, + { + "epoch": 0.014080618442849255, + "grad_norm": 1.9191882610321045, + "learning_rate": 8.500000000000001e-07, + "loss": 1.1039, + "step": 17 + }, + { + "epoch": 0.014908890115958034, + "grad_norm": 1.7732031345367432, + "learning_rate": 9.000000000000001e-07, + "loss": 1.0819, + "step": 18 + }, + { + "epoch": 0.015737161789066815, + "grad_norm": 1.7775251865386963, + "learning_rate": 9.500000000000001e-07, + "loss": 1.0936, + "step": 19 + }, + { + "epoch": 0.016565433462175594, + "grad_norm": 1.6648598909378052, + "learning_rate": 1.0000000000000002e-06, + "loss": 1.0881, + "step": 20 + }, + { + "epoch": 0.017393705135284373, + "grad_norm": 1.5726720094680786, + "learning_rate": 1.0500000000000001e-06, + "loss": 1.07, + "step": 21 + }, + { + "epoch": 0.018221976808393152, + "grad_norm": 1.561417818069458, + "learning_rate": 1.1e-06, + "loss": 1.0668, + "step": 22 + }, + { + "epoch": 0.019050248481501934, + "grad_norm": 1.4953842163085938, + "learning_rate": 1.1500000000000002e-06, + "loss": 1.054, + "step": 23 + }, + { + "epoch": 0.019878520154610713, + "grad_norm": 1.4247829914093018, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.0363, + "step": 24 + }, + { + "epoch": 0.020706791827719492, + "grad_norm": 1.3483426570892334, + "learning_rate": 1.25e-06, + "loss": 1.0143, + "step": 25 + }, + { + "epoch": 0.02153506350082827, + "grad_norm": 1.3412944078445435, + "learning_rate": 1.3e-06, + "loss": 1.025, + "step": 26 + }, + { + "epoch": 0.02236333517393705, + "grad_norm": 1.2834237813949585, + "learning_rate": 1.3500000000000002e-06, + "loss": 1.0227, + "step": 27 + }, + { + "epoch": 0.02319160684704583, + "grad_norm": 1.2128241062164307, + "learning_rate": 1.4000000000000001e-06, + "loss": 1.0161, + "step": 28 + }, + { + "epoch": 0.024019878520154612, + "grad_norm": 1.1665573120117188, + "learning_rate": 1.45e-06, + "loss": 1.0034, + "step": 29 + }, + { + "epoch": 0.02484815019326339, + "grad_norm": 1.1575899124145508, + "learning_rate": 1.5e-06, + "loss": 0.9968, + "step": 30 + }, + { + "epoch": 0.02567642186637217, + "grad_norm": 1.195304274559021, + "learning_rate": 1.5500000000000002e-06, + "loss": 1.0176, + "step": 31 + }, + { + "epoch": 0.02650469353948095, + "grad_norm": 1.1321678161621094, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.0113, + "step": 32 + }, + { + "epoch": 0.027332965212589728, + "grad_norm": 1.1076338291168213, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.9971, + "step": 33 + }, + { + "epoch": 0.02816123688569851, + "grad_norm": 1.0770610570907593, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.9994, + "step": 34 + }, + { + "epoch": 0.02898950855880729, + "grad_norm": 1.0767767429351807, + "learning_rate": 1.75e-06, + "loss": 0.978, + "step": 35 + }, + { + "epoch": 0.02981778023191607, + "grad_norm": 1.054800033569336, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.9829, + "step": 36 + }, + { + "epoch": 0.030646051905024847, + "grad_norm": 1.031015157699585, + "learning_rate": 1.85e-06, + "loss": 0.97, + "step": 37 + }, + { + "epoch": 0.03147432357813363, + "grad_norm": 1.0182578563690186, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.9549, + "step": 38 + }, + { + "epoch": 0.032302595251242405, + "grad_norm": 0.9796578288078308, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.9762, + "step": 39 + }, + { + "epoch": 0.03313086692435119, + "grad_norm": 0.9956578016281128, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.9693, + "step": 40 + }, + { + "epoch": 0.03395913859745997, + "grad_norm": 0.9383260011672974, + "learning_rate": 2.05e-06, + "loss": 0.9556, + "step": 41 + }, + { + "epoch": 0.034787410270568746, + "grad_norm": 0.9372819066047668, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.9419, + "step": 42 + }, + { + "epoch": 0.03561568194367753, + "grad_norm": 0.9334004521369934, + "learning_rate": 2.15e-06, + "loss": 0.9587, + "step": 43 + }, + { + "epoch": 0.036443953616786304, + "grad_norm": 0.8916004300117493, + "learning_rate": 2.2e-06, + "loss": 0.9227, + "step": 44 + }, + { + "epoch": 0.037272225289895086, + "grad_norm": 0.8698756694793701, + "learning_rate": 2.25e-06, + "loss": 0.9275, + "step": 45 + }, + { + "epoch": 0.03810049696300387, + "grad_norm": 0.8775100708007812, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.941, + "step": 46 + }, + { + "epoch": 0.038928768636112644, + "grad_norm": 0.8763384222984314, + "learning_rate": 2.35e-06, + "loss": 0.9253, + "step": 47 + }, + { + "epoch": 0.03975704030922143, + "grad_norm": 0.8534179925918579, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.9327, + "step": 48 + }, + { + "epoch": 0.0405853119823302, + "grad_norm": 0.8432146906852722, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.9236, + "step": 49 + }, + { + "epoch": 0.041413583655438985, + "grad_norm": 0.9899975061416626, + "learning_rate": 2.5e-06, + "loss": 0.8839, + "step": 50 + }, + { + "epoch": 0.04224185532854776, + "grad_norm": 0.8370104432106018, + "learning_rate": 2.55e-06, + "loss": 0.906, + "step": 51 + }, + { + "epoch": 0.04307012700165654, + "grad_norm": 0.8207446336746216, + "learning_rate": 2.6e-06, + "loss": 0.8755, + "step": 52 + }, + { + "epoch": 0.043898398674765325, + "grad_norm": 0.805584728717804, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.9049, + "step": 53 + }, + { + "epoch": 0.0447266703478741, + "grad_norm": 0.8162860870361328, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.8993, + "step": 54 + }, + { + "epoch": 0.04555494202098288, + "grad_norm": 0.8151668906211853, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.9282, + "step": 55 + }, + { + "epoch": 0.04638321369409166, + "grad_norm": 0.7996147274971008, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.8946, + "step": 56 + }, + { + "epoch": 0.04721148536720044, + "grad_norm": 0.8223204016685486, + "learning_rate": 2.85e-06, + "loss": 0.897, + "step": 57 + }, + { + "epoch": 0.048039757040309224, + "grad_norm": 0.8117988109588623, + "learning_rate": 2.9e-06, + "loss": 0.8845, + "step": 58 + }, + { + "epoch": 0.048868028713418, + "grad_norm": 0.7902786731719971, + "learning_rate": 2.95e-06, + "loss": 0.8846, + "step": 59 + }, + { + "epoch": 0.04969630038652678, + "grad_norm": 0.804192304611206, + "learning_rate": 3e-06, + "loss": 0.8722, + "step": 60 + }, + { + "epoch": 0.05052457205963556, + "grad_norm": 0.7852040529251099, + "learning_rate": 3.05e-06, + "loss": 0.8821, + "step": 61 + }, + { + "epoch": 0.05135284373274434, + "grad_norm": 0.8045080304145813, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.8661, + "step": 62 + }, + { + "epoch": 0.05218111540585312, + "grad_norm": 0.7920243144035339, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.8712, + "step": 63 + }, + { + "epoch": 0.0530093870789619, + "grad_norm": 0.7601844072341919, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.8792, + "step": 64 + }, + { + "epoch": 0.05383765875207068, + "grad_norm": 0.7833033800125122, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.8811, + "step": 65 + }, + { + "epoch": 0.054665930425179456, + "grad_norm": 0.7691664099693298, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.8701, + "step": 66 + }, + { + "epoch": 0.05549420209828824, + "grad_norm": 0.7845340967178345, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.8929, + "step": 67 + }, + { + "epoch": 0.05632247377139702, + "grad_norm": 0.9076205492019653, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.8352, + "step": 68 + }, + { + "epoch": 0.057150745444505796, + "grad_norm": 0.7614927887916565, + "learning_rate": 3.45e-06, + "loss": 0.8766, + "step": 69 + }, + { + "epoch": 0.05797901711761458, + "grad_norm": 0.7722934484481812, + "learning_rate": 3.5e-06, + "loss": 0.8666, + "step": 70 + }, + { + "epoch": 0.058807288790723354, + "grad_norm": 0.7603748440742493, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.84, + "step": 71 + }, + { + "epoch": 0.05963556046383214, + "grad_norm": 0.7563591003417969, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.8784, + "step": 72 + }, + { + "epoch": 0.06046383213694092, + "grad_norm": 0.7835114002227783, + "learning_rate": 3.65e-06, + "loss": 0.8516, + "step": 73 + }, + { + "epoch": 0.061292103810049695, + "grad_norm": 0.7760982513427734, + "learning_rate": 3.7e-06, + "loss": 0.8265, + "step": 74 + }, + { + "epoch": 0.06212037548315848, + "grad_norm": 0.7978456020355225, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.8564, + "step": 75 + }, + { + "epoch": 0.06294864715626726, + "grad_norm": 0.7940747141838074, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.8635, + "step": 76 + }, + { + "epoch": 0.06377691882937604, + "grad_norm": 0.8055519461631775, + "learning_rate": 3.85e-06, + "loss": 0.8445, + "step": 77 + }, + { + "epoch": 0.06460519050248481, + "grad_norm": 0.7532406449317932, + "learning_rate": 3.900000000000001e-06, + "loss": 0.8772, + "step": 78 + }, + { + "epoch": 0.06543346217559359, + "grad_norm": 0.763933539390564, + "learning_rate": 3.95e-06, + "loss": 0.8574, + "step": 79 + }, + { + "epoch": 0.06626173384870238, + "grad_norm": 0.7833206057548523, + "learning_rate": 4.000000000000001e-06, + "loss": 0.8643, + "step": 80 + }, + { + "epoch": 0.06709000552181116, + "grad_norm": 0.7825527787208557, + "learning_rate": 4.05e-06, + "loss": 0.8518, + "step": 81 + }, + { + "epoch": 0.06791827719491994, + "grad_norm": 0.7749581336975098, + "learning_rate": 4.1e-06, + "loss": 0.8146, + "step": 82 + }, + { + "epoch": 0.06874654886802871, + "grad_norm": 0.8117144703865051, + "learning_rate": 4.15e-06, + "loss": 0.8719, + "step": 83 + }, + { + "epoch": 0.06957482054113749, + "grad_norm": 0.8440507054328918, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.8665, + "step": 84 + }, + { + "epoch": 0.07040309221424627, + "grad_norm": 0.7941984534263611, + "learning_rate": 4.25e-06, + "loss": 0.8492, + "step": 85 + }, + { + "epoch": 0.07123136388735506, + "grad_norm": 0.8301989436149597, + "learning_rate": 4.3e-06, + "loss": 0.8544, + "step": 86 + }, + { + "epoch": 0.07205963556046384, + "grad_norm": 0.757798969745636, + "learning_rate": 4.350000000000001e-06, + "loss": 0.8426, + "step": 87 + }, + { + "epoch": 0.07288790723357261, + "grad_norm": 0.7817673683166504, + "learning_rate": 4.4e-06, + "loss": 0.8554, + "step": 88 + }, + { + "epoch": 0.07371617890668139, + "grad_norm": 0.7609818577766418, + "learning_rate": 4.450000000000001e-06, + "loss": 0.8473, + "step": 89 + }, + { + "epoch": 0.07454445057979017, + "grad_norm": 0.8146091103553772, + "learning_rate": 4.5e-06, + "loss": 0.8317, + "step": 90 + }, + { + "epoch": 0.07537272225289895, + "grad_norm": 0.8058239817619324, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.8411, + "step": 91 + }, + { + "epoch": 0.07620099392600774, + "grad_norm": 0.7700346112251282, + "learning_rate": 4.600000000000001e-06, + "loss": 0.815, + "step": 92 + }, + { + "epoch": 0.0770292655991165, + "grad_norm": 0.7949067950248718, + "learning_rate": 4.65e-06, + "loss": 0.8512, + "step": 93 + }, + { + "epoch": 0.07785753727222529, + "grad_norm": 0.8112176060676575, + "learning_rate": 4.7e-06, + "loss": 0.8398, + "step": 94 + }, + { + "epoch": 0.07868580894533407, + "grad_norm": 0.7990434765815735, + "learning_rate": 4.75e-06, + "loss": 0.8276, + "step": 95 + }, + { + "epoch": 0.07951408061844285, + "grad_norm": 0.7942920923233032, + "learning_rate": 4.800000000000001e-06, + "loss": 0.8333, + "step": 96 + }, + { + "epoch": 0.08034235229155164, + "grad_norm": 0.8005534410476685, + "learning_rate": 4.85e-06, + "loss": 0.8453, + "step": 97 + }, + { + "epoch": 0.0811706239646604, + "grad_norm": 0.8103216290473938, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8303, + "step": 98 + }, + { + "epoch": 0.08199889563776919, + "grad_norm": 0.8217238187789917, + "learning_rate": 4.95e-06, + "loss": 0.8204, + "step": 99 + }, + { + "epoch": 0.08282716731087797, + "grad_norm": 0.7958654761314392, + "learning_rate": 5e-06, + "loss": 0.8237, + "step": 100 + }, + { + "epoch": 0.08365543898398675, + "grad_norm": 0.8631463050842285, + "learning_rate": 4.999999758136652e-06, + "loss": 0.821, + "step": 101 + }, + { + "epoch": 0.08448371065709552, + "grad_norm": 0.7666735053062439, + "learning_rate": 4.999999032546657e-06, + "loss": 0.8307, + "step": 102 + }, + { + "epoch": 0.0853119823302043, + "grad_norm": 0.7709226012229919, + "learning_rate": 4.999997823230153e-06, + "loss": 0.8299, + "step": 103 + }, + { + "epoch": 0.08614025400331309, + "grad_norm": 0.8020651340484619, + "learning_rate": 4.9999961301873736e-06, + "loss": 0.8075, + "step": 104 + }, + { + "epoch": 0.08696852567642187, + "grad_norm": 0.8019537329673767, + "learning_rate": 4.999993953418649e-06, + "loss": 0.8615, + "step": 105 + }, + { + "epoch": 0.08779679734953065, + "grad_norm": 0.7904282808303833, + "learning_rate": 4.999991292924399e-06, + "loss": 0.8353, + "step": 106 + }, + { + "epoch": 0.08862506902263942, + "grad_norm": 0.7854570150375366, + "learning_rate": 4.999988148705138e-06, + "loss": 0.8154, + "step": 107 + }, + { + "epoch": 0.0894533406957482, + "grad_norm": 0.8044222593307495, + "learning_rate": 4.999984520761475e-06, + "loss": 0.8117, + "step": 108 + }, + { + "epoch": 0.09028161236885698, + "grad_norm": 0.7986987829208374, + "learning_rate": 4.999980409094112e-06, + "loss": 0.8407, + "step": 109 + }, + { + "epoch": 0.09110988404196577, + "grad_norm": 0.7976786494255066, + "learning_rate": 4.999975813703844e-06, + "loss": 0.7871, + "step": 110 + }, + { + "epoch": 0.09193815571507455, + "grad_norm": 0.8330315351486206, + "learning_rate": 4.9999707345915605e-06, + "loss": 0.8129, + "step": 111 + }, + { + "epoch": 0.09276642738818332, + "grad_norm": 0.8350826501846313, + "learning_rate": 4.999965171758245e-06, + "loss": 0.8144, + "step": 112 + }, + { + "epoch": 0.0935946990612921, + "grad_norm": 0.8054768443107605, + "learning_rate": 4.999959125204973e-06, + "loss": 0.809, + "step": 113 + }, + { + "epoch": 0.09442297073440088, + "grad_norm": 0.8101645708084106, + "learning_rate": 4.999952594932914e-06, + "loss": 0.8081, + "step": 114 + }, + { + "epoch": 0.09525124240750966, + "grad_norm": 0.905028760433197, + "learning_rate": 4.999945580943332e-06, + "loss": 0.8251, + "step": 115 + }, + { + "epoch": 0.09607951408061845, + "grad_norm": 0.8529418706893921, + "learning_rate": 4.999938083237585e-06, + "loss": 0.7984, + "step": 116 + }, + { + "epoch": 0.09690778575372722, + "grad_norm": 0.8047451376914978, + "learning_rate": 4.9999301018171225e-06, + "loss": 0.8222, + "step": 117 + }, + { + "epoch": 0.097736057426836, + "grad_norm": 0.8073400259017944, + "learning_rate": 4.99992163668349e-06, + "loss": 0.8251, + "step": 118 + }, + { + "epoch": 0.09856432909994478, + "grad_norm": 0.8470693826675415, + "learning_rate": 4.999912687838324e-06, + "loss": 0.8291, + "step": 119 + }, + { + "epoch": 0.09939260077305356, + "grad_norm": 0.8261650800704956, + "learning_rate": 4.999903255283357e-06, + "loss": 0.808, + "step": 120 + }, + { + "epoch": 0.10022087244616235, + "grad_norm": 0.8421704769134521, + "learning_rate": 4.999893339020414e-06, + "loss": 0.8295, + "step": 121 + }, + { + "epoch": 0.10104914411927111, + "grad_norm": 0.8244299292564392, + "learning_rate": 4.9998829390514134e-06, + "loss": 0.8215, + "step": 122 + }, + { + "epoch": 0.1018774157923799, + "grad_norm": 0.8365658521652222, + "learning_rate": 4.999872055378368e-06, + "loss": 0.8245, + "step": 123 + }, + { + "epoch": 0.10270568746548868, + "grad_norm": 0.9089561104774475, + "learning_rate": 4.999860688003384e-06, + "loss": 0.8125, + "step": 124 + }, + { + "epoch": 0.10353395913859746, + "grad_norm": 0.8784728646278381, + "learning_rate": 4.9998488369286595e-06, + "loss": 0.8321, + "step": 125 + }, + { + "epoch": 0.10436223081170624, + "grad_norm": 0.8596073389053345, + "learning_rate": 4.999836502156487e-06, + "loss": 0.8333, + "step": 126 + }, + { + "epoch": 0.10519050248481501, + "grad_norm": 0.8472688794136047, + "learning_rate": 4.9998236836892566e-06, + "loss": 0.8068, + "step": 127 + }, + { + "epoch": 0.1060187741579238, + "grad_norm": 0.8716503381729126, + "learning_rate": 4.999810381529447e-06, + "loss": 0.8005, + "step": 128 + }, + { + "epoch": 0.10684704583103258, + "grad_norm": 0.8416318893432617, + "learning_rate": 4.999796595679629e-06, + "loss": 0.7914, + "step": 129 + }, + { + "epoch": 0.10767531750414136, + "grad_norm": 0.838101863861084, + "learning_rate": 4.999782326142474e-06, + "loss": 0.7957, + "step": 130 + }, + { + "epoch": 0.10850358917725014, + "grad_norm": 0.8640785217285156, + "learning_rate": 4.999767572920741e-06, + "loss": 0.8077, + "step": 131 + }, + { + "epoch": 0.10933186085035891, + "grad_norm": 0.8207422494888306, + "learning_rate": 4.999752336017286e-06, + "loss": 0.7973, + "step": 132 + }, + { + "epoch": 0.1101601325234677, + "grad_norm": 0.8640758395195007, + "learning_rate": 4.9997366154350566e-06, + "loss": 0.8165, + "step": 133 + }, + { + "epoch": 0.11098840419657648, + "grad_norm": 0.7965807914733887, + "learning_rate": 4.999720411177094e-06, + "loss": 0.8059, + "step": 134 + }, + { + "epoch": 0.11181667586968526, + "grad_norm": 0.8463795185089111, + "learning_rate": 4.999703723246533e-06, + "loss": 0.8106, + "step": 135 + }, + { + "epoch": 0.11264494754279404, + "grad_norm": 0.8118059635162354, + "learning_rate": 4.999686551646604e-06, + "loss": 0.8099, + "step": 136 + }, + { + "epoch": 0.11347321921590281, + "grad_norm": 0.8256771564483643, + "learning_rate": 4.999668896380629e-06, + "loss": 0.8043, + "step": 137 + }, + { + "epoch": 0.11430149088901159, + "grad_norm": 0.8203126192092896, + "learning_rate": 4.999650757452025e-06, + "loss": 0.8029, + "step": 138 + }, + { + "epoch": 0.11512976256212037, + "grad_norm": 0.8442854881286621, + "learning_rate": 4.9996321348643e-06, + "loss": 0.8004, + "step": 139 + }, + { + "epoch": 0.11595803423522916, + "grad_norm": 0.8523190021514893, + "learning_rate": 4.999613028621059e-06, + "loss": 0.8092, + "step": 140 + }, + { + "epoch": 0.11678630590833794, + "grad_norm": 0.821428656578064, + "learning_rate": 4.999593438725997e-06, + "loss": 0.8105, + "step": 141 + }, + { + "epoch": 0.11761457758144671, + "grad_norm": 0.8406105041503906, + "learning_rate": 4.999573365182906e-06, + "loss": 0.7757, + "step": 142 + }, + { + "epoch": 0.11844284925455549, + "grad_norm": 0.8620600700378418, + "learning_rate": 4.999552807995669e-06, + "loss": 0.781, + "step": 143 + }, + { + "epoch": 0.11927112092766427, + "grad_norm": 0.920437216758728, + "learning_rate": 4.9995317671682655e-06, + "loss": 0.7579, + "step": 144 + }, + { + "epoch": 0.12009939260077306, + "grad_norm": 0.8414936661720276, + "learning_rate": 4.999510242704765e-06, + "loss": 0.8031, + "step": 145 + }, + { + "epoch": 0.12092766427388184, + "grad_norm": 0.8372176289558411, + "learning_rate": 4.999488234609332e-06, + "loss": 0.7931, + "step": 146 + }, + { + "epoch": 0.1217559359469906, + "grad_norm": 0.8785244226455688, + "learning_rate": 4.999465742886226e-06, + "loss": 0.8189, + "step": 147 + }, + { + "epoch": 0.12258420762009939, + "grad_norm": 0.8404353260993958, + "learning_rate": 4.999442767539799e-06, + "loss": 0.7899, + "step": 148 + }, + { + "epoch": 0.12341247929320817, + "grad_norm": 0.8541820049285889, + "learning_rate": 4.999419308574494e-06, + "loss": 0.7764, + "step": 149 + }, + { + "epoch": 0.12424075096631695, + "grad_norm": 0.8220177292823792, + "learning_rate": 4.999395365994854e-06, + "loss": 0.7942, + "step": 150 + }, + { + "epoch": 0.12506902263942574, + "grad_norm": 0.8823731541633606, + "learning_rate": 4.999370939805509e-06, + "loss": 0.793, + "step": 151 + }, + { + "epoch": 0.12589729431253452, + "grad_norm": 0.838860034942627, + "learning_rate": 4.999346030011186e-06, + "loss": 0.7741, + "step": 152 + }, + { + "epoch": 0.1267255659856433, + "grad_norm": 0.8162702322006226, + "learning_rate": 4.999320636616705e-06, + "loss": 0.7989, + "step": 153 + }, + { + "epoch": 0.12755383765875208, + "grad_norm": 0.8328465819358826, + "learning_rate": 4.99929475962698e-06, + "loss": 0.8055, + "step": 154 + }, + { + "epoch": 0.12838210933186084, + "grad_norm": 0.8373605608940125, + "learning_rate": 4.999268399047016e-06, + "loss": 0.801, + "step": 155 + }, + { + "epoch": 0.12921038100496962, + "grad_norm": 0.8408295512199402, + "learning_rate": 4.999241554881915e-06, + "loss": 0.7749, + "step": 156 + }, + { + "epoch": 0.1300386526780784, + "grad_norm": 0.8326342701911926, + "learning_rate": 4.999214227136871e-06, + "loss": 0.7818, + "step": 157 + }, + { + "epoch": 0.13086692435118719, + "grad_norm": 0.8064060211181641, + "learning_rate": 4.99918641581717e-06, + "loss": 0.7755, + "step": 158 + }, + { + "epoch": 0.13169519602429597, + "grad_norm": 0.8274610042572021, + "learning_rate": 4.999158120928196e-06, + "loss": 0.8008, + "step": 159 + }, + { + "epoch": 0.13252346769740475, + "grad_norm": 0.8121576309204102, + "learning_rate": 4.999129342475422e-06, + "loss": 0.7713, + "step": 160 + }, + { + "epoch": 0.13335173937051353, + "grad_norm": 0.8227710723876953, + "learning_rate": 4.9991000804644176e-06, + "loss": 0.7902, + "step": 161 + }, + { + "epoch": 0.13418001104362232, + "grad_norm": 0.8437026739120483, + "learning_rate": 4.999070334900843e-06, + "loss": 0.8176, + "step": 162 + }, + { + "epoch": 0.1350082827167311, + "grad_norm": 0.8379645943641663, + "learning_rate": 4.999040105790455e-06, + "loss": 0.7973, + "step": 163 + }, + { + "epoch": 0.13583655438983988, + "grad_norm": 0.822542130947113, + "learning_rate": 4.999009393139103e-06, + "loss": 0.7955, + "step": 164 + }, + { + "epoch": 0.13666482606294864, + "grad_norm": 0.8141928315162659, + "learning_rate": 4.9989781969527274e-06, + "loss": 0.7766, + "step": 165 + }, + { + "epoch": 0.13749309773605742, + "grad_norm": 0.8325410485267639, + "learning_rate": 4.998946517237367e-06, + "loss": 0.7737, + "step": 166 + }, + { + "epoch": 0.1383213694091662, + "grad_norm": 0.8399263024330139, + "learning_rate": 4.99891435399915e-06, + "loss": 0.7746, + "step": 167 + }, + { + "epoch": 0.13914964108227498, + "grad_norm": 0.8694303035736084, + "learning_rate": 4.9988817072443e-06, + "loss": 0.776, + "step": 168 + }, + { + "epoch": 0.13997791275538377, + "grad_norm": 0.8650498390197754, + "learning_rate": 4.998848576979135e-06, + "loss": 0.7941, + "step": 169 + }, + { + "epoch": 0.14080618442849255, + "grad_norm": 0.8246620893478394, + "learning_rate": 4.998814963210062e-06, + "loss": 0.7604, + "step": 170 + }, + { + "epoch": 0.14163445610160133, + "grad_norm": 0.8137409090995789, + "learning_rate": 4.99878086594359e-06, + "loss": 0.7681, + "step": 171 + }, + { + "epoch": 0.1424627277747101, + "grad_norm": 0.9755353927612305, + "learning_rate": 4.998746285186312e-06, + "loss": 0.7887, + "step": 172 + }, + { + "epoch": 0.1432909994478189, + "grad_norm": 0.8673408031463623, + "learning_rate": 4.998711220944921e-06, + "loss": 0.7803, + "step": 173 + }, + { + "epoch": 0.14411927112092768, + "grad_norm": 0.8312270045280457, + "learning_rate": 4.998675673226202e-06, + "loss": 0.7709, + "step": 174 + }, + { + "epoch": 0.14494754279403643, + "grad_norm": 0.8126094341278076, + "learning_rate": 4.998639642037032e-06, + "loss": 0.7822, + "step": 175 + }, + { + "epoch": 0.14577581446714521, + "grad_norm": 0.8561585545539856, + "learning_rate": 4.9986031273843834e-06, + "loss": 0.7829, + "step": 176 + }, + { + "epoch": 0.146604086140254, + "grad_norm": 0.8787612915039062, + "learning_rate": 4.9985661292753214e-06, + "loss": 0.7913, + "step": 177 + }, + { + "epoch": 0.14743235781336278, + "grad_norm": 0.8798345327377319, + "learning_rate": 4.998528647717006e-06, + "loss": 0.797, + "step": 178 + }, + { + "epoch": 0.14826062948647156, + "grad_norm": 0.7956589460372925, + "learning_rate": 4.998490682716687e-06, + "loss": 0.7571, + "step": 179 + }, + { + "epoch": 0.14908890115958034, + "grad_norm": 0.8505003452301025, + "learning_rate": 4.998452234281712e-06, + "loss": 0.7736, + "step": 180 + }, + { + "epoch": 0.14991717283268913, + "grad_norm": 0.8857074975967407, + "learning_rate": 4.998413302419519e-06, + "loss": 0.7863, + "step": 181 + }, + { + "epoch": 0.1507454445057979, + "grad_norm": 0.8762485980987549, + "learning_rate": 4.998373887137642e-06, + "loss": 0.7588, + "step": 182 + }, + { + "epoch": 0.1515737161789067, + "grad_norm": 0.8497205972671509, + "learning_rate": 4.998333988443709e-06, + "loss": 0.7773, + "step": 183 + }, + { + "epoch": 0.15240198785201547, + "grad_norm": 0.8720596432685852, + "learning_rate": 4.998293606345437e-06, + "loss": 0.7834, + "step": 184 + }, + { + "epoch": 0.15323025952512423, + "grad_norm": 0.8567108511924744, + "learning_rate": 4.998252740850641e-06, + "loss": 0.7794, + "step": 185 + }, + { + "epoch": 0.154058531198233, + "grad_norm": 0.8227219581604004, + "learning_rate": 4.998211391967228e-06, + "loss": 0.7688, + "step": 186 + }, + { + "epoch": 0.1548868028713418, + "grad_norm": 0.8321112394332886, + "learning_rate": 4.998169559703199e-06, + "loss": 0.7747, + "step": 187 + }, + { + "epoch": 0.15571507454445058, + "grad_norm": 0.8556115627288818, + "learning_rate": 4.9981272440666486e-06, + "loss": 0.7916, + "step": 188 + }, + { + "epoch": 0.15654334621755936, + "grad_norm": 0.8473951816558838, + "learning_rate": 4.998084445065763e-06, + "loss": 0.7823, + "step": 189 + }, + { + "epoch": 0.15737161789066814, + "grad_norm": 0.8551077246665955, + "learning_rate": 4.998041162708823e-06, + "loss": 0.7816, + "step": 190 + }, + { + "epoch": 0.15819988956377692, + "grad_norm": 0.945099413394928, + "learning_rate": 4.997997397004206e-06, + "loss": 0.7828, + "step": 191 + }, + { + "epoch": 0.1590281612368857, + "grad_norm": 0.8747237920761108, + "learning_rate": 4.9979531479603785e-06, + "loss": 0.7559, + "step": 192 + }, + { + "epoch": 0.1598564329099945, + "grad_norm": 0.8479958176612854, + "learning_rate": 4.997908415585902e-06, + "loss": 0.7694, + "step": 193 + }, + { + "epoch": 0.16068470458310327, + "grad_norm": 0.8659047484397888, + "learning_rate": 4.997863199889433e-06, + "loss": 0.7534, + "step": 194 + }, + { + "epoch": 0.16151297625621203, + "grad_norm": 0.8744727969169617, + "learning_rate": 4.997817500879719e-06, + "loss": 0.7499, + "step": 195 + }, + { + "epoch": 0.1623412479293208, + "grad_norm": 0.896468997001648, + "learning_rate": 4.997771318565603e-06, + "loss": 0.7974, + "step": 196 + }, + { + "epoch": 0.1631695196024296, + "grad_norm": 0.9058924913406372, + "learning_rate": 4.997724652956021e-06, + "loss": 0.7753, + "step": 197 + }, + { + "epoch": 0.16399779127553837, + "grad_norm": 0.8904663920402527, + "learning_rate": 4.9976775040600014e-06, + "loss": 0.7569, + "step": 198 + }, + { + "epoch": 0.16482606294864716, + "grad_norm": 0.8709238767623901, + "learning_rate": 4.997629871886669e-06, + "loss": 0.763, + "step": 199 + }, + { + "epoch": 0.16565433462175594, + "grad_norm": 0.8400478959083557, + "learning_rate": 4.997581756445238e-06, + "loss": 0.7795, + "step": 200 + }, + { + "epoch": 0.16648260629486472, + "grad_norm": 0.8458966612815857, + "learning_rate": 4.9975331577450204e-06, + "loss": 0.7597, + "step": 201 + }, + { + "epoch": 0.1673108779679735, + "grad_norm": 0.8707828521728516, + "learning_rate": 4.997484075795417e-06, + "loss": 0.759, + "step": 202 + }, + { + "epoch": 0.16813914964108229, + "grad_norm": 0.8774807453155518, + "learning_rate": 4.997434510605927e-06, + "loss": 0.7706, + "step": 203 + }, + { + "epoch": 0.16896742131419104, + "grad_norm": 0.8987929821014404, + "learning_rate": 4.99738446218614e-06, + "loss": 0.7319, + "step": 204 + }, + { + "epoch": 0.16979569298729982, + "grad_norm": 0.8869186639785767, + "learning_rate": 4.99733393054574e-06, + "loss": 0.7812, + "step": 205 + }, + { + "epoch": 0.1706239646604086, + "grad_norm": 0.860407829284668, + "learning_rate": 4.997282915694505e-06, + "loss": 0.7524, + "step": 206 + }, + { + "epoch": 0.1714522363335174, + "grad_norm": 0.8719788789749146, + "learning_rate": 4.9972314176423035e-06, + "loss": 0.7796, + "step": 207 + }, + { + "epoch": 0.17228050800662617, + "grad_norm": 0.8596137762069702, + "learning_rate": 4.997179436399103e-06, + "loss": 0.7554, + "step": 208 + }, + { + "epoch": 0.17310877967973495, + "grad_norm": 0.8505175709724426, + "learning_rate": 4.997126971974959e-06, + "loss": 0.7607, + "step": 209 + }, + { + "epoch": 0.17393705135284374, + "grad_norm": 0.7858585119247437, + "learning_rate": 4.9970740243800245e-06, + "loss": 0.7622, + "step": 210 + }, + { + "epoch": 0.17476532302595252, + "grad_norm": 0.8634621500968933, + "learning_rate": 4.997020593624543e-06, + "loss": 0.7659, + "step": 211 + }, + { + "epoch": 0.1755935946990613, + "grad_norm": 0.8650951385498047, + "learning_rate": 4.9969666797188545e-06, + "loss": 0.7689, + "step": 212 + }, + { + "epoch": 0.17642186637217008, + "grad_norm": 0.8602325916290283, + "learning_rate": 4.996912282673389e-06, + "loss": 0.7695, + "step": 213 + }, + { + "epoch": 0.17725013804527884, + "grad_norm": 0.8453742265701294, + "learning_rate": 4.9968574024986735e-06, + "loss": 0.7434, + "step": 214 + }, + { + "epoch": 0.17807840971838762, + "grad_norm": 0.8542691469192505, + "learning_rate": 4.9968020392053255e-06, + "loss": 0.7298, + "step": 215 + }, + { + "epoch": 0.1789066813914964, + "grad_norm": 0.8806673288345337, + "learning_rate": 4.996746192804058e-06, + "loss": 0.7649, + "step": 216 + }, + { + "epoch": 0.17973495306460519, + "grad_norm": 0.8451183438301086, + "learning_rate": 4.9966898633056765e-06, + "loss": 0.7681, + "step": 217 + }, + { + "epoch": 0.18056322473771397, + "grad_norm": 0.8664626479148865, + "learning_rate": 4.99663305072108e-06, + "loss": 0.7753, + "step": 218 + }, + { + "epoch": 0.18139149641082275, + "grad_norm": 0.8828144669532776, + "learning_rate": 4.996575755061262e-06, + "loss": 0.7514, + "step": 219 + }, + { + "epoch": 0.18221976808393153, + "grad_norm": 0.8493948578834534, + "learning_rate": 4.996517976337308e-06, + "loss": 0.7684, + "step": 220 + }, + { + "epoch": 0.18304803975704032, + "grad_norm": 0.8561757802963257, + "learning_rate": 4.9964597145603975e-06, + "loss": 0.7723, + "step": 221 + }, + { + "epoch": 0.1838763114301491, + "grad_norm": 0.8863773941993713, + "learning_rate": 4.996400969741805e-06, + "loss": 0.7732, + "step": 222 + }, + { + "epoch": 0.18470458310325788, + "grad_norm": 0.8571644425392151, + "learning_rate": 4.996341741892895e-06, + "loss": 0.744, + "step": 223 + }, + { + "epoch": 0.18553285477636663, + "grad_norm": 0.8622035980224609, + "learning_rate": 4.9962820310251286e-06, + "loss": 0.7607, + "step": 224 + }, + { + "epoch": 0.18636112644947542, + "grad_norm": 0.8904410600662231, + "learning_rate": 4.9962218371500595e-06, + "loss": 0.7648, + "step": 225 + }, + { + "epoch": 0.1871893981225842, + "grad_norm": 0.8975846171379089, + "learning_rate": 4.996161160279335e-06, + "loss": 0.7469, + "step": 226 + }, + { + "epoch": 0.18801766979569298, + "grad_norm": 0.8788724541664124, + "learning_rate": 4.996100000424694e-06, + "loss": 0.7473, + "step": 227 + }, + { + "epoch": 0.18884594146880176, + "grad_norm": 0.9446991086006165, + "learning_rate": 4.996038357597971e-06, + "loss": 0.7635, + "step": 228 + }, + { + "epoch": 0.18967421314191055, + "grad_norm": 0.86910480260849, + "learning_rate": 4.995976231811094e-06, + "loss": 0.7539, + "step": 229 + }, + { + "epoch": 0.19050248481501933, + "grad_norm": 0.862197756767273, + "learning_rate": 4.995913623076084e-06, + "loss": 0.7546, + "step": 230 + }, + { + "epoch": 0.1913307564881281, + "grad_norm": 0.906878650188446, + "learning_rate": 4.995850531405054e-06, + "loss": 0.7737, + "step": 231 + }, + { + "epoch": 0.1921590281612369, + "grad_norm": 0.8557183742523193, + "learning_rate": 4.995786956810212e-06, + "loss": 0.7663, + "step": 232 + }, + { + "epoch": 0.19298729983434568, + "grad_norm": 0.8125383853912354, + "learning_rate": 4.995722899303859e-06, + "loss": 0.7647, + "step": 233 + }, + { + "epoch": 0.19381557150745443, + "grad_norm": 0.8610332012176514, + "learning_rate": 4.995658358898391e-06, + "loss": 0.7541, + "step": 234 + }, + { + "epoch": 0.19464384318056321, + "grad_norm": 0.8620829582214355, + "learning_rate": 4.995593335606294e-06, + "loss": 0.7625, + "step": 235 + }, + { + "epoch": 0.195472114853672, + "grad_norm": 0.8296957612037659, + "learning_rate": 4.9955278294401496e-06, + "loss": 0.7536, + "step": 236 + }, + { + "epoch": 0.19630038652678078, + "grad_norm": 0.8974814414978027, + "learning_rate": 4.995461840412634e-06, + "loss": 0.7651, + "step": 237 + }, + { + "epoch": 0.19712865819988956, + "grad_norm": 0.8701095581054688, + "learning_rate": 4.995395368536515e-06, + "loss": 0.788, + "step": 238 + }, + { + "epoch": 0.19795692987299834, + "grad_norm": 0.8556636571884155, + "learning_rate": 4.995328413824653e-06, + "loss": 0.7649, + "step": 239 + }, + { + "epoch": 0.19878520154610713, + "grad_norm": 0.836686909198761, + "learning_rate": 4.995260976290005e-06, + "loss": 0.7583, + "step": 240 + }, + { + "epoch": 0.1996134732192159, + "grad_norm": 0.8807066082954407, + "learning_rate": 4.995193055945618e-06, + "loss": 0.7766, + "step": 241 + }, + { + "epoch": 0.2004417448923247, + "grad_norm": 0.8363582491874695, + "learning_rate": 4.995124652804635e-06, + "loss": 0.7655, + "step": 242 + }, + { + "epoch": 0.20127001656543347, + "grad_norm": 0.8699461817741394, + "learning_rate": 4.9950557668802905e-06, + "loss": 0.743, + "step": 243 + }, + { + "epoch": 0.20209828823854223, + "grad_norm": 0.8571060299873352, + "learning_rate": 4.994986398185914e-06, + "loss": 0.7512, + "step": 244 + }, + { + "epoch": 0.202926559911651, + "grad_norm": 0.8429675102233887, + "learning_rate": 4.994916546734927e-06, + "loss": 0.7589, + "step": 245 + }, + { + "epoch": 0.2037548315847598, + "grad_norm": 0.874822199344635, + "learning_rate": 4.994846212540846e-06, + "loss": 0.7534, + "step": 246 + }, + { + "epoch": 0.20458310325786858, + "grad_norm": 0.878341555595398, + "learning_rate": 4.994775395617279e-06, + "loss": 0.7709, + "step": 247 + }, + { + "epoch": 0.20541137493097736, + "grad_norm": 0.897961437702179, + "learning_rate": 4.99470409597793e-06, + "loss": 0.7768, + "step": 248 + }, + { + "epoch": 0.20623964660408614, + "grad_norm": 0.8439584374427795, + "learning_rate": 4.994632313636593e-06, + "loss": 0.7558, + "step": 249 + }, + { + "epoch": 0.20706791827719492, + "grad_norm": 0.8394468426704407, + "learning_rate": 4.994560048607159e-06, + "loss": 0.7492, + "step": 250 + }, + { + "epoch": 0.2078961899503037, + "grad_norm": 0.8489627242088318, + "learning_rate": 4.9944873009036074e-06, + "loss": 0.7365, + "step": 251 + }, + { + "epoch": 0.2087244616234125, + "grad_norm": 0.8675844669342041, + "learning_rate": 4.994414070540018e-06, + "loss": 0.762, + "step": 252 + }, + { + "epoch": 0.20955273329652127, + "grad_norm": 0.8228297829627991, + "learning_rate": 4.994340357530558e-06, + "loss": 0.7475, + "step": 253 + }, + { + "epoch": 0.21038100496963003, + "grad_norm": 0.8794129490852356, + "learning_rate": 4.9942661618894915e-06, + "loss": 0.7631, + "step": 254 + }, + { + "epoch": 0.2112092766427388, + "grad_norm": 0.8616135716438293, + "learning_rate": 4.9941914836311735e-06, + "loss": 0.7749, + "step": 255 + }, + { + "epoch": 0.2120375483158476, + "grad_norm": 0.8470495939254761, + "learning_rate": 4.994116322770054e-06, + "loss": 0.7484, + "step": 256 + }, + { + "epoch": 0.21286581998895637, + "grad_norm": 0.9346716403961182, + "learning_rate": 4.9940406793206755e-06, + "loss": 0.7557, + "step": 257 + }, + { + "epoch": 0.21369409166206516, + "grad_norm": 0.8993160724639893, + "learning_rate": 4.993964553297676e-06, + "loss": 0.7422, + "step": 258 + }, + { + "epoch": 0.21452236333517394, + "grad_norm": 0.8257086277008057, + "learning_rate": 4.993887944715783e-06, + "loss": 0.7783, + "step": 259 + }, + { + "epoch": 0.21535063500828272, + "grad_norm": 0.8950703144073486, + "learning_rate": 4.993810853589819e-06, + "loss": 0.7311, + "step": 260 + }, + { + "epoch": 0.2161789066813915, + "grad_norm": 0.8733702898025513, + "learning_rate": 4.993733279934704e-06, + "loss": 0.7572, + "step": 261 + }, + { + "epoch": 0.21700717835450029, + "grad_norm": 0.8322939276695251, + "learning_rate": 4.993655223765444e-06, + "loss": 0.7632, + "step": 262 + }, + { + "epoch": 0.21783545002760907, + "grad_norm": 0.8666137456893921, + "learning_rate": 4.993576685097145e-06, + "loss": 0.7619, + "step": 263 + }, + { + "epoch": 0.21866372170071782, + "grad_norm": 0.8770918250083923, + "learning_rate": 4.993497663945002e-06, + "loss": 0.7538, + "step": 264 + }, + { + "epoch": 0.2194919933738266, + "grad_norm": 0.8940248489379883, + "learning_rate": 4.9934181603243045e-06, + "loss": 0.7512, + "step": 265 + }, + { + "epoch": 0.2203202650469354, + "grad_norm": 0.8979714512825012, + "learning_rate": 4.993338174250437e-06, + "loss": 0.7504, + "step": 266 + }, + { + "epoch": 0.22114853672004417, + "grad_norm": 0.8802001476287842, + "learning_rate": 4.993257705738876e-06, + "loss": 0.7501, + "step": 267 + }, + { + "epoch": 0.22197680839315295, + "grad_norm": 0.8734432458877563, + "learning_rate": 4.993176754805189e-06, + "loss": 0.7452, + "step": 268 + }, + { + "epoch": 0.22280508006626173, + "grad_norm": 0.9127711057662964, + "learning_rate": 4.993095321465042e-06, + "loss": 0.7404, + "step": 269 + }, + { + "epoch": 0.22363335173937052, + "grad_norm": 0.8533952832221985, + "learning_rate": 4.993013405734191e-06, + "loss": 0.7382, + "step": 270 + }, + { + "epoch": 0.2244616234124793, + "grad_norm": 0.8463877439498901, + "learning_rate": 4.9929310076284845e-06, + "loss": 0.7574, + "step": 271 + }, + { + "epoch": 0.22528989508558808, + "grad_norm": 0.8598008155822754, + "learning_rate": 4.992848127163868e-06, + "loss": 0.7624, + "step": 272 + }, + { + "epoch": 0.22611816675869686, + "grad_norm": 0.8759154081344604, + "learning_rate": 4.992764764356376e-06, + "loss": 0.7637, + "step": 273 + }, + { + "epoch": 0.22694643843180562, + "grad_norm": 0.8880444765090942, + "learning_rate": 4.992680919222141e-06, + "loss": 0.7514, + "step": 274 + }, + { + "epoch": 0.2277747101049144, + "grad_norm": 0.871366024017334, + "learning_rate": 4.9925965917773826e-06, + "loss": 0.7569, + "step": 275 + }, + { + "epoch": 0.22860298177802318, + "grad_norm": 0.862654983997345, + "learning_rate": 4.99251178203842e-06, + "loss": 0.7349, + "step": 276 + }, + { + "epoch": 0.22943125345113197, + "grad_norm": 0.8615298867225647, + "learning_rate": 4.992426490021662e-06, + "loss": 0.7438, + "step": 277 + }, + { + "epoch": 0.23025952512424075, + "grad_norm": 0.8735869526863098, + "learning_rate": 4.992340715743614e-06, + "loss": 0.7755, + "step": 278 + }, + { + "epoch": 0.23108779679734953, + "grad_norm": 0.9082722067832947, + "learning_rate": 4.9922544592208695e-06, + "loss": 0.7561, + "step": 279 + }, + { + "epoch": 0.23191606847045831, + "grad_norm": 0.8639047145843506, + "learning_rate": 4.992167720470119e-06, + "loss": 0.7493, + "step": 280 + }, + { + "epoch": 0.2327443401435671, + "grad_norm": 0.891640305519104, + "learning_rate": 4.9920804995081475e-06, + "loss": 0.75, + "step": 281 + }, + { + "epoch": 0.23357261181667588, + "grad_norm": 0.8880137205123901, + "learning_rate": 4.99199279635183e-06, + "loss": 0.7801, + "step": 282 + }, + { + "epoch": 0.23440088348978466, + "grad_norm": 0.8600473403930664, + "learning_rate": 4.991904611018137e-06, + "loss": 0.7477, + "step": 283 + }, + { + "epoch": 0.23522915516289342, + "grad_norm": 0.8685399889945984, + "learning_rate": 4.99181594352413e-06, + "loss": 0.741, + "step": 284 + }, + { + "epoch": 0.2360574268360022, + "grad_norm": 0.9006181955337524, + "learning_rate": 4.991726793886966e-06, + "loss": 0.7288, + "step": 285 + }, + { + "epoch": 0.23688569850911098, + "grad_norm": 0.863895833492279, + "learning_rate": 4.991637162123896e-06, + "loss": 0.7375, + "step": 286 + }, + { + "epoch": 0.23771397018221976, + "grad_norm": 0.8620381951332092, + "learning_rate": 4.9915470482522625e-06, + "loss": 0.7249, + "step": 287 + }, + { + "epoch": 0.23854224185532855, + "grad_norm": 0.9634964466094971, + "learning_rate": 4.991456452289499e-06, + "loss": 0.7313, + "step": 288 + }, + { + "epoch": 0.23937051352843733, + "grad_norm": 0.9154514670372009, + "learning_rate": 4.9913653742531385e-06, + "loss": 0.782, + "step": 289 + }, + { + "epoch": 0.2401987852015461, + "grad_norm": 0.9179158806800842, + "learning_rate": 4.991273814160802e-06, + "loss": 0.7388, + "step": 290 + }, + { + "epoch": 0.2410270568746549, + "grad_norm": 0.8609046936035156, + "learning_rate": 4.991181772030206e-06, + "loss": 0.7393, + "step": 291 + }, + { + "epoch": 0.24185532854776368, + "grad_norm": 0.8689826130867004, + "learning_rate": 4.991089247879159e-06, + "loss": 0.7302, + "step": 292 + }, + { + "epoch": 0.24268360022087246, + "grad_norm": 0.8414047956466675, + "learning_rate": 4.990996241725565e-06, + "loss": 0.7496, + "step": 293 + }, + { + "epoch": 0.2435118718939812, + "grad_norm": 0.8621289134025574, + "learning_rate": 4.990902753587418e-06, + "loss": 0.7317, + "step": 294 + }, + { + "epoch": 0.24434014356709, + "grad_norm": 0.856566309928894, + "learning_rate": 4.990808783482809e-06, + "loss": 0.7169, + "step": 295 + }, + { + "epoch": 0.24516841524019878, + "grad_norm": 0.8549672961235046, + "learning_rate": 4.990714331429919e-06, + "loss": 0.7357, + "step": 296 + }, + { + "epoch": 0.24599668691330756, + "grad_norm": 0.8813024759292603, + "learning_rate": 4.990619397447024e-06, + "loss": 0.7437, + "step": 297 + }, + { + "epoch": 0.24682495858641634, + "grad_norm": 0.9008268713951111, + "learning_rate": 4.990523981552493e-06, + "loss": 0.7652, + "step": 298 + }, + { + "epoch": 0.24765323025952513, + "grad_norm": 0.910277783870697, + "learning_rate": 4.990428083764788e-06, + "loss": 0.7598, + "step": 299 + }, + { + "epoch": 0.2484815019326339, + "grad_norm": 0.9261599183082581, + "learning_rate": 4.990331704102464e-06, + "loss": 0.7199, + "step": 300 + }, + { + "epoch": 0.2493097736057427, + "grad_norm": 0.9166674613952637, + "learning_rate": 4.99023484258417e-06, + "loss": 0.7414, + "step": 301 + }, + { + "epoch": 0.2501380452788515, + "grad_norm": 0.8701214790344238, + "learning_rate": 4.990137499228648e-06, + "loss": 0.7326, + "step": 302 + }, + { + "epoch": 0.25096631695196026, + "grad_norm": 0.9120466113090515, + "learning_rate": 4.990039674054732e-06, + "loss": 0.7574, + "step": 303 + }, + { + "epoch": 0.25179458862506904, + "grad_norm": 0.9449286460876465, + "learning_rate": 4.989941367081351e-06, + "loss": 0.725, + "step": 304 + }, + { + "epoch": 0.2526228602981778, + "grad_norm": 0.8989110589027405, + "learning_rate": 4.989842578327526e-06, + "loss": 0.7569, + "step": 305 + }, + { + "epoch": 0.2534511319712866, + "grad_norm": 0.891677975654602, + "learning_rate": 4.989743307812373e-06, + "loss": 0.7135, + "step": 306 + }, + { + "epoch": 0.2542794036443954, + "grad_norm": 0.8689535856246948, + "learning_rate": 4.989643555555099e-06, + "loss": 0.7439, + "step": 307 + }, + { + "epoch": 0.25510767531750417, + "grad_norm": 0.8779630064964294, + "learning_rate": 4.989543321575004e-06, + "loss": 0.7322, + "step": 308 + }, + { + "epoch": 0.2559359469906129, + "grad_norm": 0.8977622985839844, + "learning_rate": 4.9894426058914856e-06, + "loss": 0.7544, + "step": 309 + }, + { + "epoch": 0.2567642186637217, + "grad_norm": 0.8216757774353027, + "learning_rate": 4.989341408524028e-06, + "loss": 0.7344, + "step": 310 + }, + { + "epoch": 0.25759249033683046, + "grad_norm": 0.8494759798049927, + "learning_rate": 4.9892397294922126e-06, + "loss": 0.7389, + "step": 311 + }, + { + "epoch": 0.25842076200993924, + "grad_norm": 0.881244957447052, + "learning_rate": 4.989137568815715e-06, + "loss": 0.7121, + "step": 312 + }, + { + "epoch": 0.259249033683048, + "grad_norm": 0.8679410219192505, + "learning_rate": 4.9890349265143005e-06, + "loss": 0.7469, + "step": 313 + }, + { + "epoch": 0.2600773053561568, + "grad_norm": 0.9746971130371094, + "learning_rate": 4.988931802607831e-06, + "loss": 0.7585, + "step": 314 + }, + { + "epoch": 0.2609055770292656, + "grad_norm": 0.9222471117973328, + "learning_rate": 4.988828197116259e-06, + "loss": 0.761, + "step": 315 + }, + { + "epoch": 0.26173384870237437, + "grad_norm": 0.9104124903678894, + "learning_rate": 4.988724110059631e-06, + "loss": 0.7652, + "step": 316 + }, + { + "epoch": 0.26256212037548315, + "grad_norm": 0.8876475095748901, + "learning_rate": 4.988619541458088e-06, + "loss": 0.7529, + "step": 317 + }, + { + "epoch": 0.26339039204859194, + "grad_norm": 0.8705421686172485, + "learning_rate": 4.988514491331861e-06, + "loss": 0.7437, + "step": 318 + }, + { + "epoch": 0.2642186637217007, + "grad_norm": 0.8594399690628052, + "learning_rate": 4.9884089597012784e-06, + "loss": 0.7406, + "step": 319 + }, + { + "epoch": 0.2650469353948095, + "grad_norm": 0.8807039856910706, + "learning_rate": 4.988302946586759e-06, + "loss": 0.7429, + "step": 320 + }, + { + "epoch": 0.2658752070679183, + "grad_norm": 0.8881412148475647, + "learning_rate": 4.9881964520088144e-06, + "loss": 0.7234, + "step": 321 + }, + { + "epoch": 0.26670347874102707, + "grad_norm": 0.8764941096305847, + "learning_rate": 4.988089475988052e-06, + "loss": 0.7514, + "step": 322 + }, + { + "epoch": 0.26753175041413585, + "grad_norm": 0.8992927670478821, + "learning_rate": 4.987982018545169e-06, + "loss": 0.7143, + "step": 323 + }, + { + "epoch": 0.26836002208724463, + "grad_norm": 0.8961546421051025, + "learning_rate": 4.9878740797009585e-06, + "loss": 0.7396, + "step": 324 + }, + { + "epoch": 0.2691882937603534, + "grad_norm": 0.9031578302383423, + "learning_rate": 4.987765659476304e-06, + "loss": 0.7407, + "step": 325 + }, + { + "epoch": 0.2700165654334622, + "grad_norm": 0.8693655729293823, + "learning_rate": 4.987656757892186e-06, + "loss": 0.7215, + "step": 326 + }, + { + "epoch": 0.270844837106571, + "grad_norm": 0.8739856481552124, + "learning_rate": 4.987547374969676e-06, + "loss": 0.7471, + "step": 327 + }, + { + "epoch": 0.27167310877967976, + "grad_norm": 0.8916322588920593, + "learning_rate": 4.987437510729936e-06, + "loss": 0.7459, + "step": 328 + }, + { + "epoch": 0.2725013804527885, + "grad_norm": 0.890178382396698, + "learning_rate": 4.987327165194225e-06, + "loss": 0.7322, + "step": 329 + }, + { + "epoch": 0.27332965212589727, + "grad_norm": 0.8611598014831543, + "learning_rate": 4.987216338383896e-06, + "loss": 0.733, + "step": 330 + }, + { + "epoch": 0.27415792379900605, + "grad_norm": 0.87791907787323, + "learning_rate": 4.98710503032039e-06, + "loss": 0.7445, + "step": 331 + }, + { + "epoch": 0.27498619547211484, + "grad_norm": 0.963479220867157, + "learning_rate": 4.986993241025245e-06, + "loss": 0.7376, + "step": 332 + }, + { + "epoch": 0.2758144671452236, + "grad_norm": 0.9083762764930725, + "learning_rate": 4.986880970520092e-06, + "loss": 0.7258, + "step": 333 + }, + { + "epoch": 0.2766427388183324, + "grad_norm": 1.048991084098816, + "learning_rate": 4.9867682188266524e-06, + "loss": 0.7399, + "step": 334 + }, + { + "epoch": 0.2774710104914412, + "grad_norm": 0.9404491782188416, + "learning_rate": 4.986654985966745e-06, + "loss": 0.7133, + "step": 335 + }, + { + "epoch": 0.27829928216454997, + "grad_norm": 0.9168346524238586, + "learning_rate": 4.986541271962278e-06, + "loss": 0.74, + "step": 336 + }, + { + "epoch": 0.27912755383765875, + "grad_norm": 0.8716689944267273, + "learning_rate": 4.986427076835253e-06, + "loss": 0.7249, + "step": 337 + }, + { + "epoch": 0.27995582551076753, + "grad_norm": 0.9097291827201843, + "learning_rate": 4.986312400607767e-06, + "loss": 0.7151, + "step": 338 + }, + { + "epoch": 0.2807840971838763, + "grad_norm": 0.8621960282325745, + "learning_rate": 4.98619724330201e-06, + "loss": 0.7125, + "step": 339 + }, + { + "epoch": 0.2816123688569851, + "grad_norm": 0.9823791980743408, + "learning_rate": 4.986081604940261e-06, + "loss": 0.7191, + "step": 340 + }, + { + "epoch": 0.2824406405300939, + "grad_norm": 0.8999047875404358, + "learning_rate": 4.9859654855448966e-06, + "loss": 0.7175, + "step": 341 + }, + { + "epoch": 0.28326891220320266, + "grad_norm": 0.8690859079360962, + "learning_rate": 4.985848885138386e-06, + "loss": 0.7309, + "step": 342 + }, + { + "epoch": 0.28409718387631144, + "grad_norm": 0.889500617980957, + "learning_rate": 4.985731803743287e-06, + "loss": 0.7391, + "step": 343 + }, + { + "epoch": 0.2849254555494202, + "grad_norm": 0.8992016911506653, + "learning_rate": 4.985614241382257e-06, + "loss": 0.7437, + "step": 344 + }, + { + "epoch": 0.285753727222529, + "grad_norm": 0.9100068211555481, + "learning_rate": 4.985496198078041e-06, + "loss": 0.722, + "step": 345 + }, + { + "epoch": 0.2865819988956378, + "grad_norm": 0.9013107419013977, + "learning_rate": 4.985377673853481e-06, + "loss": 0.7434, + "step": 346 + }, + { + "epoch": 0.2874102705687466, + "grad_norm": 0.8794704675674438, + "learning_rate": 4.9852586687315094e-06, + "loss": 0.7387, + "step": 347 + }, + { + "epoch": 0.28823854224185536, + "grad_norm": 0.9288201928138733, + "learning_rate": 4.9851391827351526e-06, + "loss": 0.7165, + "step": 348 + }, + { + "epoch": 0.2890668139149641, + "grad_norm": 0.8735847473144531, + "learning_rate": 4.98501921588753e-06, + "loss": 0.7479, + "step": 349 + }, + { + "epoch": 0.28989508558807286, + "grad_norm": 0.890616238117218, + "learning_rate": 4.984898768211853e-06, + "loss": 0.76, + "step": 350 + }, + { + "epoch": 0.29072335726118165, + "grad_norm": 0.9181028604507446, + "learning_rate": 4.98477783973143e-06, + "loss": 0.7366, + "step": 351 + }, + { + "epoch": 0.29155162893429043, + "grad_norm": 0.9947429299354553, + "learning_rate": 4.984656430469657e-06, + "loss": 0.7199, + "step": 352 + }, + { + "epoch": 0.2923799006073992, + "grad_norm": 0.8730769157409668, + "learning_rate": 4.984534540450027e-06, + "loss": 0.7153, + "step": 353 + }, + { + "epoch": 0.293208172280508, + "grad_norm": 0.8675997853279114, + "learning_rate": 4.9844121696961235e-06, + "loss": 0.7196, + "step": 354 + }, + { + "epoch": 0.2940364439536168, + "grad_norm": 0.9558542966842651, + "learning_rate": 4.984289318231624e-06, + "loss": 0.7311, + "step": 355 + }, + { + "epoch": 0.29486471562672556, + "grad_norm": 0.8817659616470337, + "learning_rate": 4.9841659860803e-06, + "loss": 0.7367, + "step": 356 + }, + { + "epoch": 0.29569298729983434, + "grad_norm": 0.8679526448249817, + "learning_rate": 4.984042173266016e-06, + "loss": 0.719, + "step": 357 + }, + { + "epoch": 0.2965212589729431, + "grad_norm": 0.9102020859718323, + "learning_rate": 4.983917879812726e-06, + "loss": 0.7301, + "step": 358 + }, + { + "epoch": 0.2973495306460519, + "grad_norm": 0.8987215757369995, + "learning_rate": 4.983793105744482e-06, + "loss": 0.7557, + "step": 359 + }, + { + "epoch": 0.2981778023191607, + "grad_norm": 0.8779832720756531, + "learning_rate": 4.983667851085426e-06, + "loss": 0.7471, + "step": 360 + }, + { + "epoch": 0.29900607399226947, + "grad_norm": 0.8762708306312561, + "learning_rate": 4.983542115859792e-06, + "loss": 0.7676, + "step": 361 + }, + { + "epoch": 0.29983434566537825, + "grad_norm": 0.910353422164917, + "learning_rate": 4.98341590009191e-06, + "loss": 0.7285, + "step": 362 + }, + { + "epoch": 0.30066261733848704, + "grad_norm": 0.9385405778884888, + "learning_rate": 4.983289203806202e-06, + "loss": 0.7142, + "step": 363 + }, + { + "epoch": 0.3014908890115958, + "grad_norm": 1.016781210899353, + "learning_rate": 4.983162027027182e-06, + "loss": 0.7408, + "step": 364 + }, + { + "epoch": 0.3023191606847046, + "grad_norm": 0.8642508387565613, + "learning_rate": 4.983034369779457e-06, + "loss": 0.7397, + "step": 365 + }, + { + "epoch": 0.3031474323578134, + "grad_norm": 0.9469170570373535, + "learning_rate": 4.982906232087728e-06, + "loss": 0.7158, + "step": 366 + }, + { + "epoch": 0.30397570403092217, + "grad_norm": 0.8818212151527405, + "learning_rate": 4.982777613976789e-06, + "loss": 0.7408, + "step": 367 + }, + { + "epoch": 0.30480397570403095, + "grad_norm": 0.8639838099479675, + "learning_rate": 4.982648515471526e-06, + "loss": 0.7217, + "step": 368 + }, + { + "epoch": 0.3056322473771397, + "grad_norm": 0.923780083656311, + "learning_rate": 4.982518936596917e-06, + "loss": 0.7284, + "step": 369 + }, + { + "epoch": 0.30646051905024846, + "grad_norm": 0.8992834687232971, + "learning_rate": 4.982388877378037e-06, + "loss": 0.7395, + "step": 370 + }, + { + "epoch": 0.30728879072335724, + "grad_norm": 0.8671404123306274, + "learning_rate": 4.982258337840049e-06, + "loss": 0.7157, + "step": 371 + }, + { + "epoch": 0.308117062396466, + "grad_norm": 0.8689168691635132, + "learning_rate": 4.982127318008212e-06, + "loss": 0.7398, + "step": 372 + }, + { + "epoch": 0.3089453340695748, + "grad_norm": 0.8964621424674988, + "learning_rate": 4.9819958179078765e-06, + "loss": 0.7385, + "step": 373 + }, + { + "epoch": 0.3097736057426836, + "grad_norm": 0.8984093070030212, + "learning_rate": 4.981863837564488e-06, + "loss": 0.7253, + "step": 374 + }, + { + "epoch": 0.31060187741579237, + "grad_norm": 0.8776745200157166, + "learning_rate": 4.981731377003583e-06, + "loss": 0.7126, + "step": 375 + }, + { + "epoch": 0.31143014908890115, + "grad_norm": 0.8782477378845215, + "learning_rate": 4.98159843625079e-06, + "loss": 0.7164, + "step": 376 + }, + { + "epoch": 0.31225842076200994, + "grad_norm": 0.8993703722953796, + "learning_rate": 4.981465015331832e-06, + "loss": 0.7391, + "step": 377 + }, + { + "epoch": 0.3130866924351187, + "grad_norm": 0.8975916504859924, + "learning_rate": 4.981331114272527e-06, + "loss": 0.7385, + "step": 378 + }, + { + "epoch": 0.3139149641082275, + "grad_norm": 0.8726824522018433, + "learning_rate": 4.9811967330987816e-06, + "loss": 0.7426, + "step": 379 + }, + { + "epoch": 0.3147432357813363, + "grad_norm": 0.8741514682769775, + "learning_rate": 4.981061871836597e-06, + "loss": 0.7246, + "step": 380 + }, + { + "epoch": 0.31557150745444507, + "grad_norm": 0.8807122707366943, + "learning_rate": 4.98092653051207e-06, + "loss": 0.7364, + "step": 381 + }, + { + "epoch": 0.31639977912755385, + "grad_norm": 0.835006058216095, + "learning_rate": 4.980790709151385e-06, + "loss": 0.7248, + "step": 382 + }, + { + "epoch": 0.31722805080066263, + "grad_norm": 0.9473997354507446, + "learning_rate": 4.980654407780824e-06, + "loss": 0.7331, + "step": 383 + }, + { + "epoch": 0.3180563224737714, + "grad_norm": 0.8332815170288086, + "learning_rate": 4.980517626426758e-06, + "loss": 0.7299, + "step": 384 + }, + { + "epoch": 0.3188845941468802, + "grad_norm": 0.8536360859870911, + "learning_rate": 4.980380365115656e-06, + "loss": 0.7095, + "step": 385 + }, + { + "epoch": 0.319712865819989, + "grad_norm": 0.9488157629966736, + "learning_rate": 4.980242623874075e-06, + "loss": 0.7382, + "step": 386 + }, + { + "epoch": 0.32054113749309776, + "grad_norm": 0.8623067736625671, + "learning_rate": 4.980104402728666e-06, + "loss": 0.7187, + "step": 387 + }, + { + "epoch": 0.32136940916620654, + "grad_norm": 0.9110252857208252, + "learning_rate": 4.979965701706175e-06, + "loss": 0.7354, + "step": 388 + }, + { + "epoch": 0.32219768083931527, + "grad_norm": 0.8966642618179321, + "learning_rate": 4.979826520833439e-06, + "loss": 0.7251, + "step": 389 + }, + { + "epoch": 0.32302595251242405, + "grad_norm": 0.856674075126648, + "learning_rate": 4.979686860137387e-06, + "loss": 0.723, + "step": 390 + }, + { + "epoch": 0.32385422418553284, + "grad_norm": 0.8623953461647034, + "learning_rate": 4.979546719645043e-06, + "loss": 0.7224, + "step": 391 + }, + { + "epoch": 0.3246824958586416, + "grad_norm": 0.8685624599456787, + "learning_rate": 4.979406099383523e-06, + "loss": 0.7283, + "step": 392 + }, + { + "epoch": 0.3255107675317504, + "grad_norm": 0.8610525727272034, + "learning_rate": 4.979264999380035e-06, + "loss": 0.7385, + "step": 393 + }, + { + "epoch": 0.3263390392048592, + "grad_norm": 0.8860796093940735, + "learning_rate": 4.979123419661882e-06, + "loss": 0.743, + "step": 394 + }, + { + "epoch": 0.32716731087796797, + "grad_norm": 0.8748682737350464, + "learning_rate": 4.978981360256457e-06, + "loss": 0.7447, + "step": 395 + }, + { + "epoch": 0.32799558255107675, + "grad_norm": 0.8750934600830078, + "learning_rate": 4.978838821191247e-06, + "loss": 0.7259, + "step": 396 + }, + { + "epoch": 0.32882385422418553, + "grad_norm": 0.9335777163505554, + "learning_rate": 4.978695802493834e-06, + "loss": 0.7271, + "step": 397 + }, + { + "epoch": 0.3296521258972943, + "grad_norm": 0.8823036551475525, + "learning_rate": 4.978552304191888e-06, + "loss": 0.723, + "step": 398 + }, + { + "epoch": 0.3304803975704031, + "grad_norm": 0.9503542184829712, + "learning_rate": 4.9784083263131764e-06, + "loss": 0.752, + "step": 399 + }, + { + "epoch": 0.3313086692435119, + "grad_norm": 0.8779938220977783, + "learning_rate": 4.978263868885557e-06, + "loss": 0.7098, + "step": 400 + }, + { + "epoch": 0.33213694091662066, + "grad_norm": 0.873857855796814, + "learning_rate": 4.978118931936981e-06, + "loss": 0.7143, + "step": 401 + }, + { + "epoch": 0.33296521258972944, + "grad_norm": 0.8437919020652771, + "learning_rate": 4.9779735154954924e-06, + "loss": 0.7275, + "step": 402 + }, + { + "epoch": 0.3337934842628382, + "grad_norm": 0.8914555311203003, + "learning_rate": 4.977827619589229e-06, + "loss": 0.722, + "step": 403 + }, + { + "epoch": 0.334621755935947, + "grad_norm": 0.9062597751617432, + "learning_rate": 4.977681244246417e-06, + "loss": 0.7067, + "step": 404 + }, + { + "epoch": 0.3354500276090558, + "grad_norm": 0.8555092811584473, + "learning_rate": 4.977534389495382e-06, + "loss": 0.7318, + "step": 405 + }, + { + "epoch": 0.33627829928216457, + "grad_norm": 0.8717684745788574, + "learning_rate": 4.9773870553645386e-06, + "loss": 0.7398, + "step": 406 + }, + { + "epoch": 0.33710657095527335, + "grad_norm": 0.8746877908706665, + "learning_rate": 4.977239241882393e-06, + "loss": 0.7154, + "step": 407 + }, + { + "epoch": 0.3379348426283821, + "grad_norm": 0.8905707597732544, + "learning_rate": 4.977090949077547e-06, + "loss": 0.7345, + "step": 408 + }, + { + "epoch": 0.33876311430149086, + "grad_norm": 0.9037347435951233, + "learning_rate": 4.9769421769786935e-06, + "loss": 0.7271, + "step": 409 + }, + { + "epoch": 0.33959138597459965, + "grad_norm": 0.8588722944259644, + "learning_rate": 4.976792925614619e-06, + "loss": 0.7335, + "step": 410 + }, + { + "epoch": 0.34041965764770843, + "grad_norm": 0.8786173462867737, + "learning_rate": 4.9766431950142005e-06, + "loss": 0.7254, + "step": 411 + }, + { + "epoch": 0.3412479293208172, + "grad_norm": 0.9019079804420471, + "learning_rate": 4.976492985206412e-06, + "loss": 0.715, + "step": 412 + }, + { + "epoch": 0.342076200993926, + "grad_norm": 0.878132700920105, + "learning_rate": 4.976342296220316e-06, + "loss": 0.7063, + "step": 413 + }, + { + "epoch": 0.3429044726670348, + "grad_norm": 0.8722684979438782, + "learning_rate": 4.976191128085069e-06, + "loss": 0.7481, + "step": 414 + }, + { + "epoch": 0.34373274434014356, + "grad_norm": 0.9076220393180847, + "learning_rate": 4.976039480829922e-06, + "loss": 0.7197, + "step": 415 + }, + { + "epoch": 0.34456101601325234, + "grad_norm": 0.8674967288970947, + "learning_rate": 4.975887354484217e-06, + "loss": 0.7282, + "step": 416 + }, + { + "epoch": 0.3453892876863611, + "grad_norm": 0.9406816959381104, + "learning_rate": 4.9757347490773885e-06, + "loss": 0.7222, + "step": 417 + }, + { + "epoch": 0.3462175593594699, + "grad_norm": 0.8747051358222961, + "learning_rate": 4.975581664638965e-06, + "loss": 0.7283, + "step": 418 + }, + { + "epoch": 0.3470458310325787, + "grad_norm": 1.0587687492370605, + "learning_rate": 4.9754281011985666e-06, + "loss": 0.7272, + "step": 419 + }, + { + "epoch": 0.34787410270568747, + "grad_norm": 0.8596946597099304, + "learning_rate": 4.975274058785906e-06, + "loss": 0.7407, + "step": 420 + }, + { + "epoch": 0.34870237437879625, + "grad_norm": 0.9230688214302063, + "learning_rate": 4.975119537430789e-06, + "loss": 0.7303, + "step": 421 + }, + { + "epoch": 0.34953064605190504, + "grad_norm": 0.9276477694511414, + "learning_rate": 4.974964537163115e-06, + "loss": 0.7296, + "step": 422 + }, + { + "epoch": 0.3503589177250138, + "grad_norm": 0.8969418406486511, + "learning_rate": 4.974809058012874e-06, + "loss": 0.7275, + "step": 423 + }, + { + "epoch": 0.3511871893981226, + "grad_norm": 0.8454467058181763, + "learning_rate": 4.97465310001015e-06, + "loss": 0.7144, + "step": 424 + }, + { + "epoch": 0.3520154610712314, + "grad_norm": 0.8833656907081604, + "learning_rate": 4.97449666318512e-06, + "loss": 0.7073, + "step": 425 + }, + { + "epoch": 0.35284373274434017, + "grad_norm": 0.9251653552055359, + "learning_rate": 4.974339747568052e-06, + "loss": 0.7351, + "step": 426 + }, + { + "epoch": 0.35367200441744895, + "grad_norm": 0.8938575983047485, + "learning_rate": 4.974182353189309e-06, + "loss": 0.7031, + "step": 427 + }, + { + "epoch": 0.3545002760905577, + "grad_norm": 0.8487107753753662, + "learning_rate": 4.974024480079343e-06, + "loss": 0.704, + "step": 428 + }, + { + "epoch": 0.35532854776366646, + "grad_norm": 0.8515727519989014, + "learning_rate": 4.973866128268705e-06, + "loss": 0.7092, + "step": 429 + }, + { + "epoch": 0.35615681943677524, + "grad_norm": 0.8869542479515076, + "learning_rate": 4.97370729778803e-06, + "loss": 0.7099, + "step": 430 + }, + { + "epoch": 0.356985091109884, + "grad_norm": 0.8830757141113281, + "learning_rate": 4.973547988668055e-06, + "loss": 0.7294, + "step": 431 + }, + { + "epoch": 0.3578133627829928, + "grad_norm": 0.9321587681770325, + "learning_rate": 4.973388200939601e-06, + "loss": 0.7276, + "step": 432 + }, + { + "epoch": 0.3586416344561016, + "grad_norm": 0.9408143758773804, + "learning_rate": 4.9732279346335866e-06, + "loss": 0.7216, + "step": 433 + }, + { + "epoch": 0.35946990612921037, + "grad_norm": 0.9189308285713196, + "learning_rate": 4.973067189781022e-06, + "loss": 0.7079, + "step": 434 + }, + { + "epoch": 0.36029817780231915, + "grad_norm": 0.9198120832443237, + "learning_rate": 4.97290596641301e-06, + "loss": 0.7188, + "step": 435 + }, + { + "epoch": 0.36112644947542794, + "grad_norm": 0.9123285412788391, + "learning_rate": 4.972744264560746e-06, + "loss": 0.7328, + "step": 436 + }, + { + "epoch": 0.3619547211485367, + "grad_norm": 0.8722008466720581, + "learning_rate": 4.972582084255517e-06, + "loss": 0.7228, + "step": 437 + }, + { + "epoch": 0.3627829928216455, + "grad_norm": 1.035325288772583, + "learning_rate": 4.972419425528704e-06, + "loss": 0.72, + "step": 438 + }, + { + "epoch": 0.3636112644947543, + "grad_norm": 0.9577400088310242, + "learning_rate": 4.97225628841178e-06, + "loss": 0.7382, + "step": 439 + }, + { + "epoch": 0.36443953616786307, + "grad_norm": 0.9206854104995728, + "learning_rate": 4.972092672936309e-06, + "loss": 0.7294, + "step": 440 + }, + { + "epoch": 0.36526780784097185, + "grad_norm": 0.9281194806098938, + "learning_rate": 4.971928579133952e-06, + "loss": 0.7137, + "step": 441 + }, + { + "epoch": 0.36609607951408063, + "grad_norm": 0.951049268245697, + "learning_rate": 4.971764007036457e-06, + "loss": 0.7354, + "step": 442 + }, + { + "epoch": 0.3669243511871894, + "grad_norm": 0.8779566287994385, + "learning_rate": 4.971598956675668e-06, + "loss": 0.6858, + "step": 443 + }, + { + "epoch": 0.3677526228602982, + "grad_norm": 0.9221998453140259, + "learning_rate": 4.971433428083522e-06, + "loss": 0.7116, + "step": 444 + }, + { + "epoch": 0.368580894533407, + "grad_norm": 0.9608840346336365, + "learning_rate": 4.971267421292045e-06, + "loss": 0.7129, + "step": 445 + }, + { + "epoch": 0.36940916620651576, + "grad_norm": 0.8761363625526428, + "learning_rate": 4.971100936333359e-06, + "loss": 0.718, + "step": 446 + }, + { + "epoch": 0.37023743787962454, + "grad_norm": 0.8808798789978027, + "learning_rate": 4.970933973239677e-06, + "loss": 0.6986, + "step": 447 + }, + { + "epoch": 0.37106570955273327, + "grad_norm": 0.8818857669830322, + "learning_rate": 4.9707665320433055e-06, + "loss": 0.7465, + "step": 448 + }, + { + "epoch": 0.37189398122584205, + "grad_norm": 0.9144098162651062, + "learning_rate": 4.970598612776641e-06, + "loss": 0.7187, + "step": 449 + }, + { + "epoch": 0.37272225289895083, + "grad_norm": 0.8439158797264099, + "learning_rate": 4.9704302154721776e-06, + "loss": 0.7195, + "step": 450 + }, + { + "epoch": 0.3735505245720596, + "grad_norm": 0.8661074638366699, + "learning_rate": 4.970261340162495e-06, + "loss": 0.712, + "step": 451 + }, + { + "epoch": 0.3743787962451684, + "grad_norm": 0.8818789124488831, + "learning_rate": 4.970091986880271e-06, + "loss": 0.7199, + "step": 452 + }, + { + "epoch": 0.3752070679182772, + "grad_norm": 0.9560503363609314, + "learning_rate": 4.969922155658273e-06, + "loss": 0.7145, + "step": 453 + }, + { + "epoch": 0.37603533959138596, + "grad_norm": 0.8723775744438171, + "learning_rate": 4.969751846529363e-06, + "loss": 0.7152, + "step": 454 + }, + { + "epoch": 0.37686361126449475, + "grad_norm": 0.8612253069877625, + "learning_rate": 4.9695810595264935e-06, + "loss": 0.728, + "step": 455 + }, + { + "epoch": 0.37769188293760353, + "grad_norm": 0.8995444178581238, + "learning_rate": 4.969409794682709e-06, + "loss": 0.678, + "step": 456 + }, + { + "epoch": 0.3785201546107123, + "grad_norm": 0.8770595788955688, + "learning_rate": 4.96923805203115e-06, + "loss": 0.7354, + "step": 457 + }, + { + "epoch": 0.3793484262838211, + "grad_norm": 0.893329918384552, + "learning_rate": 4.969065831605045e-06, + "loss": 0.7102, + "step": 458 + }, + { + "epoch": 0.3801766979569299, + "grad_norm": 0.954058051109314, + "learning_rate": 4.968893133437718e-06, + "loss": 0.7144, + "step": 459 + }, + { + "epoch": 0.38100496963003866, + "grad_norm": 0.9711002707481384, + "learning_rate": 4.968719957562585e-06, + "loss": 0.7452, + "step": 460 + }, + { + "epoch": 0.38183324130314744, + "grad_norm": 0.9036396741867065, + "learning_rate": 4.968546304013153e-06, + "loss": 0.7535, + "step": 461 + }, + { + "epoch": 0.3826615129762562, + "grad_norm": 0.8948055505752563, + "learning_rate": 4.968372172823023e-06, + "loss": 0.7284, + "step": 462 + }, + { + "epoch": 0.383489784649365, + "grad_norm": 0.9134189486503601, + "learning_rate": 4.968197564025886e-06, + "loss": 0.723, + "step": 463 + }, + { + "epoch": 0.3843180563224738, + "grad_norm": 0.9191608428955078, + "learning_rate": 4.9680224776555305e-06, + "loss": 0.7078, + "step": 464 + }, + { + "epoch": 0.38514632799558257, + "grad_norm": 0.879683792591095, + "learning_rate": 4.9678469137458315e-06, + "loss": 0.6805, + "step": 465 + }, + { + "epoch": 0.38597459966869135, + "grad_norm": 0.8887293338775635, + "learning_rate": 4.96767087233076e-06, + "loss": 0.7296, + "step": 466 + }, + { + "epoch": 0.38680287134180014, + "grad_norm": 0.889991044998169, + "learning_rate": 4.967494353444377e-06, + "loss": 0.702, + "step": 467 + }, + { + "epoch": 0.38763114301490886, + "grad_norm": 0.9667506814002991, + "learning_rate": 4.967317357120839e-06, + "loss": 0.7305, + "step": 468 + }, + { + "epoch": 0.38845941468801765, + "grad_norm": 0.9297274351119995, + "learning_rate": 4.967139883394392e-06, + "loss": 0.7344, + "step": 469 + }, + { + "epoch": 0.38928768636112643, + "grad_norm": 0.9167365431785583, + "learning_rate": 4.966961932299377e-06, + "loss": 0.7016, + "step": 470 + }, + { + "epoch": 0.3901159580342352, + "grad_norm": 0.877378523349762, + "learning_rate": 4.966783503870224e-06, + "loss": 0.7313, + "step": 471 + }, + { + "epoch": 0.390944229707344, + "grad_norm": 0.9625933766365051, + "learning_rate": 4.966604598141458e-06, + "loss": 0.7245, + "step": 472 + }, + { + "epoch": 0.3917725013804528, + "grad_norm": 0.8859156966209412, + "learning_rate": 4.9664252151476955e-06, + "loss": 0.7094, + "step": 473 + }, + { + "epoch": 0.39260077305356156, + "grad_norm": 0.9172984957695007, + "learning_rate": 4.966245354923645e-06, + "loss": 0.6966, + "step": 474 + }, + { + "epoch": 0.39342904472667034, + "grad_norm": 0.9040864109992981, + "learning_rate": 4.96606501750411e-06, + "loss": 0.722, + "step": 475 + }, + { + "epoch": 0.3942573163997791, + "grad_norm": 0.8900951147079468, + "learning_rate": 4.965884202923982e-06, + "loss": 0.7263, + "step": 476 + }, + { + "epoch": 0.3950855880728879, + "grad_norm": 0.9096625447273254, + "learning_rate": 4.965702911218247e-06, + "loss": 0.702, + "step": 477 + }, + { + "epoch": 0.3959138597459967, + "grad_norm": 0.8801544904708862, + "learning_rate": 4.965521142421984e-06, + "loss": 0.7113, + "step": 478 + }, + { + "epoch": 0.39674213141910547, + "grad_norm": 0.9054175615310669, + "learning_rate": 4.965338896570363e-06, + "loss": 0.7135, + "step": 479 + }, + { + "epoch": 0.39757040309221425, + "grad_norm": 0.8734445571899414, + "learning_rate": 4.965156173698648e-06, + "loss": 0.7181, + "step": 480 + }, + { + "epoch": 0.39839867476532304, + "grad_norm": 0.9113215208053589, + "learning_rate": 4.964972973842191e-06, + "loss": 0.7008, + "step": 481 + }, + { + "epoch": 0.3992269464384318, + "grad_norm": 0.8941736221313477, + "learning_rate": 4.964789297036444e-06, + "loss": 0.7201, + "step": 482 + }, + { + "epoch": 0.4000552181115406, + "grad_norm": 0.8555464148521423, + "learning_rate": 4.964605143316944e-06, + "loss": 0.7173, + "step": 483 + }, + { + "epoch": 0.4008834897846494, + "grad_norm": 0.8719691634178162, + "learning_rate": 4.9644205127193225e-06, + "loss": 0.7211, + "step": 484 + }, + { + "epoch": 0.40171176145775817, + "grad_norm": 0.901626467704773, + "learning_rate": 4.964235405279306e-06, + "loss": 0.7223, + "step": 485 + }, + { + "epoch": 0.40254003313086695, + "grad_norm": 0.856769859790802, + "learning_rate": 4.9640498210327095e-06, + "loss": 0.7103, + "step": 486 + }, + { + "epoch": 0.40336830480397573, + "grad_norm": 0.9053338766098022, + "learning_rate": 4.963863760015442e-06, + "loss": 0.7301, + "step": 487 + }, + { + "epoch": 0.40419657647708446, + "grad_norm": 0.8989829421043396, + "learning_rate": 4.963677222263505e-06, + "loss": 0.7391, + "step": 488 + }, + { + "epoch": 0.40502484815019324, + "grad_norm": 0.9187886118888855, + "learning_rate": 4.963490207812991e-06, + "loss": 0.7179, + "step": 489 + }, + { + "epoch": 0.405853119823302, + "grad_norm": 0.8712778687477112, + "learning_rate": 4.963302716700087e-06, + "loss": 0.7088, + "step": 490 + }, + { + "epoch": 0.4066813914964108, + "grad_norm": 1.0485447645187378, + "learning_rate": 4.9631147489610695e-06, + "loss": 0.7068, + "step": 491 + }, + { + "epoch": 0.4075096631695196, + "grad_norm": 0.9332602024078369, + "learning_rate": 4.962926304632309e-06, + "loss": 0.7201, + "step": 492 + }, + { + "epoch": 0.40833793484262837, + "grad_norm": 0.9264400601387024, + "learning_rate": 4.962737383750268e-06, + "loss": 0.7136, + "step": 493 + }, + { + "epoch": 0.40916620651573715, + "grad_norm": 0.8812651634216309, + "learning_rate": 4.962547986351501e-06, + "loss": 0.7315, + "step": 494 + }, + { + "epoch": 0.40999447818884593, + "grad_norm": 0.8905262351036072, + "learning_rate": 4.962358112472653e-06, + "loss": 0.7127, + "step": 495 + }, + { + "epoch": 0.4108227498619547, + "grad_norm": 0.887514591217041, + "learning_rate": 4.962167762150465e-06, + "loss": 0.7602, + "step": 496 + }, + { + "epoch": 0.4116510215350635, + "grad_norm": 0.8683431148529053, + "learning_rate": 4.9619769354217665e-06, + "loss": 0.7091, + "step": 497 + }, + { + "epoch": 0.4124792932081723, + "grad_norm": 0.944595217704773, + "learning_rate": 4.961785632323481e-06, + "loss": 0.708, + "step": 498 + }, + { + "epoch": 0.41330756488128106, + "grad_norm": 0.8834611773490906, + "learning_rate": 4.961593852892625e-06, + "loss": 0.7308, + "step": 499 + }, + { + "epoch": 0.41413583655438985, + "grad_norm": 0.8479524850845337, + "learning_rate": 4.961401597166304e-06, + "loss": 0.7061, + "step": 500 + }, + { + "epoch": 0.41496410822749863, + "grad_norm": 0.9092981219291687, + "learning_rate": 4.96120886518172e-06, + "loss": 0.7176, + "step": 501 + }, + { + "epoch": 0.4157923799006074, + "grad_norm": 0.9024932384490967, + "learning_rate": 4.961015656976163e-06, + "loss": 0.6944, + "step": 502 + }, + { + "epoch": 0.4166206515737162, + "grad_norm": 0.8945576548576355, + "learning_rate": 4.960821972587018e-06, + "loss": 0.6958, + "step": 503 + }, + { + "epoch": 0.417448923246825, + "grad_norm": 0.8668336868286133, + "learning_rate": 4.9606278120517606e-06, + "loss": 0.7219, + "step": 504 + }, + { + "epoch": 0.41827719491993376, + "grad_norm": 0.8831227421760559, + "learning_rate": 4.96043317540796e-06, + "loss": 0.7124, + "step": 505 + }, + { + "epoch": 0.41910546659304254, + "grad_norm": 0.9637073874473572, + "learning_rate": 4.960238062693274e-06, + "loss": 0.7337, + "step": 506 + }, + { + "epoch": 0.4199337382661513, + "grad_norm": 0.8895899653434753, + "learning_rate": 4.960042473945459e-06, + "loss": 0.7459, + "step": 507 + }, + { + "epoch": 0.42076200993926005, + "grad_norm": 0.8762213587760925, + "learning_rate": 4.959846409202356e-06, + "loss": 0.7319, + "step": 508 + }, + { + "epoch": 0.42159028161236883, + "grad_norm": 0.9177393317222595, + "learning_rate": 4.959649868501903e-06, + "loss": 0.7114, + "step": 509 + }, + { + "epoch": 0.4224185532854776, + "grad_norm": 0.88541179895401, + "learning_rate": 4.9594528518821305e-06, + "loss": 0.716, + "step": 510 + }, + { + "epoch": 0.4232468249585864, + "grad_norm": 0.865566611289978, + "learning_rate": 4.959255359381158e-06, + "loss": 0.7098, + "step": 511 + }, + { + "epoch": 0.4240750966316952, + "grad_norm": 0.8629707098007202, + "learning_rate": 4.959057391037197e-06, + "loss": 0.7146, + "step": 512 + }, + { + "epoch": 0.42490336830480396, + "grad_norm": 0.8642273545265198, + "learning_rate": 4.958858946888554e-06, + "loss": 0.7345, + "step": 513 + }, + { + "epoch": 0.42573163997791275, + "grad_norm": 0.9157605171203613, + "learning_rate": 4.958660026973626e-06, + "loss": 0.7177, + "step": 514 + }, + { + "epoch": 0.42655991165102153, + "grad_norm": 0.9108191728591919, + "learning_rate": 4.958460631330903e-06, + "loss": 0.7317, + "step": 515 + }, + { + "epoch": 0.4273881833241303, + "grad_norm": 0.8524913191795349, + "learning_rate": 4.958260759998965e-06, + "loss": 0.7193, + "step": 516 + }, + { + "epoch": 0.4282164549972391, + "grad_norm": 0.9197820425033569, + "learning_rate": 4.958060413016485e-06, + "loss": 0.7127, + "step": 517 + }, + { + "epoch": 0.4290447266703479, + "grad_norm": 0.8744186758995056, + "learning_rate": 4.9578595904222296e-06, + "loss": 0.6968, + "step": 518 + }, + { + "epoch": 0.42987299834345666, + "grad_norm": 0.9071037769317627, + "learning_rate": 4.957658292255055e-06, + "loss": 0.6963, + "step": 519 + }, + { + "epoch": 0.43070127001656544, + "grad_norm": 0.937796950340271, + "learning_rate": 4.95745651855391e-06, + "loss": 0.7193, + "step": 520 + }, + { + "epoch": 0.4315295416896742, + "grad_norm": 0.883471667766571, + "learning_rate": 4.9572542693578375e-06, + "loss": 0.6886, + "step": 521 + }, + { + "epoch": 0.432357813362783, + "grad_norm": 0.8738589882850647, + "learning_rate": 4.9570515447059705e-06, + "loss": 0.7072, + "step": 522 + }, + { + "epoch": 0.4331860850358918, + "grad_norm": 0.9039497971534729, + "learning_rate": 4.9568483446375335e-06, + "loss": 0.7251, + "step": 523 + }, + { + "epoch": 0.43401435670900057, + "grad_norm": 0.896789014339447, + "learning_rate": 4.956644669191845e-06, + "loss": 0.7021, + "step": 524 + }, + { + "epoch": 0.43484262838210935, + "grad_norm": 0.8883570432662964, + "learning_rate": 4.9564405184083126e-06, + "loss": 0.7153, + "step": 525 + }, + { + "epoch": 0.43567090005521814, + "grad_norm": 0.8869197368621826, + "learning_rate": 4.956235892326438e-06, + "loss": 0.7061, + "step": 526 + }, + { + "epoch": 0.4364991717283269, + "grad_norm": 0.9237920641899109, + "learning_rate": 4.9560307909858165e-06, + "loss": 0.6988, + "step": 527 + }, + { + "epoch": 0.43732744340143564, + "grad_norm": 0.9359283447265625, + "learning_rate": 4.955825214426131e-06, + "loss": 0.738, + "step": 528 + }, + { + "epoch": 0.4381557150745444, + "grad_norm": 0.8602650761604309, + "learning_rate": 4.95561916268716e-06, + "loss": 0.709, + "step": 529 + }, + { + "epoch": 0.4389839867476532, + "grad_norm": 0.9013513922691345, + "learning_rate": 4.955412635808771e-06, + "loss": 0.6836, + "step": 530 + }, + { + "epoch": 0.439812258420762, + "grad_norm": 0.9136670827865601, + "learning_rate": 4.955205633830927e-06, + "loss": 0.7263, + "step": 531 + }, + { + "epoch": 0.4406405300938708, + "grad_norm": 0.9013793468475342, + "learning_rate": 4.95499815679368e-06, + "loss": 0.6937, + "step": 532 + }, + { + "epoch": 0.44146880176697956, + "grad_norm": 0.9017112851142883, + "learning_rate": 4.954790204737174e-06, + "loss": 0.6913, + "step": 533 + }, + { + "epoch": 0.44229707344008834, + "grad_norm": 0.8901436924934387, + "learning_rate": 4.954581777701647e-06, + "loss": 0.7241, + "step": 534 + }, + { + "epoch": 0.4431253451131971, + "grad_norm": 0.8882759809494019, + "learning_rate": 4.954372875727427e-06, + "loss": 0.7115, + "step": 535 + }, + { + "epoch": 0.4439536167863059, + "grad_norm": 0.8909885883331299, + "learning_rate": 4.954163498854936e-06, + "loss": 0.7146, + "step": 536 + }, + { + "epoch": 0.4447818884594147, + "grad_norm": 0.8951786756515503, + "learning_rate": 4.953953647124684e-06, + "loss": 0.699, + "step": 537 + }, + { + "epoch": 0.44561016013252347, + "grad_norm": 0.8890933990478516, + "learning_rate": 4.953743320577278e-06, + "loss": 0.7018, + "step": 538 + }, + { + "epoch": 0.44643843180563225, + "grad_norm": 0.910393238067627, + "learning_rate": 4.953532519253412e-06, + "loss": 0.7299, + "step": 539 + }, + { + "epoch": 0.44726670347874103, + "grad_norm": 0.9152857065200806, + "learning_rate": 4.953321243193876e-06, + "loss": 0.6984, + "step": 540 + }, + { + "epoch": 0.4480949751518498, + "grad_norm": 0.9001837372779846, + "learning_rate": 4.9531094924395485e-06, + "loss": 0.727, + "step": 541 + }, + { + "epoch": 0.4489232468249586, + "grad_norm": 0.882060170173645, + "learning_rate": 4.952897267031401e-06, + "loss": 0.7016, + "step": 542 + }, + { + "epoch": 0.4497515184980674, + "grad_norm": 0.8665159344673157, + "learning_rate": 4.9526845670104995e-06, + "loss": 0.7139, + "step": 543 + }, + { + "epoch": 0.45057979017117616, + "grad_norm": 0.9200844764709473, + "learning_rate": 4.952471392417997e-06, + "loss": 0.7016, + "step": 544 + }, + { + "epoch": 0.45140806184428495, + "grad_norm": 0.8852037191390991, + "learning_rate": 4.9522577432951416e-06, + "loss": 0.7048, + "step": 545 + }, + { + "epoch": 0.45223633351739373, + "grad_norm": 0.9003865122795105, + "learning_rate": 4.952043619683274e-06, + "loss": 0.7106, + "step": 546 + }, + { + "epoch": 0.4530646051905025, + "grad_norm": 0.8888446092605591, + "learning_rate": 4.951829021623822e-06, + "loss": 0.7202, + "step": 547 + }, + { + "epoch": 0.45389287686361124, + "grad_norm": 0.8940277099609375, + "learning_rate": 4.9516139491583094e-06, + "loss": 0.7156, + "step": 548 + }, + { + "epoch": 0.45472114853672, + "grad_norm": 0.9066211581230164, + "learning_rate": 4.951398402328353e-06, + "loss": 0.6865, + "step": 549 + }, + { + "epoch": 0.4555494202098288, + "grad_norm": 0.9352865815162659, + "learning_rate": 4.9511823811756565e-06, + "loss": 0.7117, + "step": 550 + }, + { + "epoch": 0.4563776918829376, + "grad_norm": 0.8672269582748413, + "learning_rate": 4.95096588574202e-06, + "loss": 0.7166, + "step": 551 + }, + { + "epoch": 0.45720596355604637, + "grad_norm": 0.8730685114860535, + "learning_rate": 4.95074891606933e-06, + "loss": 0.6998, + "step": 552 + }, + { + "epoch": 0.45803423522915515, + "grad_norm": 0.896821916103363, + "learning_rate": 4.950531472199572e-06, + "loss": 0.6839, + "step": 553 + }, + { + "epoch": 0.45886250690226393, + "grad_norm": 0.918572187423706, + "learning_rate": 4.950313554174817e-06, + "loss": 0.7155, + "step": 554 + }, + { + "epoch": 0.4596907785753727, + "grad_norm": 0.8670752048492432, + "learning_rate": 4.950095162037232e-06, + "loss": 0.708, + "step": 555 + }, + { + "epoch": 0.4605190502484815, + "grad_norm": 0.875618040561676, + "learning_rate": 4.9498762958290715e-06, + "loss": 0.6956, + "step": 556 + }, + { + "epoch": 0.4613473219215903, + "grad_norm": 0.9120676517486572, + "learning_rate": 4.949656955592685e-06, + "loss": 0.7327, + "step": 557 + }, + { + "epoch": 0.46217559359469906, + "grad_norm": 0.9670370817184448, + "learning_rate": 4.949437141370514e-06, + "loss": 0.705, + "step": 558 + }, + { + "epoch": 0.46300386526780785, + "grad_norm": 0.9155322313308716, + "learning_rate": 4.949216853205089e-06, + "loss": 0.7126, + "step": 559 + }, + { + "epoch": 0.46383213694091663, + "grad_norm": 0.8945392370223999, + "learning_rate": 4.948996091139034e-06, + "loss": 0.7083, + "step": 560 + }, + { + "epoch": 0.4646604086140254, + "grad_norm": 0.8922162055969238, + "learning_rate": 4.948774855215065e-06, + "loss": 0.7189, + "step": 561 + }, + { + "epoch": 0.4654886802871342, + "grad_norm": 0.8940657377243042, + "learning_rate": 4.948553145475989e-06, + "loss": 0.6934, + "step": 562 + }, + { + "epoch": 0.466316951960243, + "grad_norm": 0.8946855068206787, + "learning_rate": 4.948330961964703e-06, + "loss": 0.6835, + "step": 563 + }, + { + "epoch": 0.46714522363335176, + "grad_norm": 0.9002820253372192, + "learning_rate": 4.948108304724201e-06, + "loss": 0.712, + "step": 564 + }, + { + "epoch": 0.46797349530646054, + "grad_norm": 0.8715711832046509, + "learning_rate": 4.947885173797561e-06, + "loss": 0.7197, + "step": 565 + }, + { + "epoch": 0.4688017669795693, + "grad_norm": 0.9269874095916748, + "learning_rate": 4.94766156922796e-06, + "loss": 0.7311, + "step": 566 + }, + { + "epoch": 0.4696300386526781, + "grad_norm": 0.8999423980712891, + "learning_rate": 4.947437491058662e-06, + "loss": 0.7144, + "step": 567 + }, + { + "epoch": 0.47045831032578683, + "grad_norm": 0.8704255223274231, + "learning_rate": 4.947212939333024e-06, + "loss": 0.7156, + "step": 568 + }, + { + "epoch": 0.4712865819988956, + "grad_norm": 0.8770347237586975, + "learning_rate": 4.946987914094495e-06, + "loss": 0.6969, + "step": 569 + }, + { + "epoch": 0.4721148536720044, + "grad_norm": 0.8865682482719421, + "learning_rate": 4.946762415386615e-06, + "loss": 0.7089, + "step": 570 + }, + { + "epoch": 0.4729431253451132, + "grad_norm": 0.8756279349327087, + "learning_rate": 4.946536443253016e-06, + "loss": 0.7159, + "step": 571 + }, + { + "epoch": 0.47377139701822196, + "grad_norm": 0.8844839930534363, + "learning_rate": 4.946309997737422e-06, + "loss": 0.7234, + "step": 572 + }, + { + "epoch": 0.47459966869133074, + "grad_norm": 0.8549885153770447, + "learning_rate": 4.946083078883647e-06, + "loss": 0.6927, + "step": 573 + }, + { + "epoch": 0.4754279403644395, + "grad_norm": 0.8967456817626953, + "learning_rate": 4.945855686735599e-06, + "loss": 0.7082, + "step": 574 + }, + { + "epoch": 0.4762562120375483, + "grad_norm": 0.9120200276374817, + "learning_rate": 4.9456278213372755e-06, + "loss": 0.7135, + "step": 575 + }, + { + "epoch": 0.4770844837106571, + "grad_norm": 0.8780766725540161, + "learning_rate": 4.945399482732765e-06, + "loss": 0.7139, + "step": 576 + }, + { + "epoch": 0.4779127553837659, + "grad_norm": 0.8900999426841736, + "learning_rate": 4.945170670966253e-06, + "loss": 0.7132, + "step": 577 + }, + { + "epoch": 0.47874102705687466, + "grad_norm": 0.9402151107788086, + "learning_rate": 4.944941386082008e-06, + "loss": 0.7139, + "step": 578 + }, + { + "epoch": 0.47956929872998344, + "grad_norm": 0.8769465088844299, + "learning_rate": 4.944711628124396e-06, + "loss": 0.7236, + "step": 579 + }, + { + "epoch": 0.4803975704030922, + "grad_norm": 0.9185771346092224, + "learning_rate": 4.944481397137874e-06, + "loss": 0.6884, + "step": 580 + }, + { + "epoch": 0.481225842076201, + "grad_norm": 0.8775897026062012, + "learning_rate": 4.944250693166988e-06, + "loss": 0.731, + "step": 581 + }, + { + "epoch": 0.4820541137493098, + "grad_norm": 0.8751668930053711, + "learning_rate": 4.94401951625638e-06, + "loss": 0.7007, + "step": 582 + }, + { + "epoch": 0.48288238542241857, + "grad_norm": 0.9042515754699707, + "learning_rate": 4.943787866450776e-06, + "loss": 0.7064, + "step": 583 + }, + { + "epoch": 0.48371065709552735, + "grad_norm": 0.8845478892326355, + "learning_rate": 4.943555743795002e-06, + "loss": 0.7069, + "step": 584 + }, + { + "epoch": 0.48453892876863613, + "grad_norm": 0.8959096670150757, + "learning_rate": 4.943323148333969e-06, + "loss": 0.6871, + "step": 585 + }, + { + "epoch": 0.4853672004417449, + "grad_norm": 0.9021669626235962, + "learning_rate": 4.943090080112684e-06, + "loss": 0.7156, + "step": 586 + }, + { + "epoch": 0.4861954721148537, + "grad_norm": 0.8768396973609924, + "learning_rate": 4.942856539176243e-06, + "loss": 0.681, + "step": 587 + }, + { + "epoch": 0.4870237437879624, + "grad_norm": 0.8734706044197083, + "learning_rate": 4.942622525569832e-06, + "loss": 0.7371, + "step": 588 + }, + { + "epoch": 0.4878520154610712, + "grad_norm": 0.9155406355857849, + "learning_rate": 4.942388039338733e-06, + "loss": 0.7012, + "step": 589 + }, + { + "epoch": 0.48868028713418, + "grad_norm": 0.8778123259544373, + "learning_rate": 4.942153080528317e-06, + "loss": 0.7201, + "step": 590 + }, + { + "epoch": 0.4895085588072888, + "grad_norm": 0.9102092981338501, + "learning_rate": 4.941917649184045e-06, + "loss": 0.6951, + "step": 591 + }, + { + "epoch": 0.49033683048039756, + "grad_norm": 0.8855768442153931, + "learning_rate": 4.941681745351471e-06, + "loss": 0.6997, + "step": 592 + }, + { + "epoch": 0.49116510215350634, + "grad_norm": 0.8770707845687866, + "learning_rate": 4.94144536907624e-06, + "loss": 0.7092, + "step": 593 + }, + { + "epoch": 0.4919933738266151, + "grad_norm": 0.900458574295044, + "learning_rate": 4.9412085204040896e-06, + "loss": 0.6909, + "step": 594 + }, + { + "epoch": 0.4928216454997239, + "grad_norm": 0.903489887714386, + "learning_rate": 4.940971199380846e-06, + "loss": 0.7066, + "step": 595 + }, + { + "epoch": 0.4936499171728327, + "grad_norm": 0.9331871271133423, + "learning_rate": 4.940733406052431e-06, + "loss": 0.6924, + "step": 596 + }, + { + "epoch": 0.49447818884594147, + "grad_norm": 0.8855593800544739, + "learning_rate": 4.9404951404648535e-06, + "loss": 0.6991, + "step": 597 + }, + { + "epoch": 0.49530646051905025, + "grad_norm": 0.9075731039047241, + "learning_rate": 4.940256402664216e-06, + "loss": 0.6973, + "step": 598 + }, + { + "epoch": 0.49613473219215903, + "grad_norm": 0.8970743417739868, + "learning_rate": 4.940017192696713e-06, + "loss": 0.6767, + "step": 599 + }, + { + "epoch": 0.4969630038652678, + "grad_norm": 0.8644358515739441, + "learning_rate": 4.939777510608628e-06, + "loss": 0.7284, + "step": 600 + }, + { + "epoch": 0.4977912755383766, + "grad_norm": 0.8807653784751892, + "learning_rate": 4.939537356446339e-06, + "loss": 0.7143, + "step": 601 + }, + { + "epoch": 0.4986195472114854, + "grad_norm": 0.892946183681488, + "learning_rate": 4.939296730256312e-06, + "loss": 0.7021, + "step": 602 + }, + { + "epoch": 0.49944781888459416, + "grad_norm": 0.9661110639572144, + "learning_rate": 4.939055632085107e-06, + "loss": 0.7198, + "step": 603 + }, + { + "epoch": 0.500276090557703, + "grad_norm": 0.8638496398925781, + "learning_rate": 4.9388140619793735e-06, + "loss": 0.6943, + "step": 604 + }, + { + "epoch": 0.5011043622308117, + "grad_norm": 0.9025457501411438, + "learning_rate": 4.938572019985853e-06, + "loss": 0.7114, + "step": 605 + }, + { + "epoch": 0.5019326339039205, + "grad_norm": 0.9622734785079956, + "learning_rate": 4.93832950615138e-06, + "loss": 0.7086, + "step": 606 + }, + { + "epoch": 0.5027609055770292, + "grad_norm": 0.9565468430519104, + "learning_rate": 4.938086520522876e-06, + "loss": 0.7125, + "step": 607 + }, + { + "epoch": 0.5035891772501381, + "grad_norm": 0.8970094323158264, + "learning_rate": 4.937843063147359e-06, + "loss": 0.7023, + "step": 608 + }, + { + "epoch": 0.5044174489232468, + "grad_norm": 0.8920996785163879, + "learning_rate": 4.937599134071933e-06, + "loss": 0.7041, + "step": 609 + }, + { + "epoch": 0.5052457205963556, + "grad_norm": 0.9304755926132202, + "learning_rate": 4.937354733343799e-06, + "loss": 0.7165, + "step": 610 + }, + { + "epoch": 0.5060739922694644, + "grad_norm": 0.8995933532714844, + "learning_rate": 4.937109861010244e-06, + "loss": 0.6907, + "step": 611 + }, + { + "epoch": 0.5069022639425732, + "grad_norm": 0.8648850321769714, + "learning_rate": 4.936864517118649e-06, + "loss": 0.6994, + "step": 612 + }, + { + "epoch": 0.5077305356156819, + "grad_norm": 0.8414506316184998, + "learning_rate": 4.936618701716487e-06, + "loss": 0.7081, + "step": 613 + }, + { + "epoch": 0.5085588072887908, + "grad_norm": 0.9282650947570801, + "learning_rate": 4.936372414851321e-06, + "loss": 0.7303, + "step": 614 + }, + { + "epoch": 0.5093870789618995, + "grad_norm": 0.877757728099823, + "learning_rate": 4.936125656570802e-06, + "loss": 0.7132, + "step": 615 + }, + { + "epoch": 0.5102153506350083, + "grad_norm": 0.8645135164260864, + "learning_rate": 4.935878426922679e-06, + "loss": 0.723, + "step": 616 + }, + { + "epoch": 0.5110436223081171, + "grad_norm": 0.9014747738838196, + "learning_rate": 4.935630725954787e-06, + "loss": 0.7167, + "step": 617 + }, + { + "epoch": 0.5118718939812258, + "grad_norm": 0.9295386672019958, + "learning_rate": 4.935382553715055e-06, + "loss": 0.705, + "step": 618 + }, + { + "epoch": 0.5127001656543346, + "grad_norm": 0.8910616040229797, + "learning_rate": 4.935133910251499e-06, + "loss": 0.6781, + "step": 619 + }, + { + "epoch": 0.5135284373274434, + "grad_norm": 0.8804931640625, + "learning_rate": 4.934884795612234e-06, + "loss": 0.6984, + "step": 620 + }, + { + "epoch": 0.5143567090005522, + "grad_norm": 0.8223545551300049, + "learning_rate": 4.934635209845458e-06, + "loss": 0.6892, + "step": 621 + }, + { + "epoch": 0.5151849806736609, + "grad_norm": 0.8509450554847717, + "learning_rate": 4.934385152999463e-06, + "loss": 0.7139, + "step": 622 + }, + { + "epoch": 0.5160132523467698, + "grad_norm": 0.8944748044013977, + "learning_rate": 4.934134625122636e-06, + "loss": 0.6976, + "step": 623 + }, + { + "epoch": 0.5168415240198785, + "grad_norm": 0.9021086692810059, + "learning_rate": 4.933883626263448e-06, + "loss": 0.7189, + "step": 624 + }, + { + "epoch": 0.5176697956929873, + "grad_norm": 0.8939547538757324, + "learning_rate": 4.933632156470467e-06, + "loss": 0.704, + "step": 625 + }, + { + "epoch": 0.518498067366096, + "grad_norm": 0.8793498277664185, + "learning_rate": 4.93338021579235e-06, + "loss": 0.7046, + "step": 626 + }, + { + "epoch": 0.5193263390392049, + "grad_norm": 0.8728756308555603, + "learning_rate": 4.933127804277845e-06, + "loss": 0.6826, + "step": 627 + }, + { + "epoch": 0.5201546107123136, + "grad_norm": 0.8989412188529968, + "learning_rate": 4.932874921975792e-06, + "loss": 0.6842, + "step": 628 + }, + { + "epoch": 0.5209828823854225, + "grad_norm": 0.8836191296577454, + "learning_rate": 4.93262156893512e-06, + "loss": 0.7061, + "step": 629 + }, + { + "epoch": 0.5218111540585312, + "grad_norm": 0.9429040551185608, + "learning_rate": 4.9323677452048505e-06, + "loss": 0.7053, + "step": 630 + }, + { + "epoch": 0.52263942573164, + "grad_norm": 0.879463255405426, + "learning_rate": 4.932113450834097e-06, + "loss": 0.6926, + "step": 631 + }, + { + "epoch": 0.5234676974047487, + "grad_norm": 0.8566175699234009, + "learning_rate": 4.931858685872062e-06, + "loss": 0.7126, + "step": 632 + }, + { + "epoch": 0.5242959690778576, + "grad_norm": 0.8694538474082947, + "learning_rate": 4.931603450368042e-06, + "loss": 0.7153, + "step": 633 + }, + { + "epoch": 0.5251242407509663, + "grad_norm": 0.8754484057426453, + "learning_rate": 4.931347744371421e-06, + "loss": 0.711, + "step": 634 + }, + { + "epoch": 0.5259525124240751, + "grad_norm": 0.8982204794883728, + "learning_rate": 4.931091567931675e-06, + "loss": 0.7168, + "step": 635 + }, + { + "epoch": 0.5267807840971839, + "grad_norm": 0.9262725710868835, + "learning_rate": 4.930834921098374e-06, + "loss": 0.7315, + "step": 636 + }, + { + "epoch": 0.5276090557702926, + "grad_norm": 0.9271771311759949, + "learning_rate": 4.930577803921176e-06, + "loss": 0.6968, + "step": 637 + }, + { + "epoch": 0.5284373274434014, + "grad_norm": 0.8855197429656982, + "learning_rate": 4.93032021644983e-06, + "loss": 0.6986, + "step": 638 + }, + { + "epoch": 0.5292655991165102, + "grad_norm": 0.8615745902061462, + "learning_rate": 4.930062158734178e-06, + "loss": 0.7014, + "step": 639 + }, + { + "epoch": 0.530093870789619, + "grad_norm": 0.9004185795783997, + "learning_rate": 4.92980363082415e-06, + "loss": 0.7045, + "step": 640 + }, + { + "epoch": 0.5309221424627277, + "grad_norm": 0.8533529043197632, + "learning_rate": 4.92954463276977e-06, + "loss": 0.6911, + "step": 641 + }, + { + "epoch": 0.5317504141358366, + "grad_norm": 0.881901204586029, + "learning_rate": 4.929285164621152e-06, + "loss": 0.7148, + "step": 642 + }, + { + "epoch": 0.5325786858089453, + "grad_norm": 0.8778982162475586, + "learning_rate": 4.9290252264285e-06, + "loss": 0.6879, + "step": 643 + }, + { + "epoch": 0.5334069574820541, + "grad_norm": 0.9455735087394714, + "learning_rate": 4.928764818242109e-06, + "loss": 0.7088, + "step": 644 + }, + { + "epoch": 0.5342352291551629, + "grad_norm": 0.9255897998809814, + "learning_rate": 4.928503940112367e-06, + "loss": 0.6817, + "step": 645 + }, + { + "epoch": 0.5350635008282717, + "grad_norm": 0.8854040503501892, + "learning_rate": 4.928242592089752e-06, + "loss": 0.7074, + "step": 646 + }, + { + "epoch": 0.5358917725013804, + "grad_norm": 0.8824400901794434, + "learning_rate": 4.927980774224829e-06, + "loss": 0.7092, + "step": 647 + }, + { + "epoch": 0.5367200441744893, + "grad_norm": 0.8873158693313599, + "learning_rate": 4.927718486568261e-06, + "loss": 0.7214, + "step": 648 + }, + { + "epoch": 0.537548315847598, + "grad_norm": 0.9013757109642029, + "learning_rate": 4.927455729170796e-06, + "loss": 0.6955, + "step": 649 + }, + { + "epoch": 0.5383765875207068, + "grad_norm": 0.9100024104118347, + "learning_rate": 4.927192502083276e-06, + "loss": 0.7089, + "step": 650 + }, + { + "epoch": 0.5392048591938156, + "grad_norm": 0.8935851454734802, + "learning_rate": 4.9269288053566335e-06, + "loss": 0.7018, + "step": 651 + }, + { + "epoch": 0.5400331308669244, + "grad_norm": 0.8726267218589783, + "learning_rate": 4.92666463904189e-06, + "loss": 0.6829, + "step": 652 + }, + { + "epoch": 0.5408614025400331, + "grad_norm": 0.9320974349975586, + "learning_rate": 4.926400003190161e-06, + "loss": 0.6766, + "step": 653 + }, + { + "epoch": 0.541689674213142, + "grad_norm": 0.9257067441940308, + "learning_rate": 4.926134897852648e-06, + "loss": 0.6913, + "step": 654 + }, + { + "epoch": 0.5425179458862507, + "grad_norm": 0.8642577528953552, + "learning_rate": 4.92586932308065e-06, + "loss": 0.7155, + "step": 655 + }, + { + "epoch": 0.5433462175593595, + "grad_norm": 0.8861536383628845, + "learning_rate": 4.9256032789255505e-06, + "loss": 0.6695, + "step": 656 + }, + { + "epoch": 0.5441744892324683, + "grad_norm": 0.8901777267456055, + "learning_rate": 4.925336765438829e-06, + "loss": 0.6847, + "step": 657 + }, + { + "epoch": 0.545002760905577, + "grad_norm": 0.8810168504714966, + "learning_rate": 4.9250697826720515e-06, + "loss": 0.7038, + "step": 658 + }, + { + "epoch": 0.5458310325786858, + "grad_norm": 0.8655433654785156, + "learning_rate": 4.924802330676877e-06, + "loss": 0.6908, + "step": 659 + }, + { + "epoch": 0.5466593042517945, + "grad_norm": 0.8786777853965759, + "learning_rate": 4.924534409505055e-06, + "loss": 0.663, + "step": 660 + }, + { + "epoch": 0.5474875759249034, + "grad_norm": 0.8837904334068298, + "learning_rate": 4.924266019208427e-06, + "loss": 0.6843, + "step": 661 + }, + { + "epoch": 0.5483158475980121, + "grad_norm": 0.8715303540229797, + "learning_rate": 4.923997159838923e-06, + "loss": 0.6841, + "step": 662 + }, + { + "epoch": 0.549144119271121, + "grad_norm": 0.8825753927230835, + "learning_rate": 4.923727831448564e-06, + "loss": 0.7053, + "step": 663 + }, + { + "epoch": 0.5499723909442297, + "grad_norm": 0.877668559551239, + "learning_rate": 4.923458034089463e-06, + "loss": 0.6624, + "step": 664 + }, + { + "epoch": 0.5508006626173385, + "grad_norm": 0.8721411824226379, + "learning_rate": 4.923187767813825e-06, + "loss": 0.7339, + "step": 665 + }, + { + "epoch": 0.5516289342904472, + "grad_norm": 0.8607308268547058, + "learning_rate": 4.922917032673941e-06, + "loss": 0.7184, + "step": 666 + }, + { + "epoch": 0.5524572059635561, + "grad_norm": 0.8456798791885376, + "learning_rate": 4.922645828722199e-06, + "loss": 0.7056, + "step": 667 + }, + { + "epoch": 0.5532854776366648, + "grad_norm": 0.8557775020599365, + "learning_rate": 4.922374156011071e-06, + "loss": 0.6942, + "step": 668 + }, + { + "epoch": 0.5541137493097736, + "grad_norm": 0.8877674341201782, + "learning_rate": 4.922102014593126e-06, + "loss": 0.7173, + "step": 669 + }, + { + "epoch": 0.5549420209828824, + "grad_norm": 0.8662481904029846, + "learning_rate": 4.921829404521019e-06, + "loss": 0.7113, + "step": 670 + }, + { + "epoch": 0.5557702926559912, + "grad_norm": 0.8887649774551392, + "learning_rate": 4.921556325847499e-06, + "loss": 0.7028, + "step": 671 + }, + { + "epoch": 0.5565985643290999, + "grad_norm": 0.9837973713874817, + "learning_rate": 4.921282778625403e-06, + "loss": 0.6975, + "step": 672 + }, + { + "epoch": 0.5574268360022088, + "grad_norm": 0.8754701018333435, + "learning_rate": 4.921008762907661e-06, + "loss": 0.6983, + "step": 673 + }, + { + "epoch": 0.5582551076753175, + "grad_norm": 0.9125426411628723, + "learning_rate": 4.920734278747291e-06, + "loss": 0.6982, + "step": 674 + }, + { + "epoch": 0.5590833793484263, + "grad_norm": 0.9424439072608948, + "learning_rate": 4.920459326197403e-06, + "loss": 0.7304, + "step": 675 + }, + { + "epoch": 0.5599116510215351, + "grad_norm": 0.8718393445014954, + "learning_rate": 4.9201839053112e-06, + "loss": 0.6644, + "step": 676 + }, + { + "epoch": 0.5607399226946438, + "grad_norm": 0.9002220630645752, + "learning_rate": 4.919908016141972e-06, + "loss": 0.6966, + "step": 677 + }, + { + "epoch": 0.5615681943677526, + "grad_norm": 0.8925678133964539, + "learning_rate": 4.9196316587431e-06, + "loss": 0.713, + "step": 678 + }, + { + "epoch": 0.5623964660408614, + "grad_norm": 0.8854464292526245, + "learning_rate": 4.919354833168059e-06, + "loss": 0.7075, + "step": 679 + }, + { + "epoch": 0.5632247377139702, + "grad_norm": 0.8867958188056946, + "learning_rate": 4.919077539470409e-06, + "loss": 0.7166, + "step": 680 + }, + { + "epoch": 0.5640530093870789, + "grad_norm": 0.862531840801239, + "learning_rate": 4.918799777703806e-06, + "loss": 0.7119, + "step": 681 + }, + { + "epoch": 0.5648812810601878, + "grad_norm": 0.8623220920562744, + "learning_rate": 4.9185215479219944e-06, + "loss": 0.6758, + "step": 682 + }, + { + "epoch": 0.5657095527332965, + "grad_norm": 0.9171786904335022, + "learning_rate": 4.918242850178808e-06, + "loss": 0.7049, + "step": 683 + }, + { + "epoch": 0.5665378244064053, + "grad_norm": 0.8947017192840576, + "learning_rate": 4.9179636845281735e-06, + "loss": 0.7144, + "step": 684 + }, + { + "epoch": 0.567366096079514, + "grad_norm": 0.9343388080596924, + "learning_rate": 4.917684051024104e-06, + "loss": 0.7024, + "step": 685 + }, + { + "epoch": 0.5681943677526229, + "grad_norm": 0.8929025530815125, + "learning_rate": 4.917403949720709e-06, + "loss": 0.7162, + "step": 686 + }, + { + "epoch": 0.5690226394257316, + "grad_norm": 0.9000421762466431, + "learning_rate": 4.9171233806721854e-06, + "loss": 0.7085, + "step": 687 + }, + { + "epoch": 0.5698509110988405, + "grad_norm": 0.9404364228248596, + "learning_rate": 4.91684234393282e-06, + "loss": 0.6996, + "step": 688 + }, + { + "epoch": 0.5706791827719492, + "grad_norm": 0.9566411375999451, + "learning_rate": 4.916560839556989e-06, + "loss": 0.7241, + "step": 689 + }, + { + "epoch": 0.571507454445058, + "grad_norm": 0.9066181182861328, + "learning_rate": 4.916278867599163e-06, + "loss": 0.724, + "step": 690 + }, + { + "epoch": 0.5723357261181667, + "grad_norm": 0.8914163708686829, + "learning_rate": 4.915996428113901e-06, + "loss": 0.7046, + "step": 691 + }, + { + "epoch": 0.5731639977912756, + "grad_norm": 0.8977548480033875, + "learning_rate": 4.91571352115585e-06, + "loss": 0.6823, + "step": 692 + }, + { + "epoch": 0.5739922694643843, + "grad_norm": 0.8902036547660828, + "learning_rate": 4.915430146779754e-06, + "loss": 0.702, + "step": 693 + }, + { + "epoch": 0.5748205411374931, + "grad_norm": 0.8859509229660034, + "learning_rate": 4.915146305040439e-06, + "loss": 0.6996, + "step": 694 + }, + { + "epoch": 0.5756488128106019, + "grad_norm": 0.8916864991188049, + "learning_rate": 4.914861995992829e-06, + "loss": 0.6978, + "step": 695 + }, + { + "epoch": 0.5764770844837107, + "grad_norm": 0.8961358070373535, + "learning_rate": 4.9145772196919336e-06, + "loss": 0.6803, + "step": 696 + }, + { + "epoch": 0.5773053561568194, + "grad_norm": 0.8817684650421143, + "learning_rate": 4.914291976192854e-06, + "loss": 0.6891, + "step": 697 + }, + { + "epoch": 0.5781336278299282, + "grad_norm": 0.8821113705635071, + "learning_rate": 4.914006265550783e-06, + "loss": 0.7009, + "step": 698 + }, + { + "epoch": 0.578961899503037, + "grad_norm": 0.9553062319755554, + "learning_rate": 4.913720087821003e-06, + "loss": 0.7118, + "step": 699 + }, + { + "epoch": 0.5797901711761457, + "grad_norm": 0.8997507691383362, + "learning_rate": 4.913433443058887e-06, + "loss": 0.7036, + "step": 700 + }, + { + "epoch": 0.5806184428492546, + "grad_norm": 0.8659473657608032, + "learning_rate": 4.9131463313198964e-06, + "loss": 0.6766, + "step": 701 + }, + { + "epoch": 0.5814467145223633, + "grad_norm": 0.8916030526161194, + "learning_rate": 4.912858752659586e-06, + "loss": 0.6976, + "step": 702 + }, + { + "epoch": 0.5822749861954721, + "grad_norm": 0.8726498484611511, + "learning_rate": 4.912570707133599e-06, + "loss": 0.6989, + "step": 703 + }, + { + "epoch": 0.5831032578685809, + "grad_norm": 0.8725529909133911, + "learning_rate": 4.91228219479767e-06, + "loss": 0.6862, + "step": 704 + }, + { + "epoch": 0.5839315295416897, + "grad_norm": 0.8807168006896973, + "learning_rate": 4.911993215707624e-06, + "loss": 0.6964, + "step": 705 + }, + { + "epoch": 0.5847598012147984, + "grad_norm": 0.9244167804718018, + "learning_rate": 4.911703769919375e-06, + "loss": 0.6742, + "step": 706 + }, + { + "epoch": 0.5855880728879073, + "grad_norm": 0.9174355268478394, + "learning_rate": 4.9114138574889285e-06, + "loss": 0.6852, + "step": 707 + }, + { + "epoch": 0.586416344561016, + "grad_norm": 0.8894474506378174, + "learning_rate": 4.9111234784723785e-06, + "loss": 0.689, + "step": 708 + }, + { + "epoch": 0.5872446162341248, + "grad_norm": 0.8856050968170166, + "learning_rate": 4.910832632925912e-06, + "loss": 0.6742, + "step": 709 + }, + { + "epoch": 0.5880728879072336, + "grad_norm": 0.9120550751686096, + "learning_rate": 4.910541320905804e-06, + "loss": 0.7016, + "step": 710 + }, + { + "epoch": 0.5889011595803424, + "grad_norm": 0.8925642967224121, + "learning_rate": 4.910249542468422e-06, + "loss": 0.7165, + "step": 711 + }, + { + "epoch": 0.5897294312534511, + "grad_norm": 0.8786419630050659, + "learning_rate": 4.909957297670222e-06, + "loss": 0.7207, + "step": 712 + }, + { + "epoch": 0.59055770292656, + "grad_norm": 0.89600670337677, + "learning_rate": 4.909664586567749e-06, + "loss": 0.672, + "step": 713 + }, + { + "epoch": 0.5913859745996687, + "grad_norm": 0.889835000038147, + "learning_rate": 4.909371409217642e-06, + "loss": 0.7089, + "step": 714 + }, + { + "epoch": 0.5922142462727775, + "grad_norm": 0.9029288291931152, + "learning_rate": 4.909077765676627e-06, + "loss": 0.7065, + "step": 715 + }, + { + "epoch": 0.5930425179458862, + "grad_norm": 0.9212722778320312, + "learning_rate": 4.908783656001521e-06, + "loss": 0.6787, + "step": 716 + }, + { + "epoch": 0.593870789618995, + "grad_norm": 0.9079681038856506, + "learning_rate": 4.9084890802492325e-06, + "loss": 0.6996, + "step": 717 + }, + { + "epoch": 0.5946990612921038, + "grad_norm": 0.9110735654830933, + "learning_rate": 4.908194038476758e-06, + "loss": 0.6962, + "step": 718 + }, + { + "epoch": 0.5955273329652125, + "grad_norm": 0.8945547938346863, + "learning_rate": 4.9078985307411865e-06, + "loss": 0.6648, + "step": 719 + }, + { + "epoch": 0.5963556046383214, + "grad_norm": 0.8831511735916138, + "learning_rate": 4.907602557099696e-06, + "loss": 0.7077, + "step": 720 + }, + { + "epoch": 0.5971838763114301, + "grad_norm": 0.9393531680107117, + "learning_rate": 4.9073061176095525e-06, + "loss": 0.6722, + "step": 721 + }, + { + "epoch": 0.5980121479845389, + "grad_norm": 0.8807207942008972, + "learning_rate": 4.907009212328117e-06, + "loss": 0.69, + "step": 722 + }, + { + "epoch": 0.5988404196576477, + "grad_norm": 0.8811002969741821, + "learning_rate": 4.906711841312836e-06, + "loss": 0.6902, + "step": 723 + }, + { + "epoch": 0.5996686913307565, + "grad_norm": 0.8953952789306641, + "learning_rate": 4.906414004621249e-06, + "loss": 0.6747, + "step": 724 + }, + { + "epoch": 0.6004969630038652, + "grad_norm": 0.8661350607872009, + "learning_rate": 4.9061157023109855e-06, + "loss": 0.7047, + "step": 725 + }, + { + "epoch": 0.6013252346769741, + "grad_norm": 0.8722146153450012, + "learning_rate": 4.905816934439762e-06, + "loss": 0.7231, + "step": 726 + }, + { + "epoch": 0.6021535063500828, + "grad_norm": 0.8933189511299133, + "learning_rate": 4.905517701065389e-06, + "loss": 0.7029, + "step": 727 + }, + { + "epoch": 0.6029817780231916, + "grad_norm": 0.8863442540168762, + "learning_rate": 4.905218002245764e-06, + "loss": 0.6886, + "step": 728 + }, + { + "epoch": 0.6038100496963004, + "grad_norm": 0.8672969341278076, + "learning_rate": 4.904917838038877e-06, + "loss": 0.7048, + "step": 729 + }, + { + "epoch": 0.6046383213694092, + "grad_norm": 0.8892956972122192, + "learning_rate": 4.904617208502807e-06, + "loss": 0.7009, + "step": 730 + }, + { + "epoch": 0.6054665930425179, + "grad_norm": 0.9083442687988281, + "learning_rate": 4.904316113695723e-06, + "loss": 0.7169, + "step": 731 + }, + { + "epoch": 0.6062948647156268, + "grad_norm": 0.8885907530784607, + "learning_rate": 4.904014553675883e-06, + "loss": 0.6939, + "step": 732 + }, + { + "epoch": 0.6071231363887355, + "grad_norm": 0.8859131336212158, + "learning_rate": 4.903712528501637e-06, + "loss": 0.6881, + "step": 733 + }, + { + "epoch": 0.6079514080618443, + "grad_norm": 0.874146580696106, + "learning_rate": 4.903410038231425e-06, + "loss": 0.6857, + "step": 734 + }, + { + "epoch": 0.6087796797349531, + "grad_norm": 0.8719701170921326, + "learning_rate": 4.903107082923774e-06, + "loss": 0.6834, + "step": 735 + }, + { + "epoch": 0.6096079514080619, + "grad_norm": 0.9656700491905212, + "learning_rate": 4.902803662637304e-06, + "loss": 0.6933, + "step": 736 + }, + { + "epoch": 0.6104362230811706, + "grad_norm": 0.9022018313407898, + "learning_rate": 4.9024997774307235e-06, + "loss": 0.683, + "step": 737 + }, + { + "epoch": 0.6112644947542794, + "grad_norm": 0.8694130182266235, + "learning_rate": 4.902195427362832e-06, + "loss": 0.642, + "step": 738 + }, + { + "epoch": 0.6120927664273882, + "grad_norm": 0.9153549075126648, + "learning_rate": 4.901890612492519e-06, + "loss": 0.6674, + "step": 739 + }, + { + "epoch": 0.6129210381004969, + "grad_norm": 0.9431132674217224, + "learning_rate": 4.901585332878762e-06, + "loss": 0.6918, + "step": 740 + }, + { + "epoch": 0.6137493097736058, + "grad_norm": 0.881587028503418, + "learning_rate": 4.901279588580631e-06, + "loss": 0.687, + "step": 741 + }, + { + "epoch": 0.6145775814467145, + "grad_norm": 0.9289383888244629, + "learning_rate": 4.9009733796572834e-06, + "loss": 0.6789, + "step": 742 + }, + { + "epoch": 0.6154058531198233, + "grad_norm": 0.9058337211608887, + "learning_rate": 4.900666706167968e-06, + "loss": 0.6856, + "step": 743 + }, + { + "epoch": 0.616234124792932, + "grad_norm": 0.8999397158622742, + "learning_rate": 4.900359568172024e-06, + "loss": 0.7159, + "step": 744 + }, + { + "epoch": 0.6170623964660409, + "grad_norm": 0.895321249961853, + "learning_rate": 4.90005196572888e-06, + "loss": 0.7071, + "step": 745 + }, + { + "epoch": 0.6178906681391496, + "grad_norm": 0.9142102599143982, + "learning_rate": 4.899743898898054e-06, + "loss": 0.6918, + "step": 746 + }, + { + "epoch": 0.6187189398122585, + "grad_norm": 0.8867097496986389, + "learning_rate": 4.899435367739152e-06, + "loss": 0.6991, + "step": 747 + }, + { + "epoch": 0.6195472114853672, + "grad_norm": 0.8594791293144226, + "learning_rate": 4.899126372311876e-06, + "loss": 0.7153, + "step": 748 + }, + { + "epoch": 0.620375483158476, + "grad_norm": 0.8740563988685608, + "learning_rate": 4.8988169126760085e-06, + "loss": 0.6786, + "step": 749 + }, + { + "epoch": 0.6212037548315847, + "grad_norm": 0.8818353414535522, + "learning_rate": 4.898506988891432e-06, + "loss": 0.6965, + "step": 750 + }, + { + "epoch": 0.6220320265046936, + "grad_norm": 0.9459843635559082, + "learning_rate": 4.898196601018111e-06, + "loss": 0.6821, + "step": 751 + }, + { + "epoch": 0.6228602981778023, + "grad_norm": 0.923457682132721, + "learning_rate": 4.897885749116104e-06, + "loss": 0.6822, + "step": 752 + }, + { + "epoch": 0.6236885698509111, + "grad_norm": 0.9342297315597534, + "learning_rate": 4.897574433245557e-06, + "loss": 0.7001, + "step": 753 + }, + { + "epoch": 0.6245168415240199, + "grad_norm": 0.9327735900878906, + "learning_rate": 4.897262653466706e-06, + "loss": 0.6921, + "step": 754 + }, + { + "epoch": 0.6253451131971287, + "grad_norm": 0.8876991868019104, + "learning_rate": 4.8969504098398805e-06, + "loss": 0.7087, + "step": 755 + }, + { + "epoch": 0.6261733848702374, + "grad_norm": 0.8983528017997742, + "learning_rate": 4.896637702425493e-06, + "loss": 0.6824, + "step": 756 + }, + { + "epoch": 0.6270016565433462, + "grad_norm": 0.9074965715408325, + "learning_rate": 4.896324531284053e-06, + "loss": 0.6533, + "step": 757 + }, + { + "epoch": 0.627829928216455, + "grad_norm": 0.8938930630683899, + "learning_rate": 4.896010896476153e-06, + "loss": 0.6859, + "step": 758 + }, + { + "epoch": 0.6286581998895637, + "grad_norm": 0.9036867022514343, + "learning_rate": 4.8956967980624806e-06, + "loss": 0.6904, + "step": 759 + }, + { + "epoch": 0.6294864715626726, + "grad_norm": 0.9162903428077698, + "learning_rate": 4.895382236103809e-06, + "loss": 0.6995, + "step": 760 + }, + { + "epoch": 0.6303147432357813, + "grad_norm": 1.0000487565994263, + "learning_rate": 4.895067210661005e-06, + "loss": 0.671, + "step": 761 + }, + { + "epoch": 0.6311430149088901, + "grad_norm": 0.895163893699646, + "learning_rate": 4.894751721795022e-06, + "loss": 0.6871, + "step": 762 + }, + { + "epoch": 0.6319712865819989, + "grad_norm": 0.9401512742042542, + "learning_rate": 4.894435769566905e-06, + "loss": 0.6811, + "step": 763 + }, + { + "epoch": 0.6327995582551077, + "grad_norm": 0.924578845500946, + "learning_rate": 4.894119354037787e-06, + "loss": 0.692, + "step": 764 + }, + { + "epoch": 0.6336278299282164, + "grad_norm": 0.9114750623703003, + "learning_rate": 4.893802475268892e-06, + "loss": 0.6977, + "step": 765 + }, + { + "epoch": 0.6344561016013253, + "grad_norm": 1.1472550630569458, + "learning_rate": 4.893485133321533e-06, + "loss": 0.679, + "step": 766 + }, + { + "epoch": 0.635284373274434, + "grad_norm": 0.9460884928703308, + "learning_rate": 4.893167328257111e-06, + "loss": 0.6988, + "step": 767 + }, + { + "epoch": 0.6361126449475428, + "grad_norm": 0.9040783643722534, + "learning_rate": 4.8928490601371216e-06, + "loss": 0.6706, + "step": 768 + }, + { + "epoch": 0.6369409166206516, + "grad_norm": 0.9785428047180176, + "learning_rate": 4.892530329023144e-06, + "loss": 0.7185, + "step": 769 + }, + { + "epoch": 0.6377691882937604, + "grad_norm": 0.9795847535133362, + "learning_rate": 4.892211134976851e-06, + "loss": 0.703, + "step": 770 + }, + { + "epoch": 0.6385974599668691, + "grad_norm": 0.9539455771446228, + "learning_rate": 4.8918914780600025e-06, + "loss": 0.6985, + "step": 771 + }, + { + "epoch": 0.639425731639978, + "grad_norm": 0.9304025173187256, + "learning_rate": 4.8915713583344505e-06, + "loss": 0.6731, + "step": 772 + }, + { + "epoch": 0.6402540033130867, + "grad_norm": 0.9466667771339417, + "learning_rate": 4.891250775862134e-06, + "loss": 0.6972, + "step": 773 + }, + { + "epoch": 0.6410822749861955, + "grad_norm": 0.9522298574447632, + "learning_rate": 4.890929730705084e-06, + "loss": 0.6793, + "step": 774 + }, + { + "epoch": 0.6419105466593042, + "grad_norm": 0.9077293276786804, + "learning_rate": 4.890608222925419e-06, + "loss": 0.6892, + "step": 775 + }, + { + "epoch": 0.6427388183324131, + "grad_norm": 0.8763419389724731, + "learning_rate": 4.890286252585348e-06, + "loss": 0.6985, + "step": 776 + }, + { + "epoch": 0.6435670900055218, + "grad_norm": 0.9089516401290894, + "learning_rate": 4.8899638197471685e-06, + "loss": 0.6778, + "step": 777 + }, + { + "epoch": 0.6443953616786305, + "grad_norm": 0.8733325600624084, + "learning_rate": 4.88964092447327e-06, + "loss": 0.6915, + "step": 778 + }, + { + "epoch": 0.6452236333517394, + "grad_norm": 0.9472198486328125, + "learning_rate": 4.889317566826128e-06, + "loss": 0.6984, + "step": 779 + }, + { + "epoch": 0.6460519050248481, + "grad_norm": 0.8729666471481323, + "learning_rate": 4.888993746868309e-06, + "loss": 0.6724, + "step": 780 + }, + { + "epoch": 0.6468801766979569, + "grad_norm": 0.9395233988761902, + "learning_rate": 4.88866946466247e-06, + "loss": 0.6697, + "step": 781 + }, + { + "epoch": 0.6477084483710657, + "grad_norm": 0.9158744812011719, + "learning_rate": 4.888344720271357e-06, + "loss": 0.6838, + "step": 782 + }, + { + "epoch": 0.6485367200441745, + "grad_norm": 0.8961113095283508, + "learning_rate": 4.888019513757804e-06, + "loss": 0.6488, + "step": 783 + }, + { + "epoch": 0.6493649917172832, + "grad_norm": 0.8911786079406738, + "learning_rate": 4.8876938451847364e-06, + "loss": 0.7089, + "step": 784 + }, + { + "epoch": 0.6501932633903921, + "grad_norm": 0.9214697480201721, + "learning_rate": 4.887367714615167e-06, + "loss": 0.6929, + "step": 785 + }, + { + "epoch": 0.6510215350635008, + "grad_norm": 0.9161441326141357, + "learning_rate": 4.8870411221122e-06, + "loss": 0.6761, + "step": 786 + }, + { + "epoch": 0.6518498067366096, + "grad_norm": 0.9161947965621948, + "learning_rate": 4.886714067739028e-06, + "loss": 0.6804, + "step": 787 + }, + { + "epoch": 0.6526780784097184, + "grad_norm": 0.8706585168838501, + "learning_rate": 4.886386551558933e-06, + "loss": 0.6836, + "step": 788 + }, + { + "epoch": 0.6535063500828272, + "grad_norm": 0.9037661552429199, + "learning_rate": 4.886058573635284e-06, + "loss": 0.6786, + "step": 789 + }, + { + "epoch": 0.6543346217559359, + "grad_norm": 0.8977832794189453, + "learning_rate": 4.885730134031545e-06, + "loss": 0.6938, + "step": 790 + }, + { + "epoch": 0.6551628934290448, + "grad_norm": 0.9106955528259277, + "learning_rate": 4.8854012328112645e-06, + "loss": 0.678, + "step": 791 + }, + { + "epoch": 0.6559911651021535, + "grad_norm": 0.8769396543502808, + "learning_rate": 4.885071870038082e-06, + "loss": 0.6969, + "step": 792 + }, + { + "epoch": 0.6568194367752623, + "grad_norm": 0.9946926832199097, + "learning_rate": 4.884742045775725e-06, + "loss": 0.6799, + "step": 793 + }, + { + "epoch": 0.6576477084483711, + "grad_norm": 0.9399042725563049, + "learning_rate": 4.884411760088012e-06, + "loss": 0.7038, + "step": 794 + }, + { + "epoch": 0.6584759801214799, + "grad_norm": 0.8914628624916077, + "learning_rate": 4.8840810130388526e-06, + "loss": 0.6867, + "step": 795 + }, + { + "epoch": 0.6593042517945886, + "grad_norm": 0.8900558352470398, + "learning_rate": 4.88374980469224e-06, + "loss": 0.6823, + "step": 796 + }, + { + "epoch": 0.6601325234676974, + "grad_norm": 0.8608913421630859, + "learning_rate": 4.8834181351122615e-06, + "loss": 0.6705, + "step": 797 + }, + { + "epoch": 0.6609607951408062, + "grad_norm": 0.900460422039032, + "learning_rate": 4.883086004363092e-06, + "loss": 0.7029, + "step": 798 + }, + { + "epoch": 0.6617890668139149, + "grad_norm": 0.8932703733444214, + "learning_rate": 4.882753412508996e-06, + "loss": 0.6845, + "step": 799 + }, + { + "epoch": 0.6626173384870238, + "grad_norm": 0.8737874627113342, + "learning_rate": 4.882420359614325e-06, + "loss": 0.6979, + "step": 800 + }, + { + "epoch": 0.6634456101601325, + "grad_norm": 0.8805396556854248, + "learning_rate": 4.882086845743524e-06, + "loss": 0.6935, + "step": 801 + }, + { + "epoch": 0.6642738818332413, + "grad_norm": 0.8952041268348694, + "learning_rate": 4.881752870961123e-06, + "loss": 0.7137, + "step": 802 + }, + { + "epoch": 0.66510215350635, + "grad_norm": 0.889869213104248, + "learning_rate": 4.881418435331745e-06, + "loss": 0.6643, + "step": 803 + }, + { + "epoch": 0.6659304251794589, + "grad_norm": 0.8912904858589172, + "learning_rate": 4.881083538920098e-06, + "loss": 0.6712, + "step": 804 + }, + { + "epoch": 0.6667586968525676, + "grad_norm": 0.8943006992340088, + "learning_rate": 4.880748181790984e-06, + "loss": 0.7001, + "step": 805 + }, + { + "epoch": 0.6675869685256764, + "grad_norm": 0.9042482376098633, + "learning_rate": 4.880412364009289e-06, + "loss": 0.6744, + "step": 806 + }, + { + "epoch": 0.6684152401987852, + "grad_norm": 0.9052988886833191, + "learning_rate": 4.880076085639991e-06, + "loss": 0.72, + "step": 807 + }, + { + "epoch": 0.669243511871894, + "grad_norm": 0.9702905416488647, + "learning_rate": 4.879739346748158e-06, + "loss": 0.6871, + "step": 808 + }, + { + "epoch": 0.6700717835450027, + "grad_norm": 0.8741766810417175, + "learning_rate": 4.879402147398944e-06, + "loss": 0.7174, + "step": 809 + }, + { + "epoch": 0.6709000552181116, + "grad_norm": 0.9226665496826172, + "learning_rate": 4.879064487657596e-06, + "loss": 0.6991, + "step": 810 + }, + { + "epoch": 0.6717283268912203, + "grad_norm": 0.878304123878479, + "learning_rate": 4.878726367589446e-06, + "loss": 0.6925, + "step": 811 + }, + { + "epoch": 0.6725565985643291, + "grad_norm": 0.893649160861969, + "learning_rate": 4.878387787259919e-06, + "loss": 0.6689, + "step": 812 + }, + { + "epoch": 0.6733848702374379, + "grad_norm": 0.917861819267273, + "learning_rate": 4.878048746734526e-06, + "loss": 0.7171, + "step": 813 + }, + { + "epoch": 0.6742131419105467, + "grad_norm": 0.879375696182251, + "learning_rate": 4.877709246078869e-06, + "loss": 0.6662, + "step": 814 + }, + { + "epoch": 0.6750414135836554, + "grad_norm": 0.9098542928695679, + "learning_rate": 4.877369285358637e-06, + "loss": 0.6793, + "step": 815 + }, + { + "epoch": 0.6758696852567642, + "grad_norm": 0.9465262293815613, + "learning_rate": 4.877028864639609e-06, + "loss": 0.7065, + "step": 816 + }, + { + "epoch": 0.676697956929873, + "grad_norm": 0.9029957056045532, + "learning_rate": 4.876687983987654e-06, + "loss": 0.7008, + "step": 817 + }, + { + "epoch": 0.6775262286029817, + "grad_norm": 0.9073920845985413, + "learning_rate": 4.876346643468731e-06, + "loss": 0.6836, + "step": 818 + }, + { + "epoch": 0.6783545002760906, + "grad_norm": 0.9028745889663696, + "learning_rate": 4.876004843148884e-06, + "loss": 0.6965, + "step": 819 + }, + { + "epoch": 0.6791827719491993, + "grad_norm": 0.8745572566986084, + "learning_rate": 4.875662583094249e-06, + "loss": 0.6974, + "step": 820 + }, + { + "epoch": 0.6800110436223081, + "grad_norm": 0.9203384518623352, + "learning_rate": 4.875319863371048e-06, + "loss": 0.686, + "step": 821 + }, + { + "epoch": 0.6808393152954169, + "grad_norm": 0.9087005853652954, + "learning_rate": 4.874976684045597e-06, + "loss": 0.683, + "step": 822 + }, + { + "epoch": 0.6816675869685257, + "grad_norm": 0.9007741808891296, + "learning_rate": 4.874633045184297e-06, + "loss": 0.6938, + "step": 823 + }, + { + "epoch": 0.6824958586416344, + "grad_norm": 0.9854089617729187, + "learning_rate": 4.874288946853639e-06, + "loss": 0.6945, + "step": 824 + }, + { + "epoch": 0.6833241303147433, + "grad_norm": 0.9044802784919739, + "learning_rate": 4.8739443891202025e-06, + "loss": 0.714, + "step": 825 + }, + { + "epoch": 0.684152401987852, + "grad_norm": 0.9388754963874817, + "learning_rate": 4.873599372050656e-06, + "loss": 0.6838, + "step": 826 + }, + { + "epoch": 0.6849806736609608, + "grad_norm": 0.9215866327285767, + "learning_rate": 4.873253895711757e-06, + "loss": 0.6994, + "step": 827 + }, + { + "epoch": 0.6858089453340696, + "grad_norm": 0.8786854147911072, + "learning_rate": 4.872907960170353e-06, + "loss": 0.6883, + "step": 828 + }, + { + "epoch": 0.6866372170071784, + "grad_norm": 0.9032605290412903, + "learning_rate": 4.872561565493379e-06, + "loss": 0.6847, + "step": 829 + }, + { + "epoch": 0.6874654886802871, + "grad_norm": 0.8940309882164001, + "learning_rate": 4.872214711747858e-06, + "loss": 0.6877, + "step": 830 + }, + { + "epoch": 0.688293760353396, + "grad_norm": 0.8703469634056091, + "learning_rate": 4.8718673990009045e-06, + "loss": 0.6829, + "step": 831 + }, + { + "epoch": 0.6891220320265047, + "grad_norm": 0.8618217706680298, + "learning_rate": 4.871519627319719e-06, + "loss": 0.6987, + "step": 832 + }, + { + "epoch": 0.6899503036996135, + "grad_norm": 0.8871967196464539, + "learning_rate": 4.871171396771594e-06, + "loss": 0.6824, + "step": 833 + }, + { + "epoch": 0.6907785753727222, + "grad_norm": 0.9162507057189941, + "learning_rate": 4.8708227074239064e-06, + "loss": 0.7158, + "step": 834 + }, + { + "epoch": 0.6916068470458311, + "grad_norm": 0.8829584717750549, + "learning_rate": 4.8704735593441255e-06, + "loss": 0.6714, + "step": 835 + }, + { + "epoch": 0.6924351187189398, + "grad_norm": 0.9130700826644897, + "learning_rate": 4.8701239525998076e-06, + "loss": 0.6841, + "step": 836 + }, + { + "epoch": 0.6932633903920485, + "grad_norm": 0.8706396222114563, + "learning_rate": 4.8697738872586e-06, + "loss": 0.7162, + "step": 837 + }, + { + "epoch": 0.6940916620651574, + "grad_norm": 0.8906163573265076, + "learning_rate": 4.869423363388235e-06, + "loss": 0.6586, + "step": 838 + }, + { + "epoch": 0.6949199337382661, + "grad_norm": 0.9134309887886047, + "learning_rate": 4.869072381056537e-06, + "loss": 0.6842, + "step": 839 + }, + { + "epoch": 0.6957482054113749, + "grad_norm": 0.869674563407898, + "learning_rate": 4.8687209403314175e-06, + "loss": 0.7049, + "step": 840 + }, + { + "epoch": 0.6965764770844837, + "grad_norm": 0.8405192494392395, + "learning_rate": 4.8683690412808775e-06, + "loss": 0.6677, + "step": 841 + }, + { + "epoch": 0.6974047487575925, + "grad_norm": 0.9173605442047119, + "learning_rate": 4.868016683973005e-06, + "loss": 0.6784, + "step": 842 + }, + { + "epoch": 0.6982330204307012, + "grad_norm": 0.9220597743988037, + "learning_rate": 4.867663868475978e-06, + "loss": 0.6774, + "step": 843 + }, + { + "epoch": 0.6990612921038101, + "grad_norm": 0.8687174320220947, + "learning_rate": 4.867310594858065e-06, + "loss": 0.7094, + "step": 844 + }, + { + "epoch": 0.6998895637769188, + "grad_norm": 0.9161131381988525, + "learning_rate": 4.866956863187619e-06, + "loss": 0.6867, + "step": 845 + }, + { + "epoch": 0.7007178354500276, + "grad_norm": 0.9338458776473999, + "learning_rate": 4.866602673533084e-06, + "loss": 0.6783, + "step": 846 + }, + { + "epoch": 0.7015461071231364, + "grad_norm": 0.9268187284469604, + "learning_rate": 4.866248025962993e-06, + "loss": 0.6681, + "step": 847 + }, + { + "epoch": 0.7023743787962452, + "grad_norm": 0.8843579292297363, + "learning_rate": 4.865892920545967e-06, + "loss": 0.6854, + "step": 848 + }, + { + "epoch": 0.7032026504693539, + "grad_norm": 0.889299213886261, + "learning_rate": 4.865537357350716e-06, + "loss": 0.7055, + "step": 849 + }, + { + "epoch": 0.7040309221424628, + "grad_norm": 0.9062727093696594, + "learning_rate": 4.865181336446036e-06, + "loss": 0.6806, + "step": 850 + }, + { + "epoch": 0.7048591938155715, + "grad_norm": 0.9007996320724487, + "learning_rate": 4.864824857900817e-06, + "loss": 0.6937, + "step": 851 + }, + { + "epoch": 0.7056874654886803, + "grad_norm": 0.8883542418479919, + "learning_rate": 4.864467921784032e-06, + "loss": 0.6957, + "step": 852 + }, + { + "epoch": 0.7065157371617891, + "grad_norm": 0.9422078132629395, + "learning_rate": 4.864110528164746e-06, + "loss": 0.7025, + "step": 853 + }, + { + "epoch": 0.7073440088348979, + "grad_norm": 0.8961433172225952, + "learning_rate": 4.86375267711211e-06, + "loss": 0.6795, + "step": 854 + }, + { + "epoch": 0.7081722805080066, + "grad_norm": 0.925815761089325, + "learning_rate": 4.863394368695366e-06, + "loss": 0.6824, + "step": 855 + }, + { + "epoch": 0.7090005521811154, + "grad_norm": 0.9326804280281067, + "learning_rate": 4.863035602983843e-06, + "loss": 0.6829, + "step": 856 + }, + { + "epoch": 0.7098288238542242, + "grad_norm": 0.8784539699554443, + "learning_rate": 4.862676380046959e-06, + "loss": 0.6776, + "step": 857 + }, + { + "epoch": 0.7106570955273329, + "grad_norm": 0.8577290177345276, + "learning_rate": 4.86231669995422e-06, + "loss": 0.6573, + "step": 858 + }, + { + "epoch": 0.7114853672004418, + "grad_norm": 0.9261177778244019, + "learning_rate": 4.86195656277522e-06, + "loss": 0.6891, + "step": 859 + }, + { + "epoch": 0.7123136388735505, + "grad_norm": 0.9008646607398987, + "learning_rate": 4.861595968579643e-06, + "loss": 0.6714, + "step": 860 + }, + { + "epoch": 0.7131419105466593, + "grad_norm": 0.8729333877563477, + "learning_rate": 4.861234917437261e-06, + "loss": 0.6819, + "step": 861 + }, + { + "epoch": 0.713970182219768, + "grad_norm": 0.9006103873252869, + "learning_rate": 4.860873409417934e-06, + "loss": 0.6702, + "step": 862 + }, + { + "epoch": 0.7147984538928769, + "grad_norm": 0.8699921369552612, + "learning_rate": 4.860511444591609e-06, + "loss": 0.6681, + "step": 863 + }, + { + "epoch": 0.7156267255659856, + "grad_norm": 0.9359689950942993, + "learning_rate": 4.860149023028325e-06, + "loss": 0.6834, + "step": 864 + }, + { + "epoch": 0.7164549972390944, + "grad_norm": 0.874977707862854, + "learning_rate": 4.859786144798205e-06, + "loss": 0.6692, + "step": 865 + }, + { + "epoch": 0.7172832689122032, + "grad_norm": 0.8821915984153748, + "learning_rate": 4.859422809971464e-06, + "loss": 0.6865, + "step": 866 + }, + { + "epoch": 0.718111540585312, + "grad_norm": 0.905248761177063, + "learning_rate": 4.859059018618405e-06, + "loss": 0.6705, + "step": 867 + }, + { + "epoch": 0.7189398122584207, + "grad_norm": 0.9134542346000671, + "learning_rate": 4.858694770809415e-06, + "loss": 0.7039, + "step": 868 + }, + { + "epoch": 0.7197680839315296, + "grad_norm": 0.9048055410385132, + "learning_rate": 4.858330066614975e-06, + "loss": 0.6848, + "step": 869 + }, + { + "epoch": 0.7205963556046383, + "grad_norm": 0.8780005574226379, + "learning_rate": 4.857964906105651e-06, + "loss": 0.689, + "step": 870 + }, + { + "epoch": 0.7214246272777471, + "grad_norm": 0.8826956152915955, + "learning_rate": 4.857599289352098e-06, + "loss": 0.6799, + "step": 871 + }, + { + "epoch": 0.7222528989508559, + "grad_norm": 0.919905960559845, + "learning_rate": 4.857233216425061e-06, + "loss": 0.6587, + "step": 872 + }, + { + "epoch": 0.7230811706239647, + "grad_norm": 0.8711065053939819, + "learning_rate": 4.8568666873953696e-06, + "loss": 0.6754, + "step": 873 + }, + { + "epoch": 0.7239094422970734, + "grad_norm": 0.8740426898002625, + "learning_rate": 4.856499702333944e-06, + "loss": 0.6837, + "step": 874 + }, + { + "epoch": 0.7247377139701823, + "grad_norm": 0.921416699886322, + "learning_rate": 4.8561322613117945e-06, + "loss": 0.6709, + "step": 875 + }, + { + "epoch": 0.725565985643291, + "grad_norm": 0.8988221287727356, + "learning_rate": 4.855764364400016e-06, + "loss": 0.7111, + "step": 876 + }, + { + "epoch": 0.7263942573163997, + "grad_norm": 0.8838291168212891, + "learning_rate": 4.8553960116697915e-06, + "loss": 0.6821, + "step": 877 + }, + { + "epoch": 0.7272225289895086, + "grad_norm": 0.9051623940467834, + "learning_rate": 4.8550272031923965e-06, + "loss": 0.6848, + "step": 878 + }, + { + "epoch": 0.7280508006626173, + "grad_norm": 0.9308082461357117, + "learning_rate": 4.85465793903919e-06, + "loss": 0.7087, + "step": 879 + }, + { + "epoch": 0.7288790723357261, + "grad_norm": 0.873214066028595, + "learning_rate": 4.854288219281623e-06, + "loss": 0.6877, + "step": 880 + }, + { + "epoch": 0.7297073440088349, + "grad_norm": 0.8890054225921631, + "learning_rate": 4.853918043991232e-06, + "loss": 0.6719, + "step": 881 + }, + { + "epoch": 0.7305356156819437, + "grad_norm": 0.9580398201942444, + "learning_rate": 4.853547413239642e-06, + "loss": 0.6439, + "step": 882 + }, + { + "epoch": 0.7313638873550524, + "grad_norm": 0.875035285949707, + "learning_rate": 4.853176327098567e-06, + "loss": 0.6645, + "step": 883 + }, + { + "epoch": 0.7321921590281613, + "grad_norm": 0.9299217462539673, + "learning_rate": 4.85280478563981e-06, + "loss": 0.707, + "step": 884 + }, + { + "epoch": 0.73302043070127, + "grad_norm": 0.8840609192848206, + "learning_rate": 4.8524327889352585e-06, + "loss": 0.6789, + "step": 885 + }, + { + "epoch": 0.7338487023743788, + "grad_norm": 0.8983805179595947, + "learning_rate": 4.852060337056891e-06, + "loss": 0.6814, + "step": 886 + }, + { + "epoch": 0.7346769740474876, + "grad_norm": 0.8730597496032715, + "learning_rate": 4.851687430076775e-06, + "loss": 0.7006, + "step": 887 + }, + { + "epoch": 0.7355052457205964, + "grad_norm": 0.8980515003204346, + "learning_rate": 4.851314068067063e-06, + "loss": 0.6729, + "step": 888 + }, + { + "epoch": 0.7363335173937051, + "grad_norm": 0.8919468522071838, + "learning_rate": 4.850940251099997e-06, + "loss": 0.6764, + "step": 889 + }, + { + "epoch": 0.737161789066814, + "grad_norm": 0.8987070918083191, + "learning_rate": 4.850565979247908e-06, + "loss": 0.6997, + "step": 890 + }, + { + "epoch": 0.7379900607399227, + "grad_norm": 0.9056184887886047, + "learning_rate": 4.850191252583214e-06, + "loss": 0.6887, + "step": 891 + }, + { + "epoch": 0.7388183324130315, + "grad_norm": 0.8714621067047119, + "learning_rate": 4.84981607117842e-06, + "loss": 0.6946, + "step": 892 + }, + { + "epoch": 0.7396466040861402, + "grad_norm": 0.8845617175102234, + "learning_rate": 4.849440435106122e-06, + "loss": 0.6845, + "step": 893 + }, + { + "epoch": 0.7404748757592491, + "grad_norm": 0.8926048278808594, + "learning_rate": 4.849064344439e-06, + "loss": 0.6922, + "step": 894 + }, + { + "epoch": 0.7413031474323578, + "grad_norm": 0.8731569647789001, + "learning_rate": 4.848687799249826e-06, + "loss": 0.6837, + "step": 895 + }, + { + "epoch": 0.7421314191054665, + "grad_norm": 0.8801183104515076, + "learning_rate": 4.848310799611457e-06, + "loss": 0.6932, + "step": 896 + }, + { + "epoch": 0.7429596907785754, + "grad_norm": 0.8818139433860779, + "learning_rate": 4.847933345596838e-06, + "loss": 0.6758, + "step": 897 + }, + { + "epoch": 0.7437879624516841, + "grad_norm": 0.9117923974990845, + "learning_rate": 4.847555437279004e-06, + "loss": 0.6628, + "step": 898 + }, + { + "epoch": 0.7446162341247929, + "grad_norm": 0.891360342502594, + "learning_rate": 4.847177074731078e-06, + "loss": 0.6876, + "step": 899 + }, + { + "epoch": 0.7454445057979017, + "grad_norm": 0.8905263543128967, + "learning_rate": 4.846798258026267e-06, + "loss": 0.6862, + "step": 900 + }, + { + "epoch": 0.7462727774710105, + "grad_norm": 0.8853707909584045, + "learning_rate": 4.846418987237868e-06, + "loss": 0.654, + "step": 901 + }, + { + "epoch": 0.7471010491441192, + "grad_norm": 0.8782089948654175, + "learning_rate": 4.846039262439271e-06, + "loss": 0.7005, + "step": 902 + }, + { + "epoch": 0.7479293208172281, + "grad_norm": 0.8774383068084717, + "learning_rate": 4.8456590837039445e-06, + "loss": 0.7044, + "step": 903 + }, + { + "epoch": 0.7487575924903368, + "grad_norm": 0.9385423064231873, + "learning_rate": 4.845278451105451e-06, + "loss": 0.676, + "step": 904 + }, + { + "epoch": 0.7495858641634456, + "grad_norm": 0.9211432337760925, + "learning_rate": 4.8448973647174405e-06, + "loss": 0.697, + "step": 905 + }, + { + "epoch": 0.7504141358365544, + "grad_norm": 0.887610912322998, + "learning_rate": 4.844515824613649e-06, + "loss": 0.6965, + "step": 906 + }, + { + "epoch": 0.7512424075096632, + "grad_norm": 0.885761559009552, + "learning_rate": 4.8441338308679e-06, + "loss": 0.6851, + "step": 907 + }, + { + "epoch": 0.7520706791827719, + "grad_norm": 0.9209930300712585, + "learning_rate": 4.843751383554106e-06, + "loss": 0.6718, + "step": 908 + }, + { + "epoch": 0.7528989508558808, + "grad_norm": 0.9308252334594727, + "learning_rate": 4.843368482746269e-06, + "loss": 0.6802, + "step": 909 + }, + { + "epoch": 0.7537272225289895, + "grad_norm": 0.8931897282600403, + "learning_rate": 4.842985128518473e-06, + "loss": 0.7076, + "step": 910 + }, + { + "epoch": 0.7545554942020983, + "grad_norm": 0.9102733731269836, + "learning_rate": 4.842601320944898e-06, + "loss": 0.6898, + "step": 911 + }, + { + "epoch": 0.7553837658752071, + "grad_norm": 0.9336893558502197, + "learning_rate": 4.842217060099804e-06, + "loss": 0.6709, + "step": 912 + }, + { + "epoch": 0.7562120375483159, + "grad_norm": 0.9366991519927979, + "learning_rate": 4.8418323460575425e-06, + "loss": 0.6665, + "step": 913 + }, + { + "epoch": 0.7570403092214246, + "grad_norm": 0.9371415376663208, + "learning_rate": 4.841447178892552e-06, + "loss": 0.7013, + "step": 914 + }, + { + "epoch": 0.7578685808945335, + "grad_norm": 1.031123399734497, + "learning_rate": 4.84106155867936e-06, + "loss": 0.687, + "step": 915 + }, + { + "epoch": 0.7586968525676422, + "grad_norm": 0.9319218397140503, + "learning_rate": 4.8406754854925785e-06, + "loss": 0.669, + "step": 916 + }, + { + "epoch": 0.7595251242407509, + "grad_norm": 0.8563560843467712, + "learning_rate": 4.840288959406911e-06, + "loss": 0.6905, + "step": 917 + }, + { + "epoch": 0.7603533959138598, + "grad_norm": 0.9424378275871277, + "learning_rate": 4.839901980497145e-06, + "loss": 0.6685, + "step": 918 + }, + { + "epoch": 0.7611816675869685, + "grad_norm": 0.9202269315719604, + "learning_rate": 4.839514548838158e-06, + "loss": 0.6728, + "step": 919 + }, + { + "epoch": 0.7620099392600773, + "grad_norm": 0.9152582287788391, + "learning_rate": 4.839126664504915e-06, + "loss": 0.6842, + "step": 920 + }, + { + "epoch": 0.762838210933186, + "grad_norm": 0.9457129836082458, + "learning_rate": 4.838738327572468e-06, + "loss": 0.6822, + "step": 921 + }, + { + "epoch": 0.7636664826062949, + "grad_norm": 0.93124920129776, + "learning_rate": 4.838349538115955e-06, + "loss": 0.691, + "step": 922 + }, + { + "epoch": 0.7644947542794036, + "grad_norm": 0.9249301552772522, + "learning_rate": 4.837960296210605e-06, + "loss": 0.6574, + "step": 923 + }, + { + "epoch": 0.7653230259525124, + "grad_norm": 1.0551042556762695, + "learning_rate": 4.837570601931731e-06, + "loss": 0.6782, + "step": 924 + }, + { + "epoch": 0.7661512976256212, + "grad_norm": 0.9404964447021484, + "learning_rate": 4.8371804553547365e-06, + "loss": 0.6703, + "step": 925 + }, + { + "epoch": 0.76697956929873, + "grad_norm": 0.909015417098999, + "learning_rate": 4.836789856555111e-06, + "loss": 0.6812, + "step": 926 + }, + { + "epoch": 0.7678078409718387, + "grad_norm": 0.8572261929512024, + "learning_rate": 4.836398805608431e-06, + "loss": 0.6831, + "step": 927 + }, + { + "epoch": 0.7686361126449476, + "grad_norm": 0.951753556728363, + "learning_rate": 4.836007302590362e-06, + "loss": 0.6672, + "step": 928 + }, + { + "epoch": 0.7694643843180563, + "grad_norm": 0.8875932693481445, + "learning_rate": 4.8356153475766564e-06, + "loss": 0.6819, + "step": 929 + }, + { + "epoch": 0.7702926559911651, + "grad_norm": 0.9405681490898132, + "learning_rate": 4.835222940643153e-06, + "loss": 0.6917, + "step": 930 + }, + { + "epoch": 0.7711209276642739, + "grad_norm": 0.942520022392273, + "learning_rate": 4.834830081865779e-06, + "loss": 0.6657, + "step": 931 + }, + { + "epoch": 0.7719491993373827, + "grad_norm": 0.9063259363174438, + "learning_rate": 4.83443677132055e-06, + "loss": 0.6857, + "step": 932 + }, + { + "epoch": 0.7727774710104914, + "grad_norm": 0.8778199553489685, + "learning_rate": 4.834043009083566e-06, + "loss": 0.7028, + "step": 933 + }, + { + "epoch": 0.7736057426836003, + "grad_norm": 0.9215186834335327, + "learning_rate": 4.833648795231018e-06, + "loss": 0.6712, + "step": 934 + }, + { + "epoch": 0.774434014356709, + "grad_norm": 0.8992031812667847, + "learning_rate": 4.8332541298391825e-06, + "loss": 0.6827, + "step": 935 + }, + { + "epoch": 0.7752622860298177, + "grad_norm": 0.9203833341598511, + "learning_rate": 4.8328590129844224e-06, + "loss": 0.6835, + "step": 936 + }, + { + "epoch": 0.7760905577029266, + "grad_norm": 0.9109310507774353, + "learning_rate": 4.832463444743191e-06, + "loss": 0.673, + "step": 937 + }, + { + "epoch": 0.7769188293760353, + "grad_norm": 0.978349506855011, + "learning_rate": 4.832067425192024e-06, + "loss": 0.6638, + "step": 938 + }, + { + "epoch": 0.7777471010491441, + "grad_norm": 0.9552261829376221, + "learning_rate": 4.831670954407551e-06, + "loss": 0.6919, + "step": 939 + }, + { + "epoch": 0.7785753727222529, + "grad_norm": 0.9106539487838745, + "learning_rate": 4.831274032466484e-06, + "loss": 0.6964, + "step": 940 + }, + { + "epoch": 0.7794036443953617, + "grad_norm": 0.9195299744606018, + "learning_rate": 4.830876659445623e-06, + "loss": 0.6793, + "step": 941 + }, + { + "epoch": 0.7802319160684704, + "grad_norm": 0.9075512290000916, + "learning_rate": 4.8304788354218564e-06, + "loss": 0.6649, + "step": 942 + }, + { + "epoch": 0.7810601877415793, + "grad_norm": 0.9142526984214783, + "learning_rate": 4.83008056047216e-06, + "loss": 0.6811, + "step": 943 + }, + { + "epoch": 0.781888459414688, + "grad_norm": 0.8898900151252747, + "learning_rate": 4.829681834673597e-06, + "loss": 0.6906, + "step": 944 + }, + { + "epoch": 0.7827167310877968, + "grad_norm": 0.8884983062744141, + "learning_rate": 4.8292826581033135e-06, + "loss": 0.6903, + "step": 945 + }, + { + "epoch": 0.7835450027609056, + "grad_norm": 0.9075689315795898, + "learning_rate": 4.828883030838551e-06, + "loss": 0.6794, + "step": 946 + }, + { + "epoch": 0.7843732744340144, + "grad_norm": 0.8970300555229187, + "learning_rate": 4.828482952956631e-06, + "loss": 0.6715, + "step": 947 + }, + { + "epoch": 0.7852015461071231, + "grad_norm": 0.8906447887420654, + "learning_rate": 4.828082424534965e-06, + "loss": 0.6878, + "step": 948 + }, + { + "epoch": 0.786029817780232, + "grad_norm": 0.9083748459815979, + "learning_rate": 4.827681445651052e-06, + "loss": 0.6787, + "step": 949 + }, + { + "epoch": 0.7868580894533407, + "grad_norm": 0.8975276947021484, + "learning_rate": 4.8272800163824784e-06, + "loss": 0.687, + "step": 950 + }, + { + "epoch": 0.7876863611264495, + "grad_norm": 0.9573839902877808, + "learning_rate": 4.8268781368069156e-06, + "loss": 0.6996, + "step": 951 + }, + { + "epoch": 0.7885146327995582, + "grad_norm": 0.913536787033081, + "learning_rate": 4.826475807002125e-06, + "loss": 0.6844, + "step": 952 + }, + { + "epoch": 0.7893429044726671, + "grad_norm": 0.9202503561973572, + "learning_rate": 4.826073027045952e-06, + "loss": 0.6702, + "step": 953 + }, + { + "epoch": 0.7901711761457758, + "grad_norm": 0.8800237774848938, + "learning_rate": 4.8256697970163315e-06, + "loss": 0.685, + "step": 954 + }, + { + "epoch": 0.7909994478188846, + "grad_norm": 0.8923048377037048, + "learning_rate": 4.825266116991285e-06, + "loss": 0.6839, + "step": 955 + }, + { + "epoch": 0.7918277194919934, + "grad_norm": 0.879743754863739, + "learning_rate": 4.824861987048922e-06, + "loss": 0.7112, + "step": 956 + }, + { + "epoch": 0.7926559911651021, + "grad_norm": 0.8575169444084167, + "learning_rate": 4.824457407267436e-06, + "loss": 0.671, + "step": 957 + }, + { + "epoch": 0.7934842628382109, + "grad_norm": 0.8851709961891174, + "learning_rate": 4.824052377725109e-06, + "loss": 0.669, + "step": 958 + }, + { + "epoch": 0.7943125345113197, + "grad_norm": 0.9084329605102539, + "learning_rate": 4.823646898500311e-06, + "loss": 0.6731, + "step": 959 + }, + { + "epoch": 0.7951408061844285, + "grad_norm": 0.908256471157074, + "learning_rate": 4.823240969671501e-06, + "loss": 0.688, + "step": 960 + }, + { + "epoch": 0.7959690778575372, + "grad_norm": 0.8993384838104248, + "learning_rate": 4.822834591317219e-06, + "loss": 0.6736, + "step": 961 + }, + { + "epoch": 0.7967973495306461, + "grad_norm": 0.893530011177063, + "learning_rate": 4.822427763516098e-06, + "loss": 0.6787, + "step": 962 + }, + { + "epoch": 0.7976256212037548, + "grad_norm": 0.9479735493659973, + "learning_rate": 4.822020486346852e-06, + "loss": 0.6842, + "step": 963 + }, + { + "epoch": 0.7984538928768636, + "grad_norm": 0.8716689348220825, + "learning_rate": 4.821612759888289e-06, + "loss": 0.6726, + "step": 964 + }, + { + "epoch": 0.7992821645499724, + "grad_norm": 0.8902738690376282, + "learning_rate": 4.821204584219299e-06, + "loss": 0.6672, + "step": 965 + }, + { + "epoch": 0.8001104362230812, + "grad_norm": 0.8713969588279724, + "learning_rate": 4.820795959418859e-06, + "loss": 0.6783, + "step": 966 + }, + { + "epoch": 0.8009387078961899, + "grad_norm": 0.891788899898529, + "learning_rate": 4.820386885566036e-06, + "loss": 0.6691, + "step": 967 + }, + { + "epoch": 0.8017669795692988, + "grad_norm": 0.8484700918197632, + "learning_rate": 4.81997736273998e-06, + "loss": 0.6766, + "step": 968 + }, + { + "epoch": 0.8025952512424075, + "grad_norm": 0.8931223154067993, + "learning_rate": 4.819567391019931e-06, + "loss": 0.6944, + "step": 969 + }, + { + "epoch": 0.8034235229155163, + "grad_norm": 0.9707916975021362, + "learning_rate": 4.819156970485216e-06, + "loss": 0.7076, + "step": 970 + }, + { + "epoch": 0.8042517945886251, + "grad_norm": 0.9172869920730591, + "learning_rate": 4.818746101215245e-06, + "loss": 0.6864, + "step": 971 + }, + { + "epoch": 0.8050800662617339, + "grad_norm": 0.8982006311416626, + "learning_rate": 4.818334783289518e-06, + "loss": 0.6685, + "step": 972 + }, + { + "epoch": 0.8059083379348426, + "grad_norm": 0.8960995078086853, + "learning_rate": 4.817923016787623e-06, + "loss": 0.6959, + "step": 973 + }, + { + "epoch": 0.8067366096079515, + "grad_norm": 0.8806949257850647, + "learning_rate": 4.81751080178923e-06, + "loss": 0.7028, + "step": 974 + }, + { + "epoch": 0.8075648812810602, + "grad_norm": 0.8798978924751282, + "learning_rate": 4.817098138374102e-06, + "loss": 0.6718, + "step": 975 + }, + { + "epoch": 0.8083931529541689, + "grad_norm": 0.9217897653579712, + "learning_rate": 4.8166850266220835e-06, + "loss": 0.6782, + "step": 976 + }, + { + "epoch": 0.8092214246272778, + "grad_norm": 0.8915667533874512, + "learning_rate": 4.8162714666131074e-06, + "loss": 0.6917, + "step": 977 + }, + { + "epoch": 0.8100496963003865, + "grad_norm": 0.8895093202590942, + "learning_rate": 4.815857458427195e-06, + "loss": 0.698, + "step": 978 + }, + { + "epoch": 0.8108779679734953, + "grad_norm": 0.9167448282241821, + "learning_rate": 4.815443002144453e-06, + "loss": 0.6796, + "step": 979 + }, + { + "epoch": 0.811706239646604, + "grad_norm": 0.913827657699585, + "learning_rate": 4.815028097845074e-06, + "loss": 0.6731, + "step": 980 + }, + { + "epoch": 0.8125345113197129, + "grad_norm": 0.9183399081230164, + "learning_rate": 4.814612745609338e-06, + "loss": 0.7076, + "step": 981 + }, + { + "epoch": 0.8133627829928216, + "grad_norm": 0.8812996745109558, + "learning_rate": 4.814196945517613e-06, + "loss": 0.6416, + "step": 982 + }, + { + "epoch": 0.8141910546659304, + "grad_norm": 0.9032955169677734, + "learning_rate": 4.813780697650351e-06, + "loss": 0.6799, + "step": 983 + }, + { + "epoch": 0.8150193263390392, + "grad_norm": 0.9380100965499878, + "learning_rate": 4.813364002088093e-06, + "loss": 0.6771, + "step": 984 + }, + { + "epoch": 0.815847598012148, + "grad_norm": 0.9470316171646118, + "learning_rate": 4.8129468589114655e-06, + "loss": 0.6851, + "step": 985 + }, + { + "epoch": 0.8166758696852567, + "grad_norm": 0.9401133060455322, + "learning_rate": 4.812529268201183e-06, + "loss": 0.676, + "step": 986 + }, + { + "epoch": 0.8175041413583656, + "grad_norm": 0.9379092454910278, + "learning_rate": 4.812111230038043e-06, + "loss": 0.672, + "step": 987 + }, + { + "epoch": 0.8183324130314743, + "grad_norm": 0.9591637253761292, + "learning_rate": 4.811692744502933e-06, + "loss": 0.6684, + "step": 988 + }, + { + "epoch": 0.8191606847045831, + "grad_norm": 0.8911624550819397, + "learning_rate": 4.8112738116768276e-06, + "loss": 0.6448, + "step": 989 + }, + { + "epoch": 0.8199889563776919, + "grad_norm": 0.9036282896995544, + "learning_rate": 4.810854431640784e-06, + "loss": 0.703, + "step": 990 + }, + { + "epoch": 0.8208172280508007, + "grad_norm": 0.9197635650634766, + "learning_rate": 4.81043460447595e-06, + "loss": 0.6655, + "step": 991 + }, + { + "epoch": 0.8216454997239094, + "grad_norm": 0.9548643827438354, + "learning_rate": 4.8100143302635575e-06, + "loss": 0.6892, + "step": 992 + }, + { + "epoch": 0.8224737713970183, + "grad_norm": 0.9139766097068787, + "learning_rate": 4.809593609084925e-06, + "loss": 0.6723, + "step": 993 + }, + { + "epoch": 0.823302043070127, + "grad_norm": 0.9067809581756592, + "learning_rate": 4.8091724410214595e-06, + "loss": 0.6649, + "step": 994 + }, + { + "epoch": 0.8241303147432358, + "grad_norm": 0.9705583453178406, + "learning_rate": 4.808750826154652e-06, + "loss": 0.6889, + "step": 995 + }, + { + "epoch": 0.8249585864163446, + "grad_norm": 0.9664806127548218, + "learning_rate": 4.808328764566082e-06, + "loss": 0.6758, + "step": 996 + }, + { + "epoch": 0.8257868580894533, + "grad_norm": 0.8840530514717102, + "learning_rate": 4.807906256337414e-06, + "loss": 0.6719, + "step": 997 + }, + { + "epoch": 0.8266151297625621, + "grad_norm": 0.8803462386131287, + "learning_rate": 4.8074833015504e-06, + "loss": 0.6903, + "step": 998 + }, + { + "epoch": 0.8274434014356709, + "grad_norm": 0.9404571652412415, + "learning_rate": 4.8070599002868756e-06, + "loss": 0.6917, + "step": 999 + }, + { + "epoch": 0.8282716731087797, + "grad_norm": 1.0206997394561768, + "learning_rate": 4.806636052628768e-06, + "loss": 0.6854, + "step": 1000 + }, + { + "epoch": 0.8290999447818884, + "grad_norm": 0.9294790029525757, + "learning_rate": 4.806211758658086e-06, + "loss": 0.6791, + "step": 1001 + }, + { + "epoch": 0.8299282164549973, + "grad_norm": 0.874011754989624, + "learning_rate": 4.805787018456927e-06, + "loss": 0.6915, + "step": 1002 + }, + { + "epoch": 0.830756488128106, + "grad_norm": 0.9098061323165894, + "learning_rate": 4.805361832107474e-06, + "loss": 0.6643, + "step": 1003 + }, + { + "epoch": 0.8315847598012148, + "grad_norm": 0.9783986210823059, + "learning_rate": 4.804936199691997e-06, + "loss": 0.6853, + "step": 1004 + }, + { + "epoch": 0.8324130314743236, + "grad_norm": 0.9316959381103516, + "learning_rate": 4.804510121292852e-06, + "loss": 0.6696, + "step": 1005 + }, + { + "epoch": 0.8332413031474324, + "grad_norm": 0.8930666446685791, + "learning_rate": 4.80408359699248e-06, + "loss": 0.6805, + "step": 1006 + }, + { + "epoch": 0.8340695748205411, + "grad_norm": 0.9417555332183838, + "learning_rate": 4.803656626873412e-06, + "loss": 0.66, + "step": 1007 + }, + { + "epoch": 0.83489784649365, + "grad_norm": 0.9435075521469116, + "learning_rate": 4.80322921101826e-06, + "loss": 0.6664, + "step": 1008 + }, + { + "epoch": 0.8357261181667587, + "grad_norm": 0.9090076684951782, + "learning_rate": 4.8028013495097255e-06, + "loss": 0.6595, + "step": 1009 + }, + { + "epoch": 0.8365543898398675, + "grad_norm": 0.9024788737297058, + "learning_rate": 4.802373042430597e-06, + "loss": 0.6672, + "step": 1010 + }, + { + "epoch": 0.8373826615129762, + "grad_norm": 0.899813711643219, + "learning_rate": 4.801944289863747e-06, + "loss": 0.6743, + "step": 1011 + }, + { + "epoch": 0.8382109331860851, + "grad_norm": 0.887106716632843, + "learning_rate": 4.801515091892136e-06, + "loss": 0.6823, + "step": 1012 + }, + { + "epoch": 0.8390392048591938, + "grad_norm": 0.8690978288650513, + "learning_rate": 4.801085448598809e-06, + "loss": 0.6642, + "step": 1013 + }, + { + "epoch": 0.8398674765323026, + "grad_norm": 0.9131479859352112, + "learning_rate": 4.800655360066898e-06, + "loss": 0.6841, + "step": 1014 + }, + { + "epoch": 0.8406957482054114, + "grad_norm": 0.9031343460083008, + "learning_rate": 4.800224826379622e-06, + "loss": 0.6879, + "step": 1015 + }, + { + "epoch": 0.8415240198785201, + "grad_norm": 0.8914244174957275, + "learning_rate": 4.799793847620283e-06, + "loss": 0.6562, + "step": 1016 + }, + { + "epoch": 0.8423522915516289, + "grad_norm": 0.8692408800125122, + "learning_rate": 4.799362423872274e-06, + "loss": 0.6759, + "step": 1017 + }, + { + "epoch": 0.8431805632247377, + "grad_norm": 0.9511806964874268, + "learning_rate": 4.798930555219069e-06, + "loss": 0.6637, + "step": 1018 + }, + { + "epoch": 0.8440088348978465, + "grad_norm": 0.9556595683097839, + "learning_rate": 4.7984982417442325e-06, + "loss": 0.6925, + "step": 1019 + }, + { + "epoch": 0.8448371065709552, + "grad_norm": 0.9107236862182617, + "learning_rate": 4.798065483531413e-06, + "loss": 0.67, + "step": 1020 + }, + { + "epoch": 0.8456653782440641, + "grad_norm": 0.875293493270874, + "learning_rate": 4.797632280664345e-06, + "loss": 0.6755, + "step": 1021 + }, + { + "epoch": 0.8464936499171728, + "grad_norm": 0.8972271084785461, + "learning_rate": 4.797198633226849e-06, + "loss": 0.6637, + "step": 1022 + }, + { + "epoch": 0.8473219215902816, + "grad_norm": 0.8903595805168152, + "learning_rate": 4.796764541302831e-06, + "loss": 0.6997, + "step": 1023 + }, + { + "epoch": 0.8481501932633904, + "grad_norm": 0.8948636651039124, + "learning_rate": 4.796330004976285e-06, + "loss": 0.6579, + "step": 1024 + }, + { + "epoch": 0.8489784649364992, + "grad_norm": 0.8590925335884094, + "learning_rate": 4.795895024331289e-06, + "loss": 0.6651, + "step": 1025 + }, + { + "epoch": 0.8498067366096079, + "grad_norm": 0.870811402797699, + "learning_rate": 4.795459599452007e-06, + "loss": 0.6707, + "step": 1026 + }, + { + "epoch": 0.8506350082827168, + "grad_norm": 0.8807471394538879, + "learning_rate": 4.795023730422692e-06, + "loss": 0.6787, + "step": 1027 + }, + { + "epoch": 0.8514632799558255, + "grad_norm": 0.8817671537399292, + "learning_rate": 4.794587417327678e-06, + "loss": 0.66, + "step": 1028 + }, + { + "epoch": 0.8522915516289343, + "grad_norm": 0.8924570679664612, + "learning_rate": 4.794150660251389e-06, + "loss": 0.6842, + "step": 1029 + }, + { + "epoch": 0.8531198233020431, + "grad_norm": 0.8746559619903564, + "learning_rate": 4.793713459278333e-06, + "loss": 0.6582, + "step": 1030 + }, + { + "epoch": 0.8539480949751519, + "grad_norm": 0.8885277509689331, + "learning_rate": 4.793275814493104e-06, + "loss": 0.6784, + "step": 1031 + }, + { + "epoch": 0.8547763666482606, + "grad_norm": 0.8797710537910461, + "learning_rate": 4.792837725980383e-06, + "loss": 0.6638, + "step": 1032 + }, + { + "epoch": 0.8556046383213695, + "grad_norm": 0.8932235240936279, + "learning_rate": 4.7923991938249355e-06, + "loss": 0.6918, + "step": 1033 + }, + { + "epoch": 0.8564329099944782, + "grad_norm": 0.9317487478256226, + "learning_rate": 4.791960218111613e-06, + "loss": 0.6905, + "step": 1034 + }, + { + "epoch": 0.8572611816675869, + "grad_norm": 0.8760019540786743, + "learning_rate": 4.791520798925354e-06, + "loss": 0.673, + "step": 1035 + }, + { + "epoch": 0.8580894533406958, + "grad_norm": 0.8644204139709473, + "learning_rate": 4.7910809363511825e-06, + "loss": 0.6755, + "step": 1036 + }, + { + "epoch": 0.8589177250138045, + "grad_norm": 0.8761067390441895, + "learning_rate": 4.790640630474206e-06, + "loss": 0.699, + "step": 1037 + }, + { + "epoch": 0.8597459966869133, + "grad_norm": 0.9135014414787292, + "learning_rate": 4.790199881379621e-06, + "loss": 0.6659, + "step": 1038 + }, + { + "epoch": 0.860574268360022, + "grad_norm": 0.9065434336662292, + "learning_rate": 4.789758689152708e-06, + "loss": 0.6664, + "step": 1039 + }, + { + "epoch": 0.8614025400331309, + "grad_norm": 0.8814099431037903, + "learning_rate": 4.789317053878833e-06, + "loss": 0.681, + "step": 1040 + }, + { + "epoch": 0.8622308117062396, + "grad_norm": 0.872502863407135, + "learning_rate": 4.788874975643449e-06, + "loss": 0.7044, + "step": 1041 + }, + { + "epoch": 0.8630590833793484, + "grad_norm": 0.9532367587089539, + "learning_rate": 4.788432454532093e-06, + "loss": 0.655, + "step": 1042 + }, + { + "epoch": 0.8638873550524572, + "grad_norm": 0.9009373784065247, + "learning_rate": 4.787989490630391e-06, + "loss": 0.6946, + "step": 1043 + }, + { + "epoch": 0.864715626725566, + "grad_norm": 0.889933168888092, + "learning_rate": 4.787546084024051e-06, + "loss": 0.6595, + "step": 1044 + }, + { + "epoch": 0.8655438983986747, + "grad_norm": 0.9179873466491699, + "learning_rate": 4.787102234798867e-06, + "loss": 0.658, + "step": 1045 + }, + { + "epoch": 0.8663721700717836, + "grad_norm": 0.8873292207717896, + "learning_rate": 4.786657943040721e-06, + "loss": 0.7023, + "step": 1046 + }, + { + "epoch": 0.8672004417448923, + "grad_norm": 0.8388610482215881, + "learning_rate": 4.78621320883558e-06, + "loss": 0.6644, + "step": 1047 + }, + { + "epoch": 0.8680287134180011, + "grad_norm": 0.8971349596977234, + "learning_rate": 4.785768032269493e-06, + "loss": 0.6552, + "step": 1048 + }, + { + "epoch": 0.8688569850911099, + "grad_norm": 0.8789131045341492, + "learning_rate": 4.785322413428601e-06, + "loss": 0.6681, + "step": 1049 + }, + { + "epoch": 0.8696852567642187, + "grad_norm": 0.8867958784103394, + "learning_rate": 4.784876352399125e-06, + "loss": 0.6859, + "step": 1050 + }, + { + "epoch": 0.8705135284373274, + "grad_norm": 0.8584237694740295, + "learning_rate": 4.784429849267373e-06, + "loss": 0.6655, + "step": 1051 + }, + { + "epoch": 0.8713418001104363, + "grad_norm": 0.9230811595916748, + "learning_rate": 4.783982904119743e-06, + "loss": 0.6648, + "step": 1052 + }, + { + "epoch": 0.872170071783545, + "grad_norm": 0.9217789769172668, + "learning_rate": 4.783535517042709e-06, + "loss": 0.6867, + "step": 1053 + }, + { + "epoch": 0.8729983434566538, + "grad_norm": 0.9209116697311401, + "learning_rate": 4.783087688122842e-06, + "loss": 0.6972, + "step": 1054 + }, + { + "epoch": 0.8738266151297626, + "grad_norm": 0.9493632316589355, + "learning_rate": 4.782639417446789e-06, + "loss": 0.6785, + "step": 1055 + }, + { + "epoch": 0.8746548868028713, + "grad_norm": 0.8790786862373352, + "learning_rate": 4.782190705101286e-06, + "loss": 0.6925, + "step": 1056 + }, + { + "epoch": 0.8754831584759801, + "grad_norm": 0.8662691712379456, + "learning_rate": 4.781741551173157e-06, + "loss": 0.6686, + "step": 1057 + }, + { + "epoch": 0.8763114301490889, + "grad_norm": 0.919658899307251, + "learning_rate": 4.781291955749307e-06, + "loss": 0.6814, + "step": 1058 + }, + { + "epoch": 0.8771397018221977, + "grad_norm": 0.9639708399772644, + "learning_rate": 4.780841918916731e-06, + "loss": 0.6629, + "step": 1059 + }, + { + "epoch": 0.8779679734953064, + "grad_norm": 0.8694306015968323, + "learning_rate": 4.780391440762505e-06, + "loss": 0.6893, + "step": 1060 + }, + { + "epoch": 0.8787962451684153, + "grad_norm": 0.9725603461265564, + "learning_rate": 4.779940521373793e-06, + "loss": 0.67, + "step": 1061 + }, + { + "epoch": 0.879624516841524, + "grad_norm": 0.93159419298172, + "learning_rate": 4.779489160837842e-06, + "loss": 0.686, + "step": 1062 + }, + { + "epoch": 0.8804527885146328, + "grad_norm": 0.9366991519927979, + "learning_rate": 4.779037359241989e-06, + "loss": 0.649, + "step": 1063 + }, + { + "epoch": 0.8812810601877415, + "grad_norm": 0.9052289724349976, + "learning_rate": 4.7785851166736505e-06, + "loss": 0.6677, + "step": 1064 + }, + { + "epoch": 0.8821093318608504, + "grad_norm": 0.882251501083374, + "learning_rate": 4.778132433220334e-06, + "loss": 0.6886, + "step": 1065 + }, + { + "epoch": 0.8829376035339591, + "grad_norm": 0.8867698907852173, + "learning_rate": 4.777679308969627e-06, + "loss": 0.6645, + "step": 1066 + }, + { + "epoch": 0.883765875207068, + "grad_norm": 0.890396773815155, + "learning_rate": 4.777225744009207e-06, + "loss": 0.6684, + "step": 1067 + }, + { + "epoch": 0.8845941468801767, + "grad_norm": 0.8827376365661621, + "learning_rate": 4.776771738426833e-06, + "loss": 0.6738, + "step": 1068 + }, + { + "epoch": 0.8854224185532855, + "grad_norm": 0.8953343033790588, + "learning_rate": 4.776317292310352e-06, + "loss": 0.684, + "step": 1069 + }, + { + "epoch": 0.8862506902263942, + "grad_norm": 0.8751050233840942, + "learning_rate": 4.775862405747694e-06, + "loss": 0.6779, + "step": 1070 + }, + { + "epoch": 0.8870789618995031, + "grad_norm": 0.9136618971824646, + "learning_rate": 4.775407078826876e-06, + "loss": 0.6939, + "step": 1071 + }, + { + "epoch": 0.8879072335726118, + "grad_norm": 0.8887941837310791, + "learning_rate": 4.774951311635999e-06, + "loss": 0.6547, + "step": 1072 + }, + { + "epoch": 0.8887355052457206, + "grad_norm": 0.8873274326324463, + "learning_rate": 4.77449510426325e-06, + "loss": 0.6769, + "step": 1073 + }, + { + "epoch": 0.8895637769188294, + "grad_norm": 0.9307074546813965, + "learning_rate": 4.774038456796901e-06, + "loss": 0.6984, + "step": 1074 + }, + { + "epoch": 0.8903920485919381, + "grad_norm": 0.9022747874259949, + "learning_rate": 4.77358136932531e-06, + "loss": 0.6733, + "step": 1075 + }, + { + "epoch": 0.8912203202650469, + "grad_norm": 0.8779094219207764, + "learning_rate": 4.773123841936916e-06, + "loss": 0.6434, + "step": 1076 + }, + { + "epoch": 0.8920485919381557, + "grad_norm": 0.9465747475624084, + "learning_rate": 4.7726658747202505e-06, + "loss": 0.6761, + "step": 1077 + }, + { + "epoch": 0.8928768636112645, + "grad_norm": 0.8694362044334412, + "learning_rate": 4.772207467763923e-06, + "loss": 0.6756, + "step": 1078 + }, + { + "epoch": 0.8937051352843732, + "grad_norm": 0.8839150071144104, + "learning_rate": 4.771748621156631e-06, + "loss": 0.6714, + "step": 1079 + }, + { + "epoch": 0.8945334069574821, + "grad_norm": 0.9220975041389465, + "learning_rate": 4.7712893349871585e-06, + "loss": 0.6724, + "step": 1080 + }, + { + "epoch": 0.8953616786305908, + "grad_norm": 0.9234760999679565, + "learning_rate": 4.770829609344372e-06, + "loss": 0.67, + "step": 1081 + }, + { + "epoch": 0.8961899503036996, + "grad_norm": 0.9262274503707886, + "learning_rate": 4.770369444317225e-06, + "loss": 0.6619, + "step": 1082 + }, + { + "epoch": 0.8970182219768084, + "grad_norm": 0.8889190554618835, + "learning_rate": 4.769908839994753e-06, + "loss": 0.688, + "step": 1083 + }, + { + "epoch": 0.8978464936499172, + "grad_norm": 0.9028982520103455, + "learning_rate": 4.769447796466082e-06, + "loss": 0.6633, + "step": 1084 + }, + { + "epoch": 0.8986747653230259, + "grad_norm": 0.9155063033103943, + "learning_rate": 4.7689863138204185e-06, + "loss": 0.6709, + "step": 1085 + }, + { + "epoch": 0.8995030369961348, + "grad_norm": 0.9066855311393738, + "learning_rate": 4.768524392147052e-06, + "loss": 0.6713, + "step": 1086 + }, + { + "epoch": 0.9003313086692435, + "grad_norm": 0.9365159273147583, + "learning_rate": 4.768062031535365e-06, + "loss": 0.6675, + "step": 1087 + }, + { + "epoch": 0.9011595803423523, + "grad_norm": 0.9106162190437317, + "learning_rate": 4.767599232074816e-06, + "loss": 0.6804, + "step": 1088 + }, + { + "epoch": 0.9019878520154611, + "grad_norm": 0.9183253645896912, + "learning_rate": 4.767135993854955e-06, + "loss": 0.658, + "step": 1089 + }, + { + "epoch": 0.9028161236885699, + "grad_norm": 0.9543409943580627, + "learning_rate": 4.766672316965413e-06, + "loss": 0.696, + "step": 1090 + }, + { + "epoch": 0.9036443953616786, + "grad_norm": 0.8894531726837158, + "learning_rate": 4.766208201495906e-06, + "loss": 0.6781, + "step": 1091 + }, + { + "epoch": 0.9044726670347875, + "grad_norm": 0.915542721748352, + "learning_rate": 4.765743647536238e-06, + "loss": 0.6804, + "step": 1092 + }, + { + "epoch": 0.9053009387078962, + "grad_norm": 0.8959763646125793, + "learning_rate": 4.765278655176296e-06, + "loss": 0.6873, + "step": 1093 + }, + { + "epoch": 0.906129210381005, + "grad_norm": 0.881188690662384, + "learning_rate": 4.76481322450605e-06, + "loss": 0.6582, + "step": 1094 + }, + { + "epoch": 0.9069574820541138, + "grad_norm": 0.9096754193305969, + "learning_rate": 4.764347355615558e-06, + "loss": 0.693, + "step": 1095 + }, + { + "epoch": 0.9077857537272225, + "grad_norm": 0.891132652759552, + "learning_rate": 4.76388104859496e-06, + "loss": 0.6781, + "step": 1096 + }, + { + "epoch": 0.9086140254003313, + "grad_norm": 0.893624484539032, + "learning_rate": 4.763414303534484e-06, + "loss": 0.6636, + "step": 1097 + }, + { + "epoch": 0.90944229707344, + "grad_norm": 0.8795678019523621, + "learning_rate": 4.762947120524438e-06, + "loss": 0.6604, + "step": 1098 + }, + { + "epoch": 0.9102705687465489, + "grad_norm": 0.9177265763282776, + "learning_rate": 4.76247949965522e-06, + "loss": 0.6557, + "step": 1099 + }, + { + "epoch": 0.9110988404196576, + "grad_norm": 0.9416985511779785, + "learning_rate": 4.762011441017309e-06, + "loss": 0.7008, + "step": 1100 + }, + { + "epoch": 0.9119271120927664, + "grad_norm": 0.9255479574203491, + "learning_rate": 4.7615429447012704e-06, + "loss": 0.6665, + "step": 1101 + }, + { + "epoch": 0.9127553837658752, + "grad_norm": 0.8948953151702881, + "learning_rate": 4.761074010797754e-06, + "loss": 0.6699, + "step": 1102 + }, + { + "epoch": 0.913583655438984, + "grad_norm": 0.9027442336082458, + "learning_rate": 4.760604639397493e-06, + "loss": 0.6649, + "step": 1103 + }, + { + "epoch": 0.9144119271120927, + "grad_norm": 0.9094238877296448, + "learning_rate": 4.7601348305913085e-06, + "loss": 0.6753, + "step": 1104 + }, + { + "epoch": 0.9152401987852016, + "grad_norm": 0.9098204374313354, + "learning_rate": 4.759664584470103e-06, + "loss": 0.6845, + "step": 1105 + }, + { + "epoch": 0.9160684704583103, + "grad_norm": 0.9339573383331299, + "learning_rate": 4.759193901124864e-06, + "loss": 0.6285, + "step": 1106 + }, + { + "epoch": 0.9168967421314191, + "grad_norm": 0.8812451958656311, + "learning_rate": 4.758722780646666e-06, + "loss": 0.6764, + "step": 1107 + }, + { + "epoch": 0.9177250138045279, + "grad_norm": 0.8832173347473145, + "learning_rate": 4.758251223126665e-06, + "loss": 0.6908, + "step": 1108 + }, + { + "epoch": 0.9185532854776367, + "grad_norm": 0.9124787449836731, + "learning_rate": 4.757779228656103e-06, + "loss": 0.6855, + "step": 1109 + }, + { + "epoch": 0.9193815571507454, + "grad_norm": 0.8778328895568848, + "learning_rate": 4.757306797326309e-06, + "loss": 0.6404, + "step": 1110 + }, + { + "epoch": 0.9202098288238543, + "grad_norm": 0.8831816911697388, + "learning_rate": 4.75683392922869e-06, + "loss": 0.6425, + "step": 1111 + }, + { + "epoch": 0.921038100496963, + "grad_norm": 0.9060006141662598, + "learning_rate": 4.756360624454744e-06, + "loss": 0.6738, + "step": 1112 + }, + { + "epoch": 0.9218663721700718, + "grad_norm": 0.8746777176856995, + "learning_rate": 4.755886883096051e-06, + "loss": 0.6665, + "step": 1113 + }, + { + "epoch": 0.9226946438431806, + "grad_norm": 0.9185945987701416, + "learning_rate": 4.755412705244274e-06, + "loss": 0.6564, + "step": 1114 + }, + { + "epoch": 0.9235229155162893, + "grad_norm": 0.9648581743240356, + "learning_rate": 4.754938090991165e-06, + "loss": 0.6597, + "step": 1115 + }, + { + "epoch": 0.9243511871893981, + "grad_norm": 0.8800092935562134, + "learning_rate": 4.754463040428555e-06, + "loss": 0.6884, + "step": 1116 + }, + { + "epoch": 0.9251794588625069, + "grad_norm": 0.8659892082214355, + "learning_rate": 4.753987553648363e-06, + "loss": 0.68, + "step": 1117 + }, + { + "epoch": 0.9260077305356157, + "grad_norm": 0.88991379737854, + "learning_rate": 4.75351163074259e-06, + "loss": 0.668, + "step": 1118 + }, + { + "epoch": 0.9268360022087244, + "grad_norm": 0.9683054089546204, + "learning_rate": 4.753035271803323e-06, + "loss": 0.6789, + "step": 1119 + }, + { + "epoch": 0.9276642738818333, + "grad_norm": 0.928001880645752, + "learning_rate": 4.752558476922735e-06, + "loss": 0.6786, + "step": 1120 + }, + { + "epoch": 0.928492545554942, + "grad_norm": 0.8760973215103149, + "learning_rate": 4.752081246193078e-06, + "loss": 0.6716, + "step": 1121 + }, + { + "epoch": 0.9293208172280508, + "grad_norm": 0.9171329140663147, + "learning_rate": 4.751603579706695e-06, + "loss": 0.662, + "step": 1122 + }, + { + "epoch": 0.9301490889011595, + "grad_norm": 0.9380231499671936, + "learning_rate": 4.751125477556008e-06, + "loss": 0.6836, + "step": 1123 + }, + { + "epoch": 0.9309773605742684, + "grad_norm": 0.9471802711486816, + "learning_rate": 4.750646939833526e-06, + "loss": 0.668, + "step": 1124 + }, + { + "epoch": 0.9318056322473771, + "grad_norm": 0.8860809206962585, + "learning_rate": 4.750167966631842e-06, + "loss": 0.6635, + "step": 1125 + }, + { + "epoch": 0.932633903920486, + "grad_norm": 0.8694644570350647, + "learning_rate": 4.7496885580436305e-06, + "loss": 0.6505, + "step": 1126 + }, + { + "epoch": 0.9334621755935947, + "grad_norm": 0.9602686166763306, + "learning_rate": 4.749208714161655e-06, + "loss": 0.6874, + "step": 1127 + }, + { + "epoch": 0.9342904472667035, + "grad_norm": 0.9198856353759766, + "learning_rate": 4.748728435078761e-06, + "loss": 0.697, + "step": 1128 + }, + { + "epoch": 0.9351187189398122, + "grad_norm": 0.9162499904632568, + "learning_rate": 4.748247720887876e-06, + "loss": 0.6521, + "step": 1129 + }, + { + "epoch": 0.9359469906129211, + "grad_norm": 0.9064576029777527, + "learning_rate": 4.747766571682016e-06, + "loss": 0.6525, + "step": 1130 + }, + { + "epoch": 0.9367752622860298, + "grad_norm": 0.9518057703971863, + "learning_rate": 4.747284987554277e-06, + "loss": 0.6745, + "step": 1131 + }, + { + "epoch": 0.9376035339591386, + "grad_norm": 0.8963000178337097, + "learning_rate": 4.746802968597842e-06, + "loss": 0.6999, + "step": 1132 + }, + { + "epoch": 0.9384318056322474, + "grad_norm": 0.8687212467193604, + "learning_rate": 4.746320514905977e-06, + "loss": 0.6353, + "step": 1133 + }, + { + "epoch": 0.9392600773053562, + "grad_norm": 0.8598489165306091, + "learning_rate": 4.745837626572033e-06, + "loss": 0.6897, + "step": 1134 + }, + { + "epoch": 0.9400883489784649, + "grad_norm": 0.9425498843193054, + "learning_rate": 4.745354303689444e-06, + "loss": 0.667, + "step": 1135 + }, + { + "epoch": 0.9409166206515737, + "grad_norm": 0.9026349782943726, + "learning_rate": 4.744870546351727e-06, + "loss": 0.676, + "step": 1136 + }, + { + "epoch": 0.9417448923246825, + "grad_norm": 0.8824150562286377, + "learning_rate": 4.744386354652486e-06, + "loss": 0.6941, + "step": 1137 + }, + { + "epoch": 0.9425731639977912, + "grad_norm": 0.8910468816757202, + "learning_rate": 4.743901728685406e-06, + "loss": 0.6701, + "step": 1138 + }, + { + "epoch": 0.9434014356709001, + "grad_norm": 0.9114013314247131, + "learning_rate": 4.74341666854426e-06, + "loss": 0.6735, + "step": 1139 + }, + { + "epoch": 0.9442297073440088, + "grad_norm": 0.9107307195663452, + "learning_rate": 4.742931174322902e-06, + "loss": 0.6505, + "step": 1140 + }, + { + "epoch": 0.9450579790171176, + "grad_norm": 0.8906923532485962, + "learning_rate": 4.742445246115271e-06, + "loss": 0.6648, + "step": 1141 + }, + { + "epoch": 0.9458862506902264, + "grad_norm": 0.902644157409668, + "learning_rate": 4.741958884015387e-06, + "loss": 0.6674, + "step": 1142 + }, + { + "epoch": 0.9467145223633352, + "grad_norm": 0.9522087574005127, + "learning_rate": 4.741472088117358e-06, + "loss": 0.6825, + "step": 1143 + }, + { + "epoch": 0.9475427940364439, + "grad_norm": 0.9152255654335022, + "learning_rate": 4.740984858515376e-06, + "loss": 0.6708, + "step": 1144 + }, + { + "epoch": 0.9483710657095528, + "grad_norm": 0.8779835104942322, + "learning_rate": 4.740497195303713e-06, + "loss": 0.6562, + "step": 1145 + }, + { + "epoch": 0.9491993373826615, + "grad_norm": 0.8842780590057373, + "learning_rate": 4.740009098576729e-06, + "loss": 0.6769, + "step": 1146 + }, + { + "epoch": 0.9500276090557703, + "grad_norm": 0.9137415289878845, + "learning_rate": 4.739520568428866e-06, + "loss": 0.6705, + "step": 1147 + }, + { + "epoch": 0.950855880728879, + "grad_norm": 0.8867765665054321, + "learning_rate": 4.739031604954648e-06, + "loss": 0.659, + "step": 1148 + }, + { + "epoch": 0.9516841524019879, + "grad_norm": 0.8965260982513428, + "learning_rate": 4.7385422082486874e-06, + "loss": 0.6744, + "step": 1149 + }, + { + "epoch": 0.9525124240750966, + "grad_norm": 0.8861697912216187, + "learning_rate": 4.738052378405677e-06, + "loss": 0.6575, + "step": 1150 + }, + { + "epoch": 0.9533406957482055, + "grad_norm": 0.9442310333251953, + "learning_rate": 4.737562115520394e-06, + "loss": 0.7164, + "step": 1151 + }, + { + "epoch": 0.9541689674213142, + "grad_norm": 0.9340727925300598, + "learning_rate": 4.7370714196877e-06, + "loss": 0.6577, + "step": 1152 + }, + { + "epoch": 0.954997239094423, + "grad_norm": 0.9337487816810608, + "learning_rate": 4.7365802910025405e-06, + "loss": 0.6664, + "step": 1153 + }, + { + "epoch": 0.9558255107675317, + "grad_norm": 0.8849001526832581, + "learning_rate": 4.736088729559943e-06, + "loss": 0.6721, + "step": 1154 + }, + { + "epoch": 0.9566537824406405, + "grad_norm": 0.917323648929596, + "learning_rate": 4.735596735455022e-06, + "loss": 0.6715, + "step": 1155 + }, + { + "epoch": 0.9574820541137493, + "grad_norm": 0.9480959177017212, + "learning_rate": 4.735104308782972e-06, + "loss": 0.6784, + "step": 1156 + }, + { + "epoch": 0.958310325786858, + "grad_norm": 0.9519251585006714, + "learning_rate": 4.734611449639073e-06, + "loss": 0.6617, + "step": 1157 + }, + { + "epoch": 0.9591385974599669, + "grad_norm": 0.8652079105377197, + "learning_rate": 4.734118158118689e-06, + "loss": 0.6846, + "step": 1158 + }, + { + "epoch": 0.9599668691330756, + "grad_norm": 0.8860392570495605, + "learning_rate": 4.733624434317269e-06, + "loss": 0.6729, + "step": 1159 + }, + { + "epoch": 0.9607951408061844, + "grad_norm": 0.9467882513999939, + "learning_rate": 4.7331302783303416e-06, + "loss": 0.6362, + "step": 1160 + }, + { + "epoch": 0.9616234124792932, + "grad_norm": 0.9138529896736145, + "learning_rate": 4.732635690253523e-06, + "loss": 0.6613, + "step": 1161 + }, + { + "epoch": 0.962451684152402, + "grad_norm": 0.9401054978370667, + "learning_rate": 4.7321406701825105e-06, + "loss": 0.6845, + "step": 1162 + }, + { + "epoch": 0.9632799558255107, + "grad_norm": 0.9049288630485535, + "learning_rate": 4.731645218213086e-06, + "loss": 0.6857, + "step": 1163 + }, + { + "epoch": 0.9641082274986196, + "grad_norm": 0.8972534537315369, + "learning_rate": 4.731149334441114e-06, + "loss": 0.6745, + "step": 1164 + }, + { + "epoch": 0.9649364991717283, + "grad_norm": 1.0799483060836792, + "learning_rate": 4.730653018962545e-06, + "loss": 0.6754, + "step": 1165 + }, + { + "epoch": 0.9657647708448371, + "grad_norm": 0.9841833710670471, + "learning_rate": 4.730156271873412e-06, + "loss": 0.6718, + "step": 1166 + }, + { + "epoch": 0.9665930425179459, + "grad_norm": 0.9388216733932495, + "learning_rate": 4.729659093269828e-06, + "loss": 0.6622, + "step": 1167 + }, + { + "epoch": 0.9674213141910547, + "grad_norm": 0.8879573345184326, + "learning_rate": 4.729161483247994e-06, + "loss": 0.6909, + "step": 1168 + }, + { + "epoch": 0.9682495858641634, + "grad_norm": 0.9713517427444458, + "learning_rate": 4.728663441904194e-06, + "loss": 0.6956, + "step": 1169 + }, + { + "epoch": 0.9690778575372723, + "grad_norm": 0.9200823903083801, + "learning_rate": 4.7281649693347925e-06, + "loss": 0.6766, + "step": 1170 + }, + { + "epoch": 0.969906129210381, + "grad_norm": 0.8983714580535889, + "learning_rate": 4.727666065636241e-06, + "loss": 0.6626, + "step": 1171 + }, + { + "epoch": 0.9707344008834898, + "grad_norm": 0.8792792558670044, + "learning_rate": 4.72716673090507e-06, + "loss": 0.666, + "step": 1172 + }, + { + "epoch": 0.9715626725565986, + "grad_norm": 1.0116846561431885, + "learning_rate": 4.7266669652379005e-06, + "loss": 0.6574, + "step": 1173 + }, + { + "epoch": 0.9723909442297074, + "grad_norm": 0.9433228373527527, + "learning_rate": 4.7261667687314285e-06, + "loss": 0.6972, + "step": 1174 + }, + { + "epoch": 0.9732192159028161, + "grad_norm": 0.8979770541191101, + "learning_rate": 4.725666141482439e-06, + "loss": 0.6523, + "step": 1175 + }, + { + "epoch": 0.9740474875759249, + "grad_norm": 0.8990604281425476, + "learning_rate": 4.725165083587798e-06, + "loss": 0.6535, + "step": 1176 + }, + { + "epoch": 0.9748757592490337, + "grad_norm": 0.9411956667900085, + "learning_rate": 4.724663595144457e-06, + "loss": 0.6947, + "step": 1177 + }, + { + "epoch": 0.9757040309221424, + "grad_norm": 1.0203242301940918, + "learning_rate": 4.724161676249448e-06, + "loss": 0.6894, + "step": 1178 + }, + { + "epoch": 0.9765323025952513, + "grad_norm": 0.8938170075416565, + "learning_rate": 4.723659326999888e-06, + "loss": 0.6648, + "step": 1179 + }, + { + "epoch": 0.97736057426836, + "grad_norm": 0.8810412883758545, + "learning_rate": 4.7231565474929765e-06, + "loss": 0.6761, + "step": 1180 + }, + { + "epoch": 0.9781888459414688, + "grad_norm": 0.8903076648712158, + "learning_rate": 4.722653337825998e-06, + "loss": 0.6738, + "step": 1181 + }, + { + "epoch": 0.9790171176145775, + "grad_norm": 0.905060350894928, + "learning_rate": 4.7221496980963175e-06, + "loss": 0.6625, + "step": 1182 + }, + { + "epoch": 0.9798453892876864, + "grad_norm": 0.8791100978851318, + "learning_rate": 4.721645628401385e-06, + "loss": 0.6884, + "step": 1183 + }, + { + "epoch": 0.9806736609607951, + "grad_norm": 0.897101104259491, + "learning_rate": 4.721141128838733e-06, + "loss": 0.6693, + "step": 1184 + }, + { + "epoch": 0.981501932633904, + "grad_norm": 0.9102833867073059, + "learning_rate": 4.7206361995059775e-06, + "loss": 0.6682, + "step": 1185 + }, + { + "epoch": 0.9823302043070127, + "grad_norm": 0.9492197632789612, + "learning_rate": 4.720130840500819e-06, + "loss": 0.6501, + "step": 1186 + }, + { + "epoch": 0.9831584759801215, + "grad_norm": 0.9783265590667725, + "learning_rate": 4.719625051921038e-06, + "loss": 0.6454, + "step": 1187 + }, + { + "epoch": 0.9839867476532302, + "grad_norm": 0.8938109874725342, + "learning_rate": 4.719118833864501e-06, + "loss": 0.6671, + "step": 1188 + }, + { + "epoch": 0.9848150193263391, + "grad_norm": 0.897484302520752, + "learning_rate": 4.718612186429156e-06, + "loss": 0.6664, + "step": 1189 + }, + { + "epoch": 0.9856432909994478, + "grad_norm": 0.9640376567840576, + "learning_rate": 4.718105109713034e-06, + "loss": 0.6845, + "step": 1190 + }, + { + "epoch": 0.9864715626725566, + "grad_norm": 0.9495493769645691, + "learning_rate": 4.7175976038142505e-06, + "loss": 0.6864, + "step": 1191 + }, + { + "epoch": 0.9872998343456654, + "grad_norm": 0.9390130639076233, + "learning_rate": 4.717089668831002e-06, + "loss": 0.6734, + "step": 1192 + }, + { + "epoch": 0.9881281060187742, + "grad_norm": 0.8854318261146545, + "learning_rate": 4.716581304861571e-06, + "loss": 0.6546, + "step": 1193 + }, + { + "epoch": 0.9889563776918829, + "grad_norm": 0.9242726564407349, + "learning_rate": 4.71607251200432e-06, + "loss": 0.6663, + "step": 1194 + }, + { + "epoch": 0.9897846493649917, + "grad_norm": 0.9302003383636475, + "learning_rate": 4.715563290357696e-06, + "loss": 0.6844, + "step": 1195 + }, + { + "epoch": 0.9906129210381005, + "grad_norm": 0.9279932975769043, + "learning_rate": 4.715053640020228e-06, + "loss": 0.6731, + "step": 1196 + }, + { + "epoch": 0.9914411927112092, + "grad_norm": 0.8753612041473389, + "learning_rate": 4.71454356109053e-06, + "loss": 0.6738, + "step": 1197 + }, + { + "epoch": 0.9922694643843181, + "grad_norm": 0.890133261680603, + "learning_rate": 4.714033053667296e-06, + "loss": 0.6638, + "step": 1198 + }, + { + "epoch": 0.9930977360574268, + "grad_norm": 0.9343729615211487, + "learning_rate": 4.713522117849305e-06, + "loss": 0.6826, + "step": 1199 + }, + { + "epoch": 0.9939260077305356, + "grad_norm": 0.9052755832672119, + "learning_rate": 4.7130107537354185e-06, + "loss": 0.6897, + "step": 1200 + }, + { + "epoch": 0.9947542794036444, + "grad_norm": 0.8683924674987793, + "learning_rate": 4.7124989614245805e-06, + "loss": 0.6687, + "step": 1201 + }, + { + "epoch": 0.9955825510767532, + "grad_norm": 0.8677132725715637, + "learning_rate": 4.711986741015818e-06, + "loss": 0.6917, + "step": 1202 + }, + { + "epoch": 0.9964108227498619, + "grad_norm": 0.8840690851211548, + "learning_rate": 4.711474092608243e-06, + "loss": 0.6841, + "step": 1203 + }, + { + "epoch": 0.9972390944229708, + "grad_norm": 0.9750391840934753, + "learning_rate": 4.7109610163010435e-06, + "loss": 0.6574, + "step": 1204 + }, + { + "epoch": 0.9980673660960795, + "grad_norm": 0.8752943873405457, + "learning_rate": 4.710447512193499e-06, + "loss": 0.636, + "step": 1205 + }, + { + "epoch": 0.9988956377691883, + "grad_norm": 0.9100313782691956, + "learning_rate": 4.709933580384965e-06, + "loss": 0.6712, + "step": 1206 + }, + { + "epoch": 0.999723909442297, + "grad_norm": 0.9202356338500977, + "learning_rate": 4.709419220974886e-06, + "loss": 0.6641, + "step": 1207 + }, + { + "epoch": 1.0, + "grad_norm": 0.9202356338500977, + "learning_rate": 4.708904434062783e-06, + "loss": 0.6199, + "step": 1208 + }, + { + "epoch": 1.0008282716731087, + "grad_norm": 1.6839245557785034, + "learning_rate": 4.708389219748263e-06, + "loss": 0.643, + "step": 1209 + }, + { + "epoch": 1.0016565433462175, + "grad_norm": 0.8636482357978821, + "learning_rate": 4.707873578131015e-06, + "loss": 0.6467, + "step": 1210 + }, + { + "epoch": 1.0024848150193264, + "grad_norm": 0.8548988103866577, + "learning_rate": 4.707357509310811e-06, + "loss": 0.6574, + "step": 1211 + }, + { + "epoch": 1.0033130866924351, + "grad_norm": 0.8961352705955505, + "learning_rate": 4.7068410133875066e-06, + "loss": 0.6367, + "step": 1212 + }, + { + "epoch": 1.0041413583655439, + "grad_norm": 0.8914269208908081, + "learning_rate": 4.706324090461037e-06, + "loss": 0.6294, + "step": 1213 + }, + { + "epoch": 1.0049696300386526, + "grad_norm": 0.879915177822113, + "learning_rate": 4.7058067406314235e-06, + "loss": 0.6346, + "step": 1214 + }, + { + "epoch": 1.0057979017117615, + "grad_norm": 0.8798143863677979, + "learning_rate": 4.7052889639987674e-06, + "loss": 0.6559, + "step": 1215 + }, + { + "epoch": 1.0066261733848703, + "grad_norm": 0.8888561129570007, + "learning_rate": 4.704770760663255e-06, + "loss": 0.6471, + "step": 1216 + }, + { + "epoch": 1.007454445057979, + "grad_norm": 0.8707575798034668, + "learning_rate": 4.704252130725152e-06, + "loss": 0.6485, + "step": 1217 + }, + { + "epoch": 1.0082827167310877, + "grad_norm": 0.8609346151351929, + "learning_rate": 4.7037330742848105e-06, + "loss": 0.6544, + "step": 1218 + }, + { + "epoch": 1.0091109884041967, + "grad_norm": 0.8588053584098816, + "learning_rate": 4.703213591442663e-06, + "loss": 0.6421, + "step": 1219 + }, + { + "epoch": 1.0099392600773054, + "grad_norm": 0.8932695388793945, + "learning_rate": 4.702693682299222e-06, + "loss": 0.6333, + "step": 1220 + }, + { + "epoch": 1.0107675317504141, + "grad_norm": 0.8615652322769165, + "learning_rate": 4.7021733469550865e-06, + "loss": 0.6371, + "step": 1221 + }, + { + "epoch": 1.0115958034235228, + "grad_norm": 0.8876634240150452, + "learning_rate": 4.7016525855109376e-06, + "loss": 0.6368, + "step": 1222 + }, + { + "epoch": 1.0124240750966318, + "grad_norm": 0.895575225353241, + "learning_rate": 4.701131398067537e-06, + "loss": 0.6226, + "step": 1223 + }, + { + "epoch": 1.0132523467697405, + "grad_norm": 0.8836749196052551, + "learning_rate": 4.70060978472573e-06, + "loss": 0.6516, + "step": 1224 + }, + { + "epoch": 1.0140806184428492, + "grad_norm": 0.9471280574798584, + "learning_rate": 4.700087745586442e-06, + "loss": 0.633, + "step": 1225 + }, + { + "epoch": 1.014908890115958, + "grad_norm": 0.8721561431884766, + "learning_rate": 4.699565280750685e-06, + "loss": 0.6568, + "step": 1226 + }, + { + "epoch": 1.015737161789067, + "grad_norm": 0.8946447372436523, + "learning_rate": 4.69904239031955e-06, + "loss": 0.6503, + "step": 1227 + }, + { + "epoch": 1.0165654334621756, + "grad_norm": 0.8854631185531616, + "learning_rate": 4.698519074394212e-06, + "loss": 0.6341, + "step": 1228 + }, + { + "epoch": 1.0173937051352844, + "grad_norm": 0.9348397850990295, + "learning_rate": 4.697995333075927e-06, + "loss": 0.6317, + "step": 1229 + }, + { + "epoch": 1.018221976808393, + "grad_norm": 0.9037336111068726, + "learning_rate": 4.697471166466035e-06, + "loss": 0.6618, + "step": 1230 + }, + { + "epoch": 1.0190502484815018, + "grad_norm": 0.8866400718688965, + "learning_rate": 4.696946574665957e-06, + "loss": 0.6421, + "step": 1231 + }, + { + "epoch": 1.0198785201546108, + "grad_norm": 0.8897701501846313, + "learning_rate": 4.696421557777196e-06, + "loss": 0.6608, + "step": 1232 + }, + { + "epoch": 1.0207067918277195, + "grad_norm": 0.9046173095703125, + "learning_rate": 4.6958961159013386e-06, + "loss": 0.6464, + "step": 1233 + }, + { + "epoch": 1.0215350635008282, + "grad_norm": 0.9492906332015991, + "learning_rate": 4.695370249140052e-06, + "loss": 0.6578, + "step": 1234 + }, + { + "epoch": 1.022363335173937, + "grad_norm": 0.8906951546669006, + "learning_rate": 4.6948439575950885e-06, + "loss": 0.6276, + "step": 1235 + }, + { + "epoch": 1.023191606847046, + "grad_norm": 0.9218266606330872, + "learning_rate": 4.694317241368278e-06, + "loss": 0.627, + "step": 1236 + }, + { + "epoch": 1.0240198785201546, + "grad_norm": 0.901157557964325, + "learning_rate": 4.693790100561537e-06, + "loss": 0.6379, + "step": 1237 + }, + { + "epoch": 1.0248481501932634, + "grad_norm": 0.8992452025413513, + "learning_rate": 4.693262535276863e-06, + "loss": 0.6585, + "step": 1238 + }, + { + "epoch": 1.025676421866372, + "grad_norm": 0.9525597095489502, + "learning_rate": 4.692734545616331e-06, + "loss": 0.6625, + "step": 1239 + }, + { + "epoch": 1.026504693539481, + "grad_norm": 0.9229065775871277, + "learning_rate": 4.692206131682106e-06, + "loss": 0.6454, + "step": 1240 + }, + { + "epoch": 1.0273329652125898, + "grad_norm": 1.012373685836792, + "learning_rate": 4.691677293576431e-06, + "loss": 0.6492, + "step": 1241 + }, + { + "epoch": 1.0281612368856985, + "grad_norm": 0.8650060892105103, + "learning_rate": 4.69114803140163e-06, + "loss": 0.6739, + "step": 1242 + }, + { + "epoch": 1.0289895085588072, + "grad_norm": 0.9396429657936096, + "learning_rate": 4.69061834526011e-06, + "loss": 0.6729, + "step": 1243 + }, + { + "epoch": 1.0298177802319162, + "grad_norm": 0.9584783911705017, + "learning_rate": 4.6900882352543614e-06, + "loss": 0.6678, + "step": 1244 + }, + { + "epoch": 1.030646051905025, + "grad_norm": 0.985394299030304, + "learning_rate": 4.689557701486954e-06, + "loss": 0.6478, + "step": 1245 + }, + { + "epoch": 1.0314743235781336, + "grad_norm": 0.9014914631843567, + "learning_rate": 4.689026744060543e-06, + "loss": 0.6393, + "step": 1246 + }, + { + "epoch": 1.0323025952512423, + "grad_norm": 0.9904847145080566, + "learning_rate": 4.688495363077863e-06, + "loss": 0.6752, + "step": 1247 + }, + { + "epoch": 1.0331308669243513, + "grad_norm": 0.9439241886138916, + "learning_rate": 4.687963558641731e-06, + "loss": 0.6417, + "step": 1248 + }, + { + "epoch": 1.03395913859746, + "grad_norm": 0.9277704954147339, + "learning_rate": 4.687431330855046e-06, + "loss": 0.6493, + "step": 1249 + }, + { + "epoch": 1.0347874102705688, + "grad_norm": 0.9456571340560913, + "learning_rate": 4.686898679820789e-06, + "loss": 0.6774, + "step": 1250 + }, + { + "epoch": 1.0356156819436775, + "grad_norm": 0.9043509364128113, + "learning_rate": 4.686365605642025e-06, + "loss": 0.6198, + "step": 1251 + }, + { + "epoch": 1.0364439536167862, + "grad_norm": 0.9246267676353455, + "learning_rate": 4.685832108421896e-06, + "loss": 0.6513, + "step": 1252 + }, + { + "epoch": 1.0372722252898952, + "grad_norm": 0.9227612614631653, + "learning_rate": 4.685298188263631e-06, + "loss": 0.6597, + "step": 1253 + }, + { + "epoch": 1.0381004969630039, + "grad_norm": 0.8927509188652039, + "learning_rate": 4.684763845270537e-06, + "loss": 0.641, + "step": 1254 + }, + { + "epoch": 1.0389287686361126, + "grad_norm": 0.8812937140464783, + "learning_rate": 4.684229079546005e-06, + "loss": 0.6292, + "step": 1255 + }, + { + "epoch": 1.0397570403092213, + "grad_norm": 0.9186881184577942, + "learning_rate": 4.683693891193508e-06, + "loss": 0.6485, + "step": 1256 + }, + { + "epoch": 1.0405853119823303, + "grad_norm": 0.8928134441375732, + "learning_rate": 4.683158280316599e-06, + "loss": 0.6435, + "step": 1257 + }, + { + "epoch": 1.041413583655439, + "grad_norm": 0.8919920325279236, + "learning_rate": 4.682622247018913e-06, + "loss": 0.622, + "step": 1258 + }, + { + "epoch": 1.0422418553285477, + "grad_norm": 0.885259747505188, + "learning_rate": 4.682085791404169e-06, + "loss": 0.6521, + "step": 1259 + }, + { + "epoch": 1.0430701270016565, + "grad_norm": 0.9034169316291809, + "learning_rate": 4.681548913576165e-06, + "loss": 0.6437, + "step": 1260 + }, + { + "epoch": 1.0438983986747654, + "grad_norm": 0.884848952293396, + "learning_rate": 4.6810116136387825e-06, + "loss": 0.6733, + "step": 1261 + }, + { + "epoch": 1.0447266703478741, + "grad_norm": 0.9137836694717407, + "learning_rate": 4.680473891695985e-06, + "loss": 0.6415, + "step": 1262 + }, + { + "epoch": 1.0455549420209829, + "grad_norm": 0.9044780135154724, + "learning_rate": 4.6799357478518145e-06, + "loss": 0.6453, + "step": 1263 + }, + { + "epoch": 1.0463832136940916, + "grad_norm": 0.9111152291297913, + "learning_rate": 4.6793971822103985e-06, + "loss": 0.6644, + "step": 1264 + }, + { + "epoch": 1.0472114853672005, + "grad_norm": 0.9158084988594055, + "learning_rate": 4.678858194875944e-06, + "loss": 0.6434, + "step": 1265 + }, + { + "epoch": 1.0480397570403093, + "grad_norm": 0.8970723152160645, + "learning_rate": 4.678318785952739e-06, + "loss": 0.6391, + "step": 1266 + }, + { + "epoch": 1.048868028713418, + "grad_norm": 0.8958215117454529, + "learning_rate": 4.677778955545155e-06, + "loss": 0.6562, + "step": 1267 + }, + { + "epoch": 1.0496963003865267, + "grad_norm": 0.9229645729064941, + "learning_rate": 4.677238703757644e-06, + "loss": 0.646, + "step": 1268 + }, + { + "epoch": 1.0505245720596355, + "grad_norm": 0.8986689448356628, + "learning_rate": 4.676698030694741e-06, + "loss": 0.6241, + "step": 1269 + }, + { + "epoch": 1.0513528437327444, + "grad_norm": 0.8943328261375427, + "learning_rate": 4.6761569364610585e-06, + "loss": 0.6455, + "step": 1270 + }, + { + "epoch": 1.0521811154058531, + "grad_norm": 0.9147390127182007, + "learning_rate": 4.675615421161295e-06, + "loss": 0.6299, + "step": 1271 + }, + { + "epoch": 1.0530093870789619, + "grad_norm": 0.9124075174331665, + "learning_rate": 4.675073484900229e-06, + "loss": 0.6379, + "step": 1272 + }, + { + "epoch": 1.0538376587520706, + "grad_norm": 0.9267525672912598, + "learning_rate": 4.674531127782718e-06, + "loss": 0.6478, + "step": 1273 + }, + { + "epoch": 1.0546659304251795, + "grad_norm": 0.8710078001022339, + "learning_rate": 4.673988349913707e-06, + "loss": 0.6314, + "step": 1274 + }, + { + "epoch": 1.0554942020982883, + "grad_norm": 0.8905670642852783, + "learning_rate": 4.673445151398214e-06, + "loss": 0.6456, + "step": 1275 + }, + { + "epoch": 1.056322473771397, + "grad_norm": 0.9125292897224426, + "learning_rate": 4.672901532341346e-06, + "loss": 0.6439, + "step": 1276 + }, + { + "epoch": 1.0571507454445057, + "grad_norm": 0.9003625512123108, + "learning_rate": 4.6723574928482865e-06, + "loss": 0.6561, + "step": 1277 + }, + { + "epoch": 1.0579790171176147, + "grad_norm": 0.96099454164505, + "learning_rate": 4.6718130330243014e-06, + "loss": 0.6515, + "step": 1278 + }, + { + "epoch": 1.0588072887907234, + "grad_norm": 0.8841069340705872, + "learning_rate": 4.671268152974742e-06, + "loss": 0.6389, + "step": 1279 + }, + { + "epoch": 1.0596355604638321, + "grad_norm": 0.8906645178794861, + "learning_rate": 4.670722852805033e-06, + "loss": 0.6448, + "step": 1280 + }, + { + "epoch": 1.0604638321369408, + "grad_norm": 0.9538317918777466, + "learning_rate": 4.670177132620689e-06, + "loss": 0.6544, + "step": 1281 + }, + { + "epoch": 1.0612921038100498, + "grad_norm": 0.9158053994178772, + "learning_rate": 4.669630992527299e-06, + "loss": 0.6494, + "step": 1282 + }, + { + "epoch": 1.0621203754831585, + "grad_norm": 0.9756888151168823, + "learning_rate": 4.6690844326305374e-06, + "loss": 0.6419, + "step": 1283 + }, + { + "epoch": 1.0629486471562672, + "grad_norm": 0.905055820941925, + "learning_rate": 4.668537453036159e-06, + "loss": 0.6463, + "step": 1284 + }, + { + "epoch": 1.063776918829376, + "grad_norm": 0.9011061191558838, + "learning_rate": 4.667990053849997e-06, + "loss": 0.6353, + "step": 1285 + }, + { + "epoch": 1.0646051905024847, + "grad_norm": 0.955047070980072, + "learning_rate": 4.667442235177969e-06, + "loss": 0.6485, + "step": 1286 + }, + { + "epoch": 1.0654334621755936, + "grad_norm": 0.9253432750701904, + "learning_rate": 4.666893997126074e-06, + "loss": 0.6391, + "step": 1287 + }, + { + "epoch": 1.0662617338487024, + "grad_norm": 0.8840277194976807, + "learning_rate": 4.66634533980039e-06, + "loss": 0.6372, + "step": 1288 + }, + { + "epoch": 1.067090005521811, + "grad_norm": 1.0881837606430054, + "learning_rate": 4.6657962633070765e-06, + "loss": 0.6419, + "step": 1289 + }, + { + "epoch": 1.06791827719492, + "grad_norm": 0.9263116717338562, + "learning_rate": 4.665246767752376e-06, + "loss": 0.6226, + "step": 1290 + }, + { + "epoch": 1.0687465488680288, + "grad_norm": 0.9110594987869263, + "learning_rate": 4.66469685324261e-06, + "loss": 0.6495, + "step": 1291 + }, + { + "epoch": 1.0695748205411375, + "grad_norm": 0.9086888432502747, + "learning_rate": 4.664146519884182e-06, + "loss": 0.6408, + "step": 1292 + }, + { + "epoch": 1.0704030922142462, + "grad_norm": 0.8991205096244812, + "learning_rate": 4.663595767783575e-06, + "loss": 0.6424, + "step": 1293 + }, + { + "epoch": 1.071231363887355, + "grad_norm": 0.926933228969574, + "learning_rate": 4.6630445970473576e-06, + "loss": 0.6379, + "step": 1294 + }, + { + "epoch": 1.072059635560464, + "grad_norm": 0.8793080449104309, + "learning_rate": 4.662493007782174e-06, + "loss": 0.6254, + "step": 1295 + }, + { + "epoch": 1.0728879072335726, + "grad_norm": 0.9588155150413513, + "learning_rate": 4.661941000094751e-06, + "loss": 0.6307, + "step": 1296 + }, + { + "epoch": 1.0737161789066814, + "grad_norm": 0.9468860030174255, + "learning_rate": 4.661388574091898e-06, + "loss": 0.6308, + "step": 1297 + }, + { + "epoch": 1.07454445057979, + "grad_norm": 0.8930124640464783, + "learning_rate": 4.660835729880505e-06, + "loss": 0.6371, + "step": 1298 + }, + { + "epoch": 1.075372722252899, + "grad_norm": 0.9233878254890442, + "learning_rate": 4.660282467567541e-06, + "loss": 0.6327, + "step": 1299 + }, + { + "epoch": 1.0762009939260078, + "grad_norm": 0.9714951515197754, + "learning_rate": 4.659728787260057e-06, + "loss": 0.6218, + "step": 1300 + }, + { + "epoch": 1.0770292655991165, + "grad_norm": 0.9464428424835205, + "learning_rate": 4.659174689065185e-06, + "loss": 0.6406, + "step": 1301 + }, + { + "epoch": 1.0778575372722252, + "grad_norm": 0.8984559774398804, + "learning_rate": 4.6586201730901395e-06, + "loss": 0.6483, + "step": 1302 + }, + { + "epoch": 1.0786858089453342, + "grad_norm": 0.9007774591445923, + "learning_rate": 4.658065239442212e-06, + "loss": 0.6288, + "step": 1303 + }, + { + "epoch": 1.079514080618443, + "grad_norm": 0.907171368598938, + "learning_rate": 4.657509888228779e-06, + "loss": 0.6445, + "step": 1304 + }, + { + "epoch": 1.0803423522915516, + "grad_norm": 0.9207282066345215, + "learning_rate": 4.656954119557293e-06, + "loss": 0.6148, + "step": 1305 + }, + { + "epoch": 1.0811706239646603, + "grad_norm": 0.9467616677284241, + "learning_rate": 4.656397933535293e-06, + "loss": 0.6374, + "step": 1306 + }, + { + "epoch": 1.0819988956377693, + "grad_norm": 0.9310930967330933, + "learning_rate": 4.655841330270393e-06, + "loss": 0.6362, + "step": 1307 + }, + { + "epoch": 1.082827167310878, + "grad_norm": 0.890816330909729, + "learning_rate": 4.655284309870294e-06, + "loss": 0.6372, + "step": 1308 + }, + { + "epoch": 1.0836554389839868, + "grad_norm": 0.9206095933914185, + "learning_rate": 4.654726872442771e-06, + "loss": 0.6297, + "step": 1309 + }, + { + "epoch": 1.0844837106570955, + "grad_norm": 0.9015856385231018, + "learning_rate": 4.654169018095685e-06, + "loss": 0.658, + "step": 1310 + }, + { + "epoch": 1.0853119823302042, + "grad_norm": 0.8909441828727722, + "learning_rate": 4.653610746936975e-06, + "loss": 0.6382, + "step": 1311 + }, + { + "epoch": 1.0861402540033132, + "grad_norm": 0.9035969972610474, + "learning_rate": 4.653052059074661e-06, + "loss": 0.6441, + "step": 1312 + }, + { + "epoch": 1.0869685256764219, + "grad_norm": 0.8988869786262512, + "learning_rate": 4.652492954616844e-06, + "loss": 0.6359, + "step": 1313 + }, + { + "epoch": 1.0877967973495306, + "grad_norm": 0.881885826587677, + "learning_rate": 4.6519334336717055e-06, + "loss": 0.6397, + "step": 1314 + }, + { + "epoch": 1.0886250690226393, + "grad_norm": 0.980633020401001, + "learning_rate": 4.651373496347508e-06, + "loss": 0.6223, + "step": 1315 + }, + { + "epoch": 1.0894533406957483, + "grad_norm": 0.9027420878410339, + "learning_rate": 4.650813142752593e-06, + "loss": 0.6211, + "step": 1316 + }, + { + "epoch": 1.090281612368857, + "grad_norm": 0.9011778831481934, + "learning_rate": 4.6502523729953855e-06, + "loss": 0.6442, + "step": 1317 + }, + { + "epoch": 1.0911098840419657, + "grad_norm": 0.926692545413971, + "learning_rate": 4.649691187184387e-06, + "loss": 0.6661, + "step": 1318 + }, + { + "epoch": 1.0919381557150745, + "grad_norm": 0.9758741855621338, + "learning_rate": 4.649129585428184e-06, + "loss": 0.6461, + "step": 1319 + }, + { + "epoch": 1.0927664273881834, + "grad_norm": 0.967698335647583, + "learning_rate": 4.64856756783544e-06, + "loss": 0.6461, + "step": 1320 + }, + { + "epoch": 1.0935946990612921, + "grad_norm": 0.9372200965881348, + "learning_rate": 4.6480051345149e-06, + "loss": 0.6348, + "step": 1321 + }, + { + "epoch": 1.0944229707344009, + "grad_norm": 0.8976414799690247, + "learning_rate": 4.64744228557539e-06, + "loss": 0.6297, + "step": 1322 + }, + { + "epoch": 1.0952512424075096, + "grad_norm": 0.9163550734519958, + "learning_rate": 4.646879021125816e-06, + "loss": 0.6746, + "step": 1323 + }, + { + "epoch": 1.0960795140806185, + "grad_norm": 0.9064576625823975, + "learning_rate": 4.646315341275164e-06, + "loss": 0.6242, + "step": 1324 + }, + { + "epoch": 1.0969077857537273, + "grad_norm": 0.9352027773857117, + "learning_rate": 4.645751246132501e-06, + "loss": 0.6568, + "step": 1325 + }, + { + "epoch": 1.097736057426836, + "grad_norm": 0.8680023550987244, + "learning_rate": 4.645186735806976e-06, + "loss": 0.6484, + "step": 1326 + }, + { + "epoch": 1.0985643290999447, + "grad_norm": 0.8782095909118652, + "learning_rate": 4.644621810407813e-06, + "loss": 0.646, + "step": 1327 + }, + { + "epoch": 1.0993926007730535, + "grad_norm": 0.8931819796562195, + "learning_rate": 4.644056470044323e-06, + "loss": 0.6708, + "step": 1328 + }, + { + "epoch": 1.1002208724461624, + "grad_norm": 0.9228624105453491, + "learning_rate": 4.6434907148258915e-06, + "loss": 0.6255, + "step": 1329 + }, + { + "epoch": 1.1010491441192711, + "grad_norm": 0.8960729837417603, + "learning_rate": 4.642924544861988e-06, + "loss": 0.6523, + "step": 1330 + }, + { + "epoch": 1.1018774157923799, + "grad_norm": 0.9092081785202026, + "learning_rate": 4.642357960262162e-06, + "loss": 0.6534, + "step": 1331 + }, + { + "epoch": 1.1027056874654886, + "grad_norm": 0.9024838209152222, + "learning_rate": 4.641790961136041e-06, + "loss": 0.6238, + "step": 1332 + }, + { + "epoch": 1.1035339591385975, + "grad_norm": 0.9016388058662415, + "learning_rate": 4.641223547593334e-06, + "loss": 0.6643, + "step": 1333 + }, + { + "epoch": 1.1043622308117063, + "grad_norm": 0.8877143859863281, + "learning_rate": 4.6406557197438324e-06, + "loss": 0.6222, + "step": 1334 + }, + { + "epoch": 1.105190502484815, + "grad_norm": 0.9066653251647949, + "learning_rate": 4.640087477697403e-06, + "loss": 0.6205, + "step": 1335 + }, + { + "epoch": 1.1060187741579237, + "grad_norm": 0.903465747833252, + "learning_rate": 4.639518821563997e-06, + "loss": 0.6299, + "step": 1336 + }, + { + "epoch": 1.1068470458310327, + "grad_norm": 0.8908408880233765, + "learning_rate": 4.6389497514536415e-06, + "loss": 0.6247, + "step": 1337 + }, + { + "epoch": 1.1076753175041414, + "grad_norm": 0.8957250714302063, + "learning_rate": 4.63838026747645e-06, + "loss": 0.632, + "step": 1338 + }, + { + "epoch": 1.1085035891772501, + "grad_norm": 0.938494086265564, + "learning_rate": 4.637810369742609e-06, + "loss": 0.6309, + "step": 1339 + }, + { + "epoch": 1.1093318608503588, + "grad_norm": 0.9913409948348999, + "learning_rate": 4.637240058362391e-06, + "loss": 0.6571, + "step": 1340 + }, + { + "epoch": 1.1101601325234678, + "grad_norm": 0.9037449359893799, + "learning_rate": 4.636669333446145e-06, + "loss": 0.6467, + "step": 1341 + }, + { + "epoch": 1.1109884041965765, + "grad_norm": 0.911696195602417, + "learning_rate": 4.636098195104299e-06, + "loss": 0.6361, + "step": 1342 + }, + { + "epoch": 1.1118166758696852, + "grad_norm": 0.9105570316314697, + "learning_rate": 4.6355266434473665e-06, + "loss": 0.6418, + "step": 1343 + }, + { + "epoch": 1.112644947542794, + "grad_norm": 0.9101701378822327, + "learning_rate": 4.634954678585935e-06, + "loss": 0.6454, + "step": 1344 + }, + { + "epoch": 1.1134732192159027, + "grad_norm": 0.8980433344841003, + "learning_rate": 4.634382300630675e-06, + "loss": 0.6159, + "step": 1345 + }, + { + "epoch": 1.1143014908890116, + "grad_norm": 0.9253267049789429, + "learning_rate": 4.633809509692336e-06, + "loss": 0.6465, + "step": 1346 + }, + { + "epoch": 1.1151297625621204, + "grad_norm": 0.9150062203407288, + "learning_rate": 4.6332363058817484e-06, + "loss": 0.639, + "step": 1347 + }, + { + "epoch": 1.115958034235229, + "grad_norm": 0.9257630705833435, + "learning_rate": 4.632662689309821e-06, + "loss": 0.6571, + "step": 1348 + }, + { + "epoch": 1.116786305908338, + "grad_norm": 0.9183084964752197, + "learning_rate": 4.632088660087545e-06, + "loss": 0.6291, + "step": 1349 + }, + { + "epoch": 1.1176145775814468, + "grad_norm": 0.9229403138160706, + "learning_rate": 4.631514218325987e-06, + "loss": 0.6487, + "step": 1350 + }, + { + "epoch": 1.1184428492545555, + "grad_norm": 0.9379578828811646, + "learning_rate": 4.630939364136298e-06, + "loss": 0.6564, + "step": 1351 + }, + { + "epoch": 1.1192711209276642, + "grad_norm": 0.9094021320343018, + "learning_rate": 4.630364097629706e-06, + "loss": 0.6354, + "step": 1352 + }, + { + "epoch": 1.120099392600773, + "grad_norm": 0.9528741836547852, + "learning_rate": 4.62978841891752e-06, + "loss": 0.6259, + "step": 1353 + }, + { + "epoch": 1.120927664273882, + "grad_norm": 0.8859900832176208, + "learning_rate": 4.629212328111129e-06, + "loss": 0.6783, + "step": 1354 + }, + { + "epoch": 1.1217559359469906, + "grad_norm": 0.8739688396453857, + "learning_rate": 4.628635825322e-06, + "loss": 0.6518, + "step": 1355 + }, + { + "epoch": 1.1225842076200994, + "grad_norm": 0.8773418068885803, + "learning_rate": 4.628058910661683e-06, + "loss": 0.6477, + "step": 1356 + }, + { + "epoch": 1.123412479293208, + "grad_norm": 0.9025301933288574, + "learning_rate": 4.627481584241803e-06, + "loss": 0.6471, + "step": 1357 + }, + { + "epoch": 1.124240750966317, + "grad_norm": 0.89899080991745, + "learning_rate": 4.62690384617407e-06, + "loss": 0.6196, + "step": 1358 + }, + { + "epoch": 1.1250690226394258, + "grad_norm": 0.9514591693878174, + "learning_rate": 4.6263256965702684e-06, + "loss": 0.6495, + "step": 1359 + }, + { + "epoch": 1.1258972943125345, + "grad_norm": 0.908789336681366, + "learning_rate": 4.6257471355422654e-06, + "loss": 0.6606, + "step": 1360 + }, + { + "epoch": 1.1267255659856432, + "grad_norm": 0.9262552857398987, + "learning_rate": 4.625168163202009e-06, + "loss": 0.6308, + "step": 1361 + }, + { + "epoch": 1.1275538376587522, + "grad_norm": 0.9395532608032227, + "learning_rate": 4.624588779661523e-06, + "loss": 0.6196, + "step": 1362 + }, + { + "epoch": 1.128382109331861, + "grad_norm": 0.9126531481742859, + "learning_rate": 4.624008985032913e-06, + "loss": 0.6474, + "step": 1363 + }, + { + "epoch": 1.1292103810049696, + "grad_norm": 0.9105198383331299, + "learning_rate": 4.623428779428365e-06, + "loss": 0.6455, + "step": 1364 + }, + { + "epoch": 1.1300386526780783, + "grad_norm": 0.9048190712928772, + "learning_rate": 4.6228481629601415e-06, + "loss": 0.6647, + "step": 1365 + }, + { + "epoch": 1.1308669243511873, + "grad_norm": 0.9415149092674255, + "learning_rate": 4.622267135740588e-06, + "loss": 0.6403, + "step": 1366 + }, + { + "epoch": 1.131695196024296, + "grad_norm": 0.954728901386261, + "learning_rate": 4.6216856978821275e-06, + "loss": 0.6351, + "step": 1367 + }, + { + "epoch": 1.1325234676974048, + "grad_norm": 0.9392480850219727, + "learning_rate": 4.621103849497261e-06, + "loss": 0.6481, + "step": 1368 + }, + { + "epoch": 1.1333517393705135, + "grad_norm": 0.9199283719062805, + "learning_rate": 4.620521590698574e-06, + "loss": 0.6578, + "step": 1369 + }, + { + "epoch": 1.1341800110436222, + "grad_norm": 0.9175401926040649, + "learning_rate": 4.6199389215987266e-06, + "loss": 0.6508, + "step": 1370 + }, + { + "epoch": 1.1350082827167312, + "grad_norm": 0.9140470027923584, + "learning_rate": 4.619355842310459e-06, + "loss": 0.6536, + "step": 1371 + }, + { + "epoch": 1.1358365543898399, + "grad_norm": 0.8863652348518372, + "learning_rate": 4.618772352946593e-06, + "loss": 0.6376, + "step": 1372 + }, + { + "epoch": 1.1366648260629486, + "grad_norm": 0.9074656963348389, + "learning_rate": 4.618188453620026e-06, + "loss": 0.6253, + "step": 1373 + }, + { + "epoch": 1.1374930977360573, + "grad_norm": 0.891055703163147, + "learning_rate": 4.61760414444374e-06, + "loss": 0.6507, + "step": 1374 + }, + { + "epoch": 1.1383213694091663, + "grad_norm": 0.9153634905815125, + "learning_rate": 4.617019425530791e-06, + "loss": 0.6699, + "step": 1375 + }, + { + "epoch": 1.139149641082275, + "grad_norm": 0.9172940850257874, + "learning_rate": 4.6164342969943196e-06, + "loss": 0.6385, + "step": 1376 + }, + { + "epoch": 1.1399779127553837, + "grad_norm": 0.8978968262672424, + "learning_rate": 4.615848758947539e-06, + "loss": 0.6394, + "step": 1377 + }, + { + "epoch": 1.1408061844284925, + "grad_norm": 0.9372277855873108, + "learning_rate": 4.615262811503749e-06, + "loss": 0.6418, + "step": 1378 + }, + { + "epoch": 1.1416344561016014, + "grad_norm": 0.9596112370491028, + "learning_rate": 4.614676454776321e-06, + "loss": 0.6573, + "step": 1379 + }, + { + "epoch": 1.1424627277747101, + "grad_norm": 0.9243675470352173, + "learning_rate": 4.614089688878713e-06, + "loss": 0.6546, + "step": 1380 + }, + { + "epoch": 1.1432909994478189, + "grad_norm": 0.8839207291603088, + "learning_rate": 4.613502513924459e-06, + "loss": 0.654, + "step": 1381 + }, + { + "epoch": 1.1441192711209276, + "grad_norm": 0.8940262198448181, + "learning_rate": 4.612914930027168e-06, + "loss": 0.6223, + "step": 1382 + }, + { + "epoch": 1.1449475427940365, + "grad_norm": 0.9091162085533142, + "learning_rate": 4.612326937300535e-06, + "loss": 0.6512, + "step": 1383 + }, + { + "epoch": 1.1457758144671453, + "grad_norm": 0.9332976937294006, + "learning_rate": 4.6117385358583315e-06, + "loss": 0.6373, + "step": 1384 + }, + { + "epoch": 1.146604086140254, + "grad_norm": 0.9103875756263733, + "learning_rate": 4.611149725814406e-06, + "loss": 0.6502, + "step": 1385 + }, + { + "epoch": 1.1474323578133627, + "grad_norm": 0.9004862308502197, + "learning_rate": 4.610560507282688e-06, + "loss": 0.6361, + "step": 1386 + }, + { + "epoch": 1.1482606294864715, + "grad_norm": 0.880293071269989, + "learning_rate": 4.609970880377187e-06, + "loss": 0.655, + "step": 1387 + }, + { + "epoch": 1.1490889011595804, + "grad_norm": 0.9099947214126587, + "learning_rate": 4.60938084521199e-06, + "loss": 0.6517, + "step": 1388 + }, + { + "epoch": 1.1499171728326891, + "grad_norm": 0.8913806080818176, + "learning_rate": 4.608790401901262e-06, + "loss": 0.6429, + "step": 1389 + }, + { + "epoch": 1.1507454445057979, + "grad_norm": 0.9198290109634399, + "learning_rate": 4.608199550559249e-06, + "loss": 0.6537, + "step": 1390 + }, + { + "epoch": 1.1515737161789068, + "grad_norm": 0.9192278981208801, + "learning_rate": 4.607608291300275e-06, + "loss": 0.6531, + "step": 1391 + }, + { + "epoch": 1.1524019878520155, + "grad_norm": 0.8977665305137634, + "learning_rate": 4.607016624238744e-06, + "loss": 0.6358, + "step": 1392 + }, + { + "epoch": 1.1532302595251243, + "grad_norm": 0.9576854705810547, + "learning_rate": 4.606424549489138e-06, + "loss": 0.6493, + "step": 1393 + }, + { + "epoch": 1.154058531198233, + "grad_norm": 0.905611515045166, + "learning_rate": 4.605832067166017e-06, + "loss": 0.63, + "step": 1394 + }, + { + "epoch": 1.1548868028713417, + "grad_norm": 0.8840015530586243, + "learning_rate": 4.605239177384021e-06, + "loss": 0.6504, + "step": 1395 + }, + { + "epoch": 1.1557150745444507, + "grad_norm": 0.9247309565544128, + "learning_rate": 4.604645880257869e-06, + "loss": 0.6477, + "step": 1396 + }, + { + "epoch": 1.1565433462175594, + "grad_norm": 0.9473649263381958, + "learning_rate": 4.604052175902359e-06, + "loss": 0.6374, + "step": 1397 + }, + { + "epoch": 1.1573716178906681, + "grad_norm": 0.9354954361915588, + "learning_rate": 4.603458064432367e-06, + "loss": 0.6491, + "step": 1398 + }, + { + "epoch": 1.1581998895637768, + "grad_norm": 0.8788221478462219, + "learning_rate": 4.6028635459628476e-06, + "loss": 0.6187, + "step": 1399 + }, + { + "epoch": 1.1590281612368858, + "grad_norm": 0.8836075067520142, + "learning_rate": 4.602268620608834e-06, + "loss": 0.665, + "step": 1400 + }, + { + "epoch": 1.1598564329099945, + "grad_norm": 0.9195554256439209, + "learning_rate": 4.60167328848544e-06, + "loss": 0.6291, + "step": 1401 + }, + { + "epoch": 1.1606847045831032, + "grad_norm": 0.8984520435333252, + "learning_rate": 4.601077549707856e-06, + "loss": 0.6361, + "step": 1402 + }, + { + "epoch": 1.161512976256212, + "grad_norm": 0.8667109608650208, + "learning_rate": 4.600481404391352e-06, + "loss": 0.6345, + "step": 1403 + }, + { + "epoch": 1.1623412479293207, + "grad_norm": 0.9312148094177246, + "learning_rate": 4.599884852651277e-06, + "loss": 0.6388, + "step": 1404 + }, + { + "epoch": 1.1631695196024296, + "grad_norm": 0.9096866846084595, + "learning_rate": 4.5992878946030575e-06, + "loss": 0.6383, + "step": 1405 + }, + { + "epoch": 1.1639977912755384, + "grad_norm": 0.9010055661201477, + "learning_rate": 4.5986905303622e-06, + "loss": 0.6524, + "step": 1406 + }, + { + "epoch": 1.164826062948647, + "grad_norm": 0.9414717555046082, + "learning_rate": 4.598092760044288e-06, + "loss": 0.6825, + "step": 1407 + }, + { + "epoch": 1.165654334621756, + "grad_norm": 0.9162404537200928, + "learning_rate": 4.597494583764987e-06, + "loss": 0.6467, + "step": 1408 + }, + { + "epoch": 1.1664826062948648, + "grad_norm": 0.8935617208480835, + "learning_rate": 4.596896001640034e-06, + "loss": 0.6375, + "step": 1409 + }, + { + "epoch": 1.1673108779679735, + "grad_norm": 0.8976510167121887, + "learning_rate": 4.596297013785253e-06, + "loss": 0.6135, + "step": 1410 + }, + { + "epoch": 1.1681391496410822, + "grad_norm": 0.9503909349441528, + "learning_rate": 4.5956976203165414e-06, + "loss": 0.6305, + "step": 1411 + }, + { + "epoch": 1.168967421314191, + "grad_norm": 0.9147273302078247, + "learning_rate": 4.595097821349876e-06, + "loss": 0.6548, + "step": 1412 + }, + { + "epoch": 1.1697956929873, + "grad_norm": 0.9171034693717957, + "learning_rate": 4.594497617001312e-06, + "loss": 0.6253, + "step": 1413 + }, + { + "epoch": 1.1706239646604086, + "grad_norm": 0.8970120549201965, + "learning_rate": 4.593897007386985e-06, + "loss": 0.6414, + "step": 1414 + }, + { + "epoch": 1.1714522363335174, + "grad_norm": 0.9032952189445496, + "learning_rate": 4.593295992623105e-06, + "loss": 0.6595, + "step": 1415 + }, + { + "epoch": 1.172280508006626, + "grad_norm": 0.9066448211669922, + "learning_rate": 4.592694572825964e-06, + "loss": 0.6654, + "step": 1416 + }, + { + "epoch": 1.173108779679735, + "grad_norm": 0.8747309446334839, + "learning_rate": 4.592092748111931e-06, + "loss": 0.6619, + "step": 1417 + }, + { + "epoch": 1.1739370513528438, + "grad_norm": 0.8836600184440613, + "learning_rate": 4.5914905185974535e-06, + "loss": 0.6443, + "step": 1418 + }, + { + "epoch": 1.1747653230259525, + "grad_norm": 0.9340765476226807, + "learning_rate": 4.590887884399058e-06, + "loss": 0.6633, + "step": 1419 + }, + { + "epoch": 1.1755935946990612, + "grad_norm": 0.8889044523239136, + "learning_rate": 4.590284845633347e-06, + "loss": 0.6489, + "step": 1420 + }, + { + "epoch": 1.1764218663721702, + "grad_norm": 0.9043524265289307, + "learning_rate": 4.589681402417005e-06, + "loss": 0.6392, + "step": 1421 + }, + { + "epoch": 1.177250138045279, + "grad_norm": 0.9210046529769897, + "learning_rate": 4.5890775548667895e-06, + "loss": 0.6105, + "step": 1422 + }, + { + "epoch": 1.1780784097183876, + "grad_norm": 0.8954651951789856, + "learning_rate": 4.588473303099542e-06, + "loss": 0.6425, + "step": 1423 + }, + { + "epoch": 1.1789066813914963, + "grad_norm": 0.945517897605896, + "learning_rate": 4.587868647232179e-06, + "loss": 0.6404, + "step": 1424 + }, + { + "epoch": 1.1797349530646053, + "grad_norm": 0.9059463739395142, + "learning_rate": 4.587263587381696e-06, + "loss": 0.6521, + "step": 1425 + }, + { + "epoch": 1.180563224737714, + "grad_norm": 0.8876296281814575, + "learning_rate": 4.586658123665165e-06, + "loss": 0.6447, + "step": 1426 + }, + { + "epoch": 1.1813914964108228, + "grad_norm": 0.8862487077713013, + "learning_rate": 4.58605225619974e-06, + "loss": 0.6445, + "step": 1427 + }, + { + "epoch": 1.1822197680839315, + "grad_norm": 0.9066351056098938, + "learning_rate": 4.585445985102649e-06, + "loss": 0.6563, + "step": 1428 + }, + { + "epoch": 1.1830480397570402, + "grad_norm": 0.913724958896637, + "learning_rate": 4.584839310491199e-06, + "loss": 0.6507, + "step": 1429 + }, + { + "epoch": 1.1838763114301492, + "grad_norm": 0.8942010402679443, + "learning_rate": 4.584232232482778e-06, + "loss": 0.6424, + "step": 1430 + }, + { + "epoch": 1.1847045831032579, + "grad_norm": 0.8986000418663025, + "learning_rate": 4.58362475119485e-06, + "loss": 0.6491, + "step": 1431 + }, + { + "epoch": 1.1855328547763666, + "grad_norm": 0.8879220485687256, + "learning_rate": 4.5830168667449544e-06, + "loss": 0.6412, + "step": 1432 + }, + { + "epoch": 1.1863611264494753, + "grad_norm": 0.9497302770614624, + "learning_rate": 4.5824085792507135e-06, + "loss": 0.6453, + "step": 1433 + }, + { + "epoch": 1.1871893981225843, + "grad_norm": 0.8933898210525513, + "learning_rate": 4.581799888829824e-06, + "loss": 0.6381, + "step": 1434 + }, + { + "epoch": 1.188017669795693, + "grad_norm": 0.9034610986709595, + "learning_rate": 4.581190795600062e-06, + "loss": 0.6551, + "step": 1435 + }, + { + "epoch": 1.1888459414688017, + "grad_norm": 0.8968634605407715, + "learning_rate": 4.580581299679282e-06, + "loss": 0.637, + "step": 1436 + }, + { + "epoch": 1.1896742131419105, + "grad_norm": 0.8974384069442749, + "learning_rate": 4.579971401185416e-06, + "loss": 0.6301, + "step": 1437 + }, + { + "epoch": 1.1905024848150194, + "grad_norm": 0.8726125359535217, + "learning_rate": 4.579361100236473e-06, + "loss": 0.6487, + "step": 1438 + }, + { + "epoch": 1.1913307564881281, + "grad_norm": 0.9131139516830444, + "learning_rate": 4.5787503969505405e-06, + "loss": 0.6211, + "step": 1439 + }, + { + "epoch": 1.1921590281612369, + "grad_norm": 0.8924953937530518, + "learning_rate": 4.578139291445783e-06, + "loss": 0.6339, + "step": 1440 + }, + { + "epoch": 1.1929872998343456, + "grad_norm": 0.890883207321167, + "learning_rate": 4.577527783840447e-06, + "loss": 0.66, + "step": 1441 + }, + { + "epoch": 1.1938155715074545, + "grad_norm": 0.9326318502426147, + "learning_rate": 4.5769158742528494e-06, + "loss": 0.6611, + "step": 1442 + }, + { + "epoch": 1.1946438431805633, + "grad_norm": 0.9354713559150696, + "learning_rate": 4.576303562801392e-06, + "loss": 0.6445, + "step": 1443 + }, + { + "epoch": 1.195472114853672, + "grad_norm": 0.8868383169174194, + "learning_rate": 4.57569084960455e-06, + "loss": 0.6344, + "step": 1444 + }, + { + "epoch": 1.1963003865267807, + "grad_norm": 0.893924355506897, + "learning_rate": 4.5750777347808775e-06, + "loss": 0.602, + "step": 1445 + }, + { + "epoch": 1.1971286581998895, + "grad_norm": 0.9299339652061462, + "learning_rate": 4.574464218449006e-06, + "loss": 0.6206, + "step": 1446 + }, + { + "epoch": 1.1979569298729984, + "grad_norm": 0.9058142304420471, + "learning_rate": 4.573850300727648e-06, + "loss": 0.6073, + "step": 1447 + }, + { + "epoch": 1.1987852015461071, + "grad_norm": 0.8893733620643616, + "learning_rate": 4.5732359817355885e-06, + "loss": 0.6336, + "step": 1448 + }, + { + "epoch": 1.1996134732192159, + "grad_norm": 0.9018000960350037, + "learning_rate": 4.572621261591692e-06, + "loss": 0.6268, + "step": 1449 + }, + { + "epoch": 1.2004417448923248, + "grad_norm": 0.9087574481964111, + "learning_rate": 4.572006140414903e-06, + "loss": 0.6511, + "step": 1450 + }, + { + "epoch": 1.2012700165654335, + "grad_norm": 0.9092984795570374, + "learning_rate": 4.5713906183242404e-06, + "loss": 0.6424, + "step": 1451 + }, + { + "epoch": 1.2020982882385423, + "grad_norm": 0.9040535092353821, + "learning_rate": 4.5707746954388034e-06, + "loss": 0.6242, + "step": 1452 + }, + { + "epoch": 1.202926559911651, + "grad_norm": 0.891659677028656, + "learning_rate": 4.570158371877766e-06, + "loss": 0.643, + "step": 1453 + }, + { + "epoch": 1.2037548315847597, + "grad_norm": 0.9254544973373413, + "learning_rate": 4.569541647760381e-06, + "loss": 0.6191, + "step": 1454 + }, + { + "epoch": 1.2045831032578687, + "grad_norm": 0.8978536128997803, + "learning_rate": 4.568924523205979e-06, + "loss": 0.6632, + "step": 1455 + }, + { + "epoch": 1.2054113749309774, + "grad_norm": 0.9179201126098633, + "learning_rate": 4.568306998333968e-06, + "loss": 0.6485, + "step": 1456 + }, + { + "epoch": 1.2062396466040861, + "grad_norm": 0.9061882495880127, + "learning_rate": 4.567689073263834e-06, + "loss": 0.6704, + "step": 1457 + }, + { + "epoch": 1.2070679182771948, + "grad_norm": 0.9143427610397339, + "learning_rate": 4.567070748115139e-06, + "loss": 0.6453, + "step": 1458 + }, + { + "epoch": 1.2078961899503038, + "grad_norm": 0.8740119338035583, + "learning_rate": 4.566452023007523e-06, + "loss": 0.6453, + "step": 1459 + }, + { + "epoch": 1.2087244616234125, + "grad_norm": 0.8977463841438293, + "learning_rate": 4.565832898060703e-06, + "loss": 0.6451, + "step": 1460 + }, + { + "epoch": 1.2095527332965212, + "grad_norm": 0.9416611194610596, + "learning_rate": 4.565213373394476e-06, + "loss": 0.6265, + "step": 1461 + }, + { + "epoch": 1.21038100496963, + "grad_norm": 0.9132000207901001, + "learning_rate": 4.5645934491287115e-06, + "loss": 0.6412, + "step": 1462 + }, + { + "epoch": 1.2112092766427387, + "grad_norm": 0.9480498433113098, + "learning_rate": 4.563973125383362e-06, + "loss": 0.6363, + "step": 1463 + }, + { + "epoch": 1.2120375483158476, + "grad_norm": 0.9170709848403931, + "learning_rate": 4.563352402278451e-06, + "loss": 0.6201, + "step": 1464 + }, + { + "epoch": 1.2128658199889564, + "grad_norm": 0.9655918478965759, + "learning_rate": 4.562731279934086e-06, + "loss": 0.6497, + "step": 1465 + }, + { + "epoch": 1.213694091662065, + "grad_norm": 0.8885562419891357, + "learning_rate": 4.562109758470446e-06, + "loss": 0.6265, + "step": 1466 + }, + { + "epoch": 1.214522363335174, + "grad_norm": 0.9072791337966919, + "learning_rate": 4.561487838007792e-06, + "loss": 0.6257, + "step": 1467 + }, + { + "epoch": 1.2153506350082828, + "grad_norm": 0.8992362022399902, + "learning_rate": 4.560865518666458e-06, + "loss": 0.6425, + "step": 1468 + }, + { + "epoch": 1.2161789066813915, + "grad_norm": 0.8934991955757141, + "learning_rate": 4.560242800566855e-06, + "loss": 0.6583, + "step": 1469 + }, + { + "epoch": 1.2170071783545002, + "grad_norm": 0.9292325973510742, + "learning_rate": 4.559619683829477e-06, + "loss": 0.6403, + "step": 1470 + }, + { + "epoch": 1.217835450027609, + "grad_norm": 0.8756814002990723, + "learning_rate": 4.55899616857489e-06, + "loss": 0.6261, + "step": 1471 + }, + { + "epoch": 1.218663721700718, + "grad_norm": 0.9050565361976624, + "learning_rate": 4.558372254923738e-06, + "loss": 0.6325, + "step": 1472 + }, + { + "epoch": 1.2194919933738266, + "grad_norm": 0.9230931401252747, + "learning_rate": 4.557747942996742e-06, + "loss": 0.6275, + "step": 1473 + }, + { + "epoch": 1.2203202650469354, + "grad_norm": 0.894026517868042, + "learning_rate": 4.557123232914702e-06, + "loss": 0.6348, + "step": 1474 + }, + { + "epoch": 1.221148536720044, + "grad_norm": 0.8882637023925781, + "learning_rate": 4.556498124798492e-06, + "loss": 0.6123, + "step": 1475 + }, + { + "epoch": 1.221976808393153, + "grad_norm": 0.9191063642501831, + "learning_rate": 4.555872618769065e-06, + "loss": 0.6531, + "step": 1476 + }, + { + "epoch": 1.2228050800662618, + "grad_norm": 0.9398623704910278, + "learning_rate": 4.555246714947452e-06, + "loss": 0.6553, + "step": 1477 + }, + { + "epoch": 1.2236333517393705, + "grad_norm": 0.9399452805519104, + "learning_rate": 4.554620413454757e-06, + "loss": 0.6485, + "step": 1478 + }, + { + "epoch": 1.2244616234124792, + "grad_norm": 0.9001572728157043, + "learning_rate": 4.553993714412166e-06, + "loss": 0.6467, + "step": 1479 + }, + { + "epoch": 1.2252898950855882, + "grad_norm": 0.9266572594642639, + "learning_rate": 4.553366617940938e-06, + "loss": 0.6325, + "step": 1480 + }, + { + "epoch": 1.226118166758697, + "grad_norm": 0.928301215171814, + "learning_rate": 4.552739124162411e-06, + "loss": 0.6398, + "step": 1481 + }, + { + "epoch": 1.2269464384318056, + "grad_norm": 0.880954921245575, + "learning_rate": 4.552111233197999e-06, + "loss": 0.669, + "step": 1482 + }, + { + "epoch": 1.2277747101049143, + "grad_norm": 0.9089685082435608, + "learning_rate": 4.551482945169191e-06, + "loss": 0.62, + "step": 1483 + }, + { + "epoch": 1.2286029817780233, + "grad_norm": 0.9187189936637878, + "learning_rate": 4.550854260197559e-06, + "loss": 0.6381, + "step": 1484 + }, + { + "epoch": 1.229431253451132, + "grad_norm": 0.8782045245170593, + "learning_rate": 4.550225178404744e-06, + "loss": 0.6397, + "step": 1485 + }, + { + "epoch": 1.2302595251242407, + "grad_norm": 0.8798948526382446, + "learning_rate": 4.54959569991247e-06, + "loss": 0.6522, + "step": 1486 + }, + { + "epoch": 1.2310877967973495, + "grad_norm": 0.9257029294967651, + "learning_rate": 4.548965824842534e-06, + "loss": 0.6522, + "step": 1487 + }, + { + "epoch": 1.2319160684704582, + "grad_norm": 0.8923602104187012, + "learning_rate": 4.548335553316811e-06, + "loss": 0.6538, + "step": 1488 + }, + { + "epoch": 1.2327443401435672, + "grad_norm": 0.9071294069290161, + "learning_rate": 4.5477048854572524e-06, + "loss": 0.6305, + "step": 1489 + }, + { + "epoch": 1.2335726118166759, + "grad_norm": 0.9495804905891418, + "learning_rate": 4.547073821385888e-06, + "loss": 0.6398, + "step": 1490 + }, + { + "epoch": 1.2344008834897846, + "grad_norm": 0.9163299202919006, + "learning_rate": 4.54644236122482e-06, + "loss": 0.6454, + "step": 1491 + }, + { + "epoch": 1.2352291551628933, + "grad_norm": 0.9286475777626038, + "learning_rate": 4.545810505096233e-06, + "loss": 0.6345, + "step": 1492 + }, + { + "epoch": 1.2360574268360023, + "grad_norm": 0.9116879105567932, + "learning_rate": 4.545178253122384e-06, + "loss": 0.6451, + "step": 1493 + }, + { + "epoch": 1.236885698509111, + "grad_norm": 0.917826771736145, + "learning_rate": 4.544545605425607e-06, + "loss": 0.6374, + "step": 1494 + }, + { + "epoch": 1.2377139701822197, + "grad_norm": 0.9433108568191528, + "learning_rate": 4.543912562128316e-06, + "loss": 0.6321, + "step": 1495 + }, + { + "epoch": 1.2385422418553285, + "grad_norm": 0.9017132520675659, + "learning_rate": 4.543279123352996e-06, + "loss": 0.6341, + "step": 1496 + }, + { + "epoch": 1.2393705135284374, + "grad_norm": 0.9557158350944519, + "learning_rate": 4.542645289222214e-06, + "loss": 0.6364, + "step": 1497 + }, + { + "epoch": 1.2401987852015461, + "grad_norm": 0.9258891344070435, + "learning_rate": 4.542011059858609e-06, + "loss": 0.6598, + "step": 1498 + }, + { + "epoch": 1.2410270568746549, + "grad_norm": 0.8988100290298462, + "learning_rate": 4.541376435384899e-06, + "loss": 0.6496, + "step": 1499 + }, + { + "epoch": 1.2418553285477636, + "grad_norm": 0.898154079914093, + "learning_rate": 4.540741415923878e-06, + "loss": 0.6486, + "step": 1500 + }, + { + "epoch": 1.2426836002208725, + "grad_norm": 0.9216037392616272, + "learning_rate": 4.540106001598418e-06, + "loss": 0.6436, + "step": 1501 + }, + { + "epoch": 1.2435118718939813, + "grad_norm": 0.9086095690727234, + "learning_rate": 4.539470192531463e-06, + "loss": 0.6287, + "step": 1502 + }, + { + "epoch": 1.24434014356709, + "grad_norm": 0.921291172504425, + "learning_rate": 4.538833988846039e-06, + "loss": 0.629, + "step": 1503 + }, + { + "epoch": 1.2451684152401987, + "grad_norm": 0.8621019721031189, + "learning_rate": 4.538197390665242e-06, + "loss": 0.6543, + "step": 1504 + }, + { + "epoch": 1.2459966869133074, + "grad_norm": 0.9523105025291443, + "learning_rate": 4.537560398112251e-06, + "loss": 0.6482, + "step": 1505 + }, + { + "epoch": 1.2468249585864164, + "grad_norm": 0.971931517124176, + "learning_rate": 4.536923011310317e-06, + "loss": 0.6463, + "step": 1506 + }, + { + "epoch": 1.2476532302595251, + "grad_norm": 0.927219033241272, + "learning_rate": 4.536285230382767e-06, + "loss": 0.6443, + "step": 1507 + }, + { + "epoch": 1.2484815019326339, + "grad_norm": 0.9433056116104126, + "learning_rate": 4.535647055453007e-06, + "loss": 0.6301, + "step": 1508 + }, + { + "epoch": 1.2493097736057428, + "grad_norm": 0.8992047905921936, + "learning_rate": 4.5350084866445195e-06, + "loss": 0.6177, + "step": 1509 + }, + { + "epoch": 1.2501380452788515, + "grad_norm": 0.8946895003318787, + "learning_rate": 4.534369524080858e-06, + "loss": 0.6275, + "step": 1510 + }, + { + "epoch": 1.2509663169519603, + "grad_norm": 0.8732156157493591, + "learning_rate": 4.5337301678856595e-06, + "loss": 0.6572, + "step": 1511 + }, + { + "epoch": 1.251794588625069, + "grad_norm": 0.9505333304405212, + "learning_rate": 4.533090418182631e-06, + "loss": 0.6352, + "step": 1512 + }, + { + "epoch": 1.2526228602981777, + "grad_norm": 0.8931122422218323, + "learning_rate": 4.532450275095558e-06, + "loss": 0.649, + "step": 1513 + }, + { + "epoch": 1.2534511319712867, + "grad_norm": 0.8781880140304565, + "learning_rate": 4.531809738748304e-06, + "loss": 0.651, + "step": 1514 + }, + { + "epoch": 1.2542794036443954, + "grad_norm": 0.9201871156692505, + "learning_rate": 4.531168809264805e-06, + "loss": 0.6451, + "step": 1515 + }, + { + "epoch": 1.2551076753175041, + "grad_norm": 0.9318148493766785, + "learning_rate": 4.530527486769077e-06, + "loss": 0.6458, + "step": 1516 + }, + { + "epoch": 1.2559359469906128, + "grad_norm": 0.9131407141685486, + "learning_rate": 4.529885771385208e-06, + "loss": 0.6679, + "step": 1517 + }, + { + "epoch": 1.2567642186637218, + "grad_norm": 0.8942316770553589, + "learning_rate": 4.529243663237365e-06, + "loss": 0.6262, + "step": 1518 + }, + { + "epoch": 1.2575924903368305, + "grad_norm": 0.9058274030685425, + "learning_rate": 4.528601162449788e-06, + "loss": 0.635, + "step": 1519 + }, + { + "epoch": 1.2584207620099392, + "grad_norm": 0.9129453897476196, + "learning_rate": 4.527958269146798e-06, + "loss": 0.6375, + "step": 1520 + }, + { + "epoch": 1.259249033683048, + "grad_norm": 0.8737504482269287, + "learning_rate": 4.527314983452787e-06, + "loss": 0.6437, + "step": 1521 + }, + { + "epoch": 1.2600773053561567, + "grad_norm": 0.9680130481719971, + "learning_rate": 4.526671305492225e-06, + "loss": 0.6546, + "step": 1522 + }, + { + "epoch": 1.2609055770292656, + "grad_norm": 0.9186084270477295, + "learning_rate": 4.526027235389658e-06, + "loss": 0.6505, + "step": 1523 + }, + { + "epoch": 1.2617338487023744, + "grad_norm": 0.9164871573448181, + "learning_rate": 4.525382773269706e-06, + "loss": 0.6541, + "step": 1524 + }, + { + "epoch": 1.262562120375483, + "grad_norm": 0.8994123935699463, + "learning_rate": 4.5247379192570695e-06, + "loss": 0.6313, + "step": 1525 + }, + { + "epoch": 1.263390392048592, + "grad_norm": 0.8780331015586853, + "learning_rate": 4.524092673476519e-06, + "loss": 0.6496, + "step": 1526 + }, + { + "epoch": 1.2642186637217008, + "grad_norm": 0.8670293092727661, + "learning_rate": 4.5234470360529045e-06, + "loss": 0.6422, + "step": 1527 + }, + { + "epoch": 1.2650469353948095, + "grad_norm": 0.9305918216705322, + "learning_rate": 4.522801007111152e-06, + "loss": 0.6352, + "step": 1528 + }, + { + "epoch": 1.2658752070679182, + "grad_norm": 0.9444066286087036, + "learning_rate": 4.52215458677626e-06, + "loss": 0.6357, + "step": 1529 + }, + { + "epoch": 1.266703478741027, + "grad_norm": 0.8987369537353516, + "learning_rate": 4.521507775173305e-06, + "loss": 0.6407, + "step": 1530 + }, + { + "epoch": 1.267531750414136, + "grad_norm": 0.9200116395950317, + "learning_rate": 4.5208605724274415e-06, + "loss": 0.6366, + "step": 1531 + }, + { + "epoch": 1.2683600220872446, + "grad_norm": 0.8680561780929565, + "learning_rate": 4.520212978663894e-06, + "loss": 0.6389, + "step": 1532 + }, + { + "epoch": 1.2691882937603534, + "grad_norm": 0.8928949236869812, + "learning_rate": 4.519564994007968e-06, + "loss": 0.6552, + "step": 1533 + }, + { + "epoch": 1.2700165654334623, + "grad_norm": 0.8986184000968933, + "learning_rate": 4.518916618585042e-06, + "loss": 0.6554, + "step": 1534 + }, + { + "epoch": 1.270844837106571, + "grad_norm": 0.9239423274993896, + "learning_rate": 4.51826785252057e-06, + "loss": 0.6448, + "step": 1535 + }, + { + "epoch": 1.2716731087796798, + "grad_norm": 0.9534674286842346, + "learning_rate": 4.517618695940082e-06, + "loss": 0.6303, + "step": 1536 + }, + { + "epoch": 1.2725013804527885, + "grad_norm": 0.9341598749160767, + "learning_rate": 4.516969148969186e-06, + "loss": 0.6335, + "step": 1537 + }, + { + "epoch": 1.2733296521258972, + "grad_norm": 0.9264217615127563, + "learning_rate": 4.51631921173356e-06, + "loss": 0.6543, + "step": 1538 + }, + { + "epoch": 1.274157923799006, + "grad_norm": 0.9460370540618896, + "learning_rate": 4.515668884358963e-06, + "loss": 0.6531, + "step": 1539 + }, + { + "epoch": 1.274986195472115, + "grad_norm": 0.9690954089164734, + "learning_rate": 4.515018166971227e-06, + "loss": 0.6495, + "step": 1540 + }, + { + "epoch": 1.2758144671452236, + "grad_norm": 0.8746740818023682, + "learning_rate": 4.514367059696259e-06, + "loss": 0.6169, + "step": 1541 + }, + { + "epoch": 1.2766427388183323, + "grad_norm": 0.9288576245307922, + "learning_rate": 4.513715562660043e-06, + "loss": 0.6459, + "step": 1542 + }, + { + "epoch": 1.2774710104914413, + "grad_norm": 0.9935420751571655, + "learning_rate": 4.513063675988637e-06, + "loss": 0.6471, + "step": 1543 + }, + { + "epoch": 1.27829928216455, + "grad_norm": 0.9584968686103821, + "learning_rate": 4.512411399808175e-06, + "loss": 0.6473, + "step": 1544 + }, + { + "epoch": 1.2791275538376587, + "grad_norm": 0.9049419164657593, + "learning_rate": 4.511758734244867e-06, + "loss": 0.6667, + "step": 1545 + }, + { + "epoch": 1.2799558255107675, + "grad_norm": 0.9456910490989685, + "learning_rate": 4.511105679424998e-06, + "loss": 0.625, + "step": 1546 + }, + { + "epoch": 1.2807840971838762, + "grad_norm": 0.9267060160636902, + "learning_rate": 4.510452235474926e-06, + "loss": 0.6239, + "step": 1547 + }, + { + "epoch": 1.2816123688569852, + "grad_norm": 0.9363065958023071, + "learning_rate": 4.509798402521088e-06, + "loss": 0.6416, + "step": 1548 + }, + { + "epoch": 1.2824406405300939, + "grad_norm": 0.85664302110672, + "learning_rate": 4.509144180689995e-06, + "loss": 0.6297, + "step": 1549 + }, + { + "epoch": 1.2832689122032026, + "grad_norm": 0.9298888444900513, + "learning_rate": 4.508489570108231e-06, + "loss": 0.6305, + "step": 1550 + }, + { + "epoch": 1.2840971838763116, + "grad_norm": 0.9485051035881042, + "learning_rate": 4.507834570902459e-06, + "loss": 0.6409, + "step": 1551 + }, + { + "epoch": 1.2849254555494203, + "grad_norm": 0.9129849076271057, + "learning_rate": 4.507179183199414e-06, + "loss": 0.6215, + "step": 1552 + }, + { + "epoch": 1.285753727222529, + "grad_norm": 0.9630280137062073, + "learning_rate": 4.506523407125907e-06, + "loss": 0.6229, + "step": 1553 + }, + { + "epoch": 1.2865819988956377, + "grad_norm": 0.9047848582267761, + "learning_rate": 4.505867242808826e-06, + "loss": 0.6434, + "step": 1554 + }, + { + "epoch": 1.2874102705687465, + "grad_norm": 0.9219448566436768, + "learning_rate": 4.5052106903751315e-06, + "loss": 0.6449, + "step": 1555 + }, + { + "epoch": 1.2882385422418554, + "grad_norm": 0.9347684383392334, + "learning_rate": 4.504553749951861e-06, + "loss": 0.6272, + "step": 1556 + }, + { + "epoch": 1.2890668139149641, + "grad_norm": 0.8930262923240662, + "learning_rate": 4.503896421666126e-06, + "loss": 0.6364, + "step": 1557 + }, + { + "epoch": 1.2898950855880729, + "grad_norm": 0.9067763090133667, + "learning_rate": 4.503238705645114e-06, + "loss": 0.6317, + "step": 1558 + }, + { + "epoch": 1.2907233572611816, + "grad_norm": 0.9524715542793274, + "learning_rate": 4.5025806020160865e-06, + "loss": 0.6525, + "step": 1559 + }, + { + "epoch": 1.2915516289342905, + "grad_norm": 0.927255392074585, + "learning_rate": 4.5019221109063795e-06, + "loss": 0.6343, + "step": 1560 + }, + { + "epoch": 1.2923799006073993, + "grad_norm": 0.904295802116394, + "learning_rate": 4.5012632324434065e-06, + "loss": 0.6484, + "step": 1561 + }, + { + "epoch": 1.293208172280508, + "grad_norm": 0.9138978123664856, + "learning_rate": 4.500603966754653e-06, + "loss": 0.6374, + "step": 1562 + }, + { + "epoch": 1.2940364439536167, + "grad_norm": 0.9038779139518738, + "learning_rate": 4.499944313967681e-06, + "loss": 0.6395, + "step": 1563 + }, + { + "epoch": 1.2948647156267254, + "grad_norm": 0.9218693971633911, + "learning_rate": 4.499284274210128e-06, + "loss": 0.6325, + "step": 1564 + }, + { + "epoch": 1.2956929872998344, + "grad_norm": 0.9210799932479858, + "learning_rate": 4.4986238476097055e-06, + "loss": 0.6476, + "step": 1565 + }, + { + "epoch": 1.2965212589729431, + "grad_norm": 0.8977149724960327, + "learning_rate": 4.4979630342941994e-06, + "loss": 0.6489, + "step": 1566 + }, + { + "epoch": 1.2973495306460519, + "grad_norm": 0.9423681497573853, + "learning_rate": 4.49730183439147e-06, + "loss": 0.6391, + "step": 1567 + }, + { + "epoch": 1.2981778023191608, + "grad_norm": 0.9404032826423645, + "learning_rate": 4.4966402480294545e-06, + "loss": 0.6233, + "step": 1568 + }, + { + "epoch": 1.2990060739922695, + "grad_norm": 0.9104741215705872, + "learning_rate": 4.495978275336164e-06, + "loss": 0.6369, + "step": 1569 + }, + { + "epoch": 1.2998343456653783, + "grad_norm": 0.8869790434837341, + "learning_rate": 4.4953159164396835e-06, + "loss": 0.6331, + "step": 1570 + }, + { + "epoch": 1.300662617338487, + "grad_norm": 0.9269986152648926, + "learning_rate": 4.494653171468173e-06, + "loss": 0.6361, + "step": 1571 + }, + { + "epoch": 1.3014908890115957, + "grad_norm": 0.9210803508758545, + "learning_rate": 4.4939900405498675e-06, + "loss": 0.6314, + "step": 1572 + }, + { + "epoch": 1.3023191606847047, + "grad_norm": 0.9426820874214172, + "learning_rate": 4.4933265238130775e-06, + "loss": 0.6625, + "step": 1573 + }, + { + "epoch": 1.3031474323578134, + "grad_norm": 0.9198150634765625, + "learning_rate": 4.492662621386186e-06, + "loss": 0.627, + "step": 1574 + }, + { + "epoch": 1.3039757040309221, + "grad_norm": 0.9312534928321838, + "learning_rate": 4.4919983333976525e-06, + "loss": 0.6386, + "step": 1575 + }, + { + "epoch": 1.304803975704031, + "grad_norm": 0.8926329016685486, + "learning_rate": 4.491333659976011e-06, + "loss": 0.6559, + "step": 1576 + }, + { + "epoch": 1.3056322473771398, + "grad_norm": 0.9172425270080566, + "learning_rate": 4.4906686012498705e-06, + "loss": 0.6254, + "step": 1577 + }, + { + "epoch": 1.3064605190502485, + "grad_norm": 0.9896116256713867, + "learning_rate": 4.490003157347911e-06, + "loss": 0.6331, + "step": 1578 + }, + { + "epoch": 1.3072887907233572, + "grad_norm": 0.9267039895057678, + "learning_rate": 4.489337328398891e-06, + "loss": 0.6296, + "step": 1579 + }, + { + "epoch": 1.308117062396466, + "grad_norm": 0.9033195376396179, + "learning_rate": 4.488671114531643e-06, + "loss": 0.6354, + "step": 1580 + }, + { + "epoch": 1.3089453340695747, + "grad_norm": 0.9047731161117554, + "learning_rate": 4.488004515875072e-06, + "loss": 0.6351, + "step": 1581 + }, + { + "epoch": 1.3097736057426836, + "grad_norm": 0.9096142649650574, + "learning_rate": 4.4873375325581596e-06, + "loss": 0.6504, + "step": 1582 + }, + { + "epoch": 1.3106018774157924, + "grad_norm": 0.8848153352737427, + "learning_rate": 4.48667016470996e-06, + "loss": 0.6354, + "step": 1583 + }, + { + "epoch": 1.311430149088901, + "grad_norm": 0.9036188721656799, + "learning_rate": 4.486002412459603e-06, + "loss": 0.6431, + "step": 1584 + }, + { + "epoch": 1.31225842076201, + "grad_norm": 0.8881173133850098, + "learning_rate": 4.4853342759362924e-06, + "loss": 0.6509, + "step": 1585 + }, + { + "epoch": 1.3130866924351188, + "grad_norm": 0.9195591807365417, + "learning_rate": 4.484665755269307e-06, + "loss": 0.6522, + "step": 1586 + }, + { + "epoch": 1.3139149641082275, + "grad_norm": 0.9021509885787964, + "learning_rate": 4.483996850587998e-06, + "loss": 0.6532, + "step": 1587 + }, + { + "epoch": 1.3147432357813362, + "grad_norm": 0.8902403116226196, + "learning_rate": 4.483327562021794e-06, + "loss": 0.5942, + "step": 1588 + }, + { + "epoch": 1.315571507454445, + "grad_norm": 0.8933504819869995, + "learning_rate": 4.482657889700194e-06, + "loss": 0.6305, + "step": 1589 + }, + { + "epoch": 1.316399779127554, + "grad_norm": 0.9233856201171875, + "learning_rate": 4.481987833752776e-06, + "loss": 0.6275, + "step": 1590 + }, + { + "epoch": 1.3172280508006626, + "grad_norm": 0.9482141733169556, + "learning_rate": 4.4813173943091865e-06, + "loss": 0.6232, + "step": 1591 + }, + { + "epoch": 1.3180563224737714, + "grad_norm": 0.94732266664505, + "learning_rate": 4.480646571499151e-06, + "loss": 0.6297, + "step": 1592 + }, + { + "epoch": 1.3188845941468803, + "grad_norm": 0.9066787958145142, + "learning_rate": 4.479975365452468e-06, + "loss": 0.6334, + "step": 1593 + }, + { + "epoch": 1.319712865819989, + "grad_norm": 0.9063938856124878, + "learning_rate": 4.479303776299008e-06, + "loss": 0.6415, + "step": 1594 + }, + { + "epoch": 1.3205411374930978, + "grad_norm": 0.9286819696426392, + "learning_rate": 4.478631804168719e-06, + "loss": 0.6534, + "step": 1595 + }, + { + "epoch": 1.3213694091662065, + "grad_norm": 0.9161240458488464, + "learning_rate": 4.47795944919162e-06, + "loss": 0.6401, + "step": 1596 + }, + { + "epoch": 1.3221976808393152, + "grad_norm": 0.956538200378418, + "learning_rate": 4.4772867114978056e-06, + "loss": 0.6649, + "step": 1597 + }, + { + "epoch": 1.323025952512424, + "grad_norm": 0.910504162311554, + "learning_rate": 4.4766135912174455e-06, + "loss": 0.6456, + "step": 1598 + }, + { + "epoch": 1.323854224185533, + "grad_norm": 0.9054026007652283, + "learning_rate": 4.4759400884807805e-06, + "loss": 0.6458, + "step": 1599 + }, + { + "epoch": 1.3246824958586416, + "grad_norm": 0.933506429195404, + "learning_rate": 4.475266203418127e-06, + "loss": 0.626, + "step": 1600 + }, + { + "epoch": 1.3255107675317503, + "grad_norm": 0.9042014479637146, + "learning_rate": 4.474591936159878e-06, + "loss": 0.6376, + "step": 1601 + }, + { + "epoch": 1.3263390392048593, + "grad_norm": 0.9005138278007507, + "learning_rate": 4.473917286836496e-06, + "loss": 0.6093, + "step": 1602 + }, + { + "epoch": 1.327167310877968, + "grad_norm": 0.9096264839172363, + "learning_rate": 4.4732422555785196e-06, + "loss": 0.6344, + "step": 1603 + }, + { + "epoch": 1.3279955825510767, + "grad_norm": 0.9171631336212158, + "learning_rate": 4.4725668425165605e-06, + "loss": 0.6454, + "step": 1604 + }, + { + "epoch": 1.3288238542241855, + "grad_norm": 0.8821735382080078, + "learning_rate": 4.4718910477813055e-06, + "loss": 0.6357, + "step": 1605 + }, + { + "epoch": 1.3296521258972942, + "grad_norm": 0.918499231338501, + "learning_rate": 4.471214871503514e-06, + "loss": 0.6361, + "step": 1606 + }, + { + "epoch": 1.3304803975704032, + "grad_norm": 0.9187904596328735, + "learning_rate": 4.470538313814021e-06, + "loss": 0.6362, + "step": 1607 + }, + { + "epoch": 1.3313086692435119, + "grad_norm": 0.9272632598876953, + "learning_rate": 4.469861374843734e-06, + "loss": 0.6659, + "step": 1608 + }, + { + "epoch": 1.3321369409166206, + "grad_norm": 0.8950392603874207, + "learning_rate": 4.469184054723632e-06, + "loss": 0.6551, + "step": 1609 + }, + { + "epoch": 1.3329652125897296, + "grad_norm": 0.9604961276054382, + "learning_rate": 4.468506353584773e-06, + "loss": 0.643, + "step": 1610 + }, + { + "epoch": 1.3337934842628383, + "grad_norm": 0.933073103427887, + "learning_rate": 4.467828271558286e-06, + "loss": 0.6193, + "step": 1611 + }, + { + "epoch": 1.334621755935947, + "grad_norm": 0.890964686870575, + "learning_rate": 4.467149808775371e-06, + "loss": 0.6423, + "step": 1612 + }, + { + "epoch": 1.3354500276090557, + "grad_norm": 0.8916820883750916, + "learning_rate": 4.466470965367306e-06, + "loss": 0.6399, + "step": 1613 + }, + { + "epoch": 1.3362782992821645, + "grad_norm": 0.9151459336280823, + "learning_rate": 4.465791741465439e-06, + "loss": 0.6439, + "step": 1614 + }, + { + "epoch": 1.3371065709552734, + "grad_norm": 0.9155884385108948, + "learning_rate": 4.465112137201197e-06, + "loss": 0.6516, + "step": 1615 + }, + { + "epoch": 1.3379348426283821, + "grad_norm": 0.9187855124473572, + "learning_rate": 4.464432152706073e-06, + "loss": 0.6263, + "step": 1616 + }, + { + "epoch": 1.3387631143014909, + "grad_norm": 0.9506946206092834, + "learning_rate": 4.46375178811164e-06, + "loss": 0.6586, + "step": 1617 + }, + { + "epoch": 1.3395913859745996, + "grad_norm": 0.9056413769721985, + "learning_rate": 4.463071043549543e-06, + "loss": 0.6301, + "step": 1618 + }, + { + "epoch": 1.3404196576477085, + "grad_norm": 0.9056457877159119, + "learning_rate": 4.462389919151498e-06, + "loss": 0.6352, + "step": 1619 + }, + { + "epoch": 1.3412479293208173, + "grad_norm": 0.9204022288322449, + "learning_rate": 4.461708415049297e-06, + "loss": 0.6446, + "step": 1620 + }, + { + "epoch": 1.342076200993926, + "grad_norm": 0.9411336183547974, + "learning_rate": 4.461026531374804e-06, + "loss": 0.6594, + "step": 1621 + }, + { + "epoch": 1.3429044726670347, + "grad_norm": 0.8956605792045593, + "learning_rate": 4.460344268259958e-06, + "loss": 0.6492, + "step": 1622 + }, + { + "epoch": 1.3437327443401434, + "grad_norm": 0.9010135531425476, + "learning_rate": 4.45966162583677e-06, + "loss": 0.6233, + "step": 1623 + }, + { + "epoch": 1.3445610160132524, + "grad_norm": 0.8948563933372498, + "learning_rate": 4.458978604237325e-06, + "loss": 0.6447, + "step": 1624 + }, + { + "epoch": 1.3453892876863611, + "grad_norm": 0.9227688312530518, + "learning_rate": 4.458295203593782e-06, + "loss": 0.6459, + "step": 1625 + }, + { + "epoch": 1.3462175593594699, + "grad_norm": 0.9532939195632935, + "learning_rate": 4.4576114240383725e-06, + "loss": 0.6724, + "step": 1626 + }, + { + "epoch": 1.3470458310325788, + "grad_norm": 0.9089633226394653, + "learning_rate": 4.4569272657034005e-06, + "loss": 0.6303, + "step": 1627 + }, + { + "epoch": 1.3478741027056875, + "grad_norm": 0.8903496861457825, + "learning_rate": 4.456242728721244e-06, + "loss": 0.6655, + "step": 1628 + }, + { + "epoch": 1.3487023743787963, + "grad_norm": 0.9009859561920166, + "learning_rate": 4.455557813224356e-06, + "loss": 0.6268, + "step": 1629 + }, + { + "epoch": 1.349530646051905, + "grad_norm": 0.9265439510345459, + "learning_rate": 4.454872519345261e-06, + "loss": 0.6586, + "step": 1630 + }, + { + "epoch": 1.3503589177250137, + "grad_norm": 0.8945439457893372, + "learning_rate": 4.454186847216556e-06, + "loss": 0.62, + "step": 1631 + }, + { + "epoch": 1.3511871893981227, + "grad_norm": 0.9313004016876221, + "learning_rate": 4.453500796970913e-06, + "loss": 0.6452, + "step": 1632 + }, + { + "epoch": 1.3520154610712314, + "grad_norm": 0.9005258083343506, + "learning_rate": 4.452814368741076e-06, + "loss": 0.6436, + "step": 1633 + }, + { + "epoch": 1.35284373274434, + "grad_norm": 0.8860014081001282, + "learning_rate": 4.452127562659864e-06, + "loss": 0.6423, + "step": 1634 + }, + { + "epoch": 1.353672004417449, + "grad_norm": 0.9049930572509766, + "learning_rate": 4.4514403788601645e-06, + "loss": 0.6343, + "step": 1635 + }, + { + "epoch": 1.3545002760905578, + "grad_norm": 0.8849244117736816, + "learning_rate": 4.4507528174749425e-06, + "loss": 0.6481, + "step": 1636 + }, + { + "epoch": 1.3553285477636665, + "grad_norm": 0.9214544892311096, + "learning_rate": 4.450064878637235e-06, + "loss": 0.6166, + "step": 1637 + }, + { + "epoch": 1.3561568194367752, + "grad_norm": 0.8948760628700256, + "learning_rate": 4.449376562480153e-06, + "loss": 0.6195, + "step": 1638 + }, + { + "epoch": 1.356985091109884, + "grad_norm": 0.9000245928764343, + "learning_rate": 4.448687869136878e-06, + "loss": 0.6378, + "step": 1639 + }, + { + "epoch": 1.3578133627829927, + "grad_norm": 0.8963801264762878, + "learning_rate": 4.447998798740665e-06, + "loss": 0.6118, + "step": 1640 + }, + { + "epoch": 1.3586416344561016, + "grad_norm": 0.8774341344833374, + "learning_rate": 4.447309351424843e-06, + "loss": 0.6184, + "step": 1641 + }, + { + "epoch": 1.3594699061292104, + "grad_norm": 0.8901870846748352, + "learning_rate": 4.446619527322815e-06, + "loss": 0.6499, + "step": 1642 + }, + { + "epoch": 1.360298177802319, + "grad_norm": 0.9108647108078003, + "learning_rate": 4.445929326568055e-06, + "loss": 0.6243, + "step": 1643 + }, + { + "epoch": 1.361126449475428, + "grad_norm": 0.9178787469863892, + "learning_rate": 4.445238749294109e-06, + "loss": 0.6458, + "step": 1644 + }, + { + "epoch": 1.3619547211485368, + "grad_norm": 0.9278320670127869, + "learning_rate": 4.444547795634598e-06, + "loss": 0.6281, + "step": 1645 + }, + { + "epoch": 1.3627829928216455, + "grad_norm": 0.8769298195838928, + "learning_rate": 4.443856465723216e-06, + "loss": 0.6387, + "step": 1646 + }, + { + "epoch": 1.3636112644947542, + "grad_norm": 0.9313247203826904, + "learning_rate": 4.443164759693729e-06, + "loss": 0.6304, + "step": 1647 + }, + { + "epoch": 1.364439536167863, + "grad_norm": 1.0037778615951538, + "learning_rate": 4.4424726776799745e-06, + "loss": 0.6378, + "step": 1648 + }, + { + "epoch": 1.365267807840972, + "grad_norm": 0.9090721011161804, + "learning_rate": 4.441780219815863e-06, + "loss": 0.6456, + "step": 1649 + }, + { + "epoch": 1.3660960795140806, + "grad_norm": 0.8968444466590881, + "learning_rate": 4.441087386235382e-06, + "loss": 0.6377, + "step": 1650 + }, + { + "epoch": 1.3669243511871894, + "grad_norm": 0.9638763070106506, + "learning_rate": 4.440394177072586e-06, + "loss": 0.651, + "step": 1651 + }, + { + "epoch": 1.3677526228602983, + "grad_norm": 0.9224078059196472, + "learning_rate": 4.439700592461604e-06, + "loss": 0.6492, + "step": 1652 + }, + { + "epoch": 1.368580894533407, + "grad_norm": 0.891180157661438, + "learning_rate": 4.439006632536639e-06, + "loss": 0.6402, + "step": 1653 + }, + { + "epoch": 1.3694091662065158, + "grad_norm": 0.8817652463912964, + "learning_rate": 4.438312297431967e-06, + "loss": 0.6095, + "step": 1654 + }, + { + "epoch": 1.3702374378796245, + "grad_norm": 0.8814468383789062, + "learning_rate": 4.437617587281932e-06, + "loss": 0.6334, + "step": 1655 + }, + { + "epoch": 1.3710657095527332, + "grad_norm": 0.8895065784454346, + "learning_rate": 4.436922502220958e-06, + "loss": 0.6163, + "step": 1656 + }, + { + "epoch": 1.371893981225842, + "grad_norm": 0.9043355584144592, + "learning_rate": 4.436227042383535e-06, + "loss": 0.6625, + "step": 1657 + }, + { + "epoch": 1.372722252898951, + "grad_norm": 0.942014753818512, + "learning_rate": 4.435531207904229e-06, + "loss": 0.6298, + "step": 1658 + }, + { + "epoch": 1.3735505245720596, + "grad_norm": 0.8830228447914124, + "learning_rate": 4.434834998917677e-06, + "loss": 0.6215, + "step": 1659 + }, + { + "epoch": 1.3743787962451683, + "grad_norm": 0.9138136506080627, + "learning_rate": 4.434138415558588e-06, + "loss": 0.63, + "step": 1660 + }, + { + "epoch": 1.3752070679182773, + "grad_norm": 0.8880951404571533, + "learning_rate": 4.433441457961746e-06, + "loss": 0.6411, + "step": 1661 + }, + { + "epoch": 1.376035339591386, + "grad_norm": 0.9435725808143616, + "learning_rate": 4.432744126262005e-06, + "loss": 0.6279, + "step": 1662 + }, + { + "epoch": 1.3768636112644947, + "grad_norm": 0.894296407699585, + "learning_rate": 4.432046420594292e-06, + "loss": 0.6298, + "step": 1663 + }, + { + "epoch": 1.3776918829376035, + "grad_norm": 0.8850218653678894, + "learning_rate": 4.431348341093608e-06, + "loss": 0.627, + "step": 1664 + }, + { + "epoch": 1.3785201546107122, + "grad_norm": 0.8956788182258606, + "learning_rate": 4.430649887895025e-06, + "loss": 0.6412, + "step": 1665 + }, + { + "epoch": 1.3793484262838211, + "grad_norm": 0.9128378033638, + "learning_rate": 4.429951061133683e-06, + "loss": 0.6208, + "step": 1666 + }, + { + "epoch": 1.3801766979569299, + "grad_norm": 0.8763598203659058, + "learning_rate": 4.429251860944803e-06, + "loss": 0.6317, + "step": 1667 + }, + { + "epoch": 1.3810049696300386, + "grad_norm": 0.9210475087165833, + "learning_rate": 4.428552287463672e-06, + "loss": 0.6321, + "step": 1668 + }, + { + "epoch": 1.3818332413031476, + "grad_norm": 0.9203745126724243, + "learning_rate": 4.427852340825651e-06, + "loss": 0.6442, + "step": 1669 + }, + { + "epoch": 1.3826615129762563, + "grad_norm": 0.9229859709739685, + "learning_rate": 4.4271520211661735e-06, + "loss": 0.6402, + "step": 1670 + }, + { + "epoch": 1.383489784649365, + "grad_norm": 0.9455426335334778, + "learning_rate": 4.426451328620744e-06, + "loss": 0.6402, + "step": 1671 + }, + { + "epoch": 1.3843180563224737, + "grad_norm": 0.8894345760345459, + "learning_rate": 4.425750263324942e-06, + "loss": 0.6429, + "step": 1672 + }, + { + "epoch": 1.3851463279955825, + "grad_norm": 0.871449887752533, + "learning_rate": 4.425048825414414e-06, + "loss": 0.6449, + "step": 1673 + }, + { + "epoch": 1.3859745996686914, + "grad_norm": 0.9102624654769897, + "learning_rate": 4.424347015024884e-06, + "loss": 0.634, + "step": 1674 + }, + { + "epoch": 1.3868028713418001, + "grad_norm": 0.881172776222229, + "learning_rate": 4.4236448322921445e-06, + "loss": 0.621, + "step": 1675 + }, + { + "epoch": 1.3876311430149089, + "grad_norm": 0.9143398404121399, + "learning_rate": 4.422942277352063e-06, + "loss": 0.6379, + "step": 1676 + }, + { + "epoch": 1.3884594146880176, + "grad_norm": 0.9155046343803406, + "learning_rate": 4.4222393503405754e-06, + "loss": 0.6168, + "step": 1677 + }, + { + "epoch": 1.3892876863611265, + "grad_norm": 0.9501062631607056, + "learning_rate": 4.421536051393694e-06, + "loss": 0.6238, + "step": 1678 + }, + { + "epoch": 1.3901159580342353, + "grad_norm": 0.9055907130241394, + "learning_rate": 4.420832380647498e-06, + "loss": 0.634, + "step": 1679 + }, + { + "epoch": 1.390944229707344, + "grad_norm": 0.9199142456054688, + "learning_rate": 4.420128338238141e-06, + "loss": 0.6283, + "step": 1680 + }, + { + "epoch": 1.3917725013804527, + "grad_norm": 0.8932124376296997, + "learning_rate": 4.419423924301851e-06, + "loss": 0.6146, + "step": 1681 + }, + { + "epoch": 1.3926007730535614, + "grad_norm": 0.9010130167007446, + "learning_rate": 4.418719138974924e-06, + "loss": 0.6409, + "step": 1682 + }, + { + "epoch": 1.3934290447266704, + "grad_norm": 0.9154212474822998, + "learning_rate": 4.41801398239373e-06, + "loss": 0.6115, + "step": 1683 + }, + { + "epoch": 1.3942573163997791, + "grad_norm": 0.9208158850669861, + "learning_rate": 4.417308454694709e-06, + "loss": 0.6242, + "step": 1684 + }, + { + "epoch": 1.3950855880728878, + "grad_norm": 0.932996928691864, + "learning_rate": 4.4166025560143756e-06, + "loss": 0.6277, + "step": 1685 + }, + { + "epoch": 1.3959138597459968, + "grad_norm": 0.8990349769592285, + "learning_rate": 4.415896286489314e-06, + "loss": 0.6128, + "step": 1686 + }, + { + "epoch": 1.3967421314191055, + "grad_norm": 0.926906168460846, + "learning_rate": 4.4151896462561805e-06, + "loss": 0.6442, + "step": 1687 + }, + { + "epoch": 1.3975704030922143, + "grad_norm": 0.9330859184265137, + "learning_rate": 4.414482635451704e-06, + "loss": 0.5995, + "step": 1688 + }, + { + "epoch": 1.398398674765323, + "grad_norm": 0.9039576649665833, + "learning_rate": 4.413775254212683e-06, + "loss": 0.6363, + "step": 1689 + }, + { + "epoch": 1.3992269464384317, + "grad_norm": 0.9228506684303284, + "learning_rate": 4.413067502675992e-06, + "loss": 0.6294, + "step": 1690 + }, + { + "epoch": 1.4000552181115407, + "grad_norm": 0.921620786190033, + "learning_rate": 4.412359380978571e-06, + "loss": 0.64, + "step": 1691 + }, + { + "epoch": 1.4008834897846494, + "grad_norm": 0.9379149079322815, + "learning_rate": 4.4116508892574375e-06, + "loss": 0.6207, + "step": 1692 + }, + { + "epoch": 1.401711761457758, + "grad_norm": 1.0528016090393066, + "learning_rate": 4.410942027649677e-06, + "loss": 0.6204, + "step": 1693 + }, + { + "epoch": 1.402540033130867, + "grad_norm": 0.9510579705238342, + "learning_rate": 4.410232796292447e-06, + "loss": 0.6314, + "step": 1694 + }, + { + "epoch": 1.4033683048039758, + "grad_norm": 0.8922870755195618, + "learning_rate": 4.409523195322979e-06, + "loss": 0.6056, + "step": 1695 + }, + { + "epoch": 1.4041965764770845, + "grad_norm": 0.9239864349365234, + "learning_rate": 4.408813224878573e-06, + "loss": 0.6404, + "step": 1696 + }, + { + "epoch": 1.4050248481501932, + "grad_norm": 0.8696756362915039, + "learning_rate": 4.408102885096601e-06, + "loss": 0.6553, + "step": 1697 + }, + { + "epoch": 1.405853119823302, + "grad_norm": 0.9176343679428101, + "learning_rate": 4.407392176114509e-06, + "loss": 0.6286, + "step": 1698 + }, + { + "epoch": 1.4066813914964107, + "grad_norm": 0.9825496077537537, + "learning_rate": 4.40668109806981e-06, + "loss": 0.6421, + "step": 1699 + }, + { + "epoch": 1.4075096631695196, + "grad_norm": 0.9334873557090759, + "learning_rate": 4.405969651100093e-06, + "loss": 0.6347, + "step": 1700 + }, + { + "epoch": 1.4083379348426284, + "grad_norm": 0.8881192803382874, + "learning_rate": 4.405257835343015e-06, + "loss": 0.6377, + "step": 1701 + }, + { + "epoch": 1.409166206515737, + "grad_norm": 0.8896775245666504, + "learning_rate": 4.404545650936307e-06, + "loss": 0.624, + "step": 1702 + }, + { + "epoch": 1.409994478188846, + "grad_norm": 0.9361454844474792, + "learning_rate": 4.40383309801777e-06, + "loss": 0.6561, + "step": 1703 + }, + { + "epoch": 1.4108227498619548, + "grad_norm": 0.9082345962524414, + "learning_rate": 4.403120176725276e-06, + "loss": 0.6415, + "step": 1704 + }, + { + "epoch": 1.4116510215350635, + "grad_norm": 0.9453962445259094, + "learning_rate": 4.402406887196767e-06, + "loss": 0.6443, + "step": 1705 + }, + { + "epoch": 1.4124792932081722, + "grad_norm": 0.9420861005783081, + "learning_rate": 4.401693229570261e-06, + "loss": 0.6544, + "step": 1706 + }, + { + "epoch": 1.413307564881281, + "grad_norm": 0.869742214679718, + "learning_rate": 4.400979203983841e-06, + "loss": 0.6072, + "step": 1707 + }, + { + "epoch": 1.41413583655439, + "grad_norm": 0.915322482585907, + "learning_rate": 4.4002648105756665e-06, + "loss": 0.6657, + "step": 1708 + }, + { + "epoch": 1.4149641082274986, + "grad_norm": 0.9029238224029541, + "learning_rate": 4.3995500494839656e-06, + "loss": 0.6157, + "step": 1709 + }, + { + "epoch": 1.4157923799006074, + "grad_norm": 0.8992076516151428, + "learning_rate": 4.398834920847037e-06, + "loss": 0.6321, + "step": 1710 + }, + { + "epoch": 1.4166206515737163, + "grad_norm": 0.9099612236022949, + "learning_rate": 4.398119424803252e-06, + "loss": 0.6123, + "step": 1711 + }, + { + "epoch": 1.417448923246825, + "grad_norm": 0.9081803560256958, + "learning_rate": 4.397403561491052e-06, + "loss": 0.6369, + "step": 1712 + }, + { + "epoch": 1.4182771949199338, + "grad_norm": 0.8873599767684937, + "learning_rate": 4.3966873310489505e-06, + "loss": 0.6503, + "step": 1713 + }, + { + "epoch": 1.4191054665930425, + "grad_norm": 0.8841976523399353, + "learning_rate": 4.395970733615531e-06, + "loss": 0.6303, + "step": 1714 + }, + { + "epoch": 1.4199337382661512, + "grad_norm": 0.8922518491744995, + "learning_rate": 4.395253769329449e-06, + "loss": 0.6105, + "step": 1715 + }, + { + "epoch": 1.42076200993926, + "grad_norm": 0.9023404717445374, + "learning_rate": 4.39453643832943e-06, + "loss": 0.6292, + "step": 1716 + }, + { + "epoch": 1.421590281612369, + "grad_norm": 0.9042630791664124, + "learning_rate": 4.39381874075427e-06, + "loss": 0.6675, + "step": 1717 + }, + { + "epoch": 1.4224185532854776, + "grad_norm": 0.8933060169219971, + "learning_rate": 4.393100676742838e-06, + "loss": 0.6125, + "step": 1718 + }, + { + "epoch": 1.4232468249585863, + "grad_norm": 0.8821312189102173, + "learning_rate": 4.392382246434073e-06, + "loss": 0.6266, + "step": 1719 + }, + { + "epoch": 1.4240750966316953, + "grad_norm": 0.904460608959198, + "learning_rate": 4.391663449966983e-06, + "loss": 0.5984, + "step": 1720 + }, + { + "epoch": 1.424903368304804, + "grad_norm": 0.8999350666999817, + "learning_rate": 4.39094428748065e-06, + "loss": 0.6653, + "step": 1721 + }, + { + "epoch": 1.4257316399779127, + "grad_norm": 0.9061364531517029, + "learning_rate": 4.390224759114224e-06, + "loss": 0.6306, + "step": 1722 + }, + { + "epoch": 1.4265599116510215, + "grad_norm": 0.8995294570922852, + "learning_rate": 4.389504865006927e-06, + "loss": 0.6656, + "step": 1723 + }, + { + "epoch": 1.4273881833241302, + "grad_norm": 0.904737114906311, + "learning_rate": 4.388784605298053e-06, + "loss": 0.6234, + "step": 1724 + }, + { + "epoch": 1.4282164549972391, + "grad_norm": 0.8827239871025085, + "learning_rate": 4.388063980126966e-06, + "loss": 0.6418, + "step": 1725 + }, + { + "epoch": 1.4290447266703479, + "grad_norm": 0.8791747689247131, + "learning_rate": 4.387342989633097e-06, + "loss": 0.6174, + "step": 1726 + }, + { + "epoch": 1.4298729983434566, + "grad_norm": 0.8831204175949097, + "learning_rate": 4.386621633955955e-06, + "loss": 0.6222, + "step": 1727 + }, + { + "epoch": 1.4307012700165656, + "grad_norm": 0.908511221408844, + "learning_rate": 4.3858999132351124e-06, + "loss": 0.6373, + "step": 1728 + }, + { + "epoch": 1.4315295416896743, + "grad_norm": 0.900497555732727, + "learning_rate": 4.385177827610218e-06, + "loss": 0.6291, + "step": 1729 + }, + { + "epoch": 1.432357813362783, + "grad_norm": 0.930695652961731, + "learning_rate": 4.384455377220986e-06, + "loss": 0.6408, + "step": 1730 + }, + { + "epoch": 1.4331860850358917, + "grad_norm": 0.9583060145378113, + "learning_rate": 4.383732562207206e-06, + "loss": 0.6452, + "step": 1731 + }, + { + "epoch": 1.4340143567090005, + "grad_norm": 0.9173291921615601, + "learning_rate": 4.383009382708734e-06, + "loss": 0.6491, + "step": 1732 + }, + { + "epoch": 1.4348426283821094, + "grad_norm": 0.9054985642433167, + "learning_rate": 4.382285838865501e-06, + "loss": 0.6179, + "step": 1733 + }, + { + "epoch": 1.4356709000552181, + "grad_norm": 0.9162827730178833, + "learning_rate": 4.381561930817504e-06, + "loss": 0.6428, + "step": 1734 + }, + { + "epoch": 1.4364991717283269, + "grad_norm": 0.9013632535934448, + "learning_rate": 4.380837658704813e-06, + "loss": 0.6385, + "step": 1735 + }, + { + "epoch": 1.4373274434014356, + "grad_norm": 0.9274044632911682, + "learning_rate": 4.3801130226675685e-06, + "loss": 0.6508, + "step": 1736 + }, + { + "epoch": 1.4381557150745445, + "grad_norm": 0.8884710669517517, + "learning_rate": 4.37938802284598e-06, + "loss": 0.6675, + "step": 1737 + }, + { + "epoch": 1.4389839867476533, + "grad_norm": 0.9156724214553833, + "learning_rate": 4.378662659380329e-06, + "loss": 0.6424, + "step": 1738 + }, + { + "epoch": 1.439812258420762, + "grad_norm": 0.9267123341560364, + "learning_rate": 4.377936932410966e-06, + "loss": 0.6333, + "step": 1739 + }, + { + "epoch": 1.4406405300938707, + "grad_norm": 0.8859686851501465, + "learning_rate": 4.377210842078311e-06, + "loss": 0.6343, + "step": 1740 + }, + { + "epoch": 1.4414688017669794, + "grad_norm": 0.92928147315979, + "learning_rate": 4.376484388522858e-06, + "loss": 0.6124, + "step": 1741 + }, + { + "epoch": 1.4422970734400884, + "grad_norm": 0.8908365964889526, + "learning_rate": 4.375757571885169e-06, + "loss": 0.631, + "step": 1742 + }, + { + "epoch": 1.4431253451131971, + "grad_norm": 0.8956666588783264, + "learning_rate": 4.375030392305875e-06, + "loss": 0.6314, + "step": 1743 + }, + { + "epoch": 1.4439536167863058, + "grad_norm": 0.8836198449134827, + "learning_rate": 4.374302849925679e-06, + "loss": 0.6179, + "step": 1744 + }, + { + "epoch": 1.4447818884594148, + "grad_norm": 0.87999027967453, + "learning_rate": 4.373574944885354e-06, + "loss": 0.6226, + "step": 1745 + }, + { + "epoch": 1.4456101601325235, + "grad_norm": 0.9118010997772217, + "learning_rate": 4.3728466773257414e-06, + "loss": 0.6319, + "step": 1746 + }, + { + "epoch": 1.4464384318056323, + "grad_norm": 0.8989576101303101, + "learning_rate": 4.3721180473877555e-06, + "loss": 0.6049, + "step": 1747 + }, + { + "epoch": 1.447266703478741, + "grad_norm": 0.8896577954292297, + "learning_rate": 4.371389055212379e-06, + "loss": 0.6242, + "step": 1748 + }, + { + "epoch": 1.4480949751518497, + "grad_norm": 0.8993588089942932, + "learning_rate": 4.370659700940665e-06, + "loss": 0.6255, + "step": 1749 + }, + { + "epoch": 1.4489232468249587, + "grad_norm": 0.896486759185791, + "learning_rate": 4.369929984713738e-06, + "loss": 0.6354, + "step": 1750 + }, + { + "epoch": 1.4497515184980674, + "grad_norm": 0.9003145694732666, + "learning_rate": 4.369199906672789e-06, + "loss": 0.6464, + "step": 1751 + }, + { + "epoch": 1.450579790171176, + "grad_norm": 0.8885106444358826, + "learning_rate": 4.368469466959083e-06, + "loss": 0.6567, + "step": 1752 + }, + { + "epoch": 1.451408061844285, + "grad_norm": 0.8914943933486938, + "learning_rate": 4.367738665713953e-06, + "loss": 0.6784, + "step": 1753 + }, + { + "epoch": 1.4522363335173938, + "grad_norm": 0.933029055595398, + "learning_rate": 4.367007503078802e-06, + "loss": 0.6495, + "step": 1754 + }, + { + "epoch": 1.4530646051905025, + "grad_norm": 0.8788006901741028, + "learning_rate": 4.366275979195102e-06, + "loss": 0.6352, + "step": 1755 + }, + { + "epoch": 1.4538928768636112, + "grad_norm": 0.8768455982208252, + "learning_rate": 4.365544094204399e-06, + "loss": 0.6329, + "step": 1756 + }, + { + "epoch": 1.45472114853672, + "grad_norm": 0.9021796584129333, + "learning_rate": 4.364811848248303e-06, + "loss": 0.6616, + "step": 1757 + }, + { + "epoch": 1.4555494202098287, + "grad_norm": 0.9197986125946045, + "learning_rate": 4.364079241468499e-06, + "loss": 0.6248, + "step": 1758 + }, + { + "epoch": 1.4563776918829376, + "grad_norm": 0.9095743894577026, + "learning_rate": 4.3633462740067375e-06, + "loss": 0.6722, + "step": 1759 + }, + { + "epoch": 1.4572059635560464, + "grad_norm": 0.9114678502082825, + "learning_rate": 4.362612946004843e-06, + "loss": 0.6278, + "step": 1760 + }, + { + "epoch": 1.458034235229155, + "grad_norm": 0.8978061676025391, + "learning_rate": 4.3618792576047054e-06, + "loss": 0.633, + "step": 1761 + }, + { + "epoch": 1.458862506902264, + "grad_norm": 0.8578571677207947, + "learning_rate": 4.361145208948287e-06, + "loss": 0.6263, + "step": 1762 + }, + { + "epoch": 1.4596907785753728, + "grad_norm": 0.9333001375198364, + "learning_rate": 4.360410800177622e-06, + "loss": 0.6383, + "step": 1763 + }, + { + "epoch": 1.4605190502484815, + "grad_norm": 0.9448907971382141, + "learning_rate": 4.359676031434808e-06, + "loss": 0.6324, + "step": 1764 + }, + { + "epoch": 1.4613473219215902, + "grad_norm": 0.8806003928184509, + "learning_rate": 4.358940902862019e-06, + "loss": 0.6365, + "step": 1765 + }, + { + "epoch": 1.462175593594699, + "grad_norm": 0.9975590109825134, + "learning_rate": 4.358205414601493e-06, + "loss": 0.6436, + "step": 1766 + }, + { + "epoch": 1.463003865267808, + "grad_norm": 1.0076019763946533, + "learning_rate": 4.357469566795542e-06, + "loss": 0.6234, + "step": 1767 + }, + { + "epoch": 1.4638321369409166, + "grad_norm": 0.9254529476165771, + "learning_rate": 4.356733359586545e-06, + "loss": 0.6244, + "step": 1768 + }, + { + "epoch": 1.4646604086140254, + "grad_norm": 0.9129828810691833, + "learning_rate": 4.35599679311695e-06, + "loss": 0.6519, + "step": 1769 + }, + { + "epoch": 1.4654886802871343, + "grad_norm": 0.9141498804092407, + "learning_rate": 4.355259867529278e-06, + "loss": 0.6343, + "step": 1770 + }, + { + "epoch": 1.466316951960243, + "grad_norm": 0.920790433883667, + "learning_rate": 4.354522582966116e-06, + "loss": 0.6505, + "step": 1771 + }, + { + "epoch": 1.4671452236333518, + "grad_norm": 0.9683033227920532, + "learning_rate": 4.353784939570121e-06, + "loss": 0.6238, + "step": 1772 + }, + { + "epoch": 1.4679734953064605, + "grad_norm": 0.882957935333252, + "learning_rate": 4.3530469374840226e-06, + "loss": 0.6358, + "step": 1773 + }, + { + "epoch": 1.4688017669795692, + "grad_norm": 0.943648099899292, + "learning_rate": 4.352308576850615e-06, + "loss": 0.6274, + "step": 1774 + }, + { + "epoch": 1.4696300386526782, + "grad_norm": 0.9609524607658386, + "learning_rate": 4.351569857812764e-06, + "loss": 0.6368, + "step": 1775 + }, + { + "epoch": 1.4704583103257869, + "grad_norm": 0.9170458912849426, + "learning_rate": 4.3508307805134055e-06, + "loss": 0.6515, + "step": 1776 + }, + { + "epoch": 1.4712865819988956, + "grad_norm": 0.8956430554389954, + "learning_rate": 4.350091345095545e-06, + "loss": 0.6394, + "step": 1777 + }, + { + "epoch": 1.4721148536720043, + "grad_norm": 0.9145321846008301, + "learning_rate": 4.349351551702256e-06, + "loss": 0.6572, + "step": 1778 + }, + { + "epoch": 1.4729431253451133, + "grad_norm": 0.8807089328765869, + "learning_rate": 4.34861140047668e-06, + "loss": 0.6391, + "step": 1779 + }, + { + "epoch": 1.473771397018222, + "grad_norm": 0.8828510642051697, + "learning_rate": 4.347870891562032e-06, + "loss": 0.6583, + "step": 1780 + }, + { + "epoch": 1.4745996686913307, + "grad_norm": 0.9017962217330933, + "learning_rate": 4.34713002510159e-06, + "loss": 0.6337, + "step": 1781 + }, + { + "epoch": 1.4754279403644395, + "grad_norm": 0.8738356232643127, + "learning_rate": 4.346388801238708e-06, + "loss": 0.637, + "step": 1782 + }, + { + "epoch": 1.4762562120375482, + "grad_norm": 0.8805137872695923, + "learning_rate": 4.345647220116805e-06, + "loss": 0.6306, + "step": 1783 + }, + { + "epoch": 1.4770844837106571, + "grad_norm": 0.923599362373352, + "learning_rate": 4.34490528187937e-06, + "loss": 0.6254, + "step": 1784 + }, + { + "epoch": 1.4779127553837659, + "grad_norm": 0.9181063175201416, + "learning_rate": 4.344162986669961e-06, + "loss": 0.6257, + "step": 1785 + }, + { + "epoch": 1.4787410270568746, + "grad_norm": 0.9060764312744141, + "learning_rate": 4.343420334632204e-06, + "loss": 0.6326, + "step": 1786 + }, + { + "epoch": 1.4795692987299836, + "grad_norm": 0.9101755023002625, + "learning_rate": 4.342677325909798e-06, + "loss": 0.6319, + "step": 1787 + }, + { + "epoch": 1.4803975704030923, + "grad_norm": 0.9417787790298462, + "learning_rate": 4.341933960646505e-06, + "loss": 0.6147, + "step": 1788 + }, + { + "epoch": 1.481225842076201, + "grad_norm": 0.9138630628585815, + "learning_rate": 4.341190238986162e-06, + "loss": 0.6256, + "step": 1789 + }, + { + "epoch": 1.4820541137493097, + "grad_norm": 0.9004842638969421, + "learning_rate": 4.340446161072671e-06, + "loss": 0.6443, + "step": 1790 + }, + { + "epoch": 1.4828823854224185, + "grad_norm": 0.9181780219078064, + "learning_rate": 4.339701727050004e-06, + "loss": 0.6408, + "step": 1791 + }, + { + "epoch": 1.4837106570955274, + "grad_norm": 0.880736768245697, + "learning_rate": 4.338956937062203e-06, + "loss": 0.6436, + "step": 1792 + }, + { + "epoch": 1.4845389287686361, + "grad_norm": 0.9168913960456848, + "learning_rate": 4.338211791253376e-06, + "loss": 0.6391, + "step": 1793 + }, + { + "epoch": 1.4853672004417449, + "grad_norm": 0.942925751209259, + "learning_rate": 4.337466289767704e-06, + "loss": 0.6303, + "step": 1794 + }, + { + "epoch": 1.4861954721148538, + "grad_norm": 0.8768817186355591, + "learning_rate": 4.336720432749433e-06, + "loss": 0.6239, + "step": 1795 + }, + { + "epoch": 1.4870237437879625, + "grad_norm": 0.9388988614082336, + "learning_rate": 4.33597422034288e-06, + "loss": 0.6441, + "step": 1796 + }, + { + "epoch": 1.4878520154610713, + "grad_norm": 0.9257978200912476, + "learning_rate": 4.3352276526924295e-06, + "loss": 0.6373, + "step": 1797 + }, + { + "epoch": 1.48868028713418, + "grad_norm": 0.9225389957427979, + "learning_rate": 4.334480729942537e-06, + "loss": 0.6555, + "step": 1798 + }, + { + "epoch": 1.4895085588072887, + "grad_norm": 0.8992138504981995, + "learning_rate": 4.3337334522377235e-06, + "loss": 0.6375, + "step": 1799 + }, + { + "epoch": 1.4903368304803974, + "grad_norm": 0.9207738041877747, + "learning_rate": 4.332985819722582e-06, + "loss": 0.6439, + "step": 1800 + }, + { + "epoch": 1.4911651021535064, + "grad_norm": 0.8866941332817078, + "learning_rate": 4.3322378325417705e-06, + "loss": 0.6326, + "step": 1801 + }, + { + "epoch": 1.4919933738266151, + "grad_norm": 0.9069377779960632, + "learning_rate": 4.3314894908400175e-06, + "loss": 0.6193, + "step": 1802 + }, + { + "epoch": 1.4928216454997238, + "grad_norm": 0.8918569087982178, + "learning_rate": 4.3307407947621216e-06, + "loss": 0.6374, + "step": 1803 + }, + { + "epoch": 1.4936499171728328, + "grad_norm": 0.8993483781814575, + "learning_rate": 4.329991744452948e-06, + "loss": 0.6251, + "step": 1804 + }, + { + "epoch": 1.4944781888459415, + "grad_norm": 0.918445885181427, + "learning_rate": 4.32924234005743e-06, + "loss": 0.6494, + "step": 1805 + }, + { + "epoch": 1.4953064605190503, + "grad_norm": 0.9245757460594177, + "learning_rate": 4.328492581720573e-06, + "loss": 0.6172, + "step": 1806 + }, + { + "epoch": 1.496134732192159, + "grad_norm": 0.9166370034217834, + "learning_rate": 4.327742469587445e-06, + "loss": 0.622, + "step": 1807 + }, + { + "epoch": 1.4969630038652677, + "grad_norm": 0.9083957076072693, + "learning_rate": 4.326992003803187e-06, + "loss": 0.6205, + "step": 1808 + }, + { + "epoch": 1.4977912755383767, + "grad_norm": 0.8895595073699951, + "learning_rate": 4.326241184513008e-06, + "loss": 0.6245, + "step": 1809 + }, + { + "epoch": 1.4986195472114854, + "grad_norm": 0.9845629334449768, + "learning_rate": 4.325490011862184e-06, + "loss": 0.6163, + "step": 1810 + }, + { + "epoch": 1.499447818884594, + "grad_norm": 0.9501281976699829, + "learning_rate": 4.324738485996058e-06, + "loss": 0.6218, + "step": 1811 + }, + { + "epoch": 1.500276090557703, + "grad_norm": 0.9685792922973633, + "learning_rate": 4.323986607060047e-06, + "loss": 0.634, + "step": 1812 + }, + { + "epoch": 1.5011043622308118, + "grad_norm": 0.8959722518920898, + "learning_rate": 4.32323437519963e-06, + "loss": 0.6316, + "step": 1813 + }, + { + "epoch": 1.5019326339039205, + "grad_norm": 0.914463222026825, + "learning_rate": 4.322481790560356e-06, + "loss": 0.6258, + "step": 1814 + }, + { + "epoch": 1.5027609055770292, + "grad_norm": 0.9052320718765259, + "learning_rate": 4.321728853287846e-06, + "loss": 0.6345, + "step": 1815 + }, + { + "epoch": 1.503589177250138, + "grad_norm": 0.9306650161743164, + "learning_rate": 4.320975563527785e-06, + "loss": 0.5976, + "step": 1816 + }, + { + "epoch": 1.5044174489232467, + "grad_norm": 0.8930769562721252, + "learning_rate": 4.320221921425926e-06, + "loss": 0.6427, + "step": 1817 + }, + { + "epoch": 1.5052457205963556, + "grad_norm": 0.9205043911933899, + "learning_rate": 4.3194679271280935e-06, + "loss": 0.617, + "step": 1818 + }, + { + "epoch": 1.5060739922694644, + "grad_norm": 1.0371367931365967, + "learning_rate": 4.318713580780179e-06, + "loss": 0.6301, + "step": 1819 + }, + { + "epoch": 1.5069022639425733, + "grad_norm": 0.9442588686943054, + "learning_rate": 4.317958882528139e-06, + "loss": 0.6252, + "step": 1820 + }, + { + "epoch": 1.507730535615682, + "grad_norm": 0.9214493036270142, + "learning_rate": 4.317203832518003e-06, + "loss": 0.6326, + "step": 1821 + }, + { + "epoch": 1.5085588072887908, + "grad_norm": 0.9270440340042114, + "learning_rate": 4.316448430895866e-06, + "loss": 0.6377, + "step": 1822 + }, + { + "epoch": 1.5093870789618995, + "grad_norm": 0.8822277188301086, + "learning_rate": 4.315692677807889e-06, + "loss": 0.6277, + "step": 1823 + }, + { + "epoch": 1.5102153506350082, + "grad_norm": 0.8926153779029846, + "learning_rate": 4.314936573400305e-06, + "loss": 0.6439, + "step": 1824 + }, + { + "epoch": 1.511043622308117, + "grad_norm": 0.9214072227478027, + "learning_rate": 4.314180117819412e-06, + "loss": 0.6519, + "step": 1825 + }, + { + "epoch": 1.5118718939812257, + "grad_norm": 0.881522536277771, + "learning_rate": 4.313423311211578e-06, + "loss": 0.6213, + "step": 1826 + }, + { + "epoch": 1.5127001656543346, + "grad_norm": 0.8960263729095459, + "learning_rate": 4.312666153723238e-06, + "loss": 0.6592, + "step": 1827 + }, + { + "epoch": 1.5135284373274434, + "grad_norm": 0.9176173210144043, + "learning_rate": 4.3119086455008956e-06, + "loss": 0.6569, + "step": 1828 + }, + { + "epoch": 1.5143567090005523, + "grad_norm": 0.9312153458595276, + "learning_rate": 4.311150786691119e-06, + "loss": 0.6324, + "step": 1829 + }, + { + "epoch": 1.515184980673661, + "grad_norm": 0.9202299118041992, + "learning_rate": 4.310392577440549e-06, + "loss": 0.6381, + "step": 1830 + }, + { + "epoch": 1.5160132523467698, + "grad_norm": 0.8895224928855896, + "learning_rate": 4.309634017895893e-06, + "loss": 0.6491, + "step": 1831 + }, + { + "epoch": 1.5168415240198785, + "grad_norm": 0.9424769878387451, + "learning_rate": 4.308875108203922e-06, + "loss": 0.6342, + "step": 1832 + }, + { + "epoch": 1.5176697956929872, + "grad_norm": 0.8676947951316833, + "learning_rate": 4.30811584851148e-06, + "loss": 0.6287, + "step": 1833 + }, + { + "epoch": 1.518498067366096, + "grad_norm": 0.8764673471450806, + "learning_rate": 4.3073562389654764e-06, + "loss": 0.6123, + "step": 1834 + }, + { + "epoch": 1.5193263390392049, + "grad_norm": 0.9222663640975952, + "learning_rate": 4.306596279712889e-06, + "loss": 0.6433, + "step": 1835 + }, + { + "epoch": 1.5201546107123136, + "grad_norm": 0.8839907050132751, + "learning_rate": 4.305835970900762e-06, + "loss": 0.6416, + "step": 1836 + }, + { + "epoch": 1.5209828823854226, + "grad_norm": 0.9095149636268616, + "learning_rate": 4.305075312676209e-06, + "loss": 0.6417, + "step": 1837 + }, + { + "epoch": 1.5218111540585313, + "grad_norm": 0.8718581199645996, + "learning_rate": 4.304314305186409e-06, + "loss": 0.6174, + "step": 1838 + }, + { + "epoch": 1.52263942573164, + "grad_norm": 0.9424081444740295, + "learning_rate": 4.303552948578611e-06, + "loss": 0.6064, + "step": 1839 + }, + { + "epoch": 1.5234676974047487, + "grad_norm": 0.9170297384262085, + "learning_rate": 4.30279124300013e-06, + "loss": 0.6078, + "step": 1840 + }, + { + "epoch": 1.5242959690778575, + "grad_norm": 0.8814254999160767, + "learning_rate": 4.302029188598348e-06, + "loss": 0.6194, + "step": 1841 + }, + { + "epoch": 1.5251242407509662, + "grad_norm": 0.9248694181442261, + "learning_rate": 4.301266785520718e-06, + "loss": 0.6452, + "step": 1842 + }, + { + "epoch": 1.5259525124240751, + "grad_norm": 0.8757627010345459, + "learning_rate": 4.300504033914754e-06, + "loss": 0.6486, + "step": 1843 + }, + { + "epoch": 1.5267807840971839, + "grad_norm": 0.9229605793952942, + "learning_rate": 4.299740933928045e-06, + "loss": 0.6363, + "step": 1844 + }, + { + "epoch": 1.5276090557702926, + "grad_norm": 0.9192973971366882, + "learning_rate": 4.298977485708243e-06, + "loss": 0.6383, + "step": 1845 + }, + { + "epoch": 1.5284373274434015, + "grad_norm": 0.9071361422538757, + "learning_rate": 4.2982136894030655e-06, + "loss": 0.6541, + "step": 1846 + }, + { + "epoch": 1.5292655991165103, + "grad_norm": 0.9067169427871704, + "learning_rate": 4.297449545160303e-06, + "loss": 0.6317, + "step": 1847 + }, + { + "epoch": 1.530093870789619, + "grad_norm": 0.9047722816467285, + "learning_rate": 4.2966850531278086e-06, + "loss": 0.6365, + "step": 1848 + }, + { + "epoch": 1.5309221424627277, + "grad_norm": 0.8948541879653931, + "learning_rate": 4.295920213453504e-06, + "loss": 0.6268, + "step": 1849 + }, + { + "epoch": 1.5317504141358365, + "grad_norm": 0.9061904549598694, + "learning_rate": 4.295155026285382e-06, + "loss": 0.6271, + "step": 1850 + }, + { + "epoch": 1.5325786858089452, + "grad_norm": 0.9092041850090027, + "learning_rate": 4.2943894917714946e-06, + "loss": 0.6148, + "step": 1851 + }, + { + "epoch": 1.5334069574820541, + "grad_norm": 0.932331919670105, + "learning_rate": 4.293623610059968e-06, + "loss": 0.647, + "step": 1852 + }, + { + "epoch": 1.5342352291551629, + "grad_norm": 0.9136486649513245, + "learning_rate": 4.292857381298993e-06, + "loss": 0.6326, + "step": 1853 + }, + { + "epoch": 1.5350635008282718, + "grad_norm": 0.9760140180587769, + "learning_rate": 4.292090805636827e-06, + "loss": 0.6324, + "step": 1854 + }, + { + "epoch": 1.5358917725013805, + "grad_norm": 0.9075965881347656, + "learning_rate": 4.291323883221797e-06, + "loss": 0.6194, + "step": 1855 + }, + { + "epoch": 1.5367200441744893, + "grad_norm": 0.8804707527160645, + "learning_rate": 4.290556614202292e-06, + "loss": 0.6246, + "step": 1856 + }, + { + "epoch": 1.537548315847598, + "grad_norm": 0.9123021960258484, + "learning_rate": 4.2897889987267755e-06, + "loss": 0.6156, + "step": 1857 + }, + { + "epoch": 1.5383765875207067, + "grad_norm": 0.9396944642066956, + "learning_rate": 4.289021036943771e-06, + "loss": 0.6268, + "step": 1858 + }, + { + "epoch": 1.5392048591938154, + "grad_norm": 0.858566164970398, + "learning_rate": 4.288252729001874e-06, + "loss": 0.6237, + "step": 1859 + }, + { + "epoch": 1.5400331308669244, + "grad_norm": 0.899341344833374, + "learning_rate": 4.287484075049743e-06, + "loss": 0.5973, + "step": 1860 + }, + { + "epoch": 1.5408614025400331, + "grad_norm": 0.9021979570388794, + "learning_rate": 4.286715075236106e-06, + "loss": 0.6164, + "step": 1861 + }, + { + "epoch": 1.541689674213142, + "grad_norm": 0.9693799018859863, + "learning_rate": 4.285945729709757e-06, + "loss": 0.6409, + "step": 1862 + }, + { + "epoch": 1.5425179458862508, + "grad_norm": 0.9177312254905701, + "learning_rate": 4.285176038619559e-06, + "loss": 0.6276, + "step": 1863 + }, + { + "epoch": 1.5433462175593595, + "grad_norm": 0.9382041692733765, + "learning_rate": 4.284406002114438e-06, + "loss": 0.643, + "step": 1864 + }, + { + "epoch": 1.5441744892324683, + "grad_norm": 0.8778985142707825, + "learning_rate": 4.28363562034339e-06, + "loss": 0.6119, + "step": 1865 + }, + { + "epoch": 1.545002760905577, + "grad_norm": 0.8890271782875061, + "learning_rate": 4.282864893455476e-06, + "loss": 0.6388, + "step": 1866 + }, + { + "epoch": 1.5458310325786857, + "grad_norm": 0.9081838130950928, + "learning_rate": 4.282093821599824e-06, + "loss": 0.6464, + "step": 1867 + }, + { + "epoch": 1.5466593042517944, + "grad_norm": 0.8805630803108215, + "learning_rate": 4.281322404925631e-06, + "loss": 0.6533, + "step": 1868 + }, + { + "epoch": 1.5474875759249034, + "grad_norm": 0.8817034363746643, + "learning_rate": 4.280550643582158e-06, + "loss": 0.643, + "step": 1869 + }, + { + "epoch": 1.548315847598012, + "grad_norm": 0.9003176689147949, + "learning_rate": 4.279778537718733e-06, + "loss": 0.6322, + "step": 1870 + }, + { + "epoch": 1.549144119271121, + "grad_norm": 0.8982899785041809, + "learning_rate": 4.2790060874847525e-06, + "loss": 0.6456, + "step": 1871 + }, + { + "epoch": 1.5499723909442298, + "grad_norm": 0.8730716109275818, + "learning_rate": 4.278233293029677e-06, + "loss": 0.6428, + "step": 1872 + }, + { + "epoch": 1.5508006626173385, + "grad_norm": 0.9135062098503113, + "learning_rate": 4.277460154503036e-06, + "loss": 0.645, + "step": 1873 + }, + { + "epoch": 1.5516289342904472, + "grad_norm": 0.9091652631759644, + "learning_rate": 4.276686672054424e-06, + "loss": 0.6256, + "step": 1874 + }, + { + "epoch": 1.552457205963556, + "grad_norm": 0.9090970754623413, + "learning_rate": 4.275912845833504e-06, + "loss": 0.6404, + "step": 1875 + }, + { + "epoch": 1.5532854776366647, + "grad_norm": 0.91478031873703, + "learning_rate": 4.275138675990003e-06, + "loss": 0.6498, + "step": 1876 + }, + { + "epoch": 1.5541137493097736, + "grad_norm": 0.8979254961013794, + "learning_rate": 4.274364162673715e-06, + "loss": 0.6275, + "step": 1877 + }, + { + "epoch": 1.5549420209828824, + "grad_norm": 0.92198646068573, + "learning_rate": 4.273589306034503e-06, + "loss": 0.6585, + "step": 1878 + }, + { + "epoch": 1.5557702926559913, + "grad_norm": 0.9540958404541016, + "learning_rate": 4.272814106222293e-06, + "loss": 0.6111, + "step": 1879 + }, + { + "epoch": 1.5565985643291, + "grad_norm": 0.8989667892456055, + "learning_rate": 4.272038563387079e-06, + "loss": 0.6255, + "step": 1880 + }, + { + "epoch": 1.5574268360022088, + "grad_norm": 0.9551357626914978, + "learning_rate": 4.2712626776789224e-06, + "loss": 0.6164, + "step": 1881 + }, + { + "epoch": 1.5582551076753175, + "grad_norm": 0.9278994798660278, + "learning_rate": 4.270486449247949e-06, + "loss": 0.6376, + "step": 1882 + }, + { + "epoch": 1.5590833793484262, + "grad_norm": 0.927180826663971, + "learning_rate": 4.269709878244352e-06, + "loss": 0.6602, + "step": 1883 + }, + { + "epoch": 1.559911651021535, + "grad_norm": 0.932837724685669, + "learning_rate": 4.268932964818389e-06, + "loss": 0.6229, + "step": 1884 + }, + { + "epoch": 1.5607399226946437, + "grad_norm": 0.895219087600708, + "learning_rate": 4.26815570912039e-06, + "loss": 0.6192, + "step": 1885 + }, + { + "epoch": 1.5615681943677526, + "grad_norm": 0.9075414538383484, + "learning_rate": 4.2673781113007415e-06, + "loss": 0.6418, + "step": 1886 + }, + { + "epoch": 1.5623964660408614, + "grad_norm": 0.9130772948265076, + "learning_rate": 4.266600171509905e-06, + "loss": 0.6402, + "step": 1887 + }, + { + "epoch": 1.5632247377139703, + "grad_norm": 0.8892185688018799, + "learning_rate": 4.265821889898402e-06, + "loss": 0.6213, + "step": 1888 + }, + { + "epoch": 1.564053009387079, + "grad_norm": 0.8853991031646729, + "learning_rate": 4.265043266616825e-06, + "loss": 0.6357, + "step": 1889 + }, + { + "epoch": 1.5648812810601878, + "grad_norm": 0.9273191690444946, + "learning_rate": 4.26426430181583e-06, + "loss": 0.6311, + "step": 1890 + }, + { + "epoch": 1.5657095527332965, + "grad_norm": 0.8890771865844727, + "learning_rate": 4.263484995646138e-06, + "loss": 0.6053, + "step": 1891 + }, + { + "epoch": 1.5665378244064052, + "grad_norm": 0.9022220373153687, + "learning_rate": 4.262705348258538e-06, + "loss": 0.6256, + "step": 1892 + }, + { + "epoch": 1.567366096079514, + "grad_norm": 0.9201255440711975, + "learning_rate": 4.261925359803886e-06, + "loss": 0.6302, + "step": 1893 + }, + { + "epoch": 1.5681943677526229, + "grad_norm": 0.9675379991531372, + "learning_rate": 4.261145030433101e-06, + "loss": 0.602, + "step": 1894 + }, + { + "epoch": 1.5690226394257316, + "grad_norm": 0.9039328098297119, + "learning_rate": 4.26036436029717e-06, + "loss": 0.6477, + "step": 1895 + }, + { + "epoch": 1.5698509110988406, + "grad_norm": 0.911361813545227, + "learning_rate": 4.259583349547144e-06, + "loss": 0.6211, + "step": 1896 + }, + { + "epoch": 1.5706791827719493, + "grad_norm": 0.8901238441467285, + "learning_rate": 4.258801998334143e-06, + "loss": 0.6442, + "step": 1897 + }, + { + "epoch": 1.571507454445058, + "grad_norm": 0.9290407299995422, + "learning_rate": 4.258020306809351e-06, + "loss": 0.6382, + "step": 1898 + }, + { + "epoch": 1.5723357261181667, + "grad_norm": 0.9179307222366333, + "learning_rate": 4.257238275124017e-06, + "loss": 0.6513, + "step": 1899 + }, + { + "epoch": 1.5731639977912755, + "grad_norm": 0.9212455153465271, + "learning_rate": 4.256455903429459e-06, + "loss": 0.6257, + "step": 1900 + }, + { + "epoch": 1.5739922694643842, + "grad_norm": 0.9126132130622864, + "learning_rate": 4.2556731918770565e-06, + "loss": 0.6384, + "step": 1901 + }, + { + "epoch": 1.5748205411374931, + "grad_norm": 0.8735355734825134, + "learning_rate": 4.254890140618257e-06, + "loss": 0.6031, + "step": 1902 + }, + { + "epoch": 1.5756488128106019, + "grad_norm": 0.9375724196434021, + "learning_rate": 4.254106749804575e-06, + "loss": 0.6333, + "step": 1903 + }, + { + "epoch": 1.5764770844837108, + "grad_norm": 0.9451910853385925, + "learning_rate": 4.253323019587588e-06, + "loss": 0.6186, + "step": 1904 + }, + { + "epoch": 1.5773053561568195, + "grad_norm": 0.9468234777450562, + "learning_rate": 4.25253895011894e-06, + "loss": 0.6023, + "step": 1905 + }, + { + "epoch": 1.5781336278299283, + "grad_norm": 0.8941681385040283, + "learning_rate": 4.251754541550343e-06, + "loss": 0.634, + "step": 1906 + }, + { + "epoch": 1.578961899503037, + "grad_norm": 0.9508136510848999, + "learning_rate": 4.250969794033573e-06, + "loss": 0.6346, + "step": 1907 + }, + { + "epoch": 1.5797901711761457, + "grad_norm": 0.8964453935623169, + "learning_rate": 4.2501847077204685e-06, + "loss": 0.6298, + "step": 1908 + }, + { + "epoch": 1.5806184428492545, + "grad_norm": 0.8885358572006226, + "learning_rate": 4.249399282762938e-06, + "loss": 0.6319, + "step": 1909 + }, + { + "epoch": 1.5814467145223632, + "grad_norm": 0.8929858207702637, + "learning_rate": 4.248613519312955e-06, + "loss": 0.6272, + "step": 1910 + }, + { + "epoch": 1.5822749861954721, + "grad_norm": 0.9664214849472046, + "learning_rate": 4.247827417522557e-06, + "loss": 0.6336, + "step": 1911 + }, + { + "epoch": 1.5831032578685809, + "grad_norm": 0.890846312046051, + "learning_rate": 4.247040977543845e-06, + "loss": 0.6242, + "step": 1912 + }, + { + "epoch": 1.5839315295416898, + "grad_norm": 0.8973182439804077, + "learning_rate": 4.246254199528991e-06, + "loss": 0.6336, + "step": 1913 + }, + { + "epoch": 1.5847598012147985, + "grad_norm": 0.9789074659347534, + "learning_rate": 4.245467083630227e-06, + "loss": 0.6191, + "step": 1914 + }, + { + "epoch": 1.5855880728879073, + "grad_norm": 0.9195513129234314, + "learning_rate": 4.244679629999853e-06, + "loss": 0.6333, + "step": 1915 + }, + { + "epoch": 1.586416344561016, + "grad_norm": 0.950316846370697, + "learning_rate": 4.243891838790235e-06, + "loss": 0.635, + "step": 1916 + }, + { + "epoch": 1.5872446162341247, + "grad_norm": 0.8971677422523499, + "learning_rate": 4.243103710153802e-06, + "loss": 0.6352, + "step": 1917 + }, + { + "epoch": 1.5880728879072334, + "grad_norm": 0.9198834896087646, + "learning_rate": 4.24231524424305e-06, + "loss": 0.6252, + "step": 1918 + }, + { + "epoch": 1.5889011595803424, + "grad_norm": 0.9535729885101318, + "learning_rate": 4.241526441210541e-06, + "loss": 0.6177, + "step": 1919 + }, + { + "epoch": 1.5897294312534511, + "grad_norm": 0.8910471200942993, + "learning_rate": 4.240737301208899e-06, + "loss": 0.6408, + "step": 1920 + }, + { + "epoch": 1.59055770292656, + "grad_norm": 0.862991452217102, + "learning_rate": 4.2399478243908164e-06, + "loss": 0.6221, + "step": 1921 + }, + { + "epoch": 1.5913859745996688, + "grad_norm": 0.8935141563415527, + "learning_rate": 4.239158010909049e-06, + "loss": 0.608, + "step": 1922 + }, + { + "epoch": 1.5922142462727775, + "grad_norm": 0.9177968502044678, + "learning_rate": 4.23836786091642e-06, + "loss": 0.6442, + "step": 1923 + }, + { + "epoch": 1.5930425179458862, + "grad_norm": 0.9157758951187134, + "learning_rate": 4.237577374565813e-06, + "loss": 0.6419, + "step": 1924 + }, + { + "epoch": 1.593870789618995, + "grad_norm": 0.9484630227088928, + "learning_rate": 4.236786552010182e-06, + "loss": 0.6733, + "step": 1925 + }, + { + "epoch": 1.5946990612921037, + "grad_norm": 0.9401251077651978, + "learning_rate": 4.235995393402544e-06, + "loss": 0.6325, + "step": 1926 + }, + { + "epoch": 1.5955273329652124, + "grad_norm": 0.9104607105255127, + "learning_rate": 4.23520389889598e-06, + "loss": 0.6313, + "step": 1927 + }, + { + "epoch": 1.5963556046383214, + "grad_norm": 0.9173511862754822, + "learning_rate": 4.234412068643637e-06, + "loss": 0.6044, + "step": 1928 + }, + { + "epoch": 1.59718387631143, + "grad_norm": 0.9189788699150085, + "learning_rate": 4.233619902798727e-06, + "loss": 0.6182, + "step": 1929 + }, + { + "epoch": 1.598012147984539, + "grad_norm": 0.910804808139801, + "learning_rate": 4.232827401514525e-06, + "loss": 0.6435, + "step": 1930 + }, + { + "epoch": 1.5988404196576478, + "grad_norm": 0.8986726403236389, + "learning_rate": 4.232034564944375e-06, + "loss": 0.6268, + "step": 1931 + }, + { + "epoch": 1.5996686913307565, + "grad_norm": 0.8782910108566284, + "learning_rate": 4.231241393241682e-06, + "loss": 0.6381, + "step": 1932 + }, + { + "epoch": 1.6004969630038652, + "grad_norm": 0.9176427721977234, + "learning_rate": 4.230447886559919e-06, + "loss": 0.6131, + "step": 1933 + }, + { + "epoch": 1.601325234676974, + "grad_norm": 0.8755510449409485, + "learning_rate": 4.22965404505262e-06, + "loss": 0.654, + "step": 1934 + }, + { + "epoch": 1.6021535063500827, + "grad_norm": 0.8644832968711853, + "learning_rate": 4.228859868873387e-06, + "loss": 0.627, + "step": 1935 + }, + { + "epoch": 1.6029817780231916, + "grad_norm": 0.9181885719299316, + "learning_rate": 4.228065358175886e-06, + "loss": 0.6062, + "step": 1936 + }, + { + "epoch": 1.6038100496963004, + "grad_norm": 0.9515942335128784, + "learning_rate": 4.227270513113846e-06, + "loss": 0.6262, + "step": 1937 + }, + { + "epoch": 1.6046383213694093, + "grad_norm": 0.9062270522117615, + "learning_rate": 4.226475333841063e-06, + "loss": 0.6139, + "step": 1938 + }, + { + "epoch": 1.605466593042518, + "grad_norm": 0.9040980339050293, + "learning_rate": 4.225679820511398e-06, + "loss": 0.64, + "step": 1939 + }, + { + "epoch": 1.6062948647156268, + "grad_norm": 0.8910935521125793, + "learning_rate": 4.224883973278775e-06, + "loss": 0.6239, + "step": 1940 + }, + { + "epoch": 1.6071231363887355, + "grad_norm": 0.9150440692901611, + "learning_rate": 4.224087792297181e-06, + "loss": 0.6283, + "step": 1941 + }, + { + "epoch": 1.6079514080618442, + "grad_norm": 0.9175113439559937, + "learning_rate": 4.223291277720671e-06, + "loss": 0.6498, + "step": 1942 + }, + { + "epoch": 1.608779679734953, + "grad_norm": 0.8720189332962036, + "learning_rate": 4.222494429703363e-06, + "loss": 0.623, + "step": 1943 + }, + { + "epoch": 1.609607951408062, + "grad_norm": 0.8868989944458008, + "learning_rate": 4.22169724839944e-06, + "loss": 0.6357, + "step": 1944 + }, + { + "epoch": 1.6104362230811706, + "grad_norm": 0.9084488153457642, + "learning_rate": 4.22089973396315e-06, + "loss": 0.6607, + "step": 1945 + }, + { + "epoch": 1.6112644947542794, + "grad_norm": 0.9061319231987, + "learning_rate": 4.220101886548802e-06, + "loss": 0.6309, + "step": 1946 + }, + { + "epoch": 1.6120927664273883, + "grad_norm": 0.895646870136261, + "learning_rate": 4.219303706310775e-06, + "loss": 0.6276, + "step": 1947 + }, + { + "epoch": 1.612921038100497, + "grad_norm": 0.8863298892974854, + "learning_rate": 4.218505193403506e-06, + "loss": 0.6444, + "step": 1948 + }, + { + "epoch": 1.6137493097736058, + "grad_norm": 0.9005767107009888, + "learning_rate": 4.2177063479815025e-06, + "loss": 0.6356, + "step": 1949 + }, + { + "epoch": 1.6145775814467145, + "grad_norm": 0.9094089865684509, + "learning_rate": 4.216907170199334e-06, + "loss": 0.6053, + "step": 1950 + }, + { + "epoch": 1.6154058531198232, + "grad_norm": 0.8869677782058716, + "learning_rate": 4.216107660211632e-06, + "loss": 0.6308, + "step": 1951 + }, + { + "epoch": 1.616234124792932, + "grad_norm": 0.9285000562667847, + "learning_rate": 4.215307818173095e-06, + "loss": 0.6188, + "step": 1952 + }, + { + "epoch": 1.6170623964660409, + "grad_norm": 0.9039749503135681, + "learning_rate": 4.214507644238485e-06, + "loss": 0.63, + "step": 1953 + }, + { + "epoch": 1.6178906681391496, + "grad_norm": 0.9202860593795776, + "learning_rate": 4.213707138562628e-06, + "loss": 0.632, + "step": 1954 + }, + { + "epoch": 1.6187189398122586, + "grad_norm": 0.8759778141975403, + "learning_rate": 4.212906301300415e-06, + "loss": 0.6253, + "step": 1955 + }, + { + "epoch": 1.6195472114853673, + "grad_norm": 0.8918818831443787, + "learning_rate": 4.212105132606801e-06, + "loss": 0.625, + "step": 1956 + }, + { + "epoch": 1.620375483158476, + "grad_norm": 0.9184549450874329, + "learning_rate": 4.211303632636804e-06, + "loss": 0.6293, + "step": 1957 + }, + { + "epoch": 1.6212037548315847, + "grad_norm": 0.8887206315994263, + "learning_rate": 4.210501801545506e-06, + "loss": 0.6118, + "step": 1958 + }, + { + "epoch": 1.6220320265046935, + "grad_norm": 0.9010420441627502, + "learning_rate": 4.209699639488054e-06, + "loss": 0.6209, + "step": 1959 + }, + { + "epoch": 1.6228602981778022, + "grad_norm": 0.8888309001922607, + "learning_rate": 4.20889714661966e-06, + "loss": 0.6301, + "step": 1960 + }, + { + "epoch": 1.6236885698509111, + "grad_norm": 0.9068322777748108, + "learning_rate": 4.208094323095597e-06, + "loss": 0.6234, + "step": 1961 + }, + { + "epoch": 1.6245168415240199, + "grad_norm": 0.8940340876579285, + "learning_rate": 4.207291169071207e-06, + "loss": 0.6076, + "step": 1962 + }, + { + "epoch": 1.6253451131971288, + "grad_norm": 0.907363772392273, + "learning_rate": 4.20648768470189e-06, + "loss": 0.6522, + "step": 1963 + }, + { + "epoch": 1.6261733848702375, + "grad_norm": 0.9104205965995789, + "learning_rate": 4.205683870143114e-06, + "loss": 0.6177, + "step": 1964 + }, + { + "epoch": 1.6270016565433463, + "grad_norm": 0.9051296710968018, + "learning_rate": 4.204879725550409e-06, + "loss": 0.6069, + "step": 1965 + }, + { + "epoch": 1.627829928216455, + "grad_norm": 0.9034865498542786, + "learning_rate": 4.20407525107937e-06, + "loss": 0.6099, + "step": 1966 + }, + { + "epoch": 1.6286581998895637, + "grad_norm": 0.9055590033531189, + "learning_rate": 4.203270446885655e-06, + "loss": 0.6226, + "step": 1967 + }, + { + "epoch": 1.6294864715626725, + "grad_norm": 0.897927463054657, + "learning_rate": 4.202465313124986e-06, + "loss": 0.6342, + "step": 1968 + }, + { + "epoch": 1.6303147432357812, + "grad_norm": 0.8995800018310547, + "learning_rate": 4.2016598499531495e-06, + "loss": 0.6408, + "step": 1969 + }, + { + "epoch": 1.6311430149088901, + "grad_norm": 0.8727264404296875, + "learning_rate": 4.200854057525995e-06, + "loss": 0.6263, + "step": 1970 + }, + { + "epoch": 1.6319712865819989, + "grad_norm": 0.9289891719818115, + "learning_rate": 4.200047935999435e-06, + "loss": 0.6351, + "step": 1971 + }, + { + "epoch": 1.6327995582551078, + "grad_norm": 0.8936645984649658, + "learning_rate": 4.199241485529446e-06, + "loss": 0.6205, + "step": 1972 + }, + { + "epoch": 1.6336278299282165, + "grad_norm": 0.9054183959960938, + "learning_rate": 4.198434706272072e-06, + "loss": 0.6101, + "step": 1973 + }, + { + "epoch": 1.6344561016013253, + "grad_norm": 0.9181169271469116, + "learning_rate": 4.1976275983834135e-06, + "loss": 0.6437, + "step": 1974 + }, + { + "epoch": 1.635284373274434, + "grad_norm": 0.9325777888298035, + "learning_rate": 4.1968201620196404e-06, + "loss": 0.6343, + "step": 1975 + }, + { + "epoch": 1.6361126449475427, + "grad_norm": 0.878094494342804, + "learning_rate": 4.196012397336983e-06, + "loss": 0.6298, + "step": 1976 + }, + { + "epoch": 1.6369409166206514, + "grad_norm": 0.9336429238319397, + "learning_rate": 4.195204304491737e-06, + "loss": 0.6232, + "step": 1977 + }, + { + "epoch": 1.6377691882937604, + "grad_norm": 0.9159499406814575, + "learning_rate": 4.19439588364026e-06, + "loss": 0.621, + "step": 1978 + }, + { + "epoch": 1.6385974599668691, + "grad_norm": 0.9075011610984802, + "learning_rate": 4.193587134938975e-06, + "loss": 0.6375, + "step": 1979 + }, + { + "epoch": 1.639425731639978, + "grad_norm": 0.8912638425827026, + "learning_rate": 4.192778058544367e-06, + "loss": 0.6494, + "step": 1980 + }, + { + "epoch": 1.6402540033130868, + "grad_norm": 0.9125227928161621, + "learning_rate": 4.191968654612984e-06, + "loss": 0.6238, + "step": 1981 + }, + { + "epoch": 1.6410822749861955, + "grad_norm": 0.906326174736023, + "learning_rate": 4.191158923301439e-06, + "loss": 0.5904, + "step": 1982 + }, + { + "epoch": 1.6419105466593042, + "grad_norm": 0.8858581781387329, + "learning_rate": 4.190348864766407e-06, + "loss": 0.6222, + "step": 1983 + }, + { + "epoch": 1.642738818332413, + "grad_norm": 0.864389955997467, + "learning_rate": 4.189538479164627e-06, + "loss": 0.6218, + "step": 1984 + }, + { + "epoch": 1.6435670900055217, + "grad_norm": 0.8983911871910095, + "learning_rate": 4.188727766652901e-06, + "loss": 0.6511, + "step": 1985 + }, + { + "epoch": 1.6443953616786304, + "grad_norm": 0.9633772373199463, + "learning_rate": 4.187916727388094e-06, + "loss": 0.623, + "step": 1986 + }, + { + "epoch": 1.6452236333517394, + "grad_norm": 0.9191408753395081, + "learning_rate": 4.187105361527135e-06, + "loss": 0.6245, + "step": 1987 + }, + { + "epoch": 1.646051905024848, + "grad_norm": 0.907792866230011, + "learning_rate": 4.186293669227016e-06, + "loss": 0.5994, + "step": 1988 + }, + { + "epoch": 1.646880176697957, + "grad_norm": 0.906218409538269, + "learning_rate": 4.18548165064479e-06, + "loss": 0.653, + "step": 1989 + }, + { + "epoch": 1.6477084483710658, + "grad_norm": 0.9408275485038757, + "learning_rate": 4.184669305937577e-06, + "loss": 0.5978, + "step": 1990 + }, + { + "epoch": 1.6485367200441745, + "grad_norm": 0.9278435111045837, + "learning_rate": 4.183856635262558e-06, + "loss": 0.6167, + "step": 1991 + }, + { + "epoch": 1.6493649917172832, + "grad_norm": 0.91252601146698, + "learning_rate": 4.183043638776978e-06, + "loss": 0.6229, + "step": 1992 + }, + { + "epoch": 1.650193263390392, + "grad_norm": 0.9178313612937927, + "learning_rate": 4.182230316638142e-06, + "loss": 0.6318, + "step": 1993 + }, + { + "epoch": 1.6510215350635007, + "grad_norm": 0.9056826233863831, + "learning_rate": 4.181416669003421e-06, + "loss": 0.5932, + "step": 1994 + }, + { + "epoch": 1.6518498067366096, + "grad_norm": 0.9139112830162048, + "learning_rate": 4.180602696030248e-06, + "loss": 0.6407, + "step": 1995 + }, + { + "epoch": 1.6526780784097184, + "grad_norm": 0.9195094108581543, + "learning_rate": 4.179788397876121e-06, + "loss": 0.6371, + "step": 1996 + }, + { + "epoch": 1.6535063500828273, + "grad_norm": 0.9221140146255493, + "learning_rate": 4.178973774698597e-06, + "loss": 0.611, + "step": 1997 + }, + { + "epoch": 1.654334621755936, + "grad_norm": 0.9180639386177063, + "learning_rate": 4.178158826655299e-06, + "loss": 0.6138, + "step": 1998 + }, + { + "epoch": 1.6551628934290448, + "grad_norm": 0.9255500435829163, + "learning_rate": 4.177343553903912e-06, + "loss": 0.6425, + "step": 1999 + }, + { + "epoch": 1.6559911651021535, + "grad_norm": 0.9378511905670166, + "learning_rate": 4.176527956602182e-06, + "loss": 0.6357, + "step": 2000 + }, + { + "epoch": 1.6568194367752622, + "grad_norm": 0.9212007522583008, + "learning_rate": 4.175712034907922e-06, + "loss": 0.6307, + "step": 2001 + }, + { + "epoch": 1.657647708448371, + "grad_norm": 0.9185839891433716, + "learning_rate": 4.174895788979004e-06, + "loss": 0.6173, + "step": 2002 + }, + { + "epoch": 1.65847598012148, + "grad_norm": 0.9658777117729187, + "learning_rate": 4.174079218973364e-06, + "loss": 0.622, + "step": 2003 + }, + { + "epoch": 1.6593042517945886, + "grad_norm": 0.9310017824172974, + "learning_rate": 4.173262325049e-06, + "loss": 0.6407, + "step": 2004 + }, + { + "epoch": 1.6601325234676974, + "grad_norm": 0.9048784375190735, + "learning_rate": 4.172445107363974e-06, + "loss": 0.6557, + "step": 2005 + }, + { + "epoch": 1.6609607951408063, + "grad_norm": 0.8820594549179077, + "learning_rate": 4.17162756607641e-06, + "loss": 0.6077, + "step": 2006 + }, + { + "epoch": 1.661789066813915, + "grad_norm": 0.8961437344551086, + "learning_rate": 4.170809701344496e-06, + "loss": 0.6151, + "step": 2007 + }, + { + "epoch": 1.6626173384870238, + "grad_norm": 0.9773615598678589, + "learning_rate": 4.169991513326479e-06, + "loss": 0.6339, + "step": 2008 + }, + { + "epoch": 1.6634456101601325, + "grad_norm": 0.9309290647506714, + "learning_rate": 4.169173002180672e-06, + "loss": 0.6235, + "step": 2009 + }, + { + "epoch": 1.6642738818332412, + "grad_norm": 1.0016266107559204, + "learning_rate": 4.168354168065448e-06, + "loss": 0.6355, + "step": 2010 + }, + { + "epoch": 1.66510215350635, + "grad_norm": 0.8820549249649048, + "learning_rate": 4.167535011139246e-06, + "loss": 0.6258, + "step": 2011 + }, + { + "epoch": 1.6659304251794589, + "grad_norm": 0.9030908942222595, + "learning_rate": 4.166715531560564e-06, + "loss": 0.6217, + "step": 2012 + }, + { + "epoch": 1.6667586968525676, + "grad_norm": 0.983755350112915, + "learning_rate": 4.165895729487962e-06, + "loss": 0.6281, + "step": 2013 + }, + { + "epoch": 1.6675869685256766, + "grad_norm": 0.948001503944397, + "learning_rate": 4.165075605080066e-06, + "loss": 0.6242, + "step": 2014 + }, + { + "epoch": 1.6684152401987853, + "grad_norm": 0.9468982219696045, + "learning_rate": 4.164255158495563e-06, + "loss": 0.6563, + "step": 2015 + }, + { + "epoch": 1.669243511871894, + "grad_norm": 0.9012095332145691, + "learning_rate": 4.1634343898932e-06, + "loss": 0.618, + "step": 2016 + }, + { + "epoch": 1.6700717835450027, + "grad_norm": 0.9554924964904785, + "learning_rate": 4.162613299431789e-06, + "loss": 0.6306, + "step": 2017 + }, + { + "epoch": 1.6709000552181115, + "grad_norm": 0.9587515592575073, + "learning_rate": 4.161791887270203e-06, + "loss": 0.6305, + "step": 2018 + }, + { + "epoch": 1.6717283268912202, + "grad_norm": 0.8792864084243774, + "learning_rate": 4.160970153567379e-06, + "loss": 0.6181, + "step": 2019 + }, + { + "epoch": 1.6725565985643291, + "grad_norm": 0.8958392143249512, + "learning_rate": 4.160148098482313e-06, + "loss": 0.623, + "step": 2020 + }, + { + "epoch": 1.6733848702374379, + "grad_norm": 0.9501842260360718, + "learning_rate": 4.159325722174065e-06, + "loss": 0.6454, + "step": 2021 + }, + { + "epoch": 1.6742131419105468, + "grad_norm": 0.9107666015625, + "learning_rate": 4.158503024801759e-06, + "loss": 0.6237, + "step": 2022 + }, + { + "epoch": 1.6750414135836555, + "grad_norm": 0.9405720233917236, + "learning_rate": 4.157680006524577e-06, + "loss": 0.6423, + "step": 2023 + }, + { + "epoch": 1.6758696852567643, + "grad_norm": 0.9008631706237793, + "learning_rate": 4.1568566675017675e-06, + "loss": 0.6385, + "step": 2024 + }, + { + "epoch": 1.676697956929873, + "grad_norm": 0.9056677222251892, + "learning_rate": 4.156033007892637e-06, + "loss": 0.6225, + "step": 2025 + }, + { + "epoch": 1.6775262286029817, + "grad_norm": 0.899982750415802, + "learning_rate": 4.155209027856557e-06, + "loss": 0.6268, + "step": 2026 + }, + { + "epoch": 1.6783545002760905, + "grad_norm": 0.928506076335907, + "learning_rate": 4.15438472755296e-06, + "loss": 0.6343, + "step": 2027 + }, + { + "epoch": 1.6791827719491992, + "grad_norm": 0.8831804394721985, + "learning_rate": 4.15356010714134e-06, + "loss": 0.6404, + "step": 2028 + }, + { + "epoch": 1.6800110436223081, + "grad_norm": 0.8637033700942993, + "learning_rate": 4.152735166781253e-06, + "loss": 0.6305, + "step": 2029 + }, + { + "epoch": 1.6808393152954169, + "grad_norm": 0.9164902567863464, + "learning_rate": 4.151909906632319e-06, + "loss": 0.6171, + "step": 2030 + }, + { + "epoch": 1.6816675869685258, + "grad_norm": 0.9124466776847839, + "learning_rate": 4.151084326854216e-06, + "loss": 0.6446, + "step": 2031 + }, + { + "epoch": 1.6824958586416345, + "grad_norm": 0.9212199449539185, + "learning_rate": 4.150258427606687e-06, + "loss": 0.6074, + "step": 2032 + }, + { + "epoch": 1.6833241303147433, + "grad_norm": 0.9536449313163757, + "learning_rate": 4.149432209049536e-06, + "loss": 0.6149, + "step": 2033 + }, + { + "epoch": 1.684152401987852, + "grad_norm": 0.9232591390609741, + "learning_rate": 4.148605671342629e-06, + "loss": 0.6238, + "step": 2034 + }, + { + "epoch": 1.6849806736609607, + "grad_norm": 0.8844353556632996, + "learning_rate": 4.147778814645892e-06, + "loss": 0.6434, + "step": 2035 + }, + { + "epoch": 1.6858089453340694, + "grad_norm": 0.8500795960426331, + "learning_rate": 4.1469516391193155e-06, + "loss": 0.6304, + "step": 2036 + }, + { + "epoch": 1.6866372170071784, + "grad_norm": 0.8676083087921143, + "learning_rate": 4.146124144922949e-06, + "loss": 0.6158, + "step": 2037 + }, + { + "epoch": 1.6874654886802871, + "grad_norm": 0.8914940357208252, + "learning_rate": 4.1452963322169046e-06, + "loss": 0.6336, + "step": 2038 + }, + { + "epoch": 1.688293760353396, + "grad_norm": 0.9245527386665344, + "learning_rate": 4.144468201161358e-06, + "loss": 0.6385, + "step": 2039 + }, + { + "epoch": 1.6891220320265048, + "grad_norm": 0.8946509957313538, + "learning_rate": 4.143639751916544e-06, + "loss": 0.6243, + "step": 2040 + }, + { + "epoch": 1.6899503036996135, + "grad_norm": 0.882095217704773, + "learning_rate": 4.14281098464276e-06, + "loss": 0.6171, + "step": 2041 + }, + { + "epoch": 1.6907785753727222, + "grad_norm": 0.8981113433837891, + "learning_rate": 4.141981899500364e-06, + "loss": 0.6389, + "step": 2042 + }, + { + "epoch": 1.691606847045831, + "grad_norm": 0.928701639175415, + "learning_rate": 4.1411524966497775e-06, + "loss": 0.6488, + "step": 2043 + }, + { + "epoch": 1.6924351187189397, + "grad_norm": 0.8937084674835205, + "learning_rate": 4.140322776251481e-06, + "loss": 0.6415, + "step": 2044 + }, + { + "epoch": 1.6932633903920484, + "grad_norm": 0.8940730094909668, + "learning_rate": 4.1394927384660185e-06, + "loss": 0.6419, + "step": 2045 + }, + { + "epoch": 1.6940916620651574, + "grad_norm": 0.9067767262458801, + "learning_rate": 4.138662383453994e-06, + "loss": 0.608, + "step": 2046 + }, + { + "epoch": 1.694919933738266, + "grad_norm": 0.9123063087463379, + "learning_rate": 4.137831711376074e-06, + "loss": 0.633, + "step": 2047 + }, + { + "epoch": 1.695748205411375, + "grad_norm": 0.89399653673172, + "learning_rate": 4.137000722392985e-06, + "loss": 0.6338, + "step": 2048 + }, + { + "epoch": 1.6965764770844838, + "grad_norm": 0.9146568775177002, + "learning_rate": 4.136169416665517e-06, + "loss": 0.6214, + "step": 2049 + }, + { + "epoch": 1.6974047487575925, + "grad_norm": 0.9402440786361694, + "learning_rate": 4.135337794354518e-06, + "loss": 0.6266, + "step": 2050 + }, + { + "epoch": 1.6982330204307012, + "grad_norm": 0.8790574669837952, + "learning_rate": 4.134505855620902e-06, + "loss": 0.5989, + "step": 2051 + }, + { + "epoch": 1.69906129210381, + "grad_norm": 0.9421536326408386, + "learning_rate": 4.133673600625637e-06, + "loss": 0.611, + "step": 2052 + }, + { + "epoch": 1.6998895637769187, + "grad_norm": 0.9365546703338623, + "learning_rate": 4.132841029529762e-06, + "loss": 0.6201, + "step": 2053 + }, + { + "epoch": 1.7007178354500276, + "grad_norm": 0.9213638305664062, + "learning_rate": 4.132008142494367e-06, + "loss": 0.632, + "step": 2054 + }, + { + "epoch": 1.7015461071231364, + "grad_norm": 0.9204041957855225, + "learning_rate": 4.13117493968061e-06, + "loss": 0.6086, + "step": 2055 + }, + { + "epoch": 1.7023743787962453, + "grad_norm": 0.8972902297973633, + "learning_rate": 4.130341421249708e-06, + "loss": 0.656, + "step": 2056 + }, + { + "epoch": 1.703202650469354, + "grad_norm": 0.9163963794708252, + "learning_rate": 4.129507587362939e-06, + "loss": 0.623, + "step": 2057 + }, + { + "epoch": 1.7040309221424628, + "grad_norm": 0.9178527593612671, + "learning_rate": 4.128673438181642e-06, + "loss": 0.6327, + "step": 2058 + }, + { + "epoch": 1.7048591938155715, + "grad_norm": 0.8982755541801453, + "learning_rate": 4.127838973867217e-06, + "loss": 0.6165, + "step": 2059 + }, + { + "epoch": 1.7056874654886802, + "grad_norm": 0.8995897769927979, + "learning_rate": 4.127004194581125e-06, + "loss": 0.6204, + "step": 2060 + }, + { + "epoch": 1.706515737161789, + "grad_norm": 0.9148748517036438, + "learning_rate": 4.126169100484888e-06, + "loss": 0.6177, + "step": 2061 + }, + { + "epoch": 1.707344008834898, + "grad_norm": 0.9249931573867798, + "learning_rate": 4.12533369174009e-06, + "loss": 0.6113, + "step": 2062 + }, + { + "epoch": 1.7081722805080066, + "grad_norm": 0.9162423610687256, + "learning_rate": 4.124497968508372e-06, + "loss": 0.6119, + "step": 2063 + }, + { + "epoch": 1.7090005521811154, + "grad_norm": 0.8919503688812256, + "learning_rate": 4.123661930951441e-06, + "loss": 0.6194, + "step": 2064 + }, + { + "epoch": 1.7098288238542243, + "grad_norm": 0.8884986639022827, + "learning_rate": 4.122825579231063e-06, + "loss": 0.656, + "step": 2065 + }, + { + "epoch": 1.710657095527333, + "grad_norm": 0.9072293043136597, + "learning_rate": 4.121988913509062e-06, + "loss": 0.6351, + "step": 2066 + }, + { + "epoch": 1.7114853672004418, + "grad_norm": 0.9255547523498535, + "learning_rate": 4.121151933947327e-06, + "loss": 0.601, + "step": 2067 + }, + { + "epoch": 1.7123136388735505, + "grad_norm": 0.9141303300857544, + "learning_rate": 4.120314640707804e-06, + "loss": 0.6257, + "step": 2068 + }, + { + "epoch": 1.7131419105466592, + "grad_norm": 0.8791571855545044, + "learning_rate": 4.119477033952504e-06, + "loss": 0.6355, + "step": 2069 + }, + { + "epoch": 1.713970182219768, + "grad_norm": 0.8910380005836487, + "learning_rate": 4.118639113843493e-06, + "loss": 0.6405, + "step": 2070 + }, + { + "epoch": 1.7147984538928769, + "grad_norm": 0.9583408236503601, + "learning_rate": 4.117800880542904e-06, + "loss": 0.6408, + "step": 2071 + }, + { + "epoch": 1.7156267255659856, + "grad_norm": 0.9684842824935913, + "learning_rate": 4.116962334212924e-06, + "loss": 0.6155, + "step": 2072 + }, + { + "epoch": 1.7164549972390946, + "grad_norm": 0.8882633447647095, + "learning_rate": 4.116123475015806e-06, + "loss": 0.6332, + "step": 2073 + }, + { + "epoch": 1.7172832689122033, + "grad_norm": 0.8599907755851746, + "learning_rate": 4.1152843031138604e-06, + "loss": 0.6416, + "step": 2074 + }, + { + "epoch": 1.718111540585312, + "grad_norm": 0.9216133952140808, + "learning_rate": 4.114444818669461e-06, + "loss": 0.61, + "step": 2075 + }, + { + "epoch": 1.7189398122584207, + "grad_norm": 0.943729817867279, + "learning_rate": 4.113605021845037e-06, + "loss": 0.643, + "step": 2076 + }, + { + "epoch": 1.7197680839315295, + "grad_norm": 0.9158332943916321, + "learning_rate": 4.112764912803084e-06, + "loss": 0.6353, + "step": 2077 + }, + { + "epoch": 1.7205963556046382, + "grad_norm": 0.9036465287208557, + "learning_rate": 4.111924491706154e-06, + "loss": 0.6143, + "step": 2078 + }, + { + "epoch": 1.7214246272777471, + "grad_norm": 0.9264538884162903, + "learning_rate": 4.111083758716862e-06, + "loss": 0.6302, + "step": 2079 + }, + { + "epoch": 1.7222528989508559, + "grad_norm": 0.9309650659561157, + "learning_rate": 4.11024271399788e-06, + "loss": 0.6394, + "step": 2080 + }, + { + "epoch": 1.7230811706239648, + "grad_norm": 0.9261552095413208, + "learning_rate": 4.109401357711944e-06, + "loss": 0.6338, + "step": 2081 + }, + { + "epoch": 1.7239094422970735, + "grad_norm": 0.9027671217918396, + "learning_rate": 4.1085596900218476e-06, + "loss": 0.633, + "step": 2082 + }, + { + "epoch": 1.7247377139701823, + "grad_norm": 0.8899479508399963, + "learning_rate": 4.107717711090446e-06, + "loss": 0.6372, + "step": 2083 + }, + { + "epoch": 1.725565985643291, + "grad_norm": 0.9513740539550781, + "learning_rate": 4.106875421080654e-06, + "loss": 0.6057, + "step": 2084 + }, + { + "epoch": 1.7263942573163997, + "grad_norm": 0.8926359415054321, + "learning_rate": 4.106032820155447e-06, + "loss": 0.6219, + "step": 2085 + }, + { + "epoch": 1.7272225289895085, + "grad_norm": 0.8961042165756226, + "learning_rate": 4.105189908477862e-06, + "loss": 0.5926, + "step": 2086 + }, + { + "epoch": 1.7280508006626172, + "grad_norm": 0.8870651125907898, + "learning_rate": 4.104346686210992e-06, + "loss": 0.6298, + "step": 2087 + }, + { + "epoch": 1.7288790723357261, + "grad_norm": 0.9134538173675537, + "learning_rate": 4.103503153517993e-06, + "loss": 0.6544, + "step": 2088 + }, + { + "epoch": 1.7297073440088349, + "grad_norm": 0.900499165058136, + "learning_rate": 4.102659310562082e-06, + "loss": 0.6182, + "step": 2089 + }, + { + "epoch": 1.7305356156819438, + "grad_norm": 0.9146167039871216, + "learning_rate": 4.101815157506534e-06, + "loss": 0.6297, + "step": 2090 + }, + { + "epoch": 1.7313638873550525, + "grad_norm": 0.909670352935791, + "learning_rate": 4.100970694514687e-06, + "loss": 0.6272, + "step": 2091 + }, + { + "epoch": 1.7321921590281613, + "grad_norm": 1.127607822418213, + "learning_rate": 4.100125921749932e-06, + "loss": 0.6416, + "step": 2092 + }, + { + "epoch": 1.73302043070127, + "grad_norm": 0.8950872421264648, + "learning_rate": 4.099280839375729e-06, + "loss": 0.6411, + "step": 2093 + }, + { + "epoch": 1.7338487023743787, + "grad_norm": 0.9612913727760315, + "learning_rate": 4.098435447555591e-06, + "loss": 0.626, + "step": 2094 + }, + { + "epoch": 1.7346769740474874, + "grad_norm": 0.927638053894043, + "learning_rate": 4.097589746453094e-06, + "loss": 0.6197, + "step": 2095 + }, + { + "epoch": 1.7355052457205964, + "grad_norm": 0.9428266286849976, + "learning_rate": 4.096743736231875e-06, + "loss": 0.6407, + "step": 2096 + }, + { + "epoch": 1.7363335173937051, + "grad_norm": 0.8619174361228943, + "learning_rate": 4.095897417055627e-06, + "loss": 0.6386, + "step": 2097 + }, + { + "epoch": 1.737161789066814, + "grad_norm": 0.8943668007850647, + "learning_rate": 4.095050789088106e-06, + "loss": 0.6206, + "step": 2098 + }, + { + "epoch": 1.7379900607399228, + "grad_norm": 0.9190096259117126, + "learning_rate": 4.0942038524931245e-06, + "loss": 0.6395, + "step": 2099 + }, + { + "epoch": 1.7388183324130315, + "grad_norm": 0.9055671095848083, + "learning_rate": 4.09335660743456e-06, + "loss": 0.6384, + "step": 2100 + }, + { + "epoch": 1.7396466040861402, + "grad_norm": 0.9204139709472656, + "learning_rate": 4.092509054076345e-06, + "loss": 0.6438, + "step": 2101 + }, + { + "epoch": 1.740474875759249, + "grad_norm": 0.9572716355323792, + "learning_rate": 4.091661192582473e-06, + "loss": 0.6233, + "step": 2102 + }, + { + "epoch": 1.7413031474323577, + "grad_norm": 0.9479691982269287, + "learning_rate": 4.0908130231169975e-06, + "loss": 0.6284, + "step": 2103 + }, + { + "epoch": 1.7421314191054664, + "grad_norm": 0.9081270694732666, + "learning_rate": 4.089964545844032e-06, + "loss": 0.6284, + "step": 2104 + }, + { + "epoch": 1.7429596907785754, + "grad_norm": 0.9227046966552734, + "learning_rate": 4.0891157609277475e-06, + "loss": 0.6611, + "step": 2105 + }, + { + "epoch": 1.743787962451684, + "grad_norm": 0.9033496975898743, + "learning_rate": 4.088266668532377e-06, + "loss": 0.6122, + "step": 2106 + }, + { + "epoch": 1.744616234124793, + "grad_norm": 0.933539628982544, + "learning_rate": 4.087417268822211e-06, + "loss": 0.6167, + "step": 2107 + }, + { + "epoch": 1.7454445057979018, + "grad_norm": 0.9211486577987671, + "learning_rate": 4.086567561961602e-06, + "loss": 0.6364, + "step": 2108 + }, + { + "epoch": 1.7462727774710105, + "grad_norm": 0.8782284259796143, + "learning_rate": 4.08571754811496e-06, + "loss": 0.6164, + "step": 2109 + }, + { + "epoch": 1.7471010491441192, + "grad_norm": 0.8936499357223511, + "learning_rate": 4.084867227446753e-06, + "loss": 0.6155, + "step": 2110 + }, + { + "epoch": 1.747929320817228, + "grad_norm": 0.8931904435157776, + "learning_rate": 4.084016600121513e-06, + "loss": 0.6512, + "step": 2111 + }, + { + "epoch": 1.7487575924903367, + "grad_norm": 0.9070253372192383, + "learning_rate": 4.083165666303826e-06, + "loss": 0.6197, + "step": 2112 + }, + { + "epoch": 1.7495858641634456, + "grad_norm": 0.9232584238052368, + "learning_rate": 4.08231442615834e-06, + "loss": 0.6332, + "step": 2113 + }, + { + "epoch": 1.7504141358365544, + "grad_norm": 0.92015540599823, + "learning_rate": 4.081462879849765e-06, + "loss": 0.6351, + "step": 2114 + }, + { + "epoch": 1.7512424075096633, + "grad_norm": 0.9353263974189758, + "learning_rate": 4.080611027542863e-06, + "loss": 0.6479, + "step": 2115 + }, + { + "epoch": 1.752070679182772, + "grad_norm": 0.8943057060241699, + "learning_rate": 4.079758869402462e-06, + "loss": 0.6179, + "step": 2116 + }, + { + "epoch": 1.7528989508558808, + "grad_norm": 0.9062668681144714, + "learning_rate": 4.078906405593446e-06, + "loss": 0.6166, + "step": 2117 + }, + { + "epoch": 1.7537272225289895, + "grad_norm": 0.9368826150894165, + "learning_rate": 4.078053636280759e-06, + "loss": 0.6206, + "step": 2118 + }, + { + "epoch": 1.7545554942020982, + "grad_norm": 0.9012559652328491, + "learning_rate": 4.0772005616294034e-06, + "loss": 0.6445, + "step": 2119 + }, + { + "epoch": 1.755383765875207, + "grad_norm": 0.9061569571495056, + "learning_rate": 4.076347181804442e-06, + "loss": 0.6312, + "step": 2120 + }, + { + "epoch": 1.756212037548316, + "grad_norm": 0.892900288105011, + "learning_rate": 4.075493496970996e-06, + "loss": 0.64, + "step": 2121 + }, + { + "epoch": 1.7570403092214246, + "grad_norm": 0.9200263023376465, + "learning_rate": 4.074639507294243e-06, + "loss": 0.6435, + "step": 2122 + }, + { + "epoch": 1.7578685808945336, + "grad_norm": 0.921215295791626, + "learning_rate": 4.0737852129394265e-06, + "loss": 0.6135, + "step": 2123 + }, + { + "epoch": 1.7586968525676423, + "grad_norm": 0.8980262279510498, + "learning_rate": 4.07293061407184e-06, + "loss": 0.635, + "step": 2124 + }, + { + "epoch": 1.759525124240751, + "grad_norm": 0.9192932844161987, + "learning_rate": 4.072075710856843e-06, + "loss": 0.6317, + "step": 2125 + }, + { + "epoch": 1.7603533959138598, + "grad_norm": 0.9387120008468628, + "learning_rate": 4.07122050345985e-06, + "loss": 0.6324, + "step": 2126 + }, + { + "epoch": 1.7611816675869685, + "grad_norm": 0.8787287473678589, + "learning_rate": 4.070364992046338e-06, + "loss": 0.6177, + "step": 2127 + }, + { + "epoch": 1.7620099392600772, + "grad_norm": 0.8963474631309509, + "learning_rate": 4.069509176781838e-06, + "loss": 0.6158, + "step": 2128 + }, + { + "epoch": 1.762838210933186, + "grad_norm": 0.8926964402198792, + "learning_rate": 4.068653057831942e-06, + "loss": 0.6528, + "step": 2129 + }, + { + "epoch": 1.7636664826062949, + "grad_norm": 0.9290505647659302, + "learning_rate": 4.067796635362303e-06, + "loss": 0.6274, + "step": 2130 + }, + { + "epoch": 1.7644947542794036, + "grad_norm": 0.8848921060562134, + "learning_rate": 4.06693990953863e-06, + "loss": 0.6507, + "step": 2131 + }, + { + "epoch": 1.7653230259525126, + "grad_norm": 0.8977338075637817, + "learning_rate": 4.0660828805266915e-06, + "loss": 0.6437, + "step": 2132 + }, + { + "epoch": 1.7661512976256213, + "grad_norm": 0.9125422239303589, + "learning_rate": 4.065225548492314e-06, + "loss": 0.6221, + "step": 2133 + }, + { + "epoch": 1.76697956929873, + "grad_norm": 0.9186370968818665, + "learning_rate": 4.064367913601384e-06, + "loss": 0.6174, + "step": 2134 + }, + { + "epoch": 1.7678078409718387, + "grad_norm": 0.9084877967834473, + "learning_rate": 4.063509976019845e-06, + "loss": 0.6228, + "step": 2135 + }, + { + "epoch": 1.7686361126449475, + "grad_norm": 0.9180804491043091, + "learning_rate": 4.0626517359137e-06, + "loss": 0.6126, + "step": 2136 + }, + { + "epoch": 1.7694643843180562, + "grad_norm": 0.9099741578102112, + "learning_rate": 4.061793193449013e-06, + "loss": 0.6027, + "step": 2137 + }, + { + "epoch": 1.7702926559911651, + "grad_norm": 0.9060433506965637, + "learning_rate": 4.0609343487919005e-06, + "loss": 0.632, + "step": 2138 + }, + { + "epoch": 1.7711209276642739, + "grad_norm": 0.9404996037483215, + "learning_rate": 4.060075202108542e-06, + "loss": 0.6405, + "step": 2139 + }, + { + "epoch": 1.7719491993373828, + "grad_norm": 0.9031956791877747, + "learning_rate": 4.059215753565176e-06, + "loss": 0.6137, + "step": 2140 + }, + { + "epoch": 1.7727774710104915, + "grad_norm": 0.8913756012916565, + "learning_rate": 4.058356003328095e-06, + "loss": 0.6387, + "step": 2141 + }, + { + "epoch": 1.7736057426836003, + "grad_norm": 0.8986163139343262, + "learning_rate": 4.057495951563655e-06, + "loss": 0.6151, + "step": 2142 + }, + { + "epoch": 1.774434014356709, + "grad_norm": 0.9017037153244019, + "learning_rate": 4.0566355984382676e-06, + "loss": 0.6466, + "step": 2143 + }, + { + "epoch": 1.7752622860298177, + "grad_norm": 0.9211136698722839, + "learning_rate": 4.0557749441184024e-06, + "loss": 0.638, + "step": 2144 + }, + { + "epoch": 1.7760905577029265, + "grad_norm": 0.9097506999969482, + "learning_rate": 4.054913988770588e-06, + "loss": 0.6529, + "step": 2145 + }, + { + "epoch": 1.7769188293760352, + "grad_norm": 0.8985908627510071, + "learning_rate": 4.054052732561411e-06, + "loss": 0.6388, + "step": 2146 + }, + { + "epoch": 1.7777471010491441, + "grad_norm": 0.9066522121429443, + "learning_rate": 4.053191175657518e-06, + "loss": 0.6429, + "step": 2147 + }, + { + "epoch": 1.7785753727222529, + "grad_norm": 0.9080630540847778, + "learning_rate": 4.052329318225611e-06, + "loss": 0.619, + "step": 2148 + }, + { + "epoch": 1.7794036443953618, + "grad_norm": 0.8701719641685486, + "learning_rate": 4.0514671604324505e-06, + "loss": 0.6245, + "step": 2149 + }, + { + "epoch": 1.7802319160684705, + "grad_norm": 0.9059606194496155, + "learning_rate": 4.0506047024448585e-06, + "loss": 0.6124, + "step": 2150 + }, + { + "epoch": 1.7810601877415793, + "grad_norm": 0.8912079334259033, + "learning_rate": 4.04974194442971e-06, + "loss": 0.6295, + "step": 2151 + }, + { + "epoch": 1.781888459414688, + "grad_norm": 0.9063137769699097, + "learning_rate": 4.048878886553942e-06, + "loss": 0.6299, + "step": 2152 + }, + { + "epoch": 1.7827167310877967, + "grad_norm": 0.8850398659706116, + "learning_rate": 4.048015528984548e-06, + "loss": 0.6251, + "step": 2153 + }, + { + "epoch": 1.7835450027609054, + "grad_norm": 0.9059123992919922, + "learning_rate": 4.04715187188858e-06, + "loss": 0.6278, + "step": 2154 + }, + { + "epoch": 1.7843732744340144, + "grad_norm": 0.8777562379837036, + "learning_rate": 4.046287915433147e-06, + "loss": 0.6003, + "step": 2155 + }, + { + "epoch": 1.7852015461071231, + "grad_norm": 0.9048421382904053, + "learning_rate": 4.045423659785417e-06, + "loss": 0.6143, + "step": 2156 + }, + { + "epoch": 1.786029817780232, + "grad_norm": 0.9388517737388611, + "learning_rate": 4.044559105112614e-06, + "loss": 0.6314, + "step": 2157 + }, + { + "epoch": 1.7868580894533408, + "grad_norm": 0.9405019283294678, + "learning_rate": 4.043694251582023e-06, + "loss": 0.6381, + "step": 2158 + }, + { + "epoch": 1.7876863611264495, + "grad_norm": 0.8942162990570068, + "learning_rate": 4.0428290993609845e-06, + "loss": 0.6318, + "step": 2159 + }, + { + "epoch": 1.7885146327995582, + "grad_norm": 0.9065560698509216, + "learning_rate": 4.041963648616897e-06, + "loss": 0.6258, + "step": 2160 + }, + { + "epoch": 1.789342904472667, + "grad_norm": 0.8766914010047913, + "learning_rate": 4.041097899517218e-06, + "loss": 0.6292, + "step": 2161 + }, + { + "epoch": 1.7901711761457757, + "grad_norm": 0.8942935466766357, + "learning_rate": 4.040231852229462e-06, + "loss": 0.662, + "step": 2162 + }, + { + "epoch": 1.7909994478188846, + "grad_norm": 0.8892539143562317, + "learning_rate": 4.039365506921198e-06, + "loss": 0.6251, + "step": 2163 + }, + { + "epoch": 1.7918277194919934, + "grad_norm": 0.9359427690505981, + "learning_rate": 4.038498863760061e-06, + "loss": 0.6222, + "step": 2164 + }, + { + "epoch": 1.792655991165102, + "grad_norm": 0.8899942636489868, + "learning_rate": 4.037631922913733e-06, + "loss": 0.6318, + "step": 2165 + }, + { + "epoch": 1.793484262838211, + "grad_norm": 0.9233553409576416, + "learning_rate": 4.0367646845499635e-06, + "loss": 0.6188, + "step": 2166 + }, + { + "epoch": 1.7943125345113198, + "grad_norm": 0.8946877121925354, + "learning_rate": 4.035897148836553e-06, + "loss": 0.6247, + "step": 2167 + }, + { + "epoch": 1.7951408061844285, + "grad_norm": 0.9200116991996765, + "learning_rate": 4.03502931594136e-06, + "loss": 0.5981, + "step": 2168 + }, + { + "epoch": 1.7959690778575372, + "grad_norm": 0.9193991422653198, + "learning_rate": 4.034161186032306e-06, + "loss": 0.6144, + "step": 2169 + }, + { + "epoch": 1.796797349530646, + "grad_norm": 0.9637252688407898, + "learning_rate": 4.0332927592773615e-06, + "loss": 0.6094, + "step": 2170 + }, + { + "epoch": 1.7976256212037547, + "grad_norm": 0.8958486318588257, + "learning_rate": 4.032424035844564e-06, + "loss": 0.6304, + "step": 2171 + }, + { + "epoch": 1.7984538928768636, + "grad_norm": 0.9411646127700806, + "learning_rate": 4.031555015901998e-06, + "loss": 0.5984, + "step": 2172 + }, + { + "epoch": 1.7992821645499724, + "grad_norm": 0.8594862222671509, + "learning_rate": 4.030685699617815e-06, + "loss": 0.6176, + "step": 2173 + }, + { + "epoch": 1.8001104362230813, + "grad_norm": 0.8969242572784424, + "learning_rate": 4.029816087160217e-06, + "loss": 0.6294, + "step": 2174 + }, + { + "epoch": 1.80093870789619, + "grad_norm": 0.8950173854827881, + "learning_rate": 4.028946178697467e-06, + "loss": 0.6441, + "step": 2175 + }, + { + "epoch": 1.8017669795692988, + "grad_norm": 0.9084720611572266, + "learning_rate": 4.0280759743978835e-06, + "loss": 0.6245, + "step": 2176 + }, + { + "epoch": 1.8025952512424075, + "grad_norm": 0.9174644947052002, + "learning_rate": 4.027205474429845e-06, + "loss": 0.6136, + "step": 2177 + }, + { + "epoch": 1.8034235229155162, + "grad_norm": 0.9103249907493591, + "learning_rate": 4.026334678961783e-06, + "loss": 0.6341, + "step": 2178 + }, + { + "epoch": 1.804251794588625, + "grad_norm": 0.8916360139846802, + "learning_rate": 4.025463588162189e-06, + "loss": 0.6424, + "step": 2179 + }, + { + "epoch": 1.805080066261734, + "grad_norm": 0.9347325563430786, + "learning_rate": 4.024592202199611e-06, + "loss": 0.6425, + "step": 2180 + }, + { + "epoch": 1.8059083379348426, + "grad_norm": 0.9135183691978455, + "learning_rate": 4.023720521242653e-06, + "loss": 0.6265, + "step": 2181 + }, + { + "epoch": 1.8067366096079516, + "grad_norm": 0.8996950387954712, + "learning_rate": 4.022848545459979e-06, + "loss": 0.6083, + "step": 2182 + }, + { + "epoch": 1.8075648812810603, + "grad_norm": 0.9104750752449036, + "learning_rate": 4.021976275020307e-06, + "loss": 0.6087, + "step": 2183 + }, + { + "epoch": 1.808393152954169, + "grad_norm": 0.9272193908691406, + "learning_rate": 4.0211037100924135e-06, + "loss": 0.6254, + "step": 2184 + }, + { + "epoch": 1.8092214246272778, + "grad_norm": 0.8605275750160217, + "learning_rate": 4.020230850845132e-06, + "loss": 0.6244, + "step": 2185 + }, + { + "epoch": 1.8100496963003865, + "grad_norm": 0.8817747831344604, + "learning_rate": 4.019357697447352e-06, + "loss": 0.6226, + "step": 2186 + }, + { + "epoch": 1.8108779679734952, + "grad_norm": 0.8978784084320068, + "learning_rate": 4.018484250068021e-06, + "loss": 0.615, + "step": 2187 + }, + { + "epoch": 1.811706239646604, + "grad_norm": 0.8996117115020752, + "learning_rate": 4.017610508876143e-06, + "loss": 0.6248, + "step": 2188 + }, + { + "epoch": 1.8125345113197129, + "grad_norm": 0.9133120179176331, + "learning_rate": 4.016736474040778e-06, + "loss": 0.6364, + "step": 2189 + }, + { + "epoch": 1.8133627829928216, + "grad_norm": 0.903982400894165, + "learning_rate": 4.015862145731044e-06, + "loss": 0.6078, + "step": 2190 + }, + { + "epoch": 1.8141910546659306, + "grad_norm": 0.9343582987785339, + "learning_rate": 4.014987524116115e-06, + "loss": 0.6254, + "step": 2191 + }, + { + "epoch": 1.8150193263390393, + "grad_norm": 0.883305549621582, + "learning_rate": 4.014112609365224e-06, + "loss": 0.6306, + "step": 2192 + }, + { + "epoch": 1.815847598012148, + "grad_norm": 0.9044662714004517, + "learning_rate": 4.013237401647656e-06, + "loss": 0.6288, + "step": 2193 + }, + { + "epoch": 1.8166758696852567, + "grad_norm": 0.9016374349594116, + "learning_rate": 4.012361901132758e-06, + "loss": 0.6254, + "step": 2194 + }, + { + "epoch": 1.8175041413583655, + "grad_norm": 0.9198315739631653, + "learning_rate": 4.01148610798993e-06, + "loss": 0.621, + "step": 2195 + }, + { + "epoch": 1.8183324130314742, + "grad_norm": 0.8884013891220093, + "learning_rate": 4.01061002238863e-06, + "loss": 0.6195, + "step": 2196 + }, + { + "epoch": 1.8191606847045831, + "grad_norm": 0.901606023311615, + "learning_rate": 4.009733644498372e-06, + "loss": 0.646, + "step": 2197 + }, + { + "epoch": 1.8199889563776919, + "grad_norm": 0.9003238677978516, + "learning_rate": 4.008856974488728e-06, + "loss": 0.6276, + "step": 2198 + }, + { + "epoch": 1.8208172280508008, + "grad_norm": 0.8835832476615906, + "learning_rate": 4.007980012529324e-06, + "loss": 0.6232, + "step": 2199 + }, + { + "epoch": 1.8216454997239095, + "grad_norm": 0.8751473426818848, + "learning_rate": 4.007102758789846e-06, + "loss": 0.635, + "step": 2200 + }, + { + "epoch": 1.8224737713970183, + "grad_norm": 0.8990851640701294, + "learning_rate": 4.006225213440032e-06, + "loss": 0.605, + "step": 2201 + }, + { + "epoch": 1.823302043070127, + "grad_norm": 0.8766496181488037, + "learning_rate": 4.0053473766496805e-06, + "loss": 0.6396, + "step": 2202 + }, + { + "epoch": 1.8241303147432357, + "grad_norm": 0.914273202419281, + "learning_rate": 4.004469248588644e-06, + "loss": 0.6435, + "step": 2203 + }, + { + "epoch": 1.8249585864163445, + "grad_norm": 0.9063234925270081, + "learning_rate": 4.003590829426832e-06, + "loss": 0.6144, + "step": 2204 + }, + { + "epoch": 1.8257868580894532, + "grad_norm": 0.9479252696037292, + "learning_rate": 4.002712119334212e-06, + "loss": 0.6584, + "step": 2205 + }, + { + "epoch": 1.8266151297625621, + "grad_norm": 0.9253218770027161, + "learning_rate": 4.001833118480804e-06, + "loss": 0.6203, + "step": 2206 + }, + { + "epoch": 1.8274434014356709, + "grad_norm": 0.930874764919281, + "learning_rate": 4.000953827036689e-06, + "loss": 0.6254, + "step": 2207 + }, + { + "epoch": 1.8282716731087798, + "grad_norm": 0.9202263951301575, + "learning_rate": 4.000074245171999e-06, + "loss": 0.6351, + "step": 2208 + }, + { + "epoch": 1.8290999447818885, + "grad_norm": 0.8993806838989258, + "learning_rate": 3.999194373056926e-06, + "loss": 0.6439, + "step": 2209 + }, + { + "epoch": 1.8299282164549973, + "grad_norm": 0.9138032793998718, + "learning_rate": 3.998314210861718e-06, + "loss": 0.6011, + "step": 2210 + }, + { + "epoch": 1.830756488128106, + "grad_norm": 0.8995500206947327, + "learning_rate": 3.9974337587566765e-06, + "loss": 0.6285, + "step": 2211 + }, + { + "epoch": 1.8315847598012147, + "grad_norm": 0.9092652797698975, + "learning_rate": 3.996553016912161e-06, + "loss": 0.6364, + "step": 2212 + }, + { + "epoch": 1.8324130314743234, + "grad_norm": 0.9085352420806885, + "learning_rate": 3.995671985498589e-06, + "loss": 0.6245, + "step": 2213 + }, + { + "epoch": 1.8332413031474324, + "grad_norm": 0.9546533226966858, + "learning_rate": 3.994790664686429e-06, + "loss": 0.6512, + "step": 2214 + }, + { + "epoch": 1.8340695748205411, + "grad_norm": 0.8606089353561401, + "learning_rate": 3.99390905464621e-06, + "loss": 0.6135, + "step": 2215 + }, + { + "epoch": 1.83489784649365, + "grad_norm": 0.9450663328170776, + "learning_rate": 3.993027155548515e-06, + "loss": 0.6193, + "step": 2216 + }, + { + "epoch": 1.8357261181667588, + "grad_norm": 0.9496508836746216, + "learning_rate": 3.992144967563984e-06, + "loss": 0.5978, + "step": 2217 + }, + { + "epoch": 1.8365543898398675, + "grad_norm": 0.8837412595748901, + "learning_rate": 3.99126249086331e-06, + "loss": 0.6104, + "step": 2218 + }, + { + "epoch": 1.8373826615129762, + "grad_norm": 0.8774031400680542, + "learning_rate": 3.9903797256172464e-06, + "loss": 0.6302, + "step": 2219 + }, + { + "epoch": 1.838210933186085, + "grad_norm": 0.8978205919265747, + "learning_rate": 3.989496671996599e-06, + "loss": 0.6411, + "step": 2220 + }, + { + "epoch": 1.8390392048591937, + "grad_norm": 0.9087855815887451, + "learning_rate": 3.98861333017223e-06, + "loss": 0.592, + "step": 2221 + }, + { + "epoch": 1.8398674765323026, + "grad_norm": 0.9040684700012207, + "learning_rate": 3.987729700315059e-06, + "loss": 0.6121, + "step": 2222 + }, + { + "epoch": 1.8406957482054114, + "grad_norm": 0.8980532288551331, + "learning_rate": 3.986845782596058e-06, + "loss": 0.6231, + "step": 2223 + }, + { + "epoch": 1.84152401987852, + "grad_norm": 0.9104854464530945, + "learning_rate": 3.9859615771862595e-06, + "loss": 0.6228, + "step": 2224 + }, + { + "epoch": 1.842352291551629, + "grad_norm": 0.9314625263214111, + "learning_rate": 3.985077084256747e-06, + "loss": 0.6136, + "step": 2225 + }, + { + "epoch": 1.8431805632247378, + "grad_norm": 0.9033634662628174, + "learning_rate": 3.984192303978663e-06, + "loss": 0.6408, + "step": 2226 + }, + { + "epoch": 1.8440088348978465, + "grad_norm": 0.9022733569145203, + "learning_rate": 3.983307236523203e-06, + "loss": 0.5954, + "step": 2227 + }, + { + "epoch": 1.8448371065709552, + "grad_norm": 0.8900256156921387, + "learning_rate": 3.982421882061621e-06, + "loss": 0.631, + "step": 2228 + }, + { + "epoch": 1.845665378244064, + "grad_norm": 0.8882872462272644, + "learning_rate": 3.981536240765223e-06, + "loss": 0.6204, + "step": 2229 + }, + { + "epoch": 1.8464936499171727, + "grad_norm": 0.894923985004425, + "learning_rate": 3.980650312805373e-06, + "loss": 0.6013, + "step": 2230 + }, + { + "epoch": 1.8473219215902816, + "grad_norm": 0.8910638093948364, + "learning_rate": 3.979764098353491e-06, + "loss": 0.6373, + "step": 2231 + }, + { + "epoch": 1.8481501932633904, + "grad_norm": 0.8952053189277649, + "learning_rate": 3.978877597581049e-06, + "loss": 0.6374, + "step": 2232 + }, + { + "epoch": 1.8489784649364993, + "grad_norm": 0.9100720882415771, + "learning_rate": 3.977990810659579e-06, + "loss": 0.6417, + "step": 2233 + }, + { + "epoch": 1.849806736609608, + "grad_norm": 0.8954150676727295, + "learning_rate": 3.9771037377606634e-06, + "loss": 0.6027, + "step": 2234 + }, + { + "epoch": 1.8506350082827168, + "grad_norm": 0.8989905118942261, + "learning_rate": 3.976216379055944e-06, + "loss": 0.6228, + "step": 2235 + }, + { + "epoch": 1.8514632799558255, + "grad_norm": 0.9060820937156677, + "learning_rate": 3.975328734717117e-06, + "loss": 0.6291, + "step": 2236 + }, + { + "epoch": 1.8522915516289342, + "grad_norm": 0.9164367318153381, + "learning_rate": 3.974440804915932e-06, + "loss": 0.6249, + "step": 2237 + }, + { + "epoch": 1.853119823302043, + "grad_norm": 0.9227840304374695, + "learning_rate": 3.973552589824196e-06, + "loss": 0.6328, + "step": 2238 + }, + { + "epoch": 1.853948094975152, + "grad_norm": 0.8612478971481323, + "learning_rate": 3.972664089613772e-06, + "loss": 0.5794, + "step": 2239 + }, + { + "epoch": 1.8547763666482606, + "grad_norm": 0.8819253444671631, + "learning_rate": 3.971775304456573e-06, + "loss": 0.6145, + "step": 2240 + }, + { + "epoch": 1.8556046383213696, + "grad_norm": 0.9216543436050415, + "learning_rate": 3.970886234524573e-06, + "loss": 0.609, + "step": 2241 + }, + { + "epoch": 1.8564329099944783, + "grad_norm": 0.9114699959754944, + "learning_rate": 3.969996879989797e-06, + "loss": 0.6105, + "step": 2242 + }, + { + "epoch": 1.857261181667587, + "grad_norm": 0.9464704990386963, + "learning_rate": 3.969107241024328e-06, + "loss": 0.6389, + "step": 2243 + }, + { + "epoch": 1.8580894533406958, + "grad_norm": 0.8921800255775452, + "learning_rate": 3.968217317800303e-06, + "loss": 0.6338, + "step": 2244 + }, + { + "epoch": 1.8589177250138045, + "grad_norm": 0.897452175617218, + "learning_rate": 3.967327110489913e-06, + "loss": 0.6248, + "step": 2245 + }, + { + "epoch": 1.8597459966869132, + "grad_norm": 0.9093567132949829, + "learning_rate": 3.9664366192654066e-06, + "loss": 0.6258, + "step": 2246 + }, + { + "epoch": 1.860574268360022, + "grad_norm": 0.894188642501831, + "learning_rate": 3.965545844299084e-06, + "loss": 0.6095, + "step": 2247 + }, + { + "epoch": 1.8614025400331309, + "grad_norm": 0.9238452911376953, + "learning_rate": 3.964654785763301e-06, + "loss": 0.6237, + "step": 2248 + }, + { + "epoch": 1.8622308117062396, + "grad_norm": 0.8715214133262634, + "learning_rate": 3.9637634438304715e-06, + "loss": 0.6153, + "step": 2249 + }, + { + "epoch": 1.8630590833793486, + "grad_norm": 0.8790848255157471, + "learning_rate": 3.962871818673061e-06, + "loss": 0.6389, + "step": 2250 + }, + { + "epoch": 1.8638873550524573, + "grad_norm": 0.9859224557876587, + "learning_rate": 3.961979910463589e-06, + "loss": 0.6075, + "step": 2251 + }, + { + "epoch": 1.864715626725566, + "grad_norm": 0.9127727150917053, + "learning_rate": 3.9610877193746335e-06, + "loss": 0.6332, + "step": 2252 + }, + { + "epoch": 1.8655438983986747, + "grad_norm": 0.9248382449150085, + "learning_rate": 3.960195245578825e-06, + "loss": 0.622, + "step": 2253 + }, + { + "epoch": 1.8663721700717835, + "grad_norm": 0.9030913710594177, + "learning_rate": 3.959302489248848e-06, + "loss": 0.6326, + "step": 2254 + }, + { + "epoch": 1.8672004417448922, + "grad_norm": 0.8922805190086365, + "learning_rate": 3.958409450557441e-06, + "loss": 0.6099, + "step": 2255 + }, + { + "epoch": 1.8680287134180011, + "grad_norm": 0.9165307879447937, + "learning_rate": 3.957516129677402e-06, + "loss": 0.6245, + "step": 2256 + }, + { + "epoch": 1.8688569850911099, + "grad_norm": 0.9445623159408569, + "learning_rate": 3.956622526781579e-06, + "loss": 0.6317, + "step": 2257 + }, + { + "epoch": 1.8696852567642188, + "grad_norm": 0.8833973407745361, + "learning_rate": 3.955728642042874e-06, + "loss": 0.6264, + "step": 2258 + }, + { + "epoch": 1.8705135284373275, + "grad_norm": 0.9512470364570618, + "learning_rate": 3.9548344756342486e-06, + "loss": 0.6137, + "step": 2259 + }, + { + "epoch": 1.8713418001104363, + "grad_norm": 0.8834348320960999, + "learning_rate": 3.9539400277287134e-06, + "loss": 0.6597, + "step": 2260 + }, + { + "epoch": 1.872170071783545, + "grad_norm": 0.8881180882453918, + "learning_rate": 3.953045298499335e-06, + "loss": 0.6502, + "step": 2261 + }, + { + "epoch": 1.8729983434566537, + "grad_norm": 0.8921699523925781, + "learning_rate": 3.952150288119239e-06, + "loss": 0.6216, + "step": 2262 + }, + { + "epoch": 1.8738266151297625, + "grad_norm": 0.91098552942276, + "learning_rate": 3.9512549967615965e-06, + "loss": 0.6286, + "step": 2263 + }, + { + "epoch": 1.8746548868028712, + "grad_norm": 0.9343264698982239, + "learning_rate": 3.950359424599642e-06, + "loss": 0.6368, + "step": 2264 + }, + { + "epoch": 1.8754831584759801, + "grad_norm": 0.9168746471405029, + "learning_rate": 3.949463571806658e-06, + "loss": 0.6355, + "step": 2265 + }, + { + "epoch": 1.8763114301490889, + "grad_norm": 0.9206068515777588, + "learning_rate": 3.948567438555986e-06, + "loss": 0.607, + "step": 2266 + }, + { + "epoch": 1.8771397018221978, + "grad_norm": 0.9262584447860718, + "learning_rate": 3.947671025021017e-06, + "loss": 0.6197, + "step": 2267 + }, + { + "epoch": 1.8779679734953065, + "grad_norm": 0.8699168562889099, + "learning_rate": 3.946774331375199e-06, + "loss": 0.6181, + "step": 2268 + }, + { + "epoch": 1.8787962451684153, + "grad_norm": 0.8892862200737, + "learning_rate": 3.9458773577920365e-06, + "loss": 0.6245, + "step": 2269 + }, + { + "epoch": 1.879624516841524, + "grad_norm": 0.9038919806480408, + "learning_rate": 3.944980104445082e-06, + "loss": 0.6209, + "step": 2270 + }, + { + "epoch": 1.8804527885146327, + "grad_norm": 0.8780872821807861, + "learning_rate": 3.9440825715079475e-06, + "loss": 0.6355, + "step": 2271 + }, + { + "epoch": 1.8812810601877414, + "grad_norm": 0.9379895925521851, + "learning_rate": 3.943184759154297e-06, + "loss": 0.6381, + "step": 2272 + }, + { + "epoch": 1.8821093318608504, + "grad_norm": 0.9112582802772522, + "learning_rate": 3.942286667557849e-06, + "loss": 0.6273, + "step": 2273 + }, + { + "epoch": 1.8829376035339591, + "grad_norm": 0.9367797374725342, + "learning_rate": 3.9413882968923754e-06, + "loss": 0.6111, + "step": 2274 + }, + { + "epoch": 1.883765875207068, + "grad_norm": 0.8923424482345581, + "learning_rate": 3.940489647331703e-06, + "loss": 0.6187, + "step": 2275 + }, + { + "epoch": 1.8845941468801768, + "grad_norm": 0.8797968029975891, + "learning_rate": 3.939590719049712e-06, + "loss": 0.603, + "step": 2276 + }, + { + "epoch": 1.8854224185532855, + "grad_norm": 0.8977354168891907, + "learning_rate": 3.938691512220336e-06, + "loss": 0.614, + "step": 2277 + }, + { + "epoch": 1.8862506902263942, + "grad_norm": 0.9207090139389038, + "learning_rate": 3.937792027017564e-06, + "loss": 0.6162, + "step": 2278 + }, + { + "epoch": 1.887078961899503, + "grad_norm": 0.8815562725067139, + "learning_rate": 3.936892263615438e-06, + "loss": 0.6315, + "step": 2279 + }, + { + "epoch": 1.8879072335726117, + "grad_norm": 0.9983104467391968, + "learning_rate": 3.935992222188054e-06, + "loss": 0.6337, + "step": 2280 + }, + { + "epoch": 1.8887355052457206, + "grad_norm": 0.8987579941749573, + "learning_rate": 3.93509190290956e-06, + "loss": 0.644, + "step": 2281 + }, + { + "epoch": 1.8895637769188294, + "grad_norm": 0.9132242798805237, + "learning_rate": 3.934191305954161e-06, + "loss": 0.6063, + "step": 2282 + }, + { + "epoch": 1.890392048591938, + "grad_norm": 0.8940858840942383, + "learning_rate": 3.933290431496114e-06, + "loss": 0.6204, + "step": 2283 + }, + { + "epoch": 1.891220320265047, + "grad_norm": 0.8862229585647583, + "learning_rate": 3.932389279709729e-06, + "loss": 0.6257, + "step": 2284 + }, + { + "epoch": 1.8920485919381558, + "grad_norm": 0.8841557502746582, + "learning_rate": 3.931487850769371e-06, + "loss": 0.6329, + "step": 2285 + }, + { + "epoch": 1.8928768636112645, + "grad_norm": 0.9075404405593872, + "learning_rate": 3.930586144849459e-06, + "loss": 0.6195, + "step": 2286 + }, + { + "epoch": 1.8937051352843732, + "grad_norm": 0.9349889755249023, + "learning_rate": 3.929684162124462e-06, + "loss": 0.6159, + "step": 2287 + }, + { + "epoch": 1.894533406957482, + "grad_norm": 0.9378470182418823, + "learning_rate": 3.928781902768908e-06, + "loss": 0.6221, + "step": 2288 + }, + { + "epoch": 1.8953616786305907, + "grad_norm": 0.8877297043800354, + "learning_rate": 3.927879366957374e-06, + "loss": 0.6359, + "step": 2289 + }, + { + "epoch": 1.8961899503036996, + "grad_norm": 0.9335882067680359, + "learning_rate": 3.926976554864495e-06, + "loss": 0.6397, + "step": 2290 + }, + { + "epoch": 1.8970182219768084, + "grad_norm": 0.8982004523277283, + "learning_rate": 3.926073466664953e-06, + "loss": 0.6303, + "step": 2291 + }, + { + "epoch": 1.8978464936499173, + "grad_norm": 0.8909676671028137, + "learning_rate": 3.92517010253349e-06, + "loss": 0.6476, + "step": 2292 + }, + { + "epoch": 1.898674765323026, + "grad_norm": 0.9127813577651978, + "learning_rate": 3.924266462644896e-06, + "loss": 0.6078, + "step": 2293 + }, + { + "epoch": 1.8995030369961348, + "grad_norm": 0.8923580050468445, + "learning_rate": 3.923362547174019e-06, + "loss": 0.6371, + "step": 2294 + }, + { + "epoch": 1.9003313086692435, + "grad_norm": 0.912458062171936, + "learning_rate": 3.922458356295757e-06, + "loss": 0.6162, + "step": 2295 + }, + { + "epoch": 1.9011595803423522, + "grad_norm": 0.8866751790046692, + "learning_rate": 3.921553890185064e-06, + "loss": 0.6011, + "step": 2296 + }, + { + "epoch": 1.901987852015461, + "grad_norm": 0.9114490151405334, + "learning_rate": 3.920649149016944e-06, + "loss": 0.6322, + "step": 2297 + }, + { + "epoch": 1.90281612368857, + "grad_norm": 0.912030816078186, + "learning_rate": 3.919744132966457e-06, + "loss": 0.6293, + "step": 2298 + }, + { + "epoch": 1.9036443953616786, + "grad_norm": 0.887301504611969, + "learning_rate": 3.918838842208715e-06, + "loss": 0.6288, + "step": 2299 + }, + { + "epoch": 1.9044726670347876, + "grad_norm": 0.9039818048477173, + "learning_rate": 3.917933276918884e-06, + "loss": 0.6154, + "step": 2300 + }, + { + "epoch": 1.9053009387078963, + "grad_norm": 0.9260784983634949, + "learning_rate": 3.917027437272182e-06, + "loss": 0.6111, + "step": 2301 + }, + { + "epoch": 1.906129210381005, + "grad_norm": 0.8966069221496582, + "learning_rate": 3.916121323443879e-06, + "loss": 0.6334, + "step": 2302 + }, + { + "epoch": 1.9069574820541138, + "grad_norm": 0.8983698487281799, + "learning_rate": 3.9152149356093e-06, + "loss": 0.6235, + "step": 2303 + }, + { + "epoch": 1.9077857537272225, + "grad_norm": 0.9150595664978027, + "learning_rate": 3.914308273943826e-06, + "loss": 0.6118, + "step": 2304 + }, + { + "epoch": 1.9086140254003312, + "grad_norm": 0.8768419027328491, + "learning_rate": 3.913401338622883e-06, + "loss": 0.5935, + "step": 2305 + }, + { + "epoch": 1.90944229707344, + "grad_norm": 0.887562096118927, + "learning_rate": 3.912494129821958e-06, + "loss": 0.6232, + "step": 2306 + }, + { + "epoch": 1.9102705687465489, + "grad_norm": 0.8916515111923218, + "learning_rate": 3.911586647716585e-06, + "loss": 0.6409, + "step": 2307 + }, + { + "epoch": 1.9110988404196576, + "grad_norm": 0.9207882881164551, + "learning_rate": 3.910678892482354e-06, + "loss": 0.6155, + "step": 2308 + }, + { + "epoch": 1.9119271120927666, + "grad_norm": 0.9300670027732849, + "learning_rate": 3.9097708642949085e-06, + "loss": 0.6389, + "step": 2309 + }, + { + "epoch": 1.9127553837658753, + "grad_norm": 0.8685980439186096, + "learning_rate": 3.908862563329941e-06, + "loss": 0.6392, + "step": 2310 + }, + { + "epoch": 1.913583655438984, + "grad_norm": 0.9150967001914978, + "learning_rate": 3.907953989763202e-06, + "loss": 0.6387, + "step": 2311 + }, + { + "epoch": 1.9144119271120927, + "grad_norm": 0.9094190001487732, + "learning_rate": 3.90704514377049e-06, + "loss": 0.6055, + "step": 2312 + }, + { + "epoch": 1.9152401987852015, + "grad_norm": 0.8863343596458435, + "learning_rate": 3.90613602552766e-06, + "loss": 0.6073, + "step": 2313 + }, + { + "epoch": 1.9160684704583102, + "grad_norm": 0.9878994822502136, + "learning_rate": 3.905226635210616e-06, + "loss": 0.613, + "step": 2314 + }, + { + "epoch": 1.9168967421314191, + "grad_norm": 0.8863592147827148, + "learning_rate": 3.904316972995318e-06, + "loss": 0.6102, + "step": 2315 + }, + { + "epoch": 1.9177250138045279, + "grad_norm": 0.8859980702400208, + "learning_rate": 3.9034070390577765e-06, + "loss": 0.6182, + "step": 2316 + }, + { + "epoch": 1.9185532854776368, + "grad_norm": 0.8953589200973511, + "learning_rate": 3.9024968335740556e-06, + "loss": 0.6212, + "step": 2317 + }, + { + "epoch": 1.9193815571507455, + "grad_norm": 0.873466968536377, + "learning_rate": 3.90158635672027e-06, + "loss": 0.616, + "step": 2318 + }, + { + "epoch": 1.9202098288238543, + "grad_norm": 0.8589543700218201, + "learning_rate": 3.900675608672592e-06, + "loss": 0.6141, + "step": 2319 + }, + { + "epoch": 1.921038100496963, + "grad_norm": 0.8717133402824402, + "learning_rate": 3.899764589607239e-06, + "loss": 0.6281, + "step": 2320 + }, + { + "epoch": 1.9218663721700717, + "grad_norm": 0.893227219581604, + "learning_rate": 3.898853299700486e-06, + "loss": 0.6139, + "step": 2321 + }, + { + "epoch": 1.9226946438431805, + "grad_norm": 0.8736181259155273, + "learning_rate": 3.89794173912866e-06, + "loss": 0.5987, + "step": 2322 + }, + { + "epoch": 1.9235229155162892, + "grad_norm": 0.8975626826286316, + "learning_rate": 3.897029908068139e-06, + "loss": 0.6417, + "step": 2323 + }, + { + "epoch": 1.9243511871893981, + "grad_norm": 0.8916004300117493, + "learning_rate": 3.896117806695355e-06, + "loss": 0.6259, + "step": 2324 + }, + { + "epoch": 1.9251794588625069, + "grad_norm": 0.8739155530929565, + "learning_rate": 3.8952054351867885e-06, + "loss": 0.6252, + "step": 2325 + }, + { + "epoch": 1.9260077305356158, + "grad_norm": 0.8667280673980713, + "learning_rate": 3.894292793718976e-06, + "loss": 0.6083, + "step": 2326 + }, + { + "epoch": 1.9268360022087245, + "grad_norm": 0.8749732375144958, + "learning_rate": 3.893379882468506e-06, + "loss": 0.6294, + "step": 2327 + }, + { + "epoch": 1.9276642738818333, + "grad_norm": 0.9085330963134766, + "learning_rate": 3.8924667016120175e-06, + "loss": 0.6101, + "step": 2328 + }, + { + "epoch": 1.928492545554942, + "grad_norm": 0.9394292235374451, + "learning_rate": 3.891553251326203e-06, + "loss": 0.6182, + "step": 2329 + }, + { + "epoch": 1.9293208172280507, + "grad_norm": 0.8756951093673706, + "learning_rate": 3.890639531787805e-06, + "loss": 0.6159, + "step": 2330 + }, + { + "epoch": 1.9301490889011594, + "grad_norm": 0.8765501379966736, + "learning_rate": 3.8897255431736225e-06, + "loss": 0.6182, + "step": 2331 + }, + { + "epoch": 1.9309773605742684, + "grad_norm": 0.907835841178894, + "learning_rate": 3.888811285660501e-06, + "loss": 0.6336, + "step": 2332 + }, + { + "epoch": 1.9318056322473771, + "grad_norm": 1.0617926120758057, + "learning_rate": 3.887896759425342e-06, + "loss": 0.6336, + "step": 2333 + }, + { + "epoch": 1.932633903920486, + "grad_norm": 0.8745515942573547, + "learning_rate": 3.886981964645099e-06, + "loss": 0.6244, + "step": 2334 + }, + { + "epoch": 1.9334621755935948, + "grad_norm": 0.8918715119361877, + "learning_rate": 3.886066901496775e-06, + "loss": 0.6289, + "step": 2335 + }, + { + "epoch": 1.9342904472667035, + "grad_norm": 0.8779492378234863, + "learning_rate": 3.885151570157426e-06, + "loss": 0.6335, + "step": 2336 + }, + { + "epoch": 1.9351187189398122, + "grad_norm": 0.8955790400505066, + "learning_rate": 3.88423597080416e-06, + "loss": 0.6125, + "step": 2337 + }, + { + "epoch": 1.935946990612921, + "grad_norm": 0.8905174136161804, + "learning_rate": 3.883320103614137e-06, + "loss": 0.5913, + "step": 2338 + }, + { + "epoch": 1.9367752622860297, + "grad_norm": 0.9078717231750488, + "learning_rate": 3.88240396876457e-06, + "loss": 0.6356, + "step": 2339 + }, + { + "epoch": 1.9376035339591386, + "grad_norm": 0.9211397171020508, + "learning_rate": 3.881487566432722e-06, + "loss": 0.6305, + "step": 2340 + }, + { + "epoch": 1.9384318056322474, + "grad_norm": 0.9131335020065308, + "learning_rate": 3.880570896795908e-06, + "loss": 0.6184, + "step": 2341 + }, + { + "epoch": 1.9392600773053563, + "grad_norm": 0.8874300718307495, + "learning_rate": 3.879653960031494e-06, + "loss": 0.6191, + "step": 2342 + }, + { + "epoch": 1.940088348978465, + "grad_norm": 0.9244474768638611, + "learning_rate": 3.8787367563169e-06, + "loss": 0.5994, + "step": 2343 + }, + { + "epoch": 1.9409166206515738, + "grad_norm": 0.8986138701438904, + "learning_rate": 3.877819285829596e-06, + "loss": 0.6544, + "step": 2344 + }, + { + "epoch": 1.9417448923246825, + "grad_norm": 0.9084672927856445, + "learning_rate": 3.876901548747104e-06, + "loss": 0.6248, + "step": 2345 + }, + { + "epoch": 1.9425731639977912, + "grad_norm": 0.9052600264549255, + "learning_rate": 3.875983545246997e-06, + "loss": 0.6112, + "step": 2346 + }, + { + "epoch": 1.9434014356709, + "grad_norm": 0.9029438495635986, + "learning_rate": 3.8750652755069025e-06, + "loss": 0.6157, + "step": 2347 + }, + { + "epoch": 1.9442297073440087, + "grad_norm": 0.882449746131897, + "learning_rate": 3.874146739704494e-06, + "loss": 0.5886, + "step": 2348 + }, + { + "epoch": 1.9450579790171176, + "grad_norm": 0.8628272414207458, + "learning_rate": 3.873227938017502e-06, + "loss": 0.608, + "step": 2349 + }, + { + "epoch": 1.9458862506902264, + "grad_norm": 0.8977845907211304, + "learning_rate": 3.872308870623704e-06, + "loss": 0.6475, + "step": 2350 + }, + { + "epoch": 1.9467145223633353, + "grad_norm": 0.8801501393318176, + "learning_rate": 3.871389537700933e-06, + "loss": 0.6246, + "step": 2351 + }, + { + "epoch": 1.947542794036444, + "grad_norm": 0.9245358109474182, + "learning_rate": 3.87046993942707e-06, + "loss": 0.6317, + "step": 2352 + }, + { + "epoch": 1.9483710657095528, + "grad_norm": 0.8959249258041382, + "learning_rate": 3.869550075980049e-06, + "loss": 0.6157, + "step": 2353 + }, + { + "epoch": 1.9491993373826615, + "grad_norm": 0.861315906047821, + "learning_rate": 3.868629947537855e-06, + "loss": 0.6251, + "step": 2354 + }, + { + "epoch": 1.9500276090557702, + "grad_norm": 0.8839966654777527, + "learning_rate": 3.867709554278524e-06, + "loss": 0.6289, + "step": 2355 + }, + { + "epoch": 1.950855880728879, + "grad_norm": 0.8878759741783142, + "learning_rate": 3.8667888963801445e-06, + "loss": 0.6113, + "step": 2356 + }, + { + "epoch": 1.951684152401988, + "grad_norm": 0.8725037574768066, + "learning_rate": 3.865867974020854e-06, + "loss": 0.6324, + "step": 2357 + }, + { + "epoch": 1.9525124240750966, + "grad_norm": 0.9059397578239441, + "learning_rate": 3.8649467873788436e-06, + "loss": 0.5981, + "step": 2358 + }, + { + "epoch": 1.9533406957482056, + "grad_norm": 0.8883296847343445, + "learning_rate": 3.8640253366323535e-06, + "loss": 0.598, + "step": 2359 + }, + { + "epoch": 1.9541689674213143, + "grad_norm": 0.8987225890159607, + "learning_rate": 3.863103621959676e-06, + "loss": 0.6253, + "step": 2360 + }, + { + "epoch": 1.954997239094423, + "grad_norm": 0.8977320194244385, + "learning_rate": 3.862181643539155e-06, + "loss": 0.6424, + "step": 2361 + }, + { + "epoch": 1.9558255107675317, + "grad_norm": 0.9267589449882507, + "learning_rate": 3.861259401549183e-06, + "loss": 0.647, + "step": 2362 + }, + { + "epoch": 1.9566537824406405, + "grad_norm": 0.9297959804534912, + "learning_rate": 3.860336896168207e-06, + "loss": 0.6448, + "step": 2363 + }, + { + "epoch": 1.9574820541137492, + "grad_norm": 0.9627453088760376, + "learning_rate": 3.859414127574722e-06, + "loss": 0.6154, + "step": 2364 + }, + { + "epoch": 1.958310325786858, + "grad_norm": 0.8945866227149963, + "learning_rate": 3.858491095947275e-06, + "loss": 0.6201, + "step": 2365 + }, + { + "epoch": 1.9591385974599669, + "grad_norm": 0.8859817981719971, + "learning_rate": 3.857567801464466e-06, + "loss": 0.6171, + "step": 2366 + }, + { + "epoch": 1.9599668691330756, + "grad_norm": 0.9104642271995544, + "learning_rate": 3.8566442443049425e-06, + "loss": 0.6273, + "step": 2367 + }, + { + "epoch": 1.9607951408061846, + "grad_norm": 0.8821743726730347, + "learning_rate": 3.855720424647403e-06, + "loss": 0.5993, + "step": 2368 + }, + { + "epoch": 1.9616234124792933, + "grad_norm": 0.903322160243988, + "learning_rate": 3.854796342670601e-06, + "loss": 0.6471, + "step": 2369 + }, + { + "epoch": 1.962451684152402, + "grad_norm": 0.8744645714759827, + "learning_rate": 3.853871998553334e-06, + "loss": 0.6444, + "step": 2370 + }, + { + "epoch": 1.9632799558255107, + "grad_norm": 0.9538072347640991, + "learning_rate": 3.852947392474457e-06, + "loss": 0.656, + "step": 2371 + }, + { + "epoch": 1.9641082274986195, + "grad_norm": 0.875407874584198, + "learning_rate": 3.852022524612872e-06, + "loss": 0.6479, + "step": 2372 + }, + { + "epoch": 1.9649364991717282, + "grad_norm": 0.8908282518386841, + "learning_rate": 3.8510973951475305e-06, + "loss": 0.6262, + "step": 2373 + }, + { + "epoch": 1.9657647708448371, + "grad_norm": 0.8698679208755493, + "learning_rate": 3.850172004257438e-06, + "loss": 0.6431, + "step": 2374 + }, + { + "epoch": 1.9665930425179459, + "grad_norm": 0.8997004628181458, + "learning_rate": 3.84924635212165e-06, + "loss": 0.6349, + "step": 2375 + }, + { + "epoch": 1.9674213141910548, + "grad_norm": 0.8815768957138062, + "learning_rate": 3.8483204389192695e-06, + "loss": 0.6449, + "step": 2376 + }, + { + "epoch": 1.9682495858641635, + "grad_norm": 0.9148927927017212, + "learning_rate": 3.847394264829453e-06, + "loss": 0.6106, + "step": 2377 + }, + { + "epoch": 1.9690778575372723, + "grad_norm": 0.8647779226303101, + "learning_rate": 3.846467830031407e-06, + "loss": 0.5954, + "step": 2378 + }, + { + "epoch": 1.969906129210381, + "grad_norm": 0.8877570033073425, + "learning_rate": 3.845541134704387e-06, + "loss": 0.6006, + "step": 2379 + }, + { + "epoch": 1.9707344008834897, + "grad_norm": 0.899742603302002, + "learning_rate": 3.844614179027701e-06, + "loss": 0.6069, + "step": 2380 + }, + { + "epoch": 1.9715626725565985, + "grad_norm": 0.9085001349449158, + "learning_rate": 3.843686963180705e-06, + "loss": 0.6332, + "step": 2381 + }, + { + "epoch": 1.9723909442297074, + "grad_norm": 0.8954899311065674, + "learning_rate": 3.842759487342808e-06, + "loss": 0.6186, + "step": 2382 + }, + { + "epoch": 1.9732192159028161, + "grad_norm": 0.9011021256446838, + "learning_rate": 3.841831751693466e-06, + "loss": 0.6177, + "step": 2383 + }, + { + "epoch": 1.9740474875759249, + "grad_norm": 0.8824598789215088, + "learning_rate": 3.8409037564121895e-06, + "loss": 0.6392, + "step": 2384 + }, + { + "epoch": 1.9748757592490338, + "grad_norm": 0.8936701416969299, + "learning_rate": 3.839975501678536e-06, + "loss": 0.6508, + "step": 2385 + }, + { + "epoch": 1.9757040309221425, + "grad_norm": 0.9009436368942261, + "learning_rate": 3.839046987672113e-06, + "loss": 0.6169, + "step": 2386 + }, + { + "epoch": 1.9765323025952513, + "grad_norm": 0.8839378356933594, + "learning_rate": 3.8381182145725815e-06, + "loss": 0.6107, + "step": 2387 + }, + { + "epoch": 1.97736057426836, + "grad_norm": 0.9240637421607971, + "learning_rate": 3.837189182559649e-06, + "loss": 0.6083, + "step": 2388 + }, + { + "epoch": 1.9781888459414687, + "grad_norm": 0.881908118724823, + "learning_rate": 3.836259891813075e-06, + "loss": 0.5995, + "step": 2389 + }, + { + "epoch": 1.9790171176145774, + "grad_norm": 0.8754116296768188, + "learning_rate": 3.835330342512667e-06, + "loss": 0.6111, + "step": 2390 + }, + { + "epoch": 1.9798453892876864, + "grad_norm": 0.9713384509086609, + "learning_rate": 3.834400534838287e-06, + "loss": 0.641, + "step": 2391 + }, + { + "epoch": 1.9806736609607951, + "grad_norm": 0.9263403415679932, + "learning_rate": 3.833470468969842e-06, + "loss": 0.6254, + "step": 2392 + }, + { + "epoch": 1.981501932633904, + "grad_norm": 0.9033388495445251, + "learning_rate": 3.832540145087291e-06, + "loss": 0.6049, + "step": 2393 + }, + { + "epoch": 1.9823302043070128, + "grad_norm": 0.8710628151893616, + "learning_rate": 3.8316095633706455e-06, + "loss": 0.5999, + "step": 2394 + }, + { + "epoch": 1.9831584759801215, + "grad_norm": 0.9053885340690613, + "learning_rate": 3.830678723999961e-06, + "loss": 0.6393, + "step": 2395 + }, + { + "epoch": 1.9839867476532302, + "grad_norm": 0.9075314998626709, + "learning_rate": 3.829747627155348e-06, + "loss": 0.6188, + "step": 2396 + }, + { + "epoch": 1.984815019326339, + "grad_norm": 0.9151870012283325, + "learning_rate": 3.828816273016966e-06, + "loss": 0.625, + "step": 2397 + }, + { + "epoch": 1.9856432909994477, + "grad_norm": 0.903843104839325, + "learning_rate": 3.827884661765022e-06, + "loss": 0.618, + "step": 2398 + }, + { + "epoch": 1.9864715626725566, + "grad_norm": 0.875468373298645, + "learning_rate": 3.826952793579774e-06, + "loss": 0.6121, + "step": 2399 + }, + { + "epoch": 1.9872998343456654, + "grad_norm": 0.8908163905143738, + "learning_rate": 3.8260206686415295e-06, + "loss": 0.6292, + "step": 2400 + }, + { + "epoch": 1.9881281060187743, + "grad_norm": 0.8985518217086792, + "learning_rate": 3.825088287130647e-06, + "loss": 0.6356, + "step": 2401 + }, + { + "epoch": 1.988956377691883, + "grad_norm": 0.9277573227882385, + "learning_rate": 3.8241556492275335e-06, + "loss": 0.6424, + "step": 2402 + }, + { + "epoch": 1.9897846493649918, + "grad_norm": 0.9075362682342529, + "learning_rate": 3.823222755112646e-06, + "loss": 0.623, + "step": 2403 + }, + { + "epoch": 1.9906129210381005, + "grad_norm": 0.9038227796554565, + "learning_rate": 3.822289604966489e-06, + "loss": 0.6195, + "step": 2404 + }, + { + "epoch": 1.9914411927112092, + "grad_norm": 0.8869004845619202, + "learning_rate": 3.82135619896962e-06, + "loss": 0.6034, + "step": 2405 + }, + { + "epoch": 1.992269464384318, + "grad_norm": 0.9017077684402466, + "learning_rate": 3.820422537302645e-06, + "loss": 0.6469, + "step": 2406 + }, + { + "epoch": 1.9930977360574267, + "grad_norm": 0.8785225749015808, + "learning_rate": 3.819488620146218e-06, + "loss": 0.6327, + "step": 2407 + }, + { + "epoch": 1.9939260077305356, + "grad_norm": 1.0367780923843384, + "learning_rate": 3.818554447681041e-06, + "loss": 0.6259, + "step": 2408 + }, + { + "epoch": 1.9947542794036444, + "grad_norm": 0.9392277002334595, + "learning_rate": 3.817620020087872e-06, + "loss": 0.6393, + "step": 2409 + }, + { + "epoch": 1.9955825510767533, + "grad_norm": 0.8935515880584717, + "learning_rate": 3.816685337547511e-06, + "loss": 0.6153, + "step": 2410 + }, + { + "epoch": 1.996410822749862, + "grad_norm": 0.8895331025123596, + "learning_rate": 3.815750400240811e-06, + "loss": 0.5978, + "step": 2411 + }, + { + "epoch": 1.9972390944229708, + "grad_norm": 0.8733546733856201, + "learning_rate": 3.814815208348673e-06, + "loss": 0.6254, + "step": 2412 + }, + { + "epoch": 1.9980673660960795, + "grad_norm": 0.9177356958389282, + "learning_rate": 3.8138797620520507e-06, + "loss": 0.631, + "step": 2413 + }, + { + "epoch": 1.9988956377691882, + "grad_norm": 0.9259604811668396, + "learning_rate": 3.8129440615319414e-06, + "loss": 0.6138, + "step": 2414 + }, + { + "epoch": 1.999723909442297, + "grad_norm": 0.9226608872413635, + "learning_rate": 3.812008106969395e-06, + "loss": 0.607, + "step": 2415 + }, + { + "epoch": 2.0, + "grad_norm": 0.9226608872413635, + "learning_rate": 3.8110718985455097e-06, + "loss": 0.5993, + "step": 2416 + }, + { + "epoch": 2.0008282716731087, + "grad_norm": 1.62429940700531, + "learning_rate": 3.810135436441435e-06, + "loss": 0.5954, + "step": 2417 + }, + { + "epoch": 2.0016565433462175, + "grad_norm": 0.8664761781692505, + "learning_rate": 3.8091987208383656e-06, + "loss": 0.5947, + "step": 2418 + }, + { + "epoch": 2.002484815019326, + "grad_norm": 0.9208955764770508, + "learning_rate": 3.808261751917548e-06, + "loss": 0.6028, + "step": 2419 + }, + { + "epoch": 2.003313086692435, + "grad_norm": 0.8842632174491882, + "learning_rate": 3.8073245298602775e-06, + "loss": 0.5913, + "step": 2420 + }, + { + "epoch": 2.004141358365544, + "grad_norm": 0.9232655763626099, + "learning_rate": 3.8063870548478964e-06, + "loss": 0.5977, + "step": 2421 + }, + { + "epoch": 2.004969630038653, + "grad_norm": 0.8898128271102905, + "learning_rate": 3.805449327061798e-06, + "loss": 0.6335, + "step": 2422 + }, + { + "epoch": 2.0057979017117615, + "grad_norm": 0.9356569051742554, + "learning_rate": 3.8045113466834248e-06, + "loss": 0.616, + "step": 2423 + }, + { + "epoch": 2.0066261733848703, + "grad_norm": 0.890315055847168, + "learning_rate": 3.803573113894266e-06, + "loss": 0.5874, + "step": 2424 + }, + { + "epoch": 2.007454445057979, + "grad_norm": 0.8921117186546326, + "learning_rate": 3.802634628875861e-06, + "loss": 0.5679, + "step": 2425 + }, + { + "epoch": 2.0082827167310877, + "grad_norm": 0.8871824145317078, + "learning_rate": 3.8016958918097995e-06, + "loss": 0.6066, + "step": 2426 + }, + { + "epoch": 2.0091109884041964, + "grad_norm": 0.902680516242981, + "learning_rate": 3.800756902877717e-06, + "loss": 0.5936, + "step": 2427 + }, + { + "epoch": 2.009939260077305, + "grad_norm": 0.9199647307395935, + "learning_rate": 3.799817662261298e-06, + "loss": 0.5836, + "step": 2428 + }, + { + "epoch": 2.0107675317504143, + "grad_norm": 0.9029080271720886, + "learning_rate": 3.7988781701422795e-06, + "loss": 0.5858, + "step": 2429 + }, + { + "epoch": 2.011595803423523, + "grad_norm": 0.8834925889968872, + "learning_rate": 3.7979384267024422e-06, + "loss": 0.6061, + "step": 2430 + }, + { + "epoch": 2.012424075096632, + "grad_norm": 0.9098751544952393, + "learning_rate": 3.796998432123619e-06, + "loss": 0.6238, + "step": 2431 + }, + { + "epoch": 2.0132523467697405, + "grad_norm": 0.9316835999488831, + "learning_rate": 3.7960581865876898e-06, + "loss": 0.6139, + "step": 2432 + }, + { + "epoch": 2.0140806184428492, + "grad_norm": 0.9198770523071289, + "learning_rate": 3.795117690276583e-06, + "loss": 0.6113, + "step": 2433 + }, + { + "epoch": 2.014908890115958, + "grad_norm": 0.8970876336097717, + "learning_rate": 3.794176943372275e-06, + "loss": 0.6033, + "step": 2434 + }, + { + "epoch": 2.0157371617890667, + "grad_norm": 0.9025728702545166, + "learning_rate": 3.7932359460567945e-06, + "loss": 0.586, + "step": 2435 + }, + { + "epoch": 2.0165654334621754, + "grad_norm": 0.891909122467041, + "learning_rate": 3.7922946985122124e-06, + "loss": 0.6013, + "step": 2436 + }, + { + "epoch": 2.017393705135284, + "grad_norm": 0.9115222096443176, + "learning_rate": 3.791353200920654e-06, + "loss": 0.5783, + "step": 2437 + }, + { + "epoch": 2.0182219768083933, + "grad_norm": 0.8973708152770996, + "learning_rate": 3.790411453464288e-06, + "loss": 0.5862, + "step": 2438 + }, + { + "epoch": 2.019050248481502, + "grad_norm": 0.910248875617981, + "learning_rate": 3.7894694563253352e-06, + "loss": 0.5884, + "step": 2439 + }, + { + "epoch": 2.019878520154611, + "grad_norm": 0.9227123260498047, + "learning_rate": 3.7885272096860625e-06, + "loss": 0.6185, + "step": 2440 + }, + { + "epoch": 2.0207067918277195, + "grad_norm": 0.917061448097229, + "learning_rate": 3.787584713728787e-06, + "loss": 0.5767, + "step": 2441 + }, + { + "epoch": 2.0215350635008282, + "grad_norm": 0.9035825133323669, + "learning_rate": 3.786641968635871e-06, + "loss": 0.6064, + "step": 2442 + }, + { + "epoch": 2.022363335173937, + "grad_norm": 0.8991772532463074, + "learning_rate": 3.78569897458973e-06, + "loss": 0.5912, + "step": 2443 + }, + { + "epoch": 2.0231916068470457, + "grad_norm": 0.9237397909164429, + "learning_rate": 3.78475573177282e-06, + "loss": 0.5818, + "step": 2444 + }, + { + "epoch": 2.0240198785201544, + "grad_norm": 0.8913205862045288, + "learning_rate": 3.7838122403676543e-06, + "loss": 0.5972, + "step": 2445 + }, + { + "epoch": 2.0248481501932636, + "grad_norm": 0.8960773944854736, + "learning_rate": 3.782868500556787e-06, + "loss": 0.6118, + "step": 2446 + }, + { + "epoch": 2.0256764218663723, + "grad_norm": 0.9198427200317383, + "learning_rate": 3.7819245125228228e-06, + "loss": 0.6058, + "step": 2447 + }, + { + "epoch": 2.026504693539481, + "grad_norm": 0.9039538502693176, + "learning_rate": 3.780980276448416e-06, + "loss": 0.6122, + "step": 2448 + }, + { + "epoch": 2.0273329652125898, + "grad_norm": 0.9103277325630188, + "learning_rate": 3.7800357925162675e-06, + "loss": 0.584, + "step": 2449 + }, + { + "epoch": 2.0281612368856985, + "grad_norm": 0.9114090204238892, + "learning_rate": 3.779091060909124e-06, + "loss": 0.6071, + "step": 2450 + }, + { + "epoch": 2.028989508558807, + "grad_norm": 0.9064953327178955, + "learning_rate": 3.7781460818097847e-06, + "loss": 0.6097, + "step": 2451 + }, + { + "epoch": 2.029817780231916, + "grad_norm": 0.893298864364624, + "learning_rate": 3.7772008554010937e-06, + "loss": 0.5959, + "step": 2452 + }, + { + "epoch": 2.0306460519050247, + "grad_norm": 0.8867486119270325, + "learning_rate": 3.776255381865942e-06, + "loss": 0.5788, + "step": 2453 + }, + { + "epoch": 2.031474323578134, + "grad_norm": 0.9018534421920776, + "learning_rate": 3.775309661387272e-06, + "loss": 0.6075, + "step": 2454 + }, + { + "epoch": 2.0323025952512426, + "grad_norm": 0.9353018999099731, + "learning_rate": 3.77436369414807e-06, + "loss": 0.5894, + "step": 2455 + }, + { + "epoch": 2.0331308669243513, + "grad_norm": 0.8828651309013367, + "learning_rate": 3.7734174803313733e-06, + "loss": 0.608, + "step": 2456 + }, + { + "epoch": 2.03395913859746, + "grad_norm": 0.8971118330955505, + "learning_rate": 3.7724710201202645e-06, + "loss": 0.597, + "step": 2457 + }, + { + "epoch": 2.0347874102705688, + "grad_norm": 0.9064204096794128, + "learning_rate": 3.7715243136978753e-06, + "loss": 0.61, + "step": 2458 + }, + { + "epoch": 2.0356156819436775, + "grad_norm": 0.9204864501953125, + "learning_rate": 3.770577361247384e-06, + "loss": 0.5908, + "step": 2459 + }, + { + "epoch": 2.036443953616786, + "grad_norm": 0.905817449092865, + "learning_rate": 3.7696301629520183e-06, + "loss": 0.5867, + "step": 2460 + }, + { + "epoch": 2.037272225289895, + "grad_norm": 0.8872947096824646, + "learning_rate": 3.7686827189950505e-06, + "loss": 0.5805, + "step": 2461 + }, + { + "epoch": 2.0381004969630037, + "grad_norm": 0.8944478631019592, + "learning_rate": 3.7677350295598035e-06, + "loss": 0.5835, + "step": 2462 + }, + { + "epoch": 2.038928768636113, + "grad_norm": 0.884630560874939, + "learning_rate": 3.766787094829646e-06, + "loss": 0.5909, + "step": 2463 + }, + { + "epoch": 2.0397570403092216, + "grad_norm": 0.8915899395942688, + "learning_rate": 3.7658389149879936e-06, + "loss": 0.5923, + "step": 2464 + }, + { + "epoch": 2.0405853119823303, + "grad_norm": 0.8971114158630371, + "learning_rate": 3.7648904902183128e-06, + "loss": 0.6097, + "step": 2465 + }, + { + "epoch": 2.041413583655439, + "grad_norm": 0.9223624467849731, + "learning_rate": 3.7639418207041124e-06, + "loss": 0.6048, + "step": 2466 + }, + { + "epoch": 2.0422418553285477, + "grad_norm": 0.915450930595398, + "learning_rate": 3.7629929066289517e-06, + "loss": 0.626, + "step": 2467 + }, + { + "epoch": 2.0430701270016565, + "grad_norm": 0.919333279132843, + "learning_rate": 3.762043748176437e-06, + "loss": 0.6234, + "step": 2468 + }, + { + "epoch": 2.043898398674765, + "grad_norm": 0.9282277822494507, + "learning_rate": 3.7610943455302223e-06, + "loss": 0.6008, + "step": 2469 + }, + { + "epoch": 2.044726670347874, + "grad_norm": 0.9419016242027283, + "learning_rate": 3.760144698874007e-06, + "loss": 0.5836, + "step": 2470 + }, + { + "epoch": 2.045554942020983, + "grad_norm": 0.91156405210495, + "learning_rate": 3.7591948083915403e-06, + "loss": 0.6035, + "step": 2471 + }, + { + "epoch": 2.046383213694092, + "grad_norm": 0.9019245505332947, + "learning_rate": 3.758244674266615e-06, + "loss": 0.5949, + "step": 2472 + }, + { + "epoch": 2.0472114853672005, + "grad_norm": 0.8976905345916748, + "learning_rate": 3.757294296683075e-06, + "loss": 0.5818, + "step": 2473 + }, + { + "epoch": 2.0480397570403093, + "grad_norm": 0.8748987913131714, + "learning_rate": 3.7563436758248094e-06, + "loss": 0.5816, + "step": 2474 + }, + { + "epoch": 2.048868028713418, + "grad_norm": 0.8929162621498108, + "learning_rate": 3.7553928118757532e-06, + "loss": 0.5869, + "step": 2475 + }, + { + "epoch": 2.0496963003865267, + "grad_norm": 0.9968146681785583, + "learning_rate": 3.7544417050198913e-06, + "loss": 0.6065, + "step": 2476 + }, + { + "epoch": 2.0505245720596355, + "grad_norm": 0.9206544756889343, + "learning_rate": 3.753490355441253e-06, + "loss": 0.5919, + "step": 2477 + }, + { + "epoch": 2.051352843732744, + "grad_norm": 0.8804112672805786, + "learning_rate": 3.7525387633239153e-06, + "loss": 0.5918, + "step": 2478 + }, + { + "epoch": 2.052181115405853, + "grad_norm": 0.922532320022583, + "learning_rate": 3.7515869288520027e-06, + "loss": 0.6212, + "step": 2479 + }, + { + "epoch": 2.053009387078962, + "grad_norm": 0.9353113770484924, + "learning_rate": 3.7506348522096876e-06, + "loss": 0.5695, + "step": 2480 + }, + { + "epoch": 2.053837658752071, + "grad_norm": 0.9110690951347351, + "learning_rate": 3.749682533581186e-06, + "loss": 0.6156, + "step": 2481 + }, + { + "epoch": 2.0546659304251795, + "grad_norm": 0.9076771140098572, + "learning_rate": 3.7487299731507642e-06, + "loss": 0.5987, + "step": 2482 + }, + { + "epoch": 2.0554942020982883, + "grad_norm": 0.8937394618988037, + "learning_rate": 3.747777171102732e-06, + "loss": 0.6057, + "step": 2483 + }, + { + "epoch": 2.056322473771397, + "grad_norm": 0.9488244652748108, + "learning_rate": 3.74682412762145e-06, + "loss": 0.5983, + "step": 2484 + }, + { + "epoch": 2.0571507454445057, + "grad_norm": 0.897693395614624, + "learning_rate": 3.7458708428913217e-06, + "loss": 0.606, + "step": 2485 + }, + { + "epoch": 2.0579790171176144, + "grad_norm": 0.8989591598510742, + "learning_rate": 3.744917317096799e-06, + "loss": 0.6138, + "step": 2486 + }, + { + "epoch": 2.058807288790723, + "grad_norm": 0.9028336405754089, + "learning_rate": 3.7439635504223804e-06, + "loss": 0.5714, + "step": 2487 + }, + { + "epoch": 2.0596355604638323, + "grad_norm": 0.8895571231842041, + "learning_rate": 3.7430095430526113e-06, + "loss": 0.5903, + "step": 2488 + }, + { + "epoch": 2.060463832136941, + "grad_norm": 0.926414430141449, + "learning_rate": 3.7420552951720822e-06, + "loss": 0.5841, + "step": 2489 + }, + { + "epoch": 2.06129210381005, + "grad_norm": 0.9279632568359375, + "learning_rate": 3.741100806965432e-06, + "loss": 0.6062, + "step": 2490 + }, + { + "epoch": 2.0621203754831585, + "grad_norm": 0.9196921586990356, + "learning_rate": 3.7401460786173456e-06, + "loss": 0.579, + "step": 2491 + }, + { + "epoch": 2.0629486471562672, + "grad_norm": 0.8959674835205078, + "learning_rate": 3.739191110312552e-06, + "loss": 0.5888, + "step": 2492 + }, + { + "epoch": 2.063776918829376, + "grad_norm": 0.9123788475990295, + "learning_rate": 3.7382359022358322e-06, + "loss": 0.5891, + "step": 2493 + }, + { + "epoch": 2.0646051905024847, + "grad_norm": 0.9637120366096497, + "learning_rate": 3.7372804545720067e-06, + "loss": 0.6124, + "step": 2494 + }, + { + "epoch": 2.0654334621755934, + "grad_norm": 0.9360100626945496, + "learning_rate": 3.736324767505947e-06, + "loss": 0.5884, + "step": 2495 + }, + { + "epoch": 2.0662617338487026, + "grad_norm": 0.9066753387451172, + "learning_rate": 3.7353688412225704e-06, + "loss": 0.6013, + "step": 2496 + }, + { + "epoch": 2.0670900055218113, + "grad_norm": 0.8961504101753235, + "learning_rate": 3.7344126759068382e-06, + "loss": 0.6303, + "step": 2497 + }, + { + "epoch": 2.06791827719492, + "grad_norm": 0.8995484709739685, + "learning_rate": 3.7334562717437608e-06, + "loss": 0.6031, + "step": 2498 + }, + { + "epoch": 2.068746548868029, + "grad_norm": 0.9279561042785645, + "learning_rate": 3.7324996289183927e-06, + "loss": 0.5848, + "step": 2499 + }, + { + "epoch": 2.0695748205411375, + "grad_norm": 0.9250274300575256, + "learning_rate": 3.7315427476158357e-06, + "loss": 0.5949, + "step": 2500 + }, + { + "epoch": 2.0704030922142462, + "grad_norm": 0.9003238677978516, + "learning_rate": 3.730585628021237e-06, + "loss": 0.6124, + "step": 2501 + }, + { + "epoch": 2.071231363887355, + "grad_norm": 0.9069858193397522, + "learning_rate": 3.729628270319791e-06, + "loss": 0.6051, + "step": 2502 + }, + { + "epoch": 2.0720596355604637, + "grad_norm": 0.9361990094184875, + "learning_rate": 3.7286706746967372e-06, + "loss": 0.5863, + "step": 2503 + }, + { + "epoch": 2.0728879072335724, + "grad_norm": 0.9443079233169556, + "learning_rate": 3.7277128413373607e-06, + "loss": 0.5959, + "step": 2504 + }, + { + "epoch": 2.0737161789066816, + "grad_norm": 0.9413166046142578, + "learning_rate": 3.7267547704269957e-06, + "loss": 0.5882, + "step": 2505 + }, + { + "epoch": 2.0745444505797903, + "grad_norm": 0.8937835097312927, + "learning_rate": 3.7257964621510167e-06, + "loss": 0.6108, + "step": 2506 + }, + { + "epoch": 2.075372722252899, + "grad_norm": 0.9188135862350464, + "learning_rate": 3.7248379166948497e-06, + "loss": 0.5812, + "step": 2507 + }, + { + "epoch": 2.0762009939260078, + "grad_norm": 0.916559636592865, + "learning_rate": 3.723879134243963e-06, + "loss": 0.6169, + "step": 2508 + }, + { + "epoch": 2.0770292655991165, + "grad_norm": 0.9284812211990356, + "learning_rate": 3.7229201149838727e-06, + "loss": 0.6224, + "step": 2509 + }, + { + "epoch": 2.077857537272225, + "grad_norm": 0.918048083782196, + "learning_rate": 3.721960859100141e-06, + "loss": 0.6043, + "step": 2510 + }, + { + "epoch": 2.078685808945334, + "grad_norm": 0.9250833988189697, + "learning_rate": 3.721001366778373e-06, + "loss": 0.5882, + "step": 2511 + }, + { + "epoch": 2.0795140806184427, + "grad_norm": 0.8975113034248352, + "learning_rate": 3.720041638204223e-06, + "loss": 0.6049, + "step": 2512 + }, + { + "epoch": 2.080342352291552, + "grad_norm": 0.9108197093009949, + "learning_rate": 3.7190816735633896e-06, + "loss": 0.6174, + "step": 2513 + }, + { + "epoch": 2.0811706239646606, + "grad_norm": 0.91886967420578, + "learning_rate": 3.718121473041615e-06, + "loss": 0.5854, + "step": 2514 + }, + { + "epoch": 2.0819988956377693, + "grad_norm": 0.9149959683418274, + "learning_rate": 3.717161036824692e-06, + "loss": 0.5808, + "step": 2515 + }, + { + "epoch": 2.082827167310878, + "grad_norm": 0.9402881860733032, + "learning_rate": 3.7162003650984545e-06, + "loss": 0.5936, + "step": 2516 + }, + { + "epoch": 2.0836554389839868, + "grad_norm": 0.8952768445014954, + "learning_rate": 3.7152394580487825e-06, + "loss": 0.5977, + "step": 2517 + }, + { + "epoch": 2.0844837106570955, + "grad_norm": 0.9138445258140564, + "learning_rate": 3.714278315861605e-06, + "loss": 0.5822, + "step": 2518 + }, + { + "epoch": 2.085311982330204, + "grad_norm": 0.9174036979675293, + "learning_rate": 3.713316938722892e-06, + "loss": 0.6061, + "step": 2519 + }, + { + "epoch": 2.086140254003313, + "grad_norm": 0.9291302561759949, + "learning_rate": 3.7123553268186618e-06, + "loss": 0.614, + "step": 2520 + }, + { + "epoch": 2.0869685256764217, + "grad_norm": 0.9058101773262024, + "learning_rate": 3.711393480334977e-06, + "loss": 0.6053, + "step": 2521 + }, + { + "epoch": 2.087796797349531, + "grad_norm": 0.926360011100769, + "learning_rate": 3.7104313994579473e-06, + "loss": 0.5781, + "step": 2522 + }, + { + "epoch": 2.0886250690226396, + "grad_norm": 0.898629367351532, + "learning_rate": 3.709469084373725e-06, + "loss": 0.5995, + "step": 2523 + }, + { + "epoch": 2.0894533406957483, + "grad_norm": 0.9013628363609314, + "learning_rate": 3.708506535268509e-06, + "loss": 0.5929, + "step": 2524 + }, + { + "epoch": 2.090281612368857, + "grad_norm": 0.8909673094749451, + "learning_rate": 3.7075437523285433e-06, + "loss": 0.5495, + "step": 2525 + }, + { + "epoch": 2.0911098840419657, + "grad_norm": 0.9176347255706787, + "learning_rate": 3.706580735740119e-06, + "loss": 0.6134, + "step": 2526 + }, + { + "epoch": 2.0919381557150745, + "grad_norm": 0.9174275398254395, + "learning_rate": 3.7056174856895703e-06, + "loss": 0.601, + "step": 2527 + }, + { + "epoch": 2.092766427388183, + "grad_norm": 0.9134400486946106, + "learning_rate": 3.704654002363276e-06, + "loss": 0.5952, + "step": 2528 + }, + { + "epoch": 2.093594699061292, + "grad_norm": 0.9206627607345581, + "learning_rate": 3.703690285947662e-06, + "loss": 0.6016, + "step": 2529 + }, + { + "epoch": 2.094422970734401, + "grad_norm": 0.9271345734596252, + "learning_rate": 3.702726336629199e-06, + "loss": 0.5937, + "step": 2530 + }, + { + "epoch": 2.09525124240751, + "grad_norm": 0.9390851855278015, + "learning_rate": 3.7017621545944003e-06, + "loss": 0.5858, + "step": 2531 + }, + { + "epoch": 2.0960795140806185, + "grad_norm": 0.9351701736450195, + "learning_rate": 3.700797740029828e-06, + "loss": 0.5978, + "step": 2532 + }, + { + "epoch": 2.0969077857537273, + "grad_norm": 0.9432377815246582, + "learning_rate": 3.6998330931220873e-06, + "loss": 0.5884, + "step": 2533 + }, + { + "epoch": 2.097736057426836, + "grad_norm": 0.9192067384719849, + "learning_rate": 3.698868214057827e-06, + "loss": 0.5989, + "step": 2534 + }, + { + "epoch": 2.0985643290999447, + "grad_norm": 0.9810166954994202, + "learning_rate": 3.6979031030237433e-06, + "loss": 0.6078, + "step": 2535 + }, + { + "epoch": 2.0993926007730535, + "grad_norm": 0.9065185785293579, + "learning_rate": 3.696937760206576e-06, + "loss": 0.5947, + "step": 2536 + }, + { + "epoch": 2.100220872446162, + "grad_norm": 0.9212245345115662, + "learning_rate": 3.6959721857931088e-06, + "loss": 0.5951, + "step": 2537 + }, + { + "epoch": 2.101049144119271, + "grad_norm": 0.9291266798973083, + "learning_rate": 3.695006379970173e-06, + "loss": 0.6033, + "step": 2538 + }, + { + "epoch": 2.10187741579238, + "grad_norm": 0.9154751896858215, + "learning_rate": 3.694040342924643e-06, + "loss": 0.612, + "step": 2539 + }, + { + "epoch": 2.102705687465489, + "grad_norm": 0.9028742909431458, + "learning_rate": 3.693074074843437e-06, + "loss": 0.5948, + "step": 2540 + }, + { + "epoch": 2.1035339591385975, + "grad_norm": 0.9132126569747925, + "learning_rate": 3.692107575913519e-06, + "loss": 0.6198, + "step": 2541 + }, + { + "epoch": 2.1043622308117063, + "grad_norm": 0.9116196036338806, + "learning_rate": 3.6911408463218974e-06, + "loss": 0.5927, + "step": 2542 + }, + { + "epoch": 2.105190502484815, + "grad_norm": 0.9670912027359009, + "learning_rate": 3.6901738862556263e-06, + "loss": 0.5813, + "step": 2543 + }, + { + "epoch": 2.1060187741579237, + "grad_norm": 0.9534422755241394, + "learning_rate": 3.6892066959018034e-06, + "loss": 0.5892, + "step": 2544 + }, + { + "epoch": 2.1068470458310324, + "grad_norm": 0.9175525903701782, + "learning_rate": 3.688239275447569e-06, + "loss": 0.5946, + "step": 2545 + }, + { + "epoch": 2.107675317504141, + "grad_norm": 0.8945409655570984, + "learning_rate": 3.6872716250801125e-06, + "loss": 0.6013, + "step": 2546 + }, + { + "epoch": 2.1085035891772503, + "grad_norm": 0.9498820900917053, + "learning_rate": 3.686303744986664e-06, + "loss": 0.5785, + "step": 2547 + }, + { + "epoch": 2.109331860850359, + "grad_norm": 0.9338169693946838, + "learning_rate": 3.685335635354499e-06, + "loss": 0.5959, + "step": 2548 + }, + { + "epoch": 2.110160132523468, + "grad_norm": 0.9100049734115601, + "learning_rate": 3.684367296370939e-06, + "loss": 0.5829, + "step": 2549 + }, + { + "epoch": 2.1109884041965765, + "grad_norm": 0.904279887676239, + "learning_rate": 3.6833987282233467e-06, + "loss": 0.6015, + "step": 2550 + }, + { + "epoch": 2.1118166758696852, + "grad_norm": 0.8693552613258362, + "learning_rate": 3.6824299310991324e-06, + "loss": 0.5825, + "step": 2551 + }, + { + "epoch": 2.112644947542794, + "grad_norm": 0.9180129766464233, + "learning_rate": 3.681460905185749e-06, + "loss": 0.6076, + "step": 2552 + }, + { + "epoch": 2.1134732192159027, + "grad_norm": 0.9554781913757324, + "learning_rate": 3.6804916506706936e-06, + "loss": 0.5755, + "step": 2553 + }, + { + "epoch": 2.1143014908890114, + "grad_norm": 0.9498181939125061, + "learning_rate": 3.6795221677415085e-06, + "loss": 0.6086, + "step": 2554 + }, + { + "epoch": 2.1151297625621206, + "grad_norm": 0.9110347032546997, + "learning_rate": 3.678552456585779e-06, + "loss": 0.591, + "step": 2555 + }, + { + "epoch": 2.1159580342352293, + "grad_norm": 0.8883147835731506, + "learning_rate": 3.6775825173911363e-06, + "loss": 0.5753, + "step": 2556 + }, + { + "epoch": 2.116786305908338, + "grad_norm": 0.9354976415634155, + "learning_rate": 3.676612350345253e-06, + "loss": 0.6063, + "step": 2557 + }, + { + "epoch": 2.1176145775814468, + "grad_norm": 0.9463969469070435, + "learning_rate": 3.675641955635848e-06, + "loss": 0.5983, + "step": 2558 + }, + { + "epoch": 2.1184428492545555, + "grad_norm": 0.9220679998397827, + "learning_rate": 3.674671333450684e-06, + "loss": 0.6026, + "step": 2559 + }, + { + "epoch": 2.1192711209276642, + "grad_norm": 0.9050368666648865, + "learning_rate": 3.673700483977567e-06, + "loss": 0.6075, + "step": 2560 + }, + { + "epoch": 2.120099392600773, + "grad_norm": 0.9299139976501465, + "learning_rate": 3.6727294074043478e-06, + "loss": 0.5971, + "step": 2561 + }, + { + "epoch": 2.1209276642738817, + "grad_norm": 0.9060289263725281, + "learning_rate": 3.6717581039189205e-06, + "loss": 0.5794, + "step": 2562 + }, + { + "epoch": 2.1217559359469904, + "grad_norm": 0.9253103137016296, + "learning_rate": 3.6707865737092228e-06, + "loss": 0.593, + "step": 2563 + }, + { + "epoch": 2.1225842076200996, + "grad_norm": 0.9475048780441284, + "learning_rate": 3.669814816963237e-06, + "loss": 0.5867, + "step": 2564 + }, + { + "epoch": 2.1234124792932083, + "grad_norm": 0.9546397924423218, + "learning_rate": 3.6688428338689887e-06, + "loss": 0.5766, + "step": 2565 + }, + { + "epoch": 2.124240750966317, + "grad_norm": 0.9377962350845337, + "learning_rate": 3.667870624614548e-06, + "loss": 0.5853, + "step": 2566 + }, + { + "epoch": 2.1250690226394258, + "grad_norm": 0.9562273621559143, + "learning_rate": 3.6668981893880284e-06, + "loss": 0.5992, + "step": 2567 + }, + { + "epoch": 2.1258972943125345, + "grad_norm": 0.9207432866096497, + "learning_rate": 3.6659255283775867e-06, + "loss": 0.6029, + "step": 2568 + }, + { + "epoch": 2.126725565985643, + "grad_norm": 0.9359074234962463, + "learning_rate": 3.6649526417714244e-06, + "loss": 0.6068, + "step": 2569 + }, + { + "epoch": 2.127553837658752, + "grad_norm": 0.9253597259521484, + "learning_rate": 3.663979529757784e-06, + "loss": 0.581, + "step": 2570 + }, + { + "epoch": 2.1283821093318607, + "grad_norm": 0.9551291465759277, + "learning_rate": 3.6630061925249566e-06, + "loss": 0.607, + "step": 2571 + }, + { + "epoch": 2.1292103810049694, + "grad_norm": 0.91109699010849, + "learning_rate": 3.662032630261272e-06, + "loss": 0.5724, + "step": 2572 + }, + { + "epoch": 2.1300386526780786, + "grad_norm": 0.9486557245254517, + "learning_rate": 3.661058843155105e-06, + "loss": 0.5904, + "step": 2573 + }, + { + "epoch": 2.1308669243511873, + "grad_norm": 0.9550185203552246, + "learning_rate": 3.6600848313948762e-06, + "loss": 0.5967, + "step": 2574 + }, + { + "epoch": 2.131695196024296, + "grad_norm": 0.9457482695579529, + "learning_rate": 3.6591105951690464e-06, + "loss": 0.6253, + "step": 2575 + }, + { + "epoch": 2.1325234676974048, + "grad_norm": 0.9686241149902344, + "learning_rate": 3.6581361346661204e-06, + "loss": 0.5928, + "step": 2576 + }, + { + "epoch": 2.1333517393705135, + "grad_norm": 0.9125609993934631, + "learning_rate": 3.65716145007465e-06, + "loss": 0.6039, + "step": 2577 + }, + { + "epoch": 2.134180011043622, + "grad_norm": 0.927651047706604, + "learning_rate": 3.656186541583225e-06, + "loss": 0.5881, + "step": 2578 + }, + { + "epoch": 2.135008282716731, + "grad_norm": 0.9446879029273987, + "learning_rate": 3.6552114093804825e-06, + "loss": 0.5938, + "step": 2579 + }, + { + "epoch": 2.13583655438984, + "grad_norm": 0.944601833820343, + "learning_rate": 3.6542360536551003e-06, + "loss": 0.5804, + "step": 2580 + }, + { + "epoch": 2.136664826062949, + "grad_norm": 0.9230138063430786, + "learning_rate": 3.6532604745958015e-06, + "loss": 0.6002, + "step": 2581 + }, + { + "epoch": 2.1374930977360576, + "grad_norm": 0.9163529276847839, + "learning_rate": 3.6522846723913517e-06, + "loss": 0.5811, + "step": 2582 + }, + { + "epoch": 2.1383213694091663, + "grad_norm": 0.9153618812561035, + "learning_rate": 3.6513086472305596e-06, + "loss": 0.6018, + "step": 2583 + }, + { + "epoch": 2.139149641082275, + "grad_norm": 0.9299359917640686, + "learning_rate": 3.6503323993022767e-06, + "loss": 0.5944, + "step": 2584 + }, + { + "epoch": 2.1399779127553837, + "grad_norm": 0.9075869917869568, + "learning_rate": 3.6493559287953974e-06, + "loss": 0.595, + "step": 2585 + }, + { + "epoch": 2.1408061844284925, + "grad_norm": 0.9388751983642578, + "learning_rate": 3.64837923589886e-06, + "loss": 0.6006, + "step": 2586 + }, + { + "epoch": 2.141634456101601, + "grad_norm": 0.91156405210495, + "learning_rate": 3.647402320801645e-06, + "loss": 0.5955, + "step": 2587 + }, + { + "epoch": 2.14246272777471, + "grad_norm": 0.9187628626823425, + "learning_rate": 3.6464251836927776e-06, + "loss": 0.6119, + "step": 2588 + }, + { + "epoch": 2.143290999447819, + "grad_norm": 0.9578025341033936, + "learning_rate": 3.645447824761324e-06, + "loss": 0.6076, + "step": 2589 + }, + { + "epoch": 2.144119271120928, + "grad_norm": 0.9159366488456726, + "learning_rate": 3.644470244196394e-06, + "loss": 0.6015, + "step": 2590 + }, + { + "epoch": 2.1449475427940365, + "grad_norm": 0.924027681350708, + "learning_rate": 3.6434924421871403e-06, + "loss": 0.6082, + "step": 2591 + }, + { + "epoch": 2.1457758144671453, + "grad_norm": 0.8900635242462158, + "learning_rate": 3.642514418922758e-06, + "loss": 0.611, + "step": 2592 + }, + { + "epoch": 2.146604086140254, + "grad_norm": 0.9609982371330261, + "learning_rate": 3.641536174592487e-06, + "loss": 0.5974, + "step": 2593 + }, + { + "epoch": 2.1474323578133627, + "grad_norm": 0.901737630367279, + "learning_rate": 3.6405577093856072e-06, + "loss": 0.5869, + "step": 2594 + }, + { + "epoch": 2.1482606294864715, + "grad_norm": 0.9342123866081238, + "learning_rate": 3.6395790234914423e-06, + "loss": 0.6104, + "step": 2595 + }, + { + "epoch": 2.14908890115958, + "grad_norm": 0.9063602685928345, + "learning_rate": 3.6386001170993596e-06, + "loss": 0.5801, + "step": 2596 + }, + { + "epoch": 2.149917172832689, + "grad_norm": 0.9269636869430542, + "learning_rate": 3.637620990398768e-06, + "loss": 0.5953, + "step": 2597 + }, + { + "epoch": 2.150745444505798, + "grad_norm": 0.9278765320777893, + "learning_rate": 3.63664164357912e-06, + "loss": 0.6151, + "step": 2598 + }, + { + "epoch": 2.151573716178907, + "grad_norm": 0.9786942601203918, + "learning_rate": 3.6356620768299087e-06, + "loss": 0.5773, + "step": 2599 + }, + { + "epoch": 2.1524019878520155, + "grad_norm": 0.9069851636886597, + "learning_rate": 3.6346822903406733e-06, + "loss": 0.6087, + "step": 2600 + }, + { + "epoch": 2.1532302595251243, + "grad_norm": 0.8867887854576111, + "learning_rate": 3.6337022843009913e-06, + "loss": 0.5894, + "step": 2601 + }, + { + "epoch": 2.154058531198233, + "grad_norm": 0.9444507360458374, + "learning_rate": 3.632722058900486e-06, + "loss": 0.6125, + "step": 2602 + }, + { + "epoch": 2.1548868028713417, + "grad_norm": 0.9410359859466553, + "learning_rate": 3.631741614328821e-06, + "loss": 0.5845, + "step": 2603 + }, + { + "epoch": 2.1557150745444504, + "grad_norm": 0.9473987817764282, + "learning_rate": 3.630760950775703e-06, + "loss": 0.5944, + "step": 2604 + }, + { + "epoch": 2.156543346217559, + "grad_norm": 0.913758397102356, + "learning_rate": 3.6297800684308828e-06, + "loss": 0.5998, + "step": 2605 + }, + { + "epoch": 2.1573716178906683, + "grad_norm": 0.89509516954422, + "learning_rate": 3.6287989674841505e-06, + "loss": 0.6013, + "step": 2606 + }, + { + "epoch": 2.158199889563777, + "grad_norm": 0.9042367339134216, + "learning_rate": 3.6278176481253413e-06, + "loss": 0.5915, + "step": 2607 + }, + { + "epoch": 2.159028161236886, + "grad_norm": 0.9108945727348328, + "learning_rate": 3.6268361105443297e-06, + "loss": 0.5803, + "step": 2608 + }, + { + "epoch": 2.1598564329099945, + "grad_norm": 0.8903773427009583, + "learning_rate": 3.625854354931035e-06, + "loss": 0.6025, + "step": 2609 + }, + { + "epoch": 2.1606847045831032, + "grad_norm": 0.9396197199821472, + "learning_rate": 3.6248723814754183e-06, + "loss": 0.6271, + "step": 2610 + }, + { + "epoch": 2.161512976256212, + "grad_norm": 0.9247238039970398, + "learning_rate": 3.623890190367481e-06, + "loss": 0.5945, + "step": 2611 + }, + { + "epoch": 2.1623412479293207, + "grad_norm": 0.9240396022796631, + "learning_rate": 3.622907781797269e-06, + "loss": 0.6148, + "step": 2612 + }, + { + "epoch": 2.1631695196024294, + "grad_norm": 0.9257000684738159, + "learning_rate": 3.621925155954869e-06, + "loss": 0.5964, + "step": 2613 + }, + { + "epoch": 2.1639977912755386, + "grad_norm": 0.8854860663414001, + "learning_rate": 3.62094231303041e-06, + "loss": 0.6012, + "step": 2614 + }, + { + "epoch": 2.1648260629486473, + "grad_norm": 0.9251745343208313, + "learning_rate": 3.619959253214062e-06, + "loss": 0.6001, + "step": 2615 + }, + { + "epoch": 2.165654334621756, + "grad_norm": 0.9081252813339233, + "learning_rate": 3.618975976696039e-06, + "loss": 0.5927, + "step": 2616 + }, + { + "epoch": 2.1664826062948648, + "grad_norm": 0.9065699577331543, + "learning_rate": 3.6179924836665953e-06, + "loss": 0.6079, + "step": 2617 + }, + { + "epoch": 2.1673108779679735, + "grad_norm": 0.9421206712722778, + "learning_rate": 3.617008774316028e-06, + "loss": 0.6041, + "step": 2618 + }, + { + "epoch": 2.1681391496410822, + "grad_norm": 0.8863933682441711, + "learning_rate": 3.6160248488346757e-06, + "loss": 0.5826, + "step": 2619 + }, + { + "epoch": 2.168967421314191, + "grad_norm": 0.915887713432312, + "learning_rate": 3.615040707412918e-06, + "loss": 0.5972, + "step": 2620 + }, + { + "epoch": 2.1697956929872997, + "grad_norm": 0.9150416851043701, + "learning_rate": 3.614056350241178e-06, + "loss": 0.606, + "step": 2621 + }, + { + "epoch": 2.1706239646604084, + "grad_norm": 0.9050796627998352, + "learning_rate": 3.6130717775099195e-06, + "loss": 0.5955, + "step": 2622 + }, + { + "epoch": 2.1714522363335176, + "grad_norm": 0.9249832630157471, + "learning_rate": 3.612086989409648e-06, + "loss": 0.5968, + "step": 2623 + }, + { + "epoch": 2.1722805080066263, + "grad_norm": 0.8855478763580322, + "learning_rate": 3.6111019861309105e-06, + "loss": 0.5657, + "step": 2624 + }, + { + "epoch": 2.173108779679735, + "grad_norm": 0.9182875156402588, + "learning_rate": 3.6101167678642967e-06, + "loss": 0.6096, + "step": 2625 + }, + { + "epoch": 2.1739370513528438, + "grad_norm": 0.9722150564193726, + "learning_rate": 3.609131334800436e-06, + "loss": 0.6029, + "step": 2626 + }, + { + "epoch": 2.1747653230259525, + "grad_norm": 0.8925914168357849, + "learning_rate": 3.6081456871300014e-06, + "loss": 0.5928, + "step": 2627 + }, + { + "epoch": 2.175593594699061, + "grad_norm": 0.8883031010627747, + "learning_rate": 3.6071598250437067e-06, + "loss": 0.5763, + "step": 2628 + }, + { + "epoch": 2.17642186637217, + "grad_norm": 0.8948857188224792, + "learning_rate": 3.606173748732307e-06, + "loss": 0.6078, + "step": 2629 + }, + { + "epoch": 2.1772501380452787, + "grad_norm": 0.9444127082824707, + "learning_rate": 3.6051874583865975e-06, + "loss": 0.6005, + "step": 2630 + }, + { + "epoch": 2.1780784097183874, + "grad_norm": 0.9361352920532227, + "learning_rate": 3.6042009541974176e-06, + "loss": 0.5878, + "step": 2631 + }, + { + "epoch": 2.1789066813914966, + "grad_norm": 0.9124311804771423, + "learning_rate": 3.6032142363556467e-06, + "loss": 0.6115, + "step": 2632 + }, + { + "epoch": 2.1797349530646053, + "grad_norm": 0.9168087840080261, + "learning_rate": 3.602227305052205e-06, + "loss": 0.5928, + "step": 2633 + }, + { + "epoch": 2.180563224737714, + "grad_norm": 0.9189275503158569, + "learning_rate": 3.6012401604780543e-06, + "loss": 0.6081, + "step": 2634 + }, + { + "epoch": 2.1813914964108228, + "grad_norm": 0.937911868095398, + "learning_rate": 3.6002528028241984e-06, + "loss": 0.6062, + "step": 2635 + }, + { + "epoch": 2.1822197680839315, + "grad_norm": 0.9139302968978882, + "learning_rate": 3.5992652322816817e-06, + "loss": 0.6111, + "step": 2636 + }, + { + "epoch": 2.18304803975704, + "grad_norm": 0.951528012752533, + "learning_rate": 3.598277449041589e-06, + "loss": 0.6007, + "step": 2637 + }, + { + "epoch": 2.183876311430149, + "grad_norm": 0.9146255850791931, + "learning_rate": 3.5972894532950487e-06, + "loss": 0.6094, + "step": 2638 + }, + { + "epoch": 2.184704583103258, + "grad_norm": 0.9217225313186646, + "learning_rate": 3.596301245233228e-06, + "loss": 0.6082, + "step": 2639 + }, + { + "epoch": 2.185532854776367, + "grad_norm": 0.9742298126220703, + "learning_rate": 3.5953128250473355e-06, + "loss": 0.6269, + "step": 2640 + }, + { + "epoch": 2.1863611264494756, + "grad_norm": 0.9038170576095581, + "learning_rate": 3.5943241929286223e-06, + "loss": 0.6062, + "step": 2641 + }, + { + "epoch": 2.1871893981225843, + "grad_norm": 0.8960873484611511, + "learning_rate": 3.5933353490683786e-06, + "loss": 0.5748, + "step": 2642 + }, + { + "epoch": 2.188017669795693, + "grad_norm": 0.8801732659339905, + "learning_rate": 3.5923462936579362e-06, + "loss": 0.5911, + "step": 2643 + }, + { + "epoch": 2.1888459414688017, + "grad_norm": 0.8918222188949585, + "learning_rate": 3.5913570268886695e-06, + "loss": 0.5931, + "step": 2644 + }, + { + "epoch": 2.1896742131419105, + "grad_norm": 0.9046133160591125, + "learning_rate": 3.5903675489519917e-06, + "loss": 0.5985, + "step": 2645 + }, + { + "epoch": 2.190502484815019, + "grad_norm": 0.9003219604492188, + "learning_rate": 3.589377860039357e-06, + "loss": 0.608, + "step": 2646 + }, + { + "epoch": 2.191330756488128, + "grad_norm": 0.9139525890350342, + "learning_rate": 3.5883879603422616e-06, + "loss": 0.6186, + "step": 2647 + }, + { + "epoch": 2.192159028161237, + "grad_norm": 0.9125071167945862, + "learning_rate": 3.587397850052241e-06, + "loss": 0.6129, + "step": 2648 + }, + { + "epoch": 2.192987299834346, + "grad_norm": 0.922412097454071, + "learning_rate": 3.586407529360874e-06, + "loss": 0.6008, + "step": 2649 + }, + { + "epoch": 2.1938155715074545, + "grad_norm": 0.9294847249984741, + "learning_rate": 3.5854169984597776e-06, + "loss": 0.5984, + "step": 2650 + }, + { + "epoch": 2.1946438431805633, + "grad_norm": 0.9262758493423462, + "learning_rate": 3.5844262575406084e-06, + "loss": 0.5982, + "step": 2651 + }, + { + "epoch": 2.195472114853672, + "grad_norm": 0.9311552047729492, + "learning_rate": 3.5834353067950694e-06, + "loss": 0.5965, + "step": 2652 + }, + { + "epoch": 2.1963003865267807, + "grad_norm": 0.9087333679199219, + "learning_rate": 3.5824441464148975e-06, + "loss": 0.5972, + "step": 2653 + }, + { + "epoch": 2.1971286581998895, + "grad_norm": 0.9212601184844971, + "learning_rate": 3.5814527765918733e-06, + "loss": 0.5821, + "step": 2654 + }, + { + "epoch": 2.197956929872998, + "grad_norm": 0.9208887219429016, + "learning_rate": 3.5804611975178184e-06, + "loss": 0.592, + "step": 2655 + }, + { + "epoch": 2.198785201546107, + "grad_norm": 0.923478364944458, + "learning_rate": 3.5794694093845933e-06, + "loss": 0.599, + "step": 2656 + }, + { + "epoch": 2.199613473219216, + "grad_norm": 0.9030081033706665, + "learning_rate": 3.5784774123841006e-06, + "loss": 0.5954, + "step": 2657 + }, + { + "epoch": 2.200441744892325, + "grad_norm": 0.9108036160469055, + "learning_rate": 3.5774852067082817e-06, + "loss": 0.6119, + "step": 2658 + }, + { + "epoch": 2.2012700165654335, + "grad_norm": 0.9323570728302002, + "learning_rate": 3.576492792549119e-06, + "loss": 0.6061, + "step": 2659 + }, + { + "epoch": 2.2020982882385423, + "grad_norm": 0.9002283811569214, + "learning_rate": 3.575500170098637e-06, + "loss": 0.6183, + "step": 2660 + }, + { + "epoch": 2.202926559911651, + "grad_norm": 0.9354086518287659, + "learning_rate": 3.574507339548897e-06, + "loss": 0.5989, + "step": 2661 + }, + { + "epoch": 2.2037548315847597, + "grad_norm": 0.9035785794258118, + "learning_rate": 3.5735143010920033e-06, + "loss": 0.592, + "step": 2662 + }, + { + "epoch": 2.2045831032578684, + "grad_norm": 0.9337678551673889, + "learning_rate": 3.5725210549201e-06, + "loss": 0.5881, + "step": 2663 + }, + { + "epoch": 2.205411374930977, + "grad_norm": 0.9226170182228088, + "learning_rate": 3.57152760122537e-06, + "loss": 0.5745, + "step": 2664 + }, + { + "epoch": 2.2062396466040863, + "grad_norm": 0.9272558093070984, + "learning_rate": 3.5705339402000377e-06, + "loss": 0.607, + "step": 2665 + }, + { + "epoch": 2.207067918277195, + "grad_norm": 0.930325448513031, + "learning_rate": 3.569540072036368e-06, + "loss": 0.5947, + "step": 2666 + }, + { + "epoch": 2.207896189950304, + "grad_norm": 0.9052232503890991, + "learning_rate": 3.568545996926664e-06, + "loss": 0.5883, + "step": 2667 + }, + { + "epoch": 2.2087244616234125, + "grad_norm": 0.8831207752227783, + "learning_rate": 3.5675517150632706e-06, + "loss": 0.5962, + "step": 2668 + }, + { + "epoch": 2.2095527332965212, + "grad_norm": 0.8999590873718262, + "learning_rate": 3.566557226638572e-06, + "loss": 0.6257, + "step": 2669 + }, + { + "epoch": 2.21038100496963, + "grad_norm": 0.9310323596000671, + "learning_rate": 3.565562531844992e-06, + "loss": 0.585, + "step": 2670 + }, + { + "epoch": 2.2112092766427387, + "grad_norm": 0.9469419121742249, + "learning_rate": 3.5645676308749954e-06, + "loss": 0.6163, + "step": 2671 + }, + { + "epoch": 2.2120375483158474, + "grad_norm": 0.9391891956329346, + "learning_rate": 3.563572523921086e-06, + "loss": 0.5863, + "step": 2672 + }, + { + "epoch": 2.2128658199889566, + "grad_norm": 0.9035549163818359, + "learning_rate": 3.5625772111758085e-06, + "loss": 0.6074, + "step": 2673 + }, + { + "epoch": 2.2136940916620653, + "grad_norm": 0.9008681178092957, + "learning_rate": 3.561581692831745e-06, + "loss": 0.5965, + "step": 2674 + }, + { + "epoch": 2.214522363335174, + "grad_norm": 0.9130175113677979, + "learning_rate": 3.56058596908152e-06, + "loss": 0.5913, + "step": 2675 + }, + { + "epoch": 2.2153506350082828, + "grad_norm": 0.9367324113845825, + "learning_rate": 3.5595900401177957e-06, + "loss": 0.617, + "step": 2676 + }, + { + "epoch": 2.2161789066813915, + "grad_norm": 0.9082489609718323, + "learning_rate": 3.5585939061332775e-06, + "loss": 0.6052, + "step": 2677 + }, + { + "epoch": 2.2170071783545002, + "grad_norm": 0.921264111995697, + "learning_rate": 3.557597567320706e-06, + "loss": 0.5832, + "step": 2678 + }, + { + "epoch": 2.217835450027609, + "grad_norm": 0.9040131568908691, + "learning_rate": 3.556601023872863e-06, + "loss": 0.5994, + "step": 2679 + }, + { + "epoch": 2.2186637217007177, + "grad_norm": 0.9144189357757568, + "learning_rate": 3.5556042759825733e-06, + "loss": 0.6068, + "step": 2680 + }, + { + "epoch": 2.2194919933738264, + "grad_norm": 0.91495680809021, + "learning_rate": 3.5546073238426955e-06, + "loss": 0.6154, + "step": 2681 + }, + { + "epoch": 2.2203202650469356, + "grad_norm": 0.925397515296936, + "learning_rate": 3.553610167646131e-06, + "loss": 0.6013, + "step": 2682 + }, + { + "epoch": 2.2211485367200443, + "grad_norm": 0.9633582234382629, + "learning_rate": 3.552612807585821e-06, + "loss": 0.6063, + "step": 2683 + }, + { + "epoch": 2.221976808393153, + "grad_norm": 0.9403156042098999, + "learning_rate": 3.551615243854746e-06, + "loss": 0.6141, + "step": 2684 + }, + { + "epoch": 2.2228050800662618, + "grad_norm": 0.9272189140319824, + "learning_rate": 3.550617476645923e-06, + "loss": 0.6013, + "step": 2685 + }, + { + "epoch": 2.2236333517393705, + "grad_norm": 0.92893385887146, + "learning_rate": 3.549619506152413e-06, + "loss": 0.5909, + "step": 2686 + }, + { + "epoch": 2.224461623412479, + "grad_norm": 0.9130013585090637, + "learning_rate": 3.548621332567312e-06, + "loss": 0.6005, + "step": 2687 + }, + { + "epoch": 2.225289895085588, + "grad_norm": 0.9111050367355347, + "learning_rate": 3.547622956083759e-06, + "loss": 0.5904, + "step": 2688 + }, + { + "epoch": 2.2261181667586967, + "grad_norm": 0.9379095435142517, + "learning_rate": 3.5466243768949292e-06, + "loss": 0.605, + "step": 2689 + }, + { + "epoch": 2.2269464384318054, + "grad_norm": 0.898931622505188, + "learning_rate": 3.5456255951940393e-06, + "loss": 0.5978, + "step": 2690 + }, + { + "epoch": 2.2277747101049146, + "grad_norm": 0.9569118022918701, + "learning_rate": 3.5446266111743444e-06, + "loss": 0.6162, + "step": 2691 + }, + { + "epoch": 2.2286029817780233, + "grad_norm": 0.9167311191558838, + "learning_rate": 3.543627425029138e-06, + "loss": 0.5876, + "step": 2692 + }, + { + "epoch": 2.229431253451132, + "grad_norm": 0.9584367275238037, + "learning_rate": 3.542628036951753e-06, + "loss": 0.5704, + "step": 2693 + }, + { + "epoch": 2.2302595251242407, + "grad_norm": 0.8896269202232361, + "learning_rate": 3.541628447135562e-06, + "loss": 0.5806, + "step": 2694 + }, + { + "epoch": 2.2310877967973495, + "grad_norm": 0.9340798854827881, + "learning_rate": 3.5406286557739767e-06, + "loss": 0.6218, + "step": 2695 + }, + { + "epoch": 2.231916068470458, + "grad_norm": 0.9442514777183533, + "learning_rate": 3.539628663060447e-06, + "loss": 0.5899, + "step": 2696 + }, + { + "epoch": 2.232744340143567, + "grad_norm": 0.9123642444610596, + "learning_rate": 3.5386284691884633e-06, + "loss": 0.5825, + "step": 2697 + }, + { + "epoch": 2.233572611816676, + "grad_norm": 0.9076257348060608, + "learning_rate": 3.537628074351551e-06, + "loss": 0.6001, + "step": 2698 + }, + { + "epoch": 2.234400883489785, + "grad_norm": 0.9335387945175171, + "learning_rate": 3.5366274787432804e-06, + "loss": 0.6048, + "step": 2699 + }, + { + "epoch": 2.2352291551628936, + "grad_norm": 0.9016919732093811, + "learning_rate": 3.535626682557255e-06, + "loss": 0.6175, + "step": 2700 + }, + { + "epoch": 2.2360574268360023, + "grad_norm": 0.9110099673271179, + "learning_rate": 3.5346256859871207e-06, + "loss": 0.6027, + "step": 2701 + }, + { + "epoch": 2.236885698509111, + "grad_norm": 0.9513245224952698, + "learning_rate": 3.5336244892265616e-06, + "loss": 0.6015, + "step": 2702 + }, + { + "epoch": 2.2377139701822197, + "grad_norm": 0.8986294865608215, + "learning_rate": 3.532623092469299e-06, + "loss": 0.61, + "step": 2703 + }, + { + "epoch": 2.2385422418553285, + "grad_norm": 0.8991696834564209, + "learning_rate": 3.5316214959090933e-06, + "loss": 0.6013, + "step": 2704 + }, + { + "epoch": 2.239370513528437, + "grad_norm": 0.9667947292327881, + "learning_rate": 3.5306196997397453e-06, + "loss": 0.6125, + "step": 2705 + }, + { + "epoch": 2.240198785201546, + "grad_norm": 0.9183294177055359, + "learning_rate": 3.529617704155092e-06, + "loss": 0.5901, + "step": 2706 + }, + { + "epoch": 2.241027056874655, + "grad_norm": 0.9139719605445862, + "learning_rate": 3.5286155093490117e-06, + "loss": 0.5798, + "step": 2707 + }, + { + "epoch": 2.241855328547764, + "grad_norm": 0.9529849290847778, + "learning_rate": 3.5276131155154193e-06, + "loss": 0.5763, + "step": 2708 + }, + { + "epoch": 2.2426836002208725, + "grad_norm": 0.9126972556114197, + "learning_rate": 3.526610522848267e-06, + "loss": 0.5804, + "step": 2709 + }, + { + "epoch": 2.2435118718939813, + "grad_norm": 0.9101532101631165, + "learning_rate": 3.525607731541549e-06, + "loss": 0.5815, + "step": 2710 + }, + { + "epoch": 2.24434014356709, + "grad_norm": 0.9072697162628174, + "learning_rate": 3.524604741789296e-06, + "loss": 0.6238, + "step": 2711 + }, + { + "epoch": 2.2451684152401987, + "grad_norm": 0.9228153228759766, + "learning_rate": 3.5236015537855765e-06, + "loss": 0.6011, + "step": 2712 + }, + { + "epoch": 2.2459966869133074, + "grad_norm": 0.9280546307563782, + "learning_rate": 3.522598167724498e-06, + "loss": 0.6018, + "step": 2713 + }, + { + "epoch": 2.246824958586416, + "grad_norm": 0.9077439904212952, + "learning_rate": 3.5215945838002075e-06, + "loss": 0.6231, + "step": 2714 + }, + { + "epoch": 2.247653230259525, + "grad_norm": 0.9201393127441406, + "learning_rate": 3.520590802206887e-06, + "loss": 0.591, + "step": 2715 + }, + { + "epoch": 2.248481501932634, + "grad_norm": 0.8888259530067444, + "learning_rate": 3.519586823138761e-06, + "loss": 0.585, + "step": 2716 + }, + { + "epoch": 2.249309773605743, + "grad_norm": 0.918370246887207, + "learning_rate": 3.5185826467900886e-06, + "loss": 0.6151, + "step": 2717 + }, + { + "epoch": 2.2501380452788515, + "grad_norm": 0.9071224331855774, + "learning_rate": 3.517578273355169e-06, + "loss": 0.6053, + "step": 2718 + }, + { + "epoch": 2.2509663169519603, + "grad_norm": 0.9325213432312012, + "learning_rate": 3.5165737030283404e-06, + "loss": 0.5959, + "step": 2719 + }, + { + "epoch": 2.251794588625069, + "grad_norm": 0.9014924168586731, + "learning_rate": 3.515568936003976e-06, + "loss": 0.6092, + "step": 2720 + }, + { + "epoch": 2.2526228602981777, + "grad_norm": 0.923115611076355, + "learning_rate": 3.5145639724764886e-06, + "loss": 0.6031, + "step": 2721 + }, + { + "epoch": 2.2534511319712864, + "grad_norm": 0.9452555179595947, + "learning_rate": 3.513558812640331e-06, + "loss": 0.5957, + "step": 2722 + }, + { + "epoch": 2.2542794036443956, + "grad_norm": 0.960884153842926, + "learning_rate": 3.5125534566899915e-06, + "loss": 0.5862, + "step": 2723 + }, + { + "epoch": 2.2551076753175043, + "grad_norm": 0.9164205193519592, + "learning_rate": 3.5115479048199964e-06, + "loss": 0.5992, + "step": 2724 + }, + { + "epoch": 2.255935946990613, + "grad_norm": 0.943760097026825, + "learning_rate": 3.5105421572249123e-06, + "loss": 0.5842, + "step": 2725 + }, + { + "epoch": 2.256764218663722, + "grad_norm": 0.9578623175621033, + "learning_rate": 3.50953621409934e-06, + "loss": 0.6094, + "step": 2726 + }, + { + "epoch": 2.2575924903368305, + "grad_norm": 0.9307887554168701, + "learning_rate": 3.5085300756379216e-06, + "loss": 0.5905, + "step": 2727 + }, + { + "epoch": 2.2584207620099392, + "grad_norm": 0.9150220155715942, + "learning_rate": 3.5075237420353346e-06, + "loss": 0.6035, + "step": 2728 + }, + { + "epoch": 2.259249033683048, + "grad_norm": 0.8887060880661011, + "learning_rate": 3.506517213486296e-06, + "loss": 0.6032, + "step": 2729 + }, + { + "epoch": 2.2600773053561567, + "grad_norm": 0.9477519989013672, + "learning_rate": 3.5055104901855585e-06, + "loss": 0.576, + "step": 2730 + }, + { + "epoch": 2.2609055770292654, + "grad_norm": 0.9668344855308533, + "learning_rate": 3.504503572327915e-06, + "loss": 0.5973, + "step": 2731 + }, + { + "epoch": 2.2617338487023746, + "grad_norm": 0.905135989189148, + "learning_rate": 3.503496460108194e-06, + "loss": 0.5999, + "step": 2732 + }, + { + "epoch": 2.2625621203754833, + "grad_norm": 0.933996856212616, + "learning_rate": 3.5024891537212624e-06, + "loss": 0.5803, + "step": 2733 + }, + { + "epoch": 2.263390392048592, + "grad_norm": 0.9417551159858704, + "learning_rate": 3.501481653362025e-06, + "loss": 0.5918, + "step": 2734 + }, + { + "epoch": 2.2642186637217008, + "grad_norm": 0.9239657521247864, + "learning_rate": 3.500473959225422e-06, + "loss": 0.5828, + "step": 2735 + }, + { + "epoch": 2.2650469353948095, + "grad_norm": 0.9021532535552979, + "learning_rate": 3.499466071506436e-06, + "loss": 0.5958, + "step": 2736 + }, + { + "epoch": 2.2658752070679182, + "grad_norm": 0.9159536957740784, + "learning_rate": 3.498457990400081e-06, + "loss": 0.5867, + "step": 2737 + }, + { + "epoch": 2.266703478741027, + "grad_norm": 0.9347800612449646, + "learning_rate": 3.4974497161014124e-06, + "loss": 0.6058, + "step": 2738 + }, + { + "epoch": 2.2675317504141357, + "grad_norm": 0.9229647517204285, + "learning_rate": 3.496441248805522e-06, + "loss": 0.5714, + "step": 2739 + }, + { + "epoch": 2.2683600220872444, + "grad_norm": 0.9330629706382751, + "learning_rate": 3.4954325887075386e-06, + "loss": 0.5808, + "step": 2740 + }, + { + "epoch": 2.2691882937603536, + "grad_norm": 0.8958399891853333, + "learning_rate": 3.494423736002629e-06, + "loss": 0.5793, + "step": 2741 + }, + { + "epoch": 2.2700165654334623, + "grad_norm": 0.8952097296714783, + "learning_rate": 3.493414690885997e-06, + "loss": 0.6007, + "step": 2742 + }, + { + "epoch": 2.270844837106571, + "grad_norm": 0.9001036882400513, + "learning_rate": 3.4924054535528813e-06, + "loss": 0.5821, + "step": 2743 + }, + { + "epoch": 2.2716731087796798, + "grad_norm": 0.9016004204750061, + "learning_rate": 3.4913960241985616e-06, + "loss": 0.5819, + "step": 2744 + }, + { + "epoch": 2.2725013804527885, + "grad_norm": 0.9043833017349243, + "learning_rate": 3.490386403018354e-06, + "loss": 0.6144, + "step": 2745 + }, + { + "epoch": 2.273329652125897, + "grad_norm": 0.9396303296089172, + "learning_rate": 3.489376590207608e-06, + "loss": 0.602, + "step": 2746 + }, + { + "epoch": 2.274157923799006, + "grad_norm": 0.9685373306274414, + "learning_rate": 3.488366585961716e-06, + "loss": 0.6247, + "step": 2747 + }, + { + "epoch": 2.2749861954721147, + "grad_norm": 0.9028962254524231, + "learning_rate": 3.487356390476103e-06, + "loss": 0.5959, + "step": 2748 + }, + { + "epoch": 2.2758144671452234, + "grad_norm": 0.9150850772857666, + "learning_rate": 3.486346003946232e-06, + "loss": 0.5632, + "step": 2749 + }, + { + "epoch": 2.2766427388183326, + "grad_norm": 0.902158796787262, + "learning_rate": 3.4853354265676043e-06, + "loss": 0.5901, + "step": 2750 + }, + { + "epoch": 2.2774710104914413, + "grad_norm": 0.8942626118659973, + "learning_rate": 3.4843246585357565e-06, + "loss": 0.6082, + "step": 2751 + }, + { + "epoch": 2.27829928216455, + "grad_norm": 0.912050187587738, + "learning_rate": 3.4833137000462623e-06, + "loss": 0.5812, + "step": 2752 + }, + { + "epoch": 2.2791275538376587, + "grad_norm": 0.9262410998344421, + "learning_rate": 3.482302551294735e-06, + "loss": 0.6199, + "step": 2753 + }, + { + "epoch": 2.2799558255107675, + "grad_norm": 0.9281290173530579, + "learning_rate": 3.4812912124768193e-06, + "loss": 0.591, + "step": 2754 + }, + { + "epoch": 2.280784097183876, + "grad_norm": 0.8852386474609375, + "learning_rate": 3.4802796837882023e-06, + "loss": 0.6035, + "step": 2755 + }, + { + "epoch": 2.281612368856985, + "grad_norm": 0.8982499241828918, + "learning_rate": 3.479267965424605e-06, + "loss": 0.6023, + "step": 2756 + }, + { + "epoch": 2.282440640530094, + "grad_norm": 0.9022878408432007, + "learning_rate": 3.4782560575817837e-06, + "loss": 0.6004, + "step": 2757 + }, + { + "epoch": 2.283268912203203, + "grad_norm": 0.9256078600883484, + "learning_rate": 3.4772439604555353e-06, + "loss": 0.5811, + "step": 2758 + }, + { + "epoch": 2.2840971838763116, + "grad_norm": 0.9009407758712769, + "learning_rate": 3.4762316742416913e-06, + "loss": 0.6053, + "step": 2759 + }, + { + "epoch": 2.2849254555494203, + "grad_norm": 0.9257068037986755, + "learning_rate": 3.4752191991361174e-06, + "loss": 0.5664, + "step": 2760 + }, + { + "epoch": 2.285753727222529, + "grad_norm": 0.882075309753418, + "learning_rate": 3.4742065353347193e-06, + "loss": 0.6099, + "step": 2761 + }, + { + "epoch": 2.2865819988956377, + "grad_norm": 0.9553154706954956, + "learning_rate": 3.4731936830334384e-06, + "loss": 0.6072, + "step": 2762 + }, + { + "epoch": 2.2874102705687465, + "grad_norm": 0.9329802989959717, + "learning_rate": 3.4721806424282512e-06, + "loss": 0.601, + "step": 2763 + }, + { + "epoch": 2.288238542241855, + "grad_norm": 0.9146580696105957, + "learning_rate": 3.471167413715173e-06, + "loss": 0.599, + "step": 2764 + }, + { + "epoch": 2.289066813914964, + "grad_norm": 0.9170219302177429, + "learning_rate": 3.470153997090253e-06, + "loss": 0.5969, + "step": 2765 + }, + { + "epoch": 2.289895085588073, + "grad_norm": 0.8903328776359558, + "learning_rate": 3.469140392749578e-06, + "loss": 0.6063, + "step": 2766 + }, + { + "epoch": 2.290723357261182, + "grad_norm": 0.9068611264228821, + "learning_rate": 3.468126600889272e-06, + "loss": 0.5807, + "step": 2767 + }, + { + "epoch": 2.2915516289342905, + "grad_norm": 0.9464553594589233, + "learning_rate": 3.467112621705493e-06, + "loss": 0.5955, + "step": 2768 + }, + { + "epoch": 2.2923799006073993, + "grad_norm": 0.9302126169204712, + "learning_rate": 3.4660984553944367e-06, + "loss": 0.5956, + "step": 2769 + }, + { + "epoch": 2.293208172280508, + "grad_norm": 0.8797576427459717, + "learning_rate": 3.465084102152336e-06, + "loss": 0.5964, + "step": 2770 + }, + { + "epoch": 2.2940364439536167, + "grad_norm": 0.9544422030448914, + "learning_rate": 3.4640695621754573e-06, + "loss": 0.6225, + "step": 2771 + }, + { + "epoch": 2.2948647156267254, + "grad_norm": 0.919297993183136, + "learning_rate": 3.463054835660105e-06, + "loss": 0.5901, + "step": 2772 + }, + { + "epoch": 2.295692987299834, + "grad_norm": 0.9211829900741577, + "learning_rate": 3.4620399228026203e-06, + "loss": 0.5949, + "step": 2773 + }, + { + "epoch": 2.296521258972943, + "grad_norm": 0.929501473903656, + "learning_rate": 3.4610248237993775e-06, + "loss": 0.5654, + "step": 2774 + }, + { + "epoch": 2.297349530646052, + "grad_norm": 0.9604958891868591, + "learning_rate": 3.4600095388467907e-06, + "loss": 0.6059, + "step": 2775 + }, + { + "epoch": 2.298177802319161, + "grad_norm": 0.9042990207672119, + "learning_rate": 3.458994068141307e-06, + "loss": 0.5901, + "step": 2776 + }, + { + "epoch": 2.2990060739922695, + "grad_norm": 0.8958423733711243, + "learning_rate": 3.4579784118794107e-06, + "loss": 0.5938, + "step": 2777 + }, + { + "epoch": 2.2998343456653783, + "grad_norm": 0.9388688206672668, + "learning_rate": 3.456962570257622e-06, + "loss": 0.6024, + "step": 2778 + }, + { + "epoch": 2.300662617338487, + "grad_norm": 0.9006328582763672, + "learning_rate": 3.455946543472497e-06, + "loss": 0.5929, + "step": 2779 + }, + { + "epoch": 2.3014908890115957, + "grad_norm": 0.887873113155365, + "learning_rate": 3.454930331720626e-06, + "loss": 0.599, + "step": 2780 + }, + { + "epoch": 2.3023191606847044, + "grad_norm": 0.8919411301612854, + "learning_rate": 3.4539139351986385e-06, + "loss": 0.6173, + "step": 2781 + }, + { + "epoch": 2.3031474323578136, + "grad_norm": 0.9478135704994202, + "learning_rate": 3.452897354103196e-06, + "loss": 0.6025, + "step": 2782 + }, + { + "epoch": 2.3039757040309223, + "grad_norm": 0.9009284377098083, + "learning_rate": 3.451880588630999e-06, + "loss": 0.602, + "step": 2783 + }, + { + "epoch": 2.304803975704031, + "grad_norm": 0.9043265581130981, + "learning_rate": 3.4508636389787816e-06, + "loss": 0.6002, + "step": 2784 + }, + { + "epoch": 2.30563224737714, + "grad_norm": 0.9005764126777649, + "learning_rate": 3.4498465053433126e-06, + "loss": 0.618, + "step": 2785 + }, + { + "epoch": 2.3064605190502485, + "grad_norm": 0.9352818727493286, + "learning_rate": 3.4488291879214003e-06, + "loss": 0.6069, + "step": 2786 + }, + { + "epoch": 2.3072887907233572, + "grad_norm": 0.9092200994491577, + "learning_rate": 3.4478116869098854e-06, + "loss": 0.581, + "step": 2787 + }, + { + "epoch": 2.308117062396466, + "grad_norm": 0.9055072665214539, + "learning_rate": 3.4467940025056433e-06, + "loss": 0.598, + "step": 2788 + }, + { + "epoch": 2.3089453340695747, + "grad_norm": 1.0053009986877441, + "learning_rate": 3.4457761349055885e-06, + "loss": 0.5929, + "step": 2789 + }, + { + "epoch": 2.3097736057426834, + "grad_norm": 0.9241059422492981, + "learning_rate": 3.4447580843066675e-06, + "loss": 0.5962, + "step": 2790 + }, + { + "epoch": 2.3106018774157926, + "grad_norm": 0.927590548992157, + "learning_rate": 3.443739850905864e-06, + "loss": 0.5863, + "step": 2791 + }, + { + "epoch": 2.3114301490889013, + "grad_norm": 0.9227457046508789, + "learning_rate": 3.442721434900197e-06, + "loss": 0.5928, + "step": 2792 + }, + { + "epoch": 2.31225842076201, + "grad_norm": 0.9112377762794495, + "learning_rate": 3.4417028364867205e-06, + "loss": 0.6113, + "step": 2793 + }, + { + "epoch": 2.3130866924351188, + "grad_norm": 0.9256125092506409, + "learning_rate": 3.4406840558625237e-06, + "loss": 0.6163, + "step": 2794 + }, + { + "epoch": 2.3139149641082275, + "grad_norm": 0.933461606502533, + "learning_rate": 3.4396650932247303e-06, + "loss": 0.6159, + "step": 2795 + }, + { + "epoch": 2.3147432357813362, + "grad_norm": 0.9031276702880859, + "learning_rate": 3.4386459487705016e-06, + "loss": 0.5981, + "step": 2796 + }, + { + "epoch": 2.315571507454445, + "grad_norm": 0.9465809464454651, + "learning_rate": 3.437626622697031e-06, + "loss": 0.6029, + "step": 2797 + }, + { + "epoch": 2.3163997791275537, + "grad_norm": 0.9535313248634338, + "learning_rate": 3.43660711520155e-06, + "loss": 0.5998, + "step": 2798 + }, + { + "epoch": 2.3172280508006624, + "grad_norm": 0.9048331379890442, + "learning_rate": 3.4355874264813228e-06, + "loss": 0.5991, + "step": 2799 + }, + { + "epoch": 2.3180563224737716, + "grad_norm": 0.9119572639465332, + "learning_rate": 3.4345675567336506e-06, + "loss": 0.6133, + "step": 2800 + }, + { + "epoch": 2.3188845941468803, + "grad_norm": 0.9217002391815186, + "learning_rate": 3.4335475061558676e-06, + "loss": 0.6012, + "step": 2801 + }, + { + "epoch": 2.319712865819989, + "grad_norm": 0.9501274824142456, + "learning_rate": 3.432527274945344e-06, + "loss": 0.5874, + "step": 2802 + }, + { + "epoch": 2.3205411374930978, + "grad_norm": 0.9126407504081726, + "learning_rate": 3.4315068632994864e-06, + "loss": 0.5985, + "step": 2803 + }, + { + "epoch": 2.3213694091662065, + "grad_norm": 0.9414592981338501, + "learning_rate": 3.430486271415734e-06, + "loss": 0.5989, + "step": 2804 + }, + { + "epoch": 2.322197680839315, + "grad_norm": 0.9485170841217041, + "learning_rate": 3.4294654994915623e-06, + "loss": 0.5795, + "step": 2805 + }, + { + "epoch": 2.323025952512424, + "grad_norm": 0.9311578273773193, + "learning_rate": 3.4284445477244804e-06, + "loss": 0.5606, + "step": 2806 + }, + { + "epoch": 2.3238542241855327, + "grad_norm": 0.9476608037948608, + "learning_rate": 3.4274234163120328e-06, + "loss": 0.593, + "step": 2807 + }, + { + "epoch": 2.3246824958586414, + "grad_norm": 0.9367009401321411, + "learning_rate": 3.4264021054518004e-06, + "loss": 0.6104, + "step": 2808 + }, + { + "epoch": 2.3255107675317506, + "grad_norm": 0.9663116335868835, + "learning_rate": 3.425380615341396e-06, + "loss": 0.6091, + "step": 2809 + }, + { + "epoch": 2.3263390392048593, + "grad_norm": 0.9176538586616516, + "learning_rate": 3.4243589461784687e-06, + "loss": 0.585, + "step": 2810 + }, + { + "epoch": 2.327167310877968, + "grad_norm": 0.9261791110038757, + "learning_rate": 3.423337098160703e-06, + "loss": 0.6189, + "step": 2811 + }, + { + "epoch": 2.3279955825510767, + "grad_norm": 0.9573442339897156, + "learning_rate": 3.422315071485815e-06, + "loss": 0.5925, + "step": 2812 + }, + { + "epoch": 2.3288238542241855, + "grad_norm": 0.9381136894226074, + "learning_rate": 3.4212928663515584e-06, + "loss": 0.588, + "step": 2813 + }, + { + "epoch": 2.329652125897294, + "grad_norm": 0.9276189804077148, + "learning_rate": 3.420270482955721e-06, + "loss": 0.5885, + "step": 2814 + }, + { + "epoch": 2.330480397570403, + "grad_norm": 0.9124462604522705, + "learning_rate": 3.419247921496123e-06, + "loss": 0.6004, + "step": 2815 + }, + { + "epoch": 2.331308669243512, + "grad_norm": 0.9146862626075745, + "learning_rate": 3.418225182170622e-06, + "loss": 0.6036, + "step": 2816 + }, + { + "epoch": 2.332136940916621, + "grad_norm": 0.9196732044219971, + "learning_rate": 3.4172022651771075e-06, + "loss": 0.6054, + "step": 2817 + }, + { + "epoch": 2.3329652125897296, + "grad_norm": 0.9340062737464905, + "learning_rate": 3.416179170713505e-06, + "loss": 0.6037, + "step": 2818 + }, + { + "epoch": 2.3337934842628383, + "grad_norm": 0.9004904627799988, + "learning_rate": 3.415155898977773e-06, + "loss": 0.6087, + "step": 2819 + }, + { + "epoch": 2.334621755935947, + "grad_norm": 0.8993189334869385, + "learning_rate": 3.414132450167906e-06, + "loss": 0.5972, + "step": 2820 + }, + { + "epoch": 2.3354500276090557, + "grad_norm": 0.9195947051048279, + "learning_rate": 3.413108824481931e-06, + "loss": 0.5784, + "step": 2821 + }, + { + "epoch": 2.3362782992821645, + "grad_norm": 0.9149767756462097, + "learning_rate": 3.412085022117911e-06, + "loss": 0.5838, + "step": 2822 + }, + { + "epoch": 2.337106570955273, + "grad_norm": 0.8880875110626221, + "learning_rate": 3.411061043273941e-06, + "loss": 0.6008, + "step": 2823 + }, + { + "epoch": 2.337934842628382, + "grad_norm": 0.9110493659973145, + "learning_rate": 3.4100368881481515e-06, + "loss": 0.6333, + "step": 2824 + }, + { + "epoch": 2.338763114301491, + "grad_norm": 0.9558914303779602, + "learning_rate": 3.409012556938708e-06, + "loss": 0.602, + "step": 2825 + }, + { + "epoch": 2.3395913859746, + "grad_norm": 0.9668992161750793, + "learning_rate": 3.4079880498438083e-06, + "loss": 0.5952, + "step": 2826 + }, + { + "epoch": 2.3404196576477085, + "grad_norm": 0.9352533221244812, + "learning_rate": 3.406963367061685e-06, + "loss": 0.6093, + "step": 2827 + }, + { + "epoch": 2.3412479293208173, + "grad_norm": 0.9072561264038086, + "learning_rate": 3.4059385087906055e-06, + "loss": 0.603, + "step": 2828 + }, + { + "epoch": 2.342076200993926, + "grad_norm": 0.9563443064689636, + "learning_rate": 3.4049134752288692e-06, + "loss": 0.585, + "step": 2829 + }, + { + "epoch": 2.3429044726670347, + "grad_norm": 0.9061548709869385, + "learning_rate": 3.4038882665748102e-06, + "loss": 0.6111, + "step": 2830 + }, + { + "epoch": 2.3437327443401434, + "grad_norm": 0.9177330136299133, + "learning_rate": 3.4028628830267986e-06, + "loss": 0.5872, + "step": 2831 + }, + { + "epoch": 2.344561016013252, + "grad_norm": 0.9479212164878845, + "learning_rate": 3.401837324783235e-06, + "loss": 0.5911, + "step": 2832 + }, + { + "epoch": 2.345389287686361, + "grad_norm": 0.9118002653121948, + "learning_rate": 3.4008115920425556e-06, + "loss": 0.6153, + "step": 2833 + }, + { + "epoch": 2.34621755935947, + "grad_norm": 0.9086247682571411, + "learning_rate": 3.399785685003231e-06, + "loss": 0.5913, + "step": 2834 + }, + { + "epoch": 2.347045831032579, + "grad_norm": 0.937494158744812, + "learning_rate": 3.398759603863763e-06, + "loss": 0.601, + "step": 2835 + }, + { + "epoch": 2.3478741027056875, + "grad_norm": 0.9359503388404846, + "learning_rate": 3.397733348822691e-06, + "loss": 0.5894, + "step": 2836 + }, + { + "epoch": 2.3487023743787963, + "grad_norm": 0.9384423494338989, + "learning_rate": 3.396706920078584e-06, + "loss": 0.5974, + "step": 2837 + }, + { + "epoch": 2.349530646051905, + "grad_norm": 0.9438448548316956, + "learning_rate": 3.3956803178300474e-06, + "loss": 0.6079, + "step": 2838 + }, + { + "epoch": 2.3503589177250137, + "grad_norm": 0.9250973463058472, + "learning_rate": 3.3946535422757187e-06, + "loss": 0.6011, + "step": 2839 + }, + { + "epoch": 2.3511871893981224, + "grad_norm": 0.9577994346618652, + "learning_rate": 3.3936265936142685e-06, + "loss": 0.6052, + "step": 2840 + }, + { + "epoch": 2.3520154610712316, + "grad_norm": 0.9307487607002258, + "learning_rate": 3.3925994720444033e-06, + "loss": 0.5925, + "step": 2841 + }, + { + "epoch": 2.3528437327443403, + "grad_norm": 0.9049618244171143, + "learning_rate": 3.3915721777648615e-06, + "loss": 0.6096, + "step": 2842 + }, + { + "epoch": 2.353672004417449, + "grad_norm": 0.928248405456543, + "learning_rate": 3.3905447109744137e-06, + "loss": 0.5938, + "step": 2843 + }, + { + "epoch": 2.354500276090558, + "grad_norm": 0.9107298254966736, + "learning_rate": 3.3895170718718664e-06, + "loss": 0.5947, + "step": 2844 + }, + { + "epoch": 2.3553285477636665, + "grad_norm": 0.8896966576576233, + "learning_rate": 3.388489260656057e-06, + "loss": 0.5827, + "step": 2845 + }, + { + "epoch": 2.3561568194367752, + "grad_norm": 0.9374940991401672, + "learning_rate": 3.3874612775258586e-06, + "loss": 0.6038, + "step": 2846 + }, + { + "epoch": 2.356985091109884, + "grad_norm": 0.9086610078811646, + "learning_rate": 3.3864331226801755e-06, + "loss": 0.5758, + "step": 2847 + }, + { + "epoch": 2.3578133627829927, + "grad_norm": 0.9486793875694275, + "learning_rate": 3.3854047963179464e-06, + "loss": 0.5656, + "step": 2848 + }, + { + "epoch": 2.3586416344561014, + "grad_norm": 0.9411890506744385, + "learning_rate": 3.3843762986381436e-06, + "loss": 0.605, + "step": 2849 + }, + { + "epoch": 2.3594699061292106, + "grad_norm": 0.9491676688194275, + "learning_rate": 3.3833476298397706e-06, + "loss": 0.5992, + "step": 2850 + }, + { + "epoch": 2.3602981778023193, + "grad_norm": 0.9281409382820129, + "learning_rate": 3.382318790121865e-06, + "loss": 0.5856, + "step": 2851 + }, + { + "epoch": 2.361126449475428, + "grad_norm": 0.9243955016136169, + "learning_rate": 3.381289779683499e-06, + "loss": 0.5966, + "step": 2852 + }, + { + "epoch": 2.3619547211485368, + "grad_norm": 0.8931578993797302, + "learning_rate": 3.3802605987237757e-06, + "loss": 0.6095, + "step": 2853 + }, + { + "epoch": 2.3627829928216455, + "grad_norm": 0.9200635552406311, + "learning_rate": 3.3792312474418324e-06, + "loss": 0.6014, + "step": 2854 + }, + { + "epoch": 2.3636112644947542, + "grad_norm": 0.9007267951965332, + "learning_rate": 3.378201726036839e-06, + "loss": 0.5964, + "step": 2855 + }, + { + "epoch": 2.364439536167863, + "grad_norm": 0.9420115947723389, + "learning_rate": 3.377172034707998e-06, + "loss": 0.6075, + "step": 2856 + }, + { + "epoch": 2.3652678078409717, + "grad_norm": 0.8935693502426147, + "learning_rate": 3.3761421736545453e-06, + "loss": 0.6184, + "step": 2857 + }, + { + "epoch": 2.3660960795140804, + "grad_norm": 0.9142755270004272, + "learning_rate": 3.3751121430757482e-06, + "loss": 0.6114, + "step": 2858 + }, + { + "epoch": 2.3669243511871896, + "grad_norm": 0.9280282855033875, + "learning_rate": 3.3740819431709103e-06, + "loss": 0.5922, + "step": 2859 + }, + { + "epoch": 2.3677526228602983, + "grad_norm": 0.9347119927406311, + "learning_rate": 3.3730515741393643e-06, + "loss": 0.5733, + "step": 2860 + }, + { + "epoch": 2.368580894533407, + "grad_norm": 0.9286894202232361, + "learning_rate": 3.3720210361804775e-06, + "loss": 0.597, + "step": 2861 + }, + { + "epoch": 2.3694091662065158, + "grad_norm": 0.9092636704444885, + "learning_rate": 3.3709903294936482e-06, + "loss": 0.5842, + "step": 2862 + }, + { + "epoch": 2.3702374378796245, + "grad_norm": 0.9205367565155029, + "learning_rate": 3.369959454278309e-06, + "loss": 0.6166, + "step": 2863 + }, + { + "epoch": 2.371065709552733, + "grad_norm": 0.9211122393608093, + "learning_rate": 3.3689284107339265e-06, + "loss": 0.5746, + "step": 2864 + }, + { + "epoch": 2.371893981225842, + "grad_norm": 0.908984899520874, + "learning_rate": 3.367897199059996e-06, + "loss": 0.5883, + "step": 2865 + }, + { + "epoch": 2.3727222528989507, + "grad_norm": 0.9270497560501099, + "learning_rate": 3.3668658194560477e-06, + "loss": 0.5985, + "step": 2866 + }, + { + "epoch": 2.3735505245720594, + "grad_norm": 0.9320128560066223, + "learning_rate": 3.3658342721216448e-06, + "loss": 0.5907, + "step": 2867 + }, + { + "epoch": 2.3743787962451686, + "grad_norm": 0.9428799152374268, + "learning_rate": 3.364802557256381e-06, + "loss": 0.6041, + "step": 2868 + }, + { + "epoch": 2.3752070679182773, + "grad_norm": 0.915873646736145, + "learning_rate": 3.363770675059883e-06, + "loss": 0.615, + "step": 2869 + }, + { + "epoch": 2.376035339591386, + "grad_norm": 0.9327977895736694, + "learning_rate": 3.362738625731813e-06, + "loss": 0.6147, + "step": 2870 + }, + { + "epoch": 2.3768636112644947, + "grad_norm": 0.9171217679977417, + "learning_rate": 3.3617064094718602e-06, + "loss": 0.6057, + "step": 2871 + }, + { + "epoch": 2.3776918829376035, + "grad_norm": 0.9192221760749817, + "learning_rate": 3.3606740264797504e-06, + "loss": 0.6092, + "step": 2872 + }, + { + "epoch": 2.378520154610712, + "grad_norm": 0.9024589657783508, + "learning_rate": 3.3596414769552394e-06, + "loss": 0.6047, + "step": 2873 + }, + { + "epoch": 2.379348426283821, + "grad_norm": 0.910923182964325, + "learning_rate": 3.3586087610981155e-06, + "loss": 0.5741, + "step": 2874 + }, + { + "epoch": 2.38017669795693, + "grad_norm": 0.929965078830719, + "learning_rate": 3.3575758791082014e-06, + "loss": 0.6114, + "step": 2875 + }, + { + "epoch": 2.381004969630039, + "grad_norm": 0.910308837890625, + "learning_rate": 3.356542831185348e-06, + "loss": 0.5938, + "step": 2876 + }, + { + "epoch": 2.3818332413031476, + "grad_norm": 0.9371251463890076, + "learning_rate": 3.355509617529441e-06, + "loss": 0.5883, + "step": 2877 + }, + { + "epoch": 2.3826615129762563, + "grad_norm": 0.9064118266105652, + "learning_rate": 3.3544762383403995e-06, + "loss": 0.5879, + "step": 2878 + }, + { + "epoch": 2.383489784649365, + "grad_norm": 0.954288899898529, + "learning_rate": 3.3534426938181704e-06, + "loss": 0.5981, + "step": 2879 + }, + { + "epoch": 2.3843180563224737, + "grad_norm": 0.9465821981430054, + "learning_rate": 3.3524089841627348e-06, + "loss": 0.6129, + "step": 2880 + }, + { + "epoch": 2.3851463279955825, + "grad_norm": 0.9406200051307678, + "learning_rate": 3.3513751095741084e-06, + "loss": 0.6035, + "step": 2881 + }, + { + "epoch": 2.385974599668691, + "grad_norm": 0.9627492427825928, + "learning_rate": 3.3503410702523337e-06, + "loss": 0.5701, + "step": 2882 + }, + { + "epoch": 2.3868028713418, + "grad_norm": 0.8899156451225281, + "learning_rate": 3.3493068663974892e-06, + "loss": 0.6055, + "step": 2883 + }, + { + "epoch": 2.387631143014909, + "grad_norm": 0.9304690957069397, + "learning_rate": 3.348272498209683e-06, + "loss": 0.6078, + "step": 2884 + }, + { + "epoch": 2.388459414688018, + "grad_norm": 0.9416152834892273, + "learning_rate": 3.3472379658890554e-06, + "loss": 0.609, + "step": 2885 + }, + { + "epoch": 2.3892876863611265, + "grad_norm": 0.9296873211860657, + "learning_rate": 3.34620326963578e-06, + "loss": 0.608, + "step": 2886 + }, + { + "epoch": 2.3901159580342353, + "grad_norm": 0.9163168668746948, + "learning_rate": 3.34516840965006e-06, + "loss": 0.5725, + "step": 2887 + }, + { + "epoch": 2.390944229707344, + "grad_norm": 0.9371008276939392, + "learning_rate": 3.344133386132131e-06, + "loss": 0.6223, + "step": 2888 + }, + { + "epoch": 2.3917725013804527, + "grad_norm": 0.9486922025680542, + "learning_rate": 3.3430981992822608e-06, + "loss": 0.5878, + "step": 2889 + }, + { + "epoch": 2.3926007730535614, + "grad_norm": 0.9124024510383606, + "learning_rate": 3.3420628493007483e-06, + "loss": 0.5816, + "step": 2890 + }, + { + "epoch": 2.39342904472667, + "grad_norm": 0.9193150401115417, + "learning_rate": 3.3410273363879236e-06, + "loss": 0.5909, + "step": 2891 + }, + { + "epoch": 2.394257316399779, + "grad_norm": 0.9287682175636292, + "learning_rate": 3.3399916607441502e-06, + "loss": 0.6095, + "step": 2892 + }, + { + "epoch": 2.395085588072888, + "grad_norm": 0.9130192995071411, + "learning_rate": 3.3389558225698204e-06, + "loss": 0.5782, + "step": 2893 + }, + { + "epoch": 2.395913859745997, + "grad_norm": 0.9145562052726746, + "learning_rate": 3.3379198220653587e-06, + "loss": 0.5919, + "step": 2894 + }, + { + "epoch": 2.3967421314191055, + "grad_norm": 0.9094864726066589, + "learning_rate": 3.3368836594312235e-06, + "loss": 0.6165, + "step": 2895 + }, + { + "epoch": 2.3975704030922143, + "grad_norm": 0.9446457028388977, + "learning_rate": 3.3358473348679016e-06, + "loss": 0.5967, + "step": 2896 + }, + { + "epoch": 2.398398674765323, + "grad_norm": 0.9345584511756897, + "learning_rate": 3.334810848575911e-06, + "loss": 0.6006, + "step": 2897 + }, + { + "epoch": 2.3992269464384317, + "grad_norm": 0.8865429759025574, + "learning_rate": 3.3337742007558043e-06, + "loss": 0.5766, + "step": 2898 + }, + { + "epoch": 2.4000552181115404, + "grad_norm": 0.9188973903656006, + "learning_rate": 3.3327373916081617e-06, + "loss": 0.6107, + "step": 2899 + }, + { + "epoch": 2.4008834897846496, + "grad_norm": 0.895247220993042, + "learning_rate": 3.3317004213335967e-06, + "loss": 0.6037, + "step": 2900 + }, + { + "epoch": 2.4017117614577583, + "grad_norm": 0.9131169319152832, + "learning_rate": 3.330663290132753e-06, + "loss": 0.5825, + "step": 2901 + }, + { + "epoch": 2.402540033130867, + "grad_norm": 0.9109821915626526, + "learning_rate": 3.329625998206306e-06, + "loss": 0.6194, + "step": 2902 + }, + { + "epoch": 2.403368304803976, + "grad_norm": 0.90902179479599, + "learning_rate": 3.3285885457549618e-06, + "loss": 0.5921, + "step": 2903 + }, + { + "epoch": 2.4041965764770845, + "grad_norm": 0.9419400095939636, + "learning_rate": 3.3275509329794585e-06, + "loss": 0.5973, + "step": 2904 + }, + { + "epoch": 2.4050248481501932, + "grad_norm": 0.905164897441864, + "learning_rate": 3.3265131600805636e-06, + "loss": 0.5828, + "step": 2905 + }, + { + "epoch": 2.405853119823302, + "grad_norm": 0.9232257008552551, + "learning_rate": 3.3254752272590775e-06, + "loss": 0.5972, + "step": 2906 + }, + { + "epoch": 2.4066813914964107, + "grad_norm": 0.9152960181236267, + "learning_rate": 3.3244371347158294e-06, + "loss": 0.5968, + "step": 2907 + }, + { + "epoch": 2.4075096631695194, + "grad_norm": 0.9337223172187805, + "learning_rate": 3.3233988826516807e-06, + "loss": 0.5907, + "step": 2908 + }, + { + "epoch": 2.4083379348426286, + "grad_norm": 0.9352282881736755, + "learning_rate": 3.322360471267524e-06, + "loss": 0.573, + "step": 2909 + }, + { + "epoch": 2.4091662065157373, + "grad_norm": 0.928322970867157, + "learning_rate": 3.321321900764282e-06, + "loss": 0.5836, + "step": 2910 + }, + { + "epoch": 2.409994478188846, + "grad_norm": 0.9335048794746399, + "learning_rate": 3.3202831713429084e-06, + "loss": 0.607, + "step": 2911 + }, + { + "epoch": 2.4108227498619548, + "grad_norm": 0.9128180742263794, + "learning_rate": 3.319244283204389e-06, + "loss": 0.5956, + "step": 2912 + }, + { + "epoch": 2.4116510215350635, + "grad_norm": 0.9457159042358398, + "learning_rate": 3.3182052365497364e-06, + "loss": 0.602, + "step": 2913 + }, + { + "epoch": 2.4124792932081722, + "grad_norm": 0.9508596658706665, + "learning_rate": 3.317166031579998e-06, + "loss": 0.5972, + "step": 2914 + }, + { + "epoch": 2.413307564881281, + "grad_norm": 0.9274892807006836, + "learning_rate": 3.31612666849625e-06, + "loss": 0.6015, + "step": 2915 + }, + { + "epoch": 2.4141358365543897, + "grad_norm": 0.9340214729309082, + "learning_rate": 3.3150871474995994e-06, + "loss": 0.5984, + "step": 2916 + }, + { + "epoch": 2.4149641082274984, + "grad_norm": 0.9376348853111267, + "learning_rate": 3.3140474687911844e-06, + "loss": 0.6205, + "step": 2917 + }, + { + "epoch": 2.4157923799006076, + "grad_norm": 0.9363590478897095, + "learning_rate": 3.313007632572173e-06, + "loss": 0.5926, + "step": 2918 + }, + { + "epoch": 2.4166206515737163, + "grad_norm": 0.8942010402679443, + "learning_rate": 3.3119676390437618e-06, + "loss": 0.5966, + "step": 2919 + }, + { + "epoch": 2.417448923246825, + "grad_norm": 0.9249299168586731, + "learning_rate": 3.310927488407183e-06, + "loss": 0.5951, + "step": 2920 + }, + { + "epoch": 2.4182771949199338, + "grad_norm": 0.9157326817512512, + "learning_rate": 3.3098871808636936e-06, + "loss": 0.5693, + "step": 2921 + }, + { + "epoch": 2.4191054665930425, + "grad_norm": 0.9088039994239807, + "learning_rate": 3.3088467166145843e-06, + "loss": 0.6084, + "step": 2922 + }, + { + "epoch": 2.419933738266151, + "grad_norm": 0.9258041381835938, + "learning_rate": 3.307806095861177e-06, + "loss": 0.6036, + "step": 2923 + }, + { + "epoch": 2.42076200993926, + "grad_norm": 0.905963659286499, + "learning_rate": 3.306765318804819e-06, + "loss": 0.5918, + "step": 2924 + }, + { + "epoch": 2.4215902816123687, + "grad_norm": 0.9063541293144226, + "learning_rate": 3.3057243856468924e-06, + "loss": 0.5906, + "step": 2925 + }, + { + "epoch": 2.4224185532854774, + "grad_norm": 0.9308085441589355, + "learning_rate": 3.304683296588808e-06, + "loss": 0.59, + "step": 2926 + }, + { + "epoch": 2.4232468249585866, + "grad_norm": 0.8989183902740479, + "learning_rate": 3.303642051832007e-06, + "loss": 0.5915, + "step": 2927 + }, + { + "epoch": 2.4240750966316953, + "grad_norm": 0.9413655996322632, + "learning_rate": 3.3026006515779614e-06, + "loss": 0.5952, + "step": 2928 + }, + { + "epoch": 2.424903368304804, + "grad_norm": 0.9431564211845398, + "learning_rate": 3.3015590960281708e-06, + "loss": 0.6007, + "step": 2929 + }, + { + "epoch": 2.4257316399779127, + "grad_norm": 0.8911551237106323, + "learning_rate": 3.3005173853841666e-06, + "loss": 0.573, + "step": 2930 + }, + { + "epoch": 2.4265599116510215, + "grad_norm": 0.9177538156509399, + "learning_rate": 3.299475519847511e-06, + "loss": 0.6116, + "step": 2931 + }, + { + "epoch": 2.42738818332413, + "grad_norm": 0.9592459797859192, + "learning_rate": 3.298433499619795e-06, + "loss": 0.6039, + "step": 2932 + }, + { + "epoch": 2.428216454997239, + "grad_norm": 0.9287276864051819, + "learning_rate": 3.2973913249026392e-06, + "loss": 0.5837, + "step": 2933 + }, + { + "epoch": 2.429044726670348, + "grad_norm": 0.9205804467201233, + "learning_rate": 3.296348995897696e-06, + "loss": 0.5904, + "step": 2934 + }, + { + "epoch": 2.429872998343457, + "grad_norm": 0.9075837731361389, + "learning_rate": 3.295306512806646e-06, + "loss": 0.5719, + "step": 2935 + }, + { + "epoch": 2.4307012700165656, + "grad_norm": 0.9203813672065735, + "learning_rate": 3.294263875831198e-06, + "loss": 0.604, + "step": 2936 + }, + { + "epoch": 2.4315295416896743, + "grad_norm": 0.8868334293365479, + "learning_rate": 3.2932210851730952e-06, + "loss": 0.6113, + "step": 2937 + }, + { + "epoch": 2.432357813362783, + "grad_norm": 0.9620448350906372, + "learning_rate": 3.2921781410341064e-06, + "loss": 0.5876, + "step": 2938 + }, + { + "epoch": 2.4331860850358917, + "grad_norm": 0.8991201519966125, + "learning_rate": 3.2911350436160317e-06, + "loss": 0.5994, + "step": 2939 + }, + { + "epoch": 2.4340143567090005, + "grad_norm": 0.9383120536804199, + "learning_rate": 3.2900917931207016e-06, + "loss": 0.6098, + "step": 2940 + }, + { + "epoch": 2.434842628382109, + "grad_norm": 0.9405770897865295, + "learning_rate": 3.289048389749974e-06, + "loss": 0.5978, + "step": 2941 + }, + { + "epoch": 2.435670900055218, + "grad_norm": 0.9153317213058472, + "learning_rate": 3.2880048337057387e-06, + "loss": 0.6091, + "step": 2942 + }, + { + "epoch": 2.436499171728327, + "grad_norm": 0.9396141171455383, + "learning_rate": 3.2869611251899136e-06, + "loss": 0.6143, + "step": 2943 + }, + { + "epoch": 2.437327443401436, + "grad_norm": 0.9370556473731995, + "learning_rate": 3.285917264404446e-06, + "loss": 0.61, + "step": 2944 + }, + { + "epoch": 2.4381557150745445, + "grad_norm": 0.9142118096351624, + "learning_rate": 3.2848732515513153e-06, + "loss": 0.5875, + "step": 2945 + }, + { + "epoch": 2.4389839867476533, + "grad_norm": 0.9196925759315491, + "learning_rate": 3.2838290868325267e-06, + "loss": 0.6058, + "step": 2946 + }, + { + "epoch": 2.439812258420762, + "grad_norm": 0.892217755317688, + "learning_rate": 3.282784770450116e-06, + "loss": 0.6005, + "step": 2947 + }, + { + "epoch": 2.4406405300938707, + "grad_norm": 0.9278191924095154, + "learning_rate": 3.2817403026061496e-06, + "loss": 0.6206, + "step": 2948 + }, + { + "epoch": 2.4414688017669794, + "grad_norm": 0.9343185424804688, + "learning_rate": 3.2806956835027216e-06, + "loss": 0.604, + "step": 2949 + }, + { + "epoch": 2.442297073440088, + "grad_norm": 0.9401823878288269, + "learning_rate": 3.2796509133419564e-06, + "loss": 0.6039, + "step": 2950 + }, + { + "epoch": 2.443125345113197, + "grad_norm": 0.9102358818054199, + "learning_rate": 3.278605992326008e-06, + "loss": 0.5892, + "step": 2951 + }, + { + "epoch": 2.443953616786306, + "grad_norm": 0.9241907000541687, + "learning_rate": 3.2775609206570573e-06, + "loss": 0.6054, + "step": 2952 + }, + { + "epoch": 2.444781888459415, + "grad_norm": 0.9492688775062561, + "learning_rate": 3.276515698537317e-06, + "loss": 0.5922, + "step": 2953 + }, + { + "epoch": 2.4456101601325235, + "grad_norm": 0.9492383003234863, + "learning_rate": 3.275470326169028e-06, + "loss": 0.5735, + "step": 2954 + }, + { + "epoch": 2.4464384318056323, + "grad_norm": 0.8857248425483704, + "learning_rate": 3.2744248037544597e-06, + "loss": 0.584, + "step": 2955 + }, + { + "epoch": 2.447266703478741, + "grad_norm": 0.9185032248497009, + "learning_rate": 3.2733791314959108e-06, + "loss": 0.5906, + "step": 2956 + }, + { + "epoch": 2.4480949751518497, + "grad_norm": 0.9257567524909973, + "learning_rate": 3.2723333095957094e-06, + "loss": 0.587, + "step": 2957 + }, + { + "epoch": 2.4489232468249584, + "grad_norm": 0.9427464008331299, + "learning_rate": 3.2712873382562117e-06, + "loss": 0.5932, + "step": 2958 + }, + { + "epoch": 2.4497515184980676, + "grad_norm": 0.9002125263214111, + "learning_rate": 3.2702412176798045e-06, + "loss": 0.58, + "step": 2959 + }, + { + "epoch": 2.4505797901711763, + "grad_norm": 0.906929075717926, + "learning_rate": 3.2691949480689018e-06, + "loss": 0.5878, + "step": 2960 + }, + { + "epoch": 2.451408061844285, + "grad_norm": 0.9162207841873169, + "learning_rate": 3.268148529625946e-06, + "loss": 0.5975, + "step": 2961 + }, + { + "epoch": 2.452236333517394, + "grad_norm": 0.9268636107444763, + "learning_rate": 3.2671019625534117e-06, + "loss": 0.6028, + "step": 2962 + }, + { + "epoch": 2.4530646051905025, + "grad_norm": 0.9216850996017456, + "learning_rate": 3.2660552470537975e-06, + "loss": 0.5938, + "step": 2963 + }, + { + "epoch": 2.4538928768636112, + "grad_norm": 0.9353698492050171, + "learning_rate": 3.2650083833296344e-06, + "loss": 0.6069, + "step": 2964 + }, + { + "epoch": 2.45472114853672, + "grad_norm": 0.9393898248672485, + "learning_rate": 3.2639613715834805e-06, + "loss": 0.5874, + "step": 2965 + }, + { + "epoch": 2.4555494202098287, + "grad_norm": 0.9410581588745117, + "learning_rate": 3.2629142120179224e-06, + "loss": 0.5918, + "step": 2966 + }, + { + "epoch": 2.4563776918829374, + "grad_norm": 0.9445161819458008, + "learning_rate": 3.2618669048355763e-06, + "loss": 0.5743, + "step": 2967 + }, + { + "epoch": 2.4572059635560466, + "grad_norm": 0.9591469764709473, + "learning_rate": 3.2608194502390867e-06, + "loss": 0.5781, + "step": 2968 + }, + { + "epoch": 2.4580342352291553, + "grad_norm": 0.9237606525421143, + "learning_rate": 3.2597718484311246e-06, + "loss": 0.6163, + "step": 2969 + }, + { + "epoch": 2.458862506902264, + "grad_norm": 0.9216769337654114, + "learning_rate": 3.2587240996143933e-06, + "loss": 0.6157, + "step": 2970 + }, + { + "epoch": 2.4596907785753728, + "grad_norm": 0.9081043004989624, + "learning_rate": 3.257676203991621e-06, + "loss": 0.5821, + "step": 2971 + }, + { + "epoch": 2.4605190502484815, + "grad_norm": 0.9316040873527527, + "learning_rate": 3.256628161765566e-06, + "loss": 0.5911, + "step": 2972 + }, + { + "epoch": 2.4613473219215902, + "grad_norm": 0.9326239824295044, + "learning_rate": 3.2555799731390158e-06, + "loss": 0.5963, + "step": 2973 + }, + { + "epoch": 2.462175593594699, + "grad_norm": 0.9429934024810791, + "learning_rate": 3.254531638314784e-06, + "loss": 0.5651, + "step": 2974 + }, + { + "epoch": 2.4630038652678077, + "grad_norm": 0.9090480804443359, + "learning_rate": 3.2534831574957134e-06, + "loss": 0.5779, + "step": 2975 + }, + { + "epoch": 2.4638321369409164, + "grad_norm": 0.9582802653312683, + "learning_rate": 3.2524345308846756e-06, + "loss": 0.5887, + "step": 2976 + }, + { + "epoch": 2.4646604086140256, + "grad_norm": 0.9248860478401184, + "learning_rate": 3.2513857586845705e-06, + "loss": 0.5936, + "step": 2977 + }, + { + "epoch": 2.4654886802871343, + "grad_norm": 0.9345134496688843, + "learning_rate": 3.250336841098325e-06, + "loss": 0.5985, + "step": 2978 + }, + { + "epoch": 2.466316951960243, + "grad_norm": 0.9092442393302917, + "learning_rate": 3.249287778328896e-06, + "loss": 0.5926, + "step": 2979 + }, + { + "epoch": 2.4671452236333518, + "grad_norm": 0.932470977306366, + "learning_rate": 3.2482385705792655e-06, + "loss": 0.5797, + "step": 2980 + }, + { + "epoch": 2.4679734953064605, + "grad_norm": 0.9335013031959534, + "learning_rate": 3.247189218052448e-06, + "loss": 0.5933, + "step": 2981 + }, + { + "epoch": 2.468801766979569, + "grad_norm": 0.9283298254013062, + "learning_rate": 3.2461397209514812e-06, + "loss": 0.6003, + "step": 2982 + }, + { + "epoch": 2.469630038652678, + "grad_norm": 0.9429066181182861, + "learning_rate": 3.245090079479434e-06, + "loss": 0.5944, + "step": 2983 + }, + { + "epoch": 2.4704583103257867, + "grad_norm": 0.9284510612487793, + "learning_rate": 3.2440402938394017e-06, + "loss": 0.5782, + "step": 2984 + }, + { + "epoch": 2.4712865819988954, + "grad_norm": 0.9150790572166443, + "learning_rate": 3.2429903642345094e-06, + "loss": 0.5862, + "step": 2985 + }, + { + "epoch": 2.4721148536720046, + "grad_norm": 0.9177762866020203, + "learning_rate": 3.2419402908679065e-06, + "loss": 0.5931, + "step": 2986 + }, + { + "epoch": 2.4729431253451133, + "grad_norm": 0.9098034501075745, + "learning_rate": 3.2408900739427744e-06, + "loss": 0.5864, + "step": 2987 + }, + { + "epoch": 2.473771397018222, + "grad_norm": 0.9237478971481323, + "learning_rate": 3.2398397136623195e-06, + "loss": 0.5873, + "step": 2988 + }, + { + "epoch": 2.4745996686913307, + "grad_norm": 0.9212510585784912, + "learning_rate": 3.2387892102297757e-06, + "loss": 0.5884, + "step": 2989 + }, + { + "epoch": 2.4754279403644395, + "grad_norm": 0.8987523913383484, + "learning_rate": 3.237738563848408e-06, + "loss": 0.5806, + "step": 2990 + }, + { + "epoch": 2.476256212037548, + "grad_norm": 0.9036835432052612, + "learning_rate": 3.236687774721504e-06, + "loss": 0.5993, + "step": 2991 + }, + { + "epoch": 2.477084483710657, + "grad_norm": 0.9291606545448303, + "learning_rate": 3.235636843052384e-06, + "loss": 0.5784, + "step": 2992 + }, + { + "epoch": 2.477912755383766, + "grad_norm": 0.9339813590049744, + "learning_rate": 3.234585769044392e-06, + "loss": 0.5951, + "step": 2993 + }, + { + "epoch": 2.478741027056875, + "grad_norm": 0.9638957977294922, + "learning_rate": 3.2335345529009017e-06, + "loss": 0.591, + "step": 2994 + }, + { + "epoch": 2.4795692987299836, + "grad_norm": 0.9618088603019714, + "learning_rate": 3.2324831948253123e-06, + "loss": 0.6102, + "step": 2995 + }, + { + "epoch": 2.4803975704030923, + "grad_norm": 0.9491007924079895, + "learning_rate": 3.2314316950210547e-06, + "loss": 0.599, + "step": 2996 + }, + { + "epoch": 2.481225842076201, + "grad_norm": 0.9213217496871948, + "learning_rate": 3.2303800536915804e-06, + "loss": 0.5955, + "step": 2997 + }, + { + "epoch": 2.4820541137493097, + "grad_norm": 0.923991322517395, + "learning_rate": 3.229328271040375e-06, + "loss": 0.5903, + "step": 2998 + }, + { + "epoch": 2.4828823854224185, + "grad_norm": 0.9173145890235901, + "learning_rate": 3.2282763472709484e-06, + "loss": 0.5899, + "step": 2999 + }, + { + "epoch": 2.483710657095527, + "grad_norm": 0.9357689023017883, + "learning_rate": 3.2272242825868363e-06, + "loss": 0.5885, + "step": 3000 + }, + { + "epoch": 2.484538928768636, + "grad_norm": 0.8902420401573181, + "learning_rate": 3.2261720771916056e-06, + "loss": 0.5899, + "step": 3001 + }, + { + "epoch": 2.485367200441745, + "grad_norm": 0.9079567790031433, + "learning_rate": 3.2251197312888475e-06, + "loss": 0.5966, + "step": 3002 + }, + { + "epoch": 2.486195472114854, + "grad_norm": 0.9068807363510132, + "learning_rate": 3.22406724508218e-06, + "loss": 0.6202, + "step": 3003 + }, + { + "epoch": 2.4870237437879625, + "grad_norm": 0.9458021521568298, + "learning_rate": 3.223014618775251e-06, + "loss": 0.5812, + "step": 3004 + }, + { + "epoch": 2.4878520154610713, + "grad_norm": 0.9506248235702515, + "learning_rate": 3.2219618525717328e-06, + "loss": 0.5819, + "step": 3005 + }, + { + "epoch": 2.48868028713418, + "grad_norm": 0.9573289155960083, + "learning_rate": 3.220908946675326e-06, + "loss": 0.581, + "step": 3006 + }, + { + "epoch": 2.4895085588072887, + "grad_norm": 0.9035286903381348, + "learning_rate": 3.219855901289758e-06, + "loss": 0.6182, + "step": 3007 + }, + { + "epoch": 2.4903368304803974, + "grad_norm": 0.9300901293754578, + "learning_rate": 3.218802716618784e-06, + "loss": 0.6079, + "step": 3008 + }, + { + "epoch": 2.491165102153506, + "grad_norm": 0.9116837382316589, + "learning_rate": 3.2177493928661854e-06, + "loss": 0.6034, + "step": 3009 + }, + { + "epoch": 2.491993373826615, + "grad_norm": 0.9291449189186096, + "learning_rate": 3.2166959302357694e-06, + "loss": 0.5889, + "step": 3010 + }, + { + "epoch": 2.492821645499724, + "grad_norm": 0.9002729654312134, + "learning_rate": 3.215642328931371e-06, + "loss": 0.5969, + "step": 3011 + }, + { + "epoch": 2.493649917172833, + "grad_norm": 0.8961556553840637, + "learning_rate": 3.214588589156854e-06, + "loss": 0.5899, + "step": 3012 + }, + { + "epoch": 2.4944781888459415, + "grad_norm": 0.9115450382232666, + "learning_rate": 3.2135347111161075e-06, + "loss": 0.5727, + "step": 3013 + }, + { + "epoch": 2.4953064605190503, + "grad_norm": 0.9060878157615662, + "learning_rate": 3.212480695013044e-06, + "loss": 0.5765, + "step": 3014 + }, + { + "epoch": 2.496134732192159, + "grad_norm": 0.902313768863678, + "learning_rate": 3.2114265410516088e-06, + "loss": 0.5913, + "step": 3015 + }, + { + "epoch": 2.4969630038652677, + "grad_norm": 0.931974470615387, + "learning_rate": 3.210372249435769e-06, + "loss": 0.5891, + "step": 3016 + }, + { + "epoch": 2.4977912755383764, + "grad_norm": 1.1711225509643555, + "learning_rate": 3.209317820369521e-06, + "loss": 0.6246, + "step": 3017 + }, + { + "epoch": 2.4986195472114856, + "grad_norm": 0.9561757445335388, + "learning_rate": 3.208263254056887e-06, + "loss": 0.5847, + "step": 3018 + }, + { + "epoch": 2.4994478188845943, + "grad_norm": 0.9621199369430542, + "learning_rate": 3.207208550701916e-06, + "loss": 0.5971, + "step": 3019 + }, + { + "epoch": 2.500276090557703, + "grad_norm": 0.9152470827102661, + "learning_rate": 3.206153710508683e-06, + "loss": 0.59, + "step": 3020 + }, + { + "epoch": 2.501104362230812, + "grad_norm": 0.9315138459205627, + "learning_rate": 3.20509873368129e-06, + "loss": 0.5841, + "step": 3021 + }, + { + "epoch": 2.5019326339039205, + "grad_norm": 0.9124136567115784, + "learning_rate": 3.204043620423864e-06, + "loss": 0.6007, + "step": 3022 + }, + { + "epoch": 2.5027609055770292, + "grad_norm": 0.8976690173149109, + "learning_rate": 3.2029883709405607e-06, + "loss": 0.5969, + "step": 3023 + }, + { + "epoch": 2.503589177250138, + "grad_norm": 0.8946467041969299, + "learning_rate": 3.201932985435561e-06, + "loss": 0.6069, + "step": 3024 + }, + { + "epoch": 2.5044174489232467, + "grad_norm": 0.9290626049041748, + "learning_rate": 3.2008774641130722e-06, + "loss": 0.591, + "step": 3025 + }, + { + "epoch": 2.5052457205963554, + "grad_norm": 0.8982367515563965, + "learning_rate": 3.1998218071773275e-06, + "loss": 0.5874, + "step": 3026 + }, + { + "epoch": 2.5060739922694646, + "grad_norm": 0.9062193036079407, + "learning_rate": 3.1987660148325867e-06, + "loss": 0.6107, + "step": 3027 + }, + { + "epoch": 2.5069022639425733, + "grad_norm": 0.9240522980690002, + "learning_rate": 3.1977100872831353e-06, + "loss": 0.5806, + "step": 3028 + }, + { + "epoch": 2.507730535615682, + "grad_norm": 0.9382046461105347, + "learning_rate": 3.1966540247332865e-06, + "loss": 0.5944, + "step": 3029 + }, + { + "epoch": 2.5085588072887908, + "grad_norm": 0.9194355010986328, + "learning_rate": 3.1955978273873776e-06, + "loss": 0.6085, + "step": 3030 + }, + { + "epoch": 2.5093870789618995, + "grad_norm": 0.9219276309013367, + "learning_rate": 3.1945414954497744e-06, + "loss": 0.5829, + "step": 3031 + }, + { + "epoch": 2.5102153506350082, + "grad_norm": 0.8998081088066101, + "learning_rate": 3.193485029124865e-06, + "loss": 0.6095, + "step": 3032 + }, + { + "epoch": 2.511043622308117, + "grad_norm": 0.9184247851371765, + "learning_rate": 3.192428428617068e-06, + "loss": 0.6129, + "step": 3033 + }, + { + "epoch": 2.5118718939812257, + "grad_norm": 0.9389240741729736, + "learning_rate": 3.1913716941308243e-06, + "loss": 0.607, + "step": 3034 + }, + { + "epoch": 2.5127001656543344, + "grad_norm": 0.9263836741447449, + "learning_rate": 3.190314825870603e-06, + "loss": 0.5761, + "step": 3035 + }, + { + "epoch": 2.5135284373274436, + "grad_norm": 0.9472658038139343, + "learning_rate": 3.1892578240408977e-06, + "loss": 0.5906, + "step": 3036 + }, + { + "epoch": 2.5143567090005523, + "grad_norm": 0.9045345783233643, + "learning_rate": 3.1882006888462285e-06, + "loss": 0.5993, + "step": 3037 + }, + { + "epoch": 2.515184980673661, + "grad_norm": 0.8930817246437073, + "learning_rate": 3.187143420491142e-06, + "loss": 0.5947, + "step": 3038 + }, + { + "epoch": 2.5160132523467698, + "grad_norm": 0.9130663871765137, + "learning_rate": 3.186086019180208e-06, + "loss": 0.5892, + "step": 3039 + }, + { + "epoch": 2.5168415240198785, + "grad_norm": 0.9008771777153015, + "learning_rate": 3.185028485118026e-06, + "loss": 0.5787, + "step": 3040 + }, + { + "epoch": 2.517669795692987, + "grad_norm": 0.9192292094230652, + "learning_rate": 3.1839708185092177e-06, + "loss": 0.5848, + "step": 3041 + }, + { + "epoch": 2.518498067366096, + "grad_norm": 0.8970836997032166, + "learning_rate": 3.1829130195584314e-06, + "loss": 0.5939, + "step": 3042 + }, + { + "epoch": 2.519326339039205, + "grad_norm": 0.9393385052680969, + "learning_rate": 3.181855088470342e-06, + "loss": 0.6096, + "step": 3043 + }, + { + "epoch": 2.5201546107123134, + "grad_norm": 0.9437024593353271, + "learning_rate": 3.180797025449649e-06, + "loss": 0.5998, + "step": 3044 + }, + { + "epoch": 2.5209828823854226, + "grad_norm": 0.9014863967895508, + "learning_rate": 3.179738830701078e-06, + "loss": 0.6015, + "step": 3045 + }, + { + "epoch": 2.5218111540585313, + "grad_norm": 0.9243669509887695, + "learning_rate": 3.1786805044293802e-06, + "loss": 0.5971, + "step": 3046 + }, + { + "epoch": 2.52263942573164, + "grad_norm": 0.9175970554351807, + "learning_rate": 3.1776220468393305e-06, + "loss": 0.6084, + "step": 3047 + }, + { + "epoch": 2.5234676974047487, + "grad_norm": 0.8962888121604919, + "learning_rate": 3.176563458135732e-06, + "loss": 0.5979, + "step": 3048 + }, + { + "epoch": 2.5242959690778575, + "grad_norm": 0.9321010708808899, + "learning_rate": 3.1755047385234117e-06, + "loss": 0.5941, + "step": 3049 + }, + { + "epoch": 2.525124240750966, + "grad_norm": 0.9147899746894836, + "learning_rate": 3.17444588820722e-06, + "loss": 0.5987, + "step": 3050 + }, + { + "epoch": 2.525952512424075, + "grad_norm": 0.9318193793296814, + "learning_rate": 3.173386907392037e-06, + "loss": 0.5658, + "step": 3051 + }, + { + "epoch": 2.526780784097184, + "grad_norm": 0.9039092659950256, + "learning_rate": 3.1723277962827647e-06, + "loss": 0.5956, + "step": 3052 + }, + { + "epoch": 2.5276090557702924, + "grad_norm": 0.9080818295478821, + "learning_rate": 3.1712685550843314e-06, + "loss": 0.5957, + "step": 3053 + }, + { + "epoch": 2.5284373274434015, + "grad_norm": 0.9120905995368958, + "learning_rate": 3.17020918400169e-06, + "loss": 0.5917, + "step": 3054 + }, + { + "epoch": 2.5292655991165103, + "grad_norm": 0.8983855843544006, + "learning_rate": 3.169149683239819e-06, + "loss": 0.6045, + "step": 3055 + }, + { + "epoch": 2.530093870789619, + "grad_norm": 0.9203280210494995, + "learning_rate": 3.168090053003721e-06, + "loss": 0.6093, + "step": 3056 + }, + { + "epoch": 2.5309221424627277, + "grad_norm": 0.9343932867050171, + "learning_rate": 3.1670302934984272e-06, + "loss": 0.5654, + "step": 3057 + }, + { + "epoch": 2.5317504141358365, + "grad_norm": 0.9054449200630188, + "learning_rate": 3.1659704049289885e-06, + "loss": 0.5878, + "step": 3058 + }, + { + "epoch": 2.532578685808945, + "grad_norm": 0.9011321067810059, + "learning_rate": 3.1649103875004856e-06, + "loss": 0.5831, + "step": 3059 + }, + { + "epoch": 2.533406957482054, + "grad_norm": 0.8886737823486328, + "learning_rate": 3.1638502414180206e-06, + "loss": 0.5915, + "step": 3060 + }, + { + "epoch": 2.534235229155163, + "grad_norm": 0.9289557337760925, + "learning_rate": 3.1627899668867214e-06, + "loss": 0.5944, + "step": 3061 + }, + { + "epoch": 2.535063500828272, + "grad_norm": 0.9269216060638428, + "learning_rate": 3.1617295641117425e-06, + "loss": 0.605, + "step": 3062 + }, + { + "epoch": 2.5358917725013805, + "grad_norm": 0.8877415657043457, + "learning_rate": 3.1606690332982613e-06, + "loss": 0.5954, + "step": 3063 + }, + { + "epoch": 2.5367200441744893, + "grad_norm": 0.9080086350440979, + "learning_rate": 3.1596083746514817e-06, + "loss": 0.5896, + "step": 3064 + }, + { + "epoch": 2.537548315847598, + "grad_norm": 0.9625695943832397, + "learning_rate": 3.15854758837663e-06, + "loss": 0.5666, + "step": 3065 + }, + { + "epoch": 2.5383765875207067, + "grad_norm": 0.9442515969276428, + "learning_rate": 3.1574866746789583e-06, + "loss": 0.5837, + "step": 3066 + }, + { + "epoch": 2.5392048591938154, + "grad_norm": 0.9054381847381592, + "learning_rate": 3.1564256337637443e-06, + "loss": 0.5799, + "step": 3067 + }, + { + "epoch": 2.5400331308669246, + "grad_norm": 0.9435595273971558, + "learning_rate": 3.1553644658362897e-06, + "loss": 0.6135, + "step": 3068 + }, + { + "epoch": 2.540861402540033, + "grad_norm": 0.9356298446655273, + "learning_rate": 3.1543031711019203e-06, + "loss": 0.5853, + "step": 3069 + }, + { + "epoch": 2.541689674213142, + "grad_norm": 0.917759895324707, + "learning_rate": 3.153241749765986e-06, + "loss": 0.5902, + "step": 3070 + }, + { + "epoch": 2.542517945886251, + "grad_norm": 0.9233857989311218, + "learning_rate": 3.1521802020338627e-06, + "loss": 0.6066, + "step": 3071 + }, + { + "epoch": 2.5433462175593595, + "grad_norm": 0.9497340321540833, + "learning_rate": 3.1511185281109504e-06, + "loss": 0.6003, + "step": 3072 + }, + { + "epoch": 2.5441744892324683, + "grad_norm": 0.9373629689216614, + "learning_rate": 3.1500567282026716e-06, + "loss": 0.5827, + "step": 3073 + }, + { + "epoch": 2.545002760905577, + "grad_norm": 0.8916833996772766, + "learning_rate": 3.148994802514476e-06, + "loss": 0.5996, + "step": 3074 + }, + { + "epoch": 2.5458310325786857, + "grad_norm": 0.9327612519264221, + "learning_rate": 3.1479327512518363e-06, + "loss": 0.6159, + "step": 3075 + }, + { + "epoch": 2.5466593042517944, + "grad_norm": 0.9380506873130798, + "learning_rate": 3.1468705746202493e-06, + "loss": 0.5924, + "step": 3076 + }, + { + "epoch": 2.5474875759249036, + "grad_norm": 0.9138150215148926, + "learning_rate": 3.1458082728252358e-06, + "loss": 0.5734, + "step": 3077 + }, + { + "epoch": 2.548315847598012, + "grad_norm": 0.8954431414604187, + "learning_rate": 3.144745846072341e-06, + "loss": 0.5902, + "step": 3078 + }, + { + "epoch": 2.549144119271121, + "grad_norm": 0.9259476661682129, + "learning_rate": 3.1436832945671362e-06, + "loss": 0.6057, + "step": 3079 + }, + { + "epoch": 2.54997239094423, + "grad_norm": 0.9295196533203125, + "learning_rate": 3.142620618515214e-06, + "loss": 0.5971, + "step": 3080 + }, + { + "epoch": 2.5508006626173385, + "grad_norm": 0.9318671822547913, + "learning_rate": 3.1415578181221928e-06, + "loss": 0.5852, + "step": 3081 + }, + { + "epoch": 2.5516289342904472, + "grad_norm": 0.9349870085716248, + "learning_rate": 3.1404948935937145e-06, + "loss": 0.5736, + "step": 3082 + }, + { + "epoch": 2.552457205963556, + "grad_norm": 0.9280831217765808, + "learning_rate": 3.1394318451354446e-06, + "loss": 0.5898, + "step": 3083 + }, + { + "epoch": 2.5532854776366647, + "grad_norm": 0.8969367742538452, + "learning_rate": 3.1383686729530727e-06, + "loss": 0.5983, + "step": 3084 + }, + { + "epoch": 2.5541137493097734, + "grad_norm": 0.9363370537757874, + "learning_rate": 3.1373053772523145e-06, + "loss": 0.6052, + "step": 3085 + }, + { + "epoch": 2.5549420209828826, + "grad_norm": 0.8944993615150452, + "learning_rate": 3.1362419582389057e-06, + "loss": 0.599, + "step": 3086 + }, + { + "epoch": 2.5557702926559913, + "grad_norm": 0.891412615776062, + "learning_rate": 3.1351784161186093e-06, + "loss": 0.5916, + "step": 3087 + }, + { + "epoch": 2.5565985643291, + "grad_norm": 0.930220365524292, + "learning_rate": 3.134114751097211e-06, + "loss": 0.5826, + "step": 3088 + }, + { + "epoch": 2.5574268360022088, + "grad_norm": 0.9193615317344666, + "learning_rate": 3.1330509633805183e-06, + "loss": 0.5896, + "step": 3089 + }, + { + "epoch": 2.5582551076753175, + "grad_norm": 0.9178160429000854, + "learning_rate": 3.131987053174367e-06, + "loss": 0.6104, + "step": 3090 + }, + { + "epoch": 2.5590833793484262, + "grad_norm": 0.9119648933410645, + "learning_rate": 3.130923020684611e-06, + "loss": 0.6198, + "step": 3091 + }, + { + "epoch": 2.559911651021535, + "grad_norm": 0.957272469997406, + "learning_rate": 3.1298588661171323e-06, + "loss": 0.5716, + "step": 3092 + }, + { + "epoch": 2.5607399226946437, + "grad_norm": 0.9073137044906616, + "learning_rate": 3.128794589677835e-06, + "loss": 0.5959, + "step": 3093 + }, + { + "epoch": 2.5615681943677524, + "grad_norm": 0.9372225403785706, + "learning_rate": 3.1277301915726456e-06, + "loss": 0.6146, + "step": 3094 + }, + { + "epoch": 2.5623964660408616, + "grad_norm": 0.9377845525741577, + "learning_rate": 3.126665672007515e-06, + "loss": 0.6055, + "step": 3095 + }, + { + "epoch": 2.5632247377139703, + "grad_norm": 0.9114481210708618, + "learning_rate": 3.1256010311884193e-06, + "loss": 0.6038, + "step": 3096 + }, + { + "epoch": 2.564053009387079, + "grad_norm": 0.9000943303108215, + "learning_rate": 3.124536269321356e-06, + "loss": 0.5806, + "step": 3097 + }, + { + "epoch": 2.5648812810601878, + "grad_norm": 0.9771347641944885, + "learning_rate": 3.123471386612347e-06, + "loss": 0.6029, + "step": 3098 + }, + { + "epoch": 2.5657095527332965, + "grad_norm": 0.9150393605232239, + "learning_rate": 3.1224063832674358e-06, + "loss": 0.6033, + "step": 3099 + }, + { + "epoch": 2.566537824406405, + "grad_norm": 0.8910076022148132, + "learning_rate": 3.1213412594926906e-06, + "loss": 0.6029, + "step": 3100 + }, + { + "epoch": 2.567366096079514, + "grad_norm": 0.9558795094490051, + "learning_rate": 3.1202760154942047e-06, + "loss": 0.5792, + "step": 3101 + }, + { + "epoch": 2.568194367752623, + "grad_norm": 0.958629310131073, + "learning_rate": 3.1192106514780924e-06, + "loss": 0.5933, + "step": 3102 + }, + { + "epoch": 2.5690226394257314, + "grad_norm": 0.9466328620910645, + "learning_rate": 3.1181451676504907e-06, + "loss": 0.6027, + "step": 3103 + }, + { + "epoch": 2.5698509110988406, + "grad_norm": 0.9397157430648804, + "learning_rate": 3.1170795642175614e-06, + "loss": 0.5894, + "step": 3104 + }, + { + "epoch": 2.5706791827719493, + "grad_norm": 0.9229122996330261, + "learning_rate": 3.116013841385489e-06, + "loss": 0.5812, + "step": 3105 + }, + { + "epoch": 2.571507454445058, + "grad_norm": 0.9332060217857361, + "learning_rate": 3.1149479993604805e-06, + "loss": 0.5738, + "step": 3106 + }, + { + "epoch": 2.5723357261181667, + "grad_norm": 0.9117311239242554, + "learning_rate": 3.113882038348766e-06, + "loss": 0.5923, + "step": 3107 + }, + { + "epoch": 2.5731639977912755, + "grad_norm": 0.9390400052070618, + "learning_rate": 3.1128159585566003e-06, + "loss": 0.6029, + "step": 3108 + }, + { + "epoch": 2.573992269464384, + "grad_norm": 0.9229103326797485, + "learning_rate": 3.1117497601902596e-06, + "loss": 0.6118, + "step": 3109 + }, + { + "epoch": 2.574820541137493, + "grad_norm": 0.8975940346717834, + "learning_rate": 3.1106834434560424e-06, + "loss": 0.6005, + "step": 3110 + }, + { + "epoch": 2.575648812810602, + "grad_norm": 0.9088030457496643, + "learning_rate": 3.1096170085602718e-06, + "loss": 0.5896, + "step": 3111 + }, + { + "epoch": 2.576477084483711, + "grad_norm": 0.982429027557373, + "learning_rate": 3.108550455709292e-06, + "loss": 0.5947, + "step": 3112 + }, + { + "epoch": 2.5773053561568195, + "grad_norm": 0.9404535293579102, + "learning_rate": 3.107483785109473e-06, + "loss": 0.6154, + "step": 3113 + }, + { + "epoch": 2.5781336278299283, + "grad_norm": 0.9074321389198303, + "learning_rate": 3.1064169969672043e-06, + "loss": 0.5889, + "step": 3114 + }, + { + "epoch": 2.578961899503037, + "grad_norm": 0.9000370502471924, + "learning_rate": 3.105350091488899e-06, + "loss": 0.5725, + "step": 3115 + }, + { + "epoch": 2.5797901711761457, + "grad_norm": 0.9296747446060181, + "learning_rate": 3.1042830688809944e-06, + "loss": 0.5983, + "step": 3116 + }, + { + "epoch": 2.5806184428492545, + "grad_norm": 0.9164032936096191, + "learning_rate": 3.1032159293499486e-06, + "loss": 0.5832, + "step": 3117 + }, + { + "epoch": 2.581446714522363, + "grad_norm": 0.9070228338241577, + "learning_rate": 3.1021486731022443e-06, + "loss": 0.5977, + "step": 3118 + }, + { + "epoch": 2.582274986195472, + "grad_norm": 0.9384707808494568, + "learning_rate": 3.1010813003443847e-06, + "loss": 0.5926, + "step": 3119 + }, + { + "epoch": 2.583103257868581, + "grad_norm": 0.937782347202301, + "learning_rate": 3.100013811282896e-06, + "loss": 0.5859, + "step": 3120 + }, + { + "epoch": 2.58393152954169, + "grad_norm": 0.9293384552001953, + "learning_rate": 3.098946206124329e-06, + "loss": 0.5826, + "step": 3121 + }, + { + "epoch": 2.5847598012147985, + "grad_norm": 0.935950756072998, + "learning_rate": 3.097878485075255e-06, + "loss": 0.589, + "step": 3122 + }, + { + "epoch": 2.5855880728879073, + "grad_norm": 0.9377515316009521, + "learning_rate": 3.0968106483422655e-06, + "loss": 0.5985, + "step": 3123 + }, + { + "epoch": 2.586416344561016, + "grad_norm": 0.9256250858306885, + "learning_rate": 3.09574269613198e-06, + "loss": 0.5944, + "step": 3124 + }, + { + "epoch": 2.5872446162341247, + "grad_norm": 0.873123288154602, + "learning_rate": 3.0946746286510364e-06, + "loss": 0.5918, + "step": 3125 + }, + { + "epoch": 2.5880728879072334, + "grad_norm": 0.9320067167282104, + "learning_rate": 3.093606446106096e-06, + "loss": 0.6365, + "step": 3126 + }, + { + "epoch": 2.5889011595803426, + "grad_norm": 0.9514285326004028, + "learning_rate": 3.0925381487038415e-06, + "loss": 0.585, + "step": 3127 + }, + { + "epoch": 2.589729431253451, + "grad_norm": 0.9028947353363037, + "learning_rate": 3.0914697366509777e-06, + "loss": 0.6063, + "step": 3128 + }, + { + "epoch": 2.59055770292656, + "grad_norm": 0.8989999294281006, + "learning_rate": 3.0904012101542346e-06, + "loss": 0.6012, + "step": 3129 + }, + { + "epoch": 2.591385974599669, + "grad_norm": 0.9131087064743042, + "learning_rate": 3.089332569420361e-06, + "loss": 0.5742, + "step": 3130 + }, + { + "epoch": 2.5922142462727775, + "grad_norm": 0.916441023349762, + "learning_rate": 3.088263814656128e-06, + "loss": 0.6011, + "step": 3131 + }, + { + "epoch": 2.5930425179458862, + "grad_norm": 0.8957144618034363, + "learning_rate": 3.087194946068332e-06, + "loss": 0.6008, + "step": 3132 + }, + { + "epoch": 2.593870789618995, + "grad_norm": 0.90632563829422, + "learning_rate": 3.086125963863787e-06, + "loss": 0.6062, + "step": 3133 + }, + { + "epoch": 2.5946990612921037, + "grad_norm": 0.94493168592453, + "learning_rate": 3.0850568682493315e-06, + "loss": 0.5667, + "step": 3134 + }, + { + "epoch": 2.5955273329652124, + "grad_norm": 0.9155673980712891, + "learning_rate": 3.0839876594318257e-06, + "loss": 0.5755, + "step": 3135 + }, + { + "epoch": 2.5963556046383216, + "grad_norm": 0.9013570547103882, + "learning_rate": 3.0829183376181524e-06, + "loss": 0.6053, + "step": 3136 + }, + { + "epoch": 2.59718387631143, + "grad_norm": 0.9294494390487671, + "learning_rate": 3.081848903015214e-06, + "loss": 0.5881, + "step": 3137 + }, + { + "epoch": 2.598012147984539, + "grad_norm": 0.8956479430198669, + "learning_rate": 3.0807793558299377e-06, + "loss": 0.5976, + "step": 3138 + }, + { + "epoch": 2.598840419657648, + "grad_norm": 0.9205741286277771, + "learning_rate": 3.079709696269269e-06, + "loss": 0.5776, + "step": 3139 + }, + { + "epoch": 2.5996686913307565, + "grad_norm": 0.901066780090332, + "learning_rate": 3.0786399245401784e-06, + "loss": 0.6101, + "step": 3140 + }, + { + "epoch": 2.6004969630038652, + "grad_norm": 0.9003482460975647, + "learning_rate": 3.0775700408496568e-06, + "loss": 0.5983, + "step": 3141 + }, + { + "epoch": 2.601325234676974, + "grad_norm": 0.9000759124755859, + "learning_rate": 3.0765000454047155e-06, + "loss": 0.605, + "step": 3142 + }, + { + "epoch": 2.6021535063500827, + "grad_norm": 0.8797931671142578, + "learning_rate": 3.0754299384123905e-06, + "loss": 0.5874, + "step": 3143 + }, + { + "epoch": 2.6029817780231914, + "grad_norm": 0.9307072758674622, + "learning_rate": 3.074359720079736e-06, + "loss": 0.6035, + "step": 3144 + }, + { + "epoch": 2.6038100496963006, + "grad_norm": 0.8928053379058838, + "learning_rate": 3.073289390613829e-06, + "loss": 0.5759, + "step": 3145 + }, + { + "epoch": 2.6046383213694093, + "grad_norm": 0.8879983425140381, + "learning_rate": 3.0722189502217705e-06, + "loss": 0.5947, + "step": 3146 + }, + { + "epoch": 2.605466593042518, + "grad_norm": 0.8889674544334412, + "learning_rate": 3.0711483991106784e-06, + "loss": 0.5939, + "step": 3147 + }, + { + "epoch": 2.6062948647156268, + "grad_norm": 0.9318514466285706, + "learning_rate": 3.070077737487695e-06, + "loss": 0.599, + "step": 3148 + }, + { + "epoch": 2.6071231363887355, + "grad_norm": 0.9251010417938232, + "learning_rate": 3.069006965559985e-06, + "loss": 0.5956, + "step": 3149 + }, + { + "epoch": 2.6079514080618442, + "grad_norm": 0.9122523069381714, + "learning_rate": 3.06793608353473e-06, + "loss": 0.6011, + "step": 3150 + }, + { + "epoch": 2.608779679734953, + "grad_norm": 0.9270116090774536, + "learning_rate": 3.0668650916191365e-06, + "loss": 0.5923, + "step": 3151 + }, + { + "epoch": 2.609607951408062, + "grad_norm": 0.939582884311676, + "learning_rate": 3.0657939900204338e-06, + "loss": 0.6053, + "step": 3152 + }, + { + "epoch": 2.6104362230811704, + "grad_norm": 0.9206486940383911, + "learning_rate": 3.064722778945867e-06, + "loss": 0.6197, + "step": 3153 + }, + { + "epoch": 2.6112644947542796, + "grad_norm": 0.9047932624816895, + "learning_rate": 3.063651458602708e-06, + "loss": 0.5774, + "step": 3154 + }, + { + "epoch": 2.6120927664273883, + "grad_norm": 0.9285469055175781, + "learning_rate": 3.062580029198245e-06, + "loss": 0.6107, + "step": 3155 + }, + { + "epoch": 2.612921038100497, + "grad_norm": 0.9187634587287903, + "learning_rate": 3.0615084909397906e-06, + "loss": 0.5919, + "step": 3156 + }, + { + "epoch": 2.6137493097736058, + "grad_norm": 0.8950710296630859, + "learning_rate": 3.060436844034678e-06, + "loss": 0.6029, + "step": 3157 + }, + { + "epoch": 2.6145775814467145, + "grad_norm": 0.9206116795539856, + "learning_rate": 3.0593650886902606e-06, + "loss": 0.5632, + "step": 3158 + }, + { + "epoch": 2.615405853119823, + "grad_norm": 0.9081333875656128, + "learning_rate": 3.058293225113912e-06, + "loss": 0.6127, + "step": 3159 + }, + { + "epoch": 2.616234124792932, + "grad_norm": 0.9028277397155762, + "learning_rate": 3.0572212535130297e-06, + "loss": 0.6089, + "step": 3160 + }, + { + "epoch": 2.617062396466041, + "grad_norm": 0.9449213147163391, + "learning_rate": 3.0561491740950288e-06, + "loss": 0.5841, + "step": 3161 + }, + { + "epoch": 2.6178906681391494, + "grad_norm": 0.9926438927650452, + "learning_rate": 3.055076987067347e-06, + "loss": 0.564, + "step": 3162 + }, + { + "epoch": 2.6187189398122586, + "grad_norm": 0.9098719954490662, + "learning_rate": 3.0540046926374422e-06, + "loss": 0.6268, + "step": 3163 + }, + { + "epoch": 2.6195472114853673, + "grad_norm": 0.8997418880462646, + "learning_rate": 3.0529322910127944e-06, + "loss": 0.5972, + "step": 3164 + }, + { + "epoch": 2.620375483158476, + "grad_norm": 0.9076839685440063, + "learning_rate": 3.0518597824009015e-06, + "loss": 0.5921, + "step": 3165 + }, + { + "epoch": 2.6212037548315847, + "grad_norm": 0.9348479509353638, + "learning_rate": 3.050787167009287e-06, + "loss": 0.5802, + "step": 3166 + }, + { + "epoch": 2.6220320265046935, + "grad_norm": 0.9381133913993835, + "learning_rate": 3.0497144450454886e-06, + "loss": 0.5943, + "step": 3167 + }, + { + "epoch": 2.622860298177802, + "grad_norm": 0.9327574968338013, + "learning_rate": 3.0486416167170694e-06, + "loss": 0.5832, + "step": 3168 + }, + { + "epoch": 2.623688569850911, + "grad_norm": 0.9142323732376099, + "learning_rate": 3.0475686822316126e-06, + "loss": 0.6048, + "step": 3169 + }, + { + "epoch": 2.62451684152402, + "grad_norm": 0.9799601435661316, + "learning_rate": 3.04649564179672e-06, + "loss": 0.5783, + "step": 3170 + }, + { + "epoch": 2.625345113197129, + "grad_norm": 0.9203731417655945, + "learning_rate": 3.045422495620015e-06, + "loss": 0.6028, + "step": 3171 + }, + { + "epoch": 2.6261733848702375, + "grad_norm": 0.9614686369895935, + "learning_rate": 3.0443492439091415e-06, + "loss": 0.5895, + "step": 3172 + }, + { + "epoch": 2.6270016565433463, + "grad_norm": 0.980112612247467, + "learning_rate": 3.0432758868717626e-06, + "loss": 0.5887, + "step": 3173 + }, + { + "epoch": 2.627829928216455, + "grad_norm": 0.9174870252609253, + "learning_rate": 3.042202424715565e-06, + "loss": 0.5831, + "step": 3174 + }, + { + "epoch": 2.6286581998895637, + "grad_norm": 0.9396457672119141, + "learning_rate": 3.0411288576482525e-06, + "loss": 0.5816, + "step": 3175 + }, + { + "epoch": 2.6294864715626725, + "grad_norm": 0.9067835211753845, + "learning_rate": 3.04005518587755e-06, + "loss": 0.5748, + "step": 3176 + }, + { + "epoch": 2.630314743235781, + "grad_norm": 0.9329981207847595, + "learning_rate": 3.038981409611204e-06, + "loss": 0.5901, + "step": 3177 + }, + { + "epoch": 2.63114301490889, + "grad_norm": 0.9418731927871704, + "learning_rate": 3.0379075290569793e-06, + "loss": 0.5981, + "step": 3178 + }, + { + "epoch": 2.631971286581999, + "grad_norm": 0.9180083274841309, + "learning_rate": 3.0368335444226614e-06, + "loss": 0.5653, + "step": 3179 + }, + { + "epoch": 2.632799558255108, + "grad_norm": 0.9566802978515625, + "learning_rate": 3.0357594559160575e-06, + "loss": 0.5968, + "step": 3180 + }, + { + "epoch": 2.6336278299282165, + "grad_norm": 0.9177946448326111, + "learning_rate": 3.034685263744994e-06, + "loss": 0.6076, + "step": 3181 + }, + { + "epoch": 2.6344561016013253, + "grad_norm": 0.905975341796875, + "learning_rate": 3.033610968117314e-06, + "loss": 0.6064, + "step": 3182 + }, + { + "epoch": 2.635284373274434, + "grad_norm": 0.9021571278572083, + "learning_rate": 3.0325365692408883e-06, + "loss": 0.5818, + "step": 3183 + }, + { + "epoch": 2.6361126449475427, + "grad_norm": 0.9211055040359497, + "learning_rate": 3.0314620673235995e-06, + "loss": 0.5811, + "step": 3184 + }, + { + "epoch": 2.6369409166206514, + "grad_norm": 0.929892897605896, + "learning_rate": 3.0303874625733548e-06, + "loss": 0.6039, + "step": 3185 + }, + { + "epoch": 2.6377691882937606, + "grad_norm": 0.9520539045333862, + "learning_rate": 3.0293127551980805e-06, + "loss": 0.6156, + "step": 3186 + }, + { + "epoch": 2.638597459966869, + "grad_norm": 0.9111108779907227, + "learning_rate": 3.028237945405722e-06, + "loss": 0.6081, + "step": 3187 + }, + { + "epoch": 2.639425731639978, + "grad_norm": 0.9411360025405884, + "learning_rate": 3.027163033404246e-06, + "loss": 0.5897, + "step": 3188 + }, + { + "epoch": 2.640254003313087, + "grad_norm": 0.9253203868865967, + "learning_rate": 3.026088019401637e-06, + "loss": 0.5957, + "step": 3189 + }, + { + "epoch": 2.6410822749861955, + "grad_norm": 1.010617971420288, + "learning_rate": 3.0250129036059e-06, + "loss": 0.602, + "step": 3190 + }, + { + "epoch": 2.6419105466593042, + "grad_norm": 0.9021222591400146, + "learning_rate": 3.02393768622506e-06, + "loss": 0.5796, + "step": 3191 + }, + { + "epoch": 2.642738818332413, + "grad_norm": 0.9229567050933838, + "learning_rate": 3.0228623674671627e-06, + "loss": 0.5827, + "step": 3192 + }, + { + "epoch": 2.6435670900055217, + "grad_norm": 0.9603396654129028, + "learning_rate": 3.0217869475402704e-06, + "loss": 0.5886, + "step": 3193 + }, + { + "epoch": 2.6443953616786304, + "grad_norm": 0.9483562707901001, + "learning_rate": 3.020711426652469e-06, + "loss": 0.6089, + "step": 3194 + }, + { + "epoch": 2.6452236333517396, + "grad_norm": 0.9240924715995789, + "learning_rate": 3.019635805011859e-06, + "loss": 0.6182, + "step": 3195 + }, + { + "epoch": 2.646051905024848, + "grad_norm": 0.9176941514015198, + "learning_rate": 3.018560082826566e-06, + "loss": 0.593, + "step": 3196 + }, + { + "epoch": 2.646880176697957, + "grad_norm": 0.9058155417442322, + "learning_rate": 3.0174842603047307e-06, + "loss": 0.5853, + "step": 3197 + }, + { + "epoch": 2.647708448371066, + "grad_norm": 0.9544078707695007, + "learning_rate": 3.016408337654514e-06, + "loss": 0.5743, + "step": 3198 + }, + { + "epoch": 2.6485367200441745, + "grad_norm": 0.944963812828064, + "learning_rate": 3.0153323150840987e-06, + "loss": 0.6007, + "step": 3199 + }, + { + "epoch": 2.6493649917172832, + "grad_norm": 0.9196237921714783, + "learning_rate": 3.014256192801685e-06, + "loss": 0.5818, + "step": 3200 + }, + { + "epoch": 2.650193263390392, + "grad_norm": 0.9315558075904846, + "learning_rate": 3.01317997101549e-06, + "loss": 0.5861, + "step": 3201 + }, + { + "epoch": 2.6510215350635007, + "grad_norm": 0.9272212982177734, + "learning_rate": 3.012103649933756e-06, + "loss": 0.597, + "step": 3202 + }, + { + "epoch": 2.6518498067366094, + "grad_norm": 0.9128772616386414, + "learning_rate": 3.011027229764739e-06, + "loss": 0.5863, + "step": 3203 + }, + { + "epoch": 2.6526780784097186, + "grad_norm": 0.9234123826026917, + "learning_rate": 3.0099507107167158e-06, + "loss": 0.589, + "step": 3204 + }, + { + "epoch": 2.6535063500828273, + "grad_norm": 0.9162564873695374, + "learning_rate": 3.0088740929979854e-06, + "loss": 0.5855, + "step": 3205 + }, + { + "epoch": 2.654334621755936, + "grad_norm": 0.8842553496360779, + "learning_rate": 3.0077973768168604e-06, + "loss": 0.5947, + "step": 3206 + }, + { + "epoch": 2.6551628934290448, + "grad_norm": 0.9413550496101379, + "learning_rate": 3.0067205623816775e-06, + "loss": 0.5882, + "step": 3207 + }, + { + "epoch": 2.6559911651021535, + "grad_norm": 0.9217298626899719, + "learning_rate": 3.005643649900789e-06, + "loss": 0.5806, + "step": 3208 + }, + { + "epoch": 2.656819436775262, + "grad_norm": 0.9214861392974854, + "learning_rate": 3.004566639582568e-06, + "loss": 0.5656, + "step": 3209 + }, + { + "epoch": 2.657647708448371, + "grad_norm": 0.9789690375328064, + "learning_rate": 3.003489531635405e-06, + "loss": 0.5897, + "step": 3210 + }, + { + "epoch": 2.65847598012148, + "grad_norm": 0.9465369582176208, + "learning_rate": 3.0024123262677123e-06, + "loss": 0.5996, + "step": 3211 + }, + { + "epoch": 2.6593042517945884, + "grad_norm": 0.9011247754096985, + "learning_rate": 3.001335023687917e-06, + "loss": 0.5849, + "step": 3212 + }, + { + "epoch": 2.6601325234676976, + "grad_norm": 0.9762759804725647, + "learning_rate": 3.0002576241044683e-06, + "loss": 0.6161, + "step": 3213 + }, + { + "epoch": 2.6609607951408063, + "grad_norm": 0.9362512230873108, + "learning_rate": 2.9991801277258327e-06, + "loss": 0.6186, + "step": 3214 + }, + { + "epoch": 2.661789066813915, + "grad_norm": 0.9391586780548096, + "learning_rate": 2.998102534760495e-06, + "loss": 0.5787, + "step": 3215 + }, + { + "epoch": 2.6626173384870238, + "grad_norm": 0.9361854195594788, + "learning_rate": 2.997024845416961e-06, + "loss": 0.5996, + "step": 3216 + }, + { + "epoch": 2.6634456101601325, + "grad_norm": 0.9153965711593628, + "learning_rate": 2.9959470599037525e-06, + "loss": 0.5777, + "step": 3217 + }, + { + "epoch": 2.664273881833241, + "grad_norm": 0.9155426621437073, + "learning_rate": 2.994869178429411e-06, + "loss": 0.5906, + "step": 3218 + }, + { + "epoch": 2.66510215350635, + "grad_norm": 0.8978738784790039, + "learning_rate": 2.993791201202496e-06, + "loss": 0.5707, + "step": 3219 + }, + { + "epoch": 2.665930425179459, + "grad_norm": 0.9147471785545349, + "learning_rate": 2.9927131284315874e-06, + "loss": 0.6003, + "step": 3220 + }, + { + "epoch": 2.6667586968525674, + "grad_norm": 0.9578046202659607, + "learning_rate": 2.991634960325281e-06, + "loss": 0.6068, + "step": 3221 + }, + { + "epoch": 2.6675869685256766, + "grad_norm": 0.9182425141334534, + "learning_rate": 2.9905566970921933e-06, + "loss": 0.5939, + "step": 3222 + }, + { + "epoch": 2.6684152401987853, + "grad_norm": 0.9168176651000977, + "learning_rate": 2.989478338940957e-06, + "loss": 0.5967, + "step": 3223 + }, + { + "epoch": 2.669243511871894, + "grad_norm": 0.913521409034729, + "learning_rate": 2.9883998860802253e-06, + "loss": 0.6081, + "step": 3224 + }, + { + "epoch": 2.6700717835450027, + "grad_norm": 0.9223765134811401, + "learning_rate": 2.9873213387186684e-06, + "loss": 0.5901, + "step": 3225 + }, + { + "epoch": 2.6709000552181115, + "grad_norm": 0.9228116273880005, + "learning_rate": 2.9862426970649743e-06, + "loss": 0.5959, + "step": 3226 + }, + { + "epoch": 2.67172832689122, + "grad_norm": 0.9379467368125916, + "learning_rate": 2.985163961327852e-06, + "loss": 0.5986, + "step": 3227 + }, + { + "epoch": 2.672556598564329, + "grad_norm": 0.921449601650238, + "learning_rate": 2.984085131716027e-06, + "loss": 0.5587, + "step": 3228 + }, + { + "epoch": 2.673384870237438, + "grad_norm": 0.9038673043251038, + "learning_rate": 2.9830062084382393e-06, + "loss": 0.5908, + "step": 3229 + }, + { + "epoch": 2.674213141910547, + "grad_norm": 0.9072180986404419, + "learning_rate": 2.981927191703254e-06, + "loss": 0.6064, + "step": 3230 + }, + { + "epoch": 2.6750414135836555, + "grad_norm": 0.9178011417388916, + "learning_rate": 2.9808480817198494e-06, + "loss": 0.5969, + "step": 3231 + }, + { + "epoch": 2.6758696852567643, + "grad_norm": 0.9064451456069946, + "learning_rate": 2.9797688786968233e-06, + "loss": 0.571, + "step": 3232 + }, + { + "epoch": 2.676697956929873, + "grad_norm": 0.9392306804656982, + "learning_rate": 2.9786895828429914e-06, + "loss": 0.5903, + "step": 3233 + }, + { + "epoch": 2.6775262286029817, + "grad_norm": 0.9398954510688782, + "learning_rate": 2.9776101943671875e-06, + "loss": 0.6163, + "step": 3234 + }, + { + "epoch": 2.6783545002760905, + "grad_norm": 0.9082908630371094, + "learning_rate": 2.9765307134782637e-06, + "loss": 0.5838, + "step": 3235 + }, + { + "epoch": 2.679182771949199, + "grad_norm": 0.926916778087616, + "learning_rate": 2.975451140385089e-06, + "loss": 0.5599, + "step": 3236 + }, + { + "epoch": 2.680011043622308, + "grad_norm": 0.9106405377388, + "learning_rate": 2.97437147529655e-06, + "loss": 0.6012, + "step": 3237 + }, + { + "epoch": 2.680839315295417, + "grad_norm": 0.9428172707557678, + "learning_rate": 2.9732917184215524e-06, + "loss": 0.5985, + "step": 3238 + }, + { + "epoch": 2.681667586968526, + "grad_norm": 0.9456861019134521, + "learning_rate": 2.9722118699690212e-06, + "loss": 0.6206, + "step": 3239 + }, + { + "epoch": 2.6824958586416345, + "grad_norm": 0.9116294384002686, + "learning_rate": 2.971131930147893e-06, + "loss": 0.6166, + "step": 3240 + }, + { + "epoch": 2.6833241303147433, + "grad_norm": 0.9061588644981384, + "learning_rate": 2.970051899167129e-06, + "loss": 0.5792, + "step": 3241 + }, + { + "epoch": 2.684152401987852, + "grad_norm": 0.947784423828125, + "learning_rate": 2.968971777235704e-06, + "loss": 0.5813, + "step": 3242 + }, + { + "epoch": 2.6849806736609607, + "grad_norm": 0.9296494722366333, + "learning_rate": 2.9678915645626115e-06, + "loss": 0.5916, + "step": 3243 + }, + { + "epoch": 2.6858089453340694, + "grad_norm": 0.9151764512062073, + "learning_rate": 2.9668112613568632e-06, + "loss": 0.6097, + "step": 3244 + }, + { + "epoch": 2.6866372170071786, + "grad_norm": 0.9887247681617737, + "learning_rate": 2.965730867827488e-06, + "loss": 0.5951, + "step": 3245 + }, + { + "epoch": 2.687465488680287, + "grad_norm": 0.9406042695045471, + "learning_rate": 2.964650384183531e-06, + "loss": 0.5918, + "step": 3246 + }, + { + "epoch": 2.688293760353396, + "grad_norm": 0.9440297484397888, + "learning_rate": 2.963569810634056e-06, + "loss": 0.6098, + "step": 3247 + }, + { + "epoch": 2.689122032026505, + "grad_norm": 0.93993079662323, + "learning_rate": 2.962489147388144e-06, + "loss": 0.6242, + "step": 3248 + }, + { + "epoch": 2.6899503036996135, + "grad_norm": 0.9704070091247559, + "learning_rate": 2.9614083946548927e-06, + "loss": 0.5764, + "step": 3249 + }, + { + "epoch": 2.6907785753727222, + "grad_norm": 0.9809379577636719, + "learning_rate": 2.9603275526434188e-06, + "loss": 0.5997, + "step": 3250 + }, + { + "epoch": 2.691606847045831, + "grad_norm": 0.9246951937675476, + "learning_rate": 2.9592466215628547e-06, + "loss": 0.577, + "step": 3251 + }, + { + "epoch": 2.6924351187189397, + "grad_norm": 0.9204201698303223, + "learning_rate": 2.95816560162235e-06, + "loss": 0.6036, + "step": 3252 + }, + { + "epoch": 2.6932633903920484, + "grad_norm": 0.9180527925491333, + "learning_rate": 2.9570844930310734e-06, + "loss": 0.5919, + "step": 3253 + }, + { + "epoch": 2.6940916620651576, + "grad_norm": 0.9165668487548828, + "learning_rate": 2.9560032959982067e-06, + "loss": 0.5646, + "step": 3254 + }, + { + "epoch": 2.694919933738266, + "grad_norm": 0.9150571823120117, + "learning_rate": 2.9549220107329536e-06, + "loss": 0.581, + "step": 3255 + }, + { + "epoch": 2.695748205411375, + "grad_norm": 0.9429931044578552, + "learning_rate": 2.9538406374445334e-06, + "loss": 0.595, + "step": 3256 + }, + { + "epoch": 2.696576477084484, + "grad_norm": 0.939118504524231, + "learning_rate": 2.9527591763421785e-06, + "loss": 0.5772, + "step": 3257 + }, + { + "epoch": 2.6974047487575925, + "grad_norm": 0.9130142331123352, + "learning_rate": 2.9516776276351446e-06, + "loss": 0.6031, + "step": 3258 + }, + { + "epoch": 2.6982330204307012, + "grad_norm": 0.9334836006164551, + "learning_rate": 2.9505959915327004e-06, + "loss": 0.5899, + "step": 3259 + }, + { + "epoch": 2.69906129210381, + "grad_norm": 0.9407964944839478, + "learning_rate": 2.9495142682441314e-06, + "loss": 0.6232, + "step": 3260 + }, + { + "epoch": 2.6998895637769187, + "grad_norm": 0.9573985934257507, + "learning_rate": 2.9484324579787427e-06, + "loss": 0.5932, + "step": 3261 + }, + { + "epoch": 2.7007178354500274, + "grad_norm": 0.9223983287811279, + "learning_rate": 2.947350560945853e-06, + "loss": 0.5861, + "step": 3262 + }, + { + "epoch": 2.7015461071231366, + "grad_norm": 0.9206790328025818, + "learning_rate": 2.9462685773548004e-06, + "loss": 0.5867, + "step": 3263 + }, + { + "epoch": 2.7023743787962453, + "grad_norm": 0.9197454452514648, + "learning_rate": 2.9451865074149378e-06, + "loss": 0.5867, + "step": 3264 + }, + { + "epoch": 2.703202650469354, + "grad_norm": 0.9365606904029846, + "learning_rate": 2.944104351335636e-06, + "loss": 0.6106, + "step": 3265 + }, + { + "epoch": 2.7040309221424628, + "grad_norm": 0.9116243720054626, + "learning_rate": 2.9430221093262823e-06, + "loss": 0.5904, + "step": 3266 + }, + { + "epoch": 2.7048591938155715, + "grad_norm": 0.9128034114837646, + "learning_rate": 2.9419397815962802e-06, + "loss": 0.6055, + "step": 3267 + }, + { + "epoch": 2.70568746548868, + "grad_norm": 0.9147705435752869, + "learning_rate": 2.94085736835505e-06, + "loss": 0.5941, + "step": 3268 + }, + { + "epoch": 2.706515737161789, + "grad_norm": 0.9419110417366028, + "learning_rate": 2.939774869812029e-06, + "loss": 0.5595, + "step": 3269 + }, + { + "epoch": 2.707344008834898, + "grad_norm": 0.9202789664268494, + "learning_rate": 2.93869228617667e-06, + "loss": 0.5788, + "step": 3270 + }, + { + "epoch": 2.7081722805080064, + "grad_norm": 0.9068663716316223, + "learning_rate": 2.9376096176584424e-06, + "loss": 0.5749, + "step": 3271 + }, + { + "epoch": 2.7090005521811156, + "grad_norm": 0.923497200012207, + "learning_rate": 2.936526864466833e-06, + "loss": 0.6029, + "step": 3272 + }, + { + "epoch": 2.7098288238542243, + "grad_norm": 0.9812871217727661, + "learning_rate": 2.935444026811346e-06, + "loss": 0.6193, + "step": 3273 + }, + { + "epoch": 2.710657095527333, + "grad_norm": 0.9538823366165161, + "learning_rate": 2.934361104901498e-06, + "loss": 0.5809, + "step": 3274 + }, + { + "epoch": 2.7114853672004418, + "grad_norm": 0.9111268520355225, + "learning_rate": 2.9332780989468256e-06, + "loss": 0.6181, + "step": 3275 + }, + { + "epoch": 2.7123136388735505, + "grad_norm": 0.9485889673233032, + "learning_rate": 2.9321950091568795e-06, + "loss": 0.5823, + "step": 3276 + }, + { + "epoch": 2.713141910546659, + "grad_norm": 0.9233952760696411, + "learning_rate": 2.931111835741229e-06, + "loss": 0.6008, + "step": 3277 + }, + { + "epoch": 2.713970182219768, + "grad_norm": 0.9075472950935364, + "learning_rate": 2.9300285789094563e-06, + "loss": 0.5808, + "step": 3278 + }, + { + "epoch": 2.714798453892877, + "grad_norm": 0.9238584041595459, + "learning_rate": 2.9289452388711624e-06, + "loss": 0.6095, + "step": 3279 + }, + { + "epoch": 2.7156267255659854, + "grad_norm": 0.889460027217865, + "learning_rate": 2.9278618158359635e-06, + "loss": 0.6015, + "step": 3280 + }, + { + "epoch": 2.7164549972390946, + "grad_norm": 0.9549591541290283, + "learning_rate": 2.9267783100134915e-06, + "loss": 0.6029, + "step": 3281 + }, + { + "epoch": 2.7172832689122033, + "grad_norm": 0.9522138237953186, + "learning_rate": 2.9256947216133946e-06, + "loss": 0.5998, + "step": 3282 + }, + { + "epoch": 2.718111540585312, + "grad_norm": 0.8992515802383423, + "learning_rate": 2.924611050845338e-06, + "loss": 0.6012, + "step": 3283 + }, + { + "epoch": 2.7189398122584207, + "grad_norm": 0.9058646559715271, + "learning_rate": 2.9235272979190014e-06, + "loss": 0.6092, + "step": 3284 + }, + { + "epoch": 2.7197680839315295, + "grad_norm": 0.916130542755127, + "learning_rate": 2.92244346304408e-06, + "loss": 0.6054, + "step": 3285 + }, + { + "epoch": 2.720596355604638, + "grad_norm": 0.9203442335128784, + "learning_rate": 2.921359546430287e-06, + "loss": 0.596, + "step": 3286 + }, + { + "epoch": 2.721424627277747, + "grad_norm": 0.9603051543235779, + "learning_rate": 2.920275548287349e-06, + "loss": 0.5976, + "step": 3287 + }, + { + "epoch": 2.722252898950856, + "grad_norm": 0.9188669919967651, + "learning_rate": 2.91919146882501e-06, + "loss": 0.5812, + "step": 3288 + }, + { + "epoch": 2.723081170623965, + "grad_norm": 0.9154419898986816, + "learning_rate": 2.9181073082530302e-06, + "loss": 0.6127, + "step": 3289 + }, + { + "epoch": 2.7239094422970735, + "grad_norm": 0.9228849411010742, + "learning_rate": 2.917023066781184e-06, + "loss": 0.6223, + "step": 3290 + }, + { + "epoch": 2.7247377139701823, + "grad_norm": 0.9034006595611572, + "learning_rate": 2.915938744619261e-06, + "loss": 0.5981, + "step": 3291 + }, + { + "epoch": 2.725565985643291, + "grad_norm": 0.9119758009910583, + "learning_rate": 2.914854341977069e-06, + "loss": 0.5974, + "step": 3292 + }, + { + "epoch": 2.7263942573163997, + "grad_norm": 0.9350270628929138, + "learning_rate": 2.9137698590644283e-06, + "loss": 0.5958, + "step": 3293 + }, + { + "epoch": 2.7272225289895085, + "grad_norm": 0.9229425191879272, + "learning_rate": 2.9126852960911773e-06, + "loss": 0.5873, + "step": 3294 + }, + { + "epoch": 2.728050800662617, + "grad_norm": 0.9276396632194519, + "learning_rate": 2.9116006532671685e-06, + "loss": 0.6069, + "step": 3295 + }, + { + "epoch": 2.728879072335726, + "grad_norm": 0.9154210090637207, + "learning_rate": 2.9105159308022705e-06, + "loss": 0.5741, + "step": 3296 + }, + { + "epoch": 2.729707344008835, + "grad_norm": 0.9545791149139404, + "learning_rate": 2.909431128906367e-06, + "loss": 0.5995, + "step": 3297 + }, + { + "epoch": 2.730535615681944, + "grad_norm": 0.9200988411903381, + "learning_rate": 2.908346247789356e-06, + "loss": 0.6152, + "step": 3298 + }, + { + "epoch": 2.7313638873550525, + "grad_norm": 0.9288673996925354, + "learning_rate": 2.9072612876611524e-06, + "loss": 0.5904, + "step": 3299 + }, + { + "epoch": 2.7321921590281613, + "grad_norm": 0.918323278427124, + "learning_rate": 2.906176248731686e-06, + "loss": 0.5999, + "step": 3300 + }, + { + "epoch": 2.73302043070127, + "grad_norm": 0.936652421951294, + "learning_rate": 2.905091131210903e-06, + "loss": 0.5928, + "step": 3301 + }, + { + "epoch": 2.7338487023743787, + "grad_norm": 0.8942639827728271, + "learning_rate": 2.9040059353087614e-06, + "loss": 0.5739, + "step": 3302 + }, + { + "epoch": 2.7346769740474874, + "grad_norm": 0.9197317957878113, + "learning_rate": 2.902920661235238e-06, + "loss": 0.5776, + "step": 3303 + }, + { + "epoch": 2.7355052457205966, + "grad_norm": 0.895505428314209, + "learning_rate": 2.9018353092003216e-06, + "loss": 0.6067, + "step": 3304 + }, + { + "epoch": 2.736333517393705, + "grad_norm": 0.947441816329956, + "learning_rate": 2.9007498794140193e-06, + "loss": 0.6072, + "step": 3305 + }, + { + "epoch": 2.737161789066814, + "grad_norm": 0.9008322358131409, + "learning_rate": 2.8996643720863515e-06, + "loss": 0.5908, + "step": 3306 + }, + { + "epoch": 2.737990060739923, + "grad_norm": 0.9137061238288879, + "learning_rate": 2.898578787427353e-06, + "loss": 0.607, + "step": 3307 + }, + { + "epoch": 2.7388183324130315, + "grad_norm": 0.9126008749008179, + "learning_rate": 2.8974931256470745e-06, + "loss": 0.5809, + "step": 3308 + }, + { + "epoch": 2.7396466040861402, + "grad_norm": 0.9215322136878967, + "learning_rate": 2.896407386955582e-06, + "loss": 0.5807, + "step": 3309 + }, + { + "epoch": 2.740474875759249, + "grad_norm": 0.8903998136520386, + "learning_rate": 2.8953215715629547e-06, + "loss": 0.5968, + "step": 3310 + }, + { + "epoch": 2.7413031474323577, + "grad_norm": 0.9156417846679688, + "learning_rate": 2.8942356796792888e-06, + "loss": 0.5809, + "step": 3311 + }, + { + "epoch": 2.7421314191054664, + "grad_norm": 0.952264666557312, + "learning_rate": 2.893149711514694e-06, + "loss": 0.5818, + "step": 3312 + }, + { + "epoch": 2.7429596907785756, + "grad_norm": 0.9201667904853821, + "learning_rate": 2.8920636672792945e-06, + "loss": 0.5789, + "step": 3313 + }, + { + "epoch": 2.743787962451684, + "grad_norm": 0.9581197500228882, + "learning_rate": 2.890977547183231e-06, + "loss": 0.6007, + "step": 3314 + }, + { + "epoch": 2.744616234124793, + "grad_norm": 0.9360357522964478, + "learning_rate": 2.889891351436656e-06, + "loss": 0.6096, + "step": 3315 + }, + { + "epoch": 2.745444505797902, + "grad_norm": 0.9209451079368591, + "learning_rate": 2.8888050802497388e-06, + "loss": 0.6015, + "step": 3316 + }, + { + "epoch": 2.7462727774710105, + "grad_norm": 0.9496802091598511, + "learning_rate": 2.8877187338326634e-06, + "loss": 0.5802, + "step": 3317 + }, + { + "epoch": 2.7471010491441192, + "grad_norm": 0.9407662153244019, + "learning_rate": 2.8866323123956274e-06, + "loss": 0.588, + "step": 3318 + }, + { + "epoch": 2.747929320817228, + "grad_norm": 0.9603942036628723, + "learning_rate": 2.885545816148843e-06, + "loss": 0.5759, + "step": 3319 + }, + { + "epoch": 2.7487575924903367, + "grad_norm": 0.930496335029602, + "learning_rate": 2.8844592453025372e-06, + "loss": 0.5916, + "step": 3320 + }, + { + "epoch": 2.7495858641634454, + "grad_norm": 0.8831289410591125, + "learning_rate": 2.883372600066951e-06, + "loss": 0.5889, + "step": 3321 + }, + { + "epoch": 2.7504141358365546, + "grad_norm": 0.9076021909713745, + "learning_rate": 2.8822858806523407e-06, + "loss": 0.5842, + "step": 3322 + }, + { + "epoch": 2.7512424075096633, + "grad_norm": 0.9093723297119141, + "learning_rate": 2.8811990872689764e-06, + "loss": 0.584, + "step": 3323 + }, + { + "epoch": 2.752070679182772, + "grad_norm": 0.9458228349685669, + "learning_rate": 2.880112220127142e-06, + "loss": 0.5812, + "step": 3324 + }, + { + "epoch": 2.7528989508558808, + "grad_norm": 0.9406691789627075, + "learning_rate": 2.8790252794371365e-06, + "loss": 0.5889, + "step": 3325 + }, + { + "epoch": 2.7537272225289895, + "grad_norm": 0.9094856381416321, + "learning_rate": 2.8779382654092726e-06, + "loss": 0.5919, + "step": 3326 + }, + { + "epoch": 2.754555494202098, + "grad_norm": 0.9350448250770569, + "learning_rate": 2.8768511782538766e-06, + "loss": 0.5738, + "step": 3327 + }, + { + "epoch": 2.755383765875207, + "grad_norm": 0.9427900314331055, + "learning_rate": 2.875764018181291e-06, + "loss": 0.5962, + "step": 3328 + }, + { + "epoch": 2.756212037548316, + "grad_norm": 0.9439454674720764, + "learning_rate": 2.8746767854018715e-06, + "loss": 0.5994, + "step": 3329 + }, + { + "epoch": 2.7570403092214244, + "grad_norm": 0.9291850328445435, + "learning_rate": 2.8735894801259863e-06, + "loss": 0.6055, + "step": 3330 + }, + { + "epoch": 2.7578685808945336, + "grad_norm": 0.9287533164024353, + "learning_rate": 2.872502102564019e-06, + "loss": 0.5798, + "step": 3331 + }, + { + "epoch": 2.7586968525676423, + "grad_norm": 0.8941923379898071, + "learning_rate": 2.871414652926367e-06, + "loss": 0.6066, + "step": 3332 + }, + { + "epoch": 2.759525124240751, + "grad_norm": 0.924333930015564, + "learning_rate": 2.8703271314234425e-06, + "loss": 0.5872, + "step": 3333 + }, + { + "epoch": 2.7603533959138598, + "grad_norm": 0.8994147777557373, + "learning_rate": 2.8692395382656702e-06, + "loss": 0.5906, + "step": 3334 + }, + { + "epoch": 2.7611816675869685, + "grad_norm": 0.9152485132217407, + "learning_rate": 2.8681518736634895e-06, + "loss": 0.5774, + "step": 3335 + }, + { + "epoch": 2.762009939260077, + "grad_norm": 0.9269137978553772, + "learning_rate": 2.867064137827353e-06, + "loss": 0.5759, + "step": 3336 + }, + { + "epoch": 2.762838210933186, + "grad_norm": 0.916441023349762, + "learning_rate": 2.8659763309677267e-06, + "loss": 0.5927, + "step": 3337 + }, + { + "epoch": 2.763666482606295, + "grad_norm": 0.9280357956886292, + "learning_rate": 2.8648884532950925e-06, + "loss": 0.6004, + "step": 3338 + }, + { + "epoch": 2.7644947542794034, + "grad_norm": 0.9075552821159363, + "learning_rate": 2.8638005050199443e-06, + "loss": 0.6185, + "step": 3339 + }, + { + "epoch": 2.7653230259525126, + "grad_norm": 0.9340760707855225, + "learning_rate": 2.8627124863527895e-06, + "loss": 0.6119, + "step": 3340 + }, + { + "epoch": 2.7661512976256213, + "grad_norm": 0.9393087029457092, + "learning_rate": 2.8616243975041495e-06, + "loss": 0.6058, + "step": 3341 + }, + { + "epoch": 2.76697956929873, + "grad_norm": 0.9380460381507874, + "learning_rate": 2.8605362386845605e-06, + "loss": 0.5955, + "step": 3342 + }, + { + "epoch": 2.7678078409718387, + "grad_norm": 0.9393378496170044, + "learning_rate": 2.8594480101045683e-06, + "loss": 0.5841, + "step": 3343 + }, + { + "epoch": 2.7686361126449475, + "grad_norm": 0.9186044335365295, + "learning_rate": 2.8583597119747385e-06, + "loss": 0.5977, + "step": 3344 + }, + { + "epoch": 2.769464384318056, + "grad_norm": 0.9224127531051636, + "learning_rate": 2.8572713445056448e-06, + "loss": 0.5863, + "step": 3345 + }, + { + "epoch": 2.770292655991165, + "grad_norm": 0.9517519474029541, + "learning_rate": 2.8561829079078764e-06, + "loss": 0.5748, + "step": 3346 + }, + { + "epoch": 2.771120927664274, + "grad_norm": 0.9624918103218079, + "learning_rate": 2.855094402392036e-06, + "loss": 0.588, + "step": 3347 + }, + { + "epoch": 2.771949199337383, + "grad_norm": 0.9524730443954468, + "learning_rate": 2.8540058281687375e-06, + "loss": 0.6032, + "step": 3348 + }, + { + "epoch": 2.7727774710104915, + "grad_norm": 0.9574846029281616, + "learning_rate": 2.852917185448612e-06, + "loss": 0.5915, + "step": 3349 + }, + { + "epoch": 2.7736057426836003, + "grad_norm": 0.9290011525154114, + "learning_rate": 2.851828474442301e-06, + "loss": 0.5715, + "step": 3350 + }, + { + "epoch": 2.774434014356709, + "grad_norm": 0.9393135905265808, + "learning_rate": 2.8507396953604597e-06, + "loss": 0.5855, + "step": 3351 + }, + { + "epoch": 2.7752622860298177, + "grad_norm": 0.9188078045845032, + "learning_rate": 2.8496508484137574e-06, + "loss": 0.598, + "step": 3352 + }, + { + "epoch": 2.7760905577029265, + "grad_norm": 0.9129231572151184, + "learning_rate": 2.8485619338128755e-06, + "loss": 0.6076, + "step": 3353 + }, + { + "epoch": 2.776918829376035, + "grad_norm": 0.941085934638977, + "learning_rate": 2.847472951768508e-06, + "loss": 0.5787, + "step": 3354 + }, + { + "epoch": 2.777747101049144, + "grad_norm": 1.0103200674057007, + "learning_rate": 2.846383902491363e-06, + "loss": 0.606, + "step": 3355 + }, + { + "epoch": 2.778575372722253, + "grad_norm": 0.9307713508605957, + "learning_rate": 2.845294786192163e-06, + "loss": 0.5969, + "step": 3356 + }, + { + "epoch": 2.779403644395362, + "grad_norm": 0.890837550163269, + "learning_rate": 2.84420560308164e-06, + "loss": 0.5826, + "step": 3357 + }, + { + "epoch": 2.7802319160684705, + "grad_norm": 0.8857460618019104, + "learning_rate": 2.843116353370542e-06, + "loss": 0.6005, + "step": 3358 + }, + { + "epoch": 2.7810601877415793, + "grad_norm": 0.8894275426864624, + "learning_rate": 2.8420270372696273e-06, + "loss": 0.6017, + "step": 3359 + }, + { + "epoch": 2.781888459414688, + "grad_norm": 0.9014497995376587, + "learning_rate": 2.8409376549896693e-06, + "loss": 0.5845, + "step": 3360 + }, + { + "epoch": 2.7827167310877967, + "grad_norm": 0.9019520282745361, + "learning_rate": 2.839848206741454e-06, + "loss": 0.5993, + "step": 3361 + }, + { + "epoch": 2.7835450027609054, + "grad_norm": 0.9594439268112183, + "learning_rate": 2.838758692735778e-06, + "loss": 0.5905, + "step": 3362 + }, + { + "epoch": 2.7843732744340146, + "grad_norm": 0.9144065976142883, + "learning_rate": 2.8376691131834525e-06, + "loss": 0.6006, + "step": 3363 + }, + { + "epoch": 2.785201546107123, + "grad_norm": 0.9487935900688171, + "learning_rate": 2.836579468295303e-06, + "loss": 0.5832, + "step": 3364 + }, + { + "epoch": 2.786029817780232, + "grad_norm": 0.8950932025909424, + "learning_rate": 2.835489758282163e-06, + "loss": 0.5792, + "step": 3365 + }, + { + "epoch": 2.786858089453341, + "grad_norm": 0.9522806406021118, + "learning_rate": 2.8343999833548814e-06, + "loss": 0.5891, + "step": 3366 + }, + { + "epoch": 2.7876863611264495, + "grad_norm": 0.964898407459259, + "learning_rate": 2.833310143724321e-06, + "loss": 0.572, + "step": 3367 + }, + { + "epoch": 2.7885146327995582, + "grad_norm": 0.9154083728790283, + "learning_rate": 2.832220239601355e-06, + "loss": 0.5982, + "step": 3368 + }, + { + "epoch": 2.789342904472667, + "grad_norm": 0.9093793034553528, + "learning_rate": 2.831130271196869e-06, + "loss": 0.5592, + "step": 3369 + }, + { + "epoch": 2.7901711761457757, + "grad_norm": 0.9114746451377869, + "learning_rate": 2.8300402387217628e-06, + "loss": 0.5531, + "step": 3370 + }, + { + "epoch": 2.7909994478188844, + "grad_norm": 0.959976077079773, + "learning_rate": 2.8289501423869464e-06, + "loss": 0.5932, + "step": 3371 + }, + { + "epoch": 2.7918277194919936, + "grad_norm": 0.9415104389190674, + "learning_rate": 2.8278599824033443e-06, + "loss": 0.5516, + "step": 3372 + }, + { + "epoch": 2.792655991165102, + "grad_norm": 0.9304796457290649, + "learning_rate": 2.826769758981892e-06, + "loss": 0.5786, + "step": 3373 + }, + { + "epoch": 2.793484262838211, + "grad_norm": 0.9369680285453796, + "learning_rate": 2.825679472333537e-06, + "loss": 0.6077, + "step": 3374 + }, + { + "epoch": 2.79431253451132, + "grad_norm": 0.9102757573127747, + "learning_rate": 2.8245891226692405e-06, + "loss": 0.606, + "step": 3375 + }, + { + "epoch": 2.7951408061844285, + "grad_norm": 0.921359121799469, + "learning_rate": 2.8234987101999747e-06, + "loss": 0.613, + "step": 3376 + }, + { + "epoch": 2.7959690778575372, + "grad_norm": 0.9242355227470398, + "learning_rate": 2.822408235136723e-06, + "loss": 0.5758, + "step": 3377 + }, + { + "epoch": 2.796797349530646, + "grad_norm": 0.9175317287445068, + "learning_rate": 2.8213176976904844e-06, + "loss": 0.5744, + "step": 3378 + }, + { + "epoch": 2.7976256212037547, + "grad_norm": 0.9504632949829102, + "learning_rate": 2.8202270980722663e-06, + "loss": 0.5871, + "step": 3379 + }, + { + "epoch": 2.7984538928768634, + "grad_norm": 0.9071774482727051, + "learning_rate": 2.8191364364930895e-06, + "loss": 0.6003, + "step": 3380 + }, + { + "epoch": 2.7992821645499726, + "grad_norm": 0.913815975189209, + "learning_rate": 2.8180457131639878e-06, + "loss": 0.596, + "step": 3381 + }, + { + "epoch": 2.8001104362230813, + "grad_norm": 0.9608022570610046, + "learning_rate": 2.8169549282960044e-06, + "loss": 0.6055, + "step": 3382 + }, + { + "epoch": 2.80093870789619, + "grad_norm": 2.0446290969848633, + "learning_rate": 2.815864082100198e-06, + "loss": 0.565, + "step": 3383 + }, + { + "epoch": 2.8017669795692988, + "grad_norm": 0.9219574928283691, + "learning_rate": 2.814773174787636e-06, + "loss": 0.5808, + "step": 3384 + }, + { + "epoch": 2.8025952512424075, + "grad_norm": 0.9542830586433411, + "learning_rate": 2.8136822065693987e-06, + "loss": 0.5992, + "step": 3385 + }, + { + "epoch": 2.803423522915516, + "grad_norm": 0.9478740096092224, + "learning_rate": 2.8125911776565785e-06, + "loss": 0.5705, + "step": 3386 + }, + { + "epoch": 2.804251794588625, + "grad_norm": 0.9299196600914001, + "learning_rate": 2.8115000882602792e-06, + "loss": 0.5773, + "step": 3387 + }, + { + "epoch": 2.805080066261734, + "grad_norm": 0.9167691469192505, + "learning_rate": 2.8104089385916168e-06, + "loss": 0.5789, + "step": 3388 + }, + { + "epoch": 2.8059083379348424, + "grad_norm": 0.9034630656242371, + "learning_rate": 2.8093177288617187e-06, + "loss": 0.5837, + "step": 3389 + }, + { + "epoch": 2.8067366096079516, + "grad_norm": 0.9147744178771973, + "learning_rate": 2.808226459281723e-06, + "loss": 0.5826, + "step": 3390 + }, + { + "epoch": 2.8075648812810603, + "grad_norm": 0.9159218072891235, + "learning_rate": 2.8071351300627803e-06, + "loss": 0.5931, + "step": 3391 + }, + { + "epoch": 2.808393152954169, + "grad_norm": 0.9260812401771545, + "learning_rate": 2.8060437414160546e-06, + "loss": 0.6182, + "step": 3392 + }, + { + "epoch": 2.8092214246272778, + "grad_norm": 0.8975473642349243, + "learning_rate": 2.804952293552717e-06, + "loss": 0.5786, + "step": 3393 + }, + { + "epoch": 2.8100496963003865, + "grad_norm": 0.9130522608757019, + "learning_rate": 2.8038607866839522e-06, + "loss": 0.572, + "step": 3394 + }, + { + "epoch": 2.810877967973495, + "grad_norm": 0.9170857667922974, + "learning_rate": 2.802769221020959e-06, + "loss": 0.569, + "step": 3395 + }, + { + "epoch": 2.811706239646604, + "grad_norm": 0.9313328266143799, + "learning_rate": 2.801677596774944e-06, + "loss": 0.568, + "step": 3396 + }, + { + "epoch": 2.812534511319713, + "grad_norm": 0.941814124584198, + "learning_rate": 2.8005859141571257e-06, + "loss": 0.5852, + "step": 3397 + }, + { + "epoch": 2.8133627829928214, + "grad_norm": 0.9179267883300781, + "learning_rate": 2.799494173378735e-06, + "loss": 0.5887, + "step": 3398 + }, + { + "epoch": 2.8141910546659306, + "grad_norm": 0.9193063378334045, + "learning_rate": 2.7984023746510135e-06, + "loss": 0.588, + "step": 3399 + }, + { + "epoch": 2.8150193263390393, + "grad_norm": 0.9381787776947021, + "learning_rate": 2.7973105181852145e-06, + "loss": 0.5938, + "step": 3400 + }, + { + "epoch": 2.815847598012148, + "grad_norm": 0.9631656408309937, + "learning_rate": 2.7962186041926015e-06, + "loss": 0.5828, + "step": 3401 + }, + { + "epoch": 2.8166758696852567, + "grad_norm": 0.9512215852737427, + "learning_rate": 2.7951266328844495e-06, + "loss": 0.5836, + "step": 3402 + }, + { + "epoch": 2.8175041413583655, + "grad_norm": 0.9171256422996521, + "learning_rate": 2.794034604472047e-06, + "loss": 0.574, + "step": 3403 + }, + { + "epoch": 2.818332413031474, + "grad_norm": 0.9122748374938965, + "learning_rate": 2.792942519166688e-06, + "loss": 0.6024, + "step": 3404 + }, + { + "epoch": 2.819160684704583, + "grad_norm": 0.9503622651100159, + "learning_rate": 2.7918503771796817e-06, + "loss": 0.5743, + "step": 3405 + }, + { + "epoch": 2.819988956377692, + "grad_norm": 0.9348991513252258, + "learning_rate": 2.7907581787223494e-06, + "loss": 0.5824, + "step": 3406 + }, + { + "epoch": 2.820817228050801, + "grad_norm": 1.0846208333969116, + "learning_rate": 2.7896659240060192e-06, + "loss": 0.5995, + "step": 3407 + }, + { + "epoch": 2.8216454997239095, + "grad_norm": 0.9554671049118042, + "learning_rate": 2.7885736132420326e-06, + "loss": 0.5903, + "step": 3408 + }, + { + "epoch": 2.8224737713970183, + "grad_norm": 0.9836533665657043, + "learning_rate": 2.7874812466417434e-06, + "loss": 0.5982, + "step": 3409 + }, + { + "epoch": 2.823302043070127, + "grad_norm": 0.9414785504341125, + "learning_rate": 2.786388824416511e-06, + "loss": 0.5743, + "step": 3410 + }, + { + "epoch": 2.8241303147432357, + "grad_norm": 0.9228699207305908, + "learning_rate": 2.7852963467777117e-06, + "loss": 0.5879, + "step": 3411 + }, + { + "epoch": 2.8249585864163445, + "grad_norm": 0.919826090335846, + "learning_rate": 2.7842038139367288e-06, + "loss": 0.6023, + "step": 3412 + }, + { + "epoch": 2.825786858089453, + "grad_norm": 0.9112126231193542, + "learning_rate": 2.783111226104957e-06, + "loss": 0.578, + "step": 3413 + }, + { + "epoch": 2.826615129762562, + "grad_norm": 0.9262592792510986, + "learning_rate": 2.7820185834938024e-06, + "loss": 0.6063, + "step": 3414 + }, + { + "epoch": 2.827443401435671, + "grad_norm": 0.9020029902458191, + "learning_rate": 2.780925886314681e-06, + "loss": 0.5837, + "step": 3415 + }, + { + "epoch": 2.82827167310878, + "grad_norm": 0.9186374545097351, + "learning_rate": 2.7798331347790185e-06, + "loss": 0.5801, + "step": 3416 + }, + { + "epoch": 2.8290999447818885, + "grad_norm": 0.9116926193237305, + "learning_rate": 2.7787403290982535e-06, + "loss": 0.5735, + "step": 3417 + }, + { + "epoch": 2.8299282164549973, + "grad_norm": 0.9124463796615601, + "learning_rate": 2.7776474694838333e-06, + "loss": 0.6064, + "step": 3418 + }, + { + "epoch": 2.830756488128106, + "grad_norm": 0.9450061321258545, + "learning_rate": 2.7765545561472153e-06, + "loss": 0.5885, + "step": 3419 + }, + { + "epoch": 2.8315847598012147, + "grad_norm": 0.8993310928344727, + "learning_rate": 2.77546158929987e-06, + "loss": 0.5767, + "step": 3420 + }, + { + "epoch": 2.8324130314743234, + "grad_norm": 0.9315094947814941, + "learning_rate": 2.7743685691532734e-06, + "loss": 0.595, + "step": 3421 + }, + { + "epoch": 2.8332413031474326, + "grad_norm": 0.9254942536354065, + "learning_rate": 2.773275495918917e-06, + "loss": 0.6071, + "step": 3422 + }, + { + "epoch": 2.834069574820541, + "grad_norm": 0.9551193118095398, + "learning_rate": 2.7721823698083e-06, + "loss": 0.588, + "step": 3423 + }, + { + "epoch": 2.83489784649365, + "grad_norm": 0.9356088042259216, + "learning_rate": 2.7710891910329317e-06, + "loss": 0.5917, + "step": 3424 + }, + { + "epoch": 2.835726118166759, + "grad_norm": 0.9254491925239563, + "learning_rate": 2.7699959598043315e-06, + "loss": 0.5754, + "step": 3425 + }, + { + "epoch": 2.8365543898398675, + "grad_norm": 0.928747832775116, + "learning_rate": 2.7689026763340302e-06, + "loss": 0.5925, + "step": 3426 + }, + { + "epoch": 2.8373826615129762, + "grad_norm": 0.9232720732688904, + "learning_rate": 2.7678093408335677e-06, + "loss": 0.594, + "step": 3427 + }, + { + "epoch": 2.838210933186085, + "grad_norm": 0.9319408535957336, + "learning_rate": 2.766715953514494e-06, + "loss": 0.607, + "step": 3428 + }, + { + "epoch": 2.8390392048591937, + "grad_norm": 0.9321261644363403, + "learning_rate": 2.76562251458837e-06, + "loss": 0.5839, + "step": 3429 + }, + { + "epoch": 2.8398674765323024, + "grad_norm": 0.8851343393325806, + "learning_rate": 2.7645290242667648e-06, + "loss": 0.5929, + "step": 3430 + }, + { + "epoch": 2.8406957482054116, + "grad_norm": 0.9318198561668396, + "learning_rate": 2.76343548276126e-06, + "loss": 0.5758, + "step": 3431 + }, + { + "epoch": 2.84152401987852, + "grad_norm": 0.9212138056755066, + "learning_rate": 2.7623418902834455e-06, + "loss": 0.5689, + "step": 3432 + }, + { + "epoch": 2.842352291551629, + "grad_norm": 0.8917051553726196, + "learning_rate": 2.761248247044919e-06, + "loss": 0.5915, + "step": 3433 + }, + { + "epoch": 2.843180563224738, + "grad_norm": 0.9266699552536011, + "learning_rate": 2.760154553257294e-06, + "loss": 0.5643, + "step": 3434 + }, + { + "epoch": 2.8440088348978465, + "grad_norm": 0.9837039709091187, + "learning_rate": 2.7590608091321867e-06, + "loss": 0.6135, + "step": 3435 + }, + { + "epoch": 2.8448371065709552, + "grad_norm": 0.9872124195098877, + "learning_rate": 2.7579670148812278e-06, + "loss": 0.5649, + "step": 3436 + }, + { + "epoch": 2.845665378244064, + "grad_norm": 0.9581724405288696, + "learning_rate": 2.756873170716057e-06, + "loss": 0.5833, + "step": 3437 + }, + { + "epoch": 2.8464936499171727, + "grad_norm": 0.9363225698471069, + "learning_rate": 2.7557792768483223e-06, + "loss": 0.5852, + "step": 3438 + }, + { + "epoch": 2.8473219215902814, + "grad_norm": 0.9177983403205872, + "learning_rate": 2.7546853334896813e-06, + "loss": 0.5959, + "step": 3439 + }, + { + "epoch": 2.8481501932633906, + "grad_norm": 0.9081945419311523, + "learning_rate": 2.7535913408518023e-06, + "loss": 0.5862, + "step": 3440 + }, + { + "epoch": 2.8489784649364993, + "grad_norm": 0.928897500038147, + "learning_rate": 2.7524972991463626e-06, + "loss": 0.5978, + "step": 3441 + }, + { + "epoch": 2.849806736609608, + "grad_norm": 0.9136974811553955, + "learning_rate": 2.75140320858505e-06, + "loss": 0.5926, + "step": 3442 + }, + { + "epoch": 2.8506350082827168, + "grad_norm": 0.9227224588394165, + "learning_rate": 2.7503090693795605e-06, + "loss": 0.6073, + "step": 3443 + }, + { + "epoch": 2.8514632799558255, + "grad_norm": 0.9390876889228821, + "learning_rate": 2.749214881741598e-06, + "loss": 0.5757, + "step": 3444 + }, + { + "epoch": 2.852291551628934, + "grad_norm": 0.9460882544517517, + "learning_rate": 2.74812064588288e-06, + "loss": 0.6172, + "step": 3445 + }, + { + "epoch": 2.853119823302043, + "grad_norm": 0.9087659120559692, + "learning_rate": 2.74702636201513e-06, + "loss": 0.5937, + "step": 3446 + }, + { + "epoch": 2.853948094975152, + "grad_norm": 0.8995270133018494, + "learning_rate": 2.745932030350081e-06, + "loss": 0.5798, + "step": 3447 + }, + { + "epoch": 2.8547763666482604, + "grad_norm": 0.9307460188865662, + "learning_rate": 2.744837651099478e-06, + "loss": 0.6009, + "step": 3448 + }, + { + "epoch": 2.8556046383213696, + "grad_norm": 0.9634310007095337, + "learning_rate": 2.7437432244750705e-06, + "loss": 0.5985, + "step": 3449 + }, + { + "epoch": 2.8564329099944783, + "grad_norm": 0.9647858142852783, + "learning_rate": 2.742648750688622e-06, + "loss": 0.5718, + "step": 3450 + }, + { + "epoch": 2.857261181667587, + "grad_norm": 0.9150539040565491, + "learning_rate": 2.7415542299519016e-06, + "loss": 0.5964, + "step": 3451 + }, + { + "epoch": 2.8580894533406958, + "grad_norm": 0.9129363298416138, + "learning_rate": 2.7404596624766892e-06, + "loss": 0.5514, + "step": 3452 + }, + { + "epoch": 2.8589177250138045, + "grad_norm": 0.9430226683616638, + "learning_rate": 2.739365048474774e-06, + "loss": 0.5795, + "step": 3453 + }, + { + "epoch": 2.859745996686913, + "grad_norm": 0.9402962327003479, + "learning_rate": 2.738270388157954e-06, + "loss": 0.5728, + "step": 3454 + }, + { + "epoch": 2.860574268360022, + "grad_norm": 0.9169720411300659, + "learning_rate": 2.7371756817380346e-06, + "loss": 0.6171, + "step": 3455 + }, + { + "epoch": 2.861402540033131, + "grad_norm": 0.9437370896339417, + "learning_rate": 2.7360809294268315e-06, + "loss": 0.5827, + "step": 3456 + }, + { + "epoch": 2.8622308117062394, + "grad_norm": 0.8999599814414978, + "learning_rate": 2.7349861314361693e-06, + "loss": 0.5847, + "step": 3457 + }, + { + "epoch": 2.8630590833793486, + "grad_norm": 0.9228375554084778, + "learning_rate": 2.7338912879778814e-06, + "loss": 0.6069, + "step": 3458 + }, + { + "epoch": 2.8638873550524573, + "grad_norm": 0.9543429017066956, + "learning_rate": 2.732796399263809e-06, + "loss": 0.6128, + "step": 3459 + }, + { + "epoch": 2.864715626725566, + "grad_norm": 0.923698902130127, + "learning_rate": 2.7317014655058038e-06, + "loss": 0.5714, + "step": 3460 + }, + { + "epoch": 2.8655438983986747, + "grad_norm": 0.95343017578125, + "learning_rate": 2.730606486915725e-06, + "loss": 0.5821, + "step": 3461 + }, + { + "epoch": 2.8663721700717835, + "grad_norm": 0.9078912734985352, + "learning_rate": 2.729511463705441e-06, + "loss": 0.5833, + "step": 3462 + }, + { + "epoch": 2.867200441744892, + "grad_norm": 0.9045196771621704, + "learning_rate": 2.7284163960868274e-06, + "loss": 0.5647, + "step": 3463 + }, + { + "epoch": 2.868028713418001, + "grad_norm": 0.9145541787147522, + "learning_rate": 2.727321284271771e-06, + "loss": 0.5731, + "step": 3464 + }, + { + "epoch": 2.86885698509111, + "grad_norm": 0.9013593792915344, + "learning_rate": 2.726226128472166e-06, + "loss": 0.5901, + "step": 3465 + }, + { + "epoch": 2.869685256764219, + "grad_norm": 0.9561366438865662, + "learning_rate": 2.7251309288999123e-06, + "loss": 0.585, + "step": 3466 + }, + { + "epoch": 2.8705135284373275, + "grad_norm": 0.9313666224479675, + "learning_rate": 2.724035685766923e-06, + "loss": 0.5755, + "step": 3467 + }, + { + "epoch": 2.8713418001104363, + "grad_norm": 0.9003223776817322, + "learning_rate": 2.722940399285118e-06, + "loss": 0.5961, + "step": 3468 + }, + { + "epoch": 2.872170071783545, + "grad_norm": 0.9185585975646973, + "learning_rate": 2.721845069666422e-06, + "loss": 0.5741, + "step": 3469 + }, + { + "epoch": 2.8729983434566537, + "grad_norm": 0.9295309782028198, + "learning_rate": 2.7207496971227743e-06, + "loss": 0.588, + "step": 3470 + }, + { + "epoch": 2.8738266151297625, + "grad_norm": 0.9500821232795715, + "learning_rate": 2.7196542818661188e-06, + "loss": 0.6026, + "step": 3471 + }, + { + "epoch": 2.874654886802871, + "grad_norm": 0.9265774488449097, + "learning_rate": 2.718558824108406e-06, + "loss": 0.5857, + "step": 3472 + }, + { + "epoch": 2.87548315847598, + "grad_norm": 0.9159520864486694, + "learning_rate": 2.717463324061599e-06, + "loss": 0.6023, + "step": 3473 + }, + { + "epoch": 2.876311430149089, + "grad_norm": 0.9114859700202942, + "learning_rate": 2.7163677819376653e-06, + "loss": 0.6144, + "step": 3474 + }, + { + "epoch": 2.877139701822198, + "grad_norm": 0.9181182980537415, + "learning_rate": 2.7152721979485824e-06, + "loss": 0.5853, + "step": 3475 + }, + { + "epoch": 2.8779679734953065, + "grad_norm": 0.9219772219657898, + "learning_rate": 2.7141765723063366e-06, + "loss": 0.5824, + "step": 3476 + }, + { + "epoch": 2.8787962451684153, + "grad_norm": 0.925654411315918, + "learning_rate": 2.713080905222921e-06, + "loss": 0.5859, + "step": 3477 + }, + { + "epoch": 2.879624516841524, + "grad_norm": 0.927891731262207, + "learning_rate": 2.711985196910336e-06, + "loss": 0.6047, + "step": 3478 + }, + { + "epoch": 2.8804527885146327, + "grad_norm": 0.9178606271743774, + "learning_rate": 2.7108894475805912e-06, + "loss": 0.5863, + "step": 3479 + }, + { + "epoch": 2.8812810601877414, + "grad_norm": 0.9341709613800049, + "learning_rate": 2.7097936574457046e-06, + "loss": 0.5718, + "step": 3480 + }, + { + "epoch": 2.8821093318608506, + "grad_norm": 0.9023662805557251, + "learning_rate": 2.7086978267177007e-06, + "loss": 0.593, + "step": 3481 + }, + { + "epoch": 2.882937603533959, + "grad_norm": 0.9269591569900513, + "learning_rate": 2.7076019556086136e-06, + "loss": 0.5955, + "step": 3482 + }, + { + "epoch": 2.883765875207068, + "grad_norm": 0.9572223424911499, + "learning_rate": 2.7065060443304823e-06, + "loss": 0.5698, + "step": 3483 + }, + { + "epoch": 2.884594146880177, + "grad_norm": 0.9251832365989685, + "learning_rate": 2.705410093095357e-06, + "loss": 0.5903, + "step": 3484 + }, + { + "epoch": 2.8854224185532855, + "grad_norm": 0.959985077381134, + "learning_rate": 2.7043141021152937e-06, + "loss": 0.5922, + "step": 3485 + }, + { + "epoch": 2.8862506902263942, + "grad_norm": 0.9356866478919983, + "learning_rate": 2.703218071602355e-06, + "loss": 0.576, + "step": 3486 + }, + { + "epoch": 2.887078961899503, + "grad_norm": 0.9041283130645752, + "learning_rate": 2.702122001768615e-06, + "loss": 0.5748, + "step": 3487 + }, + { + "epoch": 2.8879072335726117, + "grad_norm": 0.9056842923164368, + "learning_rate": 2.7010258928261514e-06, + "loss": 0.5795, + "step": 3488 + }, + { + "epoch": 2.8887355052457204, + "grad_norm": 0.9829550385475159, + "learning_rate": 2.6999297449870522e-06, + "loss": 0.6017, + "step": 3489 + }, + { + "epoch": 2.8895637769188296, + "grad_norm": 0.9483397603034973, + "learning_rate": 2.6988335584634103e-06, + "loss": 0.5849, + "step": 3490 + }, + { + "epoch": 2.890392048591938, + "grad_norm": 0.9553128480911255, + "learning_rate": 2.6977373334673286e-06, + "loss": 0.5518, + "step": 3491 + }, + { + "epoch": 2.891220320265047, + "grad_norm": 0.9298998713493347, + "learning_rate": 2.6966410702109153e-06, + "loss": 0.5947, + "step": 3492 + }, + { + "epoch": 2.8920485919381558, + "grad_norm": 0.9143223166465759, + "learning_rate": 2.6955447689062893e-06, + "loss": 0.5757, + "step": 3493 + }, + { + "epoch": 2.8928768636112645, + "grad_norm": 0.9143601059913635, + "learning_rate": 2.6944484297655727e-06, + "loss": 0.5867, + "step": 3494 + }, + { + "epoch": 2.8937051352843732, + "grad_norm": 0.9283237457275391, + "learning_rate": 2.693352053000898e-06, + "loss": 0.5928, + "step": 3495 + }, + { + "epoch": 2.894533406957482, + "grad_norm": 0.8984352946281433, + "learning_rate": 2.692255638824403e-06, + "loss": 0.5904, + "step": 3496 + }, + { + "epoch": 2.8953616786305907, + "grad_norm": 0.9056443572044373, + "learning_rate": 2.6911591874482334e-06, + "loss": 0.5847, + "step": 3497 + }, + { + "epoch": 2.8961899503036994, + "grad_norm": 0.9254133701324463, + "learning_rate": 2.6900626990845442e-06, + "loss": 0.608, + "step": 3498 + }, + { + "epoch": 2.8970182219768086, + "grad_norm": 0.9658727049827576, + "learning_rate": 2.6889661739454946e-06, + "loss": 0.6063, + "step": 3499 + }, + { + "epoch": 2.8978464936499173, + "grad_norm": 0.9350060820579529, + "learning_rate": 2.6878696122432505e-06, + "loss": 0.5902, + "step": 3500 + }, + { + "epoch": 2.898674765323026, + "grad_norm": 0.9070295691490173, + "learning_rate": 2.686773014189989e-06, + "loss": 0.5883, + "step": 3501 + }, + { + "epoch": 2.8995030369961348, + "grad_norm": 0.9471203088760376, + "learning_rate": 2.6856763799978895e-06, + "loss": 0.6138, + "step": 3502 + }, + { + "epoch": 2.9003313086692435, + "grad_norm": 0.901057243347168, + "learning_rate": 2.6845797098791415e-06, + "loss": 0.5925, + "step": 3503 + }, + { + "epoch": 2.901159580342352, + "grad_norm": 0.9290406107902527, + "learning_rate": 2.683483004045941e-06, + "loss": 0.5728, + "step": 3504 + }, + { + "epoch": 2.901987852015461, + "grad_norm": 0.9371906518936157, + "learning_rate": 2.682386262710489e-06, + "loss": 0.5894, + "step": 3505 + }, + { + "epoch": 2.90281612368857, + "grad_norm": 0.9183947443962097, + "learning_rate": 2.6812894860849953e-06, + "loss": 0.5959, + "step": 3506 + }, + { + "epoch": 2.9036443953616784, + "grad_norm": 0.9351047277450562, + "learning_rate": 2.6801926743816765e-06, + "loss": 0.5863, + "step": 3507 + }, + { + "epoch": 2.9044726670347876, + "grad_norm": 0.93602454662323, + "learning_rate": 2.6790958278127536e-06, + "loss": 0.5845, + "step": 3508 + }, + { + "epoch": 2.9053009387078963, + "grad_norm": 0.9618241786956787, + "learning_rate": 2.677998946590459e-06, + "loss": 0.6068, + "step": 3509 + }, + { + "epoch": 2.906129210381005, + "grad_norm": 0.9157661199569702, + "learning_rate": 2.6769020309270277e-06, + "loss": 0.5608, + "step": 3510 + }, + { + "epoch": 2.9069574820541138, + "grad_norm": 0.9157334566116333, + "learning_rate": 2.6758050810347012e-06, + "loss": 0.5883, + "step": 3511 + }, + { + "epoch": 2.9077857537272225, + "grad_norm": 0.9259895086288452, + "learning_rate": 2.674708097125731e-06, + "loss": 0.578, + "step": 3512 + }, + { + "epoch": 2.908614025400331, + "grad_norm": 0.9327818155288696, + "learning_rate": 2.6736110794123727e-06, + "loss": 0.6157, + "step": 3513 + }, + { + "epoch": 2.90944229707344, + "grad_norm": 0.9186248779296875, + "learning_rate": 2.6725140281068882e-06, + "loss": 0.5901, + "step": 3514 + }, + { + "epoch": 2.910270568746549, + "grad_norm": 0.9106549620628357, + "learning_rate": 2.671416943421548e-06, + "loss": 0.5983, + "step": 3515 + }, + { + "epoch": 2.9110988404196574, + "grad_norm": 0.9274003505706787, + "learning_rate": 2.6703198255686275e-06, + "loss": 0.5886, + "step": 3516 + }, + { + "epoch": 2.9119271120927666, + "grad_norm": 0.9015113115310669, + "learning_rate": 2.6692226747604082e-06, + "loss": 0.5979, + "step": 3517 + }, + { + "epoch": 2.9127553837658753, + "grad_norm": 0.9261288046836853, + "learning_rate": 2.6681254912091788e-06, + "loss": 0.5741, + "step": 3518 + }, + { + "epoch": 2.913583655438984, + "grad_norm": 0.8937399983406067, + "learning_rate": 2.6670282751272335e-06, + "loss": 0.5726, + "step": 3519 + }, + { + "epoch": 2.9144119271120927, + "grad_norm": 0.9202954769134521, + "learning_rate": 2.6659310267268744e-06, + "loss": 0.5788, + "step": 3520 + }, + { + "epoch": 2.9152401987852015, + "grad_norm": 0.9400494694709778, + "learning_rate": 2.6648337462204087e-06, + "loss": 0.568, + "step": 3521 + }, + { + "epoch": 2.91606847045831, + "grad_norm": 0.9291467070579529, + "learning_rate": 2.6637364338201497e-06, + "loss": 0.589, + "step": 3522 + }, + { + "epoch": 2.916896742131419, + "grad_norm": 0.9186866879463196, + "learning_rate": 2.662639089738417e-06, + "loss": 0.59, + "step": 3523 + }, + { + "epoch": 2.917725013804528, + "grad_norm": 0.9596429467201233, + "learning_rate": 2.6615417141875367e-06, + "loss": 0.5759, + "step": 3524 + }, + { + "epoch": 2.918553285477637, + "grad_norm": 0.9279852509498596, + "learning_rate": 2.6604443073798395e-06, + "loss": 0.5969, + "step": 3525 + }, + { + "epoch": 2.9193815571507455, + "grad_norm": 0.9540870189666748, + "learning_rate": 2.6593468695276657e-06, + "loss": 0.5696, + "step": 3526 + }, + { + "epoch": 2.9202098288238543, + "grad_norm": 0.9415677785873413, + "learning_rate": 2.658249400843358e-06, + "loss": 0.5973, + "step": 3527 + }, + { + "epoch": 2.921038100496963, + "grad_norm": 0.8950624465942383, + "learning_rate": 2.657151901539266e-06, + "loss": 0.5952, + "step": 3528 + }, + { + "epoch": 2.9218663721700717, + "grad_norm": 0.8927432298660278, + "learning_rate": 2.656054371827746e-06, + "loss": 0.59, + "step": 3529 + }, + { + "epoch": 2.9226946438431805, + "grad_norm": 0.9516159892082214, + "learning_rate": 2.65495681192116e-06, + "loss": 0.6024, + "step": 3530 + }, + { + "epoch": 2.923522915516289, + "grad_norm": 1.0094369649887085, + "learning_rate": 2.653859222031875e-06, + "loss": 0.596, + "step": 3531 + }, + { + "epoch": 2.924351187189398, + "grad_norm": 0.9162117838859558, + "learning_rate": 2.6527616023722647e-06, + "loss": 0.5739, + "step": 3532 + }, + { + "epoch": 2.925179458862507, + "grad_norm": 0.9844574332237244, + "learning_rate": 2.651663953154709e-06, + "loss": 0.6028, + "step": 3533 + }, + { + "epoch": 2.926007730535616, + "grad_norm": 0.9091664552688599, + "learning_rate": 2.6505662745915915e-06, + "loss": 0.5867, + "step": 3534 + }, + { + "epoch": 2.9268360022087245, + "grad_norm": 0.9107911586761475, + "learning_rate": 2.649468566895304e-06, + "loss": 0.5985, + "step": 3535 + }, + { + "epoch": 2.9276642738818333, + "grad_norm": 0.9392855167388916, + "learning_rate": 2.648370830278241e-06, + "loss": 0.5995, + "step": 3536 + }, + { + "epoch": 2.928492545554942, + "grad_norm": 0.9228532910346985, + "learning_rate": 2.6472730649528068e-06, + "loss": 0.5928, + "step": 3537 + }, + { + "epoch": 2.9293208172280507, + "grad_norm": 0.9555805325508118, + "learning_rate": 2.646175271131407e-06, + "loss": 0.606, + "step": 3538 + }, + { + "epoch": 2.9301490889011594, + "grad_norm": 0.9101611971855164, + "learning_rate": 2.6450774490264547e-06, + "loss": 0.5881, + "step": 3539 + }, + { + "epoch": 2.9309773605742686, + "grad_norm": 0.9572548866271973, + "learning_rate": 2.6439795988503685e-06, + "loss": 0.5861, + "step": 3540 + }, + { + "epoch": 2.931805632247377, + "grad_norm": 0.9441561102867126, + "learning_rate": 2.6428817208155728e-06, + "loss": 0.594, + "step": 3541 + }, + { + "epoch": 2.932633903920486, + "grad_norm": 0.9282614588737488, + "learning_rate": 2.6417838151344945e-06, + "loss": 0.5991, + "step": 3542 + }, + { + "epoch": 2.933462175593595, + "grad_norm": 0.9630649089813232, + "learning_rate": 2.640685882019571e-06, + "loss": 0.6147, + "step": 3543 + }, + { + "epoch": 2.9342904472667035, + "grad_norm": 0.9284330606460571, + "learning_rate": 2.63958792168324e-06, + "loss": 0.6021, + "step": 3544 + }, + { + "epoch": 2.9351187189398122, + "grad_norm": 0.9337431788444519, + "learning_rate": 2.638489934337948e-06, + "loss": 0.602, + "step": 3545 + }, + { + "epoch": 2.935946990612921, + "grad_norm": 0.8977872729301453, + "learning_rate": 2.637391920196144e-06, + "loss": 0.5901, + "step": 3546 + }, + { + "epoch": 2.9367752622860297, + "grad_norm": 0.962690532207489, + "learning_rate": 2.636293879470284e-06, + "loss": 0.5793, + "step": 3547 + }, + { + "epoch": 2.9376035339591384, + "grad_norm": 0.8999794721603394, + "learning_rate": 2.6351958123728294e-06, + "loss": 0.6027, + "step": 3548 + }, + { + "epoch": 2.9384318056322476, + "grad_norm": 0.9200299382209778, + "learning_rate": 2.6340977191162457e-06, + "loss": 0.5985, + "step": 3549 + }, + { + "epoch": 2.9392600773053563, + "grad_norm": 0.9161362648010254, + "learning_rate": 2.6329995999130032e-06, + "loss": 0.595, + "step": 3550 + }, + { + "epoch": 2.940088348978465, + "grad_norm": 0.9192216992378235, + "learning_rate": 2.631901454975578e-06, + "loss": 0.5844, + "step": 3551 + }, + { + "epoch": 2.9409166206515738, + "grad_norm": 0.9136098623275757, + "learning_rate": 2.6308032845164503e-06, + "loss": 0.5932, + "step": 3552 + }, + { + "epoch": 2.9417448923246825, + "grad_norm": 0.9625688791275024, + "learning_rate": 2.629705088748106e-06, + "loss": 0.5811, + "step": 3553 + }, + { + "epoch": 2.9425731639977912, + "grad_norm": 0.9300727248191833, + "learning_rate": 2.6286068678830375e-06, + "loss": 0.6035, + "step": 3554 + }, + { + "epoch": 2.9434014356709, + "grad_norm": 0.9947618246078491, + "learning_rate": 2.627508622133738e-06, + "loss": 0.5822, + "step": 3555 + }, + { + "epoch": 2.9442297073440087, + "grad_norm": 0.9255596995353699, + "learning_rate": 2.626410351712709e-06, + "loss": 0.5824, + "step": 3556 + }, + { + "epoch": 2.9450579790171174, + "grad_norm": 0.9395889043807983, + "learning_rate": 2.6253120568324557e-06, + "loss": 0.6121, + "step": 3557 + }, + { + "epoch": 2.9458862506902266, + "grad_norm": 0.9438552856445312, + "learning_rate": 2.6242137377054865e-06, + "loss": 0.5843, + "step": 3558 + }, + { + "epoch": 2.9467145223633353, + "grad_norm": 0.9153048992156982, + "learning_rate": 2.623115394544318e-06, + "loss": 0.5893, + "step": 3559 + }, + { + "epoch": 2.947542794036444, + "grad_norm": 0.9119593501091003, + "learning_rate": 2.6220170275614677e-06, + "loss": 0.6047, + "step": 3560 + }, + { + "epoch": 2.9483710657095528, + "grad_norm": 0.9174615144729614, + "learning_rate": 2.6209186369694606e-06, + "loss": 0.5949, + "step": 3561 + }, + { + "epoch": 2.9491993373826615, + "grad_norm": 0.9030957818031311, + "learning_rate": 2.619820222980824e-06, + "loss": 0.6146, + "step": 3562 + }, + { + "epoch": 2.95002760905577, + "grad_norm": 0.9036881923675537, + "learning_rate": 2.618721785808091e-06, + "loss": 0.5946, + "step": 3563 + }, + { + "epoch": 2.950855880728879, + "grad_norm": 0.8928536772727966, + "learning_rate": 2.617623325663799e-06, + "loss": 0.5679, + "step": 3564 + }, + { + "epoch": 2.951684152401988, + "grad_norm": 0.9225850701332092, + "learning_rate": 2.6165248427604907e-06, + "loss": 0.5938, + "step": 3565 + }, + { + "epoch": 2.9525124240750964, + "grad_norm": 0.9023733139038086, + "learning_rate": 2.6154263373107113e-06, + "loss": 0.5896, + "step": 3566 + }, + { + "epoch": 2.9533406957482056, + "grad_norm": 0.9055750370025635, + "learning_rate": 2.6143278095270113e-06, + "loss": 0.6007, + "step": 3567 + }, + { + "epoch": 2.9541689674213143, + "grad_norm": 0.8988812565803528, + "learning_rate": 2.613229259621947e-06, + "loss": 0.6097, + "step": 3568 + }, + { + "epoch": 2.954997239094423, + "grad_norm": 0.9200608730316162, + "learning_rate": 2.6121306878080753e-06, + "loss": 0.58, + "step": 3569 + }, + { + "epoch": 2.9558255107675317, + "grad_norm": 0.9378446936607361, + "learning_rate": 2.6110320942979606e-06, + "loss": 0.6043, + "step": 3570 + }, + { + "epoch": 2.9566537824406405, + "grad_norm": 0.9543853402137756, + "learning_rate": 2.6099334793041714e-06, + "loss": 0.58, + "step": 3571 + }, + { + "epoch": 2.957482054113749, + "grad_norm": 0.9590248465538025, + "learning_rate": 2.6088348430392784e-06, + "loss": 0.6004, + "step": 3572 + }, + { + "epoch": 2.958310325786858, + "grad_norm": 0.980459988117218, + "learning_rate": 2.607736185715858e-06, + "loss": 0.6118, + "step": 3573 + }, + { + "epoch": 2.959138597459967, + "grad_norm": 0.9050843715667725, + "learning_rate": 2.60663750754649e-06, + "loss": 0.5913, + "step": 3574 + }, + { + "epoch": 2.9599668691330754, + "grad_norm": 0.9189395904541016, + "learning_rate": 2.6055388087437576e-06, + "loss": 0.5997, + "step": 3575 + }, + { + "epoch": 2.9607951408061846, + "grad_norm": 0.9273754954338074, + "learning_rate": 2.6044400895202502e-06, + "loss": 0.586, + "step": 3576 + }, + { + "epoch": 2.9616234124792933, + "grad_norm": 0.90342116355896, + "learning_rate": 2.6033413500885594e-06, + "loss": 0.5846, + "step": 3577 + }, + { + "epoch": 2.962451684152402, + "grad_norm": 0.9073745012283325, + "learning_rate": 2.6022425906612803e-06, + "loss": 0.5899, + "step": 3578 + }, + { + "epoch": 2.9632799558255107, + "grad_norm": 0.9221594333648682, + "learning_rate": 2.6011438114510135e-06, + "loss": 0.5837, + "step": 3579 + }, + { + "epoch": 2.9641082274986195, + "grad_norm": 0.9228225350379944, + "learning_rate": 2.6000450126703613e-06, + "loss": 0.5761, + "step": 3580 + }, + { + "epoch": 2.964936499171728, + "grad_norm": 0.9136195182800293, + "learning_rate": 2.5989461945319315e-06, + "loss": 0.5977, + "step": 3581 + }, + { + "epoch": 2.965764770844837, + "grad_norm": 0.9152894616127014, + "learning_rate": 2.5978473572483366e-06, + "loss": 0.5829, + "step": 3582 + }, + { + "epoch": 2.966593042517946, + "grad_norm": 0.9124417304992676, + "learning_rate": 2.596748501032189e-06, + "loss": 0.5718, + "step": 3583 + }, + { + "epoch": 2.967421314191055, + "grad_norm": 0.9212865829467773, + "learning_rate": 2.5956496260961094e-06, + "loss": 0.6056, + "step": 3584 + }, + { + "epoch": 2.9682495858641635, + "grad_norm": 0.9101969003677368, + "learning_rate": 2.594550732652718e-06, + "loss": 0.6075, + "step": 3585 + }, + { + "epoch": 2.9690778575372723, + "grad_norm": 0.9225336313247681, + "learning_rate": 2.5934518209146414e-06, + "loss": 0.5894, + "step": 3586 + }, + { + "epoch": 2.969906129210381, + "grad_norm": 0.9271754026412964, + "learning_rate": 2.592352891094509e-06, + "loss": 0.5939, + "step": 3587 + }, + { + "epoch": 2.9707344008834897, + "grad_norm": 0.9217578768730164, + "learning_rate": 2.591253943404953e-06, + "loss": 0.5858, + "step": 3588 + }, + { + "epoch": 2.9715626725565985, + "grad_norm": 0.8871975541114807, + "learning_rate": 2.5901549780586085e-06, + "loss": 0.6052, + "step": 3589 + }, + { + "epoch": 2.9723909442297076, + "grad_norm": 0.9371136426925659, + "learning_rate": 2.5890559952681177e-06, + "loss": 0.6069, + "step": 3590 + }, + { + "epoch": 2.973219215902816, + "grad_norm": 0.9185692667961121, + "learning_rate": 2.5879569952461215e-06, + "loss": 0.5913, + "step": 3591 + }, + { + "epoch": 2.974047487575925, + "grad_norm": 0.9043979048728943, + "learning_rate": 2.586857978205266e-06, + "loss": 0.5747, + "step": 3592 + }, + { + "epoch": 2.974875759249034, + "grad_norm": 0.8951820731163025, + "learning_rate": 2.5857589443582014e-06, + "loss": 0.5904, + "step": 3593 + }, + { + "epoch": 2.9757040309221425, + "grad_norm": 0.9490121006965637, + "learning_rate": 2.5846598939175816e-06, + "loss": 0.583, + "step": 3594 + }, + { + "epoch": 2.9765323025952513, + "grad_norm": 0.9266814589500427, + "learning_rate": 2.583560827096061e-06, + "loss": 0.6039, + "step": 3595 + }, + { + "epoch": 2.97736057426836, + "grad_norm": 0.9517508149147034, + "learning_rate": 2.582461744106299e-06, + "loss": 0.572, + "step": 3596 + }, + { + "epoch": 2.9781888459414687, + "grad_norm": 0.9260841608047485, + "learning_rate": 2.581362645160958e-06, + "loss": 0.5898, + "step": 3597 + }, + { + "epoch": 2.9790171176145774, + "grad_norm": 0.9134445190429688, + "learning_rate": 2.5802635304727043e-06, + "loss": 0.6037, + "step": 3598 + }, + { + "epoch": 2.9798453892876866, + "grad_norm": 0.9307652115821838, + "learning_rate": 2.5791644002542054e-06, + "loss": 0.5941, + "step": 3599 + }, + { + "epoch": 2.980673660960795, + "grad_norm": 0.9665980935096741, + "learning_rate": 2.578065254718133e-06, + "loss": 0.5816, + "step": 3600 + }, + { + "epoch": 2.981501932633904, + "grad_norm": 0.9137181639671326, + "learning_rate": 2.5769660940771614e-06, + "loss": 0.5962, + "step": 3601 + }, + { + "epoch": 2.982330204307013, + "grad_norm": 0.9031209349632263, + "learning_rate": 2.575866918543968e-06, + "loss": 0.5853, + "step": 3602 + }, + { + "epoch": 2.9831584759801215, + "grad_norm": 0.9292592406272888, + "learning_rate": 2.5747677283312332e-06, + "loss": 0.5639, + "step": 3603 + }, + { + "epoch": 2.9839867476532302, + "grad_norm": 0.9193013906478882, + "learning_rate": 2.57366852365164e-06, + "loss": 0.5942, + "step": 3604 + }, + { + "epoch": 2.984815019326339, + "grad_norm": 0.8945804238319397, + "learning_rate": 2.572569304717874e-06, + "loss": 0.5882, + "step": 3605 + }, + { + "epoch": 2.9856432909994477, + "grad_norm": 0.9231047034263611, + "learning_rate": 2.5714700717426237e-06, + "loss": 0.5768, + "step": 3606 + }, + { + "epoch": 2.9864715626725564, + "grad_norm": 0.9362046718597412, + "learning_rate": 2.5703708249385824e-06, + "loss": 0.6145, + "step": 3607 + }, + { + "epoch": 2.9872998343456656, + "grad_norm": 0.9291548728942871, + "learning_rate": 2.5692715645184412e-06, + "loss": 0.5959, + "step": 3608 + }, + { + "epoch": 2.9881281060187743, + "grad_norm": 0.9203015565872192, + "learning_rate": 2.5681722906948977e-06, + "loss": 0.5906, + "step": 3609 + }, + { + "epoch": 2.988956377691883, + "grad_norm": 0.9317998886108398, + "learning_rate": 2.567073003680652e-06, + "loss": 0.5845, + "step": 3610 + }, + { + "epoch": 2.9897846493649918, + "grad_norm": 0.9188578128814697, + "learning_rate": 2.5659737036884054e-06, + "loss": 0.5901, + "step": 3611 + }, + { + "epoch": 2.9906129210381005, + "grad_norm": 0.9100846648216248, + "learning_rate": 2.564874390930862e-06, + "loss": 0.5952, + "step": 3612 + }, + { + "epoch": 2.9914411927112092, + "grad_norm": 0.9292994737625122, + "learning_rate": 2.563775065620729e-06, + "loss": 0.6112, + "step": 3613 + }, + { + "epoch": 2.992269464384318, + "grad_norm": 0.9025140404701233, + "learning_rate": 2.5626757279707143e-06, + "loss": 0.5691, + "step": 3614 + }, + { + "epoch": 2.9930977360574267, + "grad_norm": 0.9413566589355469, + "learning_rate": 2.561576378193531e-06, + "loss": 0.574, + "step": 3615 + }, + { + "epoch": 2.9939260077305354, + "grad_norm": 0.9244924783706665, + "learning_rate": 2.560477016501893e-06, + "loss": 0.606, + "step": 3616 + }, + { + "epoch": 2.9947542794036446, + "grad_norm": 0.8985790014266968, + "learning_rate": 2.5593776431085154e-06, + "loss": 0.5823, + "step": 3617 + }, + { + "epoch": 2.9955825510767533, + "grad_norm": 0.8941566348075867, + "learning_rate": 2.5582782582261182e-06, + "loss": 0.5643, + "step": 3618 + }, + { + "epoch": 2.996410822749862, + "grad_norm": 0.8992639183998108, + "learning_rate": 2.557178862067421e-06, + "loss": 0.6073, + "step": 3619 + }, + { + "epoch": 2.9972390944229708, + "grad_norm": 0.9369136691093445, + "learning_rate": 2.5560794548451464e-06, + "loss": 0.6031, + "step": 3620 + }, + { + "epoch": 2.9980673660960795, + "grad_norm": 0.8975582718849182, + "learning_rate": 2.5549800367720207e-06, + "loss": 0.6034, + "step": 3621 + } + ], + "logging_steps": 1, + "max_steps": 7242, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 1207, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 9.143410830040105e+18, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-3621/training_args.bin b/checkpoint-3621/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..7017eeafdf1f43b26668f0e0e5710ca52345a005 --- /dev/null +++ b/checkpoint-3621/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fdc878d2541c09353dd68f4f8a2464140b28fd84e1bc48b59914a6e84328e85e +size 8056 diff --git a/checkpoint-3621/zero_to_fp32.py b/checkpoint-3621/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-3621/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-4828/README.md b/checkpoint-4828/README.md new file mode 100644 index 0000000000000000000000000000000000000000..049d467664ca6172b7ffbe6ba60b3eac7479cac4 --- /dev/null +++ b/checkpoint-4828/README.md @@ -0,0 +1,202 @@ +--- +base_model: meta-llama/Llama-3.1-8B +library_name: peft +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.14.0 \ No newline at end of file diff --git a/checkpoint-4828/adapter_config.json b/checkpoint-4828/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d6fca8a067dce19d599e7f1e24bea8019968fe0f --- /dev/null +++ b/checkpoint-4828/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.1-8B", + "bias": "none", + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "up_proj", + "q_proj", + "k_proj", + "o_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-4828/adapter_model.safetensors b/checkpoint-4828/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1a9c436df61c3e88b5b1c5e972b7a2b3bf11b4c5 --- /dev/null +++ b/checkpoint-4828/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c9e0d2664892c77d6a71de4b69af8f2bf463c08be0935688c86ed325d6ac23c0 +size 3443586272 diff --git a/checkpoint-4828/global_step4826/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-4828/global_step4826/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..e8c2ca0029796ee7801c2da926ebfde928020844 --- /dev/null +++ b/checkpoint-4828/global_step4826/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:cbbe9d0e2e31e279e0d3b25bc3aeac674f6057cd80d4e3912544bc0423f143fc +size 20661195036 diff --git a/checkpoint-4828/global_step4826/mp_rank_00_model_states.pt b/checkpoint-4828/global_step4826/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..c862f69684bd81019646a6512ff29509c962f926 --- /dev/null +++ b/checkpoint-4828/global_step4826/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d4d3535e206cf22602d6f26b7510d8e77d2593a92207c602c680bc6b322b68c9 +size 3555326777 diff --git a/checkpoint-4828/latest b/checkpoint-4828/latest new file mode 100644 index 0000000000000000000000000000000000000000..19337dfd52dd49df3f3f829fbff2b82e372c738b --- /dev/null +++ b/checkpoint-4828/latest @@ -0,0 +1 @@ +global_step4826 \ No newline at end of file diff --git a/checkpoint-4828/rng_state.pth b/checkpoint-4828/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..d5bcc1081f87f010809156d485b70291f371af1c --- /dev/null +++ b/checkpoint-4828/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:32a0489209bb5e9aede37c126408c063c8067098acaae398670804365a4c0ec6 +size 14244 diff --git a/checkpoint-4828/scheduler.pt b/checkpoint-4828/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..24cbf12945897acae626ba0b6f07a8c70c7333f9 --- /dev/null +++ b/checkpoint-4828/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4b4a7f88e8481505d5f5b6c81c4e9ad24ba1323927c74c3fe1349bd3c526863e +size 1064 diff --git a/checkpoint-4828/special_tokens_map.json b/checkpoint-4828/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..e5b39b6305d89284b04934011c68dbb26bf588ca --- /dev/null +++ b/checkpoint-4828/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-4828/tokenizer.json b/checkpoint-4828/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-4828/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-4828/tokenizer_config.json b/checkpoint-4828/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..81dd14db6632ad5b35b9d447732e37ac074873a5 --- /dev/null +++ b/checkpoint-4828/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +} diff --git a/checkpoint-4828/trainer_state.json b/checkpoint-4828/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..d62eff43c7a7179c085e663cbf43f4f82a4512c1 --- /dev/null +++ b/checkpoint-4828/trainer_state.json @@ -0,0 +1,33829 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 3.9972390944229708, + "eval_steps": 500, + "global_step": 4828, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0008282716731087797, + "grad_norm": 2.4795358180999756, + "learning_rate": 5.0000000000000004e-08, + "loss": 1.1215, + "step": 1 + }, + { + "epoch": 0.0016565433462175593, + "grad_norm": 2.6060028076171875, + "learning_rate": 1.0000000000000001e-07, + "loss": 1.1219, + "step": 2 + }, + { + "epoch": 0.002484815019326339, + "grad_norm": 2.750868082046509, + "learning_rate": 1.5000000000000002e-07, + "loss": 1.125, + "step": 3 + }, + { + "epoch": 0.0033130866924351186, + "grad_norm": 2.7321369647979736, + "learning_rate": 2.0000000000000002e-07, + "loss": 1.1164, + "step": 4 + }, + { + "epoch": 0.0041413583655438985, + "grad_norm": 2.8193929195404053, + "learning_rate": 2.5000000000000004e-07, + "loss": 1.1263, + "step": 5 + }, + { + "epoch": 0.004969630038652678, + "grad_norm": 2.648301839828491, + "learning_rate": 3.0000000000000004e-07, + "loss": 1.1092, + "step": 6 + }, + { + "epoch": 0.005797901711761457, + "grad_norm": 2.751993179321289, + "learning_rate": 3.5000000000000004e-07, + "loss": 1.0997, + "step": 7 + }, + { + "epoch": 0.006626173384870237, + "grad_norm": 2.592381000518799, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.112, + "step": 8 + }, + { + "epoch": 0.007454445057979017, + "grad_norm": 2.3135671615600586, + "learning_rate": 4.5000000000000003e-07, + "loss": 1.1045, + "step": 9 + }, + { + "epoch": 0.008282716731087797, + "grad_norm": 2.60491943359375, + "learning_rate": 5.000000000000001e-07, + "loss": 1.0956, + "step": 10 + }, + { + "epoch": 0.009110988404196576, + "grad_norm": 2.2435271739959717, + "learning_rate": 5.5e-07, + "loss": 1.1241, + "step": 11 + }, + { + "epoch": 0.009939260077305357, + "grad_norm": 2.4412319660186768, + "learning_rate": 6.000000000000001e-07, + "loss": 1.11, + "step": 12 + }, + { + "epoch": 0.010767531750414136, + "grad_norm": 2.2075185775756836, + "learning_rate": 6.5e-07, + "loss": 1.0878, + "step": 13 + }, + { + "epoch": 0.011595803423522915, + "grad_norm": 2.101902723312378, + "learning_rate": 7.000000000000001e-07, + "loss": 1.0863, + "step": 14 + }, + { + "epoch": 0.012424075096631695, + "grad_norm": 2.134535074234009, + "learning_rate": 7.5e-07, + "loss": 1.0735, + "step": 15 + }, + { + "epoch": 0.013252346769740474, + "grad_norm": 2.000504493713379, + "learning_rate": 8.000000000000001e-07, + "loss": 1.0875, + "step": 16 + }, + { + "epoch": 0.014080618442849255, + "grad_norm": 1.9191882610321045, + "learning_rate": 8.500000000000001e-07, + "loss": 1.1039, + "step": 17 + }, + { + "epoch": 0.014908890115958034, + "grad_norm": 1.7732031345367432, + "learning_rate": 9.000000000000001e-07, + "loss": 1.0819, + "step": 18 + }, + { + "epoch": 0.015737161789066815, + "grad_norm": 1.7775251865386963, + "learning_rate": 9.500000000000001e-07, + "loss": 1.0936, + "step": 19 + }, + { + "epoch": 0.016565433462175594, + "grad_norm": 1.6648598909378052, + "learning_rate": 1.0000000000000002e-06, + "loss": 1.0881, + "step": 20 + }, + { + "epoch": 0.017393705135284373, + "grad_norm": 1.5726720094680786, + "learning_rate": 1.0500000000000001e-06, + "loss": 1.07, + "step": 21 + }, + { + "epoch": 0.018221976808393152, + "grad_norm": 1.561417818069458, + "learning_rate": 1.1e-06, + "loss": 1.0668, + "step": 22 + }, + { + "epoch": 0.019050248481501934, + "grad_norm": 1.4953842163085938, + "learning_rate": 1.1500000000000002e-06, + "loss": 1.054, + "step": 23 + }, + { + "epoch": 0.019878520154610713, + "grad_norm": 1.4247829914093018, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.0363, + "step": 24 + }, + { + "epoch": 0.020706791827719492, + "grad_norm": 1.3483426570892334, + "learning_rate": 1.25e-06, + "loss": 1.0143, + "step": 25 + }, + { + "epoch": 0.02153506350082827, + "grad_norm": 1.3412944078445435, + "learning_rate": 1.3e-06, + "loss": 1.025, + "step": 26 + }, + { + "epoch": 0.02236333517393705, + "grad_norm": 1.2834237813949585, + "learning_rate": 1.3500000000000002e-06, + "loss": 1.0227, + "step": 27 + }, + { + "epoch": 0.02319160684704583, + "grad_norm": 1.2128241062164307, + "learning_rate": 1.4000000000000001e-06, + "loss": 1.0161, + "step": 28 + }, + { + "epoch": 0.024019878520154612, + "grad_norm": 1.1665573120117188, + "learning_rate": 1.45e-06, + "loss": 1.0034, + "step": 29 + }, + { + "epoch": 0.02484815019326339, + "grad_norm": 1.1575899124145508, + "learning_rate": 1.5e-06, + "loss": 0.9968, + "step": 30 + }, + { + "epoch": 0.02567642186637217, + "grad_norm": 1.195304274559021, + "learning_rate": 1.5500000000000002e-06, + "loss": 1.0176, + "step": 31 + }, + { + "epoch": 0.02650469353948095, + "grad_norm": 1.1321678161621094, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.0113, + "step": 32 + }, + { + "epoch": 0.027332965212589728, + "grad_norm": 1.1076338291168213, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.9971, + "step": 33 + }, + { + "epoch": 0.02816123688569851, + "grad_norm": 1.0770610570907593, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.9994, + "step": 34 + }, + { + "epoch": 0.02898950855880729, + "grad_norm": 1.0767767429351807, + "learning_rate": 1.75e-06, + "loss": 0.978, + "step": 35 + }, + { + "epoch": 0.02981778023191607, + "grad_norm": 1.054800033569336, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.9829, + "step": 36 + }, + { + "epoch": 0.030646051905024847, + "grad_norm": 1.031015157699585, + "learning_rate": 1.85e-06, + "loss": 0.97, + "step": 37 + }, + { + "epoch": 0.03147432357813363, + "grad_norm": 1.0182578563690186, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.9549, + "step": 38 + }, + { + "epoch": 0.032302595251242405, + "grad_norm": 0.9796578288078308, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.9762, + "step": 39 + }, + { + "epoch": 0.03313086692435119, + "grad_norm": 0.9956578016281128, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.9693, + "step": 40 + }, + { + "epoch": 0.03395913859745997, + "grad_norm": 0.9383260011672974, + "learning_rate": 2.05e-06, + "loss": 0.9556, + "step": 41 + }, + { + "epoch": 0.034787410270568746, + "grad_norm": 0.9372819066047668, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.9419, + "step": 42 + }, + { + "epoch": 0.03561568194367753, + "grad_norm": 0.9334004521369934, + "learning_rate": 2.15e-06, + "loss": 0.9587, + "step": 43 + }, + { + "epoch": 0.036443953616786304, + "grad_norm": 0.8916004300117493, + "learning_rate": 2.2e-06, + "loss": 0.9227, + "step": 44 + }, + { + "epoch": 0.037272225289895086, + "grad_norm": 0.8698756694793701, + "learning_rate": 2.25e-06, + "loss": 0.9275, + "step": 45 + }, + { + "epoch": 0.03810049696300387, + "grad_norm": 0.8775100708007812, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.941, + "step": 46 + }, + { + "epoch": 0.038928768636112644, + "grad_norm": 0.8763384222984314, + "learning_rate": 2.35e-06, + "loss": 0.9253, + "step": 47 + }, + { + "epoch": 0.03975704030922143, + "grad_norm": 0.8534179925918579, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.9327, + "step": 48 + }, + { + "epoch": 0.0405853119823302, + "grad_norm": 0.8432146906852722, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.9236, + "step": 49 + }, + { + "epoch": 0.041413583655438985, + "grad_norm": 0.9899975061416626, + "learning_rate": 2.5e-06, + "loss": 0.8839, + "step": 50 + }, + { + "epoch": 0.04224185532854776, + "grad_norm": 0.8370104432106018, + "learning_rate": 2.55e-06, + "loss": 0.906, + "step": 51 + }, + { + "epoch": 0.04307012700165654, + "grad_norm": 0.8207446336746216, + "learning_rate": 2.6e-06, + "loss": 0.8755, + "step": 52 + }, + { + "epoch": 0.043898398674765325, + "grad_norm": 0.805584728717804, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.9049, + "step": 53 + }, + { + "epoch": 0.0447266703478741, + "grad_norm": 0.8162860870361328, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.8993, + "step": 54 + }, + { + "epoch": 0.04555494202098288, + "grad_norm": 0.8151668906211853, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.9282, + "step": 55 + }, + { + "epoch": 0.04638321369409166, + "grad_norm": 0.7996147274971008, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.8946, + "step": 56 + }, + { + "epoch": 0.04721148536720044, + "grad_norm": 0.8223204016685486, + "learning_rate": 2.85e-06, + "loss": 0.897, + "step": 57 + }, + { + "epoch": 0.048039757040309224, + "grad_norm": 0.8117988109588623, + "learning_rate": 2.9e-06, + "loss": 0.8845, + "step": 58 + }, + { + "epoch": 0.048868028713418, + "grad_norm": 0.7902786731719971, + "learning_rate": 2.95e-06, + "loss": 0.8846, + "step": 59 + }, + { + "epoch": 0.04969630038652678, + "grad_norm": 0.804192304611206, + "learning_rate": 3e-06, + "loss": 0.8722, + "step": 60 + }, + { + "epoch": 0.05052457205963556, + "grad_norm": 0.7852040529251099, + "learning_rate": 3.05e-06, + "loss": 0.8821, + "step": 61 + }, + { + "epoch": 0.05135284373274434, + "grad_norm": 0.8045080304145813, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.8661, + "step": 62 + }, + { + "epoch": 0.05218111540585312, + "grad_norm": 0.7920243144035339, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.8712, + "step": 63 + }, + { + "epoch": 0.0530093870789619, + "grad_norm": 0.7601844072341919, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.8792, + "step": 64 + }, + { + "epoch": 0.05383765875207068, + "grad_norm": 0.7833033800125122, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.8811, + "step": 65 + }, + { + "epoch": 0.054665930425179456, + "grad_norm": 0.7691664099693298, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.8701, + "step": 66 + }, + { + "epoch": 0.05549420209828824, + "grad_norm": 0.7845340967178345, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.8929, + "step": 67 + }, + { + "epoch": 0.05632247377139702, + "grad_norm": 0.9076205492019653, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.8352, + "step": 68 + }, + { + "epoch": 0.057150745444505796, + "grad_norm": 0.7614927887916565, + "learning_rate": 3.45e-06, + "loss": 0.8766, + "step": 69 + }, + { + "epoch": 0.05797901711761458, + "grad_norm": 0.7722934484481812, + "learning_rate": 3.5e-06, + "loss": 0.8666, + "step": 70 + }, + { + "epoch": 0.058807288790723354, + "grad_norm": 0.7603748440742493, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.84, + "step": 71 + }, + { + "epoch": 0.05963556046383214, + "grad_norm": 0.7563591003417969, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.8784, + "step": 72 + }, + { + "epoch": 0.06046383213694092, + "grad_norm": 0.7835114002227783, + "learning_rate": 3.65e-06, + "loss": 0.8516, + "step": 73 + }, + { + "epoch": 0.061292103810049695, + "grad_norm": 0.7760982513427734, + "learning_rate": 3.7e-06, + "loss": 0.8265, + "step": 74 + }, + { + "epoch": 0.06212037548315848, + "grad_norm": 0.7978456020355225, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.8564, + "step": 75 + }, + { + "epoch": 0.06294864715626726, + "grad_norm": 0.7940747141838074, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.8635, + "step": 76 + }, + { + "epoch": 0.06377691882937604, + "grad_norm": 0.8055519461631775, + "learning_rate": 3.85e-06, + "loss": 0.8445, + "step": 77 + }, + { + "epoch": 0.06460519050248481, + "grad_norm": 0.7532406449317932, + "learning_rate": 3.900000000000001e-06, + "loss": 0.8772, + "step": 78 + }, + { + "epoch": 0.06543346217559359, + "grad_norm": 0.763933539390564, + "learning_rate": 3.95e-06, + "loss": 0.8574, + "step": 79 + }, + { + "epoch": 0.06626173384870238, + "grad_norm": 0.7833206057548523, + "learning_rate": 4.000000000000001e-06, + "loss": 0.8643, + "step": 80 + }, + { + "epoch": 0.06709000552181116, + "grad_norm": 0.7825527787208557, + "learning_rate": 4.05e-06, + "loss": 0.8518, + "step": 81 + }, + { + "epoch": 0.06791827719491994, + "grad_norm": 0.7749581336975098, + "learning_rate": 4.1e-06, + "loss": 0.8146, + "step": 82 + }, + { + "epoch": 0.06874654886802871, + "grad_norm": 0.8117144703865051, + "learning_rate": 4.15e-06, + "loss": 0.8719, + "step": 83 + }, + { + "epoch": 0.06957482054113749, + "grad_norm": 0.8440507054328918, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.8665, + "step": 84 + }, + { + "epoch": 0.07040309221424627, + "grad_norm": 0.7941984534263611, + "learning_rate": 4.25e-06, + "loss": 0.8492, + "step": 85 + }, + { + "epoch": 0.07123136388735506, + "grad_norm": 0.8301989436149597, + "learning_rate": 4.3e-06, + "loss": 0.8544, + "step": 86 + }, + { + "epoch": 0.07205963556046384, + "grad_norm": 0.757798969745636, + "learning_rate": 4.350000000000001e-06, + "loss": 0.8426, + "step": 87 + }, + { + "epoch": 0.07288790723357261, + "grad_norm": 0.7817673683166504, + "learning_rate": 4.4e-06, + "loss": 0.8554, + "step": 88 + }, + { + "epoch": 0.07371617890668139, + "grad_norm": 0.7609818577766418, + "learning_rate": 4.450000000000001e-06, + "loss": 0.8473, + "step": 89 + }, + { + "epoch": 0.07454445057979017, + "grad_norm": 0.8146091103553772, + "learning_rate": 4.5e-06, + "loss": 0.8317, + "step": 90 + }, + { + "epoch": 0.07537272225289895, + "grad_norm": 0.8058239817619324, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.8411, + "step": 91 + }, + { + "epoch": 0.07620099392600774, + "grad_norm": 0.7700346112251282, + "learning_rate": 4.600000000000001e-06, + "loss": 0.815, + "step": 92 + }, + { + "epoch": 0.0770292655991165, + "grad_norm": 0.7949067950248718, + "learning_rate": 4.65e-06, + "loss": 0.8512, + "step": 93 + }, + { + "epoch": 0.07785753727222529, + "grad_norm": 0.8112176060676575, + "learning_rate": 4.7e-06, + "loss": 0.8398, + "step": 94 + }, + { + "epoch": 0.07868580894533407, + "grad_norm": 0.7990434765815735, + "learning_rate": 4.75e-06, + "loss": 0.8276, + "step": 95 + }, + { + "epoch": 0.07951408061844285, + "grad_norm": 0.7942920923233032, + "learning_rate": 4.800000000000001e-06, + "loss": 0.8333, + "step": 96 + }, + { + "epoch": 0.08034235229155164, + "grad_norm": 0.8005534410476685, + "learning_rate": 4.85e-06, + "loss": 0.8453, + "step": 97 + }, + { + "epoch": 0.0811706239646604, + "grad_norm": 0.8103216290473938, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8303, + "step": 98 + }, + { + "epoch": 0.08199889563776919, + "grad_norm": 0.8217238187789917, + "learning_rate": 4.95e-06, + "loss": 0.8204, + "step": 99 + }, + { + "epoch": 0.08282716731087797, + "grad_norm": 0.7958654761314392, + "learning_rate": 5e-06, + "loss": 0.8237, + "step": 100 + }, + { + "epoch": 0.08365543898398675, + "grad_norm": 0.8631463050842285, + "learning_rate": 4.999999758136652e-06, + "loss": 0.821, + "step": 101 + }, + { + "epoch": 0.08448371065709552, + "grad_norm": 0.7666735053062439, + "learning_rate": 4.999999032546657e-06, + "loss": 0.8307, + "step": 102 + }, + { + "epoch": 0.0853119823302043, + "grad_norm": 0.7709226012229919, + "learning_rate": 4.999997823230153e-06, + "loss": 0.8299, + "step": 103 + }, + { + "epoch": 0.08614025400331309, + "grad_norm": 0.8020651340484619, + "learning_rate": 4.9999961301873736e-06, + "loss": 0.8075, + "step": 104 + }, + { + "epoch": 0.08696852567642187, + "grad_norm": 0.8019537329673767, + "learning_rate": 4.999993953418649e-06, + "loss": 0.8615, + "step": 105 + }, + { + "epoch": 0.08779679734953065, + "grad_norm": 0.7904282808303833, + "learning_rate": 4.999991292924399e-06, + "loss": 0.8353, + "step": 106 + }, + { + "epoch": 0.08862506902263942, + "grad_norm": 0.7854570150375366, + "learning_rate": 4.999988148705138e-06, + "loss": 0.8154, + "step": 107 + }, + { + "epoch": 0.0894533406957482, + "grad_norm": 0.8044222593307495, + "learning_rate": 4.999984520761475e-06, + "loss": 0.8117, + "step": 108 + }, + { + "epoch": 0.09028161236885698, + "grad_norm": 0.7986987829208374, + "learning_rate": 4.999980409094112e-06, + "loss": 0.8407, + "step": 109 + }, + { + "epoch": 0.09110988404196577, + "grad_norm": 0.7976786494255066, + "learning_rate": 4.999975813703844e-06, + "loss": 0.7871, + "step": 110 + }, + { + "epoch": 0.09193815571507455, + "grad_norm": 0.8330315351486206, + "learning_rate": 4.9999707345915605e-06, + "loss": 0.8129, + "step": 111 + }, + { + "epoch": 0.09276642738818332, + "grad_norm": 0.8350826501846313, + "learning_rate": 4.999965171758245e-06, + "loss": 0.8144, + "step": 112 + }, + { + "epoch": 0.0935946990612921, + "grad_norm": 0.8054768443107605, + "learning_rate": 4.999959125204973e-06, + "loss": 0.809, + "step": 113 + }, + { + "epoch": 0.09442297073440088, + "grad_norm": 0.8101645708084106, + "learning_rate": 4.999952594932914e-06, + "loss": 0.8081, + "step": 114 + }, + { + "epoch": 0.09525124240750966, + "grad_norm": 0.905028760433197, + "learning_rate": 4.999945580943332e-06, + "loss": 0.8251, + "step": 115 + }, + { + "epoch": 0.09607951408061845, + "grad_norm": 0.8529418706893921, + "learning_rate": 4.999938083237585e-06, + "loss": 0.7984, + "step": 116 + }, + { + "epoch": 0.09690778575372722, + "grad_norm": 0.8047451376914978, + "learning_rate": 4.9999301018171225e-06, + "loss": 0.8222, + "step": 117 + }, + { + "epoch": 0.097736057426836, + "grad_norm": 0.8073400259017944, + "learning_rate": 4.99992163668349e-06, + "loss": 0.8251, + "step": 118 + }, + { + "epoch": 0.09856432909994478, + "grad_norm": 0.8470693826675415, + "learning_rate": 4.999912687838324e-06, + "loss": 0.8291, + "step": 119 + }, + { + "epoch": 0.09939260077305356, + "grad_norm": 0.8261650800704956, + "learning_rate": 4.999903255283357e-06, + "loss": 0.808, + "step": 120 + }, + { + "epoch": 0.10022087244616235, + "grad_norm": 0.8421704769134521, + "learning_rate": 4.999893339020414e-06, + "loss": 0.8295, + "step": 121 + }, + { + "epoch": 0.10104914411927111, + "grad_norm": 0.8244299292564392, + "learning_rate": 4.9998829390514134e-06, + "loss": 0.8215, + "step": 122 + }, + { + "epoch": 0.1018774157923799, + "grad_norm": 0.8365658521652222, + "learning_rate": 4.999872055378368e-06, + "loss": 0.8245, + "step": 123 + }, + { + "epoch": 0.10270568746548868, + "grad_norm": 0.9089561104774475, + "learning_rate": 4.999860688003384e-06, + "loss": 0.8125, + "step": 124 + }, + { + "epoch": 0.10353395913859746, + "grad_norm": 0.8784728646278381, + "learning_rate": 4.9998488369286595e-06, + "loss": 0.8321, + "step": 125 + }, + { + "epoch": 0.10436223081170624, + "grad_norm": 0.8596073389053345, + "learning_rate": 4.999836502156487e-06, + "loss": 0.8333, + "step": 126 + }, + { + "epoch": 0.10519050248481501, + "grad_norm": 0.8472688794136047, + "learning_rate": 4.9998236836892566e-06, + "loss": 0.8068, + "step": 127 + }, + { + "epoch": 0.1060187741579238, + "grad_norm": 0.8716503381729126, + "learning_rate": 4.999810381529447e-06, + "loss": 0.8005, + "step": 128 + }, + { + "epoch": 0.10684704583103258, + "grad_norm": 0.8416318893432617, + "learning_rate": 4.999796595679629e-06, + "loss": 0.7914, + "step": 129 + }, + { + "epoch": 0.10767531750414136, + "grad_norm": 0.838101863861084, + "learning_rate": 4.999782326142474e-06, + "loss": 0.7957, + "step": 130 + }, + { + "epoch": 0.10850358917725014, + "grad_norm": 0.8640785217285156, + "learning_rate": 4.999767572920741e-06, + "loss": 0.8077, + "step": 131 + }, + { + "epoch": 0.10933186085035891, + "grad_norm": 0.8207422494888306, + "learning_rate": 4.999752336017286e-06, + "loss": 0.7973, + "step": 132 + }, + { + "epoch": 0.1101601325234677, + "grad_norm": 0.8640758395195007, + "learning_rate": 4.9997366154350566e-06, + "loss": 0.8165, + "step": 133 + }, + { + "epoch": 0.11098840419657648, + "grad_norm": 0.7965807914733887, + "learning_rate": 4.999720411177094e-06, + "loss": 0.8059, + "step": 134 + }, + { + "epoch": 0.11181667586968526, + "grad_norm": 0.8463795185089111, + "learning_rate": 4.999703723246533e-06, + "loss": 0.8106, + "step": 135 + }, + { + "epoch": 0.11264494754279404, + "grad_norm": 0.8118059635162354, + "learning_rate": 4.999686551646604e-06, + "loss": 0.8099, + "step": 136 + }, + { + "epoch": 0.11347321921590281, + "grad_norm": 0.8256771564483643, + "learning_rate": 4.999668896380629e-06, + "loss": 0.8043, + "step": 137 + }, + { + "epoch": 0.11430149088901159, + "grad_norm": 0.8203126192092896, + "learning_rate": 4.999650757452025e-06, + "loss": 0.8029, + "step": 138 + }, + { + "epoch": 0.11512976256212037, + "grad_norm": 0.8442854881286621, + "learning_rate": 4.9996321348643e-06, + "loss": 0.8004, + "step": 139 + }, + { + "epoch": 0.11595803423522916, + "grad_norm": 0.8523190021514893, + "learning_rate": 4.999613028621059e-06, + "loss": 0.8092, + "step": 140 + }, + { + "epoch": 0.11678630590833794, + "grad_norm": 0.821428656578064, + "learning_rate": 4.999593438725997e-06, + "loss": 0.8105, + "step": 141 + }, + { + "epoch": 0.11761457758144671, + "grad_norm": 0.8406105041503906, + "learning_rate": 4.999573365182906e-06, + "loss": 0.7757, + "step": 142 + }, + { + "epoch": 0.11844284925455549, + "grad_norm": 0.8620600700378418, + "learning_rate": 4.999552807995669e-06, + "loss": 0.781, + "step": 143 + }, + { + "epoch": 0.11927112092766427, + "grad_norm": 0.920437216758728, + "learning_rate": 4.9995317671682655e-06, + "loss": 0.7579, + "step": 144 + }, + { + "epoch": 0.12009939260077306, + "grad_norm": 0.8414936661720276, + "learning_rate": 4.999510242704765e-06, + "loss": 0.8031, + "step": 145 + }, + { + "epoch": 0.12092766427388184, + "grad_norm": 0.8372176289558411, + "learning_rate": 4.999488234609332e-06, + "loss": 0.7931, + "step": 146 + }, + { + "epoch": 0.1217559359469906, + "grad_norm": 0.8785244226455688, + "learning_rate": 4.999465742886226e-06, + "loss": 0.8189, + "step": 147 + }, + { + "epoch": 0.12258420762009939, + "grad_norm": 0.8404353260993958, + "learning_rate": 4.999442767539799e-06, + "loss": 0.7899, + "step": 148 + }, + { + "epoch": 0.12341247929320817, + "grad_norm": 0.8541820049285889, + "learning_rate": 4.999419308574494e-06, + "loss": 0.7764, + "step": 149 + }, + { + "epoch": 0.12424075096631695, + "grad_norm": 0.8220177292823792, + "learning_rate": 4.999395365994854e-06, + "loss": 0.7942, + "step": 150 + }, + { + "epoch": 0.12506902263942574, + "grad_norm": 0.8823731541633606, + "learning_rate": 4.999370939805509e-06, + "loss": 0.793, + "step": 151 + }, + { + "epoch": 0.12589729431253452, + "grad_norm": 0.838860034942627, + "learning_rate": 4.999346030011186e-06, + "loss": 0.7741, + "step": 152 + }, + { + "epoch": 0.1267255659856433, + "grad_norm": 0.8162702322006226, + "learning_rate": 4.999320636616705e-06, + "loss": 0.7989, + "step": 153 + }, + { + "epoch": 0.12755383765875208, + "grad_norm": 0.8328465819358826, + "learning_rate": 4.99929475962698e-06, + "loss": 0.8055, + "step": 154 + }, + { + "epoch": 0.12838210933186084, + "grad_norm": 0.8373605608940125, + "learning_rate": 4.999268399047016e-06, + "loss": 0.801, + "step": 155 + }, + { + "epoch": 0.12921038100496962, + "grad_norm": 0.8408295512199402, + "learning_rate": 4.999241554881915e-06, + "loss": 0.7749, + "step": 156 + }, + { + "epoch": 0.1300386526780784, + "grad_norm": 0.8326342701911926, + "learning_rate": 4.999214227136871e-06, + "loss": 0.7818, + "step": 157 + }, + { + "epoch": 0.13086692435118719, + "grad_norm": 0.8064060211181641, + "learning_rate": 4.99918641581717e-06, + "loss": 0.7755, + "step": 158 + }, + { + "epoch": 0.13169519602429597, + "grad_norm": 0.8274610042572021, + "learning_rate": 4.999158120928196e-06, + "loss": 0.8008, + "step": 159 + }, + { + "epoch": 0.13252346769740475, + "grad_norm": 0.8121576309204102, + "learning_rate": 4.999129342475422e-06, + "loss": 0.7713, + "step": 160 + }, + { + "epoch": 0.13335173937051353, + "grad_norm": 0.8227710723876953, + "learning_rate": 4.9991000804644176e-06, + "loss": 0.7902, + "step": 161 + }, + { + "epoch": 0.13418001104362232, + "grad_norm": 0.8437026739120483, + "learning_rate": 4.999070334900843e-06, + "loss": 0.8176, + "step": 162 + }, + { + "epoch": 0.1350082827167311, + "grad_norm": 0.8379645943641663, + "learning_rate": 4.999040105790455e-06, + "loss": 0.7973, + "step": 163 + }, + { + "epoch": 0.13583655438983988, + "grad_norm": 0.822542130947113, + "learning_rate": 4.999009393139103e-06, + "loss": 0.7955, + "step": 164 + }, + { + "epoch": 0.13666482606294864, + "grad_norm": 0.8141928315162659, + "learning_rate": 4.9989781969527274e-06, + "loss": 0.7766, + "step": 165 + }, + { + "epoch": 0.13749309773605742, + "grad_norm": 0.8325410485267639, + "learning_rate": 4.998946517237367e-06, + "loss": 0.7737, + "step": 166 + }, + { + "epoch": 0.1383213694091662, + "grad_norm": 0.8399263024330139, + "learning_rate": 4.99891435399915e-06, + "loss": 0.7746, + "step": 167 + }, + { + "epoch": 0.13914964108227498, + "grad_norm": 0.8694303035736084, + "learning_rate": 4.9988817072443e-06, + "loss": 0.776, + "step": 168 + }, + { + "epoch": 0.13997791275538377, + "grad_norm": 0.8650498390197754, + "learning_rate": 4.998848576979135e-06, + "loss": 0.7941, + "step": 169 + }, + { + "epoch": 0.14080618442849255, + "grad_norm": 0.8246620893478394, + "learning_rate": 4.998814963210062e-06, + "loss": 0.7604, + "step": 170 + }, + { + "epoch": 0.14163445610160133, + "grad_norm": 0.8137409090995789, + "learning_rate": 4.99878086594359e-06, + "loss": 0.7681, + "step": 171 + }, + { + "epoch": 0.1424627277747101, + "grad_norm": 0.9755353927612305, + "learning_rate": 4.998746285186312e-06, + "loss": 0.7887, + "step": 172 + }, + { + "epoch": 0.1432909994478189, + "grad_norm": 0.8673408031463623, + "learning_rate": 4.998711220944921e-06, + "loss": 0.7803, + "step": 173 + }, + { + "epoch": 0.14411927112092768, + "grad_norm": 0.8312270045280457, + "learning_rate": 4.998675673226202e-06, + "loss": 0.7709, + "step": 174 + }, + { + "epoch": 0.14494754279403643, + "grad_norm": 0.8126094341278076, + "learning_rate": 4.998639642037032e-06, + "loss": 0.7822, + "step": 175 + }, + { + "epoch": 0.14577581446714521, + "grad_norm": 0.8561585545539856, + "learning_rate": 4.9986031273843834e-06, + "loss": 0.7829, + "step": 176 + }, + { + "epoch": 0.146604086140254, + "grad_norm": 0.8787612915039062, + "learning_rate": 4.9985661292753214e-06, + "loss": 0.7913, + "step": 177 + }, + { + "epoch": 0.14743235781336278, + "grad_norm": 0.8798345327377319, + "learning_rate": 4.998528647717006e-06, + "loss": 0.797, + "step": 178 + }, + { + "epoch": 0.14826062948647156, + "grad_norm": 0.7956589460372925, + "learning_rate": 4.998490682716687e-06, + "loss": 0.7571, + "step": 179 + }, + { + "epoch": 0.14908890115958034, + "grad_norm": 0.8505003452301025, + "learning_rate": 4.998452234281712e-06, + "loss": 0.7736, + "step": 180 + }, + { + "epoch": 0.14991717283268913, + "grad_norm": 0.8857074975967407, + "learning_rate": 4.998413302419519e-06, + "loss": 0.7863, + "step": 181 + }, + { + "epoch": 0.1507454445057979, + "grad_norm": 0.8762485980987549, + "learning_rate": 4.998373887137642e-06, + "loss": 0.7588, + "step": 182 + }, + { + "epoch": 0.1515737161789067, + "grad_norm": 0.8497205972671509, + "learning_rate": 4.998333988443709e-06, + "loss": 0.7773, + "step": 183 + }, + { + "epoch": 0.15240198785201547, + "grad_norm": 0.8720596432685852, + "learning_rate": 4.998293606345437e-06, + "loss": 0.7834, + "step": 184 + }, + { + "epoch": 0.15323025952512423, + "grad_norm": 0.8567108511924744, + "learning_rate": 4.998252740850641e-06, + "loss": 0.7794, + "step": 185 + }, + { + "epoch": 0.154058531198233, + "grad_norm": 0.8227219581604004, + "learning_rate": 4.998211391967228e-06, + "loss": 0.7688, + "step": 186 + }, + { + "epoch": 0.1548868028713418, + "grad_norm": 0.8321112394332886, + "learning_rate": 4.998169559703199e-06, + "loss": 0.7747, + "step": 187 + }, + { + "epoch": 0.15571507454445058, + "grad_norm": 0.8556115627288818, + "learning_rate": 4.9981272440666486e-06, + "loss": 0.7916, + "step": 188 + }, + { + "epoch": 0.15654334621755936, + "grad_norm": 0.8473951816558838, + "learning_rate": 4.998084445065763e-06, + "loss": 0.7823, + "step": 189 + }, + { + "epoch": 0.15737161789066814, + "grad_norm": 0.8551077246665955, + "learning_rate": 4.998041162708823e-06, + "loss": 0.7816, + "step": 190 + }, + { + "epoch": 0.15819988956377692, + "grad_norm": 0.945099413394928, + "learning_rate": 4.997997397004206e-06, + "loss": 0.7828, + "step": 191 + }, + { + "epoch": 0.1590281612368857, + "grad_norm": 0.8747237920761108, + "learning_rate": 4.9979531479603785e-06, + "loss": 0.7559, + "step": 192 + }, + { + "epoch": 0.1598564329099945, + "grad_norm": 0.8479958176612854, + "learning_rate": 4.997908415585902e-06, + "loss": 0.7694, + "step": 193 + }, + { + "epoch": 0.16068470458310327, + "grad_norm": 0.8659047484397888, + "learning_rate": 4.997863199889433e-06, + "loss": 0.7534, + "step": 194 + }, + { + "epoch": 0.16151297625621203, + "grad_norm": 0.8744727969169617, + "learning_rate": 4.997817500879719e-06, + "loss": 0.7499, + "step": 195 + }, + { + "epoch": 0.1623412479293208, + "grad_norm": 0.896468997001648, + "learning_rate": 4.997771318565603e-06, + "loss": 0.7974, + "step": 196 + }, + { + "epoch": 0.1631695196024296, + "grad_norm": 0.9058924913406372, + "learning_rate": 4.997724652956021e-06, + "loss": 0.7753, + "step": 197 + }, + { + "epoch": 0.16399779127553837, + "grad_norm": 0.8904663920402527, + "learning_rate": 4.9976775040600014e-06, + "loss": 0.7569, + "step": 198 + }, + { + "epoch": 0.16482606294864716, + "grad_norm": 0.8709238767623901, + "learning_rate": 4.997629871886669e-06, + "loss": 0.763, + "step": 199 + }, + { + "epoch": 0.16565433462175594, + "grad_norm": 0.8400478959083557, + "learning_rate": 4.997581756445238e-06, + "loss": 0.7795, + "step": 200 + }, + { + "epoch": 0.16648260629486472, + "grad_norm": 0.8458966612815857, + "learning_rate": 4.9975331577450204e-06, + "loss": 0.7597, + "step": 201 + }, + { + "epoch": 0.1673108779679735, + "grad_norm": 0.8707828521728516, + "learning_rate": 4.997484075795417e-06, + "loss": 0.759, + "step": 202 + }, + { + "epoch": 0.16813914964108229, + "grad_norm": 0.8774807453155518, + "learning_rate": 4.997434510605927e-06, + "loss": 0.7706, + "step": 203 + }, + { + "epoch": 0.16896742131419104, + "grad_norm": 0.8987929821014404, + "learning_rate": 4.99738446218614e-06, + "loss": 0.7319, + "step": 204 + }, + { + "epoch": 0.16979569298729982, + "grad_norm": 0.8869186639785767, + "learning_rate": 4.99733393054574e-06, + "loss": 0.7812, + "step": 205 + }, + { + "epoch": 0.1706239646604086, + "grad_norm": 0.860407829284668, + "learning_rate": 4.997282915694505e-06, + "loss": 0.7524, + "step": 206 + }, + { + "epoch": 0.1714522363335174, + "grad_norm": 0.8719788789749146, + "learning_rate": 4.9972314176423035e-06, + "loss": 0.7796, + "step": 207 + }, + { + "epoch": 0.17228050800662617, + "grad_norm": 0.8596137762069702, + "learning_rate": 4.997179436399103e-06, + "loss": 0.7554, + "step": 208 + }, + { + "epoch": 0.17310877967973495, + "grad_norm": 0.8505175709724426, + "learning_rate": 4.997126971974959e-06, + "loss": 0.7607, + "step": 209 + }, + { + "epoch": 0.17393705135284374, + "grad_norm": 0.7858585119247437, + "learning_rate": 4.9970740243800245e-06, + "loss": 0.7622, + "step": 210 + }, + { + "epoch": 0.17476532302595252, + "grad_norm": 0.8634621500968933, + "learning_rate": 4.997020593624543e-06, + "loss": 0.7659, + "step": 211 + }, + { + "epoch": 0.1755935946990613, + "grad_norm": 0.8650951385498047, + "learning_rate": 4.9969666797188545e-06, + "loss": 0.7689, + "step": 212 + }, + { + "epoch": 0.17642186637217008, + "grad_norm": 0.8602325916290283, + "learning_rate": 4.996912282673389e-06, + "loss": 0.7695, + "step": 213 + }, + { + "epoch": 0.17725013804527884, + "grad_norm": 0.8453742265701294, + "learning_rate": 4.9968574024986735e-06, + "loss": 0.7434, + "step": 214 + }, + { + "epoch": 0.17807840971838762, + "grad_norm": 0.8542691469192505, + "learning_rate": 4.9968020392053255e-06, + "loss": 0.7298, + "step": 215 + }, + { + "epoch": 0.1789066813914964, + "grad_norm": 0.8806673288345337, + "learning_rate": 4.996746192804058e-06, + "loss": 0.7649, + "step": 216 + }, + { + "epoch": 0.17973495306460519, + "grad_norm": 0.8451183438301086, + "learning_rate": 4.9966898633056765e-06, + "loss": 0.7681, + "step": 217 + }, + { + "epoch": 0.18056322473771397, + "grad_norm": 0.8664626479148865, + "learning_rate": 4.99663305072108e-06, + "loss": 0.7753, + "step": 218 + }, + { + "epoch": 0.18139149641082275, + "grad_norm": 0.8828144669532776, + "learning_rate": 4.996575755061262e-06, + "loss": 0.7514, + "step": 219 + }, + { + "epoch": 0.18221976808393153, + "grad_norm": 0.8493948578834534, + "learning_rate": 4.996517976337308e-06, + "loss": 0.7684, + "step": 220 + }, + { + "epoch": 0.18304803975704032, + "grad_norm": 0.8561757802963257, + "learning_rate": 4.9964597145603975e-06, + "loss": 0.7723, + "step": 221 + }, + { + "epoch": 0.1838763114301491, + "grad_norm": 0.8863773941993713, + "learning_rate": 4.996400969741805e-06, + "loss": 0.7732, + "step": 222 + }, + { + "epoch": 0.18470458310325788, + "grad_norm": 0.8571644425392151, + "learning_rate": 4.996341741892895e-06, + "loss": 0.744, + "step": 223 + }, + { + "epoch": 0.18553285477636663, + "grad_norm": 0.8622035980224609, + "learning_rate": 4.9962820310251286e-06, + "loss": 0.7607, + "step": 224 + }, + { + "epoch": 0.18636112644947542, + "grad_norm": 0.8904410600662231, + "learning_rate": 4.9962218371500595e-06, + "loss": 0.7648, + "step": 225 + }, + { + "epoch": 0.1871893981225842, + "grad_norm": 0.8975846171379089, + "learning_rate": 4.996161160279335e-06, + "loss": 0.7469, + "step": 226 + }, + { + "epoch": 0.18801766979569298, + "grad_norm": 0.8788724541664124, + "learning_rate": 4.996100000424694e-06, + "loss": 0.7473, + "step": 227 + }, + { + "epoch": 0.18884594146880176, + "grad_norm": 0.9446991086006165, + "learning_rate": 4.996038357597971e-06, + "loss": 0.7635, + "step": 228 + }, + { + "epoch": 0.18967421314191055, + "grad_norm": 0.86910480260849, + "learning_rate": 4.995976231811094e-06, + "loss": 0.7539, + "step": 229 + }, + { + "epoch": 0.19050248481501933, + "grad_norm": 0.862197756767273, + "learning_rate": 4.995913623076084e-06, + "loss": 0.7546, + "step": 230 + }, + { + "epoch": 0.1913307564881281, + "grad_norm": 0.906878650188446, + "learning_rate": 4.995850531405054e-06, + "loss": 0.7737, + "step": 231 + }, + { + "epoch": 0.1921590281612369, + "grad_norm": 0.8557183742523193, + "learning_rate": 4.995786956810212e-06, + "loss": 0.7663, + "step": 232 + }, + { + "epoch": 0.19298729983434568, + "grad_norm": 0.8125383853912354, + "learning_rate": 4.995722899303859e-06, + "loss": 0.7647, + "step": 233 + }, + { + "epoch": 0.19381557150745443, + "grad_norm": 0.8610332012176514, + "learning_rate": 4.995658358898391e-06, + "loss": 0.7541, + "step": 234 + }, + { + "epoch": 0.19464384318056321, + "grad_norm": 0.8620829582214355, + "learning_rate": 4.995593335606294e-06, + "loss": 0.7625, + "step": 235 + }, + { + "epoch": 0.195472114853672, + "grad_norm": 0.8296957612037659, + "learning_rate": 4.9955278294401496e-06, + "loss": 0.7536, + "step": 236 + }, + { + "epoch": 0.19630038652678078, + "grad_norm": 0.8974814414978027, + "learning_rate": 4.995461840412634e-06, + "loss": 0.7651, + "step": 237 + }, + { + "epoch": 0.19712865819988956, + "grad_norm": 0.8701095581054688, + "learning_rate": 4.995395368536515e-06, + "loss": 0.788, + "step": 238 + }, + { + "epoch": 0.19795692987299834, + "grad_norm": 0.8556636571884155, + "learning_rate": 4.995328413824653e-06, + "loss": 0.7649, + "step": 239 + }, + { + "epoch": 0.19878520154610713, + "grad_norm": 0.836686909198761, + "learning_rate": 4.995260976290005e-06, + "loss": 0.7583, + "step": 240 + }, + { + "epoch": 0.1996134732192159, + "grad_norm": 0.8807066082954407, + "learning_rate": 4.995193055945618e-06, + "loss": 0.7766, + "step": 241 + }, + { + "epoch": 0.2004417448923247, + "grad_norm": 0.8363582491874695, + "learning_rate": 4.995124652804635e-06, + "loss": 0.7655, + "step": 242 + }, + { + "epoch": 0.20127001656543347, + "grad_norm": 0.8699461817741394, + "learning_rate": 4.9950557668802905e-06, + "loss": 0.743, + "step": 243 + }, + { + "epoch": 0.20209828823854223, + "grad_norm": 0.8571060299873352, + "learning_rate": 4.994986398185914e-06, + "loss": 0.7512, + "step": 244 + }, + { + "epoch": 0.202926559911651, + "grad_norm": 0.8429675102233887, + "learning_rate": 4.994916546734927e-06, + "loss": 0.7589, + "step": 245 + }, + { + "epoch": 0.2037548315847598, + "grad_norm": 0.874822199344635, + "learning_rate": 4.994846212540846e-06, + "loss": 0.7534, + "step": 246 + }, + { + "epoch": 0.20458310325786858, + "grad_norm": 0.878341555595398, + "learning_rate": 4.994775395617279e-06, + "loss": 0.7709, + "step": 247 + }, + { + "epoch": 0.20541137493097736, + "grad_norm": 0.897961437702179, + "learning_rate": 4.99470409597793e-06, + "loss": 0.7768, + "step": 248 + }, + { + "epoch": 0.20623964660408614, + "grad_norm": 0.8439584374427795, + "learning_rate": 4.994632313636593e-06, + "loss": 0.7558, + "step": 249 + }, + { + "epoch": 0.20706791827719492, + "grad_norm": 0.8394468426704407, + "learning_rate": 4.994560048607159e-06, + "loss": 0.7492, + "step": 250 + }, + { + "epoch": 0.2078961899503037, + "grad_norm": 0.8489627242088318, + "learning_rate": 4.9944873009036074e-06, + "loss": 0.7365, + "step": 251 + }, + { + "epoch": 0.2087244616234125, + "grad_norm": 0.8675844669342041, + "learning_rate": 4.994414070540018e-06, + "loss": 0.762, + "step": 252 + }, + { + "epoch": 0.20955273329652127, + "grad_norm": 0.8228297829627991, + "learning_rate": 4.994340357530558e-06, + "loss": 0.7475, + "step": 253 + }, + { + "epoch": 0.21038100496963003, + "grad_norm": 0.8794129490852356, + "learning_rate": 4.9942661618894915e-06, + "loss": 0.7631, + "step": 254 + }, + { + "epoch": 0.2112092766427388, + "grad_norm": 0.8616135716438293, + "learning_rate": 4.9941914836311735e-06, + "loss": 0.7749, + "step": 255 + }, + { + "epoch": 0.2120375483158476, + "grad_norm": 0.8470495939254761, + "learning_rate": 4.994116322770054e-06, + "loss": 0.7484, + "step": 256 + }, + { + "epoch": 0.21286581998895637, + "grad_norm": 0.9346716403961182, + "learning_rate": 4.9940406793206755e-06, + "loss": 0.7557, + "step": 257 + }, + { + "epoch": 0.21369409166206516, + "grad_norm": 0.8993160724639893, + "learning_rate": 4.993964553297676e-06, + "loss": 0.7422, + "step": 258 + }, + { + "epoch": 0.21452236333517394, + "grad_norm": 0.8257086277008057, + "learning_rate": 4.993887944715783e-06, + "loss": 0.7783, + "step": 259 + }, + { + "epoch": 0.21535063500828272, + "grad_norm": 0.8950703144073486, + "learning_rate": 4.993810853589819e-06, + "loss": 0.7311, + "step": 260 + }, + { + "epoch": 0.2161789066813915, + "grad_norm": 0.8733702898025513, + "learning_rate": 4.993733279934704e-06, + "loss": 0.7572, + "step": 261 + }, + { + "epoch": 0.21700717835450029, + "grad_norm": 0.8322939276695251, + "learning_rate": 4.993655223765444e-06, + "loss": 0.7632, + "step": 262 + }, + { + "epoch": 0.21783545002760907, + "grad_norm": 0.8666137456893921, + "learning_rate": 4.993576685097145e-06, + "loss": 0.7619, + "step": 263 + }, + { + "epoch": 0.21866372170071782, + "grad_norm": 0.8770918250083923, + "learning_rate": 4.993497663945002e-06, + "loss": 0.7538, + "step": 264 + }, + { + "epoch": 0.2194919933738266, + "grad_norm": 0.8940248489379883, + "learning_rate": 4.9934181603243045e-06, + "loss": 0.7512, + "step": 265 + }, + { + "epoch": 0.2203202650469354, + "grad_norm": 0.8979714512825012, + "learning_rate": 4.993338174250437e-06, + "loss": 0.7504, + "step": 266 + }, + { + "epoch": 0.22114853672004417, + "grad_norm": 0.8802001476287842, + "learning_rate": 4.993257705738876e-06, + "loss": 0.7501, + "step": 267 + }, + { + "epoch": 0.22197680839315295, + "grad_norm": 0.8734432458877563, + "learning_rate": 4.993176754805189e-06, + "loss": 0.7452, + "step": 268 + }, + { + "epoch": 0.22280508006626173, + "grad_norm": 0.9127711057662964, + "learning_rate": 4.993095321465042e-06, + "loss": 0.7404, + "step": 269 + }, + { + "epoch": 0.22363335173937052, + "grad_norm": 0.8533952832221985, + "learning_rate": 4.993013405734191e-06, + "loss": 0.7382, + "step": 270 + }, + { + "epoch": 0.2244616234124793, + "grad_norm": 0.8463877439498901, + "learning_rate": 4.9929310076284845e-06, + "loss": 0.7574, + "step": 271 + }, + { + "epoch": 0.22528989508558808, + "grad_norm": 0.8598008155822754, + "learning_rate": 4.992848127163868e-06, + "loss": 0.7624, + "step": 272 + }, + { + "epoch": 0.22611816675869686, + "grad_norm": 0.8759154081344604, + "learning_rate": 4.992764764356376e-06, + "loss": 0.7637, + "step": 273 + }, + { + "epoch": 0.22694643843180562, + "grad_norm": 0.8880444765090942, + "learning_rate": 4.992680919222141e-06, + "loss": 0.7514, + "step": 274 + }, + { + "epoch": 0.2277747101049144, + "grad_norm": 0.871366024017334, + "learning_rate": 4.9925965917773826e-06, + "loss": 0.7569, + "step": 275 + }, + { + "epoch": 0.22860298177802318, + "grad_norm": 0.862654983997345, + "learning_rate": 4.99251178203842e-06, + "loss": 0.7349, + "step": 276 + }, + { + "epoch": 0.22943125345113197, + "grad_norm": 0.8615298867225647, + "learning_rate": 4.992426490021662e-06, + "loss": 0.7438, + "step": 277 + }, + { + "epoch": 0.23025952512424075, + "grad_norm": 0.8735869526863098, + "learning_rate": 4.992340715743614e-06, + "loss": 0.7755, + "step": 278 + }, + { + "epoch": 0.23108779679734953, + "grad_norm": 0.9082722067832947, + "learning_rate": 4.9922544592208695e-06, + "loss": 0.7561, + "step": 279 + }, + { + "epoch": 0.23191606847045831, + "grad_norm": 0.8639047145843506, + "learning_rate": 4.992167720470119e-06, + "loss": 0.7493, + "step": 280 + }, + { + "epoch": 0.2327443401435671, + "grad_norm": 0.891640305519104, + "learning_rate": 4.9920804995081475e-06, + "loss": 0.75, + "step": 281 + }, + { + "epoch": 0.23357261181667588, + "grad_norm": 0.8880137205123901, + "learning_rate": 4.99199279635183e-06, + "loss": 0.7801, + "step": 282 + }, + { + "epoch": 0.23440088348978466, + "grad_norm": 0.8600473403930664, + "learning_rate": 4.991904611018137e-06, + "loss": 0.7477, + "step": 283 + }, + { + "epoch": 0.23522915516289342, + "grad_norm": 0.8685399889945984, + "learning_rate": 4.99181594352413e-06, + "loss": 0.741, + "step": 284 + }, + { + "epoch": 0.2360574268360022, + "grad_norm": 0.9006181955337524, + "learning_rate": 4.991726793886966e-06, + "loss": 0.7288, + "step": 285 + }, + { + "epoch": 0.23688569850911098, + "grad_norm": 0.863895833492279, + "learning_rate": 4.991637162123896e-06, + "loss": 0.7375, + "step": 286 + }, + { + "epoch": 0.23771397018221976, + "grad_norm": 0.8620381951332092, + "learning_rate": 4.9915470482522625e-06, + "loss": 0.7249, + "step": 287 + }, + { + "epoch": 0.23854224185532855, + "grad_norm": 0.9634964466094971, + "learning_rate": 4.991456452289499e-06, + "loss": 0.7313, + "step": 288 + }, + { + "epoch": 0.23937051352843733, + "grad_norm": 0.9154514670372009, + "learning_rate": 4.9913653742531385e-06, + "loss": 0.782, + "step": 289 + }, + { + "epoch": 0.2401987852015461, + "grad_norm": 0.9179158806800842, + "learning_rate": 4.991273814160802e-06, + "loss": 0.7388, + "step": 290 + }, + { + "epoch": 0.2410270568746549, + "grad_norm": 0.8609046936035156, + "learning_rate": 4.991181772030206e-06, + "loss": 0.7393, + "step": 291 + }, + { + "epoch": 0.24185532854776368, + "grad_norm": 0.8689826130867004, + "learning_rate": 4.991089247879159e-06, + "loss": 0.7302, + "step": 292 + }, + { + "epoch": 0.24268360022087246, + "grad_norm": 0.8414047956466675, + "learning_rate": 4.990996241725565e-06, + "loss": 0.7496, + "step": 293 + }, + { + "epoch": 0.2435118718939812, + "grad_norm": 0.8621289134025574, + "learning_rate": 4.990902753587418e-06, + "loss": 0.7317, + "step": 294 + }, + { + "epoch": 0.24434014356709, + "grad_norm": 0.856566309928894, + "learning_rate": 4.990808783482809e-06, + "loss": 0.7169, + "step": 295 + }, + { + "epoch": 0.24516841524019878, + "grad_norm": 0.8549672961235046, + "learning_rate": 4.990714331429919e-06, + "loss": 0.7357, + "step": 296 + }, + { + "epoch": 0.24599668691330756, + "grad_norm": 0.8813024759292603, + "learning_rate": 4.990619397447024e-06, + "loss": 0.7437, + "step": 297 + }, + { + "epoch": 0.24682495858641634, + "grad_norm": 0.9008268713951111, + "learning_rate": 4.990523981552493e-06, + "loss": 0.7652, + "step": 298 + }, + { + "epoch": 0.24765323025952513, + "grad_norm": 0.910277783870697, + "learning_rate": 4.990428083764788e-06, + "loss": 0.7598, + "step": 299 + }, + { + "epoch": 0.2484815019326339, + "grad_norm": 0.9261599183082581, + "learning_rate": 4.990331704102464e-06, + "loss": 0.7199, + "step": 300 + }, + { + "epoch": 0.2493097736057427, + "grad_norm": 0.9166674613952637, + "learning_rate": 4.99023484258417e-06, + "loss": 0.7414, + "step": 301 + }, + { + "epoch": 0.2501380452788515, + "grad_norm": 0.8701214790344238, + "learning_rate": 4.990137499228648e-06, + "loss": 0.7326, + "step": 302 + }, + { + "epoch": 0.25096631695196026, + "grad_norm": 0.9120466113090515, + "learning_rate": 4.990039674054732e-06, + "loss": 0.7574, + "step": 303 + }, + { + "epoch": 0.25179458862506904, + "grad_norm": 0.9449286460876465, + "learning_rate": 4.989941367081351e-06, + "loss": 0.725, + "step": 304 + }, + { + "epoch": 0.2526228602981778, + "grad_norm": 0.8989110589027405, + "learning_rate": 4.989842578327526e-06, + "loss": 0.7569, + "step": 305 + }, + { + "epoch": 0.2534511319712866, + "grad_norm": 0.891677975654602, + "learning_rate": 4.989743307812373e-06, + "loss": 0.7135, + "step": 306 + }, + { + "epoch": 0.2542794036443954, + "grad_norm": 0.8689535856246948, + "learning_rate": 4.989643555555099e-06, + "loss": 0.7439, + "step": 307 + }, + { + "epoch": 0.25510767531750417, + "grad_norm": 0.8779630064964294, + "learning_rate": 4.989543321575004e-06, + "loss": 0.7322, + "step": 308 + }, + { + "epoch": 0.2559359469906129, + "grad_norm": 0.8977622985839844, + "learning_rate": 4.9894426058914856e-06, + "loss": 0.7544, + "step": 309 + }, + { + "epoch": 0.2567642186637217, + "grad_norm": 0.8216757774353027, + "learning_rate": 4.989341408524028e-06, + "loss": 0.7344, + "step": 310 + }, + { + "epoch": 0.25759249033683046, + "grad_norm": 0.8494759798049927, + "learning_rate": 4.9892397294922126e-06, + "loss": 0.7389, + "step": 311 + }, + { + "epoch": 0.25842076200993924, + "grad_norm": 0.881244957447052, + "learning_rate": 4.989137568815715e-06, + "loss": 0.7121, + "step": 312 + }, + { + "epoch": 0.259249033683048, + "grad_norm": 0.8679410219192505, + "learning_rate": 4.9890349265143005e-06, + "loss": 0.7469, + "step": 313 + }, + { + "epoch": 0.2600773053561568, + "grad_norm": 0.9746971130371094, + "learning_rate": 4.988931802607831e-06, + "loss": 0.7585, + "step": 314 + }, + { + "epoch": 0.2609055770292656, + "grad_norm": 0.9222471117973328, + "learning_rate": 4.988828197116259e-06, + "loss": 0.761, + "step": 315 + }, + { + "epoch": 0.26173384870237437, + "grad_norm": 0.9104124903678894, + "learning_rate": 4.988724110059631e-06, + "loss": 0.7652, + "step": 316 + }, + { + "epoch": 0.26256212037548315, + "grad_norm": 0.8876475095748901, + "learning_rate": 4.988619541458088e-06, + "loss": 0.7529, + "step": 317 + }, + { + "epoch": 0.26339039204859194, + "grad_norm": 0.8705421686172485, + "learning_rate": 4.988514491331861e-06, + "loss": 0.7437, + "step": 318 + }, + { + "epoch": 0.2642186637217007, + "grad_norm": 0.8594399690628052, + "learning_rate": 4.9884089597012784e-06, + "loss": 0.7406, + "step": 319 + }, + { + "epoch": 0.2650469353948095, + "grad_norm": 0.8807039856910706, + "learning_rate": 4.988302946586759e-06, + "loss": 0.7429, + "step": 320 + }, + { + "epoch": 0.2658752070679183, + "grad_norm": 0.8881412148475647, + "learning_rate": 4.9881964520088144e-06, + "loss": 0.7234, + "step": 321 + }, + { + "epoch": 0.26670347874102707, + "grad_norm": 0.8764941096305847, + "learning_rate": 4.988089475988052e-06, + "loss": 0.7514, + "step": 322 + }, + { + "epoch": 0.26753175041413585, + "grad_norm": 0.8992927670478821, + "learning_rate": 4.987982018545169e-06, + "loss": 0.7143, + "step": 323 + }, + { + "epoch": 0.26836002208724463, + "grad_norm": 0.8961546421051025, + "learning_rate": 4.9878740797009585e-06, + "loss": 0.7396, + "step": 324 + }, + { + "epoch": 0.2691882937603534, + "grad_norm": 0.9031578302383423, + "learning_rate": 4.987765659476304e-06, + "loss": 0.7407, + "step": 325 + }, + { + "epoch": 0.2700165654334622, + "grad_norm": 0.8693655729293823, + "learning_rate": 4.987656757892186e-06, + "loss": 0.7215, + "step": 326 + }, + { + "epoch": 0.270844837106571, + "grad_norm": 0.8739856481552124, + "learning_rate": 4.987547374969676e-06, + "loss": 0.7471, + "step": 327 + }, + { + "epoch": 0.27167310877967976, + "grad_norm": 0.8916322588920593, + "learning_rate": 4.987437510729936e-06, + "loss": 0.7459, + "step": 328 + }, + { + "epoch": 0.2725013804527885, + "grad_norm": 0.890178382396698, + "learning_rate": 4.987327165194225e-06, + "loss": 0.7322, + "step": 329 + }, + { + "epoch": 0.27332965212589727, + "grad_norm": 0.8611598014831543, + "learning_rate": 4.987216338383896e-06, + "loss": 0.733, + "step": 330 + }, + { + "epoch": 0.27415792379900605, + "grad_norm": 0.87791907787323, + "learning_rate": 4.98710503032039e-06, + "loss": 0.7445, + "step": 331 + }, + { + "epoch": 0.27498619547211484, + "grad_norm": 0.963479220867157, + "learning_rate": 4.986993241025245e-06, + "loss": 0.7376, + "step": 332 + }, + { + "epoch": 0.2758144671452236, + "grad_norm": 0.9083762764930725, + "learning_rate": 4.986880970520092e-06, + "loss": 0.7258, + "step": 333 + }, + { + "epoch": 0.2766427388183324, + "grad_norm": 1.048991084098816, + "learning_rate": 4.9867682188266524e-06, + "loss": 0.7399, + "step": 334 + }, + { + "epoch": 0.2774710104914412, + "grad_norm": 0.9404491782188416, + "learning_rate": 4.986654985966745e-06, + "loss": 0.7133, + "step": 335 + }, + { + "epoch": 0.27829928216454997, + "grad_norm": 0.9168346524238586, + "learning_rate": 4.986541271962278e-06, + "loss": 0.74, + "step": 336 + }, + { + "epoch": 0.27912755383765875, + "grad_norm": 0.8716689944267273, + "learning_rate": 4.986427076835253e-06, + "loss": 0.7249, + "step": 337 + }, + { + "epoch": 0.27995582551076753, + "grad_norm": 0.9097291827201843, + "learning_rate": 4.986312400607767e-06, + "loss": 0.7151, + "step": 338 + }, + { + "epoch": 0.2807840971838763, + "grad_norm": 0.8621960282325745, + "learning_rate": 4.98619724330201e-06, + "loss": 0.7125, + "step": 339 + }, + { + "epoch": 0.2816123688569851, + "grad_norm": 0.9823791980743408, + "learning_rate": 4.986081604940261e-06, + "loss": 0.7191, + "step": 340 + }, + { + "epoch": 0.2824406405300939, + "grad_norm": 0.8999047875404358, + "learning_rate": 4.9859654855448966e-06, + "loss": 0.7175, + "step": 341 + }, + { + "epoch": 0.28326891220320266, + "grad_norm": 0.8690859079360962, + "learning_rate": 4.985848885138386e-06, + "loss": 0.7309, + "step": 342 + }, + { + "epoch": 0.28409718387631144, + "grad_norm": 0.889500617980957, + "learning_rate": 4.985731803743287e-06, + "loss": 0.7391, + "step": 343 + }, + { + "epoch": 0.2849254555494202, + "grad_norm": 0.8992016911506653, + "learning_rate": 4.985614241382257e-06, + "loss": 0.7437, + "step": 344 + }, + { + "epoch": 0.285753727222529, + "grad_norm": 0.9100068211555481, + "learning_rate": 4.985496198078041e-06, + "loss": 0.722, + "step": 345 + }, + { + "epoch": 0.2865819988956378, + "grad_norm": 0.9013107419013977, + "learning_rate": 4.985377673853481e-06, + "loss": 0.7434, + "step": 346 + }, + { + "epoch": 0.2874102705687466, + "grad_norm": 0.8794704675674438, + "learning_rate": 4.9852586687315094e-06, + "loss": 0.7387, + "step": 347 + }, + { + "epoch": 0.28823854224185536, + "grad_norm": 0.9288201928138733, + "learning_rate": 4.9851391827351526e-06, + "loss": 0.7165, + "step": 348 + }, + { + "epoch": 0.2890668139149641, + "grad_norm": 0.8735847473144531, + "learning_rate": 4.98501921588753e-06, + "loss": 0.7479, + "step": 349 + }, + { + "epoch": 0.28989508558807286, + "grad_norm": 0.890616238117218, + "learning_rate": 4.984898768211853e-06, + "loss": 0.76, + "step": 350 + }, + { + "epoch": 0.29072335726118165, + "grad_norm": 0.9181028604507446, + "learning_rate": 4.98477783973143e-06, + "loss": 0.7366, + "step": 351 + }, + { + "epoch": 0.29155162893429043, + "grad_norm": 0.9947429299354553, + "learning_rate": 4.984656430469657e-06, + "loss": 0.7199, + "step": 352 + }, + { + "epoch": 0.2923799006073992, + "grad_norm": 0.8730769157409668, + "learning_rate": 4.984534540450027e-06, + "loss": 0.7153, + "step": 353 + }, + { + "epoch": 0.293208172280508, + "grad_norm": 0.8675997853279114, + "learning_rate": 4.9844121696961235e-06, + "loss": 0.7196, + "step": 354 + }, + { + "epoch": 0.2940364439536168, + "grad_norm": 0.9558542966842651, + "learning_rate": 4.984289318231624e-06, + "loss": 0.7311, + "step": 355 + }, + { + "epoch": 0.29486471562672556, + "grad_norm": 0.8817659616470337, + "learning_rate": 4.9841659860803e-06, + "loss": 0.7367, + "step": 356 + }, + { + "epoch": 0.29569298729983434, + "grad_norm": 0.8679526448249817, + "learning_rate": 4.984042173266016e-06, + "loss": 0.719, + "step": 357 + }, + { + "epoch": 0.2965212589729431, + "grad_norm": 0.9102020859718323, + "learning_rate": 4.983917879812726e-06, + "loss": 0.7301, + "step": 358 + }, + { + "epoch": 0.2973495306460519, + "grad_norm": 0.8987215757369995, + "learning_rate": 4.983793105744482e-06, + "loss": 0.7557, + "step": 359 + }, + { + "epoch": 0.2981778023191607, + "grad_norm": 0.8779832720756531, + "learning_rate": 4.983667851085426e-06, + "loss": 0.7471, + "step": 360 + }, + { + "epoch": 0.29900607399226947, + "grad_norm": 0.8762708306312561, + "learning_rate": 4.983542115859792e-06, + "loss": 0.7676, + "step": 361 + }, + { + "epoch": 0.29983434566537825, + "grad_norm": 0.910353422164917, + "learning_rate": 4.98341590009191e-06, + "loss": 0.7285, + "step": 362 + }, + { + "epoch": 0.30066261733848704, + "grad_norm": 0.9385405778884888, + "learning_rate": 4.983289203806202e-06, + "loss": 0.7142, + "step": 363 + }, + { + "epoch": 0.3014908890115958, + "grad_norm": 1.016781210899353, + "learning_rate": 4.983162027027182e-06, + "loss": 0.7408, + "step": 364 + }, + { + "epoch": 0.3023191606847046, + "grad_norm": 0.8642508387565613, + "learning_rate": 4.983034369779457e-06, + "loss": 0.7397, + "step": 365 + }, + { + "epoch": 0.3031474323578134, + "grad_norm": 0.9469170570373535, + "learning_rate": 4.982906232087728e-06, + "loss": 0.7158, + "step": 366 + }, + { + "epoch": 0.30397570403092217, + "grad_norm": 0.8818212151527405, + "learning_rate": 4.982777613976789e-06, + "loss": 0.7408, + "step": 367 + }, + { + "epoch": 0.30480397570403095, + "grad_norm": 0.8639838099479675, + "learning_rate": 4.982648515471526e-06, + "loss": 0.7217, + "step": 368 + }, + { + "epoch": 0.3056322473771397, + "grad_norm": 0.923780083656311, + "learning_rate": 4.982518936596917e-06, + "loss": 0.7284, + "step": 369 + }, + { + "epoch": 0.30646051905024846, + "grad_norm": 0.8992834687232971, + "learning_rate": 4.982388877378037e-06, + "loss": 0.7395, + "step": 370 + }, + { + "epoch": 0.30728879072335724, + "grad_norm": 0.8671404123306274, + "learning_rate": 4.982258337840049e-06, + "loss": 0.7157, + "step": 371 + }, + { + "epoch": 0.308117062396466, + "grad_norm": 0.8689168691635132, + "learning_rate": 4.982127318008212e-06, + "loss": 0.7398, + "step": 372 + }, + { + "epoch": 0.3089453340695748, + "grad_norm": 0.8964621424674988, + "learning_rate": 4.9819958179078765e-06, + "loss": 0.7385, + "step": 373 + }, + { + "epoch": 0.3097736057426836, + "grad_norm": 0.8984093070030212, + "learning_rate": 4.981863837564488e-06, + "loss": 0.7253, + "step": 374 + }, + { + "epoch": 0.31060187741579237, + "grad_norm": 0.8776745200157166, + "learning_rate": 4.981731377003583e-06, + "loss": 0.7126, + "step": 375 + }, + { + "epoch": 0.31143014908890115, + "grad_norm": 0.8782477378845215, + "learning_rate": 4.98159843625079e-06, + "loss": 0.7164, + "step": 376 + }, + { + "epoch": 0.31225842076200994, + "grad_norm": 0.8993703722953796, + "learning_rate": 4.981465015331832e-06, + "loss": 0.7391, + "step": 377 + }, + { + "epoch": 0.3130866924351187, + "grad_norm": 0.8975916504859924, + "learning_rate": 4.981331114272527e-06, + "loss": 0.7385, + "step": 378 + }, + { + "epoch": 0.3139149641082275, + "grad_norm": 0.8726824522018433, + "learning_rate": 4.9811967330987816e-06, + "loss": 0.7426, + "step": 379 + }, + { + "epoch": 0.3147432357813363, + "grad_norm": 0.8741514682769775, + "learning_rate": 4.981061871836597e-06, + "loss": 0.7246, + "step": 380 + }, + { + "epoch": 0.31557150745444507, + "grad_norm": 0.8807122707366943, + "learning_rate": 4.98092653051207e-06, + "loss": 0.7364, + "step": 381 + }, + { + "epoch": 0.31639977912755385, + "grad_norm": 0.835006058216095, + "learning_rate": 4.980790709151385e-06, + "loss": 0.7248, + "step": 382 + }, + { + "epoch": 0.31722805080066263, + "grad_norm": 0.9473997354507446, + "learning_rate": 4.980654407780824e-06, + "loss": 0.7331, + "step": 383 + }, + { + "epoch": 0.3180563224737714, + "grad_norm": 0.8332815170288086, + "learning_rate": 4.980517626426758e-06, + "loss": 0.7299, + "step": 384 + }, + { + "epoch": 0.3188845941468802, + "grad_norm": 0.8536360859870911, + "learning_rate": 4.980380365115656e-06, + "loss": 0.7095, + "step": 385 + }, + { + "epoch": 0.319712865819989, + "grad_norm": 0.9488157629966736, + "learning_rate": 4.980242623874075e-06, + "loss": 0.7382, + "step": 386 + }, + { + "epoch": 0.32054113749309776, + "grad_norm": 0.8623067736625671, + "learning_rate": 4.980104402728666e-06, + "loss": 0.7187, + "step": 387 + }, + { + "epoch": 0.32136940916620654, + "grad_norm": 0.9110252857208252, + "learning_rate": 4.979965701706175e-06, + "loss": 0.7354, + "step": 388 + }, + { + "epoch": 0.32219768083931527, + "grad_norm": 0.8966642618179321, + "learning_rate": 4.979826520833439e-06, + "loss": 0.7251, + "step": 389 + }, + { + "epoch": 0.32302595251242405, + "grad_norm": 0.856674075126648, + "learning_rate": 4.979686860137387e-06, + "loss": 0.723, + "step": 390 + }, + { + "epoch": 0.32385422418553284, + "grad_norm": 0.8623953461647034, + "learning_rate": 4.979546719645043e-06, + "loss": 0.7224, + "step": 391 + }, + { + "epoch": 0.3246824958586416, + "grad_norm": 0.8685624599456787, + "learning_rate": 4.979406099383523e-06, + "loss": 0.7283, + "step": 392 + }, + { + "epoch": 0.3255107675317504, + "grad_norm": 0.8610525727272034, + "learning_rate": 4.979264999380035e-06, + "loss": 0.7385, + "step": 393 + }, + { + "epoch": 0.3263390392048592, + "grad_norm": 0.8860796093940735, + "learning_rate": 4.979123419661882e-06, + "loss": 0.743, + "step": 394 + }, + { + "epoch": 0.32716731087796797, + "grad_norm": 0.8748682737350464, + "learning_rate": 4.978981360256457e-06, + "loss": 0.7447, + "step": 395 + }, + { + "epoch": 0.32799558255107675, + "grad_norm": 0.8750934600830078, + "learning_rate": 4.978838821191247e-06, + "loss": 0.7259, + "step": 396 + }, + { + "epoch": 0.32882385422418553, + "grad_norm": 0.9335777163505554, + "learning_rate": 4.978695802493834e-06, + "loss": 0.7271, + "step": 397 + }, + { + "epoch": 0.3296521258972943, + "grad_norm": 0.8823036551475525, + "learning_rate": 4.978552304191888e-06, + "loss": 0.723, + "step": 398 + }, + { + "epoch": 0.3304803975704031, + "grad_norm": 0.9503542184829712, + "learning_rate": 4.9784083263131764e-06, + "loss": 0.752, + "step": 399 + }, + { + "epoch": 0.3313086692435119, + "grad_norm": 0.8779938220977783, + "learning_rate": 4.978263868885557e-06, + "loss": 0.7098, + "step": 400 + }, + { + "epoch": 0.33213694091662066, + "grad_norm": 0.873857855796814, + "learning_rate": 4.978118931936981e-06, + "loss": 0.7143, + "step": 401 + }, + { + "epoch": 0.33296521258972944, + "grad_norm": 0.8437919020652771, + "learning_rate": 4.9779735154954924e-06, + "loss": 0.7275, + "step": 402 + }, + { + "epoch": 0.3337934842628382, + "grad_norm": 0.8914555311203003, + "learning_rate": 4.977827619589229e-06, + "loss": 0.722, + "step": 403 + }, + { + "epoch": 0.334621755935947, + "grad_norm": 0.9062597751617432, + "learning_rate": 4.977681244246417e-06, + "loss": 0.7067, + "step": 404 + }, + { + "epoch": 0.3354500276090558, + "grad_norm": 0.8555092811584473, + "learning_rate": 4.977534389495382e-06, + "loss": 0.7318, + "step": 405 + }, + { + "epoch": 0.33627829928216457, + "grad_norm": 0.8717684745788574, + "learning_rate": 4.9773870553645386e-06, + "loss": 0.7398, + "step": 406 + }, + { + "epoch": 0.33710657095527335, + "grad_norm": 0.8746877908706665, + "learning_rate": 4.977239241882393e-06, + "loss": 0.7154, + "step": 407 + }, + { + "epoch": 0.3379348426283821, + "grad_norm": 0.8905707597732544, + "learning_rate": 4.977090949077547e-06, + "loss": 0.7345, + "step": 408 + }, + { + "epoch": 0.33876311430149086, + "grad_norm": 0.9037347435951233, + "learning_rate": 4.9769421769786935e-06, + "loss": 0.7271, + "step": 409 + }, + { + "epoch": 0.33959138597459965, + "grad_norm": 0.8588722944259644, + "learning_rate": 4.976792925614619e-06, + "loss": 0.7335, + "step": 410 + }, + { + "epoch": 0.34041965764770843, + "grad_norm": 0.8786173462867737, + "learning_rate": 4.9766431950142005e-06, + "loss": 0.7254, + "step": 411 + }, + { + "epoch": 0.3412479293208172, + "grad_norm": 0.9019079804420471, + "learning_rate": 4.976492985206412e-06, + "loss": 0.715, + "step": 412 + }, + { + "epoch": 0.342076200993926, + "grad_norm": 0.878132700920105, + "learning_rate": 4.976342296220316e-06, + "loss": 0.7063, + "step": 413 + }, + { + "epoch": 0.3429044726670348, + "grad_norm": 0.8722684979438782, + "learning_rate": 4.976191128085069e-06, + "loss": 0.7481, + "step": 414 + }, + { + "epoch": 0.34373274434014356, + "grad_norm": 0.9076220393180847, + "learning_rate": 4.976039480829922e-06, + "loss": 0.7197, + "step": 415 + }, + { + "epoch": 0.34456101601325234, + "grad_norm": 0.8674967288970947, + "learning_rate": 4.975887354484217e-06, + "loss": 0.7282, + "step": 416 + }, + { + "epoch": 0.3453892876863611, + "grad_norm": 0.9406816959381104, + "learning_rate": 4.9757347490773885e-06, + "loss": 0.7222, + "step": 417 + }, + { + "epoch": 0.3462175593594699, + "grad_norm": 0.8747051358222961, + "learning_rate": 4.975581664638965e-06, + "loss": 0.7283, + "step": 418 + }, + { + "epoch": 0.3470458310325787, + "grad_norm": 1.0587687492370605, + "learning_rate": 4.9754281011985666e-06, + "loss": 0.7272, + "step": 419 + }, + { + "epoch": 0.34787410270568747, + "grad_norm": 0.8596946597099304, + "learning_rate": 4.975274058785906e-06, + "loss": 0.7407, + "step": 420 + }, + { + "epoch": 0.34870237437879625, + "grad_norm": 0.9230688214302063, + "learning_rate": 4.975119537430789e-06, + "loss": 0.7303, + "step": 421 + }, + { + "epoch": 0.34953064605190504, + "grad_norm": 0.9276477694511414, + "learning_rate": 4.974964537163115e-06, + "loss": 0.7296, + "step": 422 + }, + { + "epoch": 0.3503589177250138, + "grad_norm": 0.8969418406486511, + "learning_rate": 4.974809058012874e-06, + "loss": 0.7275, + "step": 423 + }, + { + "epoch": 0.3511871893981226, + "grad_norm": 0.8454467058181763, + "learning_rate": 4.97465310001015e-06, + "loss": 0.7144, + "step": 424 + }, + { + "epoch": 0.3520154610712314, + "grad_norm": 0.8833656907081604, + "learning_rate": 4.97449666318512e-06, + "loss": 0.7073, + "step": 425 + }, + { + "epoch": 0.35284373274434017, + "grad_norm": 0.9251653552055359, + "learning_rate": 4.974339747568052e-06, + "loss": 0.7351, + "step": 426 + }, + { + "epoch": 0.35367200441744895, + "grad_norm": 0.8938575983047485, + "learning_rate": 4.974182353189309e-06, + "loss": 0.7031, + "step": 427 + }, + { + "epoch": 0.3545002760905577, + "grad_norm": 0.8487107753753662, + "learning_rate": 4.974024480079343e-06, + "loss": 0.704, + "step": 428 + }, + { + "epoch": 0.35532854776366646, + "grad_norm": 0.8515727519989014, + "learning_rate": 4.973866128268705e-06, + "loss": 0.7092, + "step": 429 + }, + { + "epoch": 0.35615681943677524, + "grad_norm": 0.8869542479515076, + "learning_rate": 4.97370729778803e-06, + "loss": 0.7099, + "step": 430 + }, + { + "epoch": 0.356985091109884, + "grad_norm": 0.8830757141113281, + "learning_rate": 4.973547988668055e-06, + "loss": 0.7294, + "step": 431 + }, + { + "epoch": 0.3578133627829928, + "grad_norm": 0.9321587681770325, + "learning_rate": 4.973388200939601e-06, + "loss": 0.7276, + "step": 432 + }, + { + "epoch": 0.3586416344561016, + "grad_norm": 0.9408143758773804, + "learning_rate": 4.9732279346335866e-06, + "loss": 0.7216, + "step": 433 + }, + { + "epoch": 0.35946990612921037, + "grad_norm": 0.9189308285713196, + "learning_rate": 4.973067189781022e-06, + "loss": 0.7079, + "step": 434 + }, + { + "epoch": 0.36029817780231915, + "grad_norm": 0.9198120832443237, + "learning_rate": 4.97290596641301e-06, + "loss": 0.7188, + "step": 435 + }, + { + "epoch": 0.36112644947542794, + "grad_norm": 0.9123285412788391, + "learning_rate": 4.972744264560746e-06, + "loss": 0.7328, + "step": 436 + }, + { + "epoch": 0.3619547211485367, + "grad_norm": 0.8722008466720581, + "learning_rate": 4.972582084255517e-06, + "loss": 0.7228, + "step": 437 + }, + { + "epoch": 0.3627829928216455, + "grad_norm": 1.035325288772583, + "learning_rate": 4.972419425528704e-06, + "loss": 0.72, + "step": 438 + }, + { + "epoch": 0.3636112644947543, + "grad_norm": 0.9577400088310242, + "learning_rate": 4.97225628841178e-06, + "loss": 0.7382, + "step": 439 + }, + { + "epoch": 0.36443953616786307, + "grad_norm": 0.9206854104995728, + "learning_rate": 4.972092672936309e-06, + "loss": 0.7294, + "step": 440 + }, + { + "epoch": 0.36526780784097185, + "grad_norm": 0.9281194806098938, + "learning_rate": 4.971928579133952e-06, + "loss": 0.7137, + "step": 441 + }, + { + "epoch": 0.36609607951408063, + "grad_norm": 0.951049268245697, + "learning_rate": 4.971764007036457e-06, + "loss": 0.7354, + "step": 442 + }, + { + "epoch": 0.3669243511871894, + "grad_norm": 0.8779566287994385, + "learning_rate": 4.971598956675668e-06, + "loss": 0.6858, + "step": 443 + }, + { + "epoch": 0.3677526228602982, + "grad_norm": 0.9221998453140259, + "learning_rate": 4.971433428083522e-06, + "loss": 0.7116, + "step": 444 + }, + { + "epoch": 0.368580894533407, + "grad_norm": 0.9608840346336365, + "learning_rate": 4.971267421292045e-06, + "loss": 0.7129, + "step": 445 + }, + { + "epoch": 0.36940916620651576, + "grad_norm": 0.8761363625526428, + "learning_rate": 4.971100936333359e-06, + "loss": 0.718, + "step": 446 + }, + { + "epoch": 0.37023743787962454, + "grad_norm": 0.8808798789978027, + "learning_rate": 4.970933973239677e-06, + "loss": 0.6986, + "step": 447 + }, + { + "epoch": 0.37106570955273327, + "grad_norm": 0.8818857669830322, + "learning_rate": 4.9707665320433055e-06, + "loss": 0.7465, + "step": 448 + }, + { + "epoch": 0.37189398122584205, + "grad_norm": 0.9144098162651062, + "learning_rate": 4.970598612776641e-06, + "loss": 0.7187, + "step": 449 + }, + { + "epoch": 0.37272225289895083, + "grad_norm": 0.8439158797264099, + "learning_rate": 4.9704302154721776e-06, + "loss": 0.7195, + "step": 450 + }, + { + "epoch": 0.3735505245720596, + "grad_norm": 0.8661074638366699, + "learning_rate": 4.970261340162495e-06, + "loss": 0.712, + "step": 451 + }, + { + "epoch": 0.3743787962451684, + "grad_norm": 0.8818789124488831, + "learning_rate": 4.970091986880271e-06, + "loss": 0.7199, + "step": 452 + }, + { + "epoch": 0.3752070679182772, + "grad_norm": 0.9560503363609314, + "learning_rate": 4.969922155658273e-06, + "loss": 0.7145, + "step": 453 + }, + { + "epoch": 0.37603533959138596, + "grad_norm": 0.8723775744438171, + "learning_rate": 4.969751846529363e-06, + "loss": 0.7152, + "step": 454 + }, + { + "epoch": 0.37686361126449475, + "grad_norm": 0.8612253069877625, + "learning_rate": 4.9695810595264935e-06, + "loss": 0.728, + "step": 455 + }, + { + "epoch": 0.37769188293760353, + "grad_norm": 0.8995444178581238, + "learning_rate": 4.969409794682709e-06, + "loss": 0.678, + "step": 456 + }, + { + "epoch": 0.3785201546107123, + "grad_norm": 0.8770595788955688, + "learning_rate": 4.96923805203115e-06, + "loss": 0.7354, + "step": 457 + }, + { + "epoch": 0.3793484262838211, + "grad_norm": 0.893329918384552, + "learning_rate": 4.969065831605045e-06, + "loss": 0.7102, + "step": 458 + }, + { + "epoch": 0.3801766979569299, + "grad_norm": 0.954058051109314, + "learning_rate": 4.968893133437718e-06, + "loss": 0.7144, + "step": 459 + }, + { + "epoch": 0.38100496963003866, + "grad_norm": 0.9711002707481384, + "learning_rate": 4.968719957562585e-06, + "loss": 0.7452, + "step": 460 + }, + { + "epoch": 0.38183324130314744, + "grad_norm": 0.9036396741867065, + "learning_rate": 4.968546304013153e-06, + "loss": 0.7535, + "step": 461 + }, + { + "epoch": 0.3826615129762562, + "grad_norm": 0.8948055505752563, + "learning_rate": 4.968372172823023e-06, + "loss": 0.7284, + "step": 462 + }, + { + "epoch": 0.383489784649365, + "grad_norm": 0.9134189486503601, + "learning_rate": 4.968197564025886e-06, + "loss": 0.723, + "step": 463 + }, + { + "epoch": 0.3843180563224738, + "grad_norm": 0.9191608428955078, + "learning_rate": 4.9680224776555305e-06, + "loss": 0.7078, + "step": 464 + }, + { + "epoch": 0.38514632799558257, + "grad_norm": 0.879683792591095, + "learning_rate": 4.9678469137458315e-06, + "loss": 0.6805, + "step": 465 + }, + { + "epoch": 0.38597459966869135, + "grad_norm": 0.8887293338775635, + "learning_rate": 4.96767087233076e-06, + "loss": 0.7296, + "step": 466 + }, + { + "epoch": 0.38680287134180014, + "grad_norm": 0.889991044998169, + "learning_rate": 4.967494353444377e-06, + "loss": 0.702, + "step": 467 + }, + { + "epoch": 0.38763114301490886, + "grad_norm": 0.9667506814002991, + "learning_rate": 4.967317357120839e-06, + "loss": 0.7305, + "step": 468 + }, + { + "epoch": 0.38845941468801765, + "grad_norm": 0.9297274351119995, + "learning_rate": 4.967139883394392e-06, + "loss": 0.7344, + "step": 469 + }, + { + "epoch": 0.38928768636112643, + "grad_norm": 0.9167365431785583, + "learning_rate": 4.966961932299377e-06, + "loss": 0.7016, + "step": 470 + }, + { + "epoch": 0.3901159580342352, + "grad_norm": 0.877378523349762, + "learning_rate": 4.966783503870224e-06, + "loss": 0.7313, + "step": 471 + }, + { + "epoch": 0.390944229707344, + "grad_norm": 0.9625933766365051, + "learning_rate": 4.966604598141458e-06, + "loss": 0.7245, + "step": 472 + }, + { + "epoch": 0.3917725013804528, + "grad_norm": 0.8859156966209412, + "learning_rate": 4.9664252151476955e-06, + "loss": 0.7094, + "step": 473 + }, + { + "epoch": 0.39260077305356156, + "grad_norm": 0.9172984957695007, + "learning_rate": 4.966245354923645e-06, + "loss": 0.6966, + "step": 474 + }, + { + "epoch": 0.39342904472667034, + "grad_norm": 0.9040864109992981, + "learning_rate": 4.96606501750411e-06, + "loss": 0.722, + "step": 475 + }, + { + "epoch": 0.3942573163997791, + "grad_norm": 0.8900951147079468, + "learning_rate": 4.965884202923982e-06, + "loss": 0.7263, + "step": 476 + }, + { + "epoch": 0.3950855880728879, + "grad_norm": 0.9096625447273254, + "learning_rate": 4.965702911218247e-06, + "loss": 0.702, + "step": 477 + }, + { + "epoch": 0.3959138597459967, + "grad_norm": 0.8801544904708862, + "learning_rate": 4.965521142421984e-06, + "loss": 0.7113, + "step": 478 + }, + { + "epoch": 0.39674213141910547, + "grad_norm": 0.9054175615310669, + "learning_rate": 4.965338896570363e-06, + "loss": 0.7135, + "step": 479 + }, + { + "epoch": 0.39757040309221425, + "grad_norm": 0.8734445571899414, + "learning_rate": 4.965156173698648e-06, + "loss": 0.7181, + "step": 480 + }, + { + "epoch": 0.39839867476532304, + "grad_norm": 0.9113215208053589, + "learning_rate": 4.964972973842191e-06, + "loss": 0.7008, + "step": 481 + }, + { + "epoch": 0.3992269464384318, + "grad_norm": 0.8941736221313477, + "learning_rate": 4.964789297036444e-06, + "loss": 0.7201, + "step": 482 + }, + { + "epoch": 0.4000552181115406, + "grad_norm": 0.8555464148521423, + "learning_rate": 4.964605143316944e-06, + "loss": 0.7173, + "step": 483 + }, + { + "epoch": 0.4008834897846494, + "grad_norm": 0.8719691634178162, + "learning_rate": 4.9644205127193225e-06, + "loss": 0.7211, + "step": 484 + }, + { + "epoch": 0.40171176145775817, + "grad_norm": 0.901626467704773, + "learning_rate": 4.964235405279306e-06, + "loss": 0.7223, + "step": 485 + }, + { + "epoch": 0.40254003313086695, + "grad_norm": 0.856769859790802, + "learning_rate": 4.9640498210327095e-06, + "loss": 0.7103, + "step": 486 + }, + { + "epoch": 0.40336830480397573, + "grad_norm": 0.9053338766098022, + "learning_rate": 4.963863760015442e-06, + "loss": 0.7301, + "step": 487 + }, + { + "epoch": 0.40419657647708446, + "grad_norm": 0.8989829421043396, + "learning_rate": 4.963677222263505e-06, + "loss": 0.7391, + "step": 488 + }, + { + "epoch": 0.40502484815019324, + "grad_norm": 0.9187886118888855, + "learning_rate": 4.963490207812991e-06, + "loss": 0.7179, + "step": 489 + }, + { + "epoch": 0.405853119823302, + "grad_norm": 0.8712778687477112, + "learning_rate": 4.963302716700087e-06, + "loss": 0.7088, + "step": 490 + }, + { + "epoch": 0.4066813914964108, + "grad_norm": 1.0485447645187378, + "learning_rate": 4.9631147489610695e-06, + "loss": 0.7068, + "step": 491 + }, + { + "epoch": 0.4075096631695196, + "grad_norm": 0.9332602024078369, + "learning_rate": 4.962926304632309e-06, + "loss": 0.7201, + "step": 492 + }, + { + "epoch": 0.40833793484262837, + "grad_norm": 0.9264400601387024, + "learning_rate": 4.962737383750268e-06, + "loss": 0.7136, + "step": 493 + }, + { + "epoch": 0.40916620651573715, + "grad_norm": 0.8812651634216309, + "learning_rate": 4.962547986351501e-06, + "loss": 0.7315, + "step": 494 + }, + { + "epoch": 0.40999447818884593, + "grad_norm": 0.8905262351036072, + "learning_rate": 4.962358112472653e-06, + "loss": 0.7127, + "step": 495 + }, + { + "epoch": 0.4108227498619547, + "grad_norm": 0.887514591217041, + "learning_rate": 4.962167762150465e-06, + "loss": 0.7602, + "step": 496 + }, + { + "epoch": 0.4116510215350635, + "grad_norm": 0.8683431148529053, + "learning_rate": 4.9619769354217665e-06, + "loss": 0.7091, + "step": 497 + }, + { + "epoch": 0.4124792932081723, + "grad_norm": 0.944595217704773, + "learning_rate": 4.961785632323481e-06, + "loss": 0.708, + "step": 498 + }, + { + "epoch": 0.41330756488128106, + "grad_norm": 0.8834611773490906, + "learning_rate": 4.961593852892625e-06, + "loss": 0.7308, + "step": 499 + }, + { + "epoch": 0.41413583655438985, + "grad_norm": 0.8479524850845337, + "learning_rate": 4.961401597166304e-06, + "loss": 0.7061, + "step": 500 + }, + { + "epoch": 0.41496410822749863, + "grad_norm": 0.9092981219291687, + "learning_rate": 4.96120886518172e-06, + "loss": 0.7176, + "step": 501 + }, + { + "epoch": 0.4157923799006074, + "grad_norm": 0.9024932384490967, + "learning_rate": 4.961015656976163e-06, + "loss": 0.6944, + "step": 502 + }, + { + "epoch": 0.4166206515737162, + "grad_norm": 0.8945576548576355, + "learning_rate": 4.960821972587018e-06, + "loss": 0.6958, + "step": 503 + }, + { + "epoch": 0.417448923246825, + "grad_norm": 0.8668336868286133, + "learning_rate": 4.9606278120517606e-06, + "loss": 0.7219, + "step": 504 + }, + { + "epoch": 0.41827719491993376, + "grad_norm": 0.8831227421760559, + "learning_rate": 4.96043317540796e-06, + "loss": 0.7124, + "step": 505 + }, + { + "epoch": 0.41910546659304254, + "grad_norm": 0.9637073874473572, + "learning_rate": 4.960238062693274e-06, + "loss": 0.7337, + "step": 506 + }, + { + "epoch": 0.4199337382661513, + "grad_norm": 0.8895899653434753, + "learning_rate": 4.960042473945459e-06, + "loss": 0.7459, + "step": 507 + }, + { + "epoch": 0.42076200993926005, + "grad_norm": 0.8762213587760925, + "learning_rate": 4.959846409202356e-06, + "loss": 0.7319, + "step": 508 + }, + { + "epoch": 0.42159028161236883, + "grad_norm": 0.9177393317222595, + "learning_rate": 4.959649868501903e-06, + "loss": 0.7114, + "step": 509 + }, + { + "epoch": 0.4224185532854776, + "grad_norm": 0.88541179895401, + "learning_rate": 4.9594528518821305e-06, + "loss": 0.716, + "step": 510 + }, + { + "epoch": 0.4232468249585864, + "grad_norm": 0.865566611289978, + "learning_rate": 4.959255359381158e-06, + "loss": 0.7098, + "step": 511 + }, + { + "epoch": 0.4240750966316952, + "grad_norm": 0.8629707098007202, + "learning_rate": 4.959057391037197e-06, + "loss": 0.7146, + "step": 512 + }, + { + "epoch": 0.42490336830480396, + "grad_norm": 0.8642273545265198, + "learning_rate": 4.958858946888554e-06, + "loss": 0.7345, + "step": 513 + }, + { + "epoch": 0.42573163997791275, + "grad_norm": 0.9157605171203613, + "learning_rate": 4.958660026973626e-06, + "loss": 0.7177, + "step": 514 + }, + { + "epoch": 0.42655991165102153, + "grad_norm": 0.9108191728591919, + "learning_rate": 4.958460631330903e-06, + "loss": 0.7317, + "step": 515 + }, + { + "epoch": 0.4273881833241303, + "grad_norm": 0.8524913191795349, + "learning_rate": 4.958260759998965e-06, + "loss": 0.7193, + "step": 516 + }, + { + "epoch": 0.4282164549972391, + "grad_norm": 0.9197820425033569, + "learning_rate": 4.958060413016485e-06, + "loss": 0.7127, + "step": 517 + }, + { + "epoch": 0.4290447266703479, + "grad_norm": 0.8744186758995056, + "learning_rate": 4.9578595904222296e-06, + "loss": 0.6968, + "step": 518 + }, + { + "epoch": 0.42987299834345666, + "grad_norm": 0.9071037769317627, + "learning_rate": 4.957658292255055e-06, + "loss": 0.6963, + "step": 519 + }, + { + "epoch": 0.43070127001656544, + "grad_norm": 0.937796950340271, + "learning_rate": 4.95745651855391e-06, + "loss": 0.7193, + "step": 520 + }, + { + "epoch": 0.4315295416896742, + "grad_norm": 0.883471667766571, + "learning_rate": 4.9572542693578375e-06, + "loss": 0.6886, + "step": 521 + }, + { + "epoch": 0.432357813362783, + "grad_norm": 0.8738589882850647, + "learning_rate": 4.9570515447059705e-06, + "loss": 0.7072, + "step": 522 + }, + { + "epoch": 0.4331860850358918, + "grad_norm": 0.9039497971534729, + "learning_rate": 4.9568483446375335e-06, + "loss": 0.7251, + "step": 523 + }, + { + "epoch": 0.43401435670900057, + "grad_norm": 0.896789014339447, + "learning_rate": 4.956644669191845e-06, + "loss": 0.7021, + "step": 524 + }, + { + "epoch": 0.43484262838210935, + "grad_norm": 0.8883570432662964, + "learning_rate": 4.9564405184083126e-06, + "loss": 0.7153, + "step": 525 + }, + { + "epoch": 0.43567090005521814, + "grad_norm": 0.8869197368621826, + "learning_rate": 4.956235892326438e-06, + "loss": 0.7061, + "step": 526 + }, + { + "epoch": 0.4364991717283269, + "grad_norm": 0.9237920641899109, + "learning_rate": 4.9560307909858165e-06, + "loss": 0.6988, + "step": 527 + }, + { + "epoch": 0.43732744340143564, + "grad_norm": 0.9359283447265625, + "learning_rate": 4.955825214426131e-06, + "loss": 0.738, + "step": 528 + }, + { + "epoch": 0.4381557150745444, + "grad_norm": 0.8602650761604309, + "learning_rate": 4.95561916268716e-06, + "loss": 0.709, + "step": 529 + }, + { + "epoch": 0.4389839867476532, + "grad_norm": 0.9013513922691345, + "learning_rate": 4.955412635808771e-06, + "loss": 0.6836, + "step": 530 + }, + { + "epoch": 0.439812258420762, + "grad_norm": 0.9136670827865601, + "learning_rate": 4.955205633830927e-06, + "loss": 0.7263, + "step": 531 + }, + { + "epoch": 0.4406405300938708, + "grad_norm": 0.9013793468475342, + "learning_rate": 4.95499815679368e-06, + "loss": 0.6937, + "step": 532 + }, + { + "epoch": 0.44146880176697956, + "grad_norm": 0.9017112851142883, + "learning_rate": 4.954790204737174e-06, + "loss": 0.6913, + "step": 533 + }, + { + "epoch": 0.44229707344008834, + "grad_norm": 0.8901436924934387, + "learning_rate": 4.954581777701647e-06, + "loss": 0.7241, + "step": 534 + }, + { + "epoch": 0.4431253451131971, + "grad_norm": 0.8882759809494019, + "learning_rate": 4.954372875727427e-06, + "loss": 0.7115, + "step": 535 + }, + { + "epoch": 0.4439536167863059, + "grad_norm": 0.8909885883331299, + "learning_rate": 4.954163498854936e-06, + "loss": 0.7146, + "step": 536 + }, + { + "epoch": 0.4447818884594147, + "grad_norm": 0.8951786756515503, + "learning_rate": 4.953953647124684e-06, + "loss": 0.699, + "step": 537 + }, + { + "epoch": 0.44561016013252347, + "grad_norm": 0.8890933990478516, + "learning_rate": 4.953743320577278e-06, + "loss": 0.7018, + "step": 538 + }, + { + "epoch": 0.44643843180563225, + "grad_norm": 0.910393238067627, + "learning_rate": 4.953532519253412e-06, + "loss": 0.7299, + "step": 539 + }, + { + "epoch": 0.44726670347874103, + "grad_norm": 0.9152857065200806, + "learning_rate": 4.953321243193876e-06, + "loss": 0.6984, + "step": 540 + }, + { + "epoch": 0.4480949751518498, + "grad_norm": 0.9001837372779846, + "learning_rate": 4.9531094924395485e-06, + "loss": 0.727, + "step": 541 + }, + { + "epoch": 0.4489232468249586, + "grad_norm": 0.882060170173645, + "learning_rate": 4.952897267031401e-06, + "loss": 0.7016, + "step": 542 + }, + { + "epoch": 0.4497515184980674, + "grad_norm": 0.8665159344673157, + "learning_rate": 4.9526845670104995e-06, + "loss": 0.7139, + "step": 543 + }, + { + "epoch": 0.45057979017117616, + "grad_norm": 0.9200844764709473, + "learning_rate": 4.952471392417997e-06, + "loss": 0.7016, + "step": 544 + }, + { + "epoch": 0.45140806184428495, + "grad_norm": 0.8852037191390991, + "learning_rate": 4.9522577432951416e-06, + "loss": 0.7048, + "step": 545 + }, + { + "epoch": 0.45223633351739373, + "grad_norm": 0.9003865122795105, + "learning_rate": 4.952043619683274e-06, + "loss": 0.7106, + "step": 546 + }, + { + "epoch": 0.4530646051905025, + "grad_norm": 0.8888446092605591, + "learning_rate": 4.951829021623822e-06, + "loss": 0.7202, + "step": 547 + }, + { + "epoch": 0.45389287686361124, + "grad_norm": 0.8940277099609375, + "learning_rate": 4.9516139491583094e-06, + "loss": 0.7156, + "step": 548 + }, + { + "epoch": 0.45472114853672, + "grad_norm": 0.9066211581230164, + "learning_rate": 4.951398402328353e-06, + "loss": 0.6865, + "step": 549 + }, + { + "epoch": 0.4555494202098288, + "grad_norm": 0.9352865815162659, + "learning_rate": 4.9511823811756565e-06, + "loss": 0.7117, + "step": 550 + }, + { + "epoch": 0.4563776918829376, + "grad_norm": 0.8672269582748413, + "learning_rate": 4.95096588574202e-06, + "loss": 0.7166, + "step": 551 + }, + { + "epoch": 0.45720596355604637, + "grad_norm": 0.8730685114860535, + "learning_rate": 4.95074891606933e-06, + "loss": 0.6998, + "step": 552 + }, + { + "epoch": 0.45803423522915515, + "grad_norm": 0.896821916103363, + "learning_rate": 4.950531472199572e-06, + "loss": 0.6839, + "step": 553 + }, + { + "epoch": 0.45886250690226393, + "grad_norm": 0.918572187423706, + "learning_rate": 4.950313554174817e-06, + "loss": 0.7155, + "step": 554 + }, + { + "epoch": 0.4596907785753727, + "grad_norm": 0.8670752048492432, + "learning_rate": 4.950095162037232e-06, + "loss": 0.708, + "step": 555 + }, + { + "epoch": 0.4605190502484815, + "grad_norm": 0.875618040561676, + "learning_rate": 4.9498762958290715e-06, + "loss": 0.6956, + "step": 556 + }, + { + "epoch": 0.4613473219215903, + "grad_norm": 0.9120676517486572, + "learning_rate": 4.949656955592685e-06, + "loss": 0.7327, + "step": 557 + }, + { + "epoch": 0.46217559359469906, + "grad_norm": 0.9670370817184448, + "learning_rate": 4.949437141370514e-06, + "loss": 0.705, + "step": 558 + }, + { + "epoch": 0.46300386526780785, + "grad_norm": 0.9155322313308716, + "learning_rate": 4.949216853205089e-06, + "loss": 0.7126, + "step": 559 + }, + { + "epoch": 0.46383213694091663, + "grad_norm": 0.8945392370223999, + "learning_rate": 4.948996091139034e-06, + "loss": 0.7083, + "step": 560 + }, + { + "epoch": 0.4646604086140254, + "grad_norm": 0.8922162055969238, + "learning_rate": 4.948774855215065e-06, + "loss": 0.7189, + "step": 561 + }, + { + "epoch": 0.4654886802871342, + "grad_norm": 0.8940657377243042, + "learning_rate": 4.948553145475989e-06, + "loss": 0.6934, + "step": 562 + }, + { + "epoch": 0.466316951960243, + "grad_norm": 0.8946855068206787, + "learning_rate": 4.948330961964703e-06, + "loss": 0.6835, + "step": 563 + }, + { + "epoch": 0.46714522363335176, + "grad_norm": 0.9002820253372192, + "learning_rate": 4.948108304724201e-06, + "loss": 0.712, + "step": 564 + }, + { + "epoch": 0.46797349530646054, + "grad_norm": 0.8715711832046509, + "learning_rate": 4.947885173797561e-06, + "loss": 0.7197, + "step": 565 + }, + { + "epoch": 0.4688017669795693, + "grad_norm": 0.9269874095916748, + "learning_rate": 4.94766156922796e-06, + "loss": 0.7311, + "step": 566 + }, + { + "epoch": 0.4696300386526781, + "grad_norm": 0.8999423980712891, + "learning_rate": 4.947437491058662e-06, + "loss": 0.7144, + "step": 567 + }, + { + "epoch": 0.47045831032578683, + "grad_norm": 0.8704255223274231, + "learning_rate": 4.947212939333024e-06, + "loss": 0.7156, + "step": 568 + }, + { + "epoch": 0.4712865819988956, + "grad_norm": 0.8770347237586975, + "learning_rate": 4.946987914094495e-06, + "loss": 0.6969, + "step": 569 + }, + { + "epoch": 0.4721148536720044, + "grad_norm": 0.8865682482719421, + "learning_rate": 4.946762415386615e-06, + "loss": 0.7089, + "step": 570 + }, + { + "epoch": 0.4729431253451132, + "grad_norm": 0.8756279349327087, + "learning_rate": 4.946536443253016e-06, + "loss": 0.7159, + "step": 571 + }, + { + "epoch": 0.47377139701822196, + "grad_norm": 0.8844839930534363, + "learning_rate": 4.946309997737422e-06, + "loss": 0.7234, + "step": 572 + }, + { + "epoch": 0.47459966869133074, + "grad_norm": 0.8549885153770447, + "learning_rate": 4.946083078883647e-06, + "loss": 0.6927, + "step": 573 + }, + { + "epoch": 0.4754279403644395, + "grad_norm": 0.8967456817626953, + "learning_rate": 4.945855686735599e-06, + "loss": 0.7082, + "step": 574 + }, + { + "epoch": 0.4762562120375483, + "grad_norm": 0.9120200276374817, + "learning_rate": 4.9456278213372755e-06, + "loss": 0.7135, + "step": 575 + }, + { + "epoch": 0.4770844837106571, + "grad_norm": 0.8780766725540161, + "learning_rate": 4.945399482732765e-06, + "loss": 0.7139, + "step": 576 + }, + { + "epoch": 0.4779127553837659, + "grad_norm": 0.8900999426841736, + "learning_rate": 4.945170670966253e-06, + "loss": 0.7132, + "step": 577 + }, + { + "epoch": 0.47874102705687466, + "grad_norm": 0.9402151107788086, + "learning_rate": 4.944941386082008e-06, + "loss": 0.7139, + "step": 578 + }, + { + "epoch": 0.47956929872998344, + "grad_norm": 0.8769465088844299, + "learning_rate": 4.944711628124396e-06, + "loss": 0.7236, + "step": 579 + }, + { + "epoch": 0.4803975704030922, + "grad_norm": 0.9185771346092224, + "learning_rate": 4.944481397137874e-06, + "loss": 0.6884, + "step": 580 + }, + { + "epoch": 0.481225842076201, + "grad_norm": 0.8775897026062012, + "learning_rate": 4.944250693166988e-06, + "loss": 0.731, + "step": 581 + }, + { + "epoch": 0.4820541137493098, + "grad_norm": 0.8751668930053711, + "learning_rate": 4.94401951625638e-06, + "loss": 0.7007, + "step": 582 + }, + { + "epoch": 0.48288238542241857, + "grad_norm": 0.9042515754699707, + "learning_rate": 4.943787866450776e-06, + "loss": 0.7064, + "step": 583 + }, + { + "epoch": 0.48371065709552735, + "grad_norm": 0.8845478892326355, + "learning_rate": 4.943555743795002e-06, + "loss": 0.7069, + "step": 584 + }, + { + "epoch": 0.48453892876863613, + "grad_norm": 0.8959096670150757, + "learning_rate": 4.943323148333969e-06, + "loss": 0.6871, + "step": 585 + }, + { + "epoch": 0.4853672004417449, + "grad_norm": 0.9021669626235962, + "learning_rate": 4.943090080112684e-06, + "loss": 0.7156, + "step": 586 + }, + { + "epoch": 0.4861954721148537, + "grad_norm": 0.8768396973609924, + "learning_rate": 4.942856539176243e-06, + "loss": 0.681, + "step": 587 + }, + { + "epoch": 0.4870237437879624, + "grad_norm": 0.8734706044197083, + "learning_rate": 4.942622525569832e-06, + "loss": 0.7371, + "step": 588 + }, + { + "epoch": 0.4878520154610712, + "grad_norm": 0.9155406355857849, + "learning_rate": 4.942388039338733e-06, + "loss": 0.7012, + "step": 589 + }, + { + "epoch": 0.48868028713418, + "grad_norm": 0.8778123259544373, + "learning_rate": 4.942153080528317e-06, + "loss": 0.7201, + "step": 590 + }, + { + "epoch": 0.4895085588072888, + "grad_norm": 0.9102092981338501, + "learning_rate": 4.941917649184045e-06, + "loss": 0.6951, + "step": 591 + }, + { + "epoch": 0.49033683048039756, + "grad_norm": 0.8855768442153931, + "learning_rate": 4.941681745351471e-06, + "loss": 0.6997, + "step": 592 + }, + { + "epoch": 0.49116510215350634, + "grad_norm": 0.8770707845687866, + "learning_rate": 4.94144536907624e-06, + "loss": 0.7092, + "step": 593 + }, + { + "epoch": 0.4919933738266151, + "grad_norm": 0.900458574295044, + "learning_rate": 4.9412085204040896e-06, + "loss": 0.6909, + "step": 594 + }, + { + "epoch": 0.4928216454997239, + "grad_norm": 0.903489887714386, + "learning_rate": 4.940971199380846e-06, + "loss": 0.7066, + "step": 595 + }, + { + "epoch": 0.4936499171728327, + "grad_norm": 0.9331871271133423, + "learning_rate": 4.940733406052431e-06, + "loss": 0.6924, + "step": 596 + }, + { + "epoch": 0.49447818884594147, + "grad_norm": 0.8855593800544739, + "learning_rate": 4.9404951404648535e-06, + "loss": 0.6991, + "step": 597 + }, + { + "epoch": 0.49530646051905025, + "grad_norm": 0.9075731039047241, + "learning_rate": 4.940256402664216e-06, + "loss": 0.6973, + "step": 598 + }, + { + "epoch": 0.49613473219215903, + "grad_norm": 0.8970743417739868, + "learning_rate": 4.940017192696713e-06, + "loss": 0.6767, + "step": 599 + }, + { + "epoch": 0.4969630038652678, + "grad_norm": 0.8644358515739441, + "learning_rate": 4.939777510608628e-06, + "loss": 0.7284, + "step": 600 + }, + { + "epoch": 0.4977912755383766, + "grad_norm": 0.8807653784751892, + "learning_rate": 4.939537356446339e-06, + "loss": 0.7143, + "step": 601 + }, + { + "epoch": 0.4986195472114854, + "grad_norm": 0.892946183681488, + "learning_rate": 4.939296730256312e-06, + "loss": 0.7021, + "step": 602 + }, + { + "epoch": 0.49944781888459416, + "grad_norm": 0.9661110639572144, + "learning_rate": 4.939055632085107e-06, + "loss": 0.7198, + "step": 603 + }, + { + "epoch": 0.500276090557703, + "grad_norm": 0.8638496398925781, + "learning_rate": 4.9388140619793735e-06, + "loss": 0.6943, + "step": 604 + }, + { + "epoch": 0.5011043622308117, + "grad_norm": 0.9025457501411438, + "learning_rate": 4.938572019985853e-06, + "loss": 0.7114, + "step": 605 + }, + { + "epoch": 0.5019326339039205, + "grad_norm": 0.9622734785079956, + "learning_rate": 4.93832950615138e-06, + "loss": 0.7086, + "step": 606 + }, + { + "epoch": 0.5027609055770292, + "grad_norm": 0.9565468430519104, + "learning_rate": 4.938086520522876e-06, + "loss": 0.7125, + "step": 607 + }, + { + "epoch": 0.5035891772501381, + "grad_norm": 0.8970094323158264, + "learning_rate": 4.937843063147359e-06, + "loss": 0.7023, + "step": 608 + }, + { + "epoch": 0.5044174489232468, + "grad_norm": 0.8920996785163879, + "learning_rate": 4.937599134071933e-06, + "loss": 0.7041, + "step": 609 + }, + { + "epoch": 0.5052457205963556, + "grad_norm": 0.9304755926132202, + "learning_rate": 4.937354733343799e-06, + "loss": 0.7165, + "step": 610 + }, + { + "epoch": 0.5060739922694644, + "grad_norm": 0.8995933532714844, + "learning_rate": 4.937109861010244e-06, + "loss": 0.6907, + "step": 611 + }, + { + "epoch": 0.5069022639425732, + "grad_norm": 0.8648850321769714, + "learning_rate": 4.936864517118649e-06, + "loss": 0.6994, + "step": 612 + }, + { + "epoch": 0.5077305356156819, + "grad_norm": 0.8414506316184998, + "learning_rate": 4.936618701716487e-06, + "loss": 0.7081, + "step": 613 + }, + { + "epoch": 0.5085588072887908, + "grad_norm": 0.9282650947570801, + "learning_rate": 4.936372414851321e-06, + "loss": 0.7303, + "step": 614 + }, + { + "epoch": 0.5093870789618995, + "grad_norm": 0.877757728099823, + "learning_rate": 4.936125656570802e-06, + "loss": 0.7132, + "step": 615 + }, + { + "epoch": 0.5102153506350083, + "grad_norm": 0.8645135164260864, + "learning_rate": 4.935878426922679e-06, + "loss": 0.723, + "step": 616 + }, + { + "epoch": 0.5110436223081171, + "grad_norm": 0.9014747738838196, + "learning_rate": 4.935630725954787e-06, + "loss": 0.7167, + "step": 617 + }, + { + "epoch": 0.5118718939812258, + "grad_norm": 0.9295386672019958, + "learning_rate": 4.935382553715055e-06, + "loss": 0.705, + "step": 618 + }, + { + "epoch": 0.5127001656543346, + "grad_norm": 0.8910616040229797, + "learning_rate": 4.935133910251499e-06, + "loss": 0.6781, + "step": 619 + }, + { + "epoch": 0.5135284373274434, + "grad_norm": 0.8804931640625, + "learning_rate": 4.934884795612234e-06, + "loss": 0.6984, + "step": 620 + }, + { + "epoch": 0.5143567090005522, + "grad_norm": 0.8223545551300049, + "learning_rate": 4.934635209845458e-06, + "loss": 0.6892, + "step": 621 + }, + { + "epoch": 0.5151849806736609, + "grad_norm": 0.8509450554847717, + "learning_rate": 4.934385152999463e-06, + "loss": 0.7139, + "step": 622 + }, + { + "epoch": 0.5160132523467698, + "grad_norm": 0.8944748044013977, + "learning_rate": 4.934134625122636e-06, + "loss": 0.6976, + "step": 623 + }, + { + "epoch": 0.5168415240198785, + "grad_norm": 0.9021086692810059, + "learning_rate": 4.933883626263448e-06, + "loss": 0.7189, + "step": 624 + }, + { + "epoch": 0.5176697956929873, + "grad_norm": 0.8939547538757324, + "learning_rate": 4.933632156470467e-06, + "loss": 0.704, + "step": 625 + }, + { + "epoch": 0.518498067366096, + "grad_norm": 0.8793498277664185, + "learning_rate": 4.93338021579235e-06, + "loss": 0.7046, + "step": 626 + }, + { + "epoch": 0.5193263390392049, + "grad_norm": 0.8728756308555603, + "learning_rate": 4.933127804277845e-06, + "loss": 0.6826, + "step": 627 + }, + { + "epoch": 0.5201546107123136, + "grad_norm": 0.8989412188529968, + "learning_rate": 4.932874921975792e-06, + "loss": 0.6842, + "step": 628 + }, + { + "epoch": 0.5209828823854225, + "grad_norm": 0.8836191296577454, + "learning_rate": 4.93262156893512e-06, + "loss": 0.7061, + "step": 629 + }, + { + "epoch": 0.5218111540585312, + "grad_norm": 0.9429040551185608, + "learning_rate": 4.9323677452048505e-06, + "loss": 0.7053, + "step": 630 + }, + { + "epoch": 0.52263942573164, + "grad_norm": 0.879463255405426, + "learning_rate": 4.932113450834097e-06, + "loss": 0.6926, + "step": 631 + }, + { + "epoch": 0.5234676974047487, + "grad_norm": 0.8566175699234009, + "learning_rate": 4.931858685872062e-06, + "loss": 0.7126, + "step": 632 + }, + { + "epoch": 0.5242959690778576, + "grad_norm": 0.8694538474082947, + "learning_rate": 4.931603450368042e-06, + "loss": 0.7153, + "step": 633 + }, + { + "epoch": 0.5251242407509663, + "grad_norm": 0.8754484057426453, + "learning_rate": 4.931347744371421e-06, + "loss": 0.711, + "step": 634 + }, + { + "epoch": 0.5259525124240751, + "grad_norm": 0.8982204794883728, + "learning_rate": 4.931091567931675e-06, + "loss": 0.7168, + "step": 635 + }, + { + "epoch": 0.5267807840971839, + "grad_norm": 0.9262725710868835, + "learning_rate": 4.930834921098374e-06, + "loss": 0.7315, + "step": 636 + }, + { + "epoch": 0.5276090557702926, + "grad_norm": 0.9271771311759949, + "learning_rate": 4.930577803921176e-06, + "loss": 0.6968, + "step": 637 + }, + { + "epoch": 0.5284373274434014, + "grad_norm": 0.8855197429656982, + "learning_rate": 4.93032021644983e-06, + "loss": 0.6986, + "step": 638 + }, + { + "epoch": 0.5292655991165102, + "grad_norm": 0.8615745902061462, + "learning_rate": 4.930062158734178e-06, + "loss": 0.7014, + "step": 639 + }, + { + "epoch": 0.530093870789619, + "grad_norm": 0.9004185795783997, + "learning_rate": 4.92980363082415e-06, + "loss": 0.7045, + "step": 640 + }, + { + "epoch": 0.5309221424627277, + "grad_norm": 0.8533529043197632, + "learning_rate": 4.92954463276977e-06, + "loss": 0.6911, + "step": 641 + }, + { + "epoch": 0.5317504141358366, + "grad_norm": 0.881901204586029, + "learning_rate": 4.929285164621152e-06, + "loss": 0.7148, + "step": 642 + }, + { + "epoch": 0.5325786858089453, + "grad_norm": 0.8778982162475586, + "learning_rate": 4.9290252264285e-06, + "loss": 0.6879, + "step": 643 + }, + { + "epoch": 0.5334069574820541, + "grad_norm": 0.9455735087394714, + "learning_rate": 4.928764818242109e-06, + "loss": 0.7088, + "step": 644 + }, + { + "epoch": 0.5342352291551629, + "grad_norm": 0.9255897998809814, + "learning_rate": 4.928503940112367e-06, + "loss": 0.6817, + "step": 645 + }, + { + "epoch": 0.5350635008282717, + "grad_norm": 0.8854040503501892, + "learning_rate": 4.928242592089752e-06, + "loss": 0.7074, + "step": 646 + }, + { + "epoch": 0.5358917725013804, + "grad_norm": 0.8824400901794434, + "learning_rate": 4.927980774224829e-06, + "loss": 0.7092, + "step": 647 + }, + { + "epoch": 0.5367200441744893, + "grad_norm": 0.8873158693313599, + "learning_rate": 4.927718486568261e-06, + "loss": 0.7214, + "step": 648 + }, + { + "epoch": 0.537548315847598, + "grad_norm": 0.9013757109642029, + "learning_rate": 4.927455729170796e-06, + "loss": 0.6955, + "step": 649 + }, + { + "epoch": 0.5383765875207068, + "grad_norm": 0.9100024104118347, + "learning_rate": 4.927192502083276e-06, + "loss": 0.7089, + "step": 650 + }, + { + "epoch": 0.5392048591938156, + "grad_norm": 0.8935851454734802, + "learning_rate": 4.9269288053566335e-06, + "loss": 0.7018, + "step": 651 + }, + { + "epoch": 0.5400331308669244, + "grad_norm": 0.8726267218589783, + "learning_rate": 4.92666463904189e-06, + "loss": 0.6829, + "step": 652 + }, + { + "epoch": 0.5408614025400331, + "grad_norm": 0.9320974349975586, + "learning_rate": 4.926400003190161e-06, + "loss": 0.6766, + "step": 653 + }, + { + "epoch": 0.541689674213142, + "grad_norm": 0.9257067441940308, + "learning_rate": 4.926134897852648e-06, + "loss": 0.6913, + "step": 654 + }, + { + "epoch": 0.5425179458862507, + "grad_norm": 0.8642577528953552, + "learning_rate": 4.92586932308065e-06, + "loss": 0.7155, + "step": 655 + }, + { + "epoch": 0.5433462175593595, + "grad_norm": 0.8861536383628845, + "learning_rate": 4.9256032789255505e-06, + "loss": 0.6695, + "step": 656 + }, + { + "epoch": 0.5441744892324683, + "grad_norm": 0.8901777267456055, + "learning_rate": 4.925336765438829e-06, + "loss": 0.6847, + "step": 657 + }, + { + "epoch": 0.545002760905577, + "grad_norm": 0.8810168504714966, + "learning_rate": 4.9250697826720515e-06, + "loss": 0.7038, + "step": 658 + }, + { + "epoch": 0.5458310325786858, + "grad_norm": 0.8655433654785156, + "learning_rate": 4.924802330676877e-06, + "loss": 0.6908, + "step": 659 + }, + { + "epoch": 0.5466593042517945, + "grad_norm": 0.8786777853965759, + "learning_rate": 4.924534409505055e-06, + "loss": 0.663, + "step": 660 + }, + { + "epoch": 0.5474875759249034, + "grad_norm": 0.8837904334068298, + "learning_rate": 4.924266019208427e-06, + "loss": 0.6843, + "step": 661 + }, + { + "epoch": 0.5483158475980121, + "grad_norm": 0.8715303540229797, + "learning_rate": 4.923997159838923e-06, + "loss": 0.6841, + "step": 662 + }, + { + "epoch": 0.549144119271121, + "grad_norm": 0.8825753927230835, + "learning_rate": 4.923727831448564e-06, + "loss": 0.7053, + "step": 663 + }, + { + "epoch": 0.5499723909442297, + "grad_norm": 0.877668559551239, + "learning_rate": 4.923458034089463e-06, + "loss": 0.6624, + "step": 664 + }, + { + "epoch": 0.5508006626173385, + "grad_norm": 0.8721411824226379, + "learning_rate": 4.923187767813825e-06, + "loss": 0.7339, + "step": 665 + }, + { + "epoch": 0.5516289342904472, + "grad_norm": 0.8607308268547058, + "learning_rate": 4.922917032673941e-06, + "loss": 0.7184, + "step": 666 + }, + { + "epoch": 0.5524572059635561, + "grad_norm": 0.8456798791885376, + "learning_rate": 4.922645828722199e-06, + "loss": 0.7056, + "step": 667 + }, + { + "epoch": 0.5532854776366648, + "grad_norm": 0.8557775020599365, + "learning_rate": 4.922374156011071e-06, + "loss": 0.6942, + "step": 668 + }, + { + "epoch": 0.5541137493097736, + "grad_norm": 0.8877674341201782, + "learning_rate": 4.922102014593126e-06, + "loss": 0.7173, + "step": 669 + }, + { + "epoch": 0.5549420209828824, + "grad_norm": 0.8662481904029846, + "learning_rate": 4.921829404521019e-06, + "loss": 0.7113, + "step": 670 + }, + { + "epoch": 0.5557702926559912, + "grad_norm": 0.8887649774551392, + "learning_rate": 4.921556325847499e-06, + "loss": 0.7028, + "step": 671 + }, + { + "epoch": 0.5565985643290999, + "grad_norm": 0.9837973713874817, + "learning_rate": 4.921282778625403e-06, + "loss": 0.6975, + "step": 672 + }, + { + "epoch": 0.5574268360022088, + "grad_norm": 0.8754701018333435, + "learning_rate": 4.921008762907661e-06, + "loss": 0.6983, + "step": 673 + }, + { + "epoch": 0.5582551076753175, + "grad_norm": 0.9125426411628723, + "learning_rate": 4.920734278747291e-06, + "loss": 0.6982, + "step": 674 + }, + { + "epoch": 0.5590833793484263, + "grad_norm": 0.9424439072608948, + "learning_rate": 4.920459326197403e-06, + "loss": 0.7304, + "step": 675 + }, + { + "epoch": 0.5599116510215351, + "grad_norm": 0.8718393445014954, + "learning_rate": 4.9201839053112e-06, + "loss": 0.6644, + "step": 676 + }, + { + "epoch": 0.5607399226946438, + "grad_norm": 0.9002220630645752, + "learning_rate": 4.919908016141972e-06, + "loss": 0.6966, + "step": 677 + }, + { + "epoch": 0.5615681943677526, + "grad_norm": 0.8925678133964539, + "learning_rate": 4.9196316587431e-06, + "loss": 0.713, + "step": 678 + }, + { + "epoch": 0.5623964660408614, + "grad_norm": 0.8854464292526245, + "learning_rate": 4.919354833168059e-06, + "loss": 0.7075, + "step": 679 + }, + { + "epoch": 0.5632247377139702, + "grad_norm": 0.8867958188056946, + "learning_rate": 4.919077539470409e-06, + "loss": 0.7166, + "step": 680 + }, + { + "epoch": 0.5640530093870789, + "grad_norm": 0.862531840801239, + "learning_rate": 4.918799777703806e-06, + "loss": 0.7119, + "step": 681 + }, + { + "epoch": 0.5648812810601878, + "grad_norm": 0.8623220920562744, + "learning_rate": 4.9185215479219944e-06, + "loss": 0.6758, + "step": 682 + }, + { + "epoch": 0.5657095527332965, + "grad_norm": 0.9171786904335022, + "learning_rate": 4.918242850178808e-06, + "loss": 0.7049, + "step": 683 + }, + { + "epoch": 0.5665378244064053, + "grad_norm": 0.8947017192840576, + "learning_rate": 4.9179636845281735e-06, + "loss": 0.7144, + "step": 684 + }, + { + "epoch": 0.567366096079514, + "grad_norm": 0.9343388080596924, + "learning_rate": 4.917684051024104e-06, + "loss": 0.7024, + "step": 685 + }, + { + "epoch": 0.5681943677526229, + "grad_norm": 0.8929025530815125, + "learning_rate": 4.917403949720709e-06, + "loss": 0.7162, + "step": 686 + }, + { + "epoch": 0.5690226394257316, + "grad_norm": 0.9000421762466431, + "learning_rate": 4.9171233806721854e-06, + "loss": 0.7085, + "step": 687 + }, + { + "epoch": 0.5698509110988405, + "grad_norm": 0.9404364228248596, + "learning_rate": 4.91684234393282e-06, + "loss": 0.6996, + "step": 688 + }, + { + "epoch": 0.5706791827719492, + "grad_norm": 0.9566411375999451, + "learning_rate": 4.916560839556989e-06, + "loss": 0.7241, + "step": 689 + }, + { + "epoch": 0.571507454445058, + "grad_norm": 0.9066181182861328, + "learning_rate": 4.916278867599163e-06, + "loss": 0.724, + "step": 690 + }, + { + "epoch": 0.5723357261181667, + "grad_norm": 0.8914163708686829, + "learning_rate": 4.915996428113901e-06, + "loss": 0.7046, + "step": 691 + }, + { + "epoch": 0.5731639977912756, + "grad_norm": 0.8977548480033875, + "learning_rate": 4.91571352115585e-06, + "loss": 0.6823, + "step": 692 + }, + { + "epoch": 0.5739922694643843, + "grad_norm": 0.8902036547660828, + "learning_rate": 4.915430146779754e-06, + "loss": 0.702, + "step": 693 + }, + { + "epoch": 0.5748205411374931, + "grad_norm": 0.8859509229660034, + "learning_rate": 4.915146305040439e-06, + "loss": 0.6996, + "step": 694 + }, + { + "epoch": 0.5756488128106019, + "grad_norm": 0.8916864991188049, + "learning_rate": 4.914861995992829e-06, + "loss": 0.6978, + "step": 695 + }, + { + "epoch": 0.5764770844837107, + "grad_norm": 0.8961358070373535, + "learning_rate": 4.9145772196919336e-06, + "loss": 0.6803, + "step": 696 + }, + { + "epoch": 0.5773053561568194, + "grad_norm": 0.8817684650421143, + "learning_rate": 4.914291976192854e-06, + "loss": 0.6891, + "step": 697 + }, + { + "epoch": 0.5781336278299282, + "grad_norm": 0.8821113705635071, + "learning_rate": 4.914006265550783e-06, + "loss": 0.7009, + "step": 698 + }, + { + "epoch": 0.578961899503037, + "grad_norm": 0.9553062319755554, + "learning_rate": 4.913720087821003e-06, + "loss": 0.7118, + "step": 699 + }, + { + "epoch": 0.5797901711761457, + "grad_norm": 0.8997507691383362, + "learning_rate": 4.913433443058887e-06, + "loss": 0.7036, + "step": 700 + }, + { + "epoch": 0.5806184428492546, + "grad_norm": 0.8659473657608032, + "learning_rate": 4.9131463313198964e-06, + "loss": 0.6766, + "step": 701 + }, + { + "epoch": 0.5814467145223633, + "grad_norm": 0.8916030526161194, + "learning_rate": 4.912858752659586e-06, + "loss": 0.6976, + "step": 702 + }, + { + "epoch": 0.5822749861954721, + "grad_norm": 0.8726498484611511, + "learning_rate": 4.912570707133599e-06, + "loss": 0.6989, + "step": 703 + }, + { + "epoch": 0.5831032578685809, + "grad_norm": 0.8725529909133911, + "learning_rate": 4.91228219479767e-06, + "loss": 0.6862, + "step": 704 + }, + { + "epoch": 0.5839315295416897, + "grad_norm": 0.8807168006896973, + "learning_rate": 4.911993215707624e-06, + "loss": 0.6964, + "step": 705 + }, + { + "epoch": 0.5847598012147984, + "grad_norm": 0.9244167804718018, + "learning_rate": 4.911703769919375e-06, + "loss": 0.6742, + "step": 706 + }, + { + "epoch": 0.5855880728879073, + "grad_norm": 0.9174355268478394, + "learning_rate": 4.9114138574889285e-06, + "loss": 0.6852, + "step": 707 + }, + { + "epoch": 0.586416344561016, + "grad_norm": 0.8894474506378174, + "learning_rate": 4.9111234784723785e-06, + "loss": 0.689, + "step": 708 + }, + { + "epoch": 0.5872446162341248, + "grad_norm": 0.8856050968170166, + "learning_rate": 4.910832632925912e-06, + "loss": 0.6742, + "step": 709 + }, + { + "epoch": 0.5880728879072336, + "grad_norm": 0.9120550751686096, + "learning_rate": 4.910541320905804e-06, + "loss": 0.7016, + "step": 710 + }, + { + "epoch": 0.5889011595803424, + "grad_norm": 0.8925642967224121, + "learning_rate": 4.910249542468422e-06, + "loss": 0.7165, + "step": 711 + }, + { + "epoch": 0.5897294312534511, + "grad_norm": 0.8786419630050659, + "learning_rate": 4.909957297670222e-06, + "loss": 0.7207, + "step": 712 + }, + { + "epoch": 0.59055770292656, + "grad_norm": 0.89600670337677, + "learning_rate": 4.909664586567749e-06, + "loss": 0.672, + "step": 713 + }, + { + "epoch": 0.5913859745996687, + "grad_norm": 0.889835000038147, + "learning_rate": 4.909371409217642e-06, + "loss": 0.7089, + "step": 714 + }, + { + "epoch": 0.5922142462727775, + "grad_norm": 0.9029288291931152, + "learning_rate": 4.909077765676627e-06, + "loss": 0.7065, + "step": 715 + }, + { + "epoch": 0.5930425179458862, + "grad_norm": 0.9212722778320312, + "learning_rate": 4.908783656001521e-06, + "loss": 0.6787, + "step": 716 + }, + { + "epoch": 0.593870789618995, + "grad_norm": 0.9079681038856506, + "learning_rate": 4.9084890802492325e-06, + "loss": 0.6996, + "step": 717 + }, + { + "epoch": 0.5946990612921038, + "grad_norm": 0.9110735654830933, + "learning_rate": 4.908194038476758e-06, + "loss": 0.6962, + "step": 718 + }, + { + "epoch": 0.5955273329652125, + "grad_norm": 0.8945547938346863, + "learning_rate": 4.9078985307411865e-06, + "loss": 0.6648, + "step": 719 + }, + { + "epoch": 0.5963556046383214, + "grad_norm": 0.8831511735916138, + "learning_rate": 4.907602557099696e-06, + "loss": 0.7077, + "step": 720 + }, + { + "epoch": 0.5971838763114301, + "grad_norm": 0.9393531680107117, + "learning_rate": 4.9073061176095525e-06, + "loss": 0.6722, + "step": 721 + }, + { + "epoch": 0.5980121479845389, + "grad_norm": 0.8807207942008972, + "learning_rate": 4.907009212328117e-06, + "loss": 0.69, + "step": 722 + }, + { + "epoch": 0.5988404196576477, + "grad_norm": 0.8811002969741821, + "learning_rate": 4.906711841312836e-06, + "loss": 0.6902, + "step": 723 + }, + { + "epoch": 0.5996686913307565, + "grad_norm": 0.8953952789306641, + "learning_rate": 4.906414004621249e-06, + "loss": 0.6747, + "step": 724 + }, + { + "epoch": 0.6004969630038652, + "grad_norm": 0.8661350607872009, + "learning_rate": 4.9061157023109855e-06, + "loss": 0.7047, + "step": 725 + }, + { + "epoch": 0.6013252346769741, + "grad_norm": 0.8722146153450012, + "learning_rate": 4.905816934439762e-06, + "loss": 0.7231, + "step": 726 + }, + { + "epoch": 0.6021535063500828, + "grad_norm": 0.8933189511299133, + "learning_rate": 4.905517701065389e-06, + "loss": 0.7029, + "step": 727 + }, + { + "epoch": 0.6029817780231916, + "grad_norm": 0.8863442540168762, + "learning_rate": 4.905218002245764e-06, + "loss": 0.6886, + "step": 728 + }, + { + "epoch": 0.6038100496963004, + "grad_norm": 0.8672969341278076, + "learning_rate": 4.904917838038877e-06, + "loss": 0.7048, + "step": 729 + }, + { + "epoch": 0.6046383213694092, + "grad_norm": 0.8892956972122192, + "learning_rate": 4.904617208502807e-06, + "loss": 0.7009, + "step": 730 + }, + { + "epoch": 0.6054665930425179, + "grad_norm": 0.9083442687988281, + "learning_rate": 4.904316113695723e-06, + "loss": 0.7169, + "step": 731 + }, + { + "epoch": 0.6062948647156268, + "grad_norm": 0.8885907530784607, + "learning_rate": 4.904014553675883e-06, + "loss": 0.6939, + "step": 732 + }, + { + "epoch": 0.6071231363887355, + "grad_norm": 0.8859131336212158, + "learning_rate": 4.903712528501637e-06, + "loss": 0.6881, + "step": 733 + }, + { + "epoch": 0.6079514080618443, + "grad_norm": 0.874146580696106, + "learning_rate": 4.903410038231425e-06, + "loss": 0.6857, + "step": 734 + }, + { + "epoch": 0.6087796797349531, + "grad_norm": 0.8719701170921326, + "learning_rate": 4.903107082923774e-06, + "loss": 0.6834, + "step": 735 + }, + { + "epoch": 0.6096079514080619, + "grad_norm": 0.9656700491905212, + "learning_rate": 4.902803662637304e-06, + "loss": 0.6933, + "step": 736 + }, + { + "epoch": 0.6104362230811706, + "grad_norm": 0.9022018313407898, + "learning_rate": 4.9024997774307235e-06, + "loss": 0.683, + "step": 737 + }, + { + "epoch": 0.6112644947542794, + "grad_norm": 0.8694130182266235, + "learning_rate": 4.902195427362832e-06, + "loss": 0.642, + "step": 738 + }, + { + "epoch": 0.6120927664273882, + "grad_norm": 0.9153549075126648, + "learning_rate": 4.901890612492519e-06, + "loss": 0.6674, + "step": 739 + }, + { + "epoch": 0.6129210381004969, + "grad_norm": 0.9431132674217224, + "learning_rate": 4.901585332878762e-06, + "loss": 0.6918, + "step": 740 + }, + { + "epoch": 0.6137493097736058, + "grad_norm": 0.881587028503418, + "learning_rate": 4.901279588580631e-06, + "loss": 0.687, + "step": 741 + }, + { + "epoch": 0.6145775814467145, + "grad_norm": 0.9289383888244629, + "learning_rate": 4.9009733796572834e-06, + "loss": 0.6789, + "step": 742 + }, + { + "epoch": 0.6154058531198233, + "grad_norm": 0.9058337211608887, + "learning_rate": 4.900666706167968e-06, + "loss": 0.6856, + "step": 743 + }, + { + "epoch": 0.616234124792932, + "grad_norm": 0.8999397158622742, + "learning_rate": 4.900359568172024e-06, + "loss": 0.7159, + "step": 744 + }, + { + "epoch": 0.6170623964660409, + "grad_norm": 0.895321249961853, + "learning_rate": 4.90005196572888e-06, + "loss": 0.7071, + "step": 745 + }, + { + "epoch": 0.6178906681391496, + "grad_norm": 0.9142102599143982, + "learning_rate": 4.899743898898054e-06, + "loss": 0.6918, + "step": 746 + }, + { + "epoch": 0.6187189398122585, + "grad_norm": 0.8867097496986389, + "learning_rate": 4.899435367739152e-06, + "loss": 0.6991, + "step": 747 + }, + { + "epoch": 0.6195472114853672, + "grad_norm": 0.8594791293144226, + "learning_rate": 4.899126372311876e-06, + "loss": 0.7153, + "step": 748 + }, + { + "epoch": 0.620375483158476, + "grad_norm": 0.8740563988685608, + "learning_rate": 4.8988169126760085e-06, + "loss": 0.6786, + "step": 749 + }, + { + "epoch": 0.6212037548315847, + "grad_norm": 0.8818353414535522, + "learning_rate": 4.898506988891432e-06, + "loss": 0.6965, + "step": 750 + }, + { + "epoch": 0.6220320265046936, + "grad_norm": 0.9459843635559082, + "learning_rate": 4.898196601018111e-06, + "loss": 0.6821, + "step": 751 + }, + { + "epoch": 0.6228602981778023, + "grad_norm": 0.923457682132721, + "learning_rate": 4.897885749116104e-06, + "loss": 0.6822, + "step": 752 + }, + { + "epoch": 0.6236885698509111, + "grad_norm": 0.9342297315597534, + "learning_rate": 4.897574433245557e-06, + "loss": 0.7001, + "step": 753 + }, + { + "epoch": 0.6245168415240199, + "grad_norm": 0.9327735900878906, + "learning_rate": 4.897262653466706e-06, + "loss": 0.6921, + "step": 754 + }, + { + "epoch": 0.6253451131971287, + "grad_norm": 0.8876991868019104, + "learning_rate": 4.8969504098398805e-06, + "loss": 0.7087, + "step": 755 + }, + { + "epoch": 0.6261733848702374, + "grad_norm": 0.8983528017997742, + "learning_rate": 4.896637702425493e-06, + "loss": 0.6824, + "step": 756 + }, + { + "epoch": 0.6270016565433462, + "grad_norm": 0.9074965715408325, + "learning_rate": 4.896324531284053e-06, + "loss": 0.6533, + "step": 757 + }, + { + "epoch": 0.627829928216455, + "grad_norm": 0.8938930630683899, + "learning_rate": 4.896010896476153e-06, + "loss": 0.6859, + "step": 758 + }, + { + "epoch": 0.6286581998895637, + "grad_norm": 0.9036867022514343, + "learning_rate": 4.8956967980624806e-06, + "loss": 0.6904, + "step": 759 + }, + { + "epoch": 0.6294864715626726, + "grad_norm": 0.9162903428077698, + "learning_rate": 4.895382236103809e-06, + "loss": 0.6995, + "step": 760 + }, + { + "epoch": 0.6303147432357813, + "grad_norm": 1.0000487565994263, + "learning_rate": 4.895067210661005e-06, + "loss": 0.671, + "step": 761 + }, + { + "epoch": 0.6311430149088901, + "grad_norm": 0.895163893699646, + "learning_rate": 4.894751721795022e-06, + "loss": 0.6871, + "step": 762 + }, + { + "epoch": 0.6319712865819989, + "grad_norm": 0.9401512742042542, + "learning_rate": 4.894435769566905e-06, + "loss": 0.6811, + "step": 763 + }, + { + "epoch": 0.6327995582551077, + "grad_norm": 0.924578845500946, + "learning_rate": 4.894119354037787e-06, + "loss": 0.692, + "step": 764 + }, + { + "epoch": 0.6336278299282164, + "grad_norm": 0.9114750623703003, + "learning_rate": 4.893802475268892e-06, + "loss": 0.6977, + "step": 765 + }, + { + "epoch": 0.6344561016013253, + "grad_norm": 1.1472550630569458, + "learning_rate": 4.893485133321533e-06, + "loss": 0.679, + "step": 766 + }, + { + "epoch": 0.635284373274434, + "grad_norm": 0.9460884928703308, + "learning_rate": 4.893167328257111e-06, + "loss": 0.6988, + "step": 767 + }, + { + "epoch": 0.6361126449475428, + "grad_norm": 0.9040783643722534, + "learning_rate": 4.8928490601371216e-06, + "loss": 0.6706, + "step": 768 + }, + { + "epoch": 0.6369409166206516, + "grad_norm": 0.9785428047180176, + "learning_rate": 4.892530329023144e-06, + "loss": 0.7185, + "step": 769 + }, + { + "epoch": 0.6377691882937604, + "grad_norm": 0.9795847535133362, + "learning_rate": 4.892211134976851e-06, + "loss": 0.703, + "step": 770 + }, + { + "epoch": 0.6385974599668691, + "grad_norm": 0.9539455771446228, + "learning_rate": 4.8918914780600025e-06, + "loss": 0.6985, + "step": 771 + }, + { + "epoch": 0.639425731639978, + "grad_norm": 0.9304025173187256, + "learning_rate": 4.8915713583344505e-06, + "loss": 0.6731, + "step": 772 + }, + { + "epoch": 0.6402540033130867, + "grad_norm": 0.9466667771339417, + "learning_rate": 4.891250775862134e-06, + "loss": 0.6972, + "step": 773 + }, + { + "epoch": 0.6410822749861955, + "grad_norm": 0.9522298574447632, + "learning_rate": 4.890929730705084e-06, + "loss": 0.6793, + "step": 774 + }, + { + "epoch": 0.6419105466593042, + "grad_norm": 0.9077293276786804, + "learning_rate": 4.890608222925419e-06, + "loss": 0.6892, + "step": 775 + }, + { + "epoch": 0.6427388183324131, + "grad_norm": 0.8763419389724731, + "learning_rate": 4.890286252585348e-06, + "loss": 0.6985, + "step": 776 + }, + { + "epoch": 0.6435670900055218, + "grad_norm": 0.9089516401290894, + "learning_rate": 4.8899638197471685e-06, + "loss": 0.6778, + "step": 777 + }, + { + "epoch": 0.6443953616786305, + "grad_norm": 0.8733325600624084, + "learning_rate": 4.88964092447327e-06, + "loss": 0.6915, + "step": 778 + }, + { + "epoch": 0.6452236333517394, + "grad_norm": 0.9472198486328125, + "learning_rate": 4.889317566826128e-06, + "loss": 0.6984, + "step": 779 + }, + { + "epoch": 0.6460519050248481, + "grad_norm": 0.8729666471481323, + "learning_rate": 4.888993746868309e-06, + "loss": 0.6724, + "step": 780 + }, + { + "epoch": 0.6468801766979569, + "grad_norm": 0.9395233988761902, + "learning_rate": 4.88866946466247e-06, + "loss": 0.6697, + "step": 781 + }, + { + "epoch": 0.6477084483710657, + "grad_norm": 0.9158744812011719, + "learning_rate": 4.888344720271357e-06, + "loss": 0.6838, + "step": 782 + }, + { + "epoch": 0.6485367200441745, + "grad_norm": 0.8961113095283508, + "learning_rate": 4.888019513757804e-06, + "loss": 0.6488, + "step": 783 + }, + { + "epoch": 0.6493649917172832, + "grad_norm": 0.8911786079406738, + "learning_rate": 4.8876938451847364e-06, + "loss": 0.7089, + "step": 784 + }, + { + "epoch": 0.6501932633903921, + "grad_norm": 0.9214697480201721, + "learning_rate": 4.887367714615167e-06, + "loss": 0.6929, + "step": 785 + }, + { + "epoch": 0.6510215350635008, + "grad_norm": 0.9161441326141357, + "learning_rate": 4.8870411221122e-06, + "loss": 0.6761, + "step": 786 + }, + { + "epoch": 0.6518498067366096, + "grad_norm": 0.9161947965621948, + "learning_rate": 4.886714067739028e-06, + "loss": 0.6804, + "step": 787 + }, + { + "epoch": 0.6526780784097184, + "grad_norm": 0.8706585168838501, + "learning_rate": 4.886386551558933e-06, + "loss": 0.6836, + "step": 788 + }, + { + "epoch": 0.6535063500828272, + "grad_norm": 0.9037661552429199, + "learning_rate": 4.886058573635284e-06, + "loss": 0.6786, + "step": 789 + }, + { + "epoch": 0.6543346217559359, + "grad_norm": 0.8977832794189453, + "learning_rate": 4.885730134031545e-06, + "loss": 0.6938, + "step": 790 + }, + { + "epoch": 0.6551628934290448, + "grad_norm": 0.9106955528259277, + "learning_rate": 4.8854012328112645e-06, + "loss": 0.678, + "step": 791 + }, + { + "epoch": 0.6559911651021535, + "grad_norm": 0.8769396543502808, + "learning_rate": 4.885071870038082e-06, + "loss": 0.6969, + "step": 792 + }, + { + "epoch": 0.6568194367752623, + "grad_norm": 0.9946926832199097, + "learning_rate": 4.884742045775725e-06, + "loss": 0.6799, + "step": 793 + }, + { + "epoch": 0.6576477084483711, + "grad_norm": 0.9399042725563049, + "learning_rate": 4.884411760088012e-06, + "loss": 0.7038, + "step": 794 + }, + { + "epoch": 0.6584759801214799, + "grad_norm": 0.8914628624916077, + "learning_rate": 4.8840810130388526e-06, + "loss": 0.6867, + "step": 795 + }, + { + "epoch": 0.6593042517945886, + "grad_norm": 0.8900558352470398, + "learning_rate": 4.88374980469224e-06, + "loss": 0.6823, + "step": 796 + }, + { + "epoch": 0.6601325234676974, + "grad_norm": 0.8608913421630859, + "learning_rate": 4.8834181351122615e-06, + "loss": 0.6705, + "step": 797 + }, + { + "epoch": 0.6609607951408062, + "grad_norm": 0.900460422039032, + "learning_rate": 4.883086004363092e-06, + "loss": 0.7029, + "step": 798 + }, + { + "epoch": 0.6617890668139149, + "grad_norm": 0.8932703733444214, + "learning_rate": 4.882753412508996e-06, + "loss": 0.6845, + "step": 799 + }, + { + "epoch": 0.6626173384870238, + "grad_norm": 0.8737874627113342, + "learning_rate": 4.882420359614325e-06, + "loss": 0.6979, + "step": 800 + }, + { + "epoch": 0.6634456101601325, + "grad_norm": 0.8805396556854248, + "learning_rate": 4.882086845743524e-06, + "loss": 0.6935, + "step": 801 + }, + { + "epoch": 0.6642738818332413, + "grad_norm": 0.8952041268348694, + "learning_rate": 4.881752870961123e-06, + "loss": 0.7137, + "step": 802 + }, + { + "epoch": 0.66510215350635, + "grad_norm": 0.889869213104248, + "learning_rate": 4.881418435331745e-06, + "loss": 0.6643, + "step": 803 + }, + { + "epoch": 0.6659304251794589, + "grad_norm": 0.8912904858589172, + "learning_rate": 4.881083538920098e-06, + "loss": 0.6712, + "step": 804 + }, + { + "epoch": 0.6667586968525676, + "grad_norm": 0.8943006992340088, + "learning_rate": 4.880748181790984e-06, + "loss": 0.7001, + "step": 805 + }, + { + "epoch": 0.6675869685256764, + "grad_norm": 0.9042482376098633, + "learning_rate": 4.880412364009289e-06, + "loss": 0.6744, + "step": 806 + }, + { + "epoch": 0.6684152401987852, + "grad_norm": 0.9052988886833191, + "learning_rate": 4.880076085639991e-06, + "loss": 0.72, + "step": 807 + }, + { + "epoch": 0.669243511871894, + "grad_norm": 0.9702905416488647, + "learning_rate": 4.879739346748158e-06, + "loss": 0.6871, + "step": 808 + }, + { + "epoch": 0.6700717835450027, + "grad_norm": 0.8741766810417175, + "learning_rate": 4.879402147398944e-06, + "loss": 0.7174, + "step": 809 + }, + { + "epoch": 0.6709000552181116, + "grad_norm": 0.9226665496826172, + "learning_rate": 4.879064487657596e-06, + "loss": 0.6991, + "step": 810 + }, + { + "epoch": 0.6717283268912203, + "grad_norm": 0.878304123878479, + "learning_rate": 4.878726367589446e-06, + "loss": 0.6925, + "step": 811 + }, + { + "epoch": 0.6725565985643291, + "grad_norm": 0.893649160861969, + "learning_rate": 4.878387787259919e-06, + "loss": 0.6689, + "step": 812 + }, + { + "epoch": 0.6733848702374379, + "grad_norm": 0.917861819267273, + "learning_rate": 4.878048746734526e-06, + "loss": 0.7171, + "step": 813 + }, + { + "epoch": 0.6742131419105467, + "grad_norm": 0.879375696182251, + "learning_rate": 4.877709246078869e-06, + "loss": 0.6662, + "step": 814 + }, + { + "epoch": 0.6750414135836554, + "grad_norm": 0.9098542928695679, + "learning_rate": 4.877369285358637e-06, + "loss": 0.6793, + "step": 815 + }, + { + "epoch": 0.6758696852567642, + "grad_norm": 0.9465262293815613, + "learning_rate": 4.877028864639609e-06, + "loss": 0.7065, + "step": 816 + }, + { + "epoch": 0.676697956929873, + "grad_norm": 0.9029957056045532, + "learning_rate": 4.876687983987654e-06, + "loss": 0.7008, + "step": 817 + }, + { + "epoch": 0.6775262286029817, + "grad_norm": 0.9073920845985413, + "learning_rate": 4.876346643468731e-06, + "loss": 0.6836, + "step": 818 + }, + { + "epoch": 0.6783545002760906, + "grad_norm": 0.9028745889663696, + "learning_rate": 4.876004843148884e-06, + "loss": 0.6965, + "step": 819 + }, + { + "epoch": 0.6791827719491993, + "grad_norm": 0.8745572566986084, + "learning_rate": 4.875662583094249e-06, + "loss": 0.6974, + "step": 820 + }, + { + "epoch": 0.6800110436223081, + "grad_norm": 0.9203384518623352, + "learning_rate": 4.875319863371048e-06, + "loss": 0.686, + "step": 821 + }, + { + "epoch": 0.6808393152954169, + "grad_norm": 0.9087005853652954, + "learning_rate": 4.874976684045597e-06, + "loss": 0.683, + "step": 822 + }, + { + "epoch": 0.6816675869685257, + "grad_norm": 0.9007741808891296, + "learning_rate": 4.874633045184297e-06, + "loss": 0.6938, + "step": 823 + }, + { + "epoch": 0.6824958586416344, + "grad_norm": 0.9854089617729187, + "learning_rate": 4.874288946853639e-06, + "loss": 0.6945, + "step": 824 + }, + { + "epoch": 0.6833241303147433, + "grad_norm": 0.9044802784919739, + "learning_rate": 4.8739443891202025e-06, + "loss": 0.714, + "step": 825 + }, + { + "epoch": 0.684152401987852, + "grad_norm": 0.9388754963874817, + "learning_rate": 4.873599372050656e-06, + "loss": 0.6838, + "step": 826 + }, + { + "epoch": 0.6849806736609608, + "grad_norm": 0.9215866327285767, + "learning_rate": 4.873253895711757e-06, + "loss": 0.6994, + "step": 827 + }, + { + "epoch": 0.6858089453340696, + "grad_norm": 0.8786854147911072, + "learning_rate": 4.872907960170353e-06, + "loss": 0.6883, + "step": 828 + }, + { + "epoch": 0.6866372170071784, + "grad_norm": 0.9032605290412903, + "learning_rate": 4.872561565493379e-06, + "loss": 0.6847, + "step": 829 + }, + { + "epoch": 0.6874654886802871, + "grad_norm": 0.8940309882164001, + "learning_rate": 4.872214711747858e-06, + "loss": 0.6877, + "step": 830 + }, + { + "epoch": 0.688293760353396, + "grad_norm": 0.8703469634056091, + "learning_rate": 4.8718673990009045e-06, + "loss": 0.6829, + "step": 831 + }, + { + "epoch": 0.6891220320265047, + "grad_norm": 0.8618217706680298, + "learning_rate": 4.871519627319719e-06, + "loss": 0.6987, + "step": 832 + }, + { + "epoch": 0.6899503036996135, + "grad_norm": 0.8871967196464539, + "learning_rate": 4.871171396771594e-06, + "loss": 0.6824, + "step": 833 + }, + { + "epoch": 0.6907785753727222, + "grad_norm": 0.9162507057189941, + "learning_rate": 4.8708227074239064e-06, + "loss": 0.7158, + "step": 834 + }, + { + "epoch": 0.6916068470458311, + "grad_norm": 0.8829584717750549, + "learning_rate": 4.8704735593441255e-06, + "loss": 0.6714, + "step": 835 + }, + { + "epoch": 0.6924351187189398, + "grad_norm": 0.9130700826644897, + "learning_rate": 4.8701239525998076e-06, + "loss": 0.6841, + "step": 836 + }, + { + "epoch": 0.6932633903920485, + "grad_norm": 0.8706396222114563, + "learning_rate": 4.8697738872586e-06, + "loss": 0.7162, + "step": 837 + }, + { + "epoch": 0.6940916620651574, + "grad_norm": 0.8906163573265076, + "learning_rate": 4.869423363388235e-06, + "loss": 0.6586, + "step": 838 + }, + { + "epoch": 0.6949199337382661, + "grad_norm": 0.9134309887886047, + "learning_rate": 4.869072381056537e-06, + "loss": 0.6842, + "step": 839 + }, + { + "epoch": 0.6957482054113749, + "grad_norm": 0.869674563407898, + "learning_rate": 4.8687209403314175e-06, + "loss": 0.7049, + "step": 840 + }, + { + "epoch": 0.6965764770844837, + "grad_norm": 0.8405192494392395, + "learning_rate": 4.8683690412808775e-06, + "loss": 0.6677, + "step": 841 + }, + { + "epoch": 0.6974047487575925, + "grad_norm": 0.9173605442047119, + "learning_rate": 4.868016683973005e-06, + "loss": 0.6784, + "step": 842 + }, + { + "epoch": 0.6982330204307012, + "grad_norm": 0.9220597743988037, + "learning_rate": 4.867663868475978e-06, + "loss": 0.6774, + "step": 843 + }, + { + "epoch": 0.6990612921038101, + "grad_norm": 0.8687174320220947, + "learning_rate": 4.867310594858065e-06, + "loss": 0.7094, + "step": 844 + }, + { + "epoch": 0.6998895637769188, + "grad_norm": 0.9161131381988525, + "learning_rate": 4.866956863187619e-06, + "loss": 0.6867, + "step": 845 + }, + { + "epoch": 0.7007178354500276, + "grad_norm": 0.9338458776473999, + "learning_rate": 4.866602673533084e-06, + "loss": 0.6783, + "step": 846 + }, + { + "epoch": 0.7015461071231364, + "grad_norm": 0.9268187284469604, + "learning_rate": 4.866248025962993e-06, + "loss": 0.6681, + "step": 847 + }, + { + "epoch": 0.7023743787962452, + "grad_norm": 0.8843579292297363, + "learning_rate": 4.865892920545967e-06, + "loss": 0.6854, + "step": 848 + }, + { + "epoch": 0.7032026504693539, + "grad_norm": 0.889299213886261, + "learning_rate": 4.865537357350716e-06, + "loss": 0.7055, + "step": 849 + }, + { + "epoch": 0.7040309221424628, + "grad_norm": 0.9062727093696594, + "learning_rate": 4.865181336446036e-06, + "loss": 0.6806, + "step": 850 + }, + { + "epoch": 0.7048591938155715, + "grad_norm": 0.9007996320724487, + "learning_rate": 4.864824857900817e-06, + "loss": 0.6937, + "step": 851 + }, + { + "epoch": 0.7056874654886803, + "grad_norm": 0.8883542418479919, + "learning_rate": 4.864467921784032e-06, + "loss": 0.6957, + "step": 852 + }, + { + "epoch": 0.7065157371617891, + "grad_norm": 0.9422078132629395, + "learning_rate": 4.864110528164746e-06, + "loss": 0.7025, + "step": 853 + }, + { + "epoch": 0.7073440088348979, + "grad_norm": 0.8961433172225952, + "learning_rate": 4.86375267711211e-06, + "loss": 0.6795, + "step": 854 + }, + { + "epoch": 0.7081722805080066, + "grad_norm": 0.925815761089325, + "learning_rate": 4.863394368695366e-06, + "loss": 0.6824, + "step": 855 + }, + { + "epoch": 0.7090005521811154, + "grad_norm": 0.9326804280281067, + "learning_rate": 4.863035602983843e-06, + "loss": 0.6829, + "step": 856 + }, + { + "epoch": 0.7098288238542242, + "grad_norm": 0.8784539699554443, + "learning_rate": 4.862676380046959e-06, + "loss": 0.6776, + "step": 857 + }, + { + "epoch": 0.7106570955273329, + "grad_norm": 0.8577290177345276, + "learning_rate": 4.86231669995422e-06, + "loss": 0.6573, + "step": 858 + }, + { + "epoch": 0.7114853672004418, + "grad_norm": 0.9261177778244019, + "learning_rate": 4.86195656277522e-06, + "loss": 0.6891, + "step": 859 + }, + { + "epoch": 0.7123136388735505, + "grad_norm": 0.9008646607398987, + "learning_rate": 4.861595968579643e-06, + "loss": 0.6714, + "step": 860 + }, + { + "epoch": 0.7131419105466593, + "grad_norm": 0.8729333877563477, + "learning_rate": 4.861234917437261e-06, + "loss": 0.6819, + "step": 861 + }, + { + "epoch": 0.713970182219768, + "grad_norm": 0.9006103873252869, + "learning_rate": 4.860873409417934e-06, + "loss": 0.6702, + "step": 862 + }, + { + "epoch": 0.7147984538928769, + "grad_norm": 0.8699921369552612, + "learning_rate": 4.860511444591609e-06, + "loss": 0.6681, + "step": 863 + }, + { + "epoch": 0.7156267255659856, + "grad_norm": 0.9359689950942993, + "learning_rate": 4.860149023028325e-06, + "loss": 0.6834, + "step": 864 + }, + { + "epoch": 0.7164549972390944, + "grad_norm": 0.874977707862854, + "learning_rate": 4.859786144798205e-06, + "loss": 0.6692, + "step": 865 + }, + { + "epoch": 0.7172832689122032, + "grad_norm": 0.8821915984153748, + "learning_rate": 4.859422809971464e-06, + "loss": 0.6865, + "step": 866 + }, + { + "epoch": 0.718111540585312, + "grad_norm": 0.905248761177063, + "learning_rate": 4.859059018618405e-06, + "loss": 0.6705, + "step": 867 + }, + { + "epoch": 0.7189398122584207, + "grad_norm": 0.9134542346000671, + "learning_rate": 4.858694770809415e-06, + "loss": 0.7039, + "step": 868 + }, + { + "epoch": 0.7197680839315296, + "grad_norm": 0.9048055410385132, + "learning_rate": 4.858330066614975e-06, + "loss": 0.6848, + "step": 869 + }, + { + "epoch": 0.7205963556046383, + "grad_norm": 0.8780005574226379, + "learning_rate": 4.857964906105651e-06, + "loss": 0.689, + "step": 870 + }, + { + "epoch": 0.7214246272777471, + "grad_norm": 0.8826956152915955, + "learning_rate": 4.857599289352098e-06, + "loss": 0.6799, + "step": 871 + }, + { + "epoch": 0.7222528989508559, + "grad_norm": 0.919905960559845, + "learning_rate": 4.857233216425061e-06, + "loss": 0.6587, + "step": 872 + }, + { + "epoch": 0.7230811706239647, + "grad_norm": 0.8711065053939819, + "learning_rate": 4.8568666873953696e-06, + "loss": 0.6754, + "step": 873 + }, + { + "epoch": 0.7239094422970734, + "grad_norm": 0.8740426898002625, + "learning_rate": 4.856499702333944e-06, + "loss": 0.6837, + "step": 874 + }, + { + "epoch": 0.7247377139701823, + "grad_norm": 0.921416699886322, + "learning_rate": 4.8561322613117945e-06, + "loss": 0.6709, + "step": 875 + }, + { + "epoch": 0.725565985643291, + "grad_norm": 0.8988221287727356, + "learning_rate": 4.855764364400016e-06, + "loss": 0.7111, + "step": 876 + }, + { + "epoch": 0.7263942573163997, + "grad_norm": 0.8838291168212891, + "learning_rate": 4.8553960116697915e-06, + "loss": 0.6821, + "step": 877 + }, + { + "epoch": 0.7272225289895086, + "grad_norm": 0.9051623940467834, + "learning_rate": 4.8550272031923965e-06, + "loss": 0.6848, + "step": 878 + }, + { + "epoch": 0.7280508006626173, + "grad_norm": 0.9308082461357117, + "learning_rate": 4.85465793903919e-06, + "loss": 0.7087, + "step": 879 + }, + { + "epoch": 0.7288790723357261, + "grad_norm": 0.873214066028595, + "learning_rate": 4.854288219281623e-06, + "loss": 0.6877, + "step": 880 + }, + { + "epoch": 0.7297073440088349, + "grad_norm": 0.8890054225921631, + "learning_rate": 4.853918043991232e-06, + "loss": 0.6719, + "step": 881 + }, + { + "epoch": 0.7305356156819437, + "grad_norm": 0.9580398201942444, + "learning_rate": 4.853547413239642e-06, + "loss": 0.6439, + "step": 882 + }, + { + "epoch": 0.7313638873550524, + "grad_norm": 0.875035285949707, + "learning_rate": 4.853176327098567e-06, + "loss": 0.6645, + "step": 883 + }, + { + "epoch": 0.7321921590281613, + "grad_norm": 0.9299217462539673, + "learning_rate": 4.85280478563981e-06, + "loss": 0.707, + "step": 884 + }, + { + "epoch": 0.73302043070127, + "grad_norm": 0.8840609192848206, + "learning_rate": 4.8524327889352585e-06, + "loss": 0.6789, + "step": 885 + }, + { + "epoch": 0.7338487023743788, + "grad_norm": 0.8983805179595947, + "learning_rate": 4.852060337056891e-06, + "loss": 0.6814, + "step": 886 + }, + { + "epoch": 0.7346769740474876, + "grad_norm": 0.8730597496032715, + "learning_rate": 4.851687430076775e-06, + "loss": 0.7006, + "step": 887 + }, + { + "epoch": 0.7355052457205964, + "grad_norm": 0.8980515003204346, + "learning_rate": 4.851314068067063e-06, + "loss": 0.6729, + "step": 888 + }, + { + "epoch": 0.7363335173937051, + "grad_norm": 0.8919468522071838, + "learning_rate": 4.850940251099997e-06, + "loss": 0.6764, + "step": 889 + }, + { + "epoch": 0.737161789066814, + "grad_norm": 0.8987070918083191, + "learning_rate": 4.850565979247908e-06, + "loss": 0.6997, + "step": 890 + }, + { + "epoch": 0.7379900607399227, + "grad_norm": 0.9056184887886047, + "learning_rate": 4.850191252583214e-06, + "loss": 0.6887, + "step": 891 + }, + { + "epoch": 0.7388183324130315, + "grad_norm": 0.8714621067047119, + "learning_rate": 4.84981607117842e-06, + "loss": 0.6946, + "step": 892 + }, + { + "epoch": 0.7396466040861402, + "grad_norm": 0.8845617175102234, + "learning_rate": 4.849440435106122e-06, + "loss": 0.6845, + "step": 893 + }, + { + "epoch": 0.7404748757592491, + "grad_norm": 0.8926048278808594, + "learning_rate": 4.849064344439e-06, + "loss": 0.6922, + "step": 894 + }, + { + "epoch": 0.7413031474323578, + "grad_norm": 0.8731569647789001, + "learning_rate": 4.848687799249826e-06, + "loss": 0.6837, + "step": 895 + }, + { + "epoch": 0.7421314191054665, + "grad_norm": 0.8801183104515076, + "learning_rate": 4.848310799611457e-06, + "loss": 0.6932, + "step": 896 + }, + { + "epoch": 0.7429596907785754, + "grad_norm": 0.8818139433860779, + "learning_rate": 4.847933345596838e-06, + "loss": 0.6758, + "step": 897 + }, + { + "epoch": 0.7437879624516841, + "grad_norm": 0.9117923974990845, + "learning_rate": 4.847555437279004e-06, + "loss": 0.6628, + "step": 898 + }, + { + "epoch": 0.7446162341247929, + "grad_norm": 0.891360342502594, + "learning_rate": 4.847177074731078e-06, + "loss": 0.6876, + "step": 899 + }, + { + "epoch": 0.7454445057979017, + "grad_norm": 0.8905263543128967, + "learning_rate": 4.846798258026267e-06, + "loss": 0.6862, + "step": 900 + }, + { + "epoch": 0.7462727774710105, + "grad_norm": 0.8853707909584045, + "learning_rate": 4.846418987237868e-06, + "loss": 0.654, + "step": 901 + }, + { + "epoch": 0.7471010491441192, + "grad_norm": 0.8782089948654175, + "learning_rate": 4.846039262439271e-06, + "loss": 0.7005, + "step": 902 + }, + { + "epoch": 0.7479293208172281, + "grad_norm": 0.8774383068084717, + "learning_rate": 4.8456590837039445e-06, + "loss": 0.7044, + "step": 903 + }, + { + "epoch": 0.7487575924903368, + "grad_norm": 0.9385423064231873, + "learning_rate": 4.845278451105451e-06, + "loss": 0.676, + "step": 904 + }, + { + "epoch": 0.7495858641634456, + "grad_norm": 0.9211432337760925, + "learning_rate": 4.8448973647174405e-06, + "loss": 0.697, + "step": 905 + }, + { + "epoch": 0.7504141358365544, + "grad_norm": 0.887610912322998, + "learning_rate": 4.844515824613649e-06, + "loss": 0.6965, + "step": 906 + }, + { + "epoch": 0.7512424075096632, + "grad_norm": 0.885761559009552, + "learning_rate": 4.8441338308679e-06, + "loss": 0.6851, + "step": 907 + }, + { + "epoch": 0.7520706791827719, + "grad_norm": 0.9209930300712585, + "learning_rate": 4.843751383554106e-06, + "loss": 0.6718, + "step": 908 + }, + { + "epoch": 0.7528989508558808, + "grad_norm": 0.9308252334594727, + "learning_rate": 4.843368482746269e-06, + "loss": 0.6802, + "step": 909 + }, + { + "epoch": 0.7537272225289895, + "grad_norm": 0.8931897282600403, + "learning_rate": 4.842985128518473e-06, + "loss": 0.7076, + "step": 910 + }, + { + "epoch": 0.7545554942020983, + "grad_norm": 0.9102733731269836, + "learning_rate": 4.842601320944898e-06, + "loss": 0.6898, + "step": 911 + }, + { + "epoch": 0.7553837658752071, + "grad_norm": 0.9336893558502197, + "learning_rate": 4.842217060099804e-06, + "loss": 0.6709, + "step": 912 + }, + { + "epoch": 0.7562120375483159, + "grad_norm": 0.9366991519927979, + "learning_rate": 4.8418323460575425e-06, + "loss": 0.6665, + "step": 913 + }, + { + "epoch": 0.7570403092214246, + "grad_norm": 0.9371415376663208, + "learning_rate": 4.841447178892552e-06, + "loss": 0.7013, + "step": 914 + }, + { + "epoch": 0.7578685808945335, + "grad_norm": 1.031123399734497, + "learning_rate": 4.84106155867936e-06, + "loss": 0.687, + "step": 915 + }, + { + "epoch": 0.7586968525676422, + "grad_norm": 0.9319218397140503, + "learning_rate": 4.8406754854925785e-06, + "loss": 0.669, + "step": 916 + }, + { + "epoch": 0.7595251242407509, + "grad_norm": 0.8563560843467712, + "learning_rate": 4.840288959406911e-06, + "loss": 0.6905, + "step": 917 + }, + { + "epoch": 0.7603533959138598, + "grad_norm": 0.9424378275871277, + "learning_rate": 4.839901980497145e-06, + "loss": 0.6685, + "step": 918 + }, + { + "epoch": 0.7611816675869685, + "grad_norm": 0.9202269315719604, + "learning_rate": 4.839514548838158e-06, + "loss": 0.6728, + "step": 919 + }, + { + "epoch": 0.7620099392600773, + "grad_norm": 0.9152582287788391, + "learning_rate": 4.839126664504915e-06, + "loss": 0.6842, + "step": 920 + }, + { + "epoch": 0.762838210933186, + "grad_norm": 0.9457129836082458, + "learning_rate": 4.838738327572468e-06, + "loss": 0.6822, + "step": 921 + }, + { + "epoch": 0.7636664826062949, + "grad_norm": 0.93124920129776, + "learning_rate": 4.838349538115955e-06, + "loss": 0.691, + "step": 922 + }, + { + "epoch": 0.7644947542794036, + "grad_norm": 0.9249301552772522, + "learning_rate": 4.837960296210605e-06, + "loss": 0.6574, + "step": 923 + }, + { + "epoch": 0.7653230259525124, + "grad_norm": 1.0551042556762695, + "learning_rate": 4.837570601931731e-06, + "loss": 0.6782, + "step": 924 + }, + { + "epoch": 0.7661512976256212, + "grad_norm": 0.9404964447021484, + "learning_rate": 4.8371804553547365e-06, + "loss": 0.6703, + "step": 925 + }, + { + "epoch": 0.76697956929873, + "grad_norm": 0.909015417098999, + "learning_rate": 4.836789856555111e-06, + "loss": 0.6812, + "step": 926 + }, + { + "epoch": 0.7678078409718387, + "grad_norm": 0.8572261929512024, + "learning_rate": 4.836398805608431e-06, + "loss": 0.6831, + "step": 927 + }, + { + "epoch": 0.7686361126449476, + "grad_norm": 0.951753556728363, + "learning_rate": 4.836007302590362e-06, + "loss": 0.6672, + "step": 928 + }, + { + "epoch": 0.7694643843180563, + "grad_norm": 0.8875932693481445, + "learning_rate": 4.8356153475766564e-06, + "loss": 0.6819, + "step": 929 + }, + { + "epoch": 0.7702926559911651, + "grad_norm": 0.9405681490898132, + "learning_rate": 4.835222940643153e-06, + "loss": 0.6917, + "step": 930 + }, + { + "epoch": 0.7711209276642739, + "grad_norm": 0.942520022392273, + "learning_rate": 4.834830081865779e-06, + "loss": 0.6657, + "step": 931 + }, + { + "epoch": 0.7719491993373827, + "grad_norm": 0.9063259363174438, + "learning_rate": 4.83443677132055e-06, + "loss": 0.6857, + "step": 932 + }, + { + "epoch": 0.7727774710104914, + "grad_norm": 0.8778199553489685, + "learning_rate": 4.834043009083566e-06, + "loss": 0.7028, + "step": 933 + }, + { + "epoch": 0.7736057426836003, + "grad_norm": 0.9215186834335327, + "learning_rate": 4.833648795231018e-06, + "loss": 0.6712, + "step": 934 + }, + { + "epoch": 0.774434014356709, + "grad_norm": 0.8992031812667847, + "learning_rate": 4.8332541298391825e-06, + "loss": 0.6827, + "step": 935 + }, + { + "epoch": 0.7752622860298177, + "grad_norm": 0.9203833341598511, + "learning_rate": 4.8328590129844224e-06, + "loss": 0.6835, + "step": 936 + }, + { + "epoch": 0.7760905577029266, + "grad_norm": 0.9109310507774353, + "learning_rate": 4.832463444743191e-06, + "loss": 0.673, + "step": 937 + }, + { + "epoch": 0.7769188293760353, + "grad_norm": 0.978349506855011, + "learning_rate": 4.832067425192024e-06, + "loss": 0.6638, + "step": 938 + }, + { + "epoch": 0.7777471010491441, + "grad_norm": 0.9552261829376221, + "learning_rate": 4.831670954407551e-06, + "loss": 0.6919, + "step": 939 + }, + { + "epoch": 0.7785753727222529, + "grad_norm": 0.9106539487838745, + "learning_rate": 4.831274032466484e-06, + "loss": 0.6964, + "step": 940 + }, + { + "epoch": 0.7794036443953617, + "grad_norm": 0.9195299744606018, + "learning_rate": 4.830876659445623e-06, + "loss": 0.6793, + "step": 941 + }, + { + "epoch": 0.7802319160684704, + "grad_norm": 0.9075512290000916, + "learning_rate": 4.8304788354218564e-06, + "loss": 0.6649, + "step": 942 + }, + { + "epoch": 0.7810601877415793, + "grad_norm": 0.9142526984214783, + "learning_rate": 4.83008056047216e-06, + "loss": 0.6811, + "step": 943 + }, + { + "epoch": 0.781888459414688, + "grad_norm": 0.8898900151252747, + "learning_rate": 4.829681834673597e-06, + "loss": 0.6906, + "step": 944 + }, + { + "epoch": 0.7827167310877968, + "grad_norm": 0.8884983062744141, + "learning_rate": 4.8292826581033135e-06, + "loss": 0.6903, + "step": 945 + }, + { + "epoch": 0.7835450027609056, + "grad_norm": 0.9075689315795898, + "learning_rate": 4.828883030838551e-06, + "loss": 0.6794, + "step": 946 + }, + { + "epoch": 0.7843732744340144, + "grad_norm": 0.8970300555229187, + "learning_rate": 4.828482952956631e-06, + "loss": 0.6715, + "step": 947 + }, + { + "epoch": 0.7852015461071231, + "grad_norm": 0.8906447887420654, + "learning_rate": 4.828082424534965e-06, + "loss": 0.6878, + "step": 948 + }, + { + "epoch": 0.786029817780232, + "grad_norm": 0.9083748459815979, + "learning_rate": 4.827681445651052e-06, + "loss": 0.6787, + "step": 949 + }, + { + "epoch": 0.7868580894533407, + "grad_norm": 0.8975276947021484, + "learning_rate": 4.8272800163824784e-06, + "loss": 0.687, + "step": 950 + }, + { + "epoch": 0.7876863611264495, + "grad_norm": 0.9573839902877808, + "learning_rate": 4.8268781368069156e-06, + "loss": 0.6996, + "step": 951 + }, + { + "epoch": 0.7885146327995582, + "grad_norm": 0.913536787033081, + "learning_rate": 4.826475807002125e-06, + "loss": 0.6844, + "step": 952 + }, + { + "epoch": 0.7893429044726671, + "grad_norm": 0.9202503561973572, + "learning_rate": 4.826073027045952e-06, + "loss": 0.6702, + "step": 953 + }, + { + "epoch": 0.7901711761457758, + "grad_norm": 0.8800237774848938, + "learning_rate": 4.8256697970163315e-06, + "loss": 0.685, + "step": 954 + }, + { + "epoch": 0.7909994478188846, + "grad_norm": 0.8923048377037048, + "learning_rate": 4.825266116991285e-06, + "loss": 0.6839, + "step": 955 + }, + { + "epoch": 0.7918277194919934, + "grad_norm": 0.879743754863739, + "learning_rate": 4.824861987048922e-06, + "loss": 0.7112, + "step": 956 + }, + { + "epoch": 0.7926559911651021, + "grad_norm": 0.8575169444084167, + "learning_rate": 4.824457407267436e-06, + "loss": 0.671, + "step": 957 + }, + { + "epoch": 0.7934842628382109, + "grad_norm": 0.8851709961891174, + "learning_rate": 4.824052377725109e-06, + "loss": 0.669, + "step": 958 + }, + { + "epoch": 0.7943125345113197, + "grad_norm": 0.9084329605102539, + "learning_rate": 4.823646898500311e-06, + "loss": 0.6731, + "step": 959 + }, + { + "epoch": 0.7951408061844285, + "grad_norm": 0.908256471157074, + "learning_rate": 4.823240969671501e-06, + "loss": 0.688, + "step": 960 + }, + { + "epoch": 0.7959690778575372, + "grad_norm": 0.8993384838104248, + "learning_rate": 4.822834591317219e-06, + "loss": 0.6736, + "step": 961 + }, + { + "epoch": 0.7967973495306461, + "grad_norm": 0.893530011177063, + "learning_rate": 4.822427763516098e-06, + "loss": 0.6787, + "step": 962 + }, + { + "epoch": 0.7976256212037548, + "grad_norm": 0.9479735493659973, + "learning_rate": 4.822020486346852e-06, + "loss": 0.6842, + "step": 963 + }, + { + "epoch": 0.7984538928768636, + "grad_norm": 0.8716689348220825, + "learning_rate": 4.821612759888289e-06, + "loss": 0.6726, + "step": 964 + }, + { + "epoch": 0.7992821645499724, + "grad_norm": 0.8902738690376282, + "learning_rate": 4.821204584219299e-06, + "loss": 0.6672, + "step": 965 + }, + { + "epoch": 0.8001104362230812, + "grad_norm": 0.8713969588279724, + "learning_rate": 4.820795959418859e-06, + "loss": 0.6783, + "step": 966 + }, + { + "epoch": 0.8009387078961899, + "grad_norm": 0.891788899898529, + "learning_rate": 4.820386885566036e-06, + "loss": 0.6691, + "step": 967 + }, + { + "epoch": 0.8017669795692988, + "grad_norm": 0.8484700918197632, + "learning_rate": 4.81997736273998e-06, + "loss": 0.6766, + "step": 968 + }, + { + "epoch": 0.8025952512424075, + "grad_norm": 0.8931223154067993, + "learning_rate": 4.819567391019931e-06, + "loss": 0.6944, + "step": 969 + }, + { + "epoch": 0.8034235229155163, + "grad_norm": 0.9707916975021362, + "learning_rate": 4.819156970485216e-06, + "loss": 0.7076, + "step": 970 + }, + { + "epoch": 0.8042517945886251, + "grad_norm": 0.9172869920730591, + "learning_rate": 4.818746101215245e-06, + "loss": 0.6864, + "step": 971 + }, + { + "epoch": 0.8050800662617339, + "grad_norm": 0.8982006311416626, + "learning_rate": 4.818334783289518e-06, + "loss": 0.6685, + "step": 972 + }, + { + "epoch": 0.8059083379348426, + "grad_norm": 0.8960995078086853, + "learning_rate": 4.817923016787623e-06, + "loss": 0.6959, + "step": 973 + }, + { + "epoch": 0.8067366096079515, + "grad_norm": 0.8806949257850647, + "learning_rate": 4.81751080178923e-06, + "loss": 0.7028, + "step": 974 + }, + { + "epoch": 0.8075648812810602, + "grad_norm": 0.8798978924751282, + "learning_rate": 4.817098138374102e-06, + "loss": 0.6718, + "step": 975 + }, + { + "epoch": 0.8083931529541689, + "grad_norm": 0.9217897653579712, + "learning_rate": 4.8166850266220835e-06, + "loss": 0.6782, + "step": 976 + }, + { + "epoch": 0.8092214246272778, + "grad_norm": 0.8915667533874512, + "learning_rate": 4.8162714666131074e-06, + "loss": 0.6917, + "step": 977 + }, + { + "epoch": 0.8100496963003865, + "grad_norm": 0.8895093202590942, + "learning_rate": 4.815857458427195e-06, + "loss": 0.698, + "step": 978 + }, + { + "epoch": 0.8108779679734953, + "grad_norm": 0.9167448282241821, + "learning_rate": 4.815443002144453e-06, + "loss": 0.6796, + "step": 979 + }, + { + "epoch": 0.811706239646604, + "grad_norm": 0.913827657699585, + "learning_rate": 4.815028097845074e-06, + "loss": 0.6731, + "step": 980 + }, + { + "epoch": 0.8125345113197129, + "grad_norm": 0.9183399081230164, + "learning_rate": 4.814612745609338e-06, + "loss": 0.7076, + "step": 981 + }, + { + "epoch": 0.8133627829928216, + "grad_norm": 0.8812996745109558, + "learning_rate": 4.814196945517613e-06, + "loss": 0.6416, + "step": 982 + }, + { + "epoch": 0.8141910546659304, + "grad_norm": 0.9032955169677734, + "learning_rate": 4.813780697650351e-06, + "loss": 0.6799, + "step": 983 + }, + { + "epoch": 0.8150193263390392, + "grad_norm": 0.9380100965499878, + "learning_rate": 4.813364002088093e-06, + "loss": 0.6771, + "step": 984 + }, + { + "epoch": 0.815847598012148, + "grad_norm": 0.9470316171646118, + "learning_rate": 4.8129468589114655e-06, + "loss": 0.6851, + "step": 985 + }, + { + "epoch": 0.8166758696852567, + "grad_norm": 0.9401133060455322, + "learning_rate": 4.812529268201183e-06, + "loss": 0.676, + "step": 986 + }, + { + "epoch": 0.8175041413583656, + "grad_norm": 0.9379092454910278, + "learning_rate": 4.812111230038043e-06, + "loss": 0.672, + "step": 987 + }, + { + "epoch": 0.8183324130314743, + "grad_norm": 0.9591637253761292, + "learning_rate": 4.811692744502933e-06, + "loss": 0.6684, + "step": 988 + }, + { + "epoch": 0.8191606847045831, + "grad_norm": 0.8911624550819397, + "learning_rate": 4.8112738116768276e-06, + "loss": 0.6448, + "step": 989 + }, + { + "epoch": 0.8199889563776919, + "grad_norm": 0.9036282896995544, + "learning_rate": 4.810854431640784e-06, + "loss": 0.703, + "step": 990 + }, + { + "epoch": 0.8208172280508007, + "grad_norm": 0.9197635650634766, + "learning_rate": 4.81043460447595e-06, + "loss": 0.6655, + "step": 991 + }, + { + "epoch": 0.8216454997239094, + "grad_norm": 0.9548643827438354, + "learning_rate": 4.8100143302635575e-06, + "loss": 0.6892, + "step": 992 + }, + { + "epoch": 0.8224737713970183, + "grad_norm": 0.9139766097068787, + "learning_rate": 4.809593609084925e-06, + "loss": 0.6723, + "step": 993 + }, + { + "epoch": 0.823302043070127, + "grad_norm": 0.9067809581756592, + "learning_rate": 4.8091724410214595e-06, + "loss": 0.6649, + "step": 994 + }, + { + "epoch": 0.8241303147432358, + "grad_norm": 0.9705583453178406, + "learning_rate": 4.808750826154652e-06, + "loss": 0.6889, + "step": 995 + }, + { + "epoch": 0.8249585864163446, + "grad_norm": 0.9664806127548218, + "learning_rate": 4.808328764566082e-06, + "loss": 0.6758, + "step": 996 + }, + { + "epoch": 0.8257868580894533, + "grad_norm": 0.8840530514717102, + "learning_rate": 4.807906256337414e-06, + "loss": 0.6719, + "step": 997 + }, + { + "epoch": 0.8266151297625621, + "grad_norm": 0.8803462386131287, + "learning_rate": 4.8074833015504e-06, + "loss": 0.6903, + "step": 998 + }, + { + "epoch": 0.8274434014356709, + "grad_norm": 0.9404571652412415, + "learning_rate": 4.8070599002868756e-06, + "loss": 0.6917, + "step": 999 + }, + { + "epoch": 0.8282716731087797, + "grad_norm": 1.0206997394561768, + "learning_rate": 4.806636052628768e-06, + "loss": 0.6854, + "step": 1000 + }, + { + "epoch": 0.8290999447818884, + "grad_norm": 0.9294790029525757, + "learning_rate": 4.806211758658086e-06, + "loss": 0.6791, + "step": 1001 + }, + { + "epoch": 0.8299282164549973, + "grad_norm": 0.874011754989624, + "learning_rate": 4.805787018456927e-06, + "loss": 0.6915, + "step": 1002 + }, + { + "epoch": 0.830756488128106, + "grad_norm": 0.9098061323165894, + "learning_rate": 4.805361832107474e-06, + "loss": 0.6643, + "step": 1003 + }, + { + "epoch": 0.8315847598012148, + "grad_norm": 0.9783986210823059, + "learning_rate": 4.804936199691997e-06, + "loss": 0.6853, + "step": 1004 + }, + { + "epoch": 0.8324130314743236, + "grad_norm": 0.9316959381103516, + "learning_rate": 4.804510121292852e-06, + "loss": 0.6696, + "step": 1005 + }, + { + "epoch": 0.8332413031474324, + "grad_norm": 0.8930666446685791, + "learning_rate": 4.80408359699248e-06, + "loss": 0.6805, + "step": 1006 + }, + { + "epoch": 0.8340695748205411, + "grad_norm": 0.9417555332183838, + "learning_rate": 4.803656626873412e-06, + "loss": 0.66, + "step": 1007 + }, + { + "epoch": 0.83489784649365, + "grad_norm": 0.9435075521469116, + "learning_rate": 4.80322921101826e-06, + "loss": 0.6664, + "step": 1008 + }, + { + "epoch": 0.8357261181667587, + "grad_norm": 0.9090076684951782, + "learning_rate": 4.8028013495097255e-06, + "loss": 0.6595, + "step": 1009 + }, + { + "epoch": 0.8365543898398675, + "grad_norm": 0.9024788737297058, + "learning_rate": 4.802373042430597e-06, + "loss": 0.6672, + "step": 1010 + }, + { + "epoch": 0.8373826615129762, + "grad_norm": 0.899813711643219, + "learning_rate": 4.801944289863747e-06, + "loss": 0.6743, + "step": 1011 + }, + { + "epoch": 0.8382109331860851, + "grad_norm": 0.887106716632843, + "learning_rate": 4.801515091892136e-06, + "loss": 0.6823, + "step": 1012 + }, + { + "epoch": 0.8390392048591938, + "grad_norm": 0.8690978288650513, + "learning_rate": 4.801085448598809e-06, + "loss": 0.6642, + "step": 1013 + }, + { + "epoch": 0.8398674765323026, + "grad_norm": 0.9131479859352112, + "learning_rate": 4.800655360066898e-06, + "loss": 0.6841, + "step": 1014 + }, + { + "epoch": 0.8406957482054114, + "grad_norm": 0.9031343460083008, + "learning_rate": 4.800224826379622e-06, + "loss": 0.6879, + "step": 1015 + }, + { + "epoch": 0.8415240198785201, + "grad_norm": 0.8914244174957275, + "learning_rate": 4.799793847620283e-06, + "loss": 0.6562, + "step": 1016 + }, + { + "epoch": 0.8423522915516289, + "grad_norm": 0.8692408800125122, + "learning_rate": 4.799362423872274e-06, + "loss": 0.6759, + "step": 1017 + }, + { + "epoch": 0.8431805632247377, + "grad_norm": 0.9511806964874268, + "learning_rate": 4.798930555219069e-06, + "loss": 0.6637, + "step": 1018 + }, + { + "epoch": 0.8440088348978465, + "grad_norm": 0.9556595683097839, + "learning_rate": 4.7984982417442325e-06, + "loss": 0.6925, + "step": 1019 + }, + { + "epoch": 0.8448371065709552, + "grad_norm": 0.9107236862182617, + "learning_rate": 4.798065483531413e-06, + "loss": 0.67, + "step": 1020 + }, + { + "epoch": 0.8456653782440641, + "grad_norm": 0.875293493270874, + "learning_rate": 4.797632280664345e-06, + "loss": 0.6755, + "step": 1021 + }, + { + "epoch": 0.8464936499171728, + "grad_norm": 0.8972271084785461, + "learning_rate": 4.797198633226849e-06, + "loss": 0.6637, + "step": 1022 + }, + { + "epoch": 0.8473219215902816, + "grad_norm": 0.8903595805168152, + "learning_rate": 4.796764541302831e-06, + "loss": 0.6997, + "step": 1023 + }, + { + "epoch": 0.8481501932633904, + "grad_norm": 0.8948636651039124, + "learning_rate": 4.796330004976285e-06, + "loss": 0.6579, + "step": 1024 + }, + { + "epoch": 0.8489784649364992, + "grad_norm": 0.8590925335884094, + "learning_rate": 4.795895024331289e-06, + "loss": 0.6651, + "step": 1025 + }, + { + "epoch": 0.8498067366096079, + "grad_norm": 0.870811402797699, + "learning_rate": 4.795459599452007e-06, + "loss": 0.6707, + "step": 1026 + }, + { + "epoch": 0.8506350082827168, + "grad_norm": 0.8807471394538879, + "learning_rate": 4.795023730422692e-06, + "loss": 0.6787, + "step": 1027 + }, + { + "epoch": 0.8514632799558255, + "grad_norm": 0.8817671537399292, + "learning_rate": 4.794587417327678e-06, + "loss": 0.66, + "step": 1028 + }, + { + "epoch": 0.8522915516289343, + "grad_norm": 0.8924570679664612, + "learning_rate": 4.794150660251389e-06, + "loss": 0.6842, + "step": 1029 + }, + { + "epoch": 0.8531198233020431, + "grad_norm": 0.8746559619903564, + "learning_rate": 4.793713459278333e-06, + "loss": 0.6582, + "step": 1030 + }, + { + "epoch": 0.8539480949751519, + "grad_norm": 0.8885277509689331, + "learning_rate": 4.793275814493104e-06, + "loss": 0.6784, + "step": 1031 + }, + { + "epoch": 0.8547763666482606, + "grad_norm": 0.8797710537910461, + "learning_rate": 4.792837725980383e-06, + "loss": 0.6638, + "step": 1032 + }, + { + "epoch": 0.8556046383213695, + "grad_norm": 0.8932235240936279, + "learning_rate": 4.7923991938249355e-06, + "loss": 0.6918, + "step": 1033 + }, + { + "epoch": 0.8564329099944782, + "grad_norm": 0.9317487478256226, + "learning_rate": 4.791960218111613e-06, + "loss": 0.6905, + "step": 1034 + }, + { + "epoch": 0.8572611816675869, + "grad_norm": 0.8760019540786743, + "learning_rate": 4.791520798925354e-06, + "loss": 0.673, + "step": 1035 + }, + { + "epoch": 0.8580894533406958, + "grad_norm": 0.8644204139709473, + "learning_rate": 4.7910809363511825e-06, + "loss": 0.6755, + "step": 1036 + }, + { + "epoch": 0.8589177250138045, + "grad_norm": 0.8761067390441895, + "learning_rate": 4.790640630474206e-06, + "loss": 0.699, + "step": 1037 + }, + { + "epoch": 0.8597459966869133, + "grad_norm": 0.9135014414787292, + "learning_rate": 4.790199881379621e-06, + "loss": 0.6659, + "step": 1038 + }, + { + "epoch": 0.860574268360022, + "grad_norm": 0.9065434336662292, + "learning_rate": 4.789758689152708e-06, + "loss": 0.6664, + "step": 1039 + }, + { + "epoch": 0.8614025400331309, + "grad_norm": 0.8814099431037903, + "learning_rate": 4.789317053878833e-06, + "loss": 0.681, + "step": 1040 + }, + { + "epoch": 0.8622308117062396, + "grad_norm": 0.872502863407135, + "learning_rate": 4.788874975643449e-06, + "loss": 0.7044, + "step": 1041 + }, + { + "epoch": 0.8630590833793484, + "grad_norm": 0.9532367587089539, + "learning_rate": 4.788432454532093e-06, + "loss": 0.655, + "step": 1042 + }, + { + "epoch": 0.8638873550524572, + "grad_norm": 0.9009373784065247, + "learning_rate": 4.787989490630391e-06, + "loss": 0.6946, + "step": 1043 + }, + { + "epoch": 0.864715626725566, + "grad_norm": 0.889933168888092, + "learning_rate": 4.787546084024051e-06, + "loss": 0.6595, + "step": 1044 + }, + { + "epoch": 0.8655438983986747, + "grad_norm": 0.9179873466491699, + "learning_rate": 4.787102234798867e-06, + "loss": 0.658, + "step": 1045 + }, + { + "epoch": 0.8663721700717836, + "grad_norm": 0.8873292207717896, + "learning_rate": 4.786657943040721e-06, + "loss": 0.7023, + "step": 1046 + }, + { + "epoch": 0.8672004417448923, + "grad_norm": 0.8388610482215881, + "learning_rate": 4.78621320883558e-06, + "loss": 0.6644, + "step": 1047 + }, + { + "epoch": 0.8680287134180011, + "grad_norm": 0.8971349596977234, + "learning_rate": 4.785768032269493e-06, + "loss": 0.6552, + "step": 1048 + }, + { + "epoch": 0.8688569850911099, + "grad_norm": 0.8789131045341492, + "learning_rate": 4.785322413428601e-06, + "loss": 0.6681, + "step": 1049 + }, + { + "epoch": 0.8696852567642187, + "grad_norm": 0.8867958784103394, + "learning_rate": 4.784876352399125e-06, + "loss": 0.6859, + "step": 1050 + }, + { + "epoch": 0.8705135284373274, + "grad_norm": 0.8584237694740295, + "learning_rate": 4.784429849267373e-06, + "loss": 0.6655, + "step": 1051 + }, + { + "epoch": 0.8713418001104363, + "grad_norm": 0.9230811595916748, + "learning_rate": 4.783982904119743e-06, + "loss": 0.6648, + "step": 1052 + }, + { + "epoch": 0.872170071783545, + "grad_norm": 0.9217789769172668, + "learning_rate": 4.783535517042709e-06, + "loss": 0.6867, + "step": 1053 + }, + { + "epoch": 0.8729983434566538, + "grad_norm": 0.9209116697311401, + "learning_rate": 4.783087688122842e-06, + "loss": 0.6972, + "step": 1054 + }, + { + "epoch": 0.8738266151297626, + "grad_norm": 0.9493632316589355, + "learning_rate": 4.782639417446789e-06, + "loss": 0.6785, + "step": 1055 + }, + { + "epoch": 0.8746548868028713, + "grad_norm": 0.8790786862373352, + "learning_rate": 4.782190705101286e-06, + "loss": 0.6925, + "step": 1056 + }, + { + "epoch": 0.8754831584759801, + "grad_norm": 0.8662691712379456, + "learning_rate": 4.781741551173157e-06, + "loss": 0.6686, + "step": 1057 + }, + { + "epoch": 0.8763114301490889, + "grad_norm": 0.919658899307251, + "learning_rate": 4.781291955749307e-06, + "loss": 0.6814, + "step": 1058 + }, + { + "epoch": 0.8771397018221977, + "grad_norm": 0.9639708399772644, + "learning_rate": 4.780841918916731e-06, + "loss": 0.6629, + "step": 1059 + }, + { + "epoch": 0.8779679734953064, + "grad_norm": 0.8694306015968323, + "learning_rate": 4.780391440762505e-06, + "loss": 0.6893, + "step": 1060 + }, + { + "epoch": 0.8787962451684153, + "grad_norm": 0.9725603461265564, + "learning_rate": 4.779940521373793e-06, + "loss": 0.67, + "step": 1061 + }, + { + "epoch": 0.879624516841524, + "grad_norm": 0.93159419298172, + "learning_rate": 4.779489160837842e-06, + "loss": 0.686, + "step": 1062 + }, + { + "epoch": 0.8804527885146328, + "grad_norm": 0.9366991519927979, + "learning_rate": 4.779037359241989e-06, + "loss": 0.649, + "step": 1063 + }, + { + "epoch": 0.8812810601877415, + "grad_norm": 0.9052289724349976, + "learning_rate": 4.7785851166736505e-06, + "loss": 0.6677, + "step": 1064 + }, + { + "epoch": 0.8821093318608504, + "grad_norm": 0.882251501083374, + "learning_rate": 4.778132433220334e-06, + "loss": 0.6886, + "step": 1065 + }, + { + "epoch": 0.8829376035339591, + "grad_norm": 0.8867698907852173, + "learning_rate": 4.777679308969627e-06, + "loss": 0.6645, + "step": 1066 + }, + { + "epoch": 0.883765875207068, + "grad_norm": 0.890396773815155, + "learning_rate": 4.777225744009207e-06, + "loss": 0.6684, + "step": 1067 + }, + { + "epoch": 0.8845941468801767, + "grad_norm": 0.8827376365661621, + "learning_rate": 4.776771738426833e-06, + "loss": 0.6738, + "step": 1068 + }, + { + "epoch": 0.8854224185532855, + "grad_norm": 0.8953343033790588, + "learning_rate": 4.776317292310352e-06, + "loss": 0.684, + "step": 1069 + }, + { + "epoch": 0.8862506902263942, + "grad_norm": 0.8751050233840942, + "learning_rate": 4.775862405747694e-06, + "loss": 0.6779, + "step": 1070 + }, + { + "epoch": 0.8870789618995031, + "grad_norm": 0.9136618971824646, + "learning_rate": 4.775407078826876e-06, + "loss": 0.6939, + "step": 1071 + }, + { + "epoch": 0.8879072335726118, + "grad_norm": 0.8887941837310791, + "learning_rate": 4.774951311635999e-06, + "loss": 0.6547, + "step": 1072 + }, + { + "epoch": 0.8887355052457206, + "grad_norm": 0.8873274326324463, + "learning_rate": 4.77449510426325e-06, + "loss": 0.6769, + "step": 1073 + }, + { + "epoch": 0.8895637769188294, + "grad_norm": 0.9307074546813965, + "learning_rate": 4.774038456796901e-06, + "loss": 0.6984, + "step": 1074 + }, + { + "epoch": 0.8903920485919381, + "grad_norm": 0.9022747874259949, + "learning_rate": 4.77358136932531e-06, + "loss": 0.6733, + "step": 1075 + }, + { + "epoch": 0.8912203202650469, + "grad_norm": 0.8779094219207764, + "learning_rate": 4.773123841936916e-06, + "loss": 0.6434, + "step": 1076 + }, + { + "epoch": 0.8920485919381557, + "grad_norm": 0.9465747475624084, + "learning_rate": 4.7726658747202505e-06, + "loss": 0.6761, + "step": 1077 + }, + { + "epoch": 0.8928768636112645, + "grad_norm": 0.8694362044334412, + "learning_rate": 4.772207467763923e-06, + "loss": 0.6756, + "step": 1078 + }, + { + "epoch": 0.8937051352843732, + "grad_norm": 0.8839150071144104, + "learning_rate": 4.771748621156631e-06, + "loss": 0.6714, + "step": 1079 + }, + { + "epoch": 0.8945334069574821, + "grad_norm": 0.9220975041389465, + "learning_rate": 4.7712893349871585e-06, + "loss": 0.6724, + "step": 1080 + }, + { + "epoch": 0.8953616786305908, + "grad_norm": 0.9234760999679565, + "learning_rate": 4.770829609344372e-06, + "loss": 0.67, + "step": 1081 + }, + { + "epoch": 0.8961899503036996, + "grad_norm": 0.9262274503707886, + "learning_rate": 4.770369444317225e-06, + "loss": 0.6619, + "step": 1082 + }, + { + "epoch": 0.8970182219768084, + "grad_norm": 0.8889190554618835, + "learning_rate": 4.769908839994753e-06, + "loss": 0.688, + "step": 1083 + }, + { + "epoch": 0.8978464936499172, + "grad_norm": 0.9028982520103455, + "learning_rate": 4.769447796466082e-06, + "loss": 0.6633, + "step": 1084 + }, + { + "epoch": 0.8986747653230259, + "grad_norm": 0.9155063033103943, + "learning_rate": 4.7689863138204185e-06, + "loss": 0.6709, + "step": 1085 + }, + { + "epoch": 0.8995030369961348, + "grad_norm": 0.9066855311393738, + "learning_rate": 4.768524392147052e-06, + "loss": 0.6713, + "step": 1086 + }, + { + "epoch": 0.9003313086692435, + "grad_norm": 0.9365159273147583, + "learning_rate": 4.768062031535365e-06, + "loss": 0.6675, + "step": 1087 + }, + { + "epoch": 0.9011595803423523, + "grad_norm": 0.9106162190437317, + "learning_rate": 4.767599232074816e-06, + "loss": 0.6804, + "step": 1088 + }, + { + "epoch": 0.9019878520154611, + "grad_norm": 0.9183253645896912, + "learning_rate": 4.767135993854955e-06, + "loss": 0.658, + "step": 1089 + }, + { + "epoch": 0.9028161236885699, + "grad_norm": 0.9543409943580627, + "learning_rate": 4.766672316965413e-06, + "loss": 0.696, + "step": 1090 + }, + { + "epoch": 0.9036443953616786, + "grad_norm": 0.8894531726837158, + "learning_rate": 4.766208201495906e-06, + "loss": 0.6781, + "step": 1091 + }, + { + "epoch": 0.9044726670347875, + "grad_norm": 0.915542721748352, + "learning_rate": 4.765743647536238e-06, + "loss": 0.6804, + "step": 1092 + }, + { + "epoch": 0.9053009387078962, + "grad_norm": 0.8959763646125793, + "learning_rate": 4.765278655176296e-06, + "loss": 0.6873, + "step": 1093 + }, + { + "epoch": 0.906129210381005, + "grad_norm": 0.881188690662384, + "learning_rate": 4.76481322450605e-06, + "loss": 0.6582, + "step": 1094 + }, + { + "epoch": 0.9069574820541138, + "grad_norm": 0.9096754193305969, + "learning_rate": 4.764347355615558e-06, + "loss": 0.693, + "step": 1095 + }, + { + "epoch": 0.9077857537272225, + "grad_norm": 0.891132652759552, + "learning_rate": 4.76388104859496e-06, + "loss": 0.6781, + "step": 1096 + }, + { + "epoch": 0.9086140254003313, + "grad_norm": 0.893624484539032, + "learning_rate": 4.763414303534484e-06, + "loss": 0.6636, + "step": 1097 + }, + { + "epoch": 0.90944229707344, + "grad_norm": 0.8795678019523621, + "learning_rate": 4.762947120524438e-06, + "loss": 0.6604, + "step": 1098 + }, + { + "epoch": 0.9102705687465489, + "grad_norm": 0.9177265763282776, + "learning_rate": 4.76247949965522e-06, + "loss": 0.6557, + "step": 1099 + }, + { + "epoch": 0.9110988404196576, + "grad_norm": 0.9416985511779785, + "learning_rate": 4.762011441017309e-06, + "loss": 0.7008, + "step": 1100 + }, + { + "epoch": 0.9119271120927664, + "grad_norm": 0.9255479574203491, + "learning_rate": 4.7615429447012704e-06, + "loss": 0.6665, + "step": 1101 + }, + { + "epoch": 0.9127553837658752, + "grad_norm": 0.8948953151702881, + "learning_rate": 4.761074010797754e-06, + "loss": 0.6699, + "step": 1102 + }, + { + "epoch": 0.913583655438984, + "grad_norm": 0.9027442336082458, + "learning_rate": 4.760604639397493e-06, + "loss": 0.6649, + "step": 1103 + }, + { + "epoch": 0.9144119271120927, + "grad_norm": 0.9094238877296448, + "learning_rate": 4.7601348305913085e-06, + "loss": 0.6753, + "step": 1104 + }, + { + "epoch": 0.9152401987852016, + "grad_norm": 0.9098204374313354, + "learning_rate": 4.759664584470103e-06, + "loss": 0.6845, + "step": 1105 + }, + { + "epoch": 0.9160684704583103, + "grad_norm": 0.9339573383331299, + "learning_rate": 4.759193901124864e-06, + "loss": 0.6285, + "step": 1106 + }, + { + "epoch": 0.9168967421314191, + "grad_norm": 0.8812451958656311, + "learning_rate": 4.758722780646666e-06, + "loss": 0.6764, + "step": 1107 + }, + { + "epoch": 0.9177250138045279, + "grad_norm": 0.8832173347473145, + "learning_rate": 4.758251223126665e-06, + "loss": 0.6908, + "step": 1108 + }, + { + "epoch": 0.9185532854776367, + "grad_norm": 0.9124787449836731, + "learning_rate": 4.757779228656103e-06, + "loss": 0.6855, + "step": 1109 + }, + { + "epoch": 0.9193815571507454, + "grad_norm": 0.8778328895568848, + "learning_rate": 4.757306797326309e-06, + "loss": 0.6404, + "step": 1110 + }, + { + "epoch": 0.9202098288238543, + "grad_norm": 0.8831816911697388, + "learning_rate": 4.75683392922869e-06, + "loss": 0.6425, + "step": 1111 + }, + { + "epoch": 0.921038100496963, + "grad_norm": 0.9060006141662598, + "learning_rate": 4.756360624454744e-06, + "loss": 0.6738, + "step": 1112 + }, + { + "epoch": 0.9218663721700718, + "grad_norm": 0.8746777176856995, + "learning_rate": 4.755886883096051e-06, + "loss": 0.6665, + "step": 1113 + }, + { + "epoch": 0.9226946438431806, + "grad_norm": 0.9185945987701416, + "learning_rate": 4.755412705244274e-06, + "loss": 0.6564, + "step": 1114 + }, + { + "epoch": 0.9235229155162893, + "grad_norm": 0.9648581743240356, + "learning_rate": 4.754938090991165e-06, + "loss": 0.6597, + "step": 1115 + }, + { + "epoch": 0.9243511871893981, + "grad_norm": 0.8800092935562134, + "learning_rate": 4.754463040428555e-06, + "loss": 0.6884, + "step": 1116 + }, + { + "epoch": 0.9251794588625069, + "grad_norm": 0.8659892082214355, + "learning_rate": 4.753987553648363e-06, + "loss": 0.68, + "step": 1117 + }, + { + "epoch": 0.9260077305356157, + "grad_norm": 0.88991379737854, + "learning_rate": 4.75351163074259e-06, + "loss": 0.668, + "step": 1118 + }, + { + "epoch": 0.9268360022087244, + "grad_norm": 0.9683054089546204, + "learning_rate": 4.753035271803323e-06, + "loss": 0.6789, + "step": 1119 + }, + { + "epoch": 0.9276642738818333, + "grad_norm": 0.928001880645752, + "learning_rate": 4.752558476922735e-06, + "loss": 0.6786, + "step": 1120 + }, + { + "epoch": 0.928492545554942, + "grad_norm": 0.8760973215103149, + "learning_rate": 4.752081246193078e-06, + "loss": 0.6716, + "step": 1121 + }, + { + "epoch": 0.9293208172280508, + "grad_norm": 0.9171329140663147, + "learning_rate": 4.751603579706695e-06, + "loss": 0.662, + "step": 1122 + }, + { + "epoch": 0.9301490889011595, + "grad_norm": 0.9380231499671936, + "learning_rate": 4.751125477556008e-06, + "loss": 0.6836, + "step": 1123 + }, + { + "epoch": 0.9309773605742684, + "grad_norm": 0.9471802711486816, + "learning_rate": 4.750646939833526e-06, + "loss": 0.668, + "step": 1124 + }, + { + "epoch": 0.9318056322473771, + "grad_norm": 0.8860809206962585, + "learning_rate": 4.750167966631842e-06, + "loss": 0.6635, + "step": 1125 + }, + { + "epoch": 0.932633903920486, + "grad_norm": 0.8694644570350647, + "learning_rate": 4.7496885580436305e-06, + "loss": 0.6505, + "step": 1126 + }, + { + "epoch": 0.9334621755935947, + "grad_norm": 0.9602686166763306, + "learning_rate": 4.749208714161655e-06, + "loss": 0.6874, + "step": 1127 + }, + { + "epoch": 0.9342904472667035, + "grad_norm": 0.9198856353759766, + "learning_rate": 4.748728435078761e-06, + "loss": 0.697, + "step": 1128 + }, + { + "epoch": 0.9351187189398122, + "grad_norm": 0.9162499904632568, + "learning_rate": 4.748247720887876e-06, + "loss": 0.6521, + "step": 1129 + }, + { + "epoch": 0.9359469906129211, + "grad_norm": 0.9064576029777527, + "learning_rate": 4.747766571682016e-06, + "loss": 0.6525, + "step": 1130 + }, + { + "epoch": 0.9367752622860298, + "grad_norm": 0.9518057703971863, + "learning_rate": 4.747284987554277e-06, + "loss": 0.6745, + "step": 1131 + }, + { + "epoch": 0.9376035339591386, + "grad_norm": 0.8963000178337097, + "learning_rate": 4.746802968597842e-06, + "loss": 0.6999, + "step": 1132 + }, + { + "epoch": 0.9384318056322474, + "grad_norm": 0.8687212467193604, + "learning_rate": 4.746320514905977e-06, + "loss": 0.6353, + "step": 1133 + }, + { + "epoch": 0.9392600773053562, + "grad_norm": 0.8598489165306091, + "learning_rate": 4.745837626572033e-06, + "loss": 0.6897, + "step": 1134 + }, + { + "epoch": 0.9400883489784649, + "grad_norm": 0.9425498843193054, + "learning_rate": 4.745354303689444e-06, + "loss": 0.667, + "step": 1135 + }, + { + "epoch": 0.9409166206515737, + "grad_norm": 0.9026349782943726, + "learning_rate": 4.744870546351727e-06, + "loss": 0.676, + "step": 1136 + }, + { + "epoch": 0.9417448923246825, + "grad_norm": 0.8824150562286377, + "learning_rate": 4.744386354652486e-06, + "loss": 0.6941, + "step": 1137 + }, + { + "epoch": 0.9425731639977912, + "grad_norm": 0.8910468816757202, + "learning_rate": 4.743901728685406e-06, + "loss": 0.6701, + "step": 1138 + }, + { + "epoch": 0.9434014356709001, + "grad_norm": 0.9114013314247131, + "learning_rate": 4.74341666854426e-06, + "loss": 0.6735, + "step": 1139 + }, + { + "epoch": 0.9442297073440088, + "grad_norm": 0.9107307195663452, + "learning_rate": 4.742931174322902e-06, + "loss": 0.6505, + "step": 1140 + }, + { + "epoch": 0.9450579790171176, + "grad_norm": 0.8906923532485962, + "learning_rate": 4.742445246115271e-06, + "loss": 0.6648, + "step": 1141 + }, + { + "epoch": 0.9458862506902264, + "grad_norm": 0.902644157409668, + "learning_rate": 4.741958884015387e-06, + "loss": 0.6674, + "step": 1142 + }, + { + "epoch": 0.9467145223633352, + "grad_norm": 0.9522087574005127, + "learning_rate": 4.741472088117358e-06, + "loss": 0.6825, + "step": 1143 + }, + { + "epoch": 0.9475427940364439, + "grad_norm": 0.9152255654335022, + "learning_rate": 4.740984858515376e-06, + "loss": 0.6708, + "step": 1144 + }, + { + "epoch": 0.9483710657095528, + "grad_norm": 0.8779835104942322, + "learning_rate": 4.740497195303713e-06, + "loss": 0.6562, + "step": 1145 + }, + { + "epoch": 0.9491993373826615, + "grad_norm": 0.8842780590057373, + "learning_rate": 4.740009098576729e-06, + "loss": 0.6769, + "step": 1146 + }, + { + "epoch": 0.9500276090557703, + "grad_norm": 0.9137415289878845, + "learning_rate": 4.739520568428866e-06, + "loss": 0.6705, + "step": 1147 + }, + { + "epoch": 0.950855880728879, + "grad_norm": 0.8867765665054321, + "learning_rate": 4.739031604954648e-06, + "loss": 0.659, + "step": 1148 + }, + { + "epoch": 0.9516841524019879, + "grad_norm": 0.8965260982513428, + "learning_rate": 4.7385422082486874e-06, + "loss": 0.6744, + "step": 1149 + }, + { + "epoch": 0.9525124240750966, + "grad_norm": 0.8861697912216187, + "learning_rate": 4.738052378405677e-06, + "loss": 0.6575, + "step": 1150 + }, + { + "epoch": 0.9533406957482055, + "grad_norm": 0.9442310333251953, + "learning_rate": 4.737562115520394e-06, + "loss": 0.7164, + "step": 1151 + }, + { + "epoch": 0.9541689674213142, + "grad_norm": 0.9340727925300598, + "learning_rate": 4.7370714196877e-06, + "loss": 0.6577, + "step": 1152 + }, + { + "epoch": 0.954997239094423, + "grad_norm": 0.9337487816810608, + "learning_rate": 4.7365802910025405e-06, + "loss": 0.6664, + "step": 1153 + }, + { + "epoch": 0.9558255107675317, + "grad_norm": 0.8849001526832581, + "learning_rate": 4.736088729559943e-06, + "loss": 0.6721, + "step": 1154 + }, + { + "epoch": 0.9566537824406405, + "grad_norm": 0.917323648929596, + "learning_rate": 4.735596735455022e-06, + "loss": 0.6715, + "step": 1155 + }, + { + "epoch": 0.9574820541137493, + "grad_norm": 0.9480959177017212, + "learning_rate": 4.735104308782972e-06, + "loss": 0.6784, + "step": 1156 + }, + { + "epoch": 0.958310325786858, + "grad_norm": 0.9519251585006714, + "learning_rate": 4.734611449639073e-06, + "loss": 0.6617, + "step": 1157 + }, + { + "epoch": 0.9591385974599669, + "grad_norm": 0.8652079105377197, + "learning_rate": 4.734118158118689e-06, + "loss": 0.6846, + "step": 1158 + }, + { + "epoch": 0.9599668691330756, + "grad_norm": 0.8860392570495605, + "learning_rate": 4.733624434317269e-06, + "loss": 0.6729, + "step": 1159 + }, + { + "epoch": 0.9607951408061844, + "grad_norm": 0.9467882513999939, + "learning_rate": 4.7331302783303416e-06, + "loss": 0.6362, + "step": 1160 + }, + { + "epoch": 0.9616234124792932, + "grad_norm": 0.9138529896736145, + "learning_rate": 4.732635690253523e-06, + "loss": 0.6613, + "step": 1161 + }, + { + "epoch": 0.962451684152402, + "grad_norm": 0.9401054978370667, + "learning_rate": 4.7321406701825105e-06, + "loss": 0.6845, + "step": 1162 + }, + { + "epoch": 0.9632799558255107, + "grad_norm": 0.9049288630485535, + "learning_rate": 4.731645218213086e-06, + "loss": 0.6857, + "step": 1163 + }, + { + "epoch": 0.9641082274986196, + "grad_norm": 0.8972534537315369, + "learning_rate": 4.731149334441114e-06, + "loss": 0.6745, + "step": 1164 + }, + { + "epoch": 0.9649364991717283, + "grad_norm": 1.0799483060836792, + "learning_rate": 4.730653018962545e-06, + "loss": 0.6754, + "step": 1165 + }, + { + "epoch": 0.9657647708448371, + "grad_norm": 0.9841833710670471, + "learning_rate": 4.730156271873412e-06, + "loss": 0.6718, + "step": 1166 + }, + { + "epoch": 0.9665930425179459, + "grad_norm": 0.9388216733932495, + "learning_rate": 4.729659093269828e-06, + "loss": 0.6622, + "step": 1167 + }, + { + "epoch": 0.9674213141910547, + "grad_norm": 0.8879573345184326, + "learning_rate": 4.729161483247994e-06, + "loss": 0.6909, + "step": 1168 + }, + { + "epoch": 0.9682495858641634, + "grad_norm": 0.9713517427444458, + "learning_rate": 4.728663441904194e-06, + "loss": 0.6956, + "step": 1169 + }, + { + "epoch": 0.9690778575372723, + "grad_norm": 0.9200823903083801, + "learning_rate": 4.7281649693347925e-06, + "loss": 0.6766, + "step": 1170 + }, + { + "epoch": 0.969906129210381, + "grad_norm": 0.8983714580535889, + "learning_rate": 4.727666065636241e-06, + "loss": 0.6626, + "step": 1171 + }, + { + "epoch": 0.9707344008834898, + "grad_norm": 0.8792792558670044, + "learning_rate": 4.72716673090507e-06, + "loss": 0.666, + "step": 1172 + }, + { + "epoch": 0.9715626725565986, + "grad_norm": 1.0116846561431885, + "learning_rate": 4.7266669652379005e-06, + "loss": 0.6574, + "step": 1173 + }, + { + "epoch": 0.9723909442297074, + "grad_norm": 0.9433228373527527, + "learning_rate": 4.7261667687314285e-06, + "loss": 0.6972, + "step": 1174 + }, + { + "epoch": 0.9732192159028161, + "grad_norm": 0.8979770541191101, + "learning_rate": 4.725666141482439e-06, + "loss": 0.6523, + "step": 1175 + }, + { + "epoch": 0.9740474875759249, + "grad_norm": 0.8990604281425476, + "learning_rate": 4.725165083587798e-06, + "loss": 0.6535, + "step": 1176 + }, + { + "epoch": 0.9748757592490337, + "grad_norm": 0.9411956667900085, + "learning_rate": 4.724663595144457e-06, + "loss": 0.6947, + "step": 1177 + }, + { + "epoch": 0.9757040309221424, + "grad_norm": 1.0203242301940918, + "learning_rate": 4.724161676249448e-06, + "loss": 0.6894, + "step": 1178 + }, + { + "epoch": 0.9765323025952513, + "grad_norm": 0.8938170075416565, + "learning_rate": 4.723659326999888e-06, + "loss": 0.6648, + "step": 1179 + }, + { + "epoch": 0.97736057426836, + "grad_norm": 0.8810412883758545, + "learning_rate": 4.7231565474929765e-06, + "loss": 0.6761, + "step": 1180 + }, + { + "epoch": 0.9781888459414688, + "grad_norm": 0.8903076648712158, + "learning_rate": 4.722653337825998e-06, + "loss": 0.6738, + "step": 1181 + }, + { + "epoch": 0.9790171176145775, + "grad_norm": 0.905060350894928, + "learning_rate": 4.7221496980963175e-06, + "loss": 0.6625, + "step": 1182 + }, + { + "epoch": 0.9798453892876864, + "grad_norm": 0.8791100978851318, + "learning_rate": 4.721645628401385e-06, + "loss": 0.6884, + "step": 1183 + }, + { + "epoch": 0.9806736609607951, + "grad_norm": 0.897101104259491, + "learning_rate": 4.721141128838733e-06, + "loss": 0.6693, + "step": 1184 + }, + { + "epoch": 0.981501932633904, + "grad_norm": 0.9102833867073059, + "learning_rate": 4.7206361995059775e-06, + "loss": 0.6682, + "step": 1185 + }, + { + "epoch": 0.9823302043070127, + "grad_norm": 0.9492197632789612, + "learning_rate": 4.720130840500819e-06, + "loss": 0.6501, + "step": 1186 + }, + { + "epoch": 0.9831584759801215, + "grad_norm": 0.9783265590667725, + "learning_rate": 4.719625051921038e-06, + "loss": 0.6454, + "step": 1187 + }, + { + "epoch": 0.9839867476532302, + "grad_norm": 0.8938109874725342, + "learning_rate": 4.719118833864501e-06, + "loss": 0.6671, + "step": 1188 + }, + { + "epoch": 0.9848150193263391, + "grad_norm": 0.897484302520752, + "learning_rate": 4.718612186429156e-06, + "loss": 0.6664, + "step": 1189 + }, + { + "epoch": 0.9856432909994478, + "grad_norm": 0.9640376567840576, + "learning_rate": 4.718105109713034e-06, + "loss": 0.6845, + "step": 1190 + }, + { + "epoch": 0.9864715626725566, + "grad_norm": 0.9495493769645691, + "learning_rate": 4.7175976038142505e-06, + "loss": 0.6864, + "step": 1191 + }, + { + "epoch": 0.9872998343456654, + "grad_norm": 0.9390130639076233, + "learning_rate": 4.717089668831002e-06, + "loss": 0.6734, + "step": 1192 + }, + { + "epoch": 0.9881281060187742, + "grad_norm": 0.8854318261146545, + "learning_rate": 4.716581304861571e-06, + "loss": 0.6546, + "step": 1193 + }, + { + "epoch": 0.9889563776918829, + "grad_norm": 0.9242726564407349, + "learning_rate": 4.71607251200432e-06, + "loss": 0.6663, + "step": 1194 + }, + { + "epoch": 0.9897846493649917, + "grad_norm": 0.9302003383636475, + "learning_rate": 4.715563290357696e-06, + "loss": 0.6844, + "step": 1195 + }, + { + "epoch": 0.9906129210381005, + "grad_norm": 0.9279932975769043, + "learning_rate": 4.715053640020228e-06, + "loss": 0.6731, + "step": 1196 + }, + { + "epoch": 0.9914411927112092, + "grad_norm": 0.8753612041473389, + "learning_rate": 4.71454356109053e-06, + "loss": 0.6738, + "step": 1197 + }, + { + "epoch": 0.9922694643843181, + "grad_norm": 0.890133261680603, + "learning_rate": 4.714033053667296e-06, + "loss": 0.6638, + "step": 1198 + }, + { + "epoch": 0.9930977360574268, + "grad_norm": 0.9343729615211487, + "learning_rate": 4.713522117849305e-06, + "loss": 0.6826, + "step": 1199 + }, + { + "epoch": 0.9939260077305356, + "grad_norm": 0.9052755832672119, + "learning_rate": 4.7130107537354185e-06, + "loss": 0.6897, + "step": 1200 + }, + { + "epoch": 0.9947542794036444, + "grad_norm": 0.8683924674987793, + "learning_rate": 4.7124989614245805e-06, + "loss": 0.6687, + "step": 1201 + }, + { + "epoch": 0.9955825510767532, + "grad_norm": 0.8677132725715637, + "learning_rate": 4.711986741015818e-06, + "loss": 0.6917, + "step": 1202 + }, + { + "epoch": 0.9964108227498619, + "grad_norm": 0.8840690851211548, + "learning_rate": 4.711474092608243e-06, + "loss": 0.6841, + "step": 1203 + }, + { + "epoch": 0.9972390944229708, + "grad_norm": 0.9750391840934753, + "learning_rate": 4.7109610163010435e-06, + "loss": 0.6574, + "step": 1204 + }, + { + "epoch": 0.9980673660960795, + "grad_norm": 0.8752943873405457, + "learning_rate": 4.710447512193499e-06, + "loss": 0.636, + "step": 1205 + }, + { + "epoch": 0.9988956377691883, + "grad_norm": 0.9100313782691956, + "learning_rate": 4.709933580384965e-06, + "loss": 0.6712, + "step": 1206 + }, + { + "epoch": 0.999723909442297, + "grad_norm": 0.9202356338500977, + "learning_rate": 4.709419220974886e-06, + "loss": 0.6641, + "step": 1207 + }, + { + "epoch": 1.0, + "grad_norm": 0.9202356338500977, + "learning_rate": 4.708904434062783e-06, + "loss": 0.6199, + "step": 1208 + }, + { + "epoch": 1.0008282716731087, + "grad_norm": 1.6839245557785034, + "learning_rate": 4.708389219748263e-06, + "loss": 0.643, + "step": 1209 + }, + { + "epoch": 1.0016565433462175, + "grad_norm": 0.8636482357978821, + "learning_rate": 4.707873578131015e-06, + "loss": 0.6467, + "step": 1210 + }, + { + "epoch": 1.0024848150193264, + "grad_norm": 0.8548988103866577, + "learning_rate": 4.707357509310811e-06, + "loss": 0.6574, + "step": 1211 + }, + { + "epoch": 1.0033130866924351, + "grad_norm": 0.8961352705955505, + "learning_rate": 4.7068410133875066e-06, + "loss": 0.6367, + "step": 1212 + }, + { + "epoch": 1.0041413583655439, + "grad_norm": 0.8914269208908081, + "learning_rate": 4.706324090461037e-06, + "loss": 0.6294, + "step": 1213 + }, + { + "epoch": 1.0049696300386526, + "grad_norm": 0.879915177822113, + "learning_rate": 4.7058067406314235e-06, + "loss": 0.6346, + "step": 1214 + }, + { + "epoch": 1.0057979017117615, + "grad_norm": 0.8798143863677979, + "learning_rate": 4.7052889639987674e-06, + "loss": 0.6559, + "step": 1215 + }, + { + "epoch": 1.0066261733848703, + "grad_norm": 0.8888561129570007, + "learning_rate": 4.704770760663255e-06, + "loss": 0.6471, + "step": 1216 + }, + { + "epoch": 1.007454445057979, + "grad_norm": 0.8707575798034668, + "learning_rate": 4.704252130725152e-06, + "loss": 0.6485, + "step": 1217 + }, + { + "epoch": 1.0082827167310877, + "grad_norm": 0.8609346151351929, + "learning_rate": 4.7037330742848105e-06, + "loss": 0.6544, + "step": 1218 + }, + { + "epoch": 1.0091109884041967, + "grad_norm": 0.8588053584098816, + "learning_rate": 4.703213591442663e-06, + "loss": 0.6421, + "step": 1219 + }, + { + "epoch": 1.0099392600773054, + "grad_norm": 0.8932695388793945, + "learning_rate": 4.702693682299222e-06, + "loss": 0.6333, + "step": 1220 + }, + { + "epoch": 1.0107675317504141, + "grad_norm": 0.8615652322769165, + "learning_rate": 4.7021733469550865e-06, + "loss": 0.6371, + "step": 1221 + }, + { + "epoch": 1.0115958034235228, + "grad_norm": 0.8876634240150452, + "learning_rate": 4.7016525855109376e-06, + "loss": 0.6368, + "step": 1222 + }, + { + "epoch": 1.0124240750966318, + "grad_norm": 0.895575225353241, + "learning_rate": 4.701131398067537e-06, + "loss": 0.6226, + "step": 1223 + }, + { + "epoch": 1.0132523467697405, + "grad_norm": 0.8836749196052551, + "learning_rate": 4.70060978472573e-06, + "loss": 0.6516, + "step": 1224 + }, + { + "epoch": 1.0140806184428492, + "grad_norm": 0.9471280574798584, + "learning_rate": 4.700087745586442e-06, + "loss": 0.633, + "step": 1225 + }, + { + "epoch": 1.014908890115958, + "grad_norm": 0.8721561431884766, + "learning_rate": 4.699565280750685e-06, + "loss": 0.6568, + "step": 1226 + }, + { + "epoch": 1.015737161789067, + "grad_norm": 0.8946447372436523, + "learning_rate": 4.69904239031955e-06, + "loss": 0.6503, + "step": 1227 + }, + { + "epoch": 1.0165654334621756, + "grad_norm": 0.8854631185531616, + "learning_rate": 4.698519074394212e-06, + "loss": 0.6341, + "step": 1228 + }, + { + "epoch": 1.0173937051352844, + "grad_norm": 0.9348397850990295, + "learning_rate": 4.697995333075927e-06, + "loss": 0.6317, + "step": 1229 + }, + { + "epoch": 1.018221976808393, + "grad_norm": 0.9037336111068726, + "learning_rate": 4.697471166466035e-06, + "loss": 0.6618, + "step": 1230 + }, + { + "epoch": 1.0190502484815018, + "grad_norm": 0.8866400718688965, + "learning_rate": 4.696946574665957e-06, + "loss": 0.6421, + "step": 1231 + }, + { + "epoch": 1.0198785201546108, + "grad_norm": 0.8897701501846313, + "learning_rate": 4.696421557777196e-06, + "loss": 0.6608, + "step": 1232 + }, + { + "epoch": 1.0207067918277195, + "grad_norm": 0.9046173095703125, + "learning_rate": 4.6958961159013386e-06, + "loss": 0.6464, + "step": 1233 + }, + { + "epoch": 1.0215350635008282, + "grad_norm": 0.9492906332015991, + "learning_rate": 4.695370249140052e-06, + "loss": 0.6578, + "step": 1234 + }, + { + "epoch": 1.022363335173937, + "grad_norm": 0.8906951546669006, + "learning_rate": 4.6948439575950885e-06, + "loss": 0.6276, + "step": 1235 + }, + { + "epoch": 1.023191606847046, + "grad_norm": 0.9218266606330872, + "learning_rate": 4.694317241368278e-06, + "loss": 0.627, + "step": 1236 + }, + { + "epoch": 1.0240198785201546, + "grad_norm": 0.901157557964325, + "learning_rate": 4.693790100561537e-06, + "loss": 0.6379, + "step": 1237 + }, + { + "epoch": 1.0248481501932634, + "grad_norm": 0.8992452025413513, + "learning_rate": 4.693262535276863e-06, + "loss": 0.6585, + "step": 1238 + }, + { + "epoch": 1.025676421866372, + "grad_norm": 0.9525597095489502, + "learning_rate": 4.692734545616331e-06, + "loss": 0.6625, + "step": 1239 + }, + { + "epoch": 1.026504693539481, + "grad_norm": 0.9229065775871277, + "learning_rate": 4.692206131682106e-06, + "loss": 0.6454, + "step": 1240 + }, + { + "epoch": 1.0273329652125898, + "grad_norm": 1.012373685836792, + "learning_rate": 4.691677293576431e-06, + "loss": 0.6492, + "step": 1241 + }, + { + "epoch": 1.0281612368856985, + "grad_norm": 0.8650060892105103, + "learning_rate": 4.69114803140163e-06, + "loss": 0.6739, + "step": 1242 + }, + { + "epoch": 1.0289895085588072, + "grad_norm": 0.9396429657936096, + "learning_rate": 4.69061834526011e-06, + "loss": 0.6729, + "step": 1243 + }, + { + "epoch": 1.0298177802319162, + "grad_norm": 0.9584783911705017, + "learning_rate": 4.6900882352543614e-06, + "loss": 0.6678, + "step": 1244 + }, + { + "epoch": 1.030646051905025, + "grad_norm": 0.985394299030304, + "learning_rate": 4.689557701486954e-06, + "loss": 0.6478, + "step": 1245 + }, + { + "epoch": 1.0314743235781336, + "grad_norm": 0.9014914631843567, + "learning_rate": 4.689026744060543e-06, + "loss": 0.6393, + "step": 1246 + }, + { + "epoch": 1.0323025952512423, + "grad_norm": 0.9904847145080566, + "learning_rate": 4.688495363077863e-06, + "loss": 0.6752, + "step": 1247 + }, + { + "epoch": 1.0331308669243513, + "grad_norm": 0.9439241886138916, + "learning_rate": 4.687963558641731e-06, + "loss": 0.6417, + "step": 1248 + }, + { + "epoch": 1.03395913859746, + "grad_norm": 0.9277704954147339, + "learning_rate": 4.687431330855046e-06, + "loss": 0.6493, + "step": 1249 + }, + { + "epoch": 1.0347874102705688, + "grad_norm": 0.9456571340560913, + "learning_rate": 4.686898679820789e-06, + "loss": 0.6774, + "step": 1250 + }, + { + "epoch": 1.0356156819436775, + "grad_norm": 0.9043509364128113, + "learning_rate": 4.686365605642025e-06, + "loss": 0.6198, + "step": 1251 + }, + { + "epoch": 1.0364439536167862, + "grad_norm": 0.9246267676353455, + "learning_rate": 4.685832108421896e-06, + "loss": 0.6513, + "step": 1252 + }, + { + "epoch": 1.0372722252898952, + "grad_norm": 0.9227612614631653, + "learning_rate": 4.685298188263631e-06, + "loss": 0.6597, + "step": 1253 + }, + { + "epoch": 1.0381004969630039, + "grad_norm": 0.8927509188652039, + "learning_rate": 4.684763845270537e-06, + "loss": 0.641, + "step": 1254 + }, + { + "epoch": 1.0389287686361126, + "grad_norm": 0.8812937140464783, + "learning_rate": 4.684229079546005e-06, + "loss": 0.6292, + "step": 1255 + }, + { + "epoch": 1.0397570403092213, + "grad_norm": 0.9186881184577942, + "learning_rate": 4.683693891193508e-06, + "loss": 0.6485, + "step": 1256 + }, + { + "epoch": 1.0405853119823303, + "grad_norm": 0.8928134441375732, + "learning_rate": 4.683158280316599e-06, + "loss": 0.6435, + "step": 1257 + }, + { + "epoch": 1.041413583655439, + "grad_norm": 0.8919920325279236, + "learning_rate": 4.682622247018913e-06, + "loss": 0.622, + "step": 1258 + }, + { + "epoch": 1.0422418553285477, + "grad_norm": 0.885259747505188, + "learning_rate": 4.682085791404169e-06, + "loss": 0.6521, + "step": 1259 + }, + { + "epoch": 1.0430701270016565, + "grad_norm": 0.9034169316291809, + "learning_rate": 4.681548913576165e-06, + "loss": 0.6437, + "step": 1260 + }, + { + "epoch": 1.0438983986747654, + "grad_norm": 0.884848952293396, + "learning_rate": 4.6810116136387825e-06, + "loss": 0.6733, + "step": 1261 + }, + { + "epoch": 1.0447266703478741, + "grad_norm": 0.9137836694717407, + "learning_rate": 4.680473891695985e-06, + "loss": 0.6415, + "step": 1262 + }, + { + "epoch": 1.0455549420209829, + "grad_norm": 0.9044780135154724, + "learning_rate": 4.6799357478518145e-06, + "loss": 0.6453, + "step": 1263 + }, + { + "epoch": 1.0463832136940916, + "grad_norm": 0.9111152291297913, + "learning_rate": 4.6793971822103985e-06, + "loss": 0.6644, + "step": 1264 + }, + { + "epoch": 1.0472114853672005, + "grad_norm": 0.9158084988594055, + "learning_rate": 4.678858194875944e-06, + "loss": 0.6434, + "step": 1265 + }, + { + "epoch": 1.0480397570403093, + "grad_norm": 0.8970723152160645, + "learning_rate": 4.678318785952739e-06, + "loss": 0.6391, + "step": 1266 + }, + { + "epoch": 1.048868028713418, + "grad_norm": 0.8958215117454529, + "learning_rate": 4.677778955545155e-06, + "loss": 0.6562, + "step": 1267 + }, + { + "epoch": 1.0496963003865267, + "grad_norm": 0.9229645729064941, + "learning_rate": 4.677238703757644e-06, + "loss": 0.646, + "step": 1268 + }, + { + "epoch": 1.0505245720596355, + "grad_norm": 0.8986689448356628, + "learning_rate": 4.676698030694741e-06, + "loss": 0.6241, + "step": 1269 + }, + { + "epoch": 1.0513528437327444, + "grad_norm": 0.8943328261375427, + "learning_rate": 4.6761569364610585e-06, + "loss": 0.6455, + "step": 1270 + }, + { + "epoch": 1.0521811154058531, + "grad_norm": 0.9147390127182007, + "learning_rate": 4.675615421161295e-06, + "loss": 0.6299, + "step": 1271 + }, + { + "epoch": 1.0530093870789619, + "grad_norm": 0.9124075174331665, + "learning_rate": 4.675073484900229e-06, + "loss": 0.6379, + "step": 1272 + }, + { + "epoch": 1.0538376587520706, + "grad_norm": 0.9267525672912598, + "learning_rate": 4.674531127782718e-06, + "loss": 0.6478, + "step": 1273 + }, + { + "epoch": 1.0546659304251795, + "grad_norm": 0.8710078001022339, + "learning_rate": 4.673988349913707e-06, + "loss": 0.6314, + "step": 1274 + }, + { + "epoch": 1.0554942020982883, + "grad_norm": 0.8905670642852783, + "learning_rate": 4.673445151398214e-06, + "loss": 0.6456, + "step": 1275 + }, + { + "epoch": 1.056322473771397, + "grad_norm": 0.9125292897224426, + "learning_rate": 4.672901532341346e-06, + "loss": 0.6439, + "step": 1276 + }, + { + "epoch": 1.0571507454445057, + "grad_norm": 0.9003625512123108, + "learning_rate": 4.6723574928482865e-06, + "loss": 0.6561, + "step": 1277 + }, + { + "epoch": 1.0579790171176147, + "grad_norm": 0.96099454164505, + "learning_rate": 4.6718130330243014e-06, + "loss": 0.6515, + "step": 1278 + }, + { + "epoch": 1.0588072887907234, + "grad_norm": 0.8841069340705872, + "learning_rate": 4.671268152974742e-06, + "loss": 0.6389, + "step": 1279 + }, + { + "epoch": 1.0596355604638321, + "grad_norm": 0.8906645178794861, + "learning_rate": 4.670722852805033e-06, + "loss": 0.6448, + "step": 1280 + }, + { + "epoch": 1.0604638321369408, + "grad_norm": 0.9538317918777466, + "learning_rate": 4.670177132620689e-06, + "loss": 0.6544, + "step": 1281 + }, + { + "epoch": 1.0612921038100498, + "grad_norm": 0.9158053994178772, + "learning_rate": 4.669630992527299e-06, + "loss": 0.6494, + "step": 1282 + }, + { + "epoch": 1.0621203754831585, + "grad_norm": 0.9756888151168823, + "learning_rate": 4.6690844326305374e-06, + "loss": 0.6419, + "step": 1283 + }, + { + "epoch": 1.0629486471562672, + "grad_norm": 0.905055820941925, + "learning_rate": 4.668537453036159e-06, + "loss": 0.6463, + "step": 1284 + }, + { + "epoch": 1.063776918829376, + "grad_norm": 0.9011061191558838, + "learning_rate": 4.667990053849997e-06, + "loss": 0.6353, + "step": 1285 + }, + { + "epoch": 1.0646051905024847, + "grad_norm": 0.955047070980072, + "learning_rate": 4.667442235177969e-06, + "loss": 0.6485, + "step": 1286 + }, + { + "epoch": 1.0654334621755936, + "grad_norm": 0.9253432750701904, + "learning_rate": 4.666893997126074e-06, + "loss": 0.6391, + "step": 1287 + }, + { + "epoch": 1.0662617338487024, + "grad_norm": 0.8840277194976807, + "learning_rate": 4.66634533980039e-06, + "loss": 0.6372, + "step": 1288 + }, + { + "epoch": 1.067090005521811, + "grad_norm": 1.0881837606430054, + "learning_rate": 4.6657962633070765e-06, + "loss": 0.6419, + "step": 1289 + }, + { + "epoch": 1.06791827719492, + "grad_norm": 0.9263116717338562, + "learning_rate": 4.665246767752376e-06, + "loss": 0.6226, + "step": 1290 + }, + { + "epoch": 1.0687465488680288, + "grad_norm": 0.9110594987869263, + "learning_rate": 4.66469685324261e-06, + "loss": 0.6495, + "step": 1291 + }, + { + "epoch": 1.0695748205411375, + "grad_norm": 0.9086888432502747, + "learning_rate": 4.664146519884182e-06, + "loss": 0.6408, + "step": 1292 + }, + { + "epoch": 1.0704030922142462, + "grad_norm": 0.8991205096244812, + "learning_rate": 4.663595767783575e-06, + "loss": 0.6424, + "step": 1293 + }, + { + "epoch": 1.071231363887355, + "grad_norm": 0.926933228969574, + "learning_rate": 4.6630445970473576e-06, + "loss": 0.6379, + "step": 1294 + }, + { + "epoch": 1.072059635560464, + "grad_norm": 0.8793080449104309, + "learning_rate": 4.662493007782174e-06, + "loss": 0.6254, + "step": 1295 + }, + { + "epoch": 1.0728879072335726, + "grad_norm": 0.9588155150413513, + "learning_rate": 4.661941000094751e-06, + "loss": 0.6307, + "step": 1296 + }, + { + "epoch": 1.0737161789066814, + "grad_norm": 0.9468860030174255, + "learning_rate": 4.661388574091898e-06, + "loss": 0.6308, + "step": 1297 + }, + { + "epoch": 1.07454445057979, + "grad_norm": 0.8930124640464783, + "learning_rate": 4.660835729880505e-06, + "loss": 0.6371, + "step": 1298 + }, + { + "epoch": 1.075372722252899, + "grad_norm": 0.9233878254890442, + "learning_rate": 4.660282467567541e-06, + "loss": 0.6327, + "step": 1299 + }, + { + "epoch": 1.0762009939260078, + "grad_norm": 0.9714951515197754, + "learning_rate": 4.659728787260057e-06, + "loss": 0.6218, + "step": 1300 + }, + { + "epoch": 1.0770292655991165, + "grad_norm": 0.9464428424835205, + "learning_rate": 4.659174689065185e-06, + "loss": 0.6406, + "step": 1301 + }, + { + "epoch": 1.0778575372722252, + "grad_norm": 0.8984559774398804, + "learning_rate": 4.6586201730901395e-06, + "loss": 0.6483, + "step": 1302 + }, + { + "epoch": 1.0786858089453342, + "grad_norm": 0.9007774591445923, + "learning_rate": 4.658065239442212e-06, + "loss": 0.6288, + "step": 1303 + }, + { + "epoch": 1.079514080618443, + "grad_norm": 0.907171368598938, + "learning_rate": 4.657509888228779e-06, + "loss": 0.6445, + "step": 1304 + }, + { + "epoch": 1.0803423522915516, + "grad_norm": 0.9207282066345215, + "learning_rate": 4.656954119557293e-06, + "loss": 0.6148, + "step": 1305 + }, + { + "epoch": 1.0811706239646603, + "grad_norm": 0.9467616677284241, + "learning_rate": 4.656397933535293e-06, + "loss": 0.6374, + "step": 1306 + }, + { + "epoch": 1.0819988956377693, + "grad_norm": 0.9310930967330933, + "learning_rate": 4.655841330270393e-06, + "loss": 0.6362, + "step": 1307 + }, + { + "epoch": 1.082827167310878, + "grad_norm": 0.890816330909729, + "learning_rate": 4.655284309870294e-06, + "loss": 0.6372, + "step": 1308 + }, + { + "epoch": 1.0836554389839868, + "grad_norm": 0.9206095933914185, + "learning_rate": 4.654726872442771e-06, + "loss": 0.6297, + "step": 1309 + }, + { + "epoch": 1.0844837106570955, + "grad_norm": 0.9015856385231018, + "learning_rate": 4.654169018095685e-06, + "loss": 0.658, + "step": 1310 + }, + { + "epoch": 1.0853119823302042, + "grad_norm": 0.8909441828727722, + "learning_rate": 4.653610746936975e-06, + "loss": 0.6382, + "step": 1311 + }, + { + "epoch": 1.0861402540033132, + "grad_norm": 0.9035969972610474, + "learning_rate": 4.653052059074661e-06, + "loss": 0.6441, + "step": 1312 + }, + { + "epoch": 1.0869685256764219, + "grad_norm": 0.8988869786262512, + "learning_rate": 4.652492954616844e-06, + "loss": 0.6359, + "step": 1313 + }, + { + "epoch": 1.0877967973495306, + "grad_norm": 0.881885826587677, + "learning_rate": 4.6519334336717055e-06, + "loss": 0.6397, + "step": 1314 + }, + { + "epoch": 1.0886250690226393, + "grad_norm": 0.980633020401001, + "learning_rate": 4.651373496347508e-06, + "loss": 0.6223, + "step": 1315 + }, + { + "epoch": 1.0894533406957483, + "grad_norm": 0.9027420878410339, + "learning_rate": 4.650813142752593e-06, + "loss": 0.6211, + "step": 1316 + }, + { + "epoch": 1.090281612368857, + "grad_norm": 0.9011778831481934, + "learning_rate": 4.6502523729953855e-06, + "loss": 0.6442, + "step": 1317 + }, + { + "epoch": 1.0911098840419657, + "grad_norm": 0.926692545413971, + "learning_rate": 4.649691187184387e-06, + "loss": 0.6661, + "step": 1318 + }, + { + "epoch": 1.0919381557150745, + "grad_norm": 0.9758741855621338, + "learning_rate": 4.649129585428184e-06, + "loss": 0.6461, + "step": 1319 + }, + { + "epoch": 1.0927664273881834, + "grad_norm": 0.967698335647583, + "learning_rate": 4.64856756783544e-06, + "loss": 0.6461, + "step": 1320 + }, + { + "epoch": 1.0935946990612921, + "grad_norm": 0.9372200965881348, + "learning_rate": 4.6480051345149e-06, + "loss": 0.6348, + "step": 1321 + }, + { + "epoch": 1.0944229707344009, + "grad_norm": 0.8976414799690247, + "learning_rate": 4.64744228557539e-06, + "loss": 0.6297, + "step": 1322 + }, + { + "epoch": 1.0952512424075096, + "grad_norm": 0.9163550734519958, + "learning_rate": 4.646879021125816e-06, + "loss": 0.6746, + "step": 1323 + }, + { + "epoch": 1.0960795140806185, + "grad_norm": 0.9064576625823975, + "learning_rate": 4.646315341275164e-06, + "loss": 0.6242, + "step": 1324 + }, + { + "epoch": 1.0969077857537273, + "grad_norm": 0.9352027773857117, + "learning_rate": 4.645751246132501e-06, + "loss": 0.6568, + "step": 1325 + }, + { + "epoch": 1.097736057426836, + "grad_norm": 0.8680023550987244, + "learning_rate": 4.645186735806976e-06, + "loss": 0.6484, + "step": 1326 + }, + { + "epoch": 1.0985643290999447, + "grad_norm": 0.8782095909118652, + "learning_rate": 4.644621810407813e-06, + "loss": 0.646, + "step": 1327 + }, + { + "epoch": 1.0993926007730535, + "grad_norm": 0.8931819796562195, + "learning_rate": 4.644056470044323e-06, + "loss": 0.6708, + "step": 1328 + }, + { + "epoch": 1.1002208724461624, + "grad_norm": 0.9228624105453491, + "learning_rate": 4.6434907148258915e-06, + "loss": 0.6255, + "step": 1329 + }, + { + "epoch": 1.1010491441192711, + "grad_norm": 0.8960729837417603, + "learning_rate": 4.642924544861988e-06, + "loss": 0.6523, + "step": 1330 + }, + { + "epoch": 1.1018774157923799, + "grad_norm": 0.9092081785202026, + "learning_rate": 4.642357960262162e-06, + "loss": 0.6534, + "step": 1331 + }, + { + "epoch": 1.1027056874654886, + "grad_norm": 0.9024838209152222, + "learning_rate": 4.641790961136041e-06, + "loss": 0.6238, + "step": 1332 + }, + { + "epoch": 1.1035339591385975, + "grad_norm": 0.9016388058662415, + "learning_rate": 4.641223547593334e-06, + "loss": 0.6643, + "step": 1333 + }, + { + "epoch": 1.1043622308117063, + "grad_norm": 0.8877143859863281, + "learning_rate": 4.6406557197438324e-06, + "loss": 0.6222, + "step": 1334 + }, + { + "epoch": 1.105190502484815, + "grad_norm": 0.9066653251647949, + "learning_rate": 4.640087477697403e-06, + "loss": 0.6205, + "step": 1335 + }, + { + "epoch": 1.1060187741579237, + "grad_norm": 0.903465747833252, + "learning_rate": 4.639518821563997e-06, + "loss": 0.6299, + "step": 1336 + }, + { + "epoch": 1.1068470458310327, + "grad_norm": 0.8908408880233765, + "learning_rate": 4.6389497514536415e-06, + "loss": 0.6247, + "step": 1337 + }, + { + "epoch": 1.1076753175041414, + "grad_norm": 0.8957250714302063, + "learning_rate": 4.63838026747645e-06, + "loss": 0.632, + "step": 1338 + }, + { + "epoch": 1.1085035891772501, + "grad_norm": 0.938494086265564, + "learning_rate": 4.637810369742609e-06, + "loss": 0.6309, + "step": 1339 + }, + { + "epoch": 1.1093318608503588, + "grad_norm": 0.9913409948348999, + "learning_rate": 4.637240058362391e-06, + "loss": 0.6571, + "step": 1340 + }, + { + "epoch": 1.1101601325234678, + "grad_norm": 0.9037449359893799, + "learning_rate": 4.636669333446145e-06, + "loss": 0.6467, + "step": 1341 + }, + { + "epoch": 1.1109884041965765, + "grad_norm": 0.911696195602417, + "learning_rate": 4.636098195104299e-06, + "loss": 0.6361, + "step": 1342 + }, + { + "epoch": 1.1118166758696852, + "grad_norm": 0.9105570316314697, + "learning_rate": 4.6355266434473665e-06, + "loss": 0.6418, + "step": 1343 + }, + { + "epoch": 1.112644947542794, + "grad_norm": 0.9101701378822327, + "learning_rate": 4.634954678585935e-06, + "loss": 0.6454, + "step": 1344 + }, + { + "epoch": 1.1134732192159027, + "grad_norm": 0.8980433344841003, + "learning_rate": 4.634382300630675e-06, + "loss": 0.6159, + "step": 1345 + }, + { + "epoch": 1.1143014908890116, + "grad_norm": 0.9253267049789429, + "learning_rate": 4.633809509692336e-06, + "loss": 0.6465, + "step": 1346 + }, + { + "epoch": 1.1151297625621204, + "grad_norm": 0.9150062203407288, + "learning_rate": 4.6332363058817484e-06, + "loss": 0.639, + "step": 1347 + }, + { + "epoch": 1.115958034235229, + "grad_norm": 0.9257630705833435, + "learning_rate": 4.632662689309821e-06, + "loss": 0.6571, + "step": 1348 + }, + { + "epoch": 1.116786305908338, + "grad_norm": 0.9183084964752197, + "learning_rate": 4.632088660087545e-06, + "loss": 0.6291, + "step": 1349 + }, + { + "epoch": 1.1176145775814468, + "grad_norm": 0.9229403138160706, + "learning_rate": 4.631514218325987e-06, + "loss": 0.6487, + "step": 1350 + }, + { + "epoch": 1.1184428492545555, + "grad_norm": 0.9379578828811646, + "learning_rate": 4.630939364136298e-06, + "loss": 0.6564, + "step": 1351 + }, + { + "epoch": 1.1192711209276642, + "grad_norm": 0.9094021320343018, + "learning_rate": 4.630364097629706e-06, + "loss": 0.6354, + "step": 1352 + }, + { + "epoch": 1.120099392600773, + "grad_norm": 0.9528741836547852, + "learning_rate": 4.62978841891752e-06, + "loss": 0.6259, + "step": 1353 + }, + { + "epoch": 1.120927664273882, + "grad_norm": 0.8859900832176208, + "learning_rate": 4.629212328111129e-06, + "loss": 0.6783, + "step": 1354 + }, + { + "epoch": 1.1217559359469906, + "grad_norm": 0.8739688396453857, + "learning_rate": 4.628635825322e-06, + "loss": 0.6518, + "step": 1355 + }, + { + "epoch": 1.1225842076200994, + "grad_norm": 0.8773418068885803, + "learning_rate": 4.628058910661683e-06, + "loss": 0.6477, + "step": 1356 + }, + { + "epoch": 1.123412479293208, + "grad_norm": 0.9025301933288574, + "learning_rate": 4.627481584241803e-06, + "loss": 0.6471, + "step": 1357 + }, + { + "epoch": 1.124240750966317, + "grad_norm": 0.89899080991745, + "learning_rate": 4.62690384617407e-06, + "loss": 0.6196, + "step": 1358 + }, + { + "epoch": 1.1250690226394258, + "grad_norm": 0.9514591693878174, + "learning_rate": 4.6263256965702684e-06, + "loss": 0.6495, + "step": 1359 + }, + { + "epoch": 1.1258972943125345, + "grad_norm": 0.908789336681366, + "learning_rate": 4.6257471355422654e-06, + "loss": 0.6606, + "step": 1360 + }, + { + "epoch": 1.1267255659856432, + "grad_norm": 0.9262552857398987, + "learning_rate": 4.625168163202009e-06, + "loss": 0.6308, + "step": 1361 + }, + { + "epoch": 1.1275538376587522, + "grad_norm": 0.9395532608032227, + "learning_rate": 4.624588779661523e-06, + "loss": 0.6196, + "step": 1362 + }, + { + "epoch": 1.128382109331861, + "grad_norm": 0.9126531481742859, + "learning_rate": 4.624008985032913e-06, + "loss": 0.6474, + "step": 1363 + }, + { + "epoch": 1.1292103810049696, + "grad_norm": 0.9105198383331299, + "learning_rate": 4.623428779428365e-06, + "loss": 0.6455, + "step": 1364 + }, + { + "epoch": 1.1300386526780783, + "grad_norm": 0.9048190712928772, + "learning_rate": 4.6228481629601415e-06, + "loss": 0.6647, + "step": 1365 + }, + { + "epoch": 1.1308669243511873, + "grad_norm": 0.9415149092674255, + "learning_rate": 4.622267135740588e-06, + "loss": 0.6403, + "step": 1366 + }, + { + "epoch": 1.131695196024296, + "grad_norm": 0.954728901386261, + "learning_rate": 4.6216856978821275e-06, + "loss": 0.6351, + "step": 1367 + }, + { + "epoch": 1.1325234676974048, + "grad_norm": 0.9392480850219727, + "learning_rate": 4.621103849497261e-06, + "loss": 0.6481, + "step": 1368 + }, + { + "epoch": 1.1333517393705135, + "grad_norm": 0.9199283719062805, + "learning_rate": 4.620521590698574e-06, + "loss": 0.6578, + "step": 1369 + }, + { + "epoch": 1.1341800110436222, + "grad_norm": 0.9175401926040649, + "learning_rate": 4.6199389215987266e-06, + "loss": 0.6508, + "step": 1370 + }, + { + "epoch": 1.1350082827167312, + "grad_norm": 0.9140470027923584, + "learning_rate": 4.619355842310459e-06, + "loss": 0.6536, + "step": 1371 + }, + { + "epoch": 1.1358365543898399, + "grad_norm": 0.8863652348518372, + "learning_rate": 4.618772352946593e-06, + "loss": 0.6376, + "step": 1372 + }, + { + "epoch": 1.1366648260629486, + "grad_norm": 0.9074656963348389, + "learning_rate": 4.618188453620026e-06, + "loss": 0.6253, + "step": 1373 + }, + { + "epoch": 1.1374930977360573, + "grad_norm": 0.891055703163147, + "learning_rate": 4.61760414444374e-06, + "loss": 0.6507, + "step": 1374 + }, + { + "epoch": 1.1383213694091663, + "grad_norm": 0.9153634905815125, + "learning_rate": 4.617019425530791e-06, + "loss": 0.6699, + "step": 1375 + }, + { + "epoch": 1.139149641082275, + "grad_norm": 0.9172940850257874, + "learning_rate": 4.6164342969943196e-06, + "loss": 0.6385, + "step": 1376 + }, + { + "epoch": 1.1399779127553837, + "grad_norm": 0.8978968262672424, + "learning_rate": 4.615848758947539e-06, + "loss": 0.6394, + "step": 1377 + }, + { + "epoch": 1.1408061844284925, + "grad_norm": 0.9372277855873108, + "learning_rate": 4.615262811503749e-06, + "loss": 0.6418, + "step": 1378 + }, + { + "epoch": 1.1416344561016014, + "grad_norm": 0.9596112370491028, + "learning_rate": 4.614676454776321e-06, + "loss": 0.6573, + "step": 1379 + }, + { + "epoch": 1.1424627277747101, + "grad_norm": 0.9243675470352173, + "learning_rate": 4.614089688878713e-06, + "loss": 0.6546, + "step": 1380 + }, + { + "epoch": 1.1432909994478189, + "grad_norm": 0.8839207291603088, + "learning_rate": 4.613502513924459e-06, + "loss": 0.654, + "step": 1381 + }, + { + "epoch": 1.1441192711209276, + "grad_norm": 0.8940262198448181, + "learning_rate": 4.612914930027168e-06, + "loss": 0.6223, + "step": 1382 + }, + { + "epoch": 1.1449475427940365, + "grad_norm": 0.9091162085533142, + "learning_rate": 4.612326937300535e-06, + "loss": 0.6512, + "step": 1383 + }, + { + "epoch": 1.1457758144671453, + "grad_norm": 0.9332976937294006, + "learning_rate": 4.6117385358583315e-06, + "loss": 0.6373, + "step": 1384 + }, + { + "epoch": 1.146604086140254, + "grad_norm": 0.9103875756263733, + "learning_rate": 4.611149725814406e-06, + "loss": 0.6502, + "step": 1385 + }, + { + "epoch": 1.1474323578133627, + "grad_norm": 0.9004862308502197, + "learning_rate": 4.610560507282688e-06, + "loss": 0.6361, + "step": 1386 + }, + { + "epoch": 1.1482606294864715, + "grad_norm": 0.880293071269989, + "learning_rate": 4.609970880377187e-06, + "loss": 0.655, + "step": 1387 + }, + { + "epoch": 1.1490889011595804, + "grad_norm": 0.9099947214126587, + "learning_rate": 4.60938084521199e-06, + "loss": 0.6517, + "step": 1388 + }, + { + "epoch": 1.1499171728326891, + "grad_norm": 0.8913806080818176, + "learning_rate": 4.608790401901262e-06, + "loss": 0.6429, + "step": 1389 + }, + { + "epoch": 1.1507454445057979, + "grad_norm": 0.9198290109634399, + "learning_rate": 4.608199550559249e-06, + "loss": 0.6537, + "step": 1390 + }, + { + "epoch": 1.1515737161789068, + "grad_norm": 0.9192278981208801, + "learning_rate": 4.607608291300275e-06, + "loss": 0.6531, + "step": 1391 + }, + { + "epoch": 1.1524019878520155, + "grad_norm": 0.8977665305137634, + "learning_rate": 4.607016624238744e-06, + "loss": 0.6358, + "step": 1392 + }, + { + "epoch": 1.1532302595251243, + "grad_norm": 0.9576854705810547, + "learning_rate": 4.606424549489138e-06, + "loss": 0.6493, + "step": 1393 + }, + { + "epoch": 1.154058531198233, + "grad_norm": 0.905611515045166, + "learning_rate": 4.605832067166017e-06, + "loss": 0.63, + "step": 1394 + }, + { + "epoch": 1.1548868028713417, + "grad_norm": 0.8840015530586243, + "learning_rate": 4.605239177384021e-06, + "loss": 0.6504, + "step": 1395 + }, + { + "epoch": 1.1557150745444507, + "grad_norm": 0.9247309565544128, + "learning_rate": 4.604645880257869e-06, + "loss": 0.6477, + "step": 1396 + }, + { + "epoch": 1.1565433462175594, + "grad_norm": 0.9473649263381958, + "learning_rate": 4.604052175902359e-06, + "loss": 0.6374, + "step": 1397 + }, + { + "epoch": 1.1573716178906681, + "grad_norm": 0.9354954361915588, + "learning_rate": 4.603458064432367e-06, + "loss": 0.6491, + "step": 1398 + }, + { + "epoch": 1.1581998895637768, + "grad_norm": 0.8788221478462219, + "learning_rate": 4.6028635459628476e-06, + "loss": 0.6187, + "step": 1399 + }, + { + "epoch": 1.1590281612368858, + "grad_norm": 0.8836075067520142, + "learning_rate": 4.602268620608834e-06, + "loss": 0.665, + "step": 1400 + }, + { + "epoch": 1.1598564329099945, + "grad_norm": 0.9195554256439209, + "learning_rate": 4.60167328848544e-06, + "loss": 0.6291, + "step": 1401 + }, + { + "epoch": 1.1606847045831032, + "grad_norm": 0.8984520435333252, + "learning_rate": 4.601077549707856e-06, + "loss": 0.6361, + "step": 1402 + }, + { + "epoch": 1.161512976256212, + "grad_norm": 0.8667109608650208, + "learning_rate": 4.600481404391352e-06, + "loss": 0.6345, + "step": 1403 + }, + { + "epoch": 1.1623412479293207, + "grad_norm": 0.9312148094177246, + "learning_rate": 4.599884852651277e-06, + "loss": 0.6388, + "step": 1404 + }, + { + "epoch": 1.1631695196024296, + "grad_norm": 0.9096866846084595, + "learning_rate": 4.5992878946030575e-06, + "loss": 0.6383, + "step": 1405 + }, + { + "epoch": 1.1639977912755384, + "grad_norm": 0.9010055661201477, + "learning_rate": 4.5986905303622e-06, + "loss": 0.6524, + "step": 1406 + }, + { + "epoch": 1.164826062948647, + "grad_norm": 0.9414717555046082, + "learning_rate": 4.598092760044288e-06, + "loss": 0.6825, + "step": 1407 + }, + { + "epoch": 1.165654334621756, + "grad_norm": 0.9162404537200928, + "learning_rate": 4.597494583764987e-06, + "loss": 0.6467, + "step": 1408 + }, + { + "epoch": 1.1664826062948648, + "grad_norm": 0.8935617208480835, + "learning_rate": 4.596896001640034e-06, + "loss": 0.6375, + "step": 1409 + }, + { + "epoch": 1.1673108779679735, + "grad_norm": 0.8976510167121887, + "learning_rate": 4.596297013785253e-06, + "loss": 0.6135, + "step": 1410 + }, + { + "epoch": 1.1681391496410822, + "grad_norm": 0.9503909349441528, + "learning_rate": 4.5956976203165414e-06, + "loss": 0.6305, + "step": 1411 + }, + { + "epoch": 1.168967421314191, + "grad_norm": 0.9147273302078247, + "learning_rate": 4.595097821349876e-06, + "loss": 0.6548, + "step": 1412 + }, + { + "epoch": 1.1697956929873, + "grad_norm": 0.9171034693717957, + "learning_rate": 4.594497617001312e-06, + "loss": 0.6253, + "step": 1413 + }, + { + "epoch": 1.1706239646604086, + "grad_norm": 0.8970120549201965, + "learning_rate": 4.593897007386985e-06, + "loss": 0.6414, + "step": 1414 + }, + { + "epoch": 1.1714522363335174, + "grad_norm": 0.9032952189445496, + "learning_rate": 4.593295992623105e-06, + "loss": 0.6595, + "step": 1415 + }, + { + "epoch": 1.172280508006626, + "grad_norm": 0.9066448211669922, + "learning_rate": 4.592694572825964e-06, + "loss": 0.6654, + "step": 1416 + }, + { + "epoch": 1.173108779679735, + "grad_norm": 0.8747309446334839, + "learning_rate": 4.592092748111931e-06, + "loss": 0.6619, + "step": 1417 + }, + { + "epoch": 1.1739370513528438, + "grad_norm": 0.8836600184440613, + "learning_rate": 4.5914905185974535e-06, + "loss": 0.6443, + "step": 1418 + }, + { + "epoch": 1.1747653230259525, + "grad_norm": 0.9340765476226807, + "learning_rate": 4.590887884399058e-06, + "loss": 0.6633, + "step": 1419 + }, + { + "epoch": 1.1755935946990612, + "grad_norm": 0.8889044523239136, + "learning_rate": 4.590284845633347e-06, + "loss": 0.6489, + "step": 1420 + }, + { + "epoch": 1.1764218663721702, + "grad_norm": 0.9043524265289307, + "learning_rate": 4.589681402417005e-06, + "loss": 0.6392, + "step": 1421 + }, + { + "epoch": 1.177250138045279, + "grad_norm": 0.9210046529769897, + "learning_rate": 4.5890775548667895e-06, + "loss": 0.6105, + "step": 1422 + }, + { + "epoch": 1.1780784097183876, + "grad_norm": 0.8954651951789856, + "learning_rate": 4.588473303099542e-06, + "loss": 0.6425, + "step": 1423 + }, + { + "epoch": 1.1789066813914963, + "grad_norm": 0.945517897605896, + "learning_rate": 4.587868647232179e-06, + "loss": 0.6404, + "step": 1424 + }, + { + "epoch": 1.1797349530646053, + "grad_norm": 0.9059463739395142, + "learning_rate": 4.587263587381696e-06, + "loss": 0.6521, + "step": 1425 + }, + { + "epoch": 1.180563224737714, + "grad_norm": 0.8876296281814575, + "learning_rate": 4.586658123665165e-06, + "loss": 0.6447, + "step": 1426 + }, + { + "epoch": 1.1813914964108228, + "grad_norm": 0.8862487077713013, + "learning_rate": 4.58605225619974e-06, + "loss": 0.6445, + "step": 1427 + }, + { + "epoch": 1.1822197680839315, + "grad_norm": 0.9066351056098938, + "learning_rate": 4.585445985102649e-06, + "loss": 0.6563, + "step": 1428 + }, + { + "epoch": 1.1830480397570402, + "grad_norm": 0.913724958896637, + "learning_rate": 4.584839310491199e-06, + "loss": 0.6507, + "step": 1429 + }, + { + "epoch": 1.1838763114301492, + "grad_norm": 0.8942010402679443, + "learning_rate": 4.584232232482778e-06, + "loss": 0.6424, + "step": 1430 + }, + { + "epoch": 1.1847045831032579, + "grad_norm": 0.8986000418663025, + "learning_rate": 4.58362475119485e-06, + "loss": 0.6491, + "step": 1431 + }, + { + "epoch": 1.1855328547763666, + "grad_norm": 0.8879220485687256, + "learning_rate": 4.5830168667449544e-06, + "loss": 0.6412, + "step": 1432 + }, + { + "epoch": 1.1863611264494753, + "grad_norm": 0.9497302770614624, + "learning_rate": 4.5824085792507135e-06, + "loss": 0.6453, + "step": 1433 + }, + { + "epoch": 1.1871893981225843, + "grad_norm": 0.8933898210525513, + "learning_rate": 4.581799888829824e-06, + "loss": 0.6381, + "step": 1434 + }, + { + "epoch": 1.188017669795693, + "grad_norm": 0.9034610986709595, + "learning_rate": 4.581190795600062e-06, + "loss": 0.6551, + "step": 1435 + }, + { + "epoch": 1.1888459414688017, + "grad_norm": 0.8968634605407715, + "learning_rate": 4.580581299679282e-06, + "loss": 0.637, + "step": 1436 + }, + { + "epoch": 1.1896742131419105, + "grad_norm": 0.8974384069442749, + "learning_rate": 4.579971401185416e-06, + "loss": 0.6301, + "step": 1437 + }, + { + "epoch": 1.1905024848150194, + "grad_norm": 0.8726125359535217, + "learning_rate": 4.579361100236473e-06, + "loss": 0.6487, + "step": 1438 + }, + { + "epoch": 1.1913307564881281, + "grad_norm": 0.9131139516830444, + "learning_rate": 4.5787503969505405e-06, + "loss": 0.6211, + "step": 1439 + }, + { + "epoch": 1.1921590281612369, + "grad_norm": 0.8924953937530518, + "learning_rate": 4.578139291445783e-06, + "loss": 0.6339, + "step": 1440 + }, + { + "epoch": 1.1929872998343456, + "grad_norm": 0.890883207321167, + "learning_rate": 4.577527783840447e-06, + "loss": 0.66, + "step": 1441 + }, + { + "epoch": 1.1938155715074545, + "grad_norm": 0.9326318502426147, + "learning_rate": 4.5769158742528494e-06, + "loss": 0.6611, + "step": 1442 + }, + { + "epoch": 1.1946438431805633, + "grad_norm": 0.9354713559150696, + "learning_rate": 4.576303562801392e-06, + "loss": 0.6445, + "step": 1443 + }, + { + "epoch": 1.195472114853672, + "grad_norm": 0.8868383169174194, + "learning_rate": 4.57569084960455e-06, + "loss": 0.6344, + "step": 1444 + }, + { + "epoch": 1.1963003865267807, + "grad_norm": 0.893924355506897, + "learning_rate": 4.5750777347808775e-06, + "loss": 0.602, + "step": 1445 + }, + { + "epoch": 1.1971286581998895, + "grad_norm": 0.9299339652061462, + "learning_rate": 4.574464218449006e-06, + "loss": 0.6206, + "step": 1446 + }, + { + "epoch": 1.1979569298729984, + "grad_norm": 0.9058142304420471, + "learning_rate": 4.573850300727648e-06, + "loss": 0.6073, + "step": 1447 + }, + { + "epoch": 1.1987852015461071, + "grad_norm": 0.8893733620643616, + "learning_rate": 4.5732359817355885e-06, + "loss": 0.6336, + "step": 1448 + }, + { + "epoch": 1.1996134732192159, + "grad_norm": 0.9018000960350037, + "learning_rate": 4.572621261591692e-06, + "loss": 0.6268, + "step": 1449 + }, + { + "epoch": 1.2004417448923248, + "grad_norm": 0.9087574481964111, + "learning_rate": 4.572006140414903e-06, + "loss": 0.6511, + "step": 1450 + }, + { + "epoch": 1.2012700165654335, + "grad_norm": 0.9092984795570374, + "learning_rate": 4.5713906183242404e-06, + "loss": 0.6424, + "step": 1451 + }, + { + "epoch": 1.2020982882385423, + "grad_norm": 0.9040535092353821, + "learning_rate": 4.5707746954388034e-06, + "loss": 0.6242, + "step": 1452 + }, + { + "epoch": 1.202926559911651, + "grad_norm": 0.891659677028656, + "learning_rate": 4.570158371877766e-06, + "loss": 0.643, + "step": 1453 + }, + { + "epoch": 1.2037548315847597, + "grad_norm": 0.9254544973373413, + "learning_rate": 4.569541647760381e-06, + "loss": 0.6191, + "step": 1454 + }, + { + "epoch": 1.2045831032578687, + "grad_norm": 0.8978536128997803, + "learning_rate": 4.568924523205979e-06, + "loss": 0.6632, + "step": 1455 + }, + { + "epoch": 1.2054113749309774, + "grad_norm": 0.9179201126098633, + "learning_rate": 4.568306998333968e-06, + "loss": 0.6485, + "step": 1456 + }, + { + "epoch": 1.2062396466040861, + "grad_norm": 0.9061882495880127, + "learning_rate": 4.567689073263834e-06, + "loss": 0.6704, + "step": 1457 + }, + { + "epoch": 1.2070679182771948, + "grad_norm": 0.9143427610397339, + "learning_rate": 4.567070748115139e-06, + "loss": 0.6453, + "step": 1458 + }, + { + "epoch": 1.2078961899503038, + "grad_norm": 0.8740119338035583, + "learning_rate": 4.566452023007523e-06, + "loss": 0.6453, + "step": 1459 + }, + { + "epoch": 1.2087244616234125, + "grad_norm": 0.8977463841438293, + "learning_rate": 4.565832898060703e-06, + "loss": 0.6451, + "step": 1460 + }, + { + "epoch": 1.2095527332965212, + "grad_norm": 0.9416611194610596, + "learning_rate": 4.565213373394476e-06, + "loss": 0.6265, + "step": 1461 + }, + { + "epoch": 1.21038100496963, + "grad_norm": 0.9132000207901001, + "learning_rate": 4.5645934491287115e-06, + "loss": 0.6412, + "step": 1462 + }, + { + "epoch": 1.2112092766427387, + "grad_norm": 0.9480498433113098, + "learning_rate": 4.563973125383362e-06, + "loss": 0.6363, + "step": 1463 + }, + { + "epoch": 1.2120375483158476, + "grad_norm": 0.9170709848403931, + "learning_rate": 4.563352402278451e-06, + "loss": 0.6201, + "step": 1464 + }, + { + "epoch": 1.2128658199889564, + "grad_norm": 0.9655918478965759, + "learning_rate": 4.562731279934086e-06, + "loss": 0.6497, + "step": 1465 + }, + { + "epoch": 1.213694091662065, + "grad_norm": 0.8885562419891357, + "learning_rate": 4.562109758470446e-06, + "loss": 0.6265, + "step": 1466 + }, + { + "epoch": 1.214522363335174, + "grad_norm": 0.9072791337966919, + "learning_rate": 4.561487838007792e-06, + "loss": 0.6257, + "step": 1467 + }, + { + "epoch": 1.2153506350082828, + "grad_norm": 0.8992362022399902, + "learning_rate": 4.560865518666458e-06, + "loss": 0.6425, + "step": 1468 + }, + { + "epoch": 1.2161789066813915, + "grad_norm": 0.8934991955757141, + "learning_rate": 4.560242800566855e-06, + "loss": 0.6583, + "step": 1469 + }, + { + "epoch": 1.2170071783545002, + "grad_norm": 0.9292325973510742, + "learning_rate": 4.559619683829477e-06, + "loss": 0.6403, + "step": 1470 + }, + { + "epoch": 1.217835450027609, + "grad_norm": 0.8756814002990723, + "learning_rate": 4.55899616857489e-06, + "loss": 0.6261, + "step": 1471 + }, + { + "epoch": 1.218663721700718, + "grad_norm": 0.9050565361976624, + "learning_rate": 4.558372254923738e-06, + "loss": 0.6325, + "step": 1472 + }, + { + "epoch": 1.2194919933738266, + "grad_norm": 0.9230931401252747, + "learning_rate": 4.557747942996742e-06, + "loss": 0.6275, + "step": 1473 + }, + { + "epoch": 1.2203202650469354, + "grad_norm": 0.894026517868042, + "learning_rate": 4.557123232914702e-06, + "loss": 0.6348, + "step": 1474 + }, + { + "epoch": 1.221148536720044, + "grad_norm": 0.8882637023925781, + "learning_rate": 4.556498124798492e-06, + "loss": 0.6123, + "step": 1475 + }, + { + "epoch": 1.221976808393153, + "grad_norm": 0.9191063642501831, + "learning_rate": 4.555872618769065e-06, + "loss": 0.6531, + "step": 1476 + }, + { + "epoch": 1.2228050800662618, + "grad_norm": 0.9398623704910278, + "learning_rate": 4.555246714947452e-06, + "loss": 0.6553, + "step": 1477 + }, + { + "epoch": 1.2236333517393705, + "grad_norm": 0.9399452805519104, + "learning_rate": 4.554620413454757e-06, + "loss": 0.6485, + "step": 1478 + }, + { + "epoch": 1.2244616234124792, + "grad_norm": 0.9001572728157043, + "learning_rate": 4.553993714412166e-06, + "loss": 0.6467, + "step": 1479 + }, + { + "epoch": 1.2252898950855882, + "grad_norm": 0.9266572594642639, + "learning_rate": 4.553366617940938e-06, + "loss": 0.6325, + "step": 1480 + }, + { + "epoch": 1.226118166758697, + "grad_norm": 0.928301215171814, + "learning_rate": 4.552739124162411e-06, + "loss": 0.6398, + "step": 1481 + }, + { + "epoch": 1.2269464384318056, + "grad_norm": 0.880954921245575, + "learning_rate": 4.552111233197999e-06, + "loss": 0.669, + "step": 1482 + }, + { + "epoch": 1.2277747101049143, + "grad_norm": 0.9089685082435608, + "learning_rate": 4.551482945169191e-06, + "loss": 0.62, + "step": 1483 + }, + { + "epoch": 1.2286029817780233, + "grad_norm": 0.9187189936637878, + "learning_rate": 4.550854260197559e-06, + "loss": 0.6381, + "step": 1484 + }, + { + "epoch": 1.229431253451132, + "grad_norm": 0.8782045245170593, + "learning_rate": 4.550225178404744e-06, + "loss": 0.6397, + "step": 1485 + }, + { + "epoch": 1.2302595251242407, + "grad_norm": 0.8798948526382446, + "learning_rate": 4.54959569991247e-06, + "loss": 0.6522, + "step": 1486 + }, + { + "epoch": 1.2310877967973495, + "grad_norm": 0.9257029294967651, + "learning_rate": 4.548965824842534e-06, + "loss": 0.6522, + "step": 1487 + }, + { + "epoch": 1.2319160684704582, + "grad_norm": 0.8923602104187012, + "learning_rate": 4.548335553316811e-06, + "loss": 0.6538, + "step": 1488 + }, + { + "epoch": 1.2327443401435672, + "grad_norm": 0.9071294069290161, + "learning_rate": 4.5477048854572524e-06, + "loss": 0.6305, + "step": 1489 + }, + { + "epoch": 1.2335726118166759, + "grad_norm": 0.9495804905891418, + "learning_rate": 4.547073821385888e-06, + "loss": 0.6398, + "step": 1490 + }, + { + "epoch": 1.2344008834897846, + "grad_norm": 0.9163299202919006, + "learning_rate": 4.54644236122482e-06, + "loss": 0.6454, + "step": 1491 + }, + { + "epoch": 1.2352291551628933, + "grad_norm": 0.9286475777626038, + "learning_rate": 4.545810505096233e-06, + "loss": 0.6345, + "step": 1492 + }, + { + "epoch": 1.2360574268360023, + "grad_norm": 0.9116879105567932, + "learning_rate": 4.545178253122384e-06, + "loss": 0.6451, + "step": 1493 + }, + { + "epoch": 1.236885698509111, + "grad_norm": 0.917826771736145, + "learning_rate": 4.544545605425607e-06, + "loss": 0.6374, + "step": 1494 + }, + { + "epoch": 1.2377139701822197, + "grad_norm": 0.9433108568191528, + "learning_rate": 4.543912562128316e-06, + "loss": 0.6321, + "step": 1495 + }, + { + "epoch": 1.2385422418553285, + "grad_norm": 0.9017132520675659, + "learning_rate": 4.543279123352996e-06, + "loss": 0.6341, + "step": 1496 + }, + { + "epoch": 1.2393705135284374, + "grad_norm": 0.9557158350944519, + "learning_rate": 4.542645289222214e-06, + "loss": 0.6364, + "step": 1497 + }, + { + "epoch": 1.2401987852015461, + "grad_norm": 0.9258891344070435, + "learning_rate": 4.542011059858609e-06, + "loss": 0.6598, + "step": 1498 + }, + { + "epoch": 1.2410270568746549, + "grad_norm": 0.8988100290298462, + "learning_rate": 4.541376435384899e-06, + "loss": 0.6496, + "step": 1499 + }, + { + "epoch": 1.2418553285477636, + "grad_norm": 0.898154079914093, + "learning_rate": 4.540741415923878e-06, + "loss": 0.6486, + "step": 1500 + }, + { + "epoch": 1.2426836002208725, + "grad_norm": 0.9216037392616272, + "learning_rate": 4.540106001598418e-06, + "loss": 0.6436, + "step": 1501 + }, + { + "epoch": 1.2435118718939813, + "grad_norm": 0.9086095690727234, + "learning_rate": 4.539470192531463e-06, + "loss": 0.6287, + "step": 1502 + }, + { + "epoch": 1.24434014356709, + "grad_norm": 0.921291172504425, + "learning_rate": 4.538833988846039e-06, + "loss": 0.629, + "step": 1503 + }, + { + "epoch": 1.2451684152401987, + "grad_norm": 0.8621019721031189, + "learning_rate": 4.538197390665242e-06, + "loss": 0.6543, + "step": 1504 + }, + { + "epoch": 1.2459966869133074, + "grad_norm": 0.9523105025291443, + "learning_rate": 4.537560398112251e-06, + "loss": 0.6482, + "step": 1505 + }, + { + "epoch": 1.2468249585864164, + "grad_norm": 0.971931517124176, + "learning_rate": 4.536923011310317e-06, + "loss": 0.6463, + "step": 1506 + }, + { + "epoch": 1.2476532302595251, + "grad_norm": 0.927219033241272, + "learning_rate": 4.536285230382767e-06, + "loss": 0.6443, + "step": 1507 + }, + { + "epoch": 1.2484815019326339, + "grad_norm": 0.9433056116104126, + "learning_rate": 4.535647055453007e-06, + "loss": 0.6301, + "step": 1508 + }, + { + "epoch": 1.2493097736057428, + "grad_norm": 0.8992047905921936, + "learning_rate": 4.5350084866445195e-06, + "loss": 0.6177, + "step": 1509 + }, + { + "epoch": 1.2501380452788515, + "grad_norm": 0.8946895003318787, + "learning_rate": 4.534369524080858e-06, + "loss": 0.6275, + "step": 1510 + }, + { + "epoch": 1.2509663169519603, + "grad_norm": 0.8732156157493591, + "learning_rate": 4.5337301678856595e-06, + "loss": 0.6572, + "step": 1511 + }, + { + "epoch": 1.251794588625069, + "grad_norm": 0.9505333304405212, + "learning_rate": 4.533090418182631e-06, + "loss": 0.6352, + "step": 1512 + }, + { + "epoch": 1.2526228602981777, + "grad_norm": 0.8931122422218323, + "learning_rate": 4.532450275095558e-06, + "loss": 0.649, + "step": 1513 + }, + { + "epoch": 1.2534511319712867, + "grad_norm": 0.8781880140304565, + "learning_rate": 4.531809738748304e-06, + "loss": 0.651, + "step": 1514 + }, + { + "epoch": 1.2542794036443954, + "grad_norm": 0.9201871156692505, + "learning_rate": 4.531168809264805e-06, + "loss": 0.6451, + "step": 1515 + }, + { + "epoch": 1.2551076753175041, + "grad_norm": 0.9318148493766785, + "learning_rate": 4.530527486769077e-06, + "loss": 0.6458, + "step": 1516 + }, + { + "epoch": 1.2559359469906128, + "grad_norm": 0.9131407141685486, + "learning_rate": 4.529885771385208e-06, + "loss": 0.6679, + "step": 1517 + }, + { + "epoch": 1.2567642186637218, + "grad_norm": 0.8942316770553589, + "learning_rate": 4.529243663237365e-06, + "loss": 0.6262, + "step": 1518 + }, + { + "epoch": 1.2575924903368305, + "grad_norm": 0.9058274030685425, + "learning_rate": 4.528601162449788e-06, + "loss": 0.635, + "step": 1519 + }, + { + "epoch": 1.2584207620099392, + "grad_norm": 0.9129453897476196, + "learning_rate": 4.527958269146798e-06, + "loss": 0.6375, + "step": 1520 + }, + { + "epoch": 1.259249033683048, + "grad_norm": 0.8737504482269287, + "learning_rate": 4.527314983452787e-06, + "loss": 0.6437, + "step": 1521 + }, + { + "epoch": 1.2600773053561567, + "grad_norm": 0.9680130481719971, + "learning_rate": 4.526671305492225e-06, + "loss": 0.6546, + "step": 1522 + }, + { + "epoch": 1.2609055770292656, + "grad_norm": 0.9186084270477295, + "learning_rate": 4.526027235389658e-06, + "loss": 0.6505, + "step": 1523 + }, + { + "epoch": 1.2617338487023744, + "grad_norm": 0.9164871573448181, + "learning_rate": 4.525382773269706e-06, + "loss": 0.6541, + "step": 1524 + }, + { + "epoch": 1.262562120375483, + "grad_norm": 0.8994123935699463, + "learning_rate": 4.5247379192570695e-06, + "loss": 0.6313, + "step": 1525 + }, + { + "epoch": 1.263390392048592, + "grad_norm": 0.8780331015586853, + "learning_rate": 4.524092673476519e-06, + "loss": 0.6496, + "step": 1526 + }, + { + "epoch": 1.2642186637217008, + "grad_norm": 0.8670293092727661, + "learning_rate": 4.5234470360529045e-06, + "loss": 0.6422, + "step": 1527 + }, + { + "epoch": 1.2650469353948095, + "grad_norm": 0.9305918216705322, + "learning_rate": 4.522801007111152e-06, + "loss": 0.6352, + "step": 1528 + }, + { + "epoch": 1.2658752070679182, + "grad_norm": 0.9444066286087036, + "learning_rate": 4.52215458677626e-06, + "loss": 0.6357, + "step": 1529 + }, + { + "epoch": 1.266703478741027, + "grad_norm": 0.8987369537353516, + "learning_rate": 4.521507775173305e-06, + "loss": 0.6407, + "step": 1530 + }, + { + "epoch": 1.267531750414136, + "grad_norm": 0.9200116395950317, + "learning_rate": 4.5208605724274415e-06, + "loss": 0.6366, + "step": 1531 + }, + { + "epoch": 1.2683600220872446, + "grad_norm": 0.8680561780929565, + "learning_rate": 4.520212978663894e-06, + "loss": 0.6389, + "step": 1532 + }, + { + "epoch": 1.2691882937603534, + "grad_norm": 0.8928949236869812, + "learning_rate": 4.519564994007968e-06, + "loss": 0.6552, + "step": 1533 + }, + { + "epoch": 1.2700165654334623, + "grad_norm": 0.8986184000968933, + "learning_rate": 4.518916618585042e-06, + "loss": 0.6554, + "step": 1534 + }, + { + "epoch": 1.270844837106571, + "grad_norm": 0.9239423274993896, + "learning_rate": 4.51826785252057e-06, + "loss": 0.6448, + "step": 1535 + }, + { + "epoch": 1.2716731087796798, + "grad_norm": 0.9534674286842346, + "learning_rate": 4.517618695940082e-06, + "loss": 0.6303, + "step": 1536 + }, + { + "epoch": 1.2725013804527885, + "grad_norm": 0.9341598749160767, + "learning_rate": 4.516969148969186e-06, + "loss": 0.6335, + "step": 1537 + }, + { + "epoch": 1.2733296521258972, + "grad_norm": 0.9264217615127563, + "learning_rate": 4.51631921173356e-06, + "loss": 0.6543, + "step": 1538 + }, + { + "epoch": 1.274157923799006, + "grad_norm": 0.9460370540618896, + "learning_rate": 4.515668884358963e-06, + "loss": 0.6531, + "step": 1539 + }, + { + "epoch": 1.274986195472115, + "grad_norm": 0.9690954089164734, + "learning_rate": 4.515018166971227e-06, + "loss": 0.6495, + "step": 1540 + }, + { + "epoch": 1.2758144671452236, + "grad_norm": 0.8746740818023682, + "learning_rate": 4.514367059696259e-06, + "loss": 0.6169, + "step": 1541 + }, + { + "epoch": 1.2766427388183323, + "grad_norm": 0.9288576245307922, + "learning_rate": 4.513715562660043e-06, + "loss": 0.6459, + "step": 1542 + }, + { + "epoch": 1.2774710104914413, + "grad_norm": 0.9935420751571655, + "learning_rate": 4.513063675988637e-06, + "loss": 0.6471, + "step": 1543 + }, + { + "epoch": 1.27829928216455, + "grad_norm": 0.9584968686103821, + "learning_rate": 4.512411399808175e-06, + "loss": 0.6473, + "step": 1544 + }, + { + "epoch": 1.2791275538376587, + "grad_norm": 0.9049419164657593, + "learning_rate": 4.511758734244867e-06, + "loss": 0.6667, + "step": 1545 + }, + { + "epoch": 1.2799558255107675, + "grad_norm": 0.9456910490989685, + "learning_rate": 4.511105679424998e-06, + "loss": 0.625, + "step": 1546 + }, + { + "epoch": 1.2807840971838762, + "grad_norm": 0.9267060160636902, + "learning_rate": 4.510452235474926e-06, + "loss": 0.6239, + "step": 1547 + }, + { + "epoch": 1.2816123688569852, + "grad_norm": 0.9363065958023071, + "learning_rate": 4.509798402521088e-06, + "loss": 0.6416, + "step": 1548 + }, + { + "epoch": 1.2824406405300939, + "grad_norm": 0.85664302110672, + "learning_rate": 4.509144180689995e-06, + "loss": 0.6297, + "step": 1549 + }, + { + "epoch": 1.2832689122032026, + "grad_norm": 0.9298888444900513, + "learning_rate": 4.508489570108231e-06, + "loss": 0.6305, + "step": 1550 + }, + { + "epoch": 1.2840971838763116, + "grad_norm": 0.9485051035881042, + "learning_rate": 4.507834570902459e-06, + "loss": 0.6409, + "step": 1551 + }, + { + "epoch": 1.2849254555494203, + "grad_norm": 0.9129849076271057, + "learning_rate": 4.507179183199414e-06, + "loss": 0.6215, + "step": 1552 + }, + { + "epoch": 1.285753727222529, + "grad_norm": 0.9630280137062073, + "learning_rate": 4.506523407125907e-06, + "loss": 0.6229, + "step": 1553 + }, + { + "epoch": 1.2865819988956377, + "grad_norm": 0.9047848582267761, + "learning_rate": 4.505867242808826e-06, + "loss": 0.6434, + "step": 1554 + }, + { + "epoch": 1.2874102705687465, + "grad_norm": 0.9219448566436768, + "learning_rate": 4.5052106903751315e-06, + "loss": 0.6449, + "step": 1555 + }, + { + "epoch": 1.2882385422418554, + "grad_norm": 0.9347684383392334, + "learning_rate": 4.504553749951861e-06, + "loss": 0.6272, + "step": 1556 + }, + { + "epoch": 1.2890668139149641, + "grad_norm": 0.8930262923240662, + "learning_rate": 4.503896421666126e-06, + "loss": 0.6364, + "step": 1557 + }, + { + "epoch": 1.2898950855880729, + "grad_norm": 0.9067763090133667, + "learning_rate": 4.503238705645114e-06, + "loss": 0.6317, + "step": 1558 + }, + { + "epoch": 1.2907233572611816, + "grad_norm": 0.9524715542793274, + "learning_rate": 4.5025806020160865e-06, + "loss": 0.6525, + "step": 1559 + }, + { + "epoch": 1.2915516289342905, + "grad_norm": 0.927255392074585, + "learning_rate": 4.5019221109063795e-06, + "loss": 0.6343, + "step": 1560 + }, + { + "epoch": 1.2923799006073993, + "grad_norm": 0.904295802116394, + "learning_rate": 4.5012632324434065e-06, + "loss": 0.6484, + "step": 1561 + }, + { + "epoch": 1.293208172280508, + "grad_norm": 0.9138978123664856, + "learning_rate": 4.500603966754653e-06, + "loss": 0.6374, + "step": 1562 + }, + { + "epoch": 1.2940364439536167, + "grad_norm": 0.9038779139518738, + "learning_rate": 4.499944313967681e-06, + "loss": 0.6395, + "step": 1563 + }, + { + "epoch": 1.2948647156267254, + "grad_norm": 0.9218693971633911, + "learning_rate": 4.499284274210128e-06, + "loss": 0.6325, + "step": 1564 + }, + { + "epoch": 1.2956929872998344, + "grad_norm": 0.9210799932479858, + "learning_rate": 4.4986238476097055e-06, + "loss": 0.6476, + "step": 1565 + }, + { + "epoch": 1.2965212589729431, + "grad_norm": 0.8977149724960327, + "learning_rate": 4.4979630342941994e-06, + "loss": 0.6489, + "step": 1566 + }, + { + "epoch": 1.2973495306460519, + "grad_norm": 0.9423681497573853, + "learning_rate": 4.49730183439147e-06, + "loss": 0.6391, + "step": 1567 + }, + { + "epoch": 1.2981778023191608, + "grad_norm": 0.9404032826423645, + "learning_rate": 4.4966402480294545e-06, + "loss": 0.6233, + "step": 1568 + }, + { + "epoch": 1.2990060739922695, + "grad_norm": 0.9104741215705872, + "learning_rate": 4.495978275336164e-06, + "loss": 0.6369, + "step": 1569 + }, + { + "epoch": 1.2998343456653783, + "grad_norm": 0.8869790434837341, + "learning_rate": 4.4953159164396835e-06, + "loss": 0.6331, + "step": 1570 + }, + { + "epoch": 1.300662617338487, + "grad_norm": 0.9269986152648926, + "learning_rate": 4.494653171468173e-06, + "loss": 0.6361, + "step": 1571 + }, + { + "epoch": 1.3014908890115957, + "grad_norm": 0.9210803508758545, + "learning_rate": 4.4939900405498675e-06, + "loss": 0.6314, + "step": 1572 + }, + { + "epoch": 1.3023191606847047, + "grad_norm": 0.9426820874214172, + "learning_rate": 4.4933265238130775e-06, + "loss": 0.6625, + "step": 1573 + }, + { + "epoch": 1.3031474323578134, + "grad_norm": 0.9198150634765625, + "learning_rate": 4.492662621386186e-06, + "loss": 0.627, + "step": 1574 + }, + { + "epoch": 1.3039757040309221, + "grad_norm": 0.9312534928321838, + "learning_rate": 4.4919983333976525e-06, + "loss": 0.6386, + "step": 1575 + }, + { + "epoch": 1.304803975704031, + "grad_norm": 0.8926329016685486, + "learning_rate": 4.491333659976011e-06, + "loss": 0.6559, + "step": 1576 + }, + { + "epoch": 1.3056322473771398, + "grad_norm": 0.9172425270080566, + "learning_rate": 4.4906686012498705e-06, + "loss": 0.6254, + "step": 1577 + }, + { + "epoch": 1.3064605190502485, + "grad_norm": 0.9896116256713867, + "learning_rate": 4.490003157347911e-06, + "loss": 0.6331, + "step": 1578 + }, + { + "epoch": 1.3072887907233572, + "grad_norm": 0.9267039895057678, + "learning_rate": 4.489337328398891e-06, + "loss": 0.6296, + "step": 1579 + }, + { + "epoch": 1.308117062396466, + "grad_norm": 0.9033195376396179, + "learning_rate": 4.488671114531643e-06, + "loss": 0.6354, + "step": 1580 + }, + { + "epoch": 1.3089453340695747, + "grad_norm": 0.9047731161117554, + "learning_rate": 4.488004515875072e-06, + "loss": 0.6351, + "step": 1581 + }, + { + "epoch": 1.3097736057426836, + "grad_norm": 0.9096142649650574, + "learning_rate": 4.4873375325581596e-06, + "loss": 0.6504, + "step": 1582 + }, + { + "epoch": 1.3106018774157924, + "grad_norm": 0.8848153352737427, + "learning_rate": 4.48667016470996e-06, + "loss": 0.6354, + "step": 1583 + }, + { + "epoch": 1.311430149088901, + "grad_norm": 0.9036188721656799, + "learning_rate": 4.486002412459603e-06, + "loss": 0.6431, + "step": 1584 + }, + { + "epoch": 1.31225842076201, + "grad_norm": 0.8881173133850098, + "learning_rate": 4.4853342759362924e-06, + "loss": 0.6509, + "step": 1585 + }, + { + "epoch": 1.3130866924351188, + "grad_norm": 0.9195591807365417, + "learning_rate": 4.484665755269307e-06, + "loss": 0.6522, + "step": 1586 + }, + { + "epoch": 1.3139149641082275, + "grad_norm": 0.9021509885787964, + "learning_rate": 4.483996850587998e-06, + "loss": 0.6532, + "step": 1587 + }, + { + "epoch": 1.3147432357813362, + "grad_norm": 0.8902403116226196, + "learning_rate": 4.483327562021794e-06, + "loss": 0.5942, + "step": 1588 + }, + { + "epoch": 1.315571507454445, + "grad_norm": 0.8933504819869995, + "learning_rate": 4.482657889700194e-06, + "loss": 0.6305, + "step": 1589 + }, + { + "epoch": 1.316399779127554, + "grad_norm": 0.9233856201171875, + "learning_rate": 4.481987833752776e-06, + "loss": 0.6275, + "step": 1590 + }, + { + "epoch": 1.3172280508006626, + "grad_norm": 0.9482141733169556, + "learning_rate": 4.4813173943091865e-06, + "loss": 0.6232, + "step": 1591 + }, + { + "epoch": 1.3180563224737714, + "grad_norm": 0.94732266664505, + "learning_rate": 4.480646571499151e-06, + "loss": 0.6297, + "step": 1592 + }, + { + "epoch": 1.3188845941468803, + "grad_norm": 0.9066787958145142, + "learning_rate": 4.479975365452468e-06, + "loss": 0.6334, + "step": 1593 + }, + { + "epoch": 1.319712865819989, + "grad_norm": 0.9063938856124878, + "learning_rate": 4.479303776299008e-06, + "loss": 0.6415, + "step": 1594 + }, + { + "epoch": 1.3205411374930978, + "grad_norm": 0.9286819696426392, + "learning_rate": 4.478631804168719e-06, + "loss": 0.6534, + "step": 1595 + }, + { + "epoch": 1.3213694091662065, + "grad_norm": 0.9161240458488464, + "learning_rate": 4.47795944919162e-06, + "loss": 0.6401, + "step": 1596 + }, + { + "epoch": 1.3221976808393152, + "grad_norm": 0.956538200378418, + "learning_rate": 4.4772867114978056e-06, + "loss": 0.6649, + "step": 1597 + }, + { + "epoch": 1.323025952512424, + "grad_norm": 0.910504162311554, + "learning_rate": 4.4766135912174455e-06, + "loss": 0.6456, + "step": 1598 + }, + { + "epoch": 1.323854224185533, + "grad_norm": 0.9054026007652283, + "learning_rate": 4.4759400884807805e-06, + "loss": 0.6458, + "step": 1599 + }, + { + "epoch": 1.3246824958586416, + "grad_norm": 0.933506429195404, + "learning_rate": 4.475266203418127e-06, + "loss": 0.626, + "step": 1600 + }, + { + "epoch": 1.3255107675317503, + "grad_norm": 0.9042014479637146, + "learning_rate": 4.474591936159878e-06, + "loss": 0.6376, + "step": 1601 + }, + { + "epoch": 1.3263390392048593, + "grad_norm": 0.9005138278007507, + "learning_rate": 4.473917286836496e-06, + "loss": 0.6093, + "step": 1602 + }, + { + "epoch": 1.327167310877968, + "grad_norm": 0.9096264839172363, + "learning_rate": 4.4732422555785196e-06, + "loss": 0.6344, + "step": 1603 + }, + { + "epoch": 1.3279955825510767, + "grad_norm": 0.9171631336212158, + "learning_rate": 4.4725668425165605e-06, + "loss": 0.6454, + "step": 1604 + }, + { + "epoch": 1.3288238542241855, + "grad_norm": 0.8821735382080078, + "learning_rate": 4.4718910477813055e-06, + "loss": 0.6357, + "step": 1605 + }, + { + "epoch": 1.3296521258972942, + "grad_norm": 0.918499231338501, + "learning_rate": 4.471214871503514e-06, + "loss": 0.6361, + "step": 1606 + }, + { + "epoch": 1.3304803975704032, + "grad_norm": 0.9187904596328735, + "learning_rate": 4.470538313814021e-06, + "loss": 0.6362, + "step": 1607 + }, + { + "epoch": 1.3313086692435119, + "grad_norm": 0.9272632598876953, + "learning_rate": 4.469861374843734e-06, + "loss": 0.6659, + "step": 1608 + }, + { + "epoch": 1.3321369409166206, + "grad_norm": 0.8950392603874207, + "learning_rate": 4.469184054723632e-06, + "loss": 0.6551, + "step": 1609 + }, + { + "epoch": 1.3329652125897296, + "grad_norm": 0.9604961276054382, + "learning_rate": 4.468506353584773e-06, + "loss": 0.643, + "step": 1610 + }, + { + "epoch": 1.3337934842628383, + "grad_norm": 0.933073103427887, + "learning_rate": 4.467828271558286e-06, + "loss": 0.6193, + "step": 1611 + }, + { + "epoch": 1.334621755935947, + "grad_norm": 0.890964686870575, + "learning_rate": 4.467149808775371e-06, + "loss": 0.6423, + "step": 1612 + }, + { + "epoch": 1.3354500276090557, + "grad_norm": 0.8916820883750916, + "learning_rate": 4.466470965367306e-06, + "loss": 0.6399, + "step": 1613 + }, + { + "epoch": 1.3362782992821645, + "grad_norm": 0.9151459336280823, + "learning_rate": 4.465791741465439e-06, + "loss": 0.6439, + "step": 1614 + }, + { + "epoch": 1.3371065709552734, + "grad_norm": 0.9155884385108948, + "learning_rate": 4.465112137201197e-06, + "loss": 0.6516, + "step": 1615 + }, + { + "epoch": 1.3379348426283821, + "grad_norm": 0.9187855124473572, + "learning_rate": 4.464432152706073e-06, + "loss": 0.6263, + "step": 1616 + }, + { + "epoch": 1.3387631143014909, + "grad_norm": 0.9506946206092834, + "learning_rate": 4.46375178811164e-06, + "loss": 0.6586, + "step": 1617 + }, + { + "epoch": 1.3395913859745996, + "grad_norm": 0.9056413769721985, + "learning_rate": 4.463071043549543e-06, + "loss": 0.6301, + "step": 1618 + }, + { + "epoch": 1.3404196576477085, + "grad_norm": 0.9056457877159119, + "learning_rate": 4.462389919151498e-06, + "loss": 0.6352, + "step": 1619 + }, + { + "epoch": 1.3412479293208173, + "grad_norm": 0.9204022288322449, + "learning_rate": 4.461708415049297e-06, + "loss": 0.6446, + "step": 1620 + }, + { + "epoch": 1.342076200993926, + "grad_norm": 0.9411336183547974, + "learning_rate": 4.461026531374804e-06, + "loss": 0.6594, + "step": 1621 + }, + { + "epoch": 1.3429044726670347, + "grad_norm": 0.8956605792045593, + "learning_rate": 4.460344268259958e-06, + "loss": 0.6492, + "step": 1622 + }, + { + "epoch": 1.3437327443401434, + "grad_norm": 0.9010135531425476, + "learning_rate": 4.45966162583677e-06, + "loss": 0.6233, + "step": 1623 + }, + { + "epoch": 1.3445610160132524, + "grad_norm": 0.8948563933372498, + "learning_rate": 4.458978604237325e-06, + "loss": 0.6447, + "step": 1624 + }, + { + "epoch": 1.3453892876863611, + "grad_norm": 0.9227688312530518, + "learning_rate": 4.458295203593782e-06, + "loss": 0.6459, + "step": 1625 + }, + { + "epoch": 1.3462175593594699, + "grad_norm": 0.9532939195632935, + "learning_rate": 4.4576114240383725e-06, + "loss": 0.6724, + "step": 1626 + }, + { + "epoch": 1.3470458310325788, + "grad_norm": 0.9089633226394653, + "learning_rate": 4.4569272657034005e-06, + "loss": 0.6303, + "step": 1627 + }, + { + "epoch": 1.3478741027056875, + "grad_norm": 0.8903496861457825, + "learning_rate": 4.456242728721244e-06, + "loss": 0.6655, + "step": 1628 + }, + { + "epoch": 1.3487023743787963, + "grad_norm": 0.9009859561920166, + "learning_rate": 4.455557813224356e-06, + "loss": 0.6268, + "step": 1629 + }, + { + "epoch": 1.349530646051905, + "grad_norm": 0.9265439510345459, + "learning_rate": 4.454872519345261e-06, + "loss": 0.6586, + "step": 1630 + }, + { + "epoch": 1.3503589177250137, + "grad_norm": 0.8945439457893372, + "learning_rate": 4.454186847216556e-06, + "loss": 0.62, + "step": 1631 + }, + { + "epoch": 1.3511871893981227, + "grad_norm": 0.9313004016876221, + "learning_rate": 4.453500796970913e-06, + "loss": 0.6452, + "step": 1632 + }, + { + "epoch": 1.3520154610712314, + "grad_norm": 0.9005258083343506, + "learning_rate": 4.452814368741076e-06, + "loss": 0.6436, + "step": 1633 + }, + { + "epoch": 1.35284373274434, + "grad_norm": 0.8860014081001282, + "learning_rate": 4.452127562659864e-06, + "loss": 0.6423, + "step": 1634 + }, + { + "epoch": 1.353672004417449, + "grad_norm": 0.9049930572509766, + "learning_rate": 4.4514403788601645e-06, + "loss": 0.6343, + "step": 1635 + }, + { + "epoch": 1.3545002760905578, + "grad_norm": 0.8849244117736816, + "learning_rate": 4.4507528174749425e-06, + "loss": 0.6481, + "step": 1636 + }, + { + "epoch": 1.3553285477636665, + "grad_norm": 0.9214544892311096, + "learning_rate": 4.450064878637235e-06, + "loss": 0.6166, + "step": 1637 + }, + { + "epoch": 1.3561568194367752, + "grad_norm": 0.8948760628700256, + "learning_rate": 4.449376562480153e-06, + "loss": 0.6195, + "step": 1638 + }, + { + "epoch": 1.356985091109884, + "grad_norm": 0.9000245928764343, + "learning_rate": 4.448687869136878e-06, + "loss": 0.6378, + "step": 1639 + }, + { + "epoch": 1.3578133627829927, + "grad_norm": 0.8963801264762878, + "learning_rate": 4.447998798740665e-06, + "loss": 0.6118, + "step": 1640 + }, + { + "epoch": 1.3586416344561016, + "grad_norm": 0.8774341344833374, + "learning_rate": 4.447309351424843e-06, + "loss": 0.6184, + "step": 1641 + }, + { + "epoch": 1.3594699061292104, + "grad_norm": 0.8901870846748352, + "learning_rate": 4.446619527322815e-06, + "loss": 0.6499, + "step": 1642 + }, + { + "epoch": 1.360298177802319, + "grad_norm": 0.9108647108078003, + "learning_rate": 4.445929326568055e-06, + "loss": 0.6243, + "step": 1643 + }, + { + "epoch": 1.361126449475428, + "grad_norm": 0.9178787469863892, + "learning_rate": 4.445238749294109e-06, + "loss": 0.6458, + "step": 1644 + }, + { + "epoch": 1.3619547211485368, + "grad_norm": 0.9278320670127869, + "learning_rate": 4.444547795634598e-06, + "loss": 0.6281, + "step": 1645 + }, + { + "epoch": 1.3627829928216455, + "grad_norm": 0.8769298195838928, + "learning_rate": 4.443856465723216e-06, + "loss": 0.6387, + "step": 1646 + }, + { + "epoch": 1.3636112644947542, + "grad_norm": 0.9313247203826904, + "learning_rate": 4.443164759693729e-06, + "loss": 0.6304, + "step": 1647 + }, + { + "epoch": 1.364439536167863, + "grad_norm": 1.0037778615951538, + "learning_rate": 4.4424726776799745e-06, + "loss": 0.6378, + "step": 1648 + }, + { + "epoch": 1.365267807840972, + "grad_norm": 0.9090721011161804, + "learning_rate": 4.441780219815863e-06, + "loss": 0.6456, + "step": 1649 + }, + { + "epoch": 1.3660960795140806, + "grad_norm": 0.8968444466590881, + "learning_rate": 4.441087386235382e-06, + "loss": 0.6377, + "step": 1650 + }, + { + "epoch": 1.3669243511871894, + "grad_norm": 0.9638763070106506, + "learning_rate": 4.440394177072586e-06, + "loss": 0.651, + "step": 1651 + }, + { + "epoch": 1.3677526228602983, + "grad_norm": 0.9224078059196472, + "learning_rate": 4.439700592461604e-06, + "loss": 0.6492, + "step": 1652 + }, + { + "epoch": 1.368580894533407, + "grad_norm": 0.891180157661438, + "learning_rate": 4.439006632536639e-06, + "loss": 0.6402, + "step": 1653 + }, + { + "epoch": 1.3694091662065158, + "grad_norm": 0.8817652463912964, + "learning_rate": 4.438312297431967e-06, + "loss": 0.6095, + "step": 1654 + }, + { + "epoch": 1.3702374378796245, + "grad_norm": 0.8814468383789062, + "learning_rate": 4.437617587281932e-06, + "loss": 0.6334, + "step": 1655 + }, + { + "epoch": 1.3710657095527332, + "grad_norm": 0.8895065784454346, + "learning_rate": 4.436922502220958e-06, + "loss": 0.6163, + "step": 1656 + }, + { + "epoch": 1.371893981225842, + "grad_norm": 0.9043355584144592, + "learning_rate": 4.436227042383535e-06, + "loss": 0.6625, + "step": 1657 + }, + { + "epoch": 1.372722252898951, + "grad_norm": 0.942014753818512, + "learning_rate": 4.435531207904229e-06, + "loss": 0.6298, + "step": 1658 + }, + { + "epoch": 1.3735505245720596, + "grad_norm": 0.8830228447914124, + "learning_rate": 4.434834998917677e-06, + "loss": 0.6215, + "step": 1659 + }, + { + "epoch": 1.3743787962451683, + "grad_norm": 0.9138136506080627, + "learning_rate": 4.434138415558588e-06, + "loss": 0.63, + "step": 1660 + }, + { + "epoch": 1.3752070679182773, + "grad_norm": 0.8880951404571533, + "learning_rate": 4.433441457961746e-06, + "loss": 0.6411, + "step": 1661 + }, + { + "epoch": 1.376035339591386, + "grad_norm": 0.9435725808143616, + "learning_rate": 4.432744126262005e-06, + "loss": 0.6279, + "step": 1662 + }, + { + "epoch": 1.3768636112644947, + "grad_norm": 0.894296407699585, + "learning_rate": 4.432046420594292e-06, + "loss": 0.6298, + "step": 1663 + }, + { + "epoch": 1.3776918829376035, + "grad_norm": 0.8850218653678894, + "learning_rate": 4.431348341093608e-06, + "loss": 0.627, + "step": 1664 + }, + { + "epoch": 1.3785201546107122, + "grad_norm": 0.8956788182258606, + "learning_rate": 4.430649887895025e-06, + "loss": 0.6412, + "step": 1665 + }, + { + "epoch": 1.3793484262838211, + "grad_norm": 0.9128378033638, + "learning_rate": 4.429951061133683e-06, + "loss": 0.6208, + "step": 1666 + }, + { + "epoch": 1.3801766979569299, + "grad_norm": 0.8763598203659058, + "learning_rate": 4.429251860944803e-06, + "loss": 0.6317, + "step": 1667 + }, + { + "epoch": 1.3810049696300386, + "grad_norm": 0.9210475087165833, + "learning_rate": 4.428552287463672e-06, + "loss": 0.6321, + "step": 1668 + }, + { + "epoch": 1.3818332413031476, + "grad_norm": 0.9203745126724243, + "learning_rate": 4.427852340825651e-06, + "loss": 0.6442, + "step": 1669 + }, + { + "epoch": 1.3826615129762563, + "grad_norm": 0.9229859709739685, + "learning_rate": 4.4271520211661735e-06, + "loss": 0.6402, + "step": 1670 + }, + { + "epoch": 1.383489784649365, + "grad_norm": 0.9455426335334778, + "learning_rate": 4.426451328620744e-06, + "loss": 0.6402, + "step": 1671 + }, + { + "epoch": 1.3843180563224737, + "grad_norm": 0.8894345760345459, + "learning_rate": 4.425750263324942e-06, + "loss": 0.6429, + "step": 1672 + }, + { + "epoch": 1.3851463279955825, + "grad_norm": 0.871449887752533, + "learning_rate": 4.425048825414414e-06, + "loss": 0.6449, + "step": 1673 + }, + { + "epoch": 1.3859745996686914, + "grad_norm": 0.9102624654769897, + "learning_rate": 4.424347015024884e-06, + "loss": 0.634, + "step": 1674 + }, + { + "epoch": 1.3868028713418001, + "grad_norm": 0.881172776222229, + "learning_rate": 4.4236448322921445e-06, + "loss": 0.621, + "step": 1675 + }, + { + "epoch": 1.3876311430149089, + "grad_norm": 0.9143398404121399, + "learning_rate": 4.422942277352063e-06, + "loss": 0.6379, + "step": 1676 + }, + { + "epoch": 1.3884594146880176, + "grad_norm": 0.9155046343803406, + "learning_rate": 4.4222393503405754e-06, + "loss": 0.6168, + "step": 1677 + }, + { + "epoch": 1.3892876863611265, + "grad_norm": 0.9501062631607056, + "learning_rate": 4.421536051393694e-06, + "loss": 0.6238, + "step": 1678 + }, + { + "epoch": 1.3901159580342353, + "grad_norm": 0.9055907130241394, + "learning_rate": 4.420832380647498e-06, + "loss": 0.634, + "step": 1679 + }, + { + "epoch": 1.390944229707344, + "grad_norm": 0.9199142456054688, + "learning_rate": 4.420128338238141e-06, + "loss": 0.6283, + "step": 1680 + }, + { + "epoch": 1.3917725013804527, + "grad_norm": 0.8932124376296997, + "learning_rate": 4.419423924301851e-06, + "loss": 0.6146, + "step": 1681 + }, + { + "epoch": 1.3926007730535614, + "grad_norm": 0.9010130167007446, + "learning_rate": 4.418719138974924e-06, + "loss": 0.6409, + "step": 1682 + }, + { + "epoch": 1.3934290447266704, + "grad_norm": 0.9154212474822998, + "learning_rate": 4.41801398239373e-06, + "loss": 0.6115, + "step": 1683 + }, + { + "epoch": 1.3942573163997791, + "grad_norm": 0.9208158850669861, + "learning_rate": 4.417308454694709e-06, + "loss": 0.6242, + "step": 1684 + }, + { + "epoch": 1.3950855880728878, + "grad_norm": 0.932996928691864, + "learning_rate": 4.4166025560143756e-06, + "loss": 0.6277, + "step": 1685 + }, + { + "epoch": 1.3959138597459968, + "grad_norm": 0.8990349769592285, + "learning_rate": 4.415896286489314e-06, + "loss": 0.6128, + "step": 1686 + }, + { + "epoch": 1.3967421314191055, + "grad_norm": 0.926906168460846, + "learning_rate": 4.4151896462561805e-06, + "loss": 0.6442, + "step": 1687 + }, + { + "epoch": 1.3975704030922143, + "grad_norm": 0.9330859184265137, + "learning_rate": 4.414482635451704e-06, + "loss": 0.5995, + "step": 1688 + }, + { + "epoch": 1.398398674765323, + "grad_norm": 0.9039576649665833, + "learning_rate": 4.413775254212683e-06, + "loss": 0.6363, + "step": 1689 + }, + { + "epoch": 1.3992269464384317, + "grad_norm": 0.9228506684303284, + "learning_rate": 4.413067502675992e-06, + "loss": 0.6294, + "step": 1690 + }, + { + "epoch": 1.4000552181115407, + "grad_norm": 0.921620786190033, + "learning_rate": 4.412359380978571e-06, + "loss": 0.64, + "step": 1691 + }, + { + "epoch": 1.4008834897846494, + "grad_norm": 0.9379149079322815, + "learning_rate": 4.4116508892574375e-06, + "loss": 0.6207, + "step": 1692 + }, + { + "epoch": 1.401711761457758, + "grad_norm": 1.0528016090393066, + "learning_rate": 4.410942027649677e-06, + "loss": 0.6204, + "step": 1693 + }, + { + "epoch": 1.402540033130867, + "grad_norm": 0.9510579705238342, + "learning_rate": 4.410232796292447e-06, + "loss": 0.6314, + "step": 1694 + }, + { + "epoch": 1.4033683048039758, + "grad_norm": 0.8922870755195618, + "learning_rate": 4.409523195322979e-06, + "loss": 0.6056, + "step": 1695 + }, + { + "epoch": 1.4041965764770845, + "grad_norm": 0.9239864349365234, + "learning_rate": 4.408813224878573e-06, + "loss": 0.6404, + "step": 1696 + }, + { + "epoch": 1.4050248481501932, + "grad_norm": 0.8696756362915039, + "learning_rate": 4.408102885096601e-06, + "loss": 0.6553, + "step": 1697 + }, + { + "epoch": 1.405853119823302, + "grad_norm": 0.9176343679428101, + "learning_rate": 4.407392176114509e-06, + "loss": 0.6286, + "step": 1698 + }, + { + "epoch": 1.4066813914964107, + "grad_norm": 0.9825496077537537, + "learning_rate": 4.40668109806981e-06, + "loss": 0.6421, + "step": 1699 + }, + { + "epoch": 1.4075096631695196, + "grad_norm": 0.9334873557090759, + "learning_rate": 4.405969651100093e-06, + "loss": 0.6347, + "step": 1700 + }, + { + "epoch": 1.4083379348426284, + "grad_norm": 0.8881192803382874, + "learning_rate": 4.405257835343015e-06, + "loss": 0.6377, + "step": 1701 + }, + { + "epoch": 1.409166206515737, + "grad_norm": 0.8896775245666504, + "learning_rate": 4.404545650936307e-06, + "loss": 0.624, + "step": 1702 + }, + { + "epoch": 1.409994478188846, + "grad_norm": 0.9361454844474792, + "learning_rate": 4.40383309801777e-06, + "loss": 0.6561, + "step": 1703 + }, + { + "epoch": 1.4108227498619548, + "grad_norm": 0.9082345962524414, + "learning_rate": 4.403120176725276e-06, + "loss": 0.6415, + "step": 1704 + }, + { + "epoch": 1.4116510215350635, + "grad_norm": 0.9453962445259094, + "learning_rate": 4.402406887196767e-06, + "loss": 0.6443, + "step": 1705 + }, + { + "epoch": 1.4124792932081722, + "grad_norm": 0.9420861005783081, + "learning_rate": 4.401693229570261e-06, + "loss": 0.6544, + "step": 1706 + }, + { + "epoch": 1.413307564881281, + "grad_norm": 0.869742214679718, + "learning_rate": 4.400979203983841e-06, + "loss": 0.6072, + "step": 1707 + }, + { + "epoch": 1.41413583655439, + "grad_norm": 0.915322482585907, + "learning_rate": 4.4002648105756665e-06, + "loss": 0.6657, + "step": 1708 + }, + { + "epoch": 1.4149641082274986, + "grad_norm": 0.9029238224029541, + "learning_rate": 4.3995500494839656e-06, + "loss": 0.6157, + "step": 1709 + }, + { + "epoch": 1.4157923799006074, + "grad_norm": 0.8992076516151428, + "learning_rate": 4.398834920847037e-06, + "loss": 0.6321, + "step": 1710 + }, + { + "epoch": 1.4166206515737163, + "grad_norm": 0.9099612236022949, + "learning_rate": 4.398119424803252e-06, + "loss": 0.6123, + "step": 1711 + }, + { + "epoch": 1.417448923246825, + "grad_norm": 0.9081803560256958, + "learning_rate": 4.397403561491052e-06, + "loss": 0.6369, + "step": 1712 + }, + { + "epoch": 1.4182771949199338, + "grad_norm": 0.8873599767684937, + "learning_rate": 4.3966873310489505e-06, + "loss": 0.6503, + "step": 1713 + }, + { + "epoch": 1.4191054665930425, + "grad_norm": 0.8841976523399353, + "learning_rate": 4.395970733615531e-06, + "loss": 0.6303, + "step": 1714 + }, + { + "epoch": 1.4199337382661512, + "grad_norm": 0.8922518491744995, + "learning_rate": 4.395253769329449e-06, + "loss": 0.6105, + "step": 1715 + }, + { + "epoch": 1.42076200993926, + "grad_norm": 0.9023404717445374, + "learning_rate": 4.39453643832943e-06, + "loss": 0.6292, + "step": 1716 + }, + { + "epoch": 1.421590281612369, + "grad_norm": 0.9042630791664124, + "learning_rate": 4.39381874075427e-06, + "loss": 0.6675, + "step": 1717 + }, + { + "epoch": 1.4224185532854776, + "grad_norm": 0.8933060169219971, + "learning_rate": 4.393100676742838e-06, + "loss": 0.6125, + "step": 1718 + }, + { + "epoch": 1.4232468249585863, + "grad_norm": 0.8821312189102173, + "learning_rate": 4.392382246434073e-06, + "loss": 0.6266, + "step": 1719 + }, + { + "epoch": 1.4240750966316953, + "grad_norm": 0.904460608959198, + "learning_rate": 4.391663449966983e-06, + "loss": 0.5984, + "step": 1720 + }, + { + "epoch": 1.424903368304804, + "grad_norm": 0.8999350666999817, + "learning_rate": 4.39094428748065e-06, + "loss": 0.6653, + "step": 1721 + }, + { + "epoch": 1.4257316399779127, + "grad_norm": 0.9061364531517029, + "learning_rate": 4.390224759114224e-06, + "loss": 0.6306, + "step": 1722 + }, + { + "epoch": 1.4265599116510215, + "grad_norm": 0.8995294570922852, + "learning_rate": 4.389504865006927e-06, + "loss": 0.6656, + "step": 1723 + }, + { + "epoch": 1.4273881833241302, + "grad_norm": 0.904737114906311, + "learning_rate": 4.388784605298053e-06, + "loss": 0.6234, + "step": 1724 + }, + { + "epoch": 1.4282164549972391, + "grad_norm": 0.8827239871025085, + "learning_rate": 4.388063980126966e-06, + "loss": 0.6418, + "step": 1725 + }, + { + "epoch": 1.4290447266703479, + "grad_norm": 0.8791747689247131, + "learning_rate": 4.387342989633097e-06, + "loss": 0.6174, + "step": 1726 + }, + { + "epoch": 1.4298729983434566, + "grad_norm": 0.8831204175949097, + "learning_rate": 4.386621633955955e-06, + "loss": 0.6222, + "step": 1727 + }, + { + "epoch": 1.4307012700165656, + "grad_norm": 0.908511221408844, + "learning_rate": 4.3858999132351124e-06, + "loss": 0.6373, + "step": 1728 + }, + { + "epoch": 1.4315295416896743, + "grad_norm": 0.900497555732727, + "learning_rate": 4.385177827610218e-06, + "loss": 0.6291, + "step": 1729 + }, + { + "epoch": 1.432357813362783, + "grad_norm": 0.930695652961731, + "learning_rate": 4.384455377220986e-06, + "loss": 0.6408, + "step": 1730 + }, + { + "epoch": 1.4331860850358917, + "grad_norm": 0.9583060145378113, + "learning_rate": 4.383732562207206e-06, + "loss": 0.6452, + "step": 1731 + }, + { + "epoch": 1.4340143567090005, + "grad_norm": 0.9173291921615601, + "learning_rate": 4.383009382708734e-06, + "loss": 0.6491, + "step": 1732 + }, + { + "epoch": 1.4348426283821094, + "grad_norm": 0.9054985642433167, + "learning_rate": 4.382285838865501e-06, + "loss": 0.6179, + "step": 1733 + }, + { + "epoch": 1.4356709000552181, + "grad_norm": 0.9162827730178833, + "learning_rate": 4.381561930817504e-06, + "loss": 0.6428, + "step": 1734 + }, + { + "epoch": 1.4364991717283269, + "grad_norm": 0.9013632535934448, + "learning_rate": 4.380837658704813e-06, + "loss": 0.6385, + "step": 1735 + }, + { + "epoch": 1.4373274434014356, + "grad_norm": 0.9274044632911682, + "learning_rate": 4.3801130226675685e-06, + "loss": 0.6508, + "step": 1736 + }, + { + "epoch": 1.4381557150745445, + "grad_norm": 0.8884710669517517, + "learning_rate": 4.37938802284598e-06, + "loss": 0.6675, + "step": 1737 + }, + { + "epoch": 1.4389839867476533, + "grad_norm": 0.9156724214553833, + "learning_rate": 4.378662659380329e-06, + "loss": 0.6424, + "step": 1738 + }, + { + "epoch": 1.439812258420762, + "grad_norm": 0.9267123341560364, + "learning_rate": 4.377936932410966e-06, + "loss": 0.6333, + "step": 1739 + }, + { + "epoch": 1.4406405300938707, + "grad_norm": 0.8859686851501465, + "learning_rate": 4.377210842078311e-06, + "loss": 0.6343, + "step": 1740 + }, + { + "epoch": 1.4414688017669794, + "grad_norm": 0.92928147315979, + "learning_rate": 4.376484388522858e-06, + "loss": 0.6124, + "step": 1741 + }, + { + "epoch": 1.4422970734400884, + "grad_norm": 0.8908365964889526, + "learning_rate": 4.375757571885169e-06, + "loss": 0.631, + "step": 1742 + }, + { + "epoch": 1.4431253451131971, + "grad_norm": 0.8956666588783264, + "learning_rate": 4.375030392305875e-06, + "loss": 0.6314, + "step": 1743 + }, + { + "epoch": 1.4439536167863058, + "grad_norm": 0.8836198449134827, + "learning_rate": 4.374302849925679e-06, + "loss": 0.6179, + "step": 1744 + }, + { + "epoch": 1.4447818884594148, + "grad_norm": 0.87999027967453, + "learning_rate": 4.373574944885354e-06, + "loss": 0.6226, + "step": 1745 + }, + { + "epoch": 1.4456101601325235, + "grad_norm": 0.9118010997772217, + "learning_rate": 4.3728466773257414e-06, + "loss": 0.6319, + "step": 1746 + }, + { + "epoch": 1.4464384318056323, + "grad_norm": 0.8989576101303101, + "learning_rate": 4.3721180473877555e-06, + "loss": 0.6049, + "step": 1747 + }, + { + "epoch": 1.447266703478741, + "grad_norm": 0.8896577954292297, + "learning_rate": 4.371389055212379e-06, + "loss": 0.6242, + "step": 1748 + }, + { + "epoch": 1.4480949751518497, + "grad_norm": 0.8993588089942932, + "learning_rate": 4.370659700940665e-06, + "loss": 0.6255, + "step": 1749 + }, + { + "epoch": 1.4489232468249587, + "grad_norm": 0.896486759185791, + "learning_rate": 4.369929984713738e-06, + "loss": 0.6354, + "step": 1750 + }, + { + "epoch": 1.4497515184980674, + "grad_norm": 0.9003145694732666, + "learning_rate": 4.369199906672789e-06, + "loss": 0.6464, + "step": 1751 + }, + { + "epoch": 1.450579790171176, + "grad_norm": 0.8885106444358826, + "learning_rate": 4.368469466959083e-06, + "loss": 0.6567, + "step": 1752 + }, + { + "epoch": 1.451408061844285, + "grad_norm": 0.8914943933486938, + "learning_rate": 4.367738665713953e-06, + "loss": 0.6784, + "step": 1753 + }, + { + "epoch": 1.4522363335173938, + "grad_norm": 0.933029055595398, + "learning_rate": 4.367007503078802e-06, + "loss": 0.6495, + "step": 1754 + }, + { + "epoch": 1.4530646051905025, + "grad_norm": 0.8788006901741028, + "learning_rate": 4.366275979195102e-06, + "loss": 0.6352, + "step": 1755 + }, + { + "epoch": 1.4538928768636112, + "grad_norm": 0.8768455982208252, + "learning_rate": 4.365544094204399e-06, + "loss": 0.6329, + "step": 1756 + }, + { + "epoch": 1.45472114853672, + "grad_norm": 0.9021796584129333, + "learning_rate": 4.364811848248303e-06, + "loss": 0.6616, + "step": 1757 + }, + { + "epoch": 1.4555494202098287, + "grad_norm": 0.9197986125946045, + "learning_rate": 4.364079241468499e-06, + "loss": 0.6248, + "step": 1758 + }, + { + "epoch": 1.4563776918829376, + "grad_norm": 0.9095743894577026, + "learning_rate": 4.3633462740067375e-06, + "loss": 0.6722, + "step": 1759 + }, + { + "epoch": 1.4572059635560464, + "grad_norm": 0.9114678502082825, + "learning_rate": 4.362612946004843e-06, + "loss": 0.6278, + "step": 1760 + }, + { + "epoch": 1.458034235229155, + "grad_norm": 0.8978061676025391, + "learning_rate": 4.3618792576047054e-06, + "loss": 0.633, + "step": 1761 + }, + { + "epoch": 1.458862506902264, + "grad_norm": 0.8578571677207947, + "learning_rate": 4.361145208948287e-06, + "loss": 0.6263, + "step": 1762 + }, + { + "epoch": 1.4596907785753728, + "grad_norm": 0.9333001375198364, + "learning_rate": 4.360410800177622e-06, + "loss": 0.6383, + "step": 1763 + }, + { + "epoch": 1.4605190502484815, + "grad_norm": 0.9448907971382141, + "learning_rate": 4.359676031434808e-06, + "loss": 0.6324, + "step": 1764 + }, + { + "epoch": 1.4613473219215902, + "grad_norm": 0.8806003928184509, + "learning_rate": 4.358940902862019e-06, + "loss": 0.6365, + "step": 1765 + }, + { + "epoch": 1.462175593594699, + "grad_norm": 0.9975590109825134, + "learning_rate": 4.358205414601493e-06, + "loss": 0.6436, + "step": 1766 + }, + { + "epoch": 1.463003865267808, + "grad_norm": 1.0076019763946533, + "learning_rate": 4.357469566795542e-06, + "loss": 0.6234, + "step": 1767 + }, + { + "epoch": 1.4638321369409166, + "grad_norm": 0.9254529476165771, + "learning_rate": 4.356733359586545e-06, + "loss": 0.6244, + "step": 1768 + }, + { + "epoch": 1.4646604086140254, + "grad_norm": 0.9129828810691833, + "learning_rate": 4.35599679311695e-06, + "loss": 0.6519, + "step": 1769 + }, + { + "epoch": 1.4654886802871343, + "grad_norm": 0.9141498804092407, + "learning_rate": 4.355259867529278e-06, + "loss": 0.6343, + "step": 1770 + }, + { + "epoch": 1.466316951960243, + "grad_norm": 0.920790433883667, + "learning_rate": 4.354522582966116e-06, + "loss": 0.6505, + "step": 1771 + }, + { + "epoch": 1.4671452236333518, + "grad_norm": 0.9683033227920532, + "learning_rate": 4.353784939570121e-06, + "loss": 0.6238, + "step": 1772 + }, + { + "epoch": 1.4679734953064605, + "grad_norm": 0.882957935333252, + "learning_rate": 4.3530469374840226e-06, + "loss": 0.6358, + "step": 1773 + }, + { + "epoch": 1.4688017669795692, + "grad_norm": 0.943648099899292, + "learning_rate": 4.352308576850615e-06, + "loss": 0.6274, + "step": 1774 + }, + { + "epoch": 1.4696300386526782, + "grad_norm": 0.9609524607658386, + "learning_rate": 4.351569857812764e-06, + "loss": 0.6368, + "step": 1775 + }, + { + "epoch": 1.4704583103257869, + "grad_norm": 0.9170458912849426, + "learning_rate": 4.3508307805134055e-06, + "loss": 0.6515, + "step": 1776 + }, + { + "epoch": 1.4712865819988956, + "grad_norm": 0.8956430554389954, + "learning_rate": 4.350091345095545e-06, + "loss": 0.6394, + "step": 1777 + }, + { + "epoch": 1.4721148536720043, + "grad_norm": 0.9145321846008301, + "learning_rate": 4.349351551702256e-06, + "loss": 0.6572, + "step": 1778 + }, + { + "epoch": 1.4729431253451133, + "grad_norm": 0.8807089328765869, + "learning_rate": 4.34861140047668e-06, + "loss": 0.6391, + "step": 1779 + }, + { + "epoch": 1.473771397018222, + "grad_norm": 0.8828510642051697, + "learning_rate": 4.347870891562032e-06, + "loss": 0.6583, + "step": 1780 + }, + { + "epoch": 1.4745996686913307, + "grad_norm": 0.9017962217330933, + "learning_rate": 4.34713002510159e-06, + "loss": 0.6337, + "step": 1781 + }, + { + "epoch": 1.4754279403644395, + "grad_norm": 0.8738356232643127, + "learning_rate": 4.346388801238708e-06, + "loss": 0.637, + "step": 1782 + }, + { + "epoch": 1.4762562120375482, + "grad_norm": 0.8805137872695923, + "learning_rate": 4.345647220116805e-06, + "loss": 0.6306, + "step": 1783 + }, + { + "epoch": 1.4770844837106571, + "grad_norm": 0.923599362373352, + "learning_rate": 4.34490528187937e-06, + "loss": 0.6254, + "step": 1784 + }, + { + "epoch": 1.4779127553837659, + "grad_norm": 0.9181063175201416, + "learning_rate": 4.344162986669961e-06, + "loss": 0.6257, + "step": 1785 + }, + { + "epoch": 1.4787410270568746, + "grad_norm": 0.9060764312744141, + "learning_rate": 4.343420334632204e-06, + "loss": 0.6326, + "step": 1786 + }, + { + "epoch": 1.4795692987299836, + "grad_norm": 0.9101755023002625, + "learning_rate": 4.342677325909798e-06, + "loss": 0.6319, + "step": 1787 + }, + { + "epoch": 1.4803975704030923, + "grad_norm": 0.9417787790298462, + "learning_rate": 4.341933960646505e-06, + "loss": 0.6147, + "step": 1788 + }, + { + "epoch": 1.481225842076201, + "grad_norm": 0.9138630628585815, + "learning_rate": 4.341190238986162e-06, + "loss": 0.6256, + "step": 1789 + }, + { + "epoch": 1.4820541137493097, + "grad_norm": 0.9004842638969421, + "learning_rate": 4.340446161072671e-06, + "loss": 0.6443, + "step": 1790 + }, + { + "epoch": 1.4828823854224185, + "grad_norm": 0.9181780219078064, + "learning_rate": 4.339701727050004e-06, + "loss": 0.6408, + "step": 1791 + }, + { + "epoch": 1.4837106570955274, + "grad_norm": 0.880736768245697, + "learning_rate": 4.338956937062203e-06, + "loss": 0.6436, + "step": 1792 + }, + { + "epoch": 1.4845389287686361, + "grad_norm": 0.9168913960456848, + "learning_rate": 4.338211791253376e-06, + "loss": 0.6391, + "step": 1793 + }, + { + "epoch": 1.4853672004417449, + "grad_norm": 0.942925751209259, + "learning_rate": 4.337466289767704e-06, + "loss": 0.6303, + "step": 1794 + }, + { + "epoch": 1.4861954721148538, + "grad_norm": 0.8768817186355591, + "learning_rate": 4.336720432749433e-06, + "loss": 0.6239, + "step": 1795 + }, + { + "epoch": 1.4870237437879625, + "grad_norm": 0.9388988614082336, + "learning_rate": 4.33597422034288e-06, + "loss": 0.6441, + "step": 1796 + }, + { + "epoch": 1.4878520154610713, + "grad_norm": 0.9257978200912476, + "learning_rate": 4.3352276526924295e-06, + "loss": 0.6373, + "step": 1797 + }, + { + "epoch": 1.48868028713418, + "grad_norm": 0.9225389957427979, + "learning_rate": 4.334480729942537e-06, + "loss": 0.6555, + "step": 1798 + }, + { + "epoch": 1.4895085588072887, + "grad_norm": 0.8992138504981995, + "learning_rate": 4.3337334522377235e-06, + "loss": 0.6375, + "step": 1799 + }, + { + "epoch": 1.4903368304803974, + "grad_norm": 0.9207738041877747, + "learning_rate": 4.332985819722582e-06, + "loss": 0.6439, + "step": 1800 + }, + { + "epoch": 1.4911651021535064, + "grad_norm": 0.8866941332817078, + "learning_rate": 4.3322378325417705e-06, + "loss": 0.6326, + "step": 1801 + }, + { + "epoch": 1.4919933738266151, + "grad_norm": 0.9069377779960632, + "learning_rate": 4.3314894908400175e-06, + "loss": 0.6193, + "step": 1802 + }, + { + "epoch": 1.4928216454997238, + "grad_norm": 0.8918569087982178, + "learning_rate": 4.3307407947621216e-06, + "loss": 0.6374, + "step": 1803 + }, + { + "epoch": 1.4936499171728328, + "grad_norm": 0.8993483781814575, + "learning_rate": 4.329991744452948e-06, + "loss": 0.6251, + "step": 1804 + }, + { + "epoch": 1.4944781888459415, + "grad_norm": 0.918445885181427, + "learning_rate": 4.32924234005743e-06, + "loss": 0.6494, + "step": 1805 + }, + { + "epoch": 1.4953064605190503, + "grad_norm": 0.9245757460594177, + "learning_rate": 4.328492581720573e-06, + "loss": 0.6172, + "step": 1806 + }, + { + "epoch": 1.496134732192159, + "grad_norm": 0.9166370034217834, + "learning_rate": 4.327742469587445e-06, + "loss": 0.622, + "step": 1807 + }, + { + "epoch": 1.4969630038652677, + "grad_norm": 0.9083957076072693, + "learning_rate": 4.326992003803187e-06, + "loss": 0.6205, + "step": 1808 + }, + { + "epoch": 1.4977912755383767, + "grad_norm": 0.8895595073699951, + "learning_rate": 4.326241184513008e-06, + "loss": 0.6245, + "step": 1809 + }, + { + "epoch": 1.4986195472114854, + "grad_norm": 0.9845629334449768, + "learning_rate": 4.325490011862184e-06, + "loss": 0.6163, + "step": 1810 + }, + { + "epoch": 1.499447818884594, + "grad_norm": 0.9501281976699829, + "learning_rate": 4.324738485996058e-06, + "loss": 0.6218, + "step": 1811 + }, + { + "epoch": 1.500276090557703, + "grad_norm": 0.9685792922973633, + "learning_rate": 4.323986607060047e-06, + "loss": 0.634, + "step": 1812 + }, + { + "epoch": 1.5011043622308118, + "grad_norm": 0.8959722518920898, + "learning_rate": 4.32323437519963e-06, + "loss": 0.6316, + "step": 1813 + }, + { + "epoch": 1.5019326339039205, + "grad_norm": 0.914463222026825, + "learning_rate": 4.322481790560356e-06, + "loss": 0.6258, + "step": 1814 + }, + { + "epoch": 1.5027609055770292, + "grad_norm": 0.9052320718765259, + "learning_rate": 4.321728853287846e-06, + "loss": 0.6345, + "step": 1815 + }, + { + "epoch": 1.503589177250138, + "grad_norm": 0.9306650161743164, + "learning_rate": 4.320975563527785e-06, + "loss": 0.5976, + "step": 1816 + }, + { + "epoch": 1.5044174489232467, + "grad_norm": 0.8930769562721252, + "learning_rate": 4.320221921425926e-06, + "loss": 0.6427, + "step": 1817 + }, + { + "epoch": 1.5052457205963556, + "grad_norm": 0.9205043911933899, + "learning_rate": 4.3194679271280935e-06, + "loss": 0.617, + "step": 1818 + }, + { + "epoch": 1.5060739922694644, + "grad_norm": 1.0371367931365967, + "learning_rate": 4.318713580780179e-06, + "loss": 0.6301, + "step": 1819 + }, + { + "epoch": 1.5069022639425733, + "grad_norm": 0.9442588686943054, + "learning_rate": 4.317958882528139e-06, + "loss": 0.6252, + "step": 1820 + }, + { + "epoch": 1.507730535615682, + "grad_norm": 0.9214493036270142, + "learning_rate": 4.317203832518003e-06, + "loss": 0.6326, + "step": 1821 + }, + { + "epoch": 1.5085588072887908, + "grad_norm": 0.9270440340042114, + "learning_rate": 4.316448430895866e-06, + "loss": 0.6377, + "step": 1822 + }, + { + "epoch": 1.5093870789618995, + "grad_norm": 0.8822277188301086, + "learning_rate": 4.315692677807889e-06, + "loss": 0.6277, + "step": 1823 + }, + { + "epoch": 1.5102153506350082, + "grad_norm": 0.8926153779029846, + "learning_rate": 4.314936573400305e-06, + "loss": 0.6439, + "step": 1824 + }, + { + "epoch": 1.511043622308117, + "grad_norm": 0.9214072227478027, + "learning_rate": 4.314180117819412e-06, + "loss": 0.6519, + "step": 1825 + }, + { + "epoch": 1.5118718939812257, + "grad_norm": 0.881522536277771, + "learning_rate": 4.313423311211578e-06, + "loss": 0.6213, + "step": 1826 + }, + { + "epoch": 1.5127001656543346, + "grad_norm": 0.8960263729095459, + "learning_rate": 4.312666153723238e-06, + "loss": 0.6592, + "step": 1827 + }, + { + "epoch": 1.5135284373274434, + "grad_norm": 0.9176173210144043, + "learning_rate": 4.3119086455008956e-06, + "loss": 0.6569, + "step": 1828 + }, + { + "epoch": 1.5143567090005523, + "grad_norm": 0.9312153458595276, + "learning_rate": 4.311150786691119e-06, + "loss": 0.6324, + "step": 1829 + }, + { + "epoch": 1.515184980673661, + "grad_norm": 0.9202299118041992, + "learning_rate": 4.310392577440549e-06, + "loss": 0.6381, + "step": 1830 + }, + { + "epoch": 1.5160132523467698, + "grad_norm": 0.8895224928855896, + "learning_rate": 4.309634017895893e-06, + "loss": 0.6491, + "step": 1831 + }, + { + "epoch": 1.5168415240198785, + "grad_norm": 0.9424769878387451, + "learning_rate": 4.308875108203922e-06, + "loss": 0.6342, + "step": 1832 + }, + { + "epoch": 1.5176697956929872, + "grad_norm": 0.8676947951316833, + "learning_rate": 4.30811584851148e-06, + "loss": 0.6287, + "step": 1833 + }, + { + "epoch": 1.518498067366096, + "grad_norm": 0.8764673471450806, + "learning_rate": 4.3073562389654764e-06, + "loss": 0.6123, + "step": 1834 + }, + { + "epoch": 1.5193263390392049, + "grad_norm": 0.9222663640975952, + "learning_rate": 4.306596279712889e-06, + "loss": 0.6433, + "step": 1835 + }, + { + "epoch": 1.5201546107123136, + "grad_norm": 0.8839907050132751, + "learning_rate": 4.305835970900762e-06, + "loss": 0.6416, + "step": 1836 + }, + { + "epoch": 1.5209828823854226, + "grad_norm": 0.9095149636268616, + "learning_rate": 4.305075312676209e-06, + "loss": 0.6417, + "step": 1837 + }, + { + "epoch": 1.5218111540585313, + "grad_norm": 0.8718581199645996, + "learning_rate": 4.304314305186409e-06, + "loss": 0.6174, + "step": 1838 + }, + { + "epoch": 1.52263942573164, + "grad_norm": 0.9424081444740295, + "learning_rate": 4.303552948578611e-06, + "loss": 0.6064, + "step": 1839 + }, + { + "epoch": 1.5234676974047487, + "grad_norm": 0.9170297384262085, + "learning_rate": 4.30279124300013e-06, + "loss": 0.6078, + "step": 1840 + }, + { + "epoch": 1.5242959690778575, + "grad_norm": 0.8814254999160767, + "learning_rate": 4.302029188598348e-06, + "loss": 0.6194, + "step": 1841 + }, + { + "epoch": 1.5251242407509662, + "grad_norm": 0.9248694181442261, + "learning_rate": 4.301266785520718e-06, + "loss": 0.6452, + "step": 1842 + }, + { + "epoch": 1.5259525124240751, + "grad_norm": 0.8757627010345459, + "learning_rate": 4.300504033914754e-06, + "loss": 0.6486, + "step": 1843 + }, + { + "epoch": 1.5267807840971839, + "grad_norm": 0.9229605793952942, + "learning_rate": 4.299740933928045e-06, + "loss": 0.6363, + "step": 1844 + }, + { + "epoch": 1.5276090557702926, + "grad_norm": 0.9192973971366882, + "learning_rate": 4.298977485708243e-06, + "loss": 0.6383, + "step": 1845 + }, + { + "epoch": 1.5284373274434015, + "grad_norm": 0.9071361422538757, + "learning_rate": 4.2982136894030655e-06, + "loss": 0.6541, + "step": 1846 + }, + { + "epoch": 1.5292655991165103, + "grad_norm": 0.9067169427871704, + "learning_rate": 4.297449545160303e-06, + "loss": 0.6317, + "step": 1847 + }, + { + "epoch": 1.530093870789619, + "grad_norm": 0.9047722816467285, + "learning_rate": 4.2966850531278086e-06, + "loss": 0.6365, + "step": 1848 + }, + { + "epoch": 1.5309221424627277, + "grad_norm": 0.8948541879653931, + "learning_rate": 4.295920213453504e-06, + "loss": 0.6268, + "step": 1849 + }, + { + "epoch": 1.5317504141358365, + "grad_norm": 0.9061904549598694, + "learning_rate": 4.295155026285382e-06, + "loss": 0.6271, + "step": 1850 + }, + { + "epoch": 1.5325786858089452, + "grad_norm": 0.9092041850090027, + "learning_rate": 4.2943894917714946e-06, + "loss": 0.6148, + "step": 1851 + }, + { + "epoch": 1.5334069574820541, + "grad_norm": 0.932331919670105, + "learning_rate": 4.293623610059968e-06, + "loss": 0.647, + "step": 1852 + }, + { + "epoch": 1.5342352291551629, + "grad_norm": 0.9136486649513245, + "learning_rate": 4.292857381298993e-06, + "loss": 0.6326, + "step": 1853 + }, + { + "epoch": 1.5350635008282718, + "grad_norm": 0.9760140180587769, + "learning_rate": 4.292090805636827e-06, + "loss": 0.6324, + "step": 1854 + }, + { + "epoch": 1.5358917725013805, + "grad_norm": 0.9075965881347656, + "learning_rate": 4.291323883221797e-06, + "loss": 0.6194, + "step": 1855 + }, + { + "epoch": 1.5367200441744893, + "grad_norm": 0.8804707527160645, + "learning_rate": 4.290556614202292e-06, + "loss": 0.6246, + "step": 1856 + }, + { + "epoch": 1.537548315847598, + "grad_norm": 0.9123021960258484, + "learning_rate": 4.2897889987267755e-06, + "loss": 0.6156, + "step": 1857 + }, + { + "epoch": 1.5383765875207067, + "grad_norm": 0.9396944642066956, + "learning_rate": 4.289021036943771e-06, + "loss": 0.6268, + "step": 1858 + }, + { + "epoch": 1.5392048591938154, + "grad_norm": 0.858566164970398, + "learning_rate": 4.288252729001874e-06, + "loss": 0.6237, + "step": 1859 + }, + { + "epoch": 1.5400331308669244, + "grad_norm": 0.899341344833374, + "learning_rate": 4.287484075049743e-06, + "loss": 0.5973, + "step": 1860 + }, + { + "epoch": 1.5408614025400331, + "grad_norm": 0.9021979570388794, + "learning_rate": 4.286715075236106e-06, + "loss": 0.6164, + "step": 1861 + }, + { + "epoch": 1.541689674213142, + "grad_norm": 0.9693799018859863, + "learning_rate": 4.285945729709757e-06, + "loss": 0.6409, + "step": 1862 + }, + { + "epoch": 1.5425179458862508, + "grad_norm": 0.9177312254905701, + "learning_rate": 4.285176038619559e-06, + "loss": 0.6276, + "step": 1863 + }, + { + "epoch": 1.5433462175593595, + "grad_norm": 0.9382041692733765, + "learning_rate": 4.284406002114438e-06, + "loss": 0.643, + "step": 1864 + }, + { + "epoch": 1.5441744892324683, + "grad_norm": 0.8778985142707825, + "learning_rate": 4.28363562034339e-06, + "loss": 0.6119, + "step": 1865 + }, + { + "epoch": 1.545002760905577, + "grad_norm": 0.8890271782875061, + "learning_rate": 4.282864893455476e-06, + "loss": 0.6388, + "step": 1866 + }, + { + "epoch": 1.5458310325786857, + "grad_norm": 0.9081838130950928, + "learning_rate": 4.282093821599824e-06, + "loss": 0.6464, + "step": 1867 + }, + { + "epoch": 1.5466593042517944, + "grad_norm": 0.8805630803108215, + "learning_rate": 4.281322404925631e-06, + "loss": 0.6533, + "step": 1868 + }, + { + "epoch": 1.5474875759249034, + "grad_norm": 0.8817034363746643, + "learning_rate": 4.280550643582158e-06, + "loss": 0.643, + "step": 1869 + }, + { + "epoch": 1.548315847598012, + "grad_norm": 0.9003176689147949, + "learning_rate": 4.279778537718733e-06, + "loss": 0.6322, + "step": 1870 + }, + { + "epoch": 1.549144119271121, + "grad_norm": 0.8982899785041809, + "learning_rate": 4.2790060874847525e-06, + "loss": 0.6456, + "step": 1871 + }, + { + "epoch": 1.5499723909442298, + "grad_norm": 0.8730716109275818, + "learning_rate": 4.278233293029677e-06, + "loss": 0.6428, + "step": 1872 + }, + { + "epoch": 1.5508006626173385, + "grad_norm": 0.9135062098503113, + "learning_rate": 4.277460154503036e-06, + "loss": 0.645, + "step": 1873 + }, + { + "epoch": 1.5516289342904472, + "grad_norm": 0.9091652631759644, + "learning_rate": 4.276686672054424e-06, + "loss": 0.6256, + "step": 1874 + }, + { + "epoch": 1.552457205963556, + "grad_norm": 0.9090970754623413, + "learning_rate": 4.275912845833504e-06, + "loss": 0.6404, + "step": 1875 + }, + { + "epoch": 1.5532854776366647, + "grad_norm": 0.91478031873703, + "learning_rate": 4.275138675990003e-06, + "loss": 0.6498, + "step": 1876 + }, + { + "epoch": 1.5541137493097736, + "grad_norm": 0.8979254961013794, + "learning_rate": 4.274364162673715e-06, + "loss": 0.6275, + "step": 1877 + }, + { + "epoch": 1.5549420209828824, + "grad_norm": 0.92198646068573, + "learning_rate": 4.273589306034503e-06, + "loss": 0.6585, + "step": 1878 + }, + { + "epoch": 1.5557702926559913, + "grad_norm": 0.9540958404541016, + "learning_rate": 4.272814106222293e-06, + "loss": 0.6111, + "step": 1879 + }, + { + "epoch": 1.5565985643291, + "grad_norm": 0.8989667892456055, + "learning_rate": 4.272038563387079e-06, + "loss": 0.6255, + "step": 1880 + }, + { + "epoch": 1.5574268360022088, + "grad_norm": 0.9551357626914978, + "learning_rate": 4.2712626776789224e-06, + "loss": 0.6164, + "step": 1881 + }, + { + "epoch": 1.5582551076753175, + "grad_norm": 0.9278994798660278, + "learning_rate": 4.270486449247949e-06, + "loss": 0.6376, + "step": 1882 + }, + { + "epoch": 1.5590833793484262, + "grad_norm": 0.927180826663971, + "learning_rate": 4.269709878244352e-06, + "loss": 0.6602, + "step": 1883 + }, + { + "epoch": 1.559911651021535, + "grad_norm": 0.932837724685669, + "learning_rate": 4.268932964818389e-06, + "loss": 0.6229, + "step": 1884 + }, + { + "epoch": 1.5607399226946437, + "grad_norm": 0.895219087600708, + "learning_rate": 4.26815570912039e-06, + "loss": 0.6192, + "step": 1885 + }, + { + "epoch": 1.5615681943677526, + "grad_norm": 0.9075414538383484, + "learning_rate": 4.2673781113007415e-06, + "loss": 0.6418, + "step": 1886 + }, + { + "epoch": 1.5623964660408614, + "grad_norm": 0.9130772948265076, + "learning_rate": 4.266600171509905e-06, + "loss": 0.6402, + "step": 1887 + }, + { + "epoch": 1.5632247377139703, + "grad_norm": 0.8892185688018799, + "learning_rate": 4.265821889898402e-06, + "loss": 0.6213, + "step": 1888 + }, + { + "epoch": 1.564053009387079, + "grad_norm": 0.8853991031646729, + "learning_rate": 4.265043266616825e-06, + "loss": 0.6357, + "step": 1889 + }, + { + "epoch": 1.5648812810601878, + "grad_norm": 0.9273191690444946, + "learning_rate": 4.26426430181583e-06, + "loss": 0.6311, + "step": 1890 + }, + { + "epoch": 1.5657095527332965, + "grad_norm": 0.8890771865844727, + "learning_rate": 4.263484995646138e-06, + "loss": 0.6053, + "step": 1891 + }, + { + "epoch": 1.5665378244064052, + "grad_norm": 0.9022220373153687, + "learning_rate": 4.262705348258538e-06, + "loss": 0.6256, + "step": 1892 + }, + { + "epoch": 1.567366096079514, + "grad_norm": 0.9201255440711975, + "learning_rate": 4.261925359803886e-06, + "loss": 0.6302, + "step": 1893 + }, + { + "epoch": 1.5681943677526229, + "grad_norm": 0.9675379991531372, + "learning_rate": 4.261145030433101e-06, + "loss": 0.602, + "step": 1894 + }, + { + "epoch": 1.5690226394257316, + "grad_norm": 0.9039328098297119, + "learning_rate": 4.26036436029717e-06, + "loss": 0.6477, + "step": 1895 + }, + { + "epoch": 1.5698509110988406, + "grad_norm": 0.911361813545227, + "learning_rate": 4.259583349547144e-06, + "loss": 0.6211, + "step": 1896 + }, + { + "epoch": 1.5706791827719493, + "grad_norm": 0.8901238441467285, + "learning_rate": 4.258801998334143e-06, + "loss": 0.6442, + "step": 1897 + }, + { + "epoch": 1.571507454445058, + "grad_norm": 0.9290407299995422, + "learning_rate": 4.258020306809351e-06, + "loss": 0.6382, + "step": 1898 + }, + { + "epoch": 1.5723357261181667, + "grad_norm": 0.9179307222366333, + "learning_rate": 4.257238275124017e-06, + "loss": 0.6513, + "step": 1899 + }, + { + "epoch": 1.5731639977912755, + "grad_norm": 0.9212455153465271, + "learning_rate": 4.256455903429459e-06, + "loss": 0.6257, + "step": 1900 + }, + { + "epoch": 1.5739922694643842, + "grad_norm": 0.9126132130622864, + "learning_rate": 4.2556731918770565e-06, + "loss": 0.6384, + "step": 1901 + }, + { + "epoch": 1.5748205411374931, + "grad_norm": 0.8735355734825134, + "learning_rate": 4.254890140618257e-06, + "loss": 0.6031, + "step": 1902 + }, + { + "epoch": 1.5756488128106019, + "grad_norm": 0.9375724196434021, + "learning_rate": 4.254106749804575e-06, + "loss": 0.6333, + "step": 1903 + }, + { + "epoch": 1.5764770844837108, + "grad_norm": 0.9451910853385925, + "learning_rate": 4.253323019587588e-06, + "loss": 0.6186, + "step": 1904 + }, + { + "epoch": 1.5773053561568195, + "grad_norm": 0.9468234777450562, + "learning_rate": 4.25253895011894e-06, + "loss": 0.6023, + "step": 1905 + }, + { + "epoch": 1.5781336278299283, + "grad_norm": 0.8941681385040283, + "learning_rate": 4.251754541550343e-06, + "loss": 0.634, + "step": 1906 + }, + { + "epoch": 1.578961899503037, + "grad_norm": 0.9508136510848999, + "learning_rate": 4.250969794033573e-06, + "loss": 0.6346, + "step": 1907 + }, + { + "epoch": 1.5797901711761457, + "grad_norm": 0.8964453935623169, + "learning_rate": 4.2501847077204685e-06, + "loss": 0.6298, + "step": 1908 + }, + { + "epoch": 1.5806184428492545, + "grad_norm": 0.8885358572006226, + "learning_rate": 4.249399282762938e-06, + "loss": 0.6319, + "step": 1909 + }, + { + "epoch": 1.5814467145223632, + "grad_norm": 0.8929858207702637, + "learning_rate": 4.248613519312955e-06, + "loss": 0.6272, + "step": 1910 + }, + { + "epoch": 1.5822749861954721, + "grad_norm": 0.9664214849472046, + "learning_rate": 4.247827417522557e-06, + "loss": 0.6336, + "step": 1911 + }, + { + "epoch": 1.5831032578685809, + "grad_norm": 0.890846312046051, + "learning_rate": 4.247040977543845e-06, + "loss": 0.6242, + "step": 1912 + }, + { + "epoch": 1.5839315295416898, + "grad_norm": 0.8973182439804077, + "learning_rate": 4.246254199528991e-06, + "loss": 0.6336, + "step": 1913 + }, + { + "epoch": 1.5847598012147985, + "grad_norm": 0.9789074659347534, + "learning_rate": 4.245467083630227e-06, + "loss": 0.6191, + "step": 1914 + }, + { + "epoch": 1.5855880728879073, + "grad_norm": 0.9195513129234314, + "learning_rate": 4.244679629999853e-06, + "loss": 0.6333, + "step": 1915 + }, + { + "epoch": 1.586416344561016, + "grad_norm": 0.950316846370697, + "learning_rate": 4.243891838790235e-06, + "loss": 0.635, + "step": 1916 + }, + { + "epoch": 1.5872446162341247, + "grad_norm": 0.8971677422523499, + "learning_rate": 4.243103710153802e-06, + "loss": 0.6352, + "step": 1917 + }, + { + "epoch": 1.5880728879072334, + "grad_norm": 0.9198834896087646, + "learning_rate": 4.24231524424305e-06, + "loss": 0.6252, + "step": 1918 + }, + { + "epoch": 1.5889011595803424, + "grad_norm": 0.9535729885101318, + "learning_rate": 4.241526441210541e-06, + "loss": 0.6177, + "step": 1919 + }, + { + "epoch": 1.5897294312534511, + "grad_norm": 0.8910471200942993, + "learning_rate": 4.240737301208899e-06, + "loss": 0.6408, + "step": 1920 + }, + { + "epoch": 1.59055770292656, + "grad_norm": 0.862991452217102, + "learning_rate": 4.2399478243908164e-06, + "loss": 0.6221, + "step": 1921 + }, + { + "epoch": 1.5913859745996688, + "grad_norm": 0.8935141563415527, + "learning_rate": 4.239158010909049e-06, + "loss": 0.608, + "step": 1922 + }, + { + "epoch": 1.5922142462727775, + "grad_norm": 0.9177968502044678, + "learning_rate": 4.23836786091642e-06, + "loss": 0.6442, + "step": 1923 + }, + { + "epoch": 1.5930425179458862, + "grad_norm": 0.9157758951187134, + "learning_rate": 4.237577374565813e-06, + "loss": 0.6419, + "step": 1924 + }, + { + "epoch": 1.593870789618995, + "grad_norm": 0.9484630227088928, + "learning_rate": 4.236786552010182e-06, + "loss": 0.6733, + "step": 1925 + }, + { + "epoch": 1.5946990612921037, + "grad_norm": 0.9401251077651978, + "learning_rate": 4.235995393402544e-06, + "loss": 0.6325, + "step": 1926 + }, + { + "epoch": 1.5955273329652124, + "grad_norm": 0.9104607105255127, + "learning_rate": 4.23520389889598e-06, + "loss": 0.6313, + "step": 1927 + }, + { + "epoch": 1.5963556046383214, + "grad_norm": 0.9173511862754822, + "learning_rate": 4.234412068643637e-06, + "loss": 0.6044, + "step": 1928 + }, + { + "epoch": 1.59718387631143, + "grad_norm": 0.9189788699150085, + "learning_rate": 4.233619902798727e-06, + "loss": 0.6182, + "step": 1929 + }, + { + "epoch": 1.598012147984539, + "grad_norm": 0.910804808139801, + "learning_rate": 4.232827401514525e-06, + "loss": 0.6435, + "step": 1930 + }, + { + "epoch": 1.5988404196576478, + "grad_norm": 0.8986726403236389, + "learning_rate": 4.232034564944375e-06, + "loss": 0.6268, + "step": 1931 + }, + { + "epoch": 1.5996686913307565, + "grad_norm": 0.8782910108566284, + "learning_rate": 4.231241393241682e-06, + "loss": 0.6381, + "step": 1932 + }, + { + "epoch": 1.6004969630038652, + "grad_norm": 0.9176427721977234, + "learning_rate": 4.230447886559919e-06, + "loss": 0.6131, + "step": 1933 + }, + { + "epoch": 1.601325234676974, + "grad_norm": 0.8755510449409485, + "learning_rate": 4.22965404505262e-06, + "loss": 0.654, + "step": 1934 + }, + { + "epoch": 1.6021535063500827, + "grad_norm": 0.8644832968711853, + "learning_rate": 4.228859868873387e-06, + "loss": 0.627, + "step": 1935 + }, + { + "epoch": 1.6029817780231916, + "grad_norm": 0.9181885719299316, + "learning_rate": 4.228065358175886e-06, + "loss": 0.6062, + "step": 1936 + }, + { + "epoch": 1.6038100496963004, + "grad_norm": 0.9515942335128784, + "learning_rate": 4.227270513113846e-06, + "loss": 0.6262, + "step": 1937 + }, + { + "epoch": 1.6046383213694093, + "grad_norm": 0.9062270522117615, + "learning_rate": 4.226475333841063e-06, + "loss": 0.6139, + "step": 1938 + }, + { + "epoch": 1.605466593042518, + "grad_norm": 0.9040980339050293, + "learning_rate": 4.225679820511398e-06, + "loss": 0.64, + "step": 1939 + }, + { + "epoch": 1.6062948647156268, + "grad_norm": 0.8910935521125793, + "learning_rate": 4.224883973278775e-06, + "loss": 0.6239, + "step": 1940 + }, + { + "epoch": 1.6071231363887355, + "grad_norm": 0.9150440692901611, + "learning_rate": 4.224087792297181e-06, + "loss": 0.6283, + "step": 1941 + }, + { + "epoch": 1.6079514080618442, + "grad_norm": 0.9175113439559937, + "learning_rate": 4.223291277720671e-06, + "loss": 0.6498, + "step": 1942 + }, + { + "epoch": 1.608779679734953, + "grad_norm": 0.8720189332962036, + "learning_rate": 4.222494429703363e-06, + "loss": 0.623, + "step": 1943 + }, + { + "epoch": 1.609607951408062, + "grad_norm": 0.8868989944458008, + "learning_rate": 4.22169724839944e-06, + "loss": 0.6357, + "step": 1944 + }, + { + "epoch": 1.6104362230811706, + "grad_norm": 0.9084488153457642, + "learning_rate": 4.22089973396315e-06, + "loss": 0.6607, + "step": 1945 + }, + { + "epoch": 1.6112644947542794, + "grad_norm": 0.9061319231987, + "learning_rate": 4.220101886548802e-06, + "loss": 0.6309, + "step": 1946 + }, + { + "epoch": 1.6120927664273883, + "grad_norm": 0.895646870136261, + "learning_rate": 4.219303706310775e-06, + "loss": 0.6276, + "step": 1947 + }, + { + "epoch": 1.612921038100497, + "grad_norm": 0.8863298892974854, + "learning_rate": 4.218505193403506e-06, + "loss": 0.6444, + "step": 1948 + }, + { + "epoch": 1.6137493097736058, + "grad_norm": 0.9005767107009888, + "learning_rate": 4.2177063479815025e-06, + "loss": 0.6356, + "step": 1949 + }, + { + "epoch": 1.6145775814467145, + "grad_norm": 0.9094089865684509, + "learning_rate": 4.216907170199334e-06, + "loss": 0.6053, + "step": 1950 + }, + { + "epoch": 1.6154058531198232, + "grad_norm": 0.8869677782058716, + "learning_rate": 4.216107660211632e-06, + "loss": 0.6308, + "step": 1951 + }, + { + "epoch": 1.616234124792932, + "grad_norm": 0.9285000562667847, + "learning_rate": 4.215307818173095e-06, + "loss": 0.6188, + "step": 1952 + }, + { + "epoch": 1.6170623964660409, + "grad_norm": 0.9039749503135681, + "learning_rate": 4.214507644238485e-06, + "loss": 0.63, + "step": 1953 + }, + { + "epoch": 1.6178906681391496, + "grad_norm": 0.9202860593795776, + "learning_rate": 4.213707138562628e-06, + "loss": 0.632, + "step": 1954 + }, + { + "epoch": 1.6187189398122586, + "grad_norm": 0.8759778141975403, + "learning_rate": 4.212906301300415e-06, + "loss": 0.6253, + "step": 1955 + }, + { + "epoch": 1.6195472114853673, + "grad_norm": 0.8918818831443787, + "learning_rate": 4.212105132606801e-06, + "loss": 0.625, + "step": 1956 + }, + { + "epoch": 1.620375483158476, + "grad_norm": 0.9184549450874329, + "learning_rate": 4.211303632636804e-06, + "loss": 0.6293, + "step": 1957 + }, + { + "epoch": 1.6212037548315847, + "grad_norm": 0.8887206315994263, + "learning_rate": 4.210501801545506e-06, + "loss": 0.6118, + "step": 1958 + }, + { + "epoch": 1.6220320265046935, + "grad_norm": 0.9010420441627502, + "learning_rate": 4.209699639488054e-06, + "loss": 0.6209, + "step": 1959 + }, + { + "epoch": 1.6228602981778022, + "grad_norm": 0.8888309001922607, + "learning_rate": 4.20889714661966e-06, + "loss": 0.6301, + "step": 1960 + }, + { + "epoch": 1.6236885698509111, + "grad_norm": 0.9068322777748108, + "learning_rate": 4.208094323095597e-06, + "loss": 0.6234, + "step": 1961 + }, + { + "epoch": 1.6245168415240199, + "grad_norm": 0.8940340876579285, + "learning_rate": 4.207291169071207e-06, + "loss": 0.6076, + "step": 1962 + }, + { + "epoch": 1.6253451131971288, + "grad_norm": 0.907363772392273, + "learning_rate": 4.20648768470189e-06, + "loss": 0.6522, + "step": 1963 + }, + { + "epoch": 1.6261733848702375, + "grad_norm": 0.9104205965995789, + "learning_rate": 4.205683870143114e-06, + "loss": 0.6177, + "step": 1964 + }, + { + "epoch": 1.6270016565433463, + "grad_norm": 0.9051296710968018, + "learning_rate": 4.204879725550409e-06, + "loss": 0.6069, + "step": 1965 + }, + { + "epoch": 1.627829928216455, + "grad_norm": 0.9034865498542786, + "learning_rate": 4.20407525107937e-06, + "loss": 0.6099, + "step": 1966 + }, + { + "epoch": 1.6286581998895637, + "grad_norm": 0.9055590033531189, + "learning_rate": 4.203270446885655e-06, + "loss": 0.6226, + "step": 1967 + }, + { + "epoch": 1.6294864715626725, + "grad_norm": 0.897927463054657, + "learning_rate": 4.202465313124986e-06, + "loss": 0.6342, + "step": 1968 + }, + { + "epoch": 1.6303147432357812, + "grad_norm": 0.8995800018310547, + "learning_rate": 4.2016598499531495e-06, + "loss": 0.6408, + "step": 1969 + }, + { + "epoch": 1.6311430149088901, + "grad_norm": 0.8727264404296875, + "learning_rate": 4.200854057525995e-06, + "loss": 0.6263, + "step": 1970 + }, + { + "epoch": 1.6319712865819989, + "grad_norm": 0.9289891719818115, + "learning_rate": 4.200047935999435e-06, + "loss": 0.6351, + "step": 1971 + }, + { + "epoch": 1.6327995582551078, + "grad_norm": 0.8936645984649658, + "learning_rate": 4.199241485529446e-06, + "loss": 0.6205, + "step": 1972 + }, + { + "epoch": 1.6336278299282165, + "grad_norm": 0.9054183959960938, + "learning_rate": 4.198434706272072e-06, + "loss": 0.6101, + "step": 1973 + }, + { + "epoch": 1.6344561016013253, + "grad_norm": 0.9181169271469116, + "learning_rate": 4.1976275983834135e-06, + "loss": 0.6437, + "step": 1974 + }, + { + "epoch": 1.635284373274434, + "grad_norm": 0.9325777888298035, + "learning_rate": 4.1968201620196404e-06, + "loss": 0.6343, + "step": 1975 + }, + { + "epoch": 1.6361126449475427, + "grad_norm": 0.878094494342804, + "learning_rate": 4.196012397336983e-06, + "loss": 0.6298, + "step": 1976 + }, + { + "epoch": 1.6369409166206514, + "grad_norm": 0.9336429238319397, + "learning_rate": 4.195204304491737e-06, + "loss": 0.6232, + "step": 1977 + }, + { + "epoch": 1.6377691882937604, + "grad_norm": 0.9159499406814575, + "learning_rate": 4.19439588364026e-06, + "loss": 0.621, + "step": 1978 + }, + { + "epoch": 1.6385974599668691, + "grad_norm": 0.9075011610984802, + "learning_rate": 4.193587134938975e-06, + "loss": 0.6375, + "step": 1979 + }, + { + "epoch": 1.639425731639978, + "grad_norm": 0.8912638425827026, + "learning_rate": 4.192778058544367e-06, + "loss": 0.6494, + "step": 1980 + }, + { + "epoch": 1.6402540033130868, + "grad_norm": 0.9125227928161621, + "learning_rate": 4.191968654612984e-06, + "loss": 0.6238, + "step": 1981 + }, + { + "epoch": 1.6410822749861955, + "grad_norm": 0.906326174736023, + "learning_rate": 4.191158923301439e-06, + "loss": 0.5904, + "step": 1982 + }, + { + "epoch": 1.6419105466593042, + "grad_norm": 0.8858581781387329, + "learning_rate": 4.190348864766407e-06, + "loss": 0.6222, + "step": 1983 + }, + { + "epoch": 1.642738818332413, + "grad_norm": 0.864389955997467, + "learning_rate": 4.189538479164627e-06, + "loss": 0.6218, + "step": 1984 + }, + { + "epoch": 1.6435670900055217, + "grad_norm": 0.8983911871910095, + "learning_rate": 4.188727766652901e-06, + "loss": 0.6511, + "step": 1985 + }, + { + "epoch": 1.6443953616786304, + "grad_norm": 0.9633772373199463, + "learning_rate": 4.187916727388094e-06, + "loss": 0.623, + "step": 1986 + }, + { + "epoch": 1.6452236333517394, + "grad_norm": 0.9191408753395081, + "learning_rate": 4.187105361527135e-06, + "loss": 0.6245, + "step": 1987 + }, + { + "epoch": 1.646051905024848, + "grad_norm": 0.907792866230011, + "learning_rate": 4.186293669227016e-06, + "loss": 0.5994, + "step": 1988 + }, + { + "epoch": 1.646880176697957, + "grad_norm": 0.906218409538269, + "learning_rate": 4.18548165064479e-06, + "loss": 0.653, + "step": 1989 + }, + { + "epoch": 1.6477084483710658, + "grad_norm": 0.9408275485038757, + "learning_rate": 4.184669305937577e-06, + "loss": 0.5978, + "step": 1990 + }, + { + "epoch": 1.6485367200441745, + "grad_norm": 0.9278435111045837, + "learning_rate": 4.183856635262558e-06, + "loss": 0.6167, + "step": 1991 + }, + { + "epoch": 1.6493649917172832, + "grad_norm": 0.91252601146698, + "learning_rate": 4.183043638776978e-06, + "loss": 0.6229, + "step": 1992 + }, + { + "epoch": 1.650193263390392, + "grad_norm": 0.9178313612937927, + "learning_rate": 4.182230316638142e-06, + "loss": 0.6318, + "step": 1993 + }, + { + "epoch": 1.6510215350635007, + "grad_norm": 0.9056826233863831, + "learning_rate": 4.181416669003421e-06, + "loss": 0.5932, + "step": 1994 + }, + { + "epoch": 1.6518498067366096, + "grad_norm": 0.9139112830162048, + "learning_rate": 4.180602696030248e-06, + "loss": 0.6407, + "step": 1995 + }, + { + "epoch": 1.6526780784097184, + "grad_norm": 0.9195094108581543, + "learning_rate": 4.179788397876121e-06, + "loss": 0.6371, + "step": 1996 + }, + { + "epoch": 1.6535063500828273, + "grad_norm": 0.9221140146255493, + "learning_rate": 4.178973774698597e-06, + "loss": 0.611, + "step": 1997 + }, + { + "epoch": 1.654334621755936, + "grad_norm": 0.9180639386177063, + "learning_rate": 4.178158826655299e-06, + "loss": 0.6138, + "step": 1998 + }, + { + "epoch": 1.6551628934290448, + "grad_norm": 0.9255500435829163, + "learning_rate": 4.177343553903912e-06, + "loss": 0.6425, + "step": 1999 + }, + { + "epoch": 1.6559911651021535, + "grad_norm": 0.9378511905670166, + "learning_rate": 4.176527956602182e-06, + "loss": 0.6357, + "step": 2000 + }, + { + "epoch": 1.6568194367752622, + "grad_norm": 0.9212007522583008, + "learning_rate": 4.175712034907922e-06, + "loss": 0.6307, + "step": 2001 + }, + { + "epoch": 1.657647708448371, + "grad_norm": 0.9185839891433716, + "learning_rate": 4.174895788979004e-06, + "loss": 0.6173, + "step": 2002 + }, + { + "epoch": 1.65847598012148, + "grad_norm": 0.9658777117729187, + "learning_rate": 4.174079218973364e-06, + "loss": 0.622, + "step": 2003 + }, + { + "epoch": 1.6593042517945886, + "grad_norm": 0.9310017824172974, + "learning_rate": 4.173262325049e-06, + "loss": 0.6407, + "step": 2004 + }, + { + "epoch": 1.6601325234676974, + "grad_norm": 0.9048784375190735, + "learning_rate": 4.172445107363974e-06, + "loss": 0.6557, + "step": 2005 + }, + { + "epoch": 1.6609607951408063, + "grad_norm": 0.8820594549179077, + "learning_rate": 4.17162756607641e-06, + "loss": 0.6077, + "step": 2006 + }, + { + "epoch": 1.661789066813915, + "grad_norm": 0.8961437344551086, + "learning_rate": 4.170809701344496e-06, + "loss": 0.6151, + "step": 2007 + }, + { + "epoch": 1.6626173384870238, + "grad_norm": 0.9773615598678589, + "learning_rate": 4.169991513326479e-06, + "loss": 0.6339, + "step": 2008 + }, + { + "epoch": 1.6634456101601325, + "grad_norm": 0.9309290647506714, + "learning_rate": 4.169173002180672e-06, + "loss": 0.6235, + "step": 2009 + }, + { + "epoch": 1.6642738818332412, + "grad_norm": 1.0016266107559204, + "learning_rate": 4.168354168065448e-06, + "loss": 0.6355, + "step": 2010 + }, + { + "epoch": 1.66510215350635, + "grad_norm": 0.8820549249649048, + "learning_rate": 4.167535011139246e-06, + "loss": 0.6258, + "step": 2011 + }, + { + "epoch": 1.6659304251794589, + "grad_norm": 0.9030908942222595, + "learning_rate": 4.166715531560564e-06, + "loss": 0.6217, + "step": 2012 + }, + { + "epoch": 1.6667586968525676, + "grad_norm": 0.983755350112915, + "learning_rate": 4.165895729487962e-06, + "loss": 0.6281, + "step": 2013 + }, + { + "epoch": 1.6675869685256766, + "grad_norm": 0.948001503944397, + "learning_rate": 4.165075605080066e-06, + "loss": 0.6242, + "step": 2014 + }, + { + "epoch": 1.6684152401987853, + "grad_norm": 0.9468982219696045, + "learning_rate": 4.164255158495563e-06, + "loss": 0.6563, + "step": 2015 + }, + { + "epoch": 1.669243511871894, + "grad_norm": 0.9012095332145691, + "learning_rate": 4.1634343898932e-06, + "loss": 0.618, + "step": 2016 + }, + { + "epoch": 1.6700717835450027, + "grad_norm": 0.9554924964904785, + "learning_rate": 4.162613299431789e-06, + "loss": 0.6306, + "step": 2017 + }, + { + "epoch": 1.6709000552181115, + "grad_norm": 0.9587515592575073, + "learning_rate": 4.161791887270203e-06, + "loss": 0.6305, + "step": 2018 + }, + { + "epoch": 1.6717283268912202, + "grad_norm": 0.8792864084243774, + "learning_rate": 4.160970153567379e-06, + "loss": 0.6181, + "step": 2019 + }, + { + "epoch": 1.6725565985643291, + "grad_norm": 0.8958392143249512, + "learning_rate": 4.160148098482313e-06, + "loss": 0.623, + "step": 2020 + }, + { + "epoch": 1.6733848702374379, + "grad_norm": 0.9501842260360718, + "learning_rate": 4.159325722174065e-06, + "loss": 0.6454, + "step": 2021 + }, + { + "epoch": 1.6742131419105468, + "grad_norm": 0.9107666015625, + "learning_rate": 4.158503024801759e-06, + "loss": 0.6237, + "step": 2022 + }, + { + "epoch": 1.6750414135836555, + "grad_norm": 0.9405720233917236, + "learning_rate": 4.157680006524577e-06, + "loss": 0.6423, + "step": 2023 + }, + { + "epoch": 1.6758696852567643, + "grad_norm": 0.9008631706237793, + "learning_rate": 4.1568566675017675e-06, + "loss": 0.6385, + "step": 2024 + }, + { + "epoch": 1.676697956929873, + "grad_norm": 0.9056677222251892, + "learning_rate": 4.156033007892637e-06, + "loss": 0.6225, + "step": 2025 + }, + { + "epoch": 1.6775262286029817, + "grad_norm": 0.899982750415802, + "learning_rate": 4.155209027856557e-06, + "loss": 0.6268, + "step": 2026 + }, + { + "epoch": 1.6783545002760905, + "grad_norm": 0.928506076335907, + "learning_rate": 4.15438472755296e-06, + "loss": 0.6343, + "step": 2027 + }, + { + "epoch": 1.6791827719491992, + "grad_norm": 0.8831804394721985, + "learning_rate": 4.15356010714134e-06, + "loss": 0.6404, + "step": 2028 + }, + { + "epoch": 1.6800110436223081, + "grad_norm": 0.8637033700942993, + "learning_rate": 4.152735166781253e-06, + "loss": 0.6305, + "step": 2029 + }, + { + "epoch": 1.6808393152954169, + "grad_norm": 0.9164902567863464, + "learning_rate": 4.151909906632319e-06, + "loss": 0.6171, + "step": 2030 + }, + { + "epoch": 1.6816675869685258, + "grad_norm": 0.9124466776847839, + "learning_rate": 4.151084326854216e-06, + "loss": 0.6446, + "step": 2031 + }, + { + "epoch": 1.6824958586416345, + "grad_norm": 0.9212199449539185, + "learning_rate": 4.150258427606687e-06, + "loss": 0.6074, + "step": 2032 + }, + { + "epoch": 1.6833241303147433, + "grad_norm": 0.9536449313163757, + "learning_rate": 4.149432209049536e-06, + "loss": 0.6149, + "step": 2033 + }, + { + "epoch": 1.684152401987852, + "grad_norm": 0.9232591390609741, + "learning_rate": 4.148605671342629e-06, + "loss": 0.6238, + "step": 2034 + }, + { + "epoch": 1.6849806736609607, + "grad_norm": 0.8844353556632996, + "learning_rate": 4.147778814645892e-06, + "loss": 0.6434, + "step": 2035 + }, + { + "epoch": 1.6858089453340694, + "grad_norm": 0.8500795960426331, + "learning_rate": 4.1469516391193155e-06, + "loss": 0.6304, + "step": 2036 + }, + { + "epoch": 1.6866372170071784, + "grad_norm": 0.8676083087921143, + "learning_rate": 4.146124144922949e-06, + "loss": 0.6158, + "step": 2037 + }, + { + "epoch": 1.6874654886802871, + "grad_norm": 0.8914940357208252, + "learning_rate": 4.1452963322169046e-06, + "loss": 0.6336, + "step": 2038 + }, + { + "epoch": 1.688293760353396, + "grad_norm": 0.9245527386665344, + "learning_rate": 4.144468201161358e-06, + "loss": 0.6385, + "step": 2039 + }, + { + "epoch": 1.6891220320265048, + "grad_norm": 0.8946509957313538, + "learning_rate": 4.143639751916544e-06, + "loss": 0.6243, + "step": 2040 + }, + { + "epoch": 1.6899503036996135, + "grad_norm": 0.882095217704773, + "learning_rate": 4.14281098464276e-06, + "loss": 0.6171, + "step": 2041 + }, + { + "epoch": 1.6907785753727222, + "grad_norm": 0.8981113433837891, + "learning_rate": 4.141981899500364e-06, + "loss": 0.6389, + "step": 2042 + }, + { + "epoch": 1.691606847045831, + "grad_norm": 0.928701639175415, + "learning_rate": 4.1411524966497775e-06, + "loss": 0.6488, + "step": 2043 + }, + { + "epoch": 1.6924351187189397, + "grad_norm": 0.8937084674835205, + "learning_rate": 4.140322776251481e-06, + "loss": 0.6415, + "step": 2044 + }, + { + "epoch": 1.6932633903920484, + "grad_norm": 0.8940730094909668, + "learning_rate": 4.1394927384660185e-06, + "loss": 0.6419, + "step": 2045 + }, + { + "epoch": 1.6940916620651574, + "grad_norm": 0.9067767262458801, + "learning_rate": 4.138662383453994e-06, + "loss": 0.608, + "step": 2046 + }, + { + "epoch": 1.694919933738266, + "grad_norm": 0.9123063087463379, + "learning_rate": 4.137831711376074e-06, + "loss": 0.633, + "step": 2047 + }, + { + "epoch": 1.695748205411375, + "grad_norm": 0.89399653673172, + "learning_rate": 4.137000722392985e-06, + "loss": 0.6338, + "step": 2048 + }, + { + "epoch": 1.6965764770844838, + "grad_norm": 0.9146568775177002, + "learning_rate": 4.136169416665517e-06, + "loss": 0.6214, + "step": 2049 + }, + { + "epoch": 1.6974047487575925, + "grad_norm": 0.9402440786361694, + "learning_rate": 4.135337794354518e-06, + "loss": 0.6266, + "step": 2050 + }, + { + "epoch": 1.6982330204307012, + "grad_norm": 0.8790574669837952, + "learning_rate": 4.134505855620902e-06, + "loss": 0.5989, + "step": 2051 + }, + { + "epoch": 1.69906129210381, + "grad_norm": 0.9421536326408386, + "learning_rate": 4.133673600625637e-06, + "loss": 0.611, + "step": 2052 + }, + { + "epoch": 1.6998895637769187, + "grad_norm": 0.9365546703338623, + "learning_rate": 4.132841029529762e-06, + "loss": 0.6201, + "step": 2053 + }, + { + "epoch": 1.7007178354500276, + "grad_norm": 0.9213638305664062, + "learning_rate": 4.132008142494367e-06, + "loss": 0.632, + "step": 2054 + }, + { + "epoch": 1.7015461071231364, + "grad_norm": 0.9204041957855225, + "learning_rate": 4.13117493968061e-06, + "loss": 0.6086, + "step": 2055 + }, + { + "epoch": 1.7023743787962453, + "grad_norm": 0.8972902297973633, + "learning_rate": 4.130341421249708e-06, + "loss": 0.656, + "step": 2056 + }, + { + "epoch": 1.703202650469354, + "grad_norm": 0.9163963794708252, + "learning_rate": 4.129507587362939e-06, + "loss": 0.623, + "step": 2057 + }, + { + "epoch": 1.7040309221424628, + "grad_norm": 0.9178527593612671, + "learning_rate": 4.128673438181642e-06, + "loss": 0.6327, + "step": 2058 + }, + { + "epoch": 1.7048591938155715, + "grad_norm": 0.8982755541801453, + "learning_rate": 4.127838973867217e-06, + "loss": 0.6165, + "step": 2059 + }, + { + "epoch": 1.7056874654886802, + "grad_norm": 0.8995897769927979, + "learning_rate": 4.127004194581125e-06, + "loss": 0.6204, + "step": 2060 + }, + { + "epoch": 1.706515737161789, + "grad_norm": 0.9148748517036438, + "learning_rate": 4.126169100484888e-06, + "loss": 0.6177, + "step": 2061 + }, + { + "epoch": 1.707344008834898, + "grad_norm": 0.9249931573867798, + "learning_rate": 4.12533369174009e-06, + "loss": 0.6113, + "step": 2062 + }, + { + "epoch": 1.7081722805080066, + "grad_norm": 0.9162423610687256, + "learning_rate": 4.124497968508372e-06, + "loss": 0.6119, + "step": 2063 + }, + { + "epoch": 1.7090005521811154, + "grad_norm": 0.8919503688812256, + "learning_rate": 4.123661930951441e-06, + "loss": 0.6194, + "step": 2064 + }, + { + "epoch": 1.7098288238542243, + "grad_norm": 0.8884986639022827, + "learning_rate": 4.122825579231063e-06, + "loss": 0.656, + "step": 2065 + }, + { + "epoch": 1.710657095527333, + "grad_norm": 0.9072293043136597, + "learning_rate": 4.121988913509062e-06, + "loss": 0.6351, + "step": 2066 + }, + { + "epoch": 1.7114853672004418, + "grad_norm": 0.9255547523498535, + "learning_rate": 4.121151933947327e-06, + "loss": 0.601, + "step": 2067 + }, + { + "epoch": 1.7123136388735505, + "grad_norm": 0.9141303300857544, + "learning_rate": 4.120314640707804e-06, + "loss": 0.6257, + "step": 2068 + }, + { + "epoch": 1.7131419105466592, + "grad_norm": 0.8791571855545044, + "learning_rate": 4.119477033952504e-06, + "loss": 0.6355, + "step": 2069 + }, + { + "epoch": 1.713970182219768, + "grad_norm": 0.8910380005836487, + "learning_rate": 4.118639113843493e-06, + "loss": 0.6405, + "step": 2070 + }, + { + "epoch": 1.7147984538928769, + "grad_norm": 0.9583408236503601, + "learning_rate": 4.117800880542904e-06, + "loss": 0.6408, + "step": 2071 + }, + { + "epoch": 1.7156267255659856, + "grad_norm": 0.9684842824935913, + "learning_rate": 4.116962334212924e-06, + "loss": 0.6155, + "step": 2072 + }, + { + "epoch": 1.7164549972390946, + "grad_norm": 0.8882633447647095, + "learning_rate": 4.116123475015806e-06, + "loss": 0.6332, + "step": 2073 + }, + { + "epoch": 1.7172832689122033, + "grad_norm": 0.8599907755851746, + "learning_rate": 4.1152843031138604e-06, + "loss": 0.6416, + "step": 2074 + }, + { + "epoch": 1.718111540585312, + "grad_norm": 0.9216133952140808, + "learning_rate": 4.114444818669461e-06, + "loss": 0.61, + "step": 2075 + }, + { + "epoch": 1.7189398122584207, + "grad_norm": 0.943729817867279, + "learning_rate": 4.113605021845037e-06, + "loss": 0.643, + "step": 2076 + }, + { + "epoch": 1.7197680839315295, + "grad_norm": 0.9158332943916321, + "learning_rate": 4.112764912803084e-06, + "loss": 0.6353, + "step": 2077 + }, + { + "epoch": 1.7205963556046382, + "grad_norm": 0.9036465287208557, + "learning_rate": 4.111924491706154e-06, + "loss": 0.6143, + "step": 2078 + }, + { + "epoch": 1.7214246272777471, + "grad_norm": 0.9264538884162903, + "learning_rate": 4.111083758716862e-06, + "loss": 0.6302, + "step": 2079 + }, + { + "epoch": 1.7222528989508559, + "grad_norm": 0.9309650659561157, + "learning_rate": 4.11024271399788e-06, + "loss": 0.6394, + "step": 2080 + }, + { + "epoch": 1.7230811706239648, + "grad_norm": 0.9261552095413208, + "learning_rate": 4.109401357711944e-06, + "loss": 0.6338, + "step": 2081 + }, + { + "epoch": 1.7239094422970735, + "grad_norm": 0.9027671217918396, + "learning_rate": 4.1085596900218476e-06, + "loss": 0.633, + "step": 2082 + }, + { + "epoch": 1.7247377139701823, + "grad_norm": 0.8899479508399963, + "learning_rate": 4.107717711090446e-06, + "loss": 0.6372, + "step": 2083 + }, + { + "epoch": 1.725565985643291, + "grad_norm": 0.9513740539550781, + "learning_rate": 4.106875421080654e-06, + "loss": 0.6057, + "step": 2084 + }, + { + "epoch": 1.7263942573163997, + "grad_norm": 0.8926359415054321, + "learning_rate": 4.106032820155447e-06, + "loss": 0.6219, + "step": 2085 + }, + { + "epoch": 1.7272225289895085, + "grad_norm": 0.8961042165756226, + "learning_rate": 4.105189908477862e-06, + "loss": 0.5926, + "step": 2086 + }, + { + "epoch": 1.7280508006626172, + "grad_norm": 0.8870651125907898, + "learning_rate": 4.104346686210992e-06, + "loss": 0.6298, + "step": 2087 + }, + { + "epoch": 1.7288790723357261, + "grad_norm": 0.9134538173675537, + "learning_rate": 4.103503153517993e-06, + "loss": 0.6544, + "step": 2088 + }, + { + "epoch": 1.7297073440088349, + "grad_norm": 0.900499165058136, + "learning_rate": 4.102659310562082e-06, + "loss": 0.6182, + "step": 2089 + }, + { + "epoch": 1.7305356156819438, + "grad_norm": 0.9146167039871216, + "learning_rate": 4.101815157506534e-06, + "loss": 0.6297, + "step": 2090 + }, + { + "epoch": 1.7313638873550525, + "grad_norm": 0.909670352935791, + "learning_rate": 4.100970694514687e-06, + "loss": 0.6272, + "step": 2091 + }, + { + "epoch": 1.7321921590281613, + "grad_norm": 1.127607822418213, + "learning_rate": 4.100125921749932e-06, + "loss": 0.6416, + "step": 2092 + }, + { + "epoch": 1.73302043070127, + "grad_norm": 0.8950872421264648, + "learning_rate": 4.099280839375729e-06, + "loss": 0.6411, + "step": 2093 + }, + { + "epoch": 1.7338487023743787, + "grad_norm": 0.9612913727760315, + "learning_rate": 4.098435447555591e-06, + "loss": 0.626, + "step": 2094 + }, + { + "epoch": 1.7346769740474874, + "grad_norm": 0.927638053894043, + "learning_rate": 4.097589746453094e-06, + "loss": 0.6197, + "step": 2095 + }, + { + "epoch": 1.7355052457205964, + "grad_norm": 0.9428266286849976, + "learning_rate": 4.096743736231875e-06, + "loss": 0.6407, + "step": 2096 + }, + { + "epoch": 1.7363335173937051, + "grad_norm": 0.8619174361228943, + "learning_rate": 4.095897417055627e-06, + "loss": 0.6386, + "step": 2097 + }, + { + "epoch": 1.737161789066814, + "grad_norm": 0.8943668007850647, + "learning_rate": 4.095050789088106e-06, + "loss": 0.6206, + "step": 2098 + }, + { + "epoch": 1.7379900607399228, + "grad_norm": 0.9190096259117126, + "learning_rate": 4.0942038524931245e-06, + "loss": 0.6395, + "step": 2099 + }, + { + "epoch": 1.7388183324130315, + "grad_norm": 0.9055671095848083, + "learning_rate": 4.09335660743456e-06, + "loss": 0.6384, + "step": 2100 + }, + { + "epoch": 1.7396466040861402, + "grad_norm": 0.9204139709472656, + "learning_rate": 4.092509054076345e-06, + "loss": 0.6438, + "step": 2101 + }, + { + "epoch": 1.740474875759249, + "grad_norm": 0.9572716355323792, + "learning_rate": 4.091661192582473e-06, + "loss": 0.6233, + "step": 2102 + }, + { + "epoch": 1.7413031474323577, + "grad_norm": 0.9479691982269287, + "learning_rate": 4.0908130231169975e-06, + "loss": 0.6284, + "step": 2103 + }, + { + "epoch": 1.7421314191054664, + "grad_norm": 0.9081270694732666, + "learning_rate": 4.089964545844032e-06, + "loss": 0.6284, + "step": 2104 + }, + { + "epoch": 1.7429596907785754, + "grad_norm": 0.9227046966552734, + "learning_rate": 4.0891157609277475e-06, + "loss": 0.6611, + "step": 2105 + }, + { + "epoch": 1.743787962451684, + "grad_norm": 0.9033496975898743, + "learning_rate": 4.088266668532377e-06, + "loss": 0.6122, + "step": 2106 + }, + { + "epoch": 1.744616234124793, + "grad_norm": 0.933539628982544, + "learning_rate": 4.087417268822211e-06, + "loss": 0.6167, + "step": 2107 + }, + { + "epoch": 1.7454445057979018, + "grad_norm": 0.9211486577987671, + "learning_rate": 4.086567561961602e-06, + "loss": 0.6364, + "step": 2108 + }, + { + "epoch": 1.7462727774710105, + "grad_norm": 0.8782284259796143, + "learning_rate": 4.08571754811496e-06, + "loss": 0.6164, + "step": 2109 + }, + { + "epoch": 1.7471010491441192, + "grad_norm": 0.8936499357223511, + "learning_rate": 4.084867227446753e-06, + "loss": 0.6155, + "step": 2110 + }, + { + "epoch": 1.747929320817228, + "grad_norm": 0.8931904435157776, + "learning_rate": 4.084016600121513e-06, + "loss": 0.6512, + "step": 2111 + }, + { + "epoch": 1.7487575924903367, + "grad_norm": 0.9070253372192383, + "learning_rate": 4.083165666303826e-06, + "loss": 0.6197, + "step": 2112 + }, + { + "epoch": 1.7495858641634456, + "grad_norm": 0.9232584238052368, + "learning_rate": 4.08231442615834e-06, + "loss": 0.6332, + "step": 2113 + }, + { + "epoch": 1.7504141358365544, + "grad_norm": 0.92015540599823, + "learning_rate": 4.081462879849765e-06, + "loss": 0.6351, + "step": 2114 + }, + { + "epoch": 1.7512424075096633, + "grad_norm": 0.9353263974189758, + "learning_rate": 4.080611027542863e-06, + "loss": 0.6479, + "step": 2115 + }, + { + "epoch": 1.752070679182772, + "grad_norm": 0.8943057060241699, + "learning_rate": 4.079758869402462e-06, + "loss": 0.6179, + "step": 2116 + }, + { + "epoch": 1.7528989508558808, + "grad_norm": 0.9062668681144714, + "learning_rate": 4.078906405593446e-06, + "loss": 0.6166, + "step": 2117 + }, + { + "epoch": 1.7537272225289895, + "grad_norm": 0.9368826150894165, + "learning_rate": 4.078053636280759e-06, + "loss": 0.6206, + "step": 2118 + }, + { + "epoch": 1.7545554942020982, + "grad_norm": 0.9012559652328491, + "learning_rate": 4.0772005616294034e-06, + "loss": 0.6445, + "step": 2119 + }, + { + "epoch": 1.755383765875207, + "grad_norm": 0.9061569571495056, + "learning_rate": 4.076347181804442e-06, + "loss": 0.6312, + "step": 2120 + }, + { + "epoch": 1.756212037548316, + "grad_norm": 0.892900288105011, + "learning_rate": 4.075493496970996e-06, + "loss": 0.64, + "step": 2121 + }, + { + "epoch": 1.7570403092214246, + "grad_norm": 0.9200263023376465, + "learning_rate": 4.074639507294243e-06, + "loss": 0.6435, + "step": 2122 + }, + { + "epoch": 1.7578685808945336, + "grad_norm": 0.921215295791626, + "learning_rate": 4.0737852129394265e-06, + "loss": 0.6135, + "step": 2123 + }, + { + "epoch": 1.7586968525676423, + "grad_norm": 0.8980262279510498, + "learning_rate": 4.07293061407184e-06, + "loss": 0.635, + "step": 2124 + }, + { + "epoch": 1.759525124240751, + "grad_norm": 0.9192932844161987, + "learning_rate": 4.072075710856843e-06, + "loss": 0.6317, + "step": 2125 + }, + { + "epoch": 1.7603533959138598, + "grad_norm": 0.9387120008468628, + "learning_rate": 4.07122050345985e-06, + "loss": 0.6324, + "step": 2126 + }, + { + "epoch": 1.7611816675869685, + "grad_norm": 0.8787287473678589, + "learning_rate": 4.070364992046338e-06, + "loss": 0.6177, + "step": 2127 + }, + { + "epoch": 1.7620099392600772, + "grad_norm": 0.8963474631309509, + "learning_rate": 4.069509176781838e-06, + "loss": 0.6158, + "step": 2128 + }, + { + "epoch": 1.762838210933186, + "grad_norm": 0.8926964402198792, + "learning_rate": 4.068653057831942e-06, + "loss": 0.6528, + "step": 2129 + }, + { + "epoch": 1.7636664826062949, + "grad_norm": 0.9290505647659302, + "learning_rate": 4.067796635362303e-06, + "loss": 0.6274, + "step": 2130 + }, + { + "epoch": 1.7644947542794036, + "grad_norm": 0.8848921060562134, + "learning_rate": 4.06693990953863e-06, + "loss": 0.6507, + "step": 2131 + }, + { + "epoch": 1.7653230259525126, + "grad_norm": 0.8977338075637817, + "learning_rate": 4.0660828805266915e-06, + "loss": 0.6437, + "step": 2132 + }, + { + "epoch": 1.7661512976256213, + "grad_norm": 0.9125422239303589, + "learning_rate": 4.065225548492314e-06, + "loss": 0.6221, + "step": 2133 + }, + { + "epoch": 1.76697956929873, + "grad_norm": 0.9186370968818665, + "learning_rate": 4.064367913601384e-06, + "loss": 0.6174, + "step": 2134 + }, + { + "epoch": 1.7678078409718387, + "grad_norm": 0.9084877967834473, + "learning_rate": 4.063509976019845e-06, + "loss": 0.6228, + "step": 2135 + }, + { + "epoch": 1.7686361126449475, + "grad_norm": 0.9180804491043091, + "learning_rate": 4.0626517359137e-06, + "loss": 0.6126, + "step": 2136 + }, + { + "epoch": 1.7694643843180562, + "grad_norm": 0.9099741578102112, + "learning_rate": 4.061793193449013e-06, + "loss": 0.6027, + "step": 2137 + }, + { + "epoch": 1.7702926559911651, + "grad_norm": 0.9060433506965637, + "learning_rate": 4.0609343487919005e-06, + "loss": 0.632, + "step": 2138 + }, + { + "epoch": 1.7711209276642739, + "grad_norm": 0.9404996037483215, + "learning_rate": 4.060075202108542e-06, + "loss": 0.6405, + "step": 2139 + }, + { + "epoch": 1.7719491993373828, + "grad_norm": 0.9031956791877747, + "learning_rate": 4.059215753565176e-06, + "loss": 0.6137, + "step": 2140 + }, + { + "epoch": 1.7727774710104915, + "grad_norm": 0.8913756012916565, + "learning_rate": 4.058356003328095e-06, + "loss": 0.6387, + "step": 2141 + }, + { + "epoch": 1.7736057426836003, + "grad_norm": 0.8986163139343262, + "learning_rate": 4.057495951563655e-06, + "loss": 0.6151, + "step": 2142 + }, + { + "epoch": 1.774434014356709, + "grad_norm": 0.9017037153244019, + "learning_rate": 4.0566355984382676e-06, + "loss": 0.6466, + "step": 2143 + }, + { + "epoch": 1.7752622860298177, + "grad_norm": 0.9211136698722839, + "learning_rate": 4.0557749441184024e-06, + "loss": 0.638, + "step": 2144 + }, + { + "epoch": 1.7760905577029265, + "grad_norm": 0.9097506999969482, + "learning_rate": 4.054913988770588e-06, + "loss": 0.6529, + "step": 2145 + }, + { + "epoch": 1.7769188293760352, + "grad_norm": 0.8985908627510071, + "learning_rate": 4.054052732561411e-06, + "loss": 0.6388, + "step": 2146 + }, + { + "epoch": 1.7777471010491441, + "grad_norm": 0.9066522121429443, + "learning_rate": 4.053191175657518e-06, + "loss": 0.6429, + "step": 2147 + }, + { + "epoch": 1.7785753727222529, + "grad_norm": 0.9080630540847778, + "learning_rate": 4.052329318225611e-06, + "loss": 0.619, + "step": 2148 + }, + { + "epoch": 1.7794036443953618, + "grad_norm": 0.8701719641685486, + "learning_rate": 4.0514671604324505e-06, + "loss": 0.6245, + "step": 2149 + }, + { + "epoch": 1.7802319160684705, + "grad_norm": 0.9059606194496155, + "learning_rate": 4.0506047024448585e-06, + "loss": 0.6124, + "step": 2150 + }, + { + "epoch": 1.7810601877415793, + "grad_norm": 0.8912079334259033, + "learning_rate": 4.04974194442971e-06, + "loss": 0.6295, + "step": 2151 + }, + { + "epoch": 1.781888459414688, + "grad_norm": 0.9063137769699097, + "learning_rate": 4.048878886553942e-06, + "loss": 0.6299, + "step": 2152 + }, + { + "epoch": 1.7827167310877967, + "grad_norm": 0.8850398659706116, + "learning_rate": 4.048015528984548e-06, + "loss": 0.6251, + "step": 2153 + }, + { + "epoch": 1.7835450027609054, + "grad_norm": 0.9059123992919922, + "learning_rate": 4.04715187188858e-06, + "loss": 0.6278, + "step": 2154 + }, + { + "epoch": 1.7843732744340144, + "grad_norm": 0.8777562379837036, + "learning_rate": 4.046287915433147e-06, + "loss": 0.6003, + "step": 2155 + }, + { + "epoch": 1.7852015461071231, + "grad_norm": 0.9048421382904053, + "learning_rate": 4.045423659785417e-06, + "loss": 0.6143, + "step": 2156 + }, + { + "epoch": 1.786029817780232, + "grad_norm": 0.9388517737388611, + "learning_rate": 4.044559105112614e-06, + "loss": 0.6314, + "step": 2157 + }, + { + "epoch": 1.7868580894533408, + "grad_norm": 0.9405019283294678, + "learning_rate": 4.043694251582023e-06, + "loss": 0.6381, + "step": 2158 + }, + { + "epoch": 1.7876863611264495, + "grad_norm": 0.8942162990570068, + "learning_rate": 4.0428290993609845e-06, + "loss": 0.6318, + "step": 2159 + }, + { + "epoch": 1.7885146327995582, + "grad_norm": 0.9065560698509216, + "learning_rate": 4.041963648616897e-06, + "loss": 0.6258, + "step": 2160 + }, + { + "epoch": 1.789342904472667, + "grad_norm": 0.8766914010047913, + "learning_rate": 4.041097899517218e-06, + "loss": 0.6292, + "step": 2161 + }, + { + "epoch": 1.7901711761457757, + "grad_norm": 0.8942935466766357, + "learning_rate": 4.040231852229462e-06, + "loss": 0.662, + "step": 2162 + }, + { + "epoch": 1.7909994478188846, + "grad_norm": 0.8892539143562317, + "learning_rate": 4.039365506921198e-06, + "loss": 0.6251, + "step": 2163 + }, + { + "epoch": 1.7918277194919934, + "grad_norm": 0.9359427690505981, + "learning_rate": 4.038498863760061e-06, + "loss": 0.6222, + "step": 2164 + }, + { + "epoch": 1.792655991165102, + "grad_norm": 0.8899942636489868, + "learning_rate": 4.037631922913733e-06, + "loss": 0.6318, + "step": 2165 + }, + { + "epoch": 1.793484262838211, + "grad_norm": 0.9233553409576416, + "learning_rate": 4.0367646845499635e-06, + "loss": 0.6188, + "step": 2166 + }, + { + "epoch": 1.7943125345113198, + "grad_norm": 0.8946877121925354, + "learning_rate": 4.035897148836553e-06, + "loss": 0.6247, + "step": 2167 + }, + { + "epoch": 1.7951408061844285, + "grad_norm": 0.9200116991996765, + "learning_rate": 4.03502931594136e-06, + "loss": 0.5981, + "step": 2168 + }, + { + "epoch": 1.7959690778575372, + "grad_norm": 0.9193991422653198, + "learning_rate": 4.034161186032306e-06, + "loss": 0.6144, + "step": 2169 + }, + { + "epoch": 1.796797349530646, + "grad_norm": 0.9637252688407898, + "learning_rate": 4.0332927592773615e-06, + "loss": 0.6094, + "step": 2170 + }, + { + "epoch": 1.7976256212037547, + "grad_norm": 0.8958486318588257, + "learning_rate": 4.032424035844564e-06, + "loss": 0.6304, + "step": 2171 + }, + { + "epoch": 1.7984538928768636, + "grad_norm": 0.9411646127700806, + "learning_rate": 4.031555015901998e-06, + "loss": 0.5984, + "step": 2172 + }, + { + "epoch": 1.7992821645499724, + "grad_norm": 0.8594862222671509, + "learning_rate": 4.030685699617815e-06, + "loss": 0.6176, + "step": 2173 + }, + { + "epoch": 1.8001104362230813, + "grad_norm": 0.8969242572784424, + "learning_rate": 4.029816087160217e-06, + "loss": 0.6294, + "step": 2174 + }, + { + "epoch": 1.80093870789619, + "grad_norm": 0.8950173854827881, + "learning_rate": 4.028946178697467e-06, + "loss": 0.6441, + "step": 2175 + }, + { + "epoch": 1.8017669795692988, + "grad_norm": 0.9084720611572266, + "learning_rate": 4.0280759743978835e-06, + "loss": 0.6245, + "step": 2176 + }, + { + "epoch": 1.8025952512424075, + "grad_norm": 0.9174644947052002, + "learning_rate": 4.027205474429845e-06, + "loss": 0.6136, + "step": 2177 + }, + { + "epoch": 1.8034235229155162, + "grad_norm": 0.9103249907493591, + "learning_rate": 4.026334678961783e-06, + "loss": 0.6341, + "step": 2178 + }, + { + "epoch": 1.804251794588625, + "grad_norm": 0.8916360139846802, + "learning_rate": 4.025463588162189e-06, + "loss": 0.6424, + "step": 2179 + }, + { + "epoch": 1.805080066261734, + "grad_norm": 0.9347325563430786, + "learning_rate": 4.024592202199611e-06, + "loss": 0.6425, + "step": 2180 + }, + { + "epoch": 1.8059083379348426, + "grad_norm": 0.9135183691978455, + "learning_rate": 4.023720521242653e-06, + "loss": 0.6265, + "step": 2181 + }, + { + "epoch": 1.8067366096079516, + "grad_norm": 0.8996950387954712, + "learning_rate": 4.022848545459979e-06, + "loss": 0.6083, + "step": 2182 + }, + { + "epoch": 1.8075648812810603, + "grad_norm": 0.9104750752449036, + "learning_rate": 4.021976275020307e-06, + "loss": 0.6087, + "step": 2183 + }, + { + "epoch": 1.808393152954169, + "grad_norm": 0.9272193908691406, + "learning_rate": 4.0211037100924135e-06, + "loss": 0.6254, + "step": 2184 + }, + { + "epoch": 1.8092214246272778, + "grad_norm": 0.8605275750160217, + "learning_rate": 4.020230850845132e-06, + "loss": 0.6244, + "step": 2185 + }, + { + "epoch": 1.8100496963003865, + "grad_norm": 0.8817747831344604, + "learning_rate": 4.019357697447352e-06, + "loss": 0.6226, + "step": 2186 + }, + { + "epoch": 1.8108779679734952, + "grad_norm": 0.8978784084320068, + "learning_rate": 4.018484250068021e-06, + "loss": 0.615, + "step": 2187 + }, + { + "epoch": 1.811706239646604, + "grad_norm": 0.8996117115020752, + "learning_rate": 4.017610508876143e-06, + "loss": 0.6248, + "step": 2188 + }, + { + "epoch": 1.8125345113197129, + "grad_norm": 0.9133120179176331, + "learning_rate": 4.016736474040778e-06, + "loss": 0.6364, + "step": 2189 + }, + { + "epoch": 1.8133627829928216, + "grad_norm": 0.903982400894165, + "learning_rate": 4.015862145731044e-06, + "loss": 0.6078, + "step": 2190 + }, + { + "epoch": 1.8141910546659306, + "grad_norm": 0.9343582987785339, + "learning_rate": 4.014987524116115e-06, + "loss": 0.6254, + "step": 2191 + }, + { + "epoch": 1.8150193263390393, + "grad_norm": 0.883305549621582, + "learning_rate": 4.014112609365224e-06, + "loss": 0.6306, + "step": 2192 + }, + { + "epoch": 1.815847598012148, + "grad_norm": 0.9044662714004517, + "learning_rate": 4.013237401647656e-06, + "loss": 0.6288, + "step": 2193 + }, + { + "epoch": 1.8166758696852567, + "grad_norm": 0.9016374349594116, + "learning_rate": 4.012361901132758e-06, + "loss": 0.6254, + "step": 2194 + }, + { + "epoch": 1.8175041413583655, + "grad_norm": 0.9198315739631653, + "learning_rate": 4.01148610798993e-06, + "loss": 0.621, + "step": 2195 + }, + { + "epoch": 1.8183324130314742, + "grad_norm": 0.8884013891220093, + "learning_rate": 4.01061002238863e-06, + "loss": 0.6195, + "step": 2196 + }, + { + "epoch": 1.8191606847045831, + "grad_norm": 0.901606023311615, + "learning_rate": 4.009733644498372e-06, + "loss": 0.646, + "step": 2197 + }, + { + "epoch": 1.8199889563776919, + "grad_norm": 0.9003238677978516, + "learning_rate": 4.008856974488728e-06, + "loss": 0.6276, + "step": 2198 + }, + { + "epoch": 1.8208172280508008, + "grad_norm": 0.8835832476615906, + "learning_rate": 4.007980012529324e-06, + "loss": 0.6232, + "step": 2199 + }, + { + "epoch": 1.8216454997239095, + "grad_norm": 0.8751473426818848, + "learning_rate": 4.007102758789846e-06, + "loss": 0.635, + "step": 2200 + }, + { + "epoch": 1.8224737713970183, + "grad_norm": 0.8990851640701294, + "learning_rate": 4.006225213440032e-06, + "loss": 0.605, + "step": 2201 + }, + { + "epoch": 1.823302043070127, + "grad_norm": 0.8766496181488037, + "learning_rate": 4.0053473766496805e-06, + "loss": 0.6396, + "step": 2202 + }, + { + "epoch": 1.8241303147432357, + "grad_norm": 0.914273202419281, + "learning_rate": 4.004469248588644e-06, + "loss": 0.6435, + "step": 2203 + }, + { + "epoch": 1.8249585864163445, + "grad_norm": 0.9063234925270081, + "learning_rate": 4.003590829426832e-06, + "loss": 0.6144, + "step": 2204 + }, + { + "epoch": 1.8257868580894532, + "grad_norm": 0.9479252696037292, + "learning_rate": 4.002712119334212e-06, + "loss": 0.6584, + "step": 2205 + }, + { + "epoch": 1.8266151297625621, + "grad_norm": 0.9253218770027161, + "learning_rate": 4.001833118480804e-06, + "loss": 0.6203, + "step": 2206 + }, + { + "epoch": 1.8274434014356709, + "grad_norm": 0.930874764919281, + "learning_rate": 4.000953827036689e-06, + "loss": 0.6254, + "step": 2207 + }, + { + "epoch": 1.8282716731087798, + "grad_norm": 0.9202263951301575, + "learning_rate": 4.000074245171999e-06, + "loss": 0.6351, + "step": 2208 + }, + { + "epoch": 1.8290999447818885, + "grad_norm": 0.8993806838989258, + "learning_rate": 3.999194373056926e-06, + "loss": 0.6439, + "step": 2209 + }, + { + "epoch": 1.8299282164549973, + "grad_norm": 0.9138032793998718, + "learning_rate": 3.998314210861718e-06, + "loss": 0.6011, + "step": 2210 + }, + { + "epoch": 1.830756488128106, + "grad_norm": 0.8995500206947327, + "learning_rate": 3.9974337587566765e-06, + "loss": 0.6285, + "step": 2211 + }, + { + "epoch": 1.8315847598012147, + "grad_norm": 0.9092652797698975, + "learning_rate": 3.996553016912161e-06, + "loss": 0.6364, + "step": 2212 + }, + { + "epoch": 1.8324130314743234, + "grad_norm": 0.9085352420806885, + "learning_rate": 3.995671985498589e-06, + "loss": 0.6245, + "step": 2213 + }, + { + "epoch": 1.8332413031474324, + "grad_norm": 0.9546533226966858, + "learning_rate": 3.994790664686429e-06, + "loss": 0.6512, + "step": 2214 + }, + { + "epoch": 1.8340695748205411, + "grad_norm": 0.8606089353561401, + "learning_rate": 3.99390905464621e-06, + "loss": 0.6135, + "step": 2215 + }, + { + "epoch": 1.83489784649365, + "grad_norm": 0.9450663328170776, + "learning_rate": 3.993027155548515e-06, + "loss": 0.6193, + "step": 2216 + }, + { + "epoch": 1.8357261181667588, + "grad_norm": 0.9496508836746216, + "learning_rate": 3.992144967563984e-06, + "loss": 0.5978, + "step": 2217 + }, + { + "epoch": 1.8365543898398675, + "grad_norm": 0.8837412595748901, + "learning_rate": 3.99126249086331e-06, + "loss": 0.6104, + "step": 2218 + }, + { + "epoch": 1.8373826615129762, + "grad_norm": 0.8774031400680542, + "learning_rate": 3.9903797256172464e-06, + "loss": 0.6302, + "step": 2219 + }, + { + "epoch": 1.838210933186085, + "grad_norm": 0.8978205919265747, + "learning_rate": 3.989496671996599e-06, + "loss": 0.6411, + "step": 2220 + }, + { + "epoch": 1.8390392048591937, + "grad_norm": 0.9087855815887451, + "learning_rate": 3.98861333017223e-06, + "loss": 0.592, + "step": 2221 + }, + { + "epoch": 1.8398674765323026, + "grad_norm": 0.9040684700012207, + "learning_rate": 3.987729700315059e-06, + "loss": 0.6121, + "step": 2222 + }, + { + "epoch": 1.8406957482054114, + "grad_norm": 0.8980532288551331, + "learning_rate": 3.986845782596058e-06, + "loss": 0.6231, + "step": 2223 + }, + { + "epoch": 1.84152401987852, + "grad_norm": 0.9104854464530945, + "learning_rate": 3.9859615771862595e-06, + "loss": 0.6228, + "step": 2224 + }, + { + "epoch": 1.842352291551629, + "grad_norm": 0.9314625263214111, + "learning_rate": 3.985077084256747e-06, + "loss": 0.6136, + "step": 2225 + }, + { + "epoch": 1.8431805632247378, + "grad_norm": 0.9033634662628174, + "learning_rate": 3.984192303978663e-06, + "loss": 0.6408, + "step": 2226 + }, + { + "epoch": 1.8440088348978465, + "grad_norm": 0.9022733569145203, + "learning_rate": 3.983307236523203e-06, + "loss": 0.5954, + "step": 2227 + }, + { + "epoch": 1.8448371065709552, + "grad_norm": 0.8900256156921387, + "learning_rate": 3.982421882061621e-06, + "loss": 0.631, + "step": 2228 + }, + { + "epoch": 1.845665378244064, + "grad_norm": 0.8882872462272644, + "learning_rate": 3.981536240765223e-06, + "loss": 0.6204, + "step": 2229 + }, + { + "epoch": 1.8464936499171727, + "grad_norm": 0.894923985004425, + "learning_rate": 3.980650312805373e-06, + "loss": 0.6013, + "step": 2230 + }, + { + "epoch": 1.8473219215902816, + "grad_norm": 0.8910638093948364, + "learning_rate": 3.979764098353491e-06, + "loss": 0.6373, + "step": 2231 + }, + { + "epoch": 1.8481501932633904, + "grad_norm": 0.8952053189277649, + "learning_rate": 3.978877597581049e-06, + "loss": 0.6374, + "step": 2232 + }, + { + "epoch": 1.8489784649364993, + "grad_norm": 0.9100720882415771, + "learning_rate": 3.977990810659579e-06, + "loss": 0.6417, + "step": 2233 + }, + { + "epoch": 1.849806736609608, + "grad_norm": 0.8954150676727295, + "learning_rate": 3.9771037377606634e-06, + "loss": 0.6027, + "step": 2234 + }, + { + "epoch": 1.8506350082827168, + "grad_norm": 0.8989905118942261, + "learning_rate": 3.976216379055944e-06, + "loss": 0.6228, + "step": 2235 + }, + { + "epoch": 1.8514632799558255, + "grad_norm": 0.9060820937156677, + "learning_rate": 3.975328734717117e-06, + "loss": 0.6291, + "step": 2236 + }, + { + "epoch": 1.8522915516289342, + "grad_norm": 0.9164367318153381, + "learning_rate": 3.974440804915932e-06, + "loss": 0.6249, + "step": 2237 + }, + { + "epoch": 1.853119823302043, + "grad_norm": 0.9227840304374695, + "learning_rate": 3.973552589824196e-06, + "loss": 0.6328, + "step": 2238 + }, + { + "epoch": 1.853948094975152, + "grad_norm": 0.8612478971481323, + "learning_rate": 3.972664089613772e-06, + "loss": 0.5794, + "step": 2239 + }, + { + "epoch": 1.8547763666482606, + "grad_norm": 0.8819253444671631, + "learning_rate": 3.971775304456573e-06, + "loss": 0.6145, + "step": 2240 + }, + { + "epoch": 1.8556046383213696, + "grad_norm": 0.9216543436050415, + "learning_rate": 3.970886234524573e-06, + "loss": 0.609, + "step": 2241 + }, + { + "epoch": 1.8564329099944783, + "grad_norm": 0.9114699959754944, + "learning_rate": 3.969996879989797e-06, + "loss": 0.6105, + "step": 2242 + }, + { + "epoch": 1.857261181667587, + "grad_norm": 0.9464704990386963, + "learning_rate": 3.969107241024328e-06, + "loss": 0.6389, + "step": 2243 + }, + { + "epoch": 1.8580894533406958, + "grad_norm": 0.8921800255775452, + "learning_rate": 3.968217317800303e-06, + "loss": 0.6338, + "step": 2244 + }, + { + "epoch": 1.8589177250138045, + "grad_norm": 0.897452175617218, + "learning_rate": 3.967327110489913e-06, + "loss": 0.6248, + "step": 2245 + }, + { + "epoch": 1.8597459966869132, + "grad_norm": 0.9093567132949829, + "learning_rate": 3.9664366192654066e-06, + "loss": 0.6258, + "step": 2246 + }, + { + "epoch": 1.860574268360022, + "grad_norm": 0.894188642501831, + "learning_rate": 3.965545844299084e-06, + "loss": 0.6095, + "step": 2247 + }, + { + "epoch": 1.8614025400331309, + "grad_norm": 0.9238452911376953, + "learning_rate": 3.964654785763301e-06, + "loss": 0.6237, + "step": 2248 + }, + { + "epoch": 1.8622308117062396, + "grad_norm": 0.8715214133262634, + "learning_rate": 3.9637634438304715e-06, + "loss": 0.6153, + "step": 2249 + }, + { + "epoch": 1.8630590833793486, + "grad_norm": 0.8790848255157471, + "learning_rate": 3.962871818673061e-06, + "loss": 0.6389, + "step": 2250 + }, + { + "epoch": 1.8638873550524573, + "grad_norm": 0.9859224557876587, + "learning_rate": 3.961979910463589e-06, + "loss": 0.6075, + "step": 2251 + }, + { + "epoch": 1.864715626725566, + "grad_norm": 0.9127727150917053, + "learning_rate": 3.9610877193746335e-06, + "loss": 0.6332, + "step": 2252 + }, + { + "epoch": 1.8655438983986747, + "grad_norm": 0.9248382449150085, + "learning_rate": 3.960195245578825e-06, + "loss": 0.622, + "step": 2253 + }, + { + "epoch": 1.8663721700717835, + "grad_norm": 0.9030913710594177, + "learning_rate": 3.959302489248848e-06, + "loss": 0.6326, + "step": 2254 + }, + { + "epoch": 1.8672004417448922, + "grad_norm": 0.8922805190086365, + "learning_rate": 3.958409450557441e-06, + "loss": 0.6099, + "step": 2255 + }, + { + "epoch": 1.8680287134180011, + "grad_norm": 0.9165307879447937, + "learning_rate": 3.957516129677402e-06, + "loss": 0.6245, + "step": 2256 + }, + { + "epoch": 1.8688569850911099, + "grad_norm": 0.9445623159408569, + "learning_rate": 3.956622526781579e-06, + "loss": 0.6317, + "step": 2257 + }, + { + "epoch": 1.8696852567642188, + "grad_norm": 0.8833973407745361, + "learning_rate": 3.955728642042874e-06, + "loss": 0.6264, + "step": 2258 + }, + { + "epoch": 1.8705135284373275, + "grad_norm": 0.9512470364570618, + "learning_rate": 3.9548344756342486e-06, + "loss": 0.6137, + "step": 2259 + }, + { + "epoch": 1.8713418001104363, + "grad_norm": 0.8834348320960999, + "learning_rate": 3.9539400277287134e-06, + "loss": 0.6597, + "step": 2260 + }, + { + "epoch": 1.872170071783545, + "grad_norm": 0.8881180882453918, + "learning_rate": 3.953045298499335e-06, + "loss": 0.6502, + "step": 2261 + }, + { + "epoch": 1.8729983434566537, + "grad_norm": 0.8921699523925781, + "learning_rate": 3.952150288119239e-06, + "loss": 0.6216, + "step": 2262 + }, + { + "epoch": 1.8738266151297625, + "grad_norm": 0.91098552942276, + "learning_rate": 3.9512549967615965e-06, + "loss": 0.6286, + "step": 2263 + }, + { + "epoch": 1.8746548868028712, + "grad_norm": 0.9343264698982239, + "learning_rate": 3.950359424599642e-06, + "loss": 0.6368, + "step": 2264 + }, + { + "epoch": 1.8754831584759801, + "grad_norm": 0.9168746471405029, + "learning_rate": 3.949463571806658e-06, + "loss": 0.6355, + "step": 2265 + }, + { + "epoch": 1.8763114301490889, + "grad_norm": 0.9206068515777588, + "learning_rate": 3.948567438555986e-06, + "loss": 0.607, + "step": 2266 + }, + { + "epoch": 1.8771397018221978, + "grad_norm": 0.9262584447860718, + "learning_rate": 3.947671025021017e-06, + "loss": 0.6197, + "step": 2267 + }, + { + "epoch": 1.8779679734953065, + "grad_norm": 0.8699168562889099, + "learning_rate": 3.946774331375199e-06, + "loss": 0.6181, + "step": 2268 + }, + { + "epoch": 1.8787962451684153, + "grad_norm": 0.8892862200737, + "learning_rate": 3.9458773577920365e-06, + "loss": 0.6245, + "step": 2269 + }, + { + "epoch": 1.879624516841524, + "grad_norm": 0.9038919806480408, + "learning_rate": 3.944980104445082e-06, + "loss": 0.6209, + "step": 2270 + }, + { + "epoch": 1.8804527885146327, + "grad_norm": 0.8780872821807861, + "learning_rate": 3.9440825715079475e-06, + "loss": 0.6355, + "step": 2271 + }, + { + "epoch": 1.8812810601877414, + "grad_norm": 0.9379895925521851, + "learning_rate": 3.943184759154297e-06, + "loss": 0.6381, + "step": 2272 + }, + { + "epoch": 1.8821093318608504, + "grad_norm": 0.9112582802772522, + "learning_rate": 3.942286667557849e-06, + "loss": 0.6273, + "step": 2273 + }, + { + "epoch": 1.8829376035339591, + "grad_norm": 0.9367797374725342, + "learning_rate": 3.9413882968923754e-06, + "loss": 0.6111, + "step": 2274 + }, + { + "epoch": 1.883765875207068, + "grad_norm": 0.8923424482345581, + "learning_rate": 3.940489647331703e-06, + "loss": 0.6187, + "step": 2275 + }, + { + "epoch": 1.8845941468801768, + "grad_norm": 0.8797968029975891, + "learning_rate": 3.939590719049712e-06, + "loss": 0.603, + "step": 2276 + }, + { + "epoch": 1.8854224185532855, + "grad_norm": 0.8977354168891907, + "learning_rate": 3.938691512220336e-06, + "loss": 0.614, + "step": 2277 + }, + { + "epoch": 1.8862506902263942, + "grad_norm": 0.9207090139389038, + "learning_rate": 3.937792027017564e-06, + "loss": 0.6162, + "step": 2278 + }, + { + "epoch": 1.887078961899503, + "grad_norm": 0.8815562725067139, + "learning_rate": 3.936892263615438e-06, + "loss": 0.6315, + "step": 2279 + }, + { + "epoch": 1.8879072335726117, + "grad_norm": 0.9983104467391968, + "learning_rate": 3.935992222188054e-06, + "loss": 0.6337, + "step": 2280 + }, + { + "epoch": 1.8887355052457206, + "grad_norm": 0.8987579941749573, + "learning_rate": 3.93509190290956e-06, + "loss": 0.644, + "step": 2281 + }, + { + "epoch": 1.8895637769188294, + "grad_norm": 0.9132242798805237, + "learning_rate": 3.934191305954161e-06, + "loss": 0.6063, + "step": 2282 + }, + { + "epoch": 1.890392048591938, + "grad_norm": 0.8940858840942383, + "learning_rate": 3.933290431496114e-06, + "loss": 0.6204, + "step": 2283 + }, + { + "epoch": 1.891220320265047, + "grad_norm": 0.8862229585647583, + "learning_rate": 3.932389279709729e-06, + "loss": 0.6257, + "step": 2284 + }, + { + "epoch": 1.8920485919381558, + "grad_norm": 0.8841557502746582, + "learning_rate": 3.931487850769371e-06, + "loss": 0.6329, + "step": 2285 + }, + { + "epoch": 1.8928768636112645, + "grad_norm": 0.9075404405593872, + "learning_rate": 3.930586144849459e-06, + "loss": 0.6195, + "step": 2286 + }, + { + "epoch": 1.8937051352843732, + "grad_norm": 0.9349889755249023, + "learning_rate": 3.929684162124462e-06, + "loss": 0.6159, + "step": 2287 + }, + { + "epoch": 1.894533406957482, + "grad_norm": 0.9378470182418823, + "learning_rate": 3.928781902768908e-06, + "loss": 0.6221, + "step": 2288 + }, + { + "epoch": 1.8953616786305907, + "grad_norm": 0.8877297043800354, + "learning_rate": 3.927879366957374e-06, + "loss": 0.6359, + "step": 2289 + }, + { + "epoch": 1.8961899503036996, + "grad_norm": 0.9335882067680359, + "learning_rate": 3.926976554864495e-06, + "loss": 0.6397, + "step": 2290 + }, + { + "epoch": 1.8970182219768084, + "grad_norm": 0.8982004523277283, + "learning_rate": 3.926073466664953e-06, + "loss": 0.6303, + "step": 2291 + }, + { + "epoch": 1.8978464936499173, + "grad_norm": 0.8909676671028137, + "learning_rate": 3.92517010253349e-06, + "loss": 0.6476, + "step": 2292 + }, + { + "epoch": 1.898674765323026, + "grad_norm": 0.9127813577651978, + "learning_rate": 3.924266462644896e-06, + "loss": 0.6078, + "step": 2293 + }, + { + "epoch": 1.8995030369961348, + "grad_norm": 0.8923580050468445, + "learning_rate": 3.923362547174019e-06, + "loss": 0.6371, + "step": 2294 + }, + { + "epoch": 1.9003313086692435, + "grad_norm": 0.912458062171936, + "learning_rate": 3.922458356295757e-06, + "loss": 0.6162, + "step": 2295 + }, + { + "epoch": 1.9011595803423522, + "grad_norm": 0.8866751790046692, + "learning_rate": 3.921553890185064e-06, + "loss": 0.6011, + "step": 2296 + }, + { + "epoch": 1.901987852015461, + "grad_norm": 0.9114490151405334, + "learning_rate": 3.920649149016944e-06, + "loss": 0.6322, + "step": 2297 + }, + { + "epoch": 1.90281612368857, + "grad_norm": 0.912030816078186, + "learning_rate": 3.919744132966457e-06, + "loss": 0.6293, + "step": 2298 + }, + { + "epoch": 1.9036443953616786, + "grad_norm": 0.887301504611969, + "learning_rate": 3.918838842208715e-06, + "loss": 0.6288, + "step": 2299 + }, + { + "epoch": 1.9044726670347876, + "grad_norm": 0.9039818048477173, + "learning_rate": 3.917933276918884e-06, + "loss": 0.6154, + "step": 2300 + }, + { + "epoch": 1.9053009387078963, + "grad_norm": 0.9260784983634949, + "learning_rate": 3.917027437272182e-06, + "loss": 0.6111, + "step": 2301 + }, + { + "epoch": 1.906129210381005, + "grad_norm": 0.8966069221496582, + "learning_rate": 3.916121323443879e-06, + "loss": 0.6334, + "step": 2302 + }, + { + "epoch": 1.9069574820541138, + "grad_norm": 0.8983698487281799, + "learning_rate": 3.9152149356093e-06, + "loss": 0.6235, + "step": 2303 + }, + { + "epoch": 1.9077857537272225, + "grad_norm": 0.9150595664978027, + "learning_rate": 3.914308273943826e-06, + "loss": 0.6118, + "step": 2304 + }, + { + "epoch": 1.9086140254003312, + "grad_norm": 0.8768419027328491, + "learning_rate": 3.913401338622883e-06, + "loss": 0.5935, + "step": 2305 + }, + { + "epoch": 1.90944229707344, + "grad_norm": 0.887562096118927, + "learning_rate": 3.912494129821958e-06, + "loss": 0.6232, + "step": 2306 + }, + { + "epoch": 1.9102705687465489, + "grad_norm": 0.8916515111923218, + "learning_rate": 3.911586647716585e-06, + "loss": 0.6409, + "step": 2307 + }, + { + "epoch": 1.9110988404196576, + "grad_norm": 0.9207882881164551, + "learning_rate": 3.910678892482354e-06, + "loss": 0.6155, + "step": 2308 + }, + { + "epoch": 1.9119271120927666, + "grad_norm": 0.9300670027732849, + "learning_rate": 3.9097708642949085e-06, + "loss": 0.6389, + "step": 2309 + }, + { + "epoch": 1.9127553837658753, + "grad_norm": 0.8685980439186096, + "learning_rate": 3.908862563329941e-06, + "loss": 0.6392, + "step": 2310 + }, + { + "epoch": 1.913583655438984, + "grad_norm": 0.9150967001914978, + "learning_rate": 3.907953989763202e-06, + "loss": 0.6387, + "step": 2311 + }, + { + "epoch": 1.9144119271120927, + "grad_norm": 0.9094190001487732, + "learning_rate": 3.90704514377049e-06, + "loss": 0.6055, + "step": 2312 + }, + { + "epoch": 1.9152401987852015, + "grad_norm": 0.8863343596458435, + "learning_rate": 3.90613602552766e-06, + "loss": 0.6073, + "step": 2313 + }, + { + "epoch": 1.9160684704583102, + "grad_norm": 0.9878994822502136, + "learning_rate": 3.905226635210616e-06, + "loss": 0.613, + "step": 2314 + }, + { + "epoch": 1.9168967421314191, + "grad_norm": 0.8863592147827148, + "learning_rate": 3.904316972995318e-06, + "loss": 0.6102, + "step": 2315 + }, + { + "epoch": 1.9177250138045279, + "grad_norm": 0.8859980702400208, + "learning_rate": 3.9034070390577765e-06, + "loss": 0.6182, + "step": 2316 + }, + { + "epoch": 1.9185532854776368, + "grad_norm": 0.8953589200973511, + "learning_rate": 3.9024968335740556e-06, + "loss": 0.6212, + "step": 2317 + }, + { + "epoch": 1.9193815571507455, + "grad_norm": 0.873466968536377, + "learning_rate": 3.90158635672027e-06, + "loss": 0.616, + "step": 2318 + }, + { + "epoch": 1.9202098288238543, + "grad_norm": 0.8589543700218201, + "learning_rate": 3.900675608672592e-06, + "loss": 0.6141, + "step": 2319 + }, + { + "epoch": 1.921038100496963, + "grad_norm": 0.8717133402824402, + "learning_rate": 3.899764589607239e-06, + "loss": 0.6281, + "step": 2320 + }, + { + "epoch": 1.9218663721700717, + "grad_norm": 0.893227219581604, + "learning_rate": 3.898853299700486e-06, + "loss": 0.6139, + "step": 2321 + }, + { + "epoch": 1.9226946438431805, + "grad_norm": 0.8736181259155273, + "learning_rate": 3.89794173912866e-06, + "loss": 0.5987, + "step": 2322 + }, + { + "epoch": 1.9235229155162892, + "grad_norm": 0.8975626826286316, + "learning_rate": 3.897029908068139e-06, + "loss": 0.6417, + "step": 2323 + }, + { + "epoch": 1.9243511871893981, + "grad_norm": 0.8916004300117493, + "learning_rate": 3.896117806695355e-06, + "loss": 0.6259, + "step": 2324 + }, + { + "epoch": 1.9251794588625069, + "grad_norm": 0.8739155530929565, + "learning_rate": 3.8952054351867885e-06, + "loss": 0.6252, + "step": 2325 + }, + { + "epoch": 1.9260077305356158, + "grad_norm": 0.8667280673980713, + "learning_rate": 3.894292793718976e-06, + "loss": 0.6083, + "step": 2326 + }, + { + "epoch": 1.9268360022087245, + "grad_norm": 0.8749732375144958, + "learning_rate": 3.893379882468506e-06, + "loss": 0.6294, + "step": 2327 + }, + { + "epoch": 1.9276642738818333, + "grad_norm": 0.9085330963134766, + "learning_rate": 3.8924667016120175e-06, + "loss": 0.6101, + "step": 2328 + }, + { + "epoch": 1.928492545554942, + "grad_norm": 0.9394292235374451, + "learning_rate": 3.891553251326203e-06, + "loss": 0.6182, + "step": 2329 + }, + { + "epoch": 1.9293208172280507, + "grad_norm": 0.8756951093673706, + "learning_rate": 3.890639531787805e-06, + "loss": 0.6159, + "step": 2330 + }, + { + "epoch": 1.9301490889011594, + "grad_norm": 0.8765501379966736, + "learning_rate": 3.8897255431736225e-06, + "loss": 0.6182, + "step": 2331 + }, + { + "epoch": 1.9309773605742684, + "grad_norm": 0.907835841178894, + "learning_rate": 3.888811285660501e-06, + "loss": 0.6336, + "step": 2332 + }, + { + "epoch": 1.9318056322473771, + "grad_norm": 1.0617926120758057, + "learning_rate": 3.887896759425342e-06, + "loss": 0.6336, + "step": 2333 + }, + { + "epoch": 1.932633903920486, + "grad_norm": 0.8745515942573547, + "learning_rate": 3.886981964645099e-06, + "loss": 0.6244, + "step": 2334 + }, + { + "epoch": 1.9334621755935948, + "grad_norm": 0.8918715119361877, + "learning_rate": 3.886066901496775e-06, + "loss": 0.6289, + "step": 2335 + }, + { + "epoch": 1.9342904472667035, + "grad_norm": 0.8779492378234863, + "learning_rate": 3.885151570157426e-06, + "loss": 0.6335, + "step": 2336 + }, + { + "epoch": 1.9351187189398122, + "grad_norm": 0.8955790400505066, + "learning_rate": 3.88423597080416e-06, + "loss": 0.6125, + "step": 2337 + }, + { + "epoch": 1.935946990612921, + "grad_norm": 0.8905174136161804, + "learning_rate": 3.883320103614137e-06, + "loss": 0.5913, + "step": 2338 + }, + { + "epoch": 1.9367752622860297, + "grad_norm": 0.9078717231750488, + "learning_rate": 3.88240396876457e-06, + "loss": 0.6356, + "step": 2339 + }, + { + "epoch": 1.9376035339591386, + "grad_norm": 0.9211397171020508, + "learning_rate": 3.881487566432722e-06, + "loss": 0.6305, + "step": 2340 + }, + { + "epoch": 1.9384318056322474, + "grad_norm": 0.9131335020065308, + "learning_rate": 3.880570896795908e-06, + "loss": 0.6184, + "step": 2341 + }, + { + "epoch": 1.9392600773053563, + "grad_norm": 0.8874300718307495, + "learning_rate": 3.879653960031494e-06, + "loss": 0.6191, + "step": 2342 + }, + { + "epoch": 1.940088348978465, + "grad_norm": 0.9244474768638611, + "learning_rate": 3.8787367563169e-06, + "loss": 0.5994, + "step": 2343 + }, + { + "epoch": 1.9409166206515738, + "grad_norm": 0.8986138701438904, + "learning_rate": 3.877819285829596e-06, + "loss": 0.6544, + "step": 2344 + }, + { + "epoch": 1.9417448923246825, + "grad_norm": 0.9084672927856445, + "learning_rate": 3.876901548747104e-06, + "loss": 0.6248, + "step": 2345 + }, + { + "epoch": 1.9425731639977912, + "grad_norm": 0.9052600264549255, + "learning_rate": 3.875983545246997e-06, + "loss": 0.6112, + "step": 2346 + }, + { + "epoch": 1.9434014356709, + "grad_norm": 0.9029438495635986, + "learning_rate": 3.8750652755069025e-06, + "loss": 0.6157, + "step": 2347 + }, + { + "epoch": 1.9442297073440087, + "grad_norm": 0.882449746131897, + "learning_rate": 3.874146739704494e-06, + "loss": 0.5886, + "step": 2348 + }, + { + "epoch": 1.9450579790171176, + "grad_norm": 0.8628272414207458, + "learning_rate": 3.873227938017502e-06, + "loss": 0.608, + "step": 2349 + }, + { + "epoch": 1.9458862506902264, + "grad_norm": 0.8977845907211304, + "learning_rate": 3.872308870623704e-06, + "loss": 0.6475, + "step": 2350 + }, + { + "epoch": 1.9467145223633353, + "grad_norm": 0.8801501393318176, + "learning_rate": 3.871389537700933e-06, + "loss": 0.6246, + "step": 2351 + }, + { + "epoch": 1.947542794036444, + "grad_norm": 0.9245358109474182, + "learning_rate": 3.87046993942707e-06, + "loss": 0.6317, + "step": 2352 + }, + { + "epoch": 1.9483710657095528, + "grad_norm": 0.8959249258041382, + "learning_rate": 3.869550075980049e-06, + "loss": 0.6157, + "step": 2353 + }, + { + "epoch": 1.9491993373826615, + "grad_norm": 0.861315906047821, + "learning_rate": 3.868629947537855e-06, + "loss": 0.6251, + "step": 2354 + }, + { + "epoch": 1.9500276090557702, + "grad_norm": 0.8839966654777527, + "learning_rate": 3.867709554278524e-06, + "loss": 0.6289, + "step": 2355 + }, + { + "epoch": 1.950855880728879, + "grad_norm": 0.8878759741783142, + "learning_rate": 3.8667888963801445e-06, + "loss": 0.6113, + "step": 2356 + }, + { + "epoch": 1.951684152401988, + "grad_norm": 0.8725037574768066, + "learning_rate": 3.865867974020854e-06, + "loss": 0.6324, + "step": 2357 + }, + { + "epoch": 1.9525124240750966, + "grad_norm": 0.9059397578239441, + "learning_rate": 3.8649467873788436e-06, + "loss": 0.5981, + "step": 2358 + }, + { + "epoch": 1.9533406957482056, + "grad_norm": 0.8883296847343445, + "learning_rate": 3.8640253366323535e-06, + "loss": 0.598, + "step": 2359 + }, + { + "epoch": 1.9541689674213143, + "grad_norm": 0.8987225890159607, + "learning_rate": 3.863103621959676e-06, + "loss": 0.6253, + "step": 2360 + }, + { + "epoch": 1.954997239094423, + "grad_norm": 0.8977320194244385, + "learning_rate": 3.862181643539155e-06, + "loss": 0.6424, + "step": 2361 + }, + { + "epoch": 1.9558255107675317, + "grad_norm": 0.9267589449882507, + "learning_rate": 3.861259401549183e-06, + "loss": 0.647, + "step": 2362 + }, + { + "epoch": 1.9566537824406405, + "grad_norm": 0.9297959804534912, + "learning_rate": 3.860336896168207e-06, + "loss": 0.6448, + "step": 2363 + }, + { + "epoch": 1.9574820541137492, + "grad_norm": 0.9627453088760376, + "learning_rate": 3.859414127574722e-06, + "loss": 0.6154, + "step": 2364 + }, + { + "epoch": 1.958310325786858, + "grad_norm": 0.8945866227149963, + "learning_rate": 3.858491095947275e-06, + "loss": 0.6201, + "step": 2365 + }, + { + "epoch": 1.9591385974599669, + "grad_norm": 0.8859817981719971, + "learning_rate": 3.857567801464466e-06, + "loss": 0.6171, + "step": 2366 + }, + { + "epoch": 1.9599668691330756, + "grad_norm": 0.9104642271995544, + "learning_rate": 3.8566442443049425e-06, + "loss": 0.6273, + "step": 2367 + }, + { + "epoch": 1.9607951408061846, + "grad_norm": 0.8821743726730347, + "learning_rate": 3.855720424647403e-06, + "loss": 0.5993, + "step": 2368 + }, + { + "epoch": 1.9616234124792933, + "grad_norm": 0.903322160243988, + "learning_rate": 3.854796342670601e-06, + "loss": 0.6471, + "step": 2369 + }, + { + "epoch": 1.962451684152402, + "grad_norm": 0.8744645714759827, + "learning_rate": 3.853871998553334e-06, + "loss": 0.6444, + "step": 2370 + }, + { + "epoch": 1.9632799558255107, + "grad_norm": 0.9538072347640991, + "learning_rate": 3.852947392474457e-06, + "loss": 0.656, + "step": 2371 + }, + { + "epoch": 1.9641082274986195, + "grad_norm": 0.875407874584198, + "learning_rate": 3.852022524612872e-06, + "loss": 0.6479, + "step": 2372 + }, + { + "epoch": 1.9649364991717282, + "grad_norm": 0.8908282518386841, + "learning_rate": 3.8510973951475305e-06, + "loss": 0.6262, + "step": 2373 + }, + { + "epoch": 1.9657647708448371, + "grad_norm": 0.8698679208755493, + "learning_rate": 3.850172004257438e-06, + "loss": 0.6431, + "step": 2374 + }, + { + "epoch": 1.9665930425179459, + "grad_norm": 0.8997004628181458, + "learning_rate": 3.84924635212165e-06, + "loss": 0.6349, + "step": 2375 + }, + { + "epoch": 1.9674213141910548, + "grad_norm": 0.8815768957138062, + "learning_rate": 3.8483204389192695e-06, + "loss": 0.6449, + "step": 2376 + }, + { + "epoch": 1.9682495858641635, + "grad_norm": 0.9148927927017212, + "learning_rate": 3.847394264829453e-06, + "loss": 0.6106, + "step": 2377 + }, + { + "epoch": 1.9690778575372723, + "grad_norm": 0.8647779226303101, + "learning_rate": 3.846467830031407e-06, + "loss": 0.5954, + "step": 2378 + }, + { + "epoch": 1.969906129210381, + "grad_norm": 0.8877570033073425, + "learning_rate": 3.845541134704387e-06, + "loss": 0.6006, + "step": 2379 + }, + { + "epoch": 1.9707344008834897, + "grad_norm": 0.899742603302002, + "learning_rate": 3.844614179027701e-06, + "loss": 0.6069, + "step": 2380 + }, + { + "epoch": 1.9715626725565985, + "grad_norm": 0.9085001349449158, + "learning_rate": 3.843686963180705e-06, + "loss": 0.6332, + "step": 2381 + }, + { + "epoch": 1.9723909442297074, + "grad_norm": 0.8954899311065674, + "learning_rate": 3.842759487342808e-06, + "loss": 0.6186, + "step": 2382 + }, + { + "epoch": 1.9732192159028161, + "grad_norm": 0.9011021256446838, + "learning_rate": 3.841831751693466e-06, + "loss": 0.6177, + "step": 2383 + }, + { + "epoch": 1.9740474875759249, + "grad_norm": 0.8824598789215088, + "learning_rate": 3.8409037564121895e-06, + "loss": 0.6392, + "step": 2384 + }, + { + "epoch": 1.9748757592490338, + "grad_norm": 0.8936701416969299, + "learning_rate": 3.839975501678536e-06, + "loss": 0.6508, + "step": 2385 + }, + { + "epoch": 1.9757040309221425, + "grad_norm": 0.9009436368942261, + "learning_rate": 3.839046987672113e-06, + "loss": 0.6169, + "step": 2386 + }, + { + "epoch": 1.9765323025952513, + "grad_norm": 0.8839378356933594, + "learning_rate": 3.8381182145725815e-06, + "loss": 0.6107, + "step": 2387 + }, + { + "epoch": 1.97736057426836, + "grad_norm": 0.9240637421607971, + "learning_rate": 3.837189182559649e-06, + "loss": 0.6083, + "step": 2388 + }, + { + "epoch": 1.9781888459414687, + "grad_norm": 0.881908118724823, + "learning_rate": 3.836259891813075e-06, + "loss": 0.5995, + "step": 2389 + }, + { + "epoch": 1.9790171176145774, + "grad_norm": 0.8754116296768188, + "learning_rate": 3.835330342512667e-06, + "loss": 0.6111, + "step": 2390 + }, + { + "epoch": 1.9798453892876864, + "grad_norm": 0.9713384509086609, + "learning_rate": 3.834400534838287e-06, + "loss": 0.641, + "step": 2391 + }, + { + "epoch": 1.9806736609607951, + "grad_norm": 0.9263403415679932, + "learning_rate": 3.833470468969842e-06, + "loss": 0.6254, + "step": 2392 + }, + { + "epoch": 1.981501932633904, + "grad_norm": 0.9033388495445251, + "learning_rate": 3.832540145087291e-06, + "loss": 0.6049, + "step": 2393 + }, + { + "epoch": 1.9823302043070128, + "grad_norm": 0.8710628151893616, + "learning_rate": 3.8316095633706455e-06, + "loss": 0.5999, + "step": 2394 + }, + { + "epoch": 1.9831584759801215, + "grad_norm": 0.9053885340690613, + "learning_rate": 3.830678723999961e-06, + "loss": 0.6393, + "step": 2395 + }, + { + "epoch": 1.9839867476532302, + "grad_norm": 0.9075314998626709, + "learning_rate": 3.829747627155348e-06, + "loss": 0.6188, + "step": 2396 + }, + { + "epoch": 1.984815019326339, + "grad_norm": 0.9151870012283325, + "learning_rate": 3.828816273016966e-06, + "loss": 0.625, + "step": 2397 + }, + { + "epoch": 1.9856432909994477, + "grad_norm": 0.903843104839325, + "learning_rate": 3.827884661765022e-06, + "loss": 0.618, + "step": 2398 + }, + { + "epoch": 1.9864715626725566, + "grad_norm": 0.875468373298645, + "learning_rate": 3.826952793579774e-06, + "loss": 0.6121, + "step": 2399 + }, + { + "epoch": 1.9872998343456654, + "grad_norm": 0.8908163905143738, + "learning_rate": 3.8260206686415295e-06, + "loss": 0.6292, + "step": 2400 + }, + { + "epoch": 1.9881281060187743, + "grad_norm": 0.8985518217086792, + "learning_rate": 3.825088287130647e-06, + "loss": 0.6356, + "step": 2401 + }, + { + "epoch": 1.988956377691883, + "grad_norm": 0.9277573227882385, + "learning_rate": 3.8241556492275335e-06, + "loss": 0.6424, + "step": 2402 + }, + { + "epoch": 1.9897846493649918, + "grad_norm": 0.9075362682342529, + "learning_rate": 3.823222755112646e-06, + "loss": 0.623, + "step": 2403 + }, + { + "epoch": 1.9906129210381005, + "grad_norm": 0.9038227796554565, + "learning_rate": 3.822289604966489e-06, + "loss": 0.6195, + "step": 2404 + }, + { + "epoch": 1.9914411927112092, + "grad_norm": 0.8869004845619202, + "learning_rate": 3.82135619896962e-06, + "loss": 0.6034, + "step": 2405 + }, + { + "epoch": 1.992269464384318, + "grad_norm": 0.9017077684402466, + "learning_rate": 3.820422537302645e-06, + "loss": 0.6469, + "step": 2406 + }, + { + "epoch": 1.9930977360574267, + "grad_norm": 0.8785225749015808, + "learning_rate": 3.819488620146218e-06, + "loss": 0.6327, + "step": 2407 + }, + { + "epoch": 1.9939260077305356, + "grad_norm": 1.0367780923843384, + "learning_rate": 3.818554447681041e-06, + "loss": 0.6259, + "step": 2408 + }, + { + "epoch": 1.9947542794036444, + "grad_norm": 0.9392277002334595, + "learning_rate": 3.817620020087872e-06, + "loss": 0.6393, + "step": 2409 + }, + { + "epoch": 1.9955825510767533, + "grad_norm": 0.8935515880584717, + "learning_rate": 3.816685337547511e-06, + "loss": 0.6153, + "step": 2410 + }, + { + "epoch": 1.996410822749862, + "grad_norm": 0.8895331025123596, + "learning_rate": 3.815750400240811e-06, + "loss": 0.5978, + "step": 2411 + }, + { + "epoch": 1.9972390944229708, + "grad_norm": 0.8733546733856201, + "learning_rate": 3.814815208348673e-06, + "loss": 0.6254, + "step": 2412 + }, + { + "epoch": 1.9980673660960795, + "grad_norm": 0.9177356958389282, + "learning_rate": 3.8138797620520507e-06, + "loss": 0.631, + "step": 2413 + }, + { + "epoch": 1.9988956377691882, + "grad_norm": 0.9259604811668396, + "learning_rate": 3.8129440615319414e-06, + "loss": 0.6138, + "step": 2414 + }, + { + "epoch": 1.999723909442297, + "grad_norm": 0.9226608872413635, + "learning_rate": 3.812008106969395e-06, + "loss": 0.607, + "step": 2415 + }, + { + "epoch": 2.0, + "grad_norm": 0.9226608872413635, + "learning_rate": 3.8110718985455097e-06, + "loss": 0.5993, + "step": 2416 + }, + { + "epoch": 2.0008282716731087, + "grad_norm": 1.62429940700531, + "learning_rate": 3.810135436441435e-06, + "loss": 0.5954, + "step": 2417 + }, + { + "epoch": 2.0016565433462175, + "grad_norm": 0.8664761781692505, + "learning_rate": 3.8091987208383656e-06, + "loss": 0.5947, + "step": 2418 + }, + { + "epoch": 2.002484815019326, + "grad_norm": 0.9208955764770508, + "learning_rate": 3.808261751917548e-06, + "loss": 0.6028, + "step": 2419 + }, + { + "epoch": 2.003313086692435, + "grad_norm": 0.8842632174491882, + "learning_rate": 3.8073245298602775e-06, + "loss": 0.5913, + "step": 2420 + }, + { + "epoch": 2.004141358365544, + "grad_norm": 0.9232655763626099, + "learning_rate": 3.8063870548478964e-06, + "loss": 0.5977, + "step": 2421 + }, + { + "epoch": 2.004969630038653, + "grad_norm": 0.8898128271102905, + "learning_rate": 3.805449327061798e-06, + "loss": 0.6335, + "step": 2422 + }, + { + "epoch": 2.0057979017117615, + "grad_norm": 0.9356569051742554, + "learning_rate": 3.8045113466834248e-06, + "loss": 0.616, + "step": 2423 + }, + { + "epoch": 2.0066261733848703, + "grad_norm": 0.890315055847168, + "learning_rate": 3.803573113894266e-06, + "loss": 0.5874, + "step": 2424 + }, + { + "epoch": 2.007454445057979, + "grad_norm": 0.8921117186546326, + "learning_rate": 3.802634628875861e-06, + "loss": 0.5679, + "step": 2425 + }, + { + "epoch": 2.0082827167310877, + "grad_norm": 0.8871824145317078, + "learning_rate": 3.8016958918097995e-06, + "loss": 0.6066, + "step": 2426 + }, + { + "epoch": 2.0091109884041964, + "grad_norm": 0.902680516242981, + "learning_rate": 3.800756902877717e-06, + "loss": 0.5936, + "step": 2427 + }, + { + "epoch": 2.009939260077305, + "grad_norm": 0.9199647307395935, + "learning_rate": 3.799817662261298e-06, + "loss": 0.5836, + "step": 2428 + }, + { + "epoch": 2.0107675317504143, + "grad_norm": 0.9029080271720886, + "learning_rate": 3.7988781701422795e-06, + "loss": 0.5858, + "step": 2429 + }, + { + "epoch": 2.011595803423523, + "grad_norm": 0.8834925889968872, + "learning_rate": 3.7979384267024422e-06, + "loss": 0.6061, + "step": 2430 + }, + { + "epoch": 2.012424075096632, + "grad_norm": 0.9098751544952393, + "learning_rate": 3.796998432123619e-06, + "loss": 0.6238, + "step": 2431 + }, + { + "epoch": 2.0132523467697405, + "grad_norm": 0.9316835999488831, + "learning_rate": 3.7960581865876898e-06, + "loss": 0.6139, + "step": 2432 + }, + { + "epoch": 2.0140806184428492, + "grad_norm": 0.9198770523071289, + "learning_rate": 3.795117690276583e-06, + "loss": 0.6113, + "step": 2433 + }, + { + "epoch": 2.014908890115958, + "grad_norm": 0.8970876336097717, + "learning_rate": 3.794176943372275e-06, + "loss": 0.6033, + "step": 2434 + }, + { + "epoch": 2.0157371617890667, + "grad_norm": 0.9025728702545166, + "learning_rate": 3.7932359460567945e-06, + "loss": 0.586, + "step": 2435 + }, + { + "epoch": 2.0165654334621754, + "grad_norm": 0.891909122467041, + "learning_rate": 3.7922946985122124e-06, + "loss": 0.6013, + "step": 2436 + }, + { + "epoch": 2.017393705135284, + "grad_norm": 0.9115222096443176, + "learning_rate": 3.791353200920654e-06, + "loss": 0.5783, + "step": 2437 + }, + { + "epoch": 2.0182219768083933, + "grad_norm": 0.8973708152770996, + "learning_rate": 3.790411453464288e-06, + "loss": 0.5862, + "step": 2438 + }, + { + "epoch": 2.019050248481502, + "grad_norm": 0.910248875617981, + "learning_rate": 3.7894694563253352e-06, + "loss": 0.5884, + "step": 2439 + }, + { + "epoch": 2.019878520154611, + "grad_norm": 0.9227123260498047, + "learning_rate": 3.7885272096860625e-06, + "loss": 0.6185, + "step": 2440 + }, + { + "epoch": 2.0207067918277195, + "grad_norm": 0.917061448097229, + "learning_rate": 3.787584713728787e-06, + "loss": 0.5767, + "step": 2441 + }, + { + "epoch": 2.0215350635008282, + "grad_norm": 0.9035825133323669, + "learning_rate": 3.786641968635871e-06, + "loss": 0.6064, + "step": 2442 + }, + { + "epoch": 2.022363335173937, + "grad_norm": 0.8991772532463074, + "learning_rate": 3.78569897458973e-06, + "loss": 0.5912, + "step": 2443 + }, + { + "epoch": 2.0231916068470457, + "grad_norm": 0.9237397909164429, + "learning_rate": 3.78475573177282e-06, + "loss": 0.5818, + "step": 2444 + }, + { + "epoch": 2.0240198785201544, + "grad_norm": 0.8913205862045288, + "learning_rate": 3.7838122403676543e-06, + "loss": 0.5972, + "step": 2445 + }, + { + "epoch": 2.0248481501932636, + "grad_norm": 0.8960773944854736, + "learning_rate": 3.782868500556787e-06, + "loss": 0.6118, + "step": 2446 + }, + { + "epoch": 2.0256764218663723, + "grad_norm": 0.9198427200317383, + "learning_rate": 3.7819245125228228e-06, + "loss": 0.6058, + "step": 2447 + }, + { + "epoch": 2.026504693539481, + "grad_norm": 0.9039538502693176, + "learning_rate": 3.780980276448416e-06, + "loss": 0.6122, + "step": 2448 + }, + { + "epoch": 2.0273329652125898, + "grad_norm": 0.9103277325630188, + "learning_rate": 3.7800357925162675e-06, + "loss": 0.584, + "step": 2449 + }, + { + "epoch": 2.0281612368856985, + "grad_norm": 0.9114090204238892, + "learning_rate": 3.779091060909124e-06, + "loss": 0.6071, + "step": 2450 + }, + { + "epoch": 2.028989508558807, + "grad_norm": 0.9064953327178955, + "learning_rate": 3.7781460818097847e-06, + "loss": 0.6097, + "step": 2451 + }, + { + "epoch": 2.029817780231916, + "grad_norm": 0.893298864364624, + "learning_rate": 3.7772008554010937e-06, + "loss": 0.5959, + "step": 2452 + }, + { + "epoch": 2.0306460519050247, + "grad_norm": 0.8867486119270325, + "learning_rate": 3.776255381865942e-06, + "loss": 0.5788, + "step": 2453 + }, + { + "epoch": 2.031474323578134, + "grad_norm": 0.9018534421920776, + "learning_rate": 3.775309661387272e-06, + "loss": 0.6075, + "step": 2454 + }, + { + "epoch": 2.0323025952512426, + "grad_norm": 0.9353018999099731, + "learning_rate": 3.77436369414807e-06, + "loss": 0.5894, + "step": 2455 + }, + { + "epoch": 2.0331308669243513, + "grad_norm": 0.8828651309013367, + "learning_rate": 3.7734174803313733e-06, + "loss": 0.608, + "step": 2456 + }, + { + "epoch": 2.03395913859746, + "grad_norm": 0.8971118330955505, + "learning_rate": 3.7724710201202645e-06, + "loss": 0.597, + "step": 2457 + }, + { + "epoch": 2.0347874102705688, + "grad_norm": 0.9064204096794128, + "learning_rate": 3.7715243136978753e-06, + "loss": 0.61, + "step": 2458 + }, + { + "epoch": 2.0356156819436775, + "grad_norm": 0.9204864501953125, + "learning_rate": 3.770577361247384e-06, + "loss": 0.5908, + "step": 2459 + }, + { + "epoch": 2.036443953616786, + "grad_norm": 0.905817449092865, + "learning_rate": 3.7696301629520183e-06, + "loss": 0.5867, + "step": 2460 + }, + { + "epoch": 2.037272225289895, + "grad_norm": 0.8872947096824646, + "learning_rate": 3.7686827189950505e-06, + "loss": 0.5805, + "step": 2461 + }, + { + "epoch": 2.0381004969630037, + "grad_norm": 0.8944478631019592, + "learning_rate": 3.7677350295598035e-06, + "loss": 0.5835, + "step": 2462 + }, + { + "epoch": 2.038928768636113, + "grad_norm": 0.884630560874939, + "learning_rate": 3.766787094829646e-06, + "loss": 0.5909, + "step": 2463 + }, + { + "epoch": 2.0397570403092216, + "grad_norm": 0.8915899395942688, + "learning_rate": 3.7658389149879936e-06, + "loss": 0.5923, + "step": 2464 + }, + { + "epoch": 2.0405853119823303, + "grad_norm": 0.8971114158630371, + "learning_rate": 3.7648904902183128e-06, + "loss": 0.6097, + "step": 2465 + }, + { + "epoch": 2.041413583655439, + "grad_norm": 0.9223624467849731, + "learning_rate": 3.7639418207041124e-06, + "loss": 0.6048, + "step": 2466 + }, + { + "epoch": 2.0422418553285477, + "grad_norm": 0.915450930595398, + "learning_rate": 3.7629929066289517e-06, + "loss": 0.626, + "step": 2467 + }, + { + "epoch": 2.0430701270016565, + "grad_norm": 0.919333279132843, + "learning_rate": 3.762043748176437e-06, + "loss": 0.6234, + "step": 2468 + }, + { + "epoch": 2.043898398674765, + "grad_norm": 0.9282277822494507, + "learning_rate": 3.7610943455302223e-06, + "loss": 0.6008, + "step": 2469 + }, + { + "epoch": 2.044726670347874, + "grad_norm": 0.9419016242027283, + "learning_rate": 3.760144698874007e-06, + "loss": 0.5836, + "step": 2470 + }, + { + "epoch": 2.045554942020983, + "grad_norm": 0.91156405210495, + "learning_rate": 3.7591948083915403e-06, + "loss": 0.6035, + "step": 2471 + }, + { + "epoch": 2.046383213694092, + "grad_norm": 0.9019245505332947, + "learning_rate": 3.758244674266615e-06, + "loss": 0.5949, + "step": 2472 + }, + { + "epoch": 2.0472114853672005, + "grad_norm": 0.8976905345916748, + "learning_rate": 3.757294296683075e-06, + "loss": 0.5818, + "step": 2473 + }, + { + "epoch": 2.0480397570403093, + "grad_norm": 0.8748987913131714, + "learning_rate": 3.7563436758248094e-06, + "loss": 0.5816, + "step": 2474 + }, + { + "epoch": 2.048868028713418, + "grad_norm": 0.8929162621498108, + "learning_rate": 3.7553928118757532e-06, + "loss": 0.5869, + "step": 2475 + }, + { + "epoch": 2.0496963003865267, + "grad_norm": 0.9968146681785583, + "learning_rate": 3.7544417050198913e-06, + "loss": 0.6065, + "step": 2476 + }, + { + "epoch": 2.0505245720596355, + "grad_norm": 0.9206544756889343, + "learning_rate": 3.753490355441253e-06, + "loss": 0.5919, + "step": 2477 + }, + { + "epoch": 2.051352843732744, + "grad_norm": 0.8804112672805786, + "learning_rate": 3.7525387633239153e-06, + "loss": 0.5918, + "step": 2478 + }, + { + "epoch": 2.052181115405853, + "grad_norm": 0.922532320022583, + "learning_rate": 3.7515869288520027e-06, + "loss": 0.6212, + "step": 2479 + }, + { + "epoch": 2.053009387078962, + "grad_norm": 0.9353113770484924, + "learning_rate": 3.7506348522096876e-06, + "loss": 0.5695, + "step": 2480 + }, + { + "epoch": 2.053837658752071, + "grad_norm": 0.9110690951347351, + "learning_rate": 3.749682533581186e-06, + "loss": 0.6156, + "step": 2481 + }, + { + "epoch": 2.0546659304251795, + "grad_norm": 0.9076771140098572, + "learning_rate": 3.7487299731507642e-06, + "loss": 0.5987, + "step": 2482 + }, + { + "epoch": 2.0554942020982883, + "grad_norm": 0.8937394618988037, + "learning_rate": 3.747777171102732e-06, + "loss": 0.6057, + "step": 2483 + }, + { + "epoch": 2.056322473771397, + "grad_norm": 0.9488244652748108, + "learning_rate": 3.74682412762145e-06, + "loss": 0.5983, + "step": 2484 + }, + { + "epoch": 2.0571507454445057, + "grad_norm": 0.897693395614624, + "learning_rate": 3.7458708428913217e-06, + "loss": 0.606, + "step": 2485 + }, + { + "epoch": 2.0579790171176144, + "grad_norm": 0.8989591598510742, + "learning_rate": 3.744917317096799e-06, + "loss": 0.6138, + "step": 2486 + }, + { + "epoch": 2.058807288790723, + "grad_norm": 0.9028336405754089, + "learning_rate": 3.7439635504223804e-06, + "loss": 0.5714, + "step": 2487 + }, + { + "epoch": 2.0596355604638323, + "grad_norm": 0.8895571231842041, + "learning_rate": 3.7430095430526113e-06, + "loss": 0.5903, + "step": 2488 + }, + { + "epoch": 2.060463832136941, + "grad_norm": 0.926414430141449, + "learning_rate": 3.7420552951720822e-06, + "loss": 0.5841, + "step": 2489 + }, + { + "epoch": 2.06129210381005, + "grad_norm": 0.9279632568359375, + "learning_rate": 3.741100806965432e-06, + "loss": 0.6062, + "step": 2490 + }, + { + "epoch": 2.0621203754831585, + "grad_norm": 0.9196921586990356, + "learning_rate": 3.7401460786173456e-06, + "loss": 0.579, + "step": 2491 + }, + { + "epoch": 2.0629486471562672, + "grad_norm": 0.8959674835205078, + "learning_rate": 3.739191110312552e-06, + "loss": 0.5888, + "step": 2492 + }, + { + "epoch": 2.063776918829376, + "grad_norm": 0.9123788475990295, + "learning_rate": 3.7382359022358322e-06, + "loss": 0.5891, + "step": 2493 + }, + { + "epoch": 2.0646051905024847, + "grad_norm": 0.9637120366096497, + "learning_rate": 3.7372804545720067e-06, + "loss": 0.6124, + "step": 2494 + }, + { + "epoch": 2.0654334621755934, + "grad_norm": 0.9360100626945496, + "learning_rate": 3.736324767505947e-06, + "loss": 0.5884, + "step": 2495 + }, + { + "epoch": 2.0662617338487026, + "grad_norm": 0.9066753387451172, + "learning_rate": 3.7353688412225704e-06, + "loss": 0.6013, + "step": 2496 + }, + { + "epoch": 2.0670900055218113, + "grad_norm": 0.8961504101753235, + "learning_rate": 3.7344126759068382e-06, + "loss": 0.6303, + "step": 2497 + }, + { + "epoch": 2.06791827719492, + "grad_norm": 0.8995484709739685, + "learning_rate": 3.7334562717437608e-06, + "loss": 0.6031, + "step": 2498 + }, + { + "epoch": 2.068746548868029, + "grad_norm": 0.9279561042785645, + "learning_rate": 3.7324996289183927e-06, + "loss": 0.5848, + "step": 2499 + }, + { + "epoch": 2.0695748205411375, + "grad_norm": 0.9250274300575256, + "learning_rate": 3.7315427476158357e-06, + "loss": 0.5949, + "step": 2500 + }, + { + "epoch": 2.0704030922142462, + "grad_norm": 0.9003238677978516, + "learning_rate": 3.730585628021237e-06, + "loss": 0.6124, + "step": 2501 + }, + { + "epoch": 2.071231363887355, + "grad_norm": 0.9069858193397522, + "learning_rate": 3.729628270319791e-06, + "loss": 0.6051, + "step": 2502 + }, + { + "epoch": 2.0720596355604637, + "grad_norm": 0.9361990094184875, + "learning_rate": 3.7286706746967372e-06, + "loss": 0.5863, + "step": 2503 + }, + { + "epoch": 2.0728879072335724, + "grad_norm": 0.9443079233169556, + "learning_rate": 3.7277128413373607e-06, + "loss": 0.5959, + "step": 2504 + }, + { + "epoch": 2.0737161789066816, + "grad_norm": 0.9413166046142578, + "learning_rate": 3.7267547704269957e-06, + "loss": 0.5882, + "step": 2505 + }, + { + "epoch": 2.0745444505797903, + "grad_norm": 0.8937835097312927, + "learning_rate": 3.7257964621510167e-06, + "loss": 0.6108, + "step": 2506 + }, + { + "epoch": 2.075372722252899, + "grad_norm": 0.9188135862350464, + "learning_rate": 3.7248379166948497e-06, + "loss": 0.5812, + "step": 2507 + }, + { + "epoch": 2.0762009939260078, + "grad_norm": 0.916559636592865, + "learning_rate": 3.723879134243963e-06, + "loss": 0.6169, + "step": 2508 + }, + { + "epoch": 2.0770292655991165, + "grad_norm": 0.9284812211990356, + "learning_rate": 3.7229201149838727e-06, + "loss": 0.6224, + "step": 2509 + }, + { + "epoch": 2.077857537272225, + "grad_norm": 0.918048083782196, + "learning_rate": 3.721960859100141e-06, + "loss": 0.6043, + "step": 2510 + }, + { + "epoch": 2.078685808945334, + "grad_norm": 0.9250833988189697, + "learning_rate": 3.721001366778373e-06, + "loss": 0.5882, + "step": 2511 + }, + { + "epoch": 2.0795140806184427, + "grad_norm": 0.8975113034248352, + "learning_rate": 3.720041638204223e-06, + "loss": 0.6049, + "step": 2512 + }, + { + "epoch": 2.080342352291552, + "grad_norm": 0.9108197093009949, + "learning_rate": 3.7190816735633896e-06, + "loss": 0.6174, + "step": 2513 + }, + { + "epoch": 2.0811706239646606, + "grad_norm": 0.91886967420578, + "learning_rate": 3.718121473041615e-06, + "loss": 0.5854, + "step": 2514 + }, + { + "epoch": 2.0819988956377693, + "grad_norm": 0.9149959683418274, + "learning_rate": 3.717161036824692e-06, + "loss": 0.5808, + "step": 2515 + }, + { + "epoch": 2.082827167310878, + "grad_norm": 0.9402881860733032, + "learning_rate": 3.7162003650984545e-06, + "loss": 0.5936, + "step": 2516 + }, + { + "epoch": 2.0836554389839868, + "grad_norm": 0.8952768445014954, + "learning_rate": 3.7152394580487825e-06, + "loss": 0.5977, + "step": 2517 + }, + { + "epoch": 2.0844837106570955, + "grad_norm": 0.9138445258140564, + "learning_rate": 3.714278315861605e-06, + "loss": 0.5822, + "step": 2518 + }, + { + "epoch": 2.085311982330204, + "grad_norm": 0.9174036979675293, + "learning_rate": 3.713316938722892e-06, + "loss": 0.6061, + "step": 2519 + }, + { + "epoch": 2.086140254003313, + "grad_norm": 0.9291302561759949, + "learning_rate": 3.7123553268186618e-06, + "loss": 0.614, + "step": 2520 + }, + { + "epoch": 2.0869685256764217, + "grad_norm": 0.9058101773262024, + "learning_rate": 3.711393480334977e-06, + "loss": 0.6053, + "step": 2521 + }, + { + "epoch": 2.087796797349531, + "grad_norm": 0.926360011100769, + "learning_rate": 3.7104313994579473e-06, + "loss": 0.5781, + "step": 2522 + }, + { + "epoch": 2.0886250690226396, + "grad_norm": 0.898629367351532, + "learning_rate": 3.709469084373725e-06, + "loss": 0.5995, + "step": 2523 + }, + { + "epoch": 2.0894533406957483, + "grad_norm": 0.9013628363609314, + "learning_rate": 3.708506535268509e-06, + "loss": 0.5929, + "step": 2524 + }, + { + "epoch": 2.090281612368857, + "grad_norm": 0.8909673094749451, + "learning_rate": 3.7075437523285433e-06, + "loss": 0.5495, + "step": 2525 + }, + { + "epoch": 2.0911098840419657, + "grad_norm": 0.9176347255706787, + "learning_rate": 3.706580735740119e-06, + "loss": 0.6134, + "step": 2526 + }, + { + "epoch": 2.0919381557150745, + "grad_norm": 0.9174275398254395, + "learning_rate": 3.7056174856895703e-06, + "loss": 0.601, + "step": 2527 + }, + { + "epoch": 2.092766427388183, + "grad_norm": 0.9134400486946106, + "learning_rate": 3.704654002363276e-06, + "loss": 0.5952, + "step": 2528 + }, + { + "epoch": 2.093594699061292, + "grad_norm": 0.9206627607345581, + "learning_rate": 3.703690285947662e-06, + "loss": 0.6016, + "step": 2529 + }, + { + "epoch": 2.094422970734401, + "grad_norm": 0.9271345734596252, + "learning_rate": 3.702726336629199e-06, + "loss": 0.5937, + "step": 2530 + }, + { + "epoch": 2.09525124240751, + "grad_norm": 0.9390851855278015, + "learning_rate": 3.7017621545944003e-06, + "loss": 0.5858, + "step": 2531 + }, + { + "epoch": 2.0960795140806185, + "grad_norm": 0.9351701736450195, + "learning_rate": 3.700797740029828e-06, + "loss": 0.5978, + "step": 2532 + }, + { + "epoch": 2.0969077857537273, + "grad_norm": 0.9432377815246582, + "learning_rate": 3.6998330931220873e-06, + "loss": 0.5884, + "step": 2533 + }, + { + "epoch": 2.097736057426836, + "grad_norm": 0.9192067384719849, + "learning_rate": 3.698868214057827e-06, + "loss": 0.5989, + "step": 2534 + }, + { + "epoch": 2.0985643290999447, + "grad_norm": 0.9810166954994202, + "learning_rate": 3.6979031030237433e-06, + "loss": 0.6078, + "step": 2535 + }, + { + "epoch": 2.0993926007730535, + "grad_norm": 0.9065185785293579, + "learning_rate": 3.696937760206576e-06, + "loss": 0.5947, + "step": 2536 + }, + { + "epoch": 2.100220872446162, + "grad_norm": 0.9212245345115662, + "learning_rate": 3.6959721857931088e-06, + "loss": 0.5951, + "step": 2537 + }, + { + "epoch": 2.101049144119271, + "grad_norm": 0.9291266798973083, + "learning_rate": 3.695006379970173e-06, + "loss": 0.6033, + "step": 2538 + }, + { + "epoch": 2.10187741579238, + "grad_norm": 0.9154751896858215, + "learning_rate": 3.694040342924643e-06, + "loss": 0.612, + "step": 2539 + }, + { + "epoch": 2.102705687465489, + "grad_norm": 0.9028742909431458, + "learning_rate": 3.693074074843437e-06, + "loss": 0.5948, + "step": 2540 + }, + { + "epoch": 2.1035339591385975, + "grad_norm": 0.9132126569747925, + "learning_rate": 3.692107575913519e-06, + "loss": 0.6198, + "step": 2541 + }, + { + "epoch": 2.1043622308117063, + "grad_norm": 0.9116196036338806, + "learning_rate": 3.6911408463218974e-06, + "loss": 0.5927, + "step": 2542 + }, + { + "epoch": 2.105190502484815, + "grad_norm": 0.9670912027359009, + "learning_rate": 3.6901738862556263e-06, + "loss": 0.5813, + "step": 2543 + }, + { + "epoch": 2.1060187741579237, + "grad_norm": 0.9534422755241394, + "learning_rate": 3.6892066959018034e-06, + "loss": 0.5892, + "step": 2544 + }, + { + "epoch": 2.1068470458310324, + "grad_norm": 0.9175525903701782, + "learning_rate": 3.688239275447569e-06, + "loss": 0.5946, + "step": 2545 + }, + { + "epoch": 2.107675317504141, + "grad_norm": 0.8945409655570984, + "learning_rate": 3.6872716250801125e-06, + "loss": 0.6013, + "step": 2546 + }, + { + "epoch": 2.1085035891772503, + "grad_norm": 0.9498820900917053, + "learning_rate": 3.686303744986664e-06, + "loss": 0.5785, + "step": 2547 + }, + { + "epoch": 2.109331860850359, + "grad_norm": 0.9338169693946838, + "learning_rate": 3.685335635354499e-06, + "loss": 0.5959, + "step": 2548 + }, + { + "epoch": 2.110160132523468, + "grad_norm": 0.9100049734115601, + "learning_rate": 3.684367296370939e-06, + "loss": 0.5829, + "step": 2549 + }, + { + "epoch": 2.1109884041965765, + "grad_norm": 0.904279887676239, + "learning_rate": 3.6833987282233467e-06, + "loss": 0.6015, + "step": 2550 + }, + { + "epoch": 2.1118166758696852, + "grad_norm": 0.8693552613258362, + "learning_rate": 3.6824299310991324e-06, + "loss": 0.5825, + "step": 2551 + }, + { + "epoch": 2.112644947542794, + "grad_norm": 0.9180129766464233, + "learning_rate": 3.681460905185749e-06, + "loss": 0.6076, + "step": 2552 + }, + { + "epoch": 2.1134732192159027, + "grad_norm": 0.9554781913757324, + "learning_rate": 3.6804916506706936e-06, + "loss": 0.5755, + "step": 2553 + }, + { + "epoch": 2.1143014908890114, + "grad_norm": 0.9498181939125061, + "learning_rate": 3.6795221677415085e-06, + "loss": 0.6086, + "step": 2554 + }, + { + "epoch": 2.1151297625621206, + "grad_norm": 0.9110347032546997, + "learning_rate": 3.678552456585779e-06, + "loss": 0.591, + "step": 2555 + }, + { + "epoch": 2.1159580342352293, + "grad_norm": 0.8883147835731506, + "learning_rate": 3.6775825173911363e-06, + "loss": 0.5753, + "step": 2556 + }, + { + "epoch": 2.116786305908338, + "grad_norm": 0.9354976415634155, + "learning_rate": 3.676612350345253e-06, + "loss": 0.6063, + "step": 2557 + }, + { + "epoch": 2.1176145775814468, + "grad_norm": 0.9463969469070435, + "learning_rate": 3.675641955635848e-06, + "loss": 0.5983, + "step": 2558 + }, + { + "epoch": 2.1184428492545555, + "grad_norm": 0.9220679998397827, + "learning_rate": 3.674671333450684e-06, + "loss": 0.6026, + "step": 2559 + }, + { + "epoch": 2.1192711209276642, + "grad_norm": 0.9050368666648865, + "learning_rate": 3.673700483977567e-06, + "loss": 0.6075, + "step": 2560 + }, + { + "epoch": 2.120099392600773, + "grad_norm": 0.9299139976501465, + "learning_rate": 3.6727294074043478e-06, + "loss": 0.5971, + "step": 2561 + }, + { + "epoch": 2.1209276642738817, + "grad_norm": 0.9060289263725281, + "learning_rate": 3.6717581039189205e-06, + "loss": 0.5794, + "step": 2562 + }, + { + "epoch": 2.1217559359469904, + "grad_norm": 0.9253103137016296, + "learning_rate": 3.6707865737092228e-06, + "loss": 0.593, + "step": 2563 + }, + { + "epoch": 2.1225842076200996, + "grad_norm": 0.9475048780441284, + "learning_rate": 3.669814816963237e-06, + "loss": 0.5867, + "step": 2564 + }, + { + "epoch": 2.1234124792932083, + "grad_norm": 0.9546397924423218, + "learning_rate": 3.6688428338689887e-06, + "loss": 0.5766, + "step": 2565 + }, + { + "epoch": 2.124240750966317, + "grad_norm": 0.9377962350845337, + "learning_rate": 3.667870624614548e-06, + "loss": 0.5853, + "step": 2566 + }, + { + "epoch": 2.1250690226394258, + "grad_norm": 0.9562273621559143, + "learning_rate": 3.6668981893880284e-06, + "loss": 0.5992, + "step": 2567 + }, + { + "epoch": 2.1258972943125345, + "grad_norm": 0.9207432866096497, + "learning_rate": 3.6659255283775867e-06, + "loss": 0.6029, + "step": 2568 + }, + { + "epoch": 2.126725565985643, + "grad_norm": 0.9359074234962463, + "learning_rate": 3.6649526417714244e-06, + "loss": 0.6068, + "step": 2569 + }, + { + "epoch": 2.127553837658752, + "grad_norm": 0.9253597259521484, + "learning_rate": 3.663979529757784e-06, + "loss": 0.581, + "step": 2570 + }, + { + "epoch": 2.1283821093318607, + "grad_norm": 0.9551291465759277, + "learning_rate": 3.6630061925249566e-06, + "loss": 0.607, + "step": 2571 + }, + { + "epoch": 2.1292103810049694, + "grad_norm": 0.91109699010849, + "learning_rate": 3.662032630261272e-06, + "loss": 0.5724, + "step": 2572 + }, + { + "epoch": 2.1300386526780786, + "grad_norm": 0.9486557245254517, + "learning_rate": 3.661058843155105e-06, + "loss": 0.5904, + "step": 2573 + }, + { + "epoch": 2.1308669243511873, + "grad_norm": 0.9550185203552246, + "learning_rate": 3.6600848313948762e-06, + "loss": 0.5967, + "step": 2574 + }, + { + "epoch": 2.131695196024296, + "grad_norm": 0.9457482695579529, + "learning_rate": 3.6591105951690464e-06, + "loss": 0.6253, + "step": 2575 + }, + { + "epoch": 2.1325234676974048, + "grad_norm": 0.9686241149902344, + "learning_rate": 3.6581361346661204e-06, + "loss": 0.5928, + "step": 2576 + }, + { + "epoch": 2.1333517393705135, + "grad_norm": 0.9125609993934631, + "learning_rate": 3.65716145007465e-06, + "loss": 0.6039, + "step": 2577 + }, + { + "epoch": 2.134180011043622, + "grad_norm": 0.927651047706604, + "learning_rate": 3.656186541583225e-06, + "loss": 0.5881, + "step": 2578 + }, + { + "epoch": 2.135008282716731, + "grad_norm": 0.9446879029273987, + "learning_rate": 3.6552114093804825e-06, + "loss": 0.5938, + "step": 2579 + }, + { + "epoch": 2.13583655438984, + "grad_norm": 0.944601833820343, + "learning_rate": 3.6542360536551003e-06, + "loss": 0.5804, + "step": 2580 + }, + { + "epoch": 2.136664826062949, + "grad_norm": 0.9230138063430786, + "learning_rate": 3.6532604745958015e-06, + "loss": 0.6002, + "step": 2581 + }, + { + "epoch": 2.1374930977360576, + "grad_norm": 0.9163529276847839, + "learning_rate": 3.6522846723913517e-06, + "loss": 0.5811, + "step": 2582 + }, + { + "epoch": 2.1383213694091663, + "grad_norm": 0.9153618812561035, + "learning_rate": 3.6513086472305596e-06, + "loss": 0.6018, + "step": 2583 + }, + { + "epoch": 2.139149641082275, + "grad_norm": 0.9299359917640686, + "learning_rate": 3.6503323993022767e-06, + "loss": 0.5944, + "step": 2584 + }, + { + "epoch": 2.1399779127553837, + "grad_norm": 0.9075869917869568, + "learning_rate": 3.6493559287953974e-06, + "loss": 0.595, + "step": 2585 + }, + { + "epoch": 2.1408061844284925, + "grad_norm": 0.9388751983642578, + "learning_rate": 3.64837923589886e-06, + "loss": 0.6006, + "step": 2586 + }, + { + "epoch": 2.141634456101601, + "grad_norm": 0.91156405210495, + "learning_rate": 3.647402320801645e-06, + "loss": 0.5955, + "step": 2587 + }, + { + "epoch": 2.14246272777471, + "grad_norm": 0.9187628626823425, + "learning_rate": 3.6464251836927776e-06, + "loss": 0.6119, + "step": 2588 + }, + { + "epoch": 2.143290999447819, + "grad_norm": 0.9578025341033936, + "learning_rate": 3.645447824761324e-06, + "loss": 0.6076, + "step": 2589 + }, + { + "epoch": 2.144119271120928, + "grad_norm": 0.9159366488456726, + "learning_rate": 3.644470244196394e-06, + "loss": 0.6015, + "step": 2590 + }, + { + "epoch": 2.1449475427940365, + "grad_norm": 0.924027681350708, + "learning_rate": 3.6434924421871403e-06, + "loss": 0.6082, + "step": 2591 + }, + { + "epoch": 2.1457758144671453, + "grad_norm": 0.8900635242462158, + "learning_rate": 3.642514418922758e-06, + "loss": 0.611, + "step": 2592 + }, + { + "epoch": 2.146604086140254, + "grad_norm": 0.9609982371330261, + "learning_rate": 3.641536174592487e-06, + "loss": 0.5974, + "step": 2593 + }, + { + "epoch": 2.1474323578133627, + "grad_norm": 0.901737630367279, + "learning_rate": 3.6405577093856072e-06, + "loss": 0.5869, + "step": 2594 + }, + { + "epoch": 2.1482606294864715, + "grad_norm": 0.9342123866081238, + "learning_rate": 3.6395790234914423e-06, + "loss": 0.6104, + "step": 2595 + }, + { + "epoch": 2.14908890115958, + "grad_norm": 0.9063602685928345, + "learning_rate": 3.6386001170993596e-06, + "loss": 0.5801, + "step": 2596 + }, + { + "epoch": 2.149917172832689, + "grad_norm": 0.9269636869430542, + "learning_rate": 3.637620990398768e-06, + "loss": 0.5953, + "step": 2597 + }, + { + "epoch": 2.150745444505798, + "grad_norm": 0.9278765320777893, + "learning_rate": 3.63664164357912e-06, + "loss": 0.6151, + "step": 2598 + }, + { + "epoch": 2.151573716178907, + "grad_norm": 0.9786942601203918, + "learning_rate": 3.6356620768299087e-06, + "loss": 0.5773, + "step": 2599 + }, + { + "epoch": 2.1524019878520155, + "grad_norm": 0.9069851636886597, + "learning_rate": 3.6346822903406733e-06, + "loss": 0.6087, + "step": 2600 + }, + { + "epoch": 2.1532302595251243, + "grad_norm": 0.8867887854576111, + "learning_rate": 3.6337022843009913e-06, + "loss": 0.5894, + "step": 2601 + }, + { + "epoch": 2.154058531198233, + "grad_norm": 0.9444507360458374, + "learning_rate": 3.632722058900486e-06, + "loss": 0.6125, + "step": 2602 + }, + { + "epoch": 2.1548868028713417, + "grad_norm": 0.9410359859466553, + "learning_rate": 3.631741614328821e-06, + "loss": 0.5845, + "step": 2603 + }, + { + "epoch": 2.1557150745444504, + "grad_norm": 0.9473987817764282, + "learning_rate": 3.630760950775703e-06, + "loss": 0.5944, + "step": 2604 + }, + { + "epoch": 2.156543346217559, + "grad_norm": 0.913758397102356, + "learning_rate": 3.6297800684308828e-06, + "loss": 0.5998, + "step": 2605 + }, + { + "epoch": 2.1573716178906683, + "grad_norm": 0.89509516954422, + "learning_rate": 3.6287989674841505e-06, + "loss": 0.6013, + "step": 2606 + }, + { + "epoch": 2.158199889563777, + "grad_norm": 0.9042367339134216, + "learning_rate": 3.6278176481253413e-06, + "loss": 0.5915, + "step": 2607 + }, + { + "epoch": 2.159028161236886, + "grad_norm": 0.9108945727348328, + "learning_rate": 3.6268361105443297e-06, + "loss": 0.5803, + "step": 2608 + }, + { + "epoch": 2.1598564329099945, + "grad_norm": 0.8903773427009583, + "learning_rate": 3.625854354931035e-06, + "loss": 0.6025, + "step": 2609 + }, + { + "epoch": 2.1606847045831032, + "grad_norm": 0.9396197199821472, + "learning_rate": 3.6248723814754183e-06, + "loss": 0.6271, + "step": 2610 + }, + { + "epoch": 2.161512976256212, + "grad_norm": 0.9247238039970398, + "learning_rate": 3.623890190367481e-06, + "loss": 0.5945, + "step": 2611 + }, + { + "epoch": 2.1623412479293207, + "grad_norm": 0.9240396022796631, + "learning_rate": 3.622907781797269e-06, + "loss": 0.6148, + "step": 2612 + }, + { + "epoch": 2.1631695196024294, + "grad_norm": 0.9257000684738159, + "learning_rate": 3.621925155954869e-06, + "loss": 0.5964, + "step": 2613 + }, + { + "epoch": 2.1639977912755386, + "grad_norm": 0.8854860663414001, + "learning_rate": 3.62094231303041e-06, + "loss": 0.6012, + "step": 2614 + }, + { + "epoch": 2.1648260629486473, + "grad_norm": 0.9251745343208313, + "learning_rate": 3.619959253214062e-06, + "loss": 0.6001, + "step": 2615 + }, + { + "epoch": 2.165654334621756, + "grad_norm": 0.9081252813339233, + "learning_rate": 3.618975976696039e-06, + "loss": 0.5927, + "step": 2616 + }, + { + "epoch": 2.1664826062948648, + "grad_norm": 0.9065699577331543, + "learning_rate": 3.6179924836665953e-06, + "loss": 0.6079, + "step": 2617 + }, + { + "epoch": 2.1673108779679735, + "grad_norm": 0.9421206712722778, + "learning_rate": 3.617008774316028e-06, + "loss": 0.6041, + "step": 2618 + }, + { + "epoch": 2.1681391496410822, + "grad_norm": 0.8863933682441711, + "learning_rate": 3.6160248488346757e-06, + "loss": 0.5826, + "step": 2619 + }, + { + "epoch": 2.168967421314191, + "grad_norm": 0.915887713432312, + "learning_rate": 3.615040707412918e-06, + "loss": 0.5972, + "step": 2620 + }, + { + "epoch": 2.1697956929872997, + "grad_norm": 0.9150416851043701, + "learning_rate": 3.614056350241178e-06, + "loss": 0.606, + "step": 2621 + }, + { + "epoch": 2.1706239646604084, + "grad_norm": 0.9050796627998352, + "learning_rate": 3.6130717775099195e-06, + "loss": 0.5955, + "step": 2622 + }, + { + "epoch": 2.1714522363335176, + "grad_norm": 0.9249832630157471, + "learning_rate": 3.612086989409648e-06, + "loss": 0.5968, + "step": 2623 + }, + { + "epoch": 2.1722805080066263, + "grad_norm": 0.8855478763580322, + "learning_rate": 3.6111019861309105e-06, + "loss": 0.5657, + "step": 2624 + }, + { + "epoch": 2.173108779679735, + "grad_norm": 0.9182875156402588, + "learning_rate": 3.6101167678642967e-06, + "loss": 0.6096, + "step": 2625 + }, + { + "epoch": 2.1739370513528438, + "grad_norm": 0.9722150564193726, + "learning_rate": 3.609131334800436e-06, + "loss": 0.6029, + "step": 2626 + }, + { + "epoch": 2.1747653230259525, + "grad_norm": 0.8925914168357849, + "learning_rate": 3.6081456871300014e-06, + "loss": 0.5928, + "step": 2627 + }, + { + "epoch": 2.175593594699061, + "grad_norm": 0.8883031010627747, + "learning_rate": 3.6071598250437067e-06, + "loss": 0.5763, + "step": 2628 + }, + { + "epoch": 2.17642186637217, + "grad_norm": 0.8948857188224792, + "learning_rate": 3.606173748732307e-06, + "loss": 0.6078, + "step": 2629 + }, + { + "epoch": 2.1772501380452787, + "grad_norm": 0.9444127082824707, + "learning_rate": 3.6051874583865975e-06, + "loss": 0.6005, + "step": 2630 + }, + { + "epoch": 2.1780784097183874, + "grad_norm": 0.9361352920532227, + "learning_rate": 3.6042009541974176e-06, + "loss": 0.5878, + "step": 2631 + }, + { + "epoch": 2.1789066813914966, + "grad_norm": 0.9124311804771423, + "learning_rate": 3.6032142363556467e-06, + "loss": 0.6115, + "step": 2632 + }, + { + "epoch": 2.1797349530646053, + "grad_norm": 0.9168087840080261, + "learning_rate": 3.602227305052205e-06, + "loss": 0.5928, + "step": 2633 + }, + { + "epoch": 2.180563224737714, + "grad_norm": 0.9189275503158569, + "learning_rate": 3.6012401604780543e-06, + "loss": 0.6081, + "step": 2634 + }, + { + "epoch": 2.1813914964108228, + "grad_norm": 0.937911868095398, + "learning_rate": 3.6002528028241984e-06, + "loss": 0.6062, + "step": 2635 + }, + { + "epoch": 2.1822197680839315, + "grad_norm": 0.9139302968978882, + "learning_rate": 3.5992652322816817e-06, + "loss": 0.6111, + "step": 2636 + }, + { + "epoch": 2.18304803975704, + "grad_norm": 0.951528012752533, + "learning_rate": 3.598277449041589e-06, + "loss": 0.6007, + "step": 2637 + }, + { + "epoch": 2.183876311430149, + "grad_norm": 0.9146255850791931, + "learning_rate": 3.5972894532950487e-06, + "loss": 0.6094, + "step": 2638 + }, + { + "epoch": 2.184704583103258, + "grad_norm": 0.9217225313186646, + "learning_rate": 3.596301245233228e-06, + "loss": 0.6082, + "step": 2639 + }, + { + "epoch": 2.185532854776367, + "grad_norm": 0.9742298126220703, + "learning_rate": 3.5953128250473355e-06, + "loss": 0.6269, + "step": 2640 + }, + { + "epoch": 2.1863611264494756, + "grad_norm": 0.9038170576095581, + "learning_rate": 3.5943241929286223e-06, + "loss": 0.6062, + "step": 2641 + }, + { + "epoch": 2.1871893981225843, + "grad_norm": 0.8960873484611511, + "learning_rate": 3.5933353490683786e-06, + "loss": 0.5748, + "step": 2642 + }, + { + "epoch": 2.188017669795693, + "grad_norm": 0.8801732659339905, + "learning_rate": 3.5923462936579362e-06, + "loss": 0.5911, + "step": 2643 + }, + { + "epoch": 2.1888459414688017, + "grad_norm": 0.8918222188949585, + "learning_rate": 3.5913570268886695e-06, + "loss": 0.5931, + "step": 2644 + }, + { + "epoch": 2.1896742131419105, + "grad_norm": 0.9046133160591125, + "learning_rate": 3.5903675489519917e-06, + "loss": 0.5985, + "step": 2645 + }, + { + "epoch": 2.190502484815019, + "grad_norm": 0.9003219604492188, + "learning_rate": 3.589377860039357e-06, + "loss": 0.608, + "step": 2646 + }, + { + "epoch": 2.191330756488128, + "grad_norm": 0.9139525890350342, + "learning_rate": 3.5883879603422616e-06, + "loss": 0.6186, + "step": 2647 + }, + { + "epoch": 2.192159028161237, + "grad_norm": 0.9125071167945862, + "learning_rate": 3.587397850052241e-06, + "loss": 0.6129, + "step": 2648 + }, + { + "epoch": 2.192987299834346, + "grad_norm": 0.922412097454071, + "learning_rate": 3.586407529360874e-06, + "loss": 0.6008, + "step": 2649 + }, + { + "epoch": 2.1938155715074545, + "grad_norm": 0.9294847249984741, + "learning_rate": 3.5854169984597776e-06, + "loss": 0.5984, + "step": 2650 + }, + { + "epoch": 2.1946438431805633, + "grad_norm": 0.9262758493423462, + "learning_rate": 3.5844262575406084e-06, + "loss": 0.5982, + "step": 2651 + }, + { + "epoch": 2.195472114853672, + "grad_norm": 0.9311552047729492, + "learning_rate": 3.5834353067950694e-06, + "loss": 0.5965, + "step": 2652 + }, + { + "epoch": 2.1963003865267807, + "grad_norm": 0.9087333679199219, + "learning_rate": 3.5824441464148975e-06, + "loss": 0.5972, + "step": 2653 + }, + { + "epoch": 2.1971286581998895, + "grad_norm": 0.9212601184844971, + "learning_rate": 3.5814527765918733e-06, + "loss": 0.5821, + "step": 2654 + }, + { + "epoch": 2.197956929872998, + "grad_norm": 0.9208887219429016, + "learning_rate": 3.5804611975178184e-06, + "loss": 0.592, + "step": 2655 + }, + { + "epoch": 2.198785201546107, + "grad_norm": 0.923478364944458, + "learning_rate": 3.5794694093845933e-06, + "loss": 0.599, + "step": 2656 + }, + { + "epoch": 2.199613473219216, + "grad_norm": 0.9030081033706665, + "learning_rate": 3.5784774123841006e-06, + "loss": 0.5954, + "step": 2657 + }, + { + "epoch": 2.200441744892325, + "grad_norm": 0.9108036160469055, + "learning_rate": 3.5774852067082817e-06, + "loss": 0.6119, + "step": 2658 + }, + { + "epoch": 2.2012700165654335, + "grad_norm": 0.9323570728302002, + "learning_rate": 3.576492792549119e-06, + "loss": 0.6061, + "step": 2659 + }, + { + "epoch": 2.2020982882385423, + "grad_norm": 0.9002283811569214, + "learning_rate": 3.575500170098637e-06, + "loss": 0.6183, + "step": 2660 + }, + { + "epoch": 2.202926559911651, + "grad_norm": 0.9354086518287659, + "learning_rate": 3.574507339548897e-06, + "loss": 0.5989, + "step": 2661 + }, + { + "epoch": 2.2037548315847597, + "grad_norm": 0.9035785794258118, + "learning_rate": 3.5735143010920033e-06, + "loss": 0.592, + "step": 2662 + }, + { + "epoch": 2.2045831032578684, + "grad_norm": 0.9337678551673889, + "learning_rate": 3.5725210549201e-06, + "loss": 0.5881, + "step": 2663 + }, + { + "epoch": 2.205411374930977, + "grad_norm": 0.9226170182228088, + "learning_rate": 3.57152760122537e-06, + "loss": 0.5745, + "step": 2664 + }, + { + "epoch": 2.2062396466040863, + "grad_norm": 0.9272558093070984, + "learning_rate": 3.5705339402000377e-06, + "loss": 0.607, + "step": 2665 + }, + { + "epoch": 2.207067918277195, + "grad_norm": 0.930325448513031, + "learning_rate": 3.569540072036368e-06, + "loss": 0.5947, + "step": 2666 + }, + { + "epoch": 2.207896189950304, + "grad_norm": 0.9052232503890991, + "learning_rate": 3.568545996926664e-06, + "loss": 0.5883, + "step": 2667 + }, + { + "epoch": 2.2087244616234125, + "grad_norm": 0.8831207752227783, + "learning_rate": 3.5675517150632706e-06, + "loss": 0.5962, + "step": 2668 + }, + { + "epoch": 2.2095527332965212, + "grad_norm": 0.8999590873718262, + "learning_rate": 3.566557226638572e-06, + "loss": 0.6257, + "step": 2669 + }, + { + "epoch": 2.21038100496963, + "grad_norm": 0.9310323596000671, + "learning_rate": 3.565562531844992e-06, + "loss": 0.585, + "step": 2670 + }, + { + "epoch": 2.2112092766427387, + "grad_norm": 0.9469419121742249, + "learning_rate": 3.5645676308749954e-06, + "loss": 0.6163, + "step": 2671 + }, + { + "epoch": 2.2120375483158474, + "grad_norm": 0.9391891956329346, + "learning_rate": 3.563572523921086e-06, + "loss": 0.5863, + "step": 2672 + }, + { + "epoch": 2.2128658199889566, + "grad_norm": 0.9035549163818359, + "learning_rate": 3.5625772111758085e-06, + "loss": 0.6074, + "step": 2673 + }, + { + "epoch": 2.2136940916620653, + "grad_norm": 0.9008681178092957, + "learning_rate": 3.561581692831745e-06, + "loss": 0.5965, + "step": 2674 + }, + { + "epoch": 2.214522363335174, + "grad_norm": 0.9130175113677979, + "learning_rate": 3.56058596908152e-06, + "loss": 0.5913, + "step": 2675 + }, + { + "epoch": 2.2153506350082828, + "grad_norm": 0.9367324113845825, + "learning_rate": 3.5595900401177957e-06, + "loss": 0.617, + "step": 2676 + }, + { + "epoch": 2.2161789066813915, + "grad_norm": 0.9082489609718323, + "learning_rate": 3.5585939061332775e-06, + "loss": 0.6052, + "step": 2677 + }, + { + "epoch": 2.2170071783545002, + "grad_norm": 0.921264111995697, + "learning_rate": 3.557597567320706e-06, + "loss": 0.5832, + "step": 2678 + }, + { + "epoch": 2.217835450027609, + "grad_norm": 0.9040131568908691, + "learning_rate": 3.556601023872863e-06, + "loss": 0.5994, + "step": 2679 + }, + { + "epoch": 2.2186637217007177, + "grad_norm": 0.9144189357757568, + "learning_rate": 3.5556042759825733e-06, + "loss": 0.6068, + "step": 2680 + }, + { + "epoch": 2.2194919933738264, + "grad_norm": 0.91495680809021, + "learning_rate": 3.5546073238426955e-06, + "loss": 0.6154, + "step": 2681 + }, + { + "epoch": 2.2203202650469356, + "grad_norm": 0.925397515296936, + "learning_rate": 3.553610167646131e-06, + "loss": 0.6013, + "step": 2682 + }, + { + "epoch": 2.2211485367200443, + "grad_norm": 0.9633582234382629, + "learning_rate": 3.552612807585821e-06, + "loss": 0.6063, + "step": 2683 + }, + { + "epoch": 2.221976808393153, + "grad_norm": 0.9403156042098999, + "learning_rate": 3.551615243854746e-06, + "loss": 0.6141, + "step": 2684 + }, + { + "epoch": 2.2228050800662618, + "grad_norm": 0.9272189140319824, + "learning_rate": 3.550617476645923e-06, + "loss": 0.6013, + "step": 2685 + }, + { + "epoch": 2.2236333517393705, + "grad_norm": 0.92893385887146, + "learning_rate": 3.549619506152413e-06, + "loss": 0.5909, + "step": 2686 + }, + { + "epoch": 2.224461623412479, + "grad_norm": 0.9130013585090637, + "learning_rate": 3.548621332567312e-06, + "loss": 0.6005, + "step": 2687 + }, + { + "epoch": 2.225289895085588, + "grad_norm": 0.9111050367355347, + "learning_rate": 3.547622956083759e-06, + "loss": 0.5904, + "step": 2688 + }, + { + "epoch": 2.2261181667586967, + "grad_norm": 0.9379095435142517, + "learning_rate": 3.5466243768949292e-06, + "loss": 0.605, + "step": 2689 + }, + { + "epoch": 2.2269464384318054, + "grad_norm": 0.898931622505188, + "learning_rate": 3.5456255951940393e-06, + "loss": 0.5978, + "step": 2690 + }, + { + "epoch": 2.2277747101049146, + "grad_norm": 0.9569118022918701, + "learning_rate": 3.5446266111743444e-06, + "loss": 0.6162, + "step": 2691 + }, + { + "epoch": 2.2286029817780233, + "grad_norm": 0.9167311191558838, + "learning_rate": 3.543627425029138e-06, + "loss": 0.5876, + "step": 2692 + }, + { + "epoch": 2.229431253451132, + "grad_norm": 0.9584367275238037, + "learning_rate": 3.542628036951753e-06, + "loss": 0.5704, + "step": 2693 + }, + { + "epoch": 2.2302595251242407, + "grad_norm": 0.8896269202232361, + "learning_rate": 3.541628447135562e-06, + "loss": 0.5806, + "step": 2694 + }, + { + "epoch": 2.2310877967973495, + "grad_norm": 0.9340798854827881, + "learning_rate": 3.5406286557739767e-06, + "loss": 0.6218, + "step": 2695 + }, + { + "epoch": 2.231916068470458, + "grad_norm": 0.9442514777183533, + "learning_rate": 3.539628663060447e-06, + "loss": 0.5899, + "step": 2696 + }, + { + "epoch": 2.232744340143567, + "grad_norm": 0.9123642444610596, + "learning_rate": 3.5386284691884633e-06, + "loss": 0.5825, + "step": 2697 + }, + { + "epoch": 2.233572611816676, + "grad_norm": 0.9076257348060608, + "learning_rate": 3.537628074351551e-06, + "loss": 0.6001, + "step": 2698 + }, + { + "epoch": 2.234400883489785, + "grad_norm": 0.9335387945175171, + "learning_rate": 3.5366274787432804e-06, + "loss": 0.6048, + "step": 2699 + }, + { + "epoch": 2.2352291551628936, + "grad_norm": 0.9016919732093811, + "learning_rate": 3.535626682557255e-06, + "loss": 0.6175, + "step": 2700 + }, + { + "epoch": 2.2360574268360023, + "grad_norm": 0.9110099673271179, + "learning_rate": 3.5346256859871207e-06, + "loss": 0.6027, + "step": 2701 + }, + { + "epoch": 2.236885698509111, + "grad_norm": 0.9513245224952698, + "learning_rate": 3.5336244892265616e-06, + "loss": 0.6015, + "step": 2702 + }, + { + "epoch": 2.2377139701822197, + "grad_norm": 0.8986294865608215, + "learning_rate": 3.532623092469299e-06, + "loss": 0.61, + "step": 2703 + }, + { + "epoch": 2.2385422418553285, + "grad_norm": 0.8991696834564209, + "learning_rate": 3.5316214959090933e-06, + "loss": 0.6013, + "step": 2704 + }, + { + "epoch": 2.239370513528437, + "grad_norm": 0.9667947292327881, + "learning_rate": 3.5306196997397453e-06, + "loss": 0.6125, + "step": 2705 + }, + { + "epoch": 2.240198785201546, + "grad_norm": 0.9183294177055359, + "learning_rate": 3.529617704155092e-06, + "loss": 0.5901, + "step": 2706 + }, + { + "epoch": 2.241027056874655, + "grad_norm": 0.9139719605445862, + "learning_rate": 3.5286155093490117e-06, + "loss": 0.5798, + "step": 2707 + }, + { + "epoch": 2.241855328547764, + "grad_norm": 0.9529849290847778, + "learning_rate": 3.5276131155154193e-06, + "loss": 0.5763, + "step": 2708 + }, + { + "epoch": 2.2426836002208725, + "grad_norm": 0.9126972556114197, + "learning_rate": 3.526610522848267e-06, + "loss": 0.5804, + "step": 2709 + }, + { + "epoch": 2.2435118718939813, + "grad_norm": 0.9101532101631165, + "learning_rate": 3.525607731541549e-06, + "loss": 0.5815, + "step": 2710 + }, + { + "epoch": 2.24434014356709, + "grad_norm": 0.9072697162628174, + "learning_rate": 3.524604741789296e-06, + "loss": 0.6238, + "step": 2711 + }, + { + "epoch": 2.2451684152401987, + "grad_norm": 0.9228153228759766, + "learning_rate": 3.5236015537855765e-06, + "loss": 0.6011, + "step": 2712 + }, + { + "epoch": 2.2459966869133074, + "grad_norm": 0.9280546307563782, + "learning_rate": 3.522598167724498e-06, + "loss": 0.6018, + "step": 2713 + }, + { + "epoch": 2.246824958586416, + "grad_norm": 0.9077439904212952, + "learning_rate": 3.5215945838002075e-06, + "loss": 0.6231, + "step": 2714 + }, + { + "epoch": 2.247653230259525, + "grad_norm": 0.9201393127441406, + "learning_rate": 3.520590802206887e-06, + "loss": 0.591, + "step": 2715 + }, + { + "epoch": 2.248481501932634, + "grad_norm": 0.8888259530067444, + "learning_rate": 3.519586823138761e-06, + "loss": 0.585, + "step": 2716 + }, + { + "epoch": 2.249309773605743, + "grad_norm": 0.918370246887207, + "learning_rate": 3.5185826467900886e-06, + "loss": 0.6151, + "step": 2717 + }, + { + "epoch": 2.2501380452788515, + "grad_norm": 0.9071224331855774, + "learning_rate": 3.517578273355169e-06, + "loss": 0.6053, + "step": 2718 + }, + { + "epoch": 2.2509663169519603, + "grad_norm": 0.9325213432312012, + "learning_rate": 3.5165737030283404e-06, + "loss": 0.5959, + "step": 2719 + }, + { + "epoch": 2.251794588625069, + "grad_norm": 0.9014924168586731, + "learning_rate": 3.515568936003976e-06, + "loss": 0.6092, + "step": 2720 + }, + { + "epoch": 2.2526228602981777, + "grad_norm": 0.923115611076355, + "learning_rate": 3.5145639724764886e-06, + "loss": 0.6031, + "step": 2721 + }, + { + "epoch": 2.2534511319712864, + "grad_norm": 0.9452555179595947, + "learning_rate": 3.513558812640331e-06, + "loss": 0.5957, + "step": 2722 + }, + { + "epoch": 2.2542794036443956, + "grad_norm": 0.960884153842926, + "learning_rate": 3.5125534566899915e-06, + "loss": 0.5862, + "step": 2723 + }, + { + "epoch": 2.2551076753175043, + "grad_norm": 0.9164205193519592, + "learning_rate": 3.5115479048199964e-06, + "loss": 0.5992, + "step": 2724 + }, + { + "epoch": 2.255935946990613, + "grad_norm": 0.943760097026825, + "learning_rate": 3.5105421572249123e-06, + "loss": 0.5842, + "step": 2725 + }, + { + "epoch": 2.256764218663722, + "grad_norm": 0.9578623175621033, + "learning_rate": 3.50953621409934e-06, + "loss": 0.6094, + "step": 2726 + }, + { + "epoch": 2.2575924903368305, + "grad_norm": 0.9307887554168701, + "learning_rate": 3.5085300756379216e-06, + "loss": 0.5905, + "step": 2727 + }, + { + "epoch": 2.2584207620099392, + "grad_norm": 0.9150220155715942, + "learning_rate": 3.5075237420353346e-06, + "loss": 0.6035, + "step": 2728 + }, + { + "epoch": 2.259249033683048, + "grad_norm": 0.8887060880661011, + "learning_rate": 3.506517213486296e-06, + "loss": 0.6032, + "step": 2729 + }, + { + "epoch": 2.2600773053561567, + "grad_norm": 0.9477519989013672, + "learning_rate": 3.5055104901855585e-06, + "loss": 0.576, + "step": 2730 + }, + { + "epoch": 2.2609055770292654, + "grad_norm": 0.9668344855308533, + "learning_rate": 3.504503572327915e-06, + "loss": 0.5973, + "step": 2731 + }, + { + "epoch": 2.2617338487023746, + "grad_norm": 0.905135989189148, + "learning_rate": 3.503496460108194e-06, + "loss": 0.5999, + "step": 2732 + }, + { + "epoch": 2.2625621203754833, + "grad_norm": 0.933996856212616, + "learning_rate": 3.5024891537212624e-06, + "loss": 0.5803, + "step": 2733 + }, + { + "epoch": 2.263390392048592, + "grad_norm": 0.9417551159858704, + "learning_rate": 3.501481653362025e-06, + "loss": 0.5918, + "step": 2734 + }, + { + "epoch": 2.2642186637217008, + "grad_norm": 0.9239657521247864, + "learning_rate": 3.500473959225422e-06, + "loss": 0.5828, + "step": 2735 + }, + { + "epoch": 2.2650469353948095, + "grad_norm": 0.9021532535552979, + "learning_rate": 3.499466071506436e-06, + "loss": 0.5958, + "step": 2736 + }, + { + "epoch": 2.2658752070679182, + "grad_norm": 0.9159536957740784, + "learning_rate": 3.498457990400081e-06, + "loss": 0.5867, + "step": 2737 + }, + { + "epoch": 2.266703478741027, + "grad_norm": 0.9347800612449646, + "learning_rate": 3.4974497161014124e-06, + "loss": 0.6058, + "step": 2738 + }, + { + "epoch": 2.2675317504141357, + "grad_norm": 0.9229647517204285, + "learning_rate": 3.496441248805522e-06, + "loss": 0.5714, + "step": 2739 + }, + { + "epoch": 2.2683600220872444, + "grad_norm": 0.9330629706382751, + "learning_rate": 3.4954325887075386e-06, + "loss": 0.5808, + "step": 2740 + }, + { + "epoch": 2.2691882937603536, + "grad_norm": 0.8958399891853333, + "learning_rate": 3.494423736002629e-06, + "loss": 0.5793, + "step": 2741 + }, + { + "epoch": 2.2700165654334623, + "grad_norm": 0.8952097296714783, + "learning_rate": 3.493414690885997e-06, + "loss": 0.6007, + "step": 2742 + }, + { + "epoch": 2.270844837106571, + "grad_norm": 0.9001036882400513, + "learning_rate": 3.4924054535528813e-06, + "loss": 0.5821, + "step": 2743 + }, + { + "epoch": 2.2716731087796798, + "grad_norm": 0.9016004204750061, + "learning_rate": 3.4913960241985616e-06, + "loss": 0.5819, + "step": 2744 + }, + { + "epoch": 2.2725013804527885, + "grad_norm": 0.9043833017349243, + "learning_rate": 3.490386403018354e-06, + "loss": 0.6144, + "step": 2745 + }, + { + "epoch": 2.273329652125897, + "grad_norm": 0.9396303296089172, + "learning_rate": 3.489376590207608e-06, + "loss": 0.602, + "step": 2746 + }, + { + "epoch": 2.274157923799006, + "grad_norm": 0.9685373306274414, + "learning_rate": 3.488366585961716e-06, + "loss": 0.6247, + "step": 2747 + }, + { + "epoch": 2.2749861954721147, + "grad_norm": 0.9028962254524231, + "learning_rate": 3.487356390476103e-06, + "loss": 0.5959, + "step": 2748 + }, + { + "epoch": 2.2758144671452234, + "grad_norm": 0.9150850772857666, + "learning_rate": 3.486346003946232e-06, + "loss": 0.5632, + "step": 2749 + }, + { + "epoch": 2.2766427388183326, + "grad_norm": 0.902158796787262, + "learning_rate": 3.4853354265676043e-06, + "loss": 0.5901, + "step": 2750 + }, + { + "epoch": 2.2774710104914413, + "grad_norm": 0.8942626118659973, + "learning_rate": 3.4843246585357565e-06, + "loss": 0.6082, + "step": 2751 + }, + { + "epoch": 2.27829928216455, + "grad_norm": 0.912050187587738, + "learning_rate": 3.4833137000462623e-06, + "loss": 0.5812, + "step": 2752 + }, + { + "epoch": 2.2791275538376587, + "grad_norm": 0.9262410998344421, + "learning_rate": 3.482302551294735e-06, + "loss": 0.6199, + "step": 2753 + }, + { + "epoch": 2.2799558255107675, + "grad_norm": 0.9281290173530579, + "learning_rate": 3.4812912124768193e-06, + "loss": 0.591, + "step": 2754 + }, + { + "epoch": 2.280784097183876, + "grad_norm": 0.8852386474609375, + "learning_rate": 3.4802796837882023e-06, + "loss": 0.6035, + "step": 2755 + }, + { + "epoch": 2.281612368856985, + "grad_norm": 0.8982499241828918, + "learning_rate": 3.479267965424605e-06, + "loss": 0.6023, + "step": 2756 + }, + { + "epoch": 2.282440640530094, + "grad_norm": 0.9022878408432007, + "learning_rate": 3.4782560575817837e-06, + "loss": 0.6004, + "step": 2757 + }, + { + "epoch": 2.283268912203203, + "grad_norm": 0.9256078600883484, + "learning_rate": 3.4772439604555353e-06, + "loss": 0.5811, + "step": 2758 + }, + { + "epoch": 2.2840971838763116, + "grad_norm": 0.9009407758712769, + "learning_rate": 3.4762316742416913e-06, + "loss": 0.6053, + "step": 2759 + }, + { + "epoch": 2.2849254555494203, + "grad_norm": 0.9257068037986755, + "learning_rate": 3.4752191991361174e-06, + "loss": 0.5664, + "step": 2760 + }, + { + "epoch": 2.285753727222529, + "grad_norm": 0.882075309753418, + "learning_rate": 3.4742065353347193e-06, + "loss": 0.6099, + "step": 2761 + }, + { + "epoch": 2.2865819988956377, + "grad_norm": 0.9553154706954956, + "learning_rate": 3.4731936830334384e-06, + "loss": 0.6072, + "step": 2762 + }, + { + "epoch": 2.2874102705687465, + "grad_norm": 0.9329802989959717, + "learning_rate": 3.4721806424282512e-06, + "loss": 0.601, + "step": 2763 + }, + { + "epoch": 2.288238542241855, + "grad_norm": 0.9146580696105957, + "learning_rate": 3.471167413715173e-06, + "loss": 0.599, + "step": 2764 + }, + { + "epoch": 2.289066813914964, + "grad_norm": 0.9170219302177429, + "learning_rate": 3.470153997090253e-06, + "loss": 0.5969, + "step": 2765 + }, + { + "epoch": 2.289895085588073, + "grad_norm": 0.8903328776359558, + "learning_rate": 3.469140392749578e-06, + "loss": 0.6063, + "step": 2766 + }, + { + "epoch": 2.290723357261182, + "grad_norm": 0.9068611264228821, + "learning_rate": 3.468126600889272e-06, + "loss": 0.5807, + "step": 2767 + }, + { + "epoch": 2.2915516289342905, + "grad_norm": 0.9464553594589233, + "learning_rate": 3.467112621705493e-06, + "loss": 0.5955, + "step": 2768 + }, + { + "epoch": 2.2923799006073993, + "grad_norm": 0.9302126169204712, + "learning_rate": 3.4660984553944367e-06, + "loss": 0.5956, + "step": 2769 + }, + { + "epoch": 2.293208172280508, + "grad_norm": 0.8797576427459717, + "learning_rate": 3.465084102152336e-06, + "loss": 0.5964, + "step": 2770 + }, + { + "epoch": 2.2940364439536167, + "grad_norm": 0.9544422030448914, + "learning_rate": 3.4640695621754573e-06, + "loss": 0.6225, + "step": 2771 + }, + { + "epoch": 2.2948647156267254, + "grad_norm": 0.919297993183136, + "learning_rate": 3.463054835660105e-06, + "loss": 0.5901, + "step": 2772 + }, + { + "epoch": 2.295692987299834, + "grad_norm": 0.9211829900741577, + "learning_rate": 3.4620399228026203e-06, + "loss": 0.5949, + "step": 2773 + }, + { + "epoch": 2.296521258972943, + "grad_norm": 0.929501473903656, + "learning_rate": 3.4610248237993775e-06, + "loss": 0.5654, + "step": 2774 + }, + { + "epoch": 2.297349530646052, + "grad_norm": 0.9604958891868591, + "learning_rate": 3.4600095388467907e-06, + "loss": 0.6059, + "step": 2775 + }, + { + "epoch": 2.298177802319161, + "grad_norm": 0.9042990207672119, + "learning_rate": 3.458994068141307e-06, + "loss": 0.5901, + "step": 2776 + }, + { + "epoch": 2.2990060739922695, + "grad_norm": 0.8958423733711243, + "learning_rate": 3.4579784118794107e-06, + "loss": 0.5938, + "step": 2777 + }, + { + "epoch": 2.2998343456653783, + "grad_norm": 0.9388688206672668, + "learning_rate": 3.456962570257622e-06, + "loss": 0.6024, + "step": 2778 + }, + { + "epoch": 2.300662617338487, + "grad_norm": 0.9006328582763672, + "learning_rate": 3.455946543472497e-06, + "loss": 0.5929, + "step": 2779 + }, + { + "epoch": 2.3014908890115957, + "grad_norm": 0.887873113155365, + "learning_rate": 3.454930331720626e-06, + "loss": 0.599, + "step": 2780 + }, + { + "epoch": 2.3023191606847044, + "grad_norm": 0.8919411301612854, + "learning_rate": 3.4539139351986385e-06, + "loss": 0.6173, + "step": 2781 + }, + { + "epoch": 2.3031474323578136, + "grad_norm": 0.9478135704994202, + "learning_rate": 3.452897354103196e-06, + "loss": 0.6025, + "step": 2782 + }, + { + "epoch": 2.3039757040309223, + "grad_norm": 0.9009284377098083, + "learning_rate": 3.451880588630999e-06, + "loss": 0.602, + "step": 2783 + }, + { + "epoch": 2.304803975704031, + "grad_norm": 0.9043265581130981, + "learning_rate": 3.4508636389787816e-06, + "loss": 0.6002, + "step": 2784 + }, + { + "epoch": 2.30563224737714, + "grad_norm": 0.9005764126777649, + "learning_rate": 3.4498465053433126e-06, + "loss": 0.618, + "step": 2785 + }, + { + "epoch": 2.3064605190502485, + "grad_norm": 0.9352818727493286, + "learning_rate": 3.4488291879214003e-06, + "loss": 0.6069, + "step": 2786 + }, + { + "epoch": 2.3072887907233572, + "grad_norm": 0.9092200994491577, + "learning_rate": 3.4478116869098854e-06, + "loss": 0.581, + "step": 2787 + }, + { + "epoch": 2.308117062396466, + "grad_norm": 0.9055072665214539, + "learning_rate": 3.4467940025056433e-06, + "loss": 0.598, + "step": 2788 + }, + { + "epoch": 2.3089453340695747, + "grad_norm": 1.0053009986877441, + "learning_rate": 3.4457761349055885e-06, + "loss": 0.5929, + "step": 2789 + }, + { + "epoch": 2.3097736057426834, + "grad_norm": 0.9241059422492981, + "learning_rate": 3.4447580843066675e-06, + "loss": 0.5962, + "step": 2790 + }, + { + "epoch": 2.3106018774157926, + "grad_norm": 0.927590548992157, + "learning_rate": 3.443739850905864e-06, + "loss": 0.5863, + "step": 2791 + }, + { + "epoch": 2.3114301490889013, + "grad_norm": 0.9227457046508789, + "learning_rate": 3.442721434900197e-06, + "loss": 0.5928, + "step": 2792 + }, + { + "epoch": 2.31225842076201, + "grad_norm": 0.9112377762794495, + "learning_rate": 3.4417028364867205e-06, + "loss": 0.6113, + "step": 2793 + }, + { + "epoch": 2.3130866924351188, + "grad_norm": 0.9256125092506409, + "learning_rate": 3.4406840558625237e-06, + "loss": 0.6163, + "step": 2794 + }, + { + "epoch": 2.3139149641082275, + "grad_norm": 0.933461606502533, + "learning_rate": 3.4396650932247303e-06, + "loss": 0.6159, + "step": 2795 + }, + { + "epoch": 2.3147432357813362, + "grad_norm": 0.9031276702880859, + "learning_rate": 3.4386459487705016e-06, + "loss": 0.5981, + "step": 2796 + }, + { + "epoch": 2.315571507454445, + "grad_norm": 0.9465809464454651, + "learning_rate": 3.437626622697031e-06, + "loss": 0.6029, + "step": 2797 + }, + { + "epoch": 2.3163997791275537, + "grad_norm": 0.9535313248634338, + "learning_rate": 3.43660711520155e-06, + "loss": 0.5998, + "step": 2798 + }, + { + "epoch": 2.3172280508006624, + "grad_norm": 0.9048331379890442, + "learning_rate": 3.4355874264813228e-06, + "loss": 0.5991, + "step": 2799 + }, + { + "epoch": 2.3180563224737716, + "grad_norm": 0.9119572639465332, + "learning_rate": 3.4345675567336506e-06, + "loss": 0.6133, + "step": 2800 + }, + { + "epoch": 2.3188845941468803, + "grad_norm": 0.9217002391815186, + "learning_rate": 3.4335475061558676e-06, + "loss": 0.6012, + "step": 2801 + }, + { + "epoch": 2.319712865819989, + "grad_norm": 0.9501274824142456, + "learning_rate": 3.432527274945344e-06, + "loss": 0.5874, + "step": 2802 + }, + { + "epoch": 2.3205411374930978, + "grad_norm": 0.9126407504081726, + "learning_rate": 3.4315068632994864e-06, + "loss": 0.5985, + "step": 2803 + }, + { + "epoch": 2.3213694091662065, + "grad_norm": 0.9414592981338501, + "learning_rate": 3.430486271415734e-06, + "loss": 0.5989, + "step": 2804 + }, + { + "epoch": 2.322197680839315, + "grad_norm": 0.9485170841217041, + "learning_rate": 3.4294654994915623e-06, + "loss": 0.5795, + "step": 2805 + }, + { + "epoch": 2.323025952512424, + "grad_norm": 0.9311578273773193, + "learning_rate": 3.4284445477244804e-06, + "loss": 0.5606, + "step": 2806 + }, + { + "epoch": 2.3238542241855327, + "grad_norm": 0.9476608037948608, + "learning_rate": 3.4274234163120328e-06, + "loss": 0.593, + "step": 2807 + }, + { + "epoch": 2.3246824958586414, + "grad_norm": 0.9367009401321411, + "learning_rate": 3.4264021054518004e-06, + "loss": 0.6104, + "step": 2808 + }, + { + "epoch": 2.3255107675317506, + "grad_norm": 0.9663116335868835, + "learning_rate": 3.425380615341396e-06, + "loss": 0.6091, + "step": 2809 + }, + { + "epoch": 2.3263390392048593, + "grad_norm": 0.9176538586616516, + "learning_rate": 3.4243589461784687e-06, + "loss": 0.585, + "step": 2810 + }, + { + "epoch": 2.327167310877968, + "grad_norm": 0.9261791110038757, + "learning_rate": 3.423337098160703e-06, + "loss": 0.6189, + "step": 2811 + }, + { + "epoch": 2.3279955825510767, + "grad_norm": 0.9573442339897156, + "learning_rate": 3.422315071485815e-06, + "loss": 0.5925, + "step": 2812 + }, + { + "epoch": 2.3288238542241855, + "grad_norm": 0.9381136894226074, + "learning_rate": 3.4212928663515584e-06, + "loss": 0.588, + "step": 2813 + }, + { + "epoch": 2.329652125897294, + "grad_norm": 0.9276189804077148, + "learning_rate": 3.420270482955721e-06, + "loss": 0.5885, + "step": 2814 + }, + { + "epoch": 2.330480397570403, + "grad_norm": 0.9124462604522705, + "learning_rate": 3.419247921496123e-06, + "loss": 0.6004, + "step": 2815 + }, + { + "epoch": 2.331308669243512, + "grad_norm": 0.9146862626075745, + "learning_rate": 3.418225182170622e-06, + "loss": 0.6036, + "step": 2816 + }, + { + "epoch": 2.332136940916621, + "grad_norm": 0.9196732044219971, + "learning_rate": 3.4172022651771075e-06, + "loss": 0.6054, + "step": 2817 + }, + { + "epoch": 2.3329652125897296, + "grad_norm": 0.9340062737464905, + "learning_rate": 3.416179170713505e-06, + "loss": 0.6037, + "step": 2818 + }, + { + "epoch": 2.3337934842628383, + "grad_norm": 0.9004904627799988, + "learning_rate": 3.415155898977773e-06, + "loss": 0.6087, + "step": 2819 + }, + { + "epoch": 2.334621755935947, + "grad_norm": 0.8993189334869385, + "learning_rate": 3.414132450167906e-06, + "loss": 0.5972, + "step": 2820 + }, + { + "epoch": 2.3354500276090557, + "grad_norm": 0.9195947051048279, + "learning_rate": 3.413108824481931e-06, + "loss": 0.5784, + "step": 2821 + }, + { + "epoch": 2.3362782992821645, + "grad_norm": 0.9149767756462097, + "learning_rate": 3.412085022117911e-06, + "loss": 0.5838, + "step": 2822 + }, + { + "epoch": 2.337106570955273, + "grad_norm": 0.8880875110626221, + "learning_rate": 3.411061043273941e-06, + "loss": 0.6008, + "step": 2823 + }, + { + "epoch": 2.337934842628382, + "grad_norm": 0.9110493659973145, + "learning_rate": 3.4100368881481515e-06, + "loss": 0.6333, + "step": 2824 + }, + { + "epoch": 2.338763114301491, + "grad_norm": 0.9558914303779602, + "learning_rate": 3.409012556938708e-06, + "loss": 0.602, + "step": 2825 + }, + { + "epoch": 2.3395913859746, + "grad_norm": 0.9668992161750793, + "learning_rate": 3.4079880498438083e-06, + "loss": 0.5952, + "step": 2826 + }, + { + "epoch": 2.3404196576477085, + "grad_norm": 0.9352533221244812, + "learning_rate": 3.406963367061685e-06, + "loss": 0.6093, + "step": 2827 + }, + { + "epoch": 2.3412479293208173, + "grad_norm": 0.9072561264038086, + "learning_rate": 3.4059385087906055e-06, + "loss": 0.603, + "step": 2828 + }, + { + "epoch": 2.342076200993926, + "grad_norm": 0.9563443064689636, + "learning_rate": 3.4049134752288692e-06, + "loss": 0.585, + "step": 2829 + }, + { + "epoch": 2.3429044726670347, + "grad_norm": 0.9061548709869385, + "learning_rate": 3.4038882665748102e-06, + "loss": 0.6111, + "step": 2830 + }, + { + "epoch": 2.3437327443401434, + "grad_norm": 0.9177330136299133, + "learning_rate": 3.4028628830267986e-06, + "loss": 0.5872, + "step": 2831 + }, + { + "epoch": 2.344561016013252, + "grad_norm": 0.9479212164878845, + "learning_rate": 3.401837324783235e-06, + "loss": 0.5911, + "step": 2832 + }, + { + "epoch": 2.345389287686361, + "grad_norm": 0.9118002653121948, + "learning_rate": 3.4008115920425556e-06, + "loss": 0.6153, + "step": 2833 + }, + { + "epoch": 2.34621755935947, + "grad_norm": 0.9086247682571411, + "learning_rate": 3.399785685003231e-06, + "loss": 0.5913, + "step": 2834 + }, + { + "epoch": 2.347045831032579, + "grad_norm": 0.937494158744812, + "learning_rate": 3.398759603863763e-06, + "loss": 0.601, + "step": 2835 + }, + { + "epoch": 2.3478741027056875, + "grad_norm": 0.9359503388404846, + "learning_rate": 3.397733348822691e-06, + "loss": 0.5894, + "step": 2836 + }, + { + "epoch": 2.3487023743787963, + "grad_norm": 0.9384423494338989, + "learning_rate": 3.396706920078584e-06, + "loss": 0.5974, + "step": 2837 + }, + { + "epoch": 2.349530646051905, + "grad_norm": 0.9438448548316956, + "learning_rate": 3.3956803178300474e-06, + "loss": 0.6079, + "step": 2838 + }, + { + "epoch": 2.3503589177250137, + "grad_norm": 0.9250973463058472, + "learning_rate": 3.3946535422757187e-06, + "loss": 0.6011, + "step": 2839 + }, + { + "epoch": 2.3511871893981224, + "grad_norm": 0.9577994346618652, + "learning_rate": 3.3936265936142685e-06, + "loss": 0.6052, + "step": 2840 + }, + { + "epoch": 2.3520154610712316, + "grad_norm": 0.9307487607002258, + "learning_rate": 3.3925994720444033e-06, + "loss": 0.5925, + "step": 2841 + }, + { + "epoch": 2.3528437327443403, + "grad_norm": 0.9049618244171143, + "learning_rate": 3.3915721777648615e-06, + "loss": 0.6096, + "step": 2842 + }, + { + "epoch": 2.353672004417449, + "grad_norm": 0.928248405456543, + "learning_rate": 3.3905447109744137e-06, + "loss": 0.5938, + "step": 2843 + }, + { + "epoch": 2.354500276090558, + "grad_norm": 0.9107298254966736, + "learning_rate": 3.3895170718718664e-06, + "loss": 0.5947, + "step": 2844 + }, + { + "epoch": 2.3553285477636665, + "grad_norm": 0.8896966576576233, + "learning_rate": 3.388489260656057e-06, + "loss": 0.5827, + "step": 2845 + }, + { + "epoch": 2.3561568194367752, + "grad_norm": 0.9374940991401672, + "learning_rate": 3.3874612775258586e-06, + "loss": 0.6038, + "step": 2846 + }, + { + "epoch": 2.356985091109884, + "grad_norm": 0.9086610078811646, + "learning_rate": 3.3864331226801755e-06, + "loss": 0.5758, + "step": 2847 + }, + { + "epoch": 2.3578133627829927, + "grad_norm": 0.9486793875694275, + "learning_rate": 3.3854047963179464e-06, + "loss": 0.5656, + "step": 2848 + }, + { + "epoch": 2.3586416344561014, + "grad_norm": 0.9411890506744385, + "learning_rate": 3.3843762986381436e-06, + "loss": 0.605, + "step": 2849 + }, + { + "epoch": 2.3594699061292106, + "grad_norm": 0.9491676688194275, + "learning_rate": 3.3833476298397706e-06, + "loss": 0.5992, + "step": 2850 + }, + { + "epoch": 2.3602981778023193, + "grad_norm": 0.9281409382820129, + "learning_rate": 3.382318790121865e-06, + "loss": 0.5856, + "step": 2851 + }, + { + "epoch": 2.361126449475428, + "grad_norm": 0.9243955016136169, + "learning_rate": 3.381289779683499e-06, + "loss": 0.5966, + "step": 2852 + }, + { + "epoch": 2.3619547211485368, + "grad_norm": 0.8931578993797302, + "learning_rate": 3.3802605987237757e-06, + "loss": 0.6095, + "step": 2853 + }, + { + "epoch": 2.3627829928216455, + "grad_norm": 0.9200635552406311, + "learning_rate": 3.3792312474418324e-06, + "loss": 0.6014, + "step": 2854 + }, + { + "epoch": 2.3636112644947542, + "grad_norm": 0.9007267951965332, + "learning_rate": 3.378201726036839e-06, + "loss": 0.5964, + "step": 2855 + }, + { + "epoch": 2.364439536167863, + "grad_norm": 0.9420115947723389, + "learning_rate": 3.377172034707998e-06, + "loss": 0.6075, + "step": 2856 + }, + { + "epoch": 2.3652678078409717, + "grad_norm": 0.8935693502426147, + "learning_rate": 3.3761421736545453e-06, + "loss": 0.6184, + "step": 2857 + }, + { + "epoch": 2.3660960795140804, + "grad_norm": 0.9142755270004272, + "learning_rate": 3.3751121430757482e-06, + "loss": 0.6114, + "step": 2858 + }, + { + "epoch": 2.3669243511871896, + "grad_norm": 0.9280282855033875, + "learning_rate": 3.3740819431709103e-06, + "loss": 0.5922, + "step": 2859 + }, + { + "epoch": 2.3677526228602983, + "grad_norm": 0.9347119927406311, + "learning_rate": 3.3730515741393643e-06, + "loss": 0.5733, + "step": 2860 + }, + { + "epoch": 2.368580894533407, + "grad_norm": 0.9286894202232361, + "learning_rate": 3.3720210361804775e-06, + "loss": 0.597, + "step": 2861 + }, + { + "epoch": 2.3694091662065158, + "grad_norm": 0.9092636704444885, + "learning_rate": 3.3709903294936482e-06, + "loss": 0.5842, + "step": 2862 + }, + { + "epoch": 2.3702374378796245, + "grad_norm": 0.9205367565155029, + "learning_rate": 3.369959454278309e-06, + "loss": 0.6166, + "step": 2863 + }, + { + "epoch": 2.371065709552733, + "grad_norm": 0.9211122393608093, + "learning_rate": 3.3689284107339265e-06, + "loss": 0.5746, + "step": 2864 + }, + { + "epoch": 2.371893981225842, + "grad_norm": 0.908984899520874, + "learning_rate": 3.367897199059996e-06, + "loss": 0.5883, + "step": 2865 + }, + { + "epoch": 2.3727222528989507, + "grad_norm": 0.9270497560501099, + "learning_rate": 3.3668658194560477e-06, + "loss": 0.5985, + "step": 2866 + }, + { + "epoch": 2.3735505245720594, + "grad_norm": 0.9320128560066223, + "learning_rate": 3.3658342721216448e-06, + "loss": 0.5907, + "step": 2867 + }, + { + "epoch": 2.3743787962451686, + "grad_norm": 0.9428799152374268, + "learning_rate": 3.364802557256381e-06, + "loss": 0.6041, + "step": 2868 + }, + { + "epoch": 2.3752070679182773, + "grad_norm": 0.915873646736145, + "learning_rate": 3.363770675059883e-06, + "loss": 0.615, + "step": 2869 + }, + { + "epoch": 2.376035339591386, + "grad_norm": 0.9327977895736694, + "learning_rate": 3.362738625731813e-06, + "loss": 0.6147, + "step": 2870 + }, + { + "epoch": 2.3768636112644947, + "grad_norm": 0.9171217679977417, + "learning_rate": 3.3617064094718602e-06, + "loss": 0.6057, + "step": 2871 + }, + { + "epoch": 2.3776918829376035, + "grad_norm": 0.9192221760749817, + "learning_rate": 3.3606740264797504e-06, + "loss": 0.6092, + "step": 2872 + }, + { + "epoch": 2.378520154610712, + "grad_norm": 0.9024589657783508, + "learning_rate": 3.3596414769552394e-06, + "loss": 0.6047, + "step": 2873 + }, + { + "epoch": 2.379348426283821, + "grad_norm": 0.910923182964325, + "learning_rate": 3.3586087610981155e-06, + "loss": 0.5741, + "step": 2874 + }, + { + "epoch": 2.38017669795693, + "grad_norm": 0.929965078830719, + "learning_rate": 3.3575758791082014e-06, + "loss": 0.6114, + "step": 2875 + }, + { + "epoch": 2.381004969630039, + "grad_norm": 0.910308837890625, + "learning_rate": 3.356542831185348e-06, + "loss": 0.5938, + "step": 2876 + }, + { + "epoch": 2.3818332413031476, + "grad_norm": 0.9371251463890076, + "learning_rate": 3.355509617529441e-06, + "loss": 0.5883, + "step": 2877 + }, + { + "epoch": 2.3826615129762563, + "grad_norm": 0.9064118266105652, + "learning_rate": 3.3544762383403995e-06, + "loss": 0.5879, + "step": 2878 + }, + { + "epoch": 2.383489784649365, + "grad_norm": 0.954288899898529, + "learning_rate": 3.3534426938181704e-06, + "loss": 0.5981, + "step": 2879 + }, + { + "epoch": 2.3843180563224737, + "grad_norm": 0.9465821981430054, + "learning_rate": 3.3524089841627348e-06, + "loss": 0.6129, + "step": 2880 + }, + { + "epoch": 2.3851463279955825, + "grad_norm": 0.9406200051307678, + "learning_rate": 3.3513751095741084e-06, + "loss": 0.6035, + "step": 2881 + }, + { + "epoch": 2.385974599668691, + "grad_norm": 0.9627492427825928, + "learning_rate": 3.3503410702523337e-06, + "loss": 0.5701, + "step": 2882 + }, + { + "epoch": 2.3868028713418, + "grad_norm": 0.8899156451225281, + "learning_rate": 3.3493068663974892e-06, + "loss": 0.6055, + "step": 2883 + }, + { + "epoch": 2.387631143014909, + "grad_norm": 0.9304690957069397, + "learning_rate": 3.348272498209683e-06, + "loss": 0.6078, + "step": 2884 + }, + { + "epoch": 2.388459414688018, + "grad_norm": 0.9416152834892273, + "learning_rate": 3.3472379658890554e-06, + "loss": 0.609, + "step": 2885 + }, + { + "epoch": 2.3892876863611265, + "grad_norm": 0.9296873211860657, + "learning_rate": 3.34620326963578e-06, + "loss": 0.608, + "step": 2886 + }, + { + "epoch": 2.3901159580342353, + "grad_norm": 0.9163168668746948, + "learning_rate": 3.34516840965006e-06, + "loss": 0.5725, + "step": 2887 + }, + { + "epoch": 2.390944229707344, + "grad_norm": 0.9371008276939392, + "learning_rate": 3.344133386132131e-06, + "loss": 0.6223, + "step": 2888 + }, + { + "epoch": 2.3917725013804527, + "grad_norm": 0.9486922025680542, + "learning_rate": 3.3430981992822608e-06, + "loss": 0.5878, + "step": 2889 + }, + { + "epoch": 2.3926007730535614, + "grad_norm": 0.9124024510383606, + "learning_rate": 3.3420628493007483e-06, + "loss": 0.5816, + "step": 2890 + }, + { + "epoch": 2.39342904472667, + "grad_norm": 0.9193150401115417, + "learning_rate": 3.3410273363879236e-06, + "loss": 0.5909, + "step": 2891 + }, + { + "epoch": 2.394257316399779, + "grad_norm": 0.9287682175636292, + "learning_rate": 3.3399916607441502e-06, + "loss": 0.6095, + "step": 2892 + }, + { + "epoch": 2.395085588072888, + "grad_norm": 0.9130192995071411, + "learning_rate": 3.3389558225698204e-06, + "loss": 0.5782, + "step": 2893 + }, + { + "epoch": 2.395913859745997, + "grad_norm": 0.9145562052726746, + "learning_rate": 3.3379198220653587e-06, + "loss": 0.5919, + "step": 2894 + }, + { + "epoch": 2.3967421314191055, + "grad_norm": 0.9094864726066589, + "learning_rate": 3.3368836594312235e-06, + "loss": 0.6165, + "step": 2895 + }, + { + "epoch": 2.3975704030922143, + "grad_norm": 0.9446457028388977, + "learning_rate": 3.3358473348679016e-06, + "loss": 0.5967, + "step": 2896 + }, + { + "epoch": 2.398398674765323, + "grad_norm": 0.9345584511756897, + "learning_rate": 3.334810848575911e-06, + "loss": 0.6006, + "step": 2897 + }, + { + "epoch": 2.3992269464384317, + "grad_norm": 0.8865429759025574, + "learning_rate": 3.3337742007558043e-06, + "loss": 0.5766, + "step": 2898 + }, + { + "epoch": 2.4000552181115404, + "grad_norm": 0.9188973903656006, + "learning_rate": 3.3327373916081617e-06, + "loss": 0.6107, + "step": 2899 + }, + { + "epoch": 2.4008834897846496, + "grad_norm": 0.895247220993042, + "learning_rate": 3.3317004213335967e-06, + "loss": 0.6037, + "step": 2900 + }, + { + "epoch": 2.4017117614577583, + "grad_norm": 0.9131169319152832, + "learning_rate": 3.330663290132753e-06, + "loss": 0.5825, + "step": 2901 + }, + { + "epoch": 2.402540033130867, + "grad_norm": 0.9109821915626526, + "learning_rate": 3.329625998206306e-06, + "loss": 0.6194, + "step": 2902 + }, + { + "epoch": 2.403368304803976, + "grad_norm": 0.90902179479599, + "learning_rate": 3.3285885457549618e-06, + "loss": 0.5921, + "step": 2903 + }, + { + "epoch": 2.4041965764770845, + "grad_norm": 0.9419400095939636, + "learning_rate": 3.3275509329794585e-06, + "loss": 0.5973, + "step": 2904 + }, + { + "epoch": 2.4050248481501932, + "grad_norm": 0.905164897441864, + "learning_rate": 3.3265131600805636e-06, + "loss": 0.5828, + "step": 2905 + }, + { + "epoch": 2.405853119823302, + "grad_norm": 0.9232257008552551, + "learning_rate": 3.3254752272590775e-06, + "loss": 0.5972, + "step": 2906 + }, + { + "epoch": 2.4066813914964107, + "grad_norm": 0.9152960181236267, + "learning_rate": 3.3244371347158294e-06, + "loss": 0.5968, + "step": 2907 + }, + { + "epoch": 2.4075096631695194, + "grad_norm": 0.9337223172187805, + "learning_rate": 3.3233988826516807e-06, + "loss": 0.5907, + "step": 2908 + }, + { + "epoch": 2.4083379348426286, + "grad_norm": 0.9352282881736755, + "learning_rate": 3.322360471267524e-06, + "loss": 0.573, + "step": 2909 + }, + { + "epoch": 2.4091662065157373, + "grad_norm": 0.928322970867157, + "learning_rate": 3.321321900764282e-06, + "loss": 0.5836, + "step": 2910 + }, + { + "epoch": 2.409994478188846, + "grad_norm": 0.9335048794746399, + "learning_rate": 3.3202831713429084e-06, + "loss": 0.607, + "step": 2911 + }, + { + "epoch": 2.4108227498619548, + "grad_norm": 0.9128180742263794, + "learning_rate": 3.319244283204389e-06, + "loss": 0.5956, + "step": 2912 + }, + { + "epoch": 2.4116510215350635, + "grad_norm": 0.9457159042358398, + "learning_rate": 3.3182052365497364e-06, + "loss": 0.602, + "step": 2913 + }, + { + "epoch": 2.4124792932081722, + "grad_norm": 0.9508596658706665, + "learning_rate": 3.317166031579998e-06, + "loss": 0.5972, + "step": 2914 + }, + { + "epoch": 2.413307564881281, + "grad_norm": 0.9274892807006836, + "learning_rate": 3.31612666849625e-06, + "loss": 0.6015, + "step": 2915 + }, + { + "epoch": 2.4141358365543897, + "grad_norm": 0.9340214729309082, + "learning_rate": 3.3150871474995994e-06, + "loss": 0.5984, + "step": 2916 + }, + { + "epoch": 2.4149641082274984, + "grad_norm": 0.9376348853111267, + "learning_rate": 3.3140474687911844e-06, + "loss": 0.6205, + "step": 2917 + }, + { + "epoch": 2.4157923799006076, + "grad_norm": 0.9363590478897095, + "learning_rate": 3.313007632572173e-06, + "loss": 0.5926, + "step": 2918 + }, + { + "epoch": 2.4166206515737163, + "grad_norm": 0.8942010402679443, + "learning_rate": 3.3119676390437618e-06, + "loss": 0.5966, + "step": 2919 + }, + { + "epoch": 2.417448923246825, + "grad_norm": 0.9249299168586731, + "learning_rate": 3.310927488407183e-06, + "loss": 0.5951, + "step": 2920 + }, + { + "epoch": 2.4182771949199338, + "grad_norm": 0.9157326817512512, + "learning_rate": 3.3098871808636936e-06, + "loss": 0.5693, + "step": 2921 + }, + { + "epoch": 2.4191054665930425, + "grad_norm": 0.9088039994239807, + "learning_rate": 3.3088467166145843e-06, + "loss": 0.6084, + "step": 2922 + }, + { + "epoch": 2.419933738266151, + "grad_norm": 0.9258041381835938, + "learning_rate": 3.307806095861177e-06, + "loss": 0.6036, + "step": 2923 + }, + { + "epoch": 2.42076200993926, + "grad_norm": 0.905963659286499, + "learning_rate": 3.306765318804819e-06, + "loss": 0.5918, + "step": 2924 + }, + { + "epoch": 2.4215902816123687, + "grad_norm": 0.9063541293144226, + "learning_rate": 3.3057243856468924e-06, + "loss": 0.5906, + "step": 2925 + }, + { + "epoch": 2.4224185532854774, + "grad_norm": 0.9308085441589355, + "learning_rate": 3.304683296588808e-06, + "loss": 0.59, + "step": 2926 + }, + { + "epoch": 2.4232468249585866, + "grad_norm": 0.8989183902740479, + "learning_rate": 3.303642051832007e-06, + "loss": 0.5915, + "step": 2927 + }, + { + "epoch": 2.4240750966316953, + "grad_norm": 0.9413655996322632, + "learning_rate": 3.3026006515779614e-06, + "loss": 0.5952, + "step": 2928 + }, + { + "epoch": 2.424903368304804, + "grad_norm": 0.9431564211845398, + "learning_rate": 3.3015590960281708e-06, + "loss": 0.6007, + "step": 2929 + }, + { + "epoch": 2.4257316399779127, + "grad_norm": 0.8911551237106323, + "learning_rate": 3.3005173853841666e-06, + "loss": 0.573, + "step": 2930 + }, + { + "epoch": 2.4265599116510215, + "grad_norm": 0.9177538156509399, + "learning_rate": 3.299475519847511e-06, + "loss": 0.6116, + "step": 2931 + }, + { + "epoch": 2.42738818332413, + "grad_norm": 0.9592459797859192, + "learning_rate": 3.298433499619795e-06, + "loss": 0.6039, + "step": 2932 + }, + { + "epoch": 2.428216454997239, + "grad_norm": 0.9287276864051819, + "learning_rate": 3.2973913249026392e-06, + "loss": 0.5837, + "step": 2933 + }, + { + "epoch": 2.429044726670348, + "grad_norm": 0.9205804467201233, + "learning_rate": 3.296348995897696e-06, + "loss": 0.5904, + "step": 2934 + }, + { + "epoch": 2.429872998343457, + "grad_norm": 0.9075837731361389, + "learning_rate": 3.295306512806646e-06, + "loss": 0.5719, + "step": 2935 + }, + { + "epoch": 2.4307012700165656, + "grad_norm": 0.9203813672065735, + "learning_rate": 3.294263875831198e-06, + "loss": 0.604, + "step": 2936 + }, + { + "epoch": 2.4315295416896743, + "grad_norm": 0.8868334293365479, + "learning_rate": 3.2932210851730952e-06, + "loss": 0.6113, + "step": 2937 + }, + { + "epoch": 2.432357813362783, + "grad_norm": 0.9620448350906372, + "learning_rate": 3.2921781410341064e-06, + "loss": 0.5876, + "step": 2938 + }, + { + "epoch": 2.4331860850358917, + "grad_norm": 0.8991201519966125, + "learning_rate": 3.2911350436160317e-06, + "loss": 0.5994, + "step": 2939 + }, + { + "epoch": 2.4340143567090005, + "grad_norm": 0.9383120536804199, + "learning_rate": 3.2900917931207016e-06, + "loss": 0.6098, + "step": 2940 + }, + { + "epoch": 2.434842628382109, + "grad_norm": 0.9405770897865295, + "learning_rate": 3.289048389749974e-06, + "loss": 0.5978, + "step": 2941 + }, + { + "epoch": 2.435670900055218, + "grad_norm": 0.9153317213058472, + "learning_rate": 3.2880048337057387e-06, + "loss": 0.6091, + "step": 2942 + }, + { + "epoch": 2.436499171728327, + "grad_norm": 0.9396141171455383, + "learning_rate": 3.2869611251899136e-06, + "loss": 0.6143, + "step": 2943 + }, + { + "epoch": 2.437327443401436, + "grad_norm": 0.9370556473731995, + "learning_rate": 3.285917264404446e-06, + "loss": 0.61, + "step": 2944 + }, + { + "epoch": 2.4381557150745445, + "grad_norm": 0.9142118096351624, + "learning_rate": 3.2848732515513153e-06, + "loss": 0.5875, + "step": 2945 + }, + { + "epoch": 2.4389839867476533, + "grad_norm": 0.9196925759315491, + "learning_rate": 3.2838290868325267e-06, + "loss": 0.6058, + "step": 2946 + }, + { + "epoch": 2.439812258420762, + "grad_norm": 0.892217755317688, + "learning_rate": 3.282784770450116e-06, + "loss": 0.6005, + "step": 2947 + }, + { + "epoch": 2.4406405300938707, + "grad_norm": 0.9278191924095154, + "learning_rate": 3.2817403026061496e-06, + "loss": 0.6206, + "step": 2948 + }, + { + "epoch": 2.4414688017669794, + "grad_norm": 0.9343185424804688, + "learning_rate": 3.2806956835027216e-06, + "loss": 0.604, + "step": 2949 + }, + { + "epoch": 2.442297073440088, + "grad_norm": 0.9401823878288269, + "learning_rate": 3.2796509133419564e-06, + "loss": 0.6039, + "step": 2950 + }, + { + "epoch": 2.443125345113197, + "grad_norm": 0.9102358818054199, + "learning_rate": 3.278605992326008e-06, + "loss": 0.5892, + "step": 2951 + }, + { + "epoch": 2.443953616786306, + "grad_norm": 0.9241907000541687, + "learning_rate": 3.2775609206570573e-06, + "loss": 0.6054, + "step": 2952 + }, + { + "epoch": 2.444781888459415, + "grad_norm": 0.9492688775062561, + "learning_rate": 3.276515698537317e-06, + "loss": 0.5922, + "step": 2953 + }, + { + "epoch": 2.4456101601325235, + "grad_norm": 0.9492383003234863, + "learning_rate": 3.275470326169028e-06, + "loss": 0.5735, + "step": 2954 + }, + { + "epoch": 2.4464384318056323, + "grad_norm": 0.8857248425483704, + "learning_rate": 3.2744248037544597e-06, + "loss": 0.584, + "step": 2955 + }, + { + "epoch": 2.447266703478741, + "grad_norm": 0.9185032248497009, + "learning_rate": 3.2733791314959108e-06, + "loss": 0.5906, + "step": 2956 + }, + { + "epoch": 2.4480949751518497, + "grad_norm": 0.9257567524909973, + "learning_rate": 3.2723333095957094e-06, + "loss": 0.587, + "step": 2957 + }, + { + "epoch": 2.4489232468249584, + "grad_norm": 0.9427464008331299, + "learning_rate": 3.2712873382562117e-06, + "loss": 0.5932, + "step": 2958 + }, + { + "epoch": 2.4497515184980676, + "grad_norm": 0.9002125263214111, + "learning_rate": 3.2702412176798045e-06, + "loss": 0.58, + "step": 2959 + }, + { + "epoch": 2.4505797901711763, + "grad_norm": 0.906929075717926, + "learning_rate": 3.2691949480689018e-06, + "loss": 0.5878, + "step": 2960 + }, + { + "epoch": 2.451408061844285, + "grad_norm": 0.9162207841873169, + "learning_rate": 3.268148529625946e-06, + "loss": 0.5975, + "step": 2961 + }, + { + "epoch": 2.452236333517394, + "grad_norm": 0.9268636107444763, + "learning_rate": 3.2671019625534117e-06, + "loss": 0.6028, + "step": 2962 + }, + { + "epoch": 2.4530646051905025, + "grad_norm": 0.9216850996017456, + "learning_rate": 3.2660552470537975e-06, + "loss": 0.5938, + "step": 2963 + }, + { + "epoch": 2.4538928768636112, + "grad_norm": 0.9353698492050171, + "learning_rate": 3.2650083833296344e-06, + "loss": 0.6069, + "step": 2964 + }, + { + "epoch": 2.45472114853672, + "grad_norm": 0.9393898248672485, + "learning_rate": 3.2639613715834805e-06, + "loss": 0.5874, + "step": 2965 + }, + { + "epoch": 2.4555494202098287, + "grad_norm": 0.9410581588745117, + "learning_rate": 3.2629142120179224e-06, + "loss": 0.5918, + "step": 2966 + }, + { + "epoch": 2.4563776918829374, + "grad_norm": 0.9445161819458008, + "learning_rate": 3.2618669048355763e-06, + "loss": 0.5743, + "step": 2967 + }, + { + "epoch": 2.4572059635560466, + "grad_norm": 0.9591469764709473, + "learning_rate": 3.2608194502390867e-06, + "loss": 0.5781, + "step": 2968 + }, + { + "epoch": 2.4580342352291553, + "grad_norm": 0.9237606525421143, + "learning_rate": 3.2597718484311246e-06, + "loss": 0.6163, + "step": 2969 + }, + { + "epoch": 2.458862506902264, + "grad_norm": 0.9216769337654114, + "learning_rate": 3.2587240996143933e-06, + "loss": 0.6157, + "step": 2970 + }, + { + "epoch": 2.4596907785753728, + "grad_norm": 0.9081043004989624, + "learning_rate": 3.257676203991621e-06, + "loss": 0.5821, + "step": 2971 + }, + { + "epoch": 2.4605190502484815, + "grad_norm": 0.9316040873527527, + "learning_rate": 3.256628161765566e-06, + "loss": 0.5911, + "step": 2972 + }, + { + "epoch": 2.4613473219215902, + "grad_norm": 0.9326239824295044, + "learning_rate": 3.2555799731390158e-06, + "loss": 0.5963, + "step": 2973 + }, + { + "epoch": 2.462175593594699, + "grad_norm": 0.9429934024810791, + "learning_rate": 3.254531638314784e-06, + "loss": 0.5651, + "step": 2974 + }, + { + "epoch": 2.4630038652678077, + "grad_norm": 0.9090480804443359, + "learning_rate": 3.2534831574957134e-06, + "loss": 0.5779, + "step": 2975 + }, + { + "epoch": 2.4638321369409164, + "grad_norm": 0.9582802653312683, + "learning_rate": 3.2524345308846756e-06, + "loss": 0.5887, + "step": 2976 + }, + { + "epoch": 2.4646604086140256, + "grad_norm": 0.9248860478401184, + "learning_rate": 3.2513857586845705e-06, + "loss": 0.5936, + "step": 2977 + }, + { + "epoch": 2.4654886802871343, + "grad_norm": 0.9345134496688843, + "learning_rate": 3.250336841098325e-06, + "loss": 0.5985, + "step": 2978 + }, + { + "epoch": 2.466316951960243, + "grad_norm": 0.9092442393302917, + "learning_rate": 3.249287778328896e-06, + "loss": 0.5926, + "step": 2979 + }, + { + "epoch": 2.4671452236333518, + "grad_norm": 0.932470977306366, + "learning_rate": 3.2482385705792655e-06, + "loss": 0.5797, + "step": 2980 + }, + { + "epoch": 2.4679734953064605, + "grad_norm": 0.9335013031959534, + "learning_rate": 3.247189218052448e-06, + "loss": 0.5933, + "step": 2981 + }, + { + "epoch": 2.468801766979569, + "grad_norm": 0.9283298254013062, + "learning_rate": 3.2461397209514812e-06, + "loss": 0.6003, + "step": 2982 + }, + { + "epoch": 2.469630038652678, + "grad_norm": 0.9429066181182861, + "learning_rate": 3.245090079479434e-06, + "loss": 0.5944, + "step": 2983 + }, + { + "epoch": 2.4704583103257867, + "grad_norm": 0.9284510612487793, + "learning_rate": 3.2440402938394017e-06, + "loss": 0.5782, + "step": 2984 + }, + { + "epoch": 2.4712865819988954, + "grad_norm": 0.9150790572166443, + "learning_rate": 3.2429903642345094e-06, + "loss": 0.5862, + "step": 2985 + }, + { + "epoch": 2.4721148536720046, + "grad_norm": 0.9177762866020203, + "learning_rate": 3.2419402908679065e-06, + "loss": 0.5931, + "step": 2986 + }, + { + "epoch": 2.4729431253451133, + "grad_norm": 0.9098034501075745, + "learning_rate": 3.2408900739427744e-06, + "loss": 0.5864, + "step": 2987 + }, + { + "epoch": 2.473771397018222, + "grad_norm": 0.9237478971481323, + "learning_rate": 3.2398397136623195e-06, + "loss": 0.5873, + "step": 2988 + }, + { + "epoch": 2.4745996686913307, + "grad_norm": 0.9212510585784912, + "learning_rate": 3.2387892102297757e-06, + "loss": 0.5884, + "step": 2989 + }, + { + "epoch": 2.4754279403644395, + "grad_norm": 0.8987523913383484, + "learning_rate": 3.237738563848408e-06, + "loss": 0.5806, + "step": 2990 + }, + { + "epoch": 2.476256212037548, + "grad_norm": 0.9036835432052612, + "learning_rate": 3.236687774721504e-06, + "loss": 0.5993, + "step": 2991 + }, + { + "epoch": 2.477084483710657, + "grad_norm": 0.9291606545448303, + "learning_rate": 3.235636843052384e-06, + "loss": 0.5784, + "step": 2992 + }, + { + "epoch": 2.477912755383766, + "grad_norm": 0.9339813590049744, + "learning_rate": 3.234585769044392e-06, + "loss": 0.5951, + "step": 2993 + }, + { + "epoch": 2.478741027056875, + "grad_norm": 0.9638957977294922, + "learning_rate": 3.2335345529009017e-06, + "loss": 0.591, + "step": 2994 + }, + { + "epoch": 2.4795692987299836, + "grad_norm": 0.9618088603019714, + "learning_rate": 3.2324831948253123e-06, + "loss": 0.6102, + "step": 2995 + }, + { + "epoch": 2.4803975704030923, + "grad_norm": 0.9491007924079895, + "learning_rate": 3.2314316950210547e-06, + "loss": 0.599, + "step": 2996 + }, + { + "epoch": 2.481225842076201, + "grad_norm": 0.9213217496871948, + "learning_rate": 3.2303800536915804e-06, + "loss": 0.5955, + "step": 2997 + }, + { + "epoch": 2.4820541137493097, + "grad_norm": 0.923991322517395, + "learning_rate": 3.229328271040375e-06, + "loss": 0.5903, + "step": 2998 + }, + { + "epoch": 2.4828823854224185, + "grad_norm": 0.9173145890235901, + "learning_rate": 3.2282763472709484e-06, + "loss": 0.5899, + "step": 2999 + }, + { + "epoch": 2.483710657095527, + "grad_norm": 0.9357689023017883, + "learning_rate": 3.2272242825868363e-06, + "loss": 0.5885, + "step": 3000 + }, + { + "epoch": 2.484538928768636, + "grad_norm": 0.8902420401573181, + "learning_rate": 3.2261720771916056e-06, + "loss": 0.5899, + "step": 3001 + }, + { + "epoch": 2.485367200441745, + "grad_norm": 0.9079567790031433, + "learning_rate": 3.2251197312888475e-06, + "loss": 0.5966, + "step": 3002 + }, + { + "epoch": 2.486195472114854, + "grad_norm": 0.9068807363510132, + "learning_rate": 3.22406724508218e-06, + "loss": 0.6202, + "step": 3003 + }, + { + "epoch": 2.4870237437879625, + "grad_norm": 0.9458021521568298, + "learning_rate": 3.223014618775251e-06, + "loss": 0.5812, + "step": 3004 + }, + { + "epoch": 2.4878520154610713, + "grad_norm": 0.9506248235702515, + "learning_rate": 3.2219618525717328e-06, + "loss": 0.5819, + "step": 3005 + }, + { + "epoch": 2.48868028713418, + "grad_norm": 0.9573289155960083, + "learning_rate": 3.220908946675326e-06, + "loss": 0.581, + "step": 3006 + }, + { + "epoch": 2.4895085588072887, + "grad_norm": 0.9035286903381348, + "learning_rate": 3.219855901289758e-06, + "loss": 0.6182, + "step": 3007 + }, + { + "epoch": 2.4903368304803974, + "grad_norm": 0.9300901293754578, + "learning_rate": 3.218802716618784e-06, + "loss": 0.6079, + "step": 3008 + }, + { + "epoch": 2.491165102153506, + "grad_norm": 0.9116837382316589, + "learning_rate": 3.2177493928661854e-06, + "loss": 0.6034, + "step": 3009 + }, + { + "epoch": 2.491993373826615, + "grad_norm": 0.9291449189186096, + "learning_rate": 3.2166959302357694e-06, + "loss": 0.5889, + "step": 3010 + }, + { + "epoch": 2.492821645499724, + "grad_norm": 0.9002729654312134, + "learning_rate": 3.215642328931371e-06, + "loss": 0.5969, + "step": 3011 + }, + { + "epoch": 2.493649917172833, + "grad_norm": 0.8961556553840637, + "learning_rate": 3.214588589156854e-06, + "loss": 0.5899, + "step": 3012 + }, + { + "epoch": 2.4944781888459415, + "grad_norm": 0.9115450382232666, + "learning_rate": 3.2135347111161075e-06, + "loss": 0.5727, + "step": 3013 + }, + { + "epoch": 2.4953064605190503, + "grad_norm": 0.9060878157615662, + "learning_rate": 3.212480695013044e-06, + "loss": 0.5765, + "step": 3014 + }, + { + "epoch": 2.496134732192159, + "grad_norm": 0.902313768863678, + "learning_rate": 3.2114265410516088e-06, + "loss": 0.5913, + "step": 3015 + }, + { + "epoch": 2.4969630038652677, + "grad_norm": 0.931974470615387, + "learning_rate": 3.210372249435769e-06, + "loss": 0.5891, + "step": 3016 + }, + { + "epoch": 2.4977912755383764, + "grad_norm": 1.1711225509643555, + "learning_rate": 3.209317820369521e-06, + "loss": 0.6246, + "step": 3017 + }, + { + "epoch": 2.4986195472114856, + "grad_norm": 0.9561757445335388, + "learning_rate": 3.208263254056887e-06, + "loss": 0.5847, + "step": 3018 + }, + { + "epoch": 2.4994478188845943, + "grad_norm": 0.9621199369430542, + "learning_rate": 3.207208550701916e-06, + "loss": 0.5971, + "step": 3019 + }, + { + "epoch": 2.500276090557703, + "grad_norm": 0.9152470827102661, + "learning_rate": 3.206153710508683e-06, + "loss": 0.59, + "step": 3020 + }, + { + "epoch": 2.501104362230812, + "grad_norm": 0.9315138459205627, + "learning_rate": 3.20509873368129e-06, + "loss": 0.5841, + "step": 3021 + }, + { + "epoch": 2.5019326339039205, + "grad_norm": 0.9124136567115784, + "learning_rate": 3.204043620423864e-06, + "loss": 0.6007, + "step": 3022 + }, + { + "epoch": 2.5027609055770292, + "grad_norm": 0.8976690173149109, + "learning_rate": 3.2029883709405607e-06, + "loss": 0.5969, + "step": 3023 + }, + { + "epoch": 2.503589177250138, + "grad_norm": 0.8946467041969299, + "learning_rate": 3.201932985435561e-06, + "loss": 0.6069, + "step": 3024 + }, + { + "epoch": 2.5044174489232467, + "grad_norm": 0.9290626049041748, + "learning_rate": 3.2008774641130722e-06, + "loss": 0.591, + "step": 3025 + }, + { + "epoch": 2.5052457205963554, + "grad_norm": 0.8982367515563965, + "learning_rate": 3.1998218071773275e-06, + "loss": 0.5874, + "step": 3026 + }, + { + "epoch": 2.5060739922694646, + "grad_norm": 0.9062193036079407, + "learning_rate": 3.1987660148325867e-06, + "loss": 0.6107, + "step": 3027 + }, + { + "epoch": 2.5069022639425733, + "grad_norm": 0.9240522980690002, + "learning_rate": 3.1977100872831353e-06, + "loss": 0.5806, + "step": 3028 + }, + { + "epoch": 2.507730535615682, + "grad_norm": 0.9382046461105347, + "learning_rate": 3.1966540247332865e-06, + "loss": 0.5944, + "step": 3029 + }, + { + "epoch": 2.5085588072887908, + "grad_norm": 0.9194355010986328, + "learning_rate": 3.1955978273873776e-06, + "loss": 0.6085, + "step": 3030 + }, + { + "epoch": 2.5093870789618995, + "grad_norm": 0.9219276309013367, + "learning_rate": 3.1945414954497744e-06, + "loss": 0.5829, + "step": 3031 + }, + { + "epoch": 2.5102153506350082, + "grad_norm": 0.8998081088066101, + "learning_rate": 3.193485029124865e-06, + "loss": 0.6095, + "step": 3032 + }, + { + "epoch": 2.511043622308117, + "grad_norm": 0.9184247851371765, + "learning_rate": 3.192428428617068e-06, + "loss": 0.6129, + "step": 3033 + }, + { + "epoch": 2.5118718939812257, + "grad_norm": 0.9389240741729736, + "learning_rate": 3.1913716941308243e-06, + "loss": 0.607, + "step": 3034 + }, + { + "epoch": 2.5127001656543344, + "grad_norm": 0.9263836741447449, + "learning_rate": 3.190314825870603e-06, + "loss": 0.5761, + "step": 3035 + }, + { + "epoch": 2.5135284373274436, + "grad_norm": 0.9472658038139343, + "learning_rate": 3.1892578240408977e-06, + "loss": 0.5906, + "step": 3036 + }, + { + "epoch": 2.5143567090005523, + "grad_norm": 0.9045345783233643, + "learning_rate": 3.1882006888462285e-06, + "loss": 0.5993, + "step": 3037 + }, + { + "epoch": 2.515184980673661, + "grad_norm": 0.8930817246437073, + "learning_rate": 3.187143420491142e-06, + "loss": 0.5947, + "step": 3038 + }, + { + "epoch": 2.5160132523467698, + "grad_norm": 0.9130663871765137, + "learning_rate": 3.186086019180208e-06, + "loss": 0.5892, + "step": 3039 + }, + { + "epoch": 2.5168415240198785, + "grad_norm": 0.9008771777153015, + "learning_rate": 3.185028485118026e-06, + "loss": 0.5787, + "step": 3040 + }, + { + "epoch": 2.517669795692987, + "grad_norm": 0.9192292094230652, + "learning_rate": 3.1839708185092177e-06, + "loss": 0.5848, + "step": 3041 + }, + { + "epoch": 2.518498067366096, + "grad_norm": 0.8970836997032166, + "learning_rate": 3.1829130195584314e-06, + "loss": 0.5939, + "step": 3042 + }, + { + "epoch": 2.519326339039205, + "grad_norm": 0.9393385052680969, + "learning_rate": 3.181855088470342e-06, + "loss": 0.6096, + "step": 3043 + }, + { + "epoch": 2.5201546107123134, + "grad_norm": 0.9437024593353271, + "learning_rate": 3.180797025449649e-06, + "loss": 0.5998, + "step": 3044 + }, + { + "epoch": 2.5209828823854226, + "grad_norm": 0.9014863967895508, + "learning_rate": 3.179738830701078e-06, + "loss": 0.6015, + "step": 3045 + }, + { + "epoch": 2.5218111540585313, + "grad_norm": 0.9243669509887695, + "learning_rate": 3.1786805044293802e-06, + "loss": 0.5971, + "step": 3046 + }, + { + "epoch": 2.52263942573164, + "grad_norm": 0.9175970554351807, + "learning_rate": 3.1776220468393305e-06, + "loss": 0.6084, + "step": 3047 + }, + { + "epoch": 2.5234676974047487, + "grad_norm": 0.8962888121604919, + "learning_rate": 3.176563458135732e-06, + "loss": 0.5979, + "step": 3048 + }, + { + "epoch": 2.5242959690778575, + "grad_norm": 0.9321010708808899, + "learning_rate": 3.1755047385234117e-06, + "loss": 0.5941, + "step": 3049 + }, + { + "epoch": 2.525124240750966, + "grad_norm": 0.9147899746894836, + "learning_rate": 3.17444588820722e-06, + "loss": 0.5987, + "step": 3050 + }, + { + "epoch": 2.525952512424075, + "grad_norm": 0.9318193793296814, + "learning_rate": 3.173386907392037e-06, + "loss": 0.5658, + "step": 3051 + }, + { + "epoch": 2.526780784097184, + "grad_norm": 0.9039092659950256, + "learning_rate": 3.1723277962827647e-06, + "loss": 0.5956, + "step": 3052 + }, + { + "epoch": 2.5276090557702924, + "grad_norm": 0.9080818295478821, + "learning_rate": 3.1712685550843314e-06, + "loss": 0.5957, + "step": 3053 + }, + { + "epoch": 2.5284373274434015, + "grad_norm": 0.9120905995368958, + "learning_rate": 3.17020918400169e-06, + "loss": 0.5917, + "step": 3054 + }, + { + "epoch": 2.5292655991165103, + "grad_norm": 0.8983855843544006, + "learning_rate": 3.169149683239819e-06, + "loss": 0.6045, + "step": 3055 + }, + { + "epoch": 2.530093870789619, + "grad_norm": 0.9203280210494995, + "learning_rate": 3.168090053003721e-06, + "loss": 0.6093, + "step": 3056 + }, + { + "epoch": 2.5309221424627277, + "grad_norm": 0.9343932867050171, + "learning_rate": 3.1670302934984272e-06, + "loss": 0.5654, + "step": 3057 + }, + { + "epoch": 2.5317504141358365, + "grad_norm": 0.9054449200630188, + "learning_rate": 3.1659704049289885e-06, + "loss": 0.5878, + "step": 3058 + }, + { + "epoch": 2.532578685808945, + "grad_norm": 0.9011321067810059, + "learning_rate": 3.1649103875004856e-06, + "loss": 0.5831, + "step": 3059 + }, + { + "epoch": 2.533406957482054, + "grad_norm": 0.8886737823486328, + "learning_rate": 3.1638502414180206e-06, + "loss": 0.5915, + "step": 3060 + }, + { + "epoch": 2.534235229155163, + "grad_norm": 0.9289557337760925, + "learning_rate": 3.1627899668867214e-06, + "loss": 0.5944, + "step": 3061 + }, + { + "epoch": 2.535063500828272, + "grad_norm": 0.9269216060638428, + "learning_rate": 3.1617295641117425e-06, + "loss": 0.605, + "step": 3062 + }, + { + "epoch": 2.5358917725013805, + "grad_norm": 0.8877415657043457, + "learning_rate": 3.1606690332982613e-06, + "loss": 0.5954, + "step": 3063 + }, + { + "epoch": 2.5367200441744893, + "grad_norm": 0.9080086350440979, + "learning_rate": 3.1596083746514817e-06, + "loss": 0.5896, + "step": 3064 + }, + { + "epoch": 2.537548315847598, + "grad_norm": 0.9625695943832397, + "learning_rate": 3.15854758837663e-06, + "loss": 0.5666, + "step": 3065 + }, + { + "epoch": 2.5383765875207067, + "grad_norm": 0.9442515969276428, + "learning_rate": 3.1574866746789583e-06, + "loss": 0.5837, + "step": 3066 + }, + { + "epoch": 2.5392048591938154, + "grad_norm": 0.9054381847381592, + "learning_rate": 3.1564256337637443e-06, + "loss": 0.5799, + "step": 3067 + }, + { + "epoch": 2.5400331308669246, + "grad_norm": 0.9435595273971558, + "learning_rate": 3.1553644658362897e-06, + "loss": 0.6135, + "step": 3068 + }, + { + "epoch": 2.540861402540033, + "grad_norm": 0.9356298446655273, + "learning_rate": 3.1543031711019203e-06, + "loss": 0.5853, + "step": 3069 + }, + { + "epoch": 2.541689674213142, + "grad_norm": 0.917759895324707, + "learning_rate": 3.153241749765986e-06, + "loss": 0.5902, + "step": 3070 + }, + { + "epoch": 2.542517945886251, + "grad_norm": 0.9233857989311218, + "learning_rate": 3.1521802020338627e-06, + "loss": 0.6066, + "step": 3071 + }, + { + "epoch": 2.5433462175593595, + "grad_norm": 0.9497340321540833, + "learning_rate": 3.1511185281109504e-06, + "loss": 0.6003, + "step": 3072 + }, + { + "epoch": 2.5441744892324683, + "grad_norm": 0.9373629689216614, + "learning_rate": 3.1500567282026716e-06, + "loss": 0.5827, + "step": 3073 + }, + { + "epoch": 2.545002760905577, + "grad_norm": 0.8916833996772766, + "learning_rate": 3.148994802514476e-06, + "loss": 0.5996, + "step": 3074 + }, + { + "epoch": 2.5458310325786857, + "grad_norm": 0.9327612519264221, + "learning_rate": 3.1479327512518363e-06, + "loss": 0.6159, + "step": 3075 + }, + { + "epoch": 2.5466593042517944, + "grad_norm": 0.9380506873130798, + "learning_rate": 3.1468705746202493e-06, + "loss": 0.5924, + "step": 3076 + }, + { + "epoch": 2.5474875759249036, + "grad_norm": 0.9138150215148926, + "learning_rate": 3.1458082728252358e-06, + "loss": 0.5734, + "step": 3077 + }, + { + "epoch": 2.548315847598012, + "grad_norm": 0.8954431414604187, + "learning_rate": 3.144745846072341e-06, + "loss": 0.5902, + "step": 3078 + }, + { + "epoch": 2.549144119271121, + "grad_norm": 0.9259476661682129, + "learning_rate": 3.1436832945671362e-06, + "loss": 0.6057, + "step": 3079 + }, + { + "epoch": 2.54997239094423, + "grad_norm": 0.9295196533203125, + "learning_rate": 3.142620618515214e-06, + "loss": 0.5971, + "step": 3080 + }, + { + "epoch": 2.5508006626173385, + "grad_norm": 0.9318671822547913, + "learning_rate": 3.1415578181221928e-06, + "loss": 0.5852, + "step": 3081 + }, + { + "epoch": 2.5516289342904472, + "grad_norm": 0.9349870085716248, + "learning_rate": 3.1404948935937145e-06, + "loss": 0.5736, + "step": 3082 + }, + { + "epoch": 2.552457205963556, + "grad_norm": 0.9280831217765808, + "learning_rate": 3.1394318451354446e-06, + "loss": 0.5898, + "step": 3083 + }, + { + "epoch": 2.5532854776366647, + "grad_norm": 0.8969367742538452, + "learning_rate": 3.1383686729530727e-06, + "loss": 0.5983, + "step": 3084 + }, + { + "epoch": 2.5541137493097734, + "grad_norm": 0.9363370537757874, + "learning_rate": 3.1373053772523145e-06, + "loss": 0.6052, + "step": 3085 + }, + { + "epoch": 2.5549420209828826, + "grad_norm": 0.8944993615150452, + "learning_rate": 3.1362419582389057e-06, + "loss": 0.599, + "step": 3086 + }, + { + "epoch": 2.5557702926559913, + "grad_norm": 0.891412615776062, + "learning_rate": 3.1351784161186093e-06, + "loss": 0.5916, + "step": 3087 + }, + { + "epoch": 2.5565985643291, + "grad_norm": 0.930220365524292, + "learning_rate": 3.134114751097211e-06, + "loss": 0.5826, + "step": 3088 + }, + { + "epoch": 2.5574268360022088, + "grad_norm": 0.9193615317344666, + "learning_rate": 3.1330509633805183e-06, + "loss": 0.5896, + "step": 3089 + }, + { + "epoch": 2.5582551076753175, + "grad_norm": 0.9178160429000854, + "learning_rate": 3.131987053174367e-06, + "loss": 0.6104, + "step": 3090 + }, + { + "epoch": 2.5590833793484262, + "grad_norm": 0.9119648933410645, + "learning_rate": 3.130923020684611e-06, + "loss": 0.6198, + "step": 3091 + }, + { + "epoch": 2.559911651021535, + "grad_norm": 0.957272469997406, + "learning_rate": 3.1298588661171323e-06, + "loss": 0.5716, + "step": 3092 + }, + { + "epoch": 2.5607399226946437, + "grad_norm": 0.9073137044906616, + "learning_rate": 3.128794589677835e-06, + "loss": 0.5959, + "step": 3093 + }, + { + "epoch": 2.5615681943677524, + "grad_norm": 0.9372225403785706, + "learning_rate": 3.1277301915726456e-06, + "loss": 0.6146, + "step": 3094 + }, + { + "epoch": 2.5623964660408616, + "grad_norm": 0.9377845525741577, + "learning_rate": 3.126665672007515e-06, + "loss": 0.6055, + "step": 3095 + }, + { + "epoch": 2.5632247377139703, + "grad_norm": 0.9114481210708618, + "learning_rate": 3.1256010311884193e-06, + "loss": 0.6038, + "step": 3096 + }, + { + "epoch": 2.564053009387079, + "grad_norm": 0.9000943303108215, + "learning_rate": 3.124536269321356e-06, + "loss": 0.5806, + "step": 3097 + }, + { + "epoch": 2.5648812810601878, + "grad_norm": 0.9771347641944885, + "learning_rate": 3.123471386612347e-06, + "loss": 0.6029, + "step": 3098 + }, + { + "epoch": 2.5657095527332965, + "grad_norm": 0.9150393605232239, + "learning_rate": 3.1224063832674358e-06, + "loss": 0.6033, + "step": 3099 + }, + { + "epoch": 2.566537824406405, + "grad_norm": 0.8910076022148132, + "learning_rate": 3.1213412594926906e-06, + "loss": 0.6029, + "step": 3100 + }, + { + "epoch": 2.567366096079514, + "grad_norm": 0.9558795094490051, + "learning_rate": 3.1202760154942047e-06, + "loss": 0.5792, + "step": 3101 + }, + { + "epoch": 2.568194367752623, + "grad_norm": 0.958629310131073, + "learning_rate": 3.1192106514780924e-06, + "loss": 0.5933, + "step": 3102 + }, + { + "epoch": 2.5690226394257314, + "grad_norm": 0.9466328620910645, + "learning_rate": 3.1181451676504907e-06, + "loss": 0.6027, + "step": 3103 + }, + { + "epoch": 2.5698509110988406, + "grad_norm": 0.9397157430648804, + "learning_rate": 3.1170795642175614e-06, + "loss": 0.5894, + "step": 3104 + }, + { + "epoch": 2.5706791827719493, + "grad_norm": 0.9229122996330261, + "learning_rate": 3.116013841385489e-06, + "loss": 0.5812, + "step": 3105 + }, + { + "epoch": 2.571507454445058, + "grad_norm": 0.9332060217857361, + "learning_rate": 3.1149479993604805e-06, + "loss": 0.5738, + "step": 3106 + }, + { + "epoch": 2.5723357261181667, + "grad_norm": 0.9117311239242554, + "learning_rate": 3.113882038348766e-06, + "loss": 0.5923, + "step": 3107 + }, + { + "epoch": 2.5731639977912755, + "grad_norm": 0.9390400052070618, + "learning_rate": 3.1128159585566003e-06, + "loss": 0.6029, + "step": 3108 + }, + { + "epoch": 2.573992269464384, + "grad_norm": 0.9229103326797485, + "learning_rate": 3.1117497601902596e-06, + "loss": 0.6118, + "step": 3109 + }, + { + "epoch": 2.574820541137493, + "grad_norm": 0.8975940346717834, + "learning_rate": 3.1106834434560424e-06, + "loss": 0.6005, + "step": 3110 + }, + { + "epoch": 2.575648812810602, + "grad_norm": 0.9088030457496643, + "learning_rate": 3.1096170085602718e-06, + "loss": 0.5896, + "step": 3111 + }, + { + "epoch": 2.576477084483711, + "grad_norm": 0.982429027557373, + "learning_rate": 3.108550455709292e-06, + "loss": 0.5947, + "step": 3112 + }, + { + "epoch": 2.5773053561568195, + "grad_norm": 0.9404535293579102, + "learning_rate": 3.107483785109473e-06, + "loss": 0.6154, + "step": 3113 + }, + { + "epoch": 2.5781336278299283, + "grad_norm": 0.9074321389198303, + "learning_rate": 3.1064169969672043e-06, + "loss": 0.5889, + "step": 3114 + }, + { + "epoch": 2.578961899503037, + "grad_norm": 0.9000370502471924, + "learning_rate": 3.105350091488899e-06, + "loss": 0.5725, + "step": 3115 + }, + { + "epoch": 2.5797901711761457, + "grad_norm": 0.9296747446060181, + "learning_rate": 3.1042830688809944e-06, + "loss": 0.5983, + "step": 3116 + }, + { + "epoch": 2.5806184428492545, + "grad_norm": 0.9164032936096191, + "learning_rate": 3.1032159293499486e-06, + "loss": 0.5832, + "step": 3117 + }, + { + "epoch": 2.581446714522363, + "grad_norm": 0.9070228338241577, + "learning_rate": 3.1021486731022443e-06, + "loss": 0.5977, + "step": 3118 + }, + { + "epoch": 2.582274986195472, + "grad_norm": 0.9384707808494568, + "learning_rate": 3.1010813003443847e-06, + "loss": 0.5926, + "step": 3119 + }, + { + "epoch": 2.583103257868581, + "grad_norm": 0.937782347202301, + "learning_rate": 3.100013811282896e-06, + "loss": 0.5859, + "step": 3120 + }, + { + "epoch": 2.58393152954169, + "grad_norm": 0.9293384552001953, + "learning_rate": 3.098946206124329e-06, + "loss": 0.5826, + "step": 3121 + }, + { + "epoch": 2.5847598012147985, + "grad_norm": 0.935950756072998, + "learning_rate": 3.097878485075255e-06, + "loss": 0.589, + "step": 3122 + }, + { + "epoch": 2.5855880728879073, + "grad_norm": 0.9377515316009521, + "learning_rate": 3.0968106483422655e-06, + "loss": 0.5985, + "step": 3123 + }, + { + "epoch": 2.586416344561016, + "grad_norm": 0.9256250858306885, + "learning_rate": 3.09574269613198e-06, + "loss": 0.5944, + "step": 3124 + }, + { + "epoch": 2.5872446162341247, + "grad_norm": 0.873123288154602, + "learning_rate": 3.0946746286510364e-06, + "loss": 0.5918, + "step": 3125 + }, + { + "epoch": 2.5880728879072334, + "grad_norm": 0.9320067167282104, + "learning_rate": 3.093606446106096e-06, + "loss": 0.6365, + "step": 3126 + }, + { + "epoch": 2.5889011595803426, + "grad_norm": 0.9514285326004028, + "learning_rate": 3.0925381487038415e-06, + "loss": 0.585, + "step": 3127 + }, + { + "epoch": 2.589729431253451, + "grad_norm": 0.9028947353363037, + "learning_rate": 3.0914697366509777e-06, + "loss": 0.6063, + "step": 3128 + }, + { + "epoch": 2.59055770292656, + "grad_norm": 0.8989999294281006, + "learning_rate": 3.0904012101542346e-06, + "loss": 0.6012, + "step": 3129 + }, + { + "epoch": 2.591385974599669, + "grad_norm": 0.9131087064743042, + "learning_rate": 3.089332569420361e-06, + "loss": 0.5742, + "step": 3130 + }, + { + "epoch": 2.5922142462727775, + "grad_norm": 0.916441023349762, + "learning_rate": 3.088263814656128e-06, + "loss": 0.6011, + "step": 3131 + }, + { + "epoch": 2.5930425179458862, + "grad_norm": 0.8957144618034363, + "learning_rate": 3.087194946068332e-06, + "loss": 0.6008, + "step": 3132 + }, + { + "epoch": 2.593870789618995, + "grad_norm": 0.90632563829422, + "learning_rate": 3.086125963863787e-06, + "loss": 0.6062, + "step": 3133 + }, + { + "epoch": 2.5946990612921037, + "grad_norm": 0.94493168592453, + "learning_rate": 3.0850568682493315e-06, + "loss": 0.5667, + "step": 3134 + }, + { + "epoch": 2.5955273329652124, + "grad_norm": 0.9155673980712891, + "learning_rate": 3.0839876594318257e-06, + "loss": 0.5755, + "step": 3135 + }, + { + "epoch": 2.5963556046383216, + "grad_norm": 0.9013570547103882, + "learning_rate": 3.0829183376181524e-06, + "loss": 0.6053, + "step": 3136 + }, + { + "epoch": 2.59718387631143, + "grad_norm": 0.9294494390487671, + "learning_rate": 3.081848903015214e-06, + "loss": 0.5881, + "step": 3137 + }, + { + "epoch": 2.598012147984539, + "grad_norm": 0.8956479430198669, + "learning_rate": 3.0807793558299377e-06, + "loss": 0.5976, + "step": 3138 + }, + { + "epoch": 2.598840419657648, + "grad_norm": 0.9205741286277771, + "learning_rate": 3.079709696269269e-06, + "loss": 0.5776, + "step": 3139 + }, + { + "epoch": 2.5996686913307565, + "grad_norm": 0.901066780090332, + "learning_rate": 3.0786399245401784e-06, + "loss": 0.6101, + "step": 3140 + }, + { + "epoch": 2.6004969630038652, + "grad_norm": 0.9003482460975647, + "learning_rate": 3.0775700408496568e-06, + "loss": 0.5983, + "step": 3141 + }, + { + "epoch": 2.601325234676974, + "grad_norm": 0.9000759124755859, + "learning_rate": 3.0765000454047155e-06, + "loss": 0.605, + "step": 3142 + }, + { + "epoch": 2.6021535063500827, + "grad_norm": 0.8797931671142578, + "learning_rate": 3.0754299384123905e-06, + "loss": 0.5874, + "step": 3143 + }, + { + "epoch": 2.6029817780231914, + "grad_norm": 0.9307072758674622, + "learning_rate": 3.074359720079736e-06, + "loss": 0.6035, + "step": 3144 + }, + { + "epoch": 2.6038100496963006, + "grad_norm": 0.8928053379058838, + "learning_rate": 3.073289390613829e-06, + "loss": 0.5759, + "step": 3145 + }, + { + "epoch": 2.6046383213694093, + "grad_norm": 0.8879983425140381, + "learning_rate": 3.0722189502217705e-06, + "loss": 0.5947, + "step": 3146 + }, + { + "epoch": 2.605466593042518, + "grad_norm": 0.8889674544334412, + "learning_rate": 3.0711483991106784e-06, + "loss": 0.5939, + "step": 3147 + }, + { + "epoch": 2.6062948647156268, + "grad_norm": 0.9318514466285706, + "learning_rate": 3.070077737487695e-06, + "loss": 0.599, + "step": 3148 + }, + { + "epoch": 2.6071231363887355, + "grad_norm": 0.9251010417938232, + "learning_rate": 3.069006965559985e-06, + "loss": 0.5956, + "step": 3149 + }, + { + "epoch": 2.6079514080618442, + "grad_norm": 0.9122523069381714, + "learning_rate": 3.06793608353473e-06, + "loss": 0.6011, + "step": 3150 + }, + { + "epoch": 2.608779679734953, + "grad_norm": 0.9270116090774536, + "learning_rate": 3.0668650916191365e-06, + "loss": 0.5923, + "step": 3151 + }, + { + "epoch": 2.609607951408062, + "grad_norm": 0.939582884311676, + "learning_rate": 3.0657939900204338e-06, + "loss": 0.6053, + "step": 3152 + }, + { + "epoch": 2.6104362230811704, + "grad_norm": 0.9206486940383911, + "learning_rate": 3.064722778945867e-06, + "loss": 0.6197, + "step": 3153 + }, + { + "epoch": 2.6112644947542796, + "grad_norm": 0.9047932624816895, + "learning_rate": 3.063651458602708e-06, + "loss": 0.5774, + "step": 3154 + }, + { + "epoch": 2.6120927664273883, + "grad_norm": 0.9285469055175781, + "learning_rate": 3.062580029198245e-06, + "loss": 0.6107, + "step": 3155 + }, + { + "epoch": 2.612921038100497, + "grad_norm": 0.9187634587287903, + "learning_rate": 3.0615084909397906e-06, + "loss": 0.5919, + "step": 3156 + }, + { + "epoch": 2.6137493097736058, + "grad_norm": 0.8950710296630859, + "learning_rate": 3.060436844034678e-06, + "loss": 0.6029, + "step": 3157 + }, + { + "epoch": 2.6145775814467145, + "grad_norm": 0.9206116795539856, + "learning_rate": 3.0593650886902606e-06, + "loss": 0.5632, + "step": 3158 + }, + { + "epoch": 2.615405853119823, + "grad_norm": 0.9081333875656128, + "learning_rate": 3.058293225113912e-06, + "loss": 0.6127, + "step": 3159 + }, + { + "epoch": 2.616234124792932, + "grad_norm": 0.9028277397155762, + "learning_rate": 3.0572212535130297e-06, + "loss": 0.6089, + "step": 3160 + }, + { + "epoch": 2.617062396466041, + "grad_norm": 0.9449213147163391, + "learning_rate": 3.0561491740950288e-06, + "loss": 0.5841, + "step": 3161 + }, + { + "epoch": 2.6178906681391494, + "grad_norm": 0.9926438927650452, + "learning_rate": 3.055076987067347e-06, + "loss": 0.564, + "step": 3162 + }, + { + "epoch": 2.6187189398122586, + "grad_norm": 0.9098719954490662, + "learning_rate": 3.0540046926374422e-06, + "loss": 0.6268, + "step": 3163 + }, + { + "epoch": 2.6195472114853673, + "grad_norm": 0.8997418880462646, + "learning_rate": 3.0529322910127944e-06, + "loss": 0.5972, + "step": 3164 + }, + { + "epoch": 2.620375483158476, + "grad_norm": 0.9076839685440063, + "learning_rate": 3.0518597824009015e-06, + "loss": 0.5921, + "step": 3165 + }, + { + "epoch": 2.6212037548315847, + "grad_norm": 0.9348479509353638, + "learning_rate": 3.050787167009287e-06, + "loss": 0.5802, + "step": 3166 + }, + { + "epoch": 2.6220320265046935, + "grad_norm": 0.9381133913993835, + "learning_rate": 3.0497144450454886e-06, + "loss": 0.5943, + "step": 3167 + }, + { + "epoch": 2.622860298177802, + "grad_norm": 0.9327574968338013, + "learning_rate": 3.0486416167170694e-06, + "loss": 0.5832, + "step": 3168 + }, + { + "epoch": 2.623688569850911, + "grad_norm": 0.9142323732376099, + "learning_rate": 3.0475686822316126e-06, + "loss": 0.6048, + "step": 3169 + }, + { + "epoch": 2.62451684152402, + "grad_norm": 0.9799601435661316, + "learning_rate": 3.04649564179672e-06, + "loss": 0.5783, + "step": 3170 + }, + { + "epoch": 2.625345113197129, + "grad_norm": 0.9203731417655945, + "learning_rate": 3.045422495620015e-06, + "loss": 0.6028, + "step": 3171 + }, + { + "epoch": 2.6261733848702375, + "grad_norm": 0.9614686369895935, + "learning_rate": 3.0443492439091415e-06, + "loss": 0.5895, + "step": 3172 + }, + { + "epoch": 2.6270016565433463, + "grad_norm": 0.980112612247467, + "learning_rate": 3.0432758868717626e-06, + "loss": 0.5887, + "step": 3173 + }, + { + "epoch": 2.627829928216455, + "grad_norm": 0.9174870252609253, + "learning_rate": 3.042202424715565e-06, + "loss": 0.5831, + "step": 3174 + }, + { + "epoch": 2.6286581998895637, + "grad_norm": 0.9396457672119141, + "learning_rate": 3.0411288576482525e-06, + "loss": 0.5816, + "step": 3175 + }, + { + "epoch": 2.6294864715626725, + "grad_norm": 0.9067835211753845, + "learning_rate": 3.04005518587755e-06, + "loss": 0.5748, + "step": 3176 + }, + { + "epoch": 2.630314743235781, + "grad_norm": 0.9329981207847595, + "learning_rate": 3.038981409611204e-06, + "loss": 0.5901, + "step": 3177 + }, + { + "epoch": 2.63114301490889, + "grad_norm": 0.9418731927871704, + "learning_rate": 3.0379075290569793e-06, + "loss": 0.5981, + "step": 3178 + }, + { + "epoch": 2.631971286581999, + "grad_norm": 0.9180083274841309, + "learning_rate": 3.0368335444226614e-06, + "loss": 0.5653, + "step": 3179 + }, + { + "epoch": 2.632799558255108, + "grad_norm": 0.9566802978515625, + "learning_rate": 3.0357594559160575e-06, + "loss": 0.5968, + "step": 3180 + }, + { + "epoch": 2.6336278299282165, + "grad_norm": 0.9177946448326111, + "learning_rate": 3.034685263744994e-06, + "loss": 0.6076, + "step": 3181 + }, + { + "epoch": 2.6344561016013253, + "grad_norm": 0.905975341796875, + "learning_rate": 3.033610968117314e-06, + "loss": 0.6064, + "step": 3182 + }, + { + "epoch": 2.635284373274434, + "grad_norm": 0.9021571278572083, + "learning_rate": 3.0325365692408883e-06, + "loss": 0.5818, + "step": 3183 + }, + { + "epoch": 2.6361126449475427, + "grad_norm": 0.9211055040359497, + "learning_rate": 3.0314620673235995e-06, + "loss": 0.5811, + "step": 3184 + }, + { + "epoch": 2.6369409166206514, + "grad_norm": 0.929892897605896, + "learning_rate": 3.0303874625733548e-06, + "loss": 0.6039, + "step": 3185 + }, + { + "epoch": 2.6377691882937606, + "grad_norm": 0.9520539045333862, + "learning_rate": 3.0293127551980805e-06, + "loss": 0.6156, + "step": 3186 + }, + { + "epoch": 2.638597459966869, + "grad_norm": 0.9111108779907227, + "learning_rate": 3.028237945405722e-06, + "loss": 0.6081, + "step": 3187 + }, + { + "epoch": 2.639425731639978, + "grad_norm": 0.9411360025405884, + "learning_rate": 3.027163033404246e-06, + "loss": 0.5897, + "step": 3188 + }, + { + "epoch": 2.640254003313087, + "grad_norm": 0.9253203868865967, + "learning_rate": 3.026088019401637e-06, + "loss": 0.5957, + "step": 3189 + }, + { + "epoch": 2.6410822749861955, + "grad_norm": 1.010617971420288, + "learning_rate": 3.0250129036059e-06, + "loss": 0.602, + "step": 3190 + }, + { + "epoch": 2.6419105466593042, + "grad_norm": 0.9021222591400146, + "learning_rate": 3.02393768622506e-06, + "loss": 0.5796, + "step": 3191 + }, + { + "epoch": 2.642738818332413, + "grad_norm": 0.9229567050933838, + "learning_rate": 3.0228623674671627e-06, + "loss": 0.5827, + "step": 3192 + }, + { + "epoch": 2.6435670900055217, + "grad_norm": 0.9603396654129028, + "learning_rate": 3.0217869475402704e-06, + "loss": 0.5886, + "step": 3193 + }, + { + "epoch": 2.6443953616786304, + "grad_norm": 0.9483562707901001, + "learning_rate": 3.020711426652469e-06, + "loss": 0.6089, + "step": 3194 + }, + { + "epoch": 2.6452236333517396, + "grad_norm": 0.9240924715995789, + "learning_rate": 3.019635805011859e-06, + "loss": 0.6182, + "step": 3195 + }, + { + "epoch": 2.646051905024848, + "grad_norm": 0.9176941514015198, + "learning_rate": 3.018560082826566e-06, + "loss": 0.593, + "step": 3196 + }, + { + "epoch": 2.646880176697957, + "grad_norm": 0.9058155417442322, + "learning_rate": 3.0174842603047307e-06, + "loss": 0.5853, + "step": 3197 + }, + { + "epoch": 2.647708448371066, + "grad_norm": 0.9544078707695007, + "learning_rate": 3.016408337654514e-06, + "loss": 0.5743, + "step": 3198 + }, + { + "epoch": 2.6485367200441745, + "grad_norm": 0.944963812828064, + "learning_rate": 3.0153323150840987e-06, + "loss": 0.6007, + "step": 3199 + }, + { + "epoch": 2.6493649917172832, + "grad_norm": 0.9196237921714783, + "learning_rate": 3.014256192801685e-06, + "loss": 0.5818, + "step": 3200 + }, + { + "epoch": 2.650193263390392, + "grad_norm": 0.9315558075904846, + "learning_rate": 3.01317997101549e-06, + "loss": 0.5861, + "step": 3201 + }, + { + "epoch": 2.6510215350635007, + "grad_norm": 0.9272212982177734, + "learning_rate": 3.012103649933756e-06, + "loss": 0.597, + "step": 3202 + }, + { + "epoch": 2.6518498067366094, + "grad_norm": 0.9128772616386414, + "learning_rate": 3.011027229764739e-06, + "loss": 0.5863, + "step": 3203 + }, + { + "epoch": 2.6526780784097186, + "grad_norm": 0.9234123826026917, + "learning_rate": 3.0099507107167158e-06, + "loss": 0.589, + "step": 3204 + }, + { + "epoch": 2.6535063500828273, + "grad_norm": 0.9162564873695374, + "learning_rate": 3.0088740929979854e-06, + "loss": 0.5855, + "step": 3205 + }, + { + "epoch": 2.654334621755936, + "grad_norm": 0.8842553496360779, + "learning_rate": 3.0077973768168604e-06, + "loss": 0.5947, + "step": 3206 + }, + { + "epoch": 2.6551628934290448, + "grad_norm": 0.9413550496101379, + "learning_rate": 3.0067205623816775e-06, + "loss": 0.5882, + "step": 3207 + }, + { + "epoch": 2.6559911651021535, + "grad_norm": 0.9217298626899719, + "learning_rate": 3.005643649900789e-06, + "loss": 0.5806, + "step": 3208 + }, + { + "epoch": 2.656819436775262, + "grad_norm": 0.9214861392974854, + "learning_rate": 3.004566639582568e-06, + "loss": 0.5656, + "step": 3209 + }, + { + "epoch": 2.657647708448371, + "grad_norm": 0.9789690375328064, + "learning_rate": 3.003489531635405e-06, + "loss": 0.5897, + "step": 3210 + }, + { + "epoch": 2.65847598012148, + "grad_norm": 0.9465369582176208, + "learning_rate": 3.0024123262677123e-06, + "loss": 0.5996, + "step": 3211 + }, + { + "epoch": 2.6593042517945884, + "grad_norm": 0.9011247754096985, + "learning_rate": 3.001335023687917e-06, + "loss": 0.5849, + "step": 3212 + }, + { + "epoch": 2.6601325234676976, + "grad_norm": 0.9762759804725647, + "learning_rate": 3.0002576241044683e-06, + "loss": 0.6161, + "step": 3213 + }, + { + "epoch": 2.6609607951408063, + "grad_norm": 0.9362512230873108, + "learning_rate": 2.9991801277258327e-06, + "loss": 0.6186, + "step": 3214 + }, + { + "epoch": 2.661789066813915, + "grad_norm": 0.9391586780548096, + "learning_rate": 2.998102534760495e-06, + "loss": 0.5787, + "step": 3215 + }, + { + "epoch": 2.6626173384870238, + "grad_norm": 0.9361854195594788, + "learning_rate": 2.997024845416961e-06, + "loss": 0.5996, + "step": 3216 + }, + { + "epoch": 2.6634456101601325, + "grad_norm": 0.9153965711593628, + "learning_rate": 2.9959470599037525e-06, + "loss": 0.5777, + "step": 3217 + }, + { + "epoch": 2.664273881833241, + "grad_norm": 0.9155426621437073, + "learning_rate": 2.994869178429411e-06, + "loss": 0.5906, + "step": 3218 + }, + { + "epoch": 2.66510215350635, + "grad_norm": 0.8978738784790039, + "learning_rate": 2.993791201202496e-06, + "loss": 0.5707, + "step": 3219 + }, + { + "epoch": 2.665930425179459, + "grad_norm": 0.9147471785545349, + "learning_rate": 2.9927131284315874e-06, + "loss": 0.6003, + "step": 3220 + }, + { + "epoch": 2.6667586968525674, + "grad_norm": 0.9578046202659607, + "learning_rate": 2.991634960325281e-06, + "loss": 0.6068, + "step": 3221 + }, + { + "epoch": 2.6675869685256766, + "grad_norm": 0.9182425141334534, + "learning_rate": 2.9905566970921933e-06, + "loss": 0.5939, + "step": 3222 + }, + { + "epoch": 2.6684152401987853, + "grad_norm": 0.9168176651000977, + "learning_rate": 2.989478338940957e-06, + "loss": 0.5967, + "step": 3223 + }, + { + "epoch": 2.669243511871894, + "grad_norm": 0.913521409034729, + "learning_rate": 2.9883998860802253e-06, + "loss": 0.6081, + "step": 3224 + }, + { + "epoch": 2.6700717835450027, + "grad_norm": 0.9223765134811401, + "learning_rate": 2.9873213387186684e-06, + "loss": 0.5901, + "step": 3225 + }, + { + "epoch": 2.6709000552181115, + "grad_norm": 0.9228116273880005, + "learning_rate": 2.9862426970649743e-06, + "loss": 0.5959, + "step": 3226 + }, + { + "epoch": 2.67172832689122, + "grad_norm": 0.9379467368125916, + "learning_rate": 2.985163961327852e-06, + "loss": 0.5986, + "step": 3227 + }, + { + "epoch": 2.672556598564329, + "grad_norm": 0.921449601650238, + "learning_rate": 2.984085131716027e-06, + "loss": 0.5587, + "step": 3228 + }, + { + "epoch": 2.673384870237438, + "grad_norm": 0.9038673043251038, + "learning_rate": 2.9830062084382393e-06, + "loss": 0.5908, + "step": 3229 + }, + { + "epoch": 2.674213141910547, + "grad_norm": 0.9072180986404419, + "learning_rate": 2.981927191703254e-06, + "loss": 0.6064, + "step": 3230 + }, + { + "epoch": 2.6750414135836555, + "grad_norm": 0.9178011417388916, + "learning_rate": 2.9808480817198494e-06, + "loss": 0.5969, + "step": 3231 + }, + { + "epoch": 2.6758696852567643, + "grad_norm": 0.9064451456069946, + "learning_rate": 2.9797688786968233e-06, + "loss": 0.571, + "step": 3232 + }, + { + "epoch": 2.676697956929873, + "grad_norm": 0.9392306804656982, + "learning_rate": 2.9786895828429914e-06, + "loss": 0.5903, + "step": 3233 + }, + { + "epoch": 2.6775262286029817, + "grad_norm": 0.9398954510688782, + "learning_rate": 2.9776101943671875e-06, + "loss": 0.6163, + "step": 3234 + }, + { + "epoch": 2.6783545002760905, + "grad_norm": 0.9082908630371094, + "learning_rate": 2.9765307134782637e-06, + "loss": 0.5838, + "step": 3235 + }, + { + "epoch": 2.679182771949199, + "grad_norm": 0.926916778087616, + "learning_rate": 2.975451140385089e-06, + "loss": 0.5599, + "step": 3236 + }, + { + "epoch": 2.680011043622308, + "grad_norm": 0.9106405377388, + "learning_rate": 2.97437147529655e-06, + "loss": 0.6012, + "step": 3237 + }, + { + "epoch": 2.680839315295417, + "grad_norm": 0.9428172707557678, + "learning_rate": 2.9732917184215524e-06, + "loss": 0.5985, + "step": 3238 + }, + { + "epoch": 2.681667586968526, + "grad_norm": 0.9456861019134521, + "learning_rate": 2.9722118699690212e-06, + "loss": 0.6206, + "step": 3239 + }, + { + "epoch": 2.6824958586416345, + "grad_norm": 0.9116294384002686, + "learning_rate": 2.971131930147893e-06, + "loss": 0.6166, + "step": 3240 + }, + { + "epoch": 2.6833241303147433, + "grad_norm": 0.9061588644981384, + "learning_rate": 2.970051899167129e-06, + "loss": 0.5792, + "step": 3241 + }, + { + "epoch": 2.684152401987852, + "grad_norm": 0.947784423828125, + "learning_rate": 2.968971777235704e-06, + "loss": 0.5813, + "step": 3242 + }, + { + "epoch": 2.6849806736609607, + "grad_norm": 0.9296494722366333, + "learning_rate": 2.9678915645626115e-06, + "loss": 0.5916, + "step": 3243 + }, + { + "epoch": 2.6858089453340694, + "grad_norm": 0.9151764512062073, + "learning_rate": 2.9668112613568632e-06, + "loss": 0.6097, + "step": 3244 + }, + { + "epoch": 2.6866372170071786, + "grad_norm": 0.9887247681617737, + "learning_rate": 2.965730867827488e-06, + "loss": 0.5951, + "step": 3245 + }, + { + "epoch": 2.687465488680287, + "grad_norm": 0.9406042695045471, + "learning_rate": 2.964650384183531e-06, + "loss": 0.5918, + "step": 3246 + }, + { + "epoch": 2.688293760353396, + "grad_norm": 0.9440297484397888, + "learning_rate": 2.963569810634056e-06, + "loss": 0.6098, + "step": 3247 + }, + { + "epoch": 2.689122032026505, + "grad_norm": 0.93993079662323, + "learning_rate": 2.962489147388144e-06, + "loss": 0.6242, + "step": 3248 + }, + { + "epoch": 2.6899503036996135, + "grad_norm": 0.9704070091247559, + "learning_rate": 2.9614083946548927e-06, + "loss": 0.5764, + "step": 3249 + }, + { + "epoch": 2.6907785753727222, + "grad_norm": 0.9809379577636719, + "learning_rate": 2.9603275526434188e-06, + "loss": 0.5997, + "step": 3250 + }, + { + "epoch": 2.691606847045831, + "grad_norm": 0.9246951937675476, + "learning_rate": 2.9592466215628547e-06, + "loss": 0.577, + "step": 3251 + }, + { + "epoch": 2.6924351187189397, + "grad_norm": 0.9204201698303223, + "learning_rate": 2.95816560162235e-06, + "loss": 0.6036, + "step": 3252 + }, + { + "epoch": 2.6932633903920484, + "grad_norm": 0.9180527925491333, + "learning_rate": 2.9570844930310734e-06, + "loss": 0.5919, + "step": 3253 + }, + { + "epoch": 2.6940916620651576, + "grad_norm": 0.9165668487548828, + "learning_rate": 2.9560032959982067e-06, + "loss": 0.5646, + "step": 3254 + }, + { + "epoch": 2.694919933738266, + "grad_norm": 0.9150571823120117, + "learning_rate": 2.9549220107329536e-06, + "loss": 0.581, + "step": 3255 + }, + { + "epoch": 2.695748205411375, + "grad_norm": 0.9429931044578552, + "learning_rate": 2.9538406374445334e-06, + "loss": 0.595, + "step": 3256 + }, + { + "epoch": 2.696576477084484, + "grad_norm": 0.939118504524231, + "learning_rate": 2.9527591763421785e-06, + "loss": 0.5772, + "step": 3257 + }, + { + "epoch": 2.6974047487575925, + "grad_norm": 0.9130142331123352, + "learning_rate": 2.9516776276351446e-06, + "loss": 0.6031, + "step": 3258 + }, + { + "epoch": 2.6982330204307012, + "grad_norm": 0.9334836006164551, + "learning_rate": 2.9505959915327004e-06, + "loss": 0.5899, + "step": 3259 + }, + { + "epoch": 2.69906129210381, + "grad_norm": 0.9407964944839478, + "learning_rate": 2.9495142682441314e-06, + "loss": 0.6232, + "step": 3260 + }, + { + "epoch": 2.6998895637769187, + "grad_norm": 0.9573985934257507, + "learning_rate": 2.9484324579787427e-06, + "loss": 0.5932, + "step": 3261 + }, + { + "epoch": 2.7007178354500274, + "grad_norm": 0.9223983287811279, + "learning_rate": 2.947350560945853e-06, + "loss": 0.5861, + "step": 3262 + }, + { + "epoch": 2.7015461071231366, + "grad_norm": 0.9206790328025818, + "learning_rate": 2.9462685773548004e-06, + "loss": 0.5867, + "step": 3263 + }, + { + "epoch": 2.7023743787962453, + "grad_norm": 0.9197454452514648, + "learning_rate": 2.9451865074149378e-06, + "loss": 0.5867, + "step": 3264 + }, + { + "epoch": 2.703202650469354, + "grad_norm": 0.9365606904029846, + "learning_rate": 2.944104351335636e-06, + "loss": 0.6106, + "step": 3265 + }, + { + "epoch": 2.7040309221424628, + "grad_norm": 0.9116243720054626, + "learning_rate": 2.9430221093262823e-06, + "loss": 0.5904, + "step": 3266 + }, + { + "epoch": 2.7048591938155715, + "grad_norm": 0.9128034114837646, + "learning_rate": 2.9419397815962802e-06, + "loss": 0.6055, + "step": 3267 + }, + { + "epoch": 2.70568746548868, + "grad_norm": 0.9147705435752869, + "learning_rate": 2.94085736835505e-06, + "loss": 0.5941, + "step": 3268 + }, + { + "epoch": 2.706515737161789, + "grad_norm": 0.9419110417366028, + "learning_rate": 2.939774869812029e-06, + "loss": 0.5595, + "step": 3269 + }, + { + "epoch": 2.707344008834898, + "grad_norm": 0.9202789664268494, + "learning_rate": 2.93869228617667e-06, + "loss": 0.5788, + "step": 3270 + }, + { + "epoch": 2.7081722805080064, + "grad_norm": 0.9068663716316223, + "learning_rate": 2.9376096176584424e-06, + "loss": 0.5749, + "step": 3271 + }, + { + "epoch": 2.7090005521811156, + "grad_norm": 0.923497200012207, + "learning_rate": 2.936526864466833e-06, + "loss": 0.6029, + "step": 3272 + }, + { + "epoch": 2.7098288238542243, + "grad_norm": 0.9812871217727661, + "learning_rate": 2.935444026811346e-06, + "loss": 0.6193, + "step": 3273 + }, + { + "epoch": 2.710657095527333, + "grad_norm": 0.9538823366165161, + "learning_rate": 2.934361104901498e-06, + "loss": 0.5809, + "step": 3274 + }, + { + "epoch": 2.7114853672004418, + "grad_norm": 0.9111268520355225, + "learning_rate": 2.9332780989468256e-06, + "loss": 0.6181, + "step": 3275 + }, + { + "epoch": 2.7123136388735505, + "grad_norm": 0.9485889673233032, + "learning_rate": 2.9321950091568795e-06, + "loss": 0.5823, + "step": 3276 + }, + { + "epoch": 2.713141910546659, + "grad_norm": 0.9233952760696411, + "learning_rate": 2.931111835741229e-06, + "loss": 0.6008, + "step": 3277 + }, + { + "epoch": 2.713970182219768, + "grad_norm": 0.9075472950935364, + "learning_rate": 2.9300285789094563e-06, + "loss": 0.5808, + "step": 3278 + }, + { + "epoch": 2.714798453892877, + "grad_norm": 0.9238584041595459, + "learning_rate": 2.9289452388711624e-06, + "loss": 0.6095, + "step": 3279 + }, + { + "epoch": 2.7156267255659854, + "grad_norm": 0.889460027217865, + "learning_rate": 2.9278618158359635e-06, + "loss": 0.6015, + "step": 3280 + }, + { + "epoch": 2.7164549972390946, + "grad_norm": 0.9549591541290283, + "learning_rate": 2.9267783100134915e-06, + "loss": 0.6029, + "step": 3281 + }, + { + "epoch": 2.7172832689122033, + "grad_norm": 0.9522138237953186, + "learning_rate": 2.9256947216133946e-06, + "loss": 0.5998, + "step": 3282 + }, + { + "epoch": 2.718111540585312, + "grad_norm": 0.8992515802383423, + "learning_rate": 2.924611050845338e-06, + "loss": 0.6012, + "step": 3283 + }, + { + "epoch": 2.7189398122584207, + "grad_norm": 0.9058646559715271, + "learning_rate": 2.9235272979190014e-06, + "loss": 0.6092, + "step": 3284 + }, + { + "epoch": 2.7197680839315295, + "grad_norm": 0.916130542755127, + "learning_rate": 2.92244346304408e-06, + "loss": 0.6054, + "step": 3285 + }, + { + "epoch": 2.720596355604638, + "grad_norm": 0.9203442335128784, + "learning_rate": 2.921359546430287e-06, + "loss": 0.596, + "step": 3286 + }, + { + "epoch": 2.721424627277747, + "grad_norm": 0.9603051543235779, + "learning_rate": 2.920275548287349e-06, + "loss": 0.5976, + "step": 3287 + }, + { + "epoch": 2.722252898950856, + "grad_norm": 0.9188669919967651, + "learning_rate": 2.91919146882501e-06, + "loss": 0.5812, + "step": 3288 + }, + { + "epoch": 2.723081170623965, + "grad_norm": 0.9154419898986816, + "learning_rate": 2.9181073082530302e-06, + "loss": 0.6127, + "step": 3289 + }, + { + "epoch": 2.7239094422970735, + "grad_norm": 0.9228849411010742, + "learning_rate": 2.917023066781184e-06, + "loss": 0.6223, + "step": 3290 + }, + { + "epoch": 2.7247377139701823, + "grad_norm": 0.9034006595611572, + "learning_rate": 2.915938744619261e-06, + "loss": 0.5981, + "step": 3291 + }, + { + "epoch": 2.725565985643291, + "grad_norm": 0.9119758009910583, + "learning_rate": 2.914854341977069e-06, + "loss": 0.5974, + "step": 3292 + }, + { + "epoch": 2.7263942573163997, + "grad_norm": 0.9350270628929138, + "learning_rate": 2.9137698590644283e-06, + "loss": 0.5958, + "step": 3293 + }, + { + "epoch": 2.7272225289895085, + "grad_norm": 0.9229425191879272, + "learning_rate": 2.9126852960911773e-06, + "loss": 0.5873, + "step": 3294 + }, + { + "epoch": 2.728050800662617, + "grad_norm": 0.9276396632194519, + "learning_rate": 2.9116006532671685e-06, + "loss": 0.6069, + "step": 3295 + }, + { + "epoch": 2.728879072335726, + "grad_norm": 0.9154210090637207, + "learning_rate": 2.9105159308022705e-06, + "loss": 0.5741, + "step": 3296 + }, + { + "epoch": 2.729707344008835, + "grad_norm": 0.9545791149139404, + "learning_rate": 2.909431128906367e-06, + "loss": 0.5995, + "step": 3297 + }, + { + "epoch": 2.730535615681944, + "grad_norm": 0.9200988411903381, + "learning_rate": 2.908346247789356e-06, + "loss": 0.6152, + "step": 3298 + }, + { + "epoch": 2.7313638873550525, + "grad_norm": 0.9288673996925354, + "learning_rate": 2.9072612876611524e-06, + "loss": 0.5904, + "step": 3299 + }, + { + "epoch": 2.7321921590281613, + "grad_norm": 0.918323278427124, + "learning_rate": 2.906176248731686e-06, + "loss": 0.5999, + "step": 3300 + }, + { + "epoch": 2.73302043070127, + "grad_norm": 0.936652421951294, + "learning_rate": 2.905091131210903e-06, + "loss": 0.5928, + "step": 3301 + }, + { + "epoch": 2.7338487023743787, + "grad_norm": 0.8942639827728271, + "learning_rate": 2.9040059353087614e-06, + "loss": 0.5739, + "step": 3302 + }, + { + "epoch": 2.7346769740474874, + "grad_norm": 0.9197317957878113, + "learning_rate": 2.902920661235238e-06, + "loss": 0.5776, + "step": 3303 + }, + { + "epoch": 2.7355052457205966, + "grad_norm": 0.895505428314209, + "learning_rate": 2.9018353092003216e-06, + "loss": 0.6067, + "step": 3304 + }, + { + "epoch": 2.736333517393705, + "grad_norm": 0.947441816329956, + "learning_rate": 2.9007498794140193e-06, + "loss": 0.6072, + "step": 3305 + }, + { + "epoch": 2.737161789066814, + "grad_norm": 0.9008322358131409, + "learning_rate": 2.8996643720863515e-06, + "loss": 0.5908, + "step": 3306 + }, + { + "epoch": 2.737990060739923, + "grad_norm": 0.9137061238288879, + "learning_rate": 2.898578787427353e-06, + "loss": 0.607, + "step": 3307 + }, + { + "epoch": 2.7388183324130315, + "grad_norm": 0.9126008749008179, + "learning_rate": 2.8974931256470745e-06, + "loss": 0.5809, + "step": 3308 + }, + { + "epoch": 2.7396466040861402, + "grad_norm": 0.9215322136878967, + "learning_rate": 2.896407386955582e-06, + "loss": 0.5807, + "step": 3309 + }, + { + "epoch": 2.740474875759249, + "grad_norm": 0.8903998136520386, + "learning_rate": 2.8953215715629547e-06, + "loss": 0.5968, + "step": 3310 + }, + { + "epoch": 2.7413031474323577, + "grad_norm": 0.9156417846679688, + "learning_rate": 2.8942356796792888e-06, + "loss": 0.5809, + "step": 3311 + }, + { + "epoch": 2.7421314191054664, + "grad_norm": 0.952264666557312, + "learning_rate": 2.893149711514694e-06, + "loss": 0.5818, + "step": 3312 + }, + { + "epoch": 2.7429596907785756, + "grad_norm": 0.9201667904853821, + "learning_rate": 2.8920636672792945e-06, + "loss": 0.5789, + "step": 3313 + }, + { + "epoch": 2.743787962451684, + "grad_norm": 0.9581197500228882, + "learning_rate": 2.890977547183231e-06, + "loss": 0.6007, + "step": 3314 + }, + { + "epoch": 2.744616234124793, + "grad_norm": 0.9360357522964478, + "learning_rate": 2.889891351436656e-06, + "loss": 0.6096, + "step": 3315 + }, + { + "epoch": 2.745444505797902, + "grad_norm": 0.9209451079368591, + "learning_rate": 2.8888050802497388e-06, + "loss": 0.6015, + "step": 3316 + }, + { + "epoch": 2.7462727774710105, + "grad_norm": 0.9496802091598511, + "learning_rate": 2.8877187338326634e-06, + "loss": 0.5802, + "step": 3317 + }, + { + "epoch": 2.7471010491441192, + "grad_norm": 0.9407662153244019, + "learning_rate": 2.8866323123956274e-06, + "loss": 0.588, + "step": 3318 + }, + { + "epoch": 2.747929320817228, + "grad_norm": 0.9603942036628723, + "learning_rate": 2.885545816148843e-06, + "loss": 0.5759, + "step": 3319 + }, + { + "epoch": 2.7487575924903367, + "grad_norm": 0.930496335029602, + "learning_rate": 2.8844592453025372e-06, + "loss": 0.5916, + "step": 3320 + }, + { + "epoch": 2.7495858641634454, + "grad_norm": 0.8831289410591125, + "learning_rate": 2.883372600066951e-06, + "loss": 0.5889, + "step": 3321 + }, + { + "epoch": 2.7504141358365546, + "grad_norm": 0.9076021909713745, + "learning_rate": 2.8822858806523407e-06, + "loss": 0.5842, + "step": 3322 + }, + { + "epoch": 2.7512424075096633, + "grad_norm": 0.9093723297119141, + "learning_rate": 2.8811990872689764e-06, + "loss": 0.584, + "step": 3323 + }, + { + "epoch": 2.752070679182772, + "grad_norm": 0.9458228349685669, + "learning_rate": 2.880112220127142e-06, + "loss": 0.5812, + "step": 3324 + }, + { + "epoch": 2.7528989508558808, + "grad_norm": 0.9406691789627075, + "learning_rate": 2.8790252794371365e-06, + "loss": 0.5889, + "step": 3325 + }, + { + "epoch": 2.7537272225289895, + "grad_norm": 0.9094856381416321, + "learning_rate": 2.8779382654092726e-06, + "loss": 0.5919, + "step": 3326 + }, + { + "epoch": 2.754555494202098, + "grad_norm": 0.9350448250770569, + "learning_rate": 2.8768511782538766e-06, + "loss": 0.5738, + "step": 3327 + }, + { + "epoch": 2.755383765875207, + "grad_norm": 0.9427900314331055, + "learning_rate": 2.875764018181291e-06, + "loss": 0.5962, + "step": 3328 + }, + { + "epoch": 2.756212037548316, + "grad_norm": 0.9439454674720764, + "learning_rate": 2.8746767854018715e-06, + "loss": 0.5994, + "step": 3329 + }, + { + "epoch": 2.7570403092214244, + "grad_norm": 0.9291850328445435, + "learning_rate": 2.8735894801259863e-06, + "loss": 0.6055, + "step": 3330 + }, + { + "epoch": 2.7578685808945336, + "grad_norm": 0.9287533164024353, + "learning_rate": 2.872502102564019e-06, + "loss": 0.5798, + "step": 3331 + }, + { + "epoch": 2.7586968525676423, + "grad_norm": 0.8941923379898071, + "learning_rate": 2.871414652926367e-06, + "loss": 0.6066, + "step": 3332 + }, + { + "epoch": 2.759525124240751, + "grad_norm": 0.924333930015564, + "learning_rate": 2.8703271314234425e-06, + "loss": 0.5872, + "step": 3333 + }, + { + "epoch": 2.7603533959138598, + "grad_norm": 0.8994147777557373, + "learning_rate": 2.8692395382656702e-06, + "loss": 0.5906, + "step": 3334 + }, + { + "epoch": 2.7611816675869685, + "grad_norm": 0.9152485132217407, + "learning_rate": 2.8681518736634895e-06, + "loss": 0.5774, + "step": 3335 + }, + { + "epoch": 2.762009939260077, + "grad_norm": 0.9269137978553772, + "learning_rate": 2.867064137827353e-06, + "loss": 0.5759, + "step": 3336 + }, + { + "epoch": 2.762838210933186, + "grad_norm": 0.916441023349762, + "learning_rate": 2.8659763309677267e-06, + "loss": 0.5927, + "step": 3337 + }, + { + "epoch": 2.763666482606295, + "grad_norm": 0.9280357956886292, + "learning_rate": 2.8648884532950925e-06, + "loss": 0.6004, + "step": 3338 + }, + { + "epoch": 2.7644947542794034, + "grad_norm": 0.9075552821159363, + "learning_rate": 2.8638005050199443e-06, + "loss": 0.6185, + "step": 3339 + }, + { + "epoch": 2.7653230259525126, + "grad_norm": 0.9340760707855225, + "learning_rate": 2.8627124863527895e-06, + "loss": 0.6119, + "step": 3340 + }, + { + "epoch": 2.7661512976256213, + "grad_norm": 0.9393087029457092, + "learning_rate": 2.8616243975041495e-06, + "loss": 0.6058, + "step": 3341 + }, + { + "epoch": 2.76697956929873, + "grad_norm": 0.9380460381507874, + "learning_rate": 2.8605362386845605e-06, + "loss": 0.5955, + "step": 3342 + }, + { + "epoch": 2.7678078409718387, + "grad_norm": 0.9393378496170044, + "learning_rate": 2.8594480101045683e-06, + "loss": 0.5841, + "step": 3343 + }, + { + "epoch": 2.7686361126449475, + "grad_norm": 0.9186044335365295, + "learning_rate": 2.8583597119747385e-06, + "loss": 0.5977, + "step": 3344 + }, + { + "epoch": 2.769464384318056, + "grad_norm": 0.9224127531051636, + "learning_rate": 2.8572713445056448e-06, + "loss": 0.5863, + "step": 3345 + }, + { + "epoch": 2.770292655991165, + "grad_norm": 0.9517519474029541, + "learning_rate": 2.8561829079078764e-06, + "loss": 0.5748, + "step": 3346 + }, + { + "epoch": 2.771120927664274, + "grad_norm": 0.9624918103218079, + "learning_rate": 2.855094402392036e-06, + "loss": 0.588, + "step": 3347 + }, + { + "epoch": 2.771949199337383, + "grad_norm": 0.9524730443954468, + "learning_rate": 2.8540058281687375e-06, + "loss": 0.6032, + "step": 3348 + }, + { + "epoch": 2.7727774710104915, + "grad_norm": 0.9574846029281616, + "learning_rate": 2.852917185448612e-06, + "loss": 0.5915, + "step": 3349 + }, + { + "epoch": 2.7736057426836003, + "grad_norm": 0.9290011525154114, + "learning_rate": 2.851828474442301e-06, + "loss": 0.5715, + "step": 3350 + }, + { + "epoch": 2.774434014356709, + "grad_norm": 0.9393135905265808, + "learning_rate": 2.8507396953604597e-06, + "loss": 0.5855, + "step": 3351 + }, + { + "epoch": 2.7752622860298177, + "grad_norm": 0.9188078045845032, + "learning_rate": 2.8496508484137574e-06, + "loss": 0.598, + "step": 3352 + }, + { + "epoch": 2.7760905577029265, + "grad_norm": 0.9129231572151184, + "learning_rate": 2.8485619338128755e-06, + "loss": 0.6076, + "step": 3353 + }, + { + "epoch": 2.776918829376035, + "grad_norm": 0.941085934638977, + "learning_rate": 2.847472951768508e-06, + "loss": 0.5787, + "step": 3354 + }, + { + "epoch": 2.777747101049144, + "grad_norm": 1.0103200674057007, + "learning_rate": 2.846383902491363e-06, + "loss": 0.606, + "step": 3355 + }, + { + "epoch": 2.778575372722253, + "grad_norm": 0.9307713508605957, + "learning_rate": 2.845294786192163e-06, + "loss": 0.5969, + "step": 3356 + }, + { + "epoch": 2.779403644395362, + "grad_norm": 0.890837550163269, + "learning_rate": 2.84420560308164e-06, + "loss": 0.5826, + "step": 3357 + }, + { + "epoch": 2.7802319160684705, + "grad_norm": 0.8857460618019104, + "learning_rate": 2.843116353370542e-06, + "loss": 0.6005, + "step": 3358 + }, + { + "epoch": 2.7810601877415793, + "grad_norm": 0.8894275426864624, + "learning_rate": 2.8420270372696273e-06, + "loss": 0.6017, + "step": 3359 + }, + { + "epoch": 2.781888459414688, + "grad_norm": 0.9014497995376587, + "learning_rate": 2.8409376549896693e-06, + "loss": 0.5845, + "step": 3360 + }, + { + "epoch": 2.7827167310877967, + "grad_norm": 0.9019520282745361, + "learning_rate": 2.839848206741454e-06, + "loss": 0.5993, + "step": 3361 + }, + { + "epoch": 2.7835450027609054, + "grad_norm": 0.9594439268112183, + "learning_rate": 2.838758692735778e-06, + "loss": 0.5905, + "step": 3362 + }, + { + "epoch": 2.7843732744340146, + "grad_norm": 0.9144065976142883, + "learning_rate": 2.8376691131834525e-06, + "loss": 0.6006, + "step": 3363 + }, + { + "epoch": 2.785201546107123, + "grad_norm": 0.9487935900688171, + "learning_rate": 2.836579468295303e-06, + "loss": 0.5832, + "step": 3364 + }, + { + "epoch": 2.786029817780232, + "grad_norm": 0.8950932025909424, + "learning_rate": 2.835489758282163e-06, + "loss": 0.5792, + "step": 3365 + }, + { + "epoch": 2.786858089453341, + "grad_norm": 0.9522806406021118, + "learning_rate": 2.8343999833548814e-06, + "loss": 0.5891, + "step": 3366 + }, + { + "epoch": 2.7876863611264495, + "grad_norm": 0.964898407459259, + "learning_rate": 2.833310143724321e-06, + "loss": 0.572, + "step": 3367 + }, + { + "epoch": 2.7885146327995582, + "grad_norm": 0.9154083728790283, + "learning_rate": 2.832220239601355e-06, + "loss": 0.5982, + "step": 3368 + }, + { + "epoch": 2.789342904472667, + "grad_norm": 0.9093793034553528, + "learning_rate": 2.831130271196869e-06, + "loss": 0.5592, + "step": 3369 + }, + { + "epoch": 2.7901711761457757, + "grad_norm": 0.9114746451377869, + "learning_rate": 2.8300402387217628e-06, + "loss": 0.5531, + "step": 3370 + }, + { + "epoch": 2.7909994478188844, + "grad_norm": 0.959976077079773, + "learning_rate": 2.8289501423869464e-06, + "loss": 0.5932, + "step": 3371 + }, + { + "epoch": 2.7918277194919936, + "grad_norm": 0.9415104389190674, + "learning_rate": 2.8278599824033443e-06, + "loss": 0.5516, + "step": 3372 + }, + { + "epoch": 2.792655991165102, + "grad_norm": 0.9304796457290649, + "learning_rate": 2.826769758981892e-06, + "loss": 0.5786, + "step": 3373 + }, + { + "epoch": 2.793484262838211, + "grad_norm": 0.9369680285453796, + "learning_rate": 2.825679472333537e-06, + "loss": 0.6077, + "step": 3374 + }, + { + "epoch": 2.79431253451132, + "grad_norm": 0.9102757573127747, + "learning_rate": 2.8245891226692405e-06, + "loss": 0.606, + "step": 3375 + }, + { + "epoch": 2.7951408061844285, + "grad_norm": 0.921359121799469, + "learning_rate": 2.8234987101999747e-06, + "loss": 0.613, + "step": 3376 + }, + { + "epoch": 2.7959690778575372, + "grad_norm": 0.9242355227470398, + "learning_rate": 2.822408235136723e-06, + "loss": 0.5758, + "step": 3377 + }, + { + "epoch": 2.796797349530646, + "grad_norm": 0.9175317287445068, + "learning_rate": 2.8213176976904844e-06, + "loss": 0.5744, + "step": 3378 + }, + { + "epoch": 2.7976256212037547, + "grad_norm": 0.9504632949829102, + "learning_rate": 2.8202270980722663e-06, + "loss": 0.5871, + "step": 3379 + }, + { + "epoch": 2.7984538928768634, + "grad_norm": 0.9071774482727051, + "learning_rate": 2.8191364364930895e-06, + "loss": 0.6003, + "step": 3380 + }, + { + "epoch": 2.7992821645499726, + "grad_norm": 0.913815975189209, + "learning_rate": 2.8180457131639878e-06, + "loss": 0.596, + "step": 3381 + }, + { + "epoch": 2.8001104362230813, + "grad_norm": 0.9608022570610046, + "learning_rate": 2.8169549282960044e-06, + "loss": 0.6055, + "step": 3382 + }, + { + "epoch": 2.80093870789619, + "grad_norm": 2.0446290969848633, + "learning_rate": 2.815864082100198e-06, + "loss": 0.565, + "step": 3383 + }, + { + "epoch": 2.8017669795692988, + "grad_norm": 0.9219574928283691, + "learning_rate": 2.814773174787636e-06, + "loss": 0.5808, + "step": 3384 + }, + { + "epoch": 2.8025952512424075, + "grad_norm": 0.9542830586433411, + "learning_rate": 2.8136822065693987e-06, + "loss": 0.5992, + "step": 3385 + }, + { + "epoch": 2.803423522915516, + "grad_norm": 0.9478740096092224, + "learning_rate": 2.8125911776565785e-06, + "loss": 0.5705, + "step": 3386 + }, + { + "epoch": 2.804251794588625, + "grad_norm": 0.9299196600914001, + "learning_rate": 2.8115000882602792e-06, + "loss": 0.5773, + "step": 3387 + }, + { + "epoch": 2.805080066261734, + "grad_norm": 0.9167691469192505, + "learning_rate": 2.8104089385916168e-06, + "loss": 0.5789, + "step": 3388 + }, + { + "epoch": 2.8059083379348424, + "grad_norm": 0.9034630656242371, + "learning_rate": 2.8093177288617187e-06, + "loss": 0.5837, + "step": 3389 + }, + { + "epoch": 2.8067366096079516, + "grad_norm": 0.9147744178771973, + "learning_rate": 2.808226459281723e-06, + "loss": 0.5826, + "step": 3390 + }, + { + "epoch": 2.8075648812810603, + "grad_norm": 0.9159218072891235, + "learning_rate": 2.8071351300627803e-06, + "loss": 0.5931, + "step": 3391 + }, + { + "epoch": 2.808393152954169, + "grad_norm": 0.9260812401771545, + "learning_rate": 2.8060437414160546e-06, + "loss": 0.6182, + "step": 3392 + }, + { + "epoch": 2.8092214246272778, + "grad_norm": 0.8975473642349243, + "learning_rate": 2.804952293552717e-06, + "loss": 0.5786, + "step": 3393 + }, + { + "epoch": 2.8100496963003865, + "grad_norm": 0.9130522608757019, + "learning_rate": 2.8038607866839522e-06, + "loss": 0.572, + "step": 3394 + }, + { + "epoch": 2.810877967973495, + "grad_norm": 0.9170857667922974, + "learning_rate": 2.802769221020959e-06, + "loss": 0.569, + "step": 3395 + }, + { + "epoch": 2.811706239646604, + "grad_norm": 0.9313328266143799, + "learning_rate": 2.801677596774944e-06, + "loss": 0.568, + "step": 3396 + }, + { + "epoch": 2.812534511319713, + "grad_norm": 0.941814124584198, + "learning_rate": 2.8005859141571257e-06, + "loss": 0.5852, + "step": 3397 + }, + { + "epoch": 2.8133627829928214, + "grad_norm": 0.9179267883300781, + "learning_rate": 2.799494173378735e-06, + "loss": 0.5887, + "step": 3398 + }, + { + "epoch": 2.8141910546659306, + "grad_norm": 0.9193063378334045, + "learning_rate": 2.7984023746510135e-06, + "loss": 0.588, + "step": 3399 + }, + { + "epoch": 2.8150193263390393, + "grad_norm": 0.9381787776947021, + "learning_rate": 2.7973105181852145e-06, + "loss": 0.5938, + "step": 3400 + }, + { + "epoch": 2.815847598012148, + "grad_norm": 0.9631656408309937, + "learning_rate": 2.7962186041926015e-06, + "loss": 0.5828, + "step": 3401 + }, + { + "epoch": 2.8166758696852567, + "grad_norm": 0.9512215852737427, + "learning_rate": 2.7951266328844495e-06, + "loss": 0.5836, + "step": 3402 + }, + { + "epoch": 2.8175041413583655, + "grad_norm": 0.9171256422996521, + "learning_rate": 2.794034604472047e-06, + "loss": 0.574, + "step": 3403 + }, + { + "epoch": 2.818332413031474, + "grad_norm": 0.9122748374938965, + "learning_rate": 2.792942519166688e-06, + "loss": 0.6024, + "step": 3404 + }, + { + "epoch": 2.819160684704583, + "grad_norm": 0.9503622651100159, + "learning_rate": 2.7918503771796817e-06, + "loss": 0.5743, + "step": 3405 + }, + { + "epoch": 2.819988956377692, + "grad_norm": 0.9348991513252258, + "learning_rate": 2.7907581787223494e-06, + "loss": 0.5824, + "step": 3406 + }, + { + "epoch": 2.820817228050801, + "grad_norm": 1.0846208333969116, + "learning_rate": 2.7896659240060192e-06, + "loss": 0.5995, + "step": 3407 + }, + { + "epoch": 2.8216454997239095, + "grad_norm": 0.9554671049118042, + "learning_rate": 2.7885736132420326e-06, + "loss": 0.5903, + "step": 3408 + }, + { + "epoch": 2.8224737713970183, + "grad_norm": 0.9836533665657043, + "learning_rate": 2.7874812466417434e-06, + "loss": 0.5982, + "step": 3409 + }, + { + "epoch": 2.823302043070127, + "grad_norm": 0.9414785504341125, + "learning_rate": 2.786388824416511e-06, + "loss": 0.5743, + "step": 3410 + }, + { + "epoch": 2.8241303147432357, + "grad_norm": 0.9228699207305908, + "learning_rate": 2.7852963467777117e-06, + "loss": 0.5879, + "step": 3411 + }, + { + "epoch": 2.8249585864163445, + "grad_norm": 0.919826090335846, + "learning_rate": 2.7842038139367288e-06, + "loss": 0.6023, + "step": 3412 + }, + { + "epoch": 2.825786858089453, + "grad_norm": 0.9112126231193542, + "learning_rate": 2.783111226104957e-06, + "loss": 0.578, + "step": 3413 + }, + { + "epoch": 2.826615129762562, + "grad_norm": 0.9262592792510986, + "learning_rate": 2.7820185834938024e-06, + "loss": 0.6063, + "step": 3414 + }, + { + "epoch": 2.827443401435671, + "grad_norm": 0.9020029902458191, + "learning_rate": 2.780925886314681e-06, + "loss": 0.5837, + "step": 3415 + }, + { + "epoch": 2.82827167310878, + "grad_norm": 0.9186374545097351, + "learning_rate": 2.7798331347790185e-06, + "loss": 0.5801, + "step": 3416 + }, + { + "epoch": 2.8290999447818885, + "grad_norm": 0.9116926193237305, + "learning_rate": 2.7787403290982535e-06, + "loss": 0.5735, + "step": 3417 + }, + { + "epoch": 2.8299282164549973, + "grad_norm": 0.9124463796615601, + "learning_rate": 2.7776474694838333e-06, + "loss": 0.6064, + "step": 3418 + }, + { + "epoch": 2.830756488128106, + "grad_norm": 0.9450061321258545, + "learning_rate": 2.7765545561472153e-06, + "loss": 0.5885, + "step": 3419 + }, + { + "epoch": 2.8315847598012147, + "grad_norm": 0.8993310928344727, + "learning_rate": 2.77546158929987e-06, + "loss": 0.5767, + "step": 3420 + }, + { + "epoch": 2.8324130314743234, + "grad_norm": 0.9315094947814941, + "learning_rate": 2.7743685691532734e-06, + "loss": 0.595, + "step": 3421 + }, + { + "epoch": 2.8332413031474326, + "grad_norm": 0.9254942536354065, + "learning_rate": 2.773275495918917e-06, + "loss": 0.6071, + "step": 3422 + }, + { + "epoch": 2.834069574820541, + "grad_norm": 0.9551193118095398, + "learning_rate": 2.7721823698083e-06, + "loss": 0.588, + "step": 3423 + }, + { + "epoch": 2.83489784649365, + "grad_norm": 0.9356088042259216, + "learning_rate": 2.7710891910329317e-06, + "loss": 0.5917, + "step": 3424 + }, + { + "epoch": 2.835726118166759, + "grad_norm": 0.9254491925239563, + "learning_rate": 2.7699959598043315e-06, + "loss": 0.5754, + "step": 3425 + }, + { + "epoch": 2.8365543898398675, + "grad_norm": 0.928747832775116, + "learning_rate": 2.7689026763340302e-06, + "loss": 0.5925, + "step": 3426 + }, + { + "epoch": 2.8373826615129762, + "grad_norm": 0.9232720732688904, + "learning_rate": 2.7678093408335677e-06, + "loss": 0.594, + "step": 3427 + }, + { + "epoch": 2.838210933186085, + "grad_norm": 0.9319408535957336, + "learning_rate": 2.766715953514494e-06, + "loss": 0.607, + "step": 3428 + }, + { + "epoch": 2.8390392048591937, + "grad_norm": 0.9321261644363403, + "learning_rate": 2.76562251458837e-06, + "loss": 0.5839, + "step": 3429 + }, + { + "epoch": 2.8398674765323024, + "grad_norm": 0.8851343393325806, + "learning_rate": 2.7645290242667648e-06, + "loss": 0.5929, + "step": 3430 + }, + { + "epoch": 2.8406957482054116, + "grad_norm": 0.9318198561668396, + "learning_rate": 2.76343548276126e-06, + "loss": 0.5758, + "step": 3431 + }, + { + "epoch": 2.84152401987852, + "grad_norm": 0.9212138056755066, + "learning_rate": 2.7623418902834455e-06, + "loss": 0.5689, + "step": 3432 + }, + { + "epoch": 2.842352291551629, + "grad_norm": 0.8917051553726196, + "learning_rate": 2.761248247044919e-06, + "loss": 0.5915, + "step": 3433 + }, + { + "epoch": 2.843180563224738, + "grad_norm": 0.9266699552536011, + "learning_rate": 2.760154553257294e-06, + "loss": 0.5643, + "step": 3434 + }, + { + "epoch": 2.8440088348978465, + "grad_norm": 0.9837039709091187, + "learning_rate": 2.7590608091321867e-06, + "loss": 0.6135, + "step": 3435 + }, + { + "epoch": 2.8448371065709552, + "grad_norm": 0.9872124195098877, + "learning_rate": 2.7579670148812278e-06, + "loss": 0.5649, + "step": 3436 + }, + { + "epoch": 2.845665378244064, + "grad_norm": 0.9581724405288696, + "learning_rate": 2.756873170716057e-06, + "loss": 0.5833, + "step": 3437 + }, + { + "epoch": 2.8464936499171727, + "grad_norm": 0.9363225698471069, + "learning_rate": 2.7557792768483223e-06, + "loss": 0.5852, + "step": 3438 + }, + { + "epoch": 2.8473219215902814, + "grad_norm": 0.9177983403205872, + "learning_rate": 2.7546853334896813e-06, + "loss": 0.5959, + "step": 3439 + }, + { + "epoch": 2.8481501932633906, + "grad_norm": 0.9081945419311523, + "learning_rate": 2.7535913408518023e-06, + "loss": 0.5862, + "step": 3440 + }, + { + "epoch": 2.8489784649364993, + "grad_norm": 0.928897500038147, + "learning_rate": 2.7524972991463626e-06, + "loss": 0.5978, + "step": 3441 + }, + { + "epoch": 2.849806736609608, + "grad_norm": 0.9136974811553955, + "learning_rate": 2.75140320858505e-06, + "loss": 0.5926, + "step": 3442 + }, + { + "epoch": 2.8506350082827168, + "grad_norm": 0.9227224588394165, + "learning_rate": 2.7503090693795605e-06, + "loss": 0.6073, + "step": 3443 + }, + { + "epoch": 2.8514632799558255, + "grad_norm": 0.9390876889228821, + "learning_rate": 2.749214881741598e-06, + "loss": 0.5757, + "step": 3444 + }, + { + "epoch": 2.852291551628934, + "grad_norm": 0.9460882544517517, + "learning_rate": 2.74812064588288e-06, + "loss": 0.6172, + "step": 3445 + }, + { + "epoch": 2.853119823302043, + "grad_norm": 0.9087659120559692, + "learning_rate": 2.74702636201513e-06, + "loss": 0.5937, + "step": 3446 + }, + { + "epoch": 2.853948094975152, + "grad_norm": 0.8995270133018494, + "learning_rate": 2.745932030350081e-06, + "loss": 0.5798, + "step": 3447 + }, + { + "epoch": 2.8547763666482604, + "grad_norm": 0.9307460188865662, + "learning_rate": 2.744837651099478e-06, + "loss": 0.6009, + "step": 3448 + }, + { + "epoch": 2.8556046383213696, + "grad_norm": 0.9634310007095337, + "learning_rate": 2.7437432244750705e-06, + "loss": 0.5985, + "step": 3449 + }, + { + "epoch": 2.8564329099944783, + "grad_norm": 0.9647858142852783, + "learning_rate": 2.742648750688622e-06, + "loss": 0.5718, + "step": 3450 + }, + { + "epoch": 2.857261181667587, + "grad_norm": 0.9150539040565491, + "learning_rate": 2.7415542299519016e-06, + "loss": 0.5964, + "step": 3451 + }, + { + "epoch": 2.8580894533406958, + "grad_norm": 0.9129363298416138, + "learning_rate": 2.7404596624766892e-06, + "loss": 0.5514, + "step": 3452 + }, + { + "epoch": 2.8589177250138045, + "grad_norm": 0.9430226683616638, + "learning_rate": 2.739365048474774e-06, + "loss": 0.5795, + "step": 3453 + }, + { + "epoch": 2.859745996686913, + "grad_norm": 0.9402962327003479, + "learning_rate": 2.738270388157954e-06, + "loss": 0.5728, + "step": 3454 + }, + { + "epoch": 2.860574268360022, + "grad_norm": 0.9169720411300659, + "learning_rate": 2.7371756817380346e-06, + "loss": 0.6171, + "step": 3455 + }, + { + "epoch": 2.861402540033131, + "grad_norm": 0.9437370896339417, + "learning_rate": 2.7360809294268315e-06, + "loss": 0.5827, + "step": 3456 + }, + { + "epoch": 2.8622308117062394, + "grad_norm": 0.8999599814414978, + "learning_rate": 2.7349861314361693e-06, + "loss": 0.5847, + "step": 3457 + }, + { + "epoch": 2.8630590833793486, + "grad_norm": 0.9228375554084778, + "learning_rate": 2.7338912879778814e-06, + "loss": 0.6069, + "step": 3458 + }, + { + "epoch": 2.8638873550524573, + "grad_norm": 0.9543429017066956, + "learning_rate": 2.732796399263809e-06, + "loss": 0.6128, + "step": 3459 + }, + { + "epoch": 2.864715626725566, + "grad_norm": 0.923698902130127, + "learning_rate": 2.7317014655058038e-06, + "loss": 0.5714, + "step": 3460 + }, + { + "epoch": 2.8655438983986747, + "grad_norm": 0.95343017578125, + "learning_rate": 2.730606486915725e-06, + "loss": 0.5821, + "step": 3461 + }, + { + "epoch": 2.8663721700717835, + "grad_norm": 0.9078912734985352, + "learning_rate": 2.729511463705441e-06, + "loss": 0.5833, + "step": 3462 + }, + { + "epoch": 2.867200441744892, + "grad_norm": 0.9045196771621704, + "learning_rate": 2.7284163960868274e-06, + "loss": 0.5647, + "step": 3463 + }, + { + "epoch": 2.868028713418001, + "grad_norm": 0.9145541787147522, + "learning_rate": 2.727321284271771e-06, + "loss": 0.5731, + "step": 3464 + }, + { + "epoch": 2.86885698509111, + "grad_norm": 0.9013593792915344, + "learning_rate": 2.726226128472166e-06, + "loss": 0.5901, + "step": 3465 + }, + { + "epoch": 2.869685256764219, + "grad_norm": 0.9561366438865662, + "learning_rate": 2.7251309288999123e-06, + "loss": 0.585, + "step": 3466 + }, + { + "epoch": 2.8705135284373275, + "grad_norm": 0.9313666224479675, + "learning_rate": 2.724035685766923e-06, + "loss": 0.5755, + "step": 3467 + }, + { + "epoch": 2.8713418001104363, + "grad_norm": 0.9003223776817322, + "learning_rate": 2.722940399285118e-06, + "loss": 0.5961, + "step": 3468 + }, + { + "epoch": 2.872170071783545, + "grad_norm": 0.9185585975646973, + "learning_rate": 2.721845069666422e-06, + "loss": 0.5741, + "step": 3469 + }, + { + "epoch": 2.8729983434566537, + "grad_norm": 0.9295309782028198, + "learning_rate": 2.7207496971227743e-06, + "loss": 0.588, + "step": 3470 + }, + { + "epoch": 2.8738266151297625, + "grad_norm": 0.9500821232795715, + "learning_rate": 2.7196542818661188e-06, + "loss": 0.6026, + "step": 3471 + }, + { + "epoch": 2.874654886802871, + "grad_norm": 0.9265774488449097, + "learning_rate": 2.718558824108406e-06, + "loss": 0.5857, + "step": 3472 + }, + { + "epoch": 2.87548315847598, + "grad_norm": 0.9159520864486694, + "learning_rate": 2.717463324061599e-06, + "loss": 0.6023, + "step": 3473 + }, + { + "epoch": 2.876311430149089, + "grad_norm": 0.9114859700202942, + "learning_rate": 2.7163677819376653e-06, + "loss": 0.6144, + "step": 3474 + }, + { + "epoch": 2.877139701822198, + "grad_norm": 0.9181182980537415, + "learning_rate": 2.7152721979485824e-06, + "loss": 0.5853, + "step": 3475 + }, + { + "epoch": 2.8779679734953065, + "grad_norm": 0.9219772219657898, + "learning_rate": 2.7141765723063366e-06, + "loss": 0.5824, + "step": 3476 + }, + { + "epoch": 2.8787962451684153, + "grad_norm": 0.925654411315918, + "learning_rate": 2.713080905222921e-06, + "loss": 0.5859, + "step": 3477 + }, + { + "epoch": 2.879624516841524, + "grad_norm": 0.927891731262207, + "learning_rate": 2.711985196910336e-06, + "loss": 0.6047, + "step": 3478 + }, + { + "epoch": 2.8804527885146327, + "grad_norm": 0.9178606271743774, + "learning_rate": 2.7108894475805912e-06, + "loss": 0.5863, + "step": 3479 + }, + { + "epoch": 2.8812810601877414, + "grad_norm": 0.9341709613800049, + "learning_rate": 2.7097936574457046e-06, + "loss": 0.5718, + "step": 3480 + }, + { + "epoch": 2.8821093318608506, + "grad_norm": 0.9023662805557251, + "learning_rate": 2.7086978267177007e-06, + "loss": 0.593, + "step": 3481 + }, + { + "epoch": 2.882937603533959, + "grad_norm": 0.9269591569900513, + "learning_rate": 2.7076019556086136e-06, + "loss": 0.5955, + "step": 3482 + }, + { + "epoch": 2.883765875207068, + "grad_norm": 0.9572223424911499, + "learning_rate": 2.7065060443304823e-06, + "loss": 0.5698, + "step": 3483 + }, + { + "epoch": 2.884594146880177, + "grad_norm": 0.9251832365989685, + "learning_rate": 2.705410093095357e-06, + "loss": 0.5903, + "step": 3484 + }, + { + "epoch": 2.8854224185532855, + "grad_norm": 0.959985077381134, + "learning_rate": 2.7043141021152937e-06, + "loss": 0.5922, + "step": 3485 + }, + { + "epoch": 2.8862506902263942, + "grad_norm": 0.9356866478919983, + "learning_rate": 2.703218071602355e-06, + "loss": 0.576, + "step": 3486 + }, + { + "epoch": 2.887078961899503, + "grad_norm": 0.9041283130645752, + "learning_rate": 2.702122001768615e-06, + "loss": 0.5748, + "step": 3487 + }, + { + "epoch": 2.8879072335726117, + "grad_norm": 0.9056842923164368, + "learning_rate": 2.7010258928261514e-06, + "loss": 0.5795, + "step": 3488 + }, + { + "epoch": 2.8887355052457204, + "grad_norm": 0.9829550385475159, + "learning_rate": 2.6999297449870522e-06, + "loss": 0.6017, + "step": 3489 + }, + { + "epoch": 2.8895637769188296, + "grad_norm": 0.9483397603034973, + "learning_rate": 2.6988335584634103e-06, + "loss": 0.5849, + "step": 3490 + }, + { + "epoch": 2.890392048591938, + "grad_norm": 0.9553128480911255, + "learning_rate": 2.6977373334673286e-06, + "loss": 0.5518, + "step": 3491 + }, + { + "epoch": 2.891220320265047, + "grad_norm": 0.9298998713493347, + "learning_rate": 2.6966410702109153e-06, + "loss": 0.5947, + "step": 3492 + }, + { + "epoch": 2.8920485919381558, + "grad_norm": 0.9143223166465759, + "learning_rate": 2.6955447689062893e-06, + "loss": 0.5757, + "step": 3493 + }, + { + "epoch": 2.8928768636112645, + "grad_norm": 0.9143601059913635, + "learning_rate": 2.6944484297655727e-06, + "loss": 0.5867, + "step": 3494 + }, + { + "epoch": 2.8937051352843732, + "grad_norm": 0.9283237457275391, + "learning_rate": 2.693352053000898e-06, + "loss": 0.5928, + "step": 3495 + }, + { + "epoch": 2.894533406957482, + "grad_norm": 0.8984352946281433, + "learning_rate": 2.692255638824403e-06, + "loss": 0.5904, + "step": 3496 + }, + { + "epoch": 2.8953616786305907, + "grad_norm": 0.9056443572044373, + "learning_rate": 2.6911591874482334e-06, + "loss": 0.5847, + "step": 3497 + }, + { + "epoch": 2.8961899503036994, + "grad_norm": 0.9254133701324463, + "learning_rate": 2.6900626990845442e-06, + "loss": 0.608, + "step": 3498 + }, + { + "epoch": 2.8970182219768086, + "grad_norm": 0.9658727049827576, + "learning_rate": 2.6889661739454946e-06, + "loss": 0.6063, + "step": 3499 + }, + { + "epoch": 2.8978464936499173, + "grad_norm": 0.9350060820579529, + "learning_rate": 2.6878696122432505e-06, + "loss": 0.5902, + "step": 3500 + }, + { + "epoch": 2.898674765323026, + "grad_norm": 0.9070295691490173, + "learning_rate": 2.686773014189989e-06, + "loss": 0.5883, + "step": 3501 + }, + { + "epoch": 2.8995030369961348, + "grad_norm": 0.9471203088760376, + "learning_rate": 2.6856763799978895e-06, + "loss": 0.6138, + "step": 3502 + }, + { + "epoch": 2.9003313086692435, + "grad_norm": 0.901057243347168, + "learning_rate": 2.6845797098791415e-06, + "loss": 0.5925, + "step": 3503 + }, + { + "epoch": 2.901159580342352, + "grad_norm": 0.9290406107902527, + "learning_rate": 2.683483004045941e-06, + "loss": 0.5728, + "step": 3504 + }, + { + "epoch": 2.901987852015461, + "grad_norm": 0.9371906518936157, + "learning_rate": 2.682386262710489e-06, + "loss": 0.5894, + "step": 3505 + }, + { + "epoch": 2.90281612368857, + "grad_norm": 0.9183947443962097, + "learning_rate": 2.6812894860849953e-06, + "loss": 0.5959, + "step": 3506 + }, + { + "epoch": 2.9036443953616784, + "grad_norm": 0.9351047277450562, + "learning_rate": 2.6801926743816765e-06, + "loss": 0.5863, + "step": 3507 + }, + { + "epoch": 2.9044726670347876, + "grad_norm": 0.93602454662323, + "learning_rate": 2.6790958278127536e-06, + "loss": 0.5845, + "step": 3508 + }, + { + "epoch": 2.9053009387078963, + "grad_norm": 0.9618241786956787, + "learning_rate": 2.677998946590459e-06, + "loss": 0.6068, + "step": 3509 + }, + { + "epoch": 2.906129210381005, + "grad_norm": 0.9157661199569702, + "learning_rate": 2.6769020309270277e-06, + "loss": 0.5608, + "step": 3510 + }, + { + "epoch": 2.9069574820541138, + "grad_norm": 0.9157334566116333, + "learning_rate": 2.6758050810347012e-06, + "loss": 0.5883, + "step": 3511 + }, + { + "epoch": 2.9077857537272225, + "grad_norm": 0.9259895086288452, + "learning_rate": 2.674708097125731e-06, + "loss": 0.578, + "step": 3512 + }, + { + "epoch": 2.908614025400331, + "grad_norm": 0.9327818155288696, + "learning_rate": 2.6736110794123727e-06, + "loss": 0.6157, + "step": 3513 + }, + { + "epoch": 2.90944229707344, + "grad_norm": 0.9186248779296875, + "learning_rate": 2.6725140281068882e-06, + "loss": 0.5901, + "step": 3514 + }, + { + "epoch": 2.910270568746549, + "grad_norm": 0.9106549620628357, + "learning_rate": 2.671416943421548e-06, + "loss": 0.5983, + "step": 3515 + }, + { + "epoch": 2.9110988404196574, + "grad_norm": 0.9274003505706787, + "learning_rate": 2.6703198255686275e-06, + "loss": 0.5886, + "step": 3516 + }, + { + "epoch": 2.9119271120927666, + "grad_norm": 0.9015113115310669, + "learning_rate": 2.6692226747604082e-06, + "loss": 0.5979, + "step": 3517 + }, + { + "epoch": 2.9127553837658753, + "grad_norm": 0.9261288046836853, + "learning_rate": 2.6681254912091788e-06, + "loss": 0.5741, + "step": 3518 + }, + { + "epoch": 2.913583655438984, + "grad_norm": 0.8937399983406067, + "learning_rate": 2.6670282751272335e-06, + "loss": 0.5726, + "step": 3519 + }, + { + "epoch": 2.9144119271120927, + "grad_norm": 0.9202954769134521, + "learning_rate": 2.6659310267268744e-06, + "loss": 0.5788, + "step": 3520 + }, + { + "epoch": 2.9152401987852015, + "grad_norm": 0.9400494694709778, + "learning_rate": 2.6648337462204087e-06, + "loss": 0.568, + "step": 3521 + }, + { + "epoch": 2.91606847045831, + "grad_norm": 0.9291467070579529, + "learning_rate": 2.6637364338201497e-06, + "loss": 0.589, + "step": 3522 + }, + { + "epoch": 2.916896742131419, + "grad_norm": 0.9186866879463196, + "learning_rate": 2.662639089738417e-06, + "loss": 0.59, + "step": 3523 + }, + { + "epoch": 2.917725013804528, + "grad_norm": 0.9596429467201233, + "learning_rate": 2.6615417141875367e-06, + "loss": 0.5759, + "step": 3524 + }, + { + "epoch": 2.918553285477637, + "grad_norm": 0.9279852509498596, + "learning_rate": 2.6604443073798395e-06, + "loss": 0.5969, + "step": 3525 + }, + { + "epoch": 2.9193815571507455, + "grad_norm": 0.9540870189666748, + "learning_rate": 2.6593468695276657e-06, + "loss": 0.5696, + "step": 3526 + }, + { + "epoch": 2.9202098288238543, + "grad_norm": 0.9415677785873413, + "learning_rate": 2.658249400843358e-06, + "loss": 0.5973, + "step": 3527 + }, + { + "epoch": 2.921038100496963, + "grad_norm": 0.8950624465942383, + "learning_rate": 2.657151901539266e-06, + "loss": 0.5952, + "step": 3528 + }, + { + "epoch": 2.9218663721700717, + "grad_norm": 0.8927432298660278, + "learning_rate": 2.656054371827746e-06, + "loss": 0.59, + "step": 3529 + }, + { + "epoch": 2.9226946438431805, + "grad_norm": 0.9516159892082214, + "learning_rate": 2.65495681192116e-06, + "loss": 0.6024, + "step": 3530 + }, + { + "epoch": 2.923522915516289, + "grad_norm": 1.0094369649887085, + "learning_rate": 2.653859222031875e-06, + "loss": 0.596, + "step": 3531 + }, + { + "epoch": 2.924351187189398, + "grad_norm": 0.9162117838859558, + "learning_rate": 2.6527616023722647e-06, + "loss": 0.5739, + "step": 3532 + }, + { + "epoch": 2.925179458862507, + "grad_norm": 0.9844574332237244, + "learning_rate": 2.651663953154709e-06, + "loss": 0.6028, + "step": 3533 + }, + { + "epoch": 2.926007730535616, + "grad_norm": 0.9091664552688599, + "learning_rate": 2.6505662745915915e-06, + "loss": 0.5867, + "step": 3534 + }, + { + "epoch": 2.9268360022087245, + "grad_norm": 0.9107911586761475, + "learning_rate": 2.649468566895304e-06, + "loss": 0.5985, + "step": 3535 + }, + { + "epoch": 2.9276642738818333, + "grad_norm": 0.9392855167388916, + "learning_rate": 2.648370830278241e-06, + "loss": 0.5995, + "step": 3536 + }, + { + "epoch": 2.928492545554942, + "grad_norm": 0.9228532910346985, + "learning_rate": 2.6472730649528068e-06, + "loss": 0.5928, + "step": 3537 + }, + { + "epoch": 2.9293208172280507, + "grad_norm": 0.9555805325508118, + "learning_rate": 2.646175271131407e-06, + "loss": 0.606, + "step": 3538 + }, + { + "epoch": 2.9301490889011594, + "grad_norm": 0.9101611971855164, + "learning_rate": 2.6450774490264547e-06, + "loss": 0.5881, + "step": 3539 + }, + { + "epoch": 2.9309773605742686, + "grad_norm": 0.9572548866271973, + "learning_rate": 2.6439795988503685e-06, + "loss": 0.5861, + "step": 3540 + }, + { + "epoch": 2.931805632247377, + "grad_norm": 0.9441561102867126, + "learning_rate": 2.6428817208155728e-06, + "loss": 0.594, + "step": 3541 + }, + { + "epoch": 2.932633903920486, + "grad_norm": 0.9282614588737488, + "learning_rate": 2.6417838151344945e-06, + "loss": 0.5991, + "step": 3542 + }, + { + "epoch": 2.933462175593595, + "grad_norm": 0.9630649089813232, + "learning_rate": 2.640685882019571e-06, + "loss": 0.6147, + "step": 3543 + }, + { + "epoch": 2.9342904472667035, + "grad_norm": 0.9284330606460571, + "learning_rate": 2.63958792168324e-06, + "loss": 0.6021, + "step": 3544 + }, + { + "epoch": 2.9351187189398122, + "grad_norm": 0.9337431788444519, + "learning_rate": 2.638489934337948e-06, + "loss": 0.602, + "step": 3545 + }, + { + "epoch": 2.935946990612921, + "grad_norm": 0.8977872729301453, + "learning_rate": 2.637391920196144e-06, + "loss": 0.5901, + "step": 3546 + }, + { + "epoch": 2.9367752622860297, + "grad_norm": 0.962690532207489, + "learning_rate": 2.636293879470284e-06, + "loss": 0.5793, + "step": 3547 + }, + { + "epoch": 2.9376035339591384, + "grad_norm": 0.8999794721603394, + "learning_rate": 2.6351958123728294e-06, + "loss": 0.6027, + "step": 3548 + }, + { + "epoch": 2.9384318056322476, + "grad_norm": 0.9200299382209778, + "learning_rate": 2.6340977191162457e-06, + "loss": 0.5985, + "step": 3549 + }, + { + "epoch": 2.9392600773053563, + "grad_norm": 0.9161362648010254, + "learning_rate": 2.6329995999130032e-06, + "loss": 0.595, + "step": 3550 + }, + { + "epoch": 2.940088348978465, + "grad_norm": 0.9192216992378235, + "learning_rate": 2.631901454975578e-06, + "loss": 0.5844, + "step": 3551 + }, + { + "epoch": 2.9409166206515738, + "grad_norm": 0.9136098623275757, + "learning_rate": 2.6308032845164503e-06, + "loss": 0.5932, + "step": 3552 + }, + { + "epoch": 2.9417448923246825, + "grad_norm": 0.9625688791275024, + "learning_rate": 2.629705088748106e-06, + "loss": 0.5811, + "step": 3553 + }, + { + "epoch": 2.9425731639977912, + "grad_norm": 0.9300727248191833, + "learning_rate": 2.6286068678830375e-06, + "loss": 0.6035, + "step": 3554 + }, + { + "epoch": 2.9434014356709, + "grad_norm": 0.9947618246078491, + "learning_rate": 2.627508622133738e-06, + "loss": 0.5822, + "step": 3555 + }, + { + "epoch": 2.9442297073440087, + "grad_norm": 0.9255596995353699, + "learning_rate": 2.626410351712709e-06, + "loss": 0.5824, + "step": 3556 + }, + { + "epoch": 2.9450579790171174, + "grad_norm": 0.9395889043807983, + "learning_rate": 2.6253120568324557e-06, + "loss": 0.6121, + "step": 3557 + }, + { + "epoch": 2.9458862506902266, + "grad_norm": 0.9438552856445312, + "learning_rate": 2.6242137377054865e-06, + "loss": 0.5843, + "step": 3558 + }, + { + "epoch": 2.9467145223633353, + "grad_norm": 0.9153048992156982, + "learning_rate": 2.623115394544318e-06, + "loss": 0.5893, + "step": 3559 + }, + { + "epoch": 2.947542794036444, + "grad_norm": 0.9119593501091003, + "learning_rate": 2.6220170275614677e-06, + "loss": 0.6047, + "step": 3560 + }, + { + "epoch": 2.9483710657095528, + "grad_norm": 0.9174615144729614, + "learning_rate": 2.6209186369694606e-06, + "loss": 0.5949, + "step": 3561 + }, + { + "epoch": 2.9491993373826615, + "grad_norm": 0.9030957818031311, + "learning_rate": 2.619820222980824e-06, + "loss": 0.6146, + "step": 3562 + }, + { + "epoch": 2.95002760905577, + "grad_norm": 0.9036881923675537, + "learning_rate": 2.618721785808091e-06, + "loss": 0.5946, + "step": 3563 + }, + { + "epoch": 2.950855880728879, + "grad_norm": 0.8928536772727966, + "learning_rate": 2.617623325663799e-06, + "loss": 0.5679, + "step": 3564 + }, + { + "epoch": 2.951684152401988, + "grad_norm": 0.9225850701332092, + "learning_rate": 2.6165248427604907e-06, + "loss": 0.5938, + "step": 3565 + }, + { + "epoch": 2.9525124240750964, + "grad_norm": 0.9023733139038086, + "learning_rate": 2.6154263373107113e-06, + "loss": 0.5896, + "step": 3566 + }, + { + "epoch": 2.9533406957482056, + "grad_norm": 0.9055750370025635, + "learning_rate": 2.6143278095270113e-06, + "loss": 0.6007, + "step": 3567 + }, + { + "epoch": 2.9541689674213143, + "grad_norm": 0.8988812565803528, + "learning_rate": 2.613229259621947e-06, + "loss": 0.6097, + "step": 3568 + }, + { + "epoch": 2.954997239094423, + "grad_norm": 0.9200608730316162, + "learning_rate": 2.6121306878080753e-06, + "loss": 0.58, + "step": 3569 + }, + { + "epoch": 2.9558255107675317, + "grad_norm": 0.9378446936607361, + "learning_rate": 2.6110320942979606e-06, + "loss": 0.6043, + "step": 3570 + }, + { + "epoch": 2.9566537824406405, + "grad_norm": 0.9543853402137756, + "learning_rate": 2.6099334793041714e-06, + "loss": 0.58, + "step": 3571 + }, + { + "epoch": 2.957482054113749, + "grad_norm": 0.9590248465538025, + "learning_rate": 2.6088348430392784e-06, + "loss": 0.6004, + "step": 3572 + }, + { + "epoch": 2.958310325786858, + "grad_norm": 0.980459988117218, + "learning_rate": 2.607736185715858e-06, + "loss": 0.6118, + "step": 3573 + }, + { + "epoch": 2.959138597459967, + "grad_norm": 0.9050843715667725, + "learning_rate": 2.60663750754649e-06, + "loss": 0.5913, + "step": 3574 + }, + { + "epoch": 2.9599668691330754, + "grad_norm": 0.9189395904541016, + "learning_rate": 2.6055388087437576e-06, + "loss": 0.5997, + "step": 3575 + }, + { + "epoch": 2.9607951408061846, + "grad_norm": 0.9273754954338074, + "learning_rate": 2.6044400895202502e-06, + "loss": 0.586, + "step": 3576 + }, + { + "epoch": 2.9616234124792933, + "grad_norm": 0.90342116355896, + "learning_rate": 2.6033413500885594e-06, + "loss": 0.5846, + "step": 3577 + }, + { + "epoch": 2.962451684152402, + "grad_norm": 0.9073745012283325, + "learning_rate": 2.6022425906612803e-06, + "loss": 0.5899, + "step": 3578 + }, + { + "epoch": 2.9632799558255107, + "grad_norm": 0.9221594333648682, + "learning_rate": 2.6011438114510135e-06, + "loss": 0.5837, + "step": 3579 + }, + { + "epoch": 2.9641082274986195, + "grad_norm": 0.9228225350379944, + "learning_rate": 2.6000450126703613e-06, + "loss": 0.5761, + "step": 3580 + }, + { + "epoch": 2.964936499171728, + "grad_norm": 0.9136195182800293, + "learning_rate": 2.5989461945319315e-06, + "loss": 0.5977, + "step": 3581 + }, + { + "epoch": 2.965764770844837, + "grad_norm": 0.9152894616127014, + "learning_rate": 2.5978473572483366e-06, + "loss": 0.5829, + "step": 3582 + }, + { + "epoch": 2.966593042517946, + "grad_norm": 0.9124417304992676, + "learning_rate": 2.596748501032189e-06, + "loss": 0.5718, + "step": 3583 + }, + { + "epoch": 2.967421314191055, + "grad_norm": 0.9212865829467773, + "learning_rate": 2.5956496260961094e-06, + "loss": 0.6056, + "step": 3584 + }, + { + "epoch": 2.9682495858641635, + "grad_norm": 0.9101969003677368, + "learning_rate": 2.594550732652718e-06, + "loss": 0.6075, + "step": 3585 + }, + { + "epoch": 2.9690778575372723, + "grad_norm": 0.9225336313247681, + "learning_rate": 2.5934518209146414e-06, + "loss": 0.5894, + "step": 3586 + }, + { + "epoch": 2.969906129210381, + "grad_norm": 0.9271754026412964, + "learning_rate": 2.592352891094509e-06, + "loss": 0.5939, + "step": 3587 + }, + { + "epoch": 2.9707344008834897, + "grad_norm": 0.9217578768730164, + "learning_rate": 2.591253943404953e-06, + "loss": 0.5858, + "step": 3588 + }, + { + "epoch": 2.9715626725565985, + "grad_norm": 0.8871975541114807, + "learning_rate": 2.5901549780586085e-06, + "loss": 0.6052, + "step": 3589 + }, + { + "epoch": 2.9723909442297076, + "grad_norm": 0.9371136426925659, + "learning_rate": 2.5890559952681177e-06, + "loss": 0.6069, + "step": 3590 + }, + { + "epoch": 2.973219215902816, + "grad_norm": 0.9185692667961121, + "learning_rate": 2.5879569952461215e-06, + "loss": 0.5913, + "step": 3591 + }, + { + "epoch": 2.974047487575925, + "grad_norm": 0.9043979048728943, + "learning_rate": 2.586857978205266e-06, + "loss": 0.5747, + "step": 3592 + }, + { + "epoch": 2.974875759249034, + "grad_norm": 0.8951820731163025, + "learning_rate": 2.5857589443582014e-06, + "loss": 0.5904, + "step": 3593 + }, + { + "epoch": 2.9757040309221425, + "grad_norm": 0.9490121006965637, + "learning_rate": 2.5846598939175816e-06, + "loss": 0.583, + "step": 3594 + }, + { + "epoch": 2.9765323025952513, + "grad_norm": 0.9266814589500427, + "learning_rate": 2.583560827096061e-06, + "loss": 0.6039, + "step": 3595 + }, + { + "epoch": 2.97736057426836, + "grad_norm": 0.9517508149147034, + "learning_rate": 2.582461744106299e-06, + "loss": 0.572, + "step": 3596 + }, + { + "epoch": 2.9781888459414687, + "grad_norm": 0.9260841608047485, + "learning_rate": 2.581362645160958e-06, + "loss": 0.5898, + "step": 3597 + }, + { + "epoch": 2.9790171176145774, + "grad_norm": 0.9134445190429688, + "learning_rate": 2.5802635304727043e-06, + "loss": 0.6037, + "step": 3598 + }, + { + "epoch": 2.9798453892876866, + "grad_norm": 0.9307652115821838, + "learning_rate": 2.5791644002542054e-06, + "loss": 0.5941, + "step": 3599 + }, + { + "epoch": 2.980673660960795, + "grad_norm": 0.9665980935096741, + "learning_rate": 2.578065254718133e-06, + "loss": 0.5816, + "step": 3600 + }, + { + "epoch": 2.981501932633904, + "grad_norm": 0.9137181639671326, + "learning_rate": 2.5769660940771614e-06, + "loss": 0.5962, + "step": 3601 + }, + { + "epoch": 2.982330204307013, + "grad_norm": 0.9031209349632263, + "learning_rate": 2.575866918543968e-06, + "loss": 0.5853, + "step": 3602 + }, + { + "epoch": 2.9831584759801215, + "grad_norm": 0.9292592406272888, + "learning_rate": 2.5747677283312332e-06, + "loss": 0.5639, + "step": 3603 + }, + { + "epoch": 2.9839867476532302, + "grad_norm": 0.9193013906478882, + "learning_rate": 2.57366852365164e-06, + "loss": 0.5942, + "step": 3604 + }, + { + "epoch": 2.984815019326339, + "grad_norm": 0.8945804238319397, + "learning_rate": 2.572569304717874e-06, + "loss": 0.5882, + "step": 3605 + }, + { + "epoch": 2.9856432909994477, + "grad_norm": 0.9231047034263611, + "learning_rate": 2.5714700717426237e-06, + "loss": 0.5768, + "step": 3606 + }, + { + "epoch": 2.9864715626725564, + "grad_norm": 0.9362046718597412, + "learning_rate": 2.5703708249385824e-06, + "loss": 0.6145, + "step": 3607 + }, + { + "epoch": 2.9872998343456656, + "grad_norm": 0.9291548728942871, + "learning_rate": 2.5692715645184412e-06, + "loss": 0.5959, + "step": 3608 + }, + { + "epoch": 2.9881281060187743, + "grad_norm": 0.9203015565872192, + "learning_rate": 2.5681722906948977e-06, + "loss": 0.5906, + "step": 3609 + }, + { + "epoch": 2.988956377691883, + "grad_norm": 0.9317998886108398, + "learning_rate": 2.567073003680652e-06, + "loss": 0.5845, + "step": 3610 + }, + { + "epoch": 2.9897846493649918, + "grad_norm": 0.9188578128814697, + "learning_rate": 2.5659737036884054e-06, + "loss": 0.5901, + "step": 3611 + }, + { + "epoch": 2.9906129210381005, + "grad_norm": 0.9100846648216248, + "learning_rate": 2.564874390930862e-06, + "loss": 0.5952, + "step": 3612 + }, + { + "epoch": 2.9914411927112092, + "grad_norm": 0.9292994737625122, + "learning_rate": 2.563775065620729e-06, + "loss": 0.6112, + "step": 3613 + }, + { + "epoch": 2.992269464384318, + "grad_norm": 0.9025140404701233, + "learning_rate": 2.5626757279707143e-06, + "loss": 0.5691, + "step": 3614 + }, + { + "epoch": 2.9930977360574267, + "grad_norm": 0.9413566589355469, + "learning_rate": 2.561576378193531e-06, + "loss": 0.574, + "step": 3615 + }, + { + "epoch": 2.9939260077305354, + "grad_norm": 0.9244924783706665, + "learning_rate": 2.560477016501893e-06, + "loss": 0.606, + "step": 3616 + }, + { + "epoch": 2.9947542794036446, + "grad_norm": 0.8985790014266968, + "learning_rate": 2.5593776431085154e-06, + "loss": 0.5823, + "step": 3617 + }, + { + "epoch": 2.9955825510767533, + "grad_norm": 0.8941566348075867, + "learning_rate": 2.5582782582261182e-06, + "loss": 0.5643, + "step": 3618 + }, + { + "epoch": 2.996410822749862, + "grad_norm": 0.8992639183998108, + "learning_rate": 2.557178862067421e-06, + "loss": 0.6073, + "step": 3619 + }, + { + "epoch": 2.9972390944229708, + "grad_norm": 0.9369136691093445, + "learning_rate": 2.5560794548451464e-06, + "loss": 0.6031, + "step": 3620 + }, + { + "epoch": 2.9980673660960795, + "grad_norm": 0.8975582718849182, + "learning_rate": 2.5549800367720207e-06, + "loss": 0.6034, + "step": 3621 + }, + { + "epoch": 2.998895637769188, + "grad_norm": 0.8998291492462158, + "learning_rate": 2.5538806080607697e-06, + "loss": 0.5782, + "step": 3622 + }, + { + "epoch": 2.999723909442297, + "grad_norm": 0.9796467423439026, + "learning_rate": 2.5527811689241246e-06, + "loss": 0.581, + "step": 3623 + }, + { + "epoch": 3.0, + "grad_norm": 1.6821987628936768, + "learning_rate": 2.5516817195748143e-06, + "loss": 0.5955, + "step": 3624 + }, + { + "epoch": 3.0008282716731087, + "grad_norm": 0.9084760546684265, + "learning_rate": 2.5505822602255727e-06, + "loss": 0.5821, + "step": 3625 + }, + { + "epoch": 3.0016565433462175, + "grad_norm": 0.9643696546554565, + "learning_rate": 2.5494827910891357e-06, + "loss": 0.5627, + "step": 3626 + }, + { + "epoch": 3.002484815019326, + "grad_norm": 0.8933544754981995, + "learning_rate": 2.5483833123782405e-06, + "loss": 0.5573, + "step": 3627 + }, + { + "epoch": 3.003313086692435, + "grad_norm": 0.939877450466156, + "learning_rate": 2.5472838243056242e-06, + "loss": 0.5641, + "step": 3628 + }, + { + "epoch": 3.004141358365544, + "grad_norm": 0.9558863043785095, + "learning_rate": 2.54618432708403e-06, + "loss": 0.5753, + "step": 3629 + }, + { + "epoch": 3.004969630038653, + "grad_norm": 0.9423503875732422, + "learning_rate": 2.5450848209261987e-06, + "loss": 0.5786, + "step": 3630 + }, + { + "epoch": 3.0057979017117615, + "grad_norm": 0.9265878796577454, + "learning_rate": 2.5439853060448737e-06, + "loss": 0.5791, + "step": 3631 + }, + { + "epoch": 3.0066261733848703, + "grad_norm": 0.9355331063270569, + "learning_rate": 2.542885782652803e-06, + "loss": 0.5535, + "step": 3632 + }, + { + "epoch": 3.007454445057979, + "grad_norm": 0.9215316772460938, + "learning_rate": 2.5417862509627333e-06, + "loss": 0.5775, + "step": 3633 + }, + { + "epoch": 3.0082827167310877, + "grad_norm": 0.9157372713088989, + "learning_rate": 2.5406867111874127e-06, + "loss": 0.5547, + "step": 3634 + }, + { + "epoch": 3.0091109884041964, + "grad_norm": 0.9143363237380981, + "learning_rate": 2.539587163539594e-06, + "loss": 0.562, + "step": 3635 + }, + { + "epoch": 3.009939260077305, + "grad_norm": 0.942833423614502, + "learning_rate": 2.538487608232027e-06, + "loss": 0.5852, + "step": 3636 + }, + { + "epoch": 3.0107675317504143, + "grad_norm": 0.9196774363517761, + "learning_rate": 2.5373880454774667e-06, + "loss": 0.573, + "step": 3637 + }, + { + "epoch": 3.011595803423523, + "grad_norm": 0.9290075302124023, + "learning_rate": 2.536288475488669e-06, + "loss": 0.5672, + "step": 3638 + }, + { + "epoch": 3.012424075096632, + "grad_norm": 0.9856467843055725, + "learning_rate": 2.535188898478389e-06, + "loss": 0.5685, + "step": 3639 + }, + { + "epoch": 3.0132523467697405, + "grad_norm": 0.9258086085319519, + "learning_rate": 2.5340893146593855e-06, + "loss": 0.5705, + "step": 3640 + }, + { + "epoch": 3.0140806184428492, + "grad_norm": 0.8912391662597656, + "learning_rate": 2.5329897242444174e-06, + "loss": 0.5653, + "step": 3641 + }, + { + "epoch": 3.014908890115958, + "grad_norm": 0.9336050152778625, + "learning_rate": 2.531890127446244e-06, + "loss": 0.5515, + "step": 3642 + }, + { + "epoch": 3.0157371617890667, + "grad_norm": 0.9292400479316711, + "learning_rate": 2.5307905244776295e-06, + "loss": 0.5715, + "step": 3643 + }, + { + "epoch": 3.0165654334621754, + "grad_norm": 0.9212039709091187, + "learning_rate": 2.529690915551335e-06, + "loss": 0.5842, + "step": 3644 + }, + { + "epoch": 3.017393705135284, + "grad_norm": 0.9198867082595825, + "learning_rate": 2.528591300880125e-06, + "loss": 0.5498, + "step": 3645 + }, + { + "epoch": 3.0182219768083933, + "grad_norm": 0.9194673299789429, + "learning_rate": 2.5274916806767656e-06, + "loss": 0.5696, + "step": 3646 + }, + { + "epoch": 3.019050248481502, + "grad_norm": 0.9359262585639954, + "learning_rate": 2.5263920551540217e-06, + "loss": 0.5635, + "step": 3647 + }, + { + "epoch": 3.019878520154611, + "grad_norm": 0.9225771427154541, + "learning_rate": 2.525292424524661e-06, + "loss": 0.5489, + "step": 3648 + }, + { + "epoch": 3.0207067918277195, + "grad_norm": 0.9490647315979004, + "learning_rate": 2.524192789001452e-06, + "loss": 0.5692, + "step": 3649 + }, + { + "epoch": 3.0215350635008282, + "grad_norm": 0.8977577686309814, + "learning_rate": 2.5230931487971645e-06, + "loss": 0.5652, + "step": 3650 + }, + { + "epoch": 3.022363335173937, + "grad_norm": 0.963139533996582, + "learning_rate": 2.521993504124567e-06, + "loss": 0.5373, + "step": 3651 + }, + { + "epoch": 3.0231916068470457, + "grad_norm": 0.9195258617401123, + "learning_rate": 2.5208938551964326e-06, + "loss": 0.574, + "step": 3652 + }, + { + "epoch": 3.0240198785201544, + "grad_norm": 0.9074851274490356, + "learning_rate": 2.5197942022255305e-06, + "loss": 0.557, + "step": 3653 + }, + { + "epoch": 3.0248481501932636, + "grad_norm": 0.9083223342895508, + "learning_rate": 2.518694545424636e-06, + "loss": 0.5641, + "step": 3654 + }, + { + "epoch": 3.0256764218663723, + "grad_norm": 0.9047815203666687, + "learning_rate": 2.5175948850065207e-06, + "loss": 0.563, + "step": 3655 + }, + { + "epoch": 3.026504693539481, + "grad_norm": 0.918021559715271, + "learning_rate": 2.516495221183959e-06, + "loss": 0.5889, + "step": 3656 + }, + { + "epoch": 3.0273329652125898, + "grad_norm": 0.935458242893219, + "learning_rate": 2.5153955541697266e-06, + "loss": 0.5486, + "step": 3657 + }, + { + "epoch": 3.0281612368856985, + "grad_norm": 0.9182383418083191, + "learning_rate": 2.5142958841765976e-06, + "loss": 0.563, + "step": 3658 + }, + { + "epoch": 3.028989508558807, + "grad_norm": 0.9635967016220093, + "learning_rate": 2.5131962114173474e-06, + "loss": 0.557, + "step": 3659 + }, + { + "epoch": 3.029817780231916, + "grad_norm": 0.9150283932685852, + "learning_rate": 2.512096536104754e-06, + "loss": 0.5582, + "step": 3660 + }, + { + "epoch": 3.0306460519050247, + "grad_norm": 0.9288967847824097, + "learning_rate": 2.5109968584515935e-06, + "loss": 0.5628, + "step": 3661 + }, + { + "epoch": 3.031474323578134, + "grad_norm": 0.936221718788147, + "learning_rate": 2.5098971786706434e-06, + "loss": 0.564, + "step": 3662 + }, + { + "epoch": 3.0323025952512426, + "grad_norm": 0.9719551801681519, + "learning_rate": 2.5087974969746828e-06, + "loss": 0.5509, + "step": 3663 + }, + { + "epoch": 3.0331308669243513, + "grad_norm": 0.9522861838340759, + "learning_rate": 2.507697813576487e-06, + "loss": 0.5919, + "step": 3664 + }, + { + "epoch": 3.03395913859746, + "grad_norm": 0.9098208546638489, + "learning_rate": 2.506598128688836e-06, + "loss": 0.5661, + "step": 3665 + }, + { + "epoch": 3.0347874102705688, + "grad_norm": 0.9221910238265991, + "learning_rate": 2.5054984425245095e-06, + "loss": 0.5849, + "step": 3666 + }, + { + "epoch": 3.0356156819436775, + "grad_norm": 0.9386799335479736, + "learning_rate": 2.5043987552962844e-06, + "loss": 0.5765, + "step": 3667 + }, + { + "epoch": 3.036443953616786, + "grad_norm": 0.9208933711051941, + "learning_rate": 2.503299067216942e-06, + "loss": 0.5689, + "step": 3668 + }, + { + "epoch": 3.037272225289895, + "grad_norm": 0.9429235458374023, + "learning_rate": 2.5021993784992616e-06, + "loss": 0.555, + "step": 3669 + }, + { + "epoch": 3.0381004969630037, + "grad_norm": 0.9263848066329956, + "learning_rate": 2.5010996893560203e-06, + "loss": 0.562, + "step": 3670 + }, + { + "epoch": 3.038928768636113, + "grad_norm": 0.9523965120315552, + "learning_rate": 2.5e-06, + "loss": 0.5677, + "step": 3671 + }, + { + "epoch": 3.0397570403092216, + "grad_norm": 0.9585251212120056, + "learning_rate": 2.49890031064398e-06, + "loss": 0.5821, + "step": 3672 + }, + { + "epoch": 3.0405853119823303, + "grad_norm": 0.9311655163764954, + "learning_rate": 2.4978006215007393e-06, + "loss": 0.5752, + "step": 3673 + }, + { + "epoch": 3.041413583655439, + "grad_norm": 0.9548931121826172, + "learning_rate": 2.4967009327830584e-06, + "loss": 0.5719, + "step": 3674 + }, + { + "epoch": 3.0422418553285477, + "grad_norm": 0.9751641154289246, + "learning_rate": 2.495601244703716e-06, + "loss": 0.5869, + "step": 3675 + }, + { + "epoch": 3.0430701270016565, + "grad_norm": 0.9528283476829529, + "learning_rate": 2.494501557475491e-06, + "loss": 0.571, + "step": 3676 + }, + { + "epoch": 3.043898398674765, + "grad_norm": 0.9225226044654846, + "learning_rate": 2.493401871311165e-06, + "loss": 0.5635, + "step": 3677 + }, + { + "epoch": 3.044726670347874, + "grad_norm": 0.9444342851638794, + "learning_rate": 2.4923021864235143e-06, + "loss": 0.5593, + "step": 3678 + }, + { + "epoch": 3.045554942020983, + "grad_norm": 0.9299778938293457, + "learning_rate": 2.491202503025319e-06, + "loss": 0.5787, + "step": 3679 + }, + { + "epoch": 3.046383213694092, + "grad_norm": 0.913502037525177, + "learning_rate": 2.490102821329357e-06, + "loss": 0.5826, + "step": 3680 + }, + { + "epoch": 3.0472114853672005, + "grad_norm": 0.9337341785430908, + "learning_rate": 2.489003141548407e-06, + "loss": 0.5575, + "step": 3681 + }, + { + "epoch": 3.0480397570403093, + "grad_norm": 0.9271352887153625, + "learning_rate": 2.4879034638952464e-06, + "loss": 0.5759, + "step": 3682 + }, + { + "epoch": 3.048868028713418, + "grad_norm": 0.967474102973938, + "learning_rate": 2.486803788582653e-06, + "loss": 0.5542, + "step": 3683 + }, + { + "epoch": 3.0496963003865267, + "grad_norm": 0.9283983111381531, + "learning_rate": 2.4857041158234032e-06, + "loss": 0.5932, + "step": 3684 + }, + { + "epoch": 3.0505245720596355, + "grad_norm": 0.9075651168823242, + "learning_rate": 2.4846044458302742e-06, + "loss": 0.5704, + "step": 3685 + }, + { + "epoch": 3.051352843732744, + "grad_norm": 0.9511797428131104, + "learning_rate": 2.4835047788160415e-06, + "loss": 0.5788, + "step": 3686 + }, + { + "epoch": 3.052181115405853, + "grad_norm": 0.9016287922859192, + "learning_rate": 2.4824051149934797e-06, + "loss": 0.5829, + "step": 3687 + }, + { + "epoch": 3.053009387078962, + "grad_norm": 0.9479110836982727, + "learning_rate": 2.4813054545753657e-06, + "loss": 0.5721, + "step": 3688 + }, + { + "epoch": 3.053837658752071, + "grad_norm": 0.906857430934906, + "learning_rate": 2.4802057977744703e-06, + "loss": 0.5497, + "step": 3689 + }, + { + "epoch": 3.0546659304251795, + "grad_norm": 0.9445266723632812, + "learning_rate": 2.479106144803569e-06, + "loss": 0.5837, + "step": 3690 + }, + { + "epoch": 3.0554942020982883, + "grad_norm": 0.9599757194519043, + "learning_rate": 2.478006495875434e-06, + "loss": 0.5571, + "step": 3691 + }, + { + "epoch": 3.056322473771397, + "grad_norm": 0.924673318862915, + "learning_rate": 2.4769068512028368e-06, + "loss": 0.5644, + "step": 3692 + }, + { + "epoch": 3.0571507454445057, + "grad_norm": 0.9168586730957031, + "learning_rate": 2.4758072109985487e-06, + "loss": 0.5682, + "step": 3693 + }, + { + "epoch": 3.0579790171176144, + "grad_norm": 0.9143668413162231, + "learning_rate": 2.4747075754753395e-06, + "loss": 0.5521, + "step": 3694 + }, + { + "epoch": 3.058807288790723, + "grad_norm": 0.9329923391342163, + "learning_rate": 2.4736079448459787e-06, + "loss": 0.5534, + "step": 3695 + }, + { + "epoch": 3.0596355604638323, + "grad_norm": 0.9511533975601196, + "learning_rate": 2.472508319323235e-06, + "loss": 0.5795, + "step": 3696 + }, + { + "epoch": 3.060463832136941, + "grad_norm": 0.9574876427650452, + "learning_rate": 2.471408699119875e-06, + "loss": 0.5723, + "step": 3697 + }, + { + "epoch": 3.06129210381005, + "grad_norm": 0.9656964540481567, + "learning_rate": 2.470309084448665e-06, + "loss": 0.5876, + "step": 3698 + }, + { + "epoch": 3.0621203754831585, + "grad_norm": 0.9273092150688171, + "learning_rate": 2.469209475522371e-06, + "loss": 0.5591, + "step": 3699 + }, + { + "epoch": 3.0629486471562672, + "grad_norm": 0.9115599393844604, + "learning_rate": 2.4681098725537566e-06, + "loss": 0.5766, + "step": 3700 + }, + { + "epoch": 3.063776918829376, + "grad_norm": 0.9338927268981934, + "learning_rate": 2.4670102757555843e-06, + "loss": 0.5584, + "step": 3701 + }, + { + "epoch": 3.0646051905024847, + "grad_norm": 0.9274653792381287, + "learning_rate": 2.4659106853406158e-06, + "loss": 0.5545, + "step": 3702 + }, + { + "epoch": 3.0654334621755934, + "grad_norm": 0.9300533533096313, + "learning_rate": 2.464811101521612e-06, + "loss": 0.5751, + "step": 3703 + }, + { + "epoch": 3.0662617338487026, + "grad_norm": 0.9616261124610901, + "learning_rate": 2.463711524511332e-06, + "loss": 0.5635, + "step": 3704 + }, + { + "epoch": 3.0670900055218113, + "grad_norm": 1.0845082998275757, + "learning_rate": 2.4626119545225337e-06, + "loss": 0.5712, + "step": 3705 + }, + { + "epoch": 3.06791827719492, + "grad_norm": 0.9370817542076111, + "learning_rate": 2.4615123917679737e-06, + "loss": 0.5737, + "step": 3706 + }, + { + "epoch": 3.068746548868029, + "grad_norm": 0.984732985496521, + "learning_rate": 2.460412836460407e-06, + "loss": 0.5631, + "step": 3707 + }, + { + "epoch": 3.0695748205411375, + "grad_norm": 0.90977942943573, + "learning_rate": 2.4593132888125877e-06, + "loss": 0.5755, + "step": 3708 + }, + { + "epoch": 3.0704030922142462, + "grad_norm": 0.9592271447181702, + "learning_rate": 2.4582137490372676e-06, + "loss": 0.552, + "step": 3709 + }, + { + "epoch": 3.071231363887355, + "grad_norm": 0.9560198187828064, + "learning_rate": 2.457114217347197e-06, + "loss": 0.5744, + "step": 3710 + }, + { + "epoch": 3.0720596355604637, + "grad_norm": 0.9570426940917969, + "learning_rate": 2.4560146939551268e-06, + "loss": 0.5769, + "step": 3711 + }, + { + "epoch": 3.0728879072335724, + "grad_norm": 0.9208659529685974, + "learning_rate": 2.4549151790738025e-06, + "loss": 0.5697, + "step": 3712 + }, + { + "epoch": 3.0737161789066816, + "grad_norm": 0.9417058229446411, + "learning_rate": 2.453815672915971e-06, + "loss": 0.5697, + "step": 3713 + }, + { + "epoch": 3.0745444505797903, + "grad_norm": 0.9602729082107544, + "learning_rate": 2.4527161756943766e-06, + "loss": 0.5737, + "step": 3714 + }, + { + "epoch": 3.075372722252899, + "grad_norm": 0.931280255317688, + "learning_rate": 2.4516166876217603e-06, + "loss": 0.5704, + "step": 3715 + }, + { + "epoch": 3.0762009939260078, + "grad_norm": 0.920081377029419, + "learning_rate": 2.4505172089108648e-06, + "loss": 0.57, + "step": 3716 + }, + { + "epoch": 3.0770292655991165, + "grad_norm": 0.929806649684906, + "learning_rate": 2.449417739774428e-06, + "loss": 0.5648, + "step": 3717 + }, + { + "epoch": 3.077857537272225, + "grad_norm": 0.9344229698181152, + "learning_rate": 2.4483182804251866e-06, + "loss": 0.5666, + "step": 3718 + }, + { + "epoch": 3.078685808945334, + "grad_norm": 0.9535424113273621, + "learning_rate": 2.4472188310758766e-06, + "loss": 0.5777, + "step": 3719 + }, + { + "epoch": 3.0795140806184427, + "grad_norm": 0.9594505429267883, + "learning_rate": 2.4461193919392307e-06, + "loss": 0.5614, + "step": 3720 + }, + { + "epoch": 3.080342352291552, + "grad_norm": 0.9511420726776123, + "learning_rate": 2.4450199632279797e-06, + "loss": 0.5925, + "step": 3721 + }, + { + "epoch": 3.0811706239646606, + "grad_norm": 0.9244424104690552, + "learning_rate": 2.443920545154855e-06, + "loss": 0.5694, + "step": 3722 + }, + { + "epoch": 3.0819988956377693, + "grad_norm": 0.9384777545928955, + "learning_rate": 2.4428211379325804e-06, + "loss": 0.5443, + "step": 3723 + }, + { + "epoch": 3.082827167310878, + "grad_norm": 0.9536921381950378, + "learning_rate": 2.441721741773883e-06, + "loss": 0.5817, + "step": 3724 + }, + { + "epoch": 3.0836554389839868, + "grad_norm": 0.9547799825668335, + "learning_rate": 2.4406223568914854e-06, + "loss": 0.5565, + "step": 3725 + }, + { + "epoch": 3.0844837106570955, + "grad_norm": 0.9509586691856384, + "learning_rate": 2.4395229834981076e-06, + "loss": 0.5712, + "step": 3726 + }, + { + "epoch": 3.085311982330204, + "grad_norm": 0.9227497577667236, + "learning_rate": 2.4384236218064692e-06, + "loss": 0.5602, + "step": 3727 + }, + { + "epoch": 3.086140254003313, + "grad_norm": 0.9162245392799377, + "learning_rate": 2.4373242720292865e-06, + "loss": 0.555, + "step": 3728 + }, + { + "epoch": 3.0869685256764217, + "grad_norm": 0.9121986627578735, + "learning_rate": 2.4362249343792722e-06, + "loss": 0.5748, + "step": 3729 + }, + { + "epoch": 3.087796797349531, + "grad_norm": 0.9310787320137024, + "learning_rate": 2.435125609069139e-06, + "loss": 0.5671, + "step": 3730 + }, + { + "epoch": 3.0886250690226396, + "grad_norm": 0.9523546695709229, + "learning_rate": 2.4340262963115955e-06, + "loss": 0.595, + "step": 3731 + }, + { + "epoch": 3.0894533406957483, + "grad_norm": 0.9429517388343811, + "learning_rate": 2.432926996319348e-06, + "loss": 0.5381, + "step": 3732 + }, + { + "epoch": 3.090281612368857, + "grad_norm": 0.9443867802619934, + "learning_rate": 2.4318277093051023e-06, + "loss": 0.5588, + "step": 3733 + }, + { + "epoch": 3.0911098840419657, + "grad_norm": 0.9320855140686035, + "learning_rate": 2.43072843548156e-06, + "loss": 0.5765, + "step": 3734 + }, + { + "epoch": 3.0919381557150745, + "grad_norm": 0.9335711002349854, + "learning_rate": 2.429629175061419e-06, + "loss": 0.5734, + "step": 3735 + }, + { + "epoch": 3.092766427388183, + "grad_norm": 0.9539975523948669, + "learning_rate": 2.4285299282573767e-06, + "loss": 0.5733, + "step": 3736 + }, + { + "epoch": 3.093594699061292, + "grad_norm": 0.9393835067749023, + "learning_rate": 2.4274306952821262e-06, + "loss": 0.5808, + "step": 3737 + }, + { + "epoch": 3.094422970734401, + "grad_norm": 0.9707219004631042, + "learning_rate": 2.4263314763483607e-06, + "loss": 0.5755, + "step": 3738 + }, + { + "epoch": 3.09525124240751, + "grad_norm": 0.9541774988174438, + "learning_rate": 2.4252322716687676e-06, + "loss": 0.5472, + "step": 3739 + }, + { + "epoch": 3.0960795140806185, + "grad_norm": 0.9300408959388733, + "learning_rate": 2.424133081456032e-06, + "loss": 0.5771, + "step": 3740 + }, + { + "epoch": 3.0969077857537273, + "grad_norm": 0.9487392902374268, + "learning_rate": 2.423033905922839e-06, + "loss": 0.5805, + "step": 3741 + }, + { + "epoch": 3.097736057426836, + "grad_norm": 0.9021924734115601, + "learning_rate": 2.4219347452818678e-06, + "loss": 0.5648, + "step": 3742 + }, + { + "epoch": 3.0985643290999447, + "grad_norm": 0.9229711294174194, + "learning_rate": 2.420835599745795e-06, + "loss": 0.5569, + "step": 3743 + }, + { + "epoch": 3.0993926007730535, + "grad_norm": 0.9335191249847412, + "learning_rate": 2.419736469527296e-06, + "loss": 0.5728, + "step": 3744 + }, + { + "epoch": 3.100220872446162, + "grad_norm": 0.9294394254684448, + "learning_rate": 2.4186373548390427e-06, + "loss": 0.5522, + "step": 3745 + }, + { + "epoch": 3.101049144119271, + "grad_norm": 0.931654691696167, + "learning_rate": 2.417538255893702e-06, + "loss": 0.5747, + "step": 3746 + }, + { + "epoch": 3.10187741579238, + "grad_norm": 0.9018174409866333, + "learning_rate": 2.4164391729039405e-06, + "loss": 0.5697, + "step": 3747 + }, + { + "epoch": 3.102705687465489, + "grad_norm": 0.9389755129814148, + "learning_rate": 2.4153401060824193e-06, + "loss": 0.5559, + "step": 3748 + }, + { + "epoch": 3.1035339591385975, + "grad_norm": 0.9191637635231018, + "learning_rate": 2.414241055641799e-06, + "loss": 0.5678, + "step": 3749 + }, + { + "epoch": 3.1043622308117063, + "grad_norm": 0.9245947599411011, + "learning_rate": 2.4131420217947347e-06, + "loss": 0.5677, + "step": 3750 + }, + { + "epoch": 3.105190502484815, + "grad_norm": 0.9380972981452942, + "learning_rate": 2.4120430047538797e-06, + "loss": 0.5611, + "step": 3751 + }, + { + "epoch": 3.1060187741579237, + "grad_norm": 0.9427217841148376, + "learning_rate": 2.410944004731883e-06, + "loss": 0.592, + "step": 3752 + }, + { + "epoch": 3.1068470458310324, + "grad_norm": 0.9142959117889404, + "learning_rate": 2.4098450219413915e-06, + "loss": 0.5588, + "step": 3753 + }, + { + "epoch": 3.107675317504141, + "grad_norm": 0.9501208066940308, + "learning_rate": 2.4087460565950475e-06, + "loss": 0.5735, + "step": 3754 + }, + { + "epoch": 3.1085035891772503, + "grad_norm": 0.9072937369346619, + "learning_rate": 2.4076471089054916e-06, + "loss": 0.557, + "step": 3755 + }, + { + "epoch": 3.109331860850359, + "grad_norm": 0.9310604333877563, + "learning_rate": 2.4065481790853595e-06, + "loss": 0.5616, + "step": 3756 + }, + { + "epoch": 3.110160132523468, + "grad_norm": 0.9229986071586609, + "learning_rate": 2.405449267347283e-06, + "loss": 0.5806, + "step": 3757 + }, + { + "epoch": 3.1109884041965765, + "grad_norm": 0.9491189122200012, + "learning_rate": 2.4043503739038922e-06, + "loss": 0.571, + "step": 3758 + }, + { + "epoch": 3.1118166758696852, + "grad_norm": 0.9373705983161926, + "learning_rate": 2.4032514989678113e-06, + "loss": 0.557, + "step": 3759 + }, + { + "epoch": 3.112644947542794, + "grad_norm": 0.9476242661476135, + "learning_rate": 2.4021526427516647e-06, + "loss": 0.5749, + "step": 3760 + }, + { + "epoch": 3.1134732192159027, + "grad_norm": 0.9147936701774597, + "learning_rate": 2.401053805468069e-06, + "loss": 0.5579, + "step": 3761 + }, + { + "epoch": 3.1143014908890114, + "grad_norm": 0.9299486875534058, + "learning_rate": 2.399954987329639e-06, + "loss": 0.5707, + "step": 3762 + }, + { + "epoch": 3.1151297625621206, + "grad_norm": 0.9515376091003418, + "learning_rate": 2.3988561885489874e-06, + "loss": 0.5923, + "step": 3763 + }, + { + "epoch": 3.1159580342352293, + "grad_norm": 0.9455316662788391, + "learning_rate": 2.39775740933872e-06, + "loss": 0.5686, + "step": 3764 + }, + { + "epoch": 3.116786305908338, + "grad_norm": 0.952534019947052, + "learning_rate": 2.3966586499114406e-06, + "loss": 0.581, + "step": 3765 + }, + { + "epoch": 3.1176145775814468, + "grad_norm": 0.9540290832519531, + "learning_rate": 2.3955599104797493e-06, + "loss": 0.5612, + "step": 3766 + }, + { + "epoch": 3.1184428492545555, + "grad_norm": 0.9238110780715942, + "learning_rate": 2.3944611912562423e-06, + "loss": 0.5608, + "step": 3767 + }, + { + "epoch": 3.1192711209276642, + "grad_norm": 0.9312852025032043, + "learning_rate": 2.3933624924535113e-06, + "loss": 0.5752, + "step": 3768 + }, + { + "epoch": 3.120099392600773, + "grad_norm": 0.9356149435043335, + "learning_rate": 2.3922638142841433e-06, + "loss": 0.5719, + "step": 3769 + }, + { + "epoch": 3.1209276642738817, + "grad_norm": 0.9629190564155579, + "learning_rate": 2.391165156960722e-06, + "loss": 0.5791, + "step": 3770 + }, + { + "epoch": 3.1217559359469904, + "grad_norm": 0.9337695240974426, + "learning_rate": 2.3900665206958294e-06, + "loss": 0.5636, + "step": 3771 + }, + { + "epoch": 3.1225842076200996, + "grad_norm": 0.9723961353302002, + "learning_rate": 2.3889679057020402e-06, + "loss": 0.5897, + "step": 3772 + }, + { + "epoch": 3.1234124792932083, + "grad_norm": 0.9874154925346375, + "learning_rate": 2.3878693121919255e-06, + "loss": 0.5846, + "step": 3773 + }, + { + "epoch": 3.124240750966317, + "grad_norm": 0.9433327913284302, + "learning_rate": 2.386770740378054e-06, + "loss": 0.5631, + "step": 3774 + }, + { + "epoch": 3.1250690226394258, + "grad_norm": 0.9409958124160767, + "learning_rate": 2.385672190472989e-06, + "loss": 0.548, + "step": 3775 + }, + { + "epoch": 3.1258972943125345, + "grad_norm": 0.9370530247688293, + "learning_rate": 2.3845736626892895e-06, + "loss": 0.5277, + "step": 3776 + }, + { + "epoch": 3.126725565985643, + "grad_norm": 0.9603546857833862, + "learning_rate": 2.3834751572395097e-06, + "loss": 0.5635, + "step": 3777 + }, + { + "epoch": 3.127553837658752, + "grad_norm": 0.9234820008277893, + "learning_rate": 2.382376674336201e-06, + "loss": 0.5743, + "step": 3778 + }, + { + "epoch": 3.1283821093318607, + "grad_norm": 0.9674520492553711, + "learning_rate": 2.38127821419191e-06, + "loss": 0.5694, + "step": 3779 + }, + { + "epoch": 3.1292103810049694, + "grad_norm": 0.9508289098739624, + "learning_rate": 2.380179777019177e-06, + "loss": 0.5625, + "step": 3780 + }, + { + "epoch": 3.1300386526780786, + "grad_norm": 0.9463353753089905, + "learning_rate": 2.3790813630305403e-06, + "loss": 0.5859, + "step": 3781 + }, + { + "epoch": 3.1308669243511873, + "grad_norm": 0.9280157089233398, + "learning_rate": 2.3779829724385327e-06, + "loss": 0.5632, + "step": 3782 + }, + { + "epoch": 3.131695196024296, + "grad_norm": 0.931016206741333, + "learning_rate": 2.376884605455683e-06, + "loss": 0.556, + "step": 3783 + }, + { + "epoch": 3.1325234676974048, + "grad_norm": 0.9319890141487122, + "learning_rate": 2.375786262294514e-06, + "loss": 0.563, + "step": 3784 + }, + { + "epoch": 3.1333517393705135, + "grad_norm": 0.9274983406066895, + "learning_rate": 2.374687943167545e-06, + "loss": 0.5713, + "step": 3785 + }, + { + "epoch": 3.134180011043622, + "grad_norm": 0.9298895001411438, + "learning_rate": 2.3735896482872917e-06, + "loss": 0.5705, + "step": 3786 + }, + { + "epoch": 3.135008282716731, + "grad_norm": 0.9218611121177673, + "learning_rate": 2.3724913778662624e-06, + "loss": 0.5738, + "step": 3787 + }, + { + "epoch": 3.13583655438984, + "grad_norm": 0.9677892327308655, + "learning_rate": 2.371393132116963e-06, + "loss": 0.5675, + "step": 3788 + }, + { + "epoch": 3.136664826062949, + "grad_norm": 0.9483129978179932, + "learning_rate": 2.3702949112518935e-06, + "loss": 0.5525, + "step": 3789 + }, + { + "epoch": 3.1374930977360576, + "grad_norm": 0.9588272571563721, + "learning_rate": 2.3691967154835505e-06, + "loss": 0.5819, + "step": 3790 + }, + { + "epoch": 3.1383213694091663, + "grad_norm": 0.9432960748672485, + "learning_rate": 2.3680985450244233e-06, + "loss": 0.5757, + "step": 3791 + }, + { + "epoch": 3.139149641082275, + "grad_norm": 0.952733039855957, + "learning_rate": 2.367000400086998e-06, + "loss": 0.5807, + "step": 3792 + }, + { + "epoch": 3.1399779127553837, + "grad_norm": 0.9517266154289246, + "learning_rate": 2.365902280883755e-06, + "loss": 0.5503, + "step": 3793 + }, + { + "epoch": 3.1408061844284925, + "grad_norm": 0.9321509599685669, + "learning_rate": 2.364804187627171e-06, + "loss": 0.5534, + "step": 3794 + }, + { + "epoch": 3.141634456101601, + "grad_norm": 0.9490827918052673, + "learning_rate": 2.3637061205297167e-06, + "loss": 0.582, + "step": 3795 + }, + { + "epoch": 3.14246272777471, + "grad_norm": 0.9612736701965332, + "learning_rate": 2.3626080798038565e-06, + "loss": 0.5673, + "step": 3796 + }, + { + "epoch": 3.143290999447819, + "grad_norm": 0.9634802937507629, + "learning_rate": 2.361510065662053e-06, + "loss": 0.5784, + "step": 3797 + }, + { + "epoch": 3.144119271120928, + "grad_norm": 0.9495031237602234, + "learning_rate": 2.360412078316761e-06, + "loss": 0.5661, + "step": 3798 + }, + { + "epoch": 3.1449475427940365, + "grad_norm": 0.9234485030174255, + "learning_rate": 2.3593141179804295e-06, + "loss": 0.5718, + "step": 3799 + }, + { + "epoch": 3.1457758144671453, + "grad_norm": 0.9367272257804871, + "learning_rate": 2.3582161848655054e-06, + "loss": 0.5332, + "step": 3800 + }, + { + "epoch": 3.146604086140254, + "grad_norm": 0.9217962622642517, + "learning_rate": 2.357118279184429e-06, + "loss": 0.586, + "step": 3801 + }, + { + "epoch": 3.1474323578133627, + "grad_norm": 0.943759560585022, + "learning_rate": 2.3560204011496324e-06, + "loss": 0.5606, + "step": 3802 + }, + { + "epoch": 3.1482606294864715, + "grad_norm": 0.959769070148468, + "learning_rate": 2.354922550973546e-06, + "loss": 0.5858, + "step": 3803 + }, + { + "epoch": 3.14908890115958, + "grad_norm": 0.9223443865776062, + "learning_rate": 2.3538247288685935e-06, + "loss": 0.5924, + "step": 3804 + }, + { + "epoch": 3.149917172832689, + "grad_norm": 0.958400547504425, + "learning_rate": 2.3527269350471936e-06, + "loss": 0.5669, + "step": 3805 + }, + { + "epoch": 3.150745444505798, + "grad_norm": 0.9326207041740417, + "learning_rate": 2.351629169721759e-06, + "loss": 0.5691, + "step": 3806 + }, + { + "epoch": 3.151573716178907, + "grad_norm": 0.9581301808357239, + "learning_rate": 2.3505314331046964e-06, + "loss": 0.5767, + "step": 3807 + }, + { + "epoch": 3.1524019878520155, + "grad_norm": 0.921093225479126, + "learning_rate": 2.349433725408409e-06, + "loss": 0.5773, + "step": 3808 + }, + { + "epoch": 3.1532302595251243, + "grad_norm": 0.9264110326766968, + "learning_rate": 2.348336046845292e-06, + "loss": 0.5388, + "step": 3809 + }, + { + "epoch": 3.154058531198233, + "grad_norm": 0.9559600949287415, + "learning_rate": 2.3472383976277353e-06, + "loss": 0.5601, + "step": 3810 + }, + { + "epoch": 3.1548868028713417, + "grad_norm": 0.8962564468383789, + "learning_rate": 2.3461407779681255e-06, + "loss": 0.5718, + "step": 3811 + }, + { + "epoch": 3.1557150745444504, + "grad_norm": 0.9371088743209839, + "learning_rate": 2.345043188078841e-06, + "loss": 0.5841, + "step": 3812 + }, + { + "epoch": 3.156543346217559, + "grad_norm": 0.952690839767456, + "learning_rate": 2.343945628172255e-06, + "loss": 0.5517, + "step": 3813 + }, + { + "epoch": 3.1573716178906683, + "grad_norm": 0.9315347075462341, + "learning_rate": 2.3428480984607356e-06, + "loss": 0.5773, + "step": 3814 + }, + { + "epoch": 3.158199889563777, + "grad_norm": 0.9349079728126526, + "learning_rate": 2.341750599156643e-06, + "loss": 0.568, + "step": 3815 + }, + { + "epoch": 3.159028161236886, + "grad_norm": 0.9342600703239441, + "learning_rate": 2.340653130472335e-06, + "loss": 0.5495, + "step": 3816 + }, + { + "epoch": 3.1598564329099945, + "grad_norm": 0.9394509792327881, + "learning_rate": 2.339555692620161e-06, + "loss": 0.6008, + "step": 3817 + }, + { + "epoch": 3.1606847045831032, + "grad_norm": 0.9620388150215149, + "learning_rate": 2.338458285812464e-06, + "loss": 0.5652, + "step": 3818 + }, + { + "epoch": 3.161512976256212, + "grad_norm": 0.9644536375999451, + "learning_rate": 2.3373609102615834e-06, + "loss": 0.5556, + "step": 3819 + }, + { + "epoch": 3.1623412479293207, + "grad_norm": 0.9285350441932678, + "learning_rate": 2.3362635661798507e-06, + "loss": 0.5758, + "step": 3820 + }, + { + "epoch": 3.1631695196024294, + "grad_norm": 0.9304336905479431, + "learning_rate": 2.335166253779591e-06, + "loss": 0.6108, + "step": 3821 + }, + { + "epoch": 3.1639977912755386, + "grad_norm": 0.9232069253921509, + "learning_rate": 2.334068973273125e-06, + "loss": 0.5835, + "step": 3822 + }, + { + "epoch": 3.1648260629486473, + "grad_norm": 0.9440987706184387, + "learning_rate": 2.3329717248727665e-06, + "loss": 0.5778, + "step": 3823 + }, + { + "epoch": 3.165654334621756, + "grad_norm": 0.9345200657844543, + "learning_rate": 2.3318745087908225e-06, + "loss": 0.56, + "step": 3824 + }, + { + "epoch": 3.1664826062948648, + "grad_norm": 0.9822887778282166, + "learning_rate": 2.330777325239593e-06, + "loss": 0.556, + "step": 3825 + }, + { + "epoch": 3.1673108779679735, + "grad_norm": 0.9316609501838684, + "learning_rate": 2.3296801744313733e-06, + "loss": 0.5749, + "step": 3826 + }, + { + "epoch": 3.1681391496410822, + "grad_norm": 0.933875322341919, + "learning_rate": 2.3285830565784528e-06, + "loss": 0.5686, + "step": 3827 + }, + { + "epoch": 3.168967421314191, + "grad_norm": 0.9843499660491943, + "learning_rate": 2.327485971893112e-06, + "loss": 0.6005, + "step": 3828 + }, + { + "epoch": 3.1697956929872997, + "grad_norm": 0.9594619870185852, + "learning_rate": 2.326388920587628e-06, + "loss": 0.5754, + "step": 3829 + }, + { + "epoch": 3.1706239646604084, + "grad_norm": 0.959398090839386, + "learning_rate": 2.3252919028742698e-06, + "loss": 0.5762, + "step": 3830 + }, + { + "epoch": 3.1714522363335176, + "grad_norm": 0.9445636868476868, + "learning_rate": 2.3241949189652996e-06, + "loss": 0.5757, + "step": 3831 + }, + { + "epoch": 3.1722805080066263, + "grad_norm": 0.9507540464401245, + "learning_rate": 2.323097969072973e-06, + "loss": 0.5695, + "step": 3832 + }, + { + "epoch": 3.173108779679735, + "grad_norm": 0.9695854187011719, + "learning_rate": 2.322001053409541e-06, + "loss": 0.5665, + "step": 3833 + }, + { + "epoch": 3.1739370513528438, + "grad_norm": 0.9487887620925903, + "learning_rate": 2.3209041721872464e-06, + "loss": 0.5651, + "step": 3834 + }, + { + "epoch": 3.1747653230259525, + "grad_norm": 0.9431695938110352, + "learning_rate": 2.319807325618325e-06, + "loss": 0.5516, + "step": 3835 + }, + { + "epoch": 3.175593594699061, + "grad_norm": 0.9398961067199707, + "learning_rate": 2.318710513915006e-06, + "loss": 0.571, + "step": 3836 + }, + { + "epoch": 3.17642186637217, + "grad_norm": 0.9242047667503357, + "learning_rate": 2.317613737289512e-06, + "loss": 0.5557, + "step": 3837 + }, + { + "epoch": 3.1772501380452787, + "grad_norm": 0.9494120478630066, + "learning_rate": 2.31651699595406e-06, + "loss": 0.568, + "step": 3838 + }, + { + "epoch": 3.1780784097183874, + "grad_norm": 0.9196777939796448, + "learning_rate": 2.3154202901208593e-06, + "loss": 0.5656, + "step": 3839 + }, + { + "epoch": 3.1789066813914966, + "grad_norm": 0.9382702112197876, + "learning_rate": 2.314323620002111e-06, + "loss": 0.5792, + "step": 3840 + }, + { + "epoch": 3.1797349530646053, + "grad_norm": 1.018182635307312, + "learning_rate": 2.3132269858100118e-06, + "loss": 0.5804, + "step": 3841 + }, + { + "epoch": 3.180563224737714, + "grad_norm": 0.9466580748558044, + "learning_rate": 2.31213038775675e-06, + "loss": 0.565, + "step": 3842 + }, + { + "epoch": 3.1813914964108228, + "grad_norm": 0.9668580889701843, + "learning_rate": 2.3110338260545062e-06, + "loss": 0.5724, + "step": 3843 + }, + { + "epoch": 3.1822197680839315, + "grad_norm": 0.9272748231887817, + "learning_rate": 2.309937300915456e-06, + "loss": 0.579, + "step": 3844 + }, + { + "epoch": 3.18304803975704, + "grad_norm": 0.9182299375534058, + "learning_rate": 2.3088408125517666e-06, + "loss": 0.5906, + "step": 3845 + }, + { + "epoch": 3.183876311430149, + "grad_norm": 1.034818410873413, + "learning_rate": 2.307744361175598e-06, + "loss": 0.5709, + "step": 3846 + }, + { + "epoch": 3.184704583103258, + "grad_norm": 0.9287042617797852, + "learning_rate": 2.306647946999104e-06, + "loss": 0.5855, + "step": 3847 + }, + { + "epoch": 3.185532854776367, + "grad_norm": 0.9378973841667175, + "learning_rate": 2.305551570234428e-06, + "loss": 0.5822, + "step": 3848 + }, + { + "epoch": 3.1863611264494756, + "grad_norm": 0.9447490572929382, + "learning_rate": 2.304455231093712e-06, + "loss": 0.5781, + "step": 3849 + }, + { + "epoch": 3.1871893981225843, + "grad_norm": 0.9659326076507568, + "learning_rate": 2.303358929789085e-06, + "loss": 0.6217, + "step": 3850 + }, + { + "epoch": 3.188017669795693, + "grad_norm": 0.9089621305465698, + "learning_rate": 2.302262666532672e-06, + "loss": 0.5733, + "step": 3851 + }, + { + "epoch": 3.1888459414688017, + "grad_norm": 0.9480805993080139, + "learning_rate": 2.3011664415365905e-06, + "loss": 0.5615, + "step": 3852 + }, + { + "epoch": 3.1896742131419105, + "grad_norm": 0.9534367322921753, + "learning_rate": 2.300070255012949e-06, + "loss": 0.5776, + "step": 3853 + }, + { + "epoch": 3.190502484815019, + "grad_norm": 0.9526930451393127, + "learning_rate": 2.298974107173849e-06, + "loss": 0.5541, + "step": 3854 + }, + { + "epoch": 3.191330756488128, + "grad_norm": 0.9261385202407837, + "learning_rate": 2.2978779982313853e-06, + "loss": 0.5797, + "step": 3855 + }, + { + "epoch": 3.192159028161237, + "grad_norm": 0.9275192618370056, + "learning_rate": 2.296781928397645e-06, + "loss": 0.559, + "step": 3856 + }, + { + "epoch": 3.192987299834346, + "grad_norm": 0.9341071248054504, + "learning_rate": 2.2956858978847076e-06, + "loss": 0.5523, + "step": 3857 + }, + { + "epoch": 3.1938155715074545, + "grad_norm": 0.9231052398681641, + "learning_rate": 2.2945899069046444e-06, + "loss": 0.5663, + "step": 3858 + }, + { + "epoch": 3.1946438431805633, + "grad_norm": 0.9468841552734375, + "learning_rate": 2.293493955669519e-06, + "loss": 0.5815, + "step": 3859 + }, + { + "epoch": 3.195472114853672, + "grad_norm": 0.9482542276382446, + "learning_rate": 2.2923980443913876e-06, + "loss": 0.5679, + "step": 3860 + }, + { + "epoch": 3.1963003865267807, + "grad_norm": 0.9444701075553894, + "learning_rate": 2.2913021732822997e-06, + "loss": 0.5796, + "step": 3861 + }, + { + "epoch": 3.1971286581998895, + "grad_norm": 0.9483172297477722, + "learning_rate": 2.2902063425542963e-06, + "loss": 0.5391, + "step": 3862 + }, + { + "epoch": 3.197956929872998, + "grad_norm": 0.8964424729347229, + "learning_rate": 2.289110552419409e-06, + "loss": 0.5719, + "step": 3863 + }, + { + "epoch": 3.198785201546107, + "grad_norm": 0.9366036653518677, + "learning_rate": 2.2880148030896648e-06, + "loss": 0.5729, + "step": 3864 + }, + { + "epoch": 3.199613473219216, + "grad_norm": 0.9433067440986633, + "learning_rate": 2.28691909477708e-06, + "loss": 0.5811, + "step": 3865 + }, + { + "epoch": 3.200441744892325, + "grad_norm": 1.0146312713623047, + "learning_rate": 2.285823427693664e-06, + "loss": 0.5524, + "step": 3866 + }, + { + "epoch": 3.2012700165654335, + "grad_norm": 0.9418644905090332, + "learning_rate": 2.2847278020514176e-06, + "loss": 0.5687, + "step": 3867 + }, + { + "epoch": 3.2020982882385423, + "grad_norm": 0.9421892166137695, + "learning_rate": 2.2836322180623355e-06, + "loss": 0.559, + "step": 3868 + }, + { + "epoch": 3.202926559911651, + "grad_norm": 0.9407430291175842, + "learning_rate": 2.2825366759384027e-06, + "loss": 0.5876, + "step": 3869 + }, + { + "epoch": 3.2037548315847597, + "grad_norm": 0.9176812767982483, + "learning_rate": 2.281441175891595e-06, + "loss": 0.5514, + "step": 3870 + }, + { + "epoch": 3.2045831032578684, + "grad_norm": 0.9171411991119385, + "learning_rate": 2.2803457181338825e-06, + "loss": 0.5778, + "step": 3871 + }, + { + "epoch": 3.205411374930977, + "grad_norm": 0.9477943778038025, + "learning_rate": 2.279250302877226e-06, + "loss": 0.5806, + "step": 3872 + }, + { + "epoch": 3.2062396466040863, + "grad_norm": 0.9568420052528381, + "learning_rate": 2.2781549303335783e-06, + "loss": 0.6025, + "step": 3873 + }, + { + "epoch": 3.207067918277195, + "grad_norm": 0.9442486763000488, + "learning_rate": 2.2770596007148833e-06, + "loss": 0.5678, + "step": 3874 + }, + { + "epoch": 3.207896189950304, + "grad_norm": 0.9460006356239319, + "learning_rate": 2.2759643142330775e-06, + "loss": 0.5651, + "step": 3875 + }, + { + "epoch": 3.2087244616234125, + "grad_norm": 0.9530642032623291, + "learning_rate": 2.2748690711000885e-06, + "loss": 0.5704, + "step": 3876 + }, + { + "epoch": 3.2095527332965212, + "grad_norm": 0.9378021955490112, + "learning_rate": 2.2737738715278353e-06, + "loss": 0.5605, + "step": 3877 + }, + { + "epoch": 3.21038100496963, + "grad_norm": 0.9239064455032349, + "learning_rate": 2.272678715728229e-06, + "loss": 0.5514, + "step": 3878 + }, + { + "epoch": 3.2112092766427387, + "grad_norm": 0.9366111755371094, + "learning_rate": 2.271583603913173e-06, + "loss": 0.5746, + "step": 3879 + }, + { + "epoch": 3.2120375483158474, + "grad_norm": 0.9362907409667969, + "learning_rate": 2.270488536294559e-06, + "loss": 0.5622, + "step": 3880 + }, + { + "epoch": 3.2128658199889566, + "grad_norm": 0.9395036697387695, + "learning_rate": 2.2693935130842763e-06, + "loss": 0.5765, + "step": 3881 + }, + { + "epoch": 3.2136940916620653, + "grad_norm": 0.9549781084060669, + "learning_rate": 2.2682985344941966e-06, + "loss": 0.5716, + "step": 3882 + }, + { + "epoch": 3.214522363335174, + "grad_norm": 0.9677711129188538, + "learning_rate": 2.2672036007361915e-06, + "loss": 0.5922, + "step": 3883 + }, + { + "epoch": 3.2153506350082828, + "grad_norm": 0.9358668923377991, + "learning_rate": 2.26610871202212e-06, + "loss": 0.5794, + "step": 3884 + }, + { + "epoch": 3.2161789066813915, + "grad_norm": 0.938670814037323, + "learning_rate": 2.265013868563831e-06, + "loss": 0.5809, + "step": 3885 + }, + { + "epoch": 3.2170071783545002, + "grad_norm": 0.9247921109199524, + "learning_rate": 2.2639190705731694e-06, + "loss": 0.5714, + "step": 3886 + }, + { + "epoch": 3.217835450027609, + "grad_norm": 0.9363803863525391, + "learning_rate": 2.262824318261966e-06, + "loss": 0.5714, + "step": 3887 + }, + { + "epoch": 3.2186637217007177, + "grad_norm": 0.9042513370513916, + "learning_rate": 2.2617296118420463e-06, + "loss": 0.5612, + "step": 3888 + }, + { + "epoch": 3.2194919933738264, + "grad_norm": 0.9226694703102112, + "learning_rate": 2.260634951525226e-06, + "loss": 0.5579, + "step": 3889 + }, + { + "epoch": 3.2203202650469356, + "grad_norm": 0.9465897679328918, + "learning_rate": 2.2595403375233107e-06, + "loss": 0.5507, + "step": 3890 + }, + { + "epoch": 3.2211485367200443, + "grad_norm": 0.9200103282928467, + "learning_rate": 2.2584457700480984e-06, + "loss": 0.568, + "step": 3891 + }, + { + "epoch": 3.221976808393153, + "grad_norm": 0.9169338345527649, + "learning_rate": 2.2573512493113797e-06, + "loss": 0.5588, + "step": 3892 + }, + { + "epoch": 3.2228050800662618, + "grad_norm": 0.9455548524856567, + "learning_rate": 2.2562567755249304e-06, + "loss": 0.5725, + "step": 3893 + }, + { + "epoch": 3.2236333517393705, + "grad_norm": 0.9347724318504333, + "learning_rate": 2.2551623489005233e-06, + "loss": 0.5659, + "step": 3894 + }, + { + "epoch": 3.224461623412479, + "grad_norm": 0.9664618372917175, + "learning_rate": 2.2540679696499194e-06, + "loss": 0.5584, + "step": 3895 + }, + { + "epoch": 3.225289895085588, + "grad_norm": 1.0180222988128662, + "learning_rate": 2.252973637984871e-06, + "loss": 0.5586, + "step": 3896 + }, + { + "epoch": 3.2261181667586967, + "grad_norm": 0.9532806277275085, + "learning_rate": 2.2518793541171204e-06, + "loss": 0.5583, + "step": 3897 + }, + { + "epoch": 3.2269464384318054, + "grad_norm": 0.9502059817314148, + "learning_rate": 2.2507851182584024e-06, + "loss": 0.5676, + "step": 3898 + }, + { + "epoch": 3.2277747101049146, + "grad_norm": 0.9380828738212585, + "learning_rate": 2.24969093062044e-06, + "loss": 0.5644, + "step": 3899 + }, + { + "epoch": 3.2286029817780233, + "grad_norm": 0.9528313875198364, + "learning_rate": 2.2485967914149503e-06, + "loss": 0.5704, + "step": 3900 + }, + { + "epoch": 3.229431253451132, + "grad_norm": 0.9411875009536743, + "learning_rate": 2.2475027008536374e-06, + "loss": 0.5694, + "step": 3901 + }, + { + "epoch": 3.2302595251242407, + "grad_norm": 0.9563047289848328, + "learning_rate": 2.2464086591481977e-06, + "loss": 0.5654, + "step": 3902 + }, + { + "epoch": 3.2310877967973495, + "grad_norm": 0.9029218554496765, + "learning_rate": 2.24531466651032e-06, + "loss": 0.5818, + "step": 3903 + }, + { + "epoch": 3.231916068470458, + "grad_norm": 0.9468775391578674, + "learning_rate": 2.244220723151679e-06, + "loss": 0.548, + "step": 3904 + }, + { + "epoch": 3.232744340143567, + "grad_norm": 0.9399493932723999, + "learning_rate": 2.243126829283944e-06, + "loss": 0.5746, + "step": 3905 + }, + { + "epoch": 3.233572611816676, + "grad_norm": 0.9424317479133606, + "learning_rate": 2.2420329851187726e-06, + "loss": 0.5775, + "step": 3906 + }, + { + "epoch": 3.234400883489785, + "grad_norm": 0.9486375451087952, + "learning_rate": 2.2409391908678137e-06, + "loss": 0.5815, + "step": 3907 + }, + { + "epoch": 3.2352291551628936, + "grad_norm": 0.9591338634490967, + "learning_rate": 2.2398454467427073e-06, + "loss": 0.5826, + "step": 3908 + }, + { + "epoch": 3.2360574268360023, + "grad_norm": 0.9704396724700928, + "learning_rate": 2.2387517529550813e-06, + "loss": 0.5912, + "step": 3909 + }, + { + "epoch": 3.236885698509111, + "grad_norm": 0.9416005611419678, + "learning_rate": 2.2376581097165553e-06, + "loss": 0.5614, + "step": 3910 + }, + { + "epoch": 3.2377139701822197, + "grad_norm": 1.0255080461502075, + "learning_rate": 2.2365645172387403e-06, + "loss": 0.5688, + "step": 3911 + }, + { + "epoch": 3.2385422418553285, + "grad_norm": 0.9364955425262451, + "learning_rate": 2.2354709757332356e-06, + "loss": 0.572, + "step": 3912 + }, + { + "epoch": 3.239370513528437, + "grad_norm": 0.9503747224807739, + "learning_rate": 2.2343774854116306e-06, + "loss": 0.5893, + "step": 3913 + }, + { + "epoch": 3.240198785201546, + "grad_norm": 0.9294464588165283, + "learning_rate": 2.2332840464855072e-06, + "loss": 0.5671, + "step": 3914 + }, + { + "epoch": 3.241027056874655, + "grad_norm": 0.9659671783447266, + "learning_rate": 2.232190659166433e-06, + "loss": 0.5741, + "step": 3915 + }, + { + "epoch": 3.241855328547764, + "grad_norm": 0.9359234571456909, + "learning_rate": 2.231097323665971e-06, + "loss": 0.5608, + "step": 3916 + }, + { + "epoch": 3.2426836002208725, + "grad_norm": 0.9326402544975281, + "learning_rate": 2.2300040401956693e-06, + "loss": 0.5696, + "step": 3917 + }, + { + "epoch": 3.2435118718939813, + "grad_norm": 0.9257575869560242, + "learning_rate": 2.2289108089670687e-06, + "loss": 0.5709, + "step": 3918 + }, + { + "epoch": 3.24434014356709, + "grad_norm": 0.9669248461723328, + "learning_rate": 2.2278176301917004e-06, + "loss": 0.5666, + "step": 3919 + }, + { + "epoch": 3.2451684152401987, + "grad_norm": 0.9328755736351013, + "learning_rate": 2.2267245040810832e-06, + "loss": 0.573, + "step": 3920 + }, + { + "epoch": 3.2459966869133074, + "grad_norm": 0.942805826663971, + "learning_rate": 2.225631430846727e-06, + "loss": 0.5712, + "step": 3921 + }, + { + "epoch": 3.246824958586416, + "grad_norm": 0.9514304995536804, + "learning_rate": 2.224538410700131e-06, + "loss": 0.5775, + "step": 3922 + }, + { + "epoch": 3.247653230259525, + "grad_norm": 0.9471868872642517, + "learning_rate": 2.2234454438527847e-06, + "loss": 0.5794, + "step": 3923 + }, + { + "epoch": 3.248481501932634, + "grad_norm": 0.9507884979248047, + "learning_rate": 2.2223525305161675e-06, + "loss": 0.5726, + "step": 3924 + }, + { + "epoch": 3.249309773605743, + "grad_norm": 0.9477060437202454, + "learning_rate": 2.221259670901747e-06, + "loss": 0.5895, + "step": 3925 + }, + { + "epoch": 3.2501380452788515, + "grad_norm": 0.9437663555145264, + "learning_rate": 2.2201668652209823e-06, + "loss": 0.5538, + "step": 3926 + }, + { + "epoch": 3.2509663169519603, + "grad_norm": 0.9790158271789551, + "learning_rate": 2.2190741136853205e-06, + "loss": 0.5729, + "step": 3927 + }, + { + "epoch": 3.251794588625069, + "grad_norm": 0.9320138692855835, + "learning_rate": 2.2179814165061984e-06, + "loss": 0.5807, + "step": 3928 + }, + { + "epoch": 3.2526228602981777, + "grad_norm": 0.9543532729148865, + "learning_rate": 2.2168887738950438e-06, + "loss": 0.5645, + "step": 3929 + }, + { + "epoch": 3.2534511319712864, + "grad_norm": 0.9545353055000305, + "learning_rate": 2.2157961860632716e-06, + "loss": 0.5762, + "step": 3930 + }, + { + "epoch": 3.2542794036443956, + "grad_norm": 0.9640288352966309, + "learning_rate": 2.2147036532222887e-06, + "loss": 0.5738, + "step": 3931 + }, + { + "epoch": 3.2551076753175043, + "grad_norm": 0.9393999576568604, + "learning_rate": 2.2136111755834895e-06, + "loss": 0.5692, + "step": 3932 + }, + { + "epoch": 3.255935946990613, + "grad_norm": 0.9366894364356995, + "learning_rate": 2.212518753358258e-06, + "loss": 0.5957, + "step": 3933 + }, + { + "epoch": 3.256764218663722, + "grad_norm": 0.9408980011940002, + "learning_rate": 2.2114263867579673e-06, + "loss": 0.5644, + "step": 3934 + }, + { + "epoch": 3.2575924903368305, + "grad_norm": 0.9842433333396912, + "learning_rate": 2.2103340759939816e-06, + "loss": 0.5772, + "step": 3935 + }, + { + "epoch": 3.2584207620099392, + "grad_norm": 0.9410622119903564, + "learning_rate": 2.209241821277651e-06, + "loss": 0.5637, + "step": 3936 + }, + { + "epoch": 3.259249033683048, + "grad_norm": 0.9571241736412048, + "learning_rate": 2.208149622820319e-06, + "loss": 0.5754, + "step": 3937 + }, + { + "epoch": 3.2600773053561567, + "grad_norm": 0.9451292157173157, + "learning_rate": 2.2070574808333135e-06, + "loss": 0.557, + "step": 3938 + }, + { + "epoch": 3.2609055770292654, + "grad_norm": 0.987385094165802, + "learning_rate": 2.205965395527955e-06, + "loss": 0.5458, + "step": 3939 + }, + { + "epoch": 3.2617338487023746, + "grad_norm": 0.9282477498054504, + "learning_rate": 2.204873367115551e-06, + "loss": 0.5726, + "step": 3940 + }, + { + "epoch": 3.2625621203754833, + "grad_norm": 0.9609903693199158, + "learning_rate": 2.2037813958073994e-06, + "loss": 0.5608, + "step": 3941 + }, + { + "epoch": 3.263390392048592, + "grad_norm": 0.9469774961471558, + "learning_rate": 2.2026894818147864e-06, + "loss": 0.5663, + "step": 3942 + }, + { + "epoch": 3.2642186637217008, + "grad_norm": 0.9910560846328735, + "learning_rate": 2.2015976253489873e-06, + "loss": 0.5604, + "step": 3943 + }, + { + "epoch": 3.2650469353948095, + "grad_norm": 0.949170708656311, + "learning_rate": 2.2005058266212657e-06, + "loss": 0.5795, + "step": 3944 + }, + { + "epoch": 3.2658752070679182, + "grad_norm": 0.9276071786880493, + "learning_rate": 2.199414085842875e-06, + "loss": 0.5821, + "step": 3945 + }, + { + "epoch": 3.266703478741027, + "grad_norm": 0.9555459022521973, + "learning_rate": 2.1983224032250573e-06, + "loss": 0.5832, + "step": 3946 + }, + { + "epoch": 3.2675317504141357, + "grad_norm": 0.972984790802002, + "learning_rate": 2.1972307789790413e-06, + "loss": 0.5868, + "step": 3947 + }, + { + "epoch": 3.2683600220872444, + "grad_norm": 0.9383103251457214, + "learning_rate": 2.196139213316049e-06, + "loss": 0.5624, + "step": 3948 + }, + { + "epoch": 3.2691882937603536, + "grad_norm": 0.9539969563484192, + "learning_rate": 2.195047706447285e-06, + "loss": 0.566, + "step": 3949 + }, + { + "epoch": 3.2700165654334623, + "grad_norm": 0.955715537071228, + "learning_rate": 2.193956258583947e-06, + "loss": 0.5548, + "step": 3950 + }, + { + "epoch": 3.270844837106571, + "grad_norm": 0.9660782217979431, + "learning_rate": 2.19286486993722e-06, + "loss": 0.5726, + "step": 3951 + }, + { + "epoch": 3.2716731087796798, + "grad_norm": 0.9660666584968567, + "learning_rate": 2.1917735407182777e-06, + "loss": 0.5728, + "step": 3952 + }, + { + "epoch": 3.2725013804527885, + "grad_norm": 0.9096773862838745, + "learning_rate": 2.190682271138282e-06, + "loss": 0.5549, + "step": 3953 + }, + { + "epoch": 3.273329652125897, + "grad_norm": 0.9581109285354614, + "learning_rate": 2.189591061408384e-06, + "loss": 0.5834, + "step": 3954 + }, + { + "epoch": 3.274157923799006, + "grad_norm": 0.9483600854873657, + "learning_rate": 2.188499911739721e-06, + "loss": 0.5749, + "step": 3955 + }, + { + "epoch": 3.2749861954721147, + "grad_norm": 0.9495676159858704, + "learning_rate": 2.187408822343422e-06, + "loss": 0.5907, + "step": 3956 + }, + { + "epoch": 3.2758144671452234, + "grad_norm": 0.9316544532775879, + "learning_rate": 2.186317793430602e-06, + "loss": 0.555, + "step": 3957 + }, + { + "epoch": 3.2766427388183326, + "grad_norm": 0.9391309022903442, + "learning_rate": 2.1852268252123644e-06, + "loss": 0.5751, + "step": 3958 + }, + { + "epoch": 3.2774710104914413, + "grad_norm": 0.91743403673172, + "learning_rate": 2.1841359178998024e-06, + "loss": 0.5504, + "step": 3959 + }, + { + "epoch": 3.27829928216455, + "grad_norm": 0.9177739024162292, + "learning_rate": 2.183045071703996e-06, + "loss": 0.5452, + "step": 3960 + }, + { + "epoch": 3.2791275538376587, + "grad_norm": 0.964051365852356, + "learning_rate": 2.1819542868360135e-06, + "loss": 0.5731, + "step": 3961 + }, + { + "epoch": 3.2799558255107675, + "grad_norm": 0.9402073621749878, + "learning_rate": 2.1808635635069113e-06, + "loss": 0.5655, + "step": 3962 + }, + { + "epoch": 3.280784097183876, + "grad_norm": 0.9163627624511719, + "learning_rate": 2.1797729019277346e-06, + "loss": 0.5772, + "step": 3963 + }, + { + "epoch": 3.281612368856985, + "grad_norm": 0.9189632534980774, + "learning_rate": 2.1786823023095164e-06, + "loss": 0.5738, + "step": 3964 + }, + { + "epoch": 3.282440640530094, + "grad_norm": 0.9654256701469421, + "learning_rate": 2.1775917648632772e-06, + "loss": 0.5752, + "step": 3965 + }, + { + "epoch": 3.283268912203203, + "grad_norm": 1.0078915357589722, + "learning_rate": 2.1765012898000257e-06, + "loss": 0.5847, + "step": 3966 + }, + { + "epoch": 3.2840971838763116, + "grad_norm": 0.9313945174217224, + "learning_rate": 2.17541087733076e-06, + "loss": 0.5671, + "step": 3967 + }, + { + "epoch": 3.2849254555494203, + "grad_norm": 0.9611067771911621, + "learning_rate": 2.1743205276664634e-06, + "loss": 0.5501, + "step": 3968 + }, + { + "epoch": 3.285753727222529, + "grad_norm": 0.9783805012702942, + "learning_rate": 2.1732302410181084e-06, + "loss": 0.5969, + "step": 3969 + }, + { + "epoch": 3.2865819988956377, + "grad_norm": 0.9382510781288147, + "learning_rate": 2.172140017596656e-06, + "loss": 0.5716, + "step": 3970 + }, + { + "epoch": 3.2874102705687465, + "grad_norm": 0.9219527840614319, + "learning_rate": 2.1710498576130545e-06, + "loss": 0.5916, + "step": 3971 + }, + { + "epoch": 3.288238542241855, + "grad_norm": 0.9681289792060852, + "learning_rate": 2.1699597612782385e-06, + "loss": 0.5725, + "step": 3972 + }, + { + "epoch": 3.289066813914964, + "grad_norm": 0.946723997592926, + "learning_rate": 2.168869728803132e-06, + "loss": 0.5664, + "step": 3973 + }, + { + "epoch": 3.289895085588073, + "grad_norm": 0.9302477836608887, + "learning_rate": 2.167779760398646e-06, + "loss": 0.5777, + "step": 3974 + }, + { + "epoch": 3.290723357261182, + "grad_norm": 0.9480409026145935, + "learning_rate": 2.16668985627568e-06, + "loss": 0.5696, + "step": 3975 + }, + { + "epoch": 3.2915516289342905, + "grad_norm": 0.9363623261451721, + "learning_rate": 2.1656000166451195e-06, + "loss": 0.5798, + "step": 3976 + }, + { + "epoch": 3.2923799006073993, + "grad_norm": 0.9291804432868958, + "learning_rate": 2.164510241717838e-06, + "loss": 0.5658, + "step": 3977 + }, + { + "epoch": 3.293208172280508, + "grad_norm": 0.9314830899238586, + "learning_rate": 2.163420531704698e-06, + "loss": 0.5698, + "step": 3978 + }, + { + "epoch": 3.2940364439536167, + "grad_norm": 0.974621057510376, + "learning_rate": 2.1623308868165474e-06, + "loss": 0.5644, + "step": 3979 + }, + { + "epoch": 3.2948647156267254, + "grad_norm": 0.9464275240898132, + "learning_rate": 2.161241307264222e-06, + "loss": 0.5704, + "step": 3980 + }, + { + "epoch": 3.295692987299834, + "grad_norm": 0.9375281929969788, + "learning_rate": 2.1601517932585462e-06, + "loss": 0.5714, + "step": 3981 + }, + { + "epoch": 3.296521258972943, + "grad_norm": 0.910499095916748, + "learning_rate": 2.159062345010331e-06, + "loss": 0.5639, + "step": 3982 + }, + { + "epoch": 3.297349530646052, + "grad_norm": 0.9525747895240784, + "learning_rate": 2.1579729627303735e-06, + "loss": 0.5703, + "step": 3983 + }, + { + "epoch": 3.298177802319161, + "grad_norm": 0.9756594896316528, + "learning_rate": 2.1568836466294596e-06, + "loss": 0.5862, + "step": 3984 + }, + { + "epoch": 3.2990060739922695, + "grad_norm": 0.9418894648551941, + "learning_rate": 2.1557943969183605e-06, + "loss": 0.5554, + "step": 3985 + }, + { + "epoch": 3.2998343456653783, + "grad_norm": 0.9190452694892883, + "learning_rate": 2.154705213807838e-06, + "loss": 0.5519, + "step": 3986 + }, + { + "epoch": 3.300662617338487, + "grad_norm": 0.9576460719108582, + "learning_rate": 2.153616097508637e-06, + "loss": 0.5723, + "step": 3987 + }, + { + "epoch": 3.3014908890115957, + "grad_norm": 0.9196321964263916, + "learning_rate": 2.1525270482314924e-06, + "loss": 0.6006, + "step": 3988 + }, + { + "epoch": 3.3023191606847044, + "grad_norm": 0.9210584163665771, + "learning_rate": 2.1514380661871253e-06, + "loss": 0.5569, + "step": 3989 + }, + { + "epoch": 3.3031474323578136, + "grad_norm": 0.9365322589874268, + "learning_rate": 2.150349151586243e-06, + "loss": 0.5747, + "step": 3990 + }, + { + "epoch": 3.3039757040309223, + "grad_norm": 0.948229193687439, + "learning_rate": 2.1492603046395407e-06, + "loss": 0.556, + "step": 3991 + }, + { + "epoch": 3.304803975704031, + "grad_norm": 0.9285109043121338, + "learning_rate": 2.148171525557699e-06, + "loss": 0.5747, + "step": 3992 + }, + { + "epoch": 3.30563224737714, + "grad_norm": 0.9224300384521484, + "learning_rate": 2.1470828145513882e-06, + "loss": 0.5883, + "step": 3993 + }, + { + "epoch": 3.3064605190502485, + "grad_norm": 0.9375912547111511, + "learning_rate": 2.1459941718312633e-06, + "loss": 0.5663, + "step": 3994 + }, + { + "epoch": 3.3072887907233572, + "grad_norm": 0.9629141688346863, + "learning_rate": 2.144905597607966e-06, + "loss": 0.5593, + "step": 3995 + }, + { + "epoch": 3.308117062396466, + "grad_norm": 0.9608798623085022, + "learning_rate": 2.143817092092124e-06, + "loss": 0.573, + "step": 3996 + }, + { + "epoch": 3.3089453340695747, + "grad_norm": 0.9235004186630249, + "learning_rate": 2.1427286554943557e-06, + "loss": 0.5818, + "step": 3997 + }, + { + "epoch": 3.3097736057426834, + "grad_norm": 0.9361316561698914, + "learning_rate": 2.141640288025262e-06, + "loss": 0.556, + "step": 3998 + }, + { + "epoch": 3.3106018774157926, + "grad_norm": 0.9709519743919373, + "learning_rate": 2.140551989895432e-06, + "loss": 0.5575, + "step": 3999 + }, + { + "epoch": 3.3114301490889013, + "grad_norm": 0.9461862444877625, + "learning_rate": 2.1394637613154403e-06, + "loss": 0.5676, + "step": 4000 + }, + { + "epoch": 3.31225842076201, + "grad_norm": 0.9697636961936951, + "learning_rate": 2.138375602495851e-06, + "loss": 0.5678, + "step": 4001 + }, + { + "epoch": 3.3130866924351188, + "grad_norm": 0.9442050457000732, + "learning_rate": 2.1372875136472113e-06, + "loss": 0.573, + "step": 4002 + }, + { + "epoch": 3.3139149641082275, + "grad_norm": 0.9325514435768127, + "learning_rate": 2.136199494980056e-06, + "loss": 0.5711, + "step": 4003 + }, + { + "epoch": 3.3147432357813362, + "grad_norm": 0.934421181678772, + "learning_rate": 2.135111546704908e-06, + "loss": 0.5634, + "step": 4004 + }, + { + "epoch": 3.315571507454445, + "grad_norm": 0.9201632738113403, + "learning_rate": 2.134023669032274e-06, + "loss": 0.5627, + "step": 4005 + }, + { + "epoch": 3.3163997791275537, + "grad_norm": 0.9563625454902649, + "learning_rate": 2.1329358621726484e-06, + "loss": 0.5668, + "step": 4006 + }, + { + "epoch": 3.3172280508006624, + "grad_norm": 0.9408397078514099, + "learning_rate": 2.1318481263365117e-06, + "loss": 0.5712, + "step": 4007 + }, + { + "epoch": 3.3180563224737716, + "grad_norm": 0.9219162464141846, + "learning_rate": 2.13076046173433e-06, + "loss": 0.5694, + "step": 4008 + }, + { + "epoch": 3.3188845941468803, + "grad_norm": 0.9441806077957153, + "learning_rate": 2.129672868576558e-06, + "loss": 0.5901, + "step": 4009 + }, + { + "epoch": 3.319712865819989, + "grad_norm": 0.9196688532829285, + "learning_rate": 2.1285853470736336e-06, + "loss": 0.5768, + "step": 4010 + }, + { + "epoch": 3.3205411374930978, + "grad_norm": 0.9800437688827515, + "learning_rate": 2.1274978974359813e-06, + "loss": 0.5728, + "step": 4011 + }, + { + "epoch": 3.3213694091662065, + "grad_norm": 0.9329850077629089, + "learning_rate": 2.126410519874014e-06, + "loss": 0.5806, + "step": 4012 + }, + { + "epoch": 3.322197680839315, + "grad_norm": 0.9533185362815857, + "learning_rate": 2.1253232145981293e-06, + "loss": 0.5665, + "step": 4013 + }, + { + "epoch": 3.323025952512424, + "grad_norm": 0.9292604327201843, + "learning_rate": 2.1242359818187085e-06, + "loss": 0.5552, + "step": 4014 + }, + { + "epoch": 3.3238542241855327, + "grad_norm": 0.9430820941925049, + "learning_rate": 2.1231488217461234e-06, + "loss": 0.5849, + "step": 4015 + }, + { + "epoch": 3.3246824958586414, + "grad_norm": 0.9454953670501709, + "learning_rate": 2.1220617345907287e-06, + "loss": 0.5724, + "step": 4016 + }, + { + "epoch": 3.3255107675317506, + "grad_norm": 0.9445587396621704, + "learning_rate": 2.120974720562865e-06, + "loss": 0.5908, + "step": 4017 + }, + { + "epoch": 3.3263390392048593, + "grad_norm": 0.9306840896606445, + "learning_rate": 2.1198877798728595e-06, + "loss": 0.5938, + "step": 4018 + }, + { + "epoch": 3.327167310877968, + "grad_norm": 0.9418679475784302, + "learning_rate": 2.1188009127310244e-06, + "loss": 0.5692, + "step": 4019 + }, + { + "epoch": 3.3279955825510767, + "grad_norm": 0.9448477625846863, + "learning_rate": 2.11771411934766e-06, + "loss": 0.578, + "step": 4020 + }, + { + "epoch": 3.3288238542241855, + "grad_norm": 0.9564743638038635, + "learning_rate": 2.1166273999330497e-06, + "loss": 0.5577, + "step": 4021 + }, + { + "epoch": 3.329652125897294, + "grad_norm": 0.9450514316558838, + "learning_rate": 2.115540754697463e-06, + "loss": 0.5812, + "step": 4022 + }, + { + "epoch": 3.330480397570403, + "grad_norm": 0.932601809501648, + "learning_rate": 2.1144541838511578e-06, + "loss": 0.5534, + "step": 4023 + }, + { + "epoch": 3.331308669243512, + "grad_norm": 0.9476706385612488, + "learning_rate": 2.1133676876043734e-06, + "loss": 0.5684, + "step": 4024 + }, + { + "epoch": 3.332136940916621, + "grad_norm": 0.9359769225120544, + "learning_rate": 2.1122812661673365e-06, + "loss": 0.5624, + "step": 4025 + }, + { + "epoch": 3.3329652125897296, + "grad_norm": 0.9415692687034607, + "learning_rate": 2.1111949197502612e-06, + "loss": 0.5627, + "step": 4026 + }, + { + "epoch": 3.3337934842628383, + "grad_norm": 0.9686550498008728, + "learning_rate": 2.1101086485633455e-06, + "loss": 0.5401, + "step": 4027 + }, + { + "epoch": 3.334621755935947, + "grad_norm": 0.9065554738044739, + "learning_rate": 2.1090224528167703e-06, + "loss": 0.5785, + "step": 4028 + }, + { + "epoch": 3.3354500276090557, + "grad_norm": 0.9449801445007324, + "learning_rate": 2.1079363327207067e-06, + "loss": 0.5695, + "step": 4029 + }, + { + "epoch": 3.3362782992821645, + "grad_norm": 0.9398133754730225, + "learning_rate": 2.106850288485307e-06, + "loss": 0.5605, + "step": 4030 + }, + { + "epoch": 3.337106570955273, + "grad_norm": 0.9175443649291992, + "learning_rate": 2.1057643203207125e-06, + "loss": 0.5675, + "step": 4031 + }, + { + "epoch": 3.337934842628382, + "grad_norm": 0.9433302879333496, + "learning_rate": 2.1046784284370466e-06, + "loss": 0.5501, + "step": 4032 + }, + { + "epoch": 3.338763114301491, + "grad_norm": 0.9375688433647156, + "learning_rate": 2.103592613044419e-06, + "loss": 0.579, + "step": 4033 + }, + { + "epoch": 3.3395913859746, + "grad_norm": 0.9344912171363831, + "learning_rate": 2.102506874352926e-06, + "loss": 0.5698, + "step": 4034 + }, + { + "epoch": 3.3404196576477085, + "grad_norm": 0.9581173062324524, + "learning_rate": 2.101421212572648e-06, + "loss": 0.5774, + "step": 4035 + }, + { + "epoch": 3.3412479293208173, + "grad_norm": 0.9601131081581116, + "learning_rate": 2.100335627913649e-06, + "loss": 0.5769, + "step": 4036 + }, + { + "epoch": 3.342076200993926, + "grad_norm": 0.9634785652160645, + "learning_rate": 2.0992501205859806e-06, + "loss": 0.5883, + "step": 4037 + }, + { + "epoch": 3.3429044726670347, + "grad_norm": 1.0006943941116333, + "learning_rate": 2.0981646907996784e-06, + "loss": 0.5744, + "step": 4038 + }, + { + "epoch": 3.3437327443401434, + "grad_norm": 0.9540338516235352, + "learning_rate": 2.0970793387647636e-06, + "loss": 0.5775, + "step": 4039 + }, + { + "epoch": 3.344561016013252, + "grad_norm": 0.9318099617958069, + "learning_rate": 2.09599406469124e-06, + "loss": 0.5765, + "step": 4040 + }, + { + "epoch": 3.345389287686361, + "grad_norm": 0.9622073173522949, + "learning_rate": 2.0949088687890976e-06, + "loss": 0.5532, + "step": 4041 + }, + { + "epoch": 3.34621755935947, + "grad_norm": 0.9623757600784302, + "learning_rate": 2.0938237512683142e-06, + "loss": 0.5518, + "step": 4042 + }, + { + "epoch": 3.347045831032579, + "grad_norm": 0.9352701306343079, + "learning_rate": 2.0927387123388485e-06, + "loss": 0.5488, + "step": 4043 + }, + { + "epoch": 3.3478741027056875, + "grad_norm": 0.9591230154037476, + "learning_rate": 2.0916537522106446e-06, + "loss": 0.5887, + "step": 4044 + }, + { + "epoch": 3.3487023743787963, + "grad_norm": 0.9517300724983215, + "learning_rate": 2.0905688710936338e-06, + "loss": 0.5619, + "step": 4045 + }, + { + "epoch": 3.349530646051905, + "grad_norm": 0.9537909626960754, + "learning_rate": 2.08948406919773e-06, + "loss": 0.5949, + "step": 4046 + }, + { + "epoch": 3.3503589177250137, + "grad_norm": 0.946260929107666, + "learning_rate": 2.088399346732831e-06, + "loss": 0.5989, + "step": 4047 + }, + { + "epoch": 3.3511871893981224, + "grad_norm": 0.9061339497566223, + "learning_rate": 2.0873147039088227e-06, + "loss": 0.5855, + "step": 4048 + }, + { + "epoch": 3.3520154610712316, + "grad_norm": 0.9527447819709778, + "learning_rate": 2.0862301409355717e-06, + "loss": 0.5703, + "step": 4049 + }, + { + "epoch": 3.3528437327443403, + "grad_norm": 0.9357138872146606, + "learning_rate": 2.0851456580229325e-06, + "loss": 0.551, + "step": 4050 + }, + { + "epoch": 3.353672004417449, + "grad_norm": 0.9259575009346008, + "learning_rate": 2.08406125538074e-06, + "loss": 0.5777, + "step": 4051 + }, + { + "epoch": 3.354500276090558, + "grad_norm": 0.944807231426239, + "learning_rate": 2.082976933218817e-06, + "loss": 0.5741, + "step": 4052 + }, + { + "epoch": 3.3553285477636665, + "grad_norm": 0.9251223206520081, + "learning_rate": 2.08189269174697e-06, + "loss": 0.5829, + "step": 4053 + }, + { + "epoch": 3.3561568194367752, + "grad_norm": 0.9122853875160217, + "learning_rate": 2.0808085311749904e-06, + "loss": 0.5602, + "step": 4054 + }, + { + "epoch": 3.356985091109884, + "grad_norm": 0.9430744647979736, + "learning_rate": 2.0797244517126512e-06, + "loss": 0.5633, + "step": 4055 + }, + { + "epoch": 3.3578133627829927, + "grad_norm": 0.9500707983970642, + "learning_rate": 2.078640453569714e-06, + "loss": 0.5791, + "step": 4056 + }, + { + "epoch": 3.3586416344561014, + "grad_norm": 0.970301628112793, + "learning_rate": 2.077556536955921e-06, + "loss": 0.5668, + "step": 4057 + }, + { + "epoch": 3.3594699061292106, + "grad_norm": 0.9528861045837402, + "learning_rate": 2.0764727020809995e-06, + "loss": 0.5616, + "step": 4058 + }, + { + "epoch": 3.3602981778023193, + "grad_norm": 0.9321720004081726, + "learning_rate": 2.0753889491546624e-06, + "loss": 0.5841, + "step": 4059 + }, + { + "epoch": 3.361126449475428, + "grad_norm": 0.9030178189277649, + "learning_rate": 2.0743052783866054e-06, + "loss": 0.5757, + "step": 4060 + }, + { + "epoch": 3.3619547211485368, + "grad_norm": 0.9501296877861023, + "learning_rate": 2.0732216899865098e-06, + "loss": 0.5477, + "step": 4061 + }, + { + "epoch": 3.3627829928216455, + "grad_norm": 0.9475130438804626, + "learning_rate": 2.0721381841640377e-06, + "loss": 0.5798, + "step": 4062 + }, + { + "epoch": 3.3636112644947542, + "grad_norm": 0.954949676990509, + "learning_rate": 2.0710547611288384e-06, + "loss": 0.579, + "step": 4063 + }, + { + "epoch": 3.364439536167863, + "grad_norm": 0.9534462690353394, + "learning_rate": 2.069971421090545e-06, + "loss": 0.5448, + "step": 4064 + }, + { + "epoch": 3.3652678078409717, + "grad_norm": 0.9504910111427307, + "learning_rate": 2.068888164258772e-06, + "loss": 0.5698, + "step": 4065 + }, + { + "epoch": 3.3660960795140804, + "grad_norm": 0.934983491897583, + "learning_rate": 2.067804990843121e-06, + "loss": 0.5643, + "step": 4066 + }, + { + "epoch": 3.3669243511871896, + "grad_norm": 0.9342026114463806, + "learning_rate": 2.066721901053175e-06, + "loss": 0.5938, + "step": 4067 + }, + { + "epoch": 3.3677526228602983, + "grad_norm": 0.9376976490020752, + "learning_rate": 2.0656388950985027e-06, + "loss": 0.5897, + "step": 4068 + }, + { + "epoch": 3.368580894533407, + "grad_norm": 0.9228143095970154, + "learning_rate": 2.064555973188655e-06, + "loss": 0.5789, + "step": 4069 + }, + { + "epoch": 3.3694091662065158, + "grad_norm": 0.9547181129455566, + "learning_rate": 2.0634731355331665e-06, + "loss": 0.5707, + "step": 4070 + }, + { + "epoch": 3.3702374378796245, + "grad_norm": 0.9388718605041504, + "learning_rate": 2.062390382341558e-06, + "loss": 0.5719, + "step": 4071 + }, + { + "epoch": 3.371065709552733, + "grad_norm": 0.9529668092727661, + "learning_rate": 2.0613077138233313e-06, + "loss": 0.5634, + "step": 4072 + }, + { + "epoch": 3.371893981225842, + "grad_norm": 0.9569711089134216, + "learning_rate": 2.060225130187973e-06, + "loss": 0.5565, + "step": 4073 + }, + { + "epoch": 3.3727222528989507, + "grad_norm": 0.936896026134491, + "learning_rate": 2.0591426316449513e-06, + "loss": 0.5543, + "step": 4074 + }, + { + "epoch": 3.3735505245720594, + "grad_norm": 0.9453547596931458, + "learning_rate": 2.0580602184037206e-06, + "loss": 0.5652, + "step": 4075 + }, + { + "epoch": 3.3743787962451686, + "grad_norm": 0.9260196089744568, + "learning_rate": 2.0569778906737186e-06, + "loss": 0.5765, + "step": 4076 + }, + { + "epoch": 3.3752070679182773, + "grad_norm": 0.9467774629592896, + "learning_rate": 2.055895648664365e-06, + "loss": 0.5637, + "step": 4077 + }, + { + "epoch": 3.376035339591386, + "grad_norm": 0.9502113461494446, + "learning_rate": 2.0548134925850626e-06, + "loss": 0.5725, + "step": 4078 + }, + { + "epoch": 3.3768636112644947, + "grad_norm": 0.9475006461143494, + "learning_rate": 2.0537314226452004e-06, + "loss": 0.5936, + "step": 4079 + }, + { + "epoch": 3.3776918829376035, + "grad_norm": 0.9401916861534119, + "learning_rate": 2.0526494390541478e-06, + "loss": 0.5797, + "step": 4080 + }, + { + "epoch": 3.378520154610712, + "grad_norm": 0.9407974481582642, + "learning_rate": 2.0515675420212577e-06, + "loss": 0.5653, + "step": 4081 + }, + { + "epoch": 3.379348426283821, + "grad_norm": 0.9521286487579346, + "learning_rate": 2.0504857317558686e-06, + "loss": 0.5794, + "step": 4082 + }, + { + "epoch": 3.38017669795693, + "grad_norm": 0.9705640077590942, + "learning_rate": 2.0494040084673004e-06, + "loss": 0.5821, + "step": 4083 + }, + { + "epoch": 3.381004969630039, + "grad_norm": 0.9214118719100952, + "learning_rate": 2.0483223723648567e-06, + "loss": 0.5757, + "step": 4084 + }, + { + "epoch": 3.3818332413031476, + "grad_norm": 0.9589064717292786, + "learning_rate": 2.0472408236578228e-06, + "loss": 0.5551, + "step": 4085 + }, + { + "epoch": 3.3826615129762563, + "grad_norm": 0.9503619074821472, + "learning_rate": 2.0461593625554683e-06, + "loss": 0.5805, + "step": 4086 + }, + { + "epoch": 3.383489784649365, + "grad_norm": 0.9328595995903015, + "learning_rate": 2.045077989267047e-06, + "loss": 0.5736, + "step": 4087 + }, + { + "epoch": 3.3843180563224737, + "grad_norm": 0.9475532174110413, + "learning_rate": 2.043996704001794e-06, + "loss": 0.5652, + "step": 4088 + }, + { + "epoch": 3.3851463279955825, + "grad_norm": 0.9307344555854797, + "learning_rate": 2.042915506968928e-06, + "loss": 0.5801, + "step": 4089 + }, + { + "epoch": 3.385974599668691, + "grad_norm": 0.9430568218231201, + "learning_rate": 2.04183439837765e-06, + "loss": 0.5864, + "step": 4090 + }, + { + "epoch": 3.3868028713418, + "grad_norm": 0.9192523956298828, + "learning_rate": 2.0407533784371457e-06, + "loss": 0.5486, + "step": 4091 + }, + { + "epoch": 3.387631143014909, + "grad_norm": 0.9655841588973999, + "learning_rate": 2.039672447356581e-06, + "loss": 0.5602, + "step": 4092 + }, + { + "epoch": 3.388459414688018, + "grad_norm": 0.9553002119064331, + "learning_rate": 2.0385916053451073e-06, + "loss": 0.568, + "step": 4093 + }, + { + "epoch": 3.3892876863611265, + "grad_norm": 0.9539245963096619, + "learning_rate": 2.0375108526118563e-06, + "loss": 0.5898, + "step": 4094 + }, + { + "epoch": 3.3901159580342353, + "grad_norm": 0.9733588695526123, + "learning_rate": 2.036430189365945e-06, + "loss": 0.5817, + "step": 4095 + }, + { + "epoch": 3.390944229707344, + "grad_norm": 0.9487296938896179, + "learning_rate": 2.03534961581647e-06, + "loss": 0.5841, + "step": 4096 + }, + { + "epoch": 3.3917725013804527, + "grad_norm": 0.9788404107093811, + "learning_rate": 2.0342691321725127e-06, + "loss": 0.5951, + "step": 4097 + }, + { + "epoch": 3.3926007730535614, + "grad_norm": 0.9286518692970276, + "learning_rate": 2.033188738643137e-06, + "loss": 0.5555, + "step": 4098 + }, + { + "epoch": 3.39342904472667, + "grad_norm": 0.9736733436584473, + "learning_rate": 2.032108435437389e-06, + "loss": 0.5773, + "step": 4099 + }, + { + "epoch": 3.394257316399779, + "grad_norm": 0.9288291335105896, + "learning_rate": 2.0310282227642963e-06, + "loss": 0.5562, + "step": 4100 + }, + { + "epoch": 3.395085588072888, + "grad_norm": 0.9414420127868652, + "learning_rate": 2.0299481008328716e-06, + "loss": 0.5923, + "step": 4101 + }, + { + "epoch": 3.395913859745997, + "grad_norm": 0.9327258467674255, + "learning_rate": 2.028868069852108e-06, + "loss": 0.5636, + "step": 4102 + }, + { + "epoch": 3.3967421314191055, + "grad_norm": 0.9412753582000732, + "learning_rate": 2.02778813003098e-06, + "loss": 0.5896, + "step": 4103 + }, + { + "epoch": 3.3975704030922143, + "grad_norm": 0.973572313785553, + "learning_rate": 2.026708281578447e-06, + "loss": 0.5581, + "step": 4104 + }, + { + "epoch": 3.398398674765323, + "grad_norm": 1.1189706325531006, + "learning_rate": 2.0256285247034505e-06, + "loss": 0.5541, + "step": 4105 + }, + { + "epoch": 3.3992269464384317, + "grad_norm": 0.9494182467460632, + "learning_rate": 2.0245488596149113e-06, + "loss": 0.5686, + "step": 4106 + }, + { + "epoch": 3.4000552181115404, + "grad_norm": 0.9740265011787415, + "learning_rate": 2.0234692865217375e-06, + "loss": 0.5744, + "step": 4107 + }, + { + "epoch": 3.4008834897846496, + "grad_norm": 0.9565280079841614, + "learning_rate": 2.0223898056328133e-06, + "loss": 0.589, + "step": 4108 + }, + { + "epoch": 3.4017117614577583, + "grad_norm": 0.9839944839477539, + "learning_rate": 2.0213104171570094e-06, + "loss": 0.5704, + "step": 4109 + }, + { + "epoch": 3.402540033130867, + "grad_norm": 0.9117640256881714, + "learning_rate": 2.020231121303178e-06, + "loss": 0.5591, + "step": 4110 + }, + { + "epoch": 3.403368304803976, + "grad_norm": 0.9455560445785522, + "learning_rate": 2.019151918280151e-06, + "loss": 0.592, + "step": 4111 + }, + { + "epoch": 3.4041965764770845, + "grad_norm": 0.9271209836006165, + "learning_rate": 2.0180728082967464e-06, + "loss": 0.5832, + "step": 4112 + }, + { + "epoch": 3.4050248481501932, + "grad_norm": 0.9203625321388245, + "learning_rate": 2.016993791561761e-06, + "loss": 0.576, + "step": 4113 + }, + { + "epoch": 3.405853119823302, + "grad_norm": 0.9395371675491333, + "learning_rate": 2.015914868283974e-06, + "loss": 0.5823, + "step": 4114 + }, + { + "epoch": 3.4066813914964107, + "grad_norm": 0.9351215958595276, + "learning_rate": 2.014836038672148e-06, + "loss": 0.5903, + "step": 4115 + }, + { + "epoch": 3.4075096631695194, + "grad_norm": 0.9305378198623657, + "learning_rate": 2.0137573029350253e-06, + "loss": 0.5607, + "step": 4116 + }, + { + "epoch": 3.4083379348426286, + "grad_norm": 0.9378963708877563, + "learning_rate": 2.0126786612813316e-06, + "loss": 0.5648, + "step": 4117 + }, + { + "epoch": 3.4091662065157373, + "grad_norm": 1.2515968084335327, + "learning_rate": 2.011600113919776e-06, + "loss": 0.5915, + "step": 4118 + }, + { + "epoch": 3.409994478188846, + "grad_norm": 0.9324381947517395, + "learning_rate": 2.010521661059044e-06, + "loss": 0.5573, + "step": 4119 + }, + { + "epoch": 3.4108227498619548, + "grad_norm": 0.9508990049362183, + "learning_rate": 2.009443302907808e-06, + "loss": 0.5612, + "step": 4120 + }, + { + "epoch": 3.4116510215350635, + "grad_norm": 0.9441195130348206, + "learning_rate": 2.00836503967472e-06, + "loss": 0.5561, + "step": 4121 + }, + { + "epoch": 3.4124792932081722, + "grad_norm": 0.940579354763031, + "learning_rate": 2.007286871568413e-06, + "loss": 0.5787, + "step": 4122 + }, + { + "epoch": 3.413307564881281, + "grad_norm": 0.9296087026596069, + "learning_rate": 2.0062087987975043e-06, + "loss": 0.5638, + "step": 4123 + }, + { + "epoch": 3.4141358365543897, + "grad_norm": 0.9607263803482056, + "learning_rate": 2.00513082157059e-06, + "loss": 0.5903, + "step": 4124 + }, + { + "epoch": 3.4149641082274984, + "grad_norm": 0.9708670377731323, + "learning_rate": 2.004052940096248e-06, + "loss": 0.567, + "step": 4125 + }, + { + "epoch": 3.4157923799006076, + "grad_norm": 0.9483271241188049, + "learning_rate": 2.002975154583039e-06, + "loss": 0.5687, + "step": 4126 + }, + { + "epoch": 3.4166206515737163, + "grad_norm": 0.9517681002616882, + "learning_rate": 2.001897465239505e-06, + "loss": 0.5656, + "step": 4127 + }, + { + "epoch": 3.417448923246825, + "grad_norm": 0.9440080523490906, + "learning_rate": 2.0008198722741677e-06, + "loss": 0.5751, + "step": 4128 + }, + { + "epoch": 3.4182771949199338, + "grad_norm": 0.9267042279243469, + "learning_rate": 1.999742375895533e-06, + "loss": 0.5708, + "step": 4129 + }, + { + "epoch": 3.4191054665930425, + "grad_norm": 0.9693537950515747, + "learning_rate": 1.998664976312084e-06, + "loss": 0.5859, + "step": 4130 + }, + { + "epoch": 3.419933738266151, + "grad_norm": 0.9269915223121643, + "learning_rate": 1.997587673732289e-06, + "loss": 0.5804, + "step": 4131 + }, + { + "epoch": 3.42076200993926, + "grad_norm": 0.9441049695014954, + "learning_rate": 1.9965104683645957e-06, + "loss": 0.5547, + "step": 4132 + }, + { + "epoch": 3.4215902816123687, + "grad_norm": 0.9141139984130859, + "learning_rate": 1.9954333604174325e-06, + "loss": 0.5852, + "step": 4133 + }, + { + "epoch": 3.4224185532854774, + "grad_norm": 0.9308720827102661, + "learning_rate": 1.9943563500992115e-06, + "loss": 0.5446, + "step": 4134 + }, + { + "epoch": 3.4232468249585866, + "grad_norm": 0.9180085062980652, + "learning_rate": 1.9932794376183233e-06, + "loss": 0.5837, + "step": 4135 + }, + { + "epoch": 3.4240750966316953, + "grad_norm": 0.9463955163955688, + "learning_rate": 1.99220262318314e-06, + "loss": 0.5622, + "step": 4136 + }, + { + "epoch": 3.424903368304804, + "grad_norm": 0.9471439719200134, + "learning_rate": 1.991125907002015e-06, + "loss": 0.5682, + "step": 4137 + }, + { + "epoch": 3.4257316399779127, + "grad_norm": 0.9394784569740295, + "learning_rate": 1.990049289283284e-06, + "loss": 0.5702, + "step": 4138 + }, + { + "epoch": 3.4265599116510215, + "grad_norm": 1.005246639251709, + "learning_rate": 1.9889727702352618e-06, + "loss": 0.5772, + "step": 4139 + }, + { + "epoch": 3.42738818332413, + "grad_norm": 0.9534215927124023, + "learning_rate": 1.9878963500662453e-06, + "loss": 0.5626, + "step": 4140 + }, + { + "epoch": 3.428216454997239, + "grad_norm": 1.010836124420166, + "learning_rate": 1.98682002898451e-06, + "loss": 0.5748, + "step": 4141 + }, + { + "epoch": 3.429044726670348, + "grad_norm": 0.9219326376914978, + "learning_rate": 1.9857438071983163e-06, + "loss": 0.5893, + "step": 4142 + }, + { + "epoch": 3.429872998343457, + "grad_norm": 0.9372469186782837, + "learning_rate": 1.9846676849159017e-06, + "loss": 0.5555, + "step": 4143 + }, + { + "epoch": 3.4307012700165656, + "grad_norm": 0.9133549928665161, + "learning_rate": 1.9835916623454865e-06, + "loss": 0.5536, + "step": 4144 + }, + { + "epoch": 3.4315295416896743, + "grad_norm": 0.9400215148925781, + "learning_rate": 1.98251573969527e-06, + "loss": 0.5489, + "step": 4145 + }, + { + "epoch": 3.432357813362783, + "grad_norm": 0.9408988356590271, + "learning_rate": 1.9814399171734347e-06, + "loss": 0.5533, + "step": 4146 + }, + { + "epoch": 3.4331860850358917, + "grad_norm": 0.9352306723594666, + "learning_rate": 1.9803641949881415e-06, + "loss": 0.5645, + "step": 4147 + }, + { + "epoch": 3.4340143567090005, + "grad_norm": 0.9516900181770325, + "learning_rate": 1.979288573347532e-06, + "loss": 0.5764, + "step": 4148 + }, + { + "epoch": 3.434842628382109, + "grad_norm": 0.945179283618927, + "learning_rate": 1.97821305245973e-06, + "loss": 0.5792, + "step": 4149 + }, + { + "epoch": 3.435670900055218, + "grad_norm": 0.9419721961021423, + "learning_rate": 1.9771376325328377e-06, + "loss": 0.5598, + "step": 4150 + }, + { + "epoch": 3.436499171728327, + "grad_norm": 0.9680830836296082, + "learning_rate": 1.9760623137749394e-06, + "loss": 0.5675, + "step": 4151 + }, + { + "epoch": 3.437327443401436, + "grad_norm": 0.9370323419570923, + "learning_rate": 1.974987096394101e-06, + "loss": 0.5657, + "step": 4152 + }, + { + "epoch": 3.4381557150745445, + "grad_norm": 0.9793617725372314, + "learning_rate": 1.973911980598364e-06, + "loss": 0.5692, + "step": 4153 + }, + { + "epoch": 3.4389839867476533, + "grad_norm": 0.9549874067306519, + "learning_rate": 1.9728369665957547e-06, + "loss": 0.5798, + "step": 4154 + }, + { + "epoch": 3.439812258420762, + "grad_norm": 0.9228386878967285, + "learning_rate": 1.9717620545942784e-06, + "loss": 0.5759, + "step": 4155 + }, + { + "epoch": 3.4406405300938707, + "grad_norm": 0.9429793357849121, + "learning_rate": 1.97068724480192e-06, + "loss": 0.5618, + "step": 4156 + }, + { + "epoch": 3.4414688017669794, + "grad_norm": 0.9321681261062622, + "learning_rate": 1.969612537426646e-06, + "loss": 0.5602, + "step": 4157 + }, + { + "epoch": 3.442297073440088, + "grad_norm": 0.9411936402320862, + "learning_rate": 1.9685379326764014e-06, + "loss": 0.5787, + "step": 4158 + }, + { + "epoch": 3.443125345113197, + "grad_norm": 0.9223849177360535, + "learning_rate": 1.967463430759112e-06, + "loss": 0.5565, + "step": 4159 + }, + { + "epoch": 3.443953616786306, + "grad_norm": 0.9482093453407288, + "learning_rate": 1.966389031882686e-06, + "loss": 0.5729, + "step": 4160 + }, + { + "epoch": 3.444781888459415, + "grad_norm": 0.9146328568458557, + "learning_rate": 1.9653147362550075e-06, + "loss": 0.5661, + "step": 4161 + }, + { + "epoch": 3.4456101601325235, + "grad_norm": 0.9412453770637512, + "learning_rate": 1.9642405440839425e-06, + "loss": 0.5622, + "step": 4162 + }, + { + "epoch": 3.4464384318056323, + "grad_norm": 0.9408630132675171, + "learning_rate": 1.96316645557734e-06, + "loss": 0.5542, + "step": 4163 + }, + { + "epoch": 3.447266703478741, + "grad_norm": 0.9337356090545654, + "learning_rate": 1.962092470943022e-06, + "loss": 0.5634, + "step": 4164 + }, + { + "epoch": 3.4480949751518497, + "grad_norm": 0.9458538293838501, + "learning_rate": 1.9610185903887973e-06, + "loss": 0.5796, + "step": 4165 + }, + { + "epoch": 3.4489232468249584, + "grad_norm": 0.9226371049880981, + "learning_rate": 1.9599448141224507e-06, + "loss": 0.5596, + "step": 4166 + }, + { + "epoch": 3.4497515184980676, + "grad_norm": 0.947045087814331, + "learning_rate": 1.958871142351748e-06, + "loss": 0.554, + "step": 4167 + }, + { + "epoch": 3.4505797901711763, + "grad_norm": 0.9368892908096313, + "learning_rate": 1.9577975752844357e-06, + "loss": 0.5708, + "step": 4168 + }, + { + "epoch": 3.451408061844285, + "grad_norm": 0.9360406398773193, + "learning_rate": 1.9567241131282378e-06, + "loss": 0.5664, + "step": 4169 + }, + { + "epoch": 3.452236333517394, + "grad_norm": 0.9660393595695496, + "learning_rate": 1.9556507560908594e-06, + "loss": 0.5597, + "step": 4170 + }, + { + "epoch": 3.4530646051905025, + "grad_norm": 0.9472611546516418, + "learning_rate": 1.9545775043799858e-06, + "loss": 0.5704, + "step": 4171 + }, + { + "epoch": 3.4538928768636112, + "grad_norm": 0.941719651222229, + "learning_rate": 1.953504358203281e-06, + "loss": 0.5713, + "step": 4172 + }, + { + "epoch": 3.45472114853672, + "grad_norm": 0.9945985674858093, + "learning_rate": 1.9524313177683873e-06, + "loss": 0.563, + "step": 4173 + }, + { + "epoch": 3.4555494202098287, + "grad_norm": 0.9355506300926208, + "learning_rate": 1.9513583832829314e-06, + "loss": 0.5677, + "step": 4174 + }, + { + "epoch": 3.4563776918829374, + "grad_norm": 0.9413464665412903, + "learning_rate": 1.950285554954512e-06, + "loss": 0.5416, + "step": 4175 + }, + { + "epoch": 3.4572059635560466, + "grad_norm": 0.9668782353401184, + "learning_rate": 1.949212832990714e-06, + "loss": 0.5832, + "step": 4176 + }, + { + "epoch": 3.4580342352291553, + "grad_norm": 0.9598191976547241, + "learning_rate": 1.948140217599099e-06, + "loss": 0.597, + "step": 4177 + }, + { + "epoch": 3.458862506902264, + "grad_norm": 0.9517234563827515, + "learning_rate": 1.9470677089872064e-06, + "loss": 0.5812, + "step": 4178 + }, + { + "epoch": 3.4596907785753728, + "grad_norm": 0.9452085494995117, + "learning_rate": 1.9459953073625586e-06, + "loss": 0.5593, + "step": 4179 + }, + { + "epoch": 3.4605190502484815, + "grad_norm": 0.9393478035926819, + "learning_rate": 1.9449230129326543e-06, + "loss": 0.5828, + "step": 4180 + }, + { + "epoch": 3.4613473219215902, + "grad_norm": 0.9487680792808533, + "learning_rate": 1.9438508259049716e-06, + "loss": 0.5884, + "step": 4181 + }, + { + "epoch": 3.462175593594699, + "grad_norm": 0.9236283898353577, + "learning_rate": 1.942778746486971e-06, + "loss": 0.5534, + "step": 4182 + }, + { + "epoch": 3.4630038652678077, + "grad_norm": 0.9297188520431519, + "learning_rate": 1.941706774886088e-06, + "loss": 0.5702, + "step": 4183 + }, + { + "epoch": 3.4638321369409164, + "grad_norm": 0.9298705458641052, + "learning_rate": 1.94063491130974e-06, + "loss": 0.5645, + "step": 4184 + }, + { + "epoch": 3.4646604086140256, + "grad_norm": 0.93392014503479, + "learning_rate": 1.9395631559653223e-06, + "loss": 0.5805, + "step": 4185 + }, + { + "epoch": 3.4654886802871343, + "grad_norm": 0.9221041798591614, + "learning_rate": 1.93849150906021e-06, + "loss": 0.5746, + "step": 4186 + }, + { + "epoch": 3.466316951960243, + "grad_norm": 0.9339867830276489, + "learning_rate": 1.937419970801756e-06, + "loss": 0.5557, + "step": 4187 + }, + { + "epoch": 3.4671452236333518, + "grad_norm": 0.9240173697471619, + "learning_rate": 1.9363485413972934e-06, + "loss": 0.5616, + "step": 4188 + }, + { + "epoch": 3.4679734953064605, + "grad_norm": 0.9132168889045715, + "learning_rate": 1.9352772210541332e-06, + "loss": 0.5665, + "step": 4189 + }, + { + "epoch": 3.468801766979569, + "grad_norm": 0.9252684116363525, + "learning_rate": 1.934206009979567e-06, + "loss": 0.5534, + "step": 4190 + }, + { + "epoch": 3.469630038652678, + "grad_norm": 0.940039873123169, + "learning_rate": 1.933134908380864e-06, + "loss": 0.5686, + "step": 4191 + }, + { + "epoch": 3.4704583103257867, + "grad_norm": 0.9629718065261841, + "learning_rate": 1.9320639164652706e-06, + "loss": 0.6051, + "step": 4192 + }, + { + "epoch": 3.4712865819988954, + "grad_norm": 0.9440659880638123, + "learning_rate": 1.9309930344400164e-06, + "loss": 0.5612, + "step": 4193 + }, + { + "epoch": 3.4721148536720046, + "grad_norm": 0.9437550902366638, + "learning_rate": 1.9299222625123054e-06, + "loss": 0.5848, + "step": 4194 + }, + { + "epoch": 3.4729431253451133, + "grad_norm": 1.0566316843032837, + "learning_rate": 1.928851600889322e-06, + "loss": 0.5655, + "step": 4195 + }, + { + "epoch": 3.473771397018222, + "grad_norm": 0.9688880443572998, + "learning_rate": 1.92778104977823e-06, + "loss": 0.5792, + "step": 4196 + }, + { + "epoch": 3.4745996686913307, + "grad_norm": 0.9409403204917908, + "learning_rate": 1.9267106093861716e-06, + "loss": 0.5659, + "step": 4197 + }, + { + "epoch": 3.4754279403644395, + "grad_norm": 0.948984682559967, + "learning_rate": 1.9256402799202657e-06, + "loss": 0.5565, + "step": 4198 + }, + { + "epoch": 3.476256212037548, + "grad_norm": 0.9317809343338013, + "learning_rate": 1.924570061587611e-06, + "loss": 0.5739, + "step": 4199 + }, + { + "epoch": 3.477084483710657, + "grad_norm": 0.9296276569366455, + "learning_rate": 1.923499954595285e-06, + "loss": 0.5599, + "step": 4200 + }, + { + "epoch": 3.477912755383766, + "grad_norm": 0.9425923228263855, + "learning_rate": 1.922429959150344e-06, + "loss": 0.5686, + "step": 4201 + }, + { + "epoch": 3.478741027056875, + "grad_norm": 0.935650110244751, + "learning_rate": 1.9213600754598225e-06, + "loss": 0.5637, + "step": 4202 + }, + { + "epoch": 3.4795692987299836, + "grad_norm": 0.9613479375839233, + "learning_rate": 1.9202903037307312e-06, + "loss": 0.5766, + "step": 4203 + }, + { + "epoch": 3.4803975704030923, + "grad_norm": 0.9408090710639954, + "learning_rate": 1.9192206441700636e-06, + "loss": 0.5751, + "step": 4204 + }, + { + "epoch": 3.481225842076201, + "grad_norm": 0.9520177245140076, + "learning_rate": 1.9181510969847865e-06, + "loss": 0.5736, + "step": 4205 + }, + { + "epoch": 3.4820541137493097, + "grad_norm": 0.963941752910614, + "learning_rate": 1.9170816623818484e-06, + "loss": 0.5662, + "step": 4206 + }, + { + "epoch": 3.4828823854224185, + "grad_norm": 0.950707733631134, + "learning_rate": 1.9160123405681743e-06, + "loss": 0.576, + "step": 4207 + }, + { + "epoch": 3.483710657095527, + "grad_norm": 0.9403618574142456, + "learning_rate": 1.9149431317506693e-06, + "loss": 0.6119, + "step": 4208 + }, + { + "epoch": 3.484538928768636, + "grad_norm": 0.9552726745605469, + "learning_rate": 1.913874036136214e-06, + "loss": 0.5616, + "step": 4209 + }, + { + "epoch": 3.485367200441745, + "grad_norm": 0.9421485662460327, + "learning_rate": 1.912805053931669e-06, + "loss": 0.5708, + "step": 4210 + }, + { + "epoch": 3.486195472114854, + "grad_norm": 0.9480198621749878, + "learning_rate": 1.911736185343872e-06, + "loss": 0.5814, + "step": 4211 + }, + { + "epoch": 3.4870237437879625, + "grad_norm": 0.911723792552948, + "learning_rate": 1.91066743057964e-06, + "loss": 0.57, + "step": 4212 + }, + { + "epoch": 3.4878520154610713, + "grad_norm": 0.9331412315368652, + "learning_rate": 1.909598789845766e-06, + "loss": 0.5624, + "step": 4213 + }, + { + "epoch": 3.48868028713418, + "grad_norm": 1.0301762819290161, + "learning_rate": 1.9085302633490227e-06, + "loss": 0.5433, + "step": 4214 + }, + { + "epoch": 3.4895085588072887, + "grad_norm": 0.9518493413925171, + "learning_rate": 1.9074618512961597e-06, + "loss": 0.5702, + "step": 4215 + }, + { + "epoch": 3.4903368304803974, + "grad_norm": 0.9399821162223816, + "learning_rate": 1.9063935538939049e-06, + "loss": 0.5584, + "step": 4216 + }, + { + "epoch": 3.491165102153506, + "grad_norm": 0.9549731016159058, + "learning_rate": 1.9053253713489643e-06, + "loss": 0.5664, + "step": 4217 + }, + { + "epoch": 3.491993373826615, + "grad_norm": 0.9505523443222046, + "learning_rate": 1.9042573038680198e-06, + "loss": 0.5763, + "step": 4218 + }, + { + "epoch": 3.492821645499724, + "grad_norm": 0.9283603429794312, + "learning_rate": 1.9031893516577343e-06, + "loss": 0.5902, + "step": 4219 + }, + { + "epoch": 3.493649917172833, + "grad_norm": 0.9389188885688782, + "learning_rate": 1.9021215149247468e-06, + "loss": 0.5535, + "step": 4220 + }, + { + "epoch": 3.4944781888459415, + "grad_norm": 0.9431526064872742, + "learning_rate": 1.9010537938756719e-06, + "loss": 0.5698, + "step": 4221 + }, + { + "epoch": 3.4953064605190503, + "grad_norm": 0.9633996486663818, + "learning_rate": 1.8999861887171045e-06, + "loss": 0.573, + "step": 4222 + }, + { + "epoch": 3.496134732192159, + "grad_norm": 0.9711135625839233, + "learning_rate": 1.8989186996556161e-06, + "loss": 0.5603, + "step": 4223 + }, + { + "epoch": 3.4969630038652677, + "grad_norm": 0.9533023238182068, + "learning_rate": 1.8978513268977564e-06, + "loss": 0.5543, + "step": 4224 + }, + { + "epoch": 3.4977912755383764, + "grad_norm": 0.9520569443702698, + "learning_rate": 1.8967840706500519e-06, + "loss": 0.5859, + "step": 4225 + }, + { + "epoch": 3.4986195472114856, + "grad_norm": 0.952743411064148, + "learning_rate": 1.895716931119006e-06, + "loss": 0.5687, + "step": 4226 + }, + { + "epoch": 3.4994478188845943, + "grad_norm": 0.9404147267341614, + "learning_rate": 1.8946499085111014e-06, + "loss": 0.5669, + "step": 4227 + }, + { + "epoch": 3.500276090557703, + "grad_norm": 0.962302029132843, + "learning_rate": 1.8935830030327967e-06, + "loss": 0.5721, + "step": 4228 + }, + { + "epoch": 3.501104362230812, + "grad_norm": 0.9707161784172058, + "learning_rate": 1.8925162148905275e-06, + "loss": 0.5858, + "step": 4229 + }, + { + "epoch": 3.5019326339039205, + "grad_norm": 0.9720497727394104, + "learning_rate": 1.891449544290708e-06, + "loss": 0.5892, + "step": 4230 + }, + { + "epoch": 3.5027609055770292, + "grad_norm": 0.9394986629486084, + "learning_rate": 1.8903829914397295e-06, + "loss": 0.547, + "step": 4231 + }, + { + "epoch": 3.503589177250138, + "grad_norm": 0.9768742322921753, + "learning_rate": 1.8893165565439588e-06, + "loss": 0.5647, + "step": 4232 + }, + { + "epoch": 3.5044174489232467, + "grad_norm": 0.9559404253959656, + "learning_rate": 1.888250239809742e-06, + "loss": 0.556, + "step": 4233 + }, + { + "epoch": 3.5052457205963554, + "grad_norm": 0.9628231525421143, + "learning_rate": 1.8871840414434e-06, + "loss": 0.5633, + "step": 4234 + }, + { + "epoch": 3.5060739922694646, + "grad_norm": 0.9246355891227722, + "learning_rate": 1.8861179616512346e-06, + "loss": 0.5678, + "step": 4235 + }, + { + "epoch": 3.5069022639425733, + "grad_norm": 0.9590763449668884, + "learning_rate": 1.8850520006395206e-06, + "loss": 0.5612, + "step": 4236 + }, + { + "epoch": 3.507730535615682, + "grad_norm": 0.9386741518974304, + "learning_rate": 1.8839861586145117e-06, + "loss": 0.5737, + "step": 4237 + }, + { + "epoch": 3.5085588072887908, + "grad_norm": 0.9735057353973389, + "learning_rate": 1.8829204357824394e-06, + "loss": 0.5725, + "step": 4238 + }, + { + "epoch": 3.5093870789618995, + "grad_norm": 0.9464735984802246, + "learning_rate": 1.8818548323495101e-06, + "loss": 0.561, + "step": 4239 + }, + { + "epoch": 3.5102153506350082, + "grad_norm": 0.9822002053260803, + "learning_rate": 1.880789348521908e-06, + "loss": 0.562, + "step": 4240 + }, + { + "epoch": 3.511043622308117, + "grad_norm": 0.9448258876800537, + "learning_rate": 1.879723984505795e-06, + "loss": 0.5526, + "step": 4241 + }, + { + "epoch": 3.5118718939812257, + "grad_norm": 0.9565535187721252, + "learning_rate": 1.87865874050731e-06, + "loss": 0.5631, + "step": 4242 + }, + { + "epoch": 3.5127001656543344, + "grad_norm": 0.9824585318565369, + "learning_rate": 1.877593616732566e-06, + "loss": 0.5924, + "step": 4243 + }, + { + "epoch": 3.5135284373274436, + "grad_norm": 0.9289377927780151, + "learning_rate": 1.8765286133876549e-06, + "loss": 0.5667, + "step": 4244 + }, + { + "epoch": 3.5143567090005523, + "grad_norm": 0.9740702509880066, + "learning_rate": 1.8754637306786447e-06, + "loss": 0.567, + "step": 4245 + }, + { + "epoch": 3.515184980673661, + "grad_norm": 0.9304373264312744, + "learning_rate": 1.8743989688115813e-06, + "loss": 0.5973, + "step": 4246 + }, + { + "epoch": 3.5160132523467698, + "grad_norm": 0.9417708516120911, + "learning_rate": 1.8733343279924857e-06, + "loss": 0.5663, + "step": 4247 + }, + { + "epoch": 3.5168415240198785, + "grad_norm": 0.9550634622573853, + "learning_rate": 1.8722698084273555e-06, + "loss": 0.5829, + "step": 4248 + }, + { + "epoch": 3.517669795692987, + "grad_norm": 0.9428004622459412, + "learning_rate": 1.8712054103221661e-06, + "loss": 0.5595, + "step": 4249 + }, + { + "epoch": 3.518498067366096, + "grad_norm": 0.9728983640670776, + "learning_rate": 1.8701411338828685e-06, + "loss": 0.5611, + "step": 4250 + }, + { + "epoch": 3.519326339039205, + "grad_norm": 0.9784600138664246, + "learning_rate": 1.869076979315389e-06, + "loss": 0.5699, + "step": 4251 + }, + { + "epoch": 3.5201546107123134, + "grad_norm": 0.9751117825508118, + "learning_rate": 1.8680129468256336e-06, + "loss": 0.5622, + "step": 4252 + }, + { + "epoch": 3.5209828823854226, + "grad_norm": 0.9501684904098511, + "learning_rate": 1.866949036619482e-06, + "loss": 0.5594, + "step": 4253 + }, + { + "epoch": 3.5218111540585313, + "grad_norm": 0.9380241632461548, + "learning_rate": 1.86588524890279e-06, + "loss": 0.5759, + "step": 4254 + }, + { + "epoch": 3.52263942573164, + "grad_norm": 0.9474608898162842, + "learning_rate": 1.8648215838813915e-06, + "loss": 0.592, + "step": 4255 + }, + { + "epoch": 3.5234676974047487, + "grad_norm": 0.9680253863334656, + "learning_rate": 1.8637580417610949e-06, + "loss": 0.5369, + "step": 4256 + }, + { + "epoch": 3.5242959690778575, + "grad_norm": 0.9809057712554932, + "learning_rate": 1.8626946227476867e-06, + "loss": 0.5678, + "step": 4257 + }, + { + "epoch": 3.525124240750966, + "grad_norm": 0.9424263834953308, + "learning_rate": 1.8616313270469277e-06, + "loss": 0.5735, + "step": 4258 + }, + { + "epoch": 3.525952512424075, + "grad_norm": 0.9373759627342224, + "learning_rate": 1.860568154864556e-06, + "loss": 0.5572, + "step": 4259 + }, + { + "epoch": 3.526780784097184, + "grad_norm": 0.9195491075515747, + "learning_rate": 1.8595051064062864e-06, + "loss": 0.5552, + "step": 4260 + }, + { + "epoch": 3.5276090557702924, + "grad_norm": 0.9787479043006897, + "learning_rate": 1.8584421818778076e-06, + "loss": 0.5486, + "step": 4261 + }, + { + "epoch": 3.5284373274434015, + "grad_norm": 0.9167375564575195, + "learning_rate": 1.8573793814847856e-06, + "loss": 0.561, + "step": 4262 + }, + { + "epoch": 3.5292655991165103, + "grad_norm": 0.9589730501174927, + "learning_rate": 1.8563167054328635e-06, + "loss": 0.5792, + "step": 4263 + }, + { + "epoch": 3.530093870789619, + "grad_norm": 0.9304484724998474, + "learning_rate": 1.8552541539276587e-06, + "loss": 0.5628, + "step": 4264 + }, + { + "epoch": 3.5309221424627277, + "grad_norm": 0.9468231201171875, + "learning_rate": 1.8541917271747653e-06, + "loss": 0.5705, + "step": 4265 + }, + { + "epoch": 3.5317504141358365, + "grad_norm": 0.9367941617965698, + "learning_rate": 1.8531294253797522e-06, + "loss": 0.5769, + "step": 4266 + }, + { + "epoch": 3.532578685808945, + "grad_norm": 0.9852622151374817, + "learning_rate": 1.8520672487481645e-06, + "loss": 0.572, + "step": 4267 + }, + { + "epoch": 3.533406957482054, + "grad_norm": 0.9441319108009338, + "learning_rate": 1.8510051974855244e-06, + "loss": 0.5587, + "step": 4268 + }, + { + "epoch": 3.534235229155163, + "grad_norm": 0.9801215529441833, + "learning_rate": 1.8499432717973293e-06, + "loss": 0.6019, + "step": 4269 + }, + { + "epoch": 3.535063500828272, + "grad_norm": 0.9506904482841492, + "learning_rate": 1.8488814718890505e-06, + "loss": 0.569, + "step": 4270 + }, + { + "epoch": 3.5358917725013805, + "grad_norm": 0.9413098096847534, + "learning_rate": 1.8478197979661377e-06, + "loss": 0.5682, + "step": 4271 + }, + { + "epoch": 3.5367200441744893, + "grad_norm": 0.9410218596458435, + "learning_rate": 1.8467582502340149e-06, + "loss": 0.5843, + "step": 4272 + }, + { + "epoch": 3.537548315847598, + "grad_norm": 0.9326966404914856, + "learning_rate": 1.845696828898081e-06, + "loss": 0.5853, + "step": 4273 + }, + { + "epoch": 3.5383765875207067, + "grad_norm": 0.9390862584114075, + "learning_rate": 1.8446355341637107e-06, + "loss": 0.5767, + "step": 4274 + }, + { + "epoch": 3.5392048591938154, + "grad_norm": 0.9381791949272156, + "learning_rate": 1.843574366236256e-06, + "loss": 0.5534, + "step": 4275 + }, + { + "epoch": 3.5400331308669246, + "grad_norm": 1.0620906352996826, + "learning_rate": 1.8425133253210425e-06, + "loss": 0.5663, + "step": 4276 + }, + { + "epoch": 3.540861402540033, + "grad_norm": 0.945441484451294, + "learning_rate": 1.8414524116233715e-06, + "loss": 0.5496, + "step": 4277 + }, + { + "epoch": 3.541689674213142, + "grad_norm": 0.9375214576721191, + "learning_rate": 1.8403916253485193e-06, + "loss": 0.5556, + "step": 4278 + }, + { + "epoch": 3.542517945886251, + "grad_norm": 0.9695723056793213, + "learning_rate": 1.8393309667017388e-06, + "loss": 0.5457, + "step": 4279 + }, + { + "epoch": 3.5433462175593595, + "grad_norm": 0.9985876679420471, + "learning_rate": 1.8382704358882583e-06, + "loss": 0.5749, + "step": 4280 + }, + { + "epoch": 3.5441744892324683, + "grad_norm": 0.9784291386604309, + "learning_rate": 1.8372100331132797e-06, + "loss": 0.5671, + "step": 4281 + }, + { + "epoch": 3.545002760905577, + "grad_norm": 0.9699963331222534, + "learning_rate": 1.8361497585819804e-06, + "loss": 0.5742, + "step": 4282 + }, + { + "epoch": 3.5458310325786857, + "grad_norm": 0.951996386051178, + "learning_rate": 1.8350896124995155e-06, + "loss": 0.5729, + "step": 4283 + }, + { + "epoch": 3.5466593042517944, + "grad_norm": 0.9519326686859131, + "learning_rate": 1.8340295950710119e-06, + "loss": 0.5746, + "step": 4284 + }, + { + "epoch": 3.5474875759249036, + "grad_norm": 0.9227757453918457, + "learning_rate": 1.8329697065015734e-06, + "loss": 0.5441, + "step": 4285 + }, + { + "epoch": 3.548315847598012, + "grad_norm": 0.9924660325050354, + "learning_rate": 1.8319099469962787e-06, + "loss": 0.5695, + "step": 4286 + }, + { + "epoch": 3.549144119271121, + "grad_norm": 0.9582194089889526, + "learning_rate": 1.8308503167601827e-06, + "loss": 0.5757, + "step": 4287 + }, + { + "epoch": 3.54997239094423, + "grad_norm": 0.924054741859436, + "learning_rate": 1.8297908159983114e-06, + "loss": 0.5739, + "step": 4288 + }, + { + "epoch": 3.5508006626173385, + "grad_norm": 0.9424325227737427, + "learning_rate": 1.82873144491567e-06, + "loss": 0.6041, + "step": 4289 + }, + { + "epoch": 3.5516289342904472, + "grad_norm": 0.9287232160568237, + "learning_rate": 1.827672203717236e-06, + "loss": 0.5494, + "step": 4290 + }, + { + "epoch": 3.552457205963556, + "grad_norm": 0.94130539894104, + "learning_rate": 1.8266130926079637e-06, + "loss": 0.5642, + "step": 4291 + }, + { + "epoch": 3.5532854776366647, + "grad_norm": 0.933063805103302, + "learning_rate": 1.8255541117927805e-06, + "loss": 0.5759, + "step": 4292 + }, + { + "epoch": 3.5541137493097734, + "grad_norm": 0.9383630156517029, + "learning_rate": 1.8244952614765894e-06, + "loss": 0.5584, + "step": 4293 + }, + { + "epoch": 3.5549420209828826, + "grad_norm": 0.9373190999031067, + "learning_rate": 1.8234365418642685e-06, + "loss": 0.5597, + "step": 4294 + }, + { + "epoch": 3.5557702926559913, + "grad_norm": 0.9637420773506165, + "learning_rate": 1.82237795316067e-06, + "loss": 0.5736, + "step": 4295 + }, + { + "epoch": 3.5565985643291, + "grad_norm": 0.9421122670173645, + "learning_rate": 1.8213194955706204e-06, + "loss": 0.5424, + "step": 4296 + }, + { + "epoch": 3.5574268360022088, + "grad_norm": 0.9612517952919006, + "learning_rate": 1.8202611692989224e-06, + "loss": 0.5704, + "step": 4297 + }, + { + "epoch": 3.5582551076753175, + "grad_norm": 0.9415401220321655, + "learning_rate": 1.8192029745503515e-06, + "loss": 0.5774, + "step": 4298 + }, + { + "epoch": 3.5590833793484262, + "grad_norm": 0.9469128251075745, + "learning_rate": 1.8181449115296596e-06, + "loss": 0.5721, + "step": 4299 + }, + { + "epoch": 3.559911651021535, + "grad_norm": 0.9917946457862854, + "learning_rate": 1.8170869804415703e-06, + "loss": 0.5767, + "step": 4300 + }, + { + "epoch": 3.5607399226946437, + "grad_norm": 0.9557772874832153, + "learning_rate": 1.8160291814907835e-06, + "loss": 0.5677, + "step": 4301 + }, + { + "epoch": 3.5615681943677524, + "grad_norm": 0.9430398344993591, + "learning_rate": 1.8149715148819752e-06, + "loss": 0.5679, + "step": 4302 + }, + { + "epoch": 3.5623964660408616, + "grad_norm": 0.9426287412643433, + "learning_rate": 1.8139139808197926e-06, + "loss": 0.563, + "step": 4303 + }, + { + "epoch": 3.5632247377139703, + "grad_norm": 0.9437983632087708, + "learning_rate": 1.8128565795088588e-06, + "loss": 0.5703, + "step": 4304 + }, + { + "epoch": 3.564053009387079, + "grad_norm": 0.9386832118034363, + "learning_rate": 1.8117993111537717e-06, + "loss": 0.5842, + "step": 4305 + }, + { + "epoch": 3.5648812810601878, + "grad_norm": 0.954119086265564, + "learning_rate": 1.8107421759591031e-06, + "loss": 0.5515, + "step": 4306 + }, + { + "epoch": 3.5657095527332965, + "grad_norm": 0.945758581161499, + "learning_rate": 1.8096851741293974e-06, + "loss": 0.5455, + "step": 4307 + }, + { + "epoch": 3.566537824406405, + "grad_norm": 0.9291911721229553, + "learning_rate": 1.8086283058691759e-06, + "loss": 0.5843, + "step": 4308 + }, + { + "epoch": 3.567366096079514, + "grad_norm": 0.9620952606201172, + "learning_rate": 1.8075715713829323e-06, + "loss": 0.568, + "step": 4309 + }, + { + "epoch": 3.568194367752623, + "grad_norm": 0.9722322821617126, + "learning_rate": 1.8065149708751357e-06, + "loss": 0.5816, + "step": 4310 + }, + { + "epoch": 3.5690226394257314, + "grad_norm": 0.9354492425918579, + "learning_rate": 1.805458504550227e-06, + "loss": 0.5856, + "step": 4311 + }, + { + "epoch": 3.5698509110988406, + "grad_norm": 0.956453263759613, + "learning_rate": 1.8044021726126226e-06, + "loss": 0.5582, + "step": 4312 + }, + { + "epoch": 3.5706791827719493, + "grad_norm": 0.926895797252655, + "learning_rate": 1.8033459752667143e-06, + "loss": 0.5837, + "step": 4313 + }, + { + "epoch": 3.571507454445058, + "grad_norm": 0.9227427840232849, + "learning_rate": 1.8022899127168655e-06, + "loss": 0.5567, + "step": 4314 + }, + { + "epoch": 3.5723357261181667, + "grad_norm": 0.9473888278007507, + "learning_rate": 1.801233985167414e-06, + "loss": 0.5672, + "step": 4315 + }, + { + "epoch": 3.5731639977912755, + "grad_norm": 0.9397003054618835, + "learning_rate": 1.8001781928226734e-06, + "loss": 0.5663, + "step": 4316 + }, + { + "epoch": 3.573992269464384, + "grad_norm": 0.9553276300430298, + "learning_rate": 1.7991225358869286e-06, + "loss": 0.5716, + "step": 4317 + }, + { + "epoch": 3.574820541137493, + "grad_norm": 0.9149162769317627, + "learning_rate": 1.798067014564439e-06, + "loss": 0.5409, + "step": 4318 + }, + { + "epoch": 3.575648812810602, + "grad_norm": 1.0992645025253296, + "learning_rate": 1.7970116290594391e-06, + "loss": 0.5542, + "step": 4319 + }, + { + "epoch": 3.576477084483711, + "grad_norm": 0.9605867862701416, + "learning_rate": 1.7959563795761364e-06, + "loss": 0.5426, + "step": 4320 + }, + { + "epoch": 3.5773053561568195, + "grad_norm": 0.944625198841095, + "learning_rate": 1.7949012663187115e-06, + "loss": 0.5695, + "step": 4321 + }, + { + "epoch": 3.5781336278299283, + "grad_norm": 0.9884684085845947, + "learning_rate": 1.793846289491318e-06, + "loss": 0.5897, + "step": 4322 + }, + { + "epoch": 3.578961899503037, + "grad_norm": 0.9456211924552917, + "learning_rate": 1.7927914492980845e-06, + "loss": 0.5969, + "step": 4323 + }, + { + "epoch": 3.5797901711761457, + "grad_norm": 0.9710149168968201, + "learning_rate": 1.7917367459431135e-06, + "loss": 0.5797, + "step": 4324 + }, + { + "epoch": 3.5806184428492545, + "grad_norm": 0.9509037733078003, + "learning_rate": 1.7906821796304796e-06, + "loss": 0.5655, + "step": 4325 + }, + { + "epoch": 3.581446714522363, + "grad_norm": 0.9468769431114197, + "learning_rate": 1.7896277505642316e-06, + "loss": 0.5519, + "step": 4326 + }, + { + "epoch": 3.582274986195472, + "grad_norm": 0.9285255670547485, + "learning_rate": 1.7885734589483923e-06, + "loss": 0.5765, + "step": 4327 + }, + { + "epoch": 3.583103257868581, + "grad_norm": 0.9228923320770264, + "learning_rate": 1.7875193049869566e-06, + "loss": 0.5742, + "step": 4328 + }, + { + "epoch": 3.58393152954169, + "grad_norm": 0.9343339204788208, + "learning_rate": 1.7864652888838935e-06, + "loss": 0.5601, + "step": 4329 + }, + { + "epoch": 3.5847598012147985, + "grad_norm": 0.937486469745636, + "learning_rate": 1.7854114108431459e-06, + "loss": 0.5392, + "step": 4330 + }, + { + "epoch": 3.5855880728879073, + "grad_norm": 0.9428637027740479, + "learning_rate": 1.7843576710686288e-06, + "loss": 0.5502, + "step": 4331 + }, + { + "epoch": 3.586416344561016, + "grad_norm": 0.9170352220535278, + "learning_rate": 1.7833040697642312e-06, + "loss": 0.5707, + "step": 4332 + }, + { + "epoch": 3.5872446162341247, + "grad_norm": 0.9696611166000366, + "learning_rate": 1.7822506071338163e-06, + "loss": 0.5714, + "step": 4333 + }, + { + "epoch": 3.5880728879072334, + "grad_norm": 0.912272572517395, + "learning_rate": 1.7811972833812167e-06, + "loss": 0.5633, + "step": 4334 + }, + { + "epoch": 3.5889011595803426, + "grad_norm": 0.9608253836631775, + "learning_rate": 1.7801440987102424e-06, + "loss": 0.572, + "step": 4335 + }, + { + "epoch": 3.589729431253451, + "grad_norm": 0.9655832648277283, + "learning_rate": 1.7790910533246752e-06, + "loss": 0.5727, + "step": 4336 + }, + { + "epoch": 3.59055770292656, + "grad_norm": 0.9179590940475464, + "learning_rate": 1.7780381474282683e-06, + "loss": 0.5701, + "step": 4337 + }, + { + "epoch": 3.591385974599669, + "grad_norm": 0.9642294049263, + "learning_rate": 1.7769853812247501e-06, + "loss": 0.5659, + "step": 4338 + }, + { + "epoch": 3.5922142462727775, + "grad_norm": 0.9484195709228516, + "learning_rate": 1.7759327549178208e-06, + "loss": 0.5572, + "step": 4339 + }, + { + "epoch": 3.5930425179458862, + "grad_norm": 0.9366327524185181, + "learning_rate": 1.7748802687111533e-06, + "loss": 0.5596, + "step": 4340 + }, + { + "epoch": 3.593870789618995, + "grad_norm": 0.9256154894828796, + "learning_rate": 1.7738279228083946e-06, + "loss": 0.553, + "step": 4341 + }, + { + "epoch": 3.5946990612921037, + "grad_norm": 0.9243422746658325, + "learning_rate": 1.7727757174131637e-06, + "loss": 0.5703, + "step": 4342 + }, + { + "epoch": 3.5955273329652124, + "grad_norm": 0.9723832607269287, + "learning_rate": 1.7717236527290526e-06, + "loss": 0.5812, + "step": 4343 + }, + { + "epoch": 3.5963556046383216, + "grad_norm": 0.9459588527679443, + "learning_rate": 1.7706717289596263e-06, + "loss": 0.5871, + "step": 4344 + }, + { + "epoch": 3.59718387631143, + "grad_norm": 0.9725896716117859, + "learning_rate": 1.7696199463084205e-06, + "loss": 0.5594, + "step": 4345 + }, + { + "epoch": 3.598012147984539, + "grad_norm": 0.9436221122741699, + "learning_rate": 1.768568304978947e-06, + "loss": 0.5587, + "step": 4346 + }, + { + "epoch": 3.598840419657648, + "grad_norm": 0.9552313685417175, + "learning_rate": 1.7675168051746884e-06, + "loss": 0.572, + "step": 4347 + }, + { + "epoch": 3.5996686913307565, + "grad_norm": 0.9535266160964966, + "learning_rate": 1.7664654470990993e-06, + "loss": 0.5695, + "step": 4348 + }, + { + "epoch": 3.6004969630038652, + "grad_norm": 0.9626250863075256, + "learning_rate": 1.7654142309556088e-06, + "loss": 0.5695, + "step": 4349 + }, + { + "epoch": 3.601325234676974, + "grad_norm": 0.9406429529190063, + "learning_rate": 1.7643631569476166e-06, + "loss": 0.5525, + "step": 4350 + }, + { + "epoch": 3.6021535063500827, + "grad_norm": 0.946359395980835, + "learning_rate": 1.7633122252784963e-06, + "loss": 0.5626, + "step": 4351 + }, + { + "epoch": 3.6029817780231914, + "grad_norm": 0.9508920907974243, + "learning_rate": 1.7622614361515928e-06, + "loss": 0.5569, + "step": 4352 + }, + { + "epoch": 3.6038100496963006, + "grad_norm": 0.9463246464729309, + "learning_rate": 1.7612107897702245e-06, + "loss": 0.5608, + "step": 4353 + }, + { + "epoch": 3.6046383213694093, + "grad_norm": 0.9517480731010437, + "learning_rate": 1.7601602863376816e-06, + "loss": 0.5753, + "step": 4354 + }, + { + "epoch": 3.605466593042518, + "grad_norm": 0.9344833493232727, + "learning_rate": 1.759109926057227e-06, + "loss": 0.5633, + "step": 4355 + }, + { + "epoch": 3.6062948647156268, + "grad_norm": 0.9408402442932129, + "learning_rate": 1.758059709132094e-06, + "loss": 0.5829, + "step": 4356 + }, + { + "epoch": 3.6071231363887355, + "grad_norm": 0.9551196098327637, + "learning_rate": 1.7570096357654918e-06, + "loss": 0.5553, + "step": 4357 + }, + { + "epoch": 3.6079514080618442, + "grad_norm": 0.9403008222579956, + "learning_rate": 1.7559597061605987e-06, + "loss": 0.576, + "step": 4358 + }, + { + "epoch": 3.608779679734953, + "grad_norm": 0.9454553127288818, + "learning_rate": 1.7549099205205673e-06, + "loss": 0.5643, + "step": 4359 + }, + { + "epoch": 3.609607951408062, + "grad_norm": 0.9868146777153015, + "learning_rate": 1.7538602790485194e-06, + "loss": 0.591, + "step": 4360 + }, + { + "epoch": 3.6104362230811704, + "grad_norm": 0.9502206444740295, + "learning_rate": 1.752810781947553e-06, + "loss": 0.5431, + "step": 4361 + }, + { + "epoch": 3.6112644947542796, + "grad_norm": 0.9451961517333984, + "learning_rate": 1.7517614294207347e-06, + "loss": 0.5762, + "step": 4362 + }, + { + "epoch": 3.6120927664273883, + "grad_norm": 0.9424105882644653, + "learning_rate": 1.7507122216711048e-06, + "loss": 0.6107, + "step": 4363 + }, + { + "epoch": 3.612921038100497, + "grad_norm": 0.967083215713501, + "learning_rate": 1.7496631589016754e-06, + "loss": 0.5486, + "step": 4364 + }, + { + "epoch": 3.6137493097736058, + "grad_norm": 0.9303386211395264, + "learning_rate": 1.7486142413154304e-06, + "loss": 0.5345, + "step": 4365 + }, + { + "epoch": 3.6145775814467145, + "grad_norm": 0.9635606408119202, + "learning_rate": 1.7475654691153257e-06, + "loss": 0.563, + "step": 4366 + }, + { + "epoch": 3.615405853119823, + "grad_norm": 0.9364803433418274, + "learning_rate": 1.7465168425042883e-06, + "loss": 0.561, + "step": 4367 + }, + { + "epoch": 3.616234124792932, + "grad_norm": 0.9707603454589844, + "learning_rate": 1.7454683616852172e-06, + "loss": 0.585, + "step": 4368 + }, + { + "epoch": 3.617062396466041, + "grad_norm": 0.9474611282348633, + "learning_rate": 1.744420026860985e-06, + "loss": 0.5754, + "step": 4369 + }, + { + "epoch": 3.6178906681391494, + "grad_norm": 0.9372676014900208, + "learning_rate": 1.7433718382344344e-06, + "loss": 0.5624, + "step": 4370 + }, + { + "epoch": 3.6187189398122586, + "grad_norm": 0.9487974643707275, + "learning_rate": 1.7423237960083794e-06, + "loss": 0.5406, + "step": 4371 + }, + { + "epoch": 3.6195472114853673, + "grad_norm": 0.9595673084259033, + "learning_rate": 1.7412759003856073e-06, + "loss": 0.5977, + "step": 4372 + }, + { + "epoch": 3.620375483158476, + "grad_norm": 0.954731285572052, + "learning_rate": 1.7402281515688756e-06, + "loss": 0.56, + "step": 4373 + }, + { + "epoch": 3.6212037548315847, + "grad_norm": 0.9508479237556458, + "learning_rate": 1.7391805497609137e-06, + "loss": 0.5607, + "step": 4374 + }, + { + "epoch": 3.6220320265046935, + "grad_norm": 0.9357259273529053, + "learning_rate": 1.7381330951644237e-06, + "loss": 0.5545, + "step": 4375 + }, + { + "epoch": 3.622860298177802, + "grad_norm": 0.9336346983909607, + "learning_rate": 1.737085787982078e-06, + "loss": 0.5636, + "step": 4376 + }, + { + "epoch": 3.623688569850911, + "grad_norm": 0.9238201379776001, + "learning_rate": 1.73603862841652e-06, + "loss": 0.5497, + "step": 4377 + }, + { + "epoch": 3.62451684152402, + "grad_norm": 0.9505547881126404, + "learning_rate": 1.7349916166703668e-06, + "loss": 0.5872, + "step": 4378 + }, + { + "epoch": 3.625345113197129, + "grad_norm": 0.945364773273468, + "learning_rate": 1.7339447529462033e-06, + "loss": 0.539, + "step": 4379 + }, + { + "epoch": 3.6261733848702375, + "grad_norm": 0.9546788930892944, + "learning_rate": 1.7328980374465896e-06, + "loss": 0.5489, + "step": 4380 + }, + { + "epoch": 3.6270016565433463, + "grad_norm": 0.9455381035804749, + "learning_rate": 1.7318514703740546e-06, + "loss": 0.5624, + "step": 4381 + }, + { + "epoch": 3.627829928216455, + "grad_norm": 0.9461554288864136, + "learning_rate": 1.7308050519310993e-06, + "loss": 0.5651, + "step": 4382 + }, + { + "epoch": 3.6286581998895637, + "grad_norm": 0.9569746255874634, + "learning_rate": 1.7297587823201964e-06, + "loss": 0.5792, + "step": 4383 + }, + { + "epoch": 3.6294864715626725, + "grad_norm": 0.9243269562721252, + "learning_rate": 1.7287126617437889e-06, + "loss": 0.5493, + "step": 4384 + }, + { + "epoch": 3.630314743235781, + "grad_norm": 0.9244057536125183, + "learning_rate": 1.727666690404291e-06, + "loss": 0.5487, + "step": 4385 + }, + { + "epoch": 3.63114301490889, + "grad_norm": 0.9674320220947266, + "learning_rate": 1.7266208685040896e-06, + "loss": 0.5572, + "step": 4386 + }, + { + "epoch": 3.631971286581999, + "grad_norm": 0.9546018838882446, + "learning_rate": 1.7255751962455407e-06, + "loss": 0.581, + "step": 4387 + }, + { + "epoch": 3.632799558255108, + "grad_norm": 0.9394852519035339, + "learning_rate": 1.7245296738309719e-06, + "loss": 0.5742, + "step": 4388 + }, + { + "epoch": 3.6336278299282165, + "grad_norm": 0.9570075273513794, + "learning_rate": 1.723484301462684e-06, + "loss": 0.5631, + "step": 4389 + }, + { + "epoch": 3.6344561016013253, + "grad_norm": 0.9604787230491638, + "learning_rate": 1.7224390793429435e-06, + "loss": 0.5616, + "step": 4390 + }, + { + "epoch": 3.635284373274434, + "grad_norm": 0.9690744280815125, + "learning_rate": 1.7213940076739931e-06, + "loss": 0.5912, + "step": 4391 + }, + { + "epoch": 3.6361126449475427, + "grad_norm": 0.9428882598876953, + "learning_rate": 1.720349086658044e-06, + "loss": 0.5763, + "step": 4392 + }, + { + "epoch": 3.6369409166206514, + "grad_norm": 0.9487605094909668, + "learning_rate": 1.7193043164972788e-06, + "loss": 0.5652, + "step": 4393 + }, + { + "epoch": 3.6377691882937606, + "grad_norm": 0.9717780351638794, + "learning_rate": 1.7182596973938512e-06, + "loss": 0.5897, + "step": 4394 + }, + { + "epoch": 3.638597459966869, + "grad_norm": 0.9824546575546265, + "learning_rate": 1.7172152295498848e-06, + "loss": 0.5647, + "step": 4395 + }, + { + "epoch": 3.639425731639978, + "grad_norm": 0.9322198629379272, + "learning_rate": 1.716170913167474e-06, + "loss": 0.5532, + "step": 4396 + }, + { + "epoch": 3.640254003313087, + "grad_norm": 0.9403592944145203, + "learning_rate": 1.715126748448685e-06, + "loss": 0.5768, + "step": 4397 + }, + { + "epoch": 3.6410822749861955, + "grad_norm": 0.9282305836677551, + "learning_rate": 1.7140827355955541e-06, + "loss": 0.5354, + "step": 4398 + }, + { + "epoch": 3.6419105466593042, + "grad_norm": 1.0102676153182983, + "learning_rate": 1.7130388748100868e-06, + "loss": 0.5792, + "step": 4399 + }, + { + "epoch": 3.642738818332413, + "grad_norm": 0.9756748676300049, + "learning_rate": 1.711995166294263e-06, + "loss": 0.5691, + "step": 4400 + }, + { + "epoch": 3.6435670900055217, + "grad_norm": 0.9297899603843689, + "learning_rate": 1.710951610250027e-06, + "loss": 0.548, + "step": 4401 + }, + { + "epoch": 3.6443953616786304, + "grad_norm": 0.9422688484191895, + "learning_rate": 1.7099082068792999e-06, + "loss": 0.5344, + "step": 4402 + }, + { + "epoch": 3.6452236333517396, + "grad_norm": 0.9583215713500977, + "learning_rate": 1.7088649563839693e-06, + "loss": 0.5652, + "step": 4403 + }, + { + "epoch": 3.646051905024848, + "grad_norm": 0.9639995098114014, + "learning_rate": 1.7078218589658946e-06, + "loss": 0.5649, + "step": 4404 + }, + { + "epoch": 3.646880176697957, + "grad_norm": 0.9570342898368835, + "learning_rate": 1.7067789148269056e-06, + "loss": 0.5513, + "step": 4405 + }, + { + "epoch": 3.647708448371066, + "grad_norm": 0.9547348618507385, + "learning_rate": 1.7057361241688024e-06, + "loss": 0.5659, + "step": 4406 + }, + { + "epoch": 3.6485367200441745, + "grad_norm": 0.9749144911766052, + "learning_rate": 1.7046934871933551e-06, + "loss": 0.5672, + "step": 4407 + }, + { + "epoch": 3.6493649917172832, + "grad_norm": 0.9632282853126526, + "learning_rate": 1.7036510041023045e-06, + "loss": 0.5761, + "step": 4408 + }, + { + "epoch": 3.650193263390392, + "grad_norm": 0.9284268617630005, + "learning_rate": 1.7026086750973612e-06, + "loss": 0.5758, + "step": 4409 + }, + { + "epoch": 3.6510215350635007, + "grad_norm": 0.9373230934143066, + "learning_rate": 1.7015665003802053e-06, + "loss": 0.5537, + "step": 4410 + }, + { + "epoch": 3.6518498067366094, + "grad_norm": 0.9570644497871399, + "learning_rate": 1.7005244801524896e-06, + "loss": 0.5922, + "step": 4411 + }, + { + "epoch": 3.6526780784097186, + "grad_norm": 0.9559333920478821, + "learning_rate": 1.6994826146158345e-06, + "loss": 0.5854, + "step": 4412 + }, + { + "epoch": 3.6535063500828273, + "grad_norm": 0.943041205406189, + "learning_rate": 1.698440903971831e-06, + "loss": 0.5475, + "step": 4413 + }, + { + "epoch": 3.654334621755936, + "grad_norm": 0.9560774564743042, + "learning_rate": 1.6973993484220403e-06, + "loss": 0.5863, + "step": 4414 + }, + { + "epoch": 3.6551628934290448, + "grad_norm": 0.9528188109397888, + "learning_rate": 1.6963579481679932e-06, + "loss": 0.5536, + "step": 4415 + }, + { + "epoch": 3.6559911651021535, + "grad_norm": 0.9496455192565918, + "learning_rate": 1.6953167034111926e-06, + "loss": 0.5896, + "step": 4416 + }, + { + "epoch": 3.656819436775262, + "grad_norm": 0.9072014689445496, + "learning_rate": 1.6942756143531086e-06, + "loss": 0.5665, + "step": 4417 + }, + { + "epoch": 3.657647708448371, + "grad_norm": 0.9448863863945007, + "learning_rate": 1.6932346811951817e-06, + "loss": 0.5765, + "step": 4418 + }, + { + "epoch": 3.65847598012148, + "grad_norm": 0.9441279768943787, + "learning_rate": 1.692193904138824e-06, + "loss": 0.5721, + "step": 4419 + }, + { + "epoch": 3.6593042517945884, + "grad_norm": 0.9575514793395996, + "learning_rate": 1.6911532833854155e-06, + "loss": 0.5657, + "step": 4420 + }, + { + "epoch": 3.6601325234676976, + "grad_norm": 1.0165443420410156, + "learning_rate": 1.690112819136307e-06, + "loss": 0.5853, + "step": 4421 + }, + { + "epoch": 3.6609607951408063, + "grad_norm": 1.0065933465957642, + "learning_rate": 1.6890725115928176e-06, + "loss": 0.559, + "step": 4422 + }, + { + "epoch": 3.661789066813915, + "grad_norm": 0.9567911028862, + "learning_rate": 1.6880323609562388e-06, + "loss": 0.5734, + "step": 4423 + }, + { + "epoch": 3.6626173384870238, + "grad_norm": 0.9482328295707703, + "learning_rate": 1.6869923674278288e-06, + "loss": 0.5704, + "step": 4424 + }, + { + "epoch": 3.6634456101601325, + "grad_norm": 0.9326344728469849, + "learning_rate": 1.6859525312088166e-06, + "loss": 0.5858, + "step": 4425 + }, + { + "epoch": 3.664273881833241, + "grad_norm": 0.9377856254577637, + "learning_rate": 1.6849128525004016e-06, + "loss": 0.5769, + "step": 4426 + }, + { + "epoch": 3.66510215350635, + "grad_norm": 0.939058244228363, + "learning_rate": 1.6838733315037506e-06, + "loss": 0.5574, + "step": 4427 + }, + { + "epoch": 3.665930425179459, + "grad_norm": 0.9490857720375061, + "learning_rate": 1.682833968420003e-06, + "loss": 0.5636, + "step": 4428 + }, + { + "epoch": 3.6667586968525674, + "grad_norm": 0.902042806148529, + "learning_rate": 1.6817947634502646e-06, + "loss": 0.5698, + "step": 4429 + }, + { + "epoch": 3.6675869685256766, + "grad_norm": 0.9587634205818176, + "learning_rate": 1.6807557167956118e-06, + "loss": 0.5509, + "step": 4430 + }, + { + "epoch": 3.6684152401987853, + "grad_norm": 0.9205626845359802, + "learning_rate": 1.6797168286570914e-06, + "loss": 0.5712, + "step": 4431 + }, + { + "epoch": 3.669243511871894, + "grad_norm": 0.9732058048248291, + "learning_rate": 1.6786780992357182e-06, + "loss": 0.5991, + "step": 4432 + }, + { + "epoch": 3.6700717835450027, + "grad_norm": 0.9664617776870728, + "learning_rate": 1.6776395287324763e-06, + "loss": 0.5626, + "step": 4433 + }, + { + "epoch": 3.6709000552181115, + "grad_norm": 0.943077802658081, + "learning_rate": 1.676601117348321e-06, + "loss": 0.5699, + "step": 4434 + }, + { + "epoch": 3.67172832689122, + "grad_norm": 0.9534650444984436, + "learning_rate": 1.6755628652841721e-06, + "loss": 0.5783, + "step": 4435 + }, + { + "epoch": 3.672556598564329, + "grad_norm": 0.9525501132011414, + "learning_rate": 1.6745247727409242e-06, + "loss": 0.5793, + "step": 4436 + }, + { + "epoch": 3.673384870237438, + "grad_norm": 0.9676088690757751, + "learning_rate": 1.6734868399194377e-06, + "loss": 0.5638, + "step": 4437 + }, + { + "epoch": 3.674213141910547, + "grad_norm": 0.9717739820480347, + "learning_rate": 1.6724490670205423e-06, + "loss": 0.585, + "step": 4438 + }, + { + "epoch": 3.6750414135836555, + "grad_norm": 0.9457798600196838, + "learning_rate": 1.6714114542450388e-06, + "loss": 0.5558, + "step": 4439 + }, + { + "epoch": 3.6758696852567643, + "grad_norm": 0.940519392490387, + "learning_rate": 1.6703740017936948e-06, + "loss": 0.5815, + "step": 4440 + }, + { + "epoch": 3.676697956929873, + "grad_norm": 0.9216038584709167, + "learning_rate": 1.6693367098672475e-06, + "loss": 0.5629, + "step": 4441 + }, + { + "epoch": 3.6775262286029817, + "grad_norm": 0.927438497543335, + "learning_rate": 1.668299578666404e-06, + "loss": 0.5742, + "step": 4442 + }, + { + "epoch": 3.6783545002760905, + "grad_norm": 0.9554510712623596, + "learning_rate": 1.6672626083918387e-06, + "loss": 0.5709, + "step": 4443 + }, + { + "epoch": 3.679182771949199, + "grad_norm": 0.976613461971283, + "learning_rate": 1.6662257992441957e-06, + "loss": 0.5867, + "step": 4444 + }, + { + "epoch": 3.680011043622308, + "grad_norm": 0.9392770528793335, + "learning_rate": 1.6651891514240887e-06, + "loss": 0.563, + "step": 4445 + }, + { + "epoch": 3.680839315295417, + "grad_norm": 0.925324022769928, + "learning_rate": 1.6641526651320995e-06, + "loss": 0.5641, + "step": 4446 + }, + { + "epoch": 3.681667586968526, + "grad_norm": 0.9984707832336426, + "learning_rate": 1.6631163405687773e-06, + "loss": 0.5811, + "step": 4447 + }, + { + "epoch": 3.6824958586416345, + "grad_norm": 0.9349227547645569, + "learning_rate": 1.662080177934642e-06, + "loss": 0.5696, + "step": 4448 + }, + { + "epoch": 3.6833241303147433, + "grad_norm": 0.9371916651725769, + "learning_rate": 1.6610441774301806e-06, + "loss": 0.5787, + "step": 4449 + }, + { + "epoch": 3.684152401987852, + "grad_norm": 0.9504124522209167, + "learning_rate": 1.6600083392558508e-06, + "loss": 0.5706, + "step": 4450 + }, + { + "epoch": 3.6849806736609607, + "grad_norm": 0.9354562759399414, + "learning_rate": 1.6589726636120768e-06, + "loss": 0.5826, + "step": 4451 + }, + { + "epoch": 3.6858089453340694, + "grad_norm": 0.9202892780303955, + "learning_rate": 1.6579371506992523e-06, + "loss": 0.5807, + "step": 4452 + }, + { + "epoch": 3.6866372170071786, + "grad_norm": 0.9467330574989319, + "learning_rate": 1.65690180071774e-06, + "loss": 0.582, + "step": 4453 + }, + { + "epoch": 3.687465488680287, + "grad_norm": 0.9724621772766113, + "learning_rate": 1.6558666138678698e-06, + "loss": 0.5847, + "step": 4454 + }, + { + "epoch": 3.688293760353396, + "grad_norm": 0.9587774276733398, + "learning_rate": 1.6548315903499407e-06, + "loss": 0.5711, + "step": 4455 + }, + { + "epoch": 3.689122032026505, + "grad_norm": 0.9332905411720276, + "learning_rate": 1.6537967303642205e-06, + "loss": 0.5783, + "step": 4456 + }, + { + "epoch": 3.6899503036996135, + "grad_norm": 0.9627639055252075, + "learning_rate": 1.6527620341109452e-06, + "loss": 0.5741, + "step": 4457 + }, + { + "epoch": 3.6907785753727222, + "grad_norm": 0.9341310262680054, + "learning_rate": 1.6517275017903183e-06, + "loss": 0.5605, + "step": 4458 + }, + { + "epoch": 3.691606847045831, + "grad_norm": 0.9802857637405396, + "learning_rate": 1.6506931336025122e-06, + "loss": 0.5723, + "step": 4459 + }, + { + "epoch": 3.6924351187189397, + "grad_norm": 0.9871952533721924, + "learning_rate": 1.649658929747667e-06, + "loss": 0.5768, + "step": 4460 + }, + { + "epoch": 3.6932633903920484, + "grad_norm": 0.9286062717437744, + "learning_rate": 1.6486248904258929e-06, + "loss": 0.566, + "step": 4461 + }, + { + "epoch": 3.6940916620651576, + "grad_norm": 0.9631725549697876, + "learning_rate": 1.6475910158372654e-06, + "loss": 0.571, + "step": 4462 + }, + { + "epoch": 3.694919933738266, + "grad_norm": 0.9414538741111755, + "learning_rate": 1.6465573061818302e-06, + "loss": 0.5526, + "step": 4463 + }, + { + "epoch": 3.695748205411375, + "grad_norm": 0.9554463028907776, + "learning_rate": 1.6455237616596011e-06, + "loss": 0.5846, + "step": 4464 + }, + { + "epoch": 3.696576477084484, + "grad_norm": 0.9494985938072205, + "learning_rate": 1.6444903824705587e-06, + "loss": 0.5667, + "step": 4465 + }, + { + "epoch": 3.6974047487575925, + "grad_norm": 0.9633604288101196, + "learning_rate": 1.6434571688146522e-06, + "loss": 0.5862, + "step": 4466 + }, + { + "epoch": 3.6982330204307012, + "grad_norm": 0.955493152141571, + "learning_rate": 1.642424120891799e-06, + "loss": 0.5591, + "step": 4467 + }, + { + "epoch": 3.69906129210381, + "grad_norm": 0.9700822830200195, + "learning_rate": 1.641391238901885e-06, + "loss": 0.5733, + "step": 4468 + }, + { + "epoch": 3.6998895637769187, + "grad_norm": 0.9600526094436646, + "learning_rate": 1.6403585230447617e-06, + "loss": 0.5819, + "step": 4469 + }, + { + "epoch": 3.7007178354500274, + "grad_norm": 0.9373101592063904, + "learning_rate": 1.639325973520251e-06, + "loss": 0.5749, + "step": 4470 + }, + { + "epoch": 3.7015461071231366, + "grad_norm": 0.9339377284049988, + "learning_rate": 1.6382935905281406e-06, + "loss": 0.5812, + "step": 4471 + }, + { + "epoch": 3.7023743787962453, + "grad_norm": 0.9409409165382385, + "learning_rate": 1.637261374268188e-06, + "loss": 0.561, + "step": 4472 + }, + { + "epoch": 3.703202650469354, + "grad_norm": 0.9259868264198303, + "learning_rate": 1.6362293249401174e-06, + "loss": 0.5636, + "step": 4473 + }, + { + "epoch": 3.7040309221424628, + "grad_norm": 0.9868617057800293, + "learning_rate": 1.6351974427436198e-06, + "loss": 0.5753, + "step": 4474 + }, + { + "epoch": 3.7048591938155715, + "grad_norm": 1.001633882522583, + "learning_rate": 1.6341657278783563e-06, + "loss": 0.5834, + "step": 4475 + }, + { + "epoch": 3.70568746548868, + "grad_norm": 0.9641216993331909, + "learning_rate": 1.6331341805439527e-06, + "loss": 0.6006, + "step": 4476 + }, + { + "epoch": 3.706515737161789, + "grad_norm": 0.9380208253860474, + "learning_rate": 1.6321028009400041e-06, + "loss": 0.5827, + "step": 4477 + }, + { + "epoch": 3.707344008834898, + "grad_norm": 0.9479442238807678, + "learning_rate": 1.631071589266074e-06, + "loss": 0.5537, + "step": 4478 + }, + { + "epoch": 3.7081722805080064, + "grad_norm": 0.9482294321060181, + "learning_rate": 1.6300405457216912e-06, + "loss": 0.5811, + "step": 4479 + }, + { + "epoch": 3.7090005521811156, + "grad_norm": 0.9395027756690979, + "learning_rate": 1.629009670506353e-06, + "loss": 0.5683, + "step": 4480 + }, + { + "epoch": 3.7098288238542243, + "grad_norm": 0.9259159564971924, + "learning_rate": 1.6279789638195242e-06, + "loss": 0.5659, + "step": 4481 + }, + { + "epoch": 3.710657095527333, + "grad_norm": 0.9183517694473267, + "learning_rate": 1.6269484258606366e-06, + "loss": 0.5657, + "step": 4482 + }, + { + "epoch": 3.7114853672004418, + "grad_norm": 0.9426186680793762, + "learning_rate": 1.6259180568290905e-06, + "loss": 0.5749, + "step": 4483 + }, + { + "epoch": 3.7123136388735505, + "grad_norm": 0.9540229439735413, + "learning_rate": 1.6248878569242522e-06, + "loss": 0.5578, + "step": 4484 + }, + { + "epoch": 3.713141910546659, + "grad_norm": 0.9624539613723755, + "learning_rate": 1.6238578263454558e-06, + "loss": 0.5855, + "step": 4485 + }, + { + "epoch": 3.713970182219768, + "grad_norm": 0.9445024728775024, + "learning_rate": 1.6228279652920026e-06, + "loss": 0.5464, + "step": 4486 + }, + { + "epoch": 3.714798453892877, + "grad_norm": 0.9346494078636169, + "learning_rate": 1.6217982739631617e-06, + "loss": 0.5523, + "step": 4487 + }, + { + "epoch": 3.7156267255659854, + "grad_norm": 1.0044512748718262, + "learning_rate": 1.6207687525581684e-06, + "loss": 0.5586, + "step": 4488 + }, + { + "epoch": 3.7164549972390946, + "grad_norm": 0.9639768600463867, + "learning_rate": 1.6197394012762245e-06, + "loss": 0.5676, + "step": 4489 + }, + { + "epoch": 3.7172832689122033, + "grad_norm": 0.9930342435836792, + "learning_rate": 1.6187102203165011e-06, + "loss": 0.5463, + "step": 4490 + }, + { + "epoch": 3.718111540585312, + "grad_norm": 0.9646494388580322, + "learning_rate": 1.617681209878136e-06, + "loss": 0.5516, + "step": 4491 + }, + { + "epoch": 3.7189398122584207, + "grad_norm": 0.9502953886985779, + "learning_rate": 1.616652370160231e-06, + "loss": 0.5792, + "step": 4492 + }, + { + "epoch": 3.7197680839315295, + "grad_norm": 0.9354153871536255, + "learning_rate": 1.6156237013618573e-06, + "loss": 0.549, + "step": 4493 + }, + { + "epoch": 3.720596355604638, + "grad_norm": 0.9361369609832764, + "learning_rate": 1.6145952036820538e-06, + "loss": 0.566, + "step": 4494 + }, + { + "epoch": 3.721424627277747, + "grad_norm": 0.9847058653831482, + "learning_rate": 1.6135668773198252e-06, + "loss": 0.561, + "step": 4495 + }, + { + "epoch": 3.722252898950856, + "grad_norm": 0.9457981586456299, + "learning_rate": 1.6125387224741422e-06, + "loss": 0.5689, + "step": 4496 + }, + { + "epoch": 3.723081170623965, + "grad_norm": 0.9350807666778564, + "learning_rate": 1.6115107393439433e-06, + "loss": 0.5603, + "step": 4497 + }, + { + "epoch": 3.7239094422970735, + "grad_norm": 0.9243747591972351, + "learning_rate": 1.6104829281281343e-06, + "loss": 0.578, + "step": 4498 + }, + { + "epoch": 3.7247377139701823, + "grad_norm": 0.9140326976776123, + "learning_rate": 1.6094552890255869e-06, + "loss": 0.5831, + "step": 4499 + }, + { + "epoch": 3.725565985643291, + "grad_norm": 0.9408622980117798, + "learning_rate": 1.6084278222351391e-06, + "loss": 0.5684, + "step": 4500 + }, + { + "epoch": 3.7263942573163997, + "grad_norm": 0.9529958367347717, + "learning_rate": 1.6074005279555965e-06, + "loss": 0.5876, + "step": 4501 + }, + { + "epoch": 3.7272225289895085, + "grad_norm": 0.9634376764297485, + "learning_rate": 1.6063734063857323e-06, + "loss": 0.5809, + "step": 4502 + }, + { + "epoch": 3.728050800662617, + "grad_norm": 0.9465759992599487, + "learning_rate": 1.6053464577242828e-06, + "loss": 0.564, + "step": 4503 + }, + { + "epoch": 3.728879072335726, + "grad_norm": 0.943699836730957, + "learning_rate": 1.6043196821699536e-06, + "loss": 0.5797, + "step": 4504 + }, + { + "epoch": 3.729707344008835, + "grad_norm": 0.9360603094100952, + "learning_rate": 1.6032930799214164e-06, + "loss": 0.561, + "step": 4505 + }, + { + "epoch": 3.730535615681944, + "grad_norm": 0.9656838178634644, + "learning_rate": 1.6022666511773098e-06, + "loss": 0.5726, + "step": 4506 + }, + { + "epoch": 3.7313638873550525, + "grad_norm": 0.9636357426643372, + "learning_rate": 1.6012403961362372e-06, + "loss": 0.5859, + "step": 4507 + }, + { + "epoch": 3.7321921590281613, + "grad_norm": 0.9537240862846375, + "learning_rate": 1.6002143149967697e-06, + "loss": 0.5798, + "step": 4508 + }, + { + "epoch": 3.73302043070127, + "grad_norm": 0.9626145362854004, + "learning_rate": 1.5991884079574448e-06, + "loss": 0.5778, + "step": 4509 + }, + { + "epoch": 3.7338487023743787, + "grad_norm": 0.9493876099586487, + "learning_rate": 1.598162675216766e-06, + "loss": 0.5688, + "step": 4510 + }, + { + "epoch": 3.7346769740474874, + "grad_norm": 0.9417327046394348, + "learning_rate": 1.597137116973202e-06, + "loss": 0.5628, + "step": 4511 + }, + { + "epoch": 3.7355052457205966, + "grad_norm": 0.9836271405220032, + "learning_rate": 1.59611173342519e-06, + "loss": 0.5616, + "step": 4512 + }, + { + "epoch": 3.736333517393705, + "grad_norm": 0.9088386297225952, + "learning_rate": 1.5950865247711322e-06, + "loss": 0.5635, + "step": 4513 + }, + { + "epoch": 3.737161789066814, + "grad_norm": 0.9311172366142273, + "learning_rate": 1.5940614912093957e-06, + "loss": 0.5795, + "step": 4514 + }, + { + "epoch": 3.737990060739923, + "grad_norm": 0.9570069909095764, + "learning_rate": 1.5930366329383157e-06, + "loss": 0.5753, + "step": 4515 + }, + { + "epoch": 3.7388183324130315, + "grad_norm": 1.1909716129302979, + "learning_rate": 1.592011950156192e-06, + "loss": 0.5637, + "step": 4516 + }, + { + "epoch": 3.7396466040861402, + "grad_norm": 1.0059658288955688, + "learning_rate": 1.5909874430612925e-06, + "loss": 0.5894, + "step": 4517 + }, + { + "epoch": 3.740474875759249, + "grad_norm": 0.9572764039039612, + "learning_rate": 1.5899631118518491e-06, + "loss": 0.5762, + "step": 4518 + }, + { + "epoch": 3.7413031474323577, + "grad_norm": 0.9477603435516357, + "learning_rate": 1.58893895672606e-06, + "loss": 0.5602, + "step": 4519 + }, + { + "epoch": 3.7421314191054664, + "grad_norm": 0.9420804381370544, + "learning_rate": 1.5879149778820898e-06, + "loss": 0.5717, + "step": 4520 + }, + { + "epoch": 3.7429596907785756, + "grad_norm": 0.9785239100456238, + "learning_rate": 1.5868911755180694e-06, + "loss": 0.571, + "step": 4521 + }, + { + "epoch": 3.743787962451684, + "grad_norm": 0.9285955429077148, + "learning_rate": 1.585867549832094e-06, + "loss": 0.5603, + "step": 4522 + }, + { + "epoch": 3.744616234124793, + "grad_norm": 0.971987247467041, + "learning_rate": 1.584844101022227e-06, + "loss": 0.5829, + "step": 4523 + }, + { + "epoch": 3.745444505797902, + "grad_norm": 0.9516214728355408, + "learning_rate": 1.5838208292864954e-06, + "loss": 0.583, + "step": 4524 + }, + { + "epoch": 3.7462727774710105, + "grad_norm": 0.938643753528595, + "learning_rate": 1.5827977348228935e-06, + "loss": 0.5427, + "step": 4525 + }, + { + "epoch": 3.7471010491441192, + "grad_norm": 0.9617331027984619, + "learning_rate": 1.581774817829379e-06, + "loss": 0.5663, + "step": 4526 + }, + { + "epoch": 3.747929320817228, + "grad_norm": 0.9507145881652832, + "learning_rate": 1.5807520785038776e-06, + "loss": 0.5855, + "step": 4527 + }, + { + "epoch": 3.7487575924903367, + "grad_norm": 0.955471932888031, + "learning_rate": 1.57972951704428e-06, + "loss": 0.5676, + "step": 4528 + }, + { + "epoch": 3.7495858641634454, + "grad_norm": 0.9277394413948059, + "learning_rate": 1.5787071336484422e-06, + "loss": 0.5452, + "step": 4529 + }, + { + "epoch": 3.7504141358365546, + "grad_norm": 0.9697600603103638, + "learning_rate": 1.5776849285141858e-06, + "loss": 0.5598, + "step": 4530 + }, + { + "epoch": 3.7512424075096633, + "grad_norm": 0.9462587833404541, + "learning_rate": 1.5766629018392981e-06, + "loss": 0.5766, + "step": 4531 + }, + { + "epoch": 3.752070679182772, + "grad_norm": 0.9286374449729919, + "learning_rate": 1.5756410538215317e-06, + "loss": 0.5747, + "step": 4532 + }, + { + "epoch": 3.7528989508558808, + "grad_norm": 0.9680078625679016, + "learning_rate": 1.5746193846586045e-06, + "loss": 0.5789, + "step": 4533 + }, + { + "epoch": 3.7537272225289895, + "grad_norm": 0.9353684782981873, + "learning_rate": 1.5735978945482e-06, + "loss": 0.5763, + "step": 4534 + }, + { + "epoch": 3.754555494202098, + "grad_norm": 0.9365376830101013, + "learning_rate": 1.5725765836879674e-06, + "loss": 0.5504, + "step": 4535 + }, + { + "epoch": 3.755383765875207, + "grad_norm": 0.9342913627624512, + "learning_rate": 1.571555452275521e-06, + "loss": 0.558, + "step": 4536 + }, + { + "epoch": 3.756212037548316, + "grad_norm": 0.9252297878265381, + "learning_rate": 1.5705345005084394e-06, + "loss": 0.5666, + "step": 4537 + }, + { + "epoch": 3.7570403092214244, + "grad_norm": 1.0099254846572876, + "learning_rate": 1.5695137285842666e-06, + "loss": 0.5627, + "step": 4538 + }, + { + "epoch": 3.7578685808945336, + "grad_norm": 0.9464092254638672, + "learning_rate": 1.5684931367005144e-06, + "loss": 0.5642, + "step": 4539 + }, + { + "epoch": 3.7586968525676423, + "grad_norm": 0.93072509765625, + "learning_rate": 1.5674727250546568e-06, + "loss": 0.5821, + "step": 4540 + }, + { + "epoch": 3.759525124240751, + "grad_norm": 0.9577572345733643, + "learning_rate": 1.5664524938441334e-06, + "loss": 0.56, + "step": 4541 + }, + { + "epoch": 3.7603533959138598, + "grad_norm": 0.9372087121009827, + "learning_rate": 1.5654324432663503e-06, + "loss": 0.5583, + "step": 4542 + }, + { + "epoch": 3.7611816675869685, + "grad_norm": 0.9543371796607971, + "learning_rate": 1.5644125735186776e-06, + "loss": 0.5611, + "step": 4543 + }, + { + "epoch": 3.762009939260077, + "grad_norm": 0.9668126702308655, + "learning_rate": 1.5633928847984503e-06, + "loss": 0.5637, + "step": 4544 + }, + { + "epoch": 3.762838210933186, + "grad_norm": 0.9686347246170044, + "learning_rate": 1.5623733773029692e-06, + "loss": 0.5737, + "step": 4545 + }, + { + "epoch": 3.763666482606295, + "grad_norm": 0.9634156227111816, + "learning_rate": 1.561354051229499e-06, + "loss": 0.5602, + "step": 4546 + }, + { + "epoch": 3.7644947542794034, + "grad_norm": 0.9353019595146179, + "learning_rate": 1.5603349067752705e-06, + "loss": 0.5518, + "step": 4547 + }, + { + "epoch": 3.7653230259525126, + "grad_norm": 0.9436436295509338, + "learning_rate": 1.5593159441374778e-06, + "loss": 0.5652, + "step": 4548 + }, + { + "epoch": 3.7661512976256213, + "grad_norm": 1.0265226364135742, + "learning_rate": 1.55829716351328e-06, + "loss": 0.5624, + "step": 4549 + }, + { + "epoch": 3.76697956929873, + "grad_norm": 0.9422723650932312, + "learning_rate": 1.5572785650998035e-06, + "loss": 0.5649, + "step": 4550 + }, + { + "epoch": 3.7678078409718387, + "grad_norm": 0.9558239579200745, + "learning_rate": 1.5562601490941365e-06, + "loss": 0.5553, + "step": 4551 + }, + { + "epoch": 3.7686361126449475, + "grad_norm": 0.9281789660453796, + "learning_rate": 1.555241915693333e-06, + "loss": 0.5582, + "step": 4552 + }, + { + "epoch": 3.769464384318056, + "grad_norm": 0.9265138506889343, + "learning_rate": 1.5542238650944121e-06, + "loss": 0.5803, + "step": 4553 + }, + { + "epoch": 3.770292655991165, + "grad_norm": 0.9384198784828186, + "learning_rate": 1.5532059974943569e-06, + "loss": 0.5821, + "step": 4554 + }, + { + "epoch": 3.771120927664274, + "grad_norm": 0.9323135018348694, + "learning_rate": 1.552188313090115e-06, + "loss": 0.56, + "step": 4555 + }, + { + "epoch": 3.771949199337383, + "grad_norm": 0.9448620080947876, + "learning_rate": 1.5511708120785994e-06, + "loss": 0.5521, + "step": 4556 + }, + { + "epoch": 3.7727774710104915, + "grad_norm": 0.9806753993034363, + "learning_rate": 1.5501534946566871e-06, + "loss": 0.5842, + "step": 4557 + }, + { + "epoch": 3.7736057426836003, + "grad_norm": 0.9242349863052368, + "learning_rate": 1.549136361021219e-06, + "loss": 0.5764, + "step": 4558 + }, + { + "epoch": 3.774434014356709, + "grad_norm": 0.956167459487915, + "learning_rate": 1.5481194113690017e-06, + "loss": 0.5891, + "step": 4559 + }, + { + "epoch": 3.7752622860298177, + "grad_norm": 0.9456217288970947, + "learning_rate": 1.5471026458968041e-06, + "loss": 0.5821, + "step": 4560 + }, + { + "epoch": 3.7760905577029265, + "grad_norm": 0.9384570717811584, + "learning_rate": 1.5460860648013626e-06, + "loss": 0.5662, + "step": 4561 + }, + { + "epoch": 3.776918829376035, + "grad_norm": 0.9760017395019531, + "learning_rate": 1.5450696682793747e-06, + "loss": 0.5862, + "step": 4562 + }, + { + "epoch": 3.777747101049144, + "grad_norm": 0.9794814586639404, + "learning_rate": 1.5440534565275039e-06, + "loss": 0.5852, + "step": 4563 + }, + { + "epoch": 3.778575372722253, + "grad_norm": 0.9752492904663086, + "learning_rate": 1.5430374297423784e-06, + "loss": 0.5909, + "step": 4564 + }, + { + "epoch": 3.779403644395362, + "grad_norm": 0.969682514667511, + "learning_rate": 1.5420215881205897e-06, + "loss": 0.5505, + "step": 4565 + }, + { + "epoch": 3.7802319160684705, + "grad_norm": 0.9354566335678101, + "learning_rate": 1.5410059318586934e-06, + "loss": 0.5643, + "step": 4566 + }, + { + "epoch": 3.7810601877415793, + "grad_norm": 0.9543263912200928, + "learning_rate": 1.5399904611532095e-06, + "loss": 0.579, + "step": 4567 + }, + { + "epoch": 3.781888459414688, + "grad_norm": 0.9573609232902527, + "learning_rate": 1.5389751762006225e-06, + "loss": 0.5815, + "step": 4568 + }, + { + "epoch": 3.7827167310877967, + "grad_norm": 0.9919794201850891, + "learning_rate": 1.5379600771973805e-06, + "loss": 0.5643, + "step": 4569 + }, + { + "epoch": 3.7835450027609054, + "grad_norm": 0.966391921043396, + "learning_rate": 1.5369451643398959e-06, + "loss": 0.5702, + "step": 4570 + }, + { + "epoch": 3.7843732744340146, + "grad_norm": 0.9727235436439514, + "learning_rate": 1.5359304378245435e-06, + "loss": 0.5607, + "step": 4571 + }, + { + "epoch": 3.785201546107123, + "grad_norm": 0.9183128476142883, + "learning_rate": 1.5349158978476653e-06, + "loss": 0.539, + "step": 4572 + }, + { + "epoch": 3.786029817780232, + "grad_norm": 0.980755090713501, + "learning_rate": 1.533901544605564e-06, + "loss": 0.5761, + "step": 4573 + }, + { + "epoch": 3.786858089453341, + "grad_norm": 0.96429842710495, + "learning_rate": 1.532887378294508e-06, + "loss": 0.5797, + "step": 4574 + }, + { + "epoch": 3.7876863611264495, + "grad_norm": 0.9631879329681396, + "learning_rate": 1.5318733991107287e-06, + "loss": 0.5851, + "step": 4575 + }, + { + "epoch": 3.7885146327995582, + "grad_norm": 0.9594787359237671, + "learning_rate": 1.5308596072504223e-06, + "loss": 0.5656, + "step": 4576 + }, + { + "epoch": 3.789342904472667, + "grad_norm": 0.9394322633743286, + "learning_rate": 1.5298460029097479e-06, + "loss": 0.5554, + "step": 4577 + }, + { + "epoch": 3.7901711761457757, + "grad_norm": 0.9391270875930786, + "learning_rate": 1.5288325862848275e-06, + "loss": 0.5733, + "step": 4578 + }, + { + "epoch": 3.7909994478188844, + "grad_norm": 0.9480547904968262, + "learning_rate": 1.5278193575717492e-06, + "loss": 0.5629, + "step": 4579 + }, + { + "epoch": 3.7918277194919936, + "grad_norm": 0.9235183000564575, + "learning_rate": 1.5268063169665626e-06, + "loss": 0.5626, + "step": 4580 + }, + { + "epoch": 3.792655991165102, + "grad_norm": 0.9616346955299377, + "learning_rate": 1.5257934646652822e-06, + "loss": 0.557, + "step": 4581 + }, + { + "epoch": 3.793484262838211, + "grad_norm": 0.9767917394638062, + "learning_rate": 1.5247808008638843e-06, + "loss": 0.5693, + "step": 4582 + }, + { + "epoch": 3.79431253451132, + "grad_norm": 0.9262551665306091, + "learning_rate": 1.5237683257583102e-06, + "loss": 0.5638, + "step": 4583 + }, + { + "epoch": 3.7951408061844285, + "grad_norm": 0.9267538189888, + "learning_rate": 1.522756039544465e-06, + "loss": 0.5801, + "step": 4584 + }, + { + "epoch": 3.7959690778575372, + "grad_norm": 0.9917179346084595, + "learning_rate": 1.5217439424182167e-06, + "loss": 0.5715, + "step": 4585 + }, + { + "epoch": 3.796797349530646, + "grad_norm": 0.9482493996620178, + "learning_rate": 1.5207320345753957e-06, + "loss": 0.5618, + "step": 4586 + }, + { + "epoch": 3.7976256212037547, + "grad_norm": 0.9342687726020813, + "learning_rate": 1.5197203162117979e-06, + "loss": 0.5863, + "step": 4587 + }, + { + "epoch": 3.7984538928768634, + "grad_norm": 0.9577085375785828, + "learning_rate": 1.5187087875231809e-06, + "loss": 0.5501, + "step": 4588 + }, + { + "epoch": 3.7992821645499726, + "grad_norm": 0.9561404585838318, + "learning_rate": 1.5176974487052658e-06, + "loss": 0.576, + "step": 4589 + }, + { + "epoch": 3.8001104362230813, + "grad_norm": 0.9302974343299866, + "learning_rate": 1.5166862999537377e-06, + "loss": 0.5551, + "step": 4590 + }, + { + "epoch": 3.80093870789619, + "grad_norm": 0.9463371634483337, + "learning_rate": 1.5156753414642445e-06, + "loss": 0.5531, + "step": 4591 + }, + { + "epoch": 3.8017669795692988, + "grad_norm": 0.9446096420288086, + "learning_rate": 1.5146645734323961e-06, + "loss": 0.5652, + "step": 4592 + }, + { + "epoch": 3.8025952512424075, + "grad_norm": 0.9582951068878174, + "learning_rate": 1.513653996053769e-06, + "loss": 0.5709, + "step": 4593 + }, + { + "epoch": 3.803423522915516, + "grad_norm": 0.961769163608551, + "learning_rate": 1.5126436095238978e-06, + "loss": 0.576, + "step": 4594 + }, + { + "epoch": 3.804251794588625, + "grad_norm": 0.97950279712677, + "learning_rate": 1.5116334140382846e-06, + "loss": 0.562, + "step": 4595 + }, + { + "epoch": 3.805080066261734, + "grad_norm": 0.9386860132217407, + "learning_rate": 1.5106234097923922e-06, + "loss": 0.5723, + "step": 4596 + }, + { + "epoch": 3.8059083379348424, + "grad_norm": 0.9200562238693237, + "learning_rate": 1.509613596981647e-06, + "loss": 0.5708, + "step": 4597 + }, + { + "epoch": 3.8067366096079516, + "grad_norm": 0.9222211837768555, + "learning_rate": 1.5086039758014386e-06, + "loss": 0.5711, + "step": 4598 + }, + { + "epoch": 3.8075648812810603, + "grad_norm": 0.9545297026634216, + "learning_rate": 1.5075945464471198e-06, + "loss": 0.5628, + "step": 4599 + }, + { + "epoch": 3.808393152954169, + "grad_norm": 0.918067455291748, + "learning_rate": 1.5065853091140041e-06, + "loss": 0.5429, + "step": 4600 + }, + { + "epoch": 3.8092214246272778, + "grad_norm": 0.9256977438926697, + "learning_rate": 1.5055762639973714e-06, + "loss": 0.5593, + "step": 4601 + }, + { + "epoch": 3.8100496963003865, + "grad_norm": 0.9570629000663757, + "learning_rate": 1.5045674112924612e-06, + "loss": 0.5545, + "step": 4602 + }, + { + "epoch": 3.810877967973495, + "grad_norm": 0.9828989505767822, + "learning_rate": 1.5035587511944778e-06, + "loss": 0.5776, + "step": 4603 + }, + { + "epoch": 3.811706239646604, + "grad_norm": 0.9521099925041199, + "learning_rate": 1.5025502838985884e-06, + "loss": 0.5645, + "step": 4604 + }, + { + "epoch": 3.812534511319713, + "grad_norm": 0.9580879211425781, + "learning_rate": 1.5015420095999197e-06, + "loss": 0.5754, + "step": 4605 + }, + { + "epoch": 3.8133627829928214, + "grad_norm": 0.9599860906600952, + "learning_rate": 1.500533928493565e-06, + "loss": 0.5623, + "step": 4606 + }, + { + "epoch": 3.8141910546659306, + "grad_norm": 0.9527266621589661, + "learning_rate": 1.4995260407745782e-06, + "loss": 0.5931, + "step": 4607 + }, + { + "epoch": 3.8150193263390393, + "grad_norm": 0.9410575032234192, + "learning_rate": 1.498518346637976e-06, + "loss": 0.569, + "step": 4608 + }, + { + "epoch": 3.815847598012148, + "grad_norm": 0.9750106930732727, + "learning_rate": 1.4975108462787382e-06, + "loss": 0.5838, + "step": 4609 + }, + { + "epoch": 3.8166758696852567, + "grad_norm": 0.9242303967475891, + "learning_rate": 1.4965035398918065e-06, + "loss": 0.5595, + "step": 4610 + }, + { + "epoch": 3.8175041413583655, + "grad_norm": 0.9424212574958801, + "learning_rate": 1.495496427672085e-06, + "loss": 0.5949, + "step": 4611 + }, + { + "epoch": 3.818332413031474, + "grad_norm": 0.9485701322555542, + "learning_rate": 1.4944895098144419e-06, + "loss": 0.5799, + "step": 4612 + }, + { + "epoch": 3.819160684704583, + "grad_norm": 0.9476267695426941, + "learning_rate": 1.4934827865137047e-06, + "loss": 0.5584, + "step": 4613 + }, + { + "epoch": 3.819988956377692, + "grad_norm": 0.9722000360488892, + "learning_rate": 1.4924762579646653e-06, + "loss": 0.5665, + "step": 4614 + }, + { + "epoch": 3.820817228050801, + "grad_norm": 0.9728968739509583, + "learning_rate": 1.4914699243620795e-06, + "loss": 0.5786, + "step": 4615 + }, + { + "epoch": 3.8216454997239095, + "grad_norm": 0.96755051612854, + "learning_rate": 1.4904637859006605e-06, + "loss": 0.5529, + "step": 4616 + }, + { + "epoch": 3.8224737713970183, + "grad_norm": 0.9613732695579529, + "learning_rate": 1.4894578427750888e-06, + "loss": 0.5847, + "step": 4617 + }, + { + "epoch": 3.823302043070127, + "grad_norm": 0.9679767489433289, + "learning_rate": 1.488452095180004e-06, + "loss": 0.5623, + "step": 4618 + }, + { + "epoch": 3.8241303147432357, + "grad_norm": 0.9799363613128662, + "learning_rate": 1.4874465433100093e-06, + "loss": 0.5712, + "step": 4619 + }, + { + "epoch": 3.8249585864163445, + "grad_norm": 0.962107241153717, + "learning_rate": 1.4864411873596695e-06, + "loss": 0.5596, + "step": 4620 + }, + { + "epoch": 3.825786858089453, + "grad_norm": 0.9862436056137085, + "learning_rate": 1.4854360275235118e-06, + "loss": 0.5801, + "step": 4621 + }, + { + "epoch": 3.826615129762562, + "grad_norm": 0.9500226974487305, + "learning_rate": 1.484431063996025e-06, + "loss": 0.5467, + "step": 4622 + }, + { + "epoch": 3.827443401435671, + "grad_norm": 0.9733493328094482, + "learning_rate": 1.4834262969716607e-06, + "loss": 0.549, + "step": 4623 + }, + { + "epoch": 3.82827167310878, + "grad_norm": 0.9474700093269348, + "learning_rate": 1.4824217266448312e-06, + "loss": 0.5497, + "step": 4624 + }, + { + "epoch": 3.8290999447818885, + "grad_norm": 0.9376130104064941, + "learning_rate": 1.4814173532099116e-06, + "loss": 0.5681, + "step": 4625 + }, + { + "epoch": 3.8299282164549973, + "grad_norm": 0.9183100461959839, + "learning_rate": 1.4804131768612406e-06, + "loss": 0.5719, + "step": 4626 + }, + { + "epoch": 3.830756488128106, + "grad_norm": 0.9759798645973206, + "learning_rate": 1.4794091977931136e-06, + "loss": 0.5773, + "step": 4627 + }, + { + "epoch": 3.8315847598012147, + "grad_norm": 0.9638318419456482, + "learning_rate": 1.4784054161997937e-06, + "loss": 0.5681, + "step": 4628 + }, + { + "epoch": 3.8324130314743234, + "grad_norm": 0.9652249217033386, + "learning_rate": 1.4774018322755027e-06, + "loss": 0.5866, + "step": 4629 + }, + { + "epoch": 3.8332413031474326, + "grad_norm": 0.9402887225151062, + "learning_rate": 1.4763984462144237e-06, + "loss": 0.569, + "step": 4630 + }, + { + "epoch": 3.834069574820541, + "grad_norm": 0.925959587097168, + "learning_rate": 1.4753952582107046e-06, + "loss": 0.5945, + "step": 4631 + }, + { + "epoch": 3.83489784649365, + "grad_norm": 0.9718639254570007, + "learning_rate": 1.4743922684584514e-06, + "loss": 0.5523, + "step": 4632 + }, + { + "epoch": 3.835726118166759, + "grad_norm": 0.9743525981903076, + "learning_rate": 1.4733894771517337e-06, + "loss": 0.58, + "step": 4633 + }, + { + "epoch": 3.8365543898398675, + "grad_norm": 0.9634748697280884, + "learning_rate": 1.4723868844845818e-06, + "loss": 0.5715, + "step": 4634 + }, + { + "epoch": 3.8373826615129762, + "grad_norm": 0.9775907397270203, + "learning_rate": 1.471384490650989e-06, + "loss": 0.5697, + "step": 4635 + }, + { + "epoch": 3.838210933186085, + "grad_norm": 0.9504139423370361, + "learning_rate": 1.4703822958449081e-06, + "loss": 0.5797, + "step": 4636 + }, + { + "epoch": 3.8390392048591937, + "grad_norm": 0.9717209339141846, + "learning_rate": 1.4693803002602553e-06, + "loss": 0.5692, + "step": 4637 + }, + { + "epoch": 3.8398674765323024, + "grad_norm": 0.9564605951309204, + "learning_rate": 1.4683785040909078e-06, + "loss": 0.5719, + "step": 4638 + }, + { + "epoch": 3.8406957482054116, + "grad_norm": 0.951598584651947, + "learning_rate": 1.4673769075307023e-06, + "loss": 0.5481, + "step": 4639 + }, + { + "epoch": 3.84152401987852, + "grad_norm": 0.9717726111412048, + "learning_rate": 1.4663755107734395e-06, + "loss": 0.537, + "step": 4640 + }, + { + "epoch": 3.842352291551629, + "grad_norm": 0.9653568267822266, + "learning_rate": 1.4653743140128802e-06, + "loss": 0.5729, + "step": 4641 + }, + { + "epoch": 3.843180563224738, + "grad_norm": 0.9710245728492737, + "learning_rate": 1.4643733174427454e-06, + "loss": 0.5816, + "step": 4642 + }, + { + "epoch": 3.8440088348978465, + "grad_norm": 0.9537909626960754, + "learning_rate": 1.4633725212567206e-06, + "loss": 0.5585, + "step": 4643 + }, + { + "epoch": 3.8448371065709552, + "grad_norm": 0.9765188694000244, + "learning_rate": 1.4623719256484495e-06, + "loss": 0.5542, + "step": 4644 + }, + { + "epoch": 3.845665378244064, + "grad_norm": 0.9446712136268616, + "learning_rate": 1.4613715308115378e-06, + "loss": 0.5696, + "step": 4645 + }, + { + "epoch": 3.8464936499171727, + "grad_norm": 0.9561092853546143, + "learning_rate": 1.4603713369395533e-06, + "loss": 0.555, + "step": 4646 + }, + { + "epoch": 3.8473219215902814, + "grad_norm": 0.9431309700012207, + "learning_rate": 1.4593713442260233e-06, + "loss": 0.5589, + "step": 4647 + }, + { + "epoch": 3.8481501932633906, + "grad_norm": 0.9409856796264648, + "learning_rate": 1.4583715528644384e-06, + "loss": 0.5574, + "step": 4648 + }, + { + "epoch": 3.8489784649364993, + "grad_norm": 0.9805465936660767, + "learning_rate": 1.4573719630482485e-06, + "loss": 0.5555, + "step": 4649 + }, + { + "epoch": 3.849806736609608, + "grad_norm": 0.9668418765068054, + "learning_rate": 1.456372574970863e-06, + "loss": 0.5721, + "step": 4650 + }, + { + "epoch": 3.8506350082827168, + "grad_norm": 0.937393844127655, + "learning_rate": 1.455373388825656e-06, + "loss": 0.5619, + "step": 4651 + }, + { + "epoch": 3.8514632799558255, + "grad_norm": 0.956749439239502, + "learning_rate": 1.4543744048059613e-06, + "loss": 0.5882, + "step": 4652 + }, + { + "epoch": 3.852291551628934, + "grad_norm": 0.95976722240448, + "learning_rate": 1.4533756231050712e-06, + "loss": 0.5573, + "step": 4653 + }, + { + "epoch": 3.853119823302043, + "grad_norm": 0.9402530789375305, + "learning_rate": 1.4523770439162414e-06, + "loss": 0.549, + "step": 4654 + }, + { + "epoch": 3.853948094975152, + "grad_norm": 0.960904061794281, + "learning_rate": 1.4513786674326888e-06, + "loss": 0.5671, + "step": 4655 + }, + { + "epoch": 3.8547763666482604, + "grad_norm": 0.929169774055481, + "learning_rate": 1.450380493847588e-06, + "loss": 0.5482, + "step": 4656 + }, + { + "epoch": 3.8556046383213696, + "grad_norm": 0.9257175922393799, + "learning_rate": 1.449382523354077e-06, + "loss": 0.5526, + "step": 4657 + }, + { + "epoch": 3.8564329099944783, + "grad_norm": 0.9734342098236084, + "learning_rate": 1.4483847561452553e-06, + "loss": 0.5601, + "step": 4658 + }, + { + "epoch": 3.857261181667587, + "grad_norm": 0.9698025584220886, + "learning_rate": 1.4473871924141788e-06, + "loss": 0.568, + "step": 4659 + }, + { + "epoch": 3.8580894533406958, + "grad_norm": 0.9453846216201782, + "learning_rate": 1.4463898323538694e-06, + "loss": 0.5618, + "step": 4660 + }, + { + "epoch": 3.8589177250138045, + "grad_norm": 0.9435855150222778, + "learning_rate": 1.4453926761573062e-06, + "loss": 0.559, + "step": 4661 + }, + { + "epoch": 3.859745996686913, + "grad_norm": 0.9480921030044556, + "learning_rate": 1.4443957240174277e-06, + "loss": 0.5593, + "step": 4662 + }, + { + "epoch": 3.860574268360022, + "grad_norm": 0.9516395926475525, + "learning_rate": 1.443398976127137e-06, + "loss": 0.552, + "step": 4663 + }, + { + "epoch": 3.861402540033131, + "grad_norm": 0.9439289569854736, + "learning_rate": 1.4424024326792955e-06, + "loss": 0.5564, + "step": 4664 + }, + { + "epoch": 3.8622308117062394, + "grad_norm": 0.9423355460166931, + "learning_rate": 1.4414060938667236e-06, + "loss": 0.5428, + "step": 4665 + }, + { + "epoch": 3.8630590833793486, + "grad_norm": 0.9807745814323425, + "learning_rate": 1.4404099598822041e-06, + "loss": 0.5695, + "step": 4666 + }, + { + "epoch": 3.8638873550524573, + "grad_norm": 0.9256470203399658, + "learning_rate": 1.4394140309184815e-06, + "loss": 0.5766, + "step": 4667 + }, + { + "epoch": 3.864715626725566, + "grad_norm": 0.9468652009963989, + "learning_rate": 1.4384183071682562e-06, + "loss": 0.5762, + "step": 4668 + }, + { + "epoch": 3.8655438983986747, + "grad_norm": 0.9390266537666321, + "learning_rate": 1.4374227888241921e-06, + "loss": 0.5779, + "step": 4669 + }, + { + "epoch": 3.8663721700717835, + "grad_norm": 0.9410127997398376, + "learning_rate": 1.4364274760789143e-06, + "loss": 0.5448, + "step": 4670 + }, + { + "epoch": 3.867200441744892, + "grad_norm": 0.9276284575462341, + "learning_rate": 1.4354323691250048e-06, + "loss": 0.5764, + "step": 4671 + }, + { + "epoch": 3.868028713418001, + "grad_norm": 0.9957787394523621, + "learning_rate": 1.4344374681550089e-06, + "loss": 0.5819, + "step": 4672 + }, + { + "epoch": 3.86885698509111, + "grad_norm": 0.9688808917999268, + "learning_rate": 1.4334427733614287e-06, + "loss": 0.5565, + "step": 4673 + }, + { + "epoch": 3.869685256764219, + "grad_norm": 0.9403797388076782, + "learning_rate": 1.4324482849367304e-06, + "loss": 0.5538, + "step": 4674 + }, + { + "epoch": 3.8705135284373275, + "grad_norm": 0.9486374855041504, + "learning_rate": 1.4314540030733364e-06, + "loss": 0.5529, + "step": 4675 + }, + { + "epoch": 3.8713418001104363, + "grad_norm": 0.9329063296318054, + "learning_rate": 1.4304599279636333e-06, + "loss": 0.5656, + "step": 4676 + }, + { + "epoch": 3.872170071783545, + "grad_norm": 0.9217836856842041, + "learning_rate": 1.4294660597999627e-06, + "loss": 0.5557, + "step": 4677 + }, + { + "epoch": 3.8729983434566537, + "grad_norm": 0.926406741142273, + "learning_rate": 1.4284723987746305e-06, + "loss": 0.5532, + "step": 4678 + }, + { + "epoch": 3.8738266151297625, + "grad_norm": 0.9605270624160767, + "learning_rate": 1.427478945079901e-06, + "loss": 0.5652, + "step": 4679 + }, + { + "epoch": 3.874654886802871, + "grad_norm": 0.9510946273803711, + "learning_rate": 1.4264856989079973e-06, + "loss": 0.5618, + "step": 4680 + }, + { + "epoch": 3.87548315847598, + "grad_norm": 0.9574974775314331, + "learning_rate": 1.4254926604511033e-06, + "loss": 0.5822, + "step": 4681 + }, + { + "epoch": 3.876311430149089, + "grad_norm": 0.9598250985145569, + "learning_rate": 1.424499829901364e-06, + "loss": 0.5537, + "step": 4682 + }, + { + "epoch": 3.877139701822198, + "grad_norm": 0.9691941738128662, + "learning_rate": 1.4235072074508822e-06, + "loss": 0.568, + "step": 4683 + }, + { + "epoch": 3.8779679734953065, + "grad_norm": 0.9728984236717224, + "learning_rate": 1.4225147932917193e-06, + "loss": 0.567, + "step": 4684 + }, + { + "epoch": 3.8787962451684153, + "grad_norm": 0.921758234500885, + "learning_rate": 1.4215225876159003e-06, + "loss": 0.5587, + "step": 4685 + }, + { + "epoch": 3.879624516841524, + "grad_norm": 0.9491950869560242, + "learning_rate": 1.420530590615408e-06, + "loss": 0.5479, + "step": 4686 + }, + { + "epoch": 3.8804527885146327, + "grad_norm": 0.9212395548820496, + "learning_rate": 1.4195388024821827e-06, + "loss": 0.5623, + "step": 4687 + }, + { + "epoch": 3.8812810601877414, + "grad_norm": 0.970206081867218, + "learning_rate": 1.418547223408127e-06, + "loss": 0.571, + "step": 4688 + }, + { + "epoch": 3.8821093318608506, + "grad_norm": 0.9378070831298828, + "learning_rate": 1.4175558535851037e-06, + "loss": 0.5823, + "step": 4689 + }, + { + "epoch": 3.882937603533959, + "grad_norm": 0.9203671813011169, + "learning_rate": 1.416564693204931e-06, + "loss": 0.5705, + "step": 4690 + }, + { + "epoch": 3.883765875207068, + "grad_norm": 0.9873926043510437, + "learning_rate": 1.4155737424593907e-06, + "loss": 0.5612, + "step": 4691 + }, + { + "epoch": 3.884594146880177, + "grad_norm": 0.9374533295631409, + "learning_rate": 1.4145830015402235e-06, + "loss": 0.5808, + "step": 4692 + }, + { + "epoch": 3.8854224185532855, + "grad_norm": 0.9469799399375916, + "learning_rate": 1.413592470639126e-06, + "loss": 0.5673, + "step": 4693 + }, + { + "epoch": 3.8862506902263942, + "grad_norm": 0.9671788811683655, + "learning_rate": 1.4126021499477592e-06, + "loss": 0.5755, + "step": 4694 + }, + { + "epoch": 3.887078961899503, + "grad_norm": 0.9435954689979553, + "learning_rate": 1.4116120396577398e-06, + "loss": 0.5686, + "step": 4695 + }, + { + "epoch": 3.8879072335726117, + "grad_norm": 0.9866729378700256, + "learning_rate": 1.410622139960644e-06, + "loss": 0.5456, + "step": 4696 + }, + { + "epoch": 3.8887355052457204, + "grad_norm": 0.9504481554031372, + "learning_rate": 1.409632451048009e-06, + "loss": 0.5779, + "step": 4697 + }, + { + "epoch": 3.8895637769188296, + "grad_norm": 0.9329410195350647, + "learning_rate": 1.4086429731113316e-06, + "loss": 0.5625, + "step": 4698 + }, + { + "epoch": 3.890392048591938, + "grad_norm": 0.9381130933761597, + "learning_rate": 1.4076537063420642e-06, + "loss": 0.5905, + "step": 4699 + }, + { + "epoch": 3.891220320265047, + "grad_norm": 0.9857736229896545, + "learning_rate": 1.4066646509316222e-06, + "loss": 0.5887, + "step": 4700 + }, + { + "epoch": 3.8920485919381558, + "grad_norm": 0.9994931817054749, + "learning_rate": 1.4056758070713788e-06, + "loss": 0.5691, + "step": 4701 + }, + { + "epoch": 3.8928768636112645, + "grad_norm": 0.9347181916236877, + "learning_rate": 1.404687174952665e-06, + "loss": 0.5806, + "step": 4702 + }, + { + "epoch": 3.8937051352843732, + "grad_norm": 0.9398673176765442, + "learning_rate": 1.4036987547667725e-06, + "loss": 0.5626, + "step": 4703 + }, + { + "epoch": 3.894533406957482, + "grad_norm": 0.9418836832046509, + "learning_rate": 1.4027105467049521e-06, + "loss": 0.5514, + "step": 4704 + }, + { + "epoch": 3.8953616786305907, + "grad_norm": 0.9223387241363525, + "learning_rate": 1.401722550958411e-06, + "loss": 0.5374, + "step": 4705 + }, + { + "epoch": 3.8961899503036994, + "grad_norm": 0.9222207069396973, + "learning_rate": 1.4007347677183197e-06, + "loss": 0.5535, + "step": 4706 + }, + { + "epoch": 3.8970182219768086, + "grad_norm": 0.9431470036506653, + "learning_rate": 1.3997471971758022e-06, + "loss": 0.5727, + "step": 4707 + }, + { + "epoch": 3.8978464936499173, + "grad_norm": 0.9395588040351868, + "learning_rate": 1.398759839521947e-06, + "loss": 0.5484, + "step": 4708 + }, + { + "epoch": 3.898674765323026, + "grad_norm": 0.9754805564880371, + "learning_rate": 1.397772694947796e-06, + "loss": 0.5617, + "step": 4709 + }, + { + "epoch": 3.8995030369961348, + "grad_norm": 0.9596531987190247, + "learning_rate": 1.3967857636443539e-06, + "loss": 0.5472, + "step": 4710 + }, + { + "epoch": 3.9003313086692435, + "grad_norm": 0.961883008480072, + "learning_rate": 1.3957990458025833e-06, + "loss": 0.5697, + "step": 4711 + }, + { + "epoch": 3.901159580342352, + "grad_norm": 0.9568914175033569, + "learning_rate": 1.3948125416134029e-06, + "loss": 0.56, + "step": 4712 + }, + { + "epoch": 3.901987852015461, + "grad_norm": 0.9424316883087158, + "learning_rate": 1.3938262512676937e-06, + "loss": 0.5485, + "step": 4713 + }, + { + "epoch": 3.90281612368857, + "grad_norm": 0.9500327706336975, + "learning_rate": 1.392840174956294e-06, + "loss": 0.5748, + "step": 4714 + }, + { + "epoch": 3.9036443953616784, + "grad_norm": 0.9448386430740356, + "learning_rate": 1.3918543128699986e-06, + "loss": 0.5407, + "step": 4715 + }, + { + "epoch": 3.9044726670347876, + "grad_norm": 0.9605383276939392, + "learning_rate": 1.390868665199564e-06, + "loss": 0.5728, + "step": 4716 + }, + { + "epoch": 3.9053009387078963, + "grad_norm": 0.9455352425575256, + "learning_rate": 1.389883232135705e-06, + "loss": 0.5506, + "step": 4717 + }, + { + "epoch": 3.906129210381005, + "grad_norm": 0.9913429617881775, + "learning_rate": 1.3888980138690906e-06, + "loss": 0.5536, + "step": 4718 + }, + { + "epoch": 3.9069574820541138, + "grad_norm": 0.9688431024551392, + "learning_rate": 1.3879130105903526e-06, + "loss": 0.555, + "step": 4719 + }, + { + "epoch": 3.9077857537272225, + "grad_norm": 0.9419496655464172, + "learning_rate": 1.3869282224900815e-06, + "loss": 0.5727, + "step": 4720 + }, + { + "epoch": 3.908614025400331, + "grad_norm": 0.9332079291343689, + "learning_rate": 1.3859436497588225e-06, + "loss": 0.572, + "step": 4721 + }, + { + "epoch": 3.90944229707344, + "grad_norm": 0.952717125415802, + "learning_rate": 1.3849592925870821e-06, + "loss": 0.56, + "step": 4722 + }, + { + "epoch": 3.910270568746549, + "grad_norm": 0.944283127784729, + "learning_rate": 1.3839751511653253e-06, + "loss": 0.5803, + "step": 4723 + }, + { + "epoch": 3.9110988404196574, + "grad_norm": 0.9370842576026917, + "learning_rate": 1.3829912256839725e-06, + "loss": 0.5802, + "step": 4724 + }, + { + "epoch": 3.9119271120927666, + "grad_norm": 0.9332932829856873, + "learning_rate": 1.3820075163334047e-06, + "loss": 0.5648, + "step": 4725 + }, + { + "epoch": 3.9127553837658753, + "grad_norm": 0.9598782658576965, + "learning_rate": 1.3810240233039618e-06, + "loss": 0.5565, + "step": 4726 + }, + { + "epoch": 3.913583655438984, + "grad_norm": 0.9564656019210815, + "learning_rate": 1.3800407467859383e-06, + "loss": 0.5594, + "step": 4727 + }, + { + "epoch": 3.9144119271120927, + "grad_norm": 0.9756644368171692, + "learning_rate": 1.3790576869695913e-06, + "loss": 0.5425, + "step": 4728 + }, + { + "epoch": 3.9152401987852015, + "grad_norm": 0.9616433382034302, + "learning_rate": 1.3780748440451314e-06, + "loss": 0.5802, + "step": 4729 + }, + { + "epoch": 3.91606847045831, + "grad_norm": 0.9436020851135254, + "learning_rate": 1.3770922182027319e-06, + "loss": 0.5554, + "step": 4730 + }, + { + "epoch": 3.916896742131419, + "grad_norm": 0.9598777890205383, + "learning_rate": 1.3761098096325195e-06, + "loss": 0.5552, + "step": 4731 + }, + { + "epoch": 3.917725013804528, + "grad_norm": 0.9497062563896179, + "learning_rate": 1.3751276185245821e-06, + "loss": 0.5612, + "step": 4732 + }, + { + "epoch": 3.918553285477637, + "grad_norm": 0.9690037369728088, + "learning_rate": 1.3741456450689656e-06, + "loss": 0.5682, + "step": 4733 + }, + { + "epoch": 3.9193815571507455, + "grad_norm": 0.9524545073509216, + "learning_rate": 1.373163889455671e-06, + "loss": 0.5795, + "step": 4734 + }, + { + "epoch": 3.9202098288238543, + "grad_norm": 0.9572216272354126, + "learning_rate": 1.3721823518746591e-06, + "loss": 0.558, + "step": 4735 + }, + { + "epoch": 3.921038100496963, + "grad_norm": 0.9508054256439209, + "learning_rate": 1.3712010325158499e-06, + "loss": 0.5578, + "step": 4736 + }, + { + "epoch": 3.9218663721700717, + "grad_norm": 0.9361369013786316, + "learning_rate": 1.3702199315691176e-06, + "loss": 0.5661, + "step": 4737 + }, + { + "epoch": 3.9226946438431805, + "grad_norm": 0.9344385266304016, + "learning_rate": 1.3692390492242973e-06, + "loss": 0.571, + "step": 4738 + }, + { + "epoch": 3.923522915516289, + "grad_norm": 0.9559999704360962, + "learning_rate": 1.3682583856711807e-06, + "loss": 0.5441, + "step": 4739 + }, + { + "epoch": 3.924351187189398, + "grad_norm": 0.9483859539031982, + "learning_rate": 1.3672779410995155e-06, + "loss": 0.5645, + "step": 4740 + }, + { + "epoch": 3.925179458862507, + "grad_norm": 0.9504157304763794, + "learning_rate": 1.3662977156990093e-06, + "loss": 0.5512, + "step": 4741 + }, + { + "epoch": 3.926007730535616, + "grad_norm": 0.9570707678794861, + "learning_rate": 1.3653177096593282e-06, + "loss": 0.5722, + "step": 4742 + }, + { + "epoch": 3.9268360022087245, + "grad_norm": 0.9542663097381592, + "learning_rate": 1.3643379231700915e-06, + "loss": 0.5385, + "step": 4743 + }, + { + "epoch": 3.9276642738818333, + "grad_norm": 0.9425368309020996, + "learning_rate": 1.3633583564208808e-06, + "loss": 0.547, + "step": 4744 + }, + { + "epoch": 3.928492545554942, + "grad_norm": 0.9227926731109619, + "learning_rate": 1.362379009601233e-06, + "loss": 0.5664, + "step": 4745 + }, + { + "epoch": 3.9293208172280507, + "grad_norm": 0.9677916765213013, + "learning_rate": 1.3613998829006408e-06, + "loss": 0.5622, + "step": 4746 + }, + { + "epoch": 3.9301490889011594, + "grad_norm": 0.9727103114128113, + "learning_rate": 1.360420976508558e-06, + "loss": 0.5794, + "step": 4747 + }, + { + "epoch": 3.9309773605742686, + "grad_norm": 0.930270254611969, + "learning_rate": 1.359442290614394e-06, + "loss": 0.573, + "step": 4748 + }, + { + "epoch": 3.931805632247377, + "grad_norm": 1.0135208368301392, + "learning_rate": 1.3584638254075132e-06, + "loss": 0.5741, + "step": 4749 + }, + { + "epoch": 3.932633903920486, + "grad_norm": 0.992636501789093, + "learning_rate": 1.3574855810772414e-06, + "loss": 0.5592, + "step": 4750 + }, + { + "epoch": 3.933462175593595, + "grad_norm": 0.9497395157814026, + "learning_rate": 1.3565075578128603e-06, + "loss": 0.5579, + "step": 4751 + }, + { + "epoch": 3.9342904472667035, + "grad_norm": 0.9188456535339355, + "learning_rate": 1.355529755803607e-06, + "loss": 0.573, + "step": 4752 + }, + { + "epoch": 3.9351187189398122, + "grad_norm": 0.9138023257255554, + "learning_rate": 1.3545521752386765e-06, + "loss": 0.5895, + "step": 4753 + }, + { + "epoch": 3.935946990612921, + "grad_norm": 0.9659999012947083, + "learning_rate": 1.3535748163072233e-06, + "loss": 0.5506, + "step": 4754 + }, + { + "epoch": 3.9367752622860297, + "grad_norm": 0.9442874789237976, + "learning_rate": 1.3525976791983553e-06, + "loss": 0.5679, + "step": 4755 + }, + { + "epoch": 3.9376035339591384, + "grad_norm": 0.9444726705551147, + "learning_rate": 1.3516207641011407e-06, + "loss": 0.5795, + "step": 4756 + }, + { + "epoch": 3.9384318056322476, + "grad_norm": 0.9704516530036926, + "learning_rate": 1.3506440712046037e-06, + "loss": 0.568, + "step": 4757 + }, + { + "epoch": 3.9392600773053563, + "grad_norm": 0.9629958271980286, + "learning_rate": 1.349667600697724e-06, + "loss": 0.5967, + "step": 4758 + }, + { + "epoch": 3.940088348978465, + "grad_norm": 0.9573612213134766, + "learning_rate": 1.3486913527694406e-06, + "loss": 0.5553, + "step": 4759 + }, + { + "epoch": 3.9409166206515738, + "grad_norm": 0.9458691477775574, + "learning_rate": 1.3477153276086485e-06, + "loss": 0.5768, + "step": 4760 + }, + { + "epoch": 3.9417448923246825, + "grad_norm": 0.938666820526123, + "learning_rate": 1.3467395254041985e-06, + "loss": 0.5465, + "step": 4761 + }, + { + "epoch": 3.9425731639977912, + "grad_norm": 0.9396777153015137, + "learning_rate": 1.3457639463449008e-06, + "loss": 0.5699, + "step": 4762 + }, + { + "epoch": 3.9434014356709, + "grad_norm": 0.9707586765289307, + "learning_rate": 1.3447885906195186e-06, + "loss": 0.5472, + "step": 4763 + }, + { + "epoch": 3.9442297073440087, + "grad_norm": 0.9531437754631042, + "learning_rate": 1.343813458416776e-06, + "loss": 0.5712, + "step": 4764 + }, + { + "epoch": 3.9450579790171174, + "grad_norm": 0.942042350769043, + "learning_rate": 1.342838549925351e-06, + "loss": 0.5494, + "step": 4765 + }, + { + "epoch": 3.9458862506902266, + "grad_norm": 0.9602234959602356, + "learning_rate": 1.3418638653338795e-06, + "loss": 0.5627, + "step": 4766 + }, + { + "epoch": 3.9467145223633353, + "grad_norm": 0.9987154006958008, + "learning_rate": 1.340889404830955e-06, + "loss": 0.5824, + "step": 4767 + }, + { + "epoch": 3.947542794036444, + "grad_norm": 0.9674735069274902, + "learning_rate": 1.3399151686051246e-06, + "loss": 0.5772, + "step": 4768 + }, + { + "epoch": 3.9483710657095528, + "grad_norm": 0.9081838130950928, + "learning_rate": 1.3389411568448948e-06, + "loss": 0.555, + "step": 4769 + }, + { + "epoch": 3.9491993373826615, + "grad_norm": 0.9668666124343872, + "learning_rate": 1.337967369738729e-06, + "loss": 0.5655, + "step": 4770 + }, + { + "epoch": 3.95002760905577, + "grad_norm": 0.9260414838790894, + "learning_rate": 1.3369938074750438e-06, + "loss": 0.5685, + "step": 4771 + }, + { + "epoch": 3.950855880728879, + "grad_norm": 0.9354709386825562, + "learning_rate": 1.3360204702422154e-06, + "loss": 0.5623, + "step": 4772 + }, + { + "epoch": 3.951684152401988, + "grad_norm": 0.965032696723938, + "learning_rate": 1.3350473582285777e-06, + "loss": 0.5846, + "step": 4773 + }, + { + "epoch": 3.9525124240750964, + "grad_norm": 0.9616831541061401, + "learning_rate": 1.3340744716224143e-06, + "loss": 0.5576, + "step": 4774 + }, + { + "epoch": 3.9533406957482056, + "grad_norm": 0.9441544413566589, + "learning_rate": 1.3331018106119726e-06, + "loss": 0.5918, + "step": 4775 + }, + { + "epoch": 3.9541689674213143, + "grad_norm": 0.9979905486106873, + "learning_rate": 1.332129375385453e-06, + "loss": 0.5539, + "step": 4776 + }, + { + "epoch": 3.954997239094423, + "grad_norm": 0.950507402420044, + "learning_rate": 1.3311571661310124e-06, + "loss": 0.5463, + "step": 4777 + }, + { + "epoch": 3.9558255107675317, + "grad_norm": 0.9438148140907288, + "learning_rate": 1.3301851830367637e-06, + "loss": 0.5645, + "step": 4778 + }, + { + "epoch": 3.9566537824406405, + "grad_norm": 0.9625552296638489, + "learning_rate": 1.3292134262907785e-06, + "loss": 0.5782, + "step": 4779 + }, + { + "epoch": 3.957482054113749, + "grad_norm": 0.958839476108551, + "learning_rate": 1.3282418960810803e-06, + "loss": 0.562, + "step": 4780 + }, + { + "epoch": 3.958310325786858, + "grad_norm": 0.9578886032104492, + "learning_rate": 1.3272705925956522e-06, + "loss": 0.5609, + "step": 4781 + }, + { + "epoch": 3.959138597459967, + "grad_norm": 0.9848271608352661, + "learning_rate": 1.3262995160224335e-06, + "loss": 0.5627, + "step": 4782 + }, + { + "epoch": 3.9599668691330754, + "grad_norm": 0.9630492925643921, + "learning_rate": 1.3253286665493164e-06, + "loss": 0.5356, + "step": 4783 + }, + { + "epoch": 3.9607951408061846, + "grad_norm": 0.945763111114502, + "learning_rate": 1.3243580443641518e-06, + "loss": 0.5688, + "step": 4784 + }, + { + "epoch": 3.9616234124792933, + "grad_norm": 0.9506339430809021, + "learning_rate": 1.3233876496547477e-06, + "loss": 0.5725, + "step": 4785 + }, + { + "epoch": 3.962451684152402, + "grad_norm": 0.9367161393165588, + "learning_rate": 1.3224174826088654e-06, + "loss": 0.5605, + "step": 4786 + }, + { + "epoch": 3.9632799558255107, + "grad_norm": 0.9559187889099121, + "learning_rate": 1.3214475434142215e-06, + "loss": 0.5606, + "step": 4787 + }, + { + "epoch": 3.9641082274986195, + "grad_norm": 0.9559148550033569, + "learning_rate": 1.320477832258492e-06, + "loss": 0.5516, + "step": 4788 + }, + { + "epoch": 3.964936499171728, + "grad_norm": 0.9415640234947205, + "learning_rate": 1.319508349329307e-06, + "loss": 0.5607, + "step": 4789 + }, + { + "epoch": 3.965764770844837, + "grad_norm": 0.961673378944397, + "learning_rate": 1.3185390948142516e-06, + "loss": 0.571, + "step": 4790 + }, + { + "epoch": 3.966593042517946, + "grad_norm": 0.9346276521682739, + "learning_rate": 1.3175700689008678e-06, + "loss": 0.5716, + "step": 4791 + }, + { + "epoch": 3.967421314191055, + "grad_norm": 0.9483547806739807, + "learning_rate": 1.3166012717766541e-06, + "loss": 0.5654, + "step": 4792 + }, + { + "epoch": 3.9682495858641635, + "grad_norm": 0.9249748587608337, + "learning_rate": 1.3156327036290616e-06, + "loss": 0.5669, + "step": 4793 + }, + { + "epoch": 3.9690778575372723, + "grad_norm": 0.965886116027832, + "learning_rate": 1.3146643646455007e-06, + "loss": 0.5739, + "step": 4794 + }, + { + "epoch": 3.969906129210381, + "grad_norm": 0.9492125511169434, + "learning_rate": 1.3136962550133364e-06, + "loss": 0.5521, + "step": 4795 + }, + { + "epoch": 3.9707344008834897, + "grad_norm": 0.9473196864128113, + "learning_rate": 1.3127283749198885e-06, + "loss": 0.5412, + "step": 4796 + }, + { + "epoch": 3.9715626725565985, + "grad_norm": 0.9501288533210754, + "learning_rate": 1.3117607245524314e-06, + "loss": 0.5451, + "step": 4797 + }, + { + "epoch": 3.9723909442297076, + "grad_norm": 0.9439449310302734, + "learning_rate": 1.3107933040981985e-06, + "loss": 0.5702, + "step": 4798 + }, + { + "epoch": 3.973219215902816, + "grad_norm": 0.9604007005691528, + "learning_rate": 1.3098261137443744e-06, + "loss": 0.5674, + "step": 4799 + }, + { + "epoch": 3.974047487575925, + "grad_norm": 0.9559398889541626, + "learning_rate": 1.3088591536781026e-06, + "loss": 0.5618, + "step": 4800 + }, + { + "epoch": 3.974875759249034, + "grad_norm": 0.93898606300354, + "learning_rate": 1.3078924240864821e-06, + "loss": 0.5539, + "step": 4801 + }, + { + "epoch": 3.9757040309221425, + "grad_norm": 0.985381007194519, + "learning_rate": 1.306925925156564e-06, + "loss": 0.5748, + "step": 4802 + }, + { + "epoch": 3.9765323025952513, + "grad_norm": 0.9823545217514038, + "learning_rate": 1.3059596570753575e-06, + "loss": 0.569, + "step": 4803 + }, + { + "epoch": 3.97736057426836, + "grad_norm": 0.9789100289344788, + "learning_rate": 1.3049936200298274e-06, + "loss": 0.5784, + "step": 4804 + }, + { + "epoch": 3.9781888459414687, + "grad_norm": 0.9664195775985718, + "learning_rate": 1.3040278142068912e-06, + "loss": 0.574, + "step": 4805 + }, + { + "epoch": 3.9790171176145774, + "grad_norm": 0.9816479682922363, + "learning_rate": 1.3030622397934244e-06, + "loss": 0.607, + "step": 4806 + }, + { + "epoch": 3.9798453892876866, + "grad_norm": 0.9329001903533936, + "learning_rate": 1.3020968969762577e-06, + "loss": 0.5666, + "step": 4807 + }, + { + "epoch": 3.980673660960795, + "grad_norm": 0.9252684116363525, + "learning_rate": 1.3011317859421744e-06, + "loss": 0.5528, + "step": 4808 + }, + { + "epoch": 3.981501932633904, + "grad_norm": 0.9480351209640503, + "learning_rate": 1.3001669068779137e-06, + "loss": 0.5685, + "step": 4809 + }, + { + "epoch": 3.982330204307013, + "grad_norm": 0.9316738247871399, + "learning_rate": 1.2992022599701723e-06, + "loss": 0.569, + "step": 4810 + }, + { + "epoch": 3.9831584759801215, + "grad_norm": 0.9480258822441101, + "learning_rate": 1.2982378454056005e-06, + "loss": 0.5519, + "step": 4811 + }, + { + "epoch": 3.9839867476532302, + "grad_norm": 0.9391124248504639, + "learning_rate": 1.2972736633708021e-06, + "loss": 0.5657, + "step": 4812 + }, + { + "epoch": 3.984815019326339, + "grad_norm": 0.9289792776107788, + "learning_rate": 1.2963097140523389e-06, + "loss": 0.5258, + "step": 4813 + }, + { + "epoch": 3.9856432909994477, + "grad_norm": 0.9436705708503723, + "learning_rate": 1.295345997636725e-06, + "loss": 0.5672, + "step": 4814 + }, + { + "epoch": 3.9864715626725564, + "grad_norm": 0.9601346850395203, + "learning_rate": 1.2943825143104305e-06, + "loss": 0.5628, + "step": 4815 + }, + { + "epoch": 3.9872998343456656, + "grad_norm": 0.9829810261726379, + "learning_rate": 1.2934192642598818e-06, + "loss": 0.5772, + "step": 4816 + }, + { + "epoch": 3.9881281060187743, + "grad_norm": 0.9281584620475769, + "learning_rate": 1.292456247671457e-06, + "loss": 0.5763, + "step": 4817 + }, + { + "epoch": 3.988956377691883, + "grad_norm": 0.9712902903556824, + "learning_rate": 1.2914934647314914e-06, + "loss": 0.5553, + "step": 4818 + }, + { + "epoch": 3.9897846493649918, + "grad_norm": 0.9826544523239136, + "learning_rate": 1.290530915626276e-06, + "loss": 0.5802, + "step": 4819 + }, + { + "epoch": 3.9906129210381005, + "grad_norm": 0.9495535492897034, + "learning_rate": 1.289568600542054e-06, + "loss": 0.5746, + "step": 4820 + }, + { + "epoch": 3.9914411927112092, + "grad_norm": 0.9266762137413025, + "learning_rate": 1.2886065196650232e-06, + "loss": 0.5825, + "step": 4821 + }, + { + "epoch": 3.992269464384318, + "grad_norm": 0.946962833404541, + "learning_rate": 1.2876446731813387e-06, + "loss": 0.5825, + "step": 4822 + }, + { + "epoch": 3.9930977360574267, + "grad_norm": 0.9361686110496521, + "learning_rate": 1.2866830612771091e-06, + "loss": 0.5473, + "step": 4823 + }, + { + "epoch": 3.9939260077305354, + "grad_norm": 0.9517462849617004, + "learning_rate": 1.2857216841383958e-06, + "loss": 0.5628, + "step": 4824 + }, + { + "epoch": 3.9947542794036446, + "grad_norm": 0.957137405872345, + "learning_rate": 1.2847605419512174e-06, + "loss": 0.5516, + "step": 4825 + }, + { + "epoch": 3.9955825510767533, + "grad_norm": 0.9692016243934631, + "learning_rate": 1.2837996349015465e-06, + "loss": 0.5738, + "step": 4826 + }, + { + "epoch": 3.996410822749862, + "grad_norm": 0.9570938348770142, + "learning_rate": 1.2828389631753085e-06, + "loss": 0.5971, + "step": 4827 + }, + { + "epoch": 3.9972390944229708, + "grad_norm": 0.9742267727851868, + "learning_rate": 1.2818785269583845e-06, + "loss": 0.5721, + "step": 4828 + } + ], + "logging_steps": 1, + "max_steps": 7242, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 1207, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1.2190653098565304e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-4828/training_args.bin b/checkpoint-4828/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..7017eeafdf1f43b26668f0e0e5710ca52345a005 --- /dev/null +++ b/checkpoint-4828/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fdc878d2541c09353dd68f4f8a2464140b28fd84e1bc48b59914a6e84328e85e +size 8056 diff --git a/checkpoint-4828/zero_to_fp32.py b/checkpoint-4828/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-4828/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-6035/README.md b/checkpoint-6035/README.md new file mode 100644 index 0000000000000000000000000000000000000000..049d467664ca6172b7ffbe6ba60b3eac7479cac4 --- /dev/null +++ b/checkpoint-6035/README.md @@ -0,0 +1,202 @@ +--- +base_model: meta-llama/Llama-3.1-8B +library_name: peft +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.14.0 \ No newline at end of file diff --git a/checkpoint-6035/adapter_config.json b/checkpoint-6035/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d6fca8a067dce19d599e7f1e24bea8019968fe0f --- /dev/null +++ b/checkpoint-6035/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.1-8B", + "bias": "none", + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "up_proj", + "q_proj", + "k_proj", + "o_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-6035/adapter_model.safetensors b/checkpoint-6035/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..dbd45547d230b74d32a774632bea9c5a0023df93 --- /dev/null +++ b/checkpoint-6035/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9bc344ddd17ec532fee6d950983b711e4765390b028cfde45ed1d5eefed9d81e +size 3443586272 diff --git a/checkpoint-6035/global_step6032/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-6035/global_step6032/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..41db47cfb71198409b6009b2452359aef89d30ad --- /dev/null +++ b/checkpoint-6035/global_step6032/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7114a74b8d13b4e16922a7b9f70d7b7ac43feff8254915ac7d22b137289fb436 +size 20661195036 diff --git a/checkpoint-6035/global_step6032/mp_rank_00_model_states.pt b/checkpoint-6035/global_step6032/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..32bf7bb96a9eefc6d31e27112bb702c76093308d --- /dev/null +++ b/checkpoint-6035/global_step6032/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:43ba4ee3f1288342a21d3f781d802480fb59b78e886924290d7a15c22c70d357 +size 3555326841 diff --git a/checkpoint-6035/latest b/checkpoint-6035/latest new file mode 100644 index 0000000000000000000000000000000000000000..ce9ce2e6e714f6dfd8857a537375dea5f7df82f1 --- /dev/null +++ b/checkpoint-6035/latest @@ -0,0 +1 @@ +global_step6032 \ No newline at end of file diff --git a/checkpoint-6035/rng_state.pth b/checkpoint-6035/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..a004d1c83099e97bdaeea0bb059e4997fdd7451f --- /dev/null +++ b/checkpoint-6035/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:936bf69d58a1f3b831453509191ec91011520304e70290445d13b38ecd9e57e4 +size 14244 diff --git a/checkpoint-6035/scheduler.pt b/checkpoint-6035/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..593ed7f3b5102bdc38ffca152943e7fdcdc3c019 --- /dev/null +++ b/checkpoint-6035/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e46d49e1e68b490b0a0903b26b1a2f9c32bc56378b9c0aef3e19ecb3ef5d3bef +size 1064 diff --git a/checkpoint-6035/special_tokens_map.json b/checkpoint-6035/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..e5b39b6305d89284b04934011c68dbb26bf588ca --- /dev/null +++ b/checkpoint-6035/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-6035/tokenizer.json b/checkpoint-6035/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-6035/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-6035/tokenizer_config.json b/checkpoint-6035/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..81dd14db6632ad5b35b9d447732e37ac074873a5 --- /dev/null +++ b/checkpoint-6035/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +} diff --git a/checkpoint-6035/trainer_state.json b/checkpoint-6035/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..1f4586c87ce8bdcb4c869e5a502eccaea5db494c --- /dev/null +++ b/checkpoint-6035/trainer_state.json @@ -0,0 +1,42278 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 4.996410822749862, + "eval_steps": 500, + "global_step": 6035, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0008282716731087797, + "grad_norm": 2.4795358180999756, + "learning_rate": 5.0000000000000004e-08, + "loss": 1.1215, + "step": 1 + }, + { + "epoch": 0.0016565433462175593, + "grad_norm": 2.6060028076171875, + "learning_rate": 1.0000000000000001e-07, + "loss": 1.1219, + "step": 2 + }, + { + "epoch": 0.002484815019326339, + "grad_norm": 2.750868082046509, + "learning_rate": 1.5000000000000002e-07, + "loss": 1.125, + "step": 3 + }, + { + "epoch": 0.0033130866924351186, + "grad_norm": 2.7321369647979736, + "learning_rate": 2.0000000000000002e-07, + "loss": 1.1164, + "step": 4 + }, + { + "epoch": 0.0041413583655438985, + "grad_norm": 2.8193929195404053, + "learning_rate": 2.5000000000000004e-07, + "loss": 1.1263, + "step": 5 + }, + { + "epoch": 0.004969630038652678, + "grad_norm": 2.648301839828491, + "learning_rate": 3.0000000000000004e-07, + "loss": 1.1092, + "step": 6 + }, + { + "epoch": 0.005797901711761457, + "grad_norm": 2.751993179321289, + "learning_rate": 3.5000000000000004e-07, + "loss": 1.0997, + "step": 7 + }, + { + "epoch": 0.006626173384870237, + "grad_norm": 2.592381000518799, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.112, + "step": 8 + }, + { + "epoch": 0.007454445057979017, + "grad_norm": 2.3135671615600586, + "learning_rate": 4.5000000000000003e-07, + "loss": 1.1045, + "step": 9 + }, + { + "epoch": 0.008282716731087797, + "grad_norm": 2.60491943359375, + "learning_rate": 5.000000000000001e-07, + "loss": 1.0956, + "step": 10 + }, + { + "epoch": 0.009110988404196576, + "grad_norm": 2.2435271739959717, + "learning_rate": 5.5e-07, + "loss": 1.1241, + "step": 11 + }, + { + "epoch": 0.009939260077305357, + "grad_norm": 2.4412319660186768, + "learning_rate": 6.000000000000001e-07, + "loss": 1.11, + "step": 12 + }, + { + "epoch": 0.010767531750414136, + "grad_norm": 2.2075185775756836, + "learning_rate": 6.5e-07, + "loss": 1.0878, + "step": 13 + }, + { + "epoch": 0.011595803423522915, + "grad_norm": 2.101902723312378, + "learning_rate": 7.000000000000001e-07, + "loss": 1.0863, + "step": 14 + }, + { + "epoch": 0.012424075096631695, + "grad_norm": 2.134535074234009, + "learning_rate": 7.5e-07, + "loss": 1.0735, + "step": 15 + }, + { + "epoch": 0.013252346769740474, + "grad_norm": 2.000504493713379, + "learning_rate": 8.000000000000001e-07, + "loss": 1.0875, + "step": 16 + }, + { + "epoch": 0.014080618442849255, + "grad_norm": 1.9191882610321045, + "learning_rate": 8.500000000000001e-07, + "loss": 1.1039, + "step": 17 + }, + { + "epoch": 0.014908890115958034, + "grad_norm": 1.7732031345367432, + "learning_rate": 9.000000000000001e-07, + "loss": 1.0819, + "step": 18 + }, + { + "epoch": 0.015737161789066815, + "grad_norm": 1.7775251865386963, + "learning_rate": 9.500000000000001e-07, + "loss": 1.0936, + "step": 19 + }, + { + "epoch": 0.016565433462175594, + "grad_norm": 1.6648598909378052, + "learning_rate": 1.0000000000000002e-06, + "loss": 1.0881, + "step": 20 + }, + { + "epoch": 0.017393705135284373, + "grad_norm": 1.5726720094680786, + "learning_rate": 1.0500000000000001e-06, + "loss": 1.07, + "step": 21 + }, + { + "epoch": 0.018221976808393152, + "grad_norm": 1.561417818069458, + "learning_rate": 1.1e-06, + "loss": 1.0668, + "step": 22 + }, + { + "epoch": 0.019050248481501934, + "grad_norm": 1.4953842163085938, + "learning_rate": 1.1500000000000002e-06, + "loss": 1.054, + "step": 23 + }, + { + "epoch": 0.019878520154610713, + "grad_norm": 1.4247829914093018, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.0363, + "step": 24 + }, + { + "epoch": 0.020706791827719492, + "grad_norm": 1.3483426570892334, + "learning_rate": 1.25e-06, + "loss": 1.0143, + "step": 25 + }, + { + "epoch": 0.02153506350082827, + "grad_norm": 1.3412944078445435, + "learning_rate": 1.3e-06, + "loss": 1.025, + "step": 26 + }, + { + "epoch": 0.02236333517393705, + "grad_norm": 1.2834237813949585, + "learning_rate": 1.3500000000000002e-06, + "loss": 1.0227, + "step": 27 + }, + { + "epoch": 0.02319160684704583, + "grad_norm": 1.2128241062164307, + "learning_rate": 1.4000000000000001e-06, + "loss": 1.0161, + "step": 28 + }, + { + "epoch": 0.024019878520154612, + "grad_norm": 1.1665573120117188, + "learning_rate": 1.45e-06, + "loss": 1.0034, + "step": 29 + }, + { + "epoch": 0.02484815019326339, + "grad_norm": 1.1575899124145508, + "learning_rate": 1.5e-06, + "loss": 0.9968, + "step": 30 + }, + { + "epoch": 0.02567642186637217, + "grad_norm": 1.195304274559021, + "learning_rate": 1.5500000000000002e-06, + "loss": 1.0176, + "step": 31 + }, + { + "epoch": 0.02650469353948095, + "grad_norm": 1.1321678161621094, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.0113, + "step": 32 + }, + { + "epoch": 0.027332965212589728, + "grad_norm": 1.1076338291168213, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.9971, + "step": 33 + }, + { + "epoch": 0.02816123688569851, + "grad_norm": 1.0770610570907593, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.9994, + "step": 34 + }, + { + "epoch": 0.02898950855880729, + "grad_norm": 1.0767767429351807, + "learning_rate": 1.75e-06, + "loss": 0.978, + "step": 35 + }, + { + "epoch": 0.02981778023191607, + "grad_norm": 1.054800033569336, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.9829, + "step": 36 + }, + { + "epoch": 0.030646051905024847, + "grad_norm": 1.031015157699585, + "learning_rate": 1.85e-06, + "loss": 0.97, + "step": 37 + }, + { + "epoch": 0.03147432357813363, + "grad_norm": 1.0182578563690186, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.9549, + "step": 38 + }, + { + "epoch": 0.032302595251242405, + "grad_norm": 0.9796578288078308, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.9762, + "step": 39 + }, + { + "epoch": 0.03313086692435119, + "grad_norm": 0.9956578016281128, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.9693, + "step": 40 + }, + { + "epoch": 0.03395913859745997, + "grad_norm": 0.9383260011672974, + "learning_rate": 2.05e-06, + "loss": 0.9556, + "step": 41 + }, + { + "epoch": 0.034787410270568746, + "grad_norm": 0.9372819066047668, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.9419, + "step": 42 + }, + { + "epoch": 0.03561568194367753, + "grad_norm": 0.9334004521369934, + "learning_rate": 2.15e-06, + "loss": 0.9587, + "step": 43 + }, + { + "epoch": 0.036443953616786304, + "grad_norm": 0.8916004300117493, + "learning_rate": 2.2e-06, + "loss": 0.9227, + "step": 44 + }, + { + "epoch": 0.037272225289895086, + "grad_norm": 0.8698756694793701, + "learning_rate": 2.25e-06, + "loss": 0.9275, + "step": 45 + }, + { + "epoch": 0.03810049696300387, + "grad_norm": 0.8775100708007812, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.941, + "step": 46 + }, + { + "epoch": 0.038928768636112644, + "grad_norm": 0.8763384222984314, + "learning_rate": 2.35e-06, + "loss": 0.9253, + "step": 47 + }, + { + "epoch": 0.03975704030922143, + "grad_norm": 0.8534179925918579, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.9327, + "step": 48 + }, + { + "epoch": 0.0405853119823302, + "grad_norm": 0.8432146906852722, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.9236, + "step": 49 + }, + { + "epoch": 0.041413583655438985, + "grad_norm": 0.9899975061416626, + "learning_rate": 2.5e-06, + "loss": 0.8839, + "step": 50 + }, + { + "epoch": 0.04224185532854776, + "grad_norm": 0.8370104432106018, + "learning_rate": 2.55e-06, + "loss": 0.906, + "step": 51 + }, + { + "epoch": 0.04307012700165654, + "grad_norm": 0.8207446336746216, + "learning_rate": 2.6e-06, + "loss": 0.8755, + "step": 52 + }, + { + "epoch": 0.043898398674765325, + "grad_norm": 0.805584728717804, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.9049, + "step": 53 + }, + { + "epoch": 0.0447266703478741, + "grad_norm": 0.8162860870361328, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.8993, + "step": 54 + }, + { + "epoch": 0.04555494202098288, + "grad_norm": 0.8151668906211853, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.9282, + "step": 55 + }, + { + "epoch": 0.04638321369409166, + "grad_norm": 0.7996147274971008, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.8946, + "step": 56 + }, + { + "epoch": 0.04721148536720044, + "grad_norm": 0.8223204016685486, + "learning_rate": 2.85e-06, + "loss": 0.897, + "step": 57 + }, + { + "epoch": 0.048039757040309224, + "grad_norm": 0.8117988109588623, + "learning_rate": 2.9e-06, + "loss": 0.8845, + "step": 58 + }, + { + "epoch": 0.048868028713418, + "grad_norm": 0.7902786731719971, + "learning_rate": 2.95e-06, + "loss": 0.8846, + "step": 59 + }, + { + "epoch": 0.04969630038652678, + "grad_norm": 0.804192304611206, + "learning_rate": 3e-06, + "loss": 0.8722, + "step": 60 + }, + { + "epoch": 0.05052457205963556, + "grad_norm": 0.7852040529251099, + "learning_rate": 3.05e-06, + "loss": 0.8821, + "step": 61 + }, + { + "epoch": 0.05135284373274434, + "grad_norm": 0.8045080304145813, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.8661, + "step": 62 + }, + { + "epoch": 0.05218111540585312, + "grad_norm": 0.7920243144035339, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.8712, + "step": 63 + }, + { + "epoch": 0.0530093870789619, + "grad_norm": 0.7601844072341919, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.8792, + "step": 64 + }, + { + "epoch": 0.05383765875207068, + "grad_norm": 0.7833033800125122, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.8811, + "step": 65 + }, + { + "epoch": 0.054665930425179456, + "grad_norm": 0.7691664099693298, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.8701, + "step": 66 + }, + { + "epoch": 0.05549420209828824, + "grad_norm": 0.7845340967178345, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.8929, + "step": 67 + }, + { + "epoch": 0.05632247377139702, + "grad_norm": 0.9076205492019653, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.8352, + "step": 68 + }, + { + "epoch": 0.057150745444505796, + "grad_norm": 0.7614927887916565, + "learning_rate": 3.45e-06, + "loss": 0.8766, + "step": 69 + }, + { + "epoch": 0.05797901711761458, + "grad_norm": 0.7722934484481812, + "learning_rate": 3.5e-06, + "loss": 0.8666, + "step": 70 + }, + { + "epoch": 0.058807288790723354, + "grad_norm": 0.7603748440742493, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.84, + "step": 71 + }, + { + "epoch": 0.05963556046383214, + "grad_norm": 0.7563591003417969, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.8784, + "step": 72 + }, + { + "epoch": 0.06046383213694092, + "grad_norm": 0.7835114002227783, + "learning_rate": 3.65e-06, + "loss": 0.8516, + "step": 73 + }, + { + "epoch": 0.061292103810049695, + "grad_norm": 0.7760982513427734, + "learning_rate": 3.7e-06, + "loss": 0.8265, + "step": 74 + }, + { + "epoch": 0.06212037548315848, + "grad_norm": 0.7978456020355225, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.8564, + "step": 75 + }, + { + "epoch": 0.06294864715626726, + "grad_norm": 0.7940747141838074, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.8635, + "step": 76 + }, + { + "epoch": 0.06377691882937604, + "grad_norm": 0.8055519461631775, + "learning_rate": 3.85e-06, + "loss": 0.8445, + "step": 77 + }, + { + "epoch": 0.06460519050248481, + "grad_norm": 0.7532406449317932, + "learning_rate": 3.900000000000001e-06, + "loss": 0.8772, + "step": 78 + }, + { + "epoch": 0.06543346217559359, + "grad_norm": 0.763933539390564, + "learning_rate": 3.95e-06, + "loss": 0.8574, + "step": 79 + }, + { + "epoch": 0.06626173384870238, + "grad_norm": 0.7833206057548523, + "learning_rate": 4.000000000000001e-06, + "loss": 0.8643, + "step": 80 + }, + { + "epoch": 0.06709000552181116, + "grad_norm": 0.7825527787208557, + "learning_rate": 4.05e-06, + "loss": 0.8518, + "step": 81 + }, + { + "epoch": 0.06791827719491994, + "grad_norm": 0.7749581336975098, + "learning_rate": 4.1e-06, + "loss": 0.8146, + "step": 82 + }, + { + "epoch": 0.06874654886802871, + "grad_norm": 0.8117144703865051, + "learning_rate": 4.15e-06, + "loss": 0.8719, + "step": 83 + }, + { + "epoch": 0.06957482054113749, + "grad_norm": 0.8440507054328918, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.8665, + "step": 84 + }, + { + "epoch": 0.07040309221424627, + "grad_norm": 0.7941984534263611, + "learning_rate": 4.25e-06, + "loss": 0.8492, + "step": 85 + }, + { + "epoch": 0.07123136388735506, + "grad_norm": 0.8301989436149597, + "learning_rate": 4.3e-06, + "loss": 0.8544, + "step": 86 + }, + { + "epoch": 0.07205963556046384, + "grad_norm": 0.757798969745636, + "learning_rate": 4.350000000000001e-06, + "loss": 0.8426, + "step": 87 + }, + { + "epoch": 0.07288790723357261, + "grad_norm": 0.7817673683166504, + "learning_rate": 4.4e-06, + "loss": 0.8554, + "step": 88 + }, + { + "epoch": 0.07371617890668139, + "grad_norm": 0.7609818577766418, + "learning_rate": 4.450000000000001e-06, + "loss": 0.8473, + "step": 89 + }, + { + "epoch": 0.07454445057979017, + "grad_norm": 0.8146091103553772, + "learning_rate": 4.5e-06, + "loss": 0.8317, + "step": 90 + }, + { + "epoch": 0.07537272225289895, + "grad_norm": 0.8058239817619324, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.8411, + "step": 91 + }, + { + "epoch": 0.07620099392600774, + "grad_norm": 0.7700346112251282, + "learning_rate": 4.600000000000001e-06, + "loss": 0.815, + "step": 92 + }, + { + "epoch": 0.0770292655991165, + "grad_norm": 0.7949067950248718, + "learning_rate": 4.65e-06, + "loss": 0.8512, + "step": 93 + }, + { + "epoch": 0.07785753727222529, + "grad_norm": 0.8112176060676575, + "learning_rate": 4.7e-06, + "loss": 0.8398, + "step": 94 + }, + { + "epoch": 0.07868580894533407, + "grad_norm": 0.7990434765815735, + "learning_rate": 4.75e-06, + "loss": 0.8276, + "step": 95 + }, + { + "epoch": 0.07951408061844285, + "grad_norm": 0.7942920923233032, + "learning_rate": 4.800000000000001e-06, + "loss": 0.8333, + "step": 96 + }, + { + "epoch": 0.08034235229155164, + "grad_norm": 0.8005534410476685, + "learning_rate": 4.85e-06, + "loss": 0.8453, + "step": 97 + }, + { + "epoch": 0.0811706239646604, + "grad_norm": 0.8103216290473938, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8303, + "step": 98 + }, + { + "epoch": 0.08199889563776919, + "grad_norm": 0.8217238187789917, + "learning_rate": 4.95e-06, + "loss": 0.8204, + "step": 99 + }, + { + "epoch": 0.08282716731087797, + "grad_norm": 0.7958654761314392, + "learning_rate": 5e-06, + "loss": 0.8237, + "step": 100 + }, + { + "epoch": 0.08365543898398675, + "grad_norm": 0.8631463050842285, + "learning_rate": 4.999999758136652e-06, + "loss": 0.821, + "step": 101 + }, + { + "epoch": 0.08448371065709552, + "grad_norm": 0.7666735053062439, + "learning_rate": 4.999999032546657e-06, + "loss": 0.8307, + "step": 102 + }, + { + "epoch": 0.0853119823302043, + "grad_norm": 0.7709226012229919, + "learning_rate": 4.999997823230153e-06, + "loss": 0.8299, + "step": 103 + }, + { + "epoch": 0.08614025400331309, + "grad_norm": 0.8020651340484619, + "learning_rate": 4.9999961301873736e-06, + "loss": 0.8075, + "step": 104 + }, + { + "epoch": 0.08696852567642187, + "grad_norm": 0.8019537329673767, + "learning_rate": 4.999993953418649e-06, + "loss": 0.8615, + "step": 105 + }, + { + "epoch": 0.08779679734953065, + "grad_norm": 0.7904282808303833, + "learning_rate": 4.999991292924399e-06, + "loss": 0.8353, + "step": 106 + }, + { + "epoch": 0.08862506902263942, + "grad_norm": 0.7854570150375366, + "learning_rate": 4.999988148705138e-06, + "loss": 0.8154, + "step": 107 + }, + { + "epoch": 0.0894533406957482, + "grad_norm": 0.8044222593307495, + "learning_rate": 4.999984520761475e-06, + "loss": 0.8117, + "step": 108 + }, + { + "epoch": 0.09028161236885698, + "grad_norm": 0.7986987829208374, + "learning_rate": 4.999980409094112e-06, + "loss": 0.8407, + "step": 109 + }, + { + "epoch": 0.09110988404196577, + "grad_norm": 0.7976786494255066, + "learning_rate": 4.999975813703844e-06, + "loss": 0.7871, + "step": 110 + }, + { + "epoch": 0.09193815571507455, + "grad_norm": 0.8330315351486206, + "learning_rate": 4.9999707345915605e-06, + "loss": 0.8129, + "step": 111 + }, + { + "epoch": 0.09276642738818332, + "grad_norm": 0.8350826501846313, + "learning_rate": 4.999965171758245e-06, + "loss": 0.8144, + "step": 112 + }, + { + "epoch": 0.0935946990612921, + "grad_norm": 0.8054768443107605, + "learning_rate": 4.999959125204973e-06, + "loss": 0.809, + "step": 113 + }, + { + "epoch": 0.09442297073440088, + "grad_norm": 0.8101645708084106, + "learning_rate": 4.999952594932914e-06, + "loss": 0.8081, + "step": 114 + }, + { + "epoch": 0.09525124240750966, + "grad_norm": 0.905028760433197, + "learning_rate": 4.999945580943332e-06, + "loss": 0.8251, + "step": 115 + }, + { + "epoch": 0.09607951408061845, + "grad_norm": 0.8529418706893921, + "learning_rate": 4.999938083237585e-06, + "loss": 0.7984, + "step": 116 + }, + { + "epoch": 0.09690778575372722, + "grad_norm": 0.8047451376914978, + "learning_rate": 4.9999301018171225e-06, + "loss": 0.8222, + "step": 117 + }, + { + "epoch": 0.097736057426836, + "grad_norm": 0.8073400259017944, + "learning_rate": 4.99992163668349e-06, + "loss": 0.8251, + "step": 118 + }, + { + "epoch": 0.09856432909994478, + "grad_norm": 0.8470693826675415, + "learning_rate": 4.999912687838324e-06, + "loss": 0.8291, + "step": 119 + }, + { + "epoch": 0.09939260077305356, + "grad_norm": 0.8261650800704956, + "learning_rate": 4.999903255283357e-06, + "loss": 0.808, + "step": 120 + }, + { + "epoch": 0.10022087244616235, + "grad_norm": 0.8421704769134521, + "learning_rate": 4.999893339020414e-06, + "loss": 0.8295, + "step": 121 + }, + { + "epoch": 0.10104914411927111, + "grad_norm": 0.8244299292564392, + "learning_rate": 4.9998829390514134e-06, + "loss": 0.8215, + "step": 122 + }, + { + "epoch": 0.1018774157923799, + "grad_norm": 0.8365658521652222, + "learning_rate": 4.999872055378368e-06, + "loss": 0.8245, + "step": 123 + }, + { + "epoch": 0.10270568746548868, + "grad_norm": 0.9089561104774475, + "learning_rate": 4.999860688003384e-06, + "loss": 0.8125, + "step": 124 + }, + { + "epoch": 0.10353395913859746, + "grad_norm": 0.8784728646278381, + "learning_rate": 4.9998488369286595e-06, + "loss": 0.8321, + "step": 125 + }, + { + "epoch": 0.10436223081170624, + "grad_norm": 0.8596073389053345, + "learning_rate": 4.999836502156487e-06, + "loss": 0.8333, + "step": 126 + }, + { + "epoch": 0.10519050248481501, + "grad_norm": 0.8472688794136047, + "learning_rate": 4.9998236836892566e-06, + "loss": 0.8068, + "step": 127 + }, + { + "epoch": 0.1060187741579238, + "grad_norm": 0.8716503381729126, + "learning_rate": 4.999810381529447e-06, + "loss": 0.8005, + "step": 128 + }, + { + "epoch": 0.10684704583103258, + "grad_norm": 0.8416318893432617, + "learning_rate": 4.999796595679629e-06, + "loss": 0.7914, + "step": 129 + }, + { + "epoch": 0.10767531750414136, + "grad_norm": 0.838101863861084, + "learning_rate": 4.999782326142474e-06, + "loss": 0.7957, + "step": 130 + }, + { + "epoch": 0.10850358917725014, + "grad_norm": 0.8640785217285156, + "learning_rate": 4.999767572920741e-06, + "loss": 0.8077, + "step": 131 + }, + { + "epoch": 0.10933186085035891, + "grad_norm": 0.8207422494888306, + "learning_rate": 4.999752336017286e-06, + "loss": 0.7973, + "step": 132 + }, + { + "epoch": 0.1101601325234677, + "grad_norm": 0.8640758395195007, + "learning_rate": 4.9997366154350566e-06, + "loss": 0.8165, + "step": 133 + }, + { + "epoch": 0.11098840419657648, + "grad_norm": 0.7965807914733887, + "learning_rate": 4.999720411177094e-06, + "loss": 0.8059, + "step": 134 + }, + { + "epoch": 0.11181667586968526, + "grad_norm": 0.8463795185089111, + "learning_rate": 4.999703723246533e-06, + "loss": 0.8106, + "step": 135 + }, + { + "epoch": 0.11264494754279404, + "grad_norm": 0.8118059635162354, + "learning_rate": 4.999686551646604e-06, + "loss": 0.8099, + "step": 136 + }, + { + "epoch": 0.11347321921590281, + "grad_norm": 0.8256771564483643, + "learning_rate": 4.999668896380629e-06, + "loss": 0.8043, + "step": 137 + }, + { + "epoch": 0.11430149088901159, + "grad_norm": 0.8203126192092896, + "learning_rate": 4.999650757452025e-06, + "loss": 0.8029, + "step": 138 + }, + { + "epoch": 0.11512976256212037, + "grad_norm": 0.8442854881286621, + "learning_rate": 4.9996321348643e-06, + "loss": 0.8004, + "step": 139 + }, + { + "epoch": 0.11595803423522916, + "grad_norm": 0.8523190021514893, + "learning_rate": 4.999613028621059e-06, + "loss": 0.8092, + "step": 140 + }, + { + "epoch": 0.11678630590833794, + "grad_norm": 0.821428656578064, + "learning_rate": 4.999593438725997e-06, + "loss": 0.8105, + "step": 141 + }, + { + "epoch": 0.11761457758144671, + "grad_norm": 0.8406105041503906, + "learning_rate": 4.999573365182906e-06, + "loss": 0.7757, + "step": 142 + }, + { + "epoch": 0.11844284925455549, + "grad_norm": 0.8620600700378418, + "learning_rate": 4.999552807995669e-06, + "loss": 0.781, + "step": 143 + }, + { + "epoch": 0.11927112092766427, + "grad_norm": 0.920437216758728, + "learning_rate": 4.9995317671682655e-06, + "loss": 0.7579, + "step": 144 + }, + { + "epoch": 0.12009939260077306, + "grad_norm": 0.8414936661720276, + "learning_rate": 4.999510242704765e-06, + "loss": 0.8031, + "step": 145 + }, + { + "epoch": 0.12092766427388184, + "grad_norm": 0.8372176289558411, + "learning_rate": 4.999488234609332e-06, + "loss": 0.7931, + "step": 146 + }, + { + "epoch": 0.1217559359469906, + "grad_norm": 0.8785244226455688, + "learning_rate": 4.999465742886226e-06, + "loss": 0.8189, + "step": 147 + }, + { + "epoch": 0.12258420762009939, + "grad_norm": 0.8404353260993958, + "learning_rate": 4.999442767539799e-06, + "loss": 0.7899, + "step": 148 + }, + { + "epoch": 0.12341247929320817, + "grad_norm": 0.8541820049285889, + "learning_rate": 4.999419308574494e-06, + "loss": 0.7764, + "step": 149 + }, + { + "epoch": 0.12424075096631695, + "grad_norm": 0.8220177292823792, + "learning_rate": 4.999395365994854e-06, + "loss": 0.7942, + "step": 150 + }, + { + "epoch": 0.12506902263942574, + "grad_norm": 0.8823731541633606, + "learning_rate": 4.999370939805509e-06, + "loss": 0.793, + "step": 151 + }, + { + "epoch": 0.12589729431253452, + "grad_norm": 0.838860034942627, + "learning_rate": 4.999346030011186e-06, + "loss": 0.7741, + "step": 152 + }, + { + "epoch": 0.1267255659856433, + "grad_norm": 0.8162702322006226, + "learning_rate": 4.999320636616705e-06, + "loss": 0.7989, + "step": 153 + }, + { + "epoch": 0.12755383765875208, + "grad_norm": 0.8328465819358826, + "learning_rate": 4.99929475962698e-06, + "loss": 0.8055, + "step": 154 + }, + { + "epoch": 0.12838210933186084, + "grad_norm": 0.8373605608940125, + "learning_rate": 4.999268399047016e-06, + "loss": 0.801, + "step": 155 + }, + { + "epoch": 0.12921038100496962, + "grad_norm": 0.8408295512199402, + "learning_rate": 4.999241554881915e-06, + "loss": 0.7749, + "step": 156 + }, + { + "epoch": 0.1300386526780784, + "grad_norm": 0.8326342701911926, + "learning_rate": 4.999214227136871e-06, + "loss": 0.7818, + "step": 157 + }, + { + "epoch": 0.13086692435118719, + "grad_norm": 0.8064060211181641, + "learning_rate": 4.99918641581717e-06, + "loss": 0.7755, + "step": 158 + }, + { + "epoch": 0.13169519602429597, + "grad_norm": 0.8274610042572021, + "learning_rate": 4.999158120928196e-06, + "loss": 0.8008, + "step": 159 + }, + { + "epoch": 0.13252346769740475, + "grad_norm": 0.8121576309204102, + "learning_rate": 4.999129342475422e-06, + "loss": 0.7713, + "step": 160 + }, + { + "epoch": 0.13335173937051353, + "grad_norm": 0.8227710723876953, + "learning_rate": 4.9991000804644176e-06, + "loss": 0.7902, + "step": 161 + }, + { + "epoch": 0.13418001104362232, + "grad_norm": 0.8437026739120483, + "learning_rate": 4.999070334900843e-06, + "loss": 0.8176, + "step": 162 + }, + { + "epoch": 0.1350082827167311, + "grad_norm": 0.8379645943641663, + "learning_rate": 4.999040105790455e-06, + "loss": 0.7973, + "step": 163 + }, + { + "epoch": 0.13583655438983988, + "grad_norm": 0.822542130947113, + "learning_rate": 4.999009393139103e-06, + "loss": 0.7955, + "step": 164 + }, + { + "epoch": 0.13666482606294864, + "grad_norm": 0.8141928315162659, + "learning_rate": 4.9989781969527274e-06, + "loss": 0.7766, + "step": 165 + }, + { + "epoch": 0.13749309773605742, + "grad_norm": 0.8325410485267639, + "learning_rate": 4.998946517237367e-06, + "loss": 0.7737, + "step": 166 + }, + { + "epoch": 0.1383213694091662, + "grad_norm": 0.8399263024330139, + "learning_rate": 4.99891435399915e-06, + "loss": 0.7746, + "step": 167 + }, + { + "epoch": 0.13914964108227498, + "grad_norm": 0.8694303035736084, + "learning_rate": 4.9988817072443e-06, + "loss": 0.776, + "step": 168 + }, + { + "epoch": 0.13997791275538377, + "grad_norm": 0.8650498390197754, + "learning_rate": 4.998848576979135e-06, + "loss": 0.7941, + "step": 169 + }, + { + "epoch": 0.14080618442849255, + "grad_norm": 0.8246620893478394, + "learning_rate": 4.998814963210062e-06, + "loss": 0.7604, + "step": 170 + }, + { + "epoch": 0.14163445610160133, + "grad_norm": 0.8137409090995789, + "learning_rate": 4.99878086594359e-06, + "loss": 0.7681, + "step": 171 + }, + { + "epoch": 0.1424627277747101, + "grad_norm": 0.9755353927612305, + "learning_rate": 4.998746285186312e-06, + "loss": 0.7887, + "step": 172 + }, + { + "epoch": 0.1432909994478189, + "grad_norm": 0.8673408031463623, + "learning_rate": 4.998711220944921e-06, + "loss": 0.7803, + "step": 173 + }, + { + "epoch": 0.14411927112092768, + "grad_norm": 0.8312270045280457, + "learning_rate": 4.998675673226202e-06, + "loss": 0.7709, + "step": 174 + }, + { + "epoch": 0.14494754279403643, + "grad_norm": 0.8126094341278076, + "learning_rate": 4.998639642037032e-06, + "loss": 0.7822, + "step": 175 + }, + { + "epoch": 0.14577581446714521, + "grad_norm": 0.8561585545539856, + "learning_rate": 4.9986031273843834e-06, + "loss": 0.7829, + "step": 176 + }, + { + "epoch": 0.146604086140254, + "grad_norm": 0.8787612915039062, + "learning_rate": 4.9985661292753214e-06, + "loss": 0.7913, + "step": 177 + }, + { + "epoch": 0.14743235781336278, + "grad_norm": 0.8798345327377319, + "learning_rate": 4.998528647717006e-06, + "loss": 0.797, + "step": 178 + }, + { + "epoch": 0.14826062948647156, + "grad_norm": 0.7956589460372925, + "learning_rate": 4.998490682716687e-06, + "loss": 0.7571, + "step": 179 + }, + { + "epoch": 0.14908890115958034, + "grad_norm": 0.8505003452301025, + "learning_rate": 4.998452234281712e-06, + "loss": 0.7736, + "step": 180 + }, + { + "epoch": 0.14991717283268913, + "grad_norm": 0.8857074975967407, + "learning_rate": 4.998413302419519e-06, + "loss": 0.7863, + "step": 181 + }, + { + "epoch": 0.1507454445057979, + "grad_norm": 0.8762485980987549, + "learning_rate": 4.998373887137642e-06, + "loss": 0.7588, + "step": 182 + }, + { + "epoch": 0.1515737161789067, + "grad_norm": 0.8497205972671509, + "learning_rate": 4.998333988443709e-06, + "loss": 0.7773, + "step": 183 + }, + { + "epoch": 0.15240198785201547, + "grad_norm": 0.8720596432685852, + "learning_rate": 4.998293606345437e-06, + "loss": 0.7834, + "step": 184 + }, + { + "epoch": 0.15323025952512423, + "grad_norm": 0.8567108511924744, + "learning_rate": 4.998252740850641e-06, + "loss": 0.7794, + "step": 185 + }, + { + "epoch": 0.154058531198233, + "grad_norm": 0.8227219581604004, + "learning_rate": 4.998211391967228e-06, + "loss": 0.7688, + "step": 186 + }, + { + "epoch": 0.1548868028713418, + "grad_norm": 0.8321112394332886, + "learning_rate": 4.998169559703199e-06, + "loss": 0.7747, + "step": 187 + }, + { + "epoch": 0.15571507454445058, + "grad_norm": 0.8556115627288818, + "learning_rate": 4.9981272440666486e-06, + "loss": 0.7916, + "step": 188 + }, + { + "epoch": 0.15654334621755936, + "grad_norm": 0.8473951816558838, + "learning_rate": 4.998084445065763e-06, + "loss": 0.7823, + "step": 189 + }, + { + "epoch": 0.15737161789066814, + "grad_norm": 0.8551077246665955, + "learning_rate": 4.998041162708823e-06, + "loss": 0.7816, + "step": 190 + }, + { + "epoch": 0.15819988956377692, + "grad_norm": 0.945099413394928, + "learning_rate": 4.997997397004206e-06, + "loss": 0.7828, + "step": 191 + }, + { + "epoch": 0.1590281612368857, + "grad_norm": 0.8747237920761108, + "learning_rate": 4.9979531479603785e-06, + "loss": 0.7559, + "step": 192 + }, + { + "epoch": 0.1598564329099945, + "grad_norm": 0.8479958176612854, + "learning_rate": 4.997908415585902e-06, + "loss": 0.7694, + "step": 193 + }, + { + "epoch": 0.16068470458310327, + "grad_norm": 0.8659047484397888, + "learning_rate": 4.997863199889433e-06, + "loss": 0.7534, + "step": 194 + }, + { + "epoch": 0.16151297625621203, + "grad_norm": 0.8744727969169617, + "learning_rate": 4.997817500879719e-06, + "loss": 0.7499, + "step": 195 + }, + { + "epoch": 0.1623412479293208, + "grad_norm": 0.896468997001648, + "learning_rate": 4.997771318565603e-06, + "loss": 0.7974, + "step": 196 + }, + { + "epoch": 0.1631695196024296, + "grad_norm": 0.9058924913406372, + "learning_rate": 4.997724652956021e-06, + "loss": 0.7753, + "step": 197 + }, + { + "epoch": 0.16399779127553837, + "grad_norm": 0.8904663920402527, + "learning_rate": 4.9976775040600014e-06, + "loss": 0.7569, + "step": 198 + }, + { + "epoch": 0.16482606294864716, + "grad_norm": 0.8709238767623901, + "learning_rate": 4.997629871886669e-06, + "loss": 0.763, + "step": 199 + }, + { + "epoch": 0.16565433462175594, + "grad_norm": 0.8400478959083557, + "learning_rate": 4.997581756445238e-06, + "loss": 0.7795, + "step": 200 + }, + { + "epoch": 0.16648260629486472, + "grad_norm": 0.8458966612815857, + "learning_rate": 4.9975331577450204e-06, + "loss": 0.7597, + "step": 201 + }, + { + "epoch": 0.1673108779679735, + "grad_norm": 0.8707828521728516, + "learning_rate": 4.997484075795417e-06, + "loss": 0.759, + "step": 202 + }, + { + "epoch": 0.16813914964108229, + "grad_norm": 0.8774807453155518, + "learning_rate": 4.997434510605927e-06, + "loss": 0.7706, + "step": 203 + }, + { + "epoch": 0.16896742131419104, + "grad_norm": 0.8987929821014404, + "learning_rate": 4.99738446218614e-06, + "loss": 0.7319, + "step": 204 + }, + { + "epoch": 0.16979569298729982, + "grad_norm": 0.8869186639785767, + "learning_rate": 4.99733393054574e-06, + "loss": 0.7812, + "step": 205 + }, + { + "epoch": 0.1706239646604086, + "grad_norm": 0.860407829284668, + "learning_rate": 4.997282915694505e-06, + "loss": 0.7524, + "step": 206 + }, + { + "epoch": 0.1714522363335174, + "grad_norm": 0.8719788789749146, + "learning_rate": 4.9972314176423035e-06, + "loss": 0.7796, + "step": 207 + }, + { + "epoch": 0.17228050800662617, + "grad_norm": 0.8596137762069702, + "learning_rate": 4.997179436399103e-06, + "loss": 0.7554, + "step": 208 + }, + { + "epoch": 0.17310877967973495, + "grad_norm": 0.8505175709724426, + "learning_rate": 4.997126971974959e-06, + "loss": 0.7607, + "step": 209 + }, + { + "epoch": 0.17393705135284374, + "grad_norm": 0.7858585119247437, + "learning_rate": 4.9970740243800245e-06, + "loss": 0.7622, + "step": 210 + }, + { + "epoch": 0.17476532302595252, + "grad_norm": 0.8634621500968933, + "learning_rate": 4.997020593624543e-06, + "loss": 0.7659, + "step": 211 + }, + { + "epoch": 0.1755935946990613, + "grad_norm": 0.8650951385498047, + "learning_rate": 4.9969666797188545e-06, + "loss": 0.7689, + "step": 212 + }, + { + "epoch": 0.17642186637217008, + "grad_norm": 0.8602325916290283, + "learning_rate": 4.996912282673389e-06, + "loss": 0.7695, + "step": 213 + }, + { + "epoch": 0.17725013804527884, + "grad_norm": 0.8453742265701294, + "learning_rate": 4.9968574024986735e-06, + "loss": 0.7434, + "step": 214 + }, + { + "epoch": 0.17807840971838762, + "grad_norm": 0.8542691469192505, + "learning_rate": 4.9968020392053255e-06, + "loss": 0.7298, + "step": 215 + }, + { + "epoch": 0.1789066813914964, + "grad_norm": 0.8806673288345337, + "learning_rate": 4.996746192804058e-06, + "loss": 0.7649, + "step": 216 + }, + { + "epoch": 0.17973495306460519, + "grad_norm": 0.8451183438301086, + "learning_rate": 4.9966898633056765e-06, + "loss": 0.7681, + "step": 217 + }, + { + "epoch": 0.18056322473771397, + "grad_norm": 0.8664626479148865, + "learning_rate": 4.99663305072108e-06, + "loss": 0.7753, + "step": 218 + }, + { + "epoch": 0.18139149641082275, + "grad_norm": 0.8828144669532776, + "learning_rate": 4.996575755061262e-06, + "loss": 0.7514, + "step": 219 + }, + { + "epoch": 0.18221976808393153, + "grad_norm": 0.8493948578834534, + "learning_rate": 4.996517976337308e-06, + "loss": 0.7684, + "step": 220 + }, + { + "epoch": 0.18304803975704032, + "grad_norm": 0.8561757802963257, + "learning_rate": 4.9964597145603975e-06, + "loss": 0.7723, + "step": 221 + }, + { + "epoch": 0.1838763114301491, + "grad_norm": 0.8863773941993713, + "learning_rate": 4.996400969741805e-06, + "loss": 0.7732, + "step": 222 + }, + { + "epoch": 0.18470458310325788, + "grad_norm": 0.8571644425392151, + "learning_rate": 4.996341741892895e-06, + "loss": 0.744, + "step": 223 + }, + { + "epoch": 0.18553285477636663, + "grad_norm": 0.8622035980224609, + "learning_rate": 4.9962820310251286e-06, + "loss": 0.7607, + "step": 224 + }, + { + "epoch": 0.18636112644947542, + "grad_norm": 0.8904410600662231, + "learning_rate": 4.9962218371500595e-06, + "loss": 0.7648, + "step": 225 + }, + { + "epoch": 0.1871893981225842, + "grad_norm": 0.8975846171379089, + "learning_rate": 4.996161160279335e-06, + "loss": 0.7469, + "step": 226 + }, + { + "epoch": 0.18801766979569298, + "grad_norm": 0.8788724541664124, + "learning_rate": 4.996100000424694e-06, + "loss": 0.7473, + "step": 227 + }, + { + "epoch": 0.18884594146880176, + "grad_norm": 0.9446991086006165, + "learning_rate": 4.996038357597971e-06, + "loss": 0.7635, + "step": 228 + }, + { + "epoch": 0.18967421314191055, + "grad_norm": 0.86910480260849, + "learning_rate": 4.995976231811094e-06, + "loss": 0.7539, + "step": 229 + }, + { + "epoch": 0.19050248481501933, + "grad_norm": 0.862197756767273, + "learning_rate": 4.995913623076084e-06, + "loss": 0.7546, + "step": 230 + }, + { + "epoch": 0.1913307564881281, + "grad_norm": 0.906878650188446, + "learning_rate": 4.995850531405054e-06, + "loss": 0.7737, + "step": 231 + }, + { + "epoch": 0.1921590281612369, + "grad_norm": 0.8557183742523193, + "learning_rate": 4.995786956810212e-06, + "loss": 0.7663, + "step": 232 + }, + { + "epoch": 0.19298729983434568, + "grad_norm": 0.8125383853912354, + "learning_rate": 4.995722899303859e-06, + "loss": 0.7647, + "step": 233 + }, + { + "epoch": 0.19381557150745443, + "grad_norm": 0.8610332012176514, + "learning_rate": 4.995658358898391e-06, + "loss": 0.7541, + "step": 234 + }, + { + "epoch": 0.19464384318056321, + "grad_norm": 0.8620829582214355, + "learning_rate": 4.995593335606294e-06, + "loss": 0.7625, + "step": 235 + }, + { + "epoch": 0.195472114853672, + "grad_norm": 0.8296957612037659, + "learning_rate": 4.9955278294401496e-06, + "loss": 0.7536, + "step": 236 + }, + { + "epoch": 0.19630038652678078, + "grad_norm": 0.8974814414978027, + "learning_rate": 4.995461840412634e-06, + "loss": 0.7651, + "step": 237 + }, + { + "epoch": 0.19712865819988956, + "grad_norm": 0.8701095581054688, + "learning_rate": 4.995395368536515e-06, + "loss": 0.788, + "step": 238 + }, + { + "epoch": 0.19795692987299834, + "grad_norm": 0.8556636571884155, + "learning_rate": 4.995328413824653e-06, + "loss": 0.7649, + "step": 239 + }, + { + "epoch": 0.19878520154610713, + "grad_norm": 0.836686909198761, + "learning_rate": 4.995260976290005e-06, + "loss": 0.7583, + "step": 240 + }, + { + "epoch": 0.1996134732192159, + "grad_norm": 0.8807066082954407, + "learning_rate": 4.995193055945618e-06, + "loss": 0.7766, + "step": 241 + }, + { + "epoch": 0.2004417448923247, + "grad_norm": 0.8363582491874695, + "learning_rate": 4.995124652804635e-06, + "loss": 0.7655, + "step": 242 + }, + { + "epoch": 0.20127001656543347, + "grad_norm": 0.8699461817741394, + "learning_rate": 4.9950557668802905e-06, + "loss": 0.743, + "step": 243 + }, + { + "epoch": 0.20209828823854223, + "grad_norm": 0.8571060299873352, + "learning_rate": 4.994986398185914e-06, + "loss": 0.7512, + "step": 244 + }, + { + "epoch": 0.202926559911651, + "grad_norm": 0.8429675102233887, + "learning_rate": 4.994916546734927e-06, + "loss": 0.7589, + "step": 245 + }, + { + "epoch": 0.2037548315847598, + "grad_norm": 0.874822199344635, + "learning_rate": 4.994846212540846e-06, + "loss": 0.7534, + "step": 246 + }, + { + "epoch": 0.20458310325786858, + "grad_norm": 0.878341555595398, + "learning_rate": 4.994775395617279e-06, + "loss": 0.7709, + "step": 247 + }, + { + "epoch": 0.20541137493097736, + "grad_norm": 0.897961437702179, + "learning_rate": 4.99470409597793e-06, + "loss": 0.7768, + "step": 248 + }, + { + "epoch": 0.20623964660408614, + "grad_norm": 0.8439584374427795, + "learning_rate": 4.994632313636593e-06, + "loss": 0.7558, + "step": 249 + }, + { + "epoch": 0.20706791827719492, + "grad_norm": 0.8394468426704407, + "learning_rate": 4.994560048607159e-06, + "loss": 0.7492, + "step": 250 + }, + { + "epoch": 0.2078961899503037, + "grad_norm": 0.8489627242088318, + "learning_rate": 4.9944873009036074e-06, + "loss": 0.7365, + "step": 251 + }, + { + "epoch": 0.2087244616234125, + "grad_norm": 0.8675844669342041, + "learning_rate": 4.994414070540018e-06, + "loss": 0.762, + "step": 252 + }, + { + "epoch": 0.20955273329652127, + "grad_norm": 0.8228297829627991, + "learning_rate": 4.994340357530558e-06, + "loss": 0.7475, + "step": 253 + }, + { + "epoch": 0.21038100496963003, + "grad_norm": 0.8794129490852356, + "learning_rate": 4.9942661618894915e-06, + "loss": 0.7631, + "step": 254 + }, + { + "epoch": 0.2112092766427388, + "grad_norm": 0.8616135716438293, + "learning_rate": 4.9941914836311735e-06, + "loss": 0.7749, + "step": 255 + }, + { + "epoch": 0.2120375483158476, + "grad_norm": 0.8470495939254761, + "learning_rate": 4.994116322770054e-06, + "loss": 0.7484, + "step": 256 + }, + { + "epoch": 0.21286581998895637, + "grad_norm": 0.9346716403961182, + "learning_rate": 4.9940406793206755e-06, + "loss": 0.7557, + "step": 257 + }, + { + "epoch": 0.21369409166206516, + "grad_norm": 0.8993160724639893, + "learning_rate": 4.993964553297676e-06, + "loss": 0.7422, + "step": 258 + }, + { + "epoch": 0.21452236333517394, + "grad_norm": 0.8257086277008057, + "learning_rate": 4.993887944715783e-06, + "loss": 0.7783, + "step": 259 + }, + { + "epoch": 0.21535063500828272, + "grad_norm": 0.8950703144073486, + "learning_rate": 4.993810853589819e-06, + "loss": 0.7311, + "step": 260 + }, + { + "epoch": 0.2161789066813915, + "grad_norm": 0.8733702898025513, + "learning_rate": 4.993733279934704e-06, + "loss": 0.7572, + "step": 261 + }, + { + "epoch": 0.21700717835450029, + "grad_norm": 0.8322939276695251, + "learning_rate": 4.993655223765444e-06, + "loss": 0.7632, + "step": 262 + }, + { + "epoch": 0.21783545002760907, + "grad_norm": 0.8666137456893921, + "learning_rate": 4.993576685097145e-06, + "loss": 0.7619, + "step": 263 + }, + { + "epoch": 0.21866372170071782, + "grad_norm": 0.8770918250083923, + "learning_rate": 4.993497663945002e-06, + "loss": 0.7538, + "step": 264 + }, + { + "epoch": 0.2194919933738266, + "grad_norm": 0.8940248489379883, + "learning_rate": 4.9934181603243045e-06, + "loss": 0.7512, + "step": 265 + }, + { + "epoch": 0.2203202650469354, + "grad_norm": 0.8979714512825012, + "learning_rate": 4.993338174250437e-06, + "loss": 0.7504, + "step": 266 + }, + { + "epoch": 0.22114853672004417, + "grad_norm": 0.8802001476287842, + "learning_rate": 4.993257705738876e-06, + "loss": 0.7501, + "step": 267 + }, + { + "epoch": 0.22197680839315295, + "grad_norm": 0.8734432458877563, + "learning_rate": 4.993176754805189e-06, + "loss": 0.7452, + "step": 268 + }, + { + "epoch": 0.22280508006626173, + "grad_norm": 0.9127711057662964, + "learning_rate": 4.993095321465042e-06, + "loss": 0.7404, + "step": 269 + }, + { + "epoch": 0.22363335173937052, + "grad_norm": 0.8533952832221985, + "learning_rate": 4.993013405734191e-06, + "loss": 0.7382, + "step": 270 + }, + { + "epoch": 0.2244616234124793, + "grad_norm": 0.8463877439498901, + "learning_rate": 4.9929310076284845e-06, + "loss": 0.7574, + "step": 271 + }, + { + "epoch": 0.22528989508558808, + "grad_norm": 0.8598008155822754, + "learning_rate": 4.992848127163868e-06, + "loss": 0.7624, + "step": 272 + }, + { + "epoch": 0.22611816675869686, + "grad_norm": 0.8759154081344604, + "learning_rate": 4.992764764356376e-06, + "loss": 0.7637, + "step": 273 + }, + { + "epoch": 0.22694643843180562, + "grad_norm": 0.8880444765090942, + "learning_rate": 4.992680919222141e-06, + "loss": 0.7514, + "step": 274 + }, + { + "epoch": 0.2277747101049144, + "grad_norm": 0.871366024017334, + "learning_rate": 4.9925965917773826e-06, + "loss": 0.7569, + "step": 275 + }, + { + "epoch": 0.22860298177802318, + "grad_norm": 0.862654983997345, + "learning_rate": 4.99251178203842e-06, + "loss": 0.7349, + "step": 276 + }, + { + "epoch": 0.22943125345113197, + "grad_norm": 0.8615298867225647, + "learning_rate": 4.992426490021662e-06, + "loss": 0.7438, + "step": 277 + }, + { + "epoch": 0.23025952512424075, + "grad_norm": 0.8735869526863098, + "learning_rate": 4.992340715743614e-06, + "loss": 0.7755, + "step": 278 + }, + { + "epoch": 0.23108779679734953, + "grad_norm": 0.9082722067832947, + "learning_rate": 4.9922544592208695e-06, + "loss": 0.7561, + "step": 279 + }, + { + "epoch": 0.23191606847045831, + "grad_norm": 0.8639047145843506, + "learning_rate": 4.992167720470119e-06, + "loss": 0.7493, + "step": 280 + }, + { + "epoch": 0.2327443401435671, + "grad_norm": 0.891640305519104, + "learning_rate": 4.9920804995081475e-06, + "loss": 0.75, + "step": 281 + }, + { + "epoch": 0.23357261181667588, + "grad_norm": 0.8880137205123901, + "learning_rate": 4.99199279635183e-06, + "loss": 0.7801, + "step": 282 + }, + { + "epoch": 0.23440088348978466, + "grad_norm": 0.8600473403930664, + "learning_rate": 4.991904611018137e-06, + "loss": 0.7477, + "step": 283 + }, + { + "epoch": 0.23522915516289342, + "grad_norm": 0.8685399889945984, + "learning_rate": 4.99181594352413e-06, + "loss": 0.741, + "step": 284 + }, + { + "epoch": 0.2360574268360022, + "grad_norm": 0.9006181955337524, + "learning_rate": 4.991726793886966e-06, + "loss": 0.7288, + "step": 285 + }, + { + "epoch": 0.23688569850911098, + "grad_norm": 0.863895833492279, + "learning_rate": 4.991637162123896e-06, + "loss": 0.7375, + "step": 286 + }, + { + "epoch": 0.23771397018221976, + "grad_norm": 0.8620381951332092, + "learning_rate": 4.9915470482522625e-06, + "loss": 0.7249, + "step": 287 + }, + { + "epoch": 0.23854224185532855, + "grad_norm": 0.9634964466094971, + "learning_rate": 4.991456452289499e-06, + "loss": 0.7313, + "step": 288 + }, + { + "epoch": 0.23937051352843733, + "grad_norm": 0.9154514670372009, + "learning_rate": 4.9913653742531385e-06, + "loss": 0.782, + "step": 289 + }, + { + "epoch": 0.2401987852015461, + "grad_norm": 0.9179158806800842, + "learning_rate": 4.991273814160802e-06, + "loss": 0.7388, + "step": 290 + }, + { + "epoch": 0.2410270568746549, + "grad_norm": 0.8609046936035156, + "learning_rate": 4.991181772030206e-06, + "loss": 0.7393, + "step": 291 + }, + { + "epoch": 0.24185532854776368, + "grad_norm": 0.8689826130867004, + "learning_rate": 4.991089247879159e-06, + "loss": 0.7302, + "step": 292 + }, + { + "epoch": 0.24268360022087246, + "grad_norm": 0.8414047956466675, + "learning_rate": 4.990996241725565e-06, + "loss": 0.7496, + "step": 293 + }, + { + "epoch": 0.2435118718939812, + "grad_norm": 0.8621289134025574, + "learning_rate": 4.990902753587418e-06, + "loss": 0.7317, + "step": 294 + }, + { + "epoch": 0.24434014356709, + "grad_norm": 0.856566309928894, + "learning_rate": 4.990808783482809e-06, + "loss": 0.7169, + "step": 295 + }, + { + "epoch": 0.24516841524019878, + "grad_norm": 0.8549672961235046, + "learning_rate": 4.990714331429919e-06, + "loss": 0.7357, + "step": 296 + }, + { + "epoch": 0.24599668691330756, + "grad_norm": 0.8813024759292603, + "learning_rate": 4.990619397447024e-06, + "loss": 0.7437, + "step": 297 + }, + { + "epoch": 0.24682495858641634, + "grad_norm": 0.9008268713951111, + "learning_rate": 4.990523981552493e-06, + "loss": 0.7652, + "step": 298 + }, + { + "epoch": 0.24765323025952513, + "grad_norm": 0.910277783870697, + "learning_rate": 4.990428083764788e-06, + "loss": 0.7598, + "step": 299 + }, + { + "epoch": 0.2484815019326339, + "grad_norm": 0.9261599183082581, + "learning_rate": 4.990331704102464e-06, + "loss": 0.7199, + "step": 300 + }, + { + "epoch": 0.2493097736057427, + "grad_norm": 0.9166674613952637, + "learning_rate": 4.99023484258417e-06, + "loss": 0.7414, + "step": 301 + }, + { + "epoch": 0.2501380452788515, + "grad_norm": 0.8701214790344238, + "learning_rate": 4.990137499228648e-06, + "loss": 0.7326, + "step": 302 + }, + { + "epoch": 0.25096631695196026, + "grad_norm": 0.9120466113090515, + "learning_rate": 4.990039674054732e-06, + "loss": 0.7574, + "step": 303 + }, + { + "epoch": 0.25179458862506904, + "grad_norm": 0.9449286460876465, + "learning_rate": 4.989941367081351e-06, + "loss": 0.725, + "step": 304 + }, + { + "epoch": 0.2526228602981778, + "grad_norm": 0.8989110589027405, + "learning_rate": 4.989842578327526e-06, + "loss": 0.7569, + "step": 305 + }, + { + "epoch": 0.2534511319712866, + "grad_norm": 0.891677975654602, + "learning_rate": 4.989743307812373e-06, + "loss": 0.7135, + "step": 306 + }, + { + "epoch": 0.2542794036443954, + "grad_norm": 0.8689535856246948, + "learning_rate": 4.989643555555099e-06, + "loss": 0.7439, + "step": 307 + }, + { + "epoch": 0.25510767531750417, + "grad_norm": 0.8779630064964294, + "learning_rate": 4.989543321575004e-06, + "loss": 0.7322, + "step": 308 + }, + { + "epoch": 0.2559359469906129, + "grad_norm": 0.8977622985839844, + "learning_rate": 4.9894426058914856e-06, + "loss": 0.7544, + "step": 309 + }, + { + "epoch": 0.2567642186637217, + "grad_norm": 0.8216757774353027, + "learning_rate": 4.989341408524028e-06, + "loss": 0.7344, + "step": 310 + }, + { + "epoch": 0.25759249033683046, + "grad_norm": 0.8494759798049927, + "learning_rate": 4.9892397294922126e-06, + "loss": 0.7389, + "step": 311 + }, + { + "epoch": 0.25842076200993924, + "grad_norm": 0.881244957447052, + "learning_rate": 4.989137568815715e-06, + "loss": 0.7121, + "step": 312 + }, + { + "epoch": 0.259249033683048, + "grad_norm": 0.8679410219192505, + "learning_rate": 4.9890349265143005e-06, + "loss": 0.7469, + "step": 313 + }, + { + "epoch": 0.2600773053561568, + "grad_norm": 0.9746971130371094, + "learning_rate": 4.988931802607831e-06, + "loss": 0.7585, + "step": 314 + }, + { + "epoch": 0.2609055770292656, + "grad_norm": 0.9222471117973328, + "learning_rate": 4.988828197116259e-06, + "loss": 0.761, + "step": 315 + }, + { + "epoch": 0.26173384870237437, + "grad_norm": 0.9104124903678894, + "learning_rate": 4.988724110059631e-06, + "loss": 0.7652, + "step": 316 + }, + { + "epoch": 0.26256212037548315, + "grad_norm": 0.8876475095748901, + "learning_rate": 4.988619541458088e-06, + "loss": 0.7529, + "step": 317 + }, + { + "epoch": 0.26339039204859194, + "grad_norm": 0.8705421686172485, + "learning_rate": 4.988514491331861e-06, + "loss": 0.7437, + "step": 318 + }, + { + "epoch": 0.2642186637217007, + "grad_norm": 0.8594399690628052, + "learning_rate": 4.9884089597012784e-06, + "loss": 0.7406, + "step": 319 + }, + { + "epoch": 0.2650469353948095, + "grad_norm": 0.8807039856910706, + "learning_rate": 4.988302946586759e-06, + "loss": 0.7429, + "step": 320 + }, + { + "epoch": 0.2658752070679183, + "grad_norm": 0.8881412148475647, + "learning_rate": 4.9881964520088144e-06, + "loss": 0.7234, + "step": 321 + }, + { + "epoch": 0.26670347874102707, + "grad_norm": 0.8764941096305847, + "learning_rate": 4.988089475988052e-06, + "loss": 0.7514, + "step": 322 + }, + { + "epoch": 0.26753175041413585, + "grad_norm": 0.8992927670478821, + "learning_rate": 4.987982018545169e-06, + "loss": 0.7143, + "step": 323 + }, + { + "epoch": 0.26836002208724463, + "grad_norm": 0.8961546421051025, + "learning_rate": 4.9878740797009585e-06, + "loss": 0.7396, + "step": 324 + }, + { + "epoch": 0.2691882937603534, + "grad_norm": 0.9031578302383423, + "learning_rate": 4.987765659476304e-06, + "loss": 0.7407, + "step": 325 + }, + { + "epoch": 0.2700165654334622, + "grad_norm": 0.8693655729293823, + "learning_rate": 4.987656757892186e-06, + "loss": 0.7215, + "step": 326 + }, + { + "epoch": 0.270844837106571, + "grad_norm": 0.8739856481552124, + "learning_rate": 4.987547374969676e-06, + "loss": 0.7471, + "step": 327 + }, + { + "epoch": 0.27167310877967976, + "grad_norm": 0.8916322588920593, + "learning_rate": 4.987437510729936e-06, + "loss": 0.7459, + "step": 328 + }, + { + "epoch": 0.2725013804527885, + "grad_norm": 0.890178382396698, + "learning_rate": 4.987327165194225e-06, + "loss": 0.7322, + "step": 329 + }, + { + "epoch": 0.27332965212589727, + "grad_norm": 0.8611598014831543, + "learning_rate": 4.987216338383896e-06, + "loss": 0.733, + "step": 330 + }, + { + "epoch": 0.27415792379900605, + "grad_norm": 0.87791907787323, + "learning_rate": 4.98710503032039e-06, + "loss": 0.7445, + "step": 331 + }, + { + "epoch": 0.27498619547211484, + "grad_norm": 0.963479220867157, + "learning_rate": 4.986993241025245e-06, + "loss": 0.7376, + "step": 332 + }, + { + "epoch": 0.2758144671452236, + "grad_norm": 0.9083762764930725, + "learning_rate": 4.986880970520092e-06, + "loss": 0.7258, + "step": 333 + }, + { + "epoch": 0.2766427388183324, + "grad_norm": 1.048991084098816, + "learning_rate": 4.9867682188266524e-06, + "loss": 0.7399, + "step": 334 + }, + { + "epoch": 0.2774710104914412, + "grad_norm": 0.9404491782188416, + "learning_rate": 4.986654985966745e-06, + "loss": 0.7133, + "step": 335 + }, + { + "epoch": 0.27829928216454997, + "grad_norm": 0.9168346524238586, + "learning_rate": 4.986541271962278e-06, + "loss": 0.74, + "step": 336 + }, + { + "epoch": 0.27912755383765875, + "grad_norm": 0.8716689944267273, + "learning_rate": 4.986427076835253e-06, + "loss": 0.7249, + "step": 337 + }, + { + "epoch": 0.27995582551076753, + "grad_norm": 0.9097291827201843, + "learning_rate": 4.986312400607767e-06, + "loss": 0.7151, + "step": 338 + }, + { + "epoch": 0.2807840971838763, + "grad_norm": 0.8621960282325745, + "learning_rate": 4.98619724330201e-06, + "loss": 0.7125, + "step": 339 + }, + { + "epoch": 0.2816123688569851, + "grad_norm": 0.9823791980743408, + "learning_rate": 4.986081604940261e-06, + "loss": 0.7191, + "step": 340 + }, + { + "epoch": 0.2824406405300939, + "grad_norm": 0.8999047875404358, + "learning_rate": 4.9859654855448966e-06, + "loss": 0.7175, + "step": 341 + }, + { + "epoch": 0.28326891220320266, + "grad_norm": 0.8690859079360962, + "learning_rate": 4.985848885138386e-06, + "loss": 0.7309, + "step": 342 + }, + { + "epoch": 0.28409718387631144, + "grad_norm": 0.889500617980957, + "learning_rate": 4.985731803743287e-06, + "loss": 0.7391, + "step": 343 + }, + { + "epoch": 0.2849254555494202, + "grad_norm": 0.8992016911506653, + "learning_rate": 4.985614241382257e-06, + "loss": 0.7437, + "step": 344 + }, + { + "epoch": 0.285753727222529, + "grad_norm": 0.9100068211555481, + "learning_rate": 4.985496198078041e-06, + "loss": 0.722, + "step": 345 + }, + { + "epoch": 0.2865819988956378, + "grad_norm": 0.9013107419013977, + "learning_rate": 4.985377673853481e-06, + "loss": 0.7434, + "step": 346 + }, + { + "epoch": 0.2874102705687466, + "grad_norm": 0.8794704675674438, + "learning_rate": 4.9852586687315094e-06, + "loss": 0.7387, + "step": 347 + }, + { + "epoch": 0.28823854224185536, + "grad_norm": 0.9288201928138733, + "learning_rate": 4.9851391827351526e-06, + "loss": 0.7165, + "step": 348 + }, + { + "epoch": 0.2890668139149641, + "grad_norm": 0.8735847473144531, + "learning_rate": 4.98501921588753e-06, + "loss": 0.7479, + "step": 349 + }, + { + "epoch": 0.28989508558807286, + "grad_norm": 0.890616238117218, + "learning_rate": 4.984898768211853e-06, + "loss": 0.76, + "step": 350 + }, + { + "epoch": 0.29072335726118165, + "grad_norm": 0.9181028604507446, + "learning_rate": 4.98477783973143e-06, + "loss": 0.7366, + "step": 351 + }, + { + "epoch": 0.29155162893429043, + "grad_norm": 0.9947429299354553, + "learning_rate": 4.984656430469657e-06, + "loss": 0.7199, + "step": 352 + }, + { + "epoch": 0.2923799006073992, + "grad_norm": 0.8730769157409668, + "learning_rate": 4.984534540450027e-06, + "loss": 0.7153, + "step": 353 + }, + { + "epoch": 0.293208172280508, + "grad_norm": 0.8675997853279114, + "learning_rate": 4.9844121696961235e-06, + "loss": 0.7196, + "step": 354 + }, + { + "epoch": 0.2940364439536168, + "grad_norm": 0.9558542966842651, + "learning_rate": 4.984289318231624e-06, + "loss": 0.7311, + "step": 355 + }, + { + "epoch": 0.29486471562672556, + "grad_norm": 0.8817659616470337, + "learning_rate": 4.9841659860803e-06, + "loss": 0.7367, + "step": 356 + }, + { + "epoch": 0.29569298729983434, + "grad_norm": 0.8679526448249817, + "learning_rate": 4.984042173266016e-06, + "loss": 0.719, + "step": 357 + }, + { + "epoch": 0.2965212589729431, + "grad_norm": 0.9102020859718323, + "learning_rate": 4.983917879812726e-06, + "loss": 0.7301, + "step": 358 + }, + { + "epoch": 0.2973495306460519, + "grad_norm": 0.8987215757369995, + "learning_rate": 4.983793105744482e-06, + "loss": 0.7557, + "step": 359 + }, + { + "epoch": 0.2981778023191607, + "grad_norm": 0.8779832720756531, + "learning_rate": 4.983667851085426e-06, + "loss": 0.7471, + "step": 360 + }, + { + "epoch": 0.29900607399226947, + "grad_norm": 0.8762708306312561, + "learning_rate": 4.983542115859792e-06, + "loss": 0.7676, + "step": 361 + }, + { + "epoch": 0.29983434566537825, + "grad_norm": 0.910353422164917, + "learning_rate": 4.98341590009191e-06, + "loss": 0.7285, + "step": 362 + }, + { + "epoch": 0.30066261733848704, + "grad_norm": 0.9385405778884888, + "learning_rate": 4.983289203806202e-06, + "loss": 0.7142, + "step": 363 + }, + { + "epoch": 0.3014908890115958, + "grad_norm": 1.016781210899353, + "learning_rate": 4.983162027027182e-06, + "loss": 0.7408, + "step": 364 + }, + { + "epoch": 0.3023191606847046, + "grad_norm": 0.8642508387565613, + "learning_rate": 4.983034369779457e-06, + "loss": 0.7397, + "step": 365 + }, + { + "epoch": 0.3031474323578134, + "grad_norm": 0.9469170570373535, + "learning_rate": 4.982906232087728e-06, + "loss": 0.7158, + "step": 366 + }, + { + "epoch": 0.30397570403092217, + "grad_norm": 0.8818212151527405, + "learning_rate": 4.982777613976789e-06, + "loss": 0.7408, + "step": 367 + }, + { + "epoch": 0.30480397570403095, + "grad_norm": 0.8639838099479675, + "learning_rate": 4.982648515471526e-06, + "loss": 0.7217, + "step": 368 + }, + { + "epoch": 0.3056322473771397, + "grad_norm": 0.923780083656311, + "learning_rate": 4.982518936596917e-06, + "loss": 0.7284, + "step": 369 + }, + { + "epoch": 0.30646051905024846, + "grad_norm": 0.8992834687232971, + "learning_rate": 4.982388877378037e-06, + "loss": 0.7395, + "step": 370 + }, + { + "epoch": 0.30728879072335724, + "grad_norm": 0.8671404123306274, + "learning_rate": 4.982258337840049e-06, + "loss": 0.7157, + "step": 371 + }, + { + "epoch": 0.308117062396466, + "grad_norm": 0.8689168691635132, + "learning_rate": 4.982127318008212e-06, + "loss": 0.7398, + "step": 372 + }, + { + "epoch": 0.3089453340695748, + "grad_norm": 0.8964621424674988, + "learning_rate": 4.9819958179078765e-06, + "loss": 0.7385, + "step": 373 + }, + { + "epoch": 0.3097736057426836, + "grad_norm": 0.8984093070030212, + "learning_rate": 4.981863837564488e-06, + "loss": 0.7253, + "step": 374 + }, + { + "epoch": 0.31060187741579237, + "grad_norm": 0.8776745200157166, + "learning_rate": 4.981731377003583e-06, + "loss": 0.7126, + "step": 375 + }, + { + "epoch": 0.31143014908890115, + "grad_norm": 0.8782477378845215, + "learning_rate": 4.98159843625079e-06, + "loss": 0.7164, + "step": 376 + }, + { + "epoch": 0.31225842076200994, + "grad_norm": 0.8993703722953796, + "learning_rate": 4.981465015331832e-06, + "loss": 0.7391, + "step": 377 + }, + { + "epoch": 0.3130866924351187, + "grad_norm": 0.8975916504859924, + "learning_rate": 4.981331114272527e-06, + "loss": 0.7385, + "step": 378 + }, + { + "epoch": 0.3139149641082275, + "grad_norm": 0.8726824522018433, + "learning_rate": 4.9811967330987816e-06, + "loss": 0.7426, + "step": 379 + }, + { + "epoch": 0.3147432357813363, + "grad_norm": 0.8741514682769775, + "learning_rate": 4.981061871836597e-06, + "loss": 0.7246, + "step": 380 + }, + { + "epoch": 0.31557150745444507, + "grad_norm": 0.8807122707366943, + "learning_rate": 4.98092653051207e-06, + "loss": 0.7364, + "step": 381 + }, + { + "epoch": 0.31639977912755385, + "grad_norm": 0.835006058216095, + "learning_rate": 4.980790709151385e-06, + "loss": 0.7248, + "step": 382 + }, + { + "epoch": 0.31722805080066263, + "grad_norm": 0.9473997354507446, + "learning_rate": 4.980654407780824e-06, + "loss": 0.7331, + "step": 383 + }, + { + "epoch": 0.3180563224737714, + "grad_norm": 0.8332815170288086, + "learning_rate": 4.980517626426758e-06, + "loss": 0.7299, + "step": 384 + }, + { + "epoch": 0.3188845941468802, + "grad_norm": 0.8536360859870911, + "learning_rate": 4.980380365115656e-06, + "loss": 0.7095, + "step": 385 + }, + { + "epoch": 0.319712865819989, + "grad_norm": 0.9488157629966736, + "learning_rate": 4.980242623874075e-06, + "loss": 0.7382, + "step": 386 + }, + { + "epoch": 0.32054113749309776, + "grad_norm": 0.8623067736625671, + "learning_rate": 4.980104402728666e-06, + "loss": 0.7187, + "step": 387 + }, + { + "epoch": 0.32136940916620654, + "grad_norm": 0.9110252857208252, + "learning_rate": 4.979965701706175e-06, + "loss": 0.7354, + "step": 388 + }, + { + "epoch": 0.32219768083931527, + "grad_norm": 0.8966642618179321, + "learning_rate": 4.979826520833439e-06, + "loss": 0.7251, + "step": 389 + }, + { + "epoch": 0.32302595251242405, + "grad_norm": 0.856674075126648, + "learning_rate": 4.979686860137387e-06, + "loss": 0.723, + "step": 390 + }, + { + "epoch": 0.32385422418553284, + "grad_norm": 0.8623953461647034, + "learning_rate": 4.979546719645043e-06, + "loss": 0.7224, + "step": 391 + }, + { + "epoch": 0.3246824958586416, + "grad_norm": 0.8685624599456787, + "learning_rate": 4.979406099383523e-06, + "loss": 0.7283, + "step": 392 + }, + { + "epoch": 0.3255107675317504, + "grad_norm": 0.8610525727272034, + "learning_rate": 4.979264999380035e-06, + "loss": 0.7385, + "step": 393 + }, + { + "epoch": 0.3263390392048592, + "grad_norm": 0.8860796093940735, + "learning_rate": 4.979123419661882e-06, + "loss": 0.743, + "step": 394 + }, + { + "epoch": 0.32716731087796797, + "grad_norm": 0.8748682737350464, + "learning_rate": 4.978981360256457e-06, + "loss": 0.7447, + "step": 395 + }, + { + "epoch": 0.32799558255107675, + "grad_norm": 0.8750934600830078, + "learning_rate": 4.978838821191247e-06, + "loss": 0.7259, + "step": 396 + }, + { + "epoch": 0.32882385422418553, + "grad_norm": 0.9335777163505554, + "learning_rate": 4.978695802493834e-06, + "loss": 0.7271, + "step": 397 + }, + { + "epoch": 0.3296521258972943, + "grad_norm": 0.8823036551475525, + "learning_rate": 4.978552304191888e-06, + "loss": 0.723, + "step": 398 + }, + { + "epoch": 0.3304803975704031, + "grad_norm": 0.9503542184829712, + "learning_rate": 4.9784083263131764e-06, + "loss": 0.752, + "step": 399 + }, + { + "epoch": 0.3313086692435119, + "grad_norm": 0.8779938220977783, + "learning_rate": 4.978263868885557e-06, + "loss": 0.7098, + "step": 400 + }, + { + "epoch": 0.33213694091662066, + "grad_norm": 0.873857855796814, + "learning_rate": 4.978118931936981e-06, + "loss": 0.7143, + "step": 401 + }, + { + "epoch": 0.33296521258972944, + "grad_norm": 0.8437919020652771, + "learning_rate": 4.9779735154954924e-06, + "loss": 0.7275, + "step": 402 + }, + { + "epoch": 0.3337934842628382, + "grad_norm": 0.8914555311203003, + "learning_rate": 4.977827619589229e-06, + "loss": 0.722, + "step": 403 + }, + { + "epoch": 0.334621755935947, + "grad_norm": 0.9062597751617432, + "learning_rate": 4.977681244246417e-06, + "loss": 0.7067, + "step": 404 + }, + { + "epoch": 0.3354500276090558, + "grad_norm": 0.8555092811584473, + "learning_rate": 4.977534389495382e-06, + "loss": 0.7318, + "step": 405 + }, + { + "epoch": 0.33627829928216457, + "grad_norm": 0.8717684745788574, + "learning_rate": 4.9773870553645386e-06, + "loss": 0.7398, + "step": 406 + }, + { + "epoch": 0.33710657095527335, + "grad_norm": 0.8746877908706665, + "learning_rate": 4.977239241882393e-06, + "loss": 0.7154, + "step": 407 + }, + { + "epoch": 0.3379348426283821, + "grad_norm": 0.8905707597732544, + "learning_rate": 4.977090949077547e-06, + "loss": 0.7345, + "step": 408 + }, + { + "epoch": 0.33876311430149086, + "grad_norm": 0.9037347435951233, + "learning_rate": 4.9769421769786935e-06, + "loss": 0.7271, + "step": 409 + }, + { + "epoch": 0.33959138597459965, + "grad_norm": 0.8588722944259644, + "learning_rate": 4.976792925614619e-06, + "loss": 0.7335, + "step": 410 + }, + { + "epoch": 0.34041965764770843, + "grad_norm": 0.8786173462867737, + "learning_rate": 4.9766431950142005e-06, + "loss": 0.7254, + "step": 411 + }, + { + "epoch": 0.3412479293208172, + "grad_norm": 0.9019079804420471, + "learning_rate": 4.976492985206412e-06, + "loss": 0.715, + "step": 412 + }, + { + "epoch": 0.342076200993926, + "grad_norm": 0.878132700920105, + "learning_rate": 4.976342296220316e-06, + "loss": 0.7063, + "step": 413 + }, + { + "epoch": 0.3429044726670348, + "grad_norm": 0.8722684979438782, + "learning_rate": 4.976191128085069e-06, + "loss": 0.7481, + "step": 414 + }, + { + "epoch": 0.34373274434014356, + "grad_norm": 0.9076220393180847, + "learning_rate": 4.976039480829922e-06, + "loss": 0.7197, + "step": 415 + }, + { + "epoch": 0.34456101601325234, + "grad_norm": 0.8674967288970947, + "learning_rate": 4.975887354484217e-06, + "loss": 0.7282, + "step": 416 + }, + { + "epoch": 0.3453892876863611, + "grad_norm": 0.9406816959381104, + "learning_rate": 4.9757347490773885e-06, + "loss": 0.7222, + "step": 417 + }, + { + "epoch": 0.3462175593594699, + "grad_norm": 0.8747051358222961, + "learning_rate": 4.975581664638965e-06, + "loss": 0.7283, + "step": 418 + }, + { + "epoch": 0.3470458310325787, + "grad_norm": 1.0587687492370605, + "learning_rate": 4.9754281011985666e-06, + "loss": 0.7272, + "step": 419 + }, + { + "epoch": 0.34787410270568747, + "grad_norm": 0.8596946597099304, + "learning_rate": 4.975274058785906e-06, + "loss": 0.7407, + "step": 420 + }, + { + "epoch": 0.34870237437879625, + "grad_norm": 0.9230688214302063, + "learning_rate": 4.975119537430789e-06, + "loss": 0.7303, + "step": 421 + }, + { + "epoch": 0.34953064605190504, + "grad_norm": 0.9276477694511414, + "learning_rate": 4.974964537163115e-06, + "loss": 0.7296, + "step": 422 + }, + { + "epoch": 0.3503589177250138, + "grad_norm": 0.8969418406486511, + "learning_rate": 4.974809058012874e-06, + "loss": 0.7275, + "step": 423 + }, + { + "epoch": 0.3511871893981226, + "grad_norm": 0.8454467058181763, + "learning_rate": 4.97465310001015e-06, + "loss": 0.7144, + "step": 424 + }, + { + "epoch": 0.3520154610712314, + "grad_norm": 0.8833656907081604, + "learning_rate": 4.97449666318512e-06, + "loss": 0.7073, + "step": 425 + }, + { + "epoch": 0.35284373274434017, + "grad_norm": 0.9251653552055359, + "learning_rate": 4.974339747568052e-06, + "loss": 0.7351, + "step": 426 + }, + { + "epoch": 0.35367200441744895, + "grad_norm": 0.8938575983047485, + "learning_rate": 4.974182353189309e-06, + "loss": 0.7031, + "step": 427 + }, + { + "epoch": 0.3545002760905577, + "grad_norm": 0.8487107753753662, + "learning_rate": 4.974024480079343e-06, + "loss": 0.704, + "step": 428 + }, + { + "epoch": 0.35532854776366646, + "grad_norm": 0.8515727519989014, + "learning_rate": 4.973866128268705e-06, + "loss": 0.7092, + "step": 429 + }, + { + "epoch": 0.35615681943677524, + "grad_norm": 0.8869542479515076, + "learning_rate": 4.97370729778803e-06, + "loss": 0.7099, + "step": 430 + }, + { + "epoch": 0.356985091109884, + "grad_norm": 0.8830757141113281, + "learning_rate": 4.973547988668055e-06, + "loss": 0.7294, + "step": 431 + }, + { + "epoch": 0.3578133627829928, + "grad_norm": 0.9321587681770325, + "learning_rate": 4.973388200939601e-06, + "loss": 0.7276, + "step": 432 + }, + { + "epoch": 0.3586416344561016, + "grad_norm": 0.9408143758773804, + "learning_rate": 4.9732279346335866e-06, + "loss": 0.7216, + "step": 433 + }, + { + "epoch": 0.35946990612921037, + "grad_norm": 0.9189308285713196, + "learning_rate": 4.973067189781022e-06, + "loss": 0.7079, + "step": 434 + }, + { + "epoch": 0.36029817780231915, + "grad_norm": 0.9198120832443237, + "learning_rate": 4.97290596641301e-06, + "loss": 0.7188, + "step": 435 + }, + { + "epoch": 0.36112644947542794, + "grad_norm": 0.9123285412788391, + "learning_rate": 4.972744264560746e-06, + "loss": 0.7328, + "step": 436 + }, + { + "epoch": 0.3619547211485367, + "grad_norm": 0.8722008466720581, + "learning_rate": 4.972582084255517e-06, + "loss": 0.7228, + "step": 437 + }, + { + "epoch": 0.3627829928216455, + "grad_norm": 1.035325288772583, + "learning_rate": 4.972419425528704e-06, + "loss": 0.72, + "step": 438 + }, + { + "epoch": 0.3636112644947543, + "grad_norm": 0.9577400088310242, + "learning_rate": 4.97225628841178e-06, + "loss": 0.7382, + "step": 439 + }, + { + "epoch": 0.36443953616786307, + "grad_norm": 0.9206854104995728, + "learning_rate": 4.972092672936309e-06, + "loss": 0.7294, + "step": 440 + }, + { + "epoch": 0.36526780784097185, + "grad_norm": 0.9281194806098938, + "learning_rate": 4.971928579133952e-06, + "loss": 0.7137, + "step": 441 + }, + { + "epoch": 0.36609607951408063, + "grad_norm": 0.951049268245697, + "learning_rate": 4.971764007036457e-06, + "loss": 0.7354, + "step": 442 + }, + { + "epoch": 0.3669243511871894, + "grad_norm": 0.8779566287994385, + "learning_rate": 4.971598956675668e-06, + "loss": 0.6858, + "step": 443 + }, + { + "epoch": 0.3677526228602982, + "grad_norm": 0.9221998453140259, + "learning_rate": 4.971433428083522e-06, + "loss": 0.7116, + "step": 444 + }, + { + "epoch": 0.368580894533407, + "grad_norm": 0.9608840346336365, + "learning_rate": 4.971267421292045e-06, + "loss": 0.7129, + "step": 445 + }, + { + "epoch": 0.36940916620651576, + "grad_norm": 0.8761363625526428, + "learning_rate": 4.971100936333359e-06, + "loss": 0.718, + "step": 446 + }, + { + "epoch": 0.37023743787962454, + "grad_norm": 0.8808798789978027, + "learning_rate": 4.970933973239677e-06, + "loss": 0.6986, + "step": 447 + }, + { + "epoch": 0.37106570955273327, + "grad_norm": 0.8818857669830322, + "learning_rate": 4.9707665320433055e-06, + "loss": 0.7465, + "step": 448 + }, + { + "epoch": 0.37189398122584205, + "grad_norm": 0.9144098162651062, + "learning_rate": 4.970598612776641e-06, + "loss": 0.7187, + "step": 449 + }, + { + "epoch": 0.37272225289895083, + "grad_norm": 0.8439158797264099, + "learning_rate": 4.9704302154721776e-06, + "loss": 0.7195, + "step": 450 + }, + { + "epoch": 0.3735505245720596, + "grad_norm": 0.8661074638366699, + "learning_rate": 4.970261340162495e-06, + "loss": 0.712, + "step": 451 + }, + { + "epoch": 0.3743787962451684, + "grad_norm": 0.8818789124488831, + "learning_rate": 4.970091986880271e-06, + "loss": 0.7199, + "step": 452 + }, + { + "epoch": 0.3752070679182772, + "grad_norm": 0.9560503363609314, + "learning_rate": 4.969922155658273e-06, + "loss": 0.7145, + "step": 453 + }, + { + "epoch": 0.37603533959138596, + "grad_norm": 0.8723775744438171, + "learning_rate": 4.969751846529363e-06, + "loss": 0.7152, + "step": 454 + }, + { + "epoch": 0.37686361126449475, + "grad_norm": 0.8612253069877625, + "learning_rate": 4.9695810595264935e-06, + "loss": 0.728, + "step": 455 + }, + { + "epoch": 0.37769188293760353, + "grad_norm": 0.8995444178581238, + "learning_rate": 4.969409794682709e-06, + "loss": 0.678, + "step": 456 + }, + { + "epoch": 0.3785201546107123, + "grad_norm": 0.8770595788955688, + "learning_rate": 4.96923805203115e-06, + "loss": 0.7354, + "step": 457 + }, + { + "epoch": 0.3793484262838211, + "grad_norm": 0.893329918384552, + "learning_rate": 4.969065831605045e-06, + "loss": 0.7102, + "step": 458 + }, + { + "epoch": 0.3801766979569299, + "grad_norm": 0.954058051109314, + "learning_rate": 4.968893133437718e-06, + "loss": 0.7144, + "step": 459 + }, + { + "epoch": 0.38100496963003866, + "grad_norm": 0.9711002707481384, + "learning_rate": 4.968719957562585e-06, + "loss": 0.7452, + "step": 460 + }, + { + "epoch": 0.38183324130314744, + "grad_norm": 0.9036396741867065, + "learning_rate": 4.968546304013153e-06, + "loss": 0.7535, + "step": 461 + }, + { + "epoch": 0.3826615129762562, + "grad_norm": 0.8948055505752563, + "learning_rate": 4.968372172823023e-06, + "loss": 0.7284, + "step": 462 + }, + { + "epoch": 0.383489784649365, + "grad_norm": 0.9134189486503601, + "learning_rate": 4.968197564025886e-06, + "loss": 0.723, + "step": 463 + }, + { + "epoch": 0.3843180563224738, + "grad_norm": 0.9191608428955078, + "learning_rate": 4.9680224776555305e-06, + "loss": 0.7078, + "step": 464 + }, + { + "epoch": 0.38514632799558257, + "grad_norm": 0.879683792591095, + "learning_rate": 4.9678469137458315e-06, + "loss": 0.6805, + "step": 465 + }, + { + "epoch": 0.38597459966869135, + "grad_norm": 0.8887293338775635, + "learning_rate": 4.96767087233076e-06, + "loss": 0.7296, + "step": 466 + }, + { + "epoch": 0.38680287134180014, + "grad_norm": 0.889991044998169, + "learning_rate": 4.967494353444377e-06, + "loss": 0.702, + "step": 467 + }, + { + "epoch": 0.38763114301490886, + "grad_norm": 0.9667506814002991, + "learning_rate": 4.967317357120839e-06, + "loss": 0.7305, + "step": 468 + }, + { + "epoch": 0.38845941468801765, + "grad_norm": 0.9297274351119995, + "learning_rate": 4.967139883394392e-06, + "loss": 0.7344, + "step": 469 + }, + { + "epoch": 0.38928768636112643, + "grad_norm": 0.9167365431785583, + "learning_rate": 4.966961932299377e-06, + "loss": 0.7016, + "step": 470 + }, + { + "epoch": 0.3901159580342352, + "grad_norm": 0.877378523349762, + "learning_rate": 4.966783503870224e-06, + "loss": 0.7313, + "step": 471 + }, + { + "epoch": 0.390944229707344, + "grad_norm": 0.9625933766365051, + "learning_rate": 4.966604598141458e-06, + "loss": 0.7245, + "step": 472 + }, + { + "epoch": 0.3917725013804528, + "grad_norm": 0.8859156966209412, + "learning_rate": 4.9664252151476955e-06, + "loss": 0.7094, + "step": 473 + }, + { + "epoch": 0.39260077305356156, + "grad_norm": 0.9172984957695007, + "learning_rate": 4.966245354923645e-06, + "loss": 0.6966, + "step": 474 + }, + { + "epoch": 0.39342904472667034, + "grad_norm": 0.9040864109992981, + "learning_rate": 4.96606501750411e-06, + "loss": 0.722, + "step": 475 + }, + { + "epoch": 0.3942573163997791, + "grad_norm": 0.8900951147079468, + "learning_rate": 4.965884202923982e-06, + "loss": 0.7263, + "step": 476 + }, + { + "epoch": 0.3950855880728879, + "grad_norm": 0.9096625447273254, + "learning_rate": 4.965702911218247e-06, + "loss": 0.702, + "step": 477 + }, + { + "epoch": 0.3959138597459967, + "grad_norm": 0.8801544904708862, + "learning_rate": 4.965521142421984e-06, + "loss": 0.7113, + "step": 478 + }, + { + "epoch": 0.39674213141910547, + "grad_norm": 0.9054175615310669, + "learning_rate": 4.965338896570363e-06, + "loss": 0.7135, + "step": 479 + }, + { + "epoch": 0.39757040309221425, + "grad_norm": 0.8734445571899414, + "learning_rate": 4.965156173698648e-06, + "loss": 0.7181, + "step": 480 + }, + { + "epoch": 0.39839867476532304, + "grad_norm": 0.9113215208053589, + "learning_rate": 4.964972973842191e-06, + "loss": 0.7008, + "step": 481 + }, + { + "epoch": 0.3992269464384318, + "grad_norm": 0.8941736221313477, + "learning_rate": 4.964789297036444e-06, + "loss": 0.7201, + "step": 482 + }, + { + "epoch": 0.4000552181115406, + "grad_norm": 0.8555464148521423, + "learning_rate": 4.964605143316944e-06, + "loss": 0.7173, + "step": 483 + }, + { + "epoch": 0.4008834897846494, + "grad_norm": 0.8719691634178162, + "learning_rate": 4.9644205127193225e-06, + "loss": 0.7211, + "step": 484 + }, + { + "epoch": 0.40171176145775817, + "grad_norm": 0.901626467704773, + "learning_rate": 4.964235405279306e-06, + "loss": 0.7223, + "step": 485 + }, + { + "epoch": 0.40254003313086695, + "grad_norm": 0.856769859790802, + "learning_rate": 4.9640498210327095e-06, + "loss": 0.7103, + "step": 486 + }, + { + "epoch": 0.40336830480397573, + "grad_norm": 0.9053338766098022, + "learning_rate": 4.963863760015442e-06, + "loss": 0.7301, + "step": 487 + }, + { + "epoch": 0.40419657647708446, + "grad_norm": 0.8989829421043396, + "learning_rate": 4.963677222263505e-06, + "loss": 0.7391, + "step": 488 + }, + { + "epoch": 0.40502484815019324, + "grad_norm": 0.9187886118888855, + "learning_rate": 4.963490207812991e-06, + "loss": 0.7179, + "step": 489 + }, + { + "epoch": 0.405853119823302, + "grad_norm": 0.8712778687477112, + "learning_rate": 4.963302716700087e-06, + "loss": 0.7088, + "step": 490 + }, + { + "epoch": 0.4066813914964108, + "grad_norm": 1.0485447645187378, + "learning_rate": 4.9631147489610695e-06, + "loss": 0.7068, + "step": 491 + }, + { + "epoch": 0.4075096631695196, + "grad_norm": 0.9332602024078369, + "learning_rate": 4.962926304632309e-06, + "loss": 0.7201, + "step": 492 + }, + { + "epoch": 0.40833793484262837, + "grad_norm": 0.9264400601387024, + "learning_rate": 4.962737383750268e-06, + "loss": 0.7136, + "step": 493 + }, + { + "epoch": 0.40916620651573715, + "grad_norm": 0.8812651634216309, + "learning_rate": 4.962547986351501e-06, + "loss": 0.7315, + "step": 494 + }, + { + "epoch": 0.40999447818884593, + "grad_norm": 0.8905262351036072, + "learning_rate": 4.962358112472653e-06, + "loss": 0.7127, + "step": 495 + }, + { + "epoch": 0.4108227498619547, + "grad_norm": 0.887514591217041, + "learning_rate": 4.962167762150465e-06, + "loss": 0.7602, + "step": 496 + }, + { + "epoch": 0.4116510215350635, + "grad_norm": 0.8683431148529053, + "learning_rate": 4.9619769354217665e-06, + "loss": 0.7091, + "step": 497 + }, + { + "epoch": 0.4124792932081723, + "grad_norm": 0.944595217704773, + "learning_rate": 4.961785632323481e-06, + "loss": 0.708, + "step": 498 + }, + { + "epoch": 0.41330756488128106, + "grad_norm": 0.8834611773490906, + "learning_rate": 4.961593852892625e-06, + "loss": 0.7308, + "step": 499 + }, + { + "epoch": 0.41413583655438985, + "grad_norm": 0.8479524850845337, + "learning_rate": 4.961401597166304e-06, + "loss": 0.7061, + "step": 500 + }, + { + "epoch": 0.41496410822749863, + "grad_norm": 0.9092981219291687, + "learning_rate": 4.96120886518172e-06, + "loss": 0.7176, + "step": 501 + }, + { + "epoch": 0.4157923799006074, + "grad_norm": 0.9024932384490967, + "learning_rate": 4.961015656976163e-06, + "loss": 0.6944, + "step": 502 + }, + { + "epoch": 0.4166206515737162, + "grad_norm": 0.8945576548576355, + "learning_rate": 4.960821972587018e-06, + "loss": 0.6958, + "step": 503 + }, + { + "epoch": 0.417448923246825, + "grad_norm": 0.8668336868286133, + "learning_rate": 4.9606278120517606e-06, + "loss": 0.7219, + "step": 504 + }, + { + "epoch": 0.41827719491993376, + "grad_norm": 0.8831227421760559, + "learning_rate": 4.96043317540796e-06, + "loss": 0.7124, + "step": 505 + }, + { + "epoch": 0.41910546659304254, + "grad_norm": 0.9637073874473572, + "learning_rate": 4.960238062693274e-06, + "loss": 0.7337, + "step": 506 + }, + { + "epoch": 0.4199337382661513, + "grad_norm": 0.8895899653434753, + "learning_rate": 4.960042473945459e-06, + "loss": 0.7459, + "step": 507 + }, + { + "epoch": 0.42076200993926005, + "grad_norm": 0.8762213587760925, + "learning_rate": 4.959846409202356e-06, + "loss": 0.7319, + "step": 508 + }, + { + "epoch": 0.42159028161236883, + "grad_norm": 0.9177393317222595, + "learning_rate": 4.959649868501903e-06, + "loss": 0.7114, + "step": 509 + }, + { + "epoch": 0.4224185532854776, + "grad_norm": 0.88541179895401, + "learning_rate": 4.9594528518821305e-06, + "loss": 0.716, + "step": 510 + }, + { + "epoch": 0.4232468249585864, + "grad_norm": 0.865566611289978, + "learning_rate": 4.959255359381158e-06, + "loss": 0.7098, + "step": 511 + }, + { + "epoch": 0.4240750966316952, + "grad_norm": 0.8629707098007202, + "learning_rate": 4.959057391037197e-06, + "loss": 0.7146, + "step": 512 + }, + { + "epoch": 0.42490336830480396, + "grad_norm": 0.8642273545265198, + "learning_rate": 4.958858946888554e-06, + "loss": 0.7345, + "step": 513 + }, + { + "epoch": 0.42573163997791275, + "grad_norm": 0.9157605171203613, + "learning_rate": 4.958660026973626e-06, + "loss": 0.7177, + "step": 514 + }, + { + "epoch": 0.42655991165102153, + "grad_norm": 0.9108191728591919, + "learning_rate": 4.958460631330903e-06, + "loss": 0.7317, + "step": 515 + }, + { + "epoch": 0.4273881833241303, + "grad_norm": 0.8524913191795349, + "learning_rate": 4.958260759998965e-06, + "loss": 0.7193, + "step": 516 + }, + { + "epoch": 0.4282164549972391, + "grad_norm": 0.9197820425033569, + "learning_rate": 4.958060413016485e-06, + "loss": 0.7127, + "step": 517 + }, + { + "epoch": 0.4290447266703479, + "grad_norm": 0.8744186758995056, + "learning_rate": 4.9578595904222296e-06, + "loss": 0.6968, + "step": 518 + }, + { + "epoch": 0.42987299834345666, + "grad_norm": 0.9071037769317627, + "learning_rate": 4.957658292255055e-06, + "loss": 0.6963, + "step": 519 + }, + { + "epoch": 0.43070127001656544, + "grad_norm": 0.937796950340271, + "learning_rate": 4.95745651855391e-06, + "loss": 0.7193, + "step": 520 + }, + { + "epoch": 0.4315295416896742, + "grad_norm": 0.883471667766571, + "learning_rate": 4.9572542693578375e-06, + "loss": 0.6886, + "step": 521 + }, + { + "epoch": 0.432357813362783, + "grad_norm": 0.8738589882850647, + "learning_rate": 4.9570515447059705e-06, + "loss": 0.7072, + "step": 522 + }, + { + "epoch": 0.4331860850358918, + "grad_norm": 0.9039497971534729, + "learning_rate": 4.9568483446375335e-06, + "loss": 0.7251, + "step": 523 + }, + { + "epoch": 0.43401435670900057, + "grad_norm": 0.896789014339447, + "learning_rate": 4.956644669191845e-06, + "loss": 0.7021, + "step": 524 + }, + { + "epoch": 0.43484262838210935, + "grad_norm": 0.8883570432662964, + "learning_rate": 4.9564405184083126e-06, + "loss": 0.7153, + "step": 525 + }, + { + "epoch": 0.43567090005521814, + "grad_norm": 0.8869197368621826, + "learning_rate": 4.956235892326438e-06, + "loss": 0.7061, + "step": 526 + }, + { + "epoch": 0.4364991717283269, + "grad_norm": 0.9237920641899109, + "learning_rate": 4.9560307909858165e-06, + "loss": 0.6988, + "step": 527 + }, + { + "epoch": 0.43732744340143564, + "grad_norm": 0.9359283447265625, + "learning_rate": 4.955825214426131e-06, + "loss": 0.738, + "step": 528 + }, + { + "epoch": 0.4381557150745444, + "grad_norm": 0.8602650761604309, + "learning_rate": 4.95561916268716e-06, + "loss": 0.709, + "step": 529 + }, + { + "epoch": 0.4389839867476532, + "grad_norm": 0.9013513922691345, + "learning_rate": 4.955412635808771e-06, + "loss": 0.6836, + "step": 530 + }, + { + "epoch": 0.439812258420762, + "grad_norm": 0.9136670827865601, + "learning_rate": 4.955205633830927e-06, + "loss": 0.7263, + "step": 531 + }, + { + "epoch": 0.4406405300938708, + "grad_norm": 0.9013793468475342, + "learning_rate": 4.95499815679368e-06, + "loss": 0.6937, + "step": 532 + }, + { + "epoch": 0.44146880176697956, + "grad_norm": 0.9017112851142883, + "learning_rate": 4.954790204737174e-06, + "loss": 0.6913, + "step": 533 + }, + { + "epoch": 0.44229707344008834, + "grad_norm": 0.8901436924934387, + "learning_rate": 4.954581777701647e-06, + "loss": 0.7241, + "step": 534 + }, + { + "epoch": 0.4431253451131971, + "grad_norm": 0.8882759809494019, + "learning_rate": 4.954372875727427e-06, + "loss": 0.7115, + "step": 535 + }, + { + "epoch": 0.4439536167863059, + "grad_norm": 0.8909885883331299, + "learning_rate": 4.954163498854936e-06, + "loss": 0.7146, + "step": 536 + }, + { + "epoch": 0.4447818884594147, + "grad_norm": 0.8951786756515503, + "learning_rate": 4.953953647124684e-06, + "loss": 0.699, + "step": 537 + }, + { + "epoch": 0.44561016013252347, + "grad_norm": 0.8890933990478516, + "learning_rate": 4.953743320577278e-06, + "loss": 0.7018, + "step": 538 + }, + { + "epoch": 0.44643843180563225, + "grad_norm": 0.910393238067627, + "learning_rate": 4.953532519253412e-06, + "loss": 0.7299, + "step": 539 + }, + { + "epoch": 0.44726670347874103, + "grad_norm": 0.9152857065200806, + "learning_rate": 4.953321243193876e-06, + "loss": 0.6984, + "step": 540 + }, + { + "epoch": 0.4480949751518498, + "grad_norm": 0.9001837372779846, + "learning_rate": 4.9531094924395485e-06, + "loss": 0.727, + "step": 541 + }, + { + "epoch": 0.4489232468249586, + "grad_norm": 0.882060170173645, + "learning_rate": 4.952897267031401e-06, + "loss": 0.7016, + "step": 542 + }, + { + "epoch": 0.4497515184980674, + "grad_norm": 0.8665159344673157, + "learning_rate": 4.9526845670104995e-06, + "loss": 0.7139, + "step": 543 + }, + { + "epoch": 0.45057979017117616, + "grad_norm": 0.9200844764709473, + "learning_rate": 4.952471392417997e-06, + "loss": 0.7016, + "step": 544 + }, + { + "epoch": 0.45140806184428495, + "grad_norm": 0.8852037191390991, + "learning_rate": 4.9522577432951416e-06, + "loss": 0.7048, + "step": 545 + }, + { + "epoch": 0.45223633351739373, + "grad_norm": 0.9003865122795105, + "learning_rate": 4.952043619683274e-06, + "loss": 0.7106, + "step": 546 + }, + { + "epoch": 0.4530646051905025, + "grad_norm": 0.8888446092605591, + "learning_rate": 4.951829021623822e-06, + "loss": 0.7202, + "step": 547 + }, + { + "epoch": 0.45389287686361124, + "grad_norm": 0.8940277099609375, + "learning_rate": 4.9516139491583094e-06, + "loss": 0.7156, + "step": 548 + }, + { + "epoch": 0.45472114853672, + "grad_norm": 0.9066211581230164, + "learning_rate": 4.951398402328353e-06, + "loss": 0.6865, + "step": 549 + }, + { + "epoch": 0.4555494202098288, + "grad_norm": 0.9352865815162659, + "learning_rate": 4.9511823811756565e-06, + "loss": 0.7117, + "step": 550 + }, + { + "epoch": 0.4563776918829376, + "grad_norm": 0.8672269582748413, + "learning_rate": 4.95096588574202e-06, + "loss": 0.7166, + "step": 551 + }, + { + "epoch": 0.45720596355604637, + "grad_norm": 0.8730685114860535, + "learning_rate": 4.95074891606933e-06, + "loss": 0.6998, + "step": 552 + }, + { + "epoch": 0.45803423522915515, + "grad_norm": 0.896821916103363, + "learning_rate": 4.950531472199572e-06, + "loss": 0.6839, + "step": 553 + }, + { + "epoch": 0.45886250690226393, + "grad_norm": 0.918572187423706, + "learning_rate": 4.950313554174817e-06, + "loss": 0.7155, + "step": 554 + }, + { + "epoch": 0.4596907785753727, + "grad_norm": 0.8670752048492432, + "learning_rate": 4.950095162037232e-06, + "loss": 0.708, + "step": 555 + }, + { + "epoch": 0.4605190502484815, + "grad_norm": 0.875618040561676, + "learning_rate": 4.9498762958290715e-06, + "loss": 0.6956, + "step": 556 + }, + { + "epoch": 0.4613473219215903, + "grad_norm": 0.9120676517486572, + "learning_rate": 4.949656955592685e-06, + "loss": 0.7327, + "step": 557 + }, + { + "epoch": 0.46217559359469906, + "grad_norm": 0.9670370817184448, + "learning_rate": 4.949437141370514e-06, + "loss": 0.705, + "step": 558 + }, + { + "epoch": 0.46300386526780785, + "grad_norm": 0.9155322313308716, + "learning_rate": 4.949216853205089e-06, + "loss": 0.7126, + "step": 559 + }, + { + "epoch": 0.46383213694091663, + "grad_norm": 0.8945392370223999, + "learning_rate": 4.948996091139034e-06, + "loss": 0.7083, + "step": 560 + }, + { + "epoch": 0.4646604086140254, + "grad_norm": 0.8922162055969238, + "learning_rate": 4.948774855215065e-06, + "loss": 0.7189, + "step": 561 + }, + { + "epoch": 0.4654886802871342, + "grad_norm": 0.8940657377243042, + "learning_rate": 4.948553145475989e-06, + "loss": 0.6934, + "step": 562 + }, + { + "epoch": 0.466316951960243, + "grad_norm": 0.8946855068206787, + "learning_rate": 4.948330961964703e-06, + "loss": 0.6835, + "step": 563 + }, + { + "epoch": 0.46714522363335176, + "grad_norm": 0.9002820253372192, + "learning_rate": 4.948108304724201e-06, + "loss": 0.712, + "step": 564 + }, + { + "epoch": 0.46797349530646054, + "grad_norm": 0.8715711832046509, + "learning_rate": 4.947885173797561e-06, + "loss": 0.7197, + "step": 565 + }, + { + "epoch": 0.4688017669795693, + "grad_norm": 0.9269874095916748, + "learning_rate": 4.94766156922796e-06, + "loss": 0.7311, + "step": 566 + }, + { + "epoch": 0.4696300386526781, + "grad_norm": 0.8999423980712891, + "learning_rate": 4.947437491058662e-06, + "loss": 0.7144, + "step": 567 + }, + { + "epoch": 0.47045831032578683, + "grad_norm": 0.8704255223274231, + "learning_rate": 4.947212939333024e-06, + "loss": 0.7156, + "step": 568 + }, + { + "epoch": 0.4712865819988956, + "grad_norm": 0.8770347237586975, + "learning_rate": 4.946987914094495e-06, + "loss": 0.6969, + "step": 569 + }, + { + "epoch": 0.4721148536720044, + "grad_norm": 0.8865682482719421, + "learning_rate": 4.946762415386615e-06, + "loss": 0.7089, + "step": 570 + }, + { + "epoch": 0.4729431253451132, + "grad_norm": 0.8756279349327087, + "learning_rate": 4.946536443253016e-06, + "loss": 0.7159, + "step": 571 + }, + { + "epoch": 0.47377139701822196, + "grad_norm": 0.8844839930534363, + "learning_rate": 4.946309997737422e-06, + "loss": 0.7234, + "step": 572 + }, + { + "epoch": 0.47459966869133074, + "grad_norm": 0.8549885153770447, + "learning_rate": 4.946083078883647e-06, + "loss": 0.6927, + "step": 573 + }, + { + "epoch": 0.4754279403644395, + "grad_norm": 0.8967456817626953, + "learning_rate": 4.945855686735599e-06, + "loss": 0.7082, + "step": 574 + }, + { + "epoch": 0.4762562120375483, + "grad_norm": 0.9120200276374817, + "learning_rate": 4.9456278213372755e-06, + "loss": 0.7135, + "step": 575 + }, + { + "epoch": 0.4770844837106571, + "grad_norm": 0.8780766725540161, + "learning_rate": 4.945399482732765e-06, + "loss": 0.7139, + "step": 576 + }, + { + "epoch": 0.4779127553837659, + "grad_norm": 0.8900999426841736, + "learning_rate": 4.945170670966253e-06, + "loss": 0.7132, + "step": 577 + }, + { + "epoch": 0.47874102705687466, + "grad_norm": 0.9402151107788086, + "learning_rate": 4.944941386082008e-06, + "loss": 0.7139, + "step": 578 + }, + { + "epoch": 0.47956929872998344, + "grad_norm": 0.8769465088844299, + "learning_rate": 4.944711628124396e-06, + "loss": 0.7236, + "step": 579 + }, + { + "epoch": 0.4803975704030922, + "grad_norm": 0.9185771346092224, + "learning_rate": 4.944481397137874e-06, + "loss": 0.6884, + "step": 580 + }, + { + "epoch": 0.481225842076201, + "grad_norm": 0.8775897026062012, + "learning_rate": 4.944250693166988e-06, + "loss": 0.731, + "step": 581 + }, + { + "epoch": 0.4820541137493098, + "grad_norm": 0.8751668930053711, + "learning_rate": 4.94401951625638e-06, + "loss": 0.7007, + "step": 582 + }, + { + "epoch": 0.48288238542241857, + "grad_norm": 0.9042515754699707, + "learning_rate": 4.943787866450776e-06, + "loss": 0.7064, + "step": 583 + }, + { + "epoch": 0.48371065709552735, + "grad_norm": 0.8845478892326355, + "learning_rate": 4.943555743795002e-06, + "loss": 0.7069, + "step": 584 + }, + { + "epoch": 0.48453892876863613, + "grad_norm": 0.8959096670150757, + "learning_rate": 4.943323148333969e-06, + "loss": 0.6871, + "step": 585 + }, + { + "epoch": 0.4853672004417449, + "grad_norm": 0.9021669626235962, + "learning_rate": 4.943090080112684e-06, + "loss": 0.7156, + "step": 586 + }, + { + "epoch": 0.4861954721148537, + "grad_norm": 0.8768396973609924, + "learning_rate": 4.942856539176243e-06, + "loss": 0.681, + "step": 587 + }, + { + "epoch": 0.4870237437879624, + "grad_norm": 0.8734706044197083, + "learning_rate": 4.942622525569832e-06, + "loss": 0.7371, + "step": 588 + }, + { + "epoch": 0.4878520154610712, + "grad_norm": 0.9155406355857849, + "learning_rate": 4.942388039338733e-06, + "loss": 0.7012, + "step": 589 + }, + { + "epoch": 0.48868028713418, + "grad_norm": 0.8778123259544373, + "learning_rate": 4.942153080528317e-06, + "loss": 0.7201, + "step": 590 + }, + { + "epoch": 0.4895085588072888, + "grad_norm": 0.9102092981338501, + "learning_rate": 4.941917649184045e-06, + "loss": 0.6951, + "step": 591 + }, + { + "epoch": 0.49033683048039756, + "grad_norm": 0.8855768442153931, + "learning_rate": 4.941681745351471e-06, + "loss": 0.6997, + "step": 592 + }, + { + "epoch": 0.49116510215350634, + "grad_norm": 0.8770707845687866, + "learning_rate": 4.94144536907624e-06, + "loss": 0.7092, + "step": 593 + }, + { + "epoch": 0.4919933738266151, + "grad_norm": 0.900458574295044, + "learning_rate": 4.9412085204040896e-06, + "loss": 0.6909, + "step": 594 + }, + { + "epoch": 0.4928216454997239, + "grad_norm": 0.903489887714386, + "learning_rate": 4.940971199380846e-06, + "loss": 0.7066, + "step": 595 + }, + { + "epoch": 0.4936499171728327, + "grad_norm": 0.9331871271133423, + "learning_rate": 4.940733406052431e-06, + "loss": 0.6924, + "step": 596 + }, + { + "epoch": 0.49447818884594147, + "grad_norm": 0.8855593800544739, + "learning_rate": 4.9404951404648535e-06, + "loss": 0.6991, + "step": 597 + }, + { + "epoch": 0.49530646051905025, + "grad_norm": 0.9075731039047241, + "learning_rate": 4.940256402664216e-06, + "loss": 0.6973, + "step": 598 + }, + { + "epoch": 0.49613473219215903, + "grad_norm": 0.8970743417739868, + "learning_rate": 4.940017192696713e-06, + "loss": 0.6767, + "step": 599 + }, + { + "epoch": 0.4969630038652678, + "grad_norm": 0.8644358515739441, + "learning_rate": 4.939777510608628e-06, + "loss": 0.7284, + "step": 600 + }, + { + "epoch": 0.4977912755383766, + "grad_norm": 0.8807653784751892, + "learning_rate": 4.939537356446339e-06, + "loss": 0.7143, + "step": 601 + }, + { + "epoch": 0.4986195472114854, + "grad_norm": 0.892946183681488, + "learning_rate": 4.939296730256312e-06, + "loss": 0.7021, + "step": 602 + }, + { + "epoch": 0.49944781888459416, + "grad_norm": 0.9661110639572144, + "learning_rate": 4.939055632085107e-06, + "loss": 0.7198, + "step": 603 + }, + { + "epoch": 0.500276090557703, + "grad_norm": 0.8638496398925781, + "learning_rate": 4.9388140619793735e-06, + "loss": 0.6943, + "step": 604 + }, + { + "epoch": 0.5011043622308117, + "grad_norm": 0.9025457501411438, + "learning_rate": 4.938572019985853e-06, + "loss": 0.7114, + "step": 605 + }, + { + "epoch": 0.5019326339039205, + "grad_norm": 0.9622734785079956, + "learning_rate": 4.93832950615138e-06, + "loss": 0.7086, + "step": 606 + }, + { + "epoch": 0.5027609055770292, + "grad_norm": 0.9565468430519104, + "learning_rate": 4.938086520522876e-06, + "loss": 0.7125, + "step": 607 + }, + { + "epoch": 0.5035891772501381, + "grad_norm": 0.8970094323158264, + "learning_rate": 4.937843063147359e-06, + "loss": 0.7023, + "step": 608 + }, + { + "epoch": 0.5044174489232468, + "grad_norm": 0.8920996785163879, + "learning_rate": 4.937599134071933e-06, + "loss": 0.7041, + "step": 609 + }, + { + "epoch": 0.5052457205963556, + "grad_norm": 0.9304755926132202, + "learning_rate": 4.937354733343799e-06, + "loss": 0.7165, + "step": 610 + }, + { + "epoch": 0.5060739922694644, + "grad_norm": 0.8995933532714844, + "learning_rate": 4.937109861010244e-06, + "loss": 0.6907, + "step": 611 + }, + { + "epoch": 0.5069022639425732, + "grad_norm": 0.8648850321769714, + "learning_rate": 4.936864517118649e-06, + "loss": 0.6994, + "step": 612 + }, + { + "epoch": 0.5077305356156819, + "grad_norm": 0.8414506316184998, + "learning_rate": 4.936618701716487e-06, + "loss": 0.7081, + "step": 613 + }, + { + "epoch": 0.5085588072887908, + "grad_norm": 0.9282650947570801, + "learning_rate": 4.936372414851321e-06, + "loss": 0.7303, + "step": 614 + }, + { + "epoch": 0.5093870789618995, + "grad_norm": 0.877757728099823, + "learning_rate": 4.936125656570802e-06, + "loss": 0.7132, + "step": 615 + }, + { + "epoch": 0.5102153506350083, + "grad_norm": 0.8645135164260864, + "learning_rate": 4.935878426922679e-06, + "loss": 0.723, + "step": 616 + }, + { + "epoch": 0.5110436223081171, + "grad_norm": 0.9014747738838196, + "learning_rate": 4.935630725954787e-06, + "loss": 0.7167, + "step": 617 + }, + { + "epoch": 0.5118718939812258, + "grad_norm": 0.9295386672019958, + "learning_rate": 4.935382553715055e-06, + "loss": 0.705, + "step": 618 + }, + { + "epoch": 0.5127001656543346, + "grad_norm": 0.8910616040229797, + "learning_rate": 4.935133910251499e-06, + "loss": 0.6781, + "step": 619 + }, + { + "epoch": 0.5135284373274434, + "grad_norm": 0.8804931640625, + "learning_rate": 4.934884795612234e-06, + "loss": 0.6984, + "step": 620 + }, + { + "epoch": 0.5143567090005522, + "grad_norm": 0.8223545551300049, + "learning_rate": 4.934635209845458e-06, + "loss": 0.6892, + "step": 621 + }, + { + "epoch": 0.5151849806736609, + "grad_norm": 0.8509450554847717, + "learning_rate": 4.934385152999463e-06, + "loss": 0.7139, + "step": 622 + }, + { + "epoch": 0.5160132523467698, + "grad_norm": 0.8944748044013977, + "learning_rate": 4.934134625122636e-06, + "loss": 0.6976, + "step": 623 + }, + { + "epoch": 0.5168415240198785, + "grad_norm": 0.9021086692810059, + "learning_rate": 4.933883626263448e-06, + "loss": 0.7189, + "step": 624 + }, + { + "epoch": 0.5176697956929873, + "grad_norm": 0.8939547538757324, + "learning_rate": 4.933632156470467e-06, + "loss": 0.704, + "step": 625 + }, + { + "epoch": 0.518498067366096, + "grad_norm": 0.8793498277664185, + "learning_rate": 4.93338021579235e-06, + "loss": 0.7046, + "step": 626 + }, + { + "epoch": 0.5193263390392049, + "grad_norm": 0.8728756308555603, + "learning_rate": 4.933127804277845e-06, + "loss": 0.6826, + "step": 627 + }, + { + "epoch": 0.5201546107123136, + "grad_norm": 0.8989412188529968, + "learning_rate": 4.932874921975792e-06, + "loss": 0.6842, + "step": 628 + }, + { + "epoch": 0.5209828823854225, + "grad_norm": 0.8836191296577454, + "learning_rate": 4.93262156893512e-06, + "loss": 0.7061, + "step": 629 + }, + { + "epoch": 0.5218111540585312, + "grad_norm": 0.9429040551185608, + "learning_rate": 4.9323677452048505e-06, + "loss": 0.7053, + "step": 630 + }, + { + "epoch": 0.52263942573164, + "grad_norm": 0.879463255405426, + "learning_rate": 4.932113450834097e-06, + "loss": 0.6926, + "step": 631 + }, + { + "epoch": 0.5234676974047487, + "grad_norm": 0.8566175699234009, + "learning_rate": 4.931858685872062e-06, + "loss": 0.7126, + "step": 632 + }, + { + "epoch": 0.5242959690778576, + "grad_norm": 0.8694538474082947, + "learning_rate": 4.931603450368042e-06, + "loss": 0.7153, + "step": 633 + }, + { + "epoch": 0.5251242407509663, + "grad_norm": 0.8754484057426453, + "learning_rate": 4.931347744371421e-06, + "loss": 0.711, + "step": 634 + }, + { + "epoch": 0.5259525124240751, + "grad_norm": 0.8982204794883728, + "learning_rate": 4.931091567931675e-06, + "loss": 0.7168, + "step": 635 + }, + { + "epoch": 0.5267807840971839, + "grad_norm": 0.9262725710868835, + "learning_rate": 4.930834921098374e-06, + "loss": 0.7315, + "step": 636 + }, + { + "epoch": 0.5276090557702926, + "grad_norm": 0.9271771311759949, + "learning_rate": 4.930577803921176e-06, + "loss": 0.6968, + "step": 637 + }, + { + "epoch": 0.5284373274434014, + "grad_norm": 0.8855197429656982, + "learning_rate": 4.93032021644983e-06, + "loss": 0.6986, + "step": 638 + }, + { + "epoch": 0.5292655991165102, + "grad_norm": 0.8615745902061462, + "learning_rate": 4.930062158734178e-06, + "loss": 0.7014, + "step": 639 + }, + { + "epoch": 0.530093870789619, + "grad_norm": 0.9004185795783997, + "learning_rate": 4.92980363082415e-06, + "loss": 0.7045, + "step": 640 + }, + { + "epoch": 0.5309221424627277, + "grad_norm": 0.8533529043197632, + "learning_rate": 4.92954463276977e-06, + "loss": 0.6911, + "step": 641 + }, + { + "epoch": 0.5317504141358366, + "grad_norm": 0.881901204586029, + "learning_rate": 4.929285164621152e-06, + "loss": 0.7148, + "step": 642 + }, + { + "epoch": 0.5325786858089453, + "grad_norm": 0.8778982162475586, + "learning_rate": 4.9290252264285e-06, + "loss": 0.6879, + "step": 643 + }, + { + "epoch": 0.5334069574820541, + "grad_norm": 0.9455735087394714, + "learning_rate": 4.928764818242109e-06, + "loss": 0.7088, + "step": 644 + }, + { + "epoch": 0.5342352291551629, + "grad_norm": 0.9255897998809814, + "learning_rate": 4.928503940112367e-06, + "loss": 0.6817, + "step": 645 + }, + { + "epoch": 0.5350635008282717, + "grad_norm": 0.8854040503501892, + "learning_rate": 4.928242592089752e-06, + "loss": 0.7074, + "step": 646 + }, + { + "epoch": 0.5358917725013804, + "grad_norm": 0.8824400901794434, + "learning_rate": 4.927980774224829e-06, + "loss": 0.7092, + "step": 647 + }, + { + "epoch": 0.5367200441744893, + "grad_norm": 0.8873158693313599, + "learning_rate": 4.927718486568261e-06, + "loss": 0.7214, + "step": 648 + }, + { + "epoch": 0.537548315847598, + "grad_norm": 0.9013757109642029, + "learning_rate": 4.927455729170796e-06, + "loss": 0.6955, + "step": 649 + }, + { + "epoch": 0.5383765875207068, + "grad_norm": 0.9100024104118347, + "learning_rate": 4.927192502083276e-06, + "loss": 0.7089, + "step": 650 + }, + { + "epoch": 0.5392048591938156, + "grad_norm": 0.8935851454734802, + "learning_rate": 4.9269288053566335e-06, + "loss": 0.7018, + "step": 651 + }, + { + "epoch": 0.5400331308669244, + "grad_norm": 0.8726267218589783, + "learning_rate": 4.92666463904189e-06, + "loss": 0.6829, + "step": 652 + }, + { + "epoch": 0.5408614025400331, + "grad_norm": 0.9320974349975586, + "learning_rate": 4.926400003190161e-06, + "loss": 0.6766, + "step": 653 + }, + { + "epoch": 0.541689674213142, + "grad_norm": 0.9257067441940308, + "learning_rate": 4.926134897852648e-06, + "loss": 0.6913, + "step": 654 + }, + { + "epoch": 0.5425179458862507, + "grad_norm": 0.8642577528953552, + "learning_rate": 4.92586932308065e-06, + "loss": 0.7155, + "step": 655 + }, + { + "epoch": 0.5433462175593595, + "grad_norm": 0.8861536383628845, + "learning_rate": 4.9256032789255505e-06, + "loss": 0.6695, + "step": 656 + }, + { + "epoch": 0.5441744892324683, + "grad_norm": 0.8901777267456055, + "learning_rate": 4.925336765438829e-06, + "loss": 0.6847, + "step": 657 + }, + { + "epoch": 0.545002760905577, + "grad_norm": 0.8810168504714966, + "learning_rate": 4.9250697826720515e-06, + "loss": 0.7038, + "step": 658 + }, + { + "epoch": 0.5458310325786858, + "grad_norm": 0.8655433654785156, + "learning_rate": 4.924802330676877e-06, + "loss": 0.6908, + "step": 659 + }, + { + "epoch": 0.5466593042517945, + "grad_norm": 0.8786777853965759, + "learning_rate": 4.924534409505055e-06, + "loss": 0.663, + "step": 660 + }, + { + "epoch": 0.5474875759249034, + "grad_norm": 0.8837904334068298, + "learning_rate": 4.924266019208427e-06, + "loss": 0.6843, + "step": 661 + }, + { + "epoch": 0.5483158475980121, + "grad_norm": 0.8715303540229797, + "learning_rate": 4.923997159838923e-06, + "loss": 0.6841, + "step": 662 + }, + { + "epoch": 0.549144119271121, + "grad_norm": 0.8825753927230835, + "learning_rate": 4.923727831448564e-06, + "loss": 0.7053, + "step": 663 + }, + { + "epoch": 0.5499723909442297, + "grad_norm": 0.877668559551239, + "learning_rate": 4.923458034089463e-06, + "loss": 0.6624, + "step": 664 + }, + { + "epoch": 0.5508006626173385, + "grad_norm": 0.8721411824226379, + "learning_rate": 4.923187767813825e-06, + "loss": 0.7339, + "step": 665 + }, + { + "epoch": 0.5516289342904472, + "grad_norm": 0.8607308268547058, + "learning_rate": 4.922917032673941e-06, + "loss": 0.7184, + "step": 666 + }, + { + "epoch": 0.5524572059635561, + "grad_norm": 0.8456798791885376, + "learning_rate": 4.922645828722199e-06, + "loss": 0.7056, + "step": 667 + }, + { + "epoch": 0.5532854776366648, + "grad_norm": 0.8557775020599365, + "learning_rate": 4.922374156011071e-06, + "loss": 0.6942, + "step": 668 + }, + { + "epoch": 0.5541137493097736, + "grad_norm": 0.8877674341201782, + "learning_rate": 4.922102014593126e-06, + "loss": 0.7173, + "step": 669 + }, + { + "epoch": 0.5549420209828824, + "grad_norm": 0.8662481904029846, + "learning_rate": 4.921829404521019e-06, + "loss": 0.7113, + "step": 670 + }, + { + "epoch": 0.5557702926559912, + "grad_norm": 0.8887649774551392, + "learning_rate": 4.921556325847499e-06, + "loss": 0.7028, + "step": 671 + }, + { + "epoch": 0.5565985643290999, + "grad_norm": 0.9837973713874817, + "learning_rate": 4.921282778625403e-06, + "loss": 0.6975, + "step": 672 + }, + { + "epoch": 0.5574268360022088, + "grad_norm": 0.8754701018333435, + "learning_rate": 4.921008762907661e-06, + "loss": 0.6983, + "step": 673 + }, + { + "epoch": 0.5582551076753175, + "grad_norm": 0.9125426411628723, + "learning_rate": 4.920734278747291e-06, + "loss": 0.6982, + "step": 674 + }, + { + "epoch": 0.5590833793484263, + "grad_norm": 0.9424439072608948, + "learning_rate": 4.920459326197403e-06, + "loss": 0.7304, + "step": 675 + }, + { + "epoch": 0.5599116510215351, + "grad_norm": 0.8718393445014954, + "learning_rate": 4.9201839053112e-06, + "loss": 0.6644, + "step": 676 + }, + { + "epoch": 0.5607399226946438, + "grad_norm": 0.9002220630645752, + "learning_rate": 4.919908016141972e-06, + "loss": 0.6966, + "step": 677 + }, + { + "epoch": 0.5615681943677526, + "grad_norm": 0.8925678133964539, + "learning_rate": 4.9196316587431e-06, + "loss": 0.713, + "step": 678 + }, + { + "epoch": 0.5623964660408614, + "grad_norm": 0.8854464292526245, + "learning_rate": 4.919354833168059e-06, + "loss": 0.7075, + "step": 679 + }, + { + "epoch": 0.5632247377139702, + "grad_norm": 0.8867958188056946, + "learning_rate": 4.919077539470409e-06, + "loss": 0.7166, + "step": 680 + }, + { + "epoch": 0.5640530093870789, + "grad_norm": 0.862531840801239, + "learning_rate": 4.918799777703806e-06, + "loss": 0.7119, + "step": 681 + }, + { + "epoch": 0.5648812810601878, + "grad_norm": 0.8623220920562744, + "learning_rate": 4.9185215479219944e-06, + "loss": 0.6758, + "step": 682 + }, + { + "epoch": 0.5657095527332965, + "grad_norm": 0.9171786904335022, + "learning_rate": 4.918242850178808e-06, + "loss": 0.7049, + "step": 683 + }, + { + "epoch": 0.5665378244064053, + "grad_norm": 0.8947017192840576, + "learning_rate": 4.9179636845281735e-06, + "loss": 0.7144, + "step": 684 + }, + { + "epoch": 0.567366096079514, + "grad_norm": 0.9343388080596924, + "learning_rate": 4.917684051024104e-06, + "loss": 0.7024, + "step": 685 + }, + { + "epoch": 0.5681943677526229, + "grad_norm": 0.8929025530815125, + "learning_rate": 4.917403949720709e-06, + "loss": 0.7162, + "step": 686 + }, + { + "epoch": 0.5690226394257316, + "grad_norm": 0.9000421762466431, + "learning_rate": 4.9171233806721854e-06, + "loss": 0.7085, + "step": 687 + }, + { + "epoch": 0.5698509110988405, + "grad_norm": 0.9404364228248596, + "learning_rate": 4.91684234393282e-06, + "loss": 0.6996, + "step": 688 + }, + { + "epoch": 0.5706791827719492, + "grad_norm": 0.9566411375999451, + "learning_rate": 4.916560839556989e-06, + "loss": 0.7241, + "step": 689 + }, + { + "epoch": 0.571507454445058, + "grad_norm": 0.9066181182861328, + "learning_rate": 4.916278867599163e-06, + "loss": 0.724, + "step": 690 + }, + { + "epoch": 0.5723357261181667, + "grad_norm": 0.8914163708686829, + "learning_rate": 4.915996428113901e-06, + "loss": 0.7046, + "step": 691 + }, + { + "epoch": 0.5731639977912756, + "grad_norm": 0.8977548480033875, + "learning_rate": 4.91571352115585e-06, + "loss": 0.6823, + "step": 692 + }, + { + "epoch": 0.5739922694643843, + "grad_norm": 0.8902036547660828, + "learning_rate": 4.915430146779754e-06, + "loss": 0.702, + "step": 693 + }, + { + "epoch": 0.5748205411374931, + "grad_norm": 0.8859509229660034, + "learning_rate": 4.915146305040439e-06, + "loss": 0.6996, + "step": 694 + }, + { + "epoch": 0.5756488128106019, + "grad_norm": 0.8916864991188049, + "learning_rate": 4.914861995992829e-06, + "loss": 0.6978, + "step": 695 + }, + { + "epoch": 0.5764770844837107, + "grad_norm": 0.8961358070373535, + "learning_rate": 4.9145772196919336e-06, + "loss": 0.6803, + "step": 696 + }, + { + "epoch": 0.5773053561568194, + "grad_norm": 0.8817684650421143, + "learning_rate": 4.914291976192854e-06, + "loss": 0.6891, + "step": 697 + }, + { + "epoch": 0.5781336278299282, + "grad_norm": 0.8821113705635071, + "learning_rate": 4.914006265550783e-06, + "loss": 0.7009, + "step": 698 + }, + { + "epoch": 0.578961899503037, + "grad_norm": 0.9553062319755554, + "learning_rate": 4.913720087821003e-06, + "loss": 0.7118, + "step": 699 + }, + { + "epoch": 0.5797901711761457, + "grad_norm": 0.8997507691383362, + "learning_rate": 4.913433443058887e-06, + "loss": 0.7036, + "step": 700 + }, + { + "epoch": 0.5806184428492546, + "grad_norm": 0.8659473657608032, + "learning_rate": 4.9131463313198964e-06, + "loss": 0.6766, + "step": 701 + }, + { + "epoch": 0.5814467145223633, + "grad_norm": 0.8916030526161194, + "learning_rate": 4.912858752659586e-06, + "loss": 0.6976, + "step": 702 + }, + { + "epoch": 0.5822749861954721, + "grad_norm": 0.8726498484611511, + "learning_rate": 4.912570707133599e-06, + "loss": 0.6989, + "step": 703 + }, + { + "epoch": 0.5831032578685809, + "grad_norm": 0.8725529909133911, + "learning_rate": 4.91228219479767e-06, + "loss": 0.6862, + "step": 704 + }, + { + "epoch": 0.5839315295416897, + "grad_norm": 0.8807168006896973, + "learning_rate": 4.911993215707624e-06, + "loss": 0.6964, + "step": 705 + }, + { + "epoch": 0.5847598012147984, + "grad_norm": 0.9244167804718018, + "learning_rate": 4.911703769919375e-06, + "loss": 0.6742, + "step": 706 + }, + { + "epoch": 0.5855880728879073, + "grad_norm": 0.9174355268478394, + "learning_rate": 4.9114138574889285e-06, + "loss": 0.6852, + "step": 707 + }, + { + "epoch": 0.586416344561016, + "grad_norm": 0.8894474506378174, + "learning_rate": 4.9111234784723785e-06, + "loss": 0.689, + "step": 708 + }, + { + "epoch": 0.5872446162341248, + "grad_norm": 0.8856050968170166, + "learning_rate": 4.910832632925912e-06, + "loss": 0.6742, + "step": 709 + }, + { + "epoch": 0.5880728879072336, + "grad_norm": 0.9120550751686096, + "learning_rate": 4.910541320905804e-06, + "loss": 0.7016, + "step": 710 + }, + { + "epoch": 0.5889011595803424, + "grad_norm": 0.8925642967224121, + "learning_rate": 4.910249542468422e-06, + "loss": 0.7165, + "step": 711 + }, + { + "epoch": 0.5897294312534511, + "grad_norm": 0.8786419630050659, + "learning_rate": 4.909957297670222e-06, + "loss": 0.7207, + "step": 712 + }, + { + "epoch": 0.59055770292656, + "grad_norm": 0.89600670337677, + "learning_rate": 4.909664586567749e-06, + "loss": 0.672, + "step": 713 + }, + { + "epoch": 0.5913859745996687, + "grad_norm": 0.889835000038147, + "learning_rate": 4.909371409217642e-06, + "loss": 0.7089, + "step": 714 + }, + { + "epoch": 0.5922142462727775, + "grad_norm": 0.9029288291931152, + "learning_rate": 4.909077765676627e-06, + "loss": 0.7065, + "step": 715 + }, + { + "epoch": 0.5930425179458862, + "grad_norm": 0.9212722778320312, + "learning_rate": 4.908783656001521e-06, + "loss": 0.6787, + "step": 716 + }, + { + "epoch": 0.593870789618995, + "grad_norm": 0.9079681038856506, + "learning_rate": 4.9084890802492325e-06, + "loss": 0.6996, + "step": 717 + }, + { + "epoch": 0.5946990612921038, + "grad_norm": 0.9110735654830933, + "learning_rate": 4.908194038476758e-06, + "loss": 0.6962, + "step": 718 + }, + { + "epoch": 0.5955273329652125, + "grad_norm": 0.8945547938346863, + "learning_rate": 4.9078985307411865e-06, + "loss": 0.6648, + "step": 719 + }, + { + "epoch": 0.5963556046383214, + "grad_norm": 0.8831511735916138, + "learning_rate": 4.907602557099696e-06, + "loss": 0.7077, + "step": 720 + }, + { + "epoch": 0.5971838763114301, + "grad_norm": 0.9393531680107117, + "learning_rate": 4.9073061176095525e-06, + "loss": 0.6722, + "step": 721 + }, + { + "epoch": 0.5980121479845389, + "grad_norm": 0.8807207942008972, + "learning_rate": 4.907009212328117e-06, + "loss": 0.69, + "step": 722 + }, + { + "epoch": 0.5988404196576477, + "grad_norm": 0.8811002969741821, + "learning_rate": 4.906711841312836e-06, + "loss": 0.6902, + "step": 723 + }, + { + "epoch": 0.5996686913307565, + "grad_norm": 0.8953952789306641, + "learning_rate": 4.906414004621249e-06, + "loss": 0.6747, + "step": 724 + }, + { + "epoch": 0.6004969630038652, + "grad_norm": 0.8661350607872009, + "learning_rate": 4.9061157023109855e-06, + "loss": 0.7047, + "step": 725 + }, + { + "epoch": 0.6013252346769741, + "grad_norm": 0.8722146153450012, + "learning_rate": 4.905816934439762e-06, + "loss": 0.7231, + "step": 726 + }, + { + "epoch": 0.6021535063500828, + "grad_norm": 0.8933189511299133, + "learning_rate": 4.905517701065389e-06, + "loss": 0.7029, + "step": 727 + }, + { + "epoch": 0.6029817780231916, + "grad_norm": 0.8863442540168762, + "learning_rate": 4.905218002245764e-06, + "loss": 0.6886, + "step": 728 + }, + { + "epoch": 0.6038100496963004, + "grad_norm": 0.8672969341278076, + "learning_rate": 4.904917838038877e-06, + "loss": 0.7048, + "step": 729 + }, + { + "epoch": 0.6046383213694092, + "grad_norm": 0.8892956972122192, + "learning_rate": 4.904617208502807e-06, + "loss": 0.7009, + "step": 730 + }, + { + "epoch": 0.6054665930425179, + "grad_norm": 0.9083442687988281, + "learning_rate": 4.904316113695723e-06, + "loss": 0.7169, + "step": 731 + }, + { + "epoch": 0.6062948647156268, + "grad_norm": 0.8885907530784607, + "learning_rate": 4.904014553675883e-06, + "loss": 0.6939, + "step": 732 + }, + { + "epoch": 0.6071231363887355, + "grad_norm": 0.8859131336212158, + "learning_rate": 4.903712528501637e-06, + "loss": 0.6881, + "step": 733 + }, + { + "epoch": 0.6079514080618443, + "grad_norm": 0.874146580696106, + "learning_rate": 4.903410038231425e-06, + "loss": 0.6857, + "step": 734 + }, + { + "epoch": 0.6087796797349531, + "grad_norm": 0.8719701170921326, + "learning_rate": 4.903107082923774e-06, + "loss": 0.6834, + "step": 735 + }, + { + "epoch": 0.6096079514080619, + "grad_norm": 0.9656700491905212, + "learning_rate": 4.902803662637304e-06, + "loss": 0.6933, + "step": 736 + }, + { + "epoch": 0.6104362230811706, + "grad_norm": 0.9022018313407898, + "learning_rate": 4.9024997774307235e-06, + "loss": 0.683, + "step": 737 + }, + { + "epoch": 0.6112644947542794, + "grad_norm": 0.8694130182266235, + "learning_rate": 4.902195427362832e-06, + "loss": 0.642, + "step": 738 + }, + { + "epoch": 0.6120927664273882, + "grad_norm": 0.9153549075126648, + "learning_rate": 4.901890612492519e-06, + "loss": 0.6674, + "step": 739 + }, + { + "epoch": 0.6129210381004969, + "grad_norm": 0.9431132674217224, + "learning_rate": 4.901585332878762e-06, + "loss": 0.6918, + "step": 740 + }, + { + "epoch": 0.6137493097736058, + "grad_norm": 0.881587028503418, + "learning_rate": 4.901279588580631e-06, + "loss": 0.687, + "step": 741 + }, + { + "epoch": 0.6145775814467145, + "grad_norm": 0.9289383888244629, + "learning_rate": 4.9009733796572834e-06, + "loss": 0.6789, + "step": 742 + }, + { + "epoch": 0.6154058531198233, + "grad_norm": 0.9058337211608887, + "learning_rate": 4.900666706167968e-06, + "loss": 0.6856, + "step": 743 + }, + { + "epoch": 0.616234124792932, + "grad_norm": 0.8999397158622742, + "learning_rate": 4.900359568172024e-06, + "loss": 0.7159, + "step": 744 + }, + { + "epoch": 0.6170623964660409, + "grad_norm": 0.895321249961853, + "learning_rate": 4.90005196572888e-06, + "loss": 0.7071, + "step": 745 + }, + { + "epoch": 0.6178906681391496, + "grad_norm": 0.9142102599143982, + "learning_rate": 4.899743898898054e-06, + "loss": 0.6918, + "step": 746 + }, + { + "epoch": 0.6187189398122585, + "grad_norm": 0.8867097496986389, + "learning_rate": 4.899435367739152e-06, + "loss": 0.6991, + "step": 747 + }, + { + "epoch": 0.6195472114853672, + "grad_norm": 0.8594791293144226, + "learning_rate": 4.899126372311876e-06, + "loss": 0.7153, + "step": 748 + }, + { + "epoch": 0.620375483158476, + "grad_norm": 0.8740563988685608, + "learning_rate": 4.8988169126760085e-06, + "loss": 0.6786, + "step": 749 + }, + { + "epoch": 0.6212037548315847, + "grad_norm": 0.8818353414535522, + "learning_rate": 4.898506988891432e-06, + "loss": 0.6965, + "step": 750 + }, + { + "epoch": 0.6220320265046936, + "grad_norm": 0.9459843635559082, + "learning_rate": 4.898196601018111e-06, + "loss": 0.6821, + "step": 751 + }, + { + "epoch": 0.6228602981778023, + "grad_norm": 0.923457682132721, + "learning_rate": 4.897885749116104e-06, + "loss": 0.6822, + "step": 752 + }, + { + "epoch": 0.6236885698509111, + "grad_norm": 0.9342297315597534, + "learning_rate": 4.897574433245557e-06, + "loss": 0.7001, + "step": 753 + }, + { + "epoch": 0.6245168415240199, + "grad_norm": 0.9327735900878906, + "learning_rate": 4.897262653466706e-06, + "loss": 0.6921, + "step": 754 + }, + { + "epoch": 0.6253451131971287, + "grad_norm": 0.8876991868019104, + "learning_rate": 4.8969504098398805e-06, + "loss": 0.7087, + "step": 755 + }, + { + "epoch": 0.6261733848702374, + "grad_norm": 0.8983528017997742, + "learning_rate": 4.896637702425493e-06, + "loss": 0.6824, + "step": 756 + }, + { + "epoch": 0.6270016565433462, + "grad_norm": 0.9074965715408325, + "learning_rate": 4.896324531284053e-06, + "loss": 0.6533, + "step": 757 + }, + { + "epoch": 0.627829928216455, + "grad_norm": 0.8938930630683899, + "learning_rate": 4.896010896476153e-06, + "loss": 0.6859, + "step": 758 + }, + { + "epoch": 0.6286581998895637, + "grad_norm": 0.9036867022514343, + "learning_rate": 4.8956967980624806e-06, + "loss": 0.6904, + "step": 759 + }, + { + "epoch": 0.6294864715626726, + "grad_norm": 0.9162903428077698, + "learning_rate": 4.895382236103809e-06, + "loss": 0.6995, + "step": 760 + }, + { + "epoch": 0.6303147432357813, + "grad_norm": 1.0000487565994263, + "learning_rate": 4.895067210661005e-06, + "loss": 0.671, + "step": 761 + }, + { + "epoch": 0.6311430149088901, + "grad_norm": 0.895163893699646, + "learning_rate": 4.894751721795022e-06, + "loss": 0.6871, + "step": 762 + }, + { + "epoch": 0.6319712865819989, + "grad_norm": 0.9401512742042542, + "learning_rate": 4.894435769566905e-06, + "loss": 0.6811, + "step": 763 + }, + { + "epoch": 0.6327995582551077, + "grad_norm": 0.924578845500946, + "learning_rate": 4.894119354037787e-06, + "loss": 0.692, + "step": 764 + }, + { + "epoch": 0.6336278299282164, + "grad_norm": 0.9114750623703003, + "learning_rate": 4.893802475268892e-06, + "loss": 0.6977, + "step": 765 + }, + { + "epoch": 0.6344561016013253, + "grad_norm": 1.1472550630569458, + "learning_rate": 4.893485133321533e-06, + "loss": 0.679, + "step": 766 + }, + { + "epoch": 0.635284373274434, + "grad_norm": 0.9460884928703308, + "learning_rate": 4.893167328257111e-06, + "loss": 0.6988, + "step": 767 + }, + { + "epoch": 0.6361126449475428, + "grad_norm": 0.9040783643722534, + "learning_rate": 4.8928490601371216e-06, + "loss": 0.6706, + "step": 768 + }, + { + "epoch": 0.6369409166206516, + "grad_norm": 0.9785428047180176, + "learning_rate": 4.892530329023144e-06, + "loss": 0.7185, + "step": 769 + }, + { + "epoch": 0.6377691882937604, + "grad_norm": 0.9795847535133362, + "learning_rate": 4.892211134976851e-06, + "loss": 0.703, + "step": 770 + }, + { + "epoch": 0.6385974599668691, + "grad_norm": 0.9539455771446228, + "learning_rate": 4.8918914780600025e-06, + "loss": 0.6985, + "step": 771 + }, + { + "epoch": 0.639425731639978, + "grad_norm": 0.9304025173187256, + "learning_rate": 4.8915713583344505e-06, + "loss": 0.6731, + "step": 772 + }, + { + "epoch": 0.6402540033130867, + "grad_norm": 0.9466667771339417, + "learning_rate": 4.891250775862134e-06, + "loss": 0.6972, + "step": 773 + }, + { + "epoch": 0.6410822749861955, + "grad_norm": 0.9522298574447632, + "learning_rate": 4.890929730705084e-06, + "loss": 0.6793, + "step": 774 + }, + { + "epoch": 0.6419105466593042, + "grad_norm": 0.9077293276786804, + "learning_rate": 4.890608222925419e-06, + "loss": 0.6892, + "step": 775 + }, + { + "epoch": 0.6427388183324131, + "grad_norm": 0.8763419389724731, + "learning_rate": 4.890286252585348e-06, + "loss": 0.6985, + "step": 776 + }, + { + "epoch": 0.6435670900055218, + "grad_norm": 0.9089516401290894, + "learning_rate": 4.8899638197471685e-06, + "loss": 0.6778, + "step": 777 + }, + { + "epoch": 0.6443953616786305, + "grad_norm": 0.8733325600624084, + "learning_rate": 4.88964092447327e-06, + "loss": 0.6915, + "step": 778 + }, + { + "epoch": 0.6452236333517394, + "grad_norm": 0.9472198486328125, + "learning_rate": 4.889317566826128e-06, + "loss": 0.6984, + "step": 779 + }, + { + "epoch": 0.6460519050248481, + "grad_norm": 0.8729666471481323, + "learning_rate": 4.888993746868309e-06, + "loss": 0.6724, + "step": 780 + }, + { + "epoch": 0.6468801766979569, + "grad_norm": 0.9395233988761902, + "learning_rate": 4.88866946466247e-06, + "loss": 0.6697, + "step": 781 + }, + { + "epoch": 0.6477084483710657, + "grad_norm": 0.9158744812011719, + "learning_rate": 4.888344720271357e-06, + "loss": 0.6838, + "step": 782 + }, + { + "epoch": 0.6485367200441745, + "grad_norm": 0.8961113095283508, + "learning_rate": 4.888019513757804e-06, + "loss": 0.6488, + "step": 783 + }, + { + "epoch": 0.6493649917172832, + "grad_norm": 0.8911786079406738, + "learning_rate": 4.8876938451847364e-06, + "loss": 0.7089, + "step": 784 + }, + { + "epoch": 0.6501932633903921, + "grad_norm": 0.9214697480201721, + "learning_rate": 4.887367714615167e-06, + "loss": 0.6929, + "step": 785 + }, + { + "epoch": 0.6510215350635008, + "grad_norm": 0.9161441326141357, + "learning_rate": 4.8870411221122e-06, + "loss": 0.6761, + "step": 786 + }, + { + "epoch": 0.6518498067366096, + "grad_norm": 0.9161947965621948, + "learning_rate": 4.886714067739028e-06, + "loss": 0.6804, + "step": 787 + }, + { + "epoch": 0.6526780784097184, + "grad_norm": 0.8706585168838501, + "learning_rate": 4.886386551558933e-06, + "loss": 0.6836, + "step": 788 + }, + { + "epoch": 0.6535063500828272, + "grad_norm": 0.9037661552429199, + "learning_rate": 4.886058573635284e-06, + "loss": 0.6786, + "step": 789 + }, + { + "epoch": 0.6543346217559359, + "grad_norm": 0.8977832794189453, + "learning_rate": 4.885730134031545e-06, + "loss": 0.6938, + "step": 790 + }, + { + "epoch": 0.6551628934290448, + "grad_norm": 0.9106955528259277, + "learning_rate": 4.8854012328112645e-06, + "loss": 0.678, + "step": 791 + }, + { + "epoch": 0.6559911651021535, + "grad_norm": 0.8769396543502808, + "learning_rate": 4.885071870038082e-06, + "loss": 0.6969, + "step": 792 + }, + { + "epoch": 0.6568194367752623, + "grad_norm": 0.9946926832199097, + "learning_rate": 4.884742045775725e-06, + "loss": 0.6799, + "step": 793 + }, + { + "epoch": 0.6576477084483711, + "grad_norm": 0.9399042725563049, + "learning_rate": 4.884411760088012e-06, + "loss": 0.7038, + "step": 794 + }, + { + "epoch": 0.6584759801214799, + "grad_norm": 0.8914628624916077, + "learning_rate": 4.8840810130388526e-06, + "loss": 0.6867, + "step": 795 + }, + { + "epoch": 0.6593042517945886, + "grad_norm": 0.8900558352470398, + "learning_rate": 4.88374980469224e-06, + "loss": 0.6823, + "step": 796 + }, + { + "epoch": 0.6601325234676974, + "grad_norm": 0.8608913421630859, + "learning_rate": 4.8834181351122615e-06, + "loss": 0.6705, + "step": 797 + }, + { + "epoch": 0.6609607951408062, + "grad_norm": 0.900460422039032, + "learning_rate": 4.883086004363092e-06, + "loss": 0.7029, + "step": 798 + }, + { + "epoch": 0.6617890668139149, + "grad_norm": 0.8932703733444214, + "learning_rate": 4.882753412508996e-06, + "loss": 0.6845, + "step": 799 + }, + { + "epoch": 0.6626173384870238, + "grad_norm": 0.8737874627113342, + "learning_rate": 4.882420359614325e-06, + "loss": 0.6979, + "step": 800 + }, + { + "epoch": 0.6634456101601325, + "grad_norm": 0.8805396556854248, + "learning_rate": 4.882086845743524e-06, + "loss": 0.6935, + "step": 801 + }, + { + "epoch": 0.6642738818332413, + "grad_norm": 0.8952041268348694, + "learning_rate": 4.881752870961123e-06, + "loss": 0.7137, + "step": 802 + }, + { + "epoch": 0.66510215350635, + "grad_norm": 0.889869213104248, + "learning_rate": 4.881418435331745e-06, + "loss": 0.6643, + "step": 803 + }, + { + "epoch": 0.6659304251794589, + "grad_norm": 0.8912904858589172, + "learning_rate": 4.881083538920098e-06, + "loss": 0.6712, + "step": 804 + }, + { + "epoch": 0.6667586968525676, + "grad_norm": 0.8943006992340088, + "learning_rate": 4.880748181790984e-06, + "loss": 0.7001, + "step": 805 + }, + { + "epoch": 0.6675869685256764, + "grad_norm": 0.9042482376098633, + "learning_rate": 4.880412364009289e-06, + "loss": 0.6744, + "step": 806 + }, + { + "epoch": 0.6684152401987852, + "grad_norm": 0.9052988886833191, + "learning_rate": 4.880076085639991e-06, + "loss": 0.72, + "step": 807 + }, + { + "epoch": 0.669243511871894, + "grad_norm": 0.9702905416488647, + "learning_rate": 4.879739346748158e-06, + "loss": 0.6871, + "step": 808 + }, + { + "epoch": 0.6700717835450027, + "grad_norm": 0.8741766810417175, + "learning_rate": 4.879402147398944e-06, + "loss": 0.7174, + "step": 809 + }, + { + "epoch": 0.6709000552181116, + "grad_norm": 0.9226665496826172, + "learning_rate": 4.879064487657596e-06, + "loss": 0.6991, + "step": 810 + }, + { + "epoch": 0.6717283268912203, + "grad_norm": 0.878304123878479, + "learning_rate": 4.878726367589446e-06, + "loss": 0.6925, + "step": 811 + }, + { + "epoch": 0.6725565985643291, + "grad_norm": 0.893649160861969, + "learning_rate": 4.878387787259919e-06, + "loss": 0.6689, + "step": 812 + }, + { + "epoch": 0.6733848702374379, + "grad_norm": 0.917861819267273, + "learning_rate": 4.878048746734526e-06, + "loss": 0.7171, + "step": 813 + }, + { + "epoch": 0.6742131419105467, + "grad_norm": 0.879375696182251, + "learning_rate": 4.877709246078869e-06, + "loss": 0.6662, + "step": 814 + }, + { + "epoch": 0.6750414135836554, + "grad_norm": 0.9098542928695679, + "learning_rate": 4.877369285358637e-06, + "loss": 0.6793, + "step": 815 + }, + { + "epoch": 0.6758696852567642, + "grad_norm": 0.9465262293815613, + "learning_rate": 4.877028864639609e-06, + "loss": 0.7065, + "step": 816 + }, + { + "epoch": 0.676697956929873, + "grad_norm": 0.9029957056045532, + "learning_rate": 4.876687983987654e-06, + "loss": 0.7008, + "step": 817 + }, + { + "epoch": 0.6775262286029817, + "grad_norm": 0.9073920845985413, + "learning_rate": 4.876346643468731e-06, + "loss": 0.6836, + "step": 818 + }, + { + "epoch": 0.6783545002760906, + "grad_norm": 0.9028745889663696, + "learning_rate": 4.876004843148884e-06, + "loss": 0.6965, + "step": 819 + }, + { + "epoch": 0.6791827719491993, + "grad_norm": 0.8745572566986084, + "learning_rate": 4.875662583094249e-06, + "loss": 0.6974, + "step": 820 + }, + { + "epoch": 0.6800110436223081, + "grad_norm": 0.9203384518623352, + "learning_rate": 4.875319863371048e-06, + "loss": 0.686, + "step": 821 + }, + { + "epoch": 0.6808393152954169, + "grad_norm": 0.9087005853652954, + "learning_rate": 4.874976684045597e-06, + "loss": 0.683, + "step": 822 + }, + { + "epoch": 0.6816675869685257, + "grad_norm": 0.9007741808891296, + "learning_rate": 4.874633045184297e-06, + "loss": 0.6938, + "step": 823 + }, + { + "epoch": 0.6824958586416344, + "grad_norm": 0.9854089617729187, + "learning_rate": 4.874288946853639e-06, + "loss": 0.6945, + "step": 824 + }, + { + "epoch": 0.6833241303147433, + "grad_norm": 0.9044802784919739, + "learning_rate": 4.8739443891202025e-06, + "loss": 0.714, + "step": 825 + }, + { + "epoch": 0.684152401987852, + "grad_norm": 0.9388754963874817, + "learning_rate": 4.873599372050656e-06, + "loss": 0.6838, + "step": 826 + }, + { + "epoch": 0.6849806736609608, + "grad_norm": 0.9215866327285767, + "learning_rate": 4.873253895711757e-06, + "loss": 0.6994, + "step": 827 + }, + { + "epoch": 0.6858089453340696, + "grad_norm": 0.8786854147911072, + "learning_rate": 4.872907960170353e-06, + "loss": 0.6883, + "step": 828 + }, + { + "epoch": 0.6866372170071784, + "grad_norm": 0.9032605290412903, + "learning_rate": 4.872561565493379e-06, + "loss": 0.6847, + "step": 829 + }, + { + "epoch": 0.6874654886802871, + "grad_norm": 0.8940309882164001, + "learning_rate": 4.872214711747858e-06, + "loss": 0.6877, + "step": 830 + }, + { + "epoch": 0.688293760353396, + "grad_norm": 0.8703469634056091, + "learning_rate": 4.8718673990009045e-06, + "loss": 0.6829, + "step": 831 + }, + { + "epoch": 0.6891220320265047, + "grad_norm": 0.8618217706680298, + "learning_rate": 4.871519627319719e-06, + "loss": 0.6987, + "step": 832 + }, + { + "epoch": 0.6899503036996135, + "grad_norm": 0.8871967196464539, + "learning_rate": 4.871171396771594e-06, + "loss": 0.6824, + "step": 833 + }, + { + "epoch": 0.6907785753727222, + "grad_norm": 0.9162507057189941, + "learning_rate": 4.8708227074239064e-06, + "loss": 0.7158, + "step": 834 + }, + { + "epoch": 0.6916068470458311, + "grad_norm": 0.8829584717750549, + "learning_rate": 4.8704735593441255e-06, + "loss": 0.6714, + "step": 835 + }, + { + "epoch": 0.6924351187189398, + "grad_norm": 0.9130700826644897, + "learning_rate": 4.8701239525998076e-06, + "loss": 0.6841, + "step": 836 + }, + { + "epoch": 0.6932633903920485, + "grad_norm": 0.8706396222114563, + "learning_rate": 4.8697738872586e-06, + "loss": 0.7162, + "step": 837 + }, + { + "epoch": 0.6940916620651574, + "grad_norm": 0.8906163573265076, + "learning_rate": 4.869423363388235e-06, + "loss": 0.6586, + "step": 838 + }, + { + "epoch": 0.6949199337382661, + "grad_norm": 0.9134309887886047, + "learning_rate": 4.869072381056537e-06, + "loss": 0.6842, + "step": 839 + }, + { + "epoch": 0.6957482054113749, + "grad_norm": 0.869674563407898, + "learning_rate": 4.8687209403314175e-06, + "loss": 0.7049, + "step": 840 + }, + { + "epoch": 0.6965764770844837, + "grad_norm": 0.8405192494392395, + "learning_rate": 4.8683690412808775e-06, + "loss": 0.6677, + "step": 841 + }, + { + "epoch": 0.6974047487575925, + "grad_norm": 0.9173605442047119, + "learning_rate": 4.868016683973005e-06, + "loss": 0.6784, + "step": 842 + }, + { + "epoch": 0.6982330204307012, + "grad_norm": 0.9220597743988037, + "learning_rate": 4.867663868475978e-06, + "loss": 0.6774, + "step": 843 + }, + { + "epoch": 0.6990612921038101, + "grad_norm": 0.8687174320220947, + "learning_rate": 4.867310594858065e-06, + "loss": 0.7094, + "step": 844 + }, + { + "epoch": 0.6998895637769188, + "grad_norm": 0.9161131381988525, + "learning_rate": 4.866956863187619e-06, + "loss": 0.6867, + "step": 845 + }, + { + "epoch": 0.7007178354500276, + "grad_norm": 0.9338458776473999, + "learning_rate": 4.866602673533084e-06, + "loss": 0.6783, + "step": 846 + }, + { + "epoch": 0.7015461071231364, + "grad_norm": 0.9268187284469604, + "learning_rate": 4.866248025962993e-06, + "loss": 0.6681, + "step": 847 + }, + { + "epoch": 0.7023743787962452, + "grad_norm": 0.8843579292297363, + "learning_rate": 4.865892920545967e-06, + "loss": 0.6854, + "step": 848 + }, + { + "epoch": 0.7032026504693539, + "grad_norm": 0.889299213886261, + "learning_rate": 4.865537357350716e-06, + "loss": 0.7055, + "step": 849 + }, + { + "epoch": 0.7040309221424628, + "grad_norm": 0.9062727093696594, + "learning_rate": 4.865181336446036e-06, + "loss": 0.6806, + "step": 850 + }, + { + "epoch": 0.7048591938155715, + "grad_norm": 0.9007996320724487, + "learning_rate": 4.864824857900817e-06, + "loss": 0.6937, + "step": 851 + }, + { + "epoch": 0.7056874654886803, + "grad_norm": 0.8883542418479919, + "learning_rate": 4.864467921784032e-06, + "loss": 0.6957, + "step": 852 + }, + { + "epoch": 0.7065157371617891, + "grad_norm": 0.9422078132629395, + "learning_rate": 4.864110528164746e-06, + "loss": 0.7025, + "step": 853 + }, + { + "epoch": 0.7073440088348979, + "grad_norm": 0.8961433172225952, + "learning_rate": 4.86375267711211e-06, + "loss": 0.6795, + "step": 854 + }, + { + "epoch": 0.7081722805080066, + "grad_norm": 0.925815761089325, + "learning_rate": 4.863394368695366e-06, + "loss": 0.6824, + "step": 855 + }, + { + "epoch": 0.7090005521811154, + "grad_norm": 0.9326804280281067, + "learning_rate": 4.863035602983843e-06, + "loss": 0.6829, + "step": 856 + }, + { + "epoch": 0.7098288238542242, + "grad_norm": 0.8784539699554443, + "learning_rate": 4.862676380046959e-06, + "loss": 0.6776, + "step": 857 + }, + { + "epoch": 0.7106570955273329, + "grad_norm": 0.8577290177345276, + "learning_rate": 4.86231669995422e-06, + "loss": 0.6573, + "step": 858 + }, + { + "epoch": 0.7114853672004418, + "grad_norm": 0.9261177778244019, + "learning_rate": 4.86195656277522e-06, + "loss": 0.6891, + "step": 859 + }, + { + "epoch": 0.7123136388735505, + "grad_norm": 0.9008646607398987, + "learning_rate": 4.861595968579643e-06, + "loss": 0.6714, + "step": 860 + }, + { + "epoch": 0.7131419105466593, + "grad_norm": 0.8729333877563477, + "learning_rate": 4.861234917437261e-06, + "loss": 0.6819, + "step": 861 + }, + { + "epoch": 0.713970182219768, + "grad_norm": 0.9006103873252869, + "learning_rate": 4.860873409417934e-06, + "loss": 0.6702, + "step": 862 + }, + { + "epoch": 0.7147984538928769, + "grad_norm": 0.8699921369552612, + "learning_rate": 4.860511444591609e-06, + "loss": 0.6681, + "step": 863 + }, + { + "epoch": 0.7156267255659856, + "grad_norm": 0.9359689950942993, + "learning_rate": 4.860149023028325e-06, + "loss": 0.6834, + "step": 864 + }, + { + "epoch": 0.7164549972390944, + "grad_norm": 0.874977707862854, + "learning_rate": 4.859786144798205e-06, + "loss": 0.6692, + "step": 865 + }, + { + "epoch": 0.7172832689122032, + "grad_norm": 0.8821915984153748, + "learning_rate": 4.859422809971464e-06, + "loss": 0.6865, + "step": 866 + }, + { + "epoch": 0.718111540585312, + "grad_norm": 0.905248761177063, + "learning_rate": 4.859059018618405e-06, + "loss": 0.6705, + "step": 867 + }, + { + "epoch": 0.7189398122584207, + "grad_norm": 0.9134542346000671, + "learning_rate": 4.858694770809415e-06, + "loss": 0.7039, + "step": 868 + }, + { + "epoch": 0.7197680839315296, + "grad_norm": 0.9048055410385132, + "learning_rate": 4.858330066614975e-06, + "loss": 0.6848, + "step": 869 + }, + { + "epoch": 0.7205963556046383, + "grad_norm": 0.8780005574226379, + "learning_rate": 4.857964906105651e-06, + "loss": 0.689, + "step": 870 + }, + { + "epoch": 0.7214246272777471, + "grad_norm": 0.8826956152915955, + "learning_rate": 4.857599289352098e-06, + "loss": 0.6799, + "step": 871 + }, + { + "epoch": 0.7222528989508559, + "grad_norm": 0.919905960559845, + "learning_rate": 4.857233216425061e-06, + "loss": 0.6587, + "step": 872 + }, + { + "epoch": 0.7230811706239647, + "grad_norm": 0.8711065053939819, + "learning_rate": 4.8568666873953696e-06, + "loss": 0.6754, + "step": 873 + }, + { + "epoch": 0.7239094422970734, + "grad_norm": 0.8740426898002625, + "learning_rate": 4.856499702333944e-06, + "loss": 0.6837, + "step": 874 + }, + { + "epoch": 0.7247377139701823, + "grad_norm": 0.921416699886322, + "learning_rate": 4.8561322613117945e-06, + "loss": 0.6709, + "step": 875 + }, + { + "epoch": 0.725565985643291, + "grad_norm": 0.8988221287727356, + "learning_rate": 4.855764364400016e-06, + "loss": 0.7111, + "step": 876 + }, + { + "epoch": 0.7263942573163997, + "grad_norm": 0.8838291168212891, + "learning_rate": 4.8553960116697915e-06, + "loss": 0.6821, + "step": 877 + }, + { + "epoch": 0.7272225289895086, + "grad_norm": 0.9051623940467834, + "learning_rate": 4.8550272031923965e-06, + "loss": 0.6848, + "step": 878 + }, + { + "epoch": 0.7280508006626173, + "grad_norm": 0.9308082461357117, + "learning_rate": 4.85465793903919e-06, + "loss": 0.7087, + "step": 879 + }, + { + "epoch": 0.7288790723357261, + "grad_norm": 0.873214066028595, + "learning_rate": 4.854288219281623e-06, + "loss": 0.6877, + "step": 880 + }, + { + "epoch": 0.7297073440088349, + "grad_norm": 0.8890054225921631, + "learning_rate": 4.853918043991232e-06, + "loss": 0.6719, + "step": 881 + }, + { + "epoch": 0.7305356156819437, + "grad_norm": 0.9580398201942444, + "learning_rate": 4.853547413239642e-06, + "loss": 0.6439, + "step": 882 + }, + { + "epoch": 0.7313638873550524, + "grad_norm": 0.875035285949707, + "learning_rate": 4.853176327098567e-06, + "loss": 0.6645, + "step": 883 + }, + { + "epoch": 0.7321921590281613, + "grad_norm": 0.9299217462539673, + "learning_rate": 4.85280478563981e-06, + "loss": 0.707, + "step": 884 + }, + { + "epoch": 0.73302043070127, + "grad_norm": 0.8840609192848206, + "learning_rate": 4.8524327889352585e-06, + "loss": 0.6789, + "step": 885 + }, + { + "epoch": 0.7338487023743788, + "grad_norm": 0.8983805179595947, + "learning_rate": 4.852060337056891e-06, + "loss": 0.6814, + "step": 886 + }, + { + "epoch": 0.7346769740474876, + "grad_norm": 0.8730597496032715, + "learning_rate": 4.851687430076775e-06, + "loss": 0.7006, + "step": 887 + }, + { + "epoch": 0.7355052457205964, + "grad_norm": 0.8980515003204346, + "learning_rate": 4.851314068067063e-06, + "loss": 0.6729, + "step": 888 + }, + { + "epoch": 0.7363335173937051, + "grad_norm": 0.8919468522071838, + "learning_rate": 4.850940251099997e-06, + "loss": 0.6764, + "step": 889 + }, + { + "epoch": 0.737161789066814, + "grad_norm": 0.8987070918083191, + "learning_rate": 4.850565979247908e-06, + "loss": 0.6997, + "step": 890 + }, + { + "epoch": 0.7379900607399227, + "grad_norm": 0.9056184887886047, + "learning_rate": 4.850191252583214e-06, + "loss": 0.6887, + "step": 891 + }, + { + "epoch": 0.7388183324130315, + "grad_norm": 0.8714621067047119, + "learning_rate": 4.84981607117842e-06, + "loss": 0.6946, + "step": 892 + }, + { + "epoch": 0.7396466040861402, + "grad_norm": 0.8845617175102234, + "learning_rate": 4.849440435106122e-06, + "loss": 0.6845, + "step": 893 + }, + { + "epoch": 0.7404748757592491, + "grad_norm": 0.8926048278808594, + "learning_rate": 4.849064344439e-06, + "loss": 0.6922, + "step": 894 + }, + { + "epoch": 0.7413031474323578, + "grad_norm": 0.8731569647789001, + "learning_rate": 4.848687799249826e-06, + "loss": 0.6837, + "step": 895 + }, + { + "epoch": 0.7421314191054665, + "grad_norm": 0.8801183104515076, + "learning_rate": 4.848310799611457e-06, + "loss": 0.6932, + "step": 896 + }, + { + "epoch": 0.7429596907785754, + "grad_norm": 0.8818139433860779, + "learning_rate": 4.847933345596838e-06, + "loss": 0.6758, + "step": 897 + }, + { + "epoch": 0.7437879624516841, + "grad_norm": 0.9117923974990845, + "learning_rate": 4.847555437279004e-06, + "loss": 0.6628, + "step": 898 + }, + { + "epoch": 0.7446162341247929, + "grad_norm": 0.891360342502594, + "learning_rate": 4.847177074731078e-06, + "loss": 0.6876, + "step": 899 + }, + { + "epoch": 0.7454445057979017, + "grad_norm": 0.8905263543128967, + "learning_rate": 4.846798258026267e-06, + "loss": 0.6862, + "step": 900 + }, + { + "epoch": 0.7462727774710105, + "grad_norm": 0.8853707909584045, + "learning_rate": 4.846418987237868e-06, + "loss": 0.654, + "step": 901 + }, + { + "epoch": 0.7471010491441192, + "grad_norm": 0.8782089948654175, + "learning_rate": 4.846039262439271e-06, + "loss": 0.7005, + "step": 902 + }, + { + "epoch": 0.7479293208172281, + "grad_norm": 0.8774383068084717, + "learning_rate": 4.8456590837039445e-06, + "loss": 0.7044, + "step": 903 + }, + { + "epoch": 0.7487575924903368, + "grad_norm": 0.9385423064231873, + "learning_rate": 4.845278451105451e-06, + "loss": 0.676, + "step": 904 + }, + { + "epoch": 0.7495858641634456, + "grad_norm": 0.9211432337760925, + "learning_rate": 4.8448973647174405e-06, + "loss": 0.697, + "step": 905 + }, + { + "epoch": 0.7504141358365544, + "grad_norm": 0.887610912322998, + "learning_rate": 4.844515824613649e-06, + "loss": 0.6965, + "step": 906 + }, + { + "epoch": 0.7512424075096632, + "grad_norm": 0.885761559009552, + "learning_rate": 4.8441338308679e-06, + "loss": 0.6851, + "step": 907 + }, + { + "epoch": 0.7520706791827719, + "grad_norm": 0.9209930300712585, + "learning_rate": 4.843751383554106e-06, + "loss": 0.6718, + "step": 908 + }, + { + "epoch": 0.7528989508558808, + "grad_norm": 0.9308252334594727, + "learning_rate": 4.843368482746269e-06, + "loss": 0.6802, + "step": 909 + }, + { + "epoch": 0.7537272225289895, + "grad_norm": 0.8931897282600403, + "learning_rate": 4.842985128518473e-06, + "loss": 0.7076, + "step": 910 + }, + { + "epoch": 0.7545554942020983, + "grad_norm": 0.9102733731269836, + "learning_rate": 4.842601320944898e-06, + "loss": 0.6898, + "step": 911 + }, + { + "epoch": 0.7553837658752071, + "grad_norm": 0.9336893558502197, + "learning_rate": 4.842217060099804e-06, + "loss": 0.6709, + "step": 912 + }, + { + "epoch": 0.7562120375483159, + "grad_norm": 0.9366991519927979, + "learning_rate": 4.8418323460575425e-06, + "loss": 0.6665, + "step": 913 + }, + { + "epoch": 0.7570403092214246, + "grad_norm": 0.9371415376663208, + "learning_rate": 4.841447178892552e-06, + "loss": 0.7013, + "step": 914 + }, + { + "epoch": 0.7578685808945335, + "grad_norm": 1.031123399734497, + "learning_rate": 4.84106155867936e-06, + "loss": 0.687, + "step": 915 + }, + { + "epoch": 0.7586968525676422, + "grad_norm": 0.9319218397140503, + "learning_rate": 4.8406754854925785e-06, + "loss": 0.669, + "step": 916 + }, + { + "epoch": 0.7595251242407509, + "grad_norm": 0.8563560843467712, + "learning_rate": 4.840288959406911e-06, + "loss": 0.6905, + "step": 917 + }, + { + "epoch": 0.7603533959138598, + "grad_norm": 0.9424378275871277, + "learning_rate": 4.839901980497145e-06, + "loss": 0.6685, + "step": 918 + }, + { + "epoch": 0.7611816675869685, + "grad_norm": 0.9202269315719604, + "learning_rate": 4.839514548838158e-06, + "loss": 0.6728, + "step": 919 + }, + { + "epoch": 0.7620099392600773, + "grad_norm": 0.9152582287788391, + "learning_rate": 4.839126664504915e-06, + "loss": 0.6842, + "step": 920 + }, + { + "epoch": 0.762838210933186, + "grad_norm": 0.9457129836082458, + "learning_rate": 4.838738327572468e-06, + "loss": 0.6822, + "step": 921 + }, + { + "epoch": 0.7636664826062949, + "grad_norm": 0.93124920129776, + "learning_rate": 4.838349538115955e-06, + "loss": 0.691, + "step": 922 + }, + { + "epoch": 0.7644947542794036, + "grad_norm": 0.9249301552772522, + "learning_rate": 4.837960296210605e-06, + "loss": 0.6574, + "step": 923 + }, + { + "epoch": 0.7653230259525124, + "grad_norm": 1.0551042556762695, + "learning_rate": 4.837570601931731e-06, + "loss": 0.6782, + "step": 924 + }, + { + "epoch": 0.7661512976256212, + "grad_norm": 0.9404964447021484, + "learning_rate": 4.8371804553547365e-06, + "loss": 0.6703, + "step": 925 + }, + { + "epoch": 0.76697956929873, + "grad_norm": 0.909015417098999, + "learning_rate": 4.836789856555111e-06, + "loss": 0.6812, + "step": 926 + }, + { + "epoch": 0.7678078409718387, + "grad_norm": 0.8572261929512024, + "learning_rate": 4.836398805608431e-06, + "loss": 0.6831, + "step": 927 + }, + { + "epoch": 0.7686361126449476, + "grad_norm": 0.951753556728363, + "learning_rate": 4.836007302590362e-06, + "loss": 0.6672, + "step": 928 + }, + { + "epoch": 0.7694643843180563, + "grad_norm": 0.8875932693481445, + "learning_rate": 4.8356153475766564e-06, + "loss": 0.6819, + "step": 929 + }, + { + "epoch": 0.7702926559911651, + "grad_norm": 0.9405681490898132, + "learning_rate": 4.835222940643153e-06, + "loss": 0.6917, + "step": 930 + }, + { + "epoch": 0.7711209276642739, + "grad_norm": 0.942520022392273, + "learning_rate": 4.834830081865779e-06, + "loss": 0.6657, + "step": 931 + }, + { + "epoch": 0.7719491993373827, + "grad_norm": 0.9063259363174438, + "learning_rate": 4.83443677132055e-06, + "loss": 0.6857, + "step": 932 + }, + { + "epoch": 0.7727774710104914, + "grad_norm": 0.8778199553489685, + "learning_rate": 4.834043009083566e-06, + "loss": 0.7028, + "step": 933 + }, + { + "epoch": 0.7736057426836003, + "grad_norm": 0.9215186834335327, + "learning_rate": 4.833648795231018e-06, + "loss": 0.6712, + "step": 934 + }, + { + "epoch": 0.774434014356709, + "grad_norm": 0.8992031812667847, + "learning_rate": 4.8332541298391825e-06, + "loss": 0.6827, + "step": 935 + }, + { + "epoch": 0.7752622860298177, + "grad_norm": 0.9203833341598511, + "learning_rate": 4.8328590129844224e-06, + "loss": 0.6835, + "step": 936 + }, + { + "epoch": 0.7760905577029266, + "grad_norm": 0.9109310507774353, + "learning_rate": 4.832463444743191e-06, + "loss": 0.673, + "step": 937 + }, + { + "epoch": 0.7769188293760353, + "grad_norm": 0.978349506855011, + "learning_rate": 4.832067425192024e-06, + "loss": 0.6638, + "step": 938 + }, + { + "epoch": 0.7777471010491441, + "grad_norm": 0.9552261829376221, + "learning_rate": 4.831670954407551e-06, + "loss": 0.6919, + "step": 939 + }, + { + "epoch": 0.7785753727222529, + "grad_norm": 0.9106539487838745, + "learning_rate": 4.831274032466484e-06, + "loss": 0.6964, + "step": 940 + }, + { + "epoch": 0.7794036443953617, + "grad_norm": 0.9195299744606018, + "learning_rate": 4.830876659445623e-06, + "loss": 0.6793, + "step": 941 + }, + { + "epoch": 0.7802319160684704, + "grad_norm": 0.9075512290000916, + "learning_rate": 4.8304788354218564e-06, + "loss": 0.6649, + "step": 942 + }, + { + "epoch": 0.7810601877415793, + "grad_norm": 0.9142526984214783, + "learning_rate": 4.83008056047216e-06, + "loss": 0.6811, + "step": 943 + }, + { + "epoch": 0.781888459414688, + "grad_norm": 0.8898900151252747, + "learning_rate": 4.829681834673597e-06, + "loss": 0.6906, + "step": 944 + }, + { + "epoch": 0.7827167310877968, + "grad_norm": 0.8884983062744141, + "learning_rate": 4.8292826581033135e-06, + "loss": 0.6903, + "step": 945 + }, + { + "epoch": 0.7835450027609056, + "grad_norm": 0.9075689315795898, + "learning_rate": 4.828883030838551e-06, + "loss": 0.6794, + "step": 946 + }, + { + "epoch": 0.7843732744340144, + "grad_norm": 0.8970300555229187, + "learning_rate": 4.828482952956631e-06, + "loss": 0.6715, + "step": 947 + }, + { + "epoch": 0.7852015461071231, + "grad_norm": 0.8906447887420654, + "learning_rate": 4.828082424534965e-06, + "loss": 0.6878, + "step": 948 + }, + { + "epoch": 0.786029817780232, + "grad_norm": 0.9083748459815979, + "learning_rate": 4.827681445651052e-06, + "loss": 0.6787, + "step": 949 + }, + { + "epoch": 0.7868580894533407, + "grad_norm": 0.8975276947021484, + "learning_rate": 4.8272800163824784e-06, + "loss": 0.687, + "step": 950 + }, + { + "epoch": 0.7876863611264495, + "grad_norm": 0.9573839902877808, + "learning_rate": 4.8268781368069156e-06, + "loss": 0.6996, + "step": 951 + }, + { + "epoch": 0.7885146327995582, + "grad_norm": 0.913536787033081, + "learning_rate": 4.826475807002125e-06, + "loss": 0.6844, + "step": 952 + }, + { + "epoch": 0.7893429044726671, + "grad_norm": 0.9202503561973572, + "learning_rate": 4.826073027045952e-06, + "loss": 0.6702, + "step": 953 + }, + { + "epoch": 0.7901711761457758, + "grad_norm": 0.8800237774848938, + "learning_rate": 4.8256697970163315e-06, + "loss": 0.685, + "step": 954 + }, + { + "epoch": 0.7909994478188846, + "grad_norm": 0.8923048377037048, + "learning_rate": 4.825266116991285e-06, + "loss": 0.6839, + "step": 955 + }, + { + "epoch": 0.7918277194919934, + "grad_norm": 0.879743754863739, + "learning_rate": 4.824861987048922e-06, + "loss": 0.7112, + "step": 956 + }, + { + "epoch": 0.7926559911651021, + "grad_norm": 0.8575169444084167, + "learning_rate": 4.824457407267436e-06, + "loss": 0.671, + "step": 957 + }, + { + "epoch": 0.7934842628382109, + "grad_norm": 0.8851709961891174, + "learning_rate": 4.824052377725109e-06, + "loss": 0.669, + "step": 958 + }, + { + "epoch": 0.7943125345113197, + "grad_norm": 0.9084329605102539, + "learning_rate": 4.823646898500311e-06, + "loss": 0.6731, + "step": 959 + }, + { + "epoch": 0.7951408061844285, + "grad_norm": 0.908256471157074, + "learning_rate": 4.823240969671501e-06, + "loss": 0.688, + "step": 960 + }, + { + "epoch": 0.7959690778575372, + "grad_norm": 0.8993384838104248, + "learning_rate": 4.822834591317219e-06, + "loss": 0.6736, + "step": 961 + }, + { + "epoch": 0.7967973495306461, + "grad_norm": 0.893530011177063, + "learning_rate": 4.822427763516098e-06, + "loss": 0.6787, + "step": 962 + }, + { + "epoch": 0.7976256212037548, + "grad_norm": 0.9479735493659973, + "learning_rate": 4.822020486346852e-06, + "loss": 0.6842, + "step": 963 + }, + { + "epoch": 0.7984538928768636, + "grad_norm": 0.8716689348220825, + "learning_rate": 4.821612759888289e-06, + "loss": 0.6726, + "step": 964 + }, + { + "epoch": 0.7992821645499724, + "grad_norm": 0.8902738690376282, + "learning_rate": 4.821204584219299e-06, + "loss": 0.6672, + "step": 965 + }, + { + "epoch": 0.8001104362230812, + "grad_norm": 0.8713969588279724, + "learning_rate": 4.820795959418859e-06, + "loss": 0.6783, + "step": 966 + }, + { + "epoch": 0.8009387078961899, + "grad_norm": 0.891788899898529, + "learning_rate": 4.820386885566036e-06, + "loss": 0.6691, + "step": 967 + }, + { + "epoch": 0.8017669795692988, + "grad_norm": 0.8484700918197632, + "learning_rate": 4.81997736273998e-06, + "loss": 0.6766, + "step": 968 + }, + { + "epoch": 0.8025952512424075, + "grad_norm": 0.8931223154067993, + "learning_rate": 4.819567391019931e-06, + "loss": 0.6944, + "step": 969 + }, + { + "epoch": 0.8034235229155163, + "grad_norm": 0.9707916975021362, + "learning_rate": 4.819156970485216e-06, + "loss": 0.7076, + "step": 970 + }, + { + "epoch": 0.8042517945886251, + "grad_norm": 0.9172869920730591, + "learning_rate": 4.818746101215245e-06, + "loss": 0.6864, + "step": 971 + }, + { + "epoch": 0.8050800662617339, + "grad_norm": 0.8982006311416626, + "learning_rate": 4.818334783289518e-06, + "loss": 0.6685, + "step": 972 + }, + { + "epoch": 0.8059083379348426, + "grad_norm": 0.8960995078086853, + "learning_rate": 4.817923016787623e-06, + "loss": 0.6959, + "step": 973 + }, + { + "epoch": 0.8067366096079515, + "grad_norm": 0.8806949257850647, + "learning_rate": 4.81751080178923e-06, + "loss": 0.7028, + "step": 974 + }, + { + "epoch": 0.8075648812810602, + "grad_norm": 0.8798978924751282, + "learning_rate": 4.817098138374102e-06, + "loss": 0.6718, + "step": 975 + }, + { + "epoch": 0.8083931529541689, + "grad_norm": 0.9217897653579712, + "learning_rate": 4.8166850266220835e-06, + "loss": 0.6782, + "step": 976 + }, + { + "epoch": 0.8092214246272778, + "grad_norm": 0.8915667533874512, + "learning_rate": 4.8162714666131074e-06, + "loss": 0.6917, + "step": 977 + }, + { + "epoch": 0.8100496963003865, + "grad_norm": 0.8895093202590942, + "learning_rate": 4.815857458427195e-06, + "loss": 0.698, + "step": 978 + }, + { + "epoch": 0.8108779679734953, + "grad_norm": 0.9167448282241821, + "learning_rate": 4.815443002144453e-06, + "loss": 0.6796, + "step": 979 + }, + { + "epoch": 0.811706239646604, + "grad_norm": 0.913827657699585, + "learning_rate": 4.815028097845074e-06, + "loss": 0.6731, + "step": 980 + }, + { + "epoch": 0.8125345113197129, + "grad_norm": 0.9183399081230164, + "learning_rate": 4.814612745609338e-06, + "loss": 0.7076, + "step": 981 + }, + { + "epoch": 0.8133627829928216, + "grad_norm": 0.8812996745109558, + "learning_rate": 4.814196945517613e-06, + "loss": 0.6416, + "step": 982 + }, + { + "epoch": 0.8141910546659304, + "grad_norm": 0.9032955169677734, + "learning_rate": 4.813780697650351e-06, + "loss": 0.6799, + "step": 983 + }, + { + "epoch": 0.8150193263390392, + "grad_norm": 0.9380100965499878, + "learning_rate": 4.813364002088093e-06, + "loss": 0.6771, + "step": 984 + }, + { + "epoch": 0.815847598012148, + "grad_norm": 0.9470316171646118, + "learning_rate": 4.8129468589114655e-06, + "loss": 0.6851, + "step": 985 + }, + { + "epoch": 0.8166758696852567, + "grad_norm": 0.9401133060455322, + "learning_rate": 4.812529268201183e-06, + "loss": 0.676, + "step": 986 + }, + { + "epoch": 0.8175041413583656, + "grad_norm": 0.9379092454910278, + "learning_rate": 4.812111230038043e-06, + "loss": 0.672, + "step": 987 + }, + { + "epoch": 0.8183324130314743, + "grad_norm": 0.9591637253761292, + "learning_rate": 4.811692744502933e-06, + "loss": 0.6684, + "step": 988 + }, + { + "epoch": 0.8191606847045831, + "grad_norm": 0.8911624550819397, + "learning_rate": 4.8112738116768276e-06, + "loss": 0.6448, + "step": 989 + }, + { + "epoch": 0.8199889563776919, + "grad_norm": 0.9036282896995544, + "learning_rate": 4.810854431640784e-06, + "loss": 0.703, + "step": 990 + }, + { + "epoch": 0.8208172280508007, + "grad_norm": 0.9197635650634766, + "learning_rate": 4.81043460447595e-06, + "loss": 0.6655, + "step": 991 + }, + { + "epoch": 0.8216454997239094, + "grad_norm": 0.9548643827438354, + "learning_rate": 4.8100143302635575e-06, + "loss": 0.6892, + "step": 992 + }, + { + "epoch": 0.8224737713970183, + "grad_norm": 0.9139766097068787, + "learning_rate": 4.809593609084925e-06, + "loss": 0.6723, + "step": 993 + }, + { + "epoch": 0.823302043070127, + "grad_norm": 0.9067809581756592, + "learning_rate": 4.8091724410214595e-06, + "loss": 0.6649, + "step": 994 + }, + { + "epoch": 0.8241303147432358, + "grad_norm": 0.9705583453178406, + "learning_rate": 4.808750826154652e-06, + "loss": 0.6889, + "step": 995 + }, + { + "epoch": 0.8249585864163446, + "grad_norm": 0.9664806127548218, + "learning_rate": 4.808328764566082e-06, + "loss": 0.6758, + "step": 996 + }, + { + "epoch": 0.8257868580894533, + "grad_norm": 0.8840530514717102, + "learning_rate": 4.807906256337414e-06, + "loss": 0.6719, + "step": 997 + }, + { + "epoch": 0.8266151297625621, + "grad_norm": 0.8803462386131287, + "learning_rate": 4.8074833015504e-06, + "loss": 0.6903, + "step": 998 + }, + { + "epoch": 0.8274434014356709, + "grad_norm": 0.9404571652412415, + "learning_rate": 4.8070599002868756e-06, + "loss": 0.6917, + "step": 999 + }, + { + "epoch": 0.8282716731087797, + "grad_norm": 1.0206997394561768, + "learning_rate": 4.806636052628768e-06, + "loss": 0.6854, + "step": 1000 + }, + { + "epoch": 0.8290999447818884, + "grad_norm": 0.9294790029525757, + "learning_rate": 4.806211758658086e-06, + "loss": 0.6791, + "step": 1001 + }, + { + "epoch": 0.8299282164549973, + "grad_norm": 0.874011754989624, + "learning_rate": 4.805787018456927e-06, + "loss": 0.6915, + "step": 1002 + }, + { + "epoch": 0.830756488128106, + "grad_norm": 0.9098061323165894, + "learning_rate": 4.805361832107474e-06, + "loss": 0.6643, + "step": 1003 + }, + { + "epoch": 0.8315847598012148, + "grad_norm": 0.9783986210823059, + "learning_rate": 4.804936199691997e-06, + "loss": 0.6853, + "step": 1004 + }, + { + "epoch": 0.8324130314743236, + "grad_norm": 0.9316959381103516, + "learning_rate": 4.804510121292852e-06, + "loss": 0.6696, + "step": 1005 + }, + { + "epoch": 0.8332413031474324, + "grad_norm": 0.8930666446685791, + "learning_rate": 4.80408359699248e-06, + "loss": 0.6805, + "step": 1006 + }, + { + "epoch": 0.8340695748205411, + "grad_norm": 0.9417555332183838, + "learning_rate": 4.803656626873412e-06, + "loss": 0.66, + "step": 1007 + }, + { + "epoch": 0.83489784649365, + "grad_norm": 0.9435075521469116, + "learning_rate": 4.80322921101826e-06, + "loss": 0.6664, + "step": 1008 + }, + { + "epoch": 0.8357261181667587, + "grad_norm": 0.9090076684951782, + "learning_rate": 4.8028013495097255e-06, + "loss": 0.6595, + "step": 1009 + }, + { + "epoch": 0.8365543898398675, + "grad_norm": 0.9024788737297058, + "learning_rate": 4.802373042430597e-06, + "loss": 0.6672, + "step": 1010 + }, + { + "epoch": 0.8373826615129762, + "grad_norm": 0.899813711643219, + "learning_rate": 4.801944289863747e-06, + "loss": 0.6743, + "step": 1011 + }, + { + "epoch": 0.8382109331860851, + "grad_norm": 0.887106716632843, + "learning_rate": 4.801515091892136e-06, + "loss": 0.6823, + "step": 1012 + }, + { + "epoch": 0.8390392048591938, + "grad_norm": 0.8690978288650513, + "learning_rate": 4.801085448598809e-06, + "loss": 0.6642, + "step": 1013 + }, + { + "epoch": 0.8398674765323026, + "grad_norm": 0.9131479859352112, + "learning_rate": 4.800655360066898e-06, + "loss": 0.6841, + "step": 1014 + }, + { + "epoch": 0.8406957482054114, + "grad_norm": 0.9031343460083008, + "learning_rate": 4.800224826379622e-06, + "loss": 0.6879, + "step": 1015 + }, + { + "epoch": 0.8415240198785201, + "grad_norm": 0.8914244174957275, + "learning_rate": 4.799793847620283e-06, + "loss": 0.6562, + "step": 1016 + }, + { + "epoch": 0.8423522915516289, + "grad_norm": 0.8692408800125122, + "learning_rate": 4.799362423872274e-06, + "loss": 0.6759, + "step": 1017 + }, + { + "epoch": 0.8431805632247377, + "grad_norm": 0.9511806964874268, + "learning_rate": 4.798930555219069e-06, + "loss": 0.6637, + "step": 1018 + }, + { + "epoch": 0.8440088348978465, + "grad_norm": 0.9556595683097839, + "learning_rate": 4.7984982417442325e-06, + "loss": 0.6925, + "step": 1019 + }, + { + "epoch": 0.8448371065709552, + "grad_norm": 0.9107236862182617, + "learning_rate": 4.798065483531413e-06, + "loss": 0.67, + "step": 1020 + }, + { + "epoch": 0.8456653782440641, + "grad_norm": 0.875293493270874, + "learning_rate": 4.797632280664345e-06, + "loss": 0.6755, + "step": 1021 + }, + { + "epoch": 0.8464936499171728, + "grad_norm": 0.8972271084785461, + "learning_rate": 4.797198633226849e-06, + "loss": 0.6637, + "step": 1022 + }, + { + "epoch": 0.8473219215902816, + "grad_norm": 0.8903595805168152, + "learning_rate": 4.796764541302831e-06, + "loss": 0.6997, + "step": 1023 + }, + { + "epoch": 0.8481501932633904, + "grad_norm": 0.8948636651039124, + "learning_rate": 4.796330004976285e-06, + "loss": 0.6579, + "step": 1024 + }, + { + "epoch": 0.8489784649364992, + "grad_norm": 0.8590925335884094, + "learning_rate": 4.795895024331289e-06, + "loss": 0.6651, + "step": 1025 + }, + { + "epoch": 0.8498067366096079, + "grad_norm": 0.870811402797699, + "learning_rate": 4.795459599452007e-06, + "loss": 0.6707, + "step": 1026 + }, + { + "epoch": 0.8506350082827168, + "grad_norm": 0.8807471394538879, + "learning_rate": 4.795023730422692e-06, + "loss": 0.6787, + "step": 1027 + }, + { + "epoch": 0.8514632799558255, + "grad_norm": 0.8817671537399292, + "learning_rate": 4.794587417327678e-06, + "loss": 0.66, + "step": 1028 + }, + { + "epoch": 0.8522915516289343, + "grad_norm": 0.8924570679664612, + "learning_rate": 4.794150660251389e-06, + "loss": 0.6842, + "step": 1029 + }, + { + "epoch": 0.8531198233020431, + "grad_norm": 0.8746559619903564, + "learning_rate": 4.793713459278333e-06, + "loss": 0.6582, + "step": 1030 + }, + { + "epoch": 0.8539480949751519, + "grad_norm": 0.8885277509689331, + "learning_rate": 4.793275814493104e-06, + "loss": 0.6784, + "step": 1031 + }, + { + "epoch": 0.8547763666482606, + "grad_norm": 0.8797710537910461, + "learning_rate": 4.792837725980383e-06, + "loss": 0.6638, + "step": 1032 + }, + { + "epoch": 0.8556046383213695, + "grad_norm": 0.8932235240936279, + "learning_rate": 4.7923991938249355e-06, + "loss": 0.6918, + "step": 1033 + }, + { + "epoch": 0.8564329099944782, + "grad_norm": 0.9317487478256226, + "learning_rate": 4.791960218111613e-06, + "loss": 0.6905, + "step": 1034 + }, + { + "epoch": 0.8572611816675869, + "grad_norm": 0.8760019540786743, + "learning_rate": 4.791520798925354e-06, + "loss": 0.673, + "step": 1035 + }, + { + "epoch": 0.8580894533406958, + "grad_norm": 0.8644204139709473, + "learning_rate": 4.7910809363511825e-06, + "loss": 0.6755, + "step": 1036 + }, + { + "epoch": 0.8589177250138045, + "grad_norm": 0.8761067390441895, + "learning_rate": 4.790640630474206e-06, + "loss": 0.699, + "step": 1037 + }, + { + "epoch": 0.8597459966869133, + "grad_norm": 0.9135014414787292, + "learning_rate": 4.790199881379621e-06, + "loss": 0.6659, + "step": 1038 + }, + { + "epoch": 0.860574268360022, + "grad_norm": 0.9065434336662292, + "learning_rate": 4.789758689152708e-06, + "loss": 0.6664, + "step": 1039 + }, + { + "epoch": 0.8614025400331309, + "grad_norm": 0.8814099431037903, + "learning_rate": 4.789317053878833e-06, + "loss": 0.681, + "step": 1040 + }, + { + "epoch": 0.8622308117062396, + "grad_norm": 0.872502863407135, + "learning_rate": 4.788874975643449e-06, + "loss": 0.7044, + "step": 1041 + }, + { + "epoch": 0.8630590833793484, + "grad_norm": 0.9532367587089539, + "learning_rate": 4.788432454532093e-06, + "loss": 0.655, + "step": 1042 + }, + { + "epoch": 0.8638873550524572, + "grad_norm": 0.9009373784065247, + "learning_rate": 4.787989490630391e-06, + "loss": 0.6946, + "step": 1043 + }, + { + "epoch": 0.864715626725566, + "grad_norm": 0.889933168888092, + "learning_rate": 4.787546084024051e-06, + "loss": 0.6595, + "step": 1044 + }, + { + "epoch": 0.8655438983986747, + "grad_norm": 0.9179873466491699, + "learning_rate": 4.787102234798867e-06, + "loss": 0.658, + "step": 1045 + }, + { + "epoch": 0.8663721700717836, + "grad_norm": 0.8873292207717896, + "learning_rate": 4.786657943040721e-06, + "loss": 0.7023, + "step": 1046 + }, + { + "epoch": 0.8672004417448923, + "grad_norm": 0.8388610482215881, + "learning_rate": 4.78621320883558e-06, + "loss": 0.6644, + "step": 1047 + }, + { + "epoch": 0.8680287134180011, + "grad_norm": 0.8971349596977234, + "learning_rate": 4.785768032269493e-06, + "loss": 0.6552, + "step": 1048 + }, + { + "epoch": 0.8688569850911099, + "grad_norm": 0.8789131045341492, + "learning_rate": 4.785322413428601e-06, + "loss": 0.6681, + "step": 1049 + }, + { + "epoch": 0.8696852567642187, + "grad_norm": 0.8867958784103394, + "learning_rate": 4.784876352399125e-06, + "loss": 0.6859, + "step": 1050 + }, + { + "epoch": 0.8705135284373274, + "grad_norm": 0.8584237694740295, + "learning_rate": 4.784429849267373e-06, + "loss": 0.6655, + "step": 1051 + }, + { + "epoch": 0.8713418001104363, + "grad_norm": 0.9230811595916748, + "learning_rate": 4.783982904119743e-06, + "loss": 0.6648, + "step": 1052 + }, + { + "epoch": 0.872170071783545, + "grad_norm": 0.9217789769172668, + "learning_rate": 4.783535517042709e-06, + "loss": 0.6867, + "step": 1053 + }, + { + "epoch": 0.8729983434566538, + "grad_norm": 0.9209116697311401, + "learning_rate": 4.783087688122842e-06, + "loss": 0.6972, + "step": 1054 + }, + { + "epoch": 0.8738266151297626, + "grad_norm": 0.9493632316589355, + "learning_rate": 4.782639417446789e-06, + "loss": 0.6785, + "step": 1055 + }, + { + "epoch": 0.8746548868028713, + "grad_norm": 0.8790786862373352, + "learning_rate": 4.782190705101286e-06, + "loss": 0.6925, + "step": 1056 + }, + { + "epoch": 0.8754831584759801, + "grad_norm": 0.8662691712379456, + "learning_rate": 4.781741551173157e-06, + "loss": 0.6686, + "step": 1057 + }, + { + "epoch": 0.8763114301490889, + "grad_norm": 0.919658899307251, + "learning_rate": 4.781291955749307e-06, + "loss": 0.6814, + "step": 1058 + }, + { + "epoch": 0.8771397018221977, + "grad_norm": 0.9639708399772644, + "learning_rate": 4.780841918916731e-06, + "loss": 0.6629, + "step": 1059 + }, + { + "epoch": 0.8779679734953064, + "grad_norm": 0.8694306015968323, + "learning_rate": 4.780391440762505e-06, + "loss": 0.6893, + "step": 1060 + }, + { + "epoch": 0.8787962451684153, + "grad_norm": 0.9725603461265564, + "learning_rate": 4.779940521373793e-06, + "loss": 0.67, + "step": 1061 + }, + { + "epoch": 0.879624516841524, + "grad_norm": 0.93159419298172, + "learning_rate": 4.779489160837842e-06, + "loss": 0.686, + "step": 1062 + }, + { + "epoch": 0.8804527885146328, + "grad_norm": 0.9366991519927979, + "learning_rate": 4.779037359241989e-06, + "loss": 0.649, + "step": 1063 + }, + { + "epoch": 0.8812810601877415, + "grad_norm": 0.9052289724349976, + "learning_rate": 4.7785851166736505e-06, + "loss": 0.6677, + "step": 1064 + }, + { + "epoch": 0.8821093318608504, + "grad_norm": 0.882251501083374, + "learning_rate": 4.778132433220334e-06, + "loss": 0.6886, + "step": 1065 + }, + { + "epoch": 0.8829376035339591, + "grad_norm": 0.8867698907852173, + "learning_rate": 4.777679308969627e-06, + "loss": 0.6645, + "step": 1066 + }, + { + "epoch": 0.883765875207068, + "grad_norm": 0.890396773815155, + "learning_rate": 4.777225744009207e-06, + "loss": 0.6684, + "step": 1067 + }, + { + "epoch": 0.8845941468801767, + "grad_norm": 0.8827376365661621, + "learning_rate": 4.776771738426833e-06, + "loss": 0.6738, + "step": 1068 + }, + { + "epoch": 0.8854224185532855, + "grad_norm": 0.8953343033790588, + "learning_rate": 4.776317292310352e-06, + "loss": 0.684, + "step": 1069 + }, + { + "epoch": 0.8862506902263942, + "grad_norm": 0.8751050233840942, + "learning_rate": 4.775862405747694e-06, + "loss": 0.6779, + "step": 1070 + }, + { + "epoch": 0.8870789618995031, + "grad_norm": 0.9136618971824646, + "learning_rate": 4.775407078826876e-06, + "loss": 0.6939, + "step": 1071 + }, + { + "epoch": 0.8879072335726118, + "grad_norm": 0.8887941837310791, + "learning_rate": 4.774951311635999e-06, + "loss": 0.6547, + "step": 1072 + }, + { + "epoch": 0.8887355052457206, + "grad_norm": 0.8873274326324463, + "learning_rate": 4.77449510426325e-06, + "loss": 0.6769, + "step": 1073 + }, + { + "epoch": 0.8895637769188294, + "grad_norm": 0.9307074546813965, + "learning_rate": 4.774038456796901e-06, + "loss": 0.6984, + "step": 1074 + }, + { + "epoch": 0.8903920485919381, + "grad_norm": 0.9022747874259949, + "learning_rate": 4.77358136932531e-06, + "loss": 0.6733, + "step": 1075 + }, + { + "epoch": 0.8912203202650469, + "grad_norm": 0.8779094219207764, + "learning_rate": 4.773123841936916e-06, + "loss": 0.6434, + "step": 1076 + }, + { + "epoch": 0.8920485919381557, + "grad_norm": 0.9465747475624084, + "learning_rate": 4.7726658747202505e-06, + "loss": 0.6761, + "step": 1077 + }, + { + "epoch": 0.8928768636112645, + "grad_norm": 0.8694362044334412, + "learning_rate": 4.772207467763923e-06, + "loss": 0.6756, + "step": 1078 + }, + { + "epoch": 0.8937051352843732, + "grad_norm": 0.8839150071144104, + "learning_rate": 4.771748621156631e-06, + "loss": 0.6714, + "step": 1079 + }, + { + "epoch": 0.8945334069574821, + "grad_norm": 0.9220975041389465, + "learning_rate": 4.7712893349871585e-06, + "loss": 0.6724, + "step": 1080 + }, + { + "epoch": 0.8953616786305908, + "grad_norm": 0.9234760999679565, + "learning_rate": 4.770829609344372e-06, + "loss": 0.67, + "step": 1081 + }, + { + "epoch": 0.8961899503036996, + "grad_norm": 0.9262274503707886, + "learning_rate": 4.770369444317225e-06, + "loss": 0.6619, + "step": 1082 + }, + { + "epoch": 0.8970182219768084, + "grad_norm": 0.8889190554618835, + "learning_rate": 4.769908839994753e-06, + "loss": 0.688, + "step": 1083 + }, + { + "epoch": 0.8978464936499172, + "grad_norm": 0.9028982520103455, + "learning_rate": 4.769447796466082e-06, + "loss": 0.6633, + "step": 1084 + }, + { + "epoch": 0.8986747653230259, + "grad_norm": 0.9155063033103943, + "learning_rate": 4.7689863138204185e-06, + "loss": 0.6709, + "step": 1085 + }, + { + "epoch": 0.8995030369961348, + "grad_norm": 0.9066855311393738, + "learning_rate": 4.768524392147052e-06, + "loss": 0.6713, + "step": 1086 + }, + { + "epoch": 0.9003313086692435, + "grad_norm": 0.9365159273147583, + "learning_rate": 4.768062031535365e-06, + "loss": 0.6675, + "step": 1087 + }, + { + "epoch": 0.9011595803423523, + "grad_norm": 0.9106162190437317, + "learning_rate": 4.767599232074816e-06, + "loss": 0.6804, + "step": 1088 + }, + { + "epoch": 0.9019878520154611, + "grad_norm": 0.9183253645896912, + "learning_rate": 4.767135993854955e-06, + "loss": 0.658, + "step": 1089 + }, + { + "epoch": 0.9028161236885699, + "grad_norm": 0.9543409943580627, + "learning_rate": 4.766672316965413e-06, + "loss": 0.696, + "step": 1090 + }, + { + "epoch": 0.9036443953616786, + "grad_norm": 0.8894531726837158, + "learning_rate": 4.766208201495906e-06, + "loss": 0.6781, + "step": 1091 + }, + { + "epoch": 0.9044726670347875, + "grad_norm": 0.915542721748352, + "learning_rate": 4.765743647536238e-06, + "loss": 0.6804, + "step": 1092 + }, + { + "epoch": 0.9053009387078962, + "grad_norm": 0.8959763646125793, + "learning_rate": 4.765278655176296e-06, + "loss": 0.6873, + "step": 1093 + }, + { + "epoch": 0.906129210381005, + "grad_norm": 0.881188690662384, + "learning_rate": 4.76481322450605e-06, + "loss": 0.6582, + "step": 1094 + }, + { + "epoch": 0.9069574820541138, + "grad_norm": 0.9096754193305969, + "learning_rate": 4.764347355615558e-06, + "loss": 0.693, + "step": 1095 + }, + { + "epoch": 0.9077857537272225, + "grad_norm": 0.891132652759552, + "learning_rate": 4.76388104859496e-06, + "loss": 0.6781, + "step": 1096 + }, + { + "epoch": 0.9086140254003313, + "grad_norm": 0.893624484539032, + "learning_rate": 4.763414303534484e-06, + "loss": 0.6636, + "step": 1097 + }, + { + "epoch": 0.90944229707344, + "grad_norm": 0.8795678019523621, + "learning_rate": 4.762947120524438e-06, + "loss": 0.6604, + "step": 1098 + }, + { + "epoch": 0.9102705687465489, + "grad_norm": 0.9177265763282776, + "learning_rate": 4.76247949965522e-06, + "loss": 0.6557, + "step": 1099 + }, + { + "epoch": 0.9110988404196576, + "grad_norm": 0.9416985511779785, + "learning_rate": 4.762011441017309e-06, + "loss": 0.7008, + "step": 1100 + }, + { + "epoch": 0.9119271120927664, + "grad_norm": 0.9255479574203491, + "learning_rate": 4.7615429447012704e-06, + "loss": 0.6665, + "step": 1101 + }, + { + "epoch": 0.9127553837658752, + "grad_norm": 0.8948953151702881, + "learning_rate": 4.761074010797754e-06, + "loss": 0.6699, + "step": 1102 + }, + { + "epoch": 0.913583655438984, + "grad_norm": 0.9027442336082458, + "learning_rate": 4.760604639397493e-06, + "loss": 0.6649, + "step": 1103 + }, + { + "epoch": 0.9144119271120927, + "grad_norm": 0.9094238877296448, + "learning_rate": 4.7601348305913085e-06, + "loss": 0.6753, + "step": 1104 + }, + { + "epoch": 0.9152401987852016, + "grad_norm": 0.9098204374313354, + "learning_rate": 4.759664584470103e-06, + "loss": 0.6845, + "step": 1105 + }, + { + "epoch": 0.9160684704583103, + "grad_norm": 0.9339573383331299, + "learning_rate": 4.759193901124864e-06, + "loss": 0.6285, + "step": 1106 + }, + { + "epoch": 0.9168967421314191, + "grad_norm": 0.8812451958656311, + "learning_rate": 4.758722780646666e-06, + "loss": 0.6764, + "step": 1107 + }, + { + "epoch": 0.9177250138045279, + "grad_norm": 0.8832173347473145, + "learning_rate": 4.758251223126665e-06, + "loss": 0.6908, + "step": 1108 + }, + { + "epoch": 0.9185532854776367, + "grad_norm": 0.9124787449836731, + "learning_rate": 4.757779228656103e-06, + "loss": 0.6855, + "step": 1109 + }, + { + "epoch": 0.9193815571507454, + "grad_norm": 0.8778328895568848, + "learning_rate": 4.757306797326309e-06, + "loss": 0.6404, + "step": 1110 + }, + { + "epoch": 0.9202098288238543, + "grad_norm": 0.8831816911697388, + "learning_rate": 4.75683392922869e-06, + "loss": 0.6425, + "step": 1111 + }, + { + "epoch": 0.921038100496963, + "grad_norm": 0.9060006141662598, + "learning_rate": 4.756360624454744e-06, + "loss": 0.6738, + "step": 1112 + }, + { + "epoch": 0.9218663721700718, + "grad_norm": 0.8746777176856995, + "learning_rate": 4.755886883096051e-06, + "loss": 0.6665, + "step": 1113 + }, + { + "epoch": 0.9226946438431806, + "grad_norm": 0.9185945987701416, + "learning_rate": 4.755412705244274e-06, + "loss": 0.6564, + "step": 1114 + }, + { + "epoch": 0.9235229155162893, + "grad_norm": 0.9648581743240356, + "learning_rate": 4.754938090991165e-06, + "loss": 0.6597, + "step": 1115 + }, + { + "epoch": 0.9243511871893981, + "grad_norm": 0.8800092935562134, + "learning_rate": 4.754463040428555e-06, + "loss": 0.6884, + "step": 1116 + }, + { + "epoch": 0.9251794588625069, + "grad_norm": 0.8659892082214355, + "learning_rate": 4.753987553648363e-06, + "loss": 0.68, + "step": 1117 + }, + { + "epoch": 0.9260077305356157, + "grad_norm": 0.88991379737854, + "learning_rate": 4.75351163074259e-06, + "loss": 0.668, + "step": 1118 + }, + { + "epoch": 0.9268360022087244, + "grad_norm": 0.9683054089546204, + "learning_rate": 4.753035271803323e-06, + "loss": 0.6789, + "step": 1119 + }, + { + "epoch": 0.9276642738818333, + "grad_norm": 0.928001880645752, + "learning_rate": 4.752558476922735e-06, + "loss": 0.6786, + "step": 1120 + }, + { + "epoch": 0.928492545554942, + "grad_norm": 0.8760973215103149, + "learning_rate": 4.752081246193078e-06, + "loss": 0.6716, + "step": 1121 + }, + { + "epoch": 0.9293208172280508, + "grad_norm": 0.9171329140663147, + "learning_rate": 4.751603579706695e-06, + "loss": 0.662, + "step": 1122 + }, + { + "epoch": 0.9301490889011595, + "grad_norm": 0.9380231499671936, + "learning_rate": 4.751125477556008e-06, + "loss": 0.6836, + "step": 1123 + }, + { + "epoch": 0.9309773605742684, + "grad_norm": 0.9471802711486816, + "learning_rate": 4.750646939833526e-06, + "loss": 0.668, + "step": 1124 + }, + { + "epoch": 0.9318056322473771, + "grad_norm": 0.8860809206962585, + "learning_rate": 4.750167966631842e-06, + "loss": 0.6635, + "step": 1125 + }, + { + "epoch": 0.932633903920486, + "grad_norm": 0.8694644570350647, + "learning_rate": 4.7496885580436305e-06, + "loss": 0.6505, + "step": 1126 + }, + { + "epoch": 0.9334621755935947, + "grad_norm": 0.9602686166763306, + "learning_rate": 4.749208714161655e-06, + "loss": 0.6874, + "step": 1127 + }, + { + "epoch": 0.9342904472667035, + "grad_norm": 0.9198856353759766, + "learning_rate": 4.748728435078761e-06, + "loss": 0.697, + "step": 1128 + }, + { + "epoch": 0.9351187189398122, + "grad_norm": 0.9162499904632568, + "learning_rate": 4.748247720887876e-06, + "loss": 0.6521, + "step": 1129 + }, + { + "epoch": 0.9359469906129211, + "grad_norm": 0.9064576029777527, + "learning_rate": 4.747766571682016e-06, + "loss": 0.6525, + "step": 1130 + }, + { + "epoch": 0.9367752622860298, + "grad_norm": 0.9518057703971863, + "learning_rate": 4.747284987554277e-06, + "loss": 0.6745, + "step": 1131 + }, + { + "epoch": 0.9376035339591386, + "grad_norm": 0.8963000178337097, + "learning_rate": 4.746802968597842e-06, + "loss": 0.6999, + "step": 1132 + }, + { + "epoch": 0.9384318056322474, + "grad_norm": 0.8687212467193604, + "learning_rate": 4.746320514905977e-06, + "loss": 0.6353, + "step": 1133 + }, + { + "epoch": 0.9392600773053562, + "grad_norm": 0.8598489165306091, + "learning_rate": 4.745837626572033e-06, + "loss": 0.6897, + "step": 1134 + }, + { + "epoch": 0.9400883489784649, + "grad_norm": 0.9425498843193054, + "learning_rate": 4.745354303689444e-06, + "loss": 0.667, + "step": 1135 + }, + { + "epoch": 0.9409166206515737, + "grad_norm": 0.9026349782943726, + "learning_rate": 4.744870546351727e-06, + "loss": 0.676, + "step": 1136 + }, + { + "epoch": 0.9417448923246825, + "grad_norm": 0.8824150562286377, + "learning_rate": 4.744386354652486e-06, + "loss": 0.6941, + "step": 1137 + }, + { + "epoch": 0.9425731639977912, + "grad_norm": 0.8910468816757202, + "learning_rate": 4.743901728685406e-06, + "loss": 0.6701, + "step": 1138 + }, + { + "epoch": 0.9434014356709001, + "grad_norm": 0.9114013314247131, + "learning_rate": 4.74341666854426e-06, + "loss": 0.6735, + "step": 1139 + }, + { + "epoch": 0.9442297073440088, + "grad_norm": 0.9107307195663452, + "learning_rate": 4.742931174322902e-06, + "loss": 0.6505, + "step": 1140 + }, + { + "epoch": 0.9450579790171176, + "grad_norm": 0.8906923532485962, + "learning_rate": 4.742445246115271e-06, + "loss": 0.6648, + "step": 1141 + }, + { + "epoch": 0.9458862506902264, + "grad_norm": 0.902644157409668, + "learning_rate": 4.741958884015387e-06, + "loss": 0.6674, + "step": 1142 + }, + { + "epoch": 0.9467145223633352, + "grad_norm": 0.9522087574005127, + "learning_rate": 4.741472088117358e-06, + "loss": 0.6825, + "step": 1143 + }, + { + "epoch": 0.9475427940364439, + "grad_norm": 0.9152255654335022, + "learning_rate": 4.740984858515376e-06, + "loss": 0.6708, + "step": 1144 + }, + { + "epoch": 0.9483710657095528, + "grad_norm": 0.8779835104942322, + "learning_rate": 4.740497195303713e-06, + "loss": 0.6562, + "step": 1145 + }, + { + "epoch": 0.9491993373826615, + "grad_norm": 0.8842780590057373, + "learning_rate": 4.740009098576729e-06, + "loss": 0.6769, + "step": 1146 + }, + { + "epoch": 0.9500276090557703, + "grad_norm": 0.9137415289878845, + "learning_rate": 4.739520568428866e-06, + "loss": 0.6705, + "step": 1147 + }, + { + "epoch": 0.950855880728879, + "grad_norm": 0.8867765665054321, + "learning_rate": 4.739031604954648e-06, + "loss": 0.659, + "step": 1148 + }, + { + "epoch": 0.9516841524019879, + "grad_norm": 0.8965260982513428, + "learning_rate": 4.7385422082486874e-06, + "loss": 0.6744, + "step": 1149 + }, + { + "epoch": 0.9525124240750966, + "grad_norm": 0.8861697912216187, + "learning_rate": 4.738052378405677e-06, + "loss": 0.6575, + "step": 1150 + }, + { + "epoch": 0.9533406957482055, + "grad_norm": 0.9442310333251953, + "learning_rate": 4.737562115520394e-06, + "loss": 0.7164, + "step": 1151 + }, + { + "epoch": 0.9541689674213142, + "grad_norm": 0.9340727925300598, + "learning_rate": 4.7370714196877e-06, + "loss": 0.6577, + "step": 1152 + }, + { + "epoch": 0.954997239094423, + "grad_norm": 0.9337487816810608, + "learning_rate": 4.7365802910025405e-06, + "loss": 0.6664, + "step": 1153 + }, + { + "epoch": 0.9558255107675317, + "grad_norm": 0.8849001526832581, + "learning_rate": 4.736088729559943e-06, + "loss": 0.6721, + "step": 1154 + }, + { + "epoch": 0.9566537824406405, + "grad_norm": 0.917323648929596, + "learning_rate": 4.735596735455022e-06, + "loss": 0.6715, + "step": 1155 + }, + { + "epoch": 0.9574820541137493, + "grad_norm": 0.9480959177017212, + "learning_rate": 4.735104308782972e-06, + "loss": 0.6784, + "step": 1156 + }, + { + "epoch": 0.958310325786858, + "grad_norm": 0.9519251585006714, + "learning_rate": 4.734611449639073e-06, + "loss": 0.6617, + "step": 1157 + }, + { + "epoch": 0.9591385974599669, + "grad_norm": 0.8652079105377197, + "learning_rate": 4.734118158118689e-06, + "loss": 0.6846, + "step": 1158 + }, + { + "epoch": 0.9599668691330756, + "grad_norm": 0.8860392570495605, + "learning_rate": 4.733624434317269e-06, + "loss": 0.6729, + "step": 1159 + }, + { + "epoch": 0.9607951408061844, + "grad_norm": 0.9467882513999939, + "learning_rate": 4.7331302783303416e-06, + "loss": 0.6362, + "step": 1160 + }, + { + "epoch": 0.9616234124792932, + "grad_norm": 0.9138529896736145, + "learning_rate": 4.732635690253523e-06, + "loss": 0.6613, + "step": 1161 + }, + { + "epoch": 0.962451684152402, + "grad_norm": 0.9401054978370667, + "learning_rate": 4.7321406701825105e-06, + "loss": 0.6845, + "step": 1162 + }, + { + "epoch": 0.9632799558255107, + "grad_norm": 0.9049288630485535, + "learning_rate": 4.731645218213086e-06, + "loss": 0.6857, + "step": 1163 + }, + { + "epoch": 0.9641082274986196, + "grad_norm": 0.8972534537315369, + "learning_rate": 4.731149334441114e-06, + "loss": 0.6745, + "step": 1164 + }, + { + "epoch": 0.9649364991717283, + "grad_norm": 1.0799483060836792, + "learning_rate": 4.730653018962545e-06, + "loss": 0.6754, + "step": 1165 + }, + { + "epoch": 0.9657647708448371, + "grad_norm": 0.9841833710670471, + "learning_rate": 4.730156271873412e-06, + "loss": 0.6718, + "step": 1166 + }, + { + "epoch": 0.9665930425179459, + "grad_norm": 0.9388216733932495, + "learning_rate": 4.729659093269828e-06, + "loss": 0.6622, + "step": 1167 + }, + { + "epoch": 0.9674213141910547, + "grad_norm": 0.8879573345184326, + "learning_rate": 4.729161483247994e-06, + "loss": 0.6909, + "step": 1168 + }, + { + "epoch": 0.9682495858641634, + "grad_norm": 0.9713517427444458, + "learning_rate": 4.728663441904194e-06, + "loss": 0.6956, + "step": 1169 + }, + { + "epoch": 0.9690778575372723, + "grad_norm": 0.9200823903083801, + "learning_rate": 4.7281649693347925e-06, + "loss": 0.6766, + "step": 1170 + }, + { + "epoch": 0.969906129210381, + "grad_norm": 0.8983714580535889, + "learning_rate": 4.727666065636241e-06, + "loss": 0.6626, + "step": 1171 + }, + { + "epoch": 0.9707344008834898, + "grad_norm": 0.8792792558670044, + "learning_rate": 4.72716673090507e-06, + "loss": 0.666, + "step": 1172 + }, + { + "epoch": 0.9715626725565986, + "grad_norm": 1.0116846561431885, + "learning_rate": 4.7266669652379005e-06, + "loss": 0.6574, + "step": 1173 + }, + { + "epoch": 0.9723909442297074, + "grad_norm": 0.9433228373527527, + "learning_rate": 4.7261667687314285e-06, + "loss": 0.6972, + "step": 1174 + }, + { + "epoch": 0.9732192159028161, + "grad_norm": 0.8979770541191101, + "learning_rate": 4.725666141482439e-06, + "loss": 0.6523, + "step": 1175 + }, + { + "epoch": 0.9740474875759249, + "grad_norm": 0.8990604281425476, + "learning_rate": 4.725165083587798e-06, + "loss": 0.6535, + "step": 1176 + }, + { + "epoch": 0.9748757592490337, + "grad_norm": 0.9411956667900085, + "learning_rate": 4.724663595144457e-06, + "loss": 0.6947, + "step": 1177 + }, + { + "epoch": 0.9757040309221424, + "grad_norm": 1.0203242301940918, + "learning_rate": 4.724161676249448e-06, + "loss": 0.6894, + "step": 1178 + }, + { + "epoch": 0.9765323025952513, + "grad_norm": 0.8938170075416565, + "learning_rate": 4.723659326999888e-06, + "loss": 0.6648, + "step": 1179 + }, + { + "epoch": 0.97736057426836, + "grad_norm": 0.8810412883758545, + "learning_rate": 4.7231565474929765e-06, + "loss": 0.6761, + "step": 1180 + }, + { + "epoch": 0.9781888459414688, + "grad_norm": 0.8903076648712158, + "learning_rate": 4.722653337825998e-06, + "loss": 0.6738, + "step": 1181 + }, + { + "epoch": 0.9790171176145775, + "grad_norm": 0.905060350894928, + "learning_rate": 4.7221496980963175e-06, + "loss": 0.6625, + "step": 1182 + }, + { + "epoch": 0.9798453892876864, + "grad_norm": 0.8791100978851318, + "learning_rate": 4.721645628401385e-06, + "loss": 0.6884, + "step": 1183 + }, + { + "epoch": 0.9806736609607951, + "grad_norm": 0.897101104259491, + "learning_rate": 4.721141128838733e-06, + "loss": 0.6693, + "step": 1184 + }, + { + "epoch": 0.981501932633904, + "grad_norm": 0.9102833867073059, + "learning_rate": 4.7206361995059775e-06, + "loss": 0.6682, + "step": 1185 + }, + { + "epoch": 0.9823302043070127, + "grad_norm": 0.9492197632789612, + "learning_rate": 4.720130840500819e-06, + "loss": 0.6501, + "step": 1186 + }, + { + "epoch": 0.9831584759801215, + "grad_norm": 0.9783265590667725, + "learning_rate": 4.719625051921038e-06, + "loss": 0.6454, + "step": 1187 + }, + { + "epoch": 0.9839867476532302, + "grad_norm": 0.8938109874725342, + "learning_rate": 4.719118833864501e-06, + "loss": 0.6671, + "step": 1188 + }, + { + "epoch": 0.9848150193263391, + "grad_norm": 0.897484302520752, + "learning_rate": 4.718612186429156e-06, + "loss": 0.6664, + "step": 1189 + }, + { + "epoch": 0.9856432909994478, + "grad_norm": 0.9640376567840576, + "learning_rate": 4.718105109713034e-06, + "loss": 0.6845, + "step": 1190 + }, + { + "epoch": 0.9864715626725566, + "grad_norm": 0.9495493769645691, + "learning_rate": 4.7175976038142505e-06, + "loss": 0.6864, + "step": 1191 + }, + { + "epoch": 0.9872998343456654, + "grad_norm": 0.9390130639076233, + "learning_rate": 4.717089668831002e-06, + "loss": 0.6734, + "step": 1192 + }, + { + "epoch": 0.9881281060187742, + "grad_norm": 0.8854318261146545, + "learning_rate": 4.716581304861571e-06, + "loss": 0.6546, + "step": 1193 + }, + { + "epoch": 0.9889563776918829, + "grad_norm": 0.9242726564407349, + "learning_rate": 4.71607251200432e-06, + "loss": 0.6663, + "step": 1194 + }, + { + "epoch": 0.9897846493649917, + "grad_norm": 0.9302003383636475, + "learning_rate": 4.715563290357696e-06, + "loss": 0.6844, + "step": 1195 + }, + { + "epoch": 0.9906129210381005, + "grad_norm": 0.9279932975769043, + "learning_rate": 4.715053640020228e-06, + "loss": 0.6731, + "step": 1196 + }, + { + "epoch": 0.9914411927112092, + "grad_norm": 0.8753612041473389, + "learning_rate": 4.71454356109053e-06, + "loss": 0.6738, + "step": 1197 + }, + { + "epoch": 0.9922694643843181, + "grad_norm": 0.890133261680603, + "learning_rate": 4.714033053667296e-06, + "loss": 0.6638, + "step": 1198 + }, + { + "epoch": 0.9930977360574268, + "grad_norm": 0.9343729615211487, + "learning_rate": 4.713522117849305e-06, + "loss": 0.6826, + "step": 1199 + }, + { + "epoch": 0.9939260077305356, + "grad_norm": 0.9052755832672119, + "learning_rate": 4.7130107537354185e-06, + "loss": 0.6897, + "step": 1200 + }, + { + "epoch": 0.9947542794036444, + "grad_norm": 0.8683924674987793, + "learning_rate": 4.7124989614245805e-06, + "loss": 0.6687, + "step": 1201 + }, + { + "epoch": 0.9955825510767532, + "grad_norm": 0.8677132725715637, + "learning_rate": 4.711986741015818e-06, + "loss": 0.6917, + "step": 1202 + }, + { + "epoch": 0.9964108227498619, + "grad_norm": 0.8840690851211548, + "learning_rate": 4.711474092608243e-06, + "loss": 0.6841, + "step": 1203 + }, + { + "epoch": 0.9972390944229708, + "grad_norm": 0.9750391840934753, + "learning_rate": 4.7109610163010435e-06, + "loss": 0.6574, + "step": 1204 + }, + { + "epoch": 0.9980673660960795, + "grad_norm": 0.8752943873405457, + "learning_rate": 4.710447512193499e-06, + "loss": 0.636, + "step": 1205 + }, + { + "epoch": 0.9988956377691883, + "grad_norm": 0.9100313782691956, + "learning_rate": 4.709933580384965e-06, + "loss": 0.6712, + "step": 1206 + }, + { + "epoch": 0.999723909442297, + "grad_norm": 0.9202356338500977, + "learning_rate": 4.709419220974886e-06, + "loss": 0.6641, + "step": 1207 + }, + { + "epoch": 1.0, + "grad_norm": 0.9202356338500977, + "learning_rate": 4.708904434062783e-06, + "loss": 0.6199, + "step": 1208 + }, + { + "epoch": 1.0008282716731087, + "grad_norm": 1.6839245557785034, + "learning_rate": 4.708389219748263e-06, + "loss": 0.643, + "step": 1209 + }, + { + "epoch": 1.0016565433462175, + "grad_norm": 0.8636482357978821, + "learning_rate": 4.707873578131015e-06, + "loss": 0.6467, + "step": 1210 + }, + { + "epoch": 1.0024848150193264, + "grad_norm": 0.8548988103866577, + "learning_rate": 4.707357509310811e-06, + "loss": 0.6574, + "step": 1211 + }, + { + "epoch": 1.0033130866924351, + "grad_norm": 0.8961352705955505, + "learning_rate": 4.7068410133875066e-06, + "loss": 0.6367, + "step": 1212 + }, + { + "epoch": 1.0041413583655439, + "grad_norm": 0.8914269208908081, + "learning_rate": 4.706324090461037e-06, + "loss": 0.6294, + "step": 1213 + }, + { + "epoch": 1.0049696300386526, + "grad_norm": 0.879915177822113, + "learning_rate": 4.7058067406314235e-06, + "loss": 0.6346, + "step": 1214 + }, + { + "epoch": 1.0057979017117615, + "grad_norm": 0.8798143863677979, + "learning_rate": 4.7052889639987674e-06, + "loss": 0.6559, + "step": 1215 + }, + { + "epoch": 1.0066261733848703, + "grad_norm": 0.8888561129570007, + "learning_rate": 4.704770760663255e-06, + "loss": 0.6471, + "step": 1216 + }, + { + "epoch": 1.007454445057979, + "grad_norm": 0.8707575798034668, + "learning_rate": 4.704252130725152e-06, + "loss": 0.6485, + "step": 1217 + }, + { + "epoch": 1.0082827167310877, + "grad_norm": 0.8609346151351929, + "learning_rate": 4.7037330742848105e-06, + "loss": 0.6544, + "step": 1218 + }, + { + "epoch": 1.0091109884041967, + "grad_norm": 0.8588053584098816, + "learning_rate": 4.703213591442663e-06, + "loss": 0.6421, + "step": 1219 + }, + { + "epoch": 1.0099392600773054, + "grad_norm": 0.8932695388793945, + "learning_rate": 4.702693682299222e-06, + "loss": 0.6333, + "step": 1220 + }, + { + "epoch": 1.0107675317504141, + "grad_norm": 0.8615652322769165, + "learning_rate": 4.7021733469550865e-06, + "loss": 0.6371, + "step": 1221 + }, + { + "epoch": 1.0115958034235228, + "grad_norm": 0.8876634240150452, + "learning_rate": 4.7016525855109376e-06, + "loss": 0.6368, + "step": 1222 + }, + { + "epoch": 1.0124240750966318, + "grad_norm": 0.895575225353241, + "learning_rate": 4.701131398067537e-06, + "loss": 0.6226, + "step": 1223 + }, + { + "epoch": 1.0132523467697405, + "grad_norm": 0.8836749196052551, + "learning_rate": 4.70060978472573e-06, + "loss": 0.6516, + "step": 1224 + }, + { + "epoch": 1.0140806184428492, + "grad_norm": 0.9471280574798584, + "learning_rate": 4.700087745586442e-06, + "loss": 0.633, + "step": 1225 + }, + { + "epoch": 1.014908890115958, + "grad_norm": 0.8721561431884766, + "learning_rate": 4.699565280750685e-06, + "loss": 0.6568, + "step": 1226 + }, + { + "epoch": 1.015737161789067, + "grad_norm": 0.8946447372436523, + "learning_rate": 4.69904239031955e-06, + "loss": 0.6503, + "step": 1227 + }, + { + "epoch": 1.0165654334621756, + "grad_norm": 0.8854631185531616, + "learning_rate": 4.698519074394212e-06, + "loss": 0.6341, + "step": 1228 + }, + { + "epoch": 1.0173937051352844, + "grad_norm": 0.9348397850990295, + "learning_rate": 4.697995333075927e-06, + "loss": 0.6317, + "step": 1229 + }, + { + "epoch": 1.018221976808393, + "grad_norm": 0.9037336111068726, + "learning_rate": 4.697471166466035e-06, + "loss": 0.6618, + "step": 1230 + }, + { + "epoch": 1.0190502484815018, + "grad_norm": 0.8866400718688965, + "learning_rate": 4.696946574665957e-06, + "loss": 0.6421, + "step": 1231 + }, + { + "epoch": 1.0198785201546108, + "grad_norm": 0.8897701501846313, + "learning_rate": 4.696421557777196e-06, + "loss": 0.6608, + "step": 1232 + }, + { + "epoch": 1.0207067918277195, + "grad_norm": 0.9046173095703125, + "learning_rate": 4.6958961159013386e-06, + "loss": 0.6464, + "step": 1233 + }, + { + "epoch": 1.0215350635008282, + "grad_norm": 0.9492906332015991, + "learning_rate": 4.695370249140052e-06, + "loss": 0.6578, + "step": 1234 + }, + { + "epoch": 1.022363335173937, + "grad_norm": 0.8906951546669006, + "learning_rate": 4.6948439575950885e-06, + "loss": 0.6276, + "step": 1235 + }, + { + "epoch": 1.023191606847046, + "grad_norm": 0.9218266606330872, + "learning_rate": 4.694317241368278e-06, + "loss": 0.627, + "step": 1236 + }, + { + "epoch": 1.0240198785201546, + "grad_norm": 0.901157557964325, + "learning_rate": 4.693790100561537e-06, + "loss": 0.6379, + "step": 1237 + }, + { + "epoch": 1.0248481501932634, + "grad_norm": 0.8992452025413513, + "learning_rate": 4.693262535276863e-06, + "loss": 0.6585, + "step": 1238 + }, + { + "epoch": 1.025676421866372, + "grad_norm": 0.9525597095489502, + "learning_rate": 4.692734545616331e-06, + "loss": 0.6625, + "step": 1239 + }, + { + "epoch": 1.026504693539481, + "grad_norm": 0.9229065775871277, + "learning_rate": 4.692206131682106e-06, + "loss": 0.6454, + "step": 1240 + }, + { + "epoch": 1.0273329652125898, + "grad_norm": 1.012373685836792, + "learning_rate": 4.691677293576431e-06, + "loss": 0.6492, + "step": 1241 + }, + { + "epoch": 1.0281612368856985, + "grad_norm": 0.8650060892105103, + "learning_rate": 4.69114803140163e-06, + "loss": 0.6739, + "step": 1242 + }, + { + "epoch": 1.0289895085588072, + "grad_norm": 0.9396429657936096, + "learning_rate": 4.69061834526011e-06, + "loss": 0.6729, + "step": 1243 + }, + { + "epoch": 1.0298177802319162, + "grad_norm": 0.9584783911705017, + "learning_rate": 4.6900882352543614e-06, + "loss": 0.6678, + "step": 1244 + }, + { + "epoch": 1.030646051905025, + "grad_norm": 0.985394299030304, + "learning_rate": 4.689557701486954e-06, + "loss": 0.6478, + "step": 1245 + }, + { + "epoch": 1.0314743235781336, + "grad_norm": 0.9014914631843567, + "learning_rate": 4.689026744060543e-06, + "loss": 0.6393, + "step": 1246 + }, + { + "epoch": 1.0323025952512423, + "grad_norm": 0.9904847145080566, + "learning_rate": 4.688495363077863e-06, + "loss": 0.6752, + "step": 1247 + }, + { + "epoch": 1.0331308669243513, + "grad_norm": 0.9439241886138916, + "learning_rate": 4.687963558641731e-06, + "loss": 0.6417, + "step": 1248 + }, + { + "epoch": 1.03395913859746, + "grad_norm": 0.9277704954147339, + "learning_rate": 4.687431330855046e-06, + "loss": 0.6493, + "step": 1249 + }, + { + "epoch": 1.0347874102705688, + "grad_norm": 0.9456571340560913, + "learning_rate": 4.686898679820789e-06, + "loss": 0.6774, + "step": 1250 + }, + { + "epoch": 1.0356156819436775, + "grad_norm": 0.9043509364128113, + "learning_rate": 4.686365605642025e-06, + "loss": 0.6198, + "step": 1251 + }, + { + "epoch": 1.0364439536167862, + "grad_norm": 0.9246267676353455, + "learning_rate": 4.685832108421896e-06, + "loss": 0.6513, + "step": 1252 + }, + { + "epoch": 1.0372722252898952, + "grad_norm": 0.9227612614631653, + "learning_rate": 4.685298188263631e-06, + "loss": 0.6597, + "step": 1253 + }, + { + "epoch": 1.0381004969630039, + "grad_norm": 0.8927509188652039, + "learning_rate": 4.684763845270537e-06, + "loss": 0.641, + "step": 1254 + }, + { + "epoch": 1.0389287686361126, + "grad_norm": 0.8812937140464783, + "learning_rate": 4.684229079546005e-06, + "loss": 0.6292, + "step": 1255 + }, + { + "epoch": 1.0397570403092213, + "grad_norm": 0.9186881184577942, + "learning_rate": 4.683693891193508e-06, + "loss": 0.6485, + "step": 1256 + }, + { + "epoch": 1.0405853119823303, + "grad_norm": 0.8928134441375732, + "learning_rate": 4.683158280316599e-06, + "loss": 0.6435, + "step": 1257 + }, + { + "epoch": 1.041413583655439, + "grad_norm": 0.8919920325279236, + "learning_rate": 4.682622247018913e-06, + "loss": 0.622, + "step": 1258 + }, + { + "epoch": 1.0422418553285477, + "grad_norm": 0.885259747505188, + "learning_rate": 4.682085791404169e-06, + "loss": 0.6521, + "step": 1259 + }, + { + "epoch": 1.0430701270016565, + "grad_norm": 0.9034169316291809, + "learning_rate": 4.681548913576165e-06, + "loss": 0.6437, + "step": 1260 + }, + { + "epoch": 1.0438983986747654, + "grad_norm": 0.884848952293396, + "learning_rate": 4.6810116136387825e-06, + "loss": 0.6733, + "step": 1261 + }, + { + "epoch": 1.0447266703478741, + "grad_norm": 0.9137836694717407, + "learning_rate": 4.680473891695985e-06, + "loss": 0.6415, + "step": 1262 + }, + { + "epoch": 1.0455549420209829, + "grad_norm": 0.9044780135154724, + "learning_rate": 4.6799357478518145e-06, + "loss": 0.6453, + "step": 1263 + }, + { + "epoch": 1.0463832136940916, + "grad_norm": 0.9111152291297913, + "learning_rate": 4.6793971822103985e-06, + "loss": 0.6644, + "step": 1264 + }, + { + "epoch": 1.0472114853672005, + "grad_norm": 0.9158084988594055, + "learning_rate": 4.678858194875944e-06, + "loss": 0.6434, + "step": 1265 + }, + { + "epoch": 1.0480397570403093, + "grad_norm": 0.8970723152160645, + "learning_rate": 4.678318785952739e-06, + "loss": 0.6391, + "step": 1266 + }, + { + "epoch": 1.048868028713418, + "grad_norm": 0.8958215117454529, + "learning_rate": 4.677778955545155e-06, + "loss": 0.6562, + "step": 1267 + }, + { + "epoch": 1.0496963003865267, + "grad_norm": 0.9229645729064941, + "learning_rate": 4.677238703757644e-06, + "loss": 0.646, + "step": 1268 + }, + { + "epoch": 1.0505245720596355, + "grad_norm": 0.8986689448356628, + "learning_rate": 4.676698030694741e-06, + "loss": 0.6241, + "step": 1269 + }, + { + "epoch": 1.0513528437327444, + "grad_norm": 0.8943328261375427, + "learning_rate": 4.6761569364610585e-06, + "loss": 0.6455, + "step": 1270 + }, + { + "epoch": 1.0521811154058531, + "grad_norm": 0.9147390127182007, + "learning_rate": 4.675615421161295e-06, + "loss": 0.6299, + "step": 1271 + }, + { + "epoch": 1.0530093870789619, + "grad_norm": 0.9124075174331665, + "learning_rate": 4.675073484900229e-06, + "loss": 0.6379, + "step": 1272 + }, + { + "epoch": 1.0538376587520706, + "grad_norm": 0.9267525672912598, + "learning_rate": 4.674531127782718e-06, + "loss": 0.6478, + "step": 1273 + }, + { + "epoch": 1.0546659304251795, + "grad_norm": 0.8710078001022339, + "learning_rate": 4.673988349913707e-06, + "loss": 0.6314, + "step": 1274 + }, + { + "epoch": 1.0554942020982883, + "grad_norm": 0.8905670642852783, + "learning_rate": 4.673445151398214e-06, + "loss": 0.6456, + "step": 1275 + }, + { + "epoch": 1.056322473771397, + "grad_norm": 0.9125292897224426, + "learning_rate": 4.672901532341346e-06, + "loss": 0.6439, + "step": 1276 + }, + { + "epoch": 1.0571507454445057, + "grad_norm": 0.9003625512123108, + "learning_rate": 4.6723574928482865e-06, + "loss": 0.6561, + "step": 1277 + }, + { + "epoch": 1.0579790171176147, + "grad_norm": 0.96099454164505, + "learning_rate": 4.6718130330243014e-06, + "loss": 0.6515, + "step": 1278 + }, + { + "epoch": 1.0588072887907234, + "grad_norm": 0.8841069340705872, + "learning_rate": 4.671268152974742e-06, + "loss": 0.6389, + "step": 1279 + }, + { + "epoch": 1.0596355604638321, + "grad_norm": 0.8906645178794861, + "learning_rate": 4.670722852805033e-06, + "loss": 0.6448, + "step": 1280 + }, + { + "epoch": 1.0604638321369408, + "grad_norm": 0.9538317918777466, + "learning_rate": 4.670177132620689e-06, + "loss": 0.6544, + "step": 1281 + }, + { + "epoch": 1.0612921038100498, + "grad_norm": 0.9158053994178772, + "learning_rate": 4.669630992527299e-06, + "loss": 0.6494, + "step": 1282 + }, + { + "epoch": 1.0621203754831585, + "grad_norm": 0.9756888151168823, + "learning_rate": 4.6690844326305374e-06, + "loss": 0.6419, + "step": 1283 + }, + { + "epoch": 1.0629486471562672, + "grad_norm": 0.905055820941925, + "learning_rate": 4.668537453036159e-06, + "loss": 0.6463, + "step": 1284 + }, + { + "epoch": 1.063776918829376, + "grad_norm": 0.9011061191558838, + "learning_rate": 4.667990053849997e-06, + "loss": 0.6353, + "step": 1285 + }, + { + "epoch": 1.0646051905024847, + "grad_norm": 0.955047070980072, + "learning_rate": 4.667442235177969e-06, + "loss": 0.6485, + "step": 1286 + }, + { + "epoch": 1.0654334621755936, + "grad_norm": 0.9253432750701904, + "learning_rate": 4.666893997126074e-06, + "loss": 0.6391, + "step": 1287 + }, + { + "epoch": 1.0662617338487024, + "grad_norm": 0.8840277194976807, + "learning_rate": 4.66634533980039e-06, + "loss": 0.6372, + "step": 1288 + }, + { + "epoch": 1.067090005521811, + "grad_norm": 1.0881837606430054, + "learning_rate": 4.6657962633070765e-06, + "loss": 0.6419, + "step": 1289 + }, + { + "epoch": 1.06791827719492, + "grad_norm": 0.9263116717338562, + "learning_rate": 4.665246767752376e-06, + "loss": 0.6226, + "step": 1290 + }, + { + "epoch": 1.0687465488680288, + "grad_norm": 0.9110594987869263, + "learning_rate": 4.66469685324261e-06, + "loss": 0.6495, + "step": 1291 + }, + { + "epoch": 1.0695748205411375, + "grad_norm": 0.9086888432502747, + "learning_rate": 4.664146519884182e-06, + "loss": 0.6408, + "step": 1292 + }, + { + "epoch": 1.0704030922142462, + "grad_norm": 0.8991205096244812, + "learning_rate": 4.663595767783575e-06, + "loss": 0.6424, + "step": 1293 + }, + { + "epoch": 1.071231363887355, + "grad_norm": 0.926933228969574, + "learning_rate": 4.6630445970473576e-06, + "loss": 0.6379, + "step": 1294 + }, + { + "epoch": 1.072059635560464, + "grad_norm": 0.8793080449104309, + "learning_rate": 4.662493007782174e-06, + "loss": 0.6254, + "step": 1295 + }, + { + "epoch": 1.0728879072335726, + "grad_norm": 0.9588155150413513, + "learning_rate": 4.661941000094751e-06, + "loss": 0.6307, + "step": 1296 + }, + { + "epoch": 1.0737161789066814, + "grad_norm": 0.9468860030174255, + "learning_rate": 4.661388574091898e-06, + "loss": 0.6308, + "step": 1297 + }, + { + "epoch": 1.07454445057979, + "grad_norm": 0.8930124640464783, + "learning_rate": 4.660835729880505e-06, + "loss": 0.6371, + "step": 1298 + }, + { + "epoch": 1.075372722252899, + "grad_norm": 0.9233878254890442, + "learning_rate": 4.660282467567541e-06, + "loss": 0.6327, + "step": 1299 + }, + { + "epoch": 1.0762009939260078, + "grad_norm": 0.9714951515197754, + "learning_rate": 4.659728787260057e-06, + "loss": 0.6218, + "step": 1300 + }, + { + "epoch": 1.0770292655991165, + "grad_norm": 0.9464428424835205, + "learning_rate": 4.659174689065185e-06, + "loss": 0.6406, + "step": 1301 + }, + { + "epoch": 1.0778575372722252, + "grad_norm": 0.8984559774398804, + "learning_rate": 4.6586201730901395e-06, + "loss": 0.6483, + "step": 1302 + }, + { + "epoch": 1.0786858089453342, + "grad_norm": 0.9007774591445923, + "learning_rate": 4.658065239442212e-06, + "loss": 0.6288, + "step": 1303 + }, + { + "epoch": 1.079514080618443, + "grad_norm": 0.907171368598938, + "learning_rate": 4.657509888228779e-06, + "loss": 0.6445, + "step": 1304 + }, + { + "epoch": 1.0803423522915516, + "grad_norm": 0.9207282066345215, + "learning_rate": 4.656954119557293e-06, + "loss": 0.6148, + "step": 1305 + }, + { + "epoch": 1.0811706239646603, + "grad_norm": 0.9467616677284241, + "learning_rate": 4.656397933535293e-06, + "loss": 0.6374, + "step": 1306 + }, + { + "epoch": 1.0819988956377693, + "grad_norm": 0.9310930967330933, + "learning_rate": 4.655841330270393e-06, + "loss": 0.6362, + "step": 1307 + }, + { + "epoch": 1.082827167310878, + "grad_norm": 0.890816330909729, + "learning_rate": 4.655284309870294e-06, + "loss": 0.6372, + "step": 1308 + }, + { + "epoch": 1.0836554389839868, + "grad_norm": 0.9206095933914185, + "learning_rate": 4.654726872442771e-06, + "loss": 0.6297, + "step": 1309 + }, + { + "epoch": 1.0844837106570955, + "grad_norm": 0.9015856385231018, + "learning_rate": 4.654169018095685e-06, + "loss": 0.658, + "step": 1310 + }, + { + "epoch": 1.0853119823302042, + "grad_norm": 0.8909441828727722, + "learning_rate": 4.653610746936975e-06, + "loss": 0.6382, + "step": 1311 + }, + { + "epoch": 1.0861402540033132, + "grad_norm": 0.9035969972610474, + "learning_rate": 4.653052059074661e-06, + "loss": 0.6441, + "step": 1312 + }, + { + "epoch": 1.0869685256764219, + "grad_norm": 0.8988869786262512, + "learning_rate": 4.652492954616844e-06, + "loss": 0.6359, + "step": 1313 + }, + { + "epoch": 1.0877967973495306, + "grad_norm": 0.881885826587677, + "learning_rate": 4.6519334336717055e-06, + "loss": 0.6397, + "step": 1314 + }, + { + "epoch": 1.0886250690226393, + "grad_norm": 0.980633020401001, + "learning_rate": 4.651373496347508e-06, + "loss": 0.6223, + "step": 1315 + }, + { + "epoch": 1.0894533406957483, + "grad_norm": 0.9027420878410339, + "learning_rate": 4.650813142752593e-06, + "loss": 0.6211, + "step": 1316 + }, + { + "epoch": 1.090281612368857, + "grad_norm": 0.9011778831481934, + "learning_rate": 4.6502523729953855e-06, + "loss": 0.6442, + "step": 1317 + }, + { + "epoch": 1.0911098840419657, + "grad_norm": 0.926692545413971, + "learning_rate": 4.649691187184387e-06, + "loss": 0.6661, + "step": 1318 + }, + { + "epoch": 1.0919381557150745, + "grad_norm": 0.9758741855621338, + "learning_rate": 4.649129585428184e-06, + "loss": 0.6461, + "step": 1319 + }, + { + "epoch": 1.0927664273881834, + "grad_norm": 0.967698335647583, + "learning_rate": 4.64856756783544e-06, + "loss": 0.6461, + "step": 1320 + }, + { + "epoch": 1.0935946990612921, + "grad_norm": 0.9372200965881348, + "learning_rate": 4.6480051345149e-06, + "loss": 0.6348, + "step": 1321 + }, + { + "epoch": 1.0944229707344009, + "grad_norm": 0.8976414799690247, + "learning_rate": 4.64744228557539e-06, + "loss": 0.6297, + "step": 1322 + }, + { + "epoch": 1.0952512424075096, + "grad_norm": 0.9163550734519958, + "learning_rate": 4.646879021125816e-06, + "loss": 0.6746, + "step": 1323 + }, + { + "epoch": 1.0960795140806185, + "grad_norm": 0.9064576625823975, + "learning_rate": 4.646315341275164e-06, + "loss": 0.6242, + "step": 1324 + }, + { + "epoch": 1.0969077857537273, + "grad_norm": 0.9352027773857117, + "learning_rate": 4.645751246132501e-06, + "loss": 0.6568, + "step": 1325 + }, + { + "epoch": 1.097736057426836, + "grad_norm": 0.8680023550987244, + "learning_rate": 4.645186735806976e-06, + "loss": 0.6484, + "step": 1326 + }, + { + "epoch": 1.0985643290999447, + "grad_norm": 0.8782095909118652, + "learning_rate": 4.644621810407813e-06, + "loss": 0.646, + "step": 1327 + }, + { + "epoch": 1.0993926007730535, + "grad_norm": 0.8931819796562195, + "learning_rate": 4.644056470044323e-06, + "loss": 0.6708, + "step": 1328 + }, + { + "epoch": 1.1002208724461624, + "grad_norm": 0.9228624105453491, + "learning_rate": 4.6434907148258915e-06, + "loss": 0.6255, + "step": 1329 + }, + { + "epoch": 1.1010491441192711, + "grad_norm": 0.8960729837417603, + "learning_rate": 4.642924544861988e-06, + "loss": 0.6523, + "step": 1330 + }, + { + "epoch": 1.1018774157923799, + "grad_norm": 0.9092081785202026, + "learning_rate": 4.642357960262162e-06, + "loss": 0.6534, + "step": 1331 + }, + { + "epoch": 1.1027056874654886, + "grad_norm": 0.9024838209152222, + "learning_rate": 4.641790961136041e-06, + "loss": 0.6238, + "step": 1332 + }, + { + "epoch": 1.1035339591385975, + "grad_norm": 0.9016388058662415, + "learning_rate": 4.641223547593334e-06, + "loss": 0.6643, + "step": 1333 + }, + { + "epoch": 1.1043622308117063, + "grad_norm": 0.8877143859863281, + "learning_rate": 4.6406557197438324e-06, + "loss": 0.6222, + "step": 1334 + }, + { + "epoch": 1.105190502484815, + "grad_norm": 0.9066653251647949, + "learning_rate": 4.640087477697403e-06, + "loss": 0.6205, + "step": 1335 + }, + { + "epoch": 1.1060187741579237, + "grad_norm": 0.903465747833252, + "learning_rate": 4.639518821563997e-06, + "loss": 0.6299, + "step": 1336 + }, + { + "epoch": 1.1068470458310327, + "grad_norm": 0.8908408880233765, + "learning_rate": 4.6389497514536415e-06, + "loss": 0.6247, + "step": 1337 + }, + { + "epoch": 1.1076753175041414, + "grad_norm": 0.8957250714302063, + "learning_rate": 4.63838026747645e-06, + "loss": 0.632, + "step": 1338 + }, + { + "epoch": 1.1085035891772501, + "grad_norm": 0.938494086265564, + "learning_rate": 4.637810369742609e-06, + "loss": 0.6309, + "step": 1339 + }, + { + "epoch": 1.1093318608503588, + "grad_norm": 0.9913409948348999, + "learning_rate": 4.637240058362391e-06, + "loss": 0.6571, + "step": 1340 + }, + { + "epoch": 1.1101601325234678, + "grad_norm": 0.9037449359893799, + "learning_rate": 4.636669333446145e-06, + "loss": 0.6467, + "step": 1341 + }, + { + "epoch": 1.1109884041965765, + "grad_norm": 0.911696195602417, + "learning_rate": 4.636098195104299e-06, + "loss": 0.6361, + "step": 1342 + }, + { + "epoch": 1.1118166758696852, + "grad_norm": 0.9105570316314697, + "learning_rate": 4.6355266434473665e-06, + "loss": 0.6418, + "step": 1343 + }, + { + "epoch": 1.112644947542794, + "grad_norm": 0.9101701378822327, + "learning_rate": 4.634954678585935e-06, + "loss": 0.6454, + "step": 1344 + }, + { + "epoch": 1.1134732192159027, + "grad_norm": 0.8980433344841003, + "learning_rate": 4.634382300630675e-06, + "loss": 0.6159, + "step": 1345 + }, + { + "epoch": 1.1143014908890116, + "grad_norm": 0.9253267049789429, + "learning_rate": 4.633809509692336e-06, + "loss": 0.6465, + "step": 1346 + }, + { + "epoch": 1.1151297625621204, + "grad_norm": 0.9150062203407288, + "learning_rate": 4.6332363058817484e-06, + "loss": 0.639, + "step": 1347 + }, + { + "epoch": 1.115958034235229, + "grad_norm": 0.9257630705833435, + "learning_rate": 4.632662689309821e-06, + "loss": 0.6571, + "step": 1348 + }, + { + "epoch": 1.116786305908338, + "grad_norm": 0.9183084964752197, + "learning_rate": 4.632088660087545e-06, + "loss": 0.6291, + "step": 1349 + }, + { + "epoch": 1.1176145775814468, + "grad_norm": 0.9229403138160706, + "learning_rate": 4.631514218325987e-06, + "loss": 0.6487, + "step": 1350 + }, + { + "epoch": 1.1184428492545555, + "grad_norm": 0.9379578828811646, + "learning_rate": 4.630939364136298e-06, + "loss": 0.6564, + "step": 1351 + }, + { + "epoch": 1.1192711209276642, + "grad_norm": 0.9094021320343018, + "learning_rate": 4.630364097629706e-06, + "loss": 0.6354, + "step": 1352 + }, + { + "epoch": 1.120099392600773, + "grad_norm": 0.9528741836547852, + "learning_rate": 4.62978841891752e-06, + "loss": 0.6259, + "step": 1353 + }, + { + "epoch": 1.120927664273882, + "grad_norm": 0.8859900832176208, + "learning_rate": 4.629212328111129e-06, + "loss": 0.6783, + "step": 1354 + }, + { + "epoch": 1.1217559359469906, + "grad_norm": 0.8739688396453857, + "learning_rate": 4.628635825322e-06, + "loss": 0.6518, + "step": 1355 + }, + { + "epoch": 1.1225842076200994, + "grad_norm": 0.8773418068885803, + "learning_rate": 4.628058910661683e-06, + "loss": 0.6477, + "step": 1356 + }, + { + "epoch": 1.123412479293208, + "grad_norm": 0.9025301933288574, + "learning_rate": 4.627481584241803e-06, + "loss": 0.6471, + "step": 1357 + }, + { + "epoch": 1.124240750966317, + "grad_norm": 0.89899080991745, + "learning_rate": 4.62690384617407e-06, + "loss": 0.6196, + "step": 1358 + }, + { + "epoch": 1.1250690226394258, + "grad_norm": 0.9514591693878174, + "learning_rate": 4.6263256965702684e-06, + "loss": 0.6495, + "step": 1359 + }, + { + "epoch": 1.1258972943125345, + "grad_norm": 0.908789336681366, + "learning_rate": 4.6257471355422654e-06, + "loss": 0.6606, + "step": 1360 + }, + { + "epoch": 1.1267255659856432, + "grad_norm": 0.9262552857398987, + "learning_rate": 4.625168163202009e-06, + "loss": 0.6308, + "step": 1361 + }, + { + "epoch": 1.1275538376587522, + "grad_norm": 0.9395532608032227, + "learning_rate": 4.624588779661523e-06, + "loss": 0.6196, + "step": 1362 + }, + { + "epoch": 1.128382109331861, + "grad_norm": 0.9126531481742859, + "learning_rate": 4.624008985032913e-06, + "loss": 0.6474, + "step": 1363 + }, + { + "epoch": 1.1292103810049696, + "grad_norm": 0.9105198383331299, + "learning_rate": 4.623428779428365e-06, + "loss": 0.6455, + "step": 1364 + }, + { + "epoch": 1.1300386526780783, + "grad_norm": 0.9048190712928772, + "learning_rate": 4.6228481629601415e-06, + "loss": 0.6647, + "step": 1365 + }, + { + "epoch": 1.1308669243511873, + "grad_norm": 0.9415149092674255, + "learning_rate": 4.622267135740588e-06, + "loss": 0.6403, + "step": 1366 + }, + { + "epoch": 1.131695196024296, + "grad_norm": 0.954728901386261, + "learning_rate": 4.6216856978821275e-06, + "loss": 0.6351, + "step": 1367 + }, + { + "epoch": 1.1325234676974048, + "grad_norm": 0.9392480850219727, + "learning_rate": 4.621103849497261e-06, + "loss": 0.6481, + "step": 1368 + }, + { + "epoch": 1.1333517393705135, + "grad_norm": 0.9199283719062805, + "learning_rate": 4.620521590698574e-06, + "loss": 0.6578, + "step": 1369 + }, + { + "epoch": 1.1341800110436222, + "grad_norm": 0.9175401926040649, + "learning_rate": 4.6199389215987266e-06, + "loss": 0.6508, + "step": 1370 + }, + { + "epoch": 1.1350082827167312, + "grad_norm": 0.9140470027923584, + "learning_rate": 4.619355842310459e-06, + "loss": 0.6536, + "step": 1371 + }, + { + "epoch": 1.1358365543898399, + "grad_norm": 0.8863652348518372, + "learning_rate": 4.618772352946593e-06, + "loss": 0.6376, + "step": 1372 + }, + { + "epoch": 1.1366648260629486, + "grad_norm": 0.9074656963348389, + "learning_rate": 4.618188453620026e-06, + "loss": 0.6253, + "step": 1373 + }, + { + "epoch": 1.1374930977360573, + "grad_norm": 0.891055703163147, + "learning_rate": 4.61760414444374e-06, + "loss": 0.6507, + "step": 1374 + }, + { + "epoch": 1.1383213694091663, + "grad_norm": 0.9153634905815125, + "learning_rate": 4.617019425530791e-06, + "loss": 0.6699, + "step": 1375 + }, + { + "epoch": 1.139149641082275, + "grad_norm": 0.9172940850257874, + "learning_rate": 4.6164342969943196e-06, + "loss": 0.6385, + "step": 1376 + }, + { + "epoch": 1.1399779127553837, + "grad_norm": 0.8978968262672424, + "learning_rate": 4.615848758947539e-06, + "loss": 0.6394, + "step": 1377 + }, + { + "epoch": 1.1408061844284925, + "grad_norm": 0.9372277855873108, + "learning_rate": 4.615262811503749e-06, + "loss": 0.6418, + "step": 1378 + }, + { + "epoch": 1.1416344561016014, + "grad_norm": 0.9596112370491028, + "learning_rate": 4.614676454776321e-06, + "loss": 0.6573, + "step": 1379 + }, + { + "epoch": 1.1424627277747101, + "grad_norm": 0.9243675470352173, + "learning_rate": 4.614089688878713e-06, + "loss": 0.6546, + "step": 1380 + }, + { + "epoch": 1.1432909994478189, + "grad_norm": 0.8839207291603088, + "learning_rate": 4.613502513924459e-06, + "loss": 0.654, + "step": 1381 + }, + { + "epoch": 1.1441192711209276, + "grad_norm": 0.8940262198448181, + "learning_rate": 4.612914930027168e-06, + "loss": 0.6223, + "step": 1382 + }, + { + "epoch": 1.1449475427940365, + "grad_norm": 0.9091162085533142, + "learning_rate": 4.612326937300535e-06, + "loss": 0.6512, + "step": 1383 + }, + { + "epoch": 1.1457758144671453, + "grad_norm": 0.9332976937294006, + "learning_rate": 4.6117385358583315e-06, + "loss": 0.6373, + "step": 1384 + }, + { + "epoch": 1.146604086140254, + "grad_norm": 0.9103875756263733, + "learning_rate": 4.611149725814406e-06, + "loss": 0.6502, + "step": 1385 + }, + { + "epoch": 1.1474323578133627, + "grad_norm": 0.9004862308502197, + "learning_rate": 4.610560507282688e-06, + "loss": 0.6361, + "step": 1386 + }, + { + "epoch": 1.1482606294864715, + "grad_norm": 0.880293071269989, + "learning_rate": 4.609970880377187e-06, + "loss": 0.655, + "step": 1387 + }, + { + "epoch": 1.1490889011595804, + "grad_norm": 0.9099947214126587, + "learning_rate": 4.60938084521199e-06, + "loss": 0.6517, + "step": 1388 + }, + { + "epoch": 1.1499171728326891, + "grad_norm": 0.8913806080818176, + "learning_rate": 4.608790401901262e-06, + "loss": 0.6429, + "step": 1389 + }, + { + "epoch": 1.1507454445057979, + "grad_norm": 0.9198290109634399, + "learning_rate": 4.608199550559249e-06, + "loss": 0.6537, + "step": 1390 + }, + { + "epoch": 1.1515737161789068, + "grad_norm": 0.9192278981208801, + "learning_rate": 4.607608291300275e-06, + "loss": 0.6531, + "step": 1391 + }, + { + "epoch": 1.1524019878520155, + "grad_norm": 0.8977665305137634, + "learning_rate": 4.607016624238744e-06, + "loss": 0.6358, + "step": 1392 + }, + { + "epoch": 1.1532302595251243, + "grad_norm": 0.9576854705810547, + "learning_rate": 4.606424549489138e-06, + "loss": 0.6493, + "step": 1393 + }, + { + "epoch": 1.154058531198233, + "grad_norm": 0.905611515045166, + "learning_rate": 4.605832067166017e-06, + "loss": 0.63, + "step": 1394 + }, + { + "epoch": 1.1548868028713417, + "grad_norm": 0.8840015530586243, + "learning_rate": 4.605239177384021e-06, + "loss": 0.6504, + "step": 1395 + }, + { + "epoch": 1.1557150745444507, + "grad_norm": 0.9247309565544128, + "learning_rate": 4.604645880257869e-06, + "loss": 0.6477, + "step": 1396 + }, + { + "epoch": 1.1565433462175594, + "grad_norm": 0.9473649263381958, + "learning_rate": 4.604052175902359e-06, + "loss": 0.6374, + "step": 1397 + }, + { + "epoch": 1.1573716178906681, + "grad_norm": 0.9354954361915588, + "learning_rate": 4.603458064432367e-06, + "loss": 0.6491, + "step": 1398 + }, + { + "epoch": 1.1581998895637768, + "grad_norm": 0.8788221478462219, + "learning_rate": 4.6028635459628476e-06, + "loss": 0.6187, + "step": 1399 + }, + { + "epoch": 1.1590281612368858, + "grad_norm": 0.8836075067520142, + "learning_rate": 4.602268620608834e-06, + "loss": 0.665, + "step": 1400 + }, + { + "epoch": 1.1598564329099945, + "grad_norm": 0.9195554256439209, + "learning_rate": 4.60167328848544e-06, + "loss": 0.6291, + "step": 1401 + }, + { + "epoch": 1.1606847045831032, + "grad_norm": 0.8984520435333252, + "learning_rate": 4.601077549707856e-06, + "loss": 0.6361, + "step": 1402 + }, + { + "epoch": 1.161512976256212, + "grad_norm": 0.8667109608650208, + "learning_rate": 4.600481404391352e-06, + "loss": 0.6345, + "step": 1403 + }, + { + "epoch": 1.1623412479293207, + "grad_norm": 0.9312148094177246, + "learning_rate": 4.599884852651277e-06, + "loss": 0.6388, + "step": 1404 + }, + { + "epoch": 1.1631695196024296, + "grad_norm": 0.9096866846084595, + "learning_rate": 4.5992878946030575e-06, + "loss": 0.6383, + "step": 1405 + }, + { + "epoch": 1.1639977912755384, + "grad_norm": 0.9010055661201477, + "learning_rate": 4.5986905303622e-06, + "loss": 0.6524, + "step": 1406 + }, + { + "epoch": 1.164826062948647, + "grad_norm": 0.9414717555046082, + "learning_rate": 4.598092760044288e-06, + "loss": 0.6825, + "step": 1407 + }, + { + "epoch": 1.165654334621756, + "grad_norm": 0.9162404537200928, + "learning_rate": 4.597494583764987e-06, + "loss": 0.6467, + "step": 1408 + }, + { + "epoch": 1.1664826062948648, + "grad_norm": 0.8935617208480835, + "learning_rate": 4.596896001640034e-06, + "loss": 0.6375, + "step": 1409 + }, + { + "epoch": 1.1673108779679735, + "grad_norm": 0.8976510167121887, + "learning_rate": 4.596297013785253e-06, + "loss": 0.6135, + "step": 1410 + }, + { + "epoch": 1.1681391496410822, + "grad_norm": 0.9503909349441528, + "learning_rate": 4.5956976203165414e-06, + "loss": 0.6305, + "step": 1411 + }, + { + "epoch": 1.168967421314191, + "grad_norm": 0.9147273302078247, + "learning_rate": 4.595097821349876e-06, + "loss": 0.6548, + "step": 1412 + }, + { + "epoch": 1.1697956929873, + "grad_norm": 0.9171034693717957, + "learning_rate": 4.594497617001312e-06, + "loss": 0.6253, + "step": 1413 + }, + { + "epoch": 1.1706239646604086, + "grad_norm": 0.8970120549201965, + "learning_rate": 4.593897007386985e-06, + "loss": 0.6414, + "step": 1414 + }, + { + "epoch": 1.1714522363335174, + "grad_norm": 0.9032952189445496, + "learning_rate": 4.593295992623105e-06, + "loss": 0.6595, + "step": 1415 + }, + { + "epoch": 1.172280508006626, + "grad_norm": 0.9066448211669922, + "learning_rate": 4.592694572825964e-06, + "loss": 0.6654, + "step": 1416 + }, + { + "epoch": 1.173108779679735, + "grad_norm": 0.8747309446334839, + "learning_rate": 4.592092748111931e-06, + "loss": 0.6619, + "step": 1417 + }, + { + "epoch": 1.1739370513528438, + "grad_norm": 0.8836600184440613, + "learning_rate": 4.5914905185974535e-06, + "loss": 0.6443, + "step": 1418 + }, + { + "epoch": 1.1747653230259525, + "grad_norm": 0.9340765476226807, + "learning_rate": 4.590887884399058e-06, + "loss": 0.6633, + "step": 1419 + }, + { + "epoch": 1.1755935946990612, + "grad_norm": 0.8889044523239136, + "learning_rate": 4.590284845633347e-06, + "loss": 0.6489, + "step": 1420 + }, + { + "epoch": 1.1764218663721702, + "grad_norm": 0.9043524265289307, + "learning_rate": 4.589681402417005e-06, + "loss": 0.6392, + "step": 1421 + }, + { + "epoch": 1.177250138045279, + "grad_norm": 0.9210046529769897, + "learning_rate": 4.5890775548667895e-06, + "loss": 0.6105, + "step": 1422 + }, + { + "epoch": 1.1780784097183876, + "grad_norm": 0.8954651951789856, + "learning_rate": 4.588473303099542e-06, + "loss": 0.6425, + "step": 1423 + }, + { + "epoch": 1.1789066813914963, + "grad_norm": 0.945517897605896, + "learning_rate": 4.587868647232179e-06, + "loss": 0.6404, + "step": 1424 + }, + { + "epoch": 1.1797349530646053, + "grad_norm": 0.9059463739395142, + "learning_rate": 4.587263587381696e-06, + "loss": 0.6521, + "step": 1425 + }, + { + "epoch": 1.180563224737714, + "grad_norm": 0.8876296281814575, + "learning_rate": 4.586658123665165e-06, + "loss": 0.6447, + "step": 1426 + }, + { + "epoch": 1.1813914964108228, + "grad_norm": 0.8862487077713013, + "learning_rate": 4.58605225619974e-06, + "loss": 0.6445, + "step": 1427 + }, + { + "epoch": 1.1822197680839315, + "grad_norm": 0.9066351056098938, + "learning_rate": 4.585445985102649e-06, + "loss": 0.6563, + "step": 1428 + }, + { + "epoch": 1.1830480397570402, + "grad_norm": 0.913724958896637, + "learning_rate": 4.584839310491199e-06, + "loss": 0.6507, + "step": 1429 + }, + { + "epoch": 1.1838763114301492, + "grad_norm": 0.8942010402679443, + "learning_rate": 4.584232232482778e-06, + "loss": 0.6424, + "step": 1430 + }, + { + "epoch": 1.1847045831032579, + "grad_norm": 0.8986000418663025, + "learning_rate": 4.58362475119485e-06, + "loss": 0.6491, + "step": 1431 + }, + { + "epoch": 1.1855328547763666, + "grad_norm": 0.8879220485687256, + "learning_rate": 4.5830168667449544e-06, + "loss": 0.6412, + "step": 1432 + }, + { + "epoch": 1.1863611264494753, + "grad_norm": 0.9497302770614624, + "learning_rate": 4.5824085792507135e-06, + "loss": 0.6453, + "step": 1433 + }, + { + "epoch": 1.1871893981225843, + "grad_norm": 0.8933898210525513, + "learning_rate": 4.581799888829824e-06, + "loss": 0.6381, + "step": 1434 + }, + { + "epoch": 1.188017669795693, + "grad_norm": 0.9034610986709595, + "learning_rate": 4.581190795600062e-06, + "loss": 0.6551, + "step": 1435 + }, + { + "epoch": 1.1888459414688017, + "grad_norm": 0.8968634605407715, + "learning_rate": 4.580581299679282e-06, + "loss": 0.637, + "step": 1436 + }, + { + "epoch": 1.1896742131419105, + "grad_norm": 0.8974384069442749, + "learning_rate": 4.579971401185416e-06, + "loss": 0.6301, + "step": 1437 + }, + { + "epoch": 1.1905024848150194, + "grad_norm": 0.8726125359535217, + "learning_rate": 4.579361100236473e-06, + "loss": 0.6487, + "step": 1438 + }, + { + "epoch": 1.1913307564881281, + "grad_norm": 0.9131139516830444, + "learning_rate": 4.5787503969505405e-06, + "loss": 0.6211, + "step": 1439 + }, + { + "epoch": 1.1921590281612369, + "grad_norm": 0.8924953937530518, + "learning_rate": 4.578139291445783e-06, + "loss": 0.6339, + "step": 1440 + }, + { + "epoch": 1.1929872998343456, + "grad_norm": 0.890883207321167, + "learning_rate": 4.577527783840447e-06, + "loss": 0.66, + "step": 1441 + }, + { + "epoch": 1.1938155715074545, + "grad_norm": 0.9326318502426147, + "learning_rate": 4.5769158742528494e-06, + "loss": 0.6611, + "step": 1442 + }, + { + "epoch": 1.1946438431805633, + "grad_norm": 0.9354713559150696, + "learning_rate": 4.576303562801392e-06, + "loss": 0.6445, + "step": 1443 + }, + { + "epoch": 1.195472114853672, + "grad_norm": 0.8868383169174194, + "learning_rate": 4.57569084960455e-06, + "loss": 0.6344, + "step": 1444 + }, + { + "epoch": 1.1963003865267807, + "grad_norm": 0.893924355506897, + "learning_rate": 4.5750777347808775e-06, + "loss": 0.602, + "step": 1445 + }, + { + "epoch": 1.1971286581998895, + "grad_norm": 0.9299339652061462, + "learning_rate": 4.574464218449006e-06, + "loss": 0.6206, + "step": 1446 + }, + { + "epoch": 1.1979569298729984, + "grad_norm": 0.9058142304420471, + "learning_rate": 4.573850300727648e-06, + "loss": 0.6073, + "step": 1447 + }, + { + "epoch": 1.1987852015461071, + "grad_norm": 0.8893733620643616, + "learning_rate": 4.5732359817355885e-06, + "loss": 0.6336, + "step": 1448 + }, + { + "epoch": 1.1996134732192159, + "grad_norm": 0.9018000960350037, + "learning_rate": 4.572621261591692e-06, + "loss": 0.6268, + "step": 1449 + }, + { + "epoch": 1.2004417448923248, + "grad_norm": 0.9087574481964111, + "learning_rate": 4.572006140414903e-06, + "loss": 0.6511, + "step": 1450 + }, + { + "epoch": 1.2012700165654335, + "grad_norm": 0.9092984795570374, + "learning_rate": 4.5713906183242404e-06, + "loss": 0.6424, + "step": 1451 + }, + { + "epoch": 1.2020982882385423, + "grad_norm": 0.9040535092353821, + "learning_rate": 4.5707746954388034e-06, + "loss": 0.6242, + "step": 1452 + }, + { + "epoch": 1.202926559911651, + "grad_norm": 0.891659677028656, + "learning_rate": 4.570158371877766e-06, + "loss": 0.643, + "step": 1453 + }, + { + "epoch": 1.2037548315847597, + "grad_norm": 0.9254544973373413, + "learning_rate": 4.569541647760381e-06, + "loss": 0.6191, + "step": 1454 + }, + { + "epoch": 1.2045831032578687, + "grad_norm": 0.8978536128997803, + "learning_rate": 4.568924523205979e-06, + "loss": 0.6632, + "step": 1455 + }, + { + "epoch": 1.2054113749309774, + "grad_norm": 0.9179201126098633, + "learning_rate": 4.568306998333968e-06, + "loss": 0.6485, + "step": 1456 + }, + { + "epoch": 1.2062396466040861, + "grad_norm": 0.9061882495880127, + "learning_rate": 4.567689073263834e-06, + "loss": 0.6704, + "step": 1457 + }, + { + "epoch": 1.2070679182771948, + "grad_norm": 0.9143427610397339, + "learning_rate": 4.567070748115139e-06, + "loss": 0.6453, + "step": 1458 + }, + { + "epoch": 1.2078961899503038, + "grad_norm": 0.8740119338035583, + "learning_rate": 4.566452023007523e-06, + "loss": 0.6453, + "step": 1459 + }, + { + "epoch": 1.2087244616234125, + "grad_norm": 0.8977463841438293, + "learning_rate": 4.565832898060703e-06, + "loss": 0.6451, + "step": 1460 + }, + { + "epoch": 1.2095527332965212, + "grad_norm": 0.9416611194610596, + "learning_rate": 4.565213373394476e-06, + "loss": 0.6265, + "step": 1461 + }, + { + "epoch": 1.21038100496963, + "grad_norm": 0.9132000207901001, + "learning_rate": 4.5645934491287115e-06, + "loss": 0.6412, + "step": 1462 + }, + { + "epoch": 1.2112092766427387, + "grad_norm": 0.9480498433113098, + "learning_rate": 4.563973125383362e-06, + "loss": 0.6363, + "step": 1463 + }, + { + "epoch": 1.2120375483158476, + "grad_norm": 0.9170709848403931, + "learning_rate": 4.563352402278451e-06, + "loss": 0.6201, + "step": 1464 + }, + { + "epoch": 1.2128658199889564, + "grad_norm": 0.9655918478965759, + "learning_rate": 4.562731279934086e-06, + "loss": 0.6497, + "step": 1465 + }, + { + "epoch": 1.213694091662065, + "grad_norm": 0.8885562419891357, + "learning_rate": 4.562109758470446e-06, + "loss": 0.6265, + "step": 1466 + }, + { + "epoch": 1.214522363335174, + "grad_norm": 0.9072791337966919, + "learning_rate": 4.561487838007792e-06, + "loss": 0.6257, + "step": 1467 + }, + { + "epoch": 1.2153506350082828, + "grad_norm": 0.8992362022399902, + "learning_rate": 4.560865518666458e-06, + "loss": 0.6425, + "step": 1468 + }, + { + "epoch": 1.2161789066813915, + "grad_norm": 0.8934991955757141, + "learning_rate": 4.560242800566855e-06, + "loss": 0.6583, + "step": 1469 + }, + { + "epoch": 1.2170071783545002, + "grad_norm": 0.9292325973510742, + "learning_rate": 4.559619683829477e-06, + "loss": 0.6403, + "step": 1470 + }, + { + "epoch": 1.217835450027609, + "grad_norm": 0.8756814002990723, + "learning_rate": 4.55899616857489e-06, + "loss": 0.6261, + "step": 1471 + }, + { + "epoch": 1.218663721700718, + "grad_norm": 0.9050565361976624, + "learning_rate": 4.558372254923738e-06, + "loss": 0.6325, + "step": 1472 + }, + { + "epoch": 1.2194919933738266, + "grad_norm": 0.9230931401252747, + "learning_rate": 4.557747942996742e-06, + "loss": 0.6275, + "step": 1473 + }, + { + "epoch": 1.2203202650469354, + "grad_norm": 0.894026517868042, + "learning_rate": 4.557123232914702e-06, + "loss": 0.6348, + "step": 1474 + }, + { + "epoch": 1.221148536720044, + "grad_norm": 0.8882637023925781, + "learning_rate": 4.556498124798492e-06, + "loss": 0.6123, + "step": 1475 + }, + { + "epoch": 1.221976808393153, + "grad_norm": 0.9191063642501831, + "learning_rate": 4.555872618769065e-06, + "loss": 0.6531, + "step": 1476 + }, + { + "epoch": 1.2228050800662618, + "grad_norm": 0.9398623704910278, + "learning_rate": 4.555246714947452e-06, + "loss": 0.6553, + "step": 1477 + }, + { + "epoch": 1.2236333517393705, + "grad_norm": 0.9399452805519104, + "learning_rate": 4.554620413454757e-06, + "loss": 0.6485, + "step": 1478 + }, + { + "epoch": 1.2244616234124792, + "grad_norm": 0.9001572728157043, + "learning_rate": 4.553993714412166e-06, + "loss": 0.6467, + "step": 1479 + }, + { + "epoch": 1.2252898950855882, + "grad_norm": 0.9266572594642639, + "learning_rate": 4.553366617940938e-06, + "loss": 0.6325, + "step": 1480 + }, + { + "epoch": 1.226118166758697, + "grad_norm": 0.928301215171814, + "learning_rate": 4.552739124162411e-06, + "loss": 0.6398, + "step": 1481 + }, + { + "epoch": 1.2269464384318056, + "grad_norm": 0.880954921245575, + "learning_rate": 4.552111233197999e-06, + "loss": 0.669, + "step": 1482 + }, + { + "epoch": 1.2277747101049143, + "grad_norm": 0.9089685082435608, + "learning_rate": 4.551482945169191e-06, + "loss": 0.62, + "step": 1483 + }, + { + "epoch": 1.2286029817780233, + "grad_norm": 0.9187189936637878, + "learning_rate": 4.550854260197559e-06, + "loss": 0.6381, + "step": 1484 + }, + { + "epoch": 1.229431253451132, + "grad_norm": 0.8782045245170593, + "learning_rate": 4.550225178404744e-06, + "loss": 0.6397, + "step": 1485 + }, + { + "epoch": 1.2302595251242407, + "grad_norm": 0.8798948526382446, + "learning_rate": 4.54959569991247e-06, + "loss": 0.6522, + "step": 1486 + }, + { + "epoch": 1.2310877967973495, + "grad_norm": 0.9257029294967651, + "learning_rate": 4.548965824842534e-06, + "loss": 0.6522, + "step": 1487 + }, + { + "epoch": 1.2319160684704582, + "grad_norm": 0.8923602104187012, + "learning_rate": 4.548335553316811e-06, + "loss": 0.6538, + "step": 1488 + }, + { + "epoch": 1.2327443401435672, + "grad_norm": 0.9071294069290161, + "learning_rate": 4.5477048854572524e-06, + "loss": 0.6305, + "step": 1489 + }, + { + "epoch": 1.2335726118166759, + "grad_norm": 0.9495804905891418, + "learning_rate": 4.547073821385888e-06, + "loss": 0.6398, + "step": 1490 + }, + { + "epoch": 1.2344008834897846, + "grad_norm": 0.9163299202919006, + "learning_rate": 4.54644236122482e-06, + "loss": 0.6454, + "step": 1491 + }, + { + "epoch": 1.2352291551628933, + "grad_norm": 0.9286475777626038, + "learning_rate": 4.545810505096233e-06, + "loss": 0.6345, + "step": 1492 + }, + { + "epoch": 1.2360574268360023, + "grad_norm": 0.9116879105567932, + "learning_rate": 4.545178253122384e-06, + "loss": 0.6451, + "step": 1493 + }, + { + "epoch": 1.236885698509111, + "grad_norm": 0.917826771736145, + "learning_rate": 4.544545605425607e-06, + "loss": 0.6374, + "step": 1494 + }, + { + "epoch": 1.2377139701822197, + "grad_norm": 0.9433108568191528, + "learning_rate": 4.543912562128316e-06, + "loss": 0.6321, + "step": 1495 + }, + { + "epoch": 1.2385422418553285, + "grad_norm": 0.9017132520675659, + "learning_rate": 4.543279123352996e-06, + "loss": 0.6341, + "step": 1496 + }, + { + "epoch": 1.2393705135284374, + "grad_norm": 0.9557158350944519, + "learning_rate": 4.542645289222214e-06, + "loss": 0.6364, + "step": 1497 + }, + { + "epoch": 1.2401987852015461, + "grad_norm": 0.9258891344070435, + "learning_rate": 4.542011059858609e-06, + "loss": 0.6598, + "step": 1498 + }, + { + "epoch": 1.2410270568746549, + "grad_norm": 0.8988100290298462, + "learning_rate": 4.541376435384899e-06, + "loss": 0.6496, + "step": 1499 + }, + { + "epoch": 1.2418553285477636, + "grad_norm": 0.898154079914093, + "learning_rate": 4.540741415923878e-06, + "loss": 0.6486, + "step": 1500 + }, + { + "epoch": 1.2426836002208725, + "grad_norm": 0.9216037392616272, + "learning_rate": 4.540106001598418e-06, + "loss": 0.6436, + "step": 1501 + }, + { + "epoch": 1.2435118718939813, + "grad_norm": 0.9086095690727234, + "learning_rate": 4.539470192531463e-06, + "loss": 0.6287, + "step": 1502 + }, + { + "epoch": 1.24434014356709, + "grad_norm": 0.921291172504425, + "learning_rate": 4.538833988846039e-06, + "loss": 0.629, + "step": 1503 + }, + { + "epoch": 1.2451684152401987, + "grad_norm": 0.8621019721031189, + "learning_rate": 4.538197390665242e-06, + "loss": 0.6543, + "step": 1504 + }, + { + "epoch": 1.2459966869133074, + "grad_norm": 0.9523105025291443, + "learning_rate": 4.537560398112251e-06, + "loss": 0.6482, + "step": 1505 + }, + { + "epoch": 1.2468249585864164, + "grad_norm": 0.971931517124176, + "learning_rate": 4.536923011310317e-06, + "loss": 0.6463, + "step": 1506 + }, + { + "epoch": 1.2476532302595251, + "grad_norm": 0.927219033241272, + "learning_rate": 4.536285230382767e-06, + "loss": 0.6443, + "step": 1507 + }, + { + "epoch": 1.2484815019326339, + "grad_norm": 0.9433056116104126, + "learning_rate": 4.535647055453007e-06, + "loss": 0.6301, + "step": 1508 + }, + { + "epoch": 1.2493097736057428, + "grad_norm": 0.8992047905921936, + "learning_rate": 4.5350084866445195e-06, + "loss": 0.6177, + "step": 1509 + }, + { + "epoch": 1.2501380452788515, + "grad_norm": 0.8946895003318787, + "learning_rate": 4.534369524080858e-06, + "loss": 0.6275, + "step": 1510 + }, + { + "epoch": 1.2509663169519603, + "grad_norm": 0.8732156157493591, + "learning_rate": 4.5337301678856595e-06, + "loss": 0.6572, + "step": 1511 + }, + { + "epoch": 1.251794588625069, + "grad_norm": 0.9505333304405212, + "learning_rate": 4.533090418182631e-06, + "loss": 0.6352, + "step": 1512 + }, + { + "epoch": 1.2526228602981777, + "grad_norm": 0.8931122422218323, + "learning_rate": 4.532450275095558e-06, + "loss": 0.649, + "step": 1513 + }, + { + "epoch": 1.2534511319712867, + "grad_norm": 0.8781880140304565, + "learning_rate": 4.531809738748304e-06, + "loss": 0.651, + "step": 1514 + }, + { + "epoch": 1.2542794036443954, + "grad_norm": 0.9201871156692505, + "learning_rate": 4.531168809264805e-06, + "loss": 0.6451, + "step": 1515 + }, + { + "epoch": 1.2551076753175041, + "grad_norm": 0.9318148493766785, + "learning_rate": 4.530527486769077e-06, + "loss": 0.6458, + "step": 1516 + }, + { + "epoch": 1.2559359469906128, + "grad_norm": 0.9131407141685486, + "learning_rate": 4.529885771385208e-06, + "loss": 0.6679, + "step": 1517 + }, + { + "epoch": 1.2567642186637218, + "grad_norm": 0.8942316770553589, + "learning_rate": 4.529243663237365e-06, + "loss": 0.6262, + "step": 1518 + }, + { + "epoch": 1.2575924903368305, + "grad_norm": 0.9058274030685425, + "learning_rate": 4.528601162449788e-06, + "loss": 0.635, + "step": 1519 + }, + { + "epoch": 1.2584207620099392, + "grad_norm": 0.9129453897476196, + "learning_rate": 4.527958269146798e-06, + "loss": 0.6375, + "step": 1520 + }, + { + "epoch": 1.259249033683048, + "grad_norm": 0.8737504482269287, + "learning_rate": 4.527314983452787e-06, + "loss": 0.6437, + "step": 1521 + }, + { + "epoch": 1.2600773053561567, + "grad_norm": 0.9680130481719971, + "learning_rate": 4.526671305492225e-06, + "loss": 0.6546, + "step": 1522 + }, + { + "epoch": 1.2609055770292656, + "grad_norm": 0.9186084270477295, + "learning_rate": 4.526027235389658e-06, + "loss": 0.6505, + "step": 1523 + }, + { + "epoch": 1.2617338487023744, + "grad_norm": 0.9164871573448181, + "learning_rate": 4.525382773269706e-06, + "loss": 0.6541, + "step": 1524 + }, + { + "epoch": 1.262562120375483, + "grad_norm": 0.8994123935699463, + "learning_rate": 4.5247379192570695e-06, + "loss": 0.6313, + "step": 1525 + }, + { + "epoch": 1.263390392048592, + "grad_norm": 0.8780331015586853, + "learning_rate": 4.524092673476519e-06, + "loss": 0.6496, + "step": 1526 + }, + { + "epoch": 1.2642186637217008, + "grad_norm": 0.8670293092727661, + "learning_rate": 4.5234470360529045e-06, + "loss": 0.6422, + "step": 1527 + }, + { + "epoch": 1.2650469353948095, + "grad_norm": 0.9305918216705322, + "learning_rate": 4.522801007111152e-06, + "loss": 0.6352, + "step": 1528 + }, + { + "epoch": 1.2658752070679182, + "grad_norm": 0.9444066286087036, + "learning_rate": 4.52215458677626e-06, + "loss": 0.6357, + "step": 1529 + }, + { + "epoch": 1.266703478741027, + "grad_norm": 0.8987369537353516, + "learning_rate": 4.521507775173305e-06, + "loss": 0.6407, + "step": 1530 + }, + { + "epoch": 1.267531750414136, + "grad_norm": 0.9200116395950317, + "learning_rate": 4.5208605724274415e-06, + "loss": 0.6366, + "step": 1531 + }, + { + "epoch": 1.2683600220872446, + "grad_norm": 0.8680561780929565, + "learning_rate": 4.520212978663894e-06, + "loss": 0.6389, + "step": 1532 + }, + { + "epoch": 1.2691882937603534, + "grad_norm": 0.8928949236869812, + "learning_rate": 4.519564994007968e-06, + "loss": 0.6552, + "step": 1533 + }, + { + "epoch": 1.2700165654334623, + "grad_norm": 0.8986184000968933, + "learning_rate": 4.518916618585042e-06, + "loss": 0.6554, + "step": 1534 + }, + { + "epoch": 1.270844837106571, + "grad_norm": 0.9239423274993896, + "learning_rate": 4.51826785252057e-06, + "loss": 0.6448, + "step": 1535 + }, + { + "epoch": 1.2716731087796798, + "grad_norm": 0.9534674286842346, + "learning_rate": 4.517618695940082e-06, + "loss": 0.6303, + "step": 1536 + }, + { + "epoch": 1.2725013804527885, + "grad_norm": 0.9341598749160767, + "learning_rate": 4.516969148969186e-06, + "loss": 0.6335, + "step": 1537 + }, + { + "epoch": 1.2733296521258972, + "grad_norm": 0.9264217615127563, + "learning_rate": 4.51631921173356e-06, + "loss": 0.6543, + "step": 1538 + }, + { + "epoch": 1.274157923799006, + "grad_norm": 0.9460370540618896, + "learning_rate": 4.515668884358963e-06, + "loss": 0.6531, + "step": 1539 + }, + { + "epoch": 1.274986195472115, + "grad_norm": 0.9690954089164734, + "learning_rate": 4.515018166971227e-06, + "loss": 0.6495, + "step": 1540 + }, + { + "epoch": 1.2758144671452236, + "grad_norm": 0.8746740818023682, + "learning_rate": 4.514367059696259e-06, + "loss": 0.6169, + "step": 1541 + }, + { + "epoch": 1.2766427388183323, + "grad_norm": 0.9288576245307922, + "learning_rate": 4.513715562660043e-06, + "loss": 0.6459, + "step": 1542 + }, + { + "epoch": 1.2774710104914413, + "grad_norm": 0.9935420751571655, + "learning_rate": 4.513063675988637e-06, + "loss": 0.6471, + "step": 1543 + }, + { + "epoch": 1.27829928216455, + "grad_norm": 0.9584968686103821, + "learning_rate": 4.512411399808175e-06, + "loss": 0.6473, + "step": 1544 + }, + { + "epoch": 1.2791275538376587, + "grad_norm": 0.9049419164657593, + "learning_rate": 4.511758734244867e-06, + "loss": 0.6667, + "step": 1545 + }, + { + "epoch": 1.2799558255107675, + "grad_norm": 0.9456910490989685, + "learning_rate": 4.511105679424998e-06, + "loss": 0.625, + "step": 1546 + }, + { + "epoch": 1.2807840971838762, + "grad_norm": 0.9267060160636902, + "learning_rate": 4.510452235474926e-06, + "loss": 0.6239, + "step": 1547 + }, + { + "epoch": 1.2816123688569852, + "grad_norm": 0.9363065958023071, + "learning_rate": 4.509798402521088e-06, + "loss": 0.6416, + "step": 1548 + }, + { + "epoch": 1.2824406405300939, + "grad_norm": 0.85664302110672, + "learning_rate": 4.509144180689995e-06, + "loss": 0.6297, + "step": 1549 + }, + { + "epoch": 1.2832689122032026, + "grad_norm": 0.9298888444900513, + "learning_rate": 4.508489570108231e-06, + "loss": 0.6305, + "step": 1550 + }, + { + "epoch": 1.2840971838763116, + "grad_norm": 0.9485051035881042, + "learning_rate": 4.507834570902459e-06, + "loss": 0.6409, + "step": 1551 + }, + { + "epoch": 1.2849254555494203, + "grad_norm": 0.9129849076271057, + "learning_rate": 4.507179183199414e-06, + "loss": 0.6215, + "step": 1552 + }, + { + "epoch": 1.285753727222529, + "grad_norm": 0.9630280137062073, + "learning_rate": 4.506523407125907e-06, + "loss": 0.6229, + "step": 1553 + }, + { + "epoch": 1.2865819988956377, + "grad_norm": 0.9047848582267761, + "learning_rate": 4.505867242808826e-06, + "loss": 0.6434, + "step": 1554 + }, + { + "epoch": 1.2874102705687465, + "grad_norm": 0.9219448566436768, + "learning_rate": 4.5052106903751315e-06, + "loss": 0.6449, + "step": 1555 + }, + { + "epoch": 1.2882385422418554, + "grad_norm": 0.9347684383392334, + "learning_rate": 4.504553749951861e-06, + "loss": 0.6272, + "step": 1556 + }, + { + "epoch": 1.2890668139149641, + "grad_norm": 0.8930262923240662, + "learning_rate": 4.503896421666126e-06, + "loss": 0.6364, + "step": 1557 + }, + { + "epoch": 1.2898950855880729, + "grad_norm": 0.9067763090133667, + "learning_rate": 4.503238705645114e-06, + "loss": 0.6317, + "step": 1558 + }, + { + "epoch": 1.2907233572611816, + "grad_norm": 0.9524715542793274, + "learning_rate": 4.5025806020160865e-06, + "loss": 0.6525, + "step": 1559 + }, + { + "epoch": 1.2915516289342905, + "grad_norm": 0.927255392074585, + "learning_rate": 4.5019221109063795e-06, + "loss": 0.6343, + "step": 1560 + }, + { + "epoch": 1.2923799006073993, + "grad_norm": 0.904295802116394, + "learning_rate": 4.5012632324434065e-06, + "loss": 0.6484, + "step": 1561 + }, + { + "epoch": 1.293208172280508, + "grad_norm": 0.9138978123664856, + "learning_rate": 4.500603966754653e-06, + "loss": 0.6374, + "step": 1562 + }, + { + "epoch": 1.2940364439536167, + "grad_norm": 0.9038779139518738, + "learning_rate": 4.499944313967681e-06, + "loss": 0.6395, + "step": 1563 + }, + { + "epoch": 1.2948647156267254, + "grad_norm": 0.9218693971633911, + "learning_rate": 4.499284274210128e-06, + "loss": 0.6325, + "step": 1564 + }, + { + "epoch": 1.2956929872998344, + "grad_norm": 0.9210799932479858, + "learning_rate": 4.4986238476097055e-06, + "loss": 0.6476, + "step": 1565 + }, + { + "epoch": 1.2965212589729431, + "grad_norm": 0.8977149724960327, + "learning_rate": 4.4979630342941994e-06, + "loss": 0.6489, + "step": 1566 + }, + { + "epoch": 1.2973495306460519, + "grad_norm": 0.9423681497573853, + "learning_rate": 4.49730183439147e-06, + "loss": 0.6391, + "step": 1567 + }, + { + "epoch": 1.2981778023191608, + "grad_norm": 0.9404032826423645, + "learning_rate": 4.4966402480294545e-06, + "loss": 0.6233, + "step": 1568 + }, + { + "epoch": 1.2990060739922695, + "grad_norm": 0.9104741215705872, + "learning_rate": 4.495978275336164e-06, + "loss": 0.6369, + "step": 1569 + }, + { + "epoch": 1.2998343456653783, + "grad_norm": 0.8869790434837341, + "learning_rate": 4.4953159164396835e-06, + "loss": 0.6331, + "step": 1570 + }, + { + "epoch": 1.300662617338487, + "grad_norm": 0.9269986152648926, + "learning_rate": 4.494653171468173e-06, + "loss": 0.6361, + "step": 1571 + }, + { + "epoch": 1.3014908890115957, + "grad_norm": 0.9210803508758545, + "learning_rate": 4.4939900405498675e-06, + "loss": 0.6314, + "step": 1572 + }, + { + "epoch": 1.3023191606847047, + "grad_norm": 0.9426820874214172, + "learning_rate": 4.4933265238130775e-06, + "loss": 0.6625, + "step": 1573 + }, + { + "epoch": 1.3031474323578134, + "grad_norm": 0.9198150634765625, + "learning_rate": 4.492662621386186e-06, + "loss": 0.627, + "step": 1574 + }, + { + "epoch": 1.3039757040309221, + "grad_norm": 0.9312534928321838, + "learning_rate": 4.4919983333976525e-06, + "loss": 0.6386, + "step": 1575 + }, + { + "epoch": 1.304803975704031, + "grad_norm": 0.8926329016685486, + "learning_rate": 4.491333659976011e-06, + "loss": 0.6559, + "step": 1576 + }, + { + "epoch": 1.3056322473771398, + "grad_norm": 0.9172425270080566, + "learning_rate": 4.4906686012498705e-06, + "loss": 0.6254, + "step": 1577 + }, + { + "epoch": 1.3064605190502485, + "grad_norm": 0.9896116256713867, + "learning_rate": 4.490003157347911e-06, + "loss": 0.6331, + "step": 1578 + }, + { + "epoch": 1.3072887907233572, + "grad_norm": 0.9267039895057678, + "learning_rate": 4.489337328398891e-06, + "loss": 0.6296, + "step": 1579 + }, + { + "epoch": 1.308117062396466, + "grad_norm": 0.9033195376396179, + "learning_rate": 4.488671114531643e-06, + "loss": 0.6354, + "step": 1580 + }, + { + "epoch": 1.3089453340695747, + "grad_norm": 0.9047731161117554, + "learning_rate": 4.488004515875072e-06, + "loss": 0.6351, + "step": 1581 + }, + { + "epoch": 1.3097736057426836, + "grad_norm": 0.9096142649650574, + "learning_rate": 4.4873375325581596e-06, + "loss": 0.6504, + "step": 1582 + }, + { + "epoch": 1.3106018774157924, + "grad_norm": 0.8848153352737427, + "learning_rate": 4.48667016470996e-06, + "loss": 0.6354, + "step": 1583 + }, + { + "epoch": 1.311430149088901, + "grad_norm": 0.9036188721656799, + "learning_rate": 4.486002412459603e-06, + "loss": 0.6431, + "step": 1584 + }, + { + "epoch": 1.31225842076201, + "grad_norm": 0.8881173133850098, + "learning_rate": 4.4853342759362924e-06, + "loss": 0.6509, + "step": 1585 + }, + { + "epoch": 1.3130866924351188, + "grad_norm": 0.9195591807365417, + "learning_rate": 4.484665755269307e-06, + "loss": 0.6522, + "step": 1586 + }, + { + "epoch": 1.3139149641082275, + "grad_norm": 0.9021509885787964, + "learning_rate": 4.483996850587998e-06, + "loss": 0.6532, + "step": 1587 + }, + { + "epoch": 1.3147432357813362, + "grad_norm": 0.8902403116226196, + "learning_rate": 4.483327562021794e-06, + "loss": 0.5942, + "step": 1588 + }, + { + "epoch": 1.315571507454445, + "grad_norm": 0.8933504819869995, + "learning_rate": 4.482657889700194e-06, + "loss": 0.6305, + "step": 1589 + }, + { + "epoch": 1.316399779127554, + "grad_norm": 0.9233856201171875, + "learning_rate": 4.481987833752776e-06, + "loss": 0.6275, + "step": 1590 + }, + { + "epoch": 1.3172280508006626, + "grad_norm": 0.9482141733169556, + "learning_rate": 4.4813173943091865e-06, + "loss": 0.6232, + "step": 1591 + }, + { + "epoch": 1.3180563224737714, + "grad_norm": 0.94732266664505, + "learning_rate": 4.480646571499151e-06, + "loss": 0.6297, + "step": 1592 + }, + { + "epoch": 1.3188845941468803, + "grad_norm": 0.9066787958145142, + "learning_rate": 4.479975365452468e-06, + "loss": 0.6334, + "step": 1593 + }, + { + "epoch": 1.319712865819989, + "grad_norm": 0.9063938856124878, + "learning_rate": 4.479303776299008e-06, + "loss": 0.6415, + "step": 1594 + }, + { + "epoch": 1.3205411374930978, + "grad_norm": 0.9286819696426392, + "learning_rate": 4.478631804168719e-06, + "loss": 0.6534, + "step": 1595 + }, + { + "epoch": 1.3213694091662065, + "grad_norm": 0.9161240458488464, + "learning_rate": 4.47795944919162e-06, + "loss": 0.6401, + "step": 1596 + }, + { + "epoch": 1.3221976808393152, + "grad_norm": 0.956538200378418, + "learning_rate": 4.4772867114978056e-06, + "loss": 0.6649, + "step": 1597 + }, + { + "epoch": 1.323025952512424, + "grad_norm": 0.910504162311554, + "learning_rate": 4.4766135912174455e-06, + "loss": 0.6456, + "step": 1598 + }, + { + "epoch": 1.323854224185533, + "grad_norm": 0.9054026007652283, + "learning_rate": 4.4759400884807805e-06, + "loss": 0.6458, + "step": 1599 + }, + { + "epoch": 1.3246824958586416, + "grad_norm": 0.933506429195404, + "learning_rate": 4.475266203418127e-06, + "loss": 0.626, + "step": 1600 + }, + { + "epoch": 1.3255107675317503, + "grad_norm": 0.9042014479637146, + "learning_rate": 4.474591936159878e-06, + "loss": 0.6376, + "step": 1601 + }, + { + "epoch": 1.3263390392048593, + "grad_norm": 0.9005138278007507, + "learning_rate": 4.473917286836496e-06, + "loss": 0.6093, + "step": 1602 + }, + { + "epoch": 1.327167310877968, + "grad_norm": 0.9096264839172363, + "learning_rate": 4.4732422555785196e-06, + "loss": 0.6344, + "step": 1603 + }, + { + "epoch": 1.3279955825510767, + "grad_norm": 0.9171631336212158, + "learning_rate": 4.4725668425165605e-06, + "loss": 0.6454, + "step": 1604 + }, + { + "epoch": 1.3288238542241855, + "grad_norm": 0.8821735382080078, + "learning_rate": 4.4718910477813055e-06, + "loss": 0.6357, + "step": 1605 + }, + { + "epoch": 1.3296521258972942, + "grad_norm": 0.918499231338501, + "learning_rate": 4.471214871503514e-06, + "loss": 0.6361, + "step": 1606 + }, + { + "epoch": 1.3304803975704032, + "grad_norm": 0.9187904596328735, + "learning_rate": 4.470538313814021e-06, + "loss": 0.6362, + "step": 1607 + }, + { + "epoch": 1.3313086692435119, + "grad_norm": 0.9272632598876953, + "learning_rate": 4.469861374843734e-06, + "loss": 0.6659, + "step": 1608 + }, + { + "epoch": 1.3321369409166206, + "grad_norm": 0.8950392603874207, + "learning_rate": 4.469184054723632e-06, + "loss": 0.6551, + "step": 1609 + }, + { + "epoch": 1.3329652125897296, + "grad_norm": 0.9604961276054382, + "learning_rate": 4.468506353584773e-06, + "loss": 0.643, + "step": 1610 + }, + { + "epoch": 1.3337934842628383, + "grad_norm": 0.933073103427887, + "learning_rate": 4.467828271558286e-06, + "loss": 0.6193, + "step": 1611 + }, + { + "epoch": 1.334621755935947, + "grad_norm": 0.890964686870575, + "learning_rate": 4.467149808775371e-06, + "loss": 0.6423, + "step": 1612 + }, + { + "epoch": 1.3354500276090557, + "grad_norm": 0.8916820883750916, + "learning_rate": 4.466470965367306e-06, + "loss": 0.6399, + "step": 1613 + }, + { + "epoch": 1.3362782992821645, + "grad_norm": 0.9151459336280823, + "learning_rate": 4.465791741465439e-06, + "loss": 0.6439, + "step": 1614 + }, + { + "epoch": 1.3371065709552734, + "grad_norm": 0.9155884385108948, + "learning_rate": 4.465112137201197e-06, + "loss": 0.6516, + "step": 1615 + }, + { + "epoch": 1.3379348426283821, + "grad_norm": 0.9187855124473572, + "learning_rate": 4.464432152706073e-06, + "loss": 0.6263, + "step": 1616 + }, + { + "epoch": 1.3387631143014909, + "grad_norm": 0.9506946206092834, + "learning_rate": 4.46375178811164e-06, + "loss": 0.6586, + "step": 1617 + }, + { + "epoch": 1.3395913859745996, + "grad_norm": 0.9056413769721985, + "learning_rate": 4.463071043549543e-06, + "loss": 0.6301, + "step": 1618 + }, + { + "epoch": 1.3404196576477085, + "grad_norm": 0.9056457877159119, + "learning_rate": 4.462389919151498e-06, + "loss": 0.6352, + "step": 1619 + }, + { + "epoch": 1.3412479293208173, + "grad_norm": 0.9204022288322449, + "learning_rate": 4.461708415049297e-06, + "loss": 0.6446, + "step": 1620 + }, + { + "epoch": 1.342076200993926, + "grad_norm": 0.9411336183547974, + "learning_rate": 4.461026531374804e-06, + "loss": 0.6594, + "step": 1621 + }, + { + "epoch": 1.3429044726670347, + "grad_norm": 0.8956605792045593, + "learning_rate": 4.460344268259958e-06, + "loss": 0.6492, + "step": 1622 + }, + { + "epoch": 1.3437327443401434, + "grad_norm": 0.9010135531425476, + "learning_rate": 4.45966162583677e-06, + "loss": 0.6233, + "step": 1623 + }, + { + "epoch": 1.3445610160132524, + "grad_norm": 0.8948563933372498, + "learning_rate": 4.458978604237325e-06, + "loss": 0.6447, + "step": 1624 + }, + { + "epoch": 1.3453892876863611, + "grad_norm": 0.9227688312530518, + "learning_rate": 4.458295203593782e-06, + "loss": 0.6459, + "step": 1625 + }, + { + "epoch": 1.3462175593594699, + "grad_norm": 0.9532939195632935, + "learning_rate": 4.4576114240383725e-06, + "loss": 0.6724, + "step": 1626 + }, + { + "epoch": 1.3470458310325788, + "grad_norm": 0.9089633226394653, + "learning_rate": 4.4569272657034005e-06, + "loss": 0.6303, + "step": 1627 + }, + { + "epoch": 1.3478741027056875, + "grad_norm": 0.8903496861457825, + "learning_rate": 4.456242728721244e-06, + "loss": 0.6655, + "step": 1628 + }, + { + "epoch": 1.3487023743787963, + "grad_norm": 0.9009859561920166, + "learning_rate": 4.455557813224356e-06, + "loss": 0.6268, + "step": 1629 + }, + { + "epoch": 1.349530646051905, + "grad_norm": 0.9265439510345459, + "learning_rate": 4.454872519345261e-06, + "loss": 0.6586, + "step": 1630 + }, + { + "epoch": 1.3503589177250137, + "grad_norm": 0.8945439457893372, + "learning_rate": 4.454186847216556e-06, + "loss": 0.62, + "step": 1631 + }, + { + "epoch": 1.3511871893981227, + "grad_norm": 0.9313004016876221, + "learning_rate": 4.453500796970913e-06, + "loss": 0.6452, + "step": 1632 + }, + { + "epoch": 1.3520154610712314, + "grad_norm": 0.9005258083343506, + "learning_rate": 4.452814368741076e-06, + "loss": 0.6436, + "step": 1633 + }, + { + "epoch": 1.35284373274434, + "grad_norm": 0.8860014081001282, + "learning_rate": 4.452127562659864e-06, + "loss": 0.6423, + "step": 1634 + }, + { + "epoch": 1.353672004417449, + "grad_norm": 0.9049930572509766, + "learning_rate": 4.4514403788601645e-06, + "loss": 0.6343, + "step": 1635 + }, + { + "epoch": 1.3545002760905578, + "grad_norm": 0.8849244117736816, + "learning_rate": 4.4507528174749425e-06, + "loss": 0.6481, + "step": 1636 + }, + { + "epoch": 1.3553285477636665, + "grad_norm": 0.9214544892311096, + "learning_rate": 4.450064878637235e-06, + "loss": 0.6166, + "step": 1637 + }, + { + "epoch": 1.3561568194367752, + "grad_norm": 0.8948760628700256, + "learning_rate": 4.449376562480153e-06, + "loss": 0.6195, + "step": 1638 + }, + { + "epoch": 1.356985091109884, + "grad_norm": 0.9000245928764343, + "learning_rate": 4.448687869136878e-06, + "loss": 0.6378, + "step": 1639 + }, + { + "epoch": 1.3578133627829927, + "grad_norm": 0.8963801264762878, + "learning_rate": 4.447998798740665e-06, + "loss": 0.6118, + "step": 1640 + }, + { + "epoch": 1.3586416344561016, + "grad_norm": 0.8774341344833374, + "learning_rate": 4.447309351424843e-06, + "loss": 0.6184, + "step": 1641 + }, + { + "epoch": 1.3594699061292104, + "grad_norm": 0.8901870846748352, + "learning_rate": 4.446619527322815e-06, + "loss": 0.6499, + "step": 1642 + }, + { + "epoch": 1.360298177802319, + "grad_norm": 0.9108647108078003, + "learning_rate": 4.445929326568055e-06, + "loss": 0.6243, + "step": 1643 + }, + { + "epoch": 1.361126449475428, + "grad_norm": 0.9178787469863892, + "learning_rate": 4.445238749294109e-06, + "loss": 0.6458, + "step": 1644 + }, + { + "epoch": 1.3619547211485368, + "grad_norm": 0.9278320670127869, + "learning_rate": 4.444547795634598e-06, + "loss": 0.6281, + "step": 1645 + }, + { + "epoch": 1.3627829928216455, + "grad_norm": 0.8769298195838928, + "learning_rate": 4.443856465723216e-06, + "loss": 0.6387, + "step": 1646 + }, + { + "epoch": 1.3636112644947542, + "grad_norm": 0.9313247203826904, + "learning_rate": 4.443164759693729e-06, + "loss": 0.6304, + "step": 1647 + }, + { + "epoch": 1.364439536167863, + "grad_norm": 1.0037778615951538, + "learning_rate": 4.4424726776799745e-06, + "loss": 0.6378, + "step": 1648 + }, + { + "epoch": 1.365267807840972, + "grad_norm": 0.9090721011161804, + "learning_rate": 4.441780219815863e-06, + "loss": 0.6456, + "step": 1649 + }, + { + "epoch": 1.3660960795140806, + "grad_norm": 0.8968444466590881, + "learning_rate": 4.441087386235382e-06, + "loss": 0.6377, + "step": 1650 + }, + { + "epoch": 1.3669243511871894, + "grad_norm": 0.9638763070106506, + "learning_rate": 4.440394177072586e-06, + "loss": 0.651, + "step": 1651 + }, + { + "epoch": 1.3677526228602983, + "grad_norm": 0.9224078059196472, + "learning_rate": 4.439700592461604e-06, + "loss": 0.6492, + "step": 1652 + }, + { + "epoch": 1.368580894533407, + "grad_norm": 0.891180157661438, + "learning_rate": 4.439006632536639e-06, + "loss": 0.6402, + "step": 1653 + }, + { + "epoch": 1.3694091662065158, + "grad_norm": 0.8817652463912964, + "learning_rate": 4.438312297431967e-06, + "loss": 0.6095, + "step": 1654 + }, + { + "epoch": 1.3702374378796245, + "grad_norm": 0.8814468383789062, + "learning_rate": 4.437617587281932e-06, + "loss": 0.6334, + "step": 1655 + }, + { + "epoch": 1.3710657095527332, + "grad_norm": 0.8895065784454346, + "learning_rate": 4.436922502220958e-06, + "loss": 0.6163, + "step": 1656 + }, + { + "epoch": 1.371893981225842, + "grad_norm": 0.9043355584144592, + "learning_rate": 4.436227042383535e-06, + "loss": 0.6625, + "step": 1657 + }, + { + "epoch": 1.372722252898951, + "grad_norm": 0.942014753818512, + "learning_rate": 4.435531207904229e-06, + "loss": 0.6298, + "step": 1658 + }, + { + "epoch": 1.3735505245720596, + "grad_norm": 0.8830228447914124, + "learning_rate": 4.434834998917677e-06, + "loss": 0.6215, + "step": 1659 + }, + { + "epoch": 1.3743787962451683, + "grad_norm": 0.9138136506080627, + "learning_rate": 4.434138415558588e-06, + "loss": 0.63, + "step": 1660 + }, + { + "epoch": 1.3752070679182773, + "grad_norm": 0.8880951404571533, + "learning_rate": 4.433441457961746e-06, + "loss": 0.6411, + "step": 1661 + }, + { + "epoch": 1.376035339591386, + "grad_norm": 0.9435725808143616, + "learning_rate": 4.432744126262005e-06, + "loss": 0.6279, + "step": 1662 + }, + { + "epoch": 1.3768636112644947, + "grad_norm": 0.894296407699585, + "learning_rate": 4.432046420594292e-06, + "loss": 0.6298, + "step": 1663 + }, + { + "epoch": 1.3776918829376035, + "grad_norm": 0.8850218653678894, + "learning_rate": 4.431348341093608e-06, + "loss": 0.627, + "step": 1664 + }, + { + "epoch": 1.3785201546107122, + "grad_norm": 0.8956788182258606, + "learning_rate": 4.430649887895025e-06, + "loss": 0.6412, + "step": 1665 + }, + { + "epoch": 1.3793484262838211, + "grad_norm": 0.9128378033638, + "learning_rate": 4.429951061133683e-06, + "loss": 0.6208, + "step": 1666 + }, + { + "epoch": 1.3801766979569299, + "grad_norm": 0.8763598203659058, + "learning_rate": 4.429251860944803e-06, + "loss": 0.6317, + "step": 1667 + }, + { + "epoch": 1.3810049696300386, + "grad_norm": 0.9210475087165833, + "learning_rate": 4.428552287463672e-06, + "loss": 0.6321, + "step": 1668 + }, + { + "epoch": 1.3818332413031476, + "grad_norm": 0.9203745126724243, + "learning_rate": 4.427852340825651e-06, + "loss": 0.6442, + "step": 1669 + }, + { + "epoch": 1.3826615129762563, + "grad_norm": 0.9229859709739685, + "learning_rate": 4.4271520211661735e-06, + "loss": 0.6402, + "step": 1670 + }, + { + "epoch": 1.383489784649365, + "grad_norm": 0.9455426335334778, + "learning_rate": 4.426451328620744e-06, + "loss": 0.6402, + "step": 1671 + }, + { + "epoch": 1.3843180563224737, + "grad_norm": 0.8894345760345459, + "learning_rate": 4.425750263324942e-06, + "loss": 0.6429, + "step": 1672 + }, + { + "epoch": 1.3851463279955825, + "grad_norm": 0.871449887752533, + "learning_rate": 4.425048825414414e-06, + "loss": 0.6449, + "step": 1673 + }, + { + "epoch": 1.3859745996686914, + "grad_norm": 0.9102624654769897, + "learning_rate": 4.424347015024884e-06, + "loss": 0.634, + "step": 1674 + }, + { + "epoch": 1.3868028713418001, + "grad_norm": 0.881172776222229, + "learning_rate": 4.4236448322921445e-06, + "loss": 0.621, + "step": 1675 + }, + { + "epoch": 1.3876311430149089, + "grad_norm": 0.9143398404121399, + "learning_rate": 4.422942277352063e-06, + "loss": 0.6379, + "step": 1676 + }, + { + "epoch": 1.3884594146880176, + "grad_norm": 0.9155046343803406, + "learning_rate": 4.4222393503405754e-06, + "loss": 0.6168, + "step": 1677 + }, + { + "epoch": 1.3892876863611265, + "grad_norm": 0.9501062631607056, + "learning_rate": 4.421536051393694e-06, + "loss": 0.6238, + "step": 1678 + }, + { + "epoch": 1.3901159580342353, + "grad_norm": 0.9055907130241394, + "learning_rate": 4.420832380647498e-06, + "loss": 0.634, + "step": 1679 + }, + { + "epoch": 1.390944229707344, + "grad_norm": 0.9199142456054688, + "learning_rate": 4.420128338238141e-06, + "loss": 0.6283, + "step": 1680 + }, + { + "epoch": 1.3917725013804527, + "grad_norm": 0.8932124376296997, + "learning_rate": 4.419423924301851e-06, + "loss": 0.6146, + "step": 1681 + }, + { + "epoch": 1.3926007730535614, + "grad_norm": 0.9010130167007446, + "learning_rate": 4.418719138974924e-06, + "loss": 0.6409, + "step": 1682 + }, + { + "epoch": 1.3934290447266704, + "grad_norm": 0.9154212474822998, + "learning_rate": 4.41801398239373e-06, + "loss": 0.6115, + "step": 1683 + }, + { + "epoch": 1.3942573163997791, + "grad_norm": 0.9208158850669861, + "learning_rate": 4.417308454694709e-06, + "loss": 0.6242, + "step": 1684 + }, + { + "epoch": 1.3950855880728878, + "grad_norm": 0.932996928691864, + "learning_rate": 4.4166025560143756e-06, + "loss": 0.6277, + "step": 1685 + }, + { + "epoch": 1.3959138597459968, + "grad_norm": 0.8990349769592285, + "learning_rate": 4.415896286489314e-06, + "loss": 0.6128, + "step": 1686 + }, + { + "epoch": 1.3967421314191055, + "grad_norm": 0.926906168460846, + "learning_rate": 4.4151896462561805e-06, + "loss": 0.6442, + "step": 1687 + }, + { + "epoch": 1.3975704030922143, + "grad_norm": 0.9330859184265137, + "learning_rate": 4.414482635451704e-06, + "loss": 0.5995, + "step": 1688 + }, + { + "epoch": 1.398398674765323, + "grad_norm": 0.9039576649665833, + "learning_rate": 4.413775254212683e-06, + "loss": 0.6363, + "step": 1689 + }, + { + "epoch": 1.3992269464384317, + "grad_norm": 0.9228506684303284, + "learning_rate": 4.413067502675992e-06, + "loss": 0.6294, + "step": 1690 + }, + { + "epoch": 1.4000552181115407, + "grad_norm": 0.921620786190033, + "learning_rate": 4.412359380978571e-06, + "loss": 0.64, + "step": 1691 + }, + { + "epoch": 1.4008834897846494, + "grad_norm": 0.9379149079322815, + "learning_rate": 4.4116508892574375e-06, + "loss": 0.6207, + "step": 1692 + }, + { + "epoch": 1.401711761457758, + "grad_norm": 1.0528016090393066, + "learning_rate": 4.410942027649677e-06, + "loss": 0.6204, + "step": 1693 + }, + { + "epoch": 1.402540033130867, + "grad_norm": 0.9510579705238342, + "learning_rate": 4.410232796292447e-06, + "loss": 0.6314, + "step": 1694 + }, + { + "epoch": 1.4033683048039758, + "grad_norm": 0.8922870755195618, + "learning_rate": 4.409523195322979e-06, + "loss": 0.6056, + "step": 1695 + }, + { + "epoch": 1.4041965764770845, + "grad_norm": 0.9239864349365234, + "learning_rate": 4.408813224878573e-06, + "loss": 0.6404, + "step": 1696 + }, + { + "epoch": 1.4050248481501932, + "grad_norm": 0.8696756362915039, + "learning_rate": 4.408102885096601e-06, + "loss": 0.6553, + "step": 1697 + }, + { + "epoch": 1.405853119823302, + "grad_norm": 0.9176343679428101, + "learning_rate": 4.407392176114509e-06, + "loss": 0.6286, + "step": 1698 + }, + { + "epoch": 1.4066813914964107, + "grad_norm": 0.9825496077537537, + "learning_rate": 4.40668109806981e-06, + "loss": 0.6421, + "step": 1699 + }, + { + "epoch": 1.4075096631695196, + "grad_norm": 0.9334873557090759, + "learning_rate": 4.405969651100093e-06, + "loss": 0.6347, + "step": 1700 + }, + { + "epoch": 1.4083379348426284, + "grad_norm": 0.8881192803382874, + "learning_rate": 4.405257835343015e-06, + "loss": 0.6377, + "step": 1701 + }, + { + "epoch": 1.409166206515737, + "grad_norm": 0.8896775245666504, + "learning_rate": 4.404545650936307e-06, + "loss": 0.624, + "step": 1702 + }, + { + "epoch": 1.409994478188846, + "grad_norm": 0.9361454844474792, + "learning_rate": 4.40383309801777e-06, + "loss": 0.6561, + "step": 1703 + }, + { + "epoch": 1.4108227498619548, + "grad_norm": 0.9082345962524414, + "learning_rate": 4.403120176725276e-06, + "loss": 0.6415, + "step": 1704 + }, + { + "epoch": 1.4116510215350635, + "grad_norm": 0.9453962445259094, + "learning_rate": 4.402406887196767e-06, + "loss": 0.6443, + "step": 1705 + }, + { + "epoch": 1.4124792932081722, + "grad_norm": 0.9420861005783081, + "learning_rate": 4.401693229570261e-06, + "loss": 0.6544, + "step": 1706 + }, + { + "epoch": 1.413307564881281, + "grad_norm": 0.869742214679718, + "learning_rate": 4.400979203983841e-06, + "loss": 0.6072, + "step": 1707 + }, + { + "epoch": 1.41413583655439, + "grad_norm": 0.915322482585907, + "learning_rate": 4.4002648105756665e-06, + "loss": 0.6657, + "step": 1708 + }, + { + "epoch": 1.4149641082274986, + "grad_norm": 0.9029238224029541, + "learning_rate": 4.3995500494839656e-06, + "loss": 0.6157, + "step": 1709 + }, + { + "epoch": 1.4157923799006074, + "grad_norm": 0.8992076516151428, + "learning_rate": 4.398834920847037e-06, + "loss": 0.6321, + "step": 1710 + }, + { + "epoch": 1.4166206515737163, + "grad_norm": 0.9099612236022949, + "learning_rate": 4.398119424803252e-06, + "loss": 0.6123, + "step": 1711 + }, + { + "epoch": 1.417448923246825, + "grad_norm": 0.9081803560256958, + "learning_rate": 4.397403561491052e-06, + "loss": 0.6369, + "step": 1712 + }, + { + "epoch": 1.4182771949199338, + "grad_norm": 0.8873599767684937, + "learning_rate": 4.3966873310489505e-06, + "loss": 0.6503, + "step": 1713 + }, + { + "epoch": 1.4191054665930425, + "grad_norm": 0.8841976523399353, + "learning_rate": 4.395970733615531e-06, + "loss": 0.6303, + "step": 1714 + }, + { + "epoch": 1.4199337382661512, + "grad_norm": 0.8922518491744995, + "learning_rate": 4.395253769329449e-06, + "loss": 0.6105, + "step": 1715 + }, + { + "epoch": 1.42076200993926, + "grad_norm": 0.9023404717445374, + "learning_rate": 4.39453643832943e-06, + "loss": 0.6292, + "step": 1716 + }, + { + "epoch": 1.421590281612369, + "grad_norm": 0.9042630791664124, + "learning_rate": 4.39381874075427e-06, + "loss": 0.6675, + "step": 1717 + }, + { + "epoch": 1.4224185532854776, + "grad_norm": 0.8933060169219971, + "learning_rate": 4.393100676742838e-06, + "loss": 0.6125, + "step": 1718 + }, + { + "epoch": 1.4232468249585863, + "grad_norm": 0.8821312189102173, + "learning_rate": 4.392382246434073e-06, + "loss": 0.6266, + "step": 1719 + }, + { + "epoch": 1.4240750966316953, + "grad_norm": 0.904460608959198, + "learning_rate": 4.391663449966983e-06, + "loss": 0.5984, + "step": 1720 + }, + { + "epoch": 1.424903368304804, + "grad_norm": 0.8999350666999817, + "learning_rate": 4.39094428748065e-06, + "loss": 0.6653, + "step": 1721 + }, + { + "epoch": 1.4257316399779127, + "grad_norm": 0.9061364531517029, + "learning_rate": 4.390224759114224e-06, + "loss": 0.6306, + "step": 1722 + }, + { + "epoch": 1.4265599116510215, + "grad_norm": 0.8995294570922852, + "learning_rate": 4.389504865006927e-06, + "loss": 0.6656, + "step": 1723 + }, + { + "epoch": 1.4273881833241302, + "grad_norm": 0.904737114906311, + "learning_rate": 4.388784605298053e-06, + "loss": 0.6234, + "step": 1724 + }, + { + "epoch": 1.4282164549972391, + "grad_norm": 0.8827239871025085, + "learning_rate": 4.388063980126966e-06, + "loss": 0.6418, + "step": 1725 + }, + { + "epoch": 1.4290447266703479, + "grad_norm": 0.8791747689247131, + "learning_rate": 4.387342989633097e-06, + "loss": 0.6174, + "step": 1726 + }, + { + "epoch": 1.4298729983434566, + "grad_norm": 0.8831204175949097, + "learning_rate": 4.386621633955955e-06, + "loss": 0.6222, + "step": 1727 + }, + { + "epoch": 1.4307012700165656, + "grad_norm": 0.908511221408844, + "learning_rate": 4.3858999132351124e-06, + "loss": 0.6373, + "step": 1728 + }, + { + "epoch": 1.4315295416896743, + "grad_norm": 0.900497555732727, + "learning_rate": 4.385177827610218e-06, + "loss": 0.6291, + "step": 1729 + }, + { + "epoch": 1.432357813362783, + "grad_norm": 0.930695652961731, + "learning_rate": 4.384455377220986e-06, + "loss": 0.6408, + "step": 1730 + }, + { + "epoch": 1.4331860850358917, + "grad_norm": 0.9583060145378113, + "learning_rate": 4.383732562207206e-06, + "loss": 0.6452, + "step": 1731 + }, + { + "epoch": 1.4340143567090005, + "grad_norm": 0.9173291921615601, + "learning_rate": 4.383009382708734e-06, + "loss": 0.6491, + "step": 1732 + }, + { + "epoch": 1.4348426283821094, + "grad_norm": 0.9054985642433167, + "learning_rate": 4.382285838865501e-06, + "loss": 0.6179, + "step": 1733 + }, + { + "epoch": 1.4356709000552181, + "grad_norm": 0.9162827730178833, + "learning_rate": 4.381561930817504e-06, + "loss": 0.6428, + "step": 1734 + }, + { + "epoch": 1.4364991717283269, + "grad_norm": 0.9013632535934448, + "learning_rate": 4.380837658704813e-06, + "loss": 0.6385, + "step": 1735 + }, + { + "epoch": 1.4373274434014356, + "grad_norm": 0.9274044632911682, + "learning_rate": 4.3801130226675685e-06, + "loss": 0.6508, + "step": 1736 + }, + { + "epoch": 1.4381557150745445, + "grad_norm": 0.8884710669517517, + "learning_rate": 4.37938802284598e-06, + "loss": 0.6675, + "step": 1737 + }, + { + "epoch": 1.4389839867476533, + "grad_norm": 0.9156724214553833, + "learning_rate": 4.378662659380329e-06, + "loss": 0.6424, + "step": 1738 + }, + { + "epoch": 1.439812258420762, + "grad_norm": 0.9267123341560364, + "learning_rate": 4.377936932410966e-06, + "loss": 0.6333, + "step": 1739 + }, + { + "epoch": 1.4406405300938707, + "grad_norm": 0.8859686851501465, + "learning_rate": 4.377210842078311e-06, + "loss": 0.6343, + "step": 1740 + }, + { + "epoch": 1.4414688017669794, + "grad_norm": 0.92928147315979, + "learning_rate": 4.376484388522858e-06, + "loss": 0.6124, + "step": 1741 + }, + { + "epoch": 1.4422970734400884, + "grad_norm": 0.8908365964889526, + "learning_rate": 4.375757571885169e-06, + "loss": 0.631, + "step": 1742 + }, + { + "epoch": 1.4431253451131971, + "grad_norm": 0.8956666588783264, + "learning_rate": 4.375030392305875e-06, + "loss": 0.6314, + "step": 1743 + }, + { + "epoch": 1.4439536167863058, + "grad_norm": 0.8836198449134827, + "learning_rate": 4.374302849925679e-06, + "loss": 0.6179, + "step": 1744 + }, + { + "epoch": 1.4447818884594148, + "grad_norm": 0.87999027967453, + "learning_rate": 4.373574944885354e-06, + "loss": 0.6226, + "step": 1745 + }, + { + "epoch": 1.4456101601325235, + "grad_norm": 0.9118010997772217, + "learning_rate": 4.3728466773257414e-06, + "loss": 0.6319, + "step": 1746 + }, + { + "epoch": 1.4464384318056323, + "grad_norm": 0.8989576101303101, + "learning_rate": 4.3721180473877555e-06, + "loss": 0.6049, + "step": 1747 + }, + { + "epoch": 1.447266703478741, + "grad_norm": 0.8896577954292297, + "learning_rate": 4.371389055212379e-06, + "loss": 0.6242, + "step": 1748 + }, + { + "epoch": 1.4480949751518497, + "grad_norm": 0.8993588089942932, + "learning_rate": 4.370659700940665e-06, + "loss": 0.6255, + "step": 1749 + }, + { + "epoch": 1.4489232468249587, + "grad_norm": 0.896486759185791, + "learning_rate": 4.369929984713738e-06, + "loss": 0.6354, + "step": 1750 + }, + { + "epoch": 1.4497515184980674, + "grad_norm": 0.9003145694732666, + "learning_rate": 4.369199906672789e-06, + "loss": 0.6464, + "step": 1751 + }, + { + "epoch": 1.450579790171176, + "grad_norm": 0.8885106444358826, + "learning_rate": 4.368469466959083e-06, + "loss": 0.6567, + "step": 1752 + }, + { + "epoch": 1.451408061844285, + "grad_norm": 0.8914943933486938, + "learning_rate": 4.367738665713953e-06, + "loss": 0.6784, + "step": 1753 + }, + { + "epoch": 1.4522363335173938, + "grad_norm": 0.933029055595398, + "learning_rate": 4.367007503078802e-06, + "loss": 0.6495, + "step": 1754 + }, + { + "epoch": 1.4530646051905025, + "grad_norm": 0.8788006901741028, + "learning_rate": 4.366275979195102e-06, + "loss": 0.6352, + "step": 1755 + }, + { + "epoch": 1.4538928768636112, + "grad_norm": 0.8768455982208252, + "learning_rate": 4.365544094204399e-06, + "loss": 0.6329, + "step": 1756 + }, + { + "epoch": 1.45472114853672, + "grad_norm": 0.9021796584129333, + "learning_rate": 4.364811848248303e-06, + "loss": 0.6616, + "step": 1757 + }, + { + "epoch": 1.4555494202098287, + "grad_norm": 0.9197986125946045, + "learning_rate": 4.364079241468499e-06, + "loss": 0.6248, + "step": 1758 + }, + { + "epoch": 1.4563776918829376, + "grad_norm": 0.9095743894577026, + "learning_rate": 4.3633462740067375e-06, + "loss": 0.6722, + "step": 1759 + }, + { + "epoch": 1.4572059635560464, + "grad_norm": 0.9114678502082825, + "learning_rate": 4.362612946004843e-06, + "loss": 0.6278, + "step": 1760 + }, + { + "epoch": 1.458034235229155, + "grad_norm": 0.8978061676025391, + "learning_rate": 4.3618792576047054e-06, + "loss": 0.633, + "step": 1761 + }, + { + "epoch": 1.458862506902264, + "grad_norm": 0.8578571677207947, + "learning_rate": 4.361145208948287e-06, + "loss": 0.6263, + "step": 1762 + }, + { + "epoch": 1.4596907785753728, + "grad_norm": 0.9333001375198364, + "learning_rate": 4.360410800177622e-06, + "loss": 0.6383, + "step": 1763 + }, + { + "epoch": 1.4605190502484815, + "grad_norm": 0.9448907971382141, + "learning_rate": 4.359676031434808e-06, + "loss": 0.6324, + "step": 1764 + }, + { + "epoch": 1.4613473219215902, + "grad_norm": 0.8806003928184509, + "learning_rate": 4.358940902862019e-06, + "loss": 0.6365, + "step": 1765 + }, + { + "epoch": 1.462175593594699, + "grad_norm": 0.9975590109825134, + "learning_rate": 4.358205414601493e-06, + "loss": 0.6436, + "step": 1766 + }, + { + "epoch": 1.463003865267808, + "grad_norm": 1.0076019763946533, + "learning_rate": 4.357469566795542e-06, + "loss": 0.6234, + "step": 1767 + }, + { + "epoch": 1.4638321369409166, + "grad_norm": 0.9254529476165771, + "learning_rate": 4.356733359586545e-06, + "loss": 0.6244, + "step": 1768 + }, + { + "epoch": 1.4646604086140254, + "grad_norm": 0.9129828810691833, + "learning_rate": 4.35599679311695e-06, + "loss": 0.6519, + "step": 1769 + }, + { + "epoch": 1.4654886802871343, + "grad_norm": 0.9141498804092407, + "learning_rate": 4.355259867529278e-06, + "loss": 0.6343, + "step": 1770 + }, + { + "epoch": 1.466316951960243, + "grad_norm": 0.920790433883667, + "learning_rate": 4.354522582966116e-06, + "loss": 0.6505, + "step": 1771 + }, + { + "epoch": 1.4671452236333518, + "grad_norm": 0.9683033227920532, + "learning_rate": 4.353784939570121e-06, + "loss": 0.6238, + "step": 1772 + }, + { + "epoch": 1.4679734953064605, + "grad_norm": 0.882957935333252, + "learning_rate": 4.3530469374840226e-06, + "loss": 0.6358, + "step": 1773 + }, + { + "epoch": 1.4688017669795692, + "grad_norm": 0.943648099899292, + "learning_rate": 4.352308576850615e-06, + "loss": 0.6274, + "step": 1774 + }, + { + "epoch": 1.4696300386526782, + "grad_norm": 0.9609524607658386, + "learning_rate": 4.351569857812764e-06, + "loss": 0.6368, + "step": 1775 + }, + { + "epoch": 1.4704583103257869, + "grad_norm": 0.9170458912849426, + "learning_rate": 4.3508307805134055e-06, + "loss": 0.6515, + "step": 1776 + }, + { + "epoch": 1.4712865819988956, + "grad_norm": 0.8956430554389954, + "learning_rate": 4.350091345095545e-06, + "loss": 0.6394, + "step": 1777 + }, + { + "epoch": 1.4721148536720043, + "grad_norm": 0.9145321846008301, + "learning_rate": 4.349351551702256e-06, + "loss": 0.6572, + "step": 1778 + }, + { + "epoch": 1.4729431253451133, + "grad_norm": 0.8807089328765869, + "learning_rate": 4.34861140047668e-06, + "loss": 0.6391, + "step": 1779 + }, + { + "epoch": 1.473771397018222, + "grad_norm": 0.8828510642051697, + "learning_rate": 4.347870891562032e-06, + "loss": 0.6583, + "step": 1780 + }, + { + "epoch": 1.4745996686913307, + "grad_norm": 0.9017962217330933, + "learning_rate": 4.34713002510159e-06, + "loss": 0.6337, + "step": 1781 + }, + { + "epoch": 1.4754279403644395, + "grad_norm": 0.8738356232643127, + "learning_rate": 4.346388801238708e-06, + "loss": 0.637, + "step": 1782 + }, + { + "epoch": 1.4762562120375482, + "grad_norm": 0.8805137872695923, + "learning_rate": 4.345647220116805e-06, + "loss": 0.6306, + "step": 1783 + }, + { + "epoch": 1.4770844837106571, + "grad_norm": 0.923599362373352, + "learning_rate": 4.34490528187937e-06, + "loss": 0.6254, + "step": 1784 + }, + { + "epoch": 1.4779127553837659, + "grad_norm": 0.9181063175201416, + "learning_rate": 4.344162986669961e-06, + "loss": 0.6257, + "step": 1785 + }, + { + "epoch": 1.4787410270568746, + "grad_norm": 0.9060764312744141, + "learning_rate": 4.343420334632204e-06, + "loss": 0.6326, + "step": 1786 + }, + { + "epoch": 1.4795692987299836, + "grad_norm": 0.9101755023002625, + "learning_rate": 4.342677325909798e-06, + "loss": 0.6319, + "step": 1787 + }, + { + "epoch": 1.4803975704030923, + "grad_norm": 0.9417787790298462, + "learning_rate": 4.341933960646505e-06, + "loss": 0.6147, + "step": 1788 + }, + { + "epoch": 1.481225842076201, + "grad_norm": 0.9138630628585815, + "learning_rate": 4.341190238986162e-06, + "loss": 0.6256, + "step": 1789 + }, + { + "epoch": 1.4820541137493097, + "grad_norm": 0.9004842638969421, + "learning_rate": 4.340446161072671e-06, + "loss": 0.6443, + "step": 1790 + }, + { + "epoch": 1.4828823854224185, + "grad_norm": 0.9181780219078064, + "learning_rate": 4.339701727050004e-06, + "loss": 0.6408, + "step": 1791 + }, + { + "epoch": 1.4837106570955274, + "grad_norm": 0.880736768245697, + "learning_rate": 4.338956937062203e-06, + "loss": 0.6436, + "step": 1792 + }, + { + "epoch": 1.4845389287686361, + "grad_norm": 0.9168913960456848, + "learning_rate": 4.338211791253376e-06, + "loss": 0.6391, + "step": 1793 + }, + { + "epoch": 1.4853672004417449, + "grad_norm": 0.942925751209259, + "learning_rate": 4.337466289767704e-06, + "loss": 0.6303, + "step": 1794 + }, + { + "epoch": 1.4861954721148538, + "grad_norm": 0.8768817186355591, + "learning_rate": 4.336720432749433e-06, + "loss": 0.6239, + "step": 1795 + }, + { + "epoch": 1.4870237437879625, + "grad_norm": 0.9388988614082336, + "learning_rate": 4.33597422034288e-06, + "loss": 0.6441, + "step": 1796 + }, + { + "epoch": 1.4878520154610713, + "grad_norm": 0.9257978200912476, + "learning_rate": 4.3352276526924295e-06, + "loss": 0.6373, + "step": 1797 + }, + { + "epoch": 1.48868028713418, + "grad_norm": 0.9225389957427979, + "learning_rate": 4.334480729942537e-06, + "loss": 0.6555, + "step": 1798 + }, + { + "epoch": 1.4895085588072887, + "grad_norm": 0.8992138504981995, + "learning_rate": 4.3337334522377235e-06, + "loss": 0.6375, + "step": 1799 + }, + { + "epoch": 1.4903368304803974, + "grad_norm": 0.9207738041877747, + "learning_rate": 4.332985819722582e-06, + "loss": 0.6439, + "step": 1800 + }, + { + "epoch": 1.4911651021535064, + "grad_norm": 0.8866941332817078, + "learning_rate": 4.3322378325417705e-06, + "loss": 0.6326, + "step": 1801 + }, + { + "epoch": 1.4919933738266151, + "grad_norm": 0.9069377779960632, + "learning_rate": 4.3314894908400175e-06, + "loss": 0.6193, + "step": 1802 + }, + { + "epoch": 1.4928216454997238, + "grad_norm": 0.8918569087982178, + "learning_rate": 4.3307407947621216e-06, + "loss": 0.6374, + "step": 1803 + }, + { + "epoch": 1.4936499171728328, + "grad_norm": 0.8993483781814575, + "learning_rate": 4.329991744452948e-06, + "loss": 0.6251, + "step": 1804 + }, + { + "epoch": 1.4944781888459415, + "grad_norm": 0.918445885181427, + "learning_rate": 4.32924234005743e-06, + "loss": 0.6494, + "step": 1805 + }, + { + "epoch": 1.4953064605190503, + "grad_norm": 0.9245757460594177, + "learning_rate": 4.328492581720573e-06, + "loss": 0.6172, + "step": 1806 + }, + { + "epoch": 1.496134732192159, + "grad_norm": 0.9166370034217834, + "learning_rate": 4.327742469587445e-06, + "loss": 0.622, + "step": 1807 + }, + { + "epoch": 1.4969630038652677, + "grad_norm": 0.9083957076072693, + "learning_rate": 4.326992003803187e-06, + "loss": 0.6205, + "step": 1808 + }, + { + "epoch": 1.4977912755383767, + "grad_norm": 0.8895595073699951, + "learning_rate": 4.326241184513008e-06, + "loss": 0.6245, + "step": 1809 + }, + { + "epoch": 1.4986195472114854, + "grad_norm": 0.9845629334449768, + "learning_rate": 4.325490011862184e-06, + "loss": 0.6163, + "step": 1810 + }, + { + "epoch": 1.499447818884594, + "grad_norm": 0.9501281976699829, + "learning_rate": 4.324738485996058e-06, + "loss": 0.6218, + "step": 1811 + }, + { + "epoch": 1.500276090557703, + "grad_norm": 0.9685792922973633, + "learning_rate": 4.323986607060047e-06, + "loss": 0.634, + "step": 1812 + }, + { + "epoch": 1.5011043622308118, + "grad_norm": 0.8959722518920898, + "learning_rate": 4.32323437519963e-06, + "loss": 0.6316, + "step": 1813 + }, + { + "epoch": 1.5019326339039205, + "grad_norm": 0.914463222026825, + "learning_rate": 4.322481790560356e-06, + "loss": 0.6258, + "step": 1814 + }, + { + "epoch": 1.5027609055770292, + "grad_norm": 0.9052320718765259, + "learning_rate": 4.321728853287846e-06, + "loss": 0.6345, + "step": 1815 + }, + { + "epoch": 1.503589177250138, + "grad_norm": 0.9306650161743164, + "learning_rate": 4.320975563527785e-06, + "loss": 0.5976, + "step": 1816 + }, + { + "epoch": 1.5044174489232467, + "grad_norm": 0.8930769562721252, + "learning_rate": 4.320221921425926e-06, + "loss": 0.6427, + "step": 1817 + }, + { + "epoch": 1.5052457205963556, + "grad_norm": 0.9205043911933899, + "learning_rate": 4.3194679271280935e-06, + "loss": 0.617, + "step": 1818 + }, + { + "epoch": 1.5060739922694644, + "grad_norm": 1.0371367931365967, + "learning_rate": 4.318713580780179e-06, + "loss": 0.6301, + "step": 1819 + }, + { + "epoch": 1.5069022639425733, + "grad_norm": 0.9442588686943054, + "learning_rate": 4.317958882528139e-06, + "loss": 0.6252, + "step": 1820 + }, + { + "epoch": 1.507730535615682, + "grad_norm": 0.9214493036270142, + "learning_rate": 4.317203832518003e-06, + "loss": 0.6326, + "step": 1821 + }, + { + "epoch": 1.5085588072887908, + "grad_norm": 0.9270440340042114, + "learning_rate": 4.316448430895866e-06, + "loss": 0.6377, + "step": 1822 + }, + { + "epoch": 1.5093870789618995, + "grad_norm": 0.8822277188301086, + "learning_rate": 4.315692677807889e-06, + "loss": 0.6277, + "step": 1823 + }, + { + "epoch": 1.5102153506350082, + "grad_norm": 0.8926153779029846, + "learning_rate": 4.314936573400305e-06, + "loss": 0.6439, + "step": 1824 + }, + { + "epoch": 1.511043622308117, + "grad_norm": 0.9214072227478027, + "learning_rate": 4.314180117819412e-06, + "loss": 0.6519, + "step": 1825 + }, + { + "epoch": 1.5118718939812257, + "grad_norm": 0.881522536277771, + "learning_rate": 4.313423311211578e-06, + "loss": 0.6213, + "step": 1826 + }, + { + "epoch": 1.5127001656543346, + "grad_norm": 0.8960263729095459, + "learning_rate": 4.312666153723238e-06, + "loss": 0.6592, + "step": 1827 + }, + { + "epoch": 1.5135284373274434, + "grad_norm": 0.9176173210144043, + "learning_rate": 4.3119086455008956e-06, + "loss": 0.6569, + "step": 1828 + }, + { + "epoch": 1.5143567090005523, + "grad_norm": 0.9312153458595276, + "learning_rate": 4.311150786691119e-06, + "loss": 0.6324, + "step": 1829 + }, + { + "epoch": 1.515184980673661, + "grad_norm": 0.9202299118041992, + "learning_rate": 4.310392577440549e-06, + "loss": 0.6381, + "step": 1830 + }, + { + "epoch": 1.5160132523467698, + "grad_norm": 0.8895224928855896, + "learning_rate": 4.309634017895893e-06, + "loss": 0.6491, + "step": 1831 + }, + { + "epoch": 1.5168415240198785, + "grad_norm": 0.9424769878387451, + "learning_rate": 4.308875108203922e-06, + "loss": 0.6342, + "step": 1832 + }, + { + "epoch": 1.5176697956929872, + "grad_norm": 0.8676947951316833, + "learning_rate": 4.30811584851148e-06, + "loss": 0.6287, + "step": 1833 + }, + { + "epoch": 1.518498067366096, + "grad_norm": 0.8764673471450806, + "learning_rate": 4.3073562389654764e-06, + "loss": 0.6123, + "step": 1834 + }, + { + "epoch": 1.5193263390392049, + "grad_norm": 0.9222663640975952, + "learning_rate": 4.306596279712889e-06, + "loss": 0.6433, + "step": 1835 + }, + { + "epoch": 1.5201546107123136, + "grad_norm": 0.8839907050132751, + "learning_rate": 4.305835970900762e-06, + "loss": 0.6416, + "step": 1836 + }, + { + "epoch": 1.5209828823854226, + "grad_norm": 0.9095149636268616, + "learning_rate": 4.305075312676209e-06, + "loss": 0.6417, + "step": 1837 + }, + { + "epoch": 1.5218111540585313, + "grad_norm": 0.8718581199645996, + "learning_rate": 4.304314305186409e-06, + "loss": 0.6174, + "step": 1838 + }, + { + "epoch": 1.52263942573164, + "grad_norm": 0.9424081444740295, + "learning_rate": 4.303552948578611e-06, + "loss": 0.6064, + "step": 1839 + }, + { + "epoch": 1.5234676974047487, + "grad_norm": 0.9170297384262085, + "learning_rate": 4.30279124300013e-06, + "loss": 0.6078, + "step": 1840 + }, + { + "epoch": 1.5242959690778575, + "grad_norm": 0.8814254999160767, + "learning_rate": 4.302029188598348e-06, + "loss": 0.6194, + "step": 1841 + }, + { + "epoch": 1.5251242407509662, + "grad_norm": 0.9248694181442261, + "learning_rate": 4.301266785520718e-06, + "loss": 0.6452, + "step": 1842 + }, + { + "epoch": 1.5259525124240751, + "grad_norm": 0.8757627010345459, + "learning_rate": 4.300504033914754e-06, + "loss": 0.6486, + "step": 1843 + }, + { + "epoch": 1.5267807840971839, + "grad_norm": 0.9229605793952942, + "learning_rate": 4.299740933928045e-06, + "loss": 0.6363, + "step": 1844 + }, + { + "epoch": 1.5276090557702926, + "grad_norm": 0.9192973971366882, + "learning_rate": 4.298977485708243e-06, + "loss": 0.6383, + "step": 1845 + }, + { + "epoch": 1.5284373274434015, + "grad_norm": 0.9071361422538757, + "learning_rate": 4.2982136894030655e-06, + "loss": 0.6541, + "step": 1846 + }, + { + "epoch": 1.5292655991165103, + "grad_norm": 0.9067169427871704, + "learning_rate": 4.297449545160303e-06, + "loss": 0.6317, + "step": 1847 + }, + { + "epoch": 1.530093870789619, + "grad_norm": 0.9047722816467285, + "learning_rate": 4.2966850531278086e-06, + "loss": 0.6365, + "step": 1848 + }, + { + "epoch": 1.5309221424627277, + "grad_norm": 0.8948541879653931, + "learning_rate": 4.295920213453504e-06, + "loss": 0.6268, + "step": 1849 + }, + { + "epoch": 1.5317504141358365, + "grad_norm": 0.9061904549598694, + "learning_rate": 4.295155026285382e-06, + "loss": 0.6271, + "step": 1850 + }, + { + "epoch": 1.5325786858089452, + "grad_norm": 0.9092041850090027, + "learning_rate": 4.2943894917714946e-06, + "loss": 0.6148, + "step": 1851 + }, + { + "epoch": 1.5334069574820541, + "grad_norm": 0.932331919670105, + "learning_rate": 4.293623610059968e-06, + "loss": 0.647, + "step": 1852 + }, + { + "epoch": 1.5342352291551629, + "grad_norm": 0.9136486649513245, + "learning_rate": 4.292857381298993e-06, + "loss": 0.6326, + "step": 1853 + }, + { + "epoch": 1.5350635008282718, + "grad_norm": 0.9760140180587769, + "learning_rate": 4.292090805636827e-06, + "loss": 0.6324, + "step": 1854 + }, + { + "epoch": 1.5358917725013805, + "grad_norm": 0.9075965881347656, + "learning_rate": 4.291323883221797e-06, + "loss": 0.6194, + "step": 1855 + }, + { + "epoch": 1.5367200441744893, + "grad_norm": 0.8804707527160645, + "learning_rate": 4.290556614202292e-06, + "loss": 0.6246, + "step": 1856 + }, + { + "epoch": 1.537548315847598, + "grad_norm": 0.9123021960258484, + "learning_rate": 4.2897889987267755e-06, + "loss": 0.6156, + "step": 1857 + }, + { + "epoch": 1.5383765875207067, + "grad_norm": 0.9396944642066956, + "learning_rate": 4.289021036943771e-06, + "loss": 0.6268, + "step": 1858 + }, + { + "epoch": 1.5392048591938154, + "grad_norm": 0.858566164970398, + "learning_rate": 4.288252729001874e-06, + "loss": 0.6237, + "step": 1859 + }, + { + "epoch": 1.5400331308669244, + "grad_norm": 0.899341344833374, + "learning_rate": 4.287484075049743e-06, + "loss": 0.5973, + "step": 1860 + }, + { + "epoch": 1.5408614025400331, + "grad_norm": 0.9021979570388794, + "learning_rate": 4.286715075236106e-06, + "loss": 0.6164, + "step": 1861 + }, + { + "epoch": 1.541689674213142, + "grad_norm": 0.9693799018859863, + "learning_rate": 4.285945729709757e-06, + "loss": 0.6409, + "step": 1862 + }, + { + "epoch": 1.5425179458862508, + "grad_norm": 0.9177312254905701, + "learning_rate": 4.285176038619559e-06, + "loss": 0.6276, + "step": 1863 + }, + { + "epoch": 1.5433462175593595, + "grad_norm": 0.9382041692733765, + "learning_rate": 4.284406002114438e-06, + "loss": 0.643, + "step": 1864 + }, + { + "epoch": 1.5441744892324683, + "grad_norm": 0.8778985142707825, + "learning_rate": 4.28363562034339e-06, + "loss": 0.6119, + "step": 1865 + }, + { + "epoch": 1.545002760905577, + "grad_norm": 0.8890271782875061, + "learning_rate": 4.282864893455476e-06, + "loss": 0.6388, + "step": 1866 + }, + { + "epoch": 1.5458310325786857, + "grad_norm": 0.9081838130950928, + "learning_rate": 4.282093821599824e-06, + "loss": 0.6464, + "step": 1867 + }, + { + "epoch": 1.5466593042517944, + "grad_norm": 0.8805630803108215, + "learning_rate": 4.281322404925631e-06, + "loss": 0.6533, + "step": 1868 + }, + { + "epoch": 1.5474875759249034, + "grad_norm": 0.8817034363746643, + "learning_rate": 4.280550643582158e-06, + "loss": 0.643, + "step": 1869 + }, + { + "epoch": 1.548315847598012, + "grad_norm": 0.9003176689147949, + "learning_rate": 4.279778537718733e-06, + "loss": 0.6322, + "step": 1870 + }, + { + "epoch": 1.549144119271121, + "grad_norm": 0.8982899785041809, + "learning_rate": 4.2790060874847525e-06, + "loss": 0.6456, + "step": 1871 + }, + { + "epoch": 1.5499723909442298, + "grad_norm": 0.8730716109275818, + "learning_rate": 4.278233293029677e-06, + "loss": 0.6428, + "step": 1872 + }, + { + "epoch": 1.5508006626173385, + "grad_norm": 0.9135062098503113, + "learning_rate": 4.277460154503036e-06, + "loss": 0.645, + "step": 1873 + }, + { + "epoch": 1.5516289342904472, + "grad_norm": 0.9091652631759644, + "learning_rate": 4.276686672054424e-06, + "loss": 0.6256, + "step": 1874 + }, + { + "epoch": 1.552457205963556, + "grad_norm": 0.9090970754623413, + "learning_rate": 4.275912845833504e-06, + "loss": 0.6404, + "step": 1875 + }, + { + "epoch": 1.5532854776366647, + "grad_norm": 0.91478031873703, + "learning_rate": 4.275138675990003e-06, + "loss": 0.6498, + "step": 1876 + }, + { + "epoch": 1.5541137493097736, + "grad_norm": 0.8979254961013794, + "learning_rate": 4.274364162673715e-06, + "loss": 0.6275, + "step": 1877 + }, + { + "epoch": 1.5549420209828824, + "grad_norm": 0.92198646068573, + "learning_rate": 4.273589306034503e-06, + "loss": 0.6585, + "step": 1878 + }, + { + "epoch": 1.5557702926559913, + "grad_norm": 0.9540958404541016, + "learning_rate": 4.272814106222293e-06, + "loss": 0.6111, + "step": 1879 + }, + { + "epoch": 1.5565985643291, + "grad_norm": 0.8989667892456055, + "learning_rate": 4.272038563387079e-06, + "loss": 0.6255, + "step": 1880 + }, + { + "epoch": 1.5574268360022088, + "grad_norm": 0.9551357626914978, + "learning_rate": 4.2712626776789224e-06, + "loss": 0.6164, + "step": 1881 + }, + { + "epoch": 1.5582551076753175, + "grad_norm": 0.9278994798660278, + "learning_rate": 4.270486449247949e-06, + "loss": 0.6376, + "step": 1882 + }, + { + "epoch": 1.5590833793484262, + "grad_norm": 0.927180826663971, + "learning_rate": 4.269709878244352e-06, + "loss": 0.6602, + "step": 1883 + }, + { + "epoch": 1.559911651021535, + "grad_norm": 0.932837724685669, + "learning_rate": 4.268932964818389e-06, + "loss": 0.6229, + "step": 1884 + }, + { + "epoch": 1.5607399226946437, + "grad_norm": 0.895219087600708, + "learning_rate": 4.26815570912039e-06, + "loss": 0.6192, + "step": 1885 + }, + { + "epoch": 1.5615681943677526, + "grad_norm": 0.9075414538383484, + "learning_rate": 4.2673781113007415e-06, + "loss": 0.6418, + "step": 1886 + }, + { + "epoch": 1.5623964660408614, + "grad_norm": 0.9130772948265076, + "learning_rate": 4.266600171509905e-06, + "loss": 0.6402, + "step": 1887 + }, + { + "epoch": 1.5632247377139703, + "grad_norm": 0.8892185688018799, + "learning_rate": 4.265821889898402e-06, + "loss": 0.6213, + "step": 1888 + }, + { + "epoch": 1.564053009387079, + "grad_norm": 0.8853991031646729, + "learning_rate": 4.265043266616825e-06, + "loss": 0.6357, + "step": 1889 + }, + { + "epoch": 1.5648812810601878, + "grad_norm": 0.9273191690444946, + "learning_rate": 4.26426430181583e-06, + "loss": 0.6311, + "step": 1890 + }, + { + "epoch": 1.5657095527332965, + "grad_norm": 0.8890771865844727, + "learning_rate": 4.263484995646138e-06, + "loss": 0.6053, + "step": 1891 + }, + { + "epoch": 1.5665378244064052, + "grad_norm": 0.9022220373153687, + "learning_rate": 4.262705348258538e-06, + "loss": 0.6256, + "step": 1892 + }, + { + "epoch": 1.567366096079514, + "grad_norm": 0.9201255440711975, + "learning_rate": 4.261925359803886e-06, + "loss": 0.6302, + "step": 1893 + }, + { + "epoch": 1.5681943677526229, + "grad_norm": 0.9675379991531372, + "learning_rate": 4.261145030433101e-06, + "loss": 0.602, + "step": 1894 + }, + { + "epoch": 1.5690226394257316, + "grad_norm": 0.9039328098297119, + "learning_rate": 4.26036436029717e-06, + "loss": 0.6477, + "step": 1895 + }, + { + "epoch": 1.5698509110988406, + "grad_norm": 0.911361813545227, + "learning_rate": 4.259583349547144e-06, + "loss": 0.6211, + "step": 1896 + }, + { + "epoch": 1.5706791827719493, + "grad_norm": 0.8901238441467285, + "learning_rate": 4.258801998334143e-06, + "loss": 0.6442, + "step": 1897 + }, + { + "epoch": 1.571507454445058, + "grad_norm": 0.9290407299995422, + "learning_rate": 4.258020306809351e-06, + "loss": 0.6382, + "step": 1898 + }, + { + "epoch": 1.5723357261181667, + "grad_norm": 0.9179307222366333, + "learning_rate": 4.257238275124017e-06, + "loss": 0.6513, + "step": 1899 + }, + { + "epoch": 1.5731639977912755, + "grad_norm": 0.9212455153465271, + "learning_rate": 4.256455903429459e-06, + "loss": 0.6257, + "step": 1900 + }, + { + "epoch": 1.5739922694643842, + "grad_norm": 0.9126132130622864, + "learning_rate": 4.2556731918770565e-06, + "loss": 0.6384, + "step": 1901 + }, + { + "epoch": 1.5748205411374931, + "grad_norm": 0.8735355734825134, + "learning_rate": 4.254890140618257e-06, + "loss": 0.6031, + "step": 1902 + }, + { + "epoch": 1.5756488128106019, + "grad_norm": 0.9375724196434021, + "learning_rate": 4.254106749804575e-06, + "loss": 0.6333, + "step": 1903 + }, + { + "epoch": 1.5764770844837108, + "grad_norm": 0.9451910853385925, + "learning_rate": 4.253323019587588e-06, + "loss": 0.6186, + "step": 1904 + }, + { + "epoch": 1.5773053561568195, + "grad_norm": 0.9468234777450562, + "learning_rate": 4.25253895011894e-06, + "loss": 0.6023, + "step": 1905 + }, + { + "epoch": 1.5781336278299283, + "grad_norm": 0.8941681385040283, + "learning_rate": 4.251754541550343e-06, + "loss": 0.634, + "step": 1906 + }, + { + "epoch": 1.578961899503037, + "grad_norm": 0.9508136510848999, + "learning_rate": 4.250969794033573e-06, + "loss": 0.6346, + "step": 1907 + }, + { + "epoch": 1.5797901711761457, + "grad_norm": 0.8964453935623169, + "learning_rate": 4.2501847077204685e-06, + "loss": 0.6298, + "step": 1908 + }, + { + "epoch": 1.5806184428492545, + "grad_norm": 0.8885358572006226, + "learning_rate": 4.249399282762938e-06, + "loss": 0.6319, + "step": 1909 + }, + { + "epoch": 1.5814467145223632, + "grad_norm": 0.8929858207702637, + "learning_rate": 4.248613519312955e-06, + "loss": 0.6272, + "step": 1910 + }, + { + "epoch": 1.5822749861954721, + "grad_norm": 0.9664214849472046, + "learning_rate": 4.247827417522557e-06, + "loss": 0.6336, + "step": 1911 + }, + { + "epoch": 1.5831032578685809, + "grad_norm": 0.890846312046051, + "learning_rate": 4.247040977543845e-06, + "loss": 0.6242, + "step": 1912 + }, + { + "epoch": 1.5839315295416898, + "grad_norm": 0.8973182439804077, + "learning_rate": 4.246254199528991e-06, + "loss": 0.6336, + "step": 1913 + }, + { + "epoch": 1.5847598012147985, + "grad_norm": 0.9789074659347534, + "learning_rate": 4.245467083630227e-06, + "loss": 0.6191, + "step": 1914 + }, + { + "epoch": 1.5855880728879073, + "grad_norm": 0.9195513129234314, + "learning_rate": 4.244679629999853e-06, + "loss": 0.6333, + "step": 1915 + }, + { + "epoch": 1.586416344561016, + "grad_norm": 0.950316846370697, + "learning_rate": 4.243891838790235e-06, + "loss": 0.635, + "step": 1916 + }, + { + "epoch": 1.5872446162341247, + "grad_norm": 0.8971677422523499, + "learning_rate": 4.243103710153802e-06, + "loss": 0.6352, + "step": 1917 + }, + { + "epoch": 1.5880728879072334, + "grad_norm": 0.9198834896087646, + "learning_rate": 4.24231524424305e-06, + "loss": 0.6252, + "step": 1918 + }, + { + "epoch": 1.5889011595803424, + "grad_norm": 0.9535729885101318, + "learning_rate": 4.241526441210541e-06, + "loss": 0.6177, + "step": 1919 + }, + { + "epoch": 1.5897294312534511, + "grad_norm": 0.8910471200942993, + "learning_rate": 4.240737301208899e-06, + "loss": 0.6408, + "step": 1920 + }, + { + "epoch": 1.59055770292656, + "grad_norm": 0.862991452217102, + "learning_rate": 4.2399478243908164e-06, + "loss": 0.6221, + "step": 1921 + }, + { + "epoch": 1.5913859745996688, + "grad_norm": 0.8935141563415527, + "learning_rate": 4.239158010909049e-06, + "loss": 0.608, + "step": 1922 + }, + { + "epoch": 1.5922142462727775, + "grad_norm": 0.9177968502044678, + "learning_rate": 4.23836786091642e-06, + "loss": 0.6442, + "step": 1923 + }, + { + "epoch": 1.5930425179458862, + "grad_norm": 0.9157758951187134, + "learning_rate": 4.237577374565813e-06, + "loss": 0.6419, + "step": 1924 + }, + { + "epoch": 1.593870789618995, + "grad_norm": 0.9484630227088928, + "learning_rate": 4.236786552010182e-06, + "loss": 0.6733, + "step": 1925 + }, + { + "epoch": 1.5946990612921037, + "grad_norm": 0.9401251077651978, + "learning_rate": 4.235995393402544e-06, + "loss": 0.6325, + "step": 1926 + }, + { + "epoch": 1.5955273329652124, + "grad_norm": 0.9104607105255127, + "learning_rate": 4.23520389889598e-06, + "loss": 0.6313, + "step": 1927 + }, + { + "epoch": 1.5963556046383214, + "grad_norm": 0.9173511862754822, + "learning_rate": 4.234412068643637e-06, + "loss": 0.6044, + "step": 1928 + }, + { + "epoch": 1.59718387631143, + "grad_norm": 0.9189788699150085, + "learning_rate": 4.233619902798727e-06, + "loss": 0.6182, + "step": 1929 + }, + { + "epoch": 1.598012147984539, + "grad_norm": 0.910804808139801, + "learning_rate": 4.232827401514525e-06, + "loss": 0.6435, + "step": 1930 + }, + { + "epoch": 1.5988404196576478, + "grad_norm": 0.8986726403236389, + "learning_rate": 4.232034564944375e-06, + "loss": 0.6268, + "step": 1931 + }, + { + "epoch": 1.5996686913307565, + "grad_norm": 0.8782910108566284, + "learning_rate": 4.231241393241682e-06, + "loss": 0.6381, + "step": 1932 + }, + { + "epoch": 1.6004969630038652, + "grad_norm": 0.9176427721977234, + "learning_rate": 4.230447886559919e-06, + "loss": 0.6131, + "step": 1933 + }, + { + "epoch": 1.601325234676974, + "grad_norm": 0.8755510449409485, + "learning_rate": 4.22965404505262e-06, + "loss": 0.654, + "step": 1934 + }, + { + "epoch": 1.6021535063500827, + "grad_norm": 0.8644832968711853, + "learning_rate": 4.228859868873387e-06, + "loss": 0.627, + "step": 1935 + }, + { + "epoch": 1.6029817780231916, + "grad_norm": 0.9181885719299316, + "learning_rate": 4.228065358175886e-06, + "loss": 0.6062, + "step": 1936 + }, + { + "epoch": 1.6038100496963004, + "grad_norm": 0.9515942335128784, + "learning_rate": 4.227270513113846e-06, + "loss": 0.6262, + "step": 1937 + }, + { + "epoch": 1.6046383213694093, + "grad_norm": 0.9062270522117615, + "learning_rate": 4.226475333841063e-06, + "loss": 0.6139, + "step": 1938 + }, + { + "epoch": 1.605466593042518, + "grad_norm": 0.9040980339050293, + "learning_rate": 4.225679820511398e-06, + "loss": 0.64, + "step": 1939 + }, + { + "epoch": 1.6062948647156268, + "grad_norm": 0.8910935521125793, + "learning_rate": 4.224883973278775e-06, + "loss": 0.6239, + "step": 1940 + }, + { + "epoch": 1.6071231363887355, + "grad_norm": 0.9150440692901611, + "learning_rate": 4.224087792297181e-06, + "loss": 0.6283, + "step": 1941 + }, + { + "epoch": 1.6079514080618442, + "grad_norm": 0.9175113439559937, + "learning_rate": 4.223291277720671e-06, + "loss": 0.6498, + "step": 1942 + }, + { + "epoch": 1.608779679734953, + "grad_norm": 0.8720189332962036, + "learning_rate": 4.222494429703363e-06, + "loss": 0.623, + "step": 1943 + }, + { + "epoch": 1.609607951408062, + "grad_norm": 0.8868989944458008, + "learning_rate": 4.22169724839944e-06, + "loss": 0.6357, + "step": 1944 + }, + { + "epoch": 1.6104362230811706, + "grad_norm": 0.9084488153457642, + "learning_rate": 4.22089973396315e-06, + "loss": 0.6607, + "step": 1945 + }, + { + "epoch": 1.6112644947542794, + "grad_norm": 0.9061319231987, + "learning_rate": 4.220101886548802e-06, + "loss": 0.6309, + "step": 1946 + }, + { + "epoch": 1.6120927664273883, + "grad_norm": 0.895646870136261, + "learning_rate": 4.219303706310775e-06, + "loss": 0.6276, + "step": 1947 + }, + { + "epoch": 1.612921038100497, + "grad_norm": 0.8863298892974854, + "learning_rate": 4.218505193403506e-06, + "loss": 0.6444, + "step": 1948 + }, + { + "epoch": 1.6137493097736058, + "grad_norm": 0.9005767107009888, + "learning_rate": 4.2177063479815025e-06, + "loss": 0.6356, + "step": 1949 + }, + { + "epoch": 1.6145775814467145, + "grad_norm": 0.9094089865684509, + "learning_rate": 4.216907170199334e-06, + "loss": 0.6053, + "step": 1950 + }, + { + "epoch": 1.6154058531198232, + "grad_norm": 0.8869677782058716, + "learning_rate": 4.216107660211632e-06, + "loss": 0.6308, + "step": 1951 + }, + { + "epoch": 1.616234124792932, + "grad_norm": 0.9285000562667847, + "learning_rate": 4.215307818173095e-06, + "loss": 0.6188, + "step": 1952 + }, + { + "epoch": 1.6170623964660409, + "grad_norm": 0.9039749503135681, + "learning_rate": 4.214507644238485e-06, + "loss": 0.63, + "step": 1953 + }, + { + "epoch": 1.6178906681391496, + "grad_norm": 0.9202860593795776, + "learning_rate": 4.213707138562628e-06, + "loss": 0.632, + "step": 1954 + }, + { + "epoch": 1.6187189398122586, + "grad_norm": 0.8759778141975403, + "learning_rate": 4.212906301300415e-06, + "loss": 0.6253, + "step": 1955 + }, + { + "epoch": 1.6195472114853673, + "grad_norm": 0.8918818831443787, + "learning_rate": 4.212105132606801e-06, + "loss": 0.625, + "step": 1956 + }, + { + "epoch": 1.620375483158476, + "grad_norm": 0.9184549450874329, + "learning_rate": 4.211303632636804e-06, + "loss": 0.6293, + "step": 1957 + }, + { + "epoch": 1.6212037548315847, + "grad_norm": 0.8887206315994263, + "learning_rate": 4.210501801545506e-06, + "loss": 0.6118, + "step": 1958 + }, + { + "epoch": 1.6220320265046935, + "grad_norm": 0.9010420441627502, + "learning_rate": 4.209699639488054e-06, + "loss": 0.6209, + "step": 1959 + }, + { + "epoch": 1.6228602981778022, + "grad_norm": 0.8888309001922607, + "learning_rate": 4.20889714661966e-06, + "loss": 0.6301, + "step": 1960 + }, + { + "epoch": 1.6236885698509111, + "grad_norm": 0.9068322777748108, + "learning_rate": 4.208094323095597e-06, + "loss": 0.6234, + "step": 1961 + }, + { + "epoch": 1.6245168415240199, + "grad_norm": 0.8940340876579285, + "learning_rate": 4.207291169071207e-06, + "loss": 0.6076, + "step": 1962 + }, + { + "epoch": 1.6253451131971288, + "grad_norm": 0.907363772392273, + "learning_rate": 4.20648768470189e-06, + "loss": 0.6522, + "step": 1963 + }, + { + "epoch": 1.6261733848702375, + "grad_norm": 0.9104205965995789, + "learning_rate": 4.205683870143114e-06, + "loss": 0.6177, + "step": 1964 + }, + { + "epoch": 1.6270016565433463, + "grad_norm": 0.9051296710968018, + "learning_rate": 4.204879725550409e-06, + "loss": 0.6069, + "step": 1965 + }, + { + "epoch": 1.627829928216455, + "grad_norm": 0.9034865498542786, + "learning_rate": 4.20407525107937e-06, + "loss": 0.6099, + "step": 1966 + }, + { + "epoch": 1.6286581998895637, + "grad_norm": 0.9055590033531189, + "learning_rate": 4.203270446885655e-06, + "loss": 0.6226, + "step": 1967 + }, + { + "epoch": 1.6294864715626725, + "grad_norm": 0.897927463054657, + "learning_rate": 4.202465313124986e-06, + "loss": 0.6342, + "step": 1968 + }, + { + "epoch": 1.6303147432357812, + "grad_norm": 0.8995800018310547, + "learning_rate": 4.2016598499531495e-06, + "loss": 0.6408, + "step": 1969 + }, + { + "epoch": 1.6311430149088901, + "grad_norm": 0.8727264404296875, + "learning_rate": 4.200854057525995e-06, + "loss": 0.6263, + "step": 1970 + }, + { + "epoch": 1.6319712865819989, + "grad_norm": 0.9289891719818115, + "learning_rate": 4.200047935999435e-06, + "loss": 0.6351, + "step": 1971 + }, + { + "epoch": 1.6327995582551078, + "grad_norm": 0.8936645984649658, + "learning_rate": 4.199241485529446e-06, + "loss": 0.6205, + "step": 1972 + }, + { + "epoch": 1.6336278299282165, + "grad_norm": 0.9054183959960938, + "learning_rate": 4.198434706272072e-06, + "loss": 0.6101, + "step": 1973 + }, + { + "epoch": 1.6344561016013253, + "grad_norm": 0.9181169271469116, + "learning_rate": 4.1976275983834135e-06, + "loss": 0.6437, + "step": 1974 + }, + { + "epoch": 1.635284373274434, + "grad_norm": 0.9325777888298035, + "learning_rate": 4.1968201620196404e-06, + "loss": 0.6343, + "step": 1975 + }, + { + "epoch": 1.6361126449475427, + "grad_norm": 0.878094494342804, + "learning_rate": 4.196012397336983e-06, + "loss": 0.6298, + "step": 1976 + }, + { + "epoch": 1.6369409166206514, + "grad_norm": 0.9336429238319397, + "learning_rate": 4.195204304491737e-06, + "loss": 0.6232, + "step": 1977 + }, + { + "epoch": 1.6377691882937604, + "grad_norm": 0.9159499406814575, + "learning_rate": 4.19439588364026e-06, + "loss": 0.621, + "step": 1978 + }, + { + "epoch": 1.6385974599668691, + "grad_norm": 0.9075011610984802, + "learning_rate": 4.193587134938975e-06, + "loss": 0.6375, + "step": 1979 + }, + { + "epoch": 1.639425731639978, + "grad_norm": 0.8912638425827026, + "learning_rate": 4.192778058544367e-06, + "loss": 0.6494, + "step": 1980 + }, + { + "epoch": 1.6402540033130868, + "grad_norm": 0.9125227928161621, + "learning_rate": 4.191968654612984e-06, + "loss": 0.6238, + "step": 1981 + }, + { + "epoch": 1.6410822749861955, + "grad_norm": 0.906326174736023, + "learning_rate": 4.191158923301439e-06, + "loss": 0.5904, + "step": 1982 + }, + { + "epoch": 1.6419105466593042, + "grad_norm": 0.8858581781387329, + "learning_rate": 4.190348864766407e-06, + "loss": 0.6222, + "step": 1983 + }, + { + "epoch": 1.642738818332413, + "grad_norm": 0.864389955997467, + "learning_rate": 4.189538479164627e-06, + "loss": 0.6218, + "step": 1984 + }, + { + "epoch": 1.6435670900055217, + "grad_norm": 0.8983911871910095, + "learning_rate": 4.188727766652901e-06, + "loss": 0.6511, + "step": 1985 + }, + { + "epoch": 1.6443953616786304, + "grad_norm": 0.9633772373199463, + "learning_rate": 4.187916727388094e-06, + "loss": 0.623, + "step": 1986 + }, + { + "epoch": 1.6452236333517394, + "grad_norm": 0.9191408753395081, + "learning_rate": 4.187105361527135e-06, + "loss": 0.6245, + "step": 1987 + }, + { + "epoch": 1.646051905024848, + "grad_norm": 0.907792866230011, + "learning_rate": 4.186293669227016e-06, + "loss": 0.5994, + "step": 1988 + }, + { + "epoch": 1.646880176697957, + "grad_norm": 0.906218409538269, + "learning_rate": 4.18548165064479e-06, + "loss": 0.653, + "step": 1989 + }, + { + "epoch": 1.6477084483710658, + "grad_norm": 0.9408275485038757, + "learning_rate": 4.184669305937577e-06, + "loss": 0.5978, + "step": 1990 + }, + { + "epoch": 1.6485367200441745, + "grad_norm": 0.9278435111045837, + "learning_rate": 4.183856635262558e-06, + "loss": 0.6167, + "step": 1991 + }, + { + "epoch": 1.6493649917172832, + "grad_norm": 0.91252601146698, + "learning_rate": 4.183043638776978e-06, + "loss": 0.6229, + "step": 1992 + }, + { + "epoch": 1.650193263390392, + "grad_norm": 0.9178313612937927, + "learning_rate": 4.182230316638142e-06, + "loss": 0.6318, + "step": 1993 + }, + { + "epoch": 1.6510215350635007, + "grad_norm": 0.9056826233863831, + "learning_rate": 4.181416669003421e-06, + "loss": 0.5932, + "step": 1994 + }, + { + "epoch": 1.6518498067366096, + "grad_norm": 0.9139112830162048, + "learning_rate": 4.180602696030248e-06, + "loss": 0.6407, + "step": 1995 + }, + { + "epoch": 1.6526780784097184, + "grad_norm": 0.9195094108581543, + "learning_rate": 4.179788397876121e-06, + "loss": 0.6371, + "step": 1996 + }, + { + "epoch": 1.6535063500828273, + "grad_norm": 0.9221140146255493, + "learning_rate": 4.178973774698597e-06, + "loss": 0.611, + "step": 1997 + }, + { + "epoch": 1.654334621755936, + "grad_norm": 0.9180639386177063, + "learning_rate": 4.178158826655299e-06, + "loss": 0.6138, + "step": 1998 + }, + { + "epoch": 1.6551628934290448, + "grad_norm": 0.9255500435829163, + "learning_rate": 4.177343553903912e-06, + "loss": 0.6425, + "step": 1999 + }, + { + "epoch": 1.6559911651021535, + "grad_norm": 0.9378511905670166, + "learning_rate": 4.176527956602182e-06, + "loss": 0.6357, + "step": 2000 + }, + { + "epoch": 1.6568194367752622, + "grad_norm": 0.9212007522583008, + "learning_rate": 4.175712034907922e-06, + "loss": 0.6307, + "step": 2001 + }, + { + "epoch": 1.657647708448371, + "grad_norm": 0.9185839891433716, + "learning_rate": 4.174895788979004e-06, + "loss": 0.6173, + "step": 2002 + }, + { + "epoch": 1.65847598012148, + "grad_norm": 0.9658777117729187, + "learning_rate": 4.174079218973364e-06, + "loss": 0.622, + "step": 2003 + }, + { + "epoch": 1.6593042517945886, + "grad_norm": 0.9310017824172974, + "learning_rate": 4.173262325049e-06, + "loss": 0.6407, + "step": 2004 + }, + { + "epoch": 1.6601325234676974, + "grad_norm": 0.9048784375190735, + "learning_rate": 4.172445107363974e-06, + "loss": 0.6557, + "step": 2005 + }, + { + "epoch": 1.6609607951408063, + "grad_norm": 0.8820594549179077, + "learning_rate": 4.17162756607641e-06, + "loss": 0.6077, + "step": 2006 + }, + { + "epoch": 1.661789066813915, + "grad_norm": 0.8961437344551086, + "learning_rate": 4.170809701344496e-06, + "loss": 0.6151, + "step": 2007 + }, + { + "epoch": 1.6626173384870238, + "grad_norm": 0.9773615598678589, + "learning_rate": 4.169991513326479e-06, + "loss": 0.6339, + "step": 2008 + }, + { + "epoch": 1.6634456101601325, + "grad_norm": 0.9309290647506714, + "learning_rate": 4.169173002180672e-06, + "loss": 0.6235, + "step": 2009 + }, + { + "epoch": 1.6642738818332412, + "grad_norm": 1.0016266107559204, + "learning_rate": 4.168354168065448e-06, + "loss": 0.6355, + "step": 2010 + }, + { + "epoch": 1.66510215350635, + "grad_norm": 0.8820549249649048, + "learning_rate": 4.167535011139246e-06, + "loss": 0.6258, + "step": 2011 + }, + { + "epoch": 1.6659304251794589, + "grad_norm": 0.9030908942222595, + "learning_rate": 4.166715531560564e-06, + "loss": 0.6217, + "step": 2012 + }, + { + "epoch": 1.6667586968525676, + "grad_norm": 0.983755350112915, + "learning_rate": 4.165895729487962e-06, + "loss": 0.6281, + "step": 2013 + }, + { + "epoch": 1.6675869685256766, + "grad_norm": 0.948001503944397, + "learning_rate": 4.165075605080066e-06, + "loss": 0.6242, + "step": 2014 + }, + { + "epoch": 1.6684152401987853, + "grad_norm": 0.9468982219696045, + "learning_rate": 4.164255158495563e-06, + "loss": 0.6563, + "step": 2015 + }, + { + "epoch": 1.669243511871894, + "grad_norm": 0.9012095332145691, + "learning_rate": 4.1634343898932e-06, + "loss": 0.618, + "step": 2016 + }, + { + "epoch": 1.6700717835450027, + "grad_norm": 0.9554924964904785, + "learning_rate": 4.162613299431789e-06, + "loss": 0.6306, + "step": 2017 + }, + { + "epoch": 1.6709000552181115, + "grad_norm": 0.9587515592575073, + "learning_rate": 4.161791887270203e-06, + "loss": 0.6305, + "step": 2018 + }, + { + "epoch": 1.6717283268912202, + "grad_norm": 0.8792864084243774, + "learning_rate": 4.160970153567379e-06, + "loss": 0.6181, + "step": 2019 + }, + { + "epoch": 1.6725565985643291, + "grad_norm": 0.8958392143249512, + "learning_rate": 4.160148098482313e-06, + "loss": 0.623, + "step": 2020 + }, + { + "epoch": 1.6733848702374379, + "grad_norm": 0.9501842260360718, + "learning_rate": 4.159325722174065e-06, + "loss": 0.6454, + "step": 2021 + }, + { + "epoch": 1.6742131419105468, + "grad_norm": 0.9107666015625, + "learning_rate": 4.158503024801759e-06, + "loss": 0.6237, + "step": 2022 + }, + { + "epoch": 1.6750414135836555, + "grad_norm": 0.9405720233917236, + "learning_rate": 4.157680006524577e-06, + "loss": 0.6423, + "step": 2023 + }, + { + "epoch": 1.6758696852567643, + "grad_norm": 0.9008631706237793, + "learning_rate": 4.1568566675017675e-06, + "loss": 0.6385, + "step": 2024 + }, + { + "epoch": 1.676697956929873, + "grad_norm": 0.9056677222251892, + "learning_rate": 4.156033007892637e-06, + "loss": 0.6225, + "step": 2025 + }, + { + "epoch": 1.6775262286029817, + "grad_norm": 0.899982750415802, + "learning_rate": 4.155209027856557e-06, + "loss": 0.6268, + "step": 2026 + }, + { + "epoch": 1.6783545002760905, + "grad_norm": 0.928506076335907, + "learning_rate": 4.15438472755296e-06, + "loss": 0.6343, + "step": 2027 + }, + { + "epoch": 1.6791827719491992, + "grad_norm": 0.8831804394721985, + "learning_rate": 4.15356010714134e-06, + "loss": 0.6404, + "step": 2028 + }, + { + "epoch": 1.6800110436223081, + "grad_norm": 0.8637033700942993, + "learning_rate": 4.152735166781253e-06, + "loss": 0.6305, + "step": 2029 + }, + { + "epoch": 1.6808393152954169, + "grad_norm": 0.9164902567863464, + "learning_rate": 4.151909906632319e-06, + "loss": 0.6171, + "step": 2030 + }, + { + "epoch": 1.6816675869685258, + "grad_norm": 0.9124466776847839, + "learning_rate": 4.151084326854216e-06, + "loss": 0.6446, + "step": 2031 + }, + { + "epoch": 1.6824958586416345, + "grad_norm": 0.9212199449539185, + "learning_rate": 4.150258427606687e-06, + "loss": 0.6074, + "step": 2032 + }, + { + "epoch": 1.6833241303147433, + "grad_norm": 0.9536449313163757, + "learning_rate": 4.149432209049536e-06, + "loss": 0.6149, + "step": 2033 + }, + { + "epoch": 1.684152401987852, + "grad_norm": 0.9232591390609741, + "learning_rate": 4.148605671342629e-06, + "loss": 0.6238, + "step": 2034 + }, + { + "epoch": 1.6849806736609607, + "grad_norm": 0.8844353556632996, + "learning_rate": 4.147778814645892e-06, + "loss": 0.6434, + "step": 2035 + }, + { + "epoch": 1.6858089453340694, + "grad_norm": 0.8500795960426331, + "learning_rate": 4.1469516391193155e-06, + "loss": 0.6304, + "step": 2036 + }, + { + "epoch": 1.6866372170071784, + "grad_norm": 0.8676083087921143, + "learning_rate": 4.146124144922949e-06, + "loss": 0.6158, + "step": 2037 + }, + { + "epoch": 1.6874654886802871, + "grad_norm": 0.8914940357208252, + "learning_rate": 4.1452963322169046e-06, + "loss": 0.6336, + "step": 2038 + }, + { + "epoch": 1.688293760353396, + "grad_norm": 0.9245527386665344, + "learning_rate": 4.144468201161358e-06, + "loss": 0.6385, + "step": 2039 + }, + { + "epoch": 1.6891220320265048, + "grad_norm": 0.8946509957313538, + "learning_rate": 4.143639751916544e-06, + "loss": 0.6243, + "step": 2040 + }, + { + "epoch": 1.6899503036996135, + "grad_norm": 0.882095217704773, + "learning_rate": 4.14281098464276e-06, + "loss": 0.6171, + "step": 2041 + }, + { + "epoch": 1.6907785753727222, + "grad_norm": 0.8981113433837891, + "learning_rate": 4.141981899500364e-06, + "loss": 0.6389, + "step": 2042 + }, + { + "epoch": 1.691606847045831, + "grad_norm": 0.928701639175415, + "learning_rate": 4.1411524966497775e-06, + "loss": 0.6488, + "step": 2043 + }, + { + "epoch": 1.6924351187189397, + "grad_norm": 0.8937084674835205, + "learning_rate": 4.140322776251481e-06, + "loss": 0.6415, + "step": 2044 + }, + { + "epoch": 1.6932633903920484, + "grad_norm": 0.8940730094909668, + "learning_rate": 4.1394927384660185e-06, + "loss": 0.6419, + "step": 2045 + }, + { + "epoch": 1.6940916620651574, + "grad_norm": 0.9067767262458801, + "learning_rate": 4.138662383453994e-06, + "loss": 0.608, + "step": 2046 + }, + { + "epoch": 1.694919933738266, + "grad_norm": 0.9123063087463379, + "learning_rate": 4.137831711376074e-06, + "loss": 0.633, + "step": 2047 + }, + { + "epoch": 1.695748205411375, + "grad_norm": 0.89399653673172, + "learning_rate": 4.137000722392985e-06, + "loss": 0.6338, + "step": 2048 + }, + { + "epoch": 1.6965764770844838, + "grad_norm": 0.9146568775177002, + "learning_rate": 4.136169416665517e-06, + "loss": 0.6214, + "step": 2049 + }, + { + "epoch": 1.6974047487575925, + "grad_norm": 0.9402440786361694, + "learning_rate": 4.135337794354518e-06, + "loss": 0.6266, + "step": 2050 + }, + { + "epoch": 1.6982330204307012, + "grad_norm": 0.8790574669837952, + "learning_rate": 4.134505855620902e-06, + "loss": 0.5989, + "step": 2051 + }, + { + "epoch": 1.69906129210381, + "grad_norm": 0.9421536326408386, + "learning_rate": 4.133673600625637e-06, + "loss": 0.611, + "step": 2052 + }, + { + "epoch": 1.6998895637769187, + "grad_norm": 0.9365546703338623, + "learning_rate": 4.132841029529762e-06, + "loss": 0.6201, + "step": 2053 + }, + { + "epoch": 1.7007178354500276, + "grad_norm": 0.9213638305664062, + "learning_rate": 4.132008142494367e-06, + "loss": 0.632, + "step": 2054 + }, + { + "epoch": 1.7015461071231364, + "grad_norm": 0.9204041957855225, + "learning_rate": 4.13117493968061e-06, + "loss": 0.6086, + "step": 2055 + }, + { + "epoch": 1.7023743787962453, + "grad_norm": 0.8972902297973633, + "learning_rate": 4.130341421249708e-06, + "loss": 0.656, + "step": 2056 + }, + { + "epoch": 1.703202650469354, + "grad_norm": 0.9163963794708252, + "learning_rate": 4.129507587362939e-06, + "loss": 0.623, + "step": 2057 + }, + { + "epoch": 1.7040309221424628, + "grad_norm": 0.9178527593612671, + "learning_rate": 4.128673438181642e-06, + "loss": 0.6327, + "step": 2058 + }, + { + "epoch": 1.7048591938155715, + "grad_norm": 0.8982755541801453, + "learning_rate": 4.127838973867217e-06, + "loss": 0.6165, + "step": 2059 + }, + { + "epoch": 1.7056874654886802, + "grad_norm": 0.8995897769927979, + "learning_rate": 4.127004194581125e-06, + "loss": 0.6204, + "step": 2060 + }, + { + "epoch": 1.706515737161789, + "grad_norm": 0.9148748517036438, + "learning_rate": 4.126169100484888e-06, + "loss": 0.6177, + "step": 2061 + }, + { + "epoch": 1.707344008834898, + "grad_norm": 0.9249931573867798, + "learning_rate": 4.12533369174009e-06, + "loss": 0.6113, + "step": 2062 + }, + { + "epoch": 1.7081722805080066, + "grad_norm": 0.9162423610687256, + "learning_rate": 4.124497968508372e-06, + "loss": 0.6119, + "step": 2063 + }, + { + "epoch": 1.7090005521811154, + "grad_norm": 0.8919503688812256, + "learning_rate": 4.123661930951441e-06, + "loss": 0.6194, + "step": 2064 + }, + { + "epoch": 1.7098288238542243, + "grad_norm": 0.8884986639022827, + "learning_rate": 4.122825579231063e-06, + "loss": 0.656, + "step": 2065 + }, + { + "epoch": 1.710657095527333, + "grad_norm": 0.9072293043136597, + "learning_rate": 4.121988913509062e-06, + "loss": 0.6351, + "step": 2066 + }, + { + "epoch": 1.7114853672004418, + "grad_norm": 0.9255547523498535, + "learning_rate": 4.121151933947327e-06, + "loss": 0.601, + "step": 2067 + }, + { + "epoch": 1.7123136388735505, + "grad_norm": 0.9141303300857544, + "learning_rate": 4.120314640707804e-06, + "loss": 0.6257, + "step": 2068 + }, + { + "epoch": 1.7131419105466592, + "grad_norm": 0.8791571855545044, + "learning_rate": 4.119477033952504e-06, + "loss": 0.6355, + "step": 2069 + }, + { + "epoch": 1.713970182219768, + "grad_norm": 0.8910380005836487, + "learning_rate": 4.118639113843493e-06, + "loss": 0.6405, + "step": 2070 + }, + { + "epoch": 1.7147984538928769, + "grad_norm": 0.9583408236503601, + "learning_rate": 4.117800880542904e-06, + "loss": 0.6408, + "step": 2071 + }, + { + "epoch": 1.7156267255659856, + "grad_norm": 0.9684842824935913, + "learning_rate": 4.116962334212924e-06, + "loss": 0.6155, + "step": 2072 + }, + { + "epoch": 1.7164549972390946, + "grad_norm": 0.8882633447647095, + "learning_rate": 4.116123475015806e-06, + "loss": 0.6332, + "step": 2073 + }, + { + "epoch": 1.7172832689122033, + "grad_norm": 0.8599907755851746, + "learning_rate": 4.1152843031138604e-06, + "loss": 0.6416, + "step": 2074 + }, + { + "epoch": 1.718111540585312, + "grad_norm": 0.9216133952140808, + "learning_rate": 4.114444818669461e-06, + "loss": 0.61, + "step": 2075 + }, + { + "epoch": 1.7189398122584207, + "grad_norm": 0.943729817867279, + "learning_rate": 4.113605021845037e-06, + "loss": 0.643, + "step": 2076 + }, + { + "epoch": 1.7197680839315295, + "grad_norm": 0.9158332943916321, + "learning_rate": 4.112764912803084e-06, + "loss": 0.6353, + "step": 2077 + }, + { + "epoch": 1.7205963556046382, + "grad_norm": 0.9036465287208557, + "learning_rate": 4.111924491706154e-06, + "loss": 0.6143, + "step": 2078 + }, + { + "epoch": 1.7214246272777471, + "grad_norm": 0.9264538884162903, + "learning_rate": 4.111083758716862e-06, + "loss": 0.6302, + "step": 2079 + }, + { + "epoch": 1.7222528989508559, + "grad_norm": 0.9309650659561157, + "learning_rate": 4.11024271399788e-06, + "loss": 0.6394, + "step": 2080 + }, + { + "epoch": 1.7230811706239648, + "grad_norm": 0.9261552095413208, + "learning_rate": 4.109401357711944e-06, + "loss": 0.6338, + "step": 2081 + }, + { + "epoch": 1.7239094422970735, + "grad_norm": 0.9027671217918396, + "learning_rate": 4.1085596900218476e-06, + "loss": 0.633, + "step": 2082 + }, + { + "epoch": 1.7247377139701823, + "grad_norm": 0.8899479508399963, + "learning_rate": 4.107717711090446e-06, + "loss": 0.6372, + "step": 2083 + }, + { + "epoch": 1.725565985643291, + "grad_norm": 0.9513740539550781, + "learning_rate": 4.106875421080654e-06, + "loss": 0.6057, + "step": 2084 + }, + { + "epoch": 1.7263942573163997, + "grad_norm": 0.8926359415054321, + "learning_rate": 4.106032820155447e-06, + "loss": 0.6219, + "step": 2085 + }, + { + "epoch": 1.7272225289895085, + "grad_norm": 0.8961042165756226, + "learning_rate": 4.105189908477862e-06, + "loss": 0.5926, + "step": 2086 + }, + { + "epoch": 1.7280508006626172, + "grad_norm": 0.8870651125907898, + "learning_rate": 4.104346686210992e-06, + "loss": 0.6298, + "step": 2087 + }, + { + "epoch": 1.7288790723357261, + "grad_norm": 0.9134538173675537, + "learning_rate": 4.103503153517993e-06, + "loss": 0.6544, + "step": 2088 + }, + { + "epoch": 1.7297073440088349, + "grad_norm": 0.900499165058136, + "learning_rate": 4.102659310562082e-06, + "loss": 0.6182, + "step": 2089 + }, + { + "epoch": 1.7305356156819438, + "grad_norm": 0.9146167039871216, + "learning_rate": 4.101815157506534e-06, + "loss": 0.6297, + "step": 2090 + }, + { + "epoch": 1.7313638873550525, + "grad_norm": 0.909670352935791, + "learning_rate": 4.100970694514687e-06, + "loss": 0.6272, + "step": 2091 + }, + { + "epoch": 1.7321921590281613, + "grad_norm": 1.127607822418213, + "learning_rate": 4.100125921749932e-06, + "loss": 0.6416, + "step": 2092 + }, + { + "epoch": 1.73302043070127, + "grad_norm": 0.8950872421264648, + "learning_rate": 4.099280839375729e-06, + "loss": 0.6411, + "step": 2093 + }, + { + "epoch": 1.7338487023743787, + "grad_norm": 0.9612913727760315, + "learning_rate": 4.098435447555591e-06, + "loss": 0.626, + "step": 2094 + }, + { + "epoch": 1.7346769740474874, + "grad_norm": 0.927638053894043, + "learning_rate": 4.097589746453094e-06, + "loss": 0.6197, + "step": 2095 + }, + { + "epoch": 1.7355052457205964, + "grad_norm": 0.9428266286849976, + "learning_rate": 4.096743736231875e-06, + "loss": 0.6407, + "step": 2096 + }, + { + "epoch": 1.7363335173937051, + "grad_norm": 0.8619174361228943, + "learning_rate": 4.095897417055627e-06, + "loss": 0.6386, + "step": 2097 + }, + { + "epoch": 1.737161789066814, + "grad_norm": 0.8943668007850647, + "learning_rate": 4.095050789088106e-06, + "loss": 0.6206, + "step": 2098 + }, + { + "epoch": 1.7379900607399228, + "grad_norm": 0.9190096259117126, + "learning_rate": 4.0942038524931245e-06, + "loss": 0.6395, + "step": 2099 + }, + { + "epoch": 1.7388183324130315, + "grad_norm": 0.9055671095848083, + "learning_rate": 4.09335660743456e-06, + "loss": 0.6384, + "step": 2100 + }, + { + "epoch": 1.7396466040861402, + "grad_norm": 0.9204139709472656, + "learning_rate": 4.092509054076345e-06, + "loss": 0.6438, + "step": 2101 + }, + { + "epoch": 1.740474875759249, + "grad_norm": 0.9572716355323792, + "learning_rate": 4.091661192582473e-06, + "loss": 0.6233, + "step": 2102 + }, + { + "epoch": 1.7413031474323577, + "grad_norm": 0.9479691982269287, + "learning_rate": 4.0908130231169975e-06, + "loss": 0.6284, + "step": 2103 + }, + { + "epoch": 1.7421314191054664, + "grad_norm": 0.9081270694732666, + "learning_rate": 4.089964545844032e-06, + "loss": 0.6284, + "step": 2104 + }, + { + "epoch": 1.7429596907785754, + "grad_norm": 0.9227046966552734, + "learning_rate": 4.0891157609277475e-06, + "loss": 0.6611, + "step": 2105 + }, + { + "epoch": 1.743787962451684, + "grad_norm": 0.9033496975898743, + "learning_rate": 4.088266668532377e-06, + "loss": 0.6122, + "step": 2106 + }, + { + "epoch": 1.744616234124793, + "grad_norm": 0.933539628982544, + "learning_rate": 4.087417268822211e-06, + "loss": 0.6167, + "step": 2107 + }, + { + "epoch": 1.7454445057979018, + "grad_norm": 0.9211486577987671, + "learning_rate": 4.086567561961602e-06, + "loss": 0.6364, + "step": 2108 + }, + { + "epoch": 1.7462727774710105, + "grad_norm": 0.8782284259796143, + "learning_rate": 4.08571754811496e-06, + "loss": 0.6164, + "step": 2109 + }, + { + "epoch": 1.7471010491441192, + "grad_norm": 0.8936499357223511, + "learning_rate": 4.084867227446753e-06, + "loss": 0.6155, + "step": 2110 + }, + { + "epoch": 1.747929320817228, + "grad_norm": 0.8931904435157776, + "learning_rate": 4.084016600121513e-06, + "loss": 0.6512, + "step": 2111 + }, + { + "epoch": 1.7487575924903367, + "grad_norm": 0.9070253372192383, + "learning_rate": 4.083165666303826e-06, + "loss": 0.6197, + "step": 2112 + }, + { + "epoch": 1.7495858641634456, + "grad_norm": 0.9232584238052368, + "learning_rate": 4.08231442615834e-06, + "loss": 0.6332, + "step": 2113 + }, + { + "epoch": 1.7504141358365544, + "grad_norm": 0.92015540599823, + "learning_rate": 4.081462879849765e-06, + "loss": 0.6351, + "step": 2114 + }, + { + "epoch": 1.7512424075096633, + "grad_norm": 0.9353263974189758, + "learning_rate": 4.080611027542863e-06, + "loss": 0.6479, + "step": 2115 + }, + { + "epoch": 1.752070679182772, + "grad_norm": 0.8943057060241699, + "learning_rate": 4.079758869402462e-06, + "loss": 0.6179, + "step": 2116 + }, + { + "epoch": 1.7528989508558808, + "grad_norm": 0.9062668681144714, + "learning_rate": 4.078906405593446e-06, + "loss": 0.6166, + "step": 2117 + }, + { + "epoch": 1.7537272225289895, + "grad_norm": 0.9368826150894165, + "learning_rate": 4.078053636280759e-06, + "loss": 0.6206, + "step": 2118 + }, + { + "epoch": 1.7545554942020982, + "grad_norm": 0.9012559652328491, + "learning_rate": 4.0772005616294034e-06, + "loss": 0.6445, + "step": 2119 + }, + { + "epoch": 1.755383765875207, + "grad_norm": 0.9061569571495056, + "learning_rate": 4.076347181804442e-06, + "loss": 0.6312, + "step": 2120 + }, + { + "epoch": 1.756212037548316, + "grad_norm": 0.892900288105011, + "learning_rate": 4.075493496970996e-06, + "loss": 0.64, + "step": 2121 + }, + { + "epoch": 1.7570403092214246, + "grad_norm": 0.9200263023376465, + "learning_rate": 4.074639507294243e-06, + "loss": 0.6435, + "step": 2122 + }, + { + "epoch": 1.7578685808945336, + "grad_norm": 0.921215295791626, + "learning_rate": 4.0737852129394265e-06, + "loss": 0.6135, + "step": 2123 + }, + { + "epoch": 1.7586968525676423, + "grad_norm": 0.8980262279510498, + "learning_rate": 4.07293061407184e-06, + "loss": 0.635, + "step": 2124 + }, + { + "epoch": 1.759525124240751, + "grad_norm": 0.9192932844161987, + "learning_rate": 4.072075710856843e-06, + "loss": 0.6317, + "step": 2125 + }, + { + "epoch": 1.7603533959138598, + "grad_norm": 0.9387120008468628, + "learning_rate": 4.07122050345985e-06, + "loss": 0.6324, + "step": 2126 + }, + { + "epoch": 1.7611816675869685, + "grad_norm": 0.8787287473678589, + "learning_rate": 4.070364992046338e-06, + "loss": 0.6177, + "step": 2127 + }, + { + "epoch": 1.7620099392600772, + "grad_norm": 0.8963474631309509, + "learning_rate": 4.069509176781838e-06, + "loss": 0.6158, + "step": 2128 + }, + { + "epoch": 1.762838210933186, + "grad_norm": 0.8926964402198792, + "learning_rate": 4.068653057831942e-06, + "loss": 0.6528, + "step": 2129 + }, + { + "epoch": 1.7636664826062949, + "grad_norm": 0.9290505647659302, + "learning_rate": 4.067796635362303e-06, + "loss": 0.6274, + "step": 2130 + }, + { + "epoch": 1.7644947542794036, + "grad_norm": 0.8848921060562134, + "learning_rate": 4.06693990953863e-06, + "loss": 0.6507, + "step": 2131 + }, + { + "epoch": 1.7653230259525126, + "grad_norm": 0.8977338075637817, + "learning_rate": 4.0660828805266915e-06, + "loss": 0.6437, + "step": 2132 + }, + { + "epoch": 1.7661512976256213, + "grad_norm": 0.9125422239303589, + "learning_rate": 4.065225548492314e-06, + "loss": 0.6221, + "step": 2133 + }, + { + "epoch": 1.76697956929873, + "grad_norm": 0.9186370968818665, + "learning_rate": 4.064367913601384e-06, + "loss": 0.6174, + "step": 2134 + }, + { + "epoch": 1.7678078409718387, + "grad_norm": 0.9084877967834473, + "learning_rate": 4.063509976019845e-06, + "loss": 0.6228, + "step": 2135 + }, + { + "epoch": 1.7686361126449475, + "grad_norm": 0.9180804491043091, + "learning_rate": 4.0626517359137e-06, + "loss": 0.6126, + "step": 2136 + }, + { + "epoch": 1.7694643843180562, + "grad_norm": 0.9099741578102112, + "learning_rate": 4.061793193449013e-06, + "loss": 0.6027, + "step": 2137 + }, + { + "epoch": 1.7702926559911651, + "grad_norm": 0.9060433506965637, + "learning_rate": 4.0609343487919005e-06, + "loss": 0.632, + "step": 2138 + }, + { + "epoch": 1.7711209276642739, + "grad_norm": 0.9404996037483215, + "learning_rate": 4.060075202108542e-06, + "loss": 0.6405, + "step": 2139 + }, + { + "epoch": 1.7719491993373828, + "grad_norm": 0.9031956791877747, + "learning_rate": 4.059215753565176e-06, + "loss": 0.6137, + "step": 2140 + }, + { + "epoch": 1.7727774710104915, + "grad_norm": 0.8913756012916565, + "learning_rate": 4.058356003328095e-06, + "loss": 0.6387, + "step": 2141 + }, + { + "epoch": 1.7736057426836003, + "grad_norm": 0.8986163139343262, + "learning_rate": 4.057495951563655e-06, + "loss": 0.6151, + "step": 2142 + }, + { + "epoch": 1.774434014356709, + "grad_norm": 0.9017037153244019, + "learning_rate": 4.0566355984382676e-06, + "loss": 0.6466, + "step": 2143 + }, + { + "epoch": 1.7752622860298177, + "grad_norm": 0.9211136698722839, + "learning_rate": 4.0557749441184024e-06, + "loss": 0.638, + "step": 2144 + }, + { + "epoch": 1.7760905577029265, + "grad_norm": 0.9097506999969482, + "learning_rate": 4.054913988770588e-06, + "loss": 0.6529, + "step": 2145 + }, + { + "epoch": 1.7769188293760352, + "grad_norm": 0.8985908627510071, + "learning_rate": 4.054052732561411e-06, + "loss": 0.6388, + "step": 2146 + }, + { + "epoch": 1.7777471010491441, + "grad_norm": 0.9066522121429443, + "learning_rate": 4.053191175657518e-06, + "loss": 0.6429, + "step": 2147 + }, + { + "epoch": 1.7785753727222529, + "grad_norm": 0.9080630540847778, + "learning_rate": 4.052329318225611e-06, + "loss": 0.619, + "step": 2148 + }, + { + "epoch": 1.7794036443953618, + "grad_norm": 0.8701719641685486, + "learning_rate": 4.0514671604324505e-06, + "loss": 0.6245, + "step": 2149 + }, + { + "epoch": 1.7802319160684705, + "grad_norm": 0.9059606194496155, + "learning_rate": 4.0506047024448585e-06, + "loss": 0.6124, + "step": 2150 + }, + { + "epoch": 1.7810601877415793, + "grad_norm": 0.8912079334259033, + "learning_rate": 4.04974194442971e-06, + "loss": 0.6295, + "step": 2151 + }, + { + "epoch": 1.781888459414688, + "grad_norm": 0.9063137769699097, + "learning_rate": 4.048878886553942e-06, + "loss": 0.6299, + "step": 2152 + }, + { + "epoch": 1.7827167310877967, + "grad_norm": 0.8850398659706116, + "learning_rate": 4.048015528984548e-06, + "loss": 0.6251, + "step": 2153 + }, + { + "epoch": 1.7835450027609054, + "grad_norm": 0.9059123992919922, + "learning_rate": 4.04715187188858e-06, + "loss": 0.6278, + "step": 2154 + }, + { + "epoch": 1.7843732744340144, + "grad_norm": 0.8777562379837036, + "learning_rate": 4.046287915433147e-06, + "loss": 0.6003, + "step": 2155 + }, + { + "epoch": 1.7852015461071231, + "grad_norm": 0.9048421382904053, + "learning_rate": 4.045423659785417e-06, + "loss": 0.6143, + "step": 2156 + }, + { + "epoch": 1.786029817780232, + "grad_norm": 0.9388517737388611, + "learning_rate": 4.044559105112614e-06, + "loss": 0.6314, + "step": 2157 + }, + { + "epoch": 1.7868580894533408, + "grad_norm": 0.9405019283294678, + "learning_rate": 4.043694251582023e-06, + "loss": 0.6381, + "step": 2158 + }, + { + "epoch": 1.7876863611264495, + "grad_norm": 0.8942162990570068, + "learning_rate": 4.0428290993609845e-06, + "loss": 0.6318, + "step": 2159 + }, + { + "epoch": 1.7885146327995582, + "grad_norm": 0.9065560698509216, + "learning_rate": 4.041963648616897e-06, + "loss": 0.6258, + "step": 2160 + }, + { + "epoch": 1.789342904472667, + "grad_norm": 0.8766914010047913, + "learning_rate": 4.041097899517218e-06, + "loss": 0.6292, + "step": 2161 + }, + { + "epoch": 1.7901711761457757, + "grad_norm": 0.8942935466766357, + "learning_rate": 4.040231852229462e-06, + "loss": 0.662, + "step": 2162 + }, + { + "epoch": 1.7909994478188846, + "grad_norm": 0.8892539143562317, + "learning_rate": 4.039365506921198e-06, + "loss": 0.6251, + "step": 2163 + }, + { + "epoch": 1.7918277194919934, + "grad_norm": 0.9359427690505981, + "learning_rate": 4.038498863760061e-06, + "loss": 0.6222, + "step": 2164 + }, + { + "epoch": 1.792655991165102, + "grad_norm": 0.8899942636489868, + "learning_rate": 4.037631922913733e-06, + "loss": 0.6318, + "step": 2165 + }, + { + "epoch": 1.793484262838211, + "grad_norm": 0.9233553409576416, + "learning_rate": 4.0367646845499635e-06, + "loss": 0.6188, + "step": 2166 + }, + { + "epoch": 1.7943125345113198, + "grad_norm": 0.8946877121925354, + "learning_rate": 4.035897148836553e-06, + "loss": 0.6247, + "step": 2167 + }, + { + "epoch": 1.7951408061844285, + "grad_norm": 0.9200116991996765, + "learning_rate": 4.03502931594136e-06, + "loss": 0.5981, + "step": 2168 + }, + { + "epoch": 1.7959690778575372, + "grad_norm": 0.9193991422653198, + "learning_rate": 4.034161186032306e-06, + "loss": 0.6144, + "step": 2169 + }, + { + "epoch": 1.796797349530646, + "grad_norm": 0.9637252688407898, + "learning_rate": 4.0332927592773615e-06, + "loss": 0.6094, + "step": 2170 + }, + { + "epoch": 1.7976256212037547, + "grad_norm": 0.8958486318588257, + "learning_rate": 4.032424035844564e-06, + "loss": 0.6304, + "step": 2171 + }, + { + "epoch": 1.7984538928768636, + "grad_norm": 0.9411646127700806, + "learning_rate": 4.031555015901998e-06, + "loss": 0.5984, + "step": 2172 + }, + { + "epoch": 1.7992821645499724, + "grad_norm": 0.8594862222671509, + "learning_rate": 4.030685699617815e-06, + "loss": 0.6176, + "step": 2173 + }, + { + "epoch": 1.8001104362230813, + "grad_norm": 0.8969242572784424, + "learning_rate": 4.029816087160217e-06, + "loss": 0.6294, + "step": 2174 + }, + { + "epoch": 1.80093870789619, + "grad_norm": 0.8950173854827881, + "learning_rate": 4.028946178697467e-06, + "loss": 0.6441, + "step": 2175 + }, + { + "epoch": 1.8017669795692988, + "grad_norm": 0.9084720611572266, + "learning_rate": 4.0280759743978835e-06, + "loss": 0.6245, + "step": 2176 + }, + { + "epoch": 1.8025952512424075, + "grad_norm": 0.9174644947052002, + "learning_rate": 4.027205474429845e-06, + "loss": 0.6136, + "step": 2177 + }, + { + "epoch": 1.8034235229155162, + "grad_norm": 0.9103249907493591, + "learning_rate": 4.026334678961783e-06, + "loss": 0.6341, + "step": 2178 + }, + { + "epoch": 1.804251794588625, + "grad_norm": 0.8916360139846802, + "learning_rate": 4.025463588162189e-06, + "loss": 0.6424, + "step": 2179 + }, + { + "epoch": 1.805080066261734, + "grad_norm": 0.9347325563430786, + "learning_rate": 4.024592202199611e-06, + "loss": 0.6425, + "step": 2180 + }, + { + "epoch": 1.8059083379348426, + "grad_norm": 0.9135183691978455, + "learning_rate": 4.023720521242653e-06, + "loss": 0.6265, + "step": 2181 + }, + { + "epoch": 1.8067366096079516, + "grad_norm": 0.8996950387954712, + "learning_rate": 4.022848545459979e-06, + "loss": 0.6083, + "step": 2182 + }, + { + "epoch": 1.8075648812810603, + "grad_norm": 0.9104750752449036, + "learning_rate": 4.021976275020307e-06, + "loss": 0.6087, + "step": 2183 + }, + { + "epoch": 1.808393152954169, + "grad_norm": 0.9272193908691406, + "learning_rate": 4.0211037100924135e-06, + "loss": 0.6254, + "step": 2184 + }, + { + "epoch": 1.8092214246272778, + "grad_norm": 0.8605275750160217, + "learning_rate": 4.020230850845132e-06, + "loss": 0.6244, + "step": 2185 + }, + { + "epoch": 1.8100496963003865, + "grad_norm": 0.8817747831344604, + "learning_rate": 4.019357697447352e-06, + "loss": 0.6226, + "step": 2186 + }, + { + "epoch": 1.8108779679734952, + "grad_norm": 0.8978784084320068, + "learning_rate": 4.018484250068021e-06, + "loss": 0.615, + "step": 2187 + }, + { + "epoch": 1.811706239646604, + "grad_norm": 0.8996117115020752, + "learning_rate": 4.017610508876143e-06, + "loss": 0.6248, + "step": 2188 + }, + { + "epoch": 1.8125345113197129, + "grad_norm": 0.9133120179176331, + "learning_rate": 4.016736474040778e-06, + "loss": 0.6364, + "step": 2189 + }, + { + "epoch": 1.8133627829928216, + "grad_norm": 0.903982400894165, + "learning_rate": 4.015862145731044e-06, + "loss": 0.6078, + "step": 2190 + }, + { + "epoch": 1.8141910546659306, + "grad_norm": 0.9343582987785339, + "learning_rate": 4.014987524116115e-06, + "loss": 0.6254, + "step": 2191 + }, + { + "epoch": 1.8150193263390393, + "grad_norm": 0.883305549621582, + "learning_rate": 4.014112609365224e-06, + "loss": 0.6306, + "step": 2192 + }, + { + "epoch": 1.815847598012148, + "grad_norm": 0.9044662714004517, + "learning_rate": 4.013237401647656e-06, + "loss": 0.6288, + "step": 2193 + }, + { + "epoch": 1.8166758696852567, + "grad_norm": 0.9016374349594116, + "learning_rate": 4.012361901132758e-06, + "loss": 0.6254, + "step": 2194 + }, + { + "epoch": 1.8175041413583655, + "grad_norm": 0.9198315739631653, + "learning_rate": 4.01148610798993e-06, + "loss": 0.621, + "step": 2195 + }, + { + "epoch": 1.8183324130314742, + "grad_norm": 0.8884013891220093, + "learning_rate": 4.01061002238863e-06, + "loss": 0.6195, + "step": 2196 + }, + { + "epoch": 1.8191606847045831, + "grad_norm": 0.901606023311615, + "learning_rate": 4.009733644498372e-06, + "loss": 0.646, + "step": 2197 + }, + { + "epoch": 1.8199889563776919, + "grad_norm": 0.9003238677978516, + "learning_rate": 4.008856974488728e-06, + "loss": 0.6276, + "step": 2198 + }, + { + "epoch": 1.8208172280508008, + "grad_norm": 0.8835832476615906, + "learning_rate": 4.007980012529324e-06, + "loss": 0.6232, + "step": 2199 + }, + { + "epoch": 1.8216454997239095, + "grad_norm": 0.8751473426818848, + "learning_rate": 4.007102758789846e-06, + "loss": 0.635, + "step": 2200 + }, + { + "epoch": 1.8224737713970183, + "grad_norm": 0.8990851640701294, + "learning_rate": 4.006225213440032e-06, + "loss": 0.605, + "step": 2201 + }, + { + "epoch": 1.823302043070127, + "grad_norm": 0.8766496181488037, + "learning_rate": 4.0053473766496805e-06, + "loss": 0.6396, + "step": 2202 + }, + { + "epoch": 1.8241303147432357, + "grad_norm": 0.914273202419281, + "learning_rate": 4.004469248588644e-06, + "loss": 0.6435, + "step": 2203 + }, + { + "epoch": 1.8249585864163445, + "grad_norm": 0.9063234925270081, + "learning_rate": 4.003590829426832e-06, + "loss": 0.6144, + "step": 2204 + }, + { + "epoch": 1.8257868580894532, + "grad_norm": 0.9479252696037292, + "learning_rate": 4.002712119334212e-06, + "loss": 0.6584, + "step": 2205 + }, + { + "epoch": 1.8266151297625621, + "grad_norm": 0.9253218770027161, + "learning_rate": 4.001833118480804e-06, + "loss": 0.6203, + "step": 2206 + }, + { + "epoch": 1.8274434014356709, + "grad_norm": 0.930874764919281, + "learning_rate": 4.000953827036689e-06, + "loss": 0.6254, + "step": 2207 + }, + { + "epoch": 1.8282716731087798, + "grad_norm": 0.9202263951301575, + "learning_rate": 4.000074245171999e-06, + "loss": 0.6351, + "step": 2208 + }, + { + "epoch": 1.8290999447818885, + "grad_norm": 0.8993806838989258, + "learning_rate": 3.999194373056926e-06, + "loss": 0.6439, + "step": 2209 + }, + { + "epoch": 1.8299282164549973, + "grad_norm": 0.9138032793998718, + "learning_rate": 3.998314210861718e-06, + "loss": 0.6011, + "step": 2210 + }, + { + "epoch": 1.830756488128106, + "grad_norm": 0.8995500206947327, + "learning_rate": 3.9974337587566765e-06, + "loss": 0.6285, + "step": 2211 + }, + { + "epoch": 1.8315847598012147, + "grad_norm": 0.9092652797698975, + "learning_rate": 3.996553016912161e-06, + "loss": 0.6364, + "step": 2212 + }, + { + "epoch": 1.8324130314743234, + "grad_norm": 0.9085352420806885, + "learning_rate": 3.995671985498589e-06, + "loss": 0.6245, + "step": 2213 + }, + { + "epoch": 1.8332413031474324, + "grad_norm": 0.9546533226966858, + "learning_rate": 3.994790664686429e-06, + "loss": 0.6512, + "step": 2214 + }, + { + "epoch": 1.8340695748205411, + "grad_norm": 0.8606089353561401, + "learning_rate": 3.99390905464621e-06, + "loss": 0.6135, + "step": 2215 + }, + { + "epoch": 1.83489784649365, + "grad_norm": 0.9450663328170776, + "learning_rate": 3.993027155548515e-06, + "loss": 0.6193, + "step": 2216 + }, + { + "epoch": 1.8357261181667588, + "grad_norm": 0.9496508836746216, + "learning_rate": 3.992144967563984e-06, + "loss": 0.5978, + "step": 2217 + }, + { + "epoch": 1.8365543898398675, + "grad_norm": 0.8837412595748901, + "learning_rate": 3.99126249086331e-06, + "loss": 0.6104, + "step": 2218 + }, + { + "epoch": 1.8373826615129762, + "grad_norm": 0.8774031400680542, + "learning_rate": 3.9903797256172464e-06, + "loss": 0.6302, + "step": 2219 + }, + { + "epoch": 1.838210933186085, + "grad_norm": 0.8978205919265747, + "learning_rate": 3.989496671996599e-06, + "loss": 0.6411, + "step": 2220 + }, + { + "epoch": 1.8390392048591937, + "grad_norm": 0.9087855815887451, + "learning_rate": 3.98861333017223e-06, + "loss": 0.592, + "step": 2221 + }, + { + "epoch": 1.8398674765323026, + "grad_norm": 0.9040684700012207, + "learning_rate": 3.987729700315059e-06, + "loss": 0.6121, + "step": 2222 + }, + { + "epoch": 1.8406957482054114, + "grad_norm": 0.8980532288551331, + "learning_rate": 3.986845782596058e-06, + "loss": 0.6231, + "step": 2223 + }, + { + "epoch": 1.84152401987852, + "grad_norm": 0.9104854464530945, + "learning_rate": 3.9859615771862595e-06, + "loss": 0.6228, + "step": 2224 + }, + { + "epoch": 1.842352291551629, + "grad_norm": 0.9314625263214111, + "learning_rate": 3.985077084256747e-06, + "loss": 0.6136, + "step": 2225 + }, + { + "epoch": 1.8431805632247378, + "grad_norm": 0.9033634662628174, + "learning_rate": 3.984192303978663e-06, + "loss": 0.6408, + "step": 2226 + }, + { + "epoch": 1.8440088348978465, + "grad_norm": 0.9022733569145203, + "learning_rate": 3.983307236523203e-06, + "loss": 0.5954, + "step": 2227 + }, + { + "epoch": 1.8448371065709552, + "grad_norm": 0.8900256156921387, + "learning_rate": 3.982421882061621e-06, + "loss": 0.631, + "step": 2228 + }, + { + "epoch": 1.845665378244064, + "grad_norm": 0.8882872462272644, + "learning_rate": 3.981536240765223e-06, + "loss": 0.6204, + "step": 2229 + }, + { + "epoch": 1.8464936499171727, + "grad_norm": 0.894923985004425, + "learning_rate": 3.980650312805373e-06, + "loss": 0.6013, + "step": 2230 + }, + { + "epoch": 1.8473219215902816, + "grad_norm": 0.8910638093948364, + "learning_rate": 3.979764098353491e-06, + "loss": 0.6373, + "step": 2231 + }, + { + "epoch": 1.8481501932633904, + "grad_norm": 0.8952053189277649, + "learning_rate": 3.978877597581049e-06, + "loss": 0.6374, + "step": 2232 + }, + { + "epoch": 1.8489784649364993, + "grad_norm": 0.9100720882415771, + "learning_rate": 3.977990810659579e-06, + "loss": 0.6417, + "step": 2233 + }, + { + "epoch": 1.849806736609608, + "grad_norm": 0.8954150676727295, + "learning_rate": 3.9771037377606634e-06, + "loss": 0.6027, + "step": 2234 + }, + { + "epoch": 1.8506350082827168, + "grad_norm": 0.8989905118942261, + "learning_rate": 3.976216379055944e-06, + "loss": 0.6228, + "step": 2235 + }, + { + "epoch": 1.8514632799558255, + "grad_norm": 0.9060820937156677, + "learning_rate": 3.975328734717117e-06, + "loss": 0.6291, + "step": 2236 + }, + { + "epoch": 1.8522915516289342, + "grad_norm": 0.9164367318153381, + "learning_rate": 3.974440804915932e-06, + "loss": 0.6249, + "step": 2237 + }, + { + "epoch": 1.853119823302043, + "grad_norm": 0.9227840304374695, + "learning_rate": 3.973552589824196e-06, + "loss": 0.6328, + "step": 2238 + }, + { + "epoch": 1.853948094975152, + "grad_norm": 0.8612478971481323, + "learning_rate": 3.972664089613772e-06, + "loss": 0.5794, + "step": 2239 + }, + { + "epoch": 1.8547763666482606, + "grad_norm": 0.8819253444671631, + "learning_rate": 3.971775304456573e-06, + "loss": 0.6145, + "step": 2240 + }, + { + "epoch": 1.8556046383213696, + "grad_norm": 0.9216543436050415, + "learning_rate": 3.970886234524573e-06, + "loss": 0.609, + "step": 2241 + }, + { + "epoch": 1.8564329099944783, + "grad_norm": 0.9114699959754944, + "learning_rate": 3.969996879989797e-06, + "loss": 0.6105, + "step": 2242 + }, + { + "epoch": 1.857261181667587, + "grad_norm": 0.9464704990386963, + "learning_rate": 3.969107241024328e-06, + "loss": 0.6389, + "step": 2243 + }, + { + "epoch": 1.8580894533406958, + "grad_norm": 0.8921800255775452, + "learning_rate": 3.968217317800303e-06, + "loss": 0.6338, + "step": 2244 + }, + { + "epoch": 1.8589177250138045, + "grad_norm": 0.897452175617218, + "learning_rate": 3.967327110489913e-06, + "loss": 0.6248, + "step": 2245 + }, + { + "epoch": 1.8597459966869132, + "grad_norm": 0.9093567132949829, + "learning_rate": 3.9664366192654066e-06, + "loss": 0.6258, + "step": 2246 + }, + { + "epoch": 1.860574268360022, + "grad_norm": 0.894188642501831, + "learning_rate": 3.965545844299084e-06, + "loss": 0.6095, + "step": 2247 + }, + { + "epoch": 1.8614025400331309, + "grad_norm": 0.9238452911376953, + "learning_rate": 3.964654785763301e-06, + "loss": 0.6237, + "step": 2248 + }, + { + "epoch": 1.8622308117062396, + "grad_norm": 0.8715214133262634, + "learning_rate": 3.9637634438304715e-06, + "loss": 0.6153, + "step": 2249 + }, + { + "epoch": 1.8630590833793486, + "grad_norm": 0.8790848255157471, + "learning_rate": 3.962871818673061e-06, + "loss": 0.6389, + "step": 2250 + }, + { + "epoch": 1.8638873550524573, + "grad_norm": 0.9859224557876587, + "learning_rate": 3.961979910463589e-06, + "loss": 0.6075, + "step": 2251 + }, + { + "epoch": 1.864715626725566, + "grad_norm": 0.9127727150917053, + "learning_rate": 3.9610877193746335e-06, + "loss": 0.6332, + "step": 2252 + }, + { + "epoch": 1.8655438983986747, + "grad_norm": 0.9248382449150085, + "learning_rate": 3.960195245578825e-06, + "loss": 0.622, + "step": 2253 + }, + { + "epoch": 1.8663721700717835, + "grad_norm": 0.9030913710594177, + "learning_rate": 3.959302489248848e-06, + "loss": 0.6326, + "step": 2254 + }, + { + "epoch": 1.8672004417448922, + "grad_norm": 0.8922805190086365, + "learning_rate": 3.958409450557441e-06, + "loss": 0.6099, + "step": 2255 + }, + { + "epoch": 1.8680287134180011, + "grad_norm": 0.9165307879447937, + "learning_rate": 3.957516129677402e-06, + "loss": 0.6245, + "step": 2256 + }, + { + "epoch": 1.8688569850911099, + "grad_norm": 0.9445623159408569, + "learning_rate": 3.956622526781579e-06, + "loss": 0.6317, + "step": 2257 + }, + { + "epoch": 1.8696852567642188, + "grad_norm": 0.8833973407745361, + "learning_rate": 3.955728642042874e-06, + "loss": 0.6264, + "step": 2258 + }, + { + "epoch": 1.8705135284373275, + "grad_norm": 0.9512470364570618, + "learning_rate": 3.9548344756342486e-06, + "loss": 0.6137, + "step": 2259 + }, + { + "epoch": 1.8713418001104363, + "grad_norm": 0.8834348320960999, + "learning_rate": 3.9539400277287134e-06, + "loss": 0.6597, + "step": 2260 + }, + { + "epoch": 1.872170071783545, + "grad_norm": 0.8881180882453918, + "learning_rate": 3.953045298499335e-06, + "loss": 0.6502, + "step": 2261 + }, + { + "epoch": 1.8729983434566537, + "grad_norm": 0.8921699523925781, + "learning_rate": 3.952150288119239e-06, + "loss": 0.6216, + "step": 2262 + }, + { + "epoch": 1.8738266151297625, + "grad_norm": 0.91098552942276, + "learning_rate": 3.9512549967615965e-06, + "loss": 0.6286, + "step": 2263 + }, + { + "epoch": 1.8746548868028712, + "grad_norm": 0.9343264698982239, + "learning_rate": 3.950359424599642e-06, + "loss": 0.6368, + "step": 2264 + }, + { + "epoch": 1.8754831584759801, + "grad_norm": 0.9168746471405029, + "learning_rate": 3.949463571806658e-06, + "loss": 0.6355, + "step": 2265 + }, + { + "epoch": 1.8763114301490889, + "grad_norm": 0.9206068515777588, + "learning_rate": 3.948567438555986e-06, + "loss": 0.607, + "step": 2266 + }, + { + "epoch": 1.8771397018221978, + "grad_norm": 0.9262584447860718, + "learning_rate": 3.947671025021017e-06, + "loss": 0.6197, + "step": 2267 + }, + { + "epoch": 1.8779679734953065, + "grad_norm": 0.8699168562889099, + "learning_rate": 3.946774331375199e-06, + "loss": 0.6181, + "step": 2268 + }, + { + "epoch": 1.8787962451684153, + "grad_norm": 0.8892862200737, + "learning_rate": 3.9458773577920365e-06, + "loss": 0.6245, + "step": 2269 + }, + { + "epoch": 1.879624516841524, + "grad_norm": 0.9038919806480408, + "learning_rate": 3.944980104445082e-06, + "loss": 0.6209, + "step": 2270 + }, + { + "epoch": 1.8804527885146327, + "grad_norm": 0.8780872821807861, + "learning_rate": 3.9440825715079475e-06, + "loss": 0.6355, + "step": 2271 + }, + { + "epoch": 1.8812810601877414, + "grad_norm": 0.9379895925521851, + "learning_rate": 3.943184759154297e-06, + "loss": 0.6381, + "step": 2272 + }, + { + "epoch": 1.8821093318608504, + "grad_norm": 0.9112582802772522, + "learning_rate": 3.942286667557849e-06, + "loss": 0.6273, + "step": 2273 + }, + { + "epoch": 1.8829376035339591, + "grad_norm": 0.9367797374725342, + "learning_rate": 3.9413882968923754e-06, + "loss": 0.6111, + "step": 2274 + }, + { + "epoch": 1.883765875207068, + "grad_norm": 0.8923424482345581, + "learning_rate": 3.940489647331703e-06, + "loss": 0.6187, + "step": 2275 + }, + { + "epoch": 1.8845941468801768, + "grad_norm": 0.8797968029975891, + "learning_rate": 3.939590719049712e-06, + "loss": 0.603, + "step": 2276 + }, + { + "epoch": 1.8854224185532855, + "grad_norm": 0.8977354168891907, + "learning_rate": 3.938691512220336e-06, + "loss": 0.614, + "step": 2277 + }, + { + "epoch": 1.8862506902263942, + "grad_norm": 0.9207090139389038, + "learning_rate": 3.937792027017564e-06, + "loss": 0.6162, + "step": 2278 + }, + { + "epoch": 1.887078961899503, + "grad_norm": 0.8815562725067139, + "learning_rate": 3.936892263615438e-06, + "loss": 0.6315, + "step": 2279 + }, + { + "epoch": 1.8879072335726117, + "grad_norm": 0.9983104467391968, + "learning_rate": 3.935992222188054e-06, + "loss": 0.6337, + "step": 2280 + }, + { + "epoch": 1.8887355052457206, + "grad_norm": 0.8987579941749573, + "learning_rate": 3.93509190290956e-06, + "loss": 0.644, + "step": 2281 + }, + { + "epoch": 1.8895637769188294, + "grad_norm": 0.9132242798805237, + "learning_rate": 3.934191305954161e-06, + "loss": 0.6063, + "step": 2282 + }, + { + "epoch": 1.890392048591938, + "grad_norm": 0.8940858840942383, + "learning_rate": 3.933290431496114e-06, + "loss": 0.6204, + "step": 2283 + }, + { + "epoch": 1.891220320265047, + "grad_norm": 0.8862229585647583, + "learning_rate": 3.932389279709729e-06, + "loss": 0.6257, + "step": 2284 + }, + { + "epoch": 1.8920485919381558, + "grad_norm": 0.8841557502746582, + "learning_rate": 3.931487850769371e-06, + "loss": 0.6329, + "step": 2285 + }, + { + "epoch": 1.8928768636112645, + "grad_norm": 0.9075404405593872, + "learning_rate": 3.930586144849459e-06, + "loss": 0.6195, + "step": 2286 + }, + { + "epoch": 1.8937051352843732, + "grad_norm": 0.9349889755249023, + "learning_rate": 3.929684162124462e-06, + "loss": 0.6159, + "step": 2287 + }, + { + "epoch": 1.894533406957482, + "grad_norm": 0.9378470182418823, + "learning_rate": 3.928781902768908e-06, + "loss": 0.6221, + "step": 2288 + }, + { + "epoch": 1.8953616786305907, + "grad_norm": 0.8877297043800354, + "learning_rate": 3.927879366957374e-06, + "loss": 0.6359, + "step": 2289 + }, + { + "epoch": 1.8961899503036996, + "grad_norm": 0.9335882067680359, + "learning_rate": 3.926976554864495e-06, + "loss": 0.6397, + "step": 2290 + }, + { + "epoch": 1.8970182219768084, + "grad_norm": 0.8982004523277283, + "learning_rate": 3.926073466664953e-06, + "loss": 0.6303, + "step": 2291 + }, + { + "epoch": 1.8978464936499173, + "grad_norm": 0.8909676671028137, + "learning_rate": 3.92517010253349e-06, + "loss": 0.6476, + "step": 2292 + }, + { + "epoch": 1.898674765323026, + "grad_norm": 0.9127813577651978, + "learning_rate": 3.924266462644896e-06, + "loss": 0.6078, + "step": 2293 + }, + { + "epoch": 1.8995030369961348, + "grad_norm": 0.8923580050468445, + "learning_rate": 3.923362547174019e-06, + "loss": 0.6371, + "step": 2294 + }, + { + "epoch": 1.9003313086692435, + "grad_norm": 0.912458062171936, + "learning_rate": 3.922458356295757e-06, + "loss": 0.6162, + "step": 2295 + }, + { + "epoch": 1.9011595803423522, + "grad_norm": 0.8866751790046692, + "learning_rate": 3.921553890185064e-06, + "loss": 0.6011, + "step": 2296 + }, + { + "epoch": 1.901987852015461, + "grad_norm": 0.9114490151405334, + "learning_rate": 3.920649149016944e-06, + "loss": 0.6322, + "step": 2297 + }, + { + "epoch": 1.90281612368857, + "grad_norm": 0.912030816078186, + "learning_rate": 3.919744132966457e-06, + "loss": 0.6293, + "step": 2298 + }, + { + "epoch": 1.9036443953616786, + "grad_norm": 0.887301504611969, + "learning_rate": 3.918838842208715e-06, + "loss": 0.6288, + "step": 2299 + }, + { + "epoch": 1.9044726670347876, + "grad_norm": 0.9039818048477173, + "learning_rate": 3.917933276918884e-06, + "loss": 0.6154, + "step": 2300 + }, + { + "epoch": 1.9053009387078963, + "grad_norm": 0.9260784983634949, + "learning_rate": 3.917027437272182e-06, + "loss": 0.6111, + "step": 2301 + }, + { + "epoch": 1.906129210381005, + "grad_norm": 0.8966069221496582, + "learning_rate": 3.916121323443879e-06, + "loss": 0.6334, + "step": 2302 + }, + { + "epoch": 1.9069574820541138, + "grad_norm": 0.8983698487281799, + "learning_rate": 3.9152149356093e-06, + "loss": 0.6235, + "step": 2303 + }, + { + "epoch": 1.9077857537272225, + "grad_norm": 0.9150595664978027, + "learning_rate": 3.914308273943826e-06, + "loss": 0.6118, + "step": 2304 + }, + { + "epoch": 1.9086140254003312, + "grad_norm": 0.8768419027328491, + "learning_rate": 3.913401338622883e-06, + "loss": 0.5935, + "step": 2305 + }, + { + "epoch": 1.90944229707344, + "grad_norm": 0.887562096118927, + "learning_rate": 3.912494129821958e-06, + "loss": 0.6232, + "step": 2306 + }, + { + "epoch": 1.9102705687465489, + "grad_norm": 0.8916515111923218, + "learning_rate": 3.911586647716585e-06, + "loss": 0.6409, + "step": 2307 + }, + { + "epoch": 1.9110988404196576, + "grad_norm": 0.9207882881164551, + "learning_rate": 3.910678892482354e-06, + "loss": 0.6155, + "step": 2308 + }, + { + "epoch": 1.9119271120927666, + "grad_norm": 0.9300670027732849, + "learning_rate": 3.9097708642949085e-06, + "loss": 0.6389, + "step": 2309 + }, + { + "epoch": 1.9127553837658753, + "grad_norm": 0.8685980439186096, + "learning_rate": 3.908862563329941e-06, + "loss": 0.6392, + "step": 2310 + }, + { + "epoch": 1.913583655438984, + "grad_norm": 0.9150967001914978, + "learning_rate": 3.907953989763202e-06, + "loss": 0.6387, + "step": 2311 + }, + { + "epoch": 1.9144119271120927, + "grad_norm": 0.9094190001487732, + "learning_rate": 3.90704514377049e-06, + "loss": 0.6055, + "step": 2312 + }, + { + "epoch": 1.9152401987852015, + "grad_norm": 0.8863343596458435, + "learning_rate": 3.90613602552766e-06, + "loss": 0.6073, + "step": 2313 + }, + { + "epoch": 1.9160684704583102, + "grad_norm": 0.9878994822502136, + "learning_rate": 3.905226635210616e-06, + "loss": 0.613, + "step": 2314 + }, + { + "epoch": 1.9168967421314191, + "grad_norm": 0.8863592147827148, + "learning_rate": 3.904316972995318e-06, + "loss": 0.6102, + "step": 2315 + }, + { + "epoch": 1.9177250138045279, + "grad_norm": 0.8859980702400208, + "learning_rate": 3.9034070390577765e-06, + "loss": 0.6182, + "step": 2316 + }, + { + "epoch": 1.9185532854776368, + "grad_norm": 0.8953589200973511, + "learning_rate": 3.9024968335740556e-06, + "loss": 0.6212, + "step": 2317 + }, + { + "epoch": 1.9193815571507455, + "grad_norm": 0.873466968536377, + "learning_rate": 3.90158635672027e-06, + "loss": 0.616, + "step": 2318 + }, + { + "epoch": 1.9202098288238543, + "grad_norm": 0.8589543700218201, + "learning_rate": 3.900675608672592e-06, + "loss": 0.6141, + "step": 2319 + }, + { + "epoch": 1.921038100496963, + "grad_norm": 0.8717133402824402, + "learning_rate": 3.899764589607239e-06, + "loss": 0.6281, + "step": 2320 + }, + { + "epoch": 1.9218663721700717, + "grad_norm": 0.893227219581604, + "learning_rate": 3.898853299700486e-06, + "loss": 0.6139, + "step": 2321 + }, + { + "epoch": 1.9226946438431805, + "grad_norm": 0.8736181259155273, + "learning_rate": 3.89794173912866e-06, + "loss": 0.5987, + "step": 2322 + }, + { + "epoch": 1.9235229155162892, + "grad_norm": 0.8975626826286316, + "learning_rate": 3.897029908068139e-06, + "loss": 0.6417, + "step": 2323 + }, + { + "epoch": 1.9243511871893981, + "grad_norm": 0.8916004300117493, + "learning_rate": 3.896117806695355e-06, + "loss": 0.6259, + "step": 2324 + }, + { + "epoch": 1.9251794588625069, + "grad_norm": 0.8739155530929565, + "learning_rate": 3.8952054351867885e-06, + "loss": 0.6252, + "step": 2325 + }, + { + "epoch": 1.9260077305356158, + "grad_norm": 0.8667280673980713, + "learning_rate": 3.894292793718976e-06, + "loss": 0.6083, + "step": 2326 + }, + { + "epoch": 1.9268360022087245, + "grad_norm": 0.8749732375144958, + "learning_rate": 3.893379882468506e-06, + "loss": 0.6294, + "step": 2327 + }, + { + "epoch": 1.9276642738818333, + "grad_norm": 0.9085330963134766, + "learning_rate": 3.8924667016120175e-06, + "loss": 0.6101, + "step": 2328 + }, + { + "epoch": 1.928492545554942, + "grad_norm": 0.9394292235374451, + "learning_rate": 3.891553251326203e-06, + "loss": 0.6182, + "step": 2329 + }, + { + "epoch": 1.9293208172280507, + "grad_norm": 0.8756951093673706, + "learning_rate": 3.890639531787805e-06, + "loss": 0.6159, + "step": 2330 + }, + { + "epoch": 1.9301490889011594, + "grad_norm": 0.8765501379966736, + "learning_rate": 3.8897255431736225e-06, + "loss": 0.6182, + "step": 2331 + }, + { + "epoch": 1.9309773605742684, + "grad_norm": 0.907835841178894, + "learning_rate": 3.888811285660501e-06, + "loss": 0.6336, + "step": 2332 + }, + { + "epoch": 1.9318056322473771, + "grad_norm": 1.0617926120758057, + "learning_rate": 3.887896759425342e-06, + "loss": 0.6336, + "step": 2333 + }, + { + "epoch": 1.932633903920486, + "grad_norm": 0.8745515942573547, + "learning_rate": 3.886981964645099e-06, + "loss": 0.6244, + "step": 2334 + }, + { + "epoch": 1.9334621755935948, + "grad_norm": 0.8918715119361877, + "learning_rate": 3.886066901496775e-06, + "loss": 0.6289, + "step": 2335 + }, + { + "epoch": 1.9342904472667035, + "grad_norm": 0.8779492378234863, + "learning_rate": 3.885151570157426e-06, + "loss": 0.6335, + "step": 2336 + }, + { + "epoch": 1.9351187189398122, + "grad_norm": 0.8955790400505066, + "learning_rate": 3.88423597080416e-06, + "loss": 0.6125, + "step": 2337 + }, + { + "epoch": 1.935946990612921, + "grad_norm": 0.8905174136161804, + "learning_rate": 3.883320103614137e-06, + "loss": 0.5913, + "step": 2338 + }, + { + "epoch": 1.9367752622860297, + "grad_norm": 0.9078717231750488, + "learning_rate": 3.88240396876457e-06, + "loss": 0.6356, + "step": 2339 + }, + { + "epoch": 1.9376035339591386, + "grad_norm": 0.9211397171020508, + "learning_rate": 3.881487566432722e-06, + "loss": 0.6305, + "step": 2340 + }, + { + "epoch": 1.9384318056322474, + "grad_norm": 0.9131335020065308, + "learning_rate": 3.880570896795908e-06, + "loss": 0.6184, + "step": 2341 + }, + { + "epoch": 1.9392600773053563, + "grad_norm": 0.8874300718307495, + "learning_rate": 3.879653960031494e-06, + "loss": 0.6191, + "step": 2342 + }, + { + "epoch": 1.940088348978465, + "grad_norm": 0.9244474768638611, + "learning_rate": 3.8787367563169e-06, + "loss": 0.5994, + "step": 2343 + }, + { + "epoch": 1.9409166206515738, + "grad_norm": 0.8986138701438904, + "learning_rate": 3.877819285829596e-06, + "loss": 0.6544, + "step": 2344 + }, + { + "epoch": 1.9417448923246825, + "grad_norm": 0.9084672927856445, + "learning_rate": 3.876901548747104e-06, + "loss": 0.6248, + "step": 2345 + }, + { + "epoch": 1.9425731639977912, + "grad_norm": 0.9052600264549255, + "learning_rate": 3.875983545246997e-06, + "loss": 0.6112, + "step": 2346 + }, + { + "epoch": 1.9434014356709, + "grad_norm": 0.9029438495635986, + "learning_rate": 3.8750652755069025e-06, + "loss": 0.6157, + "step": 2347 + }, + { + "epoch": 1.9442297073440087, + "grad_norm": 0.882449746131897, + "learning_rate": 3.874146739704494e-06, + "loss": 0.5886, + "step": 2348 + }, + { + "epoch": 1.9450579790171176, + "grad_norm": 0.8628272414207458, + "learning_rate": 3.873227938017502e-06, + "loss": 0.608, + "step": 2349 + }, + { + "epoch": 1.9458862506902264, + "grad_norm": 0.8977845907211304, + "learning_rate": 3.872308870623704e-06, + "loss": 0.6475, + "step": 2350 + }, + { + "epoch": 1.9467145223633353, + "grad_norm": 0.8801501393318176, + "learning_rate": 3.871389537700933e-06, + "loss": 0.6246, + "step": 2351 + }, + { + "epoch": 1.947542794036444, + "grad_norm": 0.9245358109474182, + "learning_rate": 3.87046993942707e-06, + "loss": 0.6317, + "step": 2352 + }, + { + "epoch": 1.9483710657095528, + "grad_norm": 0.8959249258041382, + "learning_rate": 3.869550075980049e-06, + "loss": 0.6157, + "step": 2353 + }, + { + "epoch": 1.9491993373826615, + "grad_norm": 0.861315906047821, + "learning_rate": 3.868629947537855e-06, + "loss": 0.6251, + "step": 2354 + }, + { + "epoch": 1.9500276090557702, + "grad_norm": 0.8839966654777527, + "learning_rate": 3.867709554278524e-06, + "loss": 0.6289, + "step": 2355 + }, + { + "epoch": 1.950855880728879, + "grad_norm": 0.8878759741783142, + "learning_rate": 3.8667888963801445e-06, + "loss": 0.6113, + "step": 2356 + }, + { + "epoch": 1.951684152401988, + "grad_norm": 0.8725037574768066, + "learning_rate": 3.865867974020854e-06, + "loss": 0.6324, + "step": 2357 + }, + { + "epoch": 1.9525124240750966, + "grad_norm": 0.9059397578239441, + "learning_rate": 3.8649467873788436e-06, + "loss": 0.5981, + "step": 2358 + }, + { + "epoch": 1.9533406957482056, + "grad_norm": 0.8883296847343445, + "learning_rate": 3.8640253366323535e-06, + "loss": 0.598, + "step": 2359 + }, + { + "epoch": 1.9541689674213143, + "grad_norm": 0.8987225890159607, + "learning_rate": 3.863103621959676e-06, + "loss": 0.6253, + "step": 2360 + }, + { + "epoch": 1.954997239094423, + "grad_norm": 0.8977320194244385, + "learning_rate": 3.862181643539155e-06, + "loss": 0.6424, + "step": 2361 + }, + { + "epoch": 1.9558255107675317, + "grad_norm": 0.9267589449882507, + "learning_rate": 3.861259401549183e-06, + "loss": 0.647, + "step": 2362 + }, + { + "epoch": 1.9566537824406405, + "grad_norm": 0.9297959804534912, + "learning_rate": 3.860336896168207e-06, + "loss": 0.6448, + "step": 2363 + }, + { + "epoch": 1.9574820541137492, + "grad_norm": 0.9627453088760376, + "learning_rate": 3.859414127574722e-06, + "loss": 0.6154, + "step": 2364 + }, + { + "epoch": 1.958310325786858, + "grad_norm": 0.8945866227149963, + "learning_rate": 3.858491095947275e-06, + "loss": 0.6201, + "step": 2365 + }, + { + "epoch": 1.9591385974599669, + "grad_norm": 0.8859817981719971, + "learning_rate": 3.857567801464466e-06, + "loss": 0.6171, + "step": 2366 + }, + { + "epoch": 1.9599668691330756, + "grad_norm": 0.9104642271995544, + "learning_rate": 3.8566442443049425e-06, + "loss": 0.6273, + "step": 2367 + }, + { + "epoch": 1.9607951408061846, + "grad_norm": 0.8821743726730347, + "learning_rate": 3.855720424647403e-06, + "loss": 0.5993, + "step": 2368 + }, + { + "epoch": 1.9616234124792933, + "grad_norm": 0.903322160243988, + "learning_rate": 3.854796342670601e-06, + "loss": 0.6471, + "step": 2369 + }, + { + "epoch": 1.962451684152402, + "grad_norm": 0.8744645714759827, + "learning_rate": 3.853871998553334e-06, + "loss": 0.6444, + "step": 2370 + }, + { + "epoch": 1.9632799558255107, + "grad_norm": 0.9538072347640991, + "learning_rate": 3.852947392474457e-06, + "loss": 0.656, + "step": 2371 + }, + { + "epoch": 1.9641082274986195, + "grad_norm": 0.875407874584198, + "learning_rate": 3.852022524612872e-06, + "loss": 0.6479, + "step": 2372 + }, + { + "epoch": 1.9649364991717282, + "grad_norm": 0.8908282518386841, + "learning_rate": 3.8510973951475305e-06, + "loss": 0.6262, + "step": 2373 + }, + { + "epoch": 1.9657647708448371, + "grad_norm": 0.8698679208755493, + "learning_rate": 3.850172004257438e-06, + "loss": 0.6431, + "step": 2374 + }, + { + "epoch": 1.9665930425179459, + "grad_norm": 0.8997004628181458, + "learning_rate": 3.84924635212165e-06, + "loss": 0.6349, + "step": 2375 + }, + { + "epoch": 1.9674213141910548, + "grad_norm": 0.8815768957138062, + "learning_rate": 3.8483204389192695e-06, + "loss": 0.6449, + "step": 2376 + }, + { + "epoch": 1.9682495858641635, + "grad_norm": 0.9148927927017212, + "learning_rate": 3.847394264829453e-06, + "loss": 0.6106, + "step": 2377 + }, + { + "epoch": 1.9690778575372723, + "grad_norm": 0.8647779226303101, + "learning_rate": 3.846467830031407e-06, + "loss": 0.5954, + "step": 2378 + }, + { + "epoch": 1.969906129210381, + "grad_norm": 0.8877570033073425, + "learning_rate": 3.845541134704387e-06, + "loss": 0.6006, + "step": 2379 + }, + { + "epoch": 1.9707344008834897, + "grad_norm": 0.899742603302002, + "learning_rate": 3.844614179027701e-06, + "loss": 0.6069, + "step": 2380 + }, + { + "epoch": 1.9715626725565985, + "grad_norm": 0.9085001349449158, + "learning_rate": 3.843686963180705e-06, + "loss": 0.6332, + "step": 2381 + }, + { + "epoch": 1.9723909442297074, + "grad_norm": 0.8954899311065674, + "learning_rate": 3.842759487342808e-06, + "loss": 0.6186, + "step": 2382 + }, + { + "epoch": 1.9732192159028161, + "grad_norm": 0.9011021256446838, + "learning_rate": 3.841831751693466e-06, + "loss": 0.6177, + "step": 2383 + }, + { + "epoch": 1.9740474875759249, + "grad_norm": 0.8824598789215088, + "learning_rate": 3.8409037564121895e-06, + "loss": 0.6392, + "step": 2384 + }, + { + "epoch": 1.9748757592490338, + "grad_norm": 0.8936701416969299, + "learning_rate": 3.839975501678536e-06, + "loss": 0.6508, + "step": 2385 + }, + { + "epoch": 1.9757040309221425, + "grad_norm": 0.9009436368942261, + "learning_rate": 3.839046987672113e-06, + "loss": 0.6169, + "step": 2386 + }, + { + "epoch": 1.9765323025952513, + "grad_norm": 0.8839378356933594, + "learning_rate": 3.8381182145725815e-06, + "loss": 0.6107, + "step": 2387 + }, + { + "epoch": 1.97736057426836, + "grad_norm": 0.9240637421607971, + "learning_rate": 3.837189182559649e-06, + "loss": 0.6083, + "step": 2388 + }, + { + "epoch": 1.9781888459414687, + "grad_norm": 0.881908118724823, + "learning_rate": 3.836259891813075e-06, + "loss": 0.5995, + "step": 2389 + }, + { + "epoch": 1.9790171176145774, + "grad_norm": 0.8754116296768188, + "learning_rate": 3.835330342512667e-06, + "loss": 0.6111, + "step": 2390 + }, + { + "epoch": 1.9798453892876864, + "grad_norm": 0.9713384509086609, + "learning_rate": 3.834400534838287e-06, + "loss": 0.641, + "step": 2391 + }, + { + "epoch": 1.9806736609607951, + "grad_norm": 0.9263403415679932, + "learning_rate": 3.833470468969842e-06, + "loss": 0.6254, + "step": 2392 + }, + { + "epoch": 1.981501932633904, + "grad_norm": 0.9033388495445251, + "learning_rate": 3.832540145087291e-06, + "loss": 0.6049, + "step": 2393 + }, + { + "epoch": 1.9823302043070128, + "grad_norm": 0.8710628151893616, + "learning_rate": 3.8316095633706455e-06, + "loss": 0.5999, + "step": 2394 + }, + { + "epoch": 1.9831584759801215, + "grad_norm": 0.9053885340690613, + "learning_rate": 3.830678723999961e-06, + "loss": 0.6393, + "step": 2395 + }, + { + "epoch": 1.9839867476532302, + "grad_norm": 0.9075314998626709, + "learning_rate": 3.829747627155348e-06, + "loss": 0.6188, + "step": 2396 + }, + { + "epoch": 1.984815019326339, + "grad_norm": 0.9151870012283325, + "learning_rate": 3.828816273016966e-06, + "loss": 0.625, + "step": 2397 + }, + { + "epoch": 1.9856432909994477, + "grad_norm": 0.903843104839325, + "learning_rate": 3.827884661765022e-06, + "loss": 0.618, + "step": 2398 + }, + { + "epoch": 1.9864715626725566, + "grad_norm": 0.875468373298645, + "learning_rate": 3.826952793579774e-06, + "loss": 0.6121, + "step": 2399 + }, + { + "epoch": 1.9872998343456654, + "grad_norm": 0.8908163905143738, + "learning_rate": 3.8260206686415295e-06, + "loss": 0.6292, + "step": 2400 + }, + { + "epoch": 1.9881281060187743, + "grad_norm": 0.8985518217086792, + "learning_rate": 3.825088287130647e-06, + "loss": 0.6356, + "step": 2401 + }, + { + "epoch": 1.988956377691883, + "grad_norm": 0.9277573227882385, + "learning_rate": 3.8241556492275335e-06, + "loss": 0.6424, + "step": 2402 + }, + { + "epoch": 1.9897846493649918, + "grad_norm": 0.9075362682342529, + "learning_rate": 3.823222755112646e-06, + "loss": 0.623, + "step": 2403 + }, + { + "epoch": 1.9906129210381005, + "grad_norm": 0.9038227796554565, + "learning_rate": 3.822289604966489e-06, + "loss": 0.6195, + "step": 2404 + }, + { + "epoch": 1.9914411927112092, + "grad_norm": 0.8869004845619202, + "learning_rate": 3.82135619896962e-06, + "loss": 0.6034, + "step": 2405 + }, + { + "epoch": 1.992269464384318, + "grad_norm": 0.9017077684402466, + "learning_rate": 3.820422537302645e-06, + "loss": 0.6469, + "step": 2406 + }, + { + "epoch": 1.9930977360574267, + "grad_norm": 0.8785225749015808, + "learning_rate": 3.819488620146218e-06, + "loss": 0.6327, + "step": 2407 + }, + { + "epoch": 1.9939260077305356, + "grad_norm": 1.0367780923843384, + "learning_rate": 3.818554447681041e-06, + "loss": 0.6259, + "step": 2408 + }, + { + "epoch": 1.9947542794036444, + "grad_norm": 0.9392277002334595, + "learning_rate": 3.817620020087872e-06, + "loss": 0.6393, + "step": 2409 + }, + { + "epoch": 1.9955825510767533, + "grad_norm": 0.8935515880584717, + "learning_rate": 3.816685337547511e-06, + "loss": 0.6153, + "step": 2410 + }, + { + "epoch": 1.996410822749862, + "grad_norm": 0.8895331025123596, + "learning_rate": 3.815750400240811e-06, + "loss": 0.5978, + "step": 2411 + }, + { + "epoch": 1.9972390944229708, + "grad_norm": 0.8733546733856201, + "learning_rate": 3.814815208348673e-06, + "loss": 0.6254, + "step": 2412 + }, + { + "epoch": 1.9980673660960795, + "grad_norm": 0.9177356958389282, + "learning_rate": 3.8138797620520507e-06, + "loss": 0.631, + "step": 2413 + }, + { + "epoch": 1.9988956377691882, + "grad_norm": 0.9259604811668396, + "learning_rate": 3.8129440615319414e-06, + "loss": 0.6138, + "step": 2414 + }, + { + "epoch": 1.999723909442297, + "grad_norm": 0.9226608872413635, + "learning_rate": 3.812008106969395e-06, + "loss": 0.607, + "step": 2415 + }, + { + "epoch": 2.0, + "grad_norm": 0.9226608872413635, + "learning_rate": 3.8110718985455097e-06, + "loss": 0.5993, + "step": 2416 + }, + { + "epoch": 2.0008282716731087, + "grad_norm": 1.62429940700531, + "learning_rate": 3.810135436441435e-06, + "loss": 0.5954, + "step": 2417 + }, + { + "epoch": 2.0016565433462175, + "grad_norm": 0.8664761781692505, + "learning_rate": 3.8091987208383656e-06, + "loss": 0.5947, + "step": 2418 + }, + { + "epoch": 2.002484815019326, + "grad_norm": 0.9208955764770508, + "learning_rate": 3.808261751917548e-06, + "loss": 0.6028, + "step": 2419 + }, + { + "epoch": 2.003313086692435, + "grad_norm": 0.8842632174491882, + "learning_rate": 3.8073245298602775e-06, + "loss": 0.5913, + "step": 2420 + }, + { + "epoch": 2.004141358365544, + "grad_norm": 0.9232655763626099, + "learning_rate": 3.8063870548478964e-06, + "loss": 0.5977, + "step": 2421 + }, + { + "epoch": 2.004969630038653, + "grad_norm": 0.8898128271102905, + "learning_rate": 3.805449327061798e-06, + "loss": 0.6335, + "step": 2422 + }, + { + "epoch": 2.0057979017117615, + "grad_norm": 0.9356569051742554, + "learning_rate": 3.8045113466834248e-06, + "loss": 0.616, + "step": 2423 + }, + { + "epoch": 2.0066261733848703, + "grad_norm": 0.890315055847168, + "learning_rate": 3.803573113894266e-06, + "loss": 0.5874, + "step": 2424 + }, + { + "epoch": 2.007454445057979, + "grad_norm": 0.8921117186546326, + "learning_rate": 3.802634628875861e-06, + "loss": 0.5679, + "step": 2425 + }, + { + "epoch": 2.0082827167310877, + "grad_norm": 0.8871824145317078, + "learning_rate": 3.8016958918097995e-06, + "loss": 0.6066, + "step": 2426 + }, + { + "epoch": 2.0091109884041964, + "grad_norm": 0.902680516242981, + "learning_rate": 3.800756902877717e-06, + "loss": 0.5936, + "step": 2427 + }, + { + "epoch": 2.009939260077305, + "grad_norm": 0.9199647307395935, + "learning_rate": 3.799817662261298e-06, + "loss": 0.5836, + "step": 2428 + }, + { + "epoch": 2.0107675317504143, + "grad_norm": 0.9029080271720886, + "learning_rate": 3.7988781701422795e-06, + "loss": 0.5858, + "step": 2429 + }, + { + "epoch": 2.011595803423523, + "grad_norm": 0.8834925889968872, + "learning_rate": 3.7979384267024422e-06, + "loss": 0.6061, + "step": 2430 + }, + { + "epoch": 2.012424075096632, + "grad_norm": 0.9098751544952393, + "learning_rate": 3.796998432123619e-06, + "loss": 0.6238, + "step": 2431 + }, + { + "epoch": 2.0132523467697405, + "grad_norm": 0.9316835999488831, + "learning_rate": 3.7960581865876898e-06, + "loss": 0.6139, + "step": 2432 + }, + { + "epoch": 2.0140806184428492, + "grad_norm": 0.9198770523071289, + "learning_rate": 3.795117690276583e-06, + "loss": 0.6113, + "step": 2433 + }, + { + "epoch": 2.014908890115958, + "grad_norm": 0.8970876336097717, + "learning_rate": 3.794176943372275e-06, + "loss": 0.6033, + "step": 2434 + }, + { + "epoch": 2.0157371617890667, + "grad_norm": 0.9025728702545166, + "learning_rate": 3.7932359460567945e-06, + "loss": 0.586, + "step": 2435 + }, + { + "epoch": 2.0165654334621754, + "grad_norm": 0.891909122467041, + "learning_rate": 3.7922946985122124e-06, + "loss": 0.6013, + "step": 2436 + }, + { + "epoch": 2.017393705135284, + "grad_norm": 0.9115222096443176, + "learning_rate": 3.791353200920654e-06, + "loss": 0.5783, + "step": 2437 + }, + { + "epoch": 2.0182219768083933, + "grad_norm": 0.8973708152770996, + "learning_rate": 3.790411453464288e-06, + "loss": 0.5862, + "step": 2438 + }, + { + "epoch": 2.019050248481502, + "grad_norm": 0.910248875617981, + "learning_rate": 3.7894694563253352e-06, + "loss": 0.5884, + "step": 2439 + }, + { + "epoch": 2.019878520154611, + "grad_norm": 0.9227123260498047, + "learning_rate": 3.7885272096860625e-06, + "loss": 0.6185, + "step": 2440 + }, + { + "epoch": 2.0207067918277195, + "grad_norm": 0.917061448097229, + "learning_rate": 3.787584713728787e-06, + "loss": 0.5767, + "step": 2441 + }, + { + "epoch": 2.0215350635008282, + "grad_norm": 0.9035825133323669, + "learning_rate": 3.786641968635871e-06, + "loss": 0.6064, + "step": 2442 + }, + { + "epoch": 2.022363335173937, + "grad_norm": 0.8991772532463074, + "learning_rate": 3.78569897458973e-06, + "loss": 0.5912, + "step": 2443 + }, + { + "epoch": 2.0231916068470457, + "grad_norm": 0.9237397909164429, + "learning_rate": 3.78475573177282e-06, + "loss": 0.5818, + "step": 2444 + }, + { + "epoch": 2.0240198785201544, + "grad_norm": 0.8913205862045288, + "learning_rate": 3.7838122403676543e-06, + "loss": 0.5972, + "step": 2445 + }, + { + "epoch": 2.0248481501932636, + "grad_norm": 0.8960773944854736, + "learning_rate": 3.782868500556787e-06, + "loss": 0.6118, + "step": 2446 + }, + { + "epoch": 2.0256764218663723, + "grad_norm": 0.9198427200317383, + "learning_rate": 3.7819245125228228e-06, + "loss": 0.6058, + "step": 2447 + }, + { + "epoch": 2.026504693539481, + "grad_norm": 0.9039538502693176, + "learning_rate": 3.780980276448416e-06, + "loss": 0.6122, + "step": 2448 + }, + { + "epoch": 2.0273329652125898, + "grad_norm": 0.9103277325630188, + "learning_rate": 3.7800357925162675e-06, + "loss": 0.584, + "step": 2449 + }, + { + "epoch": 2.0281612368856985, + "grad_norm": 0.9114090204238892, + "learning_rate": 3.779091060909124e-06, + "loss": 0.6071, + "step": 2450 + }, + { + "epoch": 2.028989508558807, + "grad_norm": 0.9064953327178955, + "learning_rate": 3.7781460818097847e-06, + "loss": 0.6097, + "step": 2451 + }, + { + "epoch": 2.029817780231916, + "grad_norm": 0.893298864364624, + "learning_rate": 3.7772008554010937e-06, + "loss": 0.5959, + "step": 2452 + }, + { + "epoch": 2.0306460519050247, + "grad_norm": 0.8867486119270325, + "learning_rate": 3.776255381865942e-06, + "loss": 0.5788, + "step": 2453 + }, + { + "epoch": 2.031474323578134, + "grad_norm": 0.9018534421920776, + "learning_rate": 3.775309661387272e-06, + "loss": 0.6075, + "step": 2454 + }, + { + "epoch": 2.0323025952512426, + "grad_norm": 0.9353018999099731, + "learning_rate": 3.77436369414807e-06, + "loss": 0.5894, + "step": 2455 + }, + { + "epoch": 2.0331308669243513, + "grad_norm": 0.8828651309013367, + "learning_rate": 3.7734174803313733e-06, + "loss": 0.608, + "step": 2456 + }, + { + "epoch": 2.03395913859746, + "grad_norm": 0.8971118330955505, + "learning_rate": 3.7724710201202645e-06, + "loss": 0.597, + "step": 2457 + }, + { + "epoch": 2.0347874102705688, + "grad_norm": 0.9064204096794128, + "learning_rate": 3.7715243136978753e-06, + "loss": 0.61, + "step": 2458 + }, + { + "epoch": 2.0356156819436775, + "grad_norm": 0.9204864501953125, + "learning_rate": 3.770577361247384e-06, + "loss": 0.5908, + "step": 2459 + }, + { + "epoch": 2.036443953616786, + "grad_norm": 0.905817449092865, + "learning_rate": 3.7696301629520183e-06, + "loss": 0.5867, + "step": 2460 + }, + { + "epoch": 2.037272225289895, + "grad_norm": 0.8872947096824646, + "learning_rate": 3.7686827189950505e-06, + "loss": 0.5805, + "step": 2461 + }, + { + "epoch": 2.0381004969630037, + "grad_norm": 0.8944478631019592, + "learning_rate": 3.7677350295598035e-06, + "loss": 0.5835, + "step": 2462 + }, + { + "epoch": 2.038928768636113, + "grad_norm": 0.884630560874939, + "learning_rate": 3.766787094829646e-06, + "loss": 0.5909, + "step": 2463 + }, + { + "epoch": 2.0397570403092216, + "grad_norm": 0.8915899395942688, + "learning_rate": 3.7658389149879936e-06, + "loss": 0.5923, + "step": 2464 + }, + { + "epoch": 2.0405853119823303, + "grad_norm": 0.8971114158630371, + "learning_rate": 3.7648904902183128e-06, + "loss": 0.6097, + "step": 2465 + }, + { + "epoch": 2.041413583655439, + "grad_norm": 0.9223624467849731, + "learning_rate": 3.7639418207041124e-06, + "loss": 0.6048, + "step": 2466 + }, + { + "epoch": 2.0422418553285477, + "grad_norm": 0.915450930595398, + "learning_rate": 3.7629929066289517e-06, + "loss": 0.626, + "step": 2467 + }, + { + "epoch": 2.0430701270016565, + "grad_norm": 0.919333279132843, + "learning_rate": 3.762043748176437e-06, + "loss": 0.6234, + "step": 2468 + }, + { + "epoch": 2.043898398674765, + "grad_norm": 0.9282277822494507, + "learning_rate": 3.7610943455302223e-06, + "loss": 0.6008, + "step": 2469 + }, + { + "epoch": 2.044726670347874, + "grad_norm": 0.9419016242027283, + "learning_rate": 3.760144698874007e-06, + "loss": 0.5836, + "step": 2470 + }, + { + "epoch": 2.045554942020983, + "grad_norm": 0.91156405210495, + "learning_rate": 3.7591948083915403e-06, + "loss": 0.6035, + "step": 2471 + }, + { + "epoch": 2.046383213694092, + "grad_norm": 0.9019245505332947, + "learning_rate": 3.758244674266615e-06, + "loss": 0.5949, + "step": 2472 + }, + { + "epoch": 2.0472114853672005, + "grad_norm": 0.8976905345916748, + "learning_rate": 3.757294296683075e-06, + "loss": 0.5818, + "step": 2473 + }, + { + "epoch": 2.0480397570403093, + "grad_norm": 0.8748987913131714, + "learning_rate": 3.7563436758248094e-06, + "loss": 0.5816, + "step": 2474 + }, + { + "epoch": 2.048868028713418, + "grad_norm": 0.8929162621498108, + "learning_rate": 3.7553928118757532e-06, + "loss": 0.5869, + "step": 2475 + }, + { + "epoch": 2.0496963003865267, + "grad_norm": 0.9968146681785583, + "learning_rate": 3.7544417050198913e-06, + "loss": 0.6065, + "step": 2476 + }, + { + "epoch": 2.0505245720596355, + "grad_norm": 0.9206544756889343, + "learning_rate": 3.753490355441253e-06, + "loss": 0.5919, + "step": 2477 + }, + { + "epoch": 2.051352843732744, + "grad_norm": 0.8804112672805786, + "learning_rate": 3.7525387633239153e-06, + "loss": 0.5918, + "step": 2478 + }, + { + "epoch": 2.052181115405853, + "grad_norm": 0.922532320022583, + "learning_rate": 3.7515869288520027e-06, + "loss": 0.6212, + "step": 2479 + }, + { + "epoch": 2.053009387078962, + "grad_norm": 0.9353113770484924, + "learning_rate": 3.7506348522096876e-06, + "loss": 0.5695, + "step": 2480 + }, + { + "epoch": 2.053837658752071, + "grad_norm": 0.9110690951347351, + "learning_rate": 3.749682533581186e-06, + "loss": 0.6156, + "step": 2481 + }, + { + "epoch": 2.0546659304251795, + "grad_norm": 0.9076771140098572, + "learning_rate": 3.7487299731507642e-06, + "loss": 0.5987, + "step": 2482 + }, + { + "epoch": 2.0554942020982883, + "grad_norm": 0.8937394618988037, + "learning_rate": 3.747777171102732e-06, + "loss": 0.6057, + "step": 2483 + }, + { + "epoch": 2.056322473771397, + "grad_norm": 0.9488244652748108, + "learning_rate": 3.74682412762145e-06, + "loss": 0.5983, + "step": 2484 + }, + { + "epoch": 2.0571507454445057, + "grad_norm": 0.897693395614624, + "learning_rate": 3.7458708428913217e-06, + "loss": 0.606, + "step": 2485 + }, + { + "epoch": 2.0579790171176144, + "grad_norm": 0.8989591598510742, + "learning_rate": 3.744917317096799e-06, + "loss": 0.6138, + "step": 2486 + }, + { + "epoch": 2.058807288790723, + "grad_norm": 0.9028336405754089, + "learning_rate": 3.7439635504223804e-06, + "loss": 0.5714, + "step": 2487 + }, + { + "epoch": 2.0596355604638323, + "grad_norm": 0.8895571231842041, + "learning_rate": 3.7430095430526113e-06, + "loss": 0.5903, + "step": 2488 + }, + { + "epoch": 2.060463832136941, + "grad_norm": 0.926414430141449, + "learning_rate": 3.7420552951720822e-06, + "loss": 0.5841, + "step": 2489 + }, + { + "epoch": 2.06129210381005, + "grad_norm": 0.9279632568359375, + "learning_rate": 3.741100806965432e-06, + "loss": 0.6062, + "step": 2490 + }, + { + "epoch": 2.0621203754831585, + "grad_norm": 0.9196921586990356, + "learning_rate": 3.7401460786173456e-06, + "loss": 0.579, + "step": 2491 + }, + { + "epoch": 2.0629486471562672, + "grad_norm": 0.8959674835205078, + "learning_rate": 3.739191110312552e-06, + "loss": 0.5888, + "step": 2492 + }, + { + "epoch": 2.063776918829376, + "grad_norm": 0.9123788475990295, + "learning_rate": 3.7382359022358322e-06, + "loss": 0.5891, + "step": 2493 + }, + { + "epoch": 2.0646051905024847, + "grad_norm": 0.9637120366096497, + "learning_rate": 3.7372804545720067e-06, + "loss": 0.6124, + "step": 2494 + }, + { + "epoch": 2.0654334621755934, + "grad_norm": 0.9360100626945496, + "learning_rate": 3.736324767505947e-06, + "loss": 0.5884, + "step": 2495 + }, + { + "epoch": 2.0662617338487026, + "grad_norm": 0.9066753387451172, + "learning_rate": 3.7353688412225704e-06, + "loss": 0.6013, + "step": 2496 + }, + { + "epoch": 2.0670900055218113, + "grad_norm": 0.8961504101753235, + "learning_rate": 3.7344126759068382e-06, + "loss": 0.6303, + "step": 2497 + }, + { + "epoch": 2.06791827719492, + "grad_norm": 0.8995484709739685, + "learning_rate": 3.7334562717437608e-06, + "loss": 0.6031, + "step": 2498 + }, + { + "epoch": 2.068746548868029, + "grad_norm": 0.9279561042785645, + "learning_rate": 3.7324996289183927e-06, + "loss": 0.5848, + "step": 2499 + }, + { + "epoch": 2.0695748205411375, + "grad_norm": 0.9250274300575256, + "learning_rate": 3.7315427476158357e-06, + "loss": 0.5949, + "step": 2500 + }, + { + "epoch": 2.0704030922142462, + "grad_norm": 0.9003238677978516, + "learning_rate": 3.730585628021237e-06, + "loss": 0.6124, + "step": 2501 + }, + { + "epoch": 2.071231363887355, + "grad_norm": 0.9069858193397522, + "learning_rate": 3.729628270319791e-06, + "loss": 0.6051, + "step": 2502 + }, + { + "epoch": 2.0720596355604637, + "grad_norm": 0.9361990094184875, + "learning_rate": 3.7286706746967372e-06, + "loss": 0.5863, + "step": 2503 + }, + { + "epoch": 2.0728879072335724, + "grad_norm": 0.9443079233169556, + "learning_rate": 3.7277128413373607e-06, + "loss": 0.5959, + "step": 2504 + }, + { + "epoch": 2.0737161789066816, + "grad_norm": 0.9413166046142578, + "learning_rate": 3.7267547704269957e-06, + "loss": 0.5882, + "step": 2505 + }, + { + "epoch": 2.0745444505797903, + "grad_norm": 0.8937835097312927, + "learning_rate": 3.7257964621510167e-06, + "loss": 0.6108, + "step": 2506 + }, + { + "epoch": 2.075372722252899, + "grad_norm": 0.9188135862350464, + "learning_rate": 3.7248379166948497e-06, + "loss": 0.5812, + "step": 2507 + }, + { + "epoch": 2.0762009939260078, + "grad_norm": 0.916559636592865, + "learning_rate": 3.723879134243963e-06, + "loss": 0.6169, + "step": 2508 + }, + { + "epoch": 2.0770292655991165, + "grad_norm": 0.9284812211990356, + "learning_rate": 3.7229201149838727e-06, + "loss": 0.6224, + "step": 2509 + }, + { + "epoch": 2.077857537272225, + "grad_norm": 0.918048083782196, + "learning_rate": 3.721960859100141e-06, + "loss": 0.6043, + "step": 2510 + }, + { + "epoch": 2.078685808945334, + "grad_norm": 0.9250833988189697, + "learning_rate": 3.721001366778373e-06, + "loss": 0.5882, + "step": 2511 + }, + { + "epoch": 2.0795140806184427, + "grad_norm": 0.8975113034248352, + "learning_rate": 3.720041638204223e-06, + "loss": 0.6049, + "step": 2512 + }, + { + "epoch": 2.080342352291552, + "grad_norm": 0.9108197093009949, + "learning_rate": 3.7190816735633896e-06, + "loss": 0.6174, + "step": 2513 + }, + { + "epoch": 2.0811706239646606, + "grad_norm": 0.91886967420578, + "learning_rate": 3.718121473041615e-06, + "loss": 0.5854, + "step": 2514 + }, + { + "epoch": 2.0819988956377693, + "grad_norm": 0.9149959683418274, + "learning_rate": 3.717161036824692e-06, + "loss": 0.5808, + "step": 2515 + }, + { + "epoch": 2.082827167310878, + "grad_norm": 0.9402881860733032, + "learning_rate": 3.7162003650984545e-06, + "loss": 0.5936, + "step": 2516 + }, + { + "epoch": 2.0836554389839868, + "grad_norm": 0.8952768445014954, + "learning_rate": 3.7152394580487825e-06, + "loss": 0.5977, + "step": 2517 + }, + { + "epoch": 2.0844837106570955, + "grad_norm": 0.9138445258140564, + "learning_rate": 3.714278315861605e-06, + "loss": 0.5822, + "step": 2518 + }, + { + "epoch": 2.085311982330204, + "grad_norm": 0.9174036979675293, + "learning_rate": 3.713316938722892e-06, + "loss": 0.6061, + "step": 2519 + }, + { + "epoch": 2.086140254003313, + "grad_norm": 0.9291302561759949, + "learning_rate": 3.7123553268186618e-06, + "loss": 0.614, + "step": 2520 + }, + { + "epoch": 2.0869685256764217, + "grad_norm": 0.9058101773262024, + "learning_rate": 3.711393480334977e-06, + "loss": 0.6053, + "step": 2521 + }, + { + "epoch": 2.087796797349531, + "grad_norm": 0.926360011100769, + "learning_rate": 3.7104313994579473e-06, + "loss": 0.5781, + "step": 2522 + }, + { + "epoch": 2.0886250690226396, + "grad_norm": 0.898629367351532, + "learning_rate": 3.709469084373725e-06, + "loss": 0.5995, + "step": 2523 + }, + { + "epoch": 2.0894533406957483, + "grad_norm": 0.9013628363609314, + "learning_rate": 3.708506535268509e-06, + "loss": 0.5929, + "step": 2524 + }, + { + "epoch": 2.090281612368857, + "grad_norm": 0.8909673094749451, + "learning_rate": 3.7075437523285433e-06, + "loss": 0.5495, + "step": 2525 + }, + { + "epoch": 2.0911098840419657, + "grad_norm": 0.9176347255706787, + "learning_rate": 3.706580735740119e-06, + "loss": 0.6134, + "step": 2526 + }, + { + "epoch": 2.0919381557150745, + "grad_norm": 0.9174275398254395, + "learning_rate": 3.7056174856895703e-06, + "loss": 0.601, + "step": 2527 + }, + { + "epoch": 2.092766427388183, + "grad_norm": 0.9134400486946106, + "learning_rate": 3.704654002363276e-06, + "loss": 0.5952, + "step": 2528 + }, + { + "epoch": 2.093594699061292, + "grad_norm": 0.9206627607345581, + "learning_rate": 3.703690285947662e-06, + "loss": 0.6016, + "step": 2529 + }, + { + "epoch": 2.094422970734401, + "grad_norm": 0.9271345734596252, + "learning_rate": 3.702726336629199e-06, + "loss": 0.5937, + "step": 2530 + }, + { + "epoch": 2.09525124240751, + "grad_norm": 0.9390851855278015, + "learning_rate": 3.7017621545944003e-06, + "loss": 0.5858, + "step": 2531 + }, + { + "epoch": 2.0960795140806185, + "grad_norm": 0.9351701736450195, + "learning_rate": 3.700797740029828e-06, + "loss": 0.5978, + "step": 2532 + }, + { + "epoch": 2.0969077857537273, + "grad_norm": 0.9432377815246582, + "learning_rate": 3.6998330931220873e-06, + "loss": 0.5884, + "step": 2533 + }, + { + "epoch": 2.097736057426836, + "grad_norm": 0.9192067384719849, + "learning_rate": 3.698868214057827e-06, + "loss": 0.5989, + "step": 2534 + }, + { + "epoch": 2.0985643290999447, + "grad_norm": 0.9810166954994202, + "learning_rate": 3.6979031030237433e-06, + "loss": 0.6078, + "step": 2535 + }, + { + "epoch": 2.0993926007730535, + "grad_norm": 0.9065185785293579, + "learning_rate": 3.696937760206576e-06, + "loss": 0.5947, + "step": 2536 + }, + { + "epoch": 2.100220872446162, + "grad_norm": 0.9212245345115662, + "learning_rate": 3.6959721857931088e-06, + "loss": 0.5951, + "step": 2537 + }, + { + "epoch": 2.101049144119271, + "grad_norm": 0.9291266798973083, + "learning_rate": 3.695006379970173e-06, + "loss": 0.6033, + "step": 2538 + }, + { + "epoch": 2.10187741579238, + "grad_norm": 0.9154751896858215, + "learning_rate": 3.694040342924643e-06, + "loss": 0.612, + "step": 2539 + }, + { + "epoch": 2.102705687465489, + "grad_norm": 0.9028742909431458, + "learning_rate": 3.693074074843437e-06, + "loss": 0.5948, + "step": 2540 + }, + { + "epoch": 2.1035339591385975, + "grad_norm": 0.9132126569747925, + "learning_rate": 3.692107575913519e-06, + "loss": 0.6198, + "step": 2541 + }, + { + "epoch": 2.1043622308117063, + "grad_norm": 0.9116196036338806, + "learning_rate": 3.6911408463218974e-06, + "loss": 0.5927, + "step": 2542 + }, + { + "epoch": 2.105190502484815, + "grad_norm": 0.9670912027359009, + "learning_rate": 3.6901738862556263e-06, + "loss": 0.5813, + "step": 2543 + }, + { + "epoch": 2.1060187741579237, + "grad_norm": 0.9534422755241394, + "learning_rate": 3.6892066959018034e-06, + "loss": 0.5892, + "step": 2544 + }, + { + "epoch": 2.1068470458310324, + "grad_norm": 0.9175525903701782, + "learning_rate": 3.688239275447569e-06, + "loss": 0.5946, + "step": 2545 + }, + { + "epoch": 2.107675317504141, + "grad_norm": 0.8945409655570984, + "learning_rate": 3.6872716250801125e-06, + "loss": 0.6013, + "step": 2546 + }, + { + "epoch": 2.1085035891772503, + "grad_norm": 0.9498820900917053, + "learning_rate": 3.686303744986664e-06, + "loss": 0.5785, + "step": 2547 + }, + { + "epoch": 2.109331860850359, + "grad_norm": 0.9338169693946838, + "learning_rate": 3.685335635354499e-06, + "loss": 0.5959, + "step": 2548 + }, + { + "epoch": 2.110160132523468, + "grad_norm": 0.9100049734115601, + "learning_rate": 3.684367296370939e-06, + "loss": 0.5829, + "step": 2549 + }, + { + "epoch": 2.1109884041965765, + "grad_norm": 0.904279887676239, + "learning_rate": 3.6833987282233467e-06, + "loss": 0.6015, + "step": 2550 + }, + { + "epoch": 2.1118166758696852, + "grad_norm": 0.8693552613258362, + "learning_rate": 3.6824299310991324e-06, + "loss": 0.5825, + "step": 2551 + }, + { + "epoch": 2.112644947542794, + "grad_norm": 0.9180129766464233, + "learning_rate": 3.681460905185749e-06, + "loss": 0.6076, + "step": 2552 + }, + { + "epoch": 2.1134732192159027, + "grad_norm": 0.9554781913757324, + "learning_rate": 3.6804916506706936e-06, + "loss": 0.5755, + "step": 2553 + }, + { + "epoch": 2.1143014908890114, + "grad_norm": 0.9498181939125061, + "learning_rate": 3.6795221677415085e-06, + "loss": 0.6086, + "step": 2554 + }, + { + "epoch": 2.1151297625621206, + "grad_norm": 0.9110347032546997, + "learning_rate": 3.678552456585779e-06, + "loss": 0.591, + "step": 2555 + }, + { + "epoch": 2.1159580342352293, + "grad_norm": 0.8883147835731506, + "learning_rate": 3.6775825173911363e-06, + "loss": 0.5753, + "step": 2556 + }, + { + "epoch": 2.116786305908338, + "grad_norm": 0.9354976415634155, + "learning_rate": 3.676612350345253e-06, + "loss": 0.6063, + "step": 2557 + }, + { + "epoch": 2.1176145775814468, + "grad_norm": 0.9463969469070435, + "learning_rate": 3.675641955635848e-06, + "loss": 0.5983, + "step": 2558 + }, + { + "epoch": 2.1184428492545555, + "grad_norm": 0.9220679998397827, + "learning_rate": 3.674671333450684e-06, + "loss": 0.6026, + "step": 2559 + }, + { + "epoch": 2.1192711209276642, + "grad_norm": 0.9050368666648865, + "learning_rate": 3.673700483977567e-06, + "loss": 0.6075, + "step": 2560 + }, + { + "epoch": 2.120099392600773, + "grad_norm": 0.9299139976501465, + "learning_rate": 3.6727294074043478e-06, + "loss": 0.5971, + "step": 2561 + }, + { + "epoch": 2.1209276642738817, + "grad_norm": 0.9060289263725281, + "learning_rate": 3.6717581039189205e-06, + "loss": 0.5794, + "step": 2562 + }, + { + "epoch": 2.1217559359469904, + "grad_norm": 0.9253103137016296, + "learning_rate": 3.6707865737092228e-06, + "loss": 0.593, + "step": 2563 + }, + { + "epoch": 2.1225842076200996, + "grad_norm": 0.9475048780441284, + "learning_rate": 3.669814816963237e-06, + "loss": 0.5867, + "step": 2564 + }, + { + "epoch": 2.1234124792932083, + "grad_norm": 0.9546397924423218, + "learning_rate": 3.6688428338689887e-06, + "loss": 0.5766, + "step": 2565 + }, + { + "epoch": 2.124240750966317, + "grad_norm": 0.9377962350845337, + "learning_rate": 3.667870624614548e-06, + "loss": 0.5853, + "step": 2566 + }, + { + "epoch": 2.1250690226394258, + "grad_norm": 0.9562273621559143, + "learning_rate": 3.6668981893880284e-06, + "loss": 0.5992, + "step": 2567 + }, + { + "epoch": 2.1258972943125345, + "grad_norm": 0.9207432866096497, + "learning_rate": 3.6659255283775867e-06, + "loss": 0.6029, + "step": 2568 + }, + { + "epoch": 2.126725565985643, + "grad_norm": 0.9359074234962463, + "learning_rate": 3.6649526417714244e-06, + "loss": 0.6068, + "step": 2569 + }, + { + "epoch": 2.127553837658752, + "grad_norm": 0.9253597259521484, + "learning_rate": 3.663979529757784e-06, + "loss": 0.581, + "step": 2570 + }, + { + "epoch": 2.1283821093318607, + "grad_norm": 0.9551291465759277, + "learning_rate": 3.6630061925249566e-06, + "loss": 0.607, + "step": 2571 + }, + { + "epoch": 2.1292103810049694, + "grad_norm": 0.91109699010849, + "learning_rate": 3.662032630261272e-06, + "loss": 0.5724, + "step": 2572 + }, + { + "epoch": 2.1300386526780786, + "grad_norm": 0.9486557245254517, + "learning_rate": 3.661058843155105e-06, + "loss": 0.5904, + "step": 2573 + }, + { + "epoch": 2.1308669243511873, + "grad_norm": 0.9550185203552246, + "learning_rate": 3.6600848313948762e-06, + "loss": 0.5967, + "step": 2574 + }, + { + "epoch": 2.131695196024296, + "grad_norm": 0.9457482695579529, + "learning_rate": 3.6591105951690464e-06, + "loss": 0.6253, + "step": 2575 + }, + { + "epoch": 2.1325234676974048, + "grad_norm": 0.9686241149902344, + "learning_rate": 3.6581361346661204e-06, + "loss": 0.5928, + "step": 2576 + }, + { + "epoch": 2.1333517393705135, + "grad_norm": 0.9125609993934631, + "learning_rate": 3.65716145007465e-06, + "loss": 0.6039, + "step": 2577 + }, + { + "epoch": 2.134180011043622, + "grad_norm": 0.927651047706604, + "learning_rate": 3.656186541583225e-06, + "loss": 0.5881, + "step": 2578 + }, + { + "epoch": 2.135008282716731, + "grad_norm": 0.9446879029273987, + "learning_rate": 3.6552114093804825e-06, + "loss": 0.5938, + "step": 2579 + }, + { + "epoch": 2.13583655438984, + "grad_norm": 0.944601833820343, + "learning_rate": 3.6542360536551003e-06, + "loss": 0.5804, + "step": 2580 + }, + { + "epoch": 2.136664826062949, + "grad_norm": 0.9230138063430786, + "learning_rate": 3.6532604745958015e-06, + "loss": 0.6002, + "step": 2581 + }, + { + "epoch": 2.1374930977360576, + "grad_norm": 0.9163529276847839, + "learning_rate": 3.6522846723913517e-06, + "loss": 0.5811, + "step": 2582 + }, + { + "epoch": 2.1383213694091663, + "grad_norm": 0.9153618812561035, + "learning_rate": 3.6513086472305596e-06, + "loss": 0.6018, + "step": 2583 + }, + { + "epoch": 2.139149641082275, + "grad_norm": 0.9299359917640686, + "learning_rate": 3.6503323993022767e-06, + "loss": 0.5944, + "step": 2584 + }, + { + "epoch": 2.1399779127553837, + "grad_norm": 0.9075869917869568, + "learning_rate": 3.6493559287953974e-06, + "loss": 0.595, + "step": 2585 + }, + { + "epoch": 2.1408061844284925, + "grad_norm": 0.9388751983642578, + "learning_rate": 3.64837923589886e-06, + "loss": 0.6006, + "step": 2586 + }, + { + "epoch": 2.141634456101601, + "grad_norm": 0.91156405210495, + "learning_rate": 3.647402320801645e-06, + "loss": 0.5955, + "step": 2587 + }, + { + "epoch": 2.14246272777471, + "grad_norm": 0.9187628626823425, + "learning_rate": 3.6464251836927776e-06, + "loss": 0.6119, + "step": 2588 + }, + { + "epoch": 2.143290999447819, + "grad_norm": 0.9578025341033936, + "learning_rate": 3.645447824761324e-06, + "loss": 0.6076, + "step": 2589 + }, + { + "epoch": 2.144119271120928, + "grad_norm": 0.9159366488456726, + "learning_rate": 3.644470244196394e-06, + "loss": 0.6015, + "step": 2590 + }, + { + "epoch": 2.1449475427940365, + "grad_norm": 0.924027681350708, + "learning_rate": 3.6434924421871403e-06, + "loss": 0.6082, + "step": 2591 + }, + { + "epoch": 2.1457758144671453, + "grad_norm": 0.8900635242462158, + "learning_rate": 3.642514418922758e-06, + "loss": 0.611, + "step": 2592 + }, + { + "epoch": 2.146604086140254, + "grad_norm": 0.9609982371330261, + "learning_rate": 3.641536174592487e-06, + "loss": 0.5974, + "step": 2593 + }, + { + "epoch": 2.1474323578133627, + "grad_norm": 0.901737630367279, + "learning_rate": 3.6405577093856072e-06, + "loss": 0.5869, + "step": 2594 + }, + { + "epoch": 2.1482606294864715, + "grad_norm": 0.9342123866081238, + "learning_rate": 3.6395790234914423e-06, + "loss": 0.6104, + "step": 2595 + }, + { + "epoch": 2.14908890115958, + "grad_norm": 0.9063602685928345, + "learning_rate": 3.6386001170993596e-06, + "loss": 0.5801, + "step": 2596 + }, + { + "epoch": 2.149917172832689, + "grad_norm": 0.9269636869430542, + "learning_rate": 3.637620990398768e-06, + "loss": 0.5953, + "step": 2597 + }, + { + "epoch": 2.150745444505798, + "grad_norm": 0.9278765320777893, + "learning_rate": 3.63664164357912e-06, + "loss": 0.6151, + "step": 2598 + }, + { + "epoch": 2.151573716178907, + "grad_norm": 0.9786942601203918, + "learning_rate": 3.6356620768299087e-06, + "loss": 0.5773, + "step": 2599 + }, + { + "epoch": 2.1524019878520155, + "grad_norm": 0.9069851636886597, + "learning_rate": 3.6346822903406733e-06, + "loss": 0.6087, + "step": 2600 + }, + { + "epoch": 2.1532302595251243, + "grad_norm": 0.8867887854576111, + "learning_rate": 3.6337022843009913e-06, + "loss": 0.5894, + "step": 2601 + }, + { + "epoch": 2.154058531198233, + "grad_norm": 0.9444507360458374, + "learning_rate": 3.632722058900486e-06, + "loss": 0.6125, + "step": 2602 + }, + { + "epoch": 2.1548868028713417, + "grad_norm": 0.9410359859466553, + "learning_rate": 3.631741614328821e-06, + "loss": 0.5845, + "step": 2603 + }, + { + "epoch": 2.1557150745444504, + "grad_norm": 0.9473987817764282, + "learning_rate": 3.630760950775703e-06, + "loss": 0.5944, + "step": 2604 + }, + { + "epoch": 2.156543346217559, + "grad_norm": 0.913758397102356, + "learning_rate": 3.6297800684308828e-06, + "loss": 0.5998, + "step": 2605 + }, + { + "epoch": 2.1573716178906683, + "grad_norm": 0.89509516954422, + "learning_rate": 3.6287989674841505e-06, + "loss": 0.6013, + "step": 2606 + }, + { + "epoch": 2.158199889563777, + "grad_norm": 0.9042367339134216, + "learning_rate": 3.6278176481253413e-06, + "loss": 0.5915, + "step": 2607 + }, + { + "epoch": 2.159028161236886, + "grad_norm": 0.9108945727348328, + "learning_rate": 3.6268361105443297e-06, + "loss": 0.5803, + "step": 2608 + }, + { + "epoch": 2.1598564329099945, + "grad_norm": 0.8903773427009583, + "learning_rate": 3.625854354931035e-06, + "loss": 0.6025, + "step": 2609 + }, + { + "epoch": 2.1606847045831032, + "grad_norm": 0.9396197199821472, + "learning_rate": 3.6248723814754183e-06, + "loss": 0.6271, + "step": 2610 + }, + { + "epoch": 2.161512976256212, + "grad_norm": 0.9247238039970398, + "learning_rate": 3.623890190367481e-06, + "loss": 0.5945, + "step": 2611 + }, + { + "epoch": 2.1623412479293207, + "grad_norm": 0.9240396022796631, + "learning_rate": 3.622907781797269e-06, + "loss": 0.6148, + "step": 2612 + }, + { + "epoch": 2.1631695196024294, + "grad_norm": 0.9257000684738159, + "learning_rate": 3.621925155954869e-06, + "loss": 0.5964, + "step": 2613 + }, + { + "epoch": 2.1639977912755386, + "grad_norm": 0.8854860663414001, + "learning_rate": 3.62094231303041e-06, + "loss": 0.6012, + "step": 2614 + }, + { + "epoch": 2.1648260629486473, + "grad_norm": 0.9251745343208313, + "learning_rate": 3.619959253214062e-06, + "loss": 0.6001, + "step": 2615 + }, + { + "epoch": 2.165654334621756, + "grad_norm": 0.9081252813339233, + "learning_rate": 3.618975976696039e-06, + "loss": 0.5927, + "step": 2616 + }, + { + "epoch": 2.1664826062948648, + "grad_norm": 0.9065699577331543, + "learning_rate": 3.6179924836665953e-06, + "loss": 0.6079, + "step": 2617 + }, + { + "epoch": 2.1673108779679735, + "grad_norm": 0.9421206712722778, + "learning_rate": 3.617008774316028e-06, + "loss": 0.6041, + "step": 2618 + }, + { + "epoch": 2.1681391496410822, + "grad_norm": 0.8863933682441711, + "learning_rate": 3.6160248488346757e-06, + "loss": 0.5826, + "step": 2619 + }, + { + "epoch": 2.168967421314191, + "grad_norm": 0.915887713432312, + "learning_rate": 3.615040707412918e-06, + "loss": 0.5972, + "step": 2620 + }, + { + "epoch": 2.1697956929872997, + "grad_norm": 0.9150416851043701, + "learning_rate": 3.614056350241178e-06, + "loss": 0.606, + "step": 2621 + }, + { + "epoch": 2.1706239646604084, + "grad_norm": 0.9050796627998352, + "learning_rate": 3.6130717775099195e-06, + "loss": 0.5955, + "step": 2622 + }, + { + "epoch": 2.1714522363335176, + "grad_norm": 0.9249832630157471, + "learning_rate": 3.612086989409648e-06, + "loss": 0.5968, + "step": 2623 + }, + { + "epoch": 2.1722805080066263, + "grad_norm": 0.8855478763580322, + "learning_rate": 3.6111019861309105e-06, + "loss": 0.5657, + "step": 2624 + }, + { + "epoch": 2.173108779679735, + "grad_norm": 0.9182875156402588, + "learning_rate": 3.6101167678642967e-06, + "loss": 0.6096, + "step": 2625 + }, + { + "epoch": 2.1739370513528438, + "grad_norm": 0.9722150564193726, + "learning_rate": 3.609131334800436e-06, + "loss": 0.6029, + "step": 2626 + }, + { + "epoch": 2.1747653230259525, + "grad_norm": 0.8925914168357849, + "learning_rate": 3.6081456871300014e-06, + "loss": 0.5928, + "step": 2627 + }, + { + "epoch": 2.175593594699061, + "grad_norm": 0.8883031010627747, + "learning_rate": 3.6071598250437067e-06, + "loss": 0.5763, + "step": 2628 + }, + { + "epoch": 2.17642186637217, + "grad_norm": 0.8948857188224792, + "learning_rate": 3.606173748732307e-06, + "loss": 0.6078, + "step": 2629 + }, + { + "epoch": 2.1772501380452787, + "grad_norm": 0.9444127082824707, + "learning_rate": 3.6051874583865975e-06, + "loss": 0.6005, + "step": 2630 + }, + { + "epoch": 2.1780784097183874, + "grad_norm": 0.9361352920532227, + "learning_rate": 3.6042009541974176e-06, + "loss": 0.5878, + "step": 2631 + }, + { + "epoch": 2.1789066813914966, + "grad_norm": 0.9124311804771423, + "learning_rate": 3.6032142363556467e-06, + "loss": 0.6115, + "step": 2632 + }, + { + "epoch": 2.1797349530646053, + "grad_norm": 0.9168087840080261, + "learning_rate": 3.602227305052205e-06, + "loss": 0.5928, + "step": 2633 + }, + { + "epoch": 2.180563224737714, + "grad_norm": 0.9189275503158569, + "learning_rate": 3.6012401604780543e-06, + "loss": 0.6081, + "step": 2634 + }, + { + "epoch": 2.1813914964108228, + "grad_norm": 0.937911868095398, + "learning_rate": 3.6002528028241984e-06, + "loss": 0.6062, + "step": 2635 + }, + { + "epoch": 2.1822197680839315, + "grad_norm": 0.9139302968978882, + "learning_rate": 3.5992652322816817e-06, + "loss": 0.6111, + "step": 2636 + }, + { + "epoch": 2.18304803975704, + "grad_norm": 0.951528012752533, + "learning_rate": 3.598277449041589e-06, + "loss": 0.6007, + "step": 2637 + }, + { + "epoch": 2.183876311430149, + "grad_norm": 0.9146255850791931, + "learning_rate": 3.5972894532950487e-06, + "loss": 0.6094, + "step": 2638 + }, + { + "epoch": 2.184704583103258, + "grad_norm": 0.9217225313186646, + "learning_rate": 3.596301245233228e-06, + "loss": 0.6082, + "step": 2639 + }, + { + "epoch": 2.185532854776367, + "grad_norm": 0.9742298126220703, + "learning_rate": 3.5953128250473355e-06, + "loss": 0.6269, + "step": 2640 + }, + { + "epoch": 2.1863611264494756, + "grad_norm": 0.9038170576095581, + "learning_rate": 3.5943241929286223e-06, + "loss": 0.6062, + "step": 2641 + }, + { + "epoch": 2.1871893981225843, + "grad_norm": 0.8960873484611511, + "learning_rate": 3.5933353490683786e-06, + "loss": 0.5748, + "step": 2642 + }, + { + "epoch": 2.188017669795693, + "grad_norm": 0.8801732659339905, + "learning_rate": 3.5923462936579362e-06, + "loss": 0.5911, + "step": 2643 + }, + { + "epoch": 2.1888459414688017, + "grad_norm": 0.8918222188949585, + "learning_rate": 3.5913570268886695e-06, + "loss": 0.5931, + "step": 2644 + }, + { + "epoch": 2.1896742131419105, + "grad_norm": 0.9046133160591125, + "learning_rate": 3.5903675489519917e-06, + "loss": 0.5985, + "step": 2645 + }, + { + "epoch": 2.190502484815019, + "grad_norm": 0.9003219604492188, + "learning_rate": 3.589377860039357e-06, + "loss": 0.608, + "step": 2646 + }, + { + "epoch": 2.191330756488128, + "grad_norm": 0.9139525890350342, + "learning_rate": 3.5883879603422616e-06, + "loss": 0.6186, + "step": 2647 + }, + { + "epoch": 2.192159028161237, + "grad_norm": 0.9125071167945862, + "learning_rate": 3.587397850052241e-06, + "loss": 0.6129, + "step": 2648 + }, + { + "epoch": 2.192987299834346, + "grad_norm": 0.922412097454071, + "learning_rate": 3.586407529360874e-06, + "loss": 0.6008, + "step": 2649 + }, + { + "epoch": 2.1938155715074545, + "grad_norm": 0.9294847249984741, + "learning_rate": 3.5854169984597776e-06, + "loss": 0.5984, + "step": 2650 + }, + { + "epoch": 2.1946438431805633, + "grad_norm": 0.9262758493423462, + "learning_rate": 3.5844262575406084e-06, + "loss": 0.5982, + "step": 2651 + }, + { + "epoch": 2.195472114853672, + "grad_norm": 0.9311552047729492, + "learning_rate": 3.5834353067950694e-06, + "loss": 0.5965, + "step": 2652 + }, + { + "epoch": 2.1963003865267807, + "grad_norm": 0.9087333679199219, + "learning_rate": 3.5824441464148975e-06, + "loss": 0.5972, + "step": 2653 + }, + { + "epoch": 2.1971286581998895, + "grad_norm": 0.9212601184844971, + "learning_rate": 3.5814527765918733e-06, + "loss": 0.5821, + "step": 2654 + }, + { + "epoch": 2.197956929872998, + "grad_norm": 0.9208887219429016, + "learning_rate": 3.5804611975178184e-06, + "loss": 0.592, + "step": 2655 + }, + { + "epoch": 2.198785201546107, + "grad_norm": 0.923478364944458, + "learning_rate": 3.5794694093845933e-06, + "loss": 0.599, + "step": 2656 + }, + { + "epoch": 2.199613473219216, + "grad_norm": 0.9030081033706665, + "learning_rate": 3.5784774123841006e-06, + "loss": 0.5954, + "step": 2657 + }, + { + "epoch": 2.200441744892325, + "grad_norm": 0.9108036160469055, + "learning_rate": 3.5774852067082817e-06, + "loss": 0.6119, + "step": 2658 + }, + { + "epoch": 2.2012700165654335, + "grad_norm": 0.9323570728302002, + "learning_rate": 3.576492792549119e-06, + "loss": 0.6061, + "step": 2659 + }, + { + "epoch": 2.2020982882385423, + "grad_norm": 0.9002283811569214, + "learning_rate": 3.575500170098637e-06, + "loss": 0.6183, + "step": 2660 + }, + { + "epoch": 2.202926559911651, + "grad_norm": 0.9354086518287659, + "learning_rate": 3.574507339548897e-06, + "loss": 0.5989, + "step": 2661 + }, + { + "epoch": 2.2037548315847597, + "grad_norm": 0.9035785794258118, + "learning_rate": 3.5735143010920033e-06, + "loss": 0.592, + "step": 2662 + }, + { + "epoch": 2.2045831032578684, + "grad_norm": 0.9337678551673889, + "learning_rate": 3.5725210549201e-06, + "loss": 0.5881, + "step": 2663 + }, + { + "epoch": 2.205411374930977, + "grad_norm": 0.9226170182228088, + "learning_rate": 3.57152760122537e-06, + "loss": 0.5745, + "step": 2664 + }, + { + "epoch": 2.2062396466040863, + "grad_norm": 0.9272558093070984, + "learning_rate": 3.5705339402000377e-06, + "loss": 0.607, + "step": 2665 + }, + { + "epoch": 2.207067918277195, + "grad_norm": 0.930325448513031, + "learning_rate": 3.569540072036368e-06, + "loss": 0.5947, + "step": 2666 + }, + { + "epoch": 2.207896189950304, + "grad_norm": 0.9052232503890991, + "learning_rate": 3.568545996926664e-06, + "loss": 0.5883, + "step": 2667 + }, + { + "epoch": 2.2087244616234125, + "grad_norm": 0.8831207752227783, + "learning_rate": 3.5675517150632706e-06, + "loss": 0.5962, + "step": 2668 + }, + { + "epoch": 2.2095527332965212, + "grad_norm": 0.8999590873718262, + "learning_rate": 3.566557226638572e-06, + "loss": 0.6257, + "step": 2669 + }, + { + "epoch": 2.21038100496963, + "grad_norm": 0.9310323596000671, + "learning_rate": 3.565562531844992e-06, + "loss": 0.585, + "step": 2670 + }, + { + "epoch": 2.2112092766427387, + "grad_norm": 0.9469419121742249, + "learning_rate": 3.5645676308749954e-06, + "loss": 0.6163, + "step": 2671 + }, + { + "epoch": 2.2120375483158474, + "grad_norm": 0.9391891956329346, + "learning_rate": 3.563572523921086e-06, + "loss": 0.5863, + "step": 2672 + }, + { + "epoch": 2.2128658199889566, + "grad_norm": 0.9035549163818359, + "learning_rate": 3.5625772111758085e-06, + "loss": 0.6074, + "step": 2673 + }, + { + "epoch": 2.2136940916620653, + "grad_norm": 0.9008681178092957, + "learning_rate": 3.561581692831745e-06, + "loss": 0.5965, + "step": 2674 + }, + { + "epoch": 2.214522363335174, + "grad_norm": 0.9130175113677979, + "learning_rate": 3.56058596908152e-06, + "loss": 0.5913, + "step": 2675 + }, + { + "epoch": 2.2153506350082828, + "grad_norm": 0.9367324113845825, + "learning_rate": 3.5595900401177957e-06, + "loss": 0.617, + "step": 2676 + }, + { + "epoch": 2.2161789066813915, + "grad_norm": 0.9082489609718323, + "learning_rate": 3.5585939061332775e-06, + "loss": 0.6052, + "step": 2677 + }, + { + "epoch": 2.2170071783545002, + "grad_norm": 0.921264111995697, + "learning_rate": 3.557597567320706e-06, + "loss": 0.5832, + "step": 2678 + }, + { + "epoch": 2.217835450027609, + "grad_norm": 0.9040131568908691, + "learning_rate": 3.556601023872863e-06, + "loss": 0.5994, + "step": 2679 + }, + { + "epoch": 2.2186637217007177, + "grad_norm": 0.9144189357757568, + "learning_rate": 3.5556042759825733e-06, + "loss": 0.6068, + "step": 2680 + }, + { + "epoch": 2.2194919933738264, + "grad_norm": 0.91495680809021, + "learning_rate": 3.5546073238426955e-06, + "loss": 0.6154, + "step": 2681 + }, + { + "epoch": 2.2203202650469356, + "grad_norm": 0.925397515296936, + "learning_rate": 3.553610167646131e-06, + "loss": 0.6013, + "step": 2682 + }, + { + "epoch": 2.2211485367200443, + "grad_norm": 0.9633582234382629, + "learning_rate": 3.552612807585821e-06, + "loss": 0.6063, + "step": 2683 + }, + { + "epoch": 2.221976808393153, + "grad_norm": 0.9403156042098999, + "learning_rate": 3.551615243854746e-06, + "loss": 0.6141, + "step": 2684 + }, + { + "epoch": 2.2228050800662618, + "grad_norm": 0.9272189140319824, + "learning_rate": 3.550617476645923e-06, + "loss": 0.6013, + "step": 2685 + }, + { + "epoch": 2.2236333517393705, + "grad_norm": 0.92893385887146, + "learning_rate": 3.549619506152413e-06, + "loss": 0.5909, + "step": 2686 + }, + { + "epoch": 2.224461623412479, + "grad_norm": 0.9130013585090637, + "learning_rate": 3.548621332567312e-06, + "loss": 0.6005, + "step": 2687 + }, + { + "epoch": 2.225289895085588, + "grad_norm": 0.9111050367355347, + "learning_rate": 3.547622956083759e-06, + "loss": 0.5904, + "step": 2688 + }, + { + "epoch": 2.2261181667586967, + "grad_norm": 0.9379095435142517, + "learning_rate": 3.5466243768949292e-06, + "loss": 0.605, + "step": 2689 + }, + { + "epoch": 2.2269464384318054, + "grad_norm": 0.898931622505188, + "learning_rate": 3.5456255951940393e-06, + "loss": 0.5978, + "step": 2690 + }, + { + "epoch": 2.2277747101049146, + "grad_norm": 0.9569118022918701, + "learning_rate": 3.5446266111743444e-06, + "loss": 0.6162, + "step": 2691 + }, + { + "epoch": 2.2286029817780233, + "grad_norm": 0.9167311191558838, + "learning_rate": 3.543627425029138e-06, + "loss": 0.5876, + "step": 2692 + }, + { + "epoch": 2.229431253451132, + "grad_norm": 0.9584367275238037, + "learning_rate": 3.542628036951753e-06, + "loss": 0.5704, + "step": 2693 + }, + { + "epoch": 2.2302595251242407, + "grad_norm": 0.8896269202232361, + "learning_rate": 3.541628447135562e-06, + "loss": 0.5806, + "step": 2694 + }, + { + "epoch": 2.2310877967973495, + "grad_norm": 0.9340798854827881, + "learning_rate": 3.5406286557739767e-06, + "loss": 0.6218, + "step": 2695 + }, + { + "epoch": 2.231916068470458, + "grad_norm": 0.9442514777183533, + "learning_rate": 3.539628663060447e-06, + "loss": 0.5899, + "step": 2696 + }, + { + "epoch": 2.232744340143567, + "grad_norm": 0.9123642444610596, + "learning_rate": 3.5386284691884633e-06, + "loss": 0.5825, + "step": 2697 + }, + { + "epoch": 2.233572611816676, + "grad_norm": 0.9076257348060608, + "learning_rate": 3.537628074351551e-06, + "loss": 0.6001, + "step": 2698 + }, + { + "epoch": 2.234400883489785, + "grad_norm": 0.9335387945175171, + "learning_rate": 3.5366274787432804e-06, + "loss": 0.6048, + "step": 2699 + }, + { + "epoch": 2.2352291551628936, + "grad_norm": 0.9016919732093811, + "learning_rate": 3.535626682557255e-06, + "loss": 0.6175, + "step": 2700 + }, + { + "epoch": 2.2360574268360023, + "grad_norm": 0.9110099673271179, + "learning_rate": 3.5346256859871207e-06, + "loss": 0.6027, + "step": 2701 + }, + { + "epoch": 2.236885698509111, + "grad_norm": 0.9513245224952698, + "learning_rate": 3.5336244892265616e-06, + "loss": 0.6015, + "step": 2702 + }, + { + "epoch": 2.2377139701822197, + "grad_norm": 0.8986294865608215, + "learning_rate": 3.532623092469299e-06, + "loss": 0.61, + "step": 2703 + }, + { + "epoch": 2.2385422418553285, + "grad_norm": 0.8991696834564209, + "learning_rate": 3.5316214959090933e-06, + "loss": 0.6013, + "step": 2704 + }, + { + "epoch": 2.239370513528437, + "grad_norm": 0.9667947292327881, + "learning_rate": 3.5306196997397453e-06, + "loss": 0.6125, + "step": 2705 + }, + { + "epoch": 2.240198785201546, + "grad_norm": 0.9183294177055359, + "learning_rate": 3.529617704155092e-06, + "loss": 0.5901, + "step": 2706 + }, + { + "epoch": 2.241027056874655, + "grad_norm": 0.9139719605445862, + "learning_rate": 3.5286155093490117e-06, + "loss": 0.5798, + "step": 2707 + }, + { + "epoch": 2.241855328547764, + "grad_norm": 0.9529849290847778, + "learning_rate": 3.5276131155154193e-06, + "loss": 0.5763, + "step": 2708 + }, + { + "epoch": 2.2426836002208725, + "grad_norm": 0.9126972556114197, + "learning_rate": 3.526610522848267e-06, + "loss": 0.5804, + "step": 2709 + }, + { + "epoch": 2.2435118718939813, + "grad_norm": 0.9101532101631165, + "learning_rate": 3.525607731541549e-06, + "loss": 0.5815, + "step": 2710 + }, + { + "epoch": 2.24434014356709, + "grad_norm": 0.9072697162628174, + "learning_rate": 3.524604741789296e-06, + "loss": 0.6238, + "step": 2711 + }, + { + "epoch": 2.2451684152401987, + "grad_norm": 0.9228153228759766, + "learning_rate": 3.5236015537855765e-06, + "loss": 0.6011, + "step": 2712 + }, + { + "epoch": 2.2459966869133074, + "grad_norm": 0.9280546307563782, + "learning_rate": 3.522598167724498e-06, + "loss": 0.6018, + "step": 2713 + }, + { + "epoch": 2.246824958586416, + "grad_norm": 0.9077439904212952, + "learning_rate": 3.5215945838002075e-06, + "loss": 0.6231, + "step": 2714 + }, + { + "epoch": 2.247653230259525, + "grad_norm": 0.9201393127441406, + "learning_rate": 3.520590802206887e-06, + "loss": 0.591, + "step": 2715 + }, + { + "epoch": 2.248481501932634, + "grad_norm": 0.8888259530067444, + "learning_rate": 3.519586823138761e-06, + "loss": 0.585, + "step": 2716 + }, + { + "epoch": 2.249309773605743, + "grad_norm": 0.918370246887207, + "learning_rate": 3.5185826467900886e-06, + "loss": 0.6151, + "step": 2717 + }, + { + "epoch": 2.2501380452788515, + "grad_norm": 0.9071224331855774, + "learning_rate": 3.517578273355169e-06, + "loss": 0.6053, + "step": 2718 + }, + { + "epoch": 2.2509663169519603, + "grad_norm": 0.9325213432312012, + "learning_rate": 3.5165737030283404e-06, + "loss": 0.5959, + "step": 2719 + }, + { + "epoch": 2.251794588625069, + "grad_norm": 0.9014924168586731, + "learning_rate": 3.515568936003976e-06, + "loss": 0.6092, + "step": 2720 + }, + { + "epoch": 2.2526228602981777, + "grad_norm": 0.923115611076355, + "learning_rate": 3.5145639724764886e-06, + "loss": 0.6031, + "step": 2721 + }, + { + "epoch": 2.2534511319712864, + "grad_norm": 0.9452555179595947, + "learning_rate": 3.513558812640331e-06, + "loss": 0.5957, + "step": 2722 + }, + { + "epoch": 2.2542794036443956, + "grad_norm": 0.960884153842926, + "learning_rate": 3.5125534566899915e-06, + "loss": 0.5862, + "step": 2723 + }, + { + "epoch": 2.2551076753175043, + "grad_norm": 0.9164205193519592, + "learning_rate": 3.5115479048199964e-06, + "loss": 0.5992, + "step": 2724 + }, + { + "epoch": 2.255935946990613, + "grad_norm": 0.943760097026825, + "learning_rate": 3.5105421572249123e-06, + "loss": 0.5842, + "step": 2725 + }, + { + "epoch": 2.256764218663722, + "grad_norm": 0.9578623175621033, + "learning_rate": 3.50953621409934e-06, + "loss": 0.6094, + "step": 2726 + }, + { + "epoch": 2.2575924903368305, + "grad_norm": 0.9307887554168701, + "learning_rate": 3.5085300756379216e-06, + "loss": 0.5905, + "step": 2727 + }, + { + "epoch": 2.2584207620099392, + "grad_norm": 0.9150220155715942, + "learning_rate": 3.5075237420353346e-06, + "loss": 0.6035, + "step": 2728 + }, + { + "epoch": 2.259249033683048, + "grad_norm": 0.8887060880661011, + "learning_rate": 3.506517213486296e-06, + "loss": 0.6032, + "step": 2729 + }, + { + "epoch": 2.2600773053561567, + "grad_norm": 0.9477519989013672, + "learning_rate": 3.5055104901855585e-06, + "loss": 0.576, + "step": 2730 + }, + { + "epoch": 2.2609055770292654, + "grad_norm": 0.9668344855308533, + "learning_rate": 3.504503572327915e-06, + "loss": 0.5973, + "step": 2731 + }, + { + "epoch": 2.2617338487023746, + "grad_norm": 0.905135989189148, + "learning_rate": 3.503496460108194e-06, + "loss": 0.5999, + "step": 2732 + }, + { + "epoch": 2.2625621203754833, + "grad_norm": 0.933996856212616, + "learning_rate": 3.5024891537212624e-06, + "loss": 0.5803, + "step": 2733 + }, + { + "epoch": 2.263390392048592, + "grad_norm": 0.9417551159858704, + "learning_rate": 3.501481653362025e-06, + "loss": 0.5918, + "step": 2734 + }, + { + "epoch": 2.2642186637217008, + "grad_norm": 0.9239657521247864, + "learning_rate": 3.500473959225422e-06, + "loss": 0.5828, + "step": 2735 + }, + { + "epoch": 2.2650469353948095, + "grad_norm": 0.9021532535552979, + "learning_rate": 3.499466071506436e-06, + "loss": 0.5958, + "step": 2736 + }, + { + "epoch": 2.2658752070679182, + "grad_norm": 0.9159536957740784, + "learning_rate": 3.498457990400081e-06, + "loss": 0.5867, + "step": 2737 + }, + { + "epoch": 2.266703478741027, + "grad_norm": 0.9347800612449646, + "learning_rate": 3.4974497161014124e-06, + "loss": 0.6058, + "step": 2738 + }, + { + "epoch": 2.2675317504141357, + "grad_norm": 0.9229647517204285, + "learning_rate": 3.496441248805522e-06, + "loss": 0.5714, + "step": 2739 + }, + { + "epoch": 2.2683600220872444, + "grad_norm": 0.9330629706382751, + "learning_rate": 3.4954325887075386e-06, + "loss": 0.5808, + "step": 2740 + }, + { + "epoch": 2.2691882937603536, + "grad_norm": 0.8958399891853333, + "learning_rate": 3.494423736002629e-06, + "loss": 0.5793, + "step": 2741 + }, + { + "epoch": 2.2700165654334623, + "grad_norm": 0.8952097296714783, + "learning_rate": 3.493414690885997e-06, + "loss": 0.6007, + "step": 2742 + }, + { + "epoch": 2.270844837106571, + "grad_norm": 0.9001036882400513, + "learning_rate": 3.4924054535528813e-06, + "loss": 0.5821, + "step": 2743 + }, + { + "epoch": 2.2716731087796798, + "grad_norm": 0.9016004204750061, + "learning_rate": 3.4913960241985616e-06, + "loss": 0.5819, + "step": 2744 + }, + { + "epoch": 2.2725013804527885, + "grad_norm": 0.9043833017349243, + "learning_rate": 3.490386403018354e-06, + "loss": 0.6144, + "step": 2745 + }, + { + "epoch": 2.273329652125897, + "grad_norm": 0.9396303296089172, + "learning_rate": 3.489376590207608e-06, + "loss": 0.602, + "step": 2746 + }, + { + "epoch": 2.274157923799006, + "grad_norm": 0.9685373306274414, + "learning_rate": 3.488366585961716e-06, + "loss": 0.6247, + "step": 2747 + }, + { + "epoch": 2.2749861954721147, + "grad_norm": 0.9028962254524231, + "learning_rate": 3.487356390476103e-06, + "loss": 0.5959, + "step": 2748 + }, + { + "epoch": 2.2758144671452234, + "grad_norm": 0.9150850772857666, + "learning_rate": 3.486346003946232e-06, + "loss": 0.5632, + "step": 2749 + }, + { + "epoch": 2.2766427388183326, + "grad_norm": 0.902158796787262, + "learning_rate": 3.4853354265676043e-06, + "loss": 0.5901, + "step": 2750 + }, + { + "epoch": 2.2774710104914413, + "grad_norm": 0.8942626118659973, + "learning_rate": 3.4843246585357565e-06, + "loss": 0.6082, + "step": 2751 + }, + { + "epoch": 2.27829928216455, + "grad_norm": 0.912050187587738, + "learning_rate": 3.4833137000462623e-06, + "loss": 0.5812, + "step": 2752 + }, + { + "epoch": 2.2791275538376587, + "grad_norm": 0.9262410998344421, + "learning_rate": 3.482302551294735e-06, + "loss": 0.6199, + "step": 2753 + }, + { + "epoch": 2.2799558255107675, + "grad_norm": 0.9281290173530579, + "learning_rate": 3.4812912124768193e-06, + "loss": 0.591, + "step": 2754 + }, + { + "epoch": 2.280784097183876, + "grad_norm": 0.8852386474609375, + "learning_rate": 3.4802796837882023e-06, + "loss": 0.6035, + "step": 2755 + }, + { + "epoch": 2.281612368856985, + "grad_norm": 0.8982499241828918, + "learning_rate": 3.479267965424605e-06, + "loss": 0.6023, + "step": 2756 + }, + { + "epoch": 2.282440640530094, + "grad_norm": 0.9022878408432007, + "learning_rate": 3.4782560575817837e-06, + "loss": 0.6004, + "step": 2757 + }, + { + "epoch": 2.283268912203203, + "grad_norm": 0.9256078600883484, + "learning_rate": 3.4772439604555353e-06, + "loss": 0.5811, + "step": 2758 + }, + { + "epoch": 2.2840971838763116, + "grad_norm": 0.9009407758712769, + "learning_rate": 3.4762316742416913e-06, + "loss": 0.6053, + "step": 2759 + }, + { + "epoch": 2.2849254555494203, + "grad_norm": 0.9257068037986755, + "learning_rate": 3.4752191991361174e-06, + "loss": 0.5664, + "step": 2760 + }, + { + "epoch": 2.285753727222529, + "grad_norm": 0.882075309753418, + "learning_rate": 3.4742065353347193e-06, + "loss": 0.6099, + "step": 2761 + }, + { + "epoch": 2.2865819988956377, + "grad_norm": 0.9553154706954956, + "learning_rate": 3.4731936830334384e-06, + "loss": 0.6072, + "step": 2762 + }, + { + "epoch": 2.2874102705687465, + "grad_norm": 0.9329802989959717, + "learning_rate": 3.4721806424282512e-06, + "loss": 0.601, + "step": 2763 + }, + { + "epoch": 2.288238542241855, + "grad_norm": 0.9146580696105957, + "learning_rate": 3.471167413715173e-06, + "loss": 0.599, + "step": 2764 + }, + { + "epoch": 2.289066813914964, + "grad_norm": 0.9170219302177429, + "learning_rate": 3.470153997090253e-06, + "loss": 0.5969, + "step": 2765 + }, + { + "epoch": 2.289895085588073, + "grad_norm": 0.8903328776359558, + "learning_rate": 3.469140392749578e-06, + "loss": 0.6063, + "step": 2766 + }, + { + "epoch": 2.290723357261182, + "grad_norm": 0.9068611264228821, + "learning_rate": 3.468126600889272e-06, + "loss": 0.5807, + "step": 2767 + }, + { + "epoch": 2.2915516289342905, + "grad_norm": 0.9464553594589233, + "learning_rate": 3.467112621705493e-06, + "loss": 0.5955, + "step": 2768 + }, + { + "epoch": 2.2923799006073993, + "grad_norm": 0.9302126169204712, + "learning_rate": 3.4660984553944367e-06, + "loss": 0.5956, + "step": 2769 + }, + { + "epoch": 2.293208172280508, + "grad_norm": 0.8797576427459717, + "learning_rate": 3.465084102152336e-06, + "loss": 0.5964, + "step": 2770 + }, + { + "epoch": 2.2940364439536167, + "grad_norm": 0.9544422030448914, + "learning_rate": 3.4640695621754573e-06, + "loss": 0.6225, + "step": 2771 + }, + { + "epoch": 2.2948647156267254, + "grad_norm": 0.919297993183136, + "learning_rate": 3.463054835660105e-06, + "loss": 0.5901, + "step": 2772 + }, + { + "epoch": 2.295692987299834, + "grad_norm": 0.9211829900741577, + "learning_rate": 3.4620399228026203e-06, + "loss": 0.5949, + "step": 2773 + }, + { + "epoch": 2.296521258972943, + "grad_norm": 0.929501473903656, + "learning_rate": 3.4610248237993775e-06, + "loss": 0.5654, + "step": 2774 + }, + { + "epoch": 2.297349530646052, + "grad_norm": 0.9604958891868591, + "learning_rate": 3.4600095388467907e-06, + "loss": 0.6059, + "step": 2775 + }, + { + "epoch": 2.298177802319161, + "grad_norm": 0.9042990207672119, + "learning_rate": 3.458994068141307e-06, + "loss": 0.5901, + "step": 2776 + }, + { + "epoch": 2.2990060739922695, + "grad_norm": 0.8958423733711243, + "learning_rate": 3.4579784118794107e-06, + "loss": 0.5938, + "step": 2777 + }, + { + "epoch": 2.2998343456653783, + "grad_norm": 0.9388688206672668, + "learning_rate": 3.456962570257622e-06, + "loss": 0.6024, + "step": 2778 + }, + { + "epoch": 2.300662617338487, + "grad_norm": 0.9006328582763672, + "learning_rate": 3.455946543472497e-06, + "loss": 0.5929, + "step": 2779 + }, + { + "epoch": 2.3014908890115957, + "grad_norm": 0.887873113155365, + "learning_rate": 3.454930331720626e-06, + "loss": 0.599, + "step": 2780 + }, + { + "epoch": 2.3023191606847044, + "grad_norm": 0.8919411301612854, + "learning_rate": 3.4539139351986385e-06, + "loss": 0.6173, + "step": 2781 + }, + { + "epoch": 2.3031474323578136, + "grad_norm": 0.9478135704994202, + "learning_rate": 3.452897354103196e-06, + "loss": 0.6025, + "step": 2782 + }, + { + "epoch": 2.3039757040309223, + "grad_norm": 0.9009284377098083, + "learning_rate": 3.451880588630999e-06, + "loss": 0.602, + "step": 2783 + }, + { + "epoch": 2.304803975704031, + "grad_norm": 0.9043265581130981, + "learning_rate": 3.4508636389787816e-06, + "loss": 0.6002, + "step": 2784 + }, + { + "epoch": 2.30563224737714, + "grad_norm": 0.9005764126777649, + "learning_rate": 3.4498465053433126e-06, + "loss": 0.618, + "step": 2785 + }, + { + "epoch": 2.3064605190502485, + "grad_norm": 0.9352818727493286, + "learning_rate": 3.4488291879214003e-06, + "loss": 0.6069, + "step": 2786 + }, + { + "epoch": 2.3072887907233572, + "grad_norm": 0.9092200994491577, + "learning_rate": 3.4478116869098854e-06, + "loss": 0.581, + "step": 2787 + }, + { + "epoch": 2.308117062396466, + "grad_norm": 0.9055072665214539, + "learning_rate": 3.4467940025056433e-06, + "loss": 0.598, + "step": 2788 + }, + { + "epoch": 2.3089453340695747, + "grad_norm": 1.0053009986877441, + "learning_rate": 3.4457761349055885e-06, + "loss": 0.5929, + "step": 2789 + }, + { + "epoch": 2.3097736057426834, + "grad_norm": 0.9241059422492981, + "learning_rate": 3.4447580843066675e-06, + "loss": 0.5962, + "step": 2790 + }, + { + "epoch": 2.3106018774157926, + "grad_norm": 0.927590548992157, + "learning_rate": 3.443739850905864e-06, + "loss": 0.5863, + "step": 2791 + }, + { + "epoch": 2.3114301490889013, + "grad_norm": 0.9227457046508789, + "learning_rate": 3.442721434900197e-06, + "loss": 0.5928, + "step": 2792 + }, + { + "epoch": 2.31225842076201, + "grad_norm": 0.9112377762794495, + "learning_rate": 3.4417028364867205e-06, + "loss": 0.6113, + "step": 2793 + }, + { + "epoch": 2.3130866924351188, + "grad_norm": 0.9256125092506409, + "learning_rate": 3.4406840558625237e-06, + "loss": 0.6163, + "step": 2794 + }, + { + "epoch": 2.3139149641082275, + "grad_norm": 0.933461606502533, + "learning_rate": 3.4396650932247303e-06, + "loss": 0.6159, + "step": 2795 + }, + { + "epoch": 2.3147432357813362, + "grad_norm": 0.9031276702880859, + "learning_rate": 3.4386459487705016e-06, + "loss": 0.5981, + "step": 2796 + }, + { + "epoch": 2.315571507454445, + "grad_norm": 0.9465809464454651, + "learning_rate": 3.437626622697031e-06, + "loss": 0.6029, + "step": 2797 + }, + { + "epoch": 2.3163997791275537, + "grad_norm": 0.9535313248634338, + "learning_rate": 3.43660711520155e-06, + "loss": 0.5998, + "step": 2798 + }, + { + "epoch": 2.3172280508006624, + "grad_norm": 0.9048331379890442, + "learning_rate": 3.4355874264813228e-06, + "loss": 0.5991, + "step": 2799 + }, + { + "epoch": 2.3180563224737716, + "grad_norm": 0.9119572639465332, + "learning_rate": 3.4345675567336506e-06, + "loss": 0.6133, + "step": 2800 + }, + { + "epoch": 2.3188845941468803, + "grad_norm": 0.9217002391815186, + "learning_rate": 3.4335475061558676e-06, + "loss": 0.6012, + "step": 2801 + }, + { + "epoch": 2.319712865819989, + "grad_norm": 0.9501274824142456, + "learning_rate": 3.432527274945344e-06, + "loss": 0.5874, + "step": 2802 + }, + { + "epoch": 2.3205411374930978, + "grad_norm": 0.9126407504081726, + "learning_rate": 3.4315068632994864e-06, + "loss": 0.5985, + "step": 2803 + }, + { + "epoch": 2.3213694091662065, + "grad_norm": 0.9414592981338501, + "learning_rate": 3.430486271415734e-06, + "loss": 0.5989, + "step": 2804 + }, + { + "epoch": 2.322197680839315, + "grad_norm": 0.9485170841217041, + "learning_rate": 3.4294654994915623e-06, + "loss": 0.5795, + "step": 2805 + }, + { + "epoch": 2.323025952512424, + "grad_norm": 0.9311578273773193, + "learning_rate": 3.4284445477244804e-06, + "loss": 0.5606, + "step": 2806 + }, + { + "epoch": 2.3238542241855327, + "grad_norm": 0.9476608037948608, + "learning_rate": 3.4274234163120328e-06, + "loss": 0.593, + "step": 2807 + }, + { + "epoch": 2.3246824958586414, + "grad_norm": 0.9367009401321411, + "learning_rate": 3.4264021054518004e-06, + "loss": 0.6104, + "step": 2808 + }, + { + "epoch": 2.3255107675317506, + "grad_norm": 0.9663116335868835, + "learning_rate": 3.425380615341396e-06, + "loss": 0.6091, + "step": 2809 + }, + { + "epoch": 2.3263390392048593, + "grad_norm": 0.9176538586616516, + "learning_rate": 3.4243589461784687e-06, + "loss": 0.585, + "step": 2810 + }, + { + "epoch": 2.327167310877968, + "grad_norm": 0.9261791110038757, + "learning_rate": 3.423337098160703e-06, + "loss": 0.6189, + "step": 2811 + }, + { + "epoch": 2.3279955825510767, + "grad_norm": 0.9573442339897156, + "learning_rate": 3.422315071485815e-06, + "loss": 0.5925, + "step": 2812 + }, + { + "epoch": 2.3288238542241855, + "grad_norm": 0.9381136894226074, + "learning_rate": 3.4212928663515584e-06, + "loss": 0.588, + "step": 2813 + }, + { + "epoch": 2.329652125897294, + "grad_norm": 0.9276189804077148, + "learning_rate": 3.420270482955721e-06, + "loss": 0.5885, + "step": 2814 + }, + { + "epoch": 2.330480397570403, + "grad_norm": 0.9124462604522705, + "learning_rate": 3.419247921496123e-06, + "loss": 0.6004, + "step": 2815 + }, + { + "epoch": 2.331308669243512, + "grad_norm": 0.9146862626075745, + "learning_rate": 3.418225182170622e-06, + "loss": 0.6036, + "step": 2816 + }, + { + "epoch": 2.332136940916621, + "grad_norm": 0.9196732044219971, + "learning_rate": 3.4172022651771075e-06, + "loss": 0.6054, + "step": 2817 + }, + { + "epoch": 2.3329652125897296, + "grad_norm": 0.9340062737464905, + "learning_rate": 3.416179170713505e-06, + "loss": 0.6037, + "step": 2818 + }, + { + "epoch": 2.3337934842628383, + "grad_norm": 0.9004904627799988, + "learning_rate": 3.415155898977773e-06, + "loss": 0.6087, + "step": 2819 + }, + { + "epoch": 2.334621755935947, + "grad_norm": 0.8993189334869385, + "learning_rate": 3.414132450167906e-06, + "loss": 0.5972, + "step": 2820 + }, + { + "epoch": 2.3354500276090557, + "grad_norm": 0.9195947051048279, + "learning_rate": 3.413108824481931e-06, + "loss": 0.5784, + "step": 2821 + }, + { + "epoch": 2.3362782992821645, + "grad_norm": 0.9149767756462097, + "learning_rate": 3.412085022117911e-06, + "loss": 0.5838, + "step": 2822 + }, + { + "epoch": 2.337106570955273, + "grad_norm": 0.8880875110626221, + "learning_rate": 3.411061043273941e-06, + "loss": 0.6008, + "step": 2823 + }, + { + "epoch": 2.337934842628382, + "grad_norm": 0.9110493659973145, + "learning_rate": 3.4100368881481515e-06, + "loss": 0.6333, + "step": 2824 + }, + { + "epoch": 2.338763114301491, + "grad_norm": 0.9558914303779602, + "learning_rate": 3.409012556938708e-06, + "loss": 0.602, + "step": 2825 + }, + { + "epoch": 2.3395913859746, + "grad_norm": 0.9668992161750793, + "learning_rate": 3.4079880498438083e-06, + "loss": 0.5952, + "step": 2826 + }, + { + "epoch": 2.3404196576477085, + "grad_norm": 0.9352533221244812, + "learning_rate": 3.406963367061685e-06, + "loss": 0.6093, + "step": 2827 + }, + { + "epoch": 2.3412479293208173, + "grad_norm": 0.9072561264038086, + "learning_rate": 3.4059385087906055e-06, + "loss": 0.603, + "step": 2828 + }, + { + "epoch": 2.342076200993926, + "grad_norm": 0.9563443064689636, + "learning_rate": 3.4049134752288692e-06, + "loss": 0.585, + "step": 2829 + }, + { + "epoch": 2.3429044726670347, + "grad_norm": 0.9061548709869385, + "learning_rate": 3.4038882665748102e-06, + "loss": 0.6111, + "step": 2830 + }, + { + "epoch": 2.3437327443401434, + "grad_norm": 0.9177330136299133, + "learning_rate": 3.4028628830267986e-06, + "loss": 0.5872, + "step": 2831 + }, + { + "epoch": 2.344561016013252, + "grad_norm": 0.9479212164878845, + "learning_rate": 3.401837324783235e-06, + "loss": 0.5911, + "step": 2832 + }, + { + "epoch": 2.345389287686361, + "grad_norm": 0.9118002653121948, + "learning_rate": 3.4008115920425556e-06, + "loss": 0.6153, + "step": 2833 + }, + { + "epoch": 2.34621755935947, + "grad_norm": 0.9086247682571411, + "learning_rate": 3.399785685003231e-06, + "loss": 0.5913, + "step": 2834 + }, + { + "epoch": 2.347045831032579, + "grad_norm": 0.937494158744812, + "learning_rate": 3.398759603863763e-06, + "loss": 0.601, + "step": 2835 + }, + { + "epoch": 2.3478741027056875, + "grad_norm": 0.9359503388404846, + "learning_rate": 3.397733348822691e-06, + "loss": 0.5894, + "step": 2836 + }, + { + "epoch": 2.3487023743787963, + "grad_norm": 0.9384423494338989, + "learning_rate": 3.396706920078584e-06, + "loss": 0.5974, + "step": 2837 + }, + { + "epoch": 2.349530646051905, + "grad_norm": 0.9438448548316956, + "learning_rate": 3.3956803178300474e-06, + "loss": 0.6079, + "step": 2838 + }, + { + "epoch": 2.3503589177250137, + "grad_norm": 0.9250973463058472, + "learning_rate": 3.3946535422757187e-06, + "loss": 0.6011, + "step": 2839 + }, + { + "epoch": 2.3511871893981224, + "grad_norm": 0.9577994346618652, + "learning_rate": 3.3936265936142685e-06, + "loss": 0.6052, + "step": 2840 + }, + { + "epoch": 2.3520154610712316, + "grad_norm": 0.9307487607002258, + "learning_rate": 3.3925994720444033e-06, + "loss": 0.5925, + "step": 2841 + }, + { + "epoch": 2.3528437327443403, + "grad_norm": 0.9049618244171143, + "learning_rate": 3.3915721777648615e-06, + "loss": 0.6096, + "step": 2842 + }, + { + "epoch": 2.353672004417449, + "grad_norm": 0.928248405456543, + "learning_rate": 3.3905447109744137e-06, + "loss": 0.5938, + "step": 2843 + }, + { + "epoch": 2.354500276090558, + "grad_norm": 0.9107298254966736, + "learning_rate": 3.3895170718718664e-06, + "loss": 0.5947, + "step": 2844 + }, + { + "epoch": 2.3553285477636665, + "grad_norm": 0.8896966576576233, + "learning_rate": 3.388489260656057e-06, + "loss": 0.5827, + "step": 2845 + }, + { + "epoch": 2.3561568194367752, + "grad_norm": 0.9374940991401672, + "learning_rate": 3.3874612775258586e-06, + "loss": 0.6038, + "step": 2846 + }, + { + "epoch": 2.356985091109884, + "grad_norm": 0.9086610078811646, + "learning_rate": 3.3864331226801755e-06, + "loss": 0.5758, + "step": 2847 + }, + { + "epoch": 2.3578133627829927, + "grad_norm": 0.9486793875694275, + "learning_rate": 3.3854047963179464e-06, + "loss": 0.5656, + "step": 2848 + }, + { + "epoch": 2.3586416344561014, + "grad_norm": 0.9411890506744385, + "learning_rate": 3.3843762986381436e-06, + "loss": 0.605, + "step": 2849 + }, + { + "epoch": 2.3594699061292106, + "grad_norm": 0.9491676688194275, + "learning_rate": 3.3833476298397706e-06, + "loss": 0.5992, + "step": 2850 + }, + { + "epoch": 2.3602981778023193, + "grad_norm": 0.9281409382820129, + "learning_rate": 3.382318790121865e-06, + "loss": 0.5856, + "step": 2851 + }, + { + "epoch": 2.361126449475428, + "grad_norm": 0.9243955016136169, + "learning_rate": 3.381289779683499e-06, + "loss": 0.5966, + "step": 2852 + }, + { + "epoch": 2.3619547211485368, + "grad_norm": 0.8931578993797302, + "learning_rate": 3.3802605987237757e-06, + "loss": 0.6095, + "step": 2853 + }, + { + "epoch": 2.3627829928216455, + "grad_norm": 0.9200635552406311, + "learning_rate": 3.3792312474418324e-06, + "loss": 0.6014, + "step": 2854 + }, + { + "epoch": 2.3636112644947542, + "grad_norm": 0.9007267951965332, + "learning_rate": 3.378201726036839e-06, + "loss": 0.5964, + "step": 2855 + }, + { + "epoch": 2.364439536167863, + "grad_norm": 0.9420115947723389, + "learning_rate": 3.377172034707998e-06, + "loss": 0.6075, + "step": 2856 + }, + { + "epoch": 2.3652678078409717, + "grad_norm": 0.8935693502426147, + "learning_rate": 3.3761421736545453e-06, + "loss": 0.6184, + "step": 2857 + }, + { + "epoch": 2.3660960795140804, + "grad_norm": 0.9142755270004272, + "learning_rate": 3.3751121430757482e-06, + "loss": 0.6114, + "step": 2858 + }, + { + "epoch": 2.3669243511871896, + "grad_norm": 0.9280282855033875, + "learning_rate": 3.3740819431709103e-06, + "loss": 0.5922, + "step": 2859 + }, + { + "epoch": 2.3677526228602983, + "grad_norm": 0.9347119927406311, + "learning_rate": 3.3730515741393643e-06, + "loss": 0.5733, + "step": 2860 + }, + { + "epoch": 2.368580894533407, + "grad_norm": 0.9286894202232361, + "learning_rate": 3.3720210361804775e-06, + "loss": 0.597, + "step": 2861 + }, + { + "epoch": 2.3694091662065158, + "grad_norm": 0.9092636704444885, + "learning_rate": 3.3709903294936482e-06, + "loss": 0.5842, + "step": 2862 + }, + { + "epoch": 2.3702374378796245, + "grad_norm": 0.9205367565155029, + "learning_rate": 3.369959454278309e-06, + "loss": 0.6166, + "step": 2863 + }, + { + "epoch": 2.371065709552733, + "grad_norm": 0.9211122393608093, + "learning_rate": 3.3689284107339265e-06, + "loss": 0.5746, + "step": 2864 + }, + { + "epoch": 2.371893981225842, + "grad_norm": 0.908984899520874, + "learning_rate": 3.367897199059996e-06, + "loss": 0.5883, + "step": 2865 + }, + { + "epoch": 2.3727222528989507, + "grad_norm": 0.9270497560501099, + "learning_rate": 3.3668658194560477e-06, + "loss": 0.5985, + "step": 2866 + }, + { + "epoch": 2.3735505245720594, + "grad_norm": 0.9320128560066223, + "learning_rate": 3.3658342721216448e-06, + "loss": 0.5907, + "step": 2867 + }, + { + "epoch": 2.3743787962451686, + "grad_norm": 0.9428799152374268, + "learning_rate": 3.364802557256381e-06, + "loss": 0.6041, + "step": 2868 + }, + { + "epoch": 2.3752070679182773, + "grad_norm": 0.915873646736145, + "learning_rate": 3.363770675059883e-06, + "loss": 0.615, + "step": 2869 + }, + { + "epoch": 2.376035339591386, + "grad_norm": 0.9327977895736694, + "learning_rate": 3.362738625731813e-06, + "loss": 0.6147, + "step": 2870 + }, + { + "epoch": 2.3768636112644947, + "grad_norm": 0.9171217679977417, + "learning_rate": 3.3617064094718602e-06, + "loss": 0.6057, + "step": 2871 + }, + { + "epoch": 2.3776918829376035, + "grad_norm": 0.9192221760749817, + "learning_rate": 3.3606740264797504e-06, + "loss": 0.6092, + "step": 2872 + }, + { + "epoch": 2.378520154610712, + "grad_norm": 0.9024589657783508, + "learning_rate": 3.3596414769552394e-06, + "loss": 0.6047, + "step": 2873 + }, + { + "epoch": 2.379348426283821, + "grad_norm": 0.910923182964325, + "learning_rate": 3.3586087610981155e-06, + "loss": 0.5741, + "step": 2874 + }, + { + "epoch": 2.38017669795693, + "grad_norm": 0.929965078830719, + "learning_rate": 3.3575758791082014e-06, + "loss": 0.6114, + "step": 2875 + }, + { + "epoch": 2.381004969630039, + "grad_norm": 0.910308837890625, + "learning_rate": 3.356542831185348e-06, + "loss": 0.5938, + "step": 2876 + }, + { + "epoch": 2.3818332413031476, + "grad_norm": 0.9371251463890076, + "learning_rate": 3.355509617529441e-06, + "loss": 0.5883, + "step": 2877 + }, + { + "epoch": 2.3826615129762563, + "grad_norm": 0.9064118266105652, + "learning_rate": 3.3544762383403995e-06, + "loss": 0.5879, + "step": 2878 + }, + { + "epoch": 2.383489784649365, + "grad_norm": 0.954288899898529, + "learning_rate": 3.3534426938181704e-06, + "loss": 0.5981, + "step": 2879 + }, + { + "epoch": 2.3843180563224737, + "grad_norm": 0.9465821981430054, + "learning_rate": 3.3524089841627348e-06, + "loss": 0.6129, + "step": 2880 + }, + { + "epoch": 2.3851463279955825, + "grad_norm": 0.9406200051307678, + "learning_rate": 3.3513751095741084e-06, + "loss": 0.6035, + "step": 2881 + }, + { + "epoch": 2.385974599668691, + "grad_norm": 0.9627492427825928, + "learning_rate": 3.3503410702523337e-06, + "loss": 0.5701, + "step": 2882 + }, + { + "epoch": 2.3868028713418, + "grad_norm": 0.8899156451225281, + "learning_rate": 3.3493068663974892e-06, + "loss": 0.6055, + "step": 2883 + }, + { + "epoch": 2.387631143014909, + "grad_norm": 0.9304690957069397, + "learning_rate": 3.348272498209683e-06, + "loss": 0.6078, + "step": 2884 + }, + { + "epoch": 2.388459414688018, + "grad_norm": 0.9416152834892273, + "learning_rate": 3.3472379658890554e-06, + "loss": 0.609, + "step": 2885 + }, + { + "epoch": 2.3892876863611265, + "grad_norm": 0.9296873211860657, + "learning_rate": 3.34620326963578e-06, + "loss": 0.608, + "step": 2886 + }, + { + "epoch": 2.3901159580342353, + "grad_norm": 0.9163168668746948, + "learning_rate": 3.34516840965006e-06, + "loss": 0.5725, + "step": 2887 + }, + { + "epoch": 2.390944229707344, + "grad_norm": 0.9371008276939392, + "learning_rate": 3.344133386132131e-06, + "loss": 0.6223, + "step": 2888 + }, + { + "epoch": 2.3917725013804527, + "grad_norm": 0.9486922025680542, + "learning_rate": 3.3430981992822608e-06, + "loss": 0.5878, + "step": 2889 + }, + { + "epoch": 2.3926007730535614, + "grad_norm": 0.9124024510383606, + "learning_rate": 3.3420628493007483e-06, + "loss": 0.5816, + "step": 2890 + }, + { + "epoch": 2.39342904472667, + "grad_norm": 0.9193150401115417, + "learning_rate": 3.3410273363879236e-06, + "loss": 0.5909, + "step": 2891 + }, + { + "epoch": 2.394257316399779, + "grad_norm": 0.9287682175636292, + "learning_rate": 3.3399916607441502e-06, + "loss": 0.6095, + "step": 2892 + }, + { + "epoch": 2.395085588072888, + "grad_norm": 0.9130192995071411, + "learning_rate": 3.3389558225698204e-06, + "loss": 0.5782, + "step": 2893 + }, + { + "epoch": 2.395913859745997, + "grad_norm": 0.9145562052726746, + "learning_rate": 3.3379198220653587e-06, + "loss": 0.5919, + "step": 2894 + }, + { + "epoch": 2.3967421314191055, + "grad_norm": 0.9094864726066589, + "learning_rate": 3.3368836594312235e-06, + "loss": 0.6165, + "step": 2895 + }, + { + "epoch": 2.3975704030922143, + "grad_norm": 0.9446457028388977, + "learning_rate": 3.3358473348679016e-06, + "loss": 0.5967, + "step": 2896 + }, + { + "epoch": 2.398398674765323, + "grad_norm": 0.9345584511756897, + "learning_rate": 3.334810848575911e-06, + "loss": 0.6006, + "step": 2897 + }, + { + "epoch": 2.3992269464384317, + "grad_norm": 0.8865429759025574, + "learning_rate": 3.3337742007558043e-06, + "loss": 0.5766, + "step": 2898 + }, + { + "epoch": 2.4000552181115404, + "grad_norm": 0.9188973903656006, + "learning_rate": 3.3327373916081617e-06, + "loss": 0.6107, + "step": 2899 + }, + { + "epoch": 2.4008834897846496, + "grad_norm": 0.895247220993042, + "learning_rate": 3.3317004213335967e-06, + "loss": 0.6037, + "step": 2900 + }, + { + "epoch": 2.4017117614577583, + "grad_norm": 0.9131169319152832, + "learning_rate": 3.330663290132753e-06, + "loss": 0.5825, + "step": 2901 + }, + { + "epoch": 2.402540033130867, + "grad_norm": 0.9109821915626526, + "learning_rate": 3.329625998206306e-06, + "loss": 0.6194, + "step": 2902 + }, + { + "epoch": 2.403368304803976, + "grad_norm": 0.90902179479599, + "learning_rate": 3.3285885457549618e-06, + "loss": 0.5921, + "step": 2903 + }, + { + "epoch": 2.4041965764770845, + "grad_norm": 0.9419400095939636, + "learning_rate": 3.3275509329794585e-06, + "loss": 0.5973, + "step": 2904 + }, + { + "epoch": 2.4050248481501932, + "grad_norm": 0.905164897441864, + "learning_rate": 3.3265131600805636e-06, + "loss": 0.5828, + "step": 2905 + }, + { + "epoch": 2.405853119823302, + "grad_norm": 0.9232257008552551, + "learning_rate": 3.3254752272590775e-06, + "loss": 0.5972, + "step": 2906 + }, + { + "epoch": 2.4066813914964107, + "grad_norm": 0.9152960181236267, + "learning_rate": 3.3244371347158294e-06, + "loss": 0.5968, + "step": 2907 + }, + { + "epoch": 2.4075096631695194, + "grad_norm": 0.9337223172187805, + "learning_rate": 3.3233988826516807e-06, + "loss": 0.5907, + "step": 2908 + }, + { + "epoch": 2.4083379348426286, + "grad_norm": 0.9352282881736755, + "learning_rate": 3.322360471267524e-06, + "loss": 0.573, + "step": 2909 + }, + { + "epoch": 2.4091662065157373, + "grad_norm": 0.928322970867157, + "learning_rate": 3.321321900764282e-06, + "loss": 0.5836, + "step": 2910 + }, + { + "epoch": 2.409994478188846, + "grad_norm": 0.9335048794746399, + "learning_rate": 3.3202831713429084e-06, + "loss": 0.607, + "step": 2911 + }, + { + "epoch": 2.4108227498619548, + "grad_norm": 0.9128180742263794, + "learning_rate": 3.319244283204389e-06, + "loss": 0.5956, + "step": 2912 + }, + { + "epoch": 2.4116510215350635, + "grad_norm": 0.9457159042358398, + "learning_rate": 3.3182052365497364e-06, + "loss": 0.602, + "step": 2913 + }, + { + "epoch": 2.4124792932081722, + "grad_norm": 0.9508596658706665, + "learning_rate": 3.317166031579998e-06, + "loss": 0.5972, + "step": 2914 + }, + { + "epoch": 2.413307564881281, + "grad_norm": 0.9274892807006836, + "learning_rate": 3.31612666849625e-06, + "loss": 0.6015, + "step": 2915 + }, + { + "epoch": 2.4141358365543897, + "grad_norm": 0.9340214729309082, + "learning_rate": 3.3150871474995994e-06, + "loss": 0.5984, + "step": 2916 + }, + { + "epoch": 2.4149641082274984, + "grad_norm": 0.9376348853111267, + "learning_rate": 3.3140474687911844e-06, + "loss": 0.6205, + "step": 2917 + }, + { + "epoch": 2.4157923799006076, + "grad_norm": 0.9363590478897095, + "learning_rate": 3.313007632572173e-06, + "loss": 0.5926, + "step": 2918 + }, + { + "epoch": 2.4166206515737163, + "grad_norm": 0.8942010402679443, + "learning_rate": 3.3119676390437618e-06, + "loss": 0.5966, + "step": 2919 + }, + { + "epoch": 2.417448923246825, + "grad_norm": 0.9249299168586731, + "learning_rate": 3.310927488407183e-06, + "loss": 0.5951, + "step": 2920 + }, + { + "epoch": 2.4182771949199338, + "grad_norm": 0.9157326817512512, + "learning_rate": 3.3098871808636936e-06, + "loss": 0.5693, + "step": 2921 + }, + { + "epoch": 2.4191054665930425, + "grad_norm": 0.9088039994239807, + "learning_rate": 3.3088467166145843e-06, + "loss": 0.6084, + "step": 2922 + }, + { + "epoch": 2.419933738266151, + "grad_norm": 0.9258041381835938, + "learning_rate": 3.307806095861177e-06, + "loss": 0.6036, + "step": 2923 + }, + { + "epoch": 2.42076200993926, + "grad_norm": 0.905963659286499, + "learning_rate": 3.306765318804819e-06, + "loss": 0.5918, + "step": 2924 + }, + { + "epoch": 2.4215902816123687, + "grad_norm": 0.9063541293144226, + "learning_rate": 3.3057243856468924e-06, + "loss": 0.5906, + "step": 2925 + }, + { + "epoch": 2.4224185532854774, + "grad_norm": 0.9308085441589355, + "learning_rate": 3.304683296588808e-06, + "loss": 0.59, + "step": 2926 + }, + { + "epoch": 2.4232468249585866, + "grad_norm": 0.8989183902740479, + "learning_rate": 3.303642051832007e-06, + "loss": 0.5915, + "step": 2927 + }, + { + "epoch": 2.4240750966316953, + "grad_norm": 0.9413655996322632, + "learning_rate": 3.3026006515779614e-06, + "loss": 0.5952, + "step": 2928 + }, + { + "epoch": 2.424903368304804, + "grad_norm": 0.9431564211845398, + "learning_rate": 3.3015590960281708e-06, + "loss": 0.6007, + "step": 2929 + }, + { + "epoch": 2.4257316399779127, + "grad_norm": 0.8911551237106323, + "learning_rate": 3.3005173853841666e-06, + "loss": 0.573, + "step": 2930 + }, + { + "epoch": 2.4265599116510215, + "grad_norm": 0.9177538156509399, + "learning_rate": 3.299475519847511e-06, + "loss": 0.6116, + "step": 2931 + }, + { + "epoch": 2.42738818332413, + "grad_norm": 0.9592459797859192, + "learning_rate": 3.298433499619795e-06, + "loss": 0.6039, + "step": 2932 + }, + { + "epoch": 2.428216454997239, + "grad_norm": 0.9287276864051819, + "learning_rate": 3.2973913249026392e-06, + "loss": 0.5837, + "step": 2933 + }, + { + "epoch": 2.429044726670348, + "grad_norm": 0.9205804467201233, + "learning_rate": 3.296348995897696e-06, + "loss": 0.5904, + "step": 2934 + }, + { + "epoch": 2.429872998343457, + "grad_norm": 0.9075837731361389, + "learning_rate": 3.295306512806646e-06, + "loss": 0.5719, + "step": 2935 + }, + { + "epoch": 2.4307012700165656, + "grad_norm": 0.9203813672065735, + "learning_rate": 3.294263875831198e-06, + "loss": 0.604, + "step": 2936 + }, + { + "epoch": 2.4315295416896743, + "grad_norm": 0.8868334293365479, + "learning_rate": 3.2932210851730952e-06, + "loss": 0.6113, + "step": 2937 + }, + { + "epoch": 2.432357813362783, + "grad_norm": 0.9620448350906372, + "learning_rate": 3.2921781410341064e-06, + "loss": 0.5876, + "step": 2938 + }, + { + "epoch": 2.4331860850358917, + "grad_norm": 0.8991201519966125, + "learning_rate": 3.2911350436160317e-06, + "loss": 0.5994, + "step": 2939 + }, + { + "epoch": 2.4340143567090005, + "grad_norm": 0.9383120536804199, + "learning_rate": 3.2900917931207016e-06, + "loss": 0.6098, + "step": 2940 + }, + { + "epoch": 2.434842628382109, + "grad_norm": 0.9405770897865295, + "learning_rate": 3.289048389749974e-06, + "loss": 0.5978, + "step": 2941 + }, + { + "epoch": 2.435670900055218, + "grad_norm": 0.9153317213058472, + "learning_rate": 3.2880048337057387e-06, + "loss": 0.6091, + "step": 2942 + }, + { + "epoch": 2.436499171728327, + "grad_norm": 0.9396141171455383, + "learning_rate": 3.2869611251899136e-06, + "loss": 0.6143, + "step": 2943 + }, + { + "epoch": 2.437327443401436, + "grad_norm": 0.9370556473731995, + "learning_rate": 3.285917264404446e-06, + "loss": 0.61, + "step": 2944 + }, + { + "epoch": 2.4381557150745445, + "grad_norm": 0.9142118096351624, + "learning_rate": 3.2848732515513153e-06, + "loss": 0.5875, + "step": 2945 + }, + { + "epoch": 2.4389839867476533, + "grad_norm": 0.9196925759315491, + "learning_rate": 3.2838290868325267e-06, + "loss": 0.6058, + "step": 2946 + }, + { + "epoch": 2.439812258420762, + "grad_norm": 0.892217755317688, + "learning_rate": 3.282784770450116e-06, + "loss": 0.6005, + "step": 2947 + }, + { + "epoch": 2.4406405300938707, + "grad_norm": 0.9278191924095154, + "learning_rate": 3.2817403026061496e-06, + "loss": 0.6206, + "step": 2948 + }, + { + "epoch": 2.4414688017669794, + "grad_norm": 0.9343185424804688, + "learning_rate": 3.2806956835027216e-06, + "loss": 0.604, + "step": 2949 + }, + { + "epoch": 2.442297073440088, + "grad_norm": 0.9401823878288269, + "learning_rate": 3.2796509133419564e-06, + "loss": 0.6039, + "step": 2950 + }, + { + "epoch": 2.443125345113197, + "grad_norm": 0.9102358818054199, + "learning_rate": 3.278605992326008e-06, + "loss": 0.5892, + "step": 2951 + }, + { + "epoch": 2.443953616786306, + "grad_norm": 0.9241907000541687, + "learning_rate": 3.2775609206570573e-06, + "loss": 0.6054, + "step": 2952 + }, + { + "epoch": 2.444781888459415, + "grad_norm": 0.9492688775062561, + "learning_rate": 3.276515698537317e-06, + "loss": 0.5922, + "step": 2953 + }, + { + "epoch": 2.4456101601325235, + "grad_norm": 0.9492383003234863, + "learning_rate": 3.275470326169028e-06, + "loss": 0.5735, + "step": 2954 + }, + { + "epoch": 2.4464384318056323, + "grad_norm": 0.8857248425483704, + "learning_rate": 3.2744248037544597e-06, + "loss": 0.584, + "step": 2955 + }, + { + "epoch": 2.447266703478741, + "grad_norm": 0.9185032248497009, + "learning_rate": 3.2733791314959108e-06, + "loss": 0.5906, + "step": 2956 + }, + { + "epoch": 2.4480949751518497, + "grad_norm": 0.9257567524909973, + "learning_rate": 3.2723333095957094e-06, + "loss": 0.587, + "step": 2957 + }, + { + "epoch": 2.4489232468249584, + "grad_norm": 0.9427464008331299, + "learning_rate": 3.2712873382562117e-06, + "loss": 0.5932, + "step": 2958 + }, + { + "epoch": 2.4497515184980676, + "grad_norm": 0.9002125263214111, + "learning_rate": 3.2702412176798045e-06, + "loss": 0.58, + "step": 2959 + }, + { + "epoch": 2.4505797901711763, + "grad_norm": 0.906929075717926, + "learning_rate": 3.2691949480689018e-06, + "loss": 0.5878, + "step": 2960 + }, + { + "epoch": 2.451408061844285, + "grad_norm": 0.9162207841873169, + "learning_rate": 3.268148529625946e-06, + "loss": 0.5975, + "step": 2961 + }, + { + "epoch": 2.452236333517394, + "grad_norm": 0.9268636107444763, + "learning_rate": 3.2671019625534117e-06, + "loss": 0.6028, + "step": 2962 + }, + { + "epoch": 2.4530646051905025, + "grad_norm": 0.9216850996017456, + "learning_rate": 3.2660552470537975e-06, + "loss": 0.5938, + "step": 2963 + }, + { + "epoch": 2.4538928768636112, + "grad_norm": 0.9353698492050171, + "learning_rate": 3.2650083833296344e-06, + "loss": 0.6069, + "step": 2964 + }, + { + "epoch": 2.45472114853672, + "grad_norm": 0.9393898248672485, + "learning_rate": 3.2639613715834805e-06, + "loss": 0.5874, + "step": 2965 + }, + { + "epoch": 2.4555494202098287, + "grad_norm": 0.9410581588745117, + "learning_rate": 3.2629142120179224e-06, + "loss": 0.5918, + "step": 2966 + }, + { + "epoch": 2.4563776918829374, + "grad_norm": 0.9445161819458008, + "learning_rate": 3.2618669048355763e-06, + "loss": 0.5743, + "step": 2967 + }, + { + "epoch": 2.4572059635560466, + "grad_norm": 0.9591469764709473, + "learning_rate": 3.2608194502390867e-06, + "loss": 0.5781, + "step": 2968 + }, + { + "epoch": 2.4580342352291553, + "grad_norm": 0.9237606525421143, + "learning_rate": 3.2597718484311246e-06, + "loss": 0.6163, + "step": 2969 + }, + { + "epoch": 2.458862506902264, + "grad_norm": 0.9216769337654114, + "learning_rate": 3.2587240996143933e-06, + "loss": 0.6157, + "step": 2970 + }, + { + "epoch": 2.4596907785753728, + "grad_norm": 0.9081043004989624, + "learning_rate": 3.257676203991621e-06, + "loss": 0.5821, + "step": 2971 + }, + { + "epoch": 2.4605190502484815, + "grad_norm": 0.9316040873527527, + "learning_rate": 3.256628161765566e-06, + "loss": 0.5911, + "step": 2972 + }, + { + "epoch": 2.4613473219215902, + "grad_norm": 0.9326239824295044, + "learning_rate": 3.2555799731390158e-06, + "loss": 0.5963, + "step": 2973 + }, + { + "epoch": 2.462175593594699, + "grad_norm": 0.9429934024810791, + "learning_rate": 3.254531638314784e-06, + "loss": 0.5651, + "step": 2974 + }, + { + "epoch": 2.4630038652678077, + "grad_norm": 0.9090480804443359, + "learning_rate": 3.2534831574957134e-06, + "loss": 0.5779, + "step": 2975 + }, + { + "epoch": 2.4638321369409164, + "grad_norm": 0.9582802653312683, + "learning_rate": 3.2524345308846756e-06, + "loss": 0.5887, + "step": 2976 + }, + { + "epoch": 2.4646604086140256, + "grad_norm": 0.9248860478401184, + "learning_rate": 3.2513857586845705e-06, + "loss": 0.5936, + "step": 2977 + }, + { + "epoch": 2.4654886802871343, + "grad_norm": 0.9345134496688843, + "learning_rate": 3.250336841098325e-06, + "loss": 0.5985, + "step": 2978 + }, + { + "epoch": 2.466316951960243, + "grad_norm": 0.9092442393302917, + "learning_rate": 3.249287778328896e-06, + "loss": 0.5926, + "step": 2979 + }, + { + "epoch": 2.4671452236333518, + "grad_norm": 0.932470977306366, + "learning_rate": 3.2482385705792655e-06, + "loss": 0.5797, + "step": 2980 + }, + { + "epoch": 2.4679734953064605, + "grad_norm": 0.9335013031959534, + "learning_rate": 3.247189218052448e-06, + "loss": 0.5933, + "step": 2981 + }, + { + "epoch": 2.468801766979569, + "grad_norm": 0.9283298254013062, + "learning_rate": 3.2461397209514812e-06, + "loss": 0.6003, + "step": 2982 + }, + { + "epoch": 2.469630038652678, + "grad_norm": 0.9429066181182861, + "learning_rate": 3.245090079479434e-06, + "loss": 0.5944, + "step": 2983 + }, + { + "epoch": 2.4704583103257867, + "grad_norm": 0.9284510612487793, + "learning_rate": 3.2440402938394017e-06, + "loss": 0.5782, + "step": 2984 + }, + { + "epoch": 2.4712865819988954, + "grad_norm": 0.9150790572166443, + "learning_rate": 3.2429903642345094e-06, + "loss": 0.5862, + "step": 2985 + }, + { + "epoch": 2.4721148536720046, + "grad_norm": 0.9177762866020203, + "learning_rate": 3.2419402908679065e-06, + "loss": 0.5931, + "step": 2986 + }, + { + "epoch": 2.4729431253451133, + "grad_norm": 0.9098034501075745, + "learning_rate": 3.2408900739427744e-06, + "loss": 0.5864, + "step": 2987 + }, + { + "epoch": 2.473771397018222, + "grad_norm": 0.9237478971481323, + "learning_rate": 3.2398397136623195e-06, + "loss": 0.5873, + "step": 2988 + }, + { + "epoch": 2.4745996686913307, + "grad_norm": 0.9212510585784912, + "learning_rate": 3.2387892102297757e-06, + "loss": 0.5884, + "step": 2989 + }, + { + "epoch": 2.4754279403644395, + "grad_norm": 0.8987523913383484, + "learning_rate": 3.237738563848408e-06, + "loss": 0.5806, + "step": 2990 + }, + { + "epoch": 2.476256212037548, + "grad_norm": 0.9036835432052612, + "learning_rate": 3.236687774721504e-06, + "loss": 0.5993, + "step": 2991 + }, + { + "epoch": 2.477084483710657, + "grad_norm": 0.9291606545448303, + "learning_rate": 3.235636843052384e-06, + "loss": 0.5784, + "step": 2992 + }, + { + "epoch": 2.477912755383766, + "grad_norm": 0.9339813590049744, + "learning_rate": 3.234585769044392e-06, + "loss": 0.5951, + "step": 2993 + }, + { + "epoch": 2.478741027056875, + "grad_norm": 0.9638957977294922, + "learning_rate": 3.2335345529009017e-06, + "loss": 0.591, + "step": 2994 + }, + { + "epoch": 2.4795692987299836, + "grad_norm": 0.9618088603019714, + "learning_rate": 3.2324831948253123e-06, + "loss": 0.6102, + "step": 2995 + }, + { + "epoch": 2.4803975704030923, + "grad_norm": 0.9491007924079895, + "learning_rate": 3.2314316950210547e-06, + "loss": 0.599, + "step": 2996 + }, + { + "epoch": 2.481225842076201, + "grad_norm": 0.9213217496871948, + "learning_rate": 3.2303800536915804e-06, + "loss": 0.5955, + "step": 2997 + }, + { + "epoch": 2.4820541137493097, + "grad_norm": 0.923991322517395, + "learning_rate": 3.229328271040375e-06, + "loss": 0.5903, + "step": 2998 + }, + { + "epoch": 2.4828823854224185, + "grad_norm": 0.9173145890235901, + "learning_rate": 3.2282763472709484e-06, + "loss": 0.5899, + "step": 2999 + }, + { + "epoch": 2.483710657095527, + "grad_norm": 0.9357689023017883, + "learning_rate": 3.2272242825868363e-06, + "loss": 0.5885, + "step": 3000 + }, + { + "epoch": 2.484538928768636, + "grad_norm": 0.8902420401573181, + "learning_rate": 3.2261720771916056e-06, + "loss": 0.5899, + "step": 3001 + }, + { + "epoch": 2.485367200441745, + "grad_norm": 0.9079567790031433, + "learning_rate": 3.2251197312888475e-06, + "loss": 0.5966, + "step": 3002 + }, + { + "epoch": 2.486195472114854, + "grad_norm": 0.9068807363510132, + "learning_rate": 3.22406724508218e-06, + "loss": 0.6202, + "step": 3003 + }, + { + "epoch": 2.4870237437879625, + "grad_norm": 0.9458021521568298, + "learning_rate": 3.223014618775251e-06, + "loss": 0.5812, + "step": 3004 + }, + { + "epoch": 2.4878520154610713, + "grad_norm": 0.9506248235702515, + "learning_rate": 3.2219618525717328e-06, + "loss": 0.5819, + "step": 3005 + }, + { + "epoch": 2.48868028713418, + "grad_norm": 0.9573289155960083, + "learning_rate": 3.220908946675326e-06, + "loss": 0.581, + "step": 3006 + }, + { + "epoch": 2.4895085588072887, + "grad_norm": 0.9035286903381348, + "learning_rate": 3.219855901289758e-06, + "loss": 0.6182, + "step": 3007 + }, + { + "epoch": 2.4903368304803974, + "grad_norm": 0.9300901293754578, + "learning_rate": 3.218802716618784e-06, + "loss": 0.6079, + "step": 3008 + }, + { + "epoch": 2.491165102153506, + "grad_norm": 0.9116837382316589, + "learning_rate": 3.2177493928661854e-06, + "loss": 0.6034, + "step": 3009 + }, + { + "epoch": 2.491993373826615, + "grad_norm": 0.9291449189186096, + "learning_rate": 3.2166959302357694e-06, + "loss": 0.5889, + "step": 3010 + }, + { + "epoch": 2.492821645499724, + "grad_norm": 0.9002729654312134, + "learning_rate": 3.215642328931371e-06, + "loss": 0.5969, + "step": 3011 + }, + { + "epoch": 2.493649917172833, + "grad_norm": 0.8961556553840637, + "learning_rate": 3.214588589156854e-06, + "loss": 0.5899, + "step": 3012 + }, + { + "epoch": 2.4944781888459415, + "grad_norm": 0.9115450382232666, + "learning_rate": 3.2135347111161075e-06, + "loss": 0.5727, + "step": 3013 + }, + { + "epoch": 2.4953064605190503, + "grad_norm": 0.9060878157615662, + "learning_rate": 3.212480695013044e-06, + "loss": 0.5765, + "step": 3014 + }, + { + "epoch": 2.496134732192159, + "grad_norm": 0.902313768863678, + "learning_rate": 3.2114265410516088e-06, + "loss": 0.5913, + "step": 3015 + }, + { + "epoch": 2.4969630038652677, + "grad_norm": 0.931974470615387, + "learning_rate": 3.210372249435769e-06, + "loss": 0.5891, + "step": 3016 + }, + { + "epoch": 2.4977912755383764, + "grad_norm": 1.1711225509643555, + "learning_rate": 3.209317820369521e-06, + "loss": 0.6246, + "step": 3017 + }, + { + "epoch": 2.4986195472114856, + "grad_norm": 0.9561757445335388, + "learning_rate": 3.208263254056887e-06, + "loss": 0.5847, + "step": 3018 + }, + { + "epoch": 2.4994478188845943, + "grad_norm": 0.9621199369430542, + "learning_rate": 3.207208550701916e-06, + "loss": 0.5971, + "step": 3019 + }, + { + "epoch": 2.500276090557703, + "grad_norm": 0.9152470827102661, + "learning_rate": 3.206153710508683e-06, + "loss": 0.59, + "step": 3020 + }, + { + "epoch": 2.501104362230812, + "grad_norm": 0.9315138459205627, + "learning_rate": 3.20509873368129e-06, + "loss": 0.5841, + "step": 3021 + }, + { + "epoch": 2.5019326339039205, + "grad_norm": 0.9124136567115784, + "learning_rate": 3.204043620423864e-06, + "loss": 0.6007, + "step": 3022 + }, + { + "epoch": 2.5027609055770292, + "grad_norm": 0.8976690173149109, + "learning_rate": 3.2029883709405607e-06, + "loss": 0.5969, + "step": 3023 + }, + { + "epoch": 2.503589177250138, + "grad_norm": 0.8946467041969299, + "learning_rate": 3.201932985435561e-06, + "loss": 0.6069, + "step": 3024 + }, + { + "epoch": 2.5044174489232467, + "grad_norm": 0.9290626049041748, + "learning_rate": 3.2008774641130722e-06, + "loss": 0.591, + "step": 3025 + }, + { + "epoch": 2.5052457205963554, + "grad_norm": 0.8982367515563965, + "learning_rate": 3.1998218071773275e-06, + "loss": 0.5874, + "step": 3026 + }, + { + "epoch": 2.5060739922694646, + "grad_norm": 0.9062193036079407, + "learning_rate": 3.1987660148325867e-06, + "loss": 0.6107, + "step": 3027 + }, + { + "epoch": 2.5069022639425733, + "grad_norm": 0.9240522980690002, + "learning_rate": 3.1977100872831353e-06, + "loss": 0.5806, + "step": 3028 + }, + { + "epoch": 2.507730535615682, + "grad_norm": 0.9382046461105347, + "learning_rate": 3.1966540247332865e-06, + "loss": 0.5944, + "step": 3029 + }, + { + "epoch": 2.5085588072887908, + "grad_norm": 0.9194355010986328, + "learning_rate": 3.1955978273873776e-06, + "loss": 0.6085, + "step": 3030 + }, + { + "epoch": 2.5093870789618995, + "grad_norm": 0.9219276309013367, + "learning_rate": 3.1945414954497744e-06, + "loss": 0.5829, + "step": 3031 + }, + { + "epoch": 2.5102153506350082, + "grad_norm": 0.8998081088066101, + "learning_rate": 3.193485029124865e-06, + "loss": 0.6095, + "step": 3032 + }, + { + "epoch": 2.511043622308117, + "grad_norm": 0.9184247851371765, + "learning_rate": 3.192428428617068e-06, + "loss": 0.6129, + "step": 3033 + }, + { + "epoch": 2.5118718939812257, + "grad_norm": 0.9389240741729736, + "learning_rate": 3.1913716941308243e-06, + "loss": 0.607, + "step": 3034 + }, + { + "epoch": 2.5127001656543344, + "grad_norm": 0.9263836741447449, + "learning_rate": 3.190314825870603e-06, + "loss": 0.5761, + "step": 3035 + }, + { + "epoch": 2.5135284373274436, + "grad_norm": 0.9472658038139343, + "learning_rate": 3.1892578240408977e-06, + "loss": 0.5906, + "step": 3036 + }, + { + "epoch": 2.5143567090005523, + "grad_norm": 0.9045345783233643, + "learning_rate": 3.1882006888462285e-06, + "loss": 0.5993, + "step": 3037 + }, + { + "epoch": 2.515184980673661, + "grad_norm": 0.8930817246437073, + "learning_rate": 3.187143420491142e-06, + "loss": 0.5947, + "step": 3038 + }, + { + "epoch": 2.5160132523467698, + "grad_norm": 0.9130663871765137, + "learning_rate": 3.186086019180208e-06, + "loss": 0.5892, + "step": 3039 + }, + { + "epoch": 2.5168415240198785, + "grad_norm": 0.9008771777153015, + "learning_rate": 3.185028485118026e-06, + "loss": 0.5787, + "step": 3040 + }, + { + "epoch": 2.517669795692987, + "grad_norm": 0.9192292094230652, + "learning_rate": 3.1839708185092177e-06, + "loss": 0.5848, + "step": 3041 + }, + { + "epoch": 2.518498067366096, + "grad_norm": 0.8970836997032166, + "learning_rate": 3.1829130195584314e-06, + "loss": 0.5939, + "step": 3042 + }, + { + "epoch": 2.519326339039205, + "grad_norm": 0.9393385052680969, + "learning_rate": 3.181855088470342e-06, + "loss": 0.6096, + "step": 3043 + }, + { + "epoch": 2.5201546107123134, + "grad_norm": 0.9437024593353271, + "learning_rate": 3.180797025449649e-06, + "loss": 0.5998, + "step": 3044 + }, + { + "epoch": 2.5209828823854226, + "grad_norm": 0.9014863967895508, + "learning_rate": 3.179738830701078e-06, + "loss": 0.6015, + "step": 3045 + }, + { + "epoch": 2.5218111540585313, + "grad_norm": 0.9243669509887695, + "learning_rate": 3.1786805044293802e-06, + "loss": 0.5971, + "step": 3046 + }, + { + "epoch": 2.52263942573164, + "grad_norm": 0.9175970554351807, + "learning_rate": 3.1776220468393305e-06, + "loss": 0.6084, + "step": 3047 + }, + { + "epoch": 2.5234676974047487, + "grad_norm": 0.8962888121604919, + "learning_rate": 3.176563458135732e-06, + "loss": 0.5979, + "step": 3048 + }, + { + "epoch": 2.5242959690778575, + "grad_norm": 0.9321010708808899, + "learning_rate": 3.1755047385234117e-06, + "loss": 0.5941, + "step": 3049 + }, + { + "epoch": 2.525124240750966, + "grad_norm": 0.9147899746894836, + "learning_rate": 3.17444588820722e-06, + "loss": 0.5987, + "step": 3050 + }, + { + "epoch": 2.525952512424075, + "grad_norm": 0.9318193793296814, + "learning_rate": 3.173386907392037e-06, + "loss": 0.5658, + "step": 3051 + }, + { + "epoch": 2.526780784097184, + "grad_norm": 0.9039092659950256, + "learning_rate": 3.1723277962827647e-06, + "loss": 0.5956, + "step": 3052 + }, + { + "epoch": 2.5276090557702924, + "grad_norm": 0.9080818295478821, + "learning_rate": 3.1712685550843314e-06, + "loss": 0.5957, + "step": 3053 + }, + { + "epoch": 2.5284373274434015, + "grad_norm": 0.9120905995368958, + "learning_rate": 3.17020918400169e-06, + "loss": 0.5917, + "step": 3054 + }, + { + "epoch": 2.5292655991165103, + "grad_norm": 0.8983855843544006, + "learning_rate": 3.169149683239819e-06, + "loss": 0.6045, + "step": 3055 + }, + { + "epoch": 2.530093870789619, + "grad_norm": 0.9203280210494995, + "learning_rate": 3.168090053003721e-06, + "loss": 0.6093, + "step": 3056 + }, + { + "epoch": 2.5309221424627277, + "grad_norm": 0.9343932867050171, + "learning_rate": 3.1670302934984272e-06, + "loss": 0.5654, + "step": 3057 + }, + { + "epoch": 2.5317504141358365, + "grad_norm": 0.9054449200630188, + "learning_rate": 3.1659704049289885e-06, + "loss": 0.5878, + "step": 3058 + }, + { + "epoch": 2.532578685808945, + "grad_norm": 0.9011321067810059, + "learning_rate": 3.1649103875004856e-06, + "loss": 0.5831, + "step": 3059 + }, + { + "epoch": 2.533406957482054, + "grad_norm": 0.8886737823486328, + "learning_rate": 3.1638502414180206e-06, + "loss": 0.5915, + "step": 3060 + }, + { + "epoch": 2.534235229155163, + "grad_norm": 0.9289557337760925, + "learning_rate": 3.1627899668867214e-06, + "loss": 0.5944, + "step": 3061 + }, + { + "epoch": 2.535063500828272, + "grad_norm": 0.9269216060638428, + "learning_rate": 3.1617295641117425e-06, + "loss": 0.605, + "step": 3062 + }, + { + "epoch": 2.5358917725013805, + "grad_norm": 0.8877415657043457, + "learning_rate": 3.1606690332982613e-06, + "loss": 0.5954, + "step": 3063 + }, + { + "epoch": 2.5367200441744893, + "grad_norm": 0.9080086350440979, + "learning_rate": 3.1596083746514817e-06, + "loss": 0.5896, + "step": 3064 + }, + { + "epoch": 2.537548315847598, + "grad_norm": 0.9625695943832397, + "learning_rate": 3.15854758837663e-06, + "loss": 0.5666, + "step": 3065 + }, + { + "epoch": 2.5383765875207067, + "grad_norm": 0.9442515969276428, + "learning_rate": 3.1574866746789583e-06, + "loss": 0.5837, + "step": 3066 + }, + { + "epoch": 2.5392048591938154, + "grad_norm": 0.9054381847381592, + "learning_rate": 3.1564256337637443e-06, + "loss": 0.5799, + "step": 3067 + }, + { + "epoch": 2.5400331308669246, + "grad_norm": 0.9435595273971558, + "learning_rate": 3.1553644658362897e-06, + "loss": 0.6135, + "step": 3068 + }, + { + "epoch": 2.540861402540033, + "grad_norm": 0.9356298446655273, + "learning_rate": 3.1543031711019203e-06, + "loss": 0.5853, + "step": 3069 + }, + { + "epoch": 2.541689674213142, + "grad_norm": 0.917759895324707, + "learning_rate": 3.153241749765986e-06, + "loss": 0.5902, + "step": 3070 + }, + { + "epoch": 2.542517945886251, + "grad_norm": 0.9233857989311218, + "learning_rate": 3.1521802020338627e-06, + "loss": 0.6066, + "step": 3071 + }, + { + "epoch": 2.5433462175593595, + "grad_norm": 0.9497340321540833, + "learning_rate": 3.1511185281109504e-06, + "loss": 0.6003, + "step": 3072 + }, + { + "epoch": 2.5441744892324683, + "grad_norm": 0.9373629689216614, + "learning_rate": 3.1500567282026716e-06, + "loss": 0.5827, + "step": 3073 + }, + { + "epoch": 2.545002760905577, + "grad_norm": 0.8916833996772766, + "learning_rate": 3.148994802514476e-06, + "loss": 0.5996, + "step": 3074 + }, + { + "epoch": 2.5458310325786857, + "grad_norm": 0.9327612519264221, + "learning_rate": 3.1479327512518363e-06, + "loss": 0.6159, + "step": 3075 + }, + { + "epoch": 2.5466593042517944, + "grad_norm": 0.9380506873130798, + "learning_rate": 3.1468705746202493e-06, + "loss": 0.5924, + "step": 3076 + }, + { + "epoch": 2.5474875759249036, + "grad_norm": 0.9138150215148926, + "learning_rate": 3.1458082728252358e-06, + "loss": 0.5734, + "step": 3077 + }, + { + "epoch": 2.548315847598012, + "grad_norm": 0.8954431414604187, + "learning_rate": 3.144745846072341e-06, + "loss": 0.5902, + "step": 3078 + }, + { + "epoch": 2.549144119271121, + "grad_norm": 0.9259476661682129, + "learning_rate": 3.1436832945671362e-06, + "loss": 0.6057, + "step": 3079 + }, + { + "epoch": 2.54997239094423, + "grad_norm": 0.9295196533203125, + "learning_rate": 3.142620618515214e-06, + "loss": 0.5971, + "step": 3080 + }, + { + "epoch": 2.5508006626173385, + "grad_norm": 0.9318671822547913, + "learning_rate": 3.1415578181221928e-06, + "loss": 0.5852, + "step": 3081 + }, + { + "epoch": 2.5516289342904472, + "grad_norm": 0.9349870085716248, + "learning_rate": 3.1404948935937145e-06, + "loss": 0.5736, + "step": 3082 + }, + { + "epoch": 2.552457205963556, + "grad_norm": 0.9280831217765808, + "learning_rate": 3.1394318451354446e-06, + "loss": 0.5898, + "step": 3083 + }, + { + "epoch": 2.5532854776366647, + "grad_norm": 0.8969367742538452, + "learning_rate": 3.1383686729530727e-06, + "loss": 0.5983, + "step": 3084 + }, + { + "epoch": 2.5541137493097734, + "grad_norm": 0.9363370537757874, + "learning_rate": 3.1373053772523145e-06, + "loss": 0.6052, + "step": 3085 + }, + { + "epoch": 2.5549420209828826, + "grad_norm": 0.8944993615150452, + "learning_rate": 3.1362419582389057e-06, + "loss": 0.599, + "step": 3086 + }, + { + "epoch": 2.5557702926559913, + "grad_norm": 0.891412615776062, + "learning_rate": 3.1351784161186093e-06, + "loss": 0.5916, + "step": 3087 + }, + { + "epoch": 2.5565985643291, + "grad_norm": 0.930220365524292, + "learning_rate": 3.134114751097211e-06, + "loss": 0.5826, + "step": 3088 + }, + { + "epoch": 2.5574268360022088, + "grad_norm": 0.9193615317344666, + "learning_rate": 3.1330509633805183e-06, + "loss": 0.5896, + "step": 3089 + }, + { + "epoch": 2.5582551076753175, + "grad_norm": 0.9178160429000854, + "learning_rate": 3.131987053174367e-06, + "loss": 0.6104, + "step": 3090 + }, + { + "epoch": 2.5590833793484262, + "grad_norm": 0.9119648933410645, + "learning_rate": 3.130923020684611e-06, + "loss": 0.6198, + "step": 3091 + }, + { + "epoch": 2.559911651021535, + "grad_norm": 0.957272469997406, + "learning_rate": 3.1298588661171323e-06, + "loss": 0.5716, + "step": 3092 + }, + { + "epoch": 2.5607399226946437, + "grad_norm": 0.9073137044906616, + "learning_rate": 3.128794589677835e-06, + "loss": 0.5959, + "step": 3093 + }, + { + "epoch": 2.5615681943677524, + "grad_norm": 0.9372225403785706, + "learning_rate": 3.1277301915726456e-06, + "loss": 0.6146, + "step": 3094 + }, + { + "epoch": 2.5623964660408616, + "grad_norm": 0.9377845525741577, + "learning_rate": 3.126665672007515e-06, + "loss": 0.6055, + "step": 3095 + }, + { + "epoch": 2.5632247377139703, + "grad_norm": 0.9114481210708618, + "learning_rate": 3.1256010311884193e-06, + "loss": 0.6038, + "step": 3096 + }, + { + "epoch": 2.564053009387079, + "grad_norm": 0.9000943303108215, + "learning_rate": 3.124536269321356e-06, + "loss": 0.5806, + "step": 3097 + }, + { + "epoch": 2.5648812810601878, + "grad_norm": 0.9771347641944885, + "learning_rate": 3.123471386612347e-06, + "loss": 0.6029, + "step": 3098 + }, + { + "epoch": 2.5657095527332965, + "grad_norm": 0.9150393605232239, + "learning_rate": 3.1224063832674358e-06, + "loss": 0.6033, + "step": 3099 + }, + { + "epoch": 2.566537824406405, + "grad_norm": 0.8910076022148132, + "learning_rate": 3.1213412594926906e-06, + "loss": 0.6029, + "step": 3100 + }, + { + "epoch": 2.567366096079514, + "grad_norm": 0.9558795094490051, + "learning_rate": 3.1202760154942047e-06, + "loss": 0.5792, + "step": 3101 + }, + { + "epoch": 2.568194367752623, + "grad_norm": 0.958629310131073, + "learning_rate": 3.1192106514780924e-06, + "loss": 0.5933, + "step": 3102 + }, + { + "epoch": 2.5690226394257314, + "grad_norm": 0.9466328620910645, + "learning_rate": 3.1181451676504907e-06, + "loss": 0.6027, + "step": 3103 + }, + { + "epoch": 2.5698509110988406, + "grad_norm": 0.9397157430648804, + "learning_rate": 3.1170795642175614e-06, + "loss": 0.5894, + "step": 3104 + }, + { + "epoch": 2.5706791827719493, + "grad_norm": 0.9229122996330261, + "learning_rate": 3.116013841385489e-06, + "loss": 0.5812, + "step": 3105 + }, + { + "epoch": 2.571507454445058, + "grad_norm": 0.9332060217857361, + "learning_rate": 3.1149479993604805e-06, + "loss": 0.5738, + "step": 3106 + }, + { + "epoch": 2.5723357261181667, + "grad_norm": 0.9117311239242554, + "learning_rate": 3.113882038348766e-06, + "loss": 0.5923, + "step": 3107 + }, + { + "epoch": 2.5731639977912755, + "grad_norm": 0.9390400052070618, + "learning_rate": 3.1128159585566003e-06, + "loss": 0.6029, + "step": 3108 + }, + { + "epoch": 2.573992269464384, + "grad_norm": 0.9229103326797485, + "learning_rate": 3.1117497601902596e-06, + "loss": 0.6118, + "step": 3109 + }, + { + "epoch": 2.574820541137493, + "grad_norm": 0.8975940346717834, + "learning_rate": 3.1106834434560424e-06, + "loss": 0.6005, + "step": 3110 + }, + { + "epoch": 2.575648812810602, + "grad_norm": 0.9088030457496643, + "learning_rate": 3.1096170085602718e-06, + "loss": 0.5896, + "step": 3111 + }, + { + "epoch": 2.576477084483711, + "grad_norm": 0.982429027557373, + "learning_rate": 3.108550455709292e-06, + "loss": 0.5947, + "step": 3112 + }, + { + "epoch": 2.5773053561568195, + "grad_norm": 0.9404535293579102, + "learning_rate": 3.107483785109473e-06, + "loss": 0.6154, + "step": 3113 + }, + { + "epoch": 2.5781336278299283, + "grad_norm": 0.9074321389198303, + "learning_rate": 3.1064169969672043e-06, + "loss": 0.5889, + "step": 3114 + }, + { + "epoch": 2.578961899503037, + "grad_norm": 0.9000370502471924, + "learning_rate": 3.105350091488899e-06, + "loss": 0.5725, + "step": 3115 + }, + { + "epoch": 2.5797901711761457, + "grad_norm": 0.9296747446060181, + "learning_rate": 3.1042830688809944e-06, + "loss": 0.5983, + "step": 3116 + }, + { + "epoch": 2.5806184428492545, + "grad_norm": 0.9164032936096191, + "learning_rate": 3.1032159293499486e-06, + "loss": 0.5832, + "step": 3117 + }, + { + "epoch": 2.581446714522363, + "grad_norm": 0.9070228338241577, + "learning_rate": 3.1021486731022443e-06, + "loss": 0.5977, + "step": 3118 + }, + { + "epoch": 2.582274986195472, + "grad_norm": 0.9384707808494568, + "learning_rate": 3.1010813003443847e-06, + "loss": 0.5926, + "step": 3119 + }, + { + "epoch": 2.583103257868581, + "grad_norm": 0.937782347202301, + "learning_rate": 3.100013811282896e-06, + "loss": 0.5859, + "step": 3120 + }, + { + "epoch": 2.58393152954169, + "grad_norm": 0.9293384552001953, + "learning_rate": 3.098946206124329e-06, + "loss": 0.5826, + "step": 3121 + }, + { + "epoch": 2.5847598012147985, + "grad_norm": 0.935950756072998, + "learning_rate": 3.097878485075255e-06, + "loss": 0.589, + "step": 3122 + }, + { + "epoch": 2.5855880728879073, + "grad_norm": 0.9377515316009521, + "learning_rate": 3.0968106483422655e-06, + "loss": 0.5985, + "step": 3123 + }, + { + "epoch": 2.586416344561016, + "grad_norm": 0.9256250858306885, + "learning_rate": 3.09574269613198e-06, + "loss": 0.5944, + "step": 3124 + }, + { + "epoch": 2.5872446162341247, + "grad_norm": 0.873123288154602, + "learning_rate": 3.0946746286510364e-06, + "loss": 0.5918, + "step": 3125 + }, + { + "epoch": 2.5880728879072334, + "grad_norm": 0.9320067167282104, + "learning_rate": 3.093606446106096e-06, + "loss": 0.6365, + "step": 3126 + }, + { + "epoch": 2.5889011595803426, + "grad_norm": 0.9514285326004028, + "learning_rate": 3.0925381487038415e-06, + "loss": 0.585, + "step": 3127 + }, + { + "epoch": 2.589729431253451, + "grad_norm": 0.9028947353363037, + "learning_rate": 3.0914697366509777e-06, + "loss": 0.6063, + "step": 3128 + }, + { + "epoch": 2.59055770292656, + "grad_norm": 0.8989999294281006, + "learning_rate": 3.0904012101542346e-06, + "loss": 0.6012, + "step": 3129 + }, + { + "epoch": 2.591385974599669, + "grad_norm": 0.9131087064743042, + "learning_rate": 3.089332569420361e-06, + "loss": 0.5742, + "step": 3130 + }, + { + "epoch": 2.5922142462727775, + "grad_norm": 0.916441023349762, + "learning_rate": 3.088263814656128e-06, + "loss": 0.6011, + "step": 3131 + }, + { + "epoch": 2.5930425179458862, + "grad_norm": 0.8957144618034363, + "learning_rate": 3.087194946068332e-06, + "loss": 0.6008, + "step": 3132 + }, + { + "epoch": 2.593870789618995, + "grad_norm": 0.90632563829422, + "learning_rate": 3.086125963863787e-06, + "loss": 0.6062, + "step": 3133 + }, + { + "epoch": 2.5946990612921037, + "grad_norm": 0.94493168592453, + "learning_rate": 3.0850568682493315e-06, + "loss": 0.5667, + "step": 3134 + }, + { + "epoch": 2.5955273329652124, + "grad_norm": 0.9155673980712891, + "learning_rate": 3.0839876594318257e-06, + "loss": 0.5755, + "step": 3135 + }, + { + "epoch": 2.5963556046383216, + "grad_norm": 0.9013570547103882, + "learning_rate": 3.0829183376181524e-06, + "loss": 0.6053, + "step": 3136 + }, + { + "epoch": 2.59718387631143, + "grad_norm": 0.9294494390487671, + "learning_rate": 3.081848903015214e-06, + "loss": 0.5881, + "step": 3137 + }, + { + "epoch": 2.598012147984539, + "grad_norm": 0.8956479430198669, + "learning_rate": 3.0807793558299377e-06, + "loss": 0.5976, + "step": 3138 + }, + { + "epoch": 2.598840419657648, + "grad_norm": 0.9205741286277771, + "learning_rate": 3.079709696269269e-06, + "loss": 0.5776, + "step": 3139 + }, + { + "epoch": 2.5996686913307565, + "grad_norm": 0.901066780090332, + "learning_rate": 3.0786399245401784e-06, + "loss": 0.6101, + "step": 3140 + }, + { + "epoch": 2.6004969630038652, + "grad_norm": 0.9003482460975647, + "learning_rate": 3.0775700408496568e-06, + "loss": 0.5983, + "step": 3141 + }, + { + "epoch": 2.601325234676974, + "grad_norm": 0.9000759124755859, + "learning_rate": 3.0765000454047155e-06, + "loss": 0.605, + "step": 3142 + }, + { + "epoch": 2.6021535063500827, + "grad_norm": 0.8797931671142578, + "learning_rate": 3.0754299384123905e-06, + "loss": 0.5874, + "step": 3143 + }, + { + "epoch": 2.6029817780231914, + "grad_norm": 0.9307072758674622, + "learning_rate": 3.074359720079736e-06, + "loss": 0.6035, + "step": 3144 + }, + { + "epoch": 2.6038100496963006, + "grad_norm": 0.8928053379058838, + "learning_rate": 3.073289390613829e-06, + "loss": 0.5759, + "step": 3145 + }, + { + "epoch": 2.6046383213694093, + "grad_norm": 0.8879983425140381, + "learning_rate": 3.0722189502217705e-06, + "loss": 0.5947, + "step": 3146 + }, + { + "epoch": 2.605466593042518, + "grad_norm": 0.8889674544334412, + "learning_rate": 3.0711483991106784e-06, + "loss": 0.5939, + "step": 3147 + }, + { + "epoch": 2.6062948647156268, + "grad_norm": 0.9318514466285706, + "learning_rate": 3.070077737487695e-06, + "loss": 0.599, + "step": 3148 + }, + { + "epoch": 2.6071231363887355, + "grad_norm": 0.9251010417938232, + "learning_rate": 3.069006965559985e-06, + "loss": 0.5956, + "step": 3149 + }, + { + "epoch": 2.6079514080618442, + "grad_norm": 0.9122523069381714, + "learning_rate": 3.06793608353473e-06, + "loss": 0.6011, + "step": 3150 + }, + { + "epoch": 2.608779679734953, + "grad_norm": 0.9270116090774536, + "learning_rate": 3.0668650916191365e-06, + "loss": 0.5923, + "step": 3151 + }, + { + "epoch": 2.609607951408062, + "grad_norm": 0.939582884311676, + "learning_rate": 3.0657939900204338e-06, + "loss": 0.6053, + "step": 3152 + }, + { + "epoch": 2.6104362230811704, + "grad_norm": 0.9206486940383911, + "learning_rate": 3.064722778945867e-06, + "loss": 0.6197, + "step": 3153 + }, + { + "epoch": 2.6112644947542796, + "grad_norm": 0.9047932624816895, + "learning_rate": 3.063651458602708e-06, + "loss": 0.5774, + "step": 3154 + }, + { + "epoch": 2.6120927664273883, + "grad_norm": 0.9285469055175781, + "learning_rate": 3.062580029198245e-06, + "loss": 0.6107, + "step": 3155 + }, + { + "epoch": 2.612921038100497, + "grad_norm": 0.9187634587287903, + "learning_rate": 3.0615084909397906e-06, + "loss": 0.5919, + "step": 3156 + }, + { + "epoch": 2.6137493097736058, + "grad_norm": 0.8950710296630859, + "learning_rate": 3.060436844034678e-06, + "loss": 0.6029, + "step": 3157 + }, + { + "epoch": 2.6145775814467145, + "grad_norm": 0.9206116795539856, + "learning_rate": 3.0593650886902606e-06, + "loss": 0.5632, + "step": 3158 + }, + { + "epoch": 2.615405853119823, + "grad_norm": 0.9081333875656128, + "learning_rate": 3.058293225113912e-06, + "loss": 0.6127, + "step": 3159 + }, + { + "epoch": 2.616234124792932, + "grad_norm": 0.9028277397155762, + "learning_rate": 3.0572212535130297e-06, + "loss": 0.6089, + "step": 3160 + }, + { + "epoch": 2.617062396466041, + "grad_norm": 0.9449213147163391, + "learning_rate": 3.0561491740950288e-06, + "loss": 0.5841, + "step": 3161 + }, + { + "epoch": 2.6178906681391494, + "grad_norm": 0.9926438927650452, + "learning_rate": 3.055076987067347e-06, + "loss": 0.564, + "step": 3162 + }, + { + "epoch": 2.6187189398122586, + "grad_norm": 0.9098719954490662, + "learning_rate": 3.0540046926374422e-06, + "loss": 0.6268, + "step": 3163 + }, + { + "epoch": 2.6195472114853673, + "grad_norm": 0.8997418880462646, + "learning_rate": 3.0529322910127944e-06, + "loss": 0.5972, + "step": 3164 + }, + { + "epoch": 2.620375483158476, + "grad_norm": 0.9076839685440063, + "learning_rate": 3.0518597824009015e-06, + "loss": 0.5921, + "step": 3165 + }, + { + "epoch": 2.6212037548315847, + "grad_norm": 0.9348479509353638, + "learning_rate": 3.050787167009287e-06, + "loss": 0.5802, + "step": 3166 + }, + { + "epoch": 2.6220320265046935, + "grad_norm": 0.9381133913993835, + "learning_rate": 3.0497144450454886e-06, + "loss": 0.5943, + "step": 3167 + }, + { + "epoch": 2.622860298177802, + "grad_norm": 0.9327574968338013, + "learning_rate": 3.0486416167170694e-06, + "loss": 0.5832, + "step": 3168 + }, + { + "epoch": 2.623688569850911, + "grad_norm": 0.9142323732376099, + "learning_rate": 3.0475686822316126e-06, + "loss": 0.6048, + "step": 3169 + }, + { + "epoch": 2.62451684152402, + "grad_norm": 0.9799601435661316, + "learning_rate": 3.04649564179672e-06, + "loss": 0.5783, + "step": 3170 + }, + { + "epoch": 2.625345113197129, + "grad_norm": 0.9203731417655945, + "learning_rate": 3.045422495620015e-06, + "loss": 0.6028, + "step": 3171 + }, + { + "epoch": 2.6261733848702375, + "grad_norm": 0.9614686369895935, + "learning_rate": 3.0443492439091415e-06, + "loss": 0.5895, + "step": 3172 + }, + { + "epoch": 2.6270016565433463, + "grad_norm": 0.980112612247467, + "learning_rate": 3.0432758868717626e-06, + "loss": 0.5887, + "step": 3173 + }, + { + "epoch": 2.627829928216455, + "grad_norm": 0.9174870252609253, + "learning_rate": 3.042202424715565e-06, + "loss": 0.5831, + "step": 3174 + }, + { + "epoch": 2.6286581998895637, + "grad_norm": 0.9396457672119141, + "learning_rate": 3.0411288576482525e-06, + "loss": 0.5816, + "step": 3175 + }, + { + "epoch": 2.6294864715626725, + "grad_norm": 0.9067835211753845, + "learning_rate": 3.04005518587755e-06, + "loss": 0.5748, + "step": 3176 + }, + { + "epoch": 2.630314743235781, + "grad_norm": 0.9329981207847595, + "learning_rate": 3.038981409611204e-06, + "loss": 0.5901, + "step": 3177 + }, + { + "epoch": 2.63114301490889, + "grad_norm": 0.9418731927871704, + "learning_rate": 3.0379075290569793e-06, + "loss": 0.5981, + "step": 3178 + }, + { + "epoch": 2.631971286581999, + "grad_norm": 0.9180083274841309, + "learning_rate": 3.0368335444226614e-06, + "loss": 0.5653, + "step": 3179 + }, + { + "epoch": 2.632799558255108, + "grad_norm": 0.9566802978515625, + "learning_rate": 3.0357594559160575e-06, + "loss": 0.5968, + "step": 3180 + }, + { + "epoch": 2.6336278299282165, + "grad_norm": 0.9177946448326111, + "learning_rate": 3.034685263744994e-06, + "loss": 0.6076, + "step": 3181 + }, + { + "epoch": 2.6344561016013253, + "grad_norm": 0.905975341796875, + "learning_rate": 3.033610968117314e-06, + "loss": 0.6064, + "step": 3182 + }, + { + "epoch": 2.635284373274434, + "grad_norm": 0.9021571278572083, + "learning_rate": 3.0325365692408883e-06, + "loss": 0.5818, + "step": 3183 + }, + { + "epoch": 2.6361126449475427, + "grad_norm": 0.9211055040359497, + "learning_rate": 3.0314620673235995e-06, + "loss": 0.5811, + "step": 3184 + }, + { + "epoch": 2.6369409166206514, + "grad_norm": 0.929892897605896, + "learning_rate": 3.0303874625733548e-06, + "loss": 0.6039, + "step": 3185 + }, + { + "epoch": 2.6377691882937606, + "grad_norm": 0.9520539045333862, + "learning_rate": 3.0293127551980805e-06, + "loss": 0.6156, + "step": 3186 + }, + { + "epoch": 2.638597459966869, + "grad_norm": 0.9111108779907227, + "learning_rate": 3.028237945405722e-06, + "loss": 0.6081, + "step": 3187 + }, + { + "epoch": 2.639425731639978, + "grad_norm": 0.9411360025405884, + "learning_rate": 3.027163033404246e-06, + "loss": 0.5897, + "step": 3188 + }, + { + "epoch": 2.640254003313087, + "grad_norm": 0.9253203868865967, + "learning_rate": 3.026088019401637e-06, + "loss": 0.5957, + "step": 3189 + }, + { + "epoch": 2.6410822749861955, + "grad_norm": 1.010617971420288, + "learning_rate": 3.0250129036059e-06, + "loss": 0.602, + "step": 3190 + }, + { + "epoch": 2.6419105466593042, + "grad_norm": 0.9021222591400146, + "learning_rate": 3.02393768622506e-06, + "loss": 0.5796, + "step": 3191 + }, + { + "epoch": 2.642738818332413, + "grad_norm": 0.9229567050933838, + "learning_rate": 3.0228623674671627e-06, + "loss": 0.5827, + "step": 3192 + }, + { + "epoch": 2.6435670900055217, + "grad_norm": 0.9603396654129028, + "learning_rate": 3.0217869475402704e-06, + "loss": 0.5886, + "step": 3193 + }, + { + "epoch": 2.6443953616786304, + "grad_norm": 0.9483562707901001, + "learning_rate": 3.020711426652469e-06, + "loss": 0.6089, + "step": 3194 + }, + { + "epoch": 2.6452236333517396, + "grad_norm": 0.9240924715995789, + "learning_rate": 3.019635805011859e-06, + "loss": 0.6182, + "step": 3195 + }, + { + "epoch": 2.646051905024848, + "grad_norm": 0.9176941514015198, + "learning_rate": 3.018560082826566e-06, + "loss": 0.593, + "step": 3196 + }, + { + "epoch": 2.646880176697957, + "grad_norm": 0.9058155417442322, + "learning_rate": 3.0174842603047307e-06, + "loss": 0.5853, + "step": 3197 + }, + { + "epoch": 2.647708448371066, + "grad_norm": 0.9544078707695007, + "learning_rate": 3.016408337654514e-06, + "loss": 0.5743, + "step": 3198 + }, + { + "epoch": 2.6485367200441745, + "grad_norm": 0.944963812828064, + "learning_rate": 3.0153323150840987e-06, + "loss": 0.6007, + "step": 3199 + }, + { + "epoch": 2.6493649917172832, + "grad_norm": 0.9196237921714783, + "learning_rate": 3.014256192801685e-06, + "loss": 0.5818, + "step": 3200 + }, + { + "epoch": 2.650193263390392, + "grad_norm": 0.9315558075904846, + "learning_rate": 3.01317997101549e-06, + "loss": 0.5861, + "step": 3201 + }, + { + "epoch": 2.6510215350635007, + "grad_norm": 0.9272212982177734, + "learning_rate": 3.012103649933756e-06, + "loss": 0.597, + "step": 3202 + }, + { + "epoch": 2.6518498067366094, + "grad_norm": 0.9128772616386414, + "learning_rate": 3.011027229764739e-06, + "loss": 0.5863, + "step": 3203 + }, + { + "epoch": 2.6526780784097186, + "grad_norm": 0.9234123826026917, + "learning_rate": 3.0099507107167158e-06, + "loss": 0.589, + "step": 3204 + }, + { + "epoch": 2.6535063500828273, + "grad_norm": 0.9162564873695374, + "learning_rate": 3.0088740929979854e-06, + "loss": 0.5855, + "step": 3205 + }, + { + "epoch": 2.654334621755936, + "grad_norm": 0.8842553496360779, + "learning_rate": 3.0077973768168604e-06, + "loss": 0.5947, + "step": 3206 + }, + { + "epoch": 2.6551628934290448, + "grad_norm": 0.9413550496101379, + "learning_rate": 3.0067205623816775e-06, + "loss": 0.5882, + "step": 3207 + }, + { + "epoch": 2.6559911651021535, + "grad_norm": 0.9217298626899719, + "learning_rate": 3.005643649900789e-06, + "loss": 0.5806, + "step": 3208 + }, + { + "epoch": 2.656819436775262, + "grad_norm": 0.9214861392974854, + "learning_rate": 3.004566639582568e-06, + "loss": 0.5656, + "step": 3209 + }, + { + "epoch": 2.657647708448371, + "grad_norm": 0.9789690375328064, + "learning_rate": 3.003489531635405e-06, + "loss": 0.5897, + "step": 3210 + }, + { + "epoch": 2.65847598012148, + "grad_norm": 0.9465369582176208, + "learning_rate": 3.0024123262677123e-06, + "loss": 0.5996, + "step": 3211 + }, + { + "epoch": 2.6593042517945884, + "grad_norm": 0.9011247754096985, + "learning_rate": 3.001335023687917e-06, + "loss": 0.5849, + "step": 3212 + }, + { + "epoch": 2.6601325234676976, + "grad_norm": 0.9762759804725647, + "learning_rate": 3.0002576241044683e-06, + "loss": 0.6161, + "step": 3213 + }, + { + "epoch": 2.6609607951408063, + "grad_norm": 0.9362512230873108, + "learning_rate": 2.9991801277258327e-06, + "loss": 0.6186, + "step": 3214 + }, + { + "epoch": 2.661789066813915, + "grad_norm": 0.9391586780548096, + "learning_rate": 2.998102534760495e-06, + "loss": 0.5787, + "step": 3215 + }, + { + "epoch": 2.6626173384870238, + "grad_norm": 0.9361854195594788, + "learning_rate": 2.997024845416961e-06, + "loss": 0.5996, + "step": 3216 + }, + { + "epoch": 2.6634456101601325, + "grad_norm": 0.9153965711593628, + "learning_rate": 2.9959470599037525e-06, + "loss": 0.5777, + "step": 3217 + }, + { + "epoch": 2.664273881833241, + "grad_norm": 0.9155426621437073, + "learning_rate": 2.994869178429411e-06, + "loss": 0.5906, + "step": 3218 + }, + { + "epoch": 2.66510215350635, + "grad_norm": 0.8978738784790039, + "learning_rate": 2.993791201202496e-06, + "loss": 0.5707, + "step": 3219 + }, + { + "epoch": 2.665930425179459, + "grad_norm": 0.9147471785545349, + "learning_rate": 2.9927131284315874e-06, + "loss": 0.6003, + "step": 3220 + }, + { + "epoch": 2.6667586968525674, + "grad_norm": 0.9578046202659607, + "learning_rate": 2.991634960325281e-06, + "loss": 0.6068, + "step": 3221 + }, + { + "epoch": 2.6675869685256766, + "grad_norm": 0.9182425141334534, + "learning_rate": 2.9905566970921933e-06, + "loss": 0.5939, + "step": 3222 + }, + { + "epoch": 2.6684152401987853, + "grad_norm": 0.9168176651000977, + "learning_rate": 2.989478338940957e-06, + "loss": 0.5967, + "step": 3223 + }, + { + "epoch": 2.669243511871894, + "grad_norm": 0.913521409034729, + "learning_rate": 2.9883998860802253e-06, + "loss": 0.6081, + "step": 3224 + }, + { + "epoch": 2.6700717835450027, + "grad_norm": 0.9223765134811401, + "learning_rate": 2.9873213387186684e-06, + "loss": 0.5901, + "step": 3225 + }, + { + "epoch": 2.6709000552181115, + "grad_norm": 0.9228116273880005, + "learning_rate": 2.9862426970649743e-06, + "loss": 0.5959, + "step": 3226 + }, + { + "epoch": 2.67172832689122, + "grad_norm": 0.9379467368125916, + "learning_rate": 2.985163961327852e-06, + "loss": 0.5986, + "step": 3227 + }, + { + "epoch": 2.672556598564329, + "grad_norm": 0.921449601650238, + "learning_rate": 2.984085131716027e-06, + "loss": 0.5587, + "step": 3228 + }, + { + "epoch": 2.673384870237438, + "grad_norm": 0.9038673043251038, + "learning_rate": 2.9830062084382393e-06, + "loss": 0.5908, + "step": 3229 + }, + { + "epoch": 2.674213141910547, + "grad_norm": 0.9072180986404419, + "learning_rate": 2.981927191703254e-06, + "loss": 0.6064, + "step": 3230 + }, + { + "epoch": 2.6750414135836555, + "grad_norm": 0.9178011417388916, + "learning_rate": 2.9808480817198494e-06, + "loss": 0.5969, + "step": 3231 + }, + { + "epoch": 2.6758696852567643, + "grad_norm": 0.9064451456069946, + "learning_rate": 2.9797688786968233e-06, + "loss": 0.571, + "step": 3232 + }, + { + "epoch": 2.676697956929873, + "grad_norm": 0.9392306804656982, + "learning_rate": 2.9786895828429914e-06, + "loss": 0.5903, + "step": 3233 + }, + { + "epoch": 2.6775262286029817, + "grad_norm": 0.9398954510688782, + "learning_rate": 2.9776101943671875e-06, + "loss": 0.6163, + "step": 3234 + }, + { + "epoch": 2.6783545002760905, + "grad_norm": 0.9082908630371094, + "learning_rate": 2.9765307134782637e-06, + "loss": 0.5838, + "step": 3235 + }, + { + "epoch": 2.679182771949199, + "grad_norm": 0.926916778087616, + "learning_rate": 2.975451140385089e-06, + "loss": 0.5599, + "step": 3236 + }, + { + "epoch": 2.680011043622308, + "grad_norm": 0.9106405377388, + "learning_rate": 2.97437147529655e-06, + "loss": 0.6012, + "step": 3237 + }, + { + "epoch": 2.680839315295417, + "grad_norm": 0.9428172707557678, + "learning_rate": 2.9732917184215524e-06, + "loss": 0.5985, + "step": 3238 + }, + { + "epoch": 2.681667586968526, + "grad_norm": 0.9456861019134521, + "learning_rate": 2.9722118699690212e-06, + "loss": 0.6206, + "step": 3239 + }, + { + "epoch": 2.6824958586416345, + "grad_norm": 0.9116294384002686, + "learning_rate": 2.971131930147893e-06, + "loss": 0.6166, + "step": 3240 + }, + { + "epoch": 2.6833241303147433, + "grad_norm": 0.9061588644981384, + "learning_rate": 2.970051899167129e-06, + "loss": 0.5792, + "step": 3241 + }, + { + "epoch": 2.684152401987852, + "grad_norm": 0.947784423828125, + "learning_rate": 2.968971777235704e-06, + "loss": 0.5813, + "step": 3242 + }, + { + "epoch": 2.6849806736609607, + "grad_norm": 0.9296494722366333, + "learning_rate": 2.9678915645626115e-06, + "loss": 0.5916, + "step": 3243 + }, + { + "epoch": 2.6858089453340694, + "grad_norm": 0.9151764512062073, + "learning_rate": 2.9668112613568632e-06, + "loss": 0.6097, + "step": 3244 + }, + { + "epoch": 2.6866372170071786, + "grad_norm": 0.9887247681617737, + "learning_rate": 2.965730867827488e-06, + "loss": 0.5951, + "step": 3245 + }, + { + "epoch": 2.687465488680287, + "grad_norm": 0.9406042695045471, + "learning_rate": 2.964650384183531e-06, + "loss": 0.5918, + "step": 3246 + }, + { + "epoch": 2.688293760353396, + "grad_norm": 0.9440297484397888, + "learning_rate": 2.963569810634056e-06, + "loss": 0.6098, + "step": 3247 + }, + { + "epoch": 2.689122032026505, + "grad_norm": 0.93993079662323, + "learning_rate": 2.962489147388144e-06, + "loss": 0.6242, + "step": 3248 + }, + { + "epoch": 2.6899503036996135, + "grad_norm": 0.9704070091247559, + "learning_rate": 2.9614083946548927e-06, + "loss": 0.5764, + "step": 3249 + }, + { + "epoch": 2.6907785753727222, + "grad_norm": 0.9809379577636719, + "learning_rate": 2.9603275526434188e-06, + "loss": 0.5997, + "step": 3250 + }, + { + "epoch": 2.691606847045831, + "grad_norm": 0.9246951937675476, + "learning_rate": 2.9592466215628547e-06, + "loss": 0.577, + "step": 3251 + }, + { + "epoch": 2.6924351187189397, + "grad_norm": 0.9204201698303223, + "learning_rate": 2.95816560162235e-06, + "loss": 0.6036, + "step": 3252 + }, + { + "epoch": 2.6932633903920484, + "grad_norm": 0.9180527925491333, + "learning_rate": 2.9570844930310734e-06, + "loss": 0.5919, + "step": 3253 + }, + { + "epoch": 2.6940916620651576, + "grad_norm": 0.9165668487548828, + "learning_rate": 2.9560032959982067e-06, + "loss": 0.5646, + "step": 3254 + }, + { + "epoch": 2.694919933738266, + "grad_norm": 0.9150571823120117, + "learning_rate": 2.9549220107329536e-06, + "loss": 0.581, + "step": 3255 + }, + { + "epoch": 2.695748205411375, + "grad_norm": 0.9429931044578552, + "learning_rate": 2.9538406374445334e-06, + "loss": 0.595, + "step": 3256 + }, + { + "epoch": 2.696576477084484, + "grad_norm": 0.939118504524231, + "learning_rate": 2.9527591763421785e-06, + "loss": 0.5772, + "step": 3257 + }, + { + "epoch": 2.6974047487575925, + "grad_norm": 0.9130142331123352, + "learning_rate": 2.9516776276351446e-06, + "loss": 0.6031, + "step": 3258 + }, + { + "epoch": 2.6982330204307012, + "grad_norm": 0.9334836006164551, + "learning_rate": 2.9505959915327004e-06, + "loss": 0.5899, + "step": 3259 + }, + { + "epoch": 2.69906129210381, + "grad_norm": 0.9407964944839478, + "learning_rate": 2.9495142682441314e-06, + "loss": 0.6232, + "step": 3260 + }, + { + "epoch": 2.6998895637769187, + "grad_norm": 0.9573985934257507, + "learning_rate": 2.9484324579787427e-06, + "loss": 0.5932, + "step": 3261 + }, + { + "epoch": 2.7007178354500274, + "grad_norm": 0.9223983287811279, + "learning_rate": 2.947350560945853e-06, + "loss": 0.5861, + "step": 3262 + }, + { + "epoch": 2.7015461071231366, + "grad_norm": 0.9206790328025818, + "learning_rate": 2.9462685773548004e-06, + "loss": 0.5867, + "step": 3263 + }, + { + "epoch": 2.7023743787962453, + "grad_norm": 0.9197454452514648, + "learning_rate": 2.9451865074149378e-06, + "loss": 0.5867, + "step": 3264 + }, + { + "epoch": 2.703202650469354, + "grad_norm": 0.9365606904029846, + "learning_rate": 2.944104351335636e-06, + "loss": 0.6106, + "step": 3265 + }, + { + "epoch": 2.7040309221424628, + "grad_norm": 0.9116243720054626, + "learning_rate": 2.9430221093262823e-06, + "loss": 0.5904, + "step": 3266 + }, + { + "epoch": 2.7048591938155715, + "grad_norm": 0.9128034114837646, + "learning_rate": 2.9419397815962802e-06, + "loss": 0.6055, + "step": 3267 + }, + { + "epoch": 2.70568746548868, + "grad_norm": 0.9147705435752869, + "learning_rate": 2.94085736835505e-06, + "loss": 0.5941, + "step": 3268 + }, + { + "epoch": 2.706515737161789, + "grad_norm": 0.9419110417366028, + "learning_rate": 2.939774869812029e-06, + "loss": 0.5595, + "step": 3269 + }, + { + "epoch": 2.707344008834898, + "grad_norm": 0.9202789664268494, + "learning_rate": 2.93869228617667e-06, + "loss": 0.5788, + "step": 3270 + }, + { + "epoch": 2.7081722805080064, + "grad_norm": 0.9068663716316223, + "learning_rate": 2.9376096176584424e-06, + "loss": 0.5749, + "step": 3271 + }, + { + "epoch": 2.7090005521811156, + "grad_norm": 0.923497200012207, + "learning_rate": 2.936526864466833e-06, + "loss": 0.6029, + "step": 3272 + }, + { + "epoch": 2.7098288238542243, + "grad_norm": 0.9812871217727661, + "learning_rate": 2.935444026811346e-06, + "loss": 0.6193, + "step": 3273 + }, + { + "epoch": 2.710657095527333, + "grad_norm": 0.9538823366165161, + "learning_rate": 2.934361104901498e-06, + "loss": 0.5809, + "step": 3274 + }, + { + "epoch": 2.7114853672004418, + "grad_norm": 0.9111268520355225, + "learning_rate": 2.9332780989468256e-06, + "loss": 0.6181, + "step": 3275 + }, + { + "epoch": 2.7123136388735505, + "grad_norm": 0.9485889673233032, + "learning_rate": 2.9321950091568795e-06, + "loss": 0.5823, + "step": 3276 + }, + { + "epoch": 2.713141910546659, + "grad_norm": 0.9233952760696411, + "learning_rate": 2.931111835741229e-06, + "loss": 0.6008, + "step": 3277 + }, + { + "epoch": 2.713970182219768, + "grad_norm": 0.9075472950935364, + "learning_rate": 2.9300285789094563e-06, + "loss": 0.5808, + "step": 3278 + }, + { + "epoch": 2.714798453892877, + "grad_norm": 0.9238584041595459, + "learning_rate": 2.9289452388711624e-06, + "loss": 0.6095, + "step": 3279 + }, + { + "epoch": 2.7156267255659854, + "grad_norm": 0.889460027217865, + "learning_rate": 2.9278618158359635e-06, + "loss": 0.6015, + "step": 3280 + }, + { + "epoch": 2.7164549972390946, + "grad_norm": 0.9549591541290283, + "learning_rate": 2.9267783100134915e-06, + "loss": 0.6029, + "step": 3281 + }, + { + "epoch": 2.7172832689122033, + "grad_norm": 0.9522138237953186, + "learning_rate": 2.9256947216133946e-06, + "loss": 0.5998, + "step": 3282 + }, + { + "epoch": 2.718111540585312, + "grad_norm": 0.8992515802383423, + "learning_rate": 2.924611050845338e-06, + "loss": 0.6012, + "step": 3283 + }, + { + "epoch": 2.7189398122584207, + "grad_norm": 0.9058646559715271, + "learning_rate": 2.9235272979190014e-06, + "loss": 0.6092, + "step": 3284 + }, + { + "epoch": 2.7197680839315295, + "grad_norm": 0.916130542755127, + "learning_rate": 2.92244346304408e-06, + "loss": 0.6054, + "step": 3285 + }, + { + "epoch": 2.720596355604638, + "grad_norm": 0.9203442335128784, + "learning_rate": 2.921359546430287e-06, + "loss": 0.596, + "step": 3286 + }, + { + "epoch": 2.721424627277747, + "grad_norm": 0.9603051543235779, + "learning_rate": 2.920275548287349e-06, + "loss": 0.5976, + "step": 3287 + }, + { + "epoch": 2.722252898950856, + "grad_norm": 0.9188669919967651, + "learning_rate": 2.91919146882501e-06, + "loss": 0.5812, + "step": 3288 + }, + { + "epoch": 2.723081170623965, + "grad_norm": 0.9154419898986816, + "learning_rate": 2.9181073082530302e-06, + "loss": 0.6127, + "step": 3289 + }, + { + "epoch": 2.7239094422970735, + "grad_norm": 0.9228849411010742, + "learning_rate": 2.917023066781184e-06, + "loss": 0.6223, + "step": 3290 + }, + { + "epoch": 2.7247377139701823, + "grad_norm": 0.9034006595611572, + "learning_rate": 2.915938744619261e-06, + "loss": 0.5981, + "step": 3291 + }, + { + "epoch": 2.725565985643291, + "grad_norm": 0.9119758009910583, + "learning_rate": 2.914854341977069e-06, + "loss": 0.5974, + "step": 3292 + }, + { + "epoch": 2.7263942573163997, + "grad_norm": 0.9350270628929138, + "learning_rate": 2.9137698590644283e-06, + "loss": 0.5958, + "step": 3293 + }, + { + "epoch": 2.7272225289895085, + "grad_norm": 0.9229425191879272, + "learning_rate": 2.9126852960911773e-06, + "loss": 0.5873, + "step": 3294 + }, + { + "epoch": 2.728050800662617, + "grad_norm": 0.9276396632194519, + "learning_rate": 2.9116006532671685e-06, + "loss": 0.6069, + "step": 3295 + }, + { + "epoch": 2.728879072335726, + "grad_norm": 0.9154210090637207, + "learning_rate": 2.9105159308022705e-06, + "loss": 0.5741, + "step": 3296 + }, + { + "epoch": 2.729707344008835, + "grad_norm": 0.9545791149139404, + "learning_rate": 2.909431128906367e-06, + "loss": 0.5995, + "step": 3297 + }, + { + "epoch": 2.730535615681944, + "grad_norm": 0.9200988411903381, + "learning_rate": 2.908346247789356e-06, + "loss": 0.6152, + "step": 3298 + }, + { + "epoch": 2.7313638873550525, + "grad_norm": 0.9288673996925354, + "learning_rate": 2.9072612876611524e-06, + "loss": 0.5904, + "step": 3299 + }, + { + "epoch": 2.7321921590281613, + "grad_norm": 0.918323278427124, + "learning_rate": 2.906176248731686e-06, + "loss": 0.5999, + "step": 3300 + }, + { + "epoch": 2.73302043070127, + "grad_norm": 0.936652421951294, + "learning_rate": 2.905091131210903e-06, + "loss": 0.5928, + "step": 3301 + }, + { + "epoch": 2.7338487023743787, + "grad_norm": 0.8942639827728271, + "learning_rate": 2.9040059353087614e-06, + "loss": 0.5739, + "step": 3302 + }, + { + "epoch": 2.7346769740474874, + "grad_norm": 0.9197317957878113, + "learning_rate": 2.902920661235238e-06, + "loss": 0.5776, + "step": 3303 + }, + { + "epoch": 2.7355052457205966, + "grad_norm": 0.895505428314209, + "learning_rate": 2.9018353092003216e-06, + "loss": 0.6067, + "step": 3304 + }, + { + "epoch": 2.736333517393705, + "grad_norm": 0.947441816329956, + "learning_rate": 2.9007498794140193e-06, + "loss": 0.6072, + "step": 3305 + }, + { + "epoch": 2.737161789066814, + "grad_norm": 0.9008322358131409, + "learning_rate": 2.8996643720863515e-06, + "loss": 0.5908, + "step": 3306 + }, + { + "epoch": 2.737990060739923, + "grad_norm": 0.9137061238288879, + "learning_rate": 2.898578787427353e-06, + "loss": 0.607, + "step": 3307 + }, + { + "epoch": 2.7388183324130315, + "grad_norm": 0.9126008749008179, + "learning_rate": 2.8974931256470745e-06, + "loss": 0.5809, + "step": 3308 + }, + { + "epoch": 2.7396466040861402, + "grad_norm": 0.9215322136878967, + "learning_rate": 2.896407386955582e-06, + "loss": 0.5807, + "step": 3309 + }, + { + "epoch": 2.740474875759249, + "grad_norm": 0.8903998136520386, + "learning_rate": 2.8953215715629547e-06, + "loss": 0.5968, + "step": 3310 + }, + { + "epoch": 2.7413031474323577, + "grad_norm": 0.9156417846679688, + "learning_rate": 2.8942356796792888e-06, + "loss": 0.5809, + "step": 3311 + }, + { + "epoch": 2.7421314191054664, + "grad_norm": 0.952264666557312, + "learning_rate": 2.893149711514694e-06, + "loss": 0.5818, + "step": 3312 + }, + { + "epoch": 2.7429596907785756, + "grad_norm": 0.9201667904853821, + "learning_rate": 2.8920636672792945e-06, + "loss": 0.5789, + "step": 3313 + }, + { + "epoch": 2.743787962451684, + "grad_norm": 0.9581197500228882, + "learning_rate": 2.890977547183231e-06, + "loss": 0.6007, + "step": 3314 + }, + { + "epoch": 2.744616234124793, + "grad_norm": 0.9360357522964478, + "learning_rate": 2.889891351436656e-06, + "loss": 0.6096, + "step": 3315 + }, + { + "epoch": 2.745444505797902, + "grad_norm": 0.9209451079368591, + "learning_rate": 2.8888050802497388e-06, + "loss": 0.6015, + "step": 3316 + }, + { + "epoch": 2.7462727774710105, + "grad_norm": 0.9496802091598511, + "learning_rate": 2.8877187338326634e-06, + "loss": 0.5802, + "step": 3317 + }, + { + "epoch": 2.7471010491441192, + "grad_norm": 0.9407662153244019, + "learning_rate": 2.8866323123956274e-06, + "loss": 0.588, + "step": 3318 + }, + { + "epoch": 2.747929320817228, + "grad_norm": 0.9603942036628723, + "learning_rate": 2.885545816148843e-06, + "loss": 0.5759, + "step": 3319 + }, + { + "epoch": 2.7487575924903367, + "grad_norm": 0.930496335029602, + "learning_rate": 2.8844592453025372e-06, + "loss": 0.5916, + "step": 3320 + }, + { + "epoch": 2.7495858641634454, + "grad_norm": 0.8831289410591125, + "learning_rate": 2.883372600066951e-06, + "loss": 0.5889, + "step": 3321 + }, + { + "epoch": 2.7504141358365546, + "grad_norm": 0.9076021909713745, + "learning_rate": 2.8822858806523407e-06, + "loss": 0.5842, + "step": 3322 + }, + { + "epoch": 2.7512424075096633, + "grad_norm": 0.9093723297119141, + "learning_rate": 2.8811990872689764e-06, + "loss": 0.584, + "step": 3323 + }, + { + "epoch": 2.752070679182772, + "grad_norm": 0.9458228349685669, + "learning_rate": 2.880112220127142e-06, + "loss": 0.5812, + "step": 3324 + }, + { + "epoch": 2.7528989508558808, + "grad_norm": 0.9406691789627075, + "learning_rate": 2.8790252794371365e-06, + "loss": 0.5889, + "step": 3325 + }, + { + "epoch": 2.7537272225289895, + "grad_norm": 0.9094856381416321, + "learning_rate": 2.8779382654092726e-06, + "loss": 0.5919, + "step": 3326 + }, + { + "epoch": 2.754555494202098, + "grad_norm": 0.9350448250770569, + "learning_rate": 2.8768511782538766e-06, + "loss": 0.5738, + "step": 3327 + }, + { + "epoch": 2.755383765875207, + "grad_norm": 0.9427900314331055, + "learning_rate": 2.875764018181291e-06, + "loss": 0.5962, + "step": 3328 + }, + { + "epoch": 2.756212037548316, + "grad_norm": 0.9439454674720764, + "learning_rate": 2.8746767854018715e-06, + "loss": 0.5994, + "step": 3329 + }, + { + "epoch": 2.7570403092214244, + "grad_norm": 0.9291850328445435, + "learning_rate": 2.8735894801259863e-06, + "loss": 0.6055, + "step": 3330 + }, + { + "epoch": 2.7578685808945336, + "grad_norm": 0.9287533164024353, + "learning_rate": 2.872502102564019e-06, + "loss": 0.5798, + "step": 3331 + }, + { + "epoch": 2.7586968525676423, + "grad_norm": 0.8941923379898071, + "learning_rate": 2.871414652926367e-06, + "loss": 0.6066, + "step": 3332 + }, + { + "epoch": 2.759525124240751, + "grad_norm": 0.924333930015564, + "learning_rate": 2.8703271314234425e-06, + "loss": 0.5872, + "step": 3333 + }, + { + "epoch": 2.7603533959138598, + "grad_norm": 0.8994147777557373, + "learning_rate": 2.8692395382656702e-06, + "loss": 0.5906, + "step": 3334 + }, + { + "epoch": 2.7611816675869685, + "grad_norm": 0.9152485132217407, + "learning_rate": 2.8681518736634895e-06, + "loss": 0.5774, + "step": 3335 + }, + { + "epoch": 2.762009939260077, + "grad_norm": 0.9269137978553772, + "learning_rate": 2.867064137827353e-06, + "loss": 0.5759, + "step": 3336 + }, + { + "epoch": 2.762838210933186, + "grad_norm": 0.916441023349762, + "learning_rate": 2.8659763309677267e-06, + "loss": 0.5927, + "step": 3337 + }, + { + "epoch": 2.763666482606295, + "grad_norm": 0.9280357956886292, + "learning_rate": 2.8648884532950925e-06, + "loss": 0.6004, + "step": 3338 + }, + { + "epoch": 2.7644947542794034, + "grad_norm": 0.9075552821159363, + "learning_rate": 2.8638005050199443e-06, + "loss": 0.6185, + "step": 3339 + }, + { + "epoch": 2.7653230259525126, + "grad_norm": 0.9340760707855225, + "learning_rate": 2.8627124863527895e-06, + "loss": 0.6119, + "step": 3340 + }, + { + "epoch": 2.7661512976256213, + "grad_norm": 0.9393087029457092, + "learning_rate": 2.8616243975041495e-06, + "loss": 0.6058, + "step": 3341 + }, + { + "epoch": 2.76697956929873, + "grad_norm": 0.9380460381507874, + "learning_rate": 2.8605362386845605e-06, + "loss": 0.5955, + "step": 3342 + }, + { + "epoch": 2.7678078409718387, + "grad_norm": 0.9393378496170044, + "learning_rate": 2.8594480101045683e-06, + "loss": 0.5841, + "step": 3343 + }, + { + "epoch": 2.7686361126449475, + "grad_norm": 0.9186044335365295, + "learning_rate": 2.8583597119747385e-06, + "loss": 0.5977, + "step": 3344 + }, + { + "epoch": 2.769464384318056, + "grad_norm": 0.9224127531051636, + "learning_rate": 2.8572713445056448e-06, + "loss": 0.5863, + "step": 3345 + }, + { + "epoch": 2.770292655991165, + "grad_norm": 0.9517519474029541, + "learning_rate": 2.8561829079078764e-06, + "loss": 0.5748, + "step": 3346 + }, + { + "epoch": 2.771120927664274, + "grad_norm": 0.9624918103218079, + "learning_rate": 2.855094402392036e-06, + "loss": 0.588, + "step": 3347 + }, + { + "epoch": 2.771949199337383, + "grad_norm": 0.9524730443954468, + "learning_rate": 2.8540058281687375e-06, + "loss": 0.6032, + "step": 3348 + }, + { + "epoch": 2.7727774710104915, + "grad_norm": 0.9574846029281616, + "learning_rate": 2.852917185448612e-06, + "loss": 0.5915, + "step": 3349 + }, + { + "epoch": 2.7736057426836003, + "grad_norm": 0.9290011525154114, + "learning_rate": 2.851828474442301e-06, + "loss": 0.5715, + "step": 3350 + }, + { + "epoch": 2.774434014356709, + "grad_norm": 0.9393135905265808, + "learning_rate": 2.8507396953604597e-06, + "loss": 0.5855, + "step": 3351 + }, + { + "epoch": 2.7752622860298177, + "grad_norm": 0.9188078045845032, + "learning_rate": 2.8496508484137574e-06, + "loss": 0.598, + "step": 3352 + }, + { + "epoch": 2.7760905577029265, + "grad_norm": 0.9129231572151184, + "learning_rate": 2.8485619338128755e-06, + "loss": 0.6076, + "step": 3353 + }, + { + "epoch": 2.776918829376035, + "grad_norm": 0.941085934638977, + "learning_rate": 2.847472951768508e-06, + "loss": 0.5787, + "step": 3354 + }, + { + "epoch": 2.777747101049144, + "grad_norm": 1.0103200674057007, + "learning_rate": 2.846383902491363e-06, + "loss": 0.606, + "step": 3355 + }, + { + "epoch": 2.778575372722253, + "grad_norm": 0.9307713508605957, + "learning_rate": 2.845294786192163e-06, + "loss": 0.5969, + "step": 3356 + }, + { + "epoch": 2.779403644395362, + "grad_norm": 0.890837550163269, + "learning_rate": 2.84420560308164e-06, + "loss": 0.5826, + "step": 3357 + }, + { + "epoch": 2.7802319160684705, + "grad_norm": 0.8857460618019104, + "learning_rate": 2.843116353370542e-06, + "loss": 0.6005, + "step": 3358 + }, + { + "epoch": 2.7810601877415793, + "grad_norm": 0.8894275426864624, + "learning_rate": 2.8420270372696273e-06, + "loss": 0.6017, + "step": 3359 + }, + { + "epoch": 2.781888459414688, + "grad_norm": 0.9014497995376587, + "learning_rate": 2.8409376549896693e-06, + "loss": 0.5845, + "step": 3360 + }, + { + "epoch": 2.7827167310877967, + "grad_norm": 0.9019520282745361, + "learning_rate": 2.839848206741454e-06, + "loss": 0.5993, + "step": 3361 + }, + { + "epoch": 2.7835450027609054, + "grad_norm": 0.9594439268112183, + "learning_rate": 2.838758692735778e-06, + "loss": 0.5905, + "step": 3362 + }, + { + "epoch": 2.7843732744340146, + "grad_norm": 0.9144065976142883, + "learning_rate": 2.8376691131834525e-06, + "loss": 0.6006, + "step": 3363 + }, + { + "epoch": 2.785201546107123, + "grad_norm": 0.9487935900688171, + "learning_rate": 2.836579468295303e-06, + "loss": 0.5832, + "step": 3364 + }, + { + "epoch": 2.786029817780232, + "grad_norm": 0.8950932025909424, + "learning_rate": 2.835489758282163e-06, + "loss": 0.5792, + "step": 3365 + }, + { + "epoch": 2.786858089453341, + "grad_norm": 0.9522806406021118, + "learning_rate": 2.8343999833548814e-06, + "loss": 0.5891, + "step": 3366 + }, + { + "epoch": 2.7876863611264495, + "grad_norm": 0.964898407459259, + "learning_rate": 2.833310143724321e-06, + "loss": 0.572, + "step": 3367 + }, + { + "epoch": 2.7885146327995582, + "grad_norm": 0.9154083728790283, + "learning_rate": 2.832220239601355e-06, + "loss": 0.5982, + "step": 3368 + }, + { + "epoch": 2.789342904472667, + "grad_norm": 0.9093793034553528, + "learning_rate": 2.831130271196869e-06, + "loss": 0.5592, + "step": 3369 + }, + { + "epoch": 2.7901711761457757, + "grad_norm": 0.9114746451377869, + "learning_rate": 2.8300402387217628e-06, + "loss": 0.5531, + "step": 3370 + }, + { + "epoch": 2.7909994478188844, + "grad_norm": 0.959976077079773, + "learning_rate": 2.8289501423869464e-06, + "loss": 0.5932, + "step": 3371 + }, + { + "epoch": 2.7918277194919936, + "grad_norm": 0.9415104389190674, + "learning_rate": 2.8278599824033443e-06, + "loss": 0.5516, + "step": 3372 + }, + { + "epoch": 2.792655991165102, + "grad_norm": 0.9304796457290649, + "learning_rate": 2.826769758981892e-06, + "loss": 0.5786, + "step": 3373 + }, + { + "epoch": 2.793484262838211, + "grad_norm": 0.9369680285453796, + "learning_rate": 2.825679472333537e-06, + "loss": 0.6077, + "step": 3374 + }, + { + "epoch": 2.79431253451132, + "grad_norm": 0.9102757573127747, + "learning_rate": 2.8245891226692405e-06, + "loss": 0.606, + "step": 3375 + }, + { + "epoch": 2.7951408061844285, + "grad_norm": 0.921359121799469, + "learning_rate": 2.8234987101999747e-06, + "loss": 0.613, + "step": 3376 + }, + { + "epoch": 2.7959690778575372, + "grad_norm": 0.9242355227470398, + "learning_rate": 2.822408235136723e-06, + "loss": 0.5758, + "step": 3377 + }, + { + "epoch": 2.796797349530646, + "grad_norm": 0.9175317287445068, + "learning_rate": 2.8213176976904844e-06, + "loss": 0.5744, + "step": 3378 + }, + { + "epoch": 2.7976256212037547, + "grad_norm": 0.9504632949829102, + "learning_rate": 2.8202270980722663e-06, + "loss": 0.5871, + "step": 3379 + }, + { + "epoch": 2.7984538928768634, + "grad_norm": 0.9071774482727051, + "learning_rate": 2.8191364364930895e-06, + "loss": 0.6003, + "step": 3380 + }, + { + "epoch": 2.7992821645499726, + "grad_norm": 0.913815975189209, + "learning_rate": 2.8180457131639878e-06, + "loss": 0.596, + "step": 3381 + }, + { + "epoch": 2.8001104362230813, + "grad_norm": 0.9608022570610046, + "learning_rate": 2.8169549282960044e-06, + "loss": 0.6055, + "step": 3382 + }, + { + "epoch": 2.80093870789619, + "grad_norm": 2.0446290969848633, + "learning_rate": 2.815864082100198e-06, + "loss": 0.565, + "step": 3383 + }, + { + "epoch": 2.8017669795692988, + "grad_norm": 0.9219574928283691, + "learning_rate": 2.814773174787636e-06, + "loss": 0.5808, + "step": 3384 + }, + { + "epoch": 2.8025952512424075, + "grad_norm": 0.9542830586433411, + "learning_rate": 2.8136822065693987e-06, + "loss": 0.5992, + "step": 3385 + }, + { + "epoch": 2.803423522915516, + "grad_norm": 0.9478740096092224, + "learning_rate": 2.8125911776565785e-06, + "loss": 0.5705, + "step": 3386 + }, + { + "epoch": 2.804251794588625, + "grad_norm": 0.9299196600914001, + "learning_rate": 2.8115000882602792e-06, + "loss": 0.5773, + "step": 3387 + }, + { + "epoch": 2.805080066261734, + "grad_norm": 0.9167691469192505, + "learning_rate": 2.8104089385916168e-06, + "loss": 0.5789, + "step": 3388 + }, + { + "epoch": 2.8059083379348424, + "grad_norm": 0.9034630656242371, + "learning_rate": 2.8093177288617187e-06, + "loss": 0.5837, + "step": 3389 + }, + { + "epoch": 2.8067366096079516, + "grad_norm": 0.9147744178771973, + "learning_rate": 2.808226459281723e-06, + "loss": 0.5826, + "step": 3390 + }, + { + "epoch": 2.8075648812810603, + "grad_norm": 0.9159218072891235, + "learning_rate": 2.8071351300627803e-06, + "loss": 0.5931, + "step": 3391 + }, + { + "epoch": 2.808393152954169, + "grad_norm": 0.9260812401771545, + "learning_rate": 2.8060437414160546e-06, + "loss": 0.6182, + "step": 3392 + }, + { + "epoch": 2.8092214246272778, + "grad_norm": 0.8975473642349243, + "learning_rate": 2.804952293552717e-06, + "loss": 0.5786, + "step": 3393 + }, + { + "epoch": 2.8100496963003865, + "grad_norm": 0.9130522608757019, + "learning_rate": 2.8038607866839522e-06, + "loss": 0.572, + "step": 3394 + }, + { + "epoch": 2.810877967973495, + "grad_norm": 0.9170857667922974, + "learning_rate": 2.802769221020959e-06, + "loss": 0.569, + "step": 3395 + }, + { + "epoch": 2.811706239646604, + "grad_norm": 0.9313328266143799, + "learning_rate": 2.801677596774944e-06, + "loss": 0.568, + "step": 3396 + }, + { + "epoch": 2.812534511319713, + "grad_norm": 0.941814124584198, + "learning_rate": 2.8005859141571257e-06, + "loss": 0.5852, + "step": 3397 + }, + { + "epoch": 2.8133627829928214, + "grad_norm": 0.9179267883300781, + "learning_rate": 2.799494173378735e-06, + "loss": 0.5887, + "step": 3398 + }, + { + "epoch": 2.8141910546659306, + "grad_norm": 0.9193063378334045, + "learning_rate": 2.7984023746510135e-06, + "loss": 0.588, + "step": 3399 + }, + { + "epoch": 2.8150193263390393, + "grad_norm": 0.9381787776947021, + "learning_rate": 2.7973105181852145e-06, + "loss": 0.5938, + "step": 3400 + }, + { + "epoch": 2.815847598012148, + "grad_norm": 0.9631656408309937, + "learning_rate": 2.7962186041926015e-06, + "loss": 0.5828, + "step": 3401 + }, + { + "epoch": 2.8166758696852567, + "grad_norm": 0.9512215852737427, + "learning_rate": 2.7951266328844495e-06, + "loss": 0.5836, + "step": 3402 + }, + { + "epoch": 2.8175041413583655, + "grad_norm": 0.9171256422996521, + "learning_rate": 2.794034604472047e-06, + "loss": 0.574, + "step": 3403 + }, + { + "epoch": 2.818332413031474, + "grad_norm": 0.9122748374938965, + "learning_rate": 2.792942519166688e-06, + "loss": 0.6024, + "step": 3404 + }, + { + "epoch": 2.819160684704583, + "grad_norm": 0.9503622651100159, + "learning_rate": 2.7918503771796817e-06, + "loss": 0.5743, + "step": 3405 + }, + { + "epoch": 2.819988956377692, + "grad_norm": 0.9348991513252258, + "learning_rate": 2.7907581787223494e-06, + "loss": 0.5824, + "step": 3406 + }, + { + "epoch": 2.820817228050801, + "grad_norm": 1.0846208333969116, + "learning_rate": 2.7896659240060192e-06, + "loss": 0.5995, + "step": 3407 + }, + { + "epoch": 2.8216454997239095, + "grad_norm": 0.9554671049118042, + "learning_rate": 2.7885736132420326e-06, + "loss": 0.5903, + "step": 3408 + }, + { + "epoch": 2.8224737713970183, + "grad_norm": 0.9836533665657043, + "learning_rate": 2.7874812466417434e-06, + "loss": 0.5982, + "step": 3409 + }, + { + "epoch": 2.823302043070127, + "grad_norm": 0.9414785504341125, + "learning_rate": 2.786388824416511e-06, + "loss": 0.5743, + "step": 3410 + }, + { + "epoch": 2.8241303147432357, + "grad_norm": 0.9228699207305908, + "learning_rate": 2.7852963467777117e-06, + "loss": 0.5879, + "step": 3411 + }, + { + "epoch": 2.8249585864163445, + "grad_norm": 0.919826090335846, + "learning_rate": 2.7842038139367288e-06, + "loss": 0.6023, + "step": 3412 + }, + { + "epoch": 2.825786858089453, + "grad_norm": 0.9112126231193542, + "learning_rate": 2.783111226104957e-06, + "loss": 0.578, + "step": 3413 + }, + { + "epoch": 2.826615129762562, + "grad_norm": 0.9262592792510986, + "learning_rate": 2.7820185834938024e-06, + "loss": 0.6063, + "step": 3414 + }, + { + "epoch": 2.827443401435671, + "grad_norm": 0.9020029902458191, + "learning_rate": 2.780925886314681e-06, + "loss": 0.5837, + "step": 3415 + }, + { + "epoch": 2.82827167310878, + "grad_norm": 0.9186374545097351, + "learning_rate": 2.7798331347790185e-06, + "loss": 0.5801, + "step": 3416 + }, + { + "epoch": 2.8290999447818885, + "grad_norm": 0.9116926193237305, + "learning_rate": 2.7787403290982535e-06, + "loss": 0.5735, + "step": 3417 + }, + { + "epoch": 2.8299282164549973, + "grad_norm": 0.9124463796615601, + "learning_rate": 2.7776474694838333e-06, + "loss": 0.6064, + "step": 3418 + }, + { + "epoch": 2.830756488128106, + "grad_norm": 0.9450061321258545, + "learning_rate": 2.7765545561472153e-06, + "loss": 0.5885, + "step": 3419 + }, + { + "epoch": 2.8315847598012147, + "grad_norm": 0.8993310928344727, + "learning_rate": 2.77546158929987e-06, + "loss": 0.5767, + "step": 3420 + }, + { + "epoch": 2.8324130314743234, + "grad_norm": 0.9315094947814941, + "learning_rate": 2.7743685691532734e-06, + "loss": 0.595, + "step": 3421 + }, + { + "epoch": 2.8332413031474326, + "grad_norm": 0.9254942536354065, + "learning_rate": 2.773275495918917e-06, + "loss": 0.6071, + "step": 3422 + }, + { + "epoch": 2.834069574820541, + "grad_norm": 0.9551193118095398, + "learning_rate": 2.7721823698083e-06, + "loss": 0.588, + "step": 3423 + }, + { + "epoch": 2.83489784649365, + "grad_norm": 0.9356088042259216, + "learning_rate": 2.7710891910329317e-06, + "loss": 0.5917, + "step": 3424 + }, + { + "epoch": 2.835726118166759, + "grad_norm": 0.9254491925239563, + "learning_rate": 2.7699959598043315e-06, + "loss": 0.5754, + "step": 3425 + }, + { + "epoch": 2.8365543898398675, + "grad_norm": 0.928747832775116, + "learning_rate": 2.7689026763340302e-06, + "loss": 0.5925, + "step": 3426 + }, + { + "epoch": 2.8373826615129762, + "grad_norm": 0.9232720732688904, + "learning_rate": 2.7678093408335677e-06, + "loss": 0.594, + "step": 3427 + }, + { + "epoch": 2.838210933186085, + "grad_norm": 0.9319408535957336, + "learning_rate": 2.766715953514494e-06, + "loss": 0.607, + "step": 3428 + }, + { + "epoch": 2.8390392048591937, + "grad_norm": 0.9321261644363403, + "learning_rate": 2.76562251458837e-06, + "loss": 0.5839, + "step": 3429 + }, + { + "epoch": 2.8398674765323024, + "grad_norm": 0.8851343393325806, + "learning_rate": 2.7645290242667648e-06, + "loss": 0.5929, + "step": 3430 + }, + { + "epoch": 2.8406957482054116, + "grad_norm": 0.9318198561668396, + "learning_rate": 2.76343548276126e-06, + "loss": 0.5758, + "step": 3431 + }, + { + "epoch": 2.84152401987852, + "grad_norm": 0.9212138056755066, + "learning_rate": 2.7623418902834455e-06, + "loss": 0.5689, + "step": 3432 + }, + { + "epoch": 2.842352291551629, + "grad_norm": 0.8917051553726196, + "learning_rate": 2.761248247044919e-06, + "loss": 0.5915, + "step": 3433 + }, + { + "epoch": 2.843180563224738, + "grad_norm": 0.9266699552536011, + "learning_rate": 2.760154553257294e-06, + "loss": 0.5643, + "step": 3434 + }, + { + "epoch": 2.8440088348978465, + "grad_norm": 0.9837039709091187, + "learning_rate": 2.7590608091321867e-06, + "loss": 0.6135, + "step": 3435 + }, + { + "epoch": 2.8448371065709552, + "grad_norm": 0.9872124195098877, + "learning_rate": 2.7579670148812278e-06, + "loss": 0.5649, + "step": 3436 + }, + { + "epoch": 2.845665378244064, + "grad_norm": 0.9581724405288696, + "learning_rate": 2.756873170716057e-06, + "loss": 0.5833, + "step": 3437 + }, + { + "epoch": 2.8464936499171727, + "grad_norm": 0.9363225698471069, + "learning_rate": 2.7557792768483223e-06, + "loss": 0.5852, + "step": 3438 + }, + { + "epoch": 2.8473219215902814, + "grad_norm": 0.9177983403205872, + "learning_rate": 2.7546853334896813e-06, + "loss": 0.5959, + "step": 3439 + }, + { + "epoch": 2.8481501932633906, + "grad_norm": 0.9081945419311523, + "learning_rate": 2.7535913408518023e-06, + "loss": 0.5862, + "step": 3440 + }, + { + "epoch": 2.8489784649364993, + "grad_norm": 0.928897500038147, + "learning_rate": 2.7524972991463626e-06, + "loss": 0.5978, + "step": 3441 + }, + { + "epoch": 2.849806736609608, + "grad_norm": 0.9136974811553955, + "learning_rate": 2.75140320858505e-06, + "loss": 0.5926, + "step": 3442 + }, + { + "epoch": 2.8506350082827168, + "grad_norm": 0.9227224588394165, + "learning_rate": 2.7503090693795605e-06, + "loss": 0.6073, + "step": 3443 + }, + { + "epoch": 2.8514632799558255, + "grad_norm": 0.9390876889228821, + "learning_rate": 2.749214881741598e-06, + "loss": 0.5757, + "step": 3444 + }, + { + "epoch": 2.852291551628934, + "grad_norm": 0.9460882544517517, + "learning_rate": 2.74812064588288e-06, + "loss": 0.6172, + "step": 3445 + }, + { + "epoch": 2.853119823302043, + "grad_norm": 0.9087659120559692, + "learning_rate": 2.74702636201513e-06, + "loss": 0.5937, + "step": 3446 + }, + { + "epoch": 2.853948094975152, + "grad_norm": 0.8995270133018494, + "learning_rate": 2.745932030350081e-06, + "loss": 0.5798, + "step": 3447 + }, + { + "epoch": 2.8547763666482604, + "grad_norm": 0.9307460188865662, + "learning_rate": 2.744837651099478e-06, + "loss": 0.6009, + "step": 3448 + }, + { + "epoch": 2.8556046383213696, + "grad_norm": 0.9634310007095337, + "learning_rate": 2.7437432244750705e-06, + "loss": 0.5985, + "step": 3449 + }, + { + "epoch": 2.8564329099944783, + "grad_norm": 0.9647858142852783, + "learning_rate": 2.742648750688622e-06, + "loss": 0.5718, + "step": 3450 + }, + { + "epoch": 2.857261181667587, + "grad_norm": 0.9150539040565491, + "learning_rate": 2.7415542299519016e-06, + "loss": 0.5964, + "step": 3451 + }, + { + "epoch": 2.8580894533406958, + "grad_norm": 0.9129363298416138, + "learning_rate": 2.7404596624766892e-06, + "loss": 0.5514, + "step": 3452 + }, + { + "epoch": 2.8589177250138045, + "grad_norm": 0.9430226683616638, + "learning_rate": 2.739365048474774e-06, + "loss": 0.5795, + "step": 3453 + }, + { + "epoch": 2.859745996686913, + "grad_norm": 0.9402962327003479, + "learning_rate": 2.738270388157954e-06, + "loss": 0.5728, + "step": 3454 + }, + { + "epoch": 2.860574268360022, + "grad_norm": 0.9169720411300659, + "learning_rate": 2.7371756817380346e-06, + "loss": 0.6171, + "step": 3455 + }, + { + "epoch": 2.861402540033131, + "grad_norm": 0.9437370896339417, + "learning_rate": 2.7360809294268315e-06, + "loss": 0.5827, + "step": 3456 + }, + { + "epoch": 2.8622308117062394, + "grad_norm": 0.8999599814414978, + "learning_rate": 2.7349861314361693e-06, + "loss": 0.5847, + "step": 3457 + }, + { + "epoch": 2.8630590833793486, + "grad_norm": 0.9228375554084778, + "learning_rate": 2.7338912879778814e-06, + "loss": 0.6069, + "step": 3458 + }, + { + "epoch": 2.8638873550524573, + "grad_norm": 0.9543429017066956, + "learning_rate": 2.732796399263809e-06, + "loss": 0.6128, + "step": 3459 + }, + { + "epoch": 2.864715626725566, + "grad_norm": 0.923698902130127, + "learning_rate": 2.7317014655058038e-06, + "loss": 0.5714, + "step": 3460 + }, + { + "epoch": 2.8655438983986747, + "grad_norm": 0.95343017578125, + "learning_rate": 2.730606486915725e-06, + "loss": 0.5821, + "step": 3461 + }, + { + "epoch": 2.8663721700717835, + "grad_norm": 0.9078912734985352, + "learning_rate": 2.729511463705441e-06, + "loss": 0.5833, + "step": 3462 + }, + { + "epoch": 2.867200441744892, + "grad_norm": 0.9045196771621704, + "learning_rate": 2.7284163960868274e-06, + "loss": 0.5647, + "step": 3463 + }, + { + "epoch": 2.868028713418001, + "grad_norm": 0.9145541787147522, + "learning_rate": 2.727321284271771e-06, + "loss": 0.5731, + "step": 3464 + }, + { + "epoch": 2.86885698509111, + "grad_norm": 0.9013593792915344, + "learning_rate": 2.726226128472166e-06, + "loss": 0.5901, + "step": 3465 + }, + { + "epoch": 2.869685256764219, + "grad_norm": 0.9561366438865662, + "learning_rate": 2.7251309288999123e-06, + "loss": 0.585, + "step": 3466 + }, + { + "epoch": 2.8705135284373275, + "grad_norm": 0.9313666224479675, + "learning_rate": 2.724035685766923e-06, + "loss": 0.5755, + "step": 3467 + }, + { + "epoch": 2.8713418001104363, + "grad_norm": 0.9003223776817322, + "learning_rate": 2.722940399285118e-06, + "loss": 0.5961, + "step": 3468 + }, + { + "epoch": 2.872170071783545, + "grad_norm": 0.9185585975646973, + "learning_rate": 2.721845069666422e-06, + "loss": 0.5741, + "step": 3469 + }, + { + "epoch": 2.8729983434566537, + "grad_norm": 0.9295309782028198, + "learning_rate": 2.7207496971227743e-06, + "loss": 0.588, + "step": 3470 + }, + { + "epoch": 2.8738266151297625, + "grad_norm": 0.9500821232795715, + "learning_rate": 2.7196542818661188e-06, + "loss": 0.6026, + "step": 3471 + }, + { + "epoch": 2.874654886802871, + "grad_norm": 0.9265774488449097, + "learning_rate": 2.718558824108406e-06, + "loss": 0.5857, + "step": 3472 + }, + { + "epoch": 2.87548315847598, + "grad_norm": 0.9159520864486694, + "learning_rate": 2.717463324061599e-06, + "loss": 0.6023, + "step": 3473 + }, + { + "epoch": 2.876311430149089, + "grad_norm": 0.9114859700202942, + "learning_rate": 2.7163677819376653e-06, + "loss": 0.6144, + "step": 3474 + }, + { + "epoch": 2.877139701822198, + "grad_norm": 0.9181182980537415, + "learning_rate": 2.7152721979485824e-06, + "loss": 0.5853, + "step": 3475 + }, + { + "epoch": 2.8779679734953065, + "grad_norm": 0.9219772219657898, + "learning_rate": 2.7141765723063366e-06, + "loss": 0.5824, + "step": 3476 + }, + { + "epoch": 2.8787962451684153, + "grad_norm": 0.925654411315918, + "learning_rate": 2.713080905222921e-06, + "loss": 0.5859, + "step": 3477 + }, + { + "epoch": 2.879624516841524, + "grad_norm": 0.927891731262207, + "learning_rate": 2.711985196910336e-06, + "loss": 0.6047, + "step": 3478 + }, + { + "epoch": 2.8804527885146327, + "grad_norm": 0.9178606271743774, + "learning_rate": 2.7108894475805912e-06, + "loss": 0.5863, + "step": 3479 + }, + { + "epoch": 2.8812810601877414, + "grad_norm": 0.9341709613800049, + "learning_rate": 2.7097936574457046e-06, + "loss": 0.5718, + "step": 3480 + }, + { + "epoch": 2.8821093318608506, + "grad_norm": 0.9023662805557251, + "learning_rate": 2.7086978267177007e-06, + "loss": 0.593, + "step": 3481 + }, + { + "epoch": 2.882937603533959, + "grad_norm": 0.9269591569900513, + "learning_rate": 2.7076019556086136e-06, + "loss": 0.5955, + "step": 3482 + }, + { + "epoch": 2.883765875207068, + "grad_norm": 0.9572223424911499, + "learning_rate": 2.7065060443304823e-06, + "loss": 0.5698, + "step": 3483 + }, + { + "epoch": 2.884594146880177, + "grad_norm": 0.9251832365989685, + "learning_rate": 2.705410093095357e-06, + "loss": 0.5903, + "step": 3484 + }, + { + "epoch": 2.8854224185532855, + "grad_norm": 0.959985077381134, + "learning_rate": 2.7043141021152937e-06, + "loss": 0.5922, + "step": 3485 + }, + { + "epoch": 2.8862506902263942, + "grad_norm": 0.9356866478919983, + "learning_rate": 2.703218071602355e-06, + "loss": 0.576, + "step": 3486 + }, + { + "epoch": 2.887078961899503, + "grad_norm": 0.9041283130645752, + "learning_rate": 2.702122001768615e-06, + "loss": 0.5748, + "step": 3487 + }, + { + "epoch": 2.8879072335726117, + "grad_norm": 0.9056842923164368, + "learning_rate": 2.7010258928261514e-06, + "loss": 0.5795, + "step": 3488 + }, + { + "epoch": 2.8887355052457204, + "grad_norm": 0.9829550385475159, + "learning_rate": 2.6999297449870522e-06, + "loss": 0.6017, + "step": 3489 + }, + { + "epoch": 2.8895637769188296, + "grad_norm": 0.9483397603034973, + "learning_rate": 2.6988335584634103e-06, + "loss": 0.5849, + "step": 3490 + }, + { + "epoch": 2.890392048591938, + "grad_norm": 0.9553128480911255, + "learning_rate": 2.6977373334673286e-06, + "loss": 0.5518, + "step": 3491 + }, + { + "epoch": 2.891220320265047, + "grad_norm": 0.9298998713493347, + "learning_rate": 2.6966410702109153e-06, + "loss": 0.5947, + "step": 3492 + }, + { + "epoch": 2.8920485919381558, + "grad_norm": 0.9143223166465759, + "learning_rate": 2.6955447689062893e-06, + "loss": 0.5757, + "step": 3493 + }, + { + "epoch": 2.8928768636112645, + "grad_norm": 0.9143601059913635, + "learning_rate": 2.6944484297655727e-06, + "loss": 0.5867, + "step": 3494 + }, + { + "epoch": 2.8937051352843732, + "grad_norm": 0.9283237457275391, + "learning_rate": 2.693352053000898e-06, + "loss": 0.5928, + "step": 3495 + }, + { + "epoch": 2.894533406957482, + "grad_norm": 0.8984352946281433, + "learning_rate": 2.692255638824403e-06, + "loss": 0.5904, + "step": 3496 + }, + { + "epoch": 2.8953616786305907, + "grad_norm": 0.9056443572044373, + "learning_rate": 2.6911591874482334e-06, + "loss": 0.5847, + "step": 3497 + }, + { + "epoch": 2.8961899503036994, + "grad_norm": 0.9254133701324463, + "learning_rate": 2.6900626990845442e-06, + "loss": 0.608, + "step": 3498 + }, + { + "epoch": 2.8970182219768086, + "grad_norm": 0.9658727049827576, + "learning_rate": 2.6889661739454946e-06, + "loss": 0.6063, + "step": 3499 + }, + { + "epoch": 2.8978464936499173, + "grad_norm": 0.9350060820579529, + "learning_rate": 2.6878696122432505e-06, + "loss": 0.5902, + "step": 3500 + }, + { + "epoch": 2.898674765323026, + "grad_norm": 0.9070295691490173, + "learning_rate": 2.686773014189989e-06, + "loss": 0.5883, + "step": 3501 + }, + { + "epoch": 2.8995030369961348, + "grad_norm": 0.9471203088760376, + "learning_rate": 2.6856763799978895e-06, + "loss": 0.6138, + "step": 3502 + }, + { + "epoch": 2.9003313086692435, + "grad_norm": 0.901057243347168, + "learning_rate": 2.6845797098791415e-06, + "loss": 0.5925, + "step": 3503 + }, + { + "epoch": 2.901159580342352, + "grad_norm": 0.9290406107902527, + "learning_rate": 2.683483004045941e-06, + "loss": 0.5728, + "step": 3504 + }, + { + "epoch": 2.901987852015461, + "grad_norm": 0.9371906518936157, + "learning_rate": 2.682386262710489e-06, + "loss": 0.5894, + "step": 3505 + }, + { + "epoch": 2.90281612368857, + "grad_norm": 0.9183947443962097, + "learning_rate": 2.6812894860849953e-06, + "loss": 0.5959, + "step": 3506 + }, + { + "epoch": 2.9036443953616784, + "grad_norm": 0.9351047277450562, + "learning_rate": 2.6801926743816765e-06, + "loss": 0.5863, + "step": 3507 + }, + { + "epoch": 2.9044726670347876, + "grad_norm": 0.93602454662323, + "learning_rate": 2.6790958278127536e-06, + "loss": 0.5845, + "step": 3508 + }, + { + "epoch": 2.9053009387078963, + "grad_norm": 0.9618241786956787, + "learning_rate": 2.677998946590459e-06, + "loss": 0.6068, + "step": 3509 + }, + { + "epoch": 2.906129210381005, + "grad_norm": 0.9157661199569702, + "learning_rate": 2.6769020309270277e-06, + "loss": 0.5608, + "step": 3510 + }, + { + "epoch": 2.9069574820541138, + "grad_norm": 0.9157334566116333, + "learning_rate": 2.6758050810347012e-06, + "loss": 0.5883, + "step": 3511 + }, + { + "epoch": 2.9077857537272225, + "grad_norm": 0.9259895086288452, + "learning_rate": 2.674708097125731e-06, + "loss": 0.578, + "step": 3512 + }, + { + "epoch": 2.908614025400331, + "grad_norm": 0.9327818155288696, + "learning_rate": 2.6736110794123727e-06, + "loss": 0.6157, + "step": 3513 + }, + { + "epoch": 2.90944229707344, + "grad_norm": 0.9186248779296875, + "learning_rate": 2.6725140281068882e-06, + "loss": 0.5901, + "step": 3514 + }, + { + "epoch": 2.910270568746549, + "grad_norm": 0.9106549620628357, + "learning_rate": 2.671416943421548e-06, + "loss": 0.5983, + "step": 3515 + }, + { + "epoch": 2.9110988404196574, + "grad_norm": 0.9274003505706787, + "learning_rate": 2.6703198255686275e-06, + "loss": 0.5886, + "step": 3516 + }, + { + "epoch": 2.9119271120927666, + "grad_norm": 0.9015113115310669, + "learning_rate": 2.6692226747604082e-06, + "loss": 0.5979, + "step": 3517 + }, + { + "epoch": 2.9127553837658753, + "grad_norm": 0.9261288046836853, + "learning_rate": 2.6681254912091788e-06, + "loss": 0.5741, + "step": 3518 + }, + { + "epoch": 2.913583655438984, + "grad_norm": 0.8937399983406067, + "learning_rate": 2.6670282751272335e-06, + "loss": 0.5726, + "step": 3519 + }, + { + "epoch": 2.9144119271120927, + "grad_norm": 0.9202954769134521, + "learning_rate": 2.6659310267268744e-06, + "loss": 0.5788, + "step": 3520 + }, + { + "epoch": 2.9152401987852015, + "grad_norm": 0.9400494694709778, + "learning_rate": 2.6648337462204087e-06, + "loss": 0.568, + "step": 3521 + }, + { + "epoch": 2.91606847045831, + "grad_norm": 0.9291467070579529, + "learning_rate": 2.6637364338201497e-06, + "loss": 0.589, + "step": 3522 + }, + { + "epoch": 2.916896742131419, + "grad_norm": 0.9186866879463196, + "learning_rate": 2.662639089738417e-06, + "loss": 0.59, + "step": 3523 + }, + { + "epoch": 2.917725013804528, + "grad_norm": 0.9596429467201233, + "learning_rate": 2.6615417141875367e-06, + "loss": 0.5759, + "step": 3524 + }, + { + "epoch": 2.918553285477637, + "grad_norm": 0.9279852509498596, + "learning_rate": 2.6604443073798395e-06, + "loss": 0.5969, + "step": 3525 + }, + { + "epoch": 2.9193815571507455, + "grad_norm": 0.9540870189666748, + "learning_rate": 2.6593468695276657e-06, + "loss": 0.5696, + "step": 3526 + }, + { + "epoch": 2.9202098288238543, + "grad_norm": 0.9415677785873413, + "learning_rate": 2.658249400843358e-06, + "loss": 0.5973, + "step": 3527 + }, + { + "epoch": 2.921038100496963, + "grad_norm": 0.8950624465942383, + "learning_rate": 2.657151901539266e-06, + "loss": 0.5952, + "step": 3528 + }, + { + "epoch": 2.9218663721700717, + "grad_norm": 0.8927432298660278, + "learning_rate": 2.656054371827746e-06, + "loss": 0.59, + "step": 3529 + }, + { + "epoch": 2.9226946438431805, + "grad_norm": 0.9516159892082214, + "learning_rate": 2.65495681192116e-06, + "loss": 0.6024, + "step": 3530 + }, + { + "epoch": 2.923522915516289, + "grad_norm": 1.0094369649887085, + "learning_rate": 2.653859222031875e-06, + "loss": 0.596, + "step": 3531 + }, + { + "epoch": 2.924351187189398, + "grad_norm": 0.9162117838859558, + "learning_rate": 2.6527616023722647e-06, + "loss": 0.5739, + "step": 3532 + }, + { + "epoch": 2.925179458862507, + "grad_norm": 0.9844574332237244, + "learning_rate": 2.651663953154709e-06, + "loss": 0.6028, + "step": 3533 + }, + { + "epoch": 2.926007730535616, + "grad_norm": 0.9091664552688599, + "learning_rate": 2.6505662745915915e-06, + "loss": 0.5867, + "step": 3534 + }, + { + "epoch": 2.9268360022087245, + "grad_norm": 0.9107911586761475, + "learning_rate": 2.649468566895304e-06, + "loss": 0.5985, + "step": 3535 + }, + { + "epoch": 2.9276642738818333, + "grad_norm": 0.9392855167388916, + "learning_rate": 2.648370830278241e-06, + "loss": 0.5995, + "step": 3536 + }, + { + "epoch": 2.928492545554942, + "grad_norm": 0.9228532910346985, + "learning_rate": 2.6472730649528068e-06, + "loss": 0.5928, + "step": 3537 + }, + { + "epoch": 2.9293208172280507, + "grad_norm": 0.9555805325508118, + "learning_rate": 2.646175271131407e-06, + "loss": 0.606, + "step": 3538 + }, + { + "epoch": 2.9301490889011594, + "grad_norm": 0.9101611971855164, + "learning_rate": 2.6450774490264547e-06, + "loss": 0.5881, + "step": 3539 + }, + { + "epoch": 2.9309773605742686, + "grad_norm": 0.9572548866271973, + "learning_rate": 2.6439795988503685e-06, + "loss": 0.5861, + "step": 3540 + }, + { + "epoch": 2.931805632247377, + "grad_norm": 0.9441561102867126, + "learning_rate": 2.6428817208155728e-06, + "loss": 0.594, + "step": 3541 + }, + { + "epoch": 2.932633903920486, + "grad_norm": 0.9282614588737488, + "learning_rate": 2.6417838151344945e-06, + "loss": 0.5991, + "step": 3542 + }, + { + "epoch": 2.933462175593595, + "grad_norm": 0.9630649089813232, + "learning_rate": 2.640685882019571e-06, + "loss": 0.6147, + "step": 3543 + }, + { + "epoch": 2.9342904472667035, + "grad_norm": 0.9284330606460571, + "learning_rate": 2.63958792168324e-06, + "loss": 0.6021, + "step": 3544 + }, + { + "epoch": 2.9351187189398122, + "grad_norm": 0.9337431788444519, + "learning_rate": 2.638489934337948e-06, + "loss": 0.602, + "step": 3545 + }, + { + "epoch": 2.935946990612921, + "grad_norm": 0.8977872729301453, + "learning_rate": 2.637391920196144e-06, + "loss": 0.5901, + "step": 3546 + }, + { + "epoch": 2.9367752622860297, + "grad_norm": 0.962690532207489, + "learning_rate": 2.636293879470284e-06, + "loss": 0.5793, + "step": 3547 + }, + { + "epoch": 2.9376035339591384, + "grad_norm": 0.8999794721603394, + "learning_rate": 2.6351958123728294e-06, + "loss": 0.6027, + "step": 3548 + }, + { + "epoch": 2.9384318056322476, + "grad_norm": 0.9200299382209778, + "learning_rate": 2.6340977191162457e-06, + "loss": 0.5985, + "step": 3549 + }, + { + "epoch": 2.9392600773053563, + "grad_norm": 0.9161362648010254, + "learning_rate": 2.6329995999130032e-06, + "loss": 0.595, + "step": 3550 + }, + { + "epoch": 2.940088348978465, + "grad_norm": 0.9192216992378235, + "learning_rate": 2.631901454975578e-06, + "loss": 0.5844, + "step": 3551 + }, + { + "epoch": 2.9409166206515738, + "grad_norm": 0.9136098623275757, + "learning_rate": 2.6308032845164503e-06, + "loss": 0.5932, + "step": 3552 + }, + { + "epoch": 2.9417448923246825, + "grad_norm": 0.9625688791275024, + "learning_rate": 2.629705088748106e-06, + "loss": 0.5811, + "step": 3553 + }, + { + "epoch": 2.9425731639977912, + "grad_norm": 0.9300727248191833, + "learning_rate": 2.6286068678830375e-06, + "loss": 0.6035, + "step": 3554 + }, + { + "epoch": 2.9434014356709, + "grad_norm": 0.9947618246078491, + "learning_rate": 2.627508622133738e-06, + "loss": 0.5822, + "step": 3555 + }, + { + "epoch": 2.9442297073440087, + "grad_norm": 0.9255596995353699, + "learning_rate": 2.626410351712709e-06, + "loss": 0.5824, + "step": 3556 + }, + { + "epoch": 2.9450579790171174, + "grad_norm": 0.9395889043807983, + "learning_rate": 2.6253120568324557e-06, + "loss": 0.6121, + "step": 3557 + }, + { + "epoch": 2.9458862506902266, + "grad_norm": 0.9438552856445312, + "learning_rate": 2.6242137377054865e-06, + "loss": 0.5843, + "step": 3558 + }, + { + "epoch": 2.9467145223633353, + "grad_norm": 0.9153048992156982, + "learning_rate": 2.623115394544318e-06, + "loss": 0.5893, + "step": 3559 + }, + { + "epoch": 2.947542794036444, + "grad_norm": 0.9119593501091003, + "learning_rate": 2.6220170275614677e-06, + "loss": 0.6047, + "step": 3560 + }, + { + "epoch": 2.9483710657095528, + "grad_norm": 0.9174615144729614, + "learning_rate": 2.6209186369694606e-06, + "loss": 0.5949, + "step": 3561 + }, + { + "epoch": 2.9491993373826615, + "grad_norm": 0.9030957818031311, + "learning_rate": 2.619820222980824e-06, + "loss": 0.6146, + "step": 3562 + }, + { + "epoch": 2.95002760905577, + "grad_norm": 0.9036881923675537, + "learning_rate": 2.618721785808091e-06, + "loss": 0.5946, + "step": 3563 + }, + { + "epoch": 2.950855880728879, + "grad_norm": 0.8928536772727966, + "learning_rate": 2.617623325663799e-06, + "loss": 0.5679, + "step": 3564 + }, + { + "epoch": 2.951684152401988, + "grad_norm": 0.9225850701332092, + "learning_rate": 2.6165248427604907e-06, + "loss": 0.5938, + "step": 3565 + }, + { + "epoch": 2.9525124240750964, + "grad_norm": 0.9023733139038086, + "learning_rate": 2.6154263373107113e-06, + "loss": 0.5896, + "step": 3566 + }, + { + "epoch": 2.9533406957482056, + "grad_norm": 0.9055750370025635, + "learning_rate": 2.6143278095270113e-06, + "loss": 0.6007, + "step": 3567 + }, + { + "epoch": 2.9541689674213143, + "grad_norm": 0.8988812565803528, + "learning_rate": 2.613229259621947e-06, + "loss": 0.6097, + "step": 3568 + }, + { + "epoch": 2.954997239094423, + "grad_norm": 0.9200608730316162, + "learning_rate": 2.6121306878080753e-06, + "loss": 0.58, + "step": 3569 + }, + { + "epoch": 2.9558255107675317, + "grad_norm": 0.9378446936607361, + "learning_rate": 2.6110320942979606e-06, + "loss": 0.6043, + "step": 3570 + }, + { + "epoch": 2.9566537824406405, + "grad_norm": 0.9543853402137756, + "learning_rate": 2.6099334793041714e-06, + "loss": 0.58, + "step": 3571 + }, + { + "epoch": 2.957482054113749, + "grad_norm": 0.9590248465538025, + "learning_rate": 2.6088348430392784e-06, + "loss": 0.6004, + "step": 3572 + }, + { + "epoch": 2.958310325786858, + "grad_norm": 0.980459988117218, + "learning_rate": 2.607736185715858e-06, + "loss": 0.6118, + "step": 3573 + }, + { + "epoch": 2.959138597459967, + "grad_norm": 0.9050843715667725, + "learning_rate": 2.60663750754649e-06, + "loss": 0.5913, + "step": 3574 + }, + { + "epoch": 2.9599668691330754, + "grad_norm": 0.9189395904541016, + "learning_rate": 2.6055388087437576e-06, + "loss": 0.5997, + "step": 3575 + }, + { + "epoch": 2.9607951408061846, + "grad_norm": 0.9273754954338074, + "learning_rate": 2.6044400895202502e-06, + "loss": 0.586, + "step": 3576 + }, + { + "epoch": 2.9616234124792933, + "grad_norm": 0.90342116355896, + "learning_rate": 2.6033413500885594e-06, + "loss": 0.5846, + "step": 3577 + }, + { + "epoch": 2.962451684152402, + "grad_norm": 0.9073745012283325, + "learning_rate": 2.6022425906612803e-06, + "loss": 0.5899, + "step": 3578 + }, + { + "epoch": 2.9632799558255107, + "grad_norm": 0.9221594333648682, + "learning_rate": 2.6011438114510135e-06, + "loss": 0.5837, + "step": 3579 + }, + { + "epoch": 2.9641082274986195, + "grad_norm": 0.9228225350379944, + "learning_rate": 2.6000450126703613e-06, + "loss": 0.5761, + "step": 3580 + }, + { + "epoch": 2.964936499171728, + "grad_norm": 0.9136195182800293, + "learning_rate": 2.5989461945319315e-06, + "loss": 0.5977, + "step": 3581 + }, + { + "epoch": 2.965764770844837, + "grad_norm": 0.9152894616127014, + "learning_rate": 2.5978473572483366e-06, + "loss": 0.5829, + "step": 3582 + }, + { + "epoch": 2.966593042517946, + "grad_norm": 0.9124417304992676, + "learning_rate": 2.596748501032189e-06, + "loss": 0.5718, + "step": 3583 + }, + { + "epoch": 2.967421314191055, + "grad_norm": 0.9212865829467773, + "learning_rate": 2.5956496260961094e-06, + "loss": 0.6056, + "step": 3584 + }, + { + "epoch": 2.9682495858641635, + "grad_norm": 0.9101969003677368, + "learning_rate": 2.594550732652718e-06, + "loss": 0.6075, + "step": 3585 + }, + { + "epoch": 2.9690778575372723, + "grad_norm": 0.9225336313247681, + "learning_rate": 2.5934518209146414e-06, + "loss": 0.5894, + "step": 3586 + }, + { + "epoch": 2.969906129210381, + "grad_norm": 0.9271754026412964, + "learning_rate": 2.592352891094509e-06, + "loss": 0.5939, + "step": 3587 + }, + { + "epoch": 2.9707344008834897, + "grad_norm": 0.9217578768730164, + "learning_rate": 2.591253943404953e-06, + "loss": 0.5858, + "step": 3588 + }, + { + "epoch": 2.9715626725565985, + "grad_norm": 0.8871975541114807, + "learning_rate": 2.5901549780586085e-06, + "loss": 0.6052, + "step": 3589 + }, + { + "epoch": 2.9723909442297076, + "grad_norm": 0.9371136426925659, + "learning_rate": 2.5890559952681177e-06, + "loss": 0.6069, + "step": 3590 + }, + { + "epoch": 2.973219215902816, + "grad_norm": 0.9185692667961121, + "learning_rate": 2.5879569952461215e-06, + "loss": 0.5913, + "step": 3591 + }, + { + "epoch": 2.974047487575925, + "grad_norm": 0.9043979048728943, + "learning_rate": 2.586857978205266e-06, + "loss": 0.5747, + "step": 3592 + }, + { + "epoch": 2.974875759249034, + "grad_norm": 0.8951820731163025, + "learning_rate": 2.5857589443582014e-06, + "loss": 0.5904, + "step": 3593 + }, + { + "epoch": 2.9757040309221425, + "grad_norm": 0.9490121006965637, + "learning_rate": 2.5846598939175816e-06, + "loss": 0.583, + "step": 3594 + }, + { + "epoch": 2.9765323025952513, + "grad_norm": 0.9266814589500427, + "learning_rate": 2.583560827096061e-06, + "loss": 0.6039, + "step": 3595 + }, + { + "epoch": 2.97736057426836, + "grad_norm": 0.9517508149147034, + "learning_rate": 2.582461744106299e-06, + "loss": 0.572, + "step": 3596 + }, + { + "epoch": 2.9781888459414687, + "grad_norm": 0.9260841608047485, + "learning_rate": 2.581362645160958e-06, + "loss": 0.5898, + "step": 3597 + }, + { + "epoch": 2.9790171176145774, + "grad_norm": 0.9134445190429688, + "learning_rate": 2.5802635304727043e-06, + "loss": 0.6037, + "step": 3598 + }, + { + "epoch": 2.9798453892876866, + "grad_norm": 0.9307652115821838, + "learning_rate": 2.5791644002542054e-06, + "loss": 0.5941, + "step": 3599 + }, + { + "epoch": 2.980673660960795, + "grad_norm": 0.9665980935096741, + "learning_rate": 2.578065254718133e-06, + "loss": 0.5816, + "step": 3600 + }, + { + "epoch": 2.981501932633904, + "grad_norm": 0.9137181639671326, + "learning_rate": 2.5769660940771614e-06, + "loss": 0.5962, + "step": 3601 + }, + { + "epoch": 2.982330204307013, + "grad_norm": 0.9031209349632263, + "learning_rate": 2.575866918543968e-06, + "loss": 0.5853, + "step": 3602 + }, + { + "epoch": 2.9831584759801215, + "grad_norm": 0.9292592406272888, + "learning_rate": 2.5747677283312332e-06, + "loss": 0.5639, + "step": 3603 + }, + { + "epoch": 2.9839867476532302, + "grad_norm": 0.9193013906478882, + "learning_rate": 2.57366852365164e-06, + "loss": 0.5942, + "step": 3604 + }, + { + "epoch": 2.984815019326339, + "grad_norm": 0.8945804238319397, + "learning_rate": 2.572569304717874e-06, + "loss": 0.5882, + "step": 3605 + }, + { + "epoch": 2.9856432909994477, + "grad_norm": 0.9231047034263611, + "learning_rate": 2.5714700717426237e-06, + "loss": 0.5768, + "step": 3606 + }, + { + "epoch": 2.9864715626725564, + "grad_norm": 0.9362046718597412, + "learning_rate": 2.5703708249385824e-06, + "loss": 0.6145, + "step": 3607 + }, + { + "epoch": 2.9872998343456656, + "grad_norm": 0.9291548728942871, + "learning_rate": 2.5692715645184412e-06, + "loss": 0.5959, + "step": 3608 + }, + { + "epoch": 2.9881281060187743, + "grad_norm": 0.9203015565872192, + "learning_rate": 2.5681722906948977e-06, + "loss": 0.5906, + "step": 3609 + }, + { + "epoch": 2.988956377691883, + "grad_norm": 0.9317998886108398, + "learning_rate": 2.567073003680652e-06, + "loss": 0.5845, + "step": 3610 + }, + { + "epoch": 2.9897846493649918, + "grad_norm": 0.9188578128814697, + "learning_rate": 2.5659737036884054e-06, + "loss": 0.5901, + "step": 3611 + }, + { + "epoch": 2.9906129210381005, + "grad_norm": 0.9100846648216248, + "learning_rate": 2.564874390930862e-06, + "loss": 0.5952, + "step": 3612 + }, + { + "epoch": 2.9914411927112092, + "grad_norm": 0.9292994737625122, + "learning_rate": 2.563775065620729e-06, + "loss": 0.6112, + "step": 3613 + }, + { + "epoch": 2.992269464384318, + "grad_norm": 0.9025140404701233, + "learning_rate": 2.5626757279707143e-06, + "loss": 0.5691, + "step": 3614 + }, + { + "epoch": 2.9930977360574267, + "grad_norm": 0.9413566589355469, + "learning_rate": 2.561576378193531e-06, + "loss": 0.574, + "step": 3615 + }, + { + "epoch": 2.9939260077305354, + "grad_norm": 0.9244924783706665, + "learning_rate": 2.560477016501893e-06, + "loss": 0.606, + "step": 3616 + }, + { + "epoch": 2.9947542794036446, + "grad_norm": 0.8985790014266968, + "learning_rate": 2.5593776431085154e-06, + "loss": 0.5823, + "step": 3617 + }, + { + "epoch": 2.9955825510767533, + "grad_norm": 0.8941566348075867, + "learning_rate": 2.5582782582261182e-06, + "loss": 0.5643, + "step": 3618 + }, + { + "epoch": 2.996410822749862, + "grad_norm": 0.8992639183998108, + "learning_rate": 2.557178862067421e-06, + "loss": 0.6073, + "step": 3619 + }, + { + "epoch": 2.9972390944229708, + "grad_norm": 0.9369136691093445, + "learning_rate": 2.5560794548451464e-06, + "loss": 0.6031, + "step": 3620 + }, + { + "epoch": 2.9980673660960795, + "grad_norm": 0.8975582718849182, + "learning_rate": 2.5549800367720207e-06, + "loss": 0.6034, + "step": 3621 + }, + { + "epoch": 2.998895637769188, + "grad_norm": 0.8998291492462158, + "learning_rate": 2.5538806080607697e-06, + "loss": 0.5782, + "step": 3622 + }, + { + "epoch": 2.999723909442297, + "grad_norm": 0.9796467423439026, + "learning_rate": 2.5527811689241246e-06, + "loss": 0.581, + "step": 3623 + }, + { + "epoch": 3.0, + "grad_norm": 1.6821987628936768, + "learning_rate": 2.5516817195748143e-06, + "loss": 0.5955, + "step": 3624 + }, + { + "epoch": 3.0008282716731087, + "grad_norm": 0.9084760546684265, + "learning_rate": 2.5505822602255727e-06, + "loss": 0.5821, + "step": 3625 + }, + { + "epoch": 3.0016565433462175, + "grad_norm": 0.9643696546554565, + "learning_rate": 2.5494827910891357e-06, + "loss": 0.5627, + "step": 3626 + }, + { + "epoch": 3.002484815019326, + "grad_norm": 0.8933544754981995, + "learning_rate": 2.5483833123782405e-06, + "loss": 0.5573, + "step": 3627 + }, + { + "epoch": 3.003313086692435, + "grad_norm": 0.939877450466156, + "learning_rate": 2.5472838243056242e-06, + "loss": 0.5641, + "step": 3628 + }, + { + "epoch": 3.004141358365544, + "grad_norm": 0.9558863043785095, + "learning_rate": 2.54618432708403e-06, + "loss": 0.5753, + "step": 3629 + }, + { + "epoch": 3.004969630038653, + "grad_norm": 0.9423503875732422, + "learning_rate": 2.5450848209261987e-06, + "loss": 0.5786, + "step": 3630 + }, + { + "epoch": 3.0057979017117615, + "grad_norm": 0.9265878796577454, + "learning_rate": 2.5439853060448737e-06, + "loss": 0.5791, + "step": 3631 + }, + { + "epoch": 3.0066261733848703, + "grad_norm": 0.9355331063270569, + "learning_rate": 2.542885782652803e-06, + "loss": 0.5535, + "step": 3632 + }, + { + "epoch": 3.007454445057979, + "grad_norm": 0.9215316772460938, + "learning_rate": 2.5417862509627333e-06, + "loss": 0.5775, + "step": 3633 + }, + { + "epoch": 3.0082827167310877, + "grad_norm": 0.9157372713088989, + "learning_rate": 2.5406867111874127e-06, + "loss": 0.5547, + "step": 3634 + }, + { + "epoch": 3.0091109884041964, + "grad_norm": 0.9143363237380981, + "learning_rate": 2.539587163539594e-06, + "loss": 0.562, + "step": 3635 + }, + { + "epoch": 3.009939260077305, + "grad_norm": 0.942833423614502, + "learning_rate": 2.538487608232027e-06, + "loss": 0.5852, + "step": 3636 + }, + { + "epoch": 3.0107675317504143, + "grad_norm": 0.9196774363517761, + "learning_rate": 2.5373880454774667e-06, + "loss": 0.573, + "step": 3637 + }, + { + "epoch": 3.011595803423523, + "grad_norm": 0.9290075302124023, + "learning_rate": 2.536288475488669e-06, + "loss": 0.5672, + "step": 3638 + }, + { + "epoch": 3.012424075096632, + "grad_norm": 0.9856467843055725, + "learning_rate": 2.535188898478389e-06, + "loss": 0.5685, + "step": 3639 + }, + { + "epoch": 3.0132523467697405, + "grad_norm": 0.9258086085319519, + "learning_rate": 2.5340893146593855e-06, + "loss": 0.5705, + "step": 3640 + }, + { + "epoch": 3.0140806184428492, + "grad_norm": 0.8912391662597656, + "learning_rate": 2.5329897242444174e-06, + "loss": 0.5653, + "step": 3641 + }, + { + "epoch": 3.014908890115958, + "grad_norm": 0.9336050152778625, + "learning_rate": 2.531890127446244e-06, + "loss": 0.5515, + "step": 3642 + }, + { + "epoch": 3.0157371617890667, + "grad_norm": 0.9292400479316711, + "learning_rate": 2.5307905244776295e-06, + "loss": 0.5715, + "step": 3643 + }, + { + "epoch": 3.0165654334621754, + "grad_norm": 0.9212039709091187, + "learning_rate": 2.529690915551335e-06, + "loss": 0.5842, + "step": 3644 + }, + { + "epoch": 3.017393705135284, + "grad_norm": 0.9198867082595825, + "learning_rate": 2.528591300880125e-06, + "loss": 0.5498, + "step": 3645 + }, + { + "epoch": 3.0182219768083933, + "grad_norm": 0.9194673299789429, + "learning_rate": 2.5274916806767656e-06, + "loss": 0.5696, + "step": 3646 + }, + { + "epoch": 3.019050248481502, + "grad_norm": 0.9359262585639954, + "learning_rate": 2.5263920551540217e-06, + "loss": 0.5635, + "step": 3647 + }, + { + "epoch": 3.019878520154611, + "grad_norm": 0.9225771427154541, + "learning_rate": 2.525292424524661e-06, + "loss": 0.5489, + "step": 3648 + }, + { + "epoch": 3.0207067918277195, + "grad_norm": 0.9490647315979004, + "learning_rate": 2.524192789001452e-06, + "loss": 0.5692, + "step": 3649 + }, + { + "epoch": 3.0215350635008282, + "grad_norm": 0.8977577686309814, + "learning_rate": 2.5230931487971645e-06, + "loss": 0.5652, + "step": 3650 + }, + { + "epoch": 3.022363335173937, + "grad_norm": 0.963139533996582, + "learning_rate": 2.521993504124567e-06, + "loss": 0.5373, + "step": 3651 + }, + { + "epoch": 3.0231916068470457, + "grad_norm": 0.9195258617401123, + "learning_rate": 2.5208938551964326e-06, + "loss": 0.574, + "step": 3652 + }, + { + "epoch": 3.0240198785201544, + "grad_norm": 0.9074851274490356, + "learning_rate": 2.5197942022255305e-06, + "loss": 0.557, + "step": 3653 + }, + { + "epoch": 3.0248481501932636, + "grad_norm": 0.9083223342895508, + "learning_rate": 2.518694545424636e-06, + "loss": 0.5641, + "step": 3654 + }, + { + "epoch": 3.0256764218663723, + "grad_norm": 0.9047815203666687, + "learning_rate": 2.5175948850065207e-06, + "loss": 0.563, + "step": 3655 + }, + { + "epoch": 3.026504693539481, + "grad_norm": 0.918021559715271, + "learning_rate": 2.516495221183959e-06, + "loss": 0.5889, + "step": 3656 + }, + { + "epoch": 3.0273329652125898, + "grad_norm": 0.935458242893219, + "learning_rate": 2.5153955541697266e-06, + "loss": 0.5486, + "step": 3657 + }, + { + "epoch": 3.0281612368856985, + "grad_norm": 0.9182383418083191, + "learning_rate": 2.5142958841765976e-06, + "loss": 0.563, + "step": 3658 + }, + { + "epoch": 3.028989508558807, + "grad_norm": 0.9635967016220093, + "learning_rate": 2.5131962114173474e-06, + "loss": 0.557, + "step": 3659 + }, + { + "epoch": 3.029817780231916, + "grad_norm": 0.9150283932685852, + "learning_rate": 2.512096536104754e-06, + "loss": 0.5582, + "step": 3660 + }, + { + "epoch": 3.0306460519050247, + "grad_norm": 0.9288967847824097, + "learning_rate": 2.5109968584515935e-06, + "loss": 0.5628, + "step": 3661 + }, + { + "epoch": 3.031474323578134, + "grad_norm": 0.936221718788147, + "learning_rate": 2.5098971786706434e-06, + "loss": 0.564, + "step": 3662 + }, + { + "epoch": 3.0323025952512426, + "grad_norm": 0.9719551801681519, + "learning_rate": 2.5087974969746828e-06, + "loss": 0.5509, + "step": 3663 + }, + { + "epoch": 3.0331308669243513, + "grad_norm": 0.9522861838340759, + "learning_rate": 2.507697813576487e-06, + "loss": 0.5919, + "step": 3664 + }, + { + "epoch": 3.03395913859746, + "grad_norm": 0.9098208546638489, + "learning_rate": 2.506598128688836e-06, + "loss": 0.5661, + "step": 3665 + }, + { + "epoch": 3.0347874102705688, + "grad_norm": 0.9221910238265991, + "learning_rate": 2.5054984425245095e-06, + "loss": 0.5849, + "step": 3666 + }, + { + "epoch": 3.0356156819436775, + "grad_norm": 0.9386799335479736, + "learning_rate": 2.5043987552962844e-06, + "loss": 0.5765, + "step": 3667 + }, + { + "epoch": 3.036443953616786, + "grad_norm": 0.9208933711051941, + "learning_rate": 2.503299067216942e-06, + "loss": 0.5689, + "step": 3668 + }, + { + "epoch": 3.037272225289895, + "grad_norm": 0.9429235458374023, + "learning_rate": 2.5021993784992616e-06, + "loss": 0.555, + "step": 3669 + }, + { + "epoch": 3.0381004969630037, + "grad_norm": 0.9263848066329956, + "learning_rate": 2.5010996893560203e-06, + "loss": 0.562, + "step": 3670 + }, + { + "epoch": 3.038928768636113, + "grad_norm": 0.9523965120315552, + "learning_rate": 2.5e-06, + "loss": 0.5677, + "step": 3671 + }, + { + "epoch": 3.0397570403092216, + "grad_norm": 0.9585251212120056, + "learning_rate": 2.49890031064398e-06, + "loss": 0.5821, + "step": 3672 + }, + { + "epoch": 3.0405853119823303, + "grad_norm": 0.9311655163764954, + "learning_rate": 2.4978006215007393e-06, + "loss": 0.5752, + "step": 3673 + }, + { + "epoch": 3.041413583655439, + "grad_norm": 0.9548931121826172, + "learning_rate": 2.4967009327830584e-06, + "loss": 0.5719, + "step": 3674 + }, + { + "epoch": 3.0422418553285477, + "grad_norm": 0.9751641154289246, + "learning_rate": 2.495601244703716e-06, + "loss": 0.5869, + "step": 3675 + }, + { + "epoch": 3.0430701270016565, + "grad_norm": 0.9528283476829529, + "learning_rate": 2.494501557475491e-06, + "loss": 0.571, + "step": 3676 + }, + { + "epoch": 3.043898398674765, + "grad_norm": 0.9225226044654846, + "learning_rate": 2.493401871311165e-06, + "loss": 0.5635, + "step": 3677 + }, + { + "epoch": 3.044726670347874, + "grad_norm": 0.9444342851638794, + "learning_rate": 2.4923021864235143e-06, + "loss": 0.5593, + "step": 3678 + }, + { + "epoch": 3.045554942020983, + "grad_norm": 0.9299778938293457, + "learning_rate": 2.491202503025319e-06, + "loss": 0.5787, + "step": 3679 + }, + { + "epoch": 3.046383213694092, + "grad_norm": 0.913502037525177, + "learning_rate": 2.490102821329357e-06, + "loss": 0.5826, + "step": 3680 + }, + { + "epoch": 3.0472114853672005, + "grad_norm": 0.9337341785430908, + "learning_rate": 2.489003141548407e-06, + "loss": 0.5575, + "step": 3681 + }, + { + "epoch": 3.0480397570403093, + "grad_norm": 0.9271352887153625, + "learning_rate": 2.4879034638952464e-06, + "loss": 0.5759, + "step": 3682 + }, + { + "epoch": 3.048868028713418, + "grad_norm": 0.967474102973938, + "learning_rate": 2.486803788582653e-06, + "loss": 0.5542, + "step": 3683 + }, + { + "epoch": 3.0496963003865267, + "grad_norm": 0.9283983111381531, + "learning_rate": 2.4857041158234032e-06, + "loss": 0.5932, + "step": 3684 + }, + { + "epoch": 3.0505245720596355, + "grad_norm": 0.9075651168823242, + "learning_rate": 2.4846044458302742e-06, + "loss": 0.5704, + "step": 3685 + }, + { + "epoch": 3.051352843732744, + "grad_norm": 0.9511797428131104, + "learning_rate": 2.4835047788160415e-06, + "loss": 0.5788, + "step": 3686 + }, + { + "epoch": 3.052181115405853, + "grad_norm": 0.9016287922859192, + "learning_rate": 2.4824051149934797e-06, + "loss": 0.5829, + "step": 3687 + }, + { + "epoch": 3.053009387078962, + "grad_norm": 0.9479110836982727, + "learning_rate": 2.4813054545753657e-06, + "loss": 0.5721, + "step": 3688 + }, + { + "epoch": 3.053837658752071, + "grad_norm": 0.906857430934906, + "learning_rate": 2.4802057977744703e-06, + "loss": 0.5497, + "step": 3689 + }, + { + "epoch": 3.0546659304251795, + "grad_norm": 0.9445266723632812, + "learning_rate": 2.479106144803569e-06, + "loss": 0.5837, + "step": 3690 + }, + { + "epoch": 3.0554942020982883, + "grad_norm": 0.9599757194519043, + "learning_rate": 2.478006495875434e-06, + "loss": 0.5571, + "step": 3691 + }, + { + "epoch": 3.056322473771397, + "grad_norm": 0.924673318862915, + "learning_rate": 2.4769068512028368e-06, + "loss": 0.5644, + "step": 3692 + }, + { + "epoch": 3.0571507454445057, + "grad_norm": 0.9168586730957031, + "learning_rate": 2.4758072109985487e-06, + "loss": 0.5682, + "step": 3693 + }, + { + "epoch": 3.0579790171176144, + "grad_norm": 0.9143668413162231, + "learning_rate": 2.4747075754753395e-06, + "loss": 0.5521, + "step": 3694 + }, + { + "epoch": 3.058807288790723, + "grad_norm": 0.9329923391342163, + "learning_rate": 2.4736079448459787e-06, + "loss": 0.5534, + "step": 3695 + }, + { + "epoch": 3.0596355604638323, + "grad_norm": 0.9511533975601196, + "learning_rate": 2.472508319323235e-06, + "loss": 0.5795, + "step": 3696 + }, + { + "epoch": 3.060463832136941, + "grad_norm": 0.9574876427650452, + "learning_rate": 2.471408699119875e-06, + "loss": 0.5723, + "step": 3697 + }, + { + "epoch": 3.06129210381005, + "grad_norm": 0.9656964540481567, + "learning_rate": 2.470309084448665e-06, + "loss": 0.5876, + "step": 3698 + }, + { + "epoch": 3.0621203754831585, + "grad_norm": 0.9273092150688171, + "learning_rate": 2.469209475522371e-06, + "loss": 0.5591, + "step": 3699 + }, + { + "epoch": 3.0629486471562672, + "grad_norm": 0.9115599393844604, + "learning_rate": 2.4681098725537566e-06, + "loss": 0.5766, + "step": 3700 + }, + { + "epoch": 3.063776918829376, + "grad_norm": 0.9338927268981934, + "learning_rate": 2.4670102757555843e-06, + "loss": 0.5584, + "step": 3701 + }, + { + "epoch": 3.0646051905024847, + "grad_norm": 0.9274653792381287, + "learning_rate": 2.4659106853406158e-06, + "loss": 0.5545, + "step": 3702 + }, + { + "epoch": 3.0654334621755934, + "grad_norm": 0.9300533533096313, + "learning_rate": 2.464811101521612e-06, + "loss": 0.5751, + "step": 3703 + }, + { + "epoch": 3.0662617338487026, + "grad_norm": 0.9616261124610901, + "learning_rate": 2.463711524511332e-06, + "loss": 0.5635, + "step": 3704 + }, + { + "epoch": 3.0670900055218113, + "grad_norm": 1.0845082998275757, + "learning_rate": 2.4626119545225337e-06, + "loss": 0.5712, + "step": 3705 + }, + { + "epoch": 3.06791827719492, + "grad_norm": 0.9370817542076111, + "learning_rate": 2.4615123917679737e-06, + "loss": 0.5737, + "step": 3706 + }, + { + "epoch": 3.068746548868029, + "grad_norm": 0.984732985496521, + "learning_rate": 2.460412836460407e-06, + "loss": 0.5631, + "step": 3707 + }, + { + "epoch": 3.0695748205411375, + "grad_norm": 0.90977942943573, + "learning_rate": 2.4593132888125877e-06, + "loss": 0.5755, + "step": 3708 + }, + { + "epoch": 3.0704030922142462, + "grad_norm": 0.9592271447181702, + "learning_rate": 2.4582137490372676e-06, + "loss": 0.552, + "step": 3709 + }, + { + "epoch": 3.071231363887355, + "grad_norm": 0.9560198187828064, + "learning_rate": 2.457114217347197e-06, + "loss": 0.5744, + "step": 3710 + }, + { + "epoch": 3.0720596355604637, + "grad_norm": 0.9570426940917969, + "learning_rate": 2.4560146939551268e-06, + "loss": 0.5769, + "step": 3711 + }, + { + "epoch": 3.0728879072335724, + "grad_norm": 0.9208659529685974, + "learning_rate": 2.4549151790738025e-06, + "loss": 0.5697, + "step": 3712 + }, + { + "epoch": 3.0737161789066816, + "grad_norm": 0.9417058229446411, + "learning_rate": 2.453815672915971e-06, + "loss": 0.5697, + "step": 3713 + }, + { + "epoch": 3.0745444505797903, + "grad_norm": 0.9602729082107544, + "learning_rate": 2.4527161756943766e-06, + "loss": 0.5737, + "step": 3714 + }, + { + "epoch": 3.075372722252899, + "grad_norm": 0.931280255317688, + "learning_rate": 2.4516166876217603e-06, + "loss": 0.5704, + "step": 3715 + }, + { + "epoch": 3.0762009939260078, + "grad_norm": 0.920081377029419, + "learning_rate": 2.4505172089108648e-06, + "loss": 0.57, + "step": 3716 + }, + { + "epoch": 3.0770292655991165, + "grad_norm": 0.929806649684906, + "learning_rate": 2.449417739774428e-06, + "loss": 0.5648, + "step": 3717 + }, + { + "epoch": 3.077857537272225, + "grad_norm": 0.9344229698181152, + "learning_rate": 2.4483182804251866e-06, + "loss": 0.5666, + "step": 3718 + }, + { + "epoch": 3.078685808945334, + "grad_norm": 0.9535424113273621, + "learning_rate": 2.4472188310758766e-06, + "loss": 0.5777, + "step": 3719 + }, + { + "epoch": 3.0795140806184427, + "grad_norm": 0.9594505429267883, + "learning_rate": 2.4461193919392307e-06, + "loss": 0.5614, + "step": 3720 + }, + { + "epoch": 3.080342352291552, + "grad_norm": 0.9511420726776123, + "learning_rate": 2.4450199632279797e-06, + "loss": 0.5925, + "step": 3721 + }, + { + "epoch": 3.0811706239646606, + "grad_norm": 0.9244424104690552, + "learning_rate": 2.443920545154855e-06, + "loss": 0.5694, + "step": 3722 + }, + { + "epoch": 3.0819988956377693, + "grad_norm": 0.9384777545928955, + "learning_rate": 2.4428211379325804e-06, + "loss": 0.5443, + "step": 3723 + }, + { + "epoch": 3.082827167310878, + "grad_norm": 0.9536921381950378, + "learning_rate": 2.441721741773883e-06, + "loss": 0.5817, + "step": 3724 + }, + { + "epoch": 3.0836554389839868, + "grad_norm": 0.9547799825668335, + "learning_rate": 2.4406223568914854e-06, + "loss": 0.5565, + "step": 3725 + }, + { + "epoch": 3.0844837106570955, + "grad_norm": 0.9509586691856384, + "learning_rate": 2.4395229834981076e-06, + "loss": 0.5712, + "step": 3726 + }, + { + "epoch": 3.085311982330204, + "grad_norm": 0.9227497577667236, + "learning_rate": 2.4384236218064692e-06, + "loss": 0.5602, + "step": 3727 + }, + { + "epoch": 3.086140254003313, + "grad_norm": 0.9162245392799377, + "learning_rate": 2.4373242720292865e-06, + "loss": 0.555, + "step": 3728 + }, + { + "epoch": 3.0869685256764217, + "grad_norm": 0.9121986627578735, + "learning_rate": 2.4362249343792722e-06, + "loss": 0.5748, + "step": 3729 + }, + { + "epoch": 3.087796797349531, + "grad_norm": 0.9310787320137024, + "learning_rate": 2.435125609069139e-06, + "loss": 0.5671, + "step": 3730 + }, + { + "epoch": 3.0886250690226396, + "grad_norm": 0.9523546695709229, + "learning_rate": 2.4340262963115955e-06, + "loss": 0.595, + "step": 3731 + }, + { + "epoch": 3.0894533406957483, + "grad_norm": 0.9429517388343811, + "learning_rate": 2.432926996319348e-06, + "loss": 0.5381, + "step": 3732 + }, + { + "epoch": 3.090281612368857, + "grad_norm": 0.9443867802619934, + "learning_rate": 2.4318277093051023e-06, + "loss": 0.5588, + "step": 3733 + }, + { + "epoch": 3.0911098840419657, + "grad_norm": 0.9320855140686035, + "learning_rate": 2.43072843548156e-06, + "loss": 0.5765, + "step": 3734 + }, + { + "epoch": 3.0919381557150745, + "grad_norm": 0.9335711002349854, + "learning_rate": 2.429629175061419e-06, + "loss": 0.5734, + "step": 3735 + }, + { + "epoch": 3.092766427388183, + "grad_norm": 0.9539975523948669, + "learning_rate": 2.4285299282573767e-06, + "loss": 0.5733, + "step": 3736 + }, + { + "epoch": 3.093594699061292, + "grad_norm": 0.9393835067749023, + "learning_rate": 2.4274306952821262e-06, + "loss": 0.5808, + "step": 3737 + }, + { + "epoch": 3.094422970734401, + "grad_norm": 0.9707219004631042, + "learning_rate": 2.4263314763483607e-06, + "loss": 0.5755, + "step": 3738 + }, + { + "epoch": 3.09525124240751, + "grad_norm": 0.9541774988174438, + "learning_rate": 2.4252322716687676e-06, + "loss": 0.5472, + "step": 3739 + }, + { + "epoch": 3.0960795140806185, + "grad_norm": 0.9300408959388733, + "learning_rate": 2.424133081456032e-06, + "loss": 0.5771, + "step": 3740 + }, + { + "epoch": 3.0969077857537273, + "grad_norm": 0.9487392902374268, + "learning_rate": 2.423033905922839e-06, + "loss": 0.5805, + "step": 3741 + }, + { + "epoch": 3.097736057426836, + "grad_norm": 0.9021924734115601, + "learning_rate": 2.4219347452818678e-06, + "loss": 0.5648, + "step": 3742 + }, + { + "epoch": 3.0985643290999447, + "grad_norm": 0.9229711294174194, + "learning_rate": 2.420835599745795e-06, + "loss": 0.5569, + "step": 3743 + }, + { + "epoch": 3.0993926007730535, + "grad_norm": 0.9335191249847412, + "learning_rate": 2.419736469527296e-06, + "loss": 0.5728, + "step": 3744 + }, + { + "epoch": 3.100220872446162, + "grad_norm": 0.9294394254684448, + "learning_rate": 2.4186373548390427e-06, + "loss": 0.5522, + "step": 3745 + }, + { + "epoch": 3.101049144119271, + "grad_norm": 0.931654691696167, + "learning_rate": 2.417538255893702e-06, + "loss": 0.5747, + "step": 3746 + }, + { + "epoch": 3.10187741579238, + "grad_norm": 0.9018174409866333, + "learning_rate": 2.4164391729039405e-06, + "loss": 0.5697, + "step": 3747 + }, + { + "epoch": 3.102705687465489, + "grad_norm": 0.9389755129814148, + "learning_rate": 2.4153401060824193e-06, + "loss": 0.5559, + "step": 3748 + }, + { + "epoch": 3.1035339591385975, + "grad_norm": 0.9191637635231018, + "learning_rate": 2.414241055641799e-06, + "loss": 0.5678, + "step": 3749 + }, + { + "epoch": 3.1043622308117063, + "grad_norm": 0.9245947599411011, + "learning_rate": 2.4131420217947347e-06, + "loss": 0.5677, + "step": 3750 + }, + { + "epoch": 3.105190502484815, + "grad_norm": 0.9380972981452942, + "learning_rate": 2.4120430047538797e-06, + "loss": 0.5611, + "step": 3751 + }, + { + "epoch": 3.1060187741579237, + "grad_norm": 0.9427217841148376, + "learning_rate": 2.410944004731883e-06, + "loss": 0.592, + "step": 3752 + }, + { + "epoch": 3.1068470458310324, + "grad_norm": 0.9142959117889404, + "learning_rate": 2.4098450219413915e-06, + "loss": 0.5588, + "step": 3753 + }, + { + "epoch": 3.107675317504141, + "grad_norm": 0.9501208066940308, + "learning_rate": 2.4087460565950475e-06, + "loss": 0.5735, + "step": 3754 + }, + { + "epoch": 3.1085035891772503, + "grad_norm": 0.9072937369346619, + "learning_rate": 2.4076471089054916e-06, + "loss": 0.557, + "step": 3755 + }, + { + "epoch": 3.109331860850359, + "grad_norm": 0.9310604333877563, + "learning_rate": 2.4065481790853595e-06, + "loss": 0.5616, + "step": 3756 + }, + { + "epoch": 3.110160132523468, + "grad_norm": 0.9229986071586609, + "learning_rate": 2.405449267347283e-06, + "loss": 0.5806, + "step": 3757 + }, + { + "epoch": 3.1109884041965765, + "grad_norm": 0.9491189122200012, + "learning_rate": 2.4043503739038922e-06, + "loss": 0.571, + "step": 3758 + }, + { + "epoch": 3.1118166758696852, + "grad_norm": 0.9373705983161926, + "learning_rate": 2.4032514989678113e-06, + "loss": 0.557, + "step": 3759 + }, + { + "epoch": 3.112644947542794, + "grad_norm": 0.9476242661476135, + "learning_rate": 2.4021526427516647e-06, + "loss": 0.5749, + "step": 3760 + }, + { + "epoch": 3.1134732192159027, + "grad_norm": 0.9147936701774597, + "learning_rate": 2.401053805468069e-06, + "loss": 0.5579, + "step": 3761 + }, + { + "epoch": 3.1143014908890114, + "grad_norm": 0.9299486875534058, + "learning_rate": 2.399954987329639e-06, + "loss": 0.5707, + "step": 3762 + }, + { + "epoch": 3.1151297625621206, + "grad_norm": 0.9515376091003418, + "learning_rate": 2.3988561885489874e-06, + "loss": 0.5923, + "step": 3763 + }, + { + "epoch": 3.1159580342352293, + "grad_norm": 0.9455316662788391, + "learning_rate": 2.39775740933872e-06, + "loss": 0.5686, + "step": 3764 + }, + { + "epoch": 3.116786305908338, + "grad_norm": 0.952534019947052, + "learning_rate": 2.3966586499114406e-06, + "loss": 0.581, + "step": 3765 + }, + { + "epoch": 3.1176145775814468, + "grad_norm": 0.9540290832519531, + "learning_rate": 2.3955599104797493e-06, + "loss": 0.5612, + "step": 3766 + }, + { + "epoch": 3.1184428492545555, + "grad_norm": 0.9238110780715942, + "learning_rate": 2.3944611912562423e-06, + "loss": 0.5608, + "step": 3767 + }, + { + "epoch": 3.1192711209276642, + "grad_norm": 0.9312852025032043, + "learning_rate": 2.3933624924535113e-06, + "loss": 0.5752, + "step": 3768 + }, + { + "epoch": 3.120099392600773, + "grad_norm": 0.9356149435043335, + "learning_rate": 2.3922638142841433e-06, + "loss": 0.5719, + "step": 3769 + }, + { + "epoch": 3.1209276642738817, + "grad_norm": 0.9629190564155579, + "learning_rate": 2.391165156960722e-06, + "loss": 0.5791, + "step": 3770 + }, + { + "epoch": 3.1217559359469904, + "grad_norm": 0.9337695240974426, + "learning_rate": 2.3900665206958294e-06, + "loss": 0.5636, + "step": 3771 + }, + { + "epoch": 3.1225842076200996, + "grad_norm": 0.9723961353302002, + "learning_rate": 2.3889679057020402e-06, + "loss": 0.5897, + "step": 3772 + }, + { + "epoch": 3.1234124792932083, + "grad_norm": 0.9874154925346375, + "learning_rate": 2.3878693121919255e-06, + "loss": 0.5846, + "step": 3773 + }, + { + "epoch": 3.124240750966317, + "grad_norm": 0.9433327913284302, + "learning_rate": 2.386770740378054e-06, + "loss": 0.5631, + "step": 3774 + }, + { + "epoch": 3.1250690226394258, + "grad_norm": 0.9409958124160767, + "learning_rate": 2.385672190472989e-06, + "loss": 0.548, + "step": 3775 + }, + { + "epoch": 3.1258972943125345, + "grad_norm": 0.9370530247688293, + "learning_rate": 2.3845736626892895e-06, + "loss": 0.5277, + "step": 3776 + }, + { + "epoch": 3.126725565985643, + "grad_norm": 0.9603546857833862, + "learning_rate": 2.3834751572395097e-06, + "loss": 0.5635, + "step": 3777 + }, + { + "epoch": 3.127553837658752, + "grad_norm": 0.9234820008277893, + "learning_rate": 2.382376674336201e-06, + "loss": 0.5743, + "step": 3778 + }, + { + "epoch": 3.1283821093318607, + "grad_norm": 0.9674520492553711, + "learning_rate": 2.38127821419191e-06, + "loss": 0.5694, + "step": 3779 + }, + { + "epoch": 3.1292103810049694, + "grad_norm": 0.9508289098739624, + "learning_rate": 2.380179777019177e-06, + "loss": 0.5625, + "step": 3780 + }, + { + "epoch": 3.1300386526780786, + "grad_norm": 0.9463353753089905, + "learning_rate": 2.3790813630305403e-06, + "loss": 0.5859, + "step": 3781 + }, + { + "epoch": 3.1308669243511873, + "grad_norm": 0.9280157089233398, + "learning_rate": 2.3779829724385327e-06, + "loss": 0.5632, + "step": 3782 + }, + { + "epoch": 3.131695196024296, + "grad_norm": 0.931016206741333, + "learning_rate": 2.376884605455683e-06, + "loss": 0.556, + "step": 3783 + }, + { + "epoch": 3.1325234676974048, + "grad_norm": 0.9319890141487122, + "learning_rate": 2.375786262294514e-06, + "loss": 0.563, + "step": 3784 + }, + { + "epoch": 3.1333517393705135, + "grad_norm": 0.9274983406066895, + "learning_rate": 2.374687943167545e-06, + "loss": 0.5713, + "step": 3785 + }, + { + "epoch": 3.134180011043622, + "grad_norm": 0.9298895001411438, + "learning_rate": 2.3735896482872917e-06, + "loss": 0.5705, + "step": 3786 + }, + { + "epoch": 3.135008282716731, + "grad_norm": 0.9218611121177673, + "learning_rate": 2.3724913778662624e-06, + "loss": 0.5738, + "step": 3787 + }, + { + "epoch": 3.13583655438984, + "grad_norm": 0.9677892327308655, + "learning_rate": 2.371393132116963e-06, + "loss": 0.5675, + "step": 3788 + }, + { + "epoch": 3.136664826062949, + "grad_norm": 0.9483129978179932, + "learning_rate": 2.3702949112518935e-06, + "loss": 0.5525, + "step": 3789 + }, + { + "epoch": 3.1374930977360576, + "grad_norm": 0.9588272571563721, + "learning_rate": 2.3691967154835505e-06, + "loss": 0.5819, + "step": 3790 + }, + { + "epoch": 3.1383213694091663, + "grad_norm": 0.9432960748672485, + "learning_rate": 2.3680985450244233e-06, + "loss": 0.5757, + "step": 3791 + }, + { + "epoch": 3.139149641082275, + "grad_norm": 0.952733039855957, + "learning_rate": 2.367000400086998e-06, + "loss": 0.5807, + "step": 3792 + }, + { + "epoch": 3.1399779127553837, + "grad_norm": 0.9517266154289246, + "learning_rate": 2.365902280883755e-06, + "loss": 0.5503, + "step": 3793 + }, + { + "epoch": 3.1408061844284925, + "grad_norm": 0.9321509599685669, + "learning_rate": 2.364804187627171e-06, + "loss": 0.5534, + "step": 3794 + }, + { + "epoch": 3.141634456101601, + "grad_norm": 0.9490827918052673, + "learning_rate": 2.3637061205297167e-06, + "loss": 0.582, + "step": 3795 + }, + { + "epoch": 3.14246272777471, + "grad_norm": 0.9612736701965332, + "learning_rate": 2.3626080798038565e-06, + "loss": 0.5673, + "step": 3796 + }, + { + "epoch": 3.143290999447819, + "grad_norm": 0.9634802937507629, + "learning_rate": 2.361510065662053e-06, + "loss": 0.5784, + "step": 3797 + }, + { + "epoch": 3.144119271120928, + "grad_norm": 0.9495031237602234, + "learning_rate": 2.360412078316761e-06, + "loss": 0.5661, + "step": 3798 + }, + { + "epoch": 3.1449475427940365, + "grad_norm": 0.9234485030174255, + "learning_rate": 2.3593141179804295e-06, + "loss": 0.5718, + "step": 3799 + }, + { + "epoch": 3.1457758144671453, + "grad_norm": 0.9367272257804871, + "learning_rate": 2.3582161848655054e-06, + "loss": 0.5332, + "step": 3800 + }, + { + "epoch": 3.146604086140254, + "grad_norm": 0.9217962622642517, + "learning_rate": 2.357118279184429e-06, + "loss": 0.586, + "step": 3801 + }, + { + "epoch": 3.1474323578133627, + "grad_norm": 0.943759560585022, + "learning_rate": 2.3560204011496324e-06, + "loss": 0.5606, + "step": 3802 + }, + { + "epoch": 3.1482606294864715, + "grad_norm": 0.959769070148468, + "learning_rate": 2.354922550973546e-06, + "loss": 0.5858, + "step": 3803 + }, + { + "epoch": 3.14908890115958, + "grad_norm": 0.9223443865776062, + "learning_rate": 2.3538247288685935e-06, + "loss": 0.5924, + "step": 3804 + }, + { + "epoch": 3.149917172832689, + "grad_norm": 0.958400547504425, + "learning_rate": 2.3527269350471936e-06, + "loss": 0.5669, + "step": 3805 + }, + { + "epoch": 3.150745444505798, + "grad_norm": 0.9326207041740417, + "learning_rate": 2.351629169721759e-06, + "loss": 0.5691, + "step": 3806 + }, + { + "epoch": 3.151573716178907, + "grad_norm": 0.9581301808357239, + "learning_rate": 2.3505314331046964e-06, + "loss": 0.5767, + "step": 3807 + }, + { + "epoch": 3.1524019878520155, + "grad_norm": 0.921093225479126, + "learning_rate": 2.349433725408409e-06, + "loss": 0.5773, + "step": 3808 + }, + { + "epoch": 3.1532302595251243, + "grad_norm": 0.9264110326766968, + "learning_rate": 2.348336046845292e-06, + "loss": 0.5388, + "step": 3809 + }, + { + "epoch": 3.154058531198233, + "grad_norm": 0.9559600949287415, + "learning_rate": 2.3472383976277353e-06, + "loss": 0.5601, + "step": 3810 + }, + { + "epoch": 3.1548868028713417, + "grad_norm": 0.8962564468383789, + "learning_rate": 2.3461407779681255e-06, + "loss": 0.5718, + "step": 3811 + }, + { + "epoch": 3.1557150745444504, + "grad_norm": 0.9371088743209839, + "learning_rate": 2.345043188078841e-06, + "loss": 0.5841, + "step": 3812 + }, + { + "epoch": 3.156543346217559, + "grad_norm": 0.952690839767456, + "learning_rate": 2.343945628172255e-06, + "loss": 0.5517, + "step": 3813 + }, + { + "epoch": 3.1573716178906683, + "grad_norm": 0.9315347075462341, + "learning_rate": 2.3428480984607356e-06, + "loss": 0.5773, + "step": 3814 + }, + { + "epoch": 3.158199889563777, + "grad_norm": 0.9349079728126526, + "learning_rate": 2.341750599156643e-06, + "loss": 0.568, + "step": 3815 + }, + { + "epoch": 3.159028161236886, + "grad_norm": 0.9342600703239441, + "learning_rate": 2.340653130472335e-06, + "loss": 0.5495, + "step": 3816 + }, + { + "epoch": 3.1598564329099945, + "grad_norm": 0.9394509792327881, + "learning_rate": 2.339555692620161e-06, + "loss": 0.6008, + "step": 3817 + }, + { + "epoch": 3.1606847045831032, + "grad_norm": 0.9620388150215149, + "learning_rate": 2.338458285812464e-06, + "loss": 0.5652, + "step": 3818 + }, + { + "epoch": 3.161512976256212, + "grad_norm": 0.9644536375999451, + "learning_rate": 2.3373609102615834e-06, + "loss": 0.5556, + "step": 3819 + }, + { + "epoch": 3.1623412479293207, + "grad_norm": 0.9285350441932678, + "learning_rate": 2.3362635661798507e-06, + "loss": 0.5758, + "step": 3820 + }, + { + "epoch": 3.1631695196024294, + "grad_norm": 0.9304336905479431, + "learning_rate": 2.335166253779591e-06, + "loss": 0.6108, + "step": 3821 + }, + { + "epoch": 3.1639977912755386, + "grad_norm": 0.9232069253921509, + "learning_rate": 2.334068973273125e-06, + "loss": 0.5835, + "step": 3822 + }, + { + "epoch": 3.1648260629486473, + "grad_norm": 0.9440987706184387, + "learning_rate": 2.3329717248727665e-06, + "loss": 0.5778, + "step": 3823 + }, + { + "epoch": 3.165654334621756, + "grad_norm": 0.9345200657844543, + "learning_rate": 2.3318745087908225e-06, + "loss": 0.56, + "step": 3824 + }, + { + "epoch": 3.1664826062948648, + "grad_norm": 0.9822887778282166, + "learning_rate": 2.330777325239593e-06, + "loss": 0.556, + "step": 3825 + }, + { + "epoch": 3.1673108779679735, + "grad_norm": 0.9316609501838684, + "learning_rate": 2.3296801744313733e-06, + "loss": 0.5749, + "step": 3826 + }, + { + "epoch": 3.1681391496410822, + "grad_norm": 0.933875322341919, + "learning_rate": 2.3285830565784528e-06, + "loss": 0.5686, + "step": 3827 + }, + { + "epoch": 3.168967421314191, + "grad_norm": 0.9843499660491943, + "learning_rate": 2.327485971893112e-06, + "loss": 0.6005, + "step": 3828 + }, + { + "epoch": 3.1697956929872997, + "grad_norm": 0.9594619870185852, + "learning_rate": 2.326388920587628e-06, + "loss": 0.5754, + "step": 3829 + }, + { + "epoch": 3.1706239646604084, + "grad_norm": 0.959398090839386, + "learning_rate": 2.3252919028742698e-06, + "loss": 0.5762, + "step": 3830 + }, + { + "epoch": 3.1714522363335176, + "grad_norm": 0.9445636868476868, + "learning_rate": 2.3241949189652996e-06, + "loss": 0.5757, + "step": 3831 + }, + { + "epoch": 3.1722805080066263, + "grad_norm": 0.9507540464401245, + "learning_rate": 2.323097969072973e-06, + "loss": 0.5695, + "step": 3832 + }, + { + "epoch": 3.173108779679735, + "grad_norm": 0.9695854187011719, + "learning_rate": 2.322001053409541e-06, + "loss": 0.5665, + "step": 3833 + }, + { + "epoch": 3.1739370513528438, + "grad_norm": 0.9487887620925903, + "learning_rate": 2.3209041721872464e-06, + "loss": 0.5651, + "step": 3834 + }, + { + "epoch": 3.1747653230259525, + "grad_norm": 0.9431695938110352, + "learning_rate": 2.319807325618325e-06, + "loss": 0.5516, + "step": 3835 + }, + { + "epoch": 3.175593594699061, + "grad_norm": 0.9398961067199707, + "learning_rate": 2.318710513915006e-06, + "loss": 0.571, + "step": 3836 + }, + { + "epoch": 3.17642186637217, + "grad_norm": 0.9242047667503357, + "learning_rate": 2.317613737289512e-06, + "loss": 0.5557, + "step": 3837 + }, + { + "epoch": 3.1772501380452787, + "grad_norm": 0.9494120478630066, + "learning_rate": 2.31651699595406e-06, + "loss": 0.568, + "step": 3838 + }, + { + "epoch": 3.1780784097183874, + "grad_norm": 0.9196777939796448, + "learning_rate": 2.3154202901208593e-06, + "loss": 0.5656, + "step": 3839 + }, + { + "epoch": 3.1789066813914966, + "grad_norm": 0.9382702112197876, + "learning_rate": 2.314323620002111e-06, + "loss": 0.5792, + "step": 3840 + }, + { + "epoch": 3.1797349530646053, + "grad_norm": 1.018182635307312, + "learning_rate": 2.3132269858100118e-06, + "loss": 0.5804, + "step": 3841 + }, + { + "epoch": 3.180563224737714, + "grad_norm": 0.9466580748558044, + "learning_rate": 2.31213038775675e-06, + "loss": 0.565, + "step": 3842 + }, + { + "epoch": 3.1813914964108228, + "grad_norm": 0.9668580889701843, + "learning_rate": 2.3110338260545062e-06, + "loss": 0.5724, + "step": 3843 + }, + { + "epoch": 3.1822197680839315, + "grad_norm": 0.9272748231887817, + "learning_rate": 2.309937300915456e-06, + "loss": 0.579, + "step": 3844 + }, + { + "epoch": 3.18304803975704, + "grad_norm": 0.9182299375534058, + "learning_rate": 2.3088408125517666e-06, + "loss": 0.5906, + "step": 3845 + }, + { + "epoch": 3.183876311430149, + "grad_norm": 1.034818410873413, + "learning_rate": 2.307744361175598e-06, + "loss": 0.5709, + "step": 3846 + }, + { + "epoch": 3.184704583103258, + "grad_norm": 0.9287042617797852, + "learning_rate": 2.306647946999104e-06, + "loss": 0.5855, + "step": 3847 + }, + { + "epoch": 3.185532854776367, + "grad_norm": 0.9378973841667175, + "learning_rate": 2.305551570234428e-06, + "loss": 0.5822, + "step": 3848 + }, + { + "epoch": 3.1863611264494756, + "grad_norm": 0.9447490572929382, + "learning_rate": 2.304455231093712e-06, + "loss": 0.5781, + "step": 3849 + }, + { + "epoch": 3.1871893981225843, + "grad_norm": 0.9659326076507568, + "learning_rate": 2.303358929789085e-06, + "loss": 0.6217, + "step": 3850 + }, + { + "epoch": 3.188017669795693, + "grad_norm": 0.9089621305465698, + "learning_rate": 2.302262666532672e-06, + "loss": 0.5733, + "step": 3851 + }, + { + "epoch": 3.1888459414688017, + "grad_norm": 0.9480805993080139, + "learning_rate": 2.3011664415365905e-06, + "loss": 0.5615, + "step": 3852 + }, + { + "epoch": 3.1896742131419105, + "grad_norm": 0.9534367322921753, + "learning_rate": 2.300070255012949e-06, + "loss": 0.5776, + "step": 3853 + }, + { + "epoch": 3.190502484815019, + "grad_norm": 0.9526930451393127, + "learning_rate": 2.298974107173849e-06, + "loss": 0.5541, + "step": 3854 + }, + { + "epoch": 3.191330756488128, + "grad_norm": 0.9261385202407837, + "learning_rate": 2.2978779982313853e-06, + "loss": 0.5797, + "step": 3855 + }, + { + "epoch": 3.192159028161237, + "grad_norm": 0.9275192618370056, + "learning_rate": 2.296781928397645e-06, + "loss": 0.559, + "step": 3856 + }, + { + "epoch": 3.192987299834346, + "grad_norm": 0.9341071248054504, + "learning_rate": 2.2956858978847076e-06, + "loss": 0.5523, + "step": 3857 + }, + { + "epoch": 3.1938155715074545, + "grad_norm": 0.9231052398681641, + "learning_rate": 2.2945899069046444e-06, + "loss": 0.5663, + "step": 3858 + }, + { + "epoch": 3.1946438431805633, + "grad_norm": 0.9468841552734375, + "learning_rate": 2.293493955669519e-06, + "loss": 0.5815, + "step": 3859 + }, + { + "epoch": 3.195472114853672, + "grad_norm": 0.9482542276382446, + "learning_rate": 2.2923980443913876e-06, + "loss": 0.5679, + "step": 3860 + }, + { + "epoch": 3.1963003865267807, + "grad_norm": 0.9444701075553894, + "learning_rate": 2.2913021732822997e-06, + "loss": 0.5796, + "step": 3861 + }, + { + "epoch": 3.1971286581998895, + "grad_norm": 0.9483172297477722, + "learning_rate": 2.2902063425542963e-06, + "loss": 0.5391, + "step": 3862 + }, + { + "epoch": 3.197956929872998, + "grad_norm": 0.8964424729347229, + "learning_rate": 2.289110552419409e-06, + "loss": 0.5719, + "step": 3863 + }, + { + "epoch": 3.198785201546107, + "grad_norm": 0.9366036653518677, + "learning_rate": 2.2880148030896648e-06, + "loss": 0.5729, + "step": 3864 + }, + { + "epoch": 3.199613473219216, + "grad_norm": 0.9433067440986633, + "learning_rate": 2.28691909477708e-06, + "loss": 0.5811, + "step": 3865 + }, + { + "epoch": 3.200441744892325, + "grad_norm": 1.0146312713623047, + "learning_rate": 2.285823427693664e-06, + "loss": 0.5524, + "step": 3866 + }, + { + "epoch": 3.2012700165654335, + "grad_norm": 0.9418644905090332, + "learning_rate": 2.2847278020514176e-06, + "loss": 0.5687, + "step": 3867 + }, + { + "epoch": 3.2020982882385423, + "grad_norm": 0.9421892166137695, + "learning_rate": 2.2836322180623355e-06, + "loss": 0.559, + "step": 3868 + }, + { + "epoch": 3.202926559911651, + "grad_norm": 0.9407430291175842, + "learning_rate": 2.2825366759384027e-06, + "loss": 0.5876, + "step": 3869 + }, + { + "epoch": 3.2037548315847597, + "grad_norm": 0.9176812767982483, + "learning_rate": 2.281441175891595e-06, + "loss": 0.5514, + "step": 3870 + }, + { + "epoch": 3.2045831032578684, + "grad_norm": 0.9171411991119385, + "learning_rate": 2.2803457181338825e-06, + "loss": 0.5778, + "step": 3871 + }, + { + "epoch": 3.205411374930977, + "grad_norm": 0.9477943778038025, + "learning_rate": 2.279250302877226e-06, + "loss": 0.5806, + "step": 3872 + }, + { + "epoch": 3.2062396466040863, + "grad_norm": 0.9568420052528381, + "learning_rate": 2.2781549303335783e-06, + "loss": 0.6025, + "step": 3873 + }, + { + "epoch": 3.207067918277195, + "grad_norm": 0.9442486763000488, + "learning_rate": 2.2770596007148833e-06, + "loss": 0.5678, + "step": 3874 + }, + { + "epoch": 3.207896189950304, + "grad_norm": 0.9460006356239319, + "learning_rate": 2.2759643142330775e-06, + "loss": 0.5651, + "step": 3875 + }, + { + "epoch": 3.2087244616234125, + "grad_norm": 0.9530642032623291, + "learning_rate": 2.2748690711000885e-06, + "loss": 0.5704, + "step": 3876 + }, + { + "epoch": 3.2095527332965212, + "grad_norm": 0.9378021955490112, + "learning_rate": 2.2737738715278353e-06, + "loss": 0.5605, + "step": 3877 + }, + { + "epoch": 3.21038100496963, + "grad_norm": 0.9239064455032349, + "learning_rate": 2.272678715728229e-06, + "loss": 0.5514, + "step": 3878 + }, + { + "epoch": 3.2112092766427387, + "grad_norm": 0.9366111755371094, + "learning_rate": 2.271583603913173e-06, + "loss": 0.5746, + "step": 3879 + }, + { + "epoch": 3.2120375483158474, + "grad_norm": 0.9362907409667969, + "learning_rate": 2.270488536294559e-06, + "loss": 0.5622, + "step": 3880 + }, + { + "epoch": 3.2128658199889566, + "grad_norm": 0.9395036697387695, + "learning_rate": 2.2693935130842763e-06, + "loss": 0.5765, + "step": 3881 + }, + { + "epoch": 3.2136940916620653, + "grad_norm": 0.9549781084060669, + "learning_rate": 2.2682985344941966e-06, + "loss": 0.5716, + "step": 3882 + }, + { + "epoch": 3.214522363335174, + "grad_norm": 0.9677711129188538, + "learning_rate": 2.2672036007361915e-06, + "loss": 0.5922, + "step": 3883 + }, + { + "epoch": 3.2153506350082828, + "grad_norm": 0.9358668923377991, + "learning_rate": 2.26610871202212e-06, + "loss": 0.5794, + "step": 3884 + }, + { + "epoch": 3.2161789066813915, + "grad_norm": 0.938670814037323, + "learning_rate": 2.265013868563831e-06, + "loss": 0.5809, + "step": 3885 + }, + { + "epoch": 3.2170071783545002, + "grad_norm": 0.9247921109199524, + "learning_rate": 2.2639190705731694e-06, + "loss": 0.5714, + "step": 3886 + }, + { + "epoch": 3.217835450027609, + "grad_norm": 0.9363803863525391, + "learning_rate": 2.262824318261966e-06, + "loss": 0.5714, + "step": 3887 + }, + { + "epoch": 3.2186637217007177, + "grad_norm": 0.9042513370513916, + "learning_rate": 2.2617296118420463e-06, + "loss": 0.5612, + "step": 3888 + }, + { + "epoch": 3.2194919933738264, + "grad_norm": 0.9226694703102112, + "learning_rate": 2.260634951525226e-06, + "loss": 0.5579, + "step": 3889 + }, + { + "epoch": 3.2203202650469356, + "grad_norm": 0.9465897679328918, + "learning_rate": 2.2595403375233107e-06, + "loss": 0.5507, + "step": 3890 + }, + { + "epoch": 3.2211485367200443, + "grad_norm": 0.9200103282928467, + "learning_rate": 2.2584457700480984e-06, + "loss": 0.568, + "step": 3891 + }, + { + "epoch": 3.221976808393153, + "grad_norm": 0.9169338345527649, + "learning_rate": 2.2573512493113797e-06, + "loss": 0.5588, + "step": 3892 + }, + { + "epoch": 3.2228050800662618, + "grad_norm": 0.9455548524856567, + "learning_rate": 2.2562567755249304e-06, + "loss": 0.5725, + "step": 3893 + }, + { + "epoch": 3.2236333517393705, + "grad_norm": 0.9347724318504333, + "learning_rate": 2.2551623489005233e-06, + "loss": 0.5659, + "step": 3894 + }, + { + "epoch": 3.224461623412479, + "grad_norm": 0.9664618372917175, + "learning_rate": 2.2540679696499194e-06, + "loss": 0.5584, + "step": 3895 + }, + { + "epoch": 3.225289895085588, + "grad_norm": 1.0180222988128662, + "learning_rate": 2.252973637984871e-06, + "loss": 0.5586, + "step": 3896 + }, + { + "epoch": 3.2261181667586967, + "grad_norm": 0.9532806277275085, + "learning_rate": 2.2518793541171204e-06, + "loss": 0.5583, + "step": 3897 + }, + { + "epoch": 3.2269464384318054, + "grad_norm": 0.9502059817314148, + "learning_rate": 2.2507851182584024e-06, + "loss": 0.5676, + "step": 3898 + }, + { + "epoch": 3.2277747101049146, + "grad_norm": 0.9380828738212585, + "learning_rate": 2.24969093062044e-06, + "loss": 0.5644, + "step": 3899 + }, + { + "epoch": 3.2286029817780233, + "grad_norm": 0.9528313875198364, + "learning_rate": 2.2485967914149503e-06, + "loss": 0.5704, + "step": 3900 + }, + { + "epoch": 3.229431253451132, + "grad_norm": 0.9411875009536743, + "learning_rate": 2.2475027008536374e-06, + "loss": 0.5694, + "step": 3901 + }, + { + "epoch": 3.2302595251242407, + "grad_norm": 0.9563047289848328, + "learning_rate": 2.2464086591481977e-06, + "loss": 0.5654, + "step": 3902 + }, + { + "epoch": 3.2310877967973495, + "grad_norm": 0.9029218554496765, + "learning_rate": 2.24531466651032e-06, + "loss": 0.5818, + "step": 3903 + }, + { + "epoch": 3.231916068470458, + "grad_norm": 0.9468775391578674, + "learning_rate": 2.244220723151679e-06, + "loss": 0.548, + "step": 3904 + }, + { + "epoch": 3.232744340143567, + "grad_norm": 0.9399493932723999, + "learning_rate": 2.243126829283944e-06, + "loss": 0.5746, + "step": 3905 + }, + { + "epoch": 3.233572611816676, + "grad_norm": 0.9424317479133606, + "learning_rate": 2.2420329851187726e-06, + "loss": 0.5775, + "step": 3906 + }, + { + "epoch": 3.234400883489785, + "grad_norm": 0.9486375451087952, + "learning_rate": 2.2409391908678137e-06, + "loss": 0.5815, + "step": 3907 + }, + { + "epoch": 3.2352291551628936, + "grad_norm": 0.9591338634490967, + "learning_rate": 2.2398454467427073e-06, + "loss": 0.5826, + "step": 3908 + }, + { + "epoch": 3.2360574268360023, + "grad_norm": 0.9704396724700928, + "learning_rate": 2.2387517529550813e-06, + "loss": 0.5912, + "step": 3909 + }, + { + "epoch": 3.236885698509111, + "grad_norm": 0.9416005611419678, + "learning_rate": 2.2376581097165553e-06, + "loss": 0.5614, + "step": 3910 + }, + { + "epoch": 3.2377139701822197, + "grad_norm": 1.0255080461502075, + "learning_rate": 2.2365645172387403e-06, + "loss": 0.5688, + "step": 3911 + }, + { + "epoch": 3.2385422418553285, + "grad_norm": 0.9364955425262451, + "learning_rate": 2.2354709757332356e-06, + "loss": 0.572, + "step": 3912 + }, + { + "epoch": 3.239370513528437, + "grad_norm": 0.9503747224807739, + "learning_rate": 2.2343774854116306e-06, + "loss": 0.5893, + "step": 3913 + }, + { + "epoch": 3.240198785201546, + "grad_norm": 0.9294464588165283, + "learning_rate": 2.2332840464855072e-06, + "loss": 0.5671, + "step": 3914 + }, + { + "epoch": 3.241027056874655, + "grad_norm": 0.9659671783447266, + "learning_rate": 2.232190659166433e-06, + "loss": 0.5741, + "step": 3915 + }, + { + "epoch": 3.241855328547764, + "grad_norm": 0.9359234571456909, + "learning_rate": 2.231097323665971e-06, + "loss": 0.5608, + "step": 3916 + }, + { + "epoch": 3.2426836002208725, + "grad_norm": 0.9326402544975281, + "learning_rate": 2.2300040401956693e-06, + "loss": 0.5696, + "step": 3917 + }, + { + "epoch": 3.2435118718939813, + "grad_norm": 0.9257575869560242, + "learning_rate": 2.2289108089670687e-06, + "loss": 0.5709, + "step": 3918 + }, + { + "epoch": 3.24434014356709, + "grad_norm": 0.9669248461723328, + "learning_rate": 2.2278176301917004e-06, + "loss": 0.5666, + "step": 3919 + }, + { + "epoch": 3.2451684152401987, + "grad_norm": 0.9328755736351013, + "learning_rate": 2.2267245040810832e-06, + "loss": 0.573, + "step": 3920 + }, + { + "epoch": 3.2459966869133074, + "grad_norm": 0.942805826663971, + "learning_rate": 2.225631430846727e-06, + "loss": 0.5712, + "step": 3921 + }, + { + "epoch": 3.246824958586416, + "grad_norm": 0.9514304995536804, + "learning_rate": 2.224538410700131e-06, + "loss": 0.5775, + "step": 3922 + }, + { + "epoch": 3.247653230259525, + "grad_norm": 0.9471868872642517, + "learning_rate": 2.2234454438527847e-06, + "loss": 0.5794, + "step": 3923 + }, + { + "epoch": 3.248481501932634, + "grad_norm": 0.9507884979248047, + "learning_rate": 2.2223525305161675e-06, + "loss": 0.5726, + "step": 3924 + }, + { + "epoch": 3.249309773605743, + "grad_norm": 0.9477060437202454, + "learning_rate": 2.221259670901747e-06, + "loss": 0.5895, + "step": 3925 + }, + { + "epoch": 3.2501380452788515, + "grad_norm": 0.9437663555145264, + "learning_rate": 2.2201668652209823e-06, + "loss": 0.5538, + "step": 3926 + }, + { + "epoch": 3.2509663169519603, + "grad_norm": 0.9790158271789551, + "learning_rate": 2.2190741136853205e-06, + "loss": 0.5729, + "step": 3927 + }, + { + "epoch": 3.251794588625069, + "grad_norm": 0.9320138692855835, + "learning_rate": 2.2179814165061984e-06, + "loss": 0.5807, + "step": 3928 + }, + { + "epoch": 3.2526228602981777, + "grad_norm": 0.9543532729148865, + "learning_rate": 2.2168887738950438e-06, + "loss": 0.5645, + "step": 3929 + }, + { + "epoch": 3.2534511319712864, + "grad_norm": 0.9545353055000305, + "learning_rate": 2.2157961860632716e-06, + "loss": 0.5762, + "step": 3930 + }, + { + "epoch": 3.2542794036443956, + "grad_norm": 0.9640288352966309, + "learning_rate": 2.2147036532222887e-06, + "loss": 0.5738, + "step": 3931 + }, + { + "epoch": 3.2551076753175043, + "grad_norm": 0.9393999576568604, + "learning_rate": 2.2136111755834895e-06, + "loss": 0.5692, + "step": 3932 + }, + { + "epoch": 3.255935946990613, + "grad_norm": 0.9366894364356995, + "learning_rate": 2.212518753358258e-06, + "loss": 0.5957, + "step": 3933 + }, + { + "epoch": 3.256764218663722, + "grad_norm": 0.9408980011940002, + "learning_rate": 2.2114263867579673e-06, + "loss": 0.5644, + "step": 3934 + }, + { + "epoch": 3.2575924903368305, + "grad_norm": 0.9842433333396912, + "learning_rate": 2.2103340759939816e-06, + "loss": 0.5772, + "step": 3935 + }, + { + "epoch": 3.2584207620099392, + "grad_norm": 0.9410622119903564, + "learning_rate": 2.209241821277651e-06, + "loss": 0.5637, + "step": 3936 + }, + { + "epoch": 3.259249033683048, + "grad_norm": 0.9571241736412048, + "learning_rate": 2.208149622820319e-06, + "loss": 0.5754, + "step": 3937 + }, + { + "epoch": 3.2600773053561567, + "grad_norm": 0.9451292157173157, + "learning_rate": 2.2070574808333135e-06, + "loss": 0.557, + "step": 3938 + }, + { + "epoch": 3.2609055770292654, + "grad_norm": 0.987385094165802, + "learning_rate": 2.205965395527955e-06, + "loss": 0.5458, + "step": 3939 + }, + { + "epoch": 3.2617338487023746, + "grad_norm": 0.9282477498054504, + "learning_rate": 2.204873367115551e-06, + "loss": 0.5726, + "step": 3940 + }, + { + "epoch": 3.2625621203754833, + "grad_norm": 0.9609903693199158, + "learning_rate": 2.2037813958073994e-06, + "loss": 0.5608, + "step": 3941 + }, + { + "epoch": 3.263390392048592, + "grad_norm": 0.9469774961471558, + "learning_rate": 2.2026894818147864e-06, + "loss": 0.5663, + "step": 3942 + }, + { + "epoch": 3.2642186637217008, + "grad_norm": 0.9910560846328735, + "learning_rate": 2.2015976253489873e-06, + "loss": 0.5604, + "step": 3943 + }, + { + "epoch": 3.2650469353948095, + "grad_norm": 0.949170708656311, + "learning_rate": 2.2005058266212657e-06, + "loss": 0.5795, + "step": 3944 + }, + { + "epoch": 3.2658752070679182, + "grad_norm": 0.9276071786880493, + "learning_rate": 2.199414085842875e-06, + "loss": 0.5821, + "step": 3945 + }, + { + "epoch": 3.266703478741027, + "grad_norm": 0.9555459022521973, + "learning_rate": 2.1983224032250573e-06, + "loss": 0.5832, + "step": 3946 + }, + { + "epoch": 3.2675317504141357, + "grad_norm": 0.972984790802002, + "learning_rate": 2.1972307789790413e-06, + "loss": 0.5868, + "step": 3947 + }, + { + "epoch": 3.2683600220872444, + "grad_norm": 0.9383103251457214, + "learning_rate": 2.196139213316049e-06, + "loss": 0.5624, + "step": 3948 + }, + { + "epoch": 3.2691882937603536, + "grad_norm": 0.9539969563484192, + "learning_rate": 2.195047706447285e-06, + "loss": 0.566, + "step": 3949 + }, + { + "epoch": 3.2700165654334623, + "grad_norm": 0.955715537071228, + "learning_rate": 2.193956258583947e-06, + "loss": 0.5548, + "step": 3950 + }, + { + "epoch": 3.270844837106571, + "grad_norm": 0.9660782217979431, + "learning_rate": 2.19286486993722e-06, + "loss": 0.5726, + "step": 3951 + }, + { + "epoch": 3.2716731087796798, + "grad_norm": 0.9660666584968567, + "learning_rate": 2.1917735407182777e-06, + "loss": 0.5728, + "step": 3952 + }, + { + "epoch": 3.2725013804527885, + "grad_norm": 0.9096773862838745, + "learning_rate": 2.190682271138282e-06, + "loss": 0.5549, + "step": 3953 + }, + { + "epoch": 3.273329652125897, + "grad_norm": 0.9581109285354614, + "learning_rate": 2.189591061408384e-06, + "loss": 0.5834, + "step": 3954 + }, + { + "epoch": 3.274157923799006, + "grad_norm": 0.9483600854873657, + "learning_rate": 2.188499911739721e-06, + "loss": 0.5749, + "step": 3955 + }, + { + "epoch": 3.2749861954721147, + "grad_norm": 0.9495676159858704, + "learning_rate": 2.187408822343422e-06, + "loss": 0.5907, + "step": 3956 + }, + { + "epoch": 3.2758144671452234, + "grad_norm": 0.9316544532775879, + "learning_rate": 2.186317793430602e-06, + "loss": 0.555, + "step": 3957 + }, + { + "epoch": 3.2766427388183326, + "grad_norm": 0.9391309022903442, + "learning_rate": 2.1852268252123644e-06, + "loss": 0.5751, + "step": 3958 + }, + { + "epoch": 3.2774710104914413, + "grad_norm": 0.91743403673172, + "learning_rate": 2.1841359178998024e-06, + "loss": 0.5504, + "step": 3959 + }, + { + "epoch": 3.27829928216455, + "grad_norm": 0.9177739024162292, + "learning_rate": 2.183045071703996e-06, + "loss": 0.5452, + "step": 3960 + }, + { + "epoch": 3.2791275538376587, + "grad_norm": 0.964051365852356, + "learning_rate": 2.1819542868360135e-06, + "loss": 0.5731, + "step": 3961 + }, + { + "epoch": 3.2799558255107675, + "grad_norm": 0.9402073621749878, + "learning_rate": 2.1808635635069113e-06, + "loss": 0.5655, + "step": 3962 + }, + { + "epoch": 3.280784097183876, + "grad_norm": 0.9163627624511719, + "learning_rate": 2.1797729019277346e-06, + "loss": 0.5772, + "step": 3963 + }, + { + "epoch": 3.281612368856985, + "grad_norm": 0.9189632534980774, + "learning_rate": 2.1786823023095164e-06, + "loss": 0.5738, + "step": 3964 + }, + { + "epoch": 3.282440640530094, + "grad_norm": 0.9654256701469421, + "learning_rate": 2.1775917648632772e-06, + "loss": 0.5752, + "step": 3965 + }, + { + "epoch": 3.283268912203203, + "grad_norm": 1.0078915357589722, + "learning_rate": 2.1765012898000257e-06, + "loss": 0.5847, + "step": 3966 + }, + { + "epoch": 3.2840971838763116, + "grad_norm": 0.9313945174217224, + "learning_rate": 2.17541087733076e-06, + "loss": 0.5671, + "step": 3967 + }, + { + "epoch": 3.2849254555494203, + "grad_norm": 0.9611067771911621, + "learning_rate": 2.1743205276664634e-06, + "loss": 0.5501, + "step": 3968 + }, + { + "epoch": 3.285753727222529, + "grad_norm": 0.9783805012702942, + "learning_rate": 2.1732302410181084e-06, + "loss": 0.5969, + "step": 3969 + }, + { + "epoch": 3.2865819988956377, + "grad_norm": 0.9382510781288147, + "learning_rate": 2.172140017596656e-06, + "loss": 0.5716, + "step": 3970 + }, + { + "epoch": 3.2874102705687465, + "grad_norm": 0.9219527840614319, + "learning_rate": 2.1710498576130545e-06, + "loss": 0.5916, + "step": 3971 + }, + { + "epoch": 3.288238542241855, + "grad_norm": 0.9681289792060852, + "learning_rate": 2.1699597612782385e-06, + "loss": 0.5725, + "step": 3972 + }, + { + "epoch": 3.289066813914964, + "grad_norm": 0.946723997592926, + "learning_rate": 2.168869728803132e-06, + "loss": 0.5664, + "step": 3973 + }, + { + "epoch": 3.289895085588073, + "grad_norm": 0.9302477836608887, + "learning_rate": 2.167779760398646e-06, + "loss": 0.5777, + "step": 3974 + }, + { + "epoch": 3.290723357261182, + "grad_norm": 0.9480409026145935, + "learning_rate": 2.16668985627568e-06, + "loss": 0.5696, + "step": 3975 + }, + { + "epoch": 3.2915516289342905, + "grad_norm": 0.9363623261451721, + "learning_rate": 2.1656000166451195e-06, + "loss": 0.5798, + "step": 3976 + }, + { + "epoch": 3.2923799006073993, + "grad_norm": 0.9291804432868958, + "learning_rate": 2.164510241717838e-06, + "loss": 0.5658, + "step": 3977 + }, + { + "epoch": 3.293208172280508, + "grad_norm": 0.9314830899238586, + "learning_rate": 2.163420531704698e-06, + "loss": 0.5698, + "step": 3978 + }, + { + "epoch": 3.2940364439536167, + "grad_norm": 0.974621057510376, + "learning_rate": 2.1623308868165474e-06, + "loss": 0.5644, + "step": 3979 + }, + { + "epoch": 3.2948647156267254, + "grad_norm": 0.9464275240898132, + "learning_rate": 2.161241307264222e-06, + "loss": 0.5704, + "step": 3980 + }, + { + "epoch": 3.295692987299834, + "grad_norm": 0.9375281929969788, + "learning_rate": 2.1601517932585462e-06, + "loss": 0.5714, + "step": 3981 + }, + { + "epoch": 3.296521258972943, + "grad_norm": 0.910499095916748, + "learning_rate": 2.159062345010331e-06, + "loss": 0.5639, + "step": 3982 + }, + { + "epoch": 3.297349530646052, + "grad_norm": 0.9525747895240784, + "learning_rate": 2.1579729627303735e-06, + "loss": 0.5703, + "step": 3983 + }, + { + "epoch": 3.298177802319161, + "grad_norm": 0.9756594896316528, + "learning_rate": 2.1568836466294596e-06, + "loss": 0.5862, + "step": 3984 + }, + { + "epoch": 3.2990060739922695, + "grad_norm": 0.9418894648551941, + "learning_rate": 2.1557943969183605e-06, + "loss": 0.5554, + "step": 3985 + }, + { + "epoch": 3.2998343456653783, + "grad_norm": 0.9190452694892883, + "learning_rate": 2.154705213807838e-06, + "loss": 0.5519, + "step": 3986 + }, + { + "epoch": 3.300662617338487, + "grad_norm": 0.9576460719108582, + "learning_rate": 2.153616097508637e-06, + "loss": 0.5723, + "step": 3987 + }, + { + "epoch": 3.3014908890115957, + "grad_norm": 0.9196321964263916, + "learning_rate": 2.1525270482314924e-06, + "loss": 0.6006, + "step": 3988 + }, + { + "epoch": 3.3023191606847044, + "grad_norm": 0.9210584163665771, + "learning_rate": 2.1514380661871253e-06, + "loss": 0.5569, + "step": 3989 + }, + { + "epoch": 3.3031474323578136, + "grad_norm": 0.9365322589874268, + "learning_rate": 2.150349151586243e-06, + "loss": 0.5747, + "step": 3990 + }, + { + "epoch": 3.3039757040309223, + "grad_norm": 0.948229193687439, + "learning_rate": 2.1492603046395407e-06, + "loss": 0.556, + "step": 3991 + }, + { + "epoch": 3.304803975704031, + "grad_norm": 0.9285109043121338, + "learning_rate": 2.148171525557699e-06, + "loss": 0.5747, + "step": 3992 + }, + { + "epoch": 3.30563224737714, + "grad_norm": 0.9224300384521484, + "learning_rate": 2.1470828145513882e-06, + "loss": 0.5883, + "step": 3993 + }, + { + "epoch": 3.3064605190502485, + "grad_norm": 0.9375912547111511, + "learning_rate": 2.1459941718312633e-06, + "loss": 0.5663, + "step": 3994 + }, + { + "epoch": 3.3072887907233572, + "grad_norm": 0.9629141688346863, + "learning_rate": 2.144905597607966e-06, + "loss": 0.5593, + "step": 3995 + }, + { + "epoch": 3.308117062396466, + "grad_norm": 0.9608798623085022, + "learning_rate": 2.143817092092124e-06, + "loss": 0.573, + "step": 3996 + }, + { + "epoch": 3.3089453340695747, + "grad_norm": 0.9235004186630249, + "learning_rate": 2.1427286554943557e-06, + "loss": 0.5818, + "step": 3997 + }, + { + "epoch": 3.3097736057426834, + "grad_norm": 0.9361316561698914, + "learning_rate": 2.141640288025262e-06, + "loss": 0.556, + "step": 3998 + }, + { + "epoch": 3.3106018774157926, + "grad_norm": 0.9709519743919373, + "learning_rate": 2.140551989895432e-06, + "loss": 0.5575, + "step": 3999 + }, + { + "epoch": 3.3114301490889013, + "grad_norm": 0.9461862444877625, + "learning_rate": 2.1394637613154403e-06, + "loss": 0.5676, + "step": 4000 + }, + { + "epoch": 3.31225842076201, + "grad_norm": 0.9697636961936951, + "learning_rate": 2.138375602495851e-06, + "loss": 0.5678, + "step": 4001 + }, + { + "epoch": 3.3130866924351188, + "grad_norm": 0.9442050457000732, + "learning_rate": 2.1372875136472113e-06, + "loss": 0.573, + "step": 4002 + }, + { + "epoch": 3.3139149641082275, + "grad_norm": 0.9325514435768127, + "learning_rate": 2.136199494980056e-06, + "loss": 0.5711, + "step": 4003 + }, + { + "epoch": 3.3147432357813362, + "grad_norm": 0.934421181678772, + "learning_rate": 2.135111546704908e-06, + "loss": 0.5634, + "step": 4004 + }, + { + "epoch": 3.315571507454445, + "grad_norm": 0.9201632738113403, + "learning_rate": 2.134023669032274e-06, + "loss": 0.5627, + "step": 4005 + }, + { + "epoch": 3.3163997791275537, + "grad_norm": 0.9563625454902649, + "learning_rate": 2.1329358621726484e-06, + "loss": 0.5668, + "step": 4006 + }, + { + "epoch": 3.3172280508006624, + "grad_norm": 0.9408397078514099, + "learning_rate": 2.1318481263365117e-06, + "loss": 0.5712, + "step": 4007 + }, + { + "epoch": 3.3180563224737716, + "grad_norm": 0.9219162464141846, + "learning_rate": 2.13076046173433e-06, + "loss": 0.5694, + "step": 4008 + }, + { + "epoch": 3.3188845941468803, + "grad_norm": 0.9441806077957153, + "learning_rate": 2.129672868576558e-06, + "loss": 0.5901, + "step": 4009 + }, + { + "epoch": 3.319712865819989, + "grad_norm": 0.9196688532829285, + "learning_rate": 2.1285853470736336e-06, + "loss": 0.5768, + "step": 4010 + }, + { + "epoch": 3.3205411374930978, + "grad_norm": 0.9800437688827515, + "learning_rate": 2.1274978974359813e-06, + "loss": 0.5728, + "step": 4011 + }, + { + "epoch": 3.3213694091662065, + "grad_norm": 0.9329850077629089, + "learning_rate": 2.126410519874014e-06, + "loss": 0.5806, + "step": 4012 + }, + { + "epoch": 3.322197680839315, + "grad_norm": 0.9533185362815857, + "learning_rate": 2.1253232145981293e-06, + "loss": 0.5665, + "step": 4013 + }, + { + "epoch": 3.323025952512424, + "grad_norm": 0.9292604327201843, + "learning_rate": 2.1242359818187085e-06, + "loss": 0.5552, + "step": 4014 + }, + { + "epoch": 3.3238542241855327, + "grad_norm": 0.9430820941925049, + "learning_rate": 2.1231488217461234e-06, + "loss": 0.5849, + "step": 4015 + }, + { + "epoch": 3.3246824958586414, + "grad_norm": 0.9454953670501709, + "learning_rate": 2.1220617345907287e-06, + "loss": 0.5724, + "step": 4016 + }, + { + "epoch": 3.3255107675317506, + "grad_norm": 0.9445587396621704, + "learning_rate": 2.120974720562865e-06, + "loss": 0.5908, + "step": 4017 + }, + { + "epoch": 3.3263390392048593, + "grad_norm": 0.9306840896606445, + "learning_rate": 2.1198877798728595e-06, + "loss": 0.5938, + "step": 4018 + }, + { + "epoch": 3.327167310877968, + "grad_norm": 0.9418679475784302, + "learning_rate": 2.1188009127310244e-06, + "loss": 0.5692, + "step": 4019 + }, + { + "epoch": 3.3279955825510767, + "grad_norm": 0.9448477625846863, + "learning_rate": 2.11771411934766e-06, + "loss": 0.578, + "step": 4020 + }, + { + "epoch": 3.3288238542241855, + "grad_norm": 0.9564743638038635, + "learning_rate": 2.1166273999330497e-06, + "loss": 0.5577, + "step": 4021 + }, + { + "epoch": 3.329652125897294, + "grad_norm": 0.9450514316558838, + "learning_rate": 2.115540754697463e-06, + "loss": 0.5812, + "step": 4022 + }, + { + "epoch": 3.330480397570403, + "grad_norm": 0.932601809501648, + "learning_rate": 2.1144541838511578e-06, + "loss": 0.5534, + "step": 4023 + }, + { + "epoch": 3.331308669243512, + "grad_norm": 0.9476706385612488, + "learning_rate": 2.1133676876043734e-06, + "loss": 0.5684, + "step": 4024 + }, + { + "epoch": 3.332136940916621, + "grad_norm": 0.9359769225120544, + "learning_rate": 2.1122812661673365e-06, + "loss": 0.5624, + "step": 4025 + }, + { + "epoch": 3.3329652125897296, + "grad_norm": 0.9415692687034607, + "learning_rate": 2.1111949197502612e-06, + "loss": 0.5627, + "step": 4026 + }, + { + "epoch": 3.3337934842628383, + "grad_norm": 0.9686550498008728, + "learning_rate": 2.1101086485633455e-06, + "loss": 0.5401, + "step": 4027 + }, + { + "epoch": 3.334621755935947, + "grad_norm": 0.9065554738044739, + "learning_rate": 2.1090224528167703e-06, + "loss": 0.5785, + "step": 4028 + }, + { + "epoch": 3.3354500276090557, + "grad_norm": 0.9449801445007324, + "learning_rate": 2.1079363327207067e-06, + "loss": 0.5695, + "step": 4029 + }, + { + "epoch": 3.3362782992821645, + "grad_norm": 0.9398133754730225, + "learning_rate": 2.106850288485307e-06, + "loss": 0.5605, + "step": 4030 + }, + { + "epoch": 3.337106570955273, + "grad_norm": 0.9175443649291992, + "learning_rate": 2.1057643203207125e-06, + "loss": 0.5675, + "step": 4031 + }, + { + "epoch": 3.337934842628382, + "grad_norm": 0.9433302879333496, + "learning_rate": 2.1046784284370466e-06, + "loss": 0.5501, + "step": 4032 + }, + { + "epoch": 3.338763114301491, + "grad_norm": 0.9375688433647156, + "learning_rate": 2.103592613044419e-06, + "loss": 0.579, + "step": 4033 + }, + { + "epoch": 3.3395913859746, + "grad_norm": 0.9344912171363831, + "learning_rate": 2.102506874352926e-06, + "loss": 0.5698, + "step": 4034 + }, + { + "epoch": 3.3404196576477085, + "grad_norm": 0.9581173062324524, + "learning_rate": 2.101421212572648e-06, + "loss": 0.5774, + "step": 4035 + }, + { + "epoch": 3.3412479293208173, + "grad_norm": 0.9601131081581116, + "learning_rate": 2.100335627913649e-06, + "loss": 0.5769, + "step": 4036 + }, + { + "epoch": 3.342076200993926, + "grad_norm": 0.9634785652160645, + "learning_rate": 2.0992501205859806e-06, + "loss": 0.5883, + "step": 4037 + }, + { + "epoch": 3.3429044726670347, + "grad_norm": 1.0006943941116333, + "learning_rate": 2.0981646907996784e-06, + "loss": 0.5744, + "step": 4038 + }, + { + "epoch": 3.3437327443401434, + "grad_norm": 0.9540338516235352, + "learning_rate": 2.0970793387647636e-06, + "loss": 0.5775, + "step": 4039 + }, + { + "epoch": 3.344561016013252, + "grad_norm": 0.9318099617958069, + "learning_rate": 2.09599406469124e-06, + "loss": 0.5765, + "step": 4040 + }, + { + "epoch": 3.345389287686361, + "grad_norm": 0.9622073173522949, + "learning_rate": 2.0949088687890976e-06, + "loss": 0.5532, + "step": 4041 + }, + { + "epoch": 3.34621755935947, + "grad_norm": 0.9623757600784302, + "learning_rate": 2.0938237512683142e-06, + "loss": 0.5518, + "step": 4042 + }, + { + "epoch": 3.347045831032579, + "grad_norm": 0.9352701306343079, + "learning_rate": 2.0927387123388485e-06, + "loss": 0.5488, + "step": 4043 + }, + { + "epoch": 3.3478741027056875, + "grad_norm": 0.9591230154037476, + "learning_rate": 2.0916537522106446e-06, + "loss": 0.5887, + "step": 4044 + }, + { + "epoch": 3.3487023743787963, + "grad_norm": 0.9517300724983215, + "learning_rate": 2.0905688710936338e-06, + "loss": 0.5619, + "step": 4045 + }, + { + "epoch": 3.349530646051905, + "grad_norm": 0.9537909626960754, + "learning_rate": 2.08948406919773e-06, + "loss": 0.5949, + "step": 4046 + }, + { + "epoch": 3.3503589177250137, + "grad_norm": 0.946260929107666, + "learning_rate": 2.088399346732831e-06, + "loss": 0.5989, + "step": 4047 + }, + { + "epoch": 3.3511871893981224, + "grad_norm": 0.9061339497566223, + "learning_rate": 2.0873147039088227e-06, + "loss": 0.5855, + "step": 4048 + }, + { + "epoch": 3.3520154610712316, + "grad_norm": 0.9527447819709778, + "learning_rate": 2.0862301409355717e-06, + "loss": 0.5703, + "step": 4049 + }, + { + "epoch": 3.3528437327443403, + "grad_norm": 0.9357138872146606, + "learning_rate": 2.0851456580229325e-06, + "loss": 0.551, + "step": 4050 + }, + { + "epoch": 3.353672004417449, + "grad_norm": 0.9259575009346008, + "learning_rate": 2.08406125538074e-06, + "loss": 0.5777, + "step": 4051 + }, + { + "epoch": 3.354500276090558, + "grad_norm": 0.944807231426239, + "learning_rate": 2.082976933218817e-06, + "loss": 0.5741, + "step": 4052 + }, + { + "epoch": 3.3553285477636665, + "grad_norm": 0.9251223206520081, + "learning_rate": 2.08189269174697e-06, + "loss": 0.5829, + "step": 4053 + }, + { + "epoch": 3.3561568194367752, + "grad_norm": 0.9122853875160217, + "learning_rate": 2.0808085311749904e-06, + "loss": 0.5602, + "step": 4054 + }, + { + "epoch": 3.356985091109884, + "grad_norm": 0.9430744647979736, + "learning_rate": 2.0797244517126512e-06, + "loss": 0.5633, + "step": 4055 + }, + { + "epoch": 3.3578133627829927, + "grad_norm": 0.9500707983970642, + "learning_rate": 2.078640453569714e-06, + "loss": 0.5791, + "step": 4056 + }, + { + "epoch": 3.3586416344561014, + "grad_norm": 0.970301628112793, + "learning_rate": 2.077556536955921e-06, + "loss": 0.5668, + "step": 4057 + }, + { + "epoch": 3.3594699061292106, + "grad_norm": 0.9528861045837402, + "learning_rate": 2.0764727020809995e-06, + "loss": 0.5616, + "step": 4058 + }, + { + "epoch": 3.3602981778023193, + "grad_norm": 0.9321720004081726, + "learning_rate": 2.0753889491546624e-06, + "loss": 0.5841, + "step": 4059 + }, + { + "epoch": 3.361126449475428, + "grad_norm": 0.9030178189277649, + "learning_rate": 2.0743052783866054e-06, + "loss": 0.5757, + "step": 4060 + }, + { + "epoch": 3.3619547211485368, + "grad_norm": 0.9501296877861023, + "learning_rate": 2.0732216899865098e-06, + "loss": 0.5477, + "step": 4061 + }, + { + "epoch": 3.3627829928216455, + "grad_norm": 0.9475130438804626, + "learning_rate": 2.0721381841640377e-06, + "loss": 0.5798, + "step": 4062 + }, + { + "epoch": 3.3636112644947542, + "grad_norm": 0.954949676990509, + "learning_rate": 2.0710547611288384e-06, + "loss": 0.579, + "step": 4063 + }, + { + "epoch": 3.364439536167863, + "grad_norm": 0.9534462690353394, + "learning_rate": 2.069971421090545e-06, + "loss": 0.5448, + "step": 4064 + }, + { + "epoch": 3.3652678078409717, + "grad_norm": 0.9504910111427307, + "learning_rate": 2.068888164258772e-06, + "loss": 0.5698, + "step": 4065 + }, + { + "epoch": 3.3660960795140804, + "grad_norm": 0.934983491897583, + "learning_rate": 2.067804990843121e-06, + "loss": 0.5643, + "step": 4066 + }, + { + "epoch": 3.3669243511871896, + "grad_norm": 0.9342026114463806, + "learning_rate": 2.066721901053175e-06, + "loss": 0.5938, + "step": 4067 + }, + { + "epoch": 3.3677526228602983, + "grad_norm": 0.9376976490020752, + "learning_rate": 2.0656388950985027e-06, + "loss": 0.5897, + "step": 4068 + }, + { + "epoch": 3.368580894533407, + "grad_norm": 0.9228143095970154, + "learning_rate": 2.064555973188655e-06, + "loss": 0.5789, + "step": 4069 + }, + { + "epoch": 3.3694091662065158, + "grad_norm": 0.9547181129455566, + "learning_rate": 2.0634731355331665e-06, + "loss": 0.5707, + "step": 4070 + }, + { + "epoch": 3.3702374378796245, + "grad_norm": 0.9388718605041504, + "learning_rate": 2.062390382341558e-06, + "loss": 0.5719, + "step": 4071 + }, + { + "epoch": 3.371065709552733, + "grad_norm": 0.9529668092727661, + "learning_rate": 2.0613077138233313e-06, + "loss": 0.5634, + "step": 4072 + }, + { + "epoch": 3.371893981225842, + "grad_norm": 0.9569711089134216, + "learning_rate": 2.060225130187973e-06, + "loss": 0.5565, + "step": 4073 + }, + { + "epoch": 3.3727222528989507, + "grad_norm": 0.936896026134491, + "learning_rate": 2.0591426316449513e-06, + "loss": 0.5543, + "step": 4074 + }, + { + "epoch": 3.3735505245720594, + "grad_norm": 0.9453547596931458, + "learning_rate": 2.0580602184037206e-06, + "loss": 0.5652, + "step": 4075 + }, + { + "epoch": 3.3743787962451686, + "grad_norm": 0.9260196089744568, + "learning_rate": 2.0569778906737186e-06, + "loss": 0.5765, + "step": 4076 + }, + { + "epoch": 3.3752070679182773, + "grad_norm": 0.9467774629592896, + "learning_rate": 2.055895648664365e-06, + "loss": 0.5637, + "step": 4077 + }, + { + "epoch": 3.376035339591386, + "grad_norm": 0.9502113461494446, + "learning_rate": 2.0548134925850626e-06, + "loss": 0.5725, + "step": 4078 + }, + { + "epoch": 3.3768636112644947, + "grad_norm": 0.9475006461143494, + "learning_rate": 2.0537314226452004e-06, + "loss": 0.5936, + "step": 4079 + }, + { + "epoch": 3.3776918829376035, + "grad_norm": 0.9401916861534119, + "learning_rate": 2.0526494390541478e-06, + "loss": 0.5797, + "step": 4080 + }, + { + "epoch": 3.378520154610712, + "grad_norm": 0.9407974481582642, + "learning_rate": 2.0515675420212577e-06, + "loss": 0.5653, + "step": 4081 + }, + { + "epoch": 3.379348426283821, + "grad_norm": 0.9521286487579346, + "learning_rate": 2.0504857317558686e-06, + "loss": 0.5794, + "step": 4082 + }, + { + "epoch": 3.38017669795693, + "grad_norm": 0.9705640077590942, + "learning_rate": 2.0494040084673004e-06, + "loss": 0.5821, + "step": 4083 + }, + { + "epoch": 3.381004969630039, + "grad_norm": 0.9214118719100952, + "learning_rate": 2.0483223723648567e-06, + "loss": 0.5757, + "step": 4084 + }, + { + "epoch": 3.3818332413031476, + "grad_norm": 0.9589064717292786, + "learning_rate": 2.0472408236578228e-06, + "loss": 0.5551, + "step": 4085 + }, + { + "epoch": 3.3826615129762563, + "grad_norm": 0.9503619074821472, + "learning_rate": 2.0461593625554683e-06, + "loss": 0.5805, + "step": 4086 + }, + { + "epoch": 3.383489784649365, + "grad_norm": 0.9328595995903015, + "learning_rate": 2.045077989267047e-06, + "loss": 0.5736, + "step": 4087 + }, + { + "epoch": 3.3843180563224737, + "grad_norm": 0.9475532174110413, + "learning_rate": 2.043996704001794e-06, + "loss": 0.5652, + "step": 4088 + }, + { + "epoch": 3.3851463279955825, + "grad_norm": 0.9307344555854797, + "learning_rate": 2.042915506968928e-06, + "loss": 0.5801, + "step": 4089 + }, + { + "epoch": 3.385974599668691, + "grad_norm": 0.9430568218231201, + "learning_rate": 2.04183439837765e-06, + "loss": 0.5864, + "step": 4090 + }, + { + "epoch": 3.3868028713418, + "grad_norm": 0.9192523956298828, + "learning_rate": 2.0407533784371457e-06, + "loss": 0.5486, + "step": 4091 + }, + { + "epoch": 3.387631143014909, + "grad_norm": 0.9655841588973999, + "learning_rate": 2.039672447356581e-06, + "loss": 0.5602, + "step": 4092 + }, + { + "epoch": 3.388459414688018, + "grad_norm": 0.9553002119064331, + "learning_rate": 2.0385916053451073e-06, + "loss": 0.568, + "step": 4093 + }, + { + "epoch": 3.3892876863611265, + "grad_norm": 0.9539245963096619, + "learning_rate": 2.0375108526118563e-06, + "loss": 0.5898, + "step": 4094 + }, + { + "epoch": 3.3901159580342353, + "grad_norm": 0.9733588695526123, + "learning_rate": 2.036430189365945e-06, + "loss": 0.5817, + "step": 4095 + }, + { + "epoch": 3.390944229707344, + "grad_norm": 0.9487296938896179, + "learning_rate": 2.03534961581647e-06, + "loss": 0.5841, + "step": 4096 + }, + { + "epoch": 3.3917725013804527, + "grad_norm": 0.9788404107093811, + "learning_rate": 2.0342691321725127e-06, + "loss": 0.5951, + "step": 4097 + }, + { + "epoch": 3.3926007730535614, + "grad_norm": 0.9286518692970276, + "learning_rate": 2.033188738643137e-06, + "loss": 0.5555, + "step": 4098 + }, + { + "epoch": 3.39342904472667, + "grad_norm": 0.9736733436584473, + "learning_rate": 2.032108435437389e-06, + "loss": 0.5773, + "step": 4099 + }, + { + "epoch": 3.394257316399779, + "grad_norm": 0.9288291335105896, + "learning_rate": 2.0310282227642963e-06, + "loss": 0.5562, + "step": 4100 + }, + { + "epoch": 3.395085588072888, + "grad_norm": 0.9414420127868652, + "learning_rate": 2.0299481008328716e-06, + "loss": 0.5923, + "step": 4101 + }, + { + "epoch": 3.395913859745997, + "grad_norm": 0.9327258467674255, + "learning_rate": 2.028868069852108e-06, + "loss": 0.5636, + "step": 4102 + }, + { + "epoch": 3.3967421314191055, + "grad_norm": 0.9412753582000732, + "learning_rate": 2.02778813003098e-06, + "loss": 0.5896, + "step": 4103 + }, + { + "epoch": 3.3975704030922143, + "grad_norm": 0.973572313785553, + "learning_rate": 2.026708281578447e-06, + "loss": 0.5581, + "step": 4104 + }, + { + "epoch": 3.398398674765323, + "grad_norm": 1.1189706325531006, + "learning_rate": 2.0256285247034505e-06, + "loss": 0.5541, + "step": 4105 + }, + { + "epoch": 3.3992269464384317, + "grad_norm": 0.9494182467460632, + "learning_rate": 2.0245488596149113e-06, + "loss": 0.5686, + "step": 4106 + }, + { + "epoch": 3.4000552181115404, + "grad_norm": 0.9740265011787415, + "learning_rate": 2.0234692865217375e-06, + "loss": 0.5744, + "step": 4107 + }, + { + "epoch": 3.4008834897846496, + "grad_norm": 0.9565280079841614, + "learning_rate": 2.0223898056328133e-06, + "loss": 0.589, + "step": 4108 + }, + { + "epoch": 3.4017117614577583, + "grad_norm": 0.9839944839477539, + "learning_rate": 2.0213104171570094e-06, + "loss": 0.5704, + "step": 4109 + }, + { + "epoch": 3.402540033130867, + "grad_norm": 0.9117640256881714, + "learning_rate": 2.020231121303178e-06, + "loss": 0.5591, + "step": 4110 + }, + { + "epoch": 3.403368304803976, + "grad_norm": 0.9455560445785522, + "learning_rate": 2.019151918280151e-06, + "loss": 0.592, + "step": 4111 + }, + { + "epoch": 3.4041965764770845, + "grad_norm": 0.9271209836006165, + "learning_rate": 2.0180728082967464e-06, + "loss": 0.5832, + "step": 4112 + }, + { + "epoch": 3.4050248481501932, + "grad_norm": 0.9203625321388245, + "learning_rate": 2.016993791561761e-06, + "loss": 0.576, + "step": 4113 + }, + { + "epoch": 3.405853119823302, + "grad_norm": 0.9395371675491333, + "learning_rate": 2.015914868283974e-06, + "loss": 0.5823, + "step": 4114 + }, + { + "epoch": 3.4066813914964107, + "grad_norm": 0.9351215958595276, + "learning_rate": 2.014836038672148e-06, + "loss": 0.5903, + "step": 4115 + }, + { + "epoch": 3.4075096631695194, + "grad_norm": 0.9305378198623657, + "learning_rate": 2.0137573029350253e-06, + "loss": 0.5607, + "step": 4116 + }, + { + "epoch": 3.4083379348426286, + "grad_norm": 0.9378963708877563, + "learning_rate": 2.0126786612813316e-06, + "loss": 0.5648, + "step": 4117 + }, + { + "epoch": 3.4091662065157373, + "grad_norm": 1.2515968084335327, + "learning_rate": 2.011600113919776e-06, + "loss": 0.5915, + "step": 4118 + }, + { + "epoch": 3.409994478188846, + "grad_norm": 0.9324381947517395, + "learning_rate": 2.010521661059044e-06, + "loss": 0.5573, + "step": 4119 + }, + { + "epoch": 3.4108227498619548, + "grad_norm": 0.9508990049362183, + "learning_rate": 2.009443302907808e-06, + "loss": 0.5612, + "step": 4120 + }, + { + "epoch": 3.4116510215350635, + "grad_norm": 0.9441195130348206, + "learning_rate": 2.00836503967472e-06, + "loss": 0.5561, + "step": 4121 + }, + { + "epoch": 3.4124792932081722, + "grad_norm": 0.940579354763031, + "learning_rate": 2.007286871568413e-06, + "loss": 0.5787, + "step": 4122 + }, + { + "epoch": 3.413307564881281, + "grad_norm": 0.9296087026596069, + "learning_rate": 2.0062087987975043e-06, + "loss": 0.5638, + "step": 4123 + }, + { + "epoch": 3.4141358365543897, + "grad_norm": 0.9607263803482056, + "learning_rate": 2.00513082157059e-06, + "loss": 0.5903, + "step": 4124 + }, + { + "epoch": 3.4149641082274984, + "grad_norm": 0.9708670377731323, + "learning_rate": 2.004052940096248e-06, + "loss": 0.567, + "step": 4125 + }, + { + "epoch": 3.4157923799006076, + "grad_norm": 0.9483271241188049, + "learning_rate": 2.002975154583039e-06, + "loss": 0.5687, + "step": 4126 + }, + { + "epoch": 3.4166206515737163, + "grad_norm": 0.9517681002616882, + "learning_rate": 2.001897465239505e-06, + "loss": 0.5656, + "step": 4127 + }, + { + "epoch": 3.417448923246825, + "grad_norm": 0.9440080523490906, + "learning_rate": 2.0008198722741677e-06, + "loss": 0.5751, + "step": 4128 + }, + { + "epoch": 3.4182771949199338, + "grad_norm": 0.9267042279243469, + "learning_rate": 1.999742375895533e-06, + "loss": 0.5708, + "step": 4129 + }, + { + "epoch": 3.4191054665930425, + "grad_norm": 0.9693537950515747, + "learning_rate": 1.998664976312084e-06, + "loss": 0.5859, + "step": 4130 + }, + { + "epoch": 3.419933738266151, + "grad_norm": 0.9269915223121643, + "learning_rate": 1.997587673732289e-06, + "loss": 0.5804, + "step": 4131 + }, + { + "epoch": 3.42076200993926, + "grad_norm": 0.9441049695014954, + "learning_rate": 1.9965104683645957e-06, + "loss": 0.5547, + "step": 4132 + }, + { + "epoch": 3.4215902816123687, + "grad_norm": 0.9141139984130859, + "learning_rate": 1.9954333604174325e-06, + "loss": 0.5852, + "step": 4133 + }, + { + "epoch": 3.4224185532854774, + "grad_norm": 0.9308720827102661, + "learning_rate": 1.9943563500992115e-06, + "loss": 0.5446, + "step": 4134 + }, + { + "epoch": 3.4232468249585866, + "grad_norm": 0.9180085062980652, + "learning_rate": 1.9932794376183233e-06, + "loss": 0.5837, + "step": 4135 + }, + { + "epoch": 3.4240750966316953, + "grad_norm": 0.9463955163955688, + "learning_rate": 1.99220262318314e-06, + "loss": 0.5622, + "step": 4136 + }, + { + "epoch": 3.424903368304804, + "grad_norm": 0.9471439719200134, + "learning_rate": 1.991125907002015e-06, + "loss": 0.5682, + "step": 4137 + }, + { + "epoch": 3.4257316399779127, + "grad_norm": 0.9394784569740295, + "learning_rate": 1.990049289283284e-06, + "loss": 0.5702, + "step": 4138 + }, + { + "epoch": 3.4265599116510215, + "grad_norm": 1.005246639251709, + "learning_rate": 1.9889727702352618e-06, + "loss": 0.5772, + "step": 4139 + }, + { + "epoch": 3.42738818332413, + "grad_norm": 0.9534215927124023, + "learning_rate": 1.9878963500662453e-06, + "loss": 0.5626, + "step": 4140 + }, + { + "epoch": 3.428216454997239, + "grad_norm": 1.010836124420166, + "learning_rate": 1.98682002898451e-06, + "loss": 0.5748, + "step": 4141 + }, + { + "epoch": 3.429044726670348, + "grad_norm": 0.9219326376914978, + "learning_rate": 1.9857438071983163e-06, + "loss": 0.5893, + "step": 4142 + }, + { + "epoch": 3.429872998343457, + "grad_norm": 0.9372469186782837, + "learning_rate": 1.9846676849159017e-06, + "loss": 0.5555, + "step": 4143 + }, + { + "epoch": 3.4307012700165656, + "grad_norm": 0.9133549928665161, + "learning_rate": 1.9835916623454865e-06, + "loss": 0.5536, + "step": 4144 + }, + { + "epoch": 3.4315295416896743, + "grad_norm": 0.9400215148925781, + "learning_rate": 1.98251573969527e-06, + "loss": 0.5489, + "step": 4145 + }, + { + "epoch": 3.432357813362783, + "grad_norm": 0.9408988356590271, + "learning_rate": 1.9814399171734347e-06, + "loss": 0.5533, + "step": 4146 + }, + { + "epoch": 3.4331860850358917, + "grad_norm": 0.9352306723594666, + "learning_rate": 1.9803641949881415e-06, + "loss": 0.5645, + "step": 4147 + }, + { + "epoch": 3.4340143567090005, + "grad_norm": 0.9516900181770325, + "learning_rate": 1.979288573347532e-06, + "loss": 0.5764, + "step": 4148 + }, + { + "epoch": 3.434842628382109, + "grad_norm": 0.945179283618927, + "learning_rate": 1.97821305245973e-06, + "loss": 0.5792, + "step": 4149 + }, + { + "epoch": 3.435670900055218, + "grad_norm": 0.9419721961021423, + "learning_rate": 1.9771376325328377e-06, + "loss": 0.5598, + "step": 4150 + }, + { + "epoch": 3.436499171728327, + "grad_norm": 0.9680830836296082, + "learning_rate": 1.9760623137749394e-06, + "loss": 0.5675, + "step": 4151 + }, + { + "epoch": 3.437327443401436, + "grad_norm": 0.9370323419570923, + "learning_rate": 1.974987096394101e-06, + "loss": 0.5657, + "step": 4152 + }, + { + "epoch": 3.4381557150745445, + "grad_norm": 0.9793617725372314, + "learning_rate": 1.973911980598364e-06, + "loss": 0.5692, + "step": 4153 + }, + { + "epoch": 3.4389839867476533, + "grad_norm": 0.9549874067306519, + "learning_rate": 1.9728369665957547e-06, + "loss": 0.5798, + "step": 4154 + }, + { + "epoch": 3.439812258420762, + "grad_norm": 0.9228386878967285, + "learning_rate": 1.9717620545942784e-06, + "loss": 0.5759, + "step": 4155 + }, + { + "epoch": 3.4406405300938707, + "grad_norm": 0.9429793357849121, + "learning_rate": 1.97068724480192e-06, + "loss": 0.5618, + "step": 4156 + }, + { + "epoch": 3.4414688017669794, + "grad_norm": 0.9321681261062622, + "learning_rate": 1.969612537426646e-06, + "loss": 0.5602, + "step": 4157 + }, + { + "epoch": 3.442297073440088, + "grad_norm": 0.9411936402320862, + "learning_rate": 1.9685379326764014e-06, + "loss": 0.5787, + "step": 4158 + }, + { + "epoch": 3.443125345113197, + "grad_norm": 0.9223849177360535, + "learning_rate": 1.967463430759112e-06, + "loss": 0.5565, + "step": 4159 + }, + { + "epoch": 3.443953616786306, + "grad_norm": 0.9482093453407288, + "learning_rate": 1.966389031882686e-06, + "loss": 0.5729, + "step": 4160 + }, + { + "epoch": 3.444781888459415, + "grad_norm": 0.9146328568458557, + "learning_rate": 1.9653147362550075e-06, + "loss": 0.5661, + "step": 4161 + }, + { + "epoch": 3.4456101601325235, + "grad_norm": 0.9412453770637512, + "learning_rate": 1.9642405440839425e-06, + "loss": 0.5622, + "step": 4162 + }, + { + "epoch": 3.4464384318056323, + "grad_norm": 0.9408630132675171, + "learning_rate": 1.96316645557734e-06, + "loss": 0.5542, + "step": 4163 + }, + { + "epoch": 3.447266703478741, + "grad_norm": 0.9337356090545654, + "learning_rate": 1.962092470943022e-06, + "loss": 0.5634, + "step": 4164 + }, + { + "epoch": 3.4480949751518497, + "grad_norm": 0.9458538293838501, + "learning_rate": 1.9610185903887973e-06, + "loss": 0.5796, + "step": 4165 + }, + { + "epoch": 3.4489232468249584, + "grad_norm": 0.9226371049880981, + "learning_rate": 1.9599448141224507e-06, + "loss": 0.5596, + "step": 4166 + }, + { + "epoch": 3.4497515184980676, + "grad_norm": 0.947045087814331, + "learning_rate": 1.958871142351748e-06, + "loss": 0.554, + "step": 4167 + }, + { + "epoch": 3.4505797901711763, + "grad_norm": 0.9368892908096313, + "learning_rate": 1.9577975752844357e-06, + "loss": 0.5708, + "step": 4168 + }, + { + "epoch": 3.451408061844285, + "grad_norm": 0.9360406398773193, + "learning_rate": 1.9567241131282378e-06, + "loss": 0.5664, + "step": 4169 + }, + { + "epoch": 3.452236333517394, + "grad_norm": 0.9660393595695496, + "learning_rate": 1.9556507560908594e-06, + "loss": 0.5597, + "step": 4170 + }, + { + "epoch": 3.4530646051905025, + "grad_norm": 0.9472611546516418, + "learning_rate": 1.9545775043799858e-06, + "loss": 0.5704, + "step": 4171 + }, + { + "epoch": 3.4538928768636112, + "grad_norm": 0.941719651222229, + "learning_rate": 1.953504358203281e-06, + "loss": 0.5713, + "step": 4172 + }, + { + "epoch": 3.45472114853672, + "grad_norm": 0.9945985674858093, + "learning_rate": 1.9524313177683873e-06, + "loss": 0.563, + "step": 4173 + }, + { + "epoch": 3.4555494202098287, + "grad_norm": 0.9355506300926208, + "learning_rate": 1.9513583832829314e-06, + "loss": 0.5677, + "step": 4174 + }, + { + "epoch": 3.4563776918829374, + "grad_norm": 0.9413464665412903, + "learning_rate": 1.950285554954512e-06, + "loss": 0.5416, + "step": 4175 + }, + { + "epoch": 3.4572059635560466, + "grad_norm": 0.9668782353401184, + "learning_rate": 1.949212832990714e-06, + "loss": 0.5832, + "step": 4176 + }, + { + "epoch": 3.4580342352291553, + "grad_norm": 0.9598191976547241, + "learning_rate": 1.948140217599099e-06, + "loss": 0.597, + "step": 4177 + }, + { + "epoch": 3.458862506902264, + "grad_norm": 0.9517234563827515, + "learning_rate": 1.9470677089872064e-06, + "loss": 0.5812, + "step": 4178 + }, + { + "epoch": 3.4596907785753728, + "grad_norm": 0.9452085494995117, + "learning_rate": 1.9459953073625586e-06, + "loss": 0.5593, + "step": 4179 + }, + { + "epoch": 3.4605190502484815, + "grad_norm": 0.9393478035926819, + "learning_rate": 1.9449230129326543e-06, + "loss": 0.5828, + "step": 4180 + }, + { + "epoch": 3.4613473219215902, + "grad_norm": 0.9487680792808533, + "learning_rate": 1.9438508259049716e-06, + "loss": 0.5884, + "step": 4181 + }, + { + "epoch": 3.462175593594699, + "grad_norm": 0.9236283898353577, + "learning_rate": 1.942778746486971e-06, + "loss": 0.5534, + "step": 4182 + }, + { + "epoch": 3.4630038652678077, + "grad_norm": 0.9297188520431519, + "learning_rate": 1.941706774886088e-06, + "loss": 0.5702, + "step": 4183 + }, + { + "epoch": 3.4638321369409164, + "grad_norm": 0.9298705458641052, + "learning_rate": 1.94063491130974e-06, + "loss": 0.5645, + "step": 4184 + }, + { + "epoch": 3.4646604086140256, + "grad_norm": 0.93392014503479, + "learning_rate": 1.9395631559653223e-06, + "loss": 0.5805, + "step": 4185 + }, + { + "epoch": 3.4654886802871343, + "grad_norm": 0.9221041798591614, + "learning_rate": 1.93849150906021e-06, + "loss": 0.5746, + "step": 4186 + }, + { + "epoch": 3.466316951960243, + "grad_norm": 0.9339867830276489, + "learning_rate": 1.937419970801756e-06, + "loss": 0.5557, + "step": 4187 + }, + { + "epoch": 3.4671452236333518, + "grad_norm": 0.9240173697471619, + "learning_rate": 1.9363485413972934e-06, + "loss": 0.5616, + "step": 4188 + }, + { + "epoch": 3.4679734953064605, + "grad_norm": 0.9132168889045715, + "learning_rate": 1.9352772210541332e-06, + "loss": 0.5665, + "step": 4189 + }, + { + "epoch": 3.468801766979569, + "grad_norm": 0.9252684116363525, + "learning_rate": 1.934206009979567e-06, + "loss": 0.5534, + "step": 4190 + }, + { + "epoch": 3.469630038652678, + "grad_norm": 0.940039873123169, + "learning_rate": 1.933134908380864e-06, + "loss": 0.5686, + "step": 4191 + }, + { + "epoch": 3.4704583103257867, + "grad_norm": 0.9629718065261841, + "learning_rate": 1.9320639164652706e-06, + "loss": 0.6051, + "step": 4192 + }, + { + "epoch": 3.4712865819988954, + "grad_norm": 0.9440659880638123, + "learning_rate": 1.9309930344400164e-06, + "loss": 0.5612, + "step": 4193 + }, + { + "epoch": 3.4721148536720046, + "grad_norm": 0.9437550902366638, + "learning_rate": 1.9299222625123054e-06, + "loss": 0.5848, + "step": 4194 + }, + { + "epoch": 3.4729431253451133, + "grad_norm": 1.0566316843032837, + "learning_rate": 1.928851600889322e-06, + "loss": 0.5655, + "step": 4195 + }, + { + "epoch": 3.473771397018222, + "grad_norm": 0.9688880443572998, + "learning_rate": 1.92778104977823e-06, + "loss": 0.5792, + "step": 4196 + }, + { + "epoch": 3.4745996686913307, + "grad_norm": 0.9409403204917908, + "learning_rate": 1.9267106093861716e-06, + "loss": 0.5659, + "step": 4197 + }, + { + "epoch": 3.4754279403644395, + "grad_norm": 0.948984682559967, + "learning_rate": 1.9256402799202657e-06, + "loss": 0.5565, + "step": 4198 + }, + { + "epoch": 3.476256212037548, + "grad_norm": 0.9317809343338013, + "learning_rate": 1.924570061587611e-06, + "loss": 0.5739, + "step": 4199 + }, + { + "epoch": 3.477084483710657, + "grad_norm": 0.9296276569366455, + "learning_rate": 1.923499954595285e-06, + "loss": 0.5599, + "step": 4200 + }, + { + "epoch": 3.477912755383766, + "grad_norm": 0.9425923228263855, + "learning_rate": 1.922429959150344e-06, + "loss": 0.5686, + "step": 4201 + }, + { + "epoch": 3.478741027056875, + "grad_norm": 0.935650110244751, + "learning_rate": 1.9213600754598225e-06, + "loss": 0.5637, + "step": 4202 + }, + { + "epoch": 3.4795692987299836, + "grad_norm": 0.9613479375839233, + "learning_rate": 1.9202903037307312e-06, + "loss": 0.5766, + "step": 4203 + }, + { + "epoch": 3.4803975704030923, + "grad_norm": 0.9408090710639954, + "learning_rate": 1.9192206441700636e-06, + "loss": 0.5751, + "step": 4204 + }, + { + "epoch": 3.481225842076201, + "grad_norm": 0.9520177245140076, + "learning_rate": 1.9181510969847865e-06, + "loss": 0.5736, + "step": 4205 + }, + { + "epoch": 3.4820541137493097, + "grad_norm": 0.963941752910614, + "learning_rate": 1.9170816623818484e-06, + "loss": 0.5662, + "step": 4206 + }, + { + "epoch": 3.4828823854224185, + "grad_norm": 0.950707733631134, + "learning_rate": 1.9160123405681743e-06, + "loss": 0.576, + "step": 4207 + }, + { + "epoch": 3.483710657095527, + "grad_norm": 0.9403618574142456, + "learning_rate": 1.9149431317506693e-06, + "loss": 0.6119, + "step": 4208 + }, + { + "epoch": 3.484538928768636, + "grad_norm": 0.9552726745605469, + "learning_rate": 1.913874036136214e-06, + "loss": 0.5616, + "step": 4209 + }, + { + "epoch": 3.485367200441745, + "grad_norm": 0.9421485662460327, + "learning_rate": 1.912805053931669e-06, + "loss": 0.5708, + "step": 4210 + }, + { + "epoch": 3.486195472114854, + "grad_norm": 0.9480198621749878, + "learning_rate": 1.911736185343872e-06, + "loss": 0.5814, + "step": 4211 + }, + { + "epoch": 3.4870237437879625, + "grad_norm": 0.911723792552948, + "learning_rate": 1.91066743057964e-06, + "loss": 0.57, + "step": 4212 + }, + { + "epoch": 3.4878520154610713, + "grad_norm": 0.9331412315368652, + "learning_rate": 1.909598789845766e-06, + "loss": 0.5624, + "step": 4213 + }, + { + "epoch": 3.48868028713418, + "grad_norm": 1.0301762819290161, + "learning_rate": 1.9085302633490227e-06, + "loss": 0.5433, + "step": 4214 + }, + { + "epoch": 3.4895085588072887, + "grad_norm": 0.9518493413925171, + "learning_rate": 1.9074618512961597e-06, + "loss": 0.5702, + "step": 4215 + }, + { + "epoch": 3.4903368304803974, + "grad_norm": 0.9399821162223816, + "learning_rate": 1.9063935538939049e-06, + "loss": 0.5584, + "step": 4216 + }, + { + "epoch": 3.491165102153506, + "grad_norm": 0.9549731016159058, + "learning_rate": 1.9053253713489643e-06, + "loss": 0.5664, + "step": 4217 + }, + { + "epoch": 3.491993373826615, + "grad_norm": 0.9505523443222046, + "learning_rate": 1.9042573038680198e-06, + "loss": 0.5763, + "step": 4218 + }, + { + "epoch": 3.492821645499724, + "grad_norm": 0.9283603429794312, + "learning_rate": 1.9031893516577343e-06, + "loss": 0.5902, + "step": 4219 + }, + { + "epoch": 3.493649917172833, + "grad_norm": 0.9389188885688782, + "learning_rate": 1.9021215149247468e-06, + "loss": 0.5535, + "step": 4220 + }, + { + "epoch": 3.4944781888459415, + "grad_norm": 0.9431526064872742, + "learning_rate": 1.9010537938756719e-06, + "loss": 0.5698, + "step": 4221 + }, + { + "epoch": 3.4953064605190503, + "grad_norm": 0.9633996486663818, + "learning_rate": 1.8999861887171045e-06, + "loss": 0.573, + "step": 4222 + }, + { + "epoch": 3.496134732192159, + "grad_norm": 0.9711135625839233, + "learning_rate": 1.8989186996556161e-06, + "loss": 0.5603, + "step": 4223 + }, + { + "epoch": 3.4969630038652677, + "grad_norm": 0.9533023238182068, + "learning_rate": 1.8978513268977564e-06, + "loss": 0.5543, + "step": 4224 + }, + { + "epoch": 3.4977912755383764, + "grad_norm": 0.9520569443702698, + "learning_rate": 1.8967840706500519e-06, + "loss": 0.5859, + "step": 4225 + }, + { + "epoch": 3.4986195472114856, + "grad_norm": 0.952743411064148, + "learning_rate": 1.895716931119006e-06, + "loss": 0.5687, + "step": 4226 + }, + { + "epoch": 3.4994478188845943, + "grad_norm": 0.9404147267341614, + "learning_rate": 1.8946499085111014e-06, + "loss": 0.5669, + "step": 4227 + }, + { + "epoch": 3.500276090557703, + "grad_norm": 0.962302029132843, + "learning_rate": 1.8935830030327967e-06, + "loss": 0.5721, + "step": 4228 + }, + { + "epoch": 3.501104362230812, + "grad_norm": 0.9707161784172058, + "learning_rate": 1.8925162148905275e-06, + "loss": 0.5858, + "step": 4229 + }, + { + "epoch": 3.5019326339039205, + "grad_norm": 0.9720497727394104, + "learning_rate": 1.891449544290708e-06, + "loss": 0.5892, + "step": 4230 + }, + { + "epoch": 3.5027609055770292, + "grad_norm": 0.9394986629486084, + "learning_rate": 1.8903829914397295e-06, + "loss": 0.547, + "step": 4231 + }, + { + "epoch": 3.503589177250138, + "grad_norm": 0.9768742322921753, + "learning_rate": 1.8893165565439588e-06, + "loss": 0.5647, + "step": 4232 + }, + { + "epoch": 3.5044174489232467, + "grad_norm": 0.9559404253959656, + "learning_rate": 1.888250239809742e-06, + "loss": 0.556, + "step": 4233 + }, + { + "epoch": 3.5052457205963554, + "grad_norm": 0.9628231525421143, + "learning_rate": 1.8871840414434e-06, + "loss": 0.5633, + "step": 4234 + }, + { + "epoch": 3.5060739922694646, + "grad_norm": 0.9246355891227722, + "learning_rate": 1.8861179616512346e-06, + "loss": 0.5678, + "step": 4235 + }, + { + "epoch": 3.5069022639425733, + "grad_norm": 0.9590763449668884, + "learning_rate": 1.8850520006395206e-06, + "loss": 0.5612, + "step": 4236 + }, + { + "epoch": 3.507730535615682, + "grad_norm": 0.9386741518974304, + "learning_rate": 1.8839861586145117e-06, + "loss": 0.5737, + "step": 4237 + }, + { + "epoch": 3.5085588072887908, + "grad_norm": 0.9735057353973389, + "learning_rate": 1.8829204357824394e-06, + "loss": 0.5725, + "step": 4238 + }, + { + "epoch": 3.5093870789618995, + "grad_norm": 0.9464735984802246, + "learning_rate": 1.8818548323495101e-06, + "loss": 0.561, + "step": 4239 + }, + { + "epoch": 3.5102153506350082, + "grad_norm": 0.9822002053260803, + "learning_rate": 1.880789348521908e-06, + "loss": 0.562, + "step": 4240 + }, + { + "epoch": 3.511043622308117, + "grad_norm": 0.9448258876800537, + "learning_rate": 1.879723984505795e-06, + "loss": 0.5526, + "step": 4241 + }, + { + "epoch": 3.5118718939812257, + "grad_norm": 0.9565535187721252, + "learning_rate": 1.87865874050731e-06, + "loss": 0.5631, + "step": 4242 + }, + { + "epoch": 3.5127001656543344, + "grad_norm": 0.9824585318565369, + "learning_rate": 1.877593616732566e-06, + "loss": 0.5924, + "step": 4243 + }, + { + "epoch": 3.5135284373274436, + "grad_norm": 0.9289377927780151, + "learning_rate": 1.8765286133876549e-06, + "loss": 0.5667, + "step": 4244 + }, + { + "epoch": 3.5143567090005523, + "grad_norm": 0.9740702509880066, + "learning_rate": 1.8754637306786447e-06, + "loss": 0.567, + "step": 4245 + }, + { + "epoch": 3.515184980673661, + "grad_norm": 0.9304373264312744, + "learning_rate": 1.8743989688115813e-06, + "loss": 0.5973, + "step": 4246 + }, + { + "epoch": 3.5160132523467698, + "grad_norm": 0.9417708516120911, + "learning_rate": 1.8733343279924857e-06, + "loss": 0.5663, + "step": 4247 + }, + { + "epoch": 3.5168415240198785, + "grad_norm": 0.9550634622573853, + "learning_rate": 1.8722698084273555e-06, + "loss": 0.5829, + "step": 4248 + }, + { + "epoch": 3.517669795692987, + "grad_norm": 0.9428004622459412, + "learning_rate": 1.8712054103221661e-06, + "loss": 0.5595, + "step": 4249 + }, + { + "epoch": 3.518498067366096, + "grad_norm": 0.9728983640670776, + "learning_rate": 1.8701411338828685e-06, + "loss": 0.5611, + "step": 4250 + }, + { + "epoch": 3.519326339039205, + "grad_norm": 0.9784600138664246, + "learning_rate": 1.869076979315389e-06, + "loss": 0.5699, + "step": 4251 + }, + { + "epoch": 3.5201546107123134, + "grad_norm": 0.9751117825508118, + "learning_rate": 1.8680129468256336e-06, + "loss": 0.5622, + "step": 4252 + }, + { + "epoch": 3.5209828823854226, + "grad_norm": 0.9501684904098511, + "learning_rate": 1.866949036619482e-06, + "loss": 0.5594, + "step": 4253 + }, + { + "epoch": 3.5218111540585313, + "grad_norm": 0.9380241632461548, + "learning_rate": 1.86588524890279e-06, + "loss": 0.5759, + "step": 4254 + }, + { + "epoch": 3.52263942573164, + "grad_norm": 0.9474608898162842, + "learning_rate": 1.8648215838813915e-06, + "loss": 0.592, + "step": 4255 + }, + { + "epoch": 3.5234676974047487, + "grad_norm": 0.9680253863334656, + "learning_rate": 1.8637580417610949e-06, + "loss": 0.5369, + "step": 4256 + }, + { + "epoch": 3.5242959690778575, + "grad_norm": 0.9809057712554932, + "learning_rate": 1.8626946227476867e-06, + "loss": 0.5678, + "step": 4257 + }, + { + "epoch": 3.525124240750966, + "grad_norm": 0.9424263834953308, + "learning_rate": 1.8616313270469277e-06, + "loss": 0.5735, + "step": 4258 + }, + { + "epoch": 3.525952512424075, + "grad_norm": 0.9373759627342224, + "learning_rate": 1.860568154864556e-06, + "loss": 0.5572, + "step": 4259 + }, + { + "epoch": 3.526780784097184, + "grad_norm": 0.9195491075515747, + "learning_rate": 1.8595051064062864e-06, + "loss": 0.5552, + "step": 4260 + }, + { + "epoch": 3.5276090557702924, + "grad_norm": 0.9787479043006897, + "learning_rate": 1.8584421818778076e-06, + "loss": 0.5486, + "step": 4261 + }, + { + "epoch": 3.5284373274434015, + "grad_norm": 0.9167375564575195, + "learning_rate": 1.8573793814847856e-06, + "loss": 0.561, + "step": 4262 + }, + { + "epoch": 3.5292655991165103, + "grad_norm": 0.9589730501174927, + "learning_rate": 1.8563167054328635e-06, + "loss": 0.5792, + "step": 4263 + }, + { + "epoch": 3.530093870789619, + "grad_norm": 0.9304484724998474, + "learning_rate": 1.8552541539276587e-06, + "loss": 0.5628, + "step": 4264 + }, + { + "epoch": 3.5309221424627277, + "grad_norm": 0.9468231201171875, + "learning_rate": 1.8541917271747653e-06, + "loss": 0.5705, + "step": 4265 + }, + { + "epoch": 3.5317504141358365, + "grad_norm": 0.9367941617965698, + "learning_rate": 1.8531294253797522e-06, + "loss": 0.5769, + "step": 4266 + }, + { + "epoch": 3.532578685808945, + "grad_norm": 0.9852622151374817, + "learning_rate": 1.8520672487481645e-06, + "loss": 0.572, + "step": 4267 + }, + { + "epoch": 3.533406957482054, + "grad_norm": 0.9441319108009338, + "learning_rate": 1.8510051974855244e-06, + "loss": 0.5587, + "step": 4268 + }, + { + "epoch": 3.534235229155163, + "grad_norm": 0.9801215529441833, + "learning_rate": 1.8499432717973293e-06, + "loss": 0.6019, + "step": 4269 + }, + { + "epoch": 3.535063500828272, + "grad_norm": 0.9506904482841492, + "learning_rate": 1.8488814718890505e-06, + "loss": 0.569, + "step": 4270 + }, + { + "epoch": 3.5358917725013805, + "grad_norm": 0.9413098096847534, + "learning_rate": 1.8478197979661377e-06, + "loss": 0.5682, + "step": 4271 + }, + { + "epoch": 3.5367200441744893, + "grad_norm": 0.9410218596458435, + "learning_rate": 1.8467582502340149e-06, + "loss": 0.5843, + "step": 4272 + }, + { + "epoch": 3.537548315847598, + "grad_norm": 0.9326966404914856, + "learning_rate": 1.845696828898081e-06, + "loss": 0.5853, + "step": 4273 + }, + { + "epoch": 3.5383765875207067, + "grad_norm": 0.9390862584114075, + "learning_rate": 1.8446355341637107e-06, + "loss": 0.5767, + "step": 4274 + }, + { + "epoch": 3.5392048591938154, + "grad_norm": 0.9381791949272156, + "learning_rate": 1.843574366236256e-06, + "loss": 0.5534, + "step": 4275 + }, + { + "epoch": 3.5400331308669246, + "grad_norm": 1.0620906352996826, + "learning_rate": 1.8425133253210425e-06, + "loss": 0.5663, + "step": 4276 + }, + { + "epoch": 3.540861402540033, + "grad_norm": 0.945441484451294, + "learning_rate": 1.8414524116233715e-06, + "loss": 0.5496, + "step": 4277 + }, + { + "epoch": 3.541689674213142, + "grad_norm": 0.9375214576721191, + "learning_rate": 1.8403916253485193e-06, + "loss": 0.5556, + "step": 4278 + }, + { + "epoch": 3.542517945886251, + "grad_norm": 0.9695723056793213, + "learning_rate": 1.8393309667017388e-06, + "loss": 0.5457, + "step": 4279 + }, + { + "epoch": 3.5433462175593595, + "grad_norm": 0.9985876679420471, + "learning_rate": 1.8382704358882583e-06, + "loss": 0.5749, + "step": 4280 + }, + { + "epoch": 3.5441744892324683, + "grad_norm": 0.9784291386604309, + "learning_rate": 1.8372100331132797e-06, + "loss": 0.5671, + "step": 4281 + }, + { + "epoch": 3.545002760905577, + "grad_norm": 0.9699963331222534, + "learning_rate": 1.8361497585819804e-06, + "loss": 0.5742, + "step": 4282 + }, + { + "epoch": 3.5458310325786857, + "grad_norm": 0.951996386051178, + "learning_rate": 1.8350896124995155e-06, + "loss": 0.5729, + "step": 4283 + }, + { + "epoch": 3.5466593042517944, + "grad_norm": 0.9519326686859131, + "learning_rate": 1.8340295950710119e-06, + "loss": 0.5746, + "step": 4284 + }, + { + "epoch": 3.5474875759249036, + "grad_norm": 0.9227757453918457, + "learning_rate": 1.8329697065015734e-06, + "loss": 0.5441, + "step": 4285 + }, + { + "epoch": 3.548315847598012, + "grad_norm": 0.9924660325050354, + "learning_rate": 1.8319099469962787e-06, + "loss": 0.5695, + "step": 4286 + }, + { + "epoch": 3.549144119271121, + "grad_norm": 0.9582194089889526, + "learning_rate": 1.8308503167601827e-06, + "loss": 0.5757, + "step": 4287 + }, + { + "epoch": 3.54997239094423, + "grad_norm": 0.924054741859436, + "learning_rate": 1.8297908159983114e-06, + "loss": 0.5739, + "step": 4288 + }, + { + "epoch": 3.5508006626173385, + "grad_norm": 0.9424325227737427, + "learning_rate": 1.82873144491567e-06, + "loss": 0.6041, + "step": 4289 + }, + { + "epoch": 3.5516289342904472, + "grad_norm": 0.9287232160568237, + "learning_rate": 1.827672203717236e-06, + "loss": 0.5494, + "step": 4290 + }, + { + "epoch": 3.552457205963556, + "grad_norm": 0.94130539894104, + "learning_rate": 1.8266130926079637e-06, + "loss": 0.5642, + "step": 4291 + }, + { + "epoch": 3.5532854776366647, + "grad_norm": 0.933063805103302, + "learning_rate": 1.8255541117927805e-06, + "loss": 0.5759, + "step": 4292 + }, + { + "epoch": 3.5541137493097734, + "grad_norm": 0.9383630156517029, + "learning_rate": 1.8244952614765894e-06, + "loss": 0.5584, + "step": 4293 + }, + { + "epoch": 3.5549420209828826, + "grad_norm": 0.9373190999031067, + "learning_rate": 1.8234365418642685e-06, + "loss": 0.5597, + "step": 4294 + }, + { + "epoch": 3.5557702926559913, + "grad_norm": 0.9637420773506165, + "learning_rate": 1.82237795316067e-06, + "loss": 0.5736, + "step": 4295 + }, + { + "epoch": 3.5565985643291, + "grad_norm": 0.9421122670173645, + "learning_rate": 1.8213194955706204e-06, + "loss": 0.5424, + "step": 4296 + }, + { + "epoch": 3.5574268360022088, + "grad_norm": 0.9612517952919006, + "learning_rate": 1.8202611692989224e-06, + "loss": 0.5704, + "step": 4297 + }, + { + "epoch": 3.5582551076753175, + "grad_norm": 0.9415401220321655, + "learning_rate": 1.8192029745503515e-06, + "loss": 0.5774, + "step": 4298 + }, + { + "epoch": 3.5590833793484262, + "grad_norm": 0.9469128251075745, + "learning_rate": 1.8181449115296596e-06, + "loss": 0.5721, + "step": 4299 + }, + { + "epoch": 3.559911651021535, + "grad_norm": 0.9917946457862854, + "learning_rate": 1.8170869804415703e-06, + "loss": 0.5767, + "step": 4300 + }, + { + "epoch": 3.5607399226946437, + "grad_norm": 0.9557772874832153, + "learning_rate": 1.8160291814907835e-06, + "loss": 0.5677, + "step": 4301 + }, + { + "epoch": 3.5615681943677524, + "grad_norm": 0.9430398344993591, + "learning_rate": 1.8149715148819752e-06, + "loss": 0.5679, + "step": 4302 + }, + { + "epoch": 3.5623964660408616, + "grad_norm": 0.9426287412643433, + "learning_rate": 1.8139139808197926e-06, + "loss": 0.563, + "step": 4303 + }, + { + "epoch": 3.5632247377139703, + "grad_norm": 0.9437983632087708, + "learning_rate": 1.8128565795088588e-06, + "loss": 0.5703, + "step": 4304 + }, + { + "epoch": 3.564053009387079, + "grad_norm": 0.9386832118034363, + "learning_rate": 1.8117993111537717e-06, + "loss": 0.5842, + "step": 4305 + }, + { + "epoch": 3.5648812810601878, + "grad_norm": 0.954119086265564, + "learning_rate": 1.8107421759591031e-06, + "loss": 0.5515, + "step": 4306 + }, + { + "epoch": 3.5657095527332965, + "grad_norm": 0.945758581161499, + "learning_rate": 1.8096851741293974e-06, + "loss": 0.5455, + "step": 4307 + }, + { + "epoch": 3.566537824406405, + "grad_norm": 0.9291911721229553, + "learning_rate": 1.8086283058691759e-06, + "loss": 0.5843, + "step": 4308 + }, + { + "epoch": 3.567366096079514, + "grad_norm": 0.9620952606201172, + "learning_rate": 1.8075715713829323e-06, + "loss": 0.568, + "step": 4309 + }, + { + "epoch": 3.568194367752623, + "grad_norm": 0.9722322821617126, + "learning_rate": 1.8065149708751357e-06, + "loss": 0.5816, + "step": 4310 + }, + { + "epoch": 3.5690226394257314, + "grad_norm": 0.9354492425918579, + "learning_rate": 1.805458504550227e-06, + "loss": 0.5856, + "step": 4311 + }, + { + "epoch": 3.5698509110988406, + "grad_norm": 0.956453263759613, + "learning_rate": 1.8044021726126226e-06, + "loss": 0.5582, + "step": 4312 + }, + { + "epoch": 3.5706791827719493, + "grad_norm": 0.926895797252655, + "learning_rate": 1.8033459752667143e-06, + "loss": 0.5837, + "step": 4313 + }, + { + "epoch": 3.571507454445058, + "grad_norm": 0.9227427840232849, + "learning_rate": 1.8022899127168655e-06, + "loss": 0.5567, + "step": 4314 + }, + { + "epoch": 3.5723357261181667, + "grad_norm": 0.9473888278007507, + "learning_rate": 1.801233985167414e-06, + "loss": 0.5672, + "step": 4315 + }, + { + "epoch": 3.5731639977912755, + "grad_norm": 0.9397003054618835, + "learning_rate": 1.8001781928226734e-06, + "loss": 0.5663, + "step": 4316 + }, + { + "epoch": 3.573992269464384, + "grad_norm": 0.9553276300430298, + "learning_rate": 1.7991225358869286e-06, + "loss": 0.5716, + "step": 4317 + }, + { + "epoch": 3.574820541137493, + "grad_norm": 0.9149162769317627, + "learning_rate": 1.798067014564439e-06, + "loss": 0.5409, + "step": 4318 + }, + { + "epoch": 3.575648812810602, + "grad_norm": 1.0992645025253296, + "learning_rate": 1.7970116290594391e-06, + "loss": 0.5542, + "step": 4319 + }, + { + "epoch": 3.576477084483711, + "grad_norm": 0.9605867862701416, + "learning_rate": 1.7959563795761364e-06, + "loss": 0.5426, + "step": 4320 + }, + { + "epoch": 3.5773053561568195, + "grad_norm": 0.944625198841095, + "learning_rate": 1.7949012663187115e-06, + "loss": 0.5695, + "step": 4321 + }, + { + "epoch": 3.5781336278299283, + "grad_norm": 0.9884684085845947, + "learning_rate": 1.793846289491318e-06, + "loss": 0.5897, + "step": 4322 + }, + { + "epoch": 3.578961899503037, + "grad_norm": 0.9456211924552917, + "learning_rate": 1.7927914492980845e-06, + "loss": 0.5969, + "step": 4323 + }, + { + "epoch": 3.5797901711761457, + "grad_norm": 0.9710149168968201, + "learning_rate": 1.7917367459431135e-06, + "loss": 0.5797, + "step": 4324 + }, + { + "epoch": 3.5806184428492545, + "grad_norm": 0.9509037733078003, + "learning_rate": 1.7906821796304796e-06, + "loss": 0.5655, + "step": 4325 + }, + { + "epoch": 3.581446714522363, + "grad_norm": 0.9468769431114197, + "learning_rate": 1.7896277505642316e-06, + "loss": 0.5519, + "step": 4326 + }, + { + "epoch": 3.582274986195472, + "grad_norm": 0.9285255670547485, + "learning_rate": 1.7885734589483923e-06, + "loss": 0.5765, + "step": 4327 + }, + { + "epoch": 3.583103257868581, + "grad_norm": 0.9228923320770264, + "learning_rate": 1.7875193049869566e-06, + "loss": 0.5742, + "step": 4328 + }, + { + "epoch": 3.58393152954169, + "grad_norm": 0.9343339204788208, + "learning_rate": 1.7864652888838935e-06, + "loss": 0.5601, + "step": 4329 + }, + { + "epoch": 3.5847598012147985, + "grad_norm": 0.937486469745636, + "learning_rate": 1.7854114108431459e-06, + "loss": 0.5392, + "step": 4330 + }, + { + "epoch": 3.5855880728879073, + "grad_norm": 0.9428637027740479, + "learning_rate": 1.7843576710686288e-06, + "loss": 0.5502, + "step": 4331 + }, + { + "epoch": 3.586416344561016, + "grad_norm": 0.9170352220535278, + "learning_rate": 1.7833040697642312e-06, + "loss": 0.5707, + "step": 4332 + }, + { + "epoch": 3.5872446162341247, + "grad_norm": 0.9696611166000366, + "learning_rate": 1.7822506071338163e-06, + "loss": 0.5714, + "step": 4333 + }, + { + "epoch": 3.5880728879072334, + "grad_norm": 0.912272572517395, + "learning_rate": 1.7811972833812167e-06, + "loss": 0.5633, + "step": 4334 + }, + { + "epoch": 3.5889011595803426, + "grad_norm": 0.9608253836631775, + "learning_rate": 1.7801440987102424e-06, + "loss": 0.572, + "step": 4335 + }, + { + "epoch": 3.589729431253451, + "grad_norm": 0.9655832648277283, + "learning_rate": 1.7790910533246752e-06, + "loss": 0.5727, + "step": 4336 + }, + { + "epoch": 3.59055770292656, + "grad_norm": 0.9179590940475464, + "learning_rate": 1.7780381474282683e-06, + "loss": 0.5701, + "step": 4337 + }, + { + "epoch": 3.591385974599669, + "grad_norm": 0.9642294049263, + "learning_rate": 1.7769853812247501e-06, + "loss": 0.5659, + "step": 4338 + }, + { + "epoch": 3.5922142462727775, + "grad_norm": 0.9484195709228516, + "learning_rate": 1.7759327549178208e-06, + "loss": 0.5572, + "step": 4339 + }, + { + "epoch": 3.5930425179458862, + "grad_norm": 0.9366327524185181, + "learning_rate": 1.7748802687111533e-06, + "loss": 0.5596, + "step": 4340 + }, + { + "epoch": 3.593870789618995, + "grad_norm": 0.9256154894828796, + "learning_rate": 1.7738279228083946e-06, + "loss": 0.553, + "step": 4341 + }, + { + "epoch": 3.5946990612921037, + "grad_norm": 0.9243422746658325, + "learning_rate": 1.7727757174131637e-06, + "loss": 0.5703, + "step": 4342 + }, + { + "epoch": 3.5955273329652124, + "grad_norm": 0.9723832607269287, + "learning_rate": 1.7717236527290526e-06, + "loss": 0.5812, + "step": 4343 + }, + { + "epoch": 3.5963556046383216, + "grad_norm": 0.9459588527679443, + "learning_rate": 1.7706717289596263e-06, + "loss": 0.5871, + "step": 4344 + }, + { + "epoch": 3.59718387631143, + "grad_norm": 0.9725896716117859, + "learning_rate": 1.7696199463084205e-06, + "loss": 0.5594, + "step": 4345 + }, + { + "epoch": 3.598012147984539, + "grad_norm": 0.9436221122741699, + "learning_rate": 1.768568304978947e-06, + "loss": 0.5587, + "step": 4346 + }, + { + "epoch": 3.598840419657648, + "grad_norm": 0.9552313685417175, + "learning_rate": 1.7675168051746884e-06, + "loss": 0.572, + "step": 4347 + }, + { + "epoch": 3.5996686913307565, + "grad_norm": 0.9535266160964966, + "learning_rate": 1.7664654470990993e-06, + "loss": 0.5695, + "step": 4348 + }, + { + "epoch": 3.6004969630038652, + "grad_norm": 0.9626250863075256, + "learning_rate": 1.7654142309556088e-06, + "loss": 0.5695, + "step": 4349 + }, + { + "epoch": 3.601325234676974, + "grad_norm": 0.9406429529190063, + "learning_rate": 1.7643631569476166e-06, + "loss": 0.5525, + "step": 4350 + }, + { + "epoch": 3.6021535063500827, + "grad_norm": 0.946359395980835, + "learning_rate": 1.7633122252784963e-06, + "loss": 0.5626, + "step": 4351 + }, + { + "epoch": 3.6029817780231914, + "grad_norm": 0.9508920907974243, + "learning_rate": 1.7622614361515928e-06, + "loss": 0.5569, + "step": 4352 + }, + { + "epoch": 3.6038100496963006, + "grad_norm": 0.9463246464729309, + "learning_rate": 1.7612107897702245e-06, + "loss": 0.5608, + "step": 4353 + }, + { + "epoch": 3.6046383213694093, + "grad_norm": 0.9517480731010437, + "learning_rate": 1.7601602863376816e-06, + "loss": 0.5753, + "step": 4354 + }, + { + "epoch": 3.605466593042518, + "grad_norm": 0.9344833493232727, + "learning_rate": 1.759109926057227e-06, + "loss": 0.5633, + "step": 4355 + }, + { + "epoch": 3.6062948647156268, + "grad_norm": 0.9408402442932129, + "learning_rate": 1.758059709132094e-06, + "loss": 0.5829, + "step": 4356 + }, + { + "epoch": 3.6071231363887355, + "grad_norm": 0.9551196098327637, + "learning_rate": 1.7570096357654918e-06, + "loss": 0.5553, + "step": 4357 + }, + { + "epoch": 3.6079514080618442, + "grad_norm": 0.9403008222579956, + "learning_rate": 1.7559597061605987e-06, + "loss": 0.576, + "step": 4358 + }, + { + "epoch": 3.608779679734953, + "grad_norm": 0.9454553127288818, + "learning_rate": 1.7549099205205673e-06, + "loss": 0.5643, + "step": 4359 + }, + { + "epoch": 3.609607951408062, + "grad_norm": 0.9868146777153015, + "learning_rate": 1.7538602790485194e-06, + "loss": 0.591, + "step": 4360 + }, + { + "epoch": 3.6104362230811704, + "grad_norm": 0.9502206444740295, + "learning_rate": 1.752810781947553e-06, + "loss": 0.5431, + "step": 4361 + }, + { + "epoch": 3.6112644947542796, + "grad_norm": 0.9451961517333984, + "learning_rate": 1.7517614294207347e-06, + "loss": 0.5762, + "step": 4362 + }, + { + "epoch": 3.6120927664273883, + "grad_norm": 0.9424105882644653, + "learning_rate": 1.7507122216711048e-06, + "loss": 0.6107, + "step": 4363 + }, + { + "epoch": 3.612921038100497, + "grad_norm": 0.967083215713501, + "learning_rate": 1.7496631589016754e-06, + "loss": 0.5486, + "step": 4364 + }, + { + "epoch": 3.6137493097736058, + "grad_norm": 0.9303386211395264, + "learning_rate": 1.7486142413154304e-06, + "loss": 0.5345, + "step": 4365 + }, + { + "epoch": 3.6145775814467145, + "grad_norm": 0.9635606408119202, + "learning_rate": 1.7475654691153257e-06, + "loss": 0.563, + "step": 4366 + }, + { + "epoch": 3.615405853119823, + "grad_norm": 0.9364803433418274, + "learning_rate": 1.7465168425042883e-06, + "loss": 0.561, + "step": 4367 + }, + { + "epoch": 3.616234124792932, + "grad_norm": 0.9707603454589844, + "learning_rate": 1.7454683616852172e-06, + "loss": 0.585, + "step": 4368 + }, + { + "epoch": 3.617062396466041, + "grad_norm": 0.9474611282348633, + "learning_rate": 1.744420026860985e-06, + "loss": 0.5754, + "step": 4369 + }, + { + "epoch": 3.6178906681391494, + "grad_norm": 0.9372676014900208, + "learning_rate": 1.7433718382344344e-06, + "loss": 0.5624, + "step": 4370 + }, + { + "epoch": 3.6187189398122586, + "grad_norm": 0.9487974643707275, + "learning_rate": 1.7423237960083794e-06, + "loss": 0.5406, + "step": 4371 + }, + { + "epoch": 3.6195472114853673, + "grad_norm": 0.9595673084259033, + "learning_rate": 1.7412759003856073e-06, + "loss": 0.5977, + "step": 4372 + }, + { + "epoch": 3.620375483158476, + "grad_norm": 0.954731285572052, + "learning_rate": 1.7402281515688756e-06, + "loss": 0.56, + "step": 4373 + }, + { + "epoch": 3.6212037548315847, + "grad_norm": 0.9508479237556458, + "learning_rate": 1.7391805497609137e-06, + "loss": 0.5607, + "step": 4374 + }, + { + "epoch": 3.6220320265046935, + "grad_norm": 0.9357259273529053, + "learning_rate": 1.7381330951644237e-06, + "loss": 0.5545, + "step": 4375 + }, + { + "epoch": 3.622860298177802, + "grad_norm": 0.9336346983909607, + "learning_rate": 1.737085787982078e-06, + "loss": 0.5636, + "step": 4376 + }, + { + "epoch": 3.623688569850911, + "grad_norm": 0.9238201379776001, + "learning_rate": 1.73603862841652e-06, + "loss": 0.5497, + "step": 4377 + }, + { + "epoch": 3.62451684152402, + "grad_norm": 0.9505547881126404, + "learning_rate": 1.7349916166703668e-06, + "loss": 0.5872, + "step": 4378 + }, + { + "epoch": 3.625345113197129, + "grad_norm": 0.945364773273468, + "learning_rate": 1.7339447529462033e-06, + "loss": 0.539, + "step": 4379 + }, + { + "epoch": 3.6261733848702375, + "grad_norm": 0.9546788930892944, + "learning_rate": 1.7328980374465896e-06, + "loss": 0.5489, + "step": 4380 + }, + { + "epoch": 3.6270016565433463, + "grad_norm": 0.9455381035804749, + "learning_rate": 1.7318514703740546e-06, + "loss": 0.5624, + "step": 4381 + }, + { + "epoch": 3.627829928216455, + "grad_norm": 0.9461554288864136, + "learning_rate": 1.7308050519310993e-06, + "loss": 0.5651, + "step": 4382 + }, + { + "epoch": 3.6286581998895637, + "grad_norm": 0.9569746255874634, + "learning_rate": 1.7297587823201964e-06, + "loss": 0.5792, + "step": 4383 + }, + { + "epoch": 3.6294864715626725, + "grad_norm": 0.9243269562721252, + "learning_rate": 1.7287126617437889e-06, + "loss": 0.5493, + "step": 4384 + }, + { + "epoch": 3.630314743235781, + "grad_norm": 0.9244057536125183, + "learning_rate": 1.727666690404291e-06, + "loss": 0.5487, + "step": 4385 + }, + { + "epoch": 3.63114301490889, + "grad_norm": 0.9674320220947266, + "learning_rate": 1.7266208685040896e-06, + "loss": 0.5572, + "step": 4386 + }, + { + "epoch": 3.631971286581999, + "grad_norm": 0.9546018838882446, + "learning_rate": 1.7255751962455407e-06, + "loss": 0.581, + "step": 4387 + }, + { + "epoch": 3.632799558255108, + "grad_norm": 0.9394852519035339, + "learning_rate": 1.7245296738309719e-06, + "loss": 0.5742, + "step": 4388 + }, + { + "epoch": 3.6336278299282165, + "grad_norm": 0.9570075273513794, + "learning_rate": 1.723484301462684e-06, + "loss": 0.5631, + "step": 4389 + }, + { + "epoch": 3.6344561016013253, + "grad_norm": 0.9604787230491638, + "learning_rate": 1.7224390793429435e-06, + "loss": 0.5616, + "step": 4390 + }, + { + "epoch": 3.635284373274434, + "grad_norm": 0.9690744280815125, + "learning_rate": 1.7213940076739931e-06, + "loss": 0.5912, + "step": 4391 + }, + { + "epoch": 3.6361126449475427, + "grad_norm": 0.9428882598876953, + "learning_rate": 1.720349086658044e-06, + "loss": 0.5763, + "step": 4392 + }, + { + "epoch": 3.6369409166206514, + "grad_norm": 0.9487605094909668, + "learning_rate": 1.7193043164972788e-06, + "loss": 0.5652, + "step": 4393 + }, + { + "epoch": 3.6377691882937606, + "grad_norm": 0.9717780351638794, + "learning_rate": 1.7182596973938512e-06, + "loss": 0.5897, + "step": 4394 + }, + { + "epoch": 3.638597459966869, + "grad_norm": 0.9824546575546265, + "learning_rate": 1.7172152295498848e-06, + "loss": 0.5647, + "step": 4395 + }, + { + "epoch": 3.639425731639978, + "grad_norm": 0.9322198629379272, + "learning_rate": 1.716170913167474e-06, + "loss": 0.5532, + "step": 4396 + }, + { + "epoch": 3.640254003313087, + "grad_norm": 0.9403592944145203, + "learning_rate": 1.715126748448685e-06, + "loss": 0.5768, + "step": 4397 + }, + { + "epoch": 3.6410822749861955, + "grad_norm": 0.9282305836677551, + "learning_rate": 1.7140827355955541e-06, + "loss": 0.5354, + "step": 4398 + }, + { + "epoch": 3.6419105466593042, + "grad_norm": 1.0102676153182983, + "learning_rate": 1.7130388748100868e-06, + "loss": 0.5792, + "step": 4399 + }, + { + "epoch": 3.642738818332413, + "grad_norm": 0.9756748676300049, + "learning_rate": 1.711995166294263e-06, + "loss": 0.5691, + "step": 4400 + }, + { + "epoch": 3.6435670900055217, + "grad_norm": 0.9297899603843689, + "learning_rate": 1.710951610250027e-06, + "loss": 0.548, + "step": 4401 + }, + { + "epoch": 3.6443953616786304, + "grad_norm": 0.9422688484191895, + "learning_rate": 1.7099082068792999e-06, + "loss": 0.5344, + "step": 4402 + }, + { + "epoch": 3.6452236333517396, + "grad_norm": 0.9583215713500977, + "learning_rate": 1.7088649563839693e-06, + "loss": 0.5652, + "step": 4403 + }, + { + "epoch": 3.646051905024848, + "grad_norm": 0.9639995098114014, + "learning_rate": 1.7078218589658946e-06, + "loss": 0.5649, + "step": 4404 + }, + { + "epoch": 3.646880176697957, + "grad_norm": 0.9570342898368835, + "learning_rate": 1.7067789148269056e-06, + "loss": 0.5513, + "step": 4405 + }, + { + "epoch": 3.647708448371066, + "grad_norm": 0.9547348618507385, + "learning_rate": 1.7057361241688024e-06, + "loss": 0.5659, + "step": 4406 + }, + { + "epoch": 3.6485367200441745, + "grad_norm": 0.9749144911766052, + "learning_rate": 1.7046934871933551e-06, + "loss": 0.5672, + "step": 4407 + }, + { + "epoch": 3.6493649917172832, + "grad_norm": 0.9632282853126526, + "learning_rate": 1.7036510041023045e-06, + "loss": 0.5761, + "step": 4408 + }, + { + "epoch": 3.650193263390392, + "grad_norm": 0.9284268617630005, + "learning_rate": 1.7026086750973612e-06, + "loss": 0.5758, + "step": 4409 + }, + { + "epoch": 3.6510215350635007, + "grad_norm": 0.9373230934143066, + "learning_rate": 1.7015665003802053e-06, + "loss": 0.5537, + "step": 4410 + }, + { + "epoch": 3.6518498067366094, + "grad_norm": 0.9570644497871399, + "learning_rate": 1.7005244801524896e-06, + "loss": 0.5922, + "step": 4411 + }, + { + "epoch": 3.6526780784097186, + "grad_norm": 0.9559333920478821, + "learning_rate": 1.6994826146158345e-06, + "loss": 0.5854, + "step": 4412 + }, + { + "epoch": 3.6535063500828273, + "grad_norm": 0.943041205406189, + "learning_rate": 1.698440903971831e-06, + "loss": 0.5475, + "step": 4413 + }, + { + "epoch": 3.654334621755936, + "grad_norm": 0.9560774564743042, + "learning_rate": 1.6973993484220403e-06, + "loss": 0.5863, + "step": 4414 + }, + { + "epoch": 3.6551628934290448, + "grad_norm": 0.9528188109397888, + "learning_rate": 1.6963579481679932e-06, + "loss": 0.5536, + "step": 4415 + }, + { + "epoch": 3.6559911651021535, + "grad_norm": 0.9496455192565918, + "learning_rate": 1.6953167034111926e-06, + "loss": 0.5896, + "step": 4416 + }, + { + "epoch": 3.656819436775262, + "grad_norm": 0.9072014689445496, + "learning_rate": 1.6942756143531086e-06, + "loss": 0.5665, + "step": 4417 + }, + { + "epoch": 3.657647708448371, + "grad_norm": 0.9448863863945007, + "learning_rate": 1.6932346811951817e-06, + "loss": 0.5765, + "step": 4418 + }, + { + "epoch": 3.65847598012148, + "grad_norm": 0.9441279768943787, + "learning_rate": 1.692193904138824e-06, + "loss": 0.5721, + "step": 4419 + }, + { + "epoch": 3.6593042517945884, + "grad_norm": 0.9575514793395996, + "learning_rate": 1.6911532833854155e-06, + "loss": 0.5657, + "step": 4420 + }, + { + "epoch": 3.6601325234676976, + "grad_norm": 1.0165443420410156, + "learning_rate": 1.690112819136307e-06, + "loss": 0.5853, + "step": 4421 + }, + { + "epoch": 3.6609607951408063, + "grad_norm": 1.0065933465957642, + "learning_rate": 1.6890725115928176e-06, + "loss": 0.559, + "step": 4422 + }, + { + "epoch": 3.661789066813915, + "grad_norm": 0.9567911028862, + "learning_rate": 1.6880323609562388e-06, + "loss": 0.5734, + "step": 4423 + }, + { + "epoch": 3.6626173384870238, + "grad_norm": 0.9482328295707703, + "learning_rate": 1.6869923674278288e-06, + "loss": 0.5704, + "step": 4424 + }, + { + "epoch": 3.6634456101601325, + "grad_norm": 0.9326344728469849, + "learning_rate": 1.6859525312088166e-06, + "loss": 0.5858, + "step": 4425 + }, + { + "epoch": 3.664273881833241, + "grad_norm": 0.9377856254577637, + "learning_rate": 1.6849128525004016e-06, + "loss": 0.5769, + "step": 4426 + }, + { + "epoch": 3.66510215350635, + "grad_norm": 0.939058244228363, + "learning_rate": 1.6838733315037506e-06, + "loss": 0.5574, + "step": 4427 + }, + { + "epoch": 3.665930425179459, + "grad_norm": 0.9490857720375061, + "learning_rate": 1.682833968420003e-06, + "loss": 0.5636, + "step": 4428 + }, + { + "epoch": 3.6667586968525674, + "grad_norm": 0.902042806148529, + "learning_rate": 1.6817947634502646e-06, + "loss": 0.5698, + "step": 4429 + }, + { + "epoch": 3.6675869685256766, + "grad_norm": 0.9587634205818176, + "learning_rate": 1.6807557167956118e-06, + "loss": 0.5509, + "step": 4430 + }, + { + "epoch": 3.6684152401987853, + "grad_norm": 0.9205626845359802, + "learning_rate": 1.6797168286570914e-06, + "loss": 0.5712, + "step": 4431 + }, + { + "epoch": 3.669243511871894, + "grad_norm": 0.9732058048248291, + "learning_rate": 1.6786780992357182e-06, + "loss": 0.5991, + "step": 4432 + }, + { + "epoch": 3.6700717835450027, + "grad_norm": 0.9664617776870728, + "learning_rate": 1.6776395287324763e-06, + "loss": 0.5626, + "step": 4433 + }, + { + "epoch": 3.6709000552181115, + "grad_norm": 0.943077802658081, + "learning_rate": 1.676601117348321e-06, + "loss": 0.5699, + "step": 4434 + }, + { + "epoch": 3.67172832689122, + "grad_norm": 0.9534650444984436, + "learning_rate": 1.6755628652841721e-06, + "loss": 0.5783, + "step": 4435 + }, + { + "epoch": 3.672556598564329, + "grad_norm": 0.9525501132011414, + "learning_rate": 1.6745247727409242e-06, + "loss": 0.5793, + "step": 4436 + }, + { + "epoch": 3.673384870237438, + "grad_norm": 0.9676088690757751, + "learning_rate": 1.6734868399194377e-06, + "loss": 0.5638, + "step": 4437 + }, + { + "epoch": 3.674213141910547, + "grad_norm": 0.9717739820480347, + "learning_rate": 1.6724490670205423e-06, + "loss": 0.585, + "step": 4438 + }, + { + "epoch": 3.6750414135836555, + "grad_norm": 0.9457798600196838, + "learning_rate": 1.6714114542450388e-06, + "loss": 0.5558, + "step": 4439 + }, + { + "epoch": 3.6758696852567643, + "grad_norm": 0.940519392490387, + "learning_rate": 1.6703740017936948e-06, + "loss": 0.5815, + "step": 4440 + }, + { + "epoch": 3.676697956929873, + "grad_norm": 0.9216038584709167, + "learning_rate": 1.6693367098672475e-06, + "loss": 0.5629, + "step": 4441 + }, + { + "epoch": 3.6775262286029817, + "grad_norm": 0.927438497543335, + "learning_rate": 1.668299578666404e-06, + "loss": 0.5742, + "step": 4442 + }, + { + "epoch": 3.6783545002760905, + "grad_norm": 0.9554510712623596, + "learning_rate": 1.6672626083918387e-06, + "loss": 0.5709, + "step": 4443 + }, + { + "epoch": 3.679182771949199, + "grad_norm": 0.976613461971283, + "learning_rate": 1.6662257992441957e-06, + "loss": 0.5867, + "step": 4444 + }, + { + "epoch": 3.680011043622308, + "grad_norm": 0.9392770528793335, + "learning_rate": 1.6651891514240887e-06, + "loss": 0.563, + "step": 4445 + }, + { + "epoch": 3.680839315295417, + "grad_norm": 0.925324022769928, + "learning_rate": 1.6641526651320995e-06, + "loss": 0.5641, + "step": 4446 + }, + { + "epoch": 3.681667586968526, + "grad_norm": 0.9984707832336426, + "learning_rate": 1.6631163405687773e-06, + "loss": 0.5811, + "step": 4447 + }, + { + "epoch": 3.6824958586416345, + "grad_norm": 0.9349227547645569, + "learning_rate": 1.662080177934642e-06, + "loss": 0.5696, + "step": 4448 + }, + { + "epoch": 3.6833241303147433, + "grad_norm": 0.9371916651725769, + "learning_rate": 1.6610441774301806e-06, + "loss": 0.5787, + "step": 4449 + }, + { + "epoch": 3.684152401987852, + "grad_norm": 0.9504124522209167, + "learning_rate": 1.6600083392558508e-06, + "loss": 0.5706, + "step": 4450 + }, + { + "epoch": 3.6849806736609607, + "grad_norm": 0.9354562759399414, + "learning_rate": 1.6589726636120768e-06, + "loss": 0.5826, + "step": 4451 + }, + { + "epoch": 3.6858089453340694, + "grad_norm": 0.9202892780303955, + "learning_rate": 1.6579371506992523e-06, + "loss": 0.5807, + "step": 4452 + }, + { + "epoch": 3.6866372170071786, + "grad_norm": 0.9467330574989319, + "learning_rate": 1.65690180071774e-06, + "loss": 0.582, + "step": 4453 + }, + { + "epoch": 3.687465488680287, + "grad_norm": 0.9724621772766113, + "learning_rate": 1.6558666138678698e-06, + "loss": 0.5847, + "step": 4454 + }, + { + "epoch": 3.688293760353396, + "grad_norm": 0.9587774276733398, + "learning_rate": 1.6548315903499407e-06, + "loss": 0.5711, + "step": 4455 + }, + { + "epoch": 3.689122032026505, + "grad_norm": 0.9332905411720276, + "learning_rate": 1.6537967303642205e-06, + "loss": 0.5783, + "step": 4456 + }, + { + "epoch": 3.6899503036996135, + "grad_norm": 0.9627639055252075, + "learning_rate": 1.6527620341109452e-06, + "loss": 0.5741, + "step": 4457 + }, + { + "epoch": 3.6907785753727222, + "grad_norm": 0.9341310262680054, + "learning_rate": 1.6517275017903183e-06, + "loss": 0.5605, + "step": 4458 + }, + { + "epoch": 3.691606847045831, + "grad_norm": 0.9802857637405396, + "learning_rate": 1.6506931336025122e-06, + "loss": 0.5723, + "step": 4459 + }, + { + "epoch": 3.6924351187189397, + "grad_norm": 0.9871952533721924, + "learning_rate": 1.649658929747667e-06, + "loss": 0.5768, + "step": 4460 + }, + { + "epoch": 3.6932633903920484, + "grad_norm": 0.9286062717437744, + "learning_rate": 1.6486248904258929e-06, + "loss": 0.566, + "step": 4461 + }, + { + "epoch": 3.6940916620651576, + "grad_norm": 0.9631725549697876, + "learning_rate": 1.6475910158372654e-06, + "loss": 0.571, + "step": 4462 + }, + { + "epoch": 3.694919933738266, + "grad_norm": 0.9414538741111755, + "learning_rate": 1.6465573061818302e-06, + "loss": 0.5526, + "step": 4463 + }, + { + "epoch": 3.695748205411375, + "grad_norm": 0.9554463028907776, + "learning_rate": 1.6455237616596011e-06, + "loss": 0.5846, + "step": 4464 + }, + { + "epoch": 3.696576477084484, + "grad_norm": 0.9494985938072205, + "learning_rate": 1.6444903824705587e-06, + "loss": 0.5667, + "step": 4465 + }, + { + "epoch": 3.6974047487575925, + "grad_norm": 0.9633604288101196, + "learning_rate": 1.6434571688146522e-06, + "loss": 0.5862, + "step": 4466 + }, + { + "epoch": 3.6982330204307012, + "grad_norm": 0.955493152141571, + "learning_rate": 1.642424120891799e-06, + "loss": 0.5591, + "step": 4467 + }, + { + "epoch": 3.69906129210381, + "grad_norm": 0.9700822830200195, + "learning_rate": 1.641391238901885e-06, + "loss": 0.5733, + "step": 4468 + }, + { + "epoch": 3.6998895637769187, + "grad_norm": 0.9600526094436646, + "learning_rate": 1.6403585230447617e-06, + "loss": 0.5819, + "step": 4469 + }, + { + "epoch": 3.7007178354500274, + "grad_norm": 0.9373101592063904, + "learning_rate": 1.639325973520251e-06, + "loss": 0.5749, + "step": 4470 + }, + { + "epoch": 3.7015461071231366, + "grad_norm": 0.9339377284049988, + "learning_rate": 1.6382935905281406e-06, + "loss": 0.5812, + "step": 4471 + }, + { + "epoch": 3.7023743787962453, + "grad_norm": 0.9409409165382385, + "learning_rate": 1.637261374268188e-06, + "loss": 0.561, + "step": 4472 + }, + { + "epoch": 3.703202650469354, + "grad_norm": 0.9259868264198303, + "learning_rate": 1.6362293249401174e-06, + "loss": 0.5636, + "step": 4473 + }, + { + "epoch": 3.7040309221424628, + "grad_norm": 0.9868617057800293, + "learning_rate": 1.6351974427436198e-06, + "loss": 0.5753, + "step": 4474 + }, + { + "epoch": 3.7048591938155715, + "grad_norm": 1.001633882522583, + "learning_rate": 1.6341657278783563e-06, + "loss": 0.5834, + "step": 4475 + }, + { + "epoch": 3.70568746548868, + "grad_norm": 0.9641216993331909, + "learning_rate": 1.6331341805439527e-06, + "loss": 0.6006, + "step": 4476 + }, + { + "epoch": 3.706515737161789, + "grad_norm": 0.9380208253860474, + "learning_rate": 1.6321028009400041e-06, + "loss": 0.5827, + "step": 4477 + }, + { + "epoch": 3.707344008834898, + "grad_norm": 0.9479442238807678, + "learning_rate": 1.631071589266074e-06, + "loss": 0.5537, + "step": 4478 + }, + { + "epoch": 3.7081722805080064, + "grad_norm": 0.9482294321060181, + "learning_rate": 1.6300405457216912e-06, + "loss": 0.5811, + "step": 4479 + }, + { + "epoch": 3.7090005521811156, + "grad_norm": 0.9395027756690979, + "learning_rate": 1.629009670506353e-06, + "loss": 0.5683, + "step": 4480 + }, + { + "epoch": 3.7098288238542243, + "grad_norm": 0.9259159564971924, + "learning_rate": 1.6279789638195242e-06, + "loss": 0.5659, + "step": 4481 + }, + { + "epoch": 3.710657095527333, + "grad_norm": 0.9183517694473267, + "learning_rate": 1.6269484258606366e-06, + "loss": 0.5657, + "step": 4482 + }, + { + "epoch": 3.7114853672004418, + "grad_norm": 0.9426186680793762, + "learning_rate": 1.6259180568290905e-06, + "loss": 0.5749, + "step": 4483 + }, + { + "epoch": 3.7123136388735505, + "grad_norm": 0.9540229439735413, + "learning_rate": 1.6248878569242522e-06, + "loss": 0.5578, + "step": 4484 + }, + { + "epoch": 3.713141910546659, + "grad_norm": 0.9624539613723755, + "learning_rate": 1.6238578263454558e-06, + "loss": 0.5855, + "step": 4485 + }, + { + "epoch": 3.713970182219768, + "grad_norm": 0.9445024728775024, + "learning_rate": 1.6228279652920026e-06, + "loss": 0.5464, + "step": 4486 + }, + { + "epoch": 3.714798453892877, + "grad_norm": 0.9346494078636169, + "learning_rate": 1.6217982739631617e-06, + "loss": 0.5523, + "step": 4487 + }, + { + "epoch": 3.7156267255659854, + "grad_norm": 1.0044512748718262, + "learning_rate": 1.6207687525581684e-06, + "loss": 0.5586, + "step": 4488 + }, + { + "epoch": 3.7164549972390946, + "grad_norm": 0.9639768600463867, + "learning_rate": 1.6197394012762245e-06, + "loss": 0.5676, + "step": 4489 + }, + { + "epoch": 3.7172832689122033, + "grad_norm": 0.9930342435836792, + "learning_rate": 1.6187102203165011e-06, + "loss": 0.5463, + "step": 4490 + }, + { + "epoch": 3.718111540585312, + "grad_norm": 0.9646494388580322, + "learning_rate": 1.617681209878136e-06, + "loss": 0.5516, + "step": 4491 + }, + { + "epoch": 3.7189398122584207, + "grad_norm": 0.9502953886985779, + "learning_rate": 1.616652370160231e-06, + "loss": 0.5792, + "step": 4492 + }, + { + "epoch": 3.7197680839315295, + "grad_norm": 0.9354153871536255, + "learning_rate": 1.6156237013618573e-06, + "loss": 0.549, + "step": 4493 + }, + { + "epoch": 3.720596355604638, + "grad_norm": 0.9361369609832764, + "learning_rate": 1.6145952036820538e-06, + "loss": 0.566, + "step": 4494 + }, + { + "epoch": 3.721424627277747, + "grad_norm": 0.9847058653831482, + "learning_rate": 1.6135668773198252e-06, + "loss": 0.561, + "step": 4495 + }, + { + "epoch": 3.722252898950856, + "grad_norm": 0.9457981586456299, + "learning_rate": 1.6125387224741422e-06, + "loss": 0.5689, + "step": 4496 + }, + { + "epoch": 3.723081170623965, + "grad_norm": 0.9350807666778564, + "learning_rate": 1.6115107393439433e-06, + "loss": 0.5603, + "step": 4497 + }, + { + "epoch": 3.7239094422970735, + "grad_norm": 0.9243747591972351, + "learning_rate": 1.6104829281281343e-06, + "loss": 0.578, + "step": 4498 + }, + { + "epoch": 3.7247377139701823, + "grad_norm": 0.9140326976776123, + "learning_rate": 1.6094552890255869e-06, + "loss": 0.5831, + "step": 4499 + }, + { + "epoch": 3.725565985643291, + "grad_norm": 0.9408622980117798, + "learning_rate": 1.6084278222351391e-06, + "loss": 0.5684, + "step": 4500 + }, + { + "epoch": 3.7263942573163997, + "grad_norm": 0.9529958367347717, + "learning_rate": 1.6074005279555965e-06, + "loss": 0.5876, + "step": 4501 + }, + { + "epoch": 3.7272225289895085, + "grad_norm": 0.9634376764297485, + "learning_rate": 1.6063734063857323e-06, + "loss": 0.5809, + "step": 4502 + }, + { + "epoch": 3.728050800662617, + "grad_norm": 0.9465759992599487, + "learning_rate": 1.6053464577242828e-06, + "loss": 0.564, + "step": 4503 + }, + { + "epoch": 3.728879072335726, + "grad_norm": 0.943699836730957, + "learning_rate": 1.6043196821699536e-06, + "loss": 0.5797, + "step": 4504 + }, + { + "epoch": 3.729707344008835, + "grad_norm": 0.9360603094100952, + "learning_rate": 1.6032930799214164e-06, + "loss": 0.561, + "step": 4505 + }, + { + "epoch": 3.730535615681944, + "grad_norm": 0.9656838178634644, + "learning_rate": 1.6022666511773098e-06, + "loss": 0.5726, + "step": 4506 + }, + { + "epoch": 3.7313638873550525, + "grad_norm": 0.9636357426643372, + "learning_rate": 1.6012403961362372e-06, + "loss": 0.5859, + "step": 4507 + }, + { + "epoch": 3.7321921590281613, + "grad_norm": 0.9537240862846375, + "learning_rate": 1.6002143149967697e-06, + "loss": 0.5798, + "step": 4508 + }, + { + "epoch": 3.73302043070127, + "grad_norm": 0.9626145362854004, + "learning_rate": 1.5991884079574448e-06, + "loss": 0.5778, + "step": 4509 + }, + { + "epoch": 3.7338487023743787, + "grad_norm": 0.9493876099586487, + "learning_rate": 1.598162675216766e-06, + "loss": 0.5688, + "step": 4510 + }, + { + "epoch": 3.7346769740474874, + "grad_norm": 0.9417327046394348, + "learning_rate": 1.597137116973202e-06, + "loss": 0.5628, + "step": 4511 + }, + { + "epoch": 3.7355052457205966, + "grad_norm": 0.9836271405220032, + "learning_rate": 1.59611173342519e-06, + "loss": 0.5616, + "step": 4512 + }, + { + "epoch": 3.736333517393705, + "grad_norm": 0.9088386297225952, + "learning_rate": 1.5950865247711322e-06, + "loss": 0.5635, + "step": 4513 + }, + { + "epoch": 3.737161789066814, + "grad_norm": 0.9311172366142273, + "learning_rate": 1.5940614912093957e-06, + "loss": 0.5795, + "step": 4514 + }, + { + "epoch": 3.737990060739923, + "grad_norm": 0.9570069909095764, + "learning_rate": 1.5930366329383157e-06, + "loss": 0.5753, + "step": 4515 + }, + { + "epoch": 3.7388183324130315, + "grad_norm": 1.1909716129302979, + "learning_rate": 1.592011950156192e-06, + "loss": 0.5637, + "step": 4516 + }, + { + "epoch": 3.7396466040861402, + "grad_norm": 1.0059658288955688, + "learning_rate": 1.5909874430612925e-06, + "loss": 0.5894, + "step": 4517 + }, + { + "epoch": 3.740474875759249, + "grad_norm": 0.9572764039039612, + "learning_rate": 1.5899631118518491e-06, + "loss": 0.5762, + "step": 4518 + }, + { + "epoch": 3.7413031474323577, + "grad_norm": 0.9477603435516357, + "learning_rate": 1.58893895672606e-06, + "loss": 0.5602, + "step": 4519 + }, + { + "epoch": 3.7421314191054664, + "grad_norm": 0.9420804381370544, + "learning_rate": 1.5879149778820898e-06, + "loss": 0.5717, + "step": 4520 + }, + { + "epoch": 3.7429596907785756, + "grad_norm": 0.9785239100456238, + "learning_rate": 1.5868911755180694e-06, + "loss": 0.571, + "step": 4521 + }, + { + "epoch": 3.743787962451684, + "grad_norm": 0.9285955429077148, + "learning_rate": 1.585867549832094e-06, + "loss": 0.5603, + "step": 4522 + }, + { + "epoch": 3.744616234124793, + "grad_norm": 0.971987247467041, + "learning_rate": 1.584844101022227e-06, + "loss": 0.5829, + "step": 4523 + }, + { + "epoch": 3.745444505797902, + "grad_norm": 0.9516214728355408, + "learning_rate": 1.5838208292864954e-06, + "loss": 0.583, + "step": 4524 + }, + { + "epoch": 3.7462727774710105, + "grad_norm": 0.938643753528595, + "learning_rate": 1.5827977348228935e-06, + "loss": 0.5427, + "step": 4525 + }, + { + "epoch": 3.7471010491441192, + "grad_norm": 0.9617331027984619, + "learning_rate": 1.581774817829379e-06, + "loss": 0.5663, + "step": 4526 + }, + { + "epoch": 3.747929320817228, + "grad_norm": 0.9507145881652832, + "learning_rate": 1.5807520785038776e-06, + "loss": 0.5855, + "step": 4527 + }, + { + "epoch": 3.7487575924903367, + "grad_norm": 0.955471932888031, + "learning_rate": 1.57972951704428e-06, + "loss": 0.5676, + "step": 4528 + }, + { + "epoch": 3.7495858641634454, + "grad_norm": 0.9277394413948059, + "learning_rate": 1.5787071336484422e-06, + "loss": 0.5452, + "step": 4529 + }, + { + "epoch": 3.7504141358365546, + "grad_norm": 0.9697600603103638, + "learning_rate": 1.5776849285141858e-06, + "loss": 0.5598, + "step": 4530 + }, + { + "epoch": 3.7512424075096633, + "grad_norm": 0.9462587833404541, + "learning_rate": 1.5766629018392981e-06, + "loss": 0.5766, + "step": 4531 + }, + { + "epoch": 3.752070679182772, + "grad_norm": 0.9286374449729919, + "learning_rate": 1.5756410538215317e-06, + "loss": 0.5747, + "step": 4532 + }, + { + "epoch": 3.7528989508558808, + "grad_norm": 0.9680078625679016, + "learning_rate": 1.5746193846586045e-06, + "loss": 0.5789, + "step": 4533 + }, + { + "epoch": 3.7537272225289895, + "grad_norm": 0.9353684782981873, + "learning_rate": 1.5735978945482e-06, + "loss": 0.5763, + "step": 4534 + }, + { + "epoch": 3.754555494202098, + "grad_norm": 0.9365376830101013, + "learning_rate": 1.5725765836879674e-06, + "loss": 0.5504, + "step": 4535 + }, + { + "epoch": 3.755383765875207, + "grad_norm": 0.9342913627624512, + "learning_rate": 1.571555452275521e-06, + "loss": 0.558, + "step": 4536 + }, + { + "epoch": 3.756212037548316, + "grad_norm": 0.9252297878265381, + "learning_rate": 1.5705345005084394e-06, + "loss": 0.5666, + "step": 4537 + }, + { + "epoch": 3.7570403092214244, + "grad_norm": 1.0099254846572876, + "learning_rate": 1.5695137285842666e-06, + "loss": 0.5627, + "step": 4538 + }, + { + "epoch": 3.7578685808945336, + "grad_norm": 0.9464092254638672, + "learning_rate": 1.5684931367005144e-06, + "loss": 0.5642, + "step": 4539 + }, + { + "epoch": 3.7586968525676423, + "grad_norm": 0.93072509765625, + "learning_rate": 1.5674727250546568e-06, + "loss": 0.5821, + "step": 4540 + }, + { + "epoch": 3.759525124240751, + "grad_norm": 0.9577572345733643, + "learning_rate": 1.5664524938441334e-06, + "loss": 0.56, + "step": 4541 + }, + { + "epoch": 3.7603533959138598, + "grad_norm": 0.9372087121009827, + "learning_rate": 1.5654324432663503e-06, + "loss": 0.5583, + "step": 4542 + }, + { + "epoch": 3.7611816675869685, + "grad_norm": 0.9543371796607971, + "learning_rate": 1.5644125735186776e-06, + "loss": 0.5611, + "step": 4543 + }, + { + "epoch": 3.762009939260077, + "grad_norm": 0.9668126702308655, + "learning_rate": 1.5633928847984503e-06, + "loss": 0.5637, + "step": 4544 + }, + { + "epoch": 3.762838210933186, + "grad_norm": 0.9686347246170044, + "learning_rate": 1.5623733773029692e-06, + "loss": 0.5737, + "step": 4545 + }, + { + "epoch": 3.763666482606295, + "grad_norm": 0.9634156227111816, + "learning_rate": 1.561354051229499e-06, + "loss": 0.5602, + "step": 4546 + }, + { + "epoch": 3.7644947542794034, + "grad_norm": 0.9353019595146179, + "learning_rate": 1.5603349067752705e-06, + "loss": 0.5518, + "step": 4547 + }, + { + "epoch": 3.7653230259525126, + "grad_norm": 0.9436436295509338, + "learning_rate": 1.5593159441374778e-06, + "loss": 0.5652, + "step": 4548 + }, + { + "epoch": 3.7661512976256213, + "grad_norm": 1.0265226364135742, + "learning_rate": 1.55829716351328e-06, + "loss": 0.5624, + "step": 4549 + }, + { + "epoch": 3.76697956929873, + "grad_norm": 0.9422723650932312, + "learning_rate": 1.5572785650998035e-06, + "loss": 0.5649, + "step": 4550 + }, + { + "epoch": 3.7678078409718387, + "grad_norm": 0.9558239579200745, + "learning_rate": 1.5562601490941365e-06, + "loss": 0.5553, + "step": 4551 + }, + { + "epoch": 3.7686361126449475, + "grad_norm": 0.9281789660453796, + "learning_rate": 1.555241915693333e-06, + "loss": 0.5582, + "step": 4552 + }, + { + "epoch": 3.769464384318056, + "grad_norm": 0.9265138506889343, + "learning_rate": 1.5542238650944121e-06, + "loss": 0.5803, + "step": 4553 + }, + { + "epoch": 3.770292655991165, + "grad_norm": 0.9384198784828186, + "learning_rate": 1.5532059974943569e-06, + "loss": 0.5821, + "step": 4554 + }, + { + "epoch": 3.771120927664274, + "grad_norm": 0.9323135018348694, + "learning_rate": 1.552188313090115e-06, + "loss": 0.56, + "step": 4555 + }, + { + "epoch": 3.771949199337383, + "grad_norm": 0.9448620080947876, + "learning_rate": 1.5511708120785994e-06, + "loss": 0.5521, + "step": 4556 + }, + { + "epoch": 3.7727774710104915, + "grad_norm": 0.9806753993034363, + "learning_rate": 1.5501534946566871e-06, + "loss": 0.5842, + "step": 4557 + }, + { + "epoch": 3.7736057426836003, + "grad_norm": 0.9242349863052368, + "learning_rate": 1.549136361021219e-06, + "loss": 0.5764, + "step": 4558 + }, + { + "epoch": 3.774434014356709, + "grad_norm": 0.956167459487915, + "learning_rate": 1.5481194113690017e-06, + "loss": 0.5891, + "step": 4559 + }, + { + "epoch": 3.7752622860298177, + "grad_norm": 0.9456217288970947, + "learning_rate": 1.5471026458968041e-06, + "loss": 0.5821, + "step": 4560 + }, + { + "epoch": 3.7760905577029265, + "grad_norm": 0.9384570717811584, + "learning_rate": 1.5460860648013626e-06, + "loss": 0.5662, + "step": 4561 + }, + { + "epoch": 3.776918829376035, + "grad_norm": 0.9760017395019531, + "learning_rate": 1.5450696682793747e-06, + "loss": 0.5862, + "step": 4562 + }, + { + "epoch": 3.777747101049144, + "grad_norm": 0.9794814586639404, + "learning_rate": 1.5440534565275039e-06, + "loss": 0.5852, + "step": 4563 + }, + { + "epoch": 3.778575372722253, + "grad_norm": 0.9752492904663086, + "learning_rate": 1.5430374297423784e-06, + "loss": 0.5909, + "step": 4564 + }, + { + "epoch": 3.779403644395362, + "grad_norm": 0.969682514667511, + "learning_rate": 1.5420215881205897e-06, + "loss": 0.5505, + "step": 4565 + }, + { + "epoch": 3.7802319160684705, + "grad_norm": 0.9354566335678101, + "learning_rate": 1.5410059318586934e-06, + "loss": 0.5643, + "step": 4566 + }, + { + "epoch": 3.7810601877415793, + "grad_norm": 0.9543263912200928, + "learning_rate": 1.5399904611532095e-06, + "loss": 0.579, + "step": 4567 + }, + { + "epoch": 3.781888459414688, + "grad_norm": 0.9573609232902527, + "learning_rate": 1.5389751762006225e-06, + "loss": 0.5815, + "step": 4568 + }, + { + "epoch": 3.7827167310877967, + "grad_norm": 0.9919794201850891, + "learning_rate": 1.5379600771973805e-06, + "loss": 0.5643, + "step": 4569 + }, + { + "epoch": 3.7835450027609054, + "grad_norm": 0.966391921043396, + "learning_rate": 1.5369451643398959e-06, + "loss": 0.5702, + "step": 4570 + }, + { + "epoch": 3.7843732744340146, + "grad_norm": 0.9727235436439514, + "learning_rate": 1.5359304378245435e-06, + "loss": 0.5607, + "step": 4571 + }, + { + "epoch": 3.785201546107123, + "grad_norm": 0.9183128476142883, + "learning_rate": 1.5349158978476653e-06, + "loss": 0.539, + "step": 4572 + }, + { + "epoch": 3.786029817780232, + "grad_norm": 0.980755090713501, + "learning_rate": 1.533901544605564e-06, + "loss": 0.5761, + "step": 4573 + }, + { + "epoch": 3.786858089453341, + "grad_norm": 0.96429842710495, + "learning_rate": 1.532887378294508e-06, + "loss": 0.5797, + "step": 4574 + }, + { + "epoch": 3.7876863611264495, + "grad_norm": 0.9631879329681396, + "learning_rate": 1.5318733991107287e-06, + "loss": 0.5851, + "step": 4575 + }, + { + "epoch": 3.7885146327995582, + "grad_norm": 0.9594787359237671, + "learning_rate": 1.5308596072504223e-06, + "loss": 0.5656, + "step": 4576 + }, + { + "epoch": 3.789342904472667, + "grad_norm": 0.9394322633743286, + "learning_rate": 1.5298460029097479e-06, + "loss": 0.5554, + "step": 4577 + }, + { + "epoch": 3.7901711761457757, + "grad_norm": 0.9391270875930786, + "learning_rate": 1.5288325862848275e-06, + "loss": 0.5733, + "step": 4578 + }, + { + "epoch": 3.7909994478188844, + "grad_norm": 0.9480547904968262, + "learning_rate": 1.5278193575717492e-06, + "loss": 0.5629, + "step": 4579 + }, + { + "epoch": 3.7918277194919936, + "grad_norm": 0.9235183000564575, + "learning_rate": 1.5268063169665626e-06, + "loss": 0.5626, + "step": 4580 + }, + { + "epoch": 3.792655991165102, + "grad_norm": 0.9616346955299377, + "learning_rate": 1.5257934646652822e-06, + "loss": 0.557, + "step": 4581 + }, + { + "epoch": 3.793484262838211, + "grad_norm": 0.9767917394638062, + "learning_rate": 1.5247808008638843e-06, + "loss": 0.5693, + "step": 4582 + }, + { + "epoch": 3.79431253451132, + "grad_norm": 0.9262551665306091, + "learning_rate": 1.5237683257583102e-06, + "loss": 0.5638, + "step": 4583 + }, + { + "epoch": 3.7951408061844285, + "grad_norm": 0.9267538189888, + "learning_rate": 1.522756039544465e-06, + "loss": 0.5801, + "step": 4584 + }, + { + "epoch": 3.7959690778575372, + "grad_norm": 0.9917179346084595, + "learning_rate": 1.5217439424182167e-06, + "loss": 0.5715, + "step": 4585 + }, + { + "epoch": 3.796797349530646, + "grad_norm": 0.9482493996620178, + "learning_rate": 1.5207320345753957e-06, + "loss": 0.5618, + "step": 4586 + }, + { + "epoch": 3.7976256212037547, + "grad_norm": 0.9342687726020813, + "learning_rate": 1.5197203162117979e-06, + "loss": 0.5863, + "step": 4587 + }, + { + "epoch": 3.7984538928768634, + "grad_norm": 0.9577085375785828, + "learning_rate": 1.5187087875231809e-06, + "loss": 0.5501, + "step": 4588 + }, + { + "epoch": 3.7992821645499726, + "grad_norm": 0.9561404585838318, + "learning_rate": 1.5176974487052658e-06, + "loss": 0.576, + "step": 4589 + }, + { + "epoch": 3.8001104362230813, + "grad_norm": 0.9302974343299866, + "learning_rate": 1.5166862999537377e-06, + "loss": 0.5551, + "step": 4590 + }, + { + "epoch": 3.80093870789619, + "grad_norm": 0.9463371634483337, + "learning_rate": 1.5156753414642445e-06, + "loss": 0.5531, + "step": 4591 + }, + { + "epoch": 3.8017669795692988, + "grad_norm": 0.9446096420288086, + "learning_rate": 1.5146645734323961e-06, + "loss": 0.5652, + "step": 4592 + }, + { + "epoch": 3.8025952512424075, + "grad_norm": 0.9582951068878174, + "learning_rate": 1.513653996053769e-06, + "loss": 0.5709, + "step": 4593 + }, + { + "epoch": 3.803423522915516, + "grad_norm": 0.961769163608551, + "learning_rate": 1.5126436095238978e-06, + "loss": 0.576, + "step": 4594 + }, + { + "epoch": 3.804251794588625, + "grad_norm": 0.97950279712677, + "learning_rate": 1.5116334140382846e-06, + "loss": 0.562, + "step": 4595 + }, + { + "epoch": 3.805080066261734, + "grad_norm": 0.9386860132217407, + "learning_rate": 1.5106234097923922e-06, + "loss": 0.5723, + "step": 4596 + }, + { + "epoch": 3.8059083379348424, + "grad_norm": 0.9200562238693237, + "learning_rate": 1.509613596981647e-06, + "loss": 0.5708, + "step": 4597 + }, + { + "epoch": 3.8067366096079516, + "grad_norm": 0.9222211837768555, + "learning_rate": 1.5086039758014386e-06, + "loss": 0.5711, + "step": 4598 + }, + { + "epoch": 3.8075648812810603, + "grad_norm": 0.9545297026634216, + "learning_rate": 1.5075945464471198e-06, + "loss": 0.5628, + "step": 4599 + }, + { + "epoch": 3.808393152954169, + "grad_norm": 0.918067455291748, + "learning_rate": 1.5065853091140041e-06, + "loss": 0.5429, + "step": 4600 + }, + { + "epoch": 3.8092214246272778, + "grad_norm": 0.9256977438926697, + "learning_rate": 1.5055762639973714e-06, + "loss": 0.5593, + "step": 4601 + }, + { + "epoch": 3.8100496963003865, + "grad_norm": 0.9570629000663757, + "learning_rate": 1.5045674112924612e-06, + "loss": 0.5545, + "step": 4602 + }, + { + "epoch": 3.810877967973495, + "grad_norm": 0.9828989505767822, + "learning_rate": 1.5035587511944778e-06, + "loss": 0.5776, + "step": 4603 + }, + { + "epoch": 3.811706239646604, + "grad_norm": 0.9521099925041199, + "learning_rate": 1.5025502838985884e-06, + "loss": 0.5645, + "step": 4604 + }, + { + "epoch": 3.812534511319713, + "grad_norm": 0.9580879211425781, + "learning_rate": 1.5015420095999197e-06, + "loss": 0.5754, + "step": 4605 + }, + { + "epoch": 3.8133627829928214, + "grad_norm": 0.9599860906600952, + "learning_rate": 1.500533928493565e-06, + "loss": 0.5623, + "step": 4606 + }, + { + "epoch": 3.8141910546659306, + "grad_norm": 0.9527266621589661, + "learning_rate": 1.4995260407745782e-06, + "loss": 0.5931, + "step": 4607 + }, + { + "epoch": 3.8150193263390393, + "grad_norm": 0.9410575032234192, + "learning_rate": 1.498518346637976e-06, + "loss": 0.569, + "step": 4608 + }, + { + "epoch": 3.815847598012148, + "grad_norm": 0.9750106930732727, + "learning_rate": 1.4975108462787382e-06, + "loss": 0.5838, + "step": 4609 + }, + { + "epoch": 3.8166758696852567, + "grad_norm": 0.9242303967475891, + "learning_rate": 1.4965035398918065e-06, + "loss": 0.5595, + "step": 4610 + }, + { + "epoch": 3.8175041413583655, + "grad_norm": 0.9424212574958801, + "learning_rate": 1.495496427672085e-06, + "loss": 0.5949, + "step": 4611 + }, + { + "epoch": 3.818332413031474, + "grad_norm": 0.9485701322555542, + "learning_rate": 1.4944895098144419e-06, + "loss": 0.5799, + "step": 4612 + }, + { + "epoch": 3.819160684704583, + "grad_norm": 0.9476267695426941, + "learning_rate": 1.4934827865137047e-06, + "loss": 0.5584, + "step": 4613 + }, + { + "epoch": 3.819988956377692, + "grad_norm": 0.9722000360488892, + "learning_rate": 1.4924762579646653e-06, + "loss": 0.5665, + "step": 4614 + }, + { + "epoch": 3.820817228050801, + "grad_norm": 0.9728968739509583, + "learning_rate": 1.4914699243620795e-06, + "loss": 0.5786, + "step": 4615 + }, + { + "epoch": 3.8216454997239095, + "grad_norm": 0.96755051612854, + "learning_rate": 1.4904637859006605e-06, + "loss": 0.5529, + "step": 4616 + }, + { + "epoch": 3.8224737713970183, + "grad_norm": 0.9613732695579529, + "learning_rate": 1.4894578427750888e-06, + "loss": 0.5847, + "step": 4617 + }, + { + "epoch": 3.823302043070127, + "grad_norm": 0.9679767489433289, + "learning_rate": 1.488452095180004e-06, + "loss": 0.5623, + "step": 4618 + }, + { + "epoch": 3.8241303147432357, + "grad_norm": 0.9799363613128662, + "learning_rate": 1.4874465433100093e-06, + "loss": 0.5712, + "step": 4619 + }, + { + "epoch": 3.8249585864163445, + "grad_norm": 0.962107241153717, + "learning_rate": 1.4864411873596695e-06, + "loss": 0.5596, + "step": 4620 + }, + { + "epoch": 3.825786858089453, + "grad_norm": 0.9862436056137085, + "learning_rate": 1.4854360275235118e-06, + "loss": 0.5801, + "step": 4621 + }, + { + "epoch": 3.826615129762562, + "grad_norm": 0.9500226974487305, + "learning_rate": 1.484431063996025e-06, + "loss": 0.5467, + "step": 4622 + }, + { + "epoch": 3.827443401435671, + "grad_norm": 0.9733493328094482, + "learning_rate": 1.4834262969716607e-06, + "loss": 0.549, + "step": 4623 + }, + { + "epoch": 3.82827167310878, + "grad_norm": 0.9474700093269348, + "learning_rate": 1.4824217266448312e-06, + "loss": 0.5497, + "step": 4624 + }, + { + "epoch": 3.8290999447818885, + "grad_norm": 0.9376130104064941, + "learning_rate": 1.4814173532099116e-06, + "loss": 0.5681, + "step": 4625 + }, + { + "epoch": 3.8299282164549973, + "grad_norm": 0.9183100461959839, + "learning_rate": 1.4804131768612406e-06, + "loss": 0.5719, + "step": 4626 + }, + { + "epoch": 3.830756488128106, + "grad_norm": 0.9759798645973206, + "learning_rate": 1.4794091977931136e-06, + "loss": 0.5773, + "step": 4627 + }, + { + "epoch": 3.8315847598012147, + "grad_norm": 0.9638318419456482, + "learning_rate": 1.4784054161997937e-06, + "loss": 0.5681, + "step": 4628 + }, + { + "epoch": 3.8324130314743234, + "grad_norm": 0.9652249217033386, + "learning_rate": 1.4774018322755027e-06, + "loss": 0.5866, + "step": 4629 + }, + { + "epoch": 3.8332413031474326, + "grad_norm": 0.9402887225151062, + "learning_rate": 1.4763984462144237e-06, + "loss": 0.569, + "step": 4630 + }, + { + "epoch": 3.834069574820541, + "grad_norm": 0.925959587097168, + "learning_rate": 1.4753952582107046e-06, + "loss": 0.5945, + "step": 4631 + }, + { + "epoch": 3.83489784649365, + "grad_norm": 0.9718639254570007, + "learning_rate": 1.4743922684584514e-06, + "loss": 0.5523, + "step": 4632 + }, + { + "epoch": 3.835726118166759, + "grad_norm": 0.9743525981903076, + "learning_rate": 1.4733894771517337e-06, + "loss": 0.58, + "step": 4633 + }, + { + "epoch": 3.8365543898398675, + "grad_norm": 0.9634748697280884, + "learning_rate": 1.4723868844845818e-06, + "loss": 0.5715, + "step": 4634 + }, + { + "epoch": 3.8373826615129762, + "grad_norm": 0.9775907397270203, + "learning_rate": 1.471384490650989e-06, + "loss": 0.5697, + "step": 4635 + }, + { + "epoch": 3.838210933186085, + "grad_norm": 0.9504139423370361, + "learning_rate": 1.4703822958449081e-06, + "loss": 0.5797, + "step": 4636 + }, + { + "epoch": 3.8390392048591937, + "grad_norm": 0.9717209339141846, + "learning_rate": 1.4693803002602553e-06, + "loss": 0.5692, + "step": 4637 + }, + { + "epoch": 3.8398674765323024, + "grad_norm": 0.9564605951309204, + "learning_rate": 1.4683785040909078e-06, + "loss": 0.5719, + "step": 4638 + }, + { + "epoch": 3.8406957482054116, + "grad_norm": 0.951598584651947, + "learning_rate": 1.4673769075307023e-06, + "loss": 0.5481, + "step": 4639 + }, + { + "epoch": 3.84152401987852, + "grad_norm": 0.9717726111412048, + "learning_rate": 1.4663755107734395e-06, + "loss": 0.537, + "step": 4640 + }, + { + "epoch": 3.842352291551629, + "grad_norm": 0.9653568267822266, + "learning_rate": 1.4653743140128802e-06, + "loss": 0.5729, + "step": 4641 + }, + { + "epoch": 3.843180563224738, + "grad_norm": 0.9710245728492737, + "learning_rate": 1.4643733174427454e-06, + "loss": 0.5816, + "step": 4642 + }, + { + "epoch": 3.8440088348978465, + "grad_norm": 0.9537909626960754, + "learning_rate": 1.4633725212567206e-06, + "loss": 0.5585, + "step": 4643 + }, + { + "epoch": 3.8448371065709552, + "grad_norm": 0.9765188694000244, + "learning_rate": 1.4623719256484495e-06, + "loss": 0.5542, + "step": 4644 + }, + { + "epoch": 3.845665378244064, + "grad_norm": 0.9446712136268616, + "learning_rate": 1.4613715308115378e-06, + "loss": 0.5696, + "step": 4645 + }, + { + "epoch": 3.8464936499171727, + "grad_norm": 0.9561092853546143, + "learning_rate": 1.4603713369395533e-06, + "loss": 0.555, + "step": 4646 + }, + { + "epoch": 3.8473219215902814, + "grad_norm": 0.9431309700012207, + "learning_rate": 1.4593713442260233e-06, + "loss": 0.5589, + "step": 4647 + }, + { + "epoch": 3.8481501932633906, + "grad_norm": 0.9409856796264648, + "learning_rate": 1.4583715528644384e-06, + "loss": 0.5574, + "step": 4648 + }, + { + "epoch": 3.8489784649364993, + "grad_norm": 0.9805465936660767, + "learning_rate": 1.4573719630482485e-06, + "loss": 0.5555, + "step": 4649 + }, + { + "epoch": 3.849806736609608, + "grad_norm": 0.9668418765068054, + "learning_rate": 1.456372574970863e-06, + "loss": 0.5721, + "step": 4650 + }, + { + "epoch": 3.8506350082827168, + "grad_norm": 0.937393844127655, + "learning_rate": 1.455373388825656e-06, + "loss": 0.5619, + "step": 4651 + }, + { + "epoch": 3.8514632799558255, + "grad_norm": 0.956749439239502, + "learning_rate": 1.4543744048059613e-06, + "loss": 0.5882, + "step": 4652 + }, + { + "epoch": 3.852291551628934, + "grad_norm": 0.95976722240448, + "learning_rate": 1.4533756231050712e-06, + "loss": 0.5573, + "step": 4653 + }, + { + "epoch": 3.853119823302043, + "grad_norm": 0.9402530789375305, + "learning_rate": 1.4523770439162414e-06, + "loss": 0.549, + "step": 4654 + }, + { + "epoch": 3.853948094975152, + "grad_norm": 0.960904061794281, + "learning_rate": 1.4513786674326888e-06, + "loss": 0.5671, + "step": 4655 + }, + { + "epoch": 3.8547763666482604, + "grad_norm": 0.929169774055481, + "learning_rate": 1.450380493847588e-06, + "loss": 0.5482, + "step": 4656 + }, + { + "epoch": 3.8556046383213696, + "grad_norm": 0.9257175922393799, + "learning_rate": 1.449382523354077e-06, + "loss": 0.5526, + "step": 4657 + }, + { + "epoch": 3.8564329099944783, + "grad_norm": 0.9734342098236084, + "learning_rate": 1.4483847561452553e-06, + "loss": 0.5601, + "step": 4658 + }, + { + "epoch": 3.857261181667587, + "grad_norm": 0.9698025584220886, + "learning_rate": 1.4473871924141788e-06, + "loss": 0.568, + "step": 4659 + }, + { + "epoch": 3.8580894533406958, + "grad_norm": 0.9453846216201782, + "learning_rate": 1.4463898323538694e-06, + "loss": 0.5618, + "step": 4660 + }, + { + "epoch": 3.8589177250138045, + "grad_norm": 0.9435855150222778, + "learning_rate": 1.4453926761573062e-06, + "loss": 0.559, + "step": 4661 + }, + { + "epoch": 3.859745996686913, + "grad_norm": 0.9480921030044556, + "learning_rate": 1.4443957240174277e-06, + "loss": 0.5593, + "step": 4662 + }, + { + "epoch": 3.860574268360022, + "grad_norm": 0.9516395926475525, + "learning_rate": 1.443398976127137e-06, + "loss": 0.552, + "step": 4663 + }, + { + "epoch": 3.861402540033131, + "grad_norm": 0.9439289569854736, + "learning_rate": 1.4424024326792955e-06, + "loss": 0.5564, + "step": 4664 + }, + { + "epoch": 3.8622308117062394, + "grad_norm": 0.9423355460166931, + "learning_rate": 1.4414060938667236e-06, + "loss": 0.5428, + "step": 4665 + }, + { + "epoch": 3.8630590833793486, + "grad_norm": 0.9807745814323425, + "learning_rate": 1.4404099598822041e-06, + "loss": 0.5695, + "step": 4666 + }, + { + "epoch": 3.8638873550524573, + "grad_norm": 0.9256470203399658, + "learning_rate": 1.4394140309184815e-06, + "loss": 0.5766, + "step": 4667 + }, + { + "epoch": 3.864715626725566, + "grad_norm": 0.9468652009963989, + "learning_rate": 1.4384183071682562e-06, + "loss": 0.5762, + "step": 4668 + }, + { + "epoch": 3.8655438983986747, + "grad_norm": 0.9390266537666321, + "learning_rate": 1.4374227888241921e-06, + "loss": 0.5779, + "step": 4669 + }, + { + "epoch": 3.8663721700717835, + "grad_norm": 0.9410127997398376, + "learning_rate": 1.4364274760789143e-06, + "loss": 0.5448, + "step": 4670 + }, + { + "epoch": 3.867200441744892, + "grad_norm": 0.9276284575462341, + "learning_rate": 1.4354323691250048e-06, + "loss": 0.5764, + "step": 4671 + }, + { + "epoch": 3.868028713418001, + "grad_norm": 0.9957787394523621, + "learning_rate": 1.4344374681550089e-06, + "loss": 0.5819, + "step": 4672 + }, + { + "epoch": 3.86885698509111, + "grad_norm": 0.9688808917999268, + "learning_rate": 1.4334427733614287e-06, + "loss": 0.5565, + "step": 4673 + }, + { + "epoch": 3.869685256764219, + "grad_norm": 0.9403797388076782, + "learning_rate": 1.4324482849367304e-06, + "loss": 0.5538, + "step": 4674 + }, + { + "epoch": 3.8705135284373275, + "grad_norm": 0.9486374855041504, + "learning_rate": 1.4314540030733364e-06, + "loss": 0.5529, + "step": 4675 + }, + { + "epoch": 3.8713418001104363, + "grad_norm": 0.9329063296318054, + "learning_rate": 1.4304599279636333e-06, + "loss": 0.5656, + "step": 4676 + }, + { + "epoch": 3.872170071783545, + "grad_norm": 0.9217836856842041, + "learning_rate": 1.4294660597999627e-06, + "loss": 0.5557, + "step": 4677 + }, + { + "epoch": 3.8729983434566537, + "grad_norm": 0.926406741142273, + "learning_rate": 1.4284723987746305e-06, + "loss": 0.5532, + "step": 4678 + }, + { + "epoch": 3.8738266151297625, + "grad_norm": 0.9605270624160767, + "learning_rate": 1.427478945079901e-06, + "loss": 0.5652, + "step": 4679 + }, + { + "epoch": 3.874654886802871, + "grad_norm": 0.9510946273803711, + "learning_rate": 1.4264856989079973e-06, + "loss": 0.5618, + "step": 4680 + }, + { + "epoch": 3.87548315847598, + "grad_norm": 0.9574974775314331, + "learning_rate": 1.4254926604511033e-06, + "loss": 0.5822, + "step": 4681 + }, + { + "epoch": 3.876311430149089, + "grad_norm": 0.9598250985145569, + "learning_rate": 1.424499829901364e-06, + "loss": 0.5537, + "step": 4682 + }, + { + "epoch": 3.877139701822198, + "grad_norm": 0.9691941738128662, + "learning_rate": 1.4235072074508822e-06, + "loss": 0.568, + "step": 4683 + }, + { + "epoch": 3.8779679734953065, + "grad_norm": 0.9728984236717224, + "learning_rate": 1.4225147932917193e-06, + "loss": 0.567, + "step": 4684 + }, + { + "epoch": 3.8787962451684153, + "grad_norm": 0.921758234500885, + "learning_rate": 1.4215225876159003e-06, + "loss": 0.5587, + "step": 4685 + }, + { + "epoch": 3.879624516841524, + "grad_norm": 0.9491950869560242, + "learning_rate": 1.420530590615408e-06, + "loss": 0.5479, + "step": 4686 + }, + { + "epoch": 3.8804527885146327, + "grad_norm": 0.9212395548820496, + "learning_rate": 1.4195388024821827e-06, + "loss": 0.5623, + "step": 4687 + }, + { + "epoch": 3.8812810601877414, + "grad_norm": 0.970206081867218, + "learning_rate": 1.418547223408127e-06, + "loss": 0.571, + "step": 4688 + }, + { + "epoch": 3.8821093318608506, + "grad_norm": 0.9378070831298828, + "learning_rate": 1.4175558535851037e-06, + "loss": 0.5823, + "step": 4689 + }, + { + "epoch": 3.882937603533959, + "grad_norm": 0.9203671813011169, + "learning_rate": 1.416564693204931e-06, + "loss": 0.5705, + "step": 4690 + }, + { + "epoch": 3.883765875207068, + "grad_norm": 0.9873926043510437, + "learning_rate": 1.4155737424593907e-06, + "loss": 0.5612, + "step": 4691 + }, + { + "epoch": 3.884594146880177, + "grad_norm": 0.9374533295631409, + "learning_rate": 1.4145830015402235e-06, + "loss": 0.5808, + "step": 4692 + }, + { + "epoch": 3.8854224185532855, + "grad_norm": 0.9469799399375916, + "learning_rate": 1.413592470639126e-06, + "loss": 0.5673, + "step": 4693 + }, + { + "epoch": 3.8862506902263942, + "grad_norm": 0.9671788811683655, + "learning_rate": 1.4126021499477592e-06, + "loss": 0.5755, + "step": 4694 + }, + { + "epoch": 3.887078961899503, + "grad_norm": 0.9435954689979553, + "learning_rate": 1.4116120396577398e-06, + "loss": 0.5686, + "step": 4695 + }, + { + "epoch": 3.8879072335726117, + "grad_norm": 0.9866729378700256, + "learning_rate": 1.410622139960644e-06, + "loss": 0.5456, + "step": 4696 + }, + { + "epoch": 3.8887355052457204, + "grad_norm": 0.9504481554031372, + "learning_rate": 1.409632451048009e-06, + "loss": 0.5779, + "step": 4697 + }, + { + "epoch": 3.8895637769188296, + "grad_norm": 0.9329410195350647, + "learning_rate": 1.4086429731113316e-06, + "loss": 0.5625, + "step": 4698 + }, + { + "epoch": 3.890392048591938, + "grad_norm": 0.9381130933761597, + "learning_rate": 1.4076537063420642e-06, + "loss": 0.5905, + "step": 4699 + }, + { + "epoch": 3.891220320265047, + "grad_norm": 0.9857736229896545, + "learning_rate": 1.4066646509316222e-06, + "loss": 0.5887, + "step": 4700 + }, + { + "epoch": 3.8920485919381558, + "grad_norm": 0.9994931817054749, + "learning_rate": 1.4056758070713788e-06, + "loss": 0.5691, + "step": 4701 + }, + { + "epoch": 3.8928768636112645, + "grad_norm": 0.9347181916236877, + "learning_rate": 1.404687174952665e-06, + "loss": 0.5806, + "step": 4702 + }, + { + "epoch": 3.8937051352843732, + "grad_norm": 0.9398673176765442, + "learning_rate": 1.4036987547667725e-06, + "loss": 0.5626, + "step": 4703 + }, + { + "epoch": 3.894533406957482, + "grad_norm": 0.9418836832046509, + "learning_rate": 1.4027105467049521e-06, + "loss": 0.5514, + "step": 4704 + }, + { + "epoch": 3.8953616786305907, + "grad_norm": 0.9223387241363525, + "learning_rate": 1.401722550958411e-06, + "loss": 0.5374, + "step": 4705 + }, + { + "epoch": 3.8961899503036994, + "grad_norm": 0.9222207069396973, + "learning_rate": 1.4007347677183197e-06, + "loss": 0.5535, + "step": 4706 + }, + { + "epoch": 3.8970182219768086, + "grad_norm": 0.9431470036506653, + "learning_rate": 1.3997471971758022e-06, + "loss": 0.5727, + "step": 4707 + }, + { + "epoch": 3.8978464936499173, + "grad_norm": 0.9395588040351868, + "learning_rate": 1.398759839521947e-06, + "loss": 0.5484, + "step": 4708 + }, + { + "epoch": 3.898674765323026, + "grad_norm": 0.9754805564880371, + "learning_rate": 1.397772694947796e-06, + "loss": 0.5617, + "step": 4709 + }, + { + "epoch": 3.8995030369961348, + "grad_norm": 0.9596531987190247, + "learning_rate": 1.3967857636443539e-06, + "loss": 0.5472, + "step": 4710 + }, + { + "epoch": 3.9003313086692435, + "grad_norm": 0.961883008480072, + "learning_rate": 1.3957990458025833e-06, + "loss": 0.5697, + "step": 4711 + }, + { + "epoch": 3.901159580342352, + "grad_norm": 0.9568914175033569, + "learning_rate": 1.3948125416134029e-06, + "loss": 0.56, + "step": 4712 + }, + { + "epoch": 3.901987852015461, + "grad_norm": 0.9424316883087158, + "learning_rate": 1.3938262512676937e-06, + "loss": 0.5485, + "step": 4713 + }, + { + "epoch": 3.90281612368857, + "grad_norm": 0.9500327706336975, + "learning_rate": 1.392840174956294e-06, + "loss": 0.5748, + "step": 4714 + }, + { + "epoch": 3.9036443953616784, + "grad_norm": 0.9448386430740356, + "learning_rate": 1.3918543128699986e-06, + "loss": 0.5407, + "step": 4715 + }, + { + "epoch": 3.9044726670347876, + "grad_norm": 0.9605383276939392, + "learning_rate": 1.390868665199564e-06, + "loss": 0.5728, + "step": 4716 + }, + { + "epoch": 3.9053009387078963, + "grad_norm": 0.9455352425575256, + "learning_rate": 1.389883232135705e-06, + "loss": 0.5506, + "step": 4717 + }, + { + "epoch": 3.906129210381005, + "grad_norm": 0.9913429617881775, + "learning_rate": 1.3888980138690906e-06, + "loss": 0.5536, + "step": 4718 + }, + { + "epoch": 3.9069574820541138, + "grad_norm": 0.9688431024551392, + "learning_rate": 1.3879130105903526e-06, + "loss": 0.555, + "step": 4719 + }, + { + "epoch": 3.9077857537272225, + "grad_norm": 0.9419496655464172, + "learning_rate": 1.3869282224900815e-06, + "loss": 0.5727, + "step": 4720 + }, + { + "epoch": 3.908614025400331, + "grad_norm": 0.9332079291343689, + "learning_rate": 1.3859436497588225e-06, + "loss": 0.572, + "step": 4721 + }, + { + "epoch": 3.90944229707344, + "grad_norm": 0.952717125415802, + "learning_rate": 1.3849592925870821e-06, + "loss": 0.56, + "step": 4722 + }, + { + "epoch": 3.910270568746549, + "grad_norm": 0.944283127784729, + "learning_rate": 1.3839751511653253e-06, + "loss": 0.5803, + "step": 4723 + }, + { + "epoch": 3.9110988404196574, + "grad_norm": 0.9370842576026917, + "learning_rate": 1.3829912256839725e-06, + "loss": 0.5802, + "step": 4724 + }, + { + "epoch": 3.9119271120927666, + "grad_norm": 0.9332932829856873, + "learning_rate": 1.3820075163334047e-06, + "loss": 0.5648, + "step": 4725 + }, + { + "epoch": 3.9127553837658753, + "grad_norm": 0.9598782658576965, + "learning_rate": 1.3810240233039618e-06, + "loss": 0.5565, + "step": 4726 + }, + { + "epoch": 3.913583655438984, + "grad_norm": 0.9564656019210815, + "learning_rate": 1.3800407467859383e-06, + "loss": 0.5594, + "step": 4727 + }, + { + "epoch": 3.9144119271120927, + "grad_norm": 0.9756644368171692, + "learning_rate": 1.3790576869695913e-06, + "loss": 0.5425, + "step": 4728 + }, + { + "epoch": 3.9152401987852015, + "grad_norm": 0.9616433382034302, + "learning_rate": 1.3780748440451314e-06, + "loss": 0.5802, + "step": 4729 + }, + { + "epoch": 3.91606847045831, + "grad_norm": 0.9436020851135254, + "learning_rate": 1.3770922182027319e-06, + "loss": 0.5554, + "step": 4730 + }, + { + "epoch": 3.916896742131419, + "grad_norm": 0.9598777890205383, + "learning_rate": 1.3761098096325195e-06, + "loss": 0.5552, + "step": 4731 + }, + { + "epoch": 3.917725013804528, + "grad_norm": 0.9497062563896179, + "learning_rate": 1.3751276185245821e-06, + "loss": 0.5612, + "step": 4732 + }, + { + "epoch": 3.918553285477637, + "grad_norm": 0.9690037369728088, + "learning_rate": 1.3741456450689656e-06, + "loss": 0.5682, + "step": 4733 + }, + { + "epoch": 3.9193815571507455, + "grad_norm": 0.9524545073509216, + "learning_rate": 1.373163889455671e-06, + "loss": 0.5795, + "step": 4734 + }, + { + "epoch": 3.9202098288238543, + "grad_norm": 0.9572216272354126, + "learning_rate": 1.3721823518746591e-06, + "loss": 0.558, + "step": 4735 + }, + { + "epoch": 3.921038100496963, + "grad_norm": 0.9508054256439209, + "learning_rate": 1.3712010325158499e-06, + "loss": 0.5578, + "step": 4736 + }, + { + "epoch": 3.9218663721700717, + "grad_norm": 0.9361369013786316, + "learning_rate": 1.3702199315691176e-06, + "loss": 0.5661, + "step": 4737 + }, + { + "epoch": 3.9226946438431805, + "grad_norm": 0.9344385266304016, + "learning_rate": 1.3692390492242973e-06, + "loss": 0.571, + "step": 4738 + }, + { + "epoch": 3.923522915516289, + "grad_norm": 0.9559999704360962, + "learning_rate": 1.3682583856711807e-06, + "loss": 0.5441, + "step": 4739 + }, + { + "epoch": 3.924351187189398, + "grad_norm": 0.9483859539031982, + "learning_rate": 1.3672779410995155e-06, + "loss": 0.5645, + "step": 4740 + }, + { + "epoch": 3.925179458862507, + "grad_norm": 0.9504157304763794, + "learning_rate": 1.3662977156990093e-06, + "loss": 0.5512, + "step": 4741 + }, + { + "epoch": 3.926007730535616, + "grad_norm": 0.9570707678794861, + "learning_rate": 1.3653177096593282e-06, + "loss": 0.5722, + "step": 4742 + }, + { + "epoch": 3.9268360022087245, + "grad_norm": 0.9542663097381592, + "learning_rate": 1.3643379231700915e-06, + "loss": 0.5385, + "step": 4743 + }, + { + "epoch": 3.9276642738818333, + "grad_norm": 0.9425368309020996, + "learning_rate": 1.3633583564208808e-06, + "loss": 0.547, + "step": 4744 + }, + { + "epoch": 3.928492545554942, + "grad_norm": 0.9227926731109619, + "learning_rate": 1.362379009601233e-06, + "loss": 0.5664, + "step": 4745 + }, + { + "epoch": 3.9293208172280507, + "grad_norm": 0.9677916765213013, + "learning_rate": 1.3613998829006408e-06, + "loss": 0.5622, + "step": 4746 + }, + { + "epoch": 3.9301490889011594, + "grad_norm": 0.9727103114128113, + "learning_rate": 1.360420976508558e-06, + "loss": 0.5794, + "step": 4747 + }, + { + "epoch": 3.9309773605742686, + "grad_norm": 0.930270254611969, + "learning_rate": 1.359442290614394e-06, + "loss": 0.573, + "step": 4748 + }, + { + "epoch": 3.931805632247377, + "grad_norm": 1.0135208368301392, + "learning_rate": 1.3584638254075132e-06, + "loss": 0.5741, + "step": 4749 + }, + { + "epoch": 3.932633903920486, + "grad_norm": 0.992636501789093, + "learning_rate": 1.3574855810772414e-06, + "loss": 0.5592, + "step": 4750 + }, + { + "epoch": 3.933462175593595, + "grad_norm": 0.9497395157814026, + "learning_rate": 1.3565075578128603e-06, + "loss": 0.5579, + "step": 4751 + }, + { + "epoch": 3.9342904472667035, + "grad_norm": 0.9188456535339355, + "learning_rate": 1.355529755803607e-06, + "loss": 0.573, + "step": 4752 + }, + { + "epoch": 3.9351187189398122, + "grad_norm": 0.9138023257255554, + "learning_rate": 1.3545521752386765e-06, + "loss": 0.5895, + "step": 4753 + }, + { + "epoch": 3.935946990612921, + "grad_norm": 0.9659999012947083, + "learning_rate": 1.3535748163072233e-06, + "loss": 0.5506, + "step": 4754 + }, + { + "epoch": 3.9367752622860297, + "grad_norm": 0.9442874789237976, + "learning_rate": 1.3525976791983553e-06, + "loss": 0.5679, + "step": 4755 + }, + { + "epoch": 3.9376035339591384, + "grad_norm": 0.9444726705551147, + "learning_rate": 1.3516207641011407e-06, + "loss": 0.5795, + "step": 4756 + }, + { + "epoch": 3.9384318056322476, + "grad_norm": 0.9704516530036926, + "learning_rate": 1.3506440712046037e-06, + "loss": 0.568, + "step": 4757 + }, + { + "epoch": 3.9392600773053563, + "grad_norm": 0.9629958271980286, + "learning_rate": 1.349667600697724e-06, + "loss": 0.5967, + "step": 4758 + }, + { + "epoch": 3.940088348978465, + "grad_norm": 0.9573612213134766, + "learning_rate": 1.3486913527694406e-06, + "loss": 0.5553, + "step": 4759 + }, + { + "epoch": 3.9409166206515738, + "grad_norm": 0.9458691477775574, + "learning_rate": 1.3477153276086485e-06, + "loss": 0.5768, + "step": 4760 + }, + { + "epoch": 3.9417448923246825, + "grad_norm": 0.938666820526123, + "learning_rate": 1.3467395254041985e-06, + "loss": 0.5465, + "step": 4761 + }, + { + "epoch": 3.9425731639977912, + "grad_norm": 0.9396777153015137, + "learning_rate": 1.3457639463449008e-06, + "loss": 0.5699, + "step": 4762 + }, + { + "epoch": 3.9434014356709, + "grad_norm": 0.9707586765289307, + "learning_rate": 1.3447885906195186e-06, + "loss": 0.5472, + "step": 4763 + }, + { + "epoch": 3.9442297073440087, + "grad_norm": 0.9531437754631042, + "learning_rate": 1.343813458416776e-06, + "loss": 0.5712, + "step": 4764 + }, + { + "epoch": 3.9450579790171174, + "grad_norm": 0.942042350769043, + "learning_rate": 1.342838549925351e-06, + "loss": 0.5494, + "step": 4765 + }, + { + "epoch": 3.9458862506902266, + "grad_norm": 0.9602234959602356, + "learning_rate": 1.3418638653338795e-06, + "loss": 0.5627, + "step": 4766 + }, + { + "epoch": 3.9467145223633353, + "grad_norm": 0.9987154006958008, + "learning_rate": 1.340889404830955e-06, + "loss": 0.5824, + "step": 4767 + }, + { + "epoch": 3.947542794036444, + "grad_norm": 0.9674735069274902, + "learning_rate": 1.3399151686051246e-06, + "loss": 0.5772, + "step": 4768 + }, + { + "epoch": 3.9483710657095528, + "grad_norm": 0.9081838130950928, + "learning_rate": 1.3389411568448948e-06, + "loss": 0.555, + "step": 4769 + }, + { + "epoch": 3.9491993373826615, + "grad_norm": 0.9668666124343872, + "learning_rate": 1.337967369738729e-06, + "loss": 0.5655, + "step": 4770 + }, + { + "epoch": 3.95002760905577, + "grad_norm": 0.9260414838790894, + "learning_rate": 1.3369938074750438e-06, + "loss": 0.5685, + "step": 4771 + }, + { + "epoch": 3.950855880728879, + "grad_norm": 0.9354709386825562, + "learning_rate": 1.3360204702422154e-06, + "loss": 0.5623, + "step": 4772 + }, + { + "epoch": 3.951684152401988, + "grad_norm": 0.965032696723938, + "learning_rate": 1.3350473582285777e-06, + "loss": 0.5846, + "step": 4773 + }, + { + "epoch": 3.9525124240750964, + "grad_norm": 0.9616831541061401, + "learning_rate": 1.3340744716224143e-06, + "loss": 0.5576, + "step": 4774 + }, + { + "epoch": 3.9533406957482056, + "grad_norm": 0.9441544413566589, + "learning_rate": 1.3331018106119726e-06, + "loss": 0.5918, + "step": 4775 + }, + { + "epoch": 3.9541689674213143, + "grad_norm": 0.9979905486106873, + "learning_rate": 1.332129375385453e-06, + "loss": 0.5539, + "step": 4776 + }, + { + "epoch": 3.954997239094423, + "grad_norm": 0.950507402420044, + "learning_rate": 1.3311571661310124e-06, + "loss": 0.5463, + "step": 4777 + }, + { + "epoch": 3.9558255107675317, + "grad_norm": 0.9438148140907288, + "learning_rate": 1.3301851830367637e-06, + "loss": 0.5645, + "step": 4778 + }, + { + "epoch": 3.9566537824406405, + "grad_norm": 0.9625552296638489, + "learning_rate": 1.3292134262907785e-06, + "loss": 0.5782, + "step": 4779 + }, + { + "epoch": 3.957482054113749, + "grad_norm": 0.958839476108551, + "learning_rate": 1.3282418960810803e-06, + "loss": 0.562, + "step": 4780 + }, + { + "epoch": 3.958310325786858, + "grad_norm": 0.9578886032104492, + "learning_rate": 1.3272705925956522e-06, + "loss": 0.5609, + "step": 4781 + }, + { + "epoch": 3.959138597459967, + "grad_norm": 0.9848271608352661, + "learning_rate": 1.3262995160224335e-06, + "loss": 0.5627, + "step": 4782 + }, + { + "epoch": 3.9599668691330754, + "grad_norm": 0.9630492925643921, + "learning_rate": 1.3253286665493164e-06, + "loss": 0.5356, + "step": 4783 + }, + { + "epoch": 3.9607951408061846, + "grad_norm": 0.945763111114502, + "learning_rate": 1.3243580443641518e-06, + "loss": 0.5688, + "step": 4784 + }, + { + "epoch": 3.9616234124792933, + "grad_norm": 0.9506339430809021, + "learning_rate": 1.3233876496547477e-06, + "loss": 0.5725, + "step": 4785 + }, + { + "epoch": 3.962451684152402, + "grad_norm": 0.9367161393165588, + "learning_rate": 1.3224174826088654e-06, + "loss": 0.5605, + "step": 4786 + }, + { + "epoch": 3.9632799558255107, + "grad_norm": 0.9559187889099121, + "learning_rate": 1.3214475434142215e-06, + "loss": 0.5606, + "step": 4787 + }, + { + "epoch": 3.9641082274986195, + "grad_norm": 0.9559148550033569, + "learning_rate": 1.320477832258492e-06, + "loss": 0.5516, + "step": 4788 + }, + { + "epoch": 3.964936499171728, + "grad_norm": 0.9415640234947205, + "learning_rate": 1.319508349329307e-06, + "loss": 0.5607, + "step": 4789 + }, + { + "epoch": 3.965764770844837, + "grad_norm": 0.961673378944397, + "learning_rate": 1.3185390948142516e-06, + "loss": 0.571, + "step": 4790 + }, + { + "epoch": 3.966593042517946, + "grad_norm": 0.9346276521682739, + "learning_rate": 1.3175700689008678e-06, + "loss": 0.5716, + "step": 4791 + }, + { + "epoch": 3.967421314191055, + "grad_norm": 0.9483547806739807, + "learning_rate": 1.3166012717766541e-06, + "loss": 0.5654, + "step": 4792 + }, + { + "epoch": 3.9682495858641635, + "grad_norm": 0.9249748587608337, + "learning_rate": 1.3156327036290616e-06, + "loss": 0.5669, + "step": 4793 + }, + { + "epoch": 3.9690778575372723, + "grad_norm": 0.965886116027832, + "learning_rate": 1.3146643646455007e-06, + "loss": 0.5739, + "step": 4794 + }, + { + "epoch": 3.969906129210381, + "grad_norm": 0.9492125511169434, + "learning_rate": 1.3136962550133364e-06, + "loss": 0.5521, + "step": 4795 + }, + { + "epoch": 3.9707344008834897, + "grad_norm": 0.9473196864128113, + "learning_rate": 1.3127283749198885e-06, + "loss": 0.5412, + "step": 4796 + }, + { + "epoch": 3.9715626725565985, + "grad_norm": 0.9501288533210754, + "learning_rate": 1.3117607245524314e-06, + "loss": 0.5451, + "step": 4797 + }, + { + "epoch": 3.9723909442297076, + "grad_norm": 0.9439449310302734, + "learning_rate": 1.3107933040981985e-06, + "loss": 0.5702, + "step": 4798 + }, + { + "epoch": 3.973219215902816, + "grad_norm": 0.9604007005691528, + "learning_rate": 1.3098261137443744e-06, + "loss": 0.5674, + "step": 4799 + }, + { + "epoch": 3.974047487575925, + "grad_norm": 0.9559398889541626, + "learning_rate": 1.3088591536781026e-06, + "loss": 0.5618, + "step": 4800 + }, + { + "epoch": 3.974875759249034, + "grad_norm": 0.93898606300354, + "learning_rate": 1.3078924240864821e-06, + "loss": 0.5539, + "step": 4801 + }, + { + "epoch": 3.9757040309221425, + "grad_norm": 0.985381007194519, + "learning_rate": 1.306925925156564e-06, + "loss": 0.5748, + "step": 4802 + }, + { + "epoch": 3.9765323025952513, + "grad_norm": 0.9823545217514038, + "learning_rate": 1.3059596570753575e-06, + "loss": 0.569, + "step": 4803 + }, + { + "epoch": 3.97736057426836, + "grad_norm": 0.9789100289344788, + "learning_rate": 1.3049936200298274e-06, + "loss": 0.5784, + "step": 4804 + }, + { + "epoch": 3.9781888459414687, + "grad_norm": 0.9664195775985718, + "learning_rate": 1.3040278142068912e-06, + "loss": 0.574, + "step": 4805 + }, + { + "epoch": 3.9790171176145774, + "grad_norm": 0.9816479682922363, + "learning_rate": 1.3030622397934244e-06, + "loss": 0.607, + "step": 4806 + }, + { + "epoch": 3.9798453892876866, + "grad_norm": 0.9329001903533936, + "learning_rate": 1.3020968969762577e-06, + "loss": 0.5666, + "step": 4807 + }, + { + "epoch": 3.980673660960795, + "grad_norm": 0.9252684116363525, + "learning_rate": 1.3011317859421744e-06, + "loss": 0.5528, + "step": 4808 + }, + { + "epoch": 3.981501932633904, + "grad_norm": 0.9480351209640503, + "learning_rate": 1.3001669068779137e-06, + "loss": 0.5685, + "step": 4809 + }, + { + "epoch": 3.982330204307013, + "grad_norm": 0.9316738247871399, + "learning_rate": 1.2992022599701723e-06, + "loss": 0.569, + "step": 4810 + }, + { + "epoch": 3.9831584759801215, + "grad_norm": 0.9480258822441101, + "learning_rate": 1.2982378454056005e-06, + "loss": 0.5519, + "step": 4811 + }, + { + "epoch": 3.9839867476532302, + "grad_norm": 0.9391124248504639, + "learning_rate": 1.2972736633708021e-06, + "loss": 0.5657, + "step": 4812 + }, + { + "epoch": 3.984815019326339, + "grad_norm": 0.9289792776107788, + "learning_rate": 1.2963097140523389e-06, + "loss": 0.5258, + "step": 4813 + }, + { + "epoch": 3.9856432909994477, + "grad_norm": 0.9436705708503723, + "learning_rate": 1.295345997636725e-06, + "loss": 0.5672, + "step": 4814 + }, + { + "epoch": 3.9864715626725564, + "grad_norm": 0.9601346850395203, + "learning_rate": 1.2943825143104305e-06, + "loss": 0.5628, + "step": 4815 + }, + { + "epoch": 3.9872998343456656, + "grad_norm": 0.9829810261726379, + "learning_rate": 1.2934192642598818e-06, + "loss": 0.5772, + "step": 4816 + }, + { + "epoch": 3.9881281060187743, + "grad_norm": 0.9281584620475769, + "learning_rate": 1.292456247671457e-06, + "loss": 0.5763, + "step": 4817 + }, + { + "epoch": 3.988956377691883, + "grad_norm": 0.9712902903556824, + "learning_rate": 1.2914934647314914e-06, + "loss": 0.5553, + "step": 4818 + }, + { + "epoch": 3.9897846493649918, + "grad_norm": 0.9826544523239136, + "learning_rate": 1.290530915626276e-06, + "loss": 0.5802, + "step": 4819 + }, + { + "epoch": 3.9906129210381005, + "grad_norm": 0.9495535492897034, + "learning_rate": 1.289568600542054e-06, + "loss": 0.5746, + "step": 4820 + }, + { + "epoch": 3.9914411927112092, + "grad_norm": 0.9266762137413025, + "learning_rate": 1.2886065196650232e-06, + "loss": 0.5825, + "step": 4821 + }, + { + "epoch": 3.992269464384318, + "grad_norm": 0.946962833404541, + "learning_rate": 1.2876446731813387e-06, + "loss": 0.5825, + "step": 4822 + }, + { + "epoch": 3.9930977360574267, + "grad_norm": 0.9361686110496521, + "learning_rate": 1.2866830612771091e-06, + "loss": 0.5473, + "step": 4823 + }, + { + "epoch": 3.9939260077305354, + "grad_norm": 0.9517462849617004, + "learning_rate": 1.2857216841383958e-06, + "loss": 0.5628, + "step": 4824 + }, + { + "epoch": 3.9947542794036446, + "grad_norm": 0.957137405872345, + "learning_rate": 1.2847605419512174e-06, + "loss": 0.5516, + "step": 4825 + }, + { + "epoch": 3.9955825510767533, + "grad_norm": 0.9692016243934631, + "learning_rate": 1.2837996349015465e-06, + "loss": 0.5738, + "step": 4826 + }, + { + "epoch": 3.996410822749862, + "grad_norm": 0.9570938348770142, + "learning_rate": 1.2828389631753085e-06, + "loss": 0.5971, + "step": 4827 + }, + { + "epoch": 3.9972390944229708, + "grad_norm": 0.9742267727851868, + "learning_rate": 1.2818785269583845e-06, + "loss": 0.5721, + "step": 4828 + }, + { + "epoch": 3.9980673660960795, + "grad_norm": 0.9525173902511597, + "learning_rate": 1.2809183264366115e-06, + "loss": 0.5548, + "step": 4829 + }, + { + "epoch": 3.998895637769188, + "grad_norm": 0.9499869346618652, + "learning_rate": 1.279958361795778e-06, + "loss": 0.5825, + "step": 4830 + }, + { + "epoch": 3.999723909442297, + "grad_norm": 0.9081862568855286, + "learning_rate": 1.2789986332216276e-06, + "loss": 0.5835, + "step": 4831 + }, + { + "epoch": 4.0, + "grad_norm": 0.9081862568855286, + "learning_rate": 1.2780391408998607e-06, + "loss": 0.5353, + "step": 4832 + }, + { + "epoch": 4.000828271673109, + "grad_norm": 1.8408721685409546, + "learning_rate": 1.2770798850161281e-06, + "loss": 0.544, + "step": 4833 + }, + { + "epoch": 4.0016565433462175, + "grad_norm": 0.950079083442688, + "learning_rate": 1.2761208657560376e-06, + "loss": 0.5485, + "step": 4834 + }, + { + "epoch": 4.002484815019327, + "grad_norm": 0.9583503603935242, + "learning_rate": 1.275162083305152e-06, + "loss": 0.5657, + "step": 4835 + }, + { + "epoch": 4.003313086692435, + "grad_norm": 0.9331668615341187, + "learning_rate": 1.2742035378489841e-06, + "loss": 0.5673, + "step": 4836 + }, + { + "epoch": 4.004141358365544, + "grad_norm": 0.9338731169700623, + "learning_rate": 1.2732452295730053e-06, + "loss": 0.5333, + "step": 4837 + }, + { + "epoch": 4.004969630038652, + "grad_norm": 0.9427415132522583, + "learning_rate": 1.2722871586626395e-06, + "loss": 0.5584, + "step": 4838 + }, + { + "epoch": 4.0057979017117615, + "grad_norm": 0.9849832653999329, + "learning_rate": 1.2713293253032632e-06, + "loss": 0.5486, + "step": 4839 + }, + { + "epoch": 4.00662617338487, + "grad_norm": 0.9528194069862366, + "learning_rate": 1.270371729680209e-06, + "loss": 0.5574, + "step": 4840 + }, + { + "epoch": 4.007454445057979, + "grad_norm": 0.9479176998138428, + "learning_rate": 1.2694143719787633e-06, + "loss": 0.5601, + "step": 4841 + }, + { + "epoch": 4.008282716731088, + "grad_norm": 0.9426152110099792, + "learning_rate": 1.2684572523841654e-06, + "loss": 0.5599, + "step": 4842 + }, + { + "epoch": 4.009110988404196, + "grad_norm": 0.9592022895812988, + "learning_rate": 1.2675003710816081e-06, + "loss": 0.5579, + "step": 4843 + }, + { + "epoch": 4.009939260077306, + "grad_norm": 0.9946890473365784, + "learning_rate": 1.2665437282562398e-06, + "loss": 0.5443, + "step": 4844 + }, + { + "epoch": 4.010767531750414, + "grad_norm": 0.9415493011474609, + "learning_rate": 1.2655873240931626e-06, + "loss": 0.5685, + "step": 4845 + }, + { + "epoch": 4.011595803423523, + "grad_norm": 0.9373499155044556, + "learning_rate": 1.2646311587774304e-06, + "loss": 0.5609, + "step": 4846 + }, + { + "epoch": 4.012424075096631, + "grad_norm": 0.9726147651672363, + "learning_rate": 1.263675232494053e-06, + "loss": 0.571, + "step": 4847 + }, + { + "epoch": 4.0132523467697405, + "grad_norm": 0.9378729462623596, + "learning_rate": 1.262719545427994e-06, + "loss": 0.5653, + "step": 4848 + }, + { + "epoch": 4.01408061844285, + "grad_norm": 0.9564341902732849, + "learning_rate": 1.2617640977641684e-06, + "loss": 0.5502, + "step": 4849 + }, + { + "epoch": 4.014908890115958, + "grad_norm": 0.9591152667999268, + "learning_rate": 1.260808889687447e-06, + "loss": 0.551, + "step": 4850 + }, + { + "epoch": 4.015737161789067, + "grad_norm": 0.9530390501022339, + "learning_rate": 1.259853921382655e-06, + "loss": 0.5297, + "step": 4851 + }, + { + "epoch": 4.016565433462175, + "grad_norm": 0.9545491933822632, + "learning_rate": 1.258899193034569e-06, + "loss": 0.5399, + "step": 4852 + }, + { + "epoch": 4.017393705135285, + "grad_norm": 0.9423142671585083, + "learning_rate": 1.2579447048279184e-06, + "loss": 0.584, + "step": 4853 + }, + { + "epoch": 4.018221976808393, + "grad_norm": 0.933307945728302, + "learning_rate": 1.2569904569473901e-06, + "loss": 0.5463, + "step": 4854 + }, + { + "epoch": 4.019050248481502, + "grad_norm": 0.9511076211929321, + "learning_rate": 1.2560364495776202e-06, + "loss": 0.5652, + "step": 4855 + }, + { + "epoch": 4.01987852015461, + "grad_norm": 0.9643457531929016, + "learning_rate": 1.2550826829032015e-06, + "loss": 0.5329, + "step": 4856 + }, + { + "epoch": 4.0207067918277195, + "grad_norm": 0.9959326386451721, + "learning_rate": 1.2541291571086794e-06, + "loss": 0.5545, + "step": 4857 + }, + { + "epoch": 4.021535063500829, + "grad_norm": 0.9828082323074341, + "learning_rate": 1.2531758723785505e-06, + "loss": 0.5396, + "step": 4858 + }, + { + "epoch": 4.022363335173937, + "grad_norm": 0.9460323452949524, + "learning_rate": 1.2522228288972676e-06, + "loss": 0.5334, + "step": 4859 + }, + { + "epoch": 4.023191606847046, + "grad_norm": 0.9510787725448608, + "learning_rate": 1.251270026849237e-06, + "loss": 0.5765, + "step": 4860 + }, + { + "epoch": 4.024019878520154, + "grad_norm": 0.9397110342979431, + "learning_rate": 1.2503174664188145e-06, + "loss": 0.52, + "step": 4861 + }, + { + "epoch": 4.024848150193264, + "grad_norm": 0.9416722655296326, + "learning_rate": 1.2493651477903126e-06, + "loss": 0.5497, + "step": 4862 + }, + { + "epoch": 4.025676421866372, + "grad_norm": 0.9606363773345947, + "learning_rate": 1.2484130711479977e-06, + "loss": 0.5442, + "step": 4863 + }, + { + "epoch": 4.026504693539481, + "grad_norm": 0.9509909749031067, + "learning_rate": 1.2474612366760861e-06, + "loss": 0.5768, + "step": 4864 + }, + { + "epoch": 4.027332965212589, + "grad_norm": 0.9398588538169861, + "learning_rate": 1.2465096445587486e-06, + "loss": 0.5337, + "step": 4865 + }, + { + "epoch": 4.0281612368856985, + "grad_norm": 1.0803090333938599, + "learning_rate": 1.2455582949801095e-06, + "loss": 0.5437, + "step": 4866 + }, + { + "epoch": 4.028989508558808, + "grad_norm": 0.9432228207588196, + "learning_rate": 1.2446071881242478e-06, + "loss": 0.5439, + "step": 4867 + }, + { + "epoch": 4.029817780231916, + "grad_norm": 0.962143063545227, + "learning_rate": 1.2436563241751914e-06, + "loss": 0.5444, + "step": 4868 + }, + { + "epoch": 4.030646051905025, + "grad_norm": 0.9307393431663513, + "learning_rate": 1.2427057033169252e-06, + "loss": 0.5684, + "step": 4869 + }, + { + "epoch": 4.031474323578133, + "grad_norm": 0.9873812198638916, + "learning_rate": 1.2417553257333856e-06, + "loss": 0.5602, + "step": 4870 + }, + { + "epoch": 4.032302595251243, + "grad_norm": 0.9589493870735168, + "learning_rate": 1.2408051916084603e-06, + "loss": 0.5257, + "step": 4871 + }, + { + "epoch": 4.033130866924351, + "grad_norm": 0.9406089186668396, + "learning_rate": 1.239855301125993e-06, + "loss": 0.5417, + "step": 4872 + }, + { + "epoch": 4.03395913859746, + "grad_norm": 0.9455104470252991, + "learning_rate": 1.2389056544697781e-06, + "loss": 0.52, + "step": 4873 + }, + { + "epoch": 4.034787410270568, + "grad_norm": 0.9494171142578125, + "learning_rate": 1.2379562518235628e-06, + "loss": 0.5469, + "step": 4874 + }, + { + "epoch": 4.0356156819436775, + "grad_norm": 0.9548773765563965, + "learning_rate": 1.2370070933710491e-06, + "loss": 0.5548, + "step": 4875 + }, + { + "epoch": 4.036443953616787, + "grad_norm": 0.9773174524307251, + "learning_rate": 1.2360581792958893e-06, + "loss": 0.5484, + "step": 4876 + }, + { + "epoch": 4.037272225289895, + "grad_norm": 0.9558678269386292, + "learning_rate": 1.2351095097816885e-06, + "loss": 0.5578, + "step": 4877 + }, + { + "epoch": 4.038100496963004, + "grad_norm": 1.1254467964172363, + "learning_rate": 1.2341610850120064e-06, + "loss": 0.5588, + "step": 4878 + }, + { + "epoch": 4.038928768636112, + "grad_norm": 0.9522780776023865, + "learning_rate": 1.2332129051703553e-06, + "loss": 0.5518, + "step": 4879 + }, + { + "epoch": 4.039757040309222, + "grad_norm": 0.9933323860168457, + "learning_rate": 1.2322649704401973e-06, + "loss": 0.556, + "step": 4880 + }, + { + "epoch": 4.04058531198233, + "grad_norm": 0.9555357098579407, + "learning_rate": 1.2313172810049497e-06, + "loss": 0.5678, + "step": 4881 + }, + { + "epoch": 4.041413583655439, + "grad_norm": 0.9392233490943909, + "learning_rate": 1.230369837047983e-06, + "loss": 0.5307, + "step": 4882 + }, + { + "epoch": 4.042241855328548, + "grad_norm": 0.9643675088882446, + "learning_rate": 1.2294226387526165e-06, + "loss": 0.5391, + "step": 4883 + }, + { + "epoch": 4.0430701270016565, + "grad_norm": 0.9979854226112366, + "learning_rate": 1.2284756863021249e-06, + "loss": 0.567, + "step": 4884 + }, + { + "epoch": 4.043898398674766, + "grad_norm": 0.9585314393043518, + "learning_rate": 1.227528979879736e-06, + "loss": 0.5315, + "step": 4885 + }, + { + "epoch": 4.044726670347874, + "grad_norm": 0.9538570642471313, + "learning_rate": 1.226582519668628e-06, + "loss": 0.5525, + "step": 4886 + }, + { + "epoch": 4.045554942020983, + "grad_norm": 0.9781275987625122, + "learning_rate": 1.2256363058519305e-06, + "loss": 0.5745, + "step": 4887 + }, + { + "epoch": 4.046383213694091, + "grad_norm": 0.9645776152610779, + "learning_rate": 1.2246903386127284e-06, + "loss": 0.5631, + "step": 4888 + }, + { + "epoch": 4.0472114853672005, + "grad_norm": 0.9528216123580933, + "learning_rate": 1.2237446181340587e-06, + "loss": 0.5697, + "step": 4889 + }, + { + "epoch": 4.048039757040309, + "grad_norm": 0.9601432085037231, + "learning_rate": 1.2227991445989072e-06, + "loss": 0.5333, + "step": 4890 + }, + { + "epoch": 4.048868028713418, + "grad_norm": 0.9265902638435364, + "learning_rate": 1.221853918190216e-06, + "loss": 0.5535, + "step": 4891 + }, + { + "epoch": 4.049696300386527, + "grad_norm": 0.940017819404602, + "learning_rate": 1.2209089390908762e-06, + "loss": 0.5493, + "step": 4892 + }, + { + "epoch": 4.0505245720596355, + "grad_norm": 0.935411274433136, + "learning_rate": 1.2199642074837334e-06, + "loss": 0.5763, + "step": 4893 + }, + { + "epoch": 4.051352843732745, + "grad_norm": 0.9772312045097351, + "learning_rate": 1.2190197235515846e-06, + "loss": 0.5326, + "step": 4894 + }, + { + "epoch": 4.052181115405853, + "grad_norm": 0.9477226138114929, + "learning_rate": 1.2180754874771774e-06, + "loss": 0.548, + "step": 4895 + }, + { + "epoch": 4.053009387078962, + "grad_norm": 0.9361460208892822, + "learning_rate": 1.2171314994432134e-06, + "loss": 0.5692, + "step": 4896 + }, + { + "epoch": 4.05383765875207, + "grad_norm": 0.9590089321136475, + "learning_rate": 1.2161877596323466e-06, + "loss": 0.5494, + "step": 4897 + }, + { + "epoch": 4.0546659304251795, + "grad_norm": 0.9448971152305603, + "learning_rate": 1.2152442682271807e-06, + "loss": 0.5623, + "step": 4898 + }, + { + "epoch": 4.055494202098288, + "grad_norm": 0.9581391215324402, + "learning_rate": 1.2143010254102716e-06, + "loss": 0.5525, + "step": 4899 + }, + { + "epoch": 4.056322473771397, + "grad_norm": 0.9311408996582031, + "learning_rate": 1.2133580313641289e-06, + "loss": 0.5317, + "step": 4900 + }, + { + "epoch": 4.057150745444506, + "grad_norm": 0.9715316295623779, + "learning_rate": 1.212415286271214e-06, + "loss": 0.5626, + "step": 4901 + }, + { + "epoch": 4.057979017117614, + "grad_norm": 0.9454675316810608, + "learning_rate": 1.2114727903139379e-06, + "loss": 0.567, + "step": 4902 + }, + { + "epoch": 4.058807288790724, + "grad_norm": 0.9326028227806091, + "learning_rate": 1.210530543674665e-06, + "loss": 0.5372, + "step": 4903 + }, + { + "epoch": 4.059635560463832, + "grad_norm": 0.9475375413894653, + "learning_rate": 1.2095885465357128e-06, + "loss": 0.5521, + "step": 4904 + }, + { + "epoch": 4.060463832136941, + "grad_norm": 0.9660394787788391, + "learning_rate": 1.2086467990793469e-06, + "loss": 0.5478, + "step": 4905 + }, + { + "epoch": 4.061292103810049, + "grad_norm": 0.9396241903305054, + "learning_rate": 1.2077053014877874e-06, + "loss": 0.5489, + "step": 4906 + }, + { + "epoch": 4.0621203754831585, + "grad_norm": 0.9579562544822693, + "learning_rate": 1.2067640539432063e-06, + "loss": 0.5455, + "step": 4907 + }, + { + "epoch": 4.062948647156268, + "grad_norm": 0.9574836492538452, + "learning_rate": 1.2058230566277246e-06, + "loss": 0.5425, + "step": 4908 + }, + { + "epoch": 4.063776918829376, + "grad_norm": 0.9654014706611633, + "learning_rate": 1.204882309723418e-06, + "loss": 0.5399, + "step": 4909 + }, + { + "epoch": 4.064605190502485, + "grad_norm": 0.9588503241539001, + "learning_rate": 1.203941813412312e-06, + "loss": 0.5633, + "step": 4910 + }, + { + "epoch": 4.065433462175593, + "grad_norm": 0.9639830589294434, + "learning_rate": 1.203001567876382e-06, + "loss": 0.5445, + "step": 4911 + }, + { + "epoch": 4.066261733848703, + "grad_norm": 0.9721859693527222, + "learning_rate": 1.2020615732975582e-06, + "loss": 0.5378, + "step": 4912 + }, + { + "epoch": 4.067090005521811, + "grad_norm": 0.9711841940879822, + "learning_rate": 1.2011218298577218e-06, + "loss": 0.5462, + "step": 4913 + }, + { + "epoch": 4.06791827719492, + "grad_norm": 0.9579488039016724, + "learning_rate": 1.2001823377387022e-06, + "loss": 0.5468, + "step": 4914 + }, + { + "epoch": 4.068746548868028, + "grad_norm": 0.9761105179786682, + "learning_rate": 1.1992430971222839e-06, + "loss": 0.5534, + "step": 4915 + }, + { + "epoch": 4.0695748205411375, + "grad_norm": 0.9403853416442871, + "learning_rate": 1.1983041081902016e-06, + "loss": 0.5701, + "step": 4916 + }, + { + "epoch": 4.070403092214247, + "grad_norm": 0.9711827039718628, + "learning_rate": 1.197365371124139e-06, + "loss": 0.5674, + "step": 4917 + }, + { + "epoch": 4.071231363887355, + "grad_norm": 0.9828927516937256, + "learning_rate": 1.1964268861057343e-06, + "loss": 0.5562, + "step": 4918 + }, + { + "epoch": 4.072059635560464, + "grad_norm": 0.9552168250083923, + "learning_rate": 1.195488653316576e-06, + "loss": 0.5549, + "step": 4919 + }, + { + "epoch": 4.072887907233572, + "grad_norm": 0.9774886965751648, + "learning_rate": 1.194550672938203e-06, + "loss": 0.5602, + "step": 4920 + }, + { + "epoch": 4.073716178906682, + "grad_norm": 0.9632622599601746, + "learning_rate": 1.1936129451521048e-06, + "loss": 0.562, + "step": 4921 + }, + { + "epoch": 4.07454445057979, + "grad_norm": 0.9444198608398438, + "learning_rate": 1.1926754701397234e-06, + "loss": 0.5557, + "step": 4922 + }, + { + "epoch": 4.075372722252899, + "grad_norm": 0.9769295454025269, + "learning_rate": 1.191738248082453e-06, + "loss": 0.5604, + "step": 4923 + }, + { + "epoch": 4.076200993926007, + "grad_norm": 0.9791744947433472, + "learning_rate": 1.190801279161635e-06, + "loss": 0.5539, + "step": 4924 + }, + { + "epoch": 4.0770292655991165, + "grad_norm": 0.9577887654304504, + "learning_rate": 1.1898645635585652e-06, + "loss": 0.5288, + "step": 4925 + }, + { + "epoch": 4.077857537272226, + "grad_norm": 0.9718350768089294, + "learning_rate": 1.1889281014544907e-06, + "loss": 0.5587, + "step": 4926 + }, + { + "epoch": 4.078685808945334, + "grad_norm": 0.9948698878288269, + "learning_rate": 1.1879918930306057e-06, + "loss": 0.5454, + "step": 4927 + }, + { + "epoch": 4.079514080618443, + "grad_norm": 0.9597614407539368, + "learning_rate": 1.187055938468059e-06, + "loss": 0.5506, + "step": 4928 + }, + { + "epoch": 4.080342352291551, + "grad_norm": 1.0018571615219116, + "learning_rate": 1.1861202379479504e-06, + "loss": 0.5659, + "step": 4929 + }, + { + "epoch": 4.081170623964661, + "grad_norm": 0.9607214331626892, + "learning_rate": 1.1851847916513265e-06, + "loss": 0.5459, + "step": 4930 + }, + { + "epoch": 4.081998895637769, + "grad_norm": 0.958617091178894, + "learning_rate": 1.1842495997591891e-06, + "loss": 0.5569, + "step": 4931 + }, + { + "epoch": 4.082827167310878, + "grad_norm": 0.9821828007698059, + "learning_rate": 1.1833146624524907e-06, + "loss": 0.5503, + "step": 4932 + }, + { + "epoch": 4.083655438983987, + "grad_norm": 0.9603815078735352, + "learning_rate": 1.1823799799121293e-06, + "loss": 0.5537, + "step": 4933 + }, + { + "epoch": 4.0844837106570955, + "grad_norm": 0.9634932279586792, + "learning_rate": 1.181445552318959e-06, + "loss": 0.5436, + "step": 4934 + }, + { + "epoch": 4.085311982330205, + "grad_norm": 0.9784684777259827, + "learning_rate": 1.180511379853784e-06, + "loss": 0.5454, + "step": 4935 + }, + { + "epoch": 4.086140254003313, + "grad_norm": 0.9718360304832458, + "learning_rate": 1.1795774626973558e-06, + "loss": 0.5568, + "step": 4936 + }, + { + "epoch": 4.086968525676422, + "grad_norm": 0.962182879447937, + "learning_rate": 1.1786438010303799e-06, + "loss": 0.5388, + "step": 4937 + }, + { + "epoch": 4.08779679734953, + "grad_norm": 0.9819624423980713, + "learning_rate": 1.1777103950335119e-06, + "loss": 0.5678, + "step": 4938 + }, + { + "epoch": 4.08862506902264, + "grad_norm": 0.9512423872947693, + "learning_rate": 1.1767772448873553e-06, + "loss": 0.54, + "step": 4939 + }, + { + "epoch": 4.089453340695748, + "grad_norm": 0.9459388256072998, + "learning_rate": 1.1758443507724667e-06, + "loss": 0.5427, + "step": 4940 + }, + { + "epoch": 4.090281612368857, + "grad_norm": 0.9718794822692871, + "learning_rate": 1.1749117128693538e-06, + "loss": 0.542, + "step": 4941 + }, + { + "epoch": 4.091109884041966, + "grad_norm": 0.9619993567466736, + "learning_rate": 1.1739793313584711e-06, + "loss": 0.556, + "step": 4942 + }, + { + "epoch": 4.0919381557150745, + "grad_norm": 0.9623501300811768, + "learning_rate": 1.1730472064202276e-06, + "loss": 0.5503, + "step": 4943 + }, + { + "epoch": 4.092766427388184, + "grad_norm": 0.9375072717666626, + "learning_rate": 1.172115338234979e-06, + "loss": 0.5542, + "step": 4944 + }, + { + "epoch": 4.093594699061292, + "grad_norm": 0.9621657133102417, + "learning_rate": 1.171183726983035e-06, + "loss": 0.5379, + "step": 4945 + }, + { + "epoch": 4.094422970734401, + "grad_norm": 0.9603687524795532, + "learning_rate": 1.1702523728446522e-06, + "loss": 0.535, + "step": 4946 + }, + { + "epoch": 4.095251242407509, + "grad_norm": 0.9662330150604248, + "learning_rate": 1.1693212760000392e-06, + "loss": 0.5437, + "step": 4947 + }, + { + "epoch": 4.0960795140806185, + "grad_norm": 0.9543896913528442, + "learning_rate": 1.1683904366293558e-06, + "loss": 0.5531, + "step": 4948 + }, + { + "epoch": 4.096907785753727, + "grad_norm": 0.9495199918746948, + "learning_rate": 1.1674598549127088e-06, + "loss": 0.5416, + "step": 4949 + }, + { + "epoch": 4.097736057426836, + "grad_norm": 1.0251307487487793, + "learning_rate": 1.1665295310301582e-06, + "loss": 0.5424, + "step": 4950 + }, + { + "epoch": 4.098564329099945, + "grad_norm": 0.9514643549919128, + "learning_rate": 1.1655994651617138e-06, + "loss": 0.5618, + "step": 4951 + }, + { + "epoch": 4.0993926007730535, + "grad_norm": 0.9422398209571838, + "learning_rate": 1.1646696574873332e-06, + "loss": 0.5412, + "step": 4952 + }, + { + "epoch": 4.100220872446163, + "grad_norm": 0.9614496231079102, + "learning_rate": 1.1637401081869266e-06, + "loss": 0.5473, + "step": 4953 + }, + { + "epoch": 4.101049144119271, + "grad_norm": 0.953124463558197, + "learning_rate": 1.1628108174403526e-06, + "loss": 0.5448, + "step": 4954 + }, + { + "epoch": 4.10187741579238, + "grad_norm": 0.9754185080528259, + "learning_rate": 1.1618817854274195e-06, + "loss": 0.5265, + "step": 4955 + }, + { + "epoch": 4.102705687465488, + "grad_norm": 0.9708113074302673, + "learning_rate": 1.160953012327887e-06, + "loss": 0.5399, + "step": 4956 + }, + { + "epoch": 4.1035339591385975, + "grad_norm": 0.9540616869926453, + "learning_rate": 1.1600244983214654e-06, + "loss": 0.5396, + "step": 4957 + }, + { + "epoch": 4.104362230811706, + "grad_norm": 0.9439674019813538, + "learning_rate": 1.159096243587811e-06, + "loss": 0.5688, + "step": 4958 + }, + { + "epoch": 4.105190502484815, + "grad_norm": 0.9611501097679138, + "learning_rate": 1.158168248306534e-06, + "loss": 0.5548, + "step": 4959 + }, + { + "epoch": 4.106018774157924, + "grad_norm": 0.9381349682807922, + "learning_rate": 1.1572405126571935e-06, + "loss": 0.5291, + "step": 4960 + }, + { + "epoch": 4.106847045831032, + "grad_norm": 0.9494814276695251, + "learning_rate": 1.1563130368192955e-06, + "loss": 0.5478, + "step": 4961 + }, + { + "epoch": 4.107675317504142, + "grad_norm": 0.9880107641220093, + "learning_rate": 1.1553858209722996e-06, + "loss": 0.5592, + "step": 4962 + }, + { + "epoch": 4.10850358917725, + "grad_norm": 0.9652872681617737, + "learning_rate": 1.1544588652956135e-06, + "loss": 0.5697, + "step": 4963 + }, + { + "epoch": 4.109331860850359, + "grad_norm": 0.9371438026428223, + "learning_rate": 1.1535321699685934e-06, + "loss": 0.5363, + "step": 4964 + }, + { + "epoch": 4.110160132523467, + "grad_norm": 0.9895375967025757, + "learning_rate": 1.1526057351705475e-06, + "loss": 0.5427, + "step": 4965 + }, + { + "epoch": 4.1109884041965765, + "grad_norm": 0.9242417812347412, + "learning_rate": 1.1516795610807318e-06, + "loss": 0.5222, + "step": 4966 + }, + { + "epoch": 4.111816675869686, + "grad_norm": 0.9818521738052368, + "learning_rate": 1.1507536478783509e-06, + "loss": 0.5729, + "step": 4967 + }, + { + "epoch": 4.112644947542794, + "grad_norm": 0.9748761653900146, + "learning_rate": 1.149827995742562e-06, + "loss": 0.5553, + "step": 4968 + }, + { + "epoch": 4.113473219215903, + "grad_norm": 0.9590509533882141, + "learning_rate": 1.1489026048524707e-06, + "loss": 0.5397, + "step": 4969 + }, + { + "epoch": 4.114301490889011, + "grad_norm": 0.956331729888916, + "learning_rate": 1.1479774753871294e-06, + "loss": 0.5818, + "step": 4970 + }, + { + "epoch": 4.115129762562121, + "grad_norm": 0.9534357786178589, + "learning_rate": 1.1470526075255434e-06, + "loss": 0.5665, + "step": 4971 + }, + { + "epoch": 4.115958034235229, + "grad_norm": 0.9746752381324768, + "learning_rate": 1.1461280014466667e-06, + "loss": 0.5518, + "step": 4972 + }, + { + "epoch": 4.116786305908338, + "grad_norm": 0.9731650948524475, + "learning_rate": 1.1452036573294e-06, + "loss": 0.5683, + "step": 4973 + }, + { + "epoch": 4.117614577581446, + "grad_norm": 0.9399099349975586, + "learning_rate": 1.1442795753525969e-06, + "loss": 0.5614, + "step": 4974 + }, + { + "epoch": 4.1184428492545555, + "grad_norm": 0.9529602527618408, + "learning_rate": 1.1433557556950583e-06, + "loss": 0.5456, + "step": 4975 + }, + { + "epoch": 4.119271120927665, + "grad_norm": 0.9645585417747498, + "learning_rate": 1.142432198535534e-06, + "loss": 0.5386, + "step": 4976 + }, + { + "epoch": 4.120099392600773, + "grad_norm": 0.9642344117164612, + "learning_rate": 1.141508904052725e-06, + "loss": 0.5716, + "step": 4977 + }, + { + "epoch": 4.120927664273882, + "grad_norm": 0.9634251594543457, + "learning_rate": 1.1405858724252787e-06, + "loss": 0.5613, + "step": 4978 + }, + { + "epoch": 4.12175593594699, + "grad_norm": 0.9380096793174744, + "learning_rate": 1.1396631038317944e-06, + "loss": 0.5439, + "step": 4979 + }, + { + "epoch": 4.1225842076201, + "grad_norm": 0.963697075843811, + "learning_rate": 1.1387405984508177e-06, + "loss": 0.5605, + "step": 4980 + }, + { + "epoch": 4.123412479293208, + "grad_norm": 0.9700136780738831, + "learning_rate": 1.1378183564608459e-06, + "loss": 0.5671, + "step": 4981 + }, + { + "epoch": 4.124240750966317, + "grad_norm": 0.9514760971069336, + "learning_rate": 1.1368963780403248e-06, + "loss": 0.5446, + "step": 4982 + }, + { + "epoch": 4.125069022639425, + "grad_norm": 0.9677363634109497, + "learning_rate": 1.135974663367647e-06, + "loss": 0.5387, + "step": 4983 + }, + { + "epoch": 4.1258972943125345, + "grad_norm": 0.9543762803077698, + "learning_rate": 1.1350532126211564e-06, + "loss": 0.5859, + "step": 4984 + }, + { + "epoch": 4.126725565985644, + "grad_norm": 0.9666879773139954, + "learning_rate": 1.1341320259791463e-06, + "loss": 0.5626, + "step": 4985 + }, + { + "epoch": 4.127553837658752, + "grad_norm": 0.9627404808998108, + "learning_rate": 1.133211103619856e-06, + "loss": 0.5593, + "step": 4986 + }, + { + "epoch": 4.128382109331861, + "grad_norm": 0.9710012674331665, + "learning_rate": 1.1322904457214759e-06, + "loss": 0.5665, + "step": 4987 + }, + { + "epoch": 4.129210381004969, + "grad_norm": 0.9189611077308655, + "learning_rate": 1.1313700524621468e-06, + "loss": 0.5435, + "step": 4988 + }, + { + "epoch": 4.130038652678079, + "grad_norm": 0.9373254776000977, + "learning_rate": 1.1304499240199523e-06, + "loss": 0.5832, + "step": 4989 + }, + { + "epoch": 4.130866924351187, + "grad_norm": 0.967099130153656, + "learning_rate": 1.1295300605729311e-06, + "loss": 0.5398, + "step": 4990 + }, + { + "epoch": 4.131695196024296, + "grad_norm": 0.9356493353843689, + "learning_rate": 1.1286104622990684e-06, + "loss": 0.5659, + "step": 4991 + }, + { + "epoch": 4.132523467697405, + "grad_norm": 0.970832347869873, + "learning_rate": 1.1276911293762968e-06, + "loss": 0.557, + "step": 4992 + }, + { + "epoch": 4.1333517393705135, + "grad_norm": 0.9779406189918518, + "learning_rate": 1.1267720619824986e-06, + "loss": 0.5918, + "step": 4993 + }, + { + "epoch": 4.134180011043623, + "grad_norm": 0.9740846157073975, + "learning_rate": 1.1258532602955069e-06, + "loss": 0.55, + "step": 4994 + }, + { + "epoch": 4.135008282716731, + "grad_norm": 0.959419846534729, + "learning_rate": 1.1249347244930983e-06, + "loss": 0.5383, + "step": 4995 + }, + { + "epoch": 4.13583655438984, + "grad_norm": 0.9859365224838257, + "learning_rate": 1.1240164547530025e-06, + "loss": 0.5651, + "step": 4996 + }, + { + "epoch": 4.136664826062948, + "grad_norm": 0.9637149572372437, + "learning_rate": 1.123098451252897e-06, + "loss": 0.5352, + "step": 4997 + }, + { + "epoch": 4.137493097736058, + "grad_norm": 0.9376708269119263, + "learning_rate": 1.1221807141704048e-06, + "loss": 0.5651, + "step": 4998 + }, + { + "epoch": 4.138321369409166, + "grad_norm": 0.971984326839447, + "learning_rate": 1.1212632436831015e-06, + "loss": 0.5429, + "step": 4999 + }, + { + "epoch": 4.139149641082275, + "grad_norm": 0.9628618955612183, + "learning_rate": 1.1203460399685071e-06, + "loss": 0.534, + "step": 5000 + }, + { + "epoch": 4.139977912755384, + "grad_norm": 0.9970594048500061, + "learning_rate": 1.119429103204094e-06, + "loss": 0.5468, + "step": 5001 + }, + { + "epoch": 4.1408061844284925, + "grad_norm": 0.9739755988121033, + "learning_rate": 1.118512433567279e-06, + "loss": 0.5457, + "step": 5002 + }, + { + "epoch": 4.141634456101602, + "grad_norm": 0.9746037125587463, + "learning_rate": 1.11759603123543e-06, + "loss": 0.5637, + "step": 5003 + }, + { + "epoch": 4.14246272777471, + "grad_norm": 0.9784834980964661, + "learning_rate": 1.1166798963858633e-06, + "loss": 0.5673, + "step": 5004 + }, + { + "epoch": 4.143290999447819, + "grad_norm": 0.9432827830314636, + "learning_rate": 1.1157640291958404e-06, + "loss": 0.5435, + "step": 5005 + }, + { + "epoch": 4.144119271120927, + "grad_norm": 0.9606834053993225, + "learning_rate": 1.1148484298425743e-06, + "loss": 0.5407, + "step": 5006 + }, + { + "epoch": 4.1449475427940365, + "grad_norm": 0.9408406019210815, + "learning_rate": 1.113933098503226e-06, + "loss": 0.5439, + "step": 5007 + }, + { + "epoch": 4.145775814467145, + "grad_norm": 0.9570668339729309, + "learning_rate": 1.1130180353549012e-06, + "loss": 0.5335, + "step": 5008 + }, + { + "epoch": 4.146604086140254, + "grad_norm": 0.9712949991226196, + "learning_rate": 1.1121032405746574e-06, + "loss": 0.5564, + "step": 5009 + }, + { + "epoch": 4.147432357813363, + "grad_norm": 0.9611795544624329, + "learning_rate": 1.1111887143394995e-06, + "loss": 0.556, + "step": 5010 + }, + { + "epoch": 4.1482606294864715, + "grad_norm": 1.120573878288269, + "learning_rate": 1.110274456826379e-06, + "loss": 0.5556, + "step": 5011 + }, + { + "epoch": 4.149088901159581, + "grad_norm": 0.979116678237915, + "learning_rate": 1.1093604682121956e-06, + "loss": 0.5448, + "step": 5012 + }, + { + "epoch": 4.149917172832689, + "grad_norm": 0.964521050453186, + "learning_rate": 1.1084467486737988e-06, + "loss": 0.5537, + "step": 5013 + }, + { + "epoch": 4.150745444505798, + "grad_norm": 0.9442644715309143, + "learning_rate": 1.1075332983879835e-06, + "loss": 0.5662, + "step": 5014 + }, + { + "epoch": 4.151573716178906, + "grad_norm": 0.9522957801818848, + "learning_rate": 1.1066201175314945e-06, + "loss": 0.5296, + "step": 5015 + }, + { + "epoch": 4.1524019878520155, + "grad_norm": 1.3315223455429077, + "learning_rate": 1.1057072062810247e-06, + "loss": 0.583, + "step": 5016 + }, + { + "epoch": 4.153230259525124, + "grad_norm": 0.9419928789138794, + "learning_rate": 1.1047945648132124e-06, + "loss": 0.5538, + "step": 5017 + }, + { + "epoch": 4.154058531198233, + "grad_norm": 0.9866177439689636, + "learning_rate": 1.1038821933046456e-06, + "loss": 0.5587, + "step": 5018 + }, + { + "epoch": 4.154886802871342, + "grad_norm": 0.9400057196617126, + "learning_rate": 1.102970091931861e-06, + "loss": 0.5498, + "step": 5019 + }, + { + "epoch": 4.15571507454445, + "grad_norm": 0.9604130387306213, + "learning_rate": 1.1020582608713398e-06, + "loss": 0.5431, + "step": 5020 + }, + { + "epoch": 4.15654334621756, + "grad_norm": 0.9357389807701111, + "learning_rate": 1.1011467002995135e-06, + "loss": 0.5611, + "step": 5021 + }, + { + "epoch": 4.157371617890668, + "grad_norm": 0.9960111379623413, + "learning_rate": 1.100235410392762e-06, + "loss": 0.5611, + "step": 5022 + }, + { + "epoch": 4.158199889563777, + "grad_norm": 0.987533450126648, + "learning_rate": 1.0993243913274099e-06, + "loss": 0.5598, + "step": 5023 + }, + { + "epoch": 4.159028161236885, + "grad_norm": 0.9703227877616882, + "learning_rate": 1.09841364327973e-06, + "loss": 0.5574, + "step": 5024 + }, + { + "epoch": 4.1598564329099945, + "grad_norm": 0.9649814963340759, + "learning_rate": 1.097503166425945e-06, + "loss": 0.5696, + "step": 5025 + }, + { + "epoch": 4.160684704583104, + "grad_norm": 0.9445302486419678, + "learning_rate": 1.0965929609422243e-06, + "loss": 0.5529, + "step": 5026 + }, + { + "epoch": 4.161512976256212, + "grad_norm": 0.9760971069335938, + "learning_rate": 1.0956830270046823e-06, + "loss": 0.5412, + "step": 5027 + }, + { + "epoch": 4.162341247929321, + "grad_norm": 0.9667978882789612, + "learning_rate": 1.0947733647893845e-06, + "loss": 0.5407, + "step": 5028 + }, + { + "epoch": 4.163169519602429, + "grad_norm": 0.9345208406448364, + "learning_rate": 1.0938639744723405e-06, + "loss": 0.5673, + "step": 5029 + }, + { + "epoch": 4.163997791275539, + "grad_norm": 0.9527465105056763, + "learning_rate": 1.0929548562295098e-06, + "loss": 0.5441, + "step": 5030 + }, + { + "epoch": 4.164826062948647, + "grad_norm": 0.9564769268035889, + "learning_rate": 1.0920460102367988e-06, + "loss": 0.5487, + "step": 5031 + }, + { + "epoch": 4.165654334621756, + "grad_norm": 0.9697523713111877, + "learning_rate": 1.0911374366700593e-06, + "loss": 0.5309, + "step": 5032 + }, + { + "epoch": 4.166482606294864, + "grad_norm": 0.9563925266265869, + "learning_rate": 1.0902291357050932e-06, + "loss": 0.5432, + "step": 5033 + }, + { + "epoch": 4.1673108779679735, + "grad_norm": 0.9529309272766113, + "learning_rate": 1.0893211075176465e-06, + "loss": 0.5528, + "step": 5034 + }, + { + "epoch": 4.168139149641083, + "grad_norm": 0.9415030479431152, + "learning_rate": 1.0884133522834168e-06, + "loss": 0.5491, + "step": 5035 + }, + { + "epoch": 4.168967421314191, + "grad_norm": 0.9647647142410278, + "learning_rate": 1.0875058701780432e-06, + "loss": 0.5509, + "step": 5036 + }, + { + "epoch": 4.1697956929873, + "grad_norm": 0.9571723937988281, + "learning_rate": 1.0865986613771173e-06, + "loss": 0.5494, + "step": 5037 + }, + { + "epoch": 4.170623964660408, + "grad_norm": 0.9916718602180481, + "learning_rate": 1.0856917260561755e-06, + "loss": 0.5801, + "step": 5038 + }, + { + "epoch": 4.171452236333518, + "grad_norm": 0.9753178358078003, + "learning_rate": 1.0847850643906997e-06, + "loss": 0.5663, + "step": 5039 + }, + { + "epoch": 4.172280508006626, + "grad_norm": 0.9760473966598511, + "learning_rate": 1.0838786765561217e-06, + "loss": 0.5528, + "step": 5040 + }, + { + "epoch": 4.173108779679735, + "grad_norm": 0.9597117304801941, + "learning_rate": 1.0829725627278195e-06, + "loss": 0.5523, + "step": 5041 + }, + { + "epoch": 4.173937051352843, + "grad_norm": 0.9663171768188477, + "learning_rate": 1.0820667230811164e-06, + "loss": 0.5468, + "step": 5042 + }, + { + "epoch": 4.1747653230259525, + "grad_norm": 0.9724160432815552, + "learning_rate": 1.0811611577912841e-06, + "loss": 0.5627, + "step": 5043 + }, + { + "epoch": 4.175593594699062, + "grad_norm": 0.9543465375900269, + "learning_rate": 1.080255867033543e-06, + "loss": 0.5518, + "step": 5044 + }, + { + "epoch": 4.17642186637217, + "grad_norm": 0.9619806408882141, + "learning_rate": 1.0793508509830567e-06, + "loss": 0.5268, + "step": 5045 + }, + { + "epoch": 4.177250138045279, + "grad_norm": 0.9508680105209351, + "learning_rate": 1.0784461098149367e-06, + "loss": 0.5328, + "step": 5046 + }, + { + "epoch": 4.178078409718387, + "grad_norm": 0.944930911064148, + "learning_rate": 1.0775416437042438e-06, + "loss": 0.5393, + "step": 5047 + }, + { + "epoch": 4.178906681391497, + "grad_norm": 0.987951397895813, + "learning_rate": 1.0766374528259822e-06, + "loss": 0.5556, + "step": 5048 + }, + { + "epoch": 4.179734953064605, + "grad_norm": 0.960111677646637, + "learning_rate": 1.0757335373551047e-06, + "loss": 0.5473, + "step": 5049 + }, + { + "epoch": 4.180563224737714, + "grad_norm": 0.9849942922592163, + "learning_rate": 1.0748298974665119e-06, + "loss": 0.5559, + "step": 5050 + }, + { + "epoch": 4.181391496410823, + "grad_norm": 0.9856019020080566, + "learning_rate": 1.073926533335048e-06, + "loss": 0.5547, + "step": 5051 + }, + { + "epoch": 4.1822197680839315, + "grad_norm": 0.9418588280677795, + "learning_rate": 1.073023445135506e-06, + "loss": 0.5377, + "step": 5052 + }, + { + "epoch": 4.183048039757041, + "grad_norm": 0.9605504274368286, + "learning_rate": 1.072120633042626e-06, + "loss": 0.5513, + "step": 5053 + }, + { + "epoch": 4.183876311430149, + "grad_norm": 0.9515420198440552, + "learning_rate": 1.0712180972310923e-06, + "loss": 0.5331, + "step": 5054 + }, + { + "epoch": 4.184704583103258, + "grad_norm": 0.9456093907356262, + "learning_rate": 1.0703158378755376e-06, + "loss": 0.5318, + "step": 5055 + }, + { + "epoch": 4.185532854776366, + "grad_norm": 0.986414909362793, + "learning_rate": 1.0694138551505424e-06, + "loss": 0.553, + "step": 5056 + }, + { + "epoch": 4.186361126449476, + "grad_norm": 0.9519484639167786, + "learning_rate": 1.0685121492306299e-06, + "loss": 0.5364, + "step": 5057 + }, + { + "epoch": 4.187189398122584, + "grad_norm": 0.9635508060455322, + "learning_rate": 1.0676107202902717e-06, + "loss": 0.5414, + "step": 5058 + }, + { + "epoch": 4.188017669795693, + "grad_norm": 0.948654294013977, + "learning_rate": 1.0667095685038864e-06, + "loss": 0.5756, + "step": 5059 + }, + { + "epoch": 4.188845941468802, + "grad_norm": 0.960182249546051, + "learning_rate": 1.0658086940458397e-06, + "loss": 0.5499, + "step": 5060 + }, + { + "epoch": 4.1896742131419105, + "grad_norm": 0.9578927755355835, + "learning_rate": 1.0649080970904405e-06, + "loss": 0.5625, + "step": 5061 + }, + { + "epoch": 4.19050248481502, + "grad_norm": 0.9450972676277161, + "learning_rate": 1.0640077778119466e-06, + "loss": 0.5544, + "step": 5062 + }, + { + "epoch": 4.191330756488128, + "grad_norm": 0.9294493794441223, + "learning_rate": 1.0631077363845624e-06, + "loss": 0.547, + "step": 5063 + }, + { + "epoch": 4.192159028161237, + "grad_norm": 0.9553670287132263, + "learning_rate": 1.062207972982436e-06, + "loss": 0.5548, + "step": 5064 + }, + { + "epoch": 4.192987299834345, + "grad_norm": 0.9614834189414978, + "learning_rate": 1.0613084877796636e-06, + "loss": 0.5564, + "step": 5065 + }, + { + "epoch": 4.1938155715074545, + "grad_norm": 0.9543241262435913, + "learning_rate": 1.0604092809502886e-06, + "loss": 0.5657, + "step": 5066 + }, + { + "epoch": 4.194643843180563, + "grad_norm": 0.9624195694923401, + "learning_rate": 1.0595103526682982e-06, + "loss": 0.5594, + "step": 5067 + }, + { + "epoch": 4.195472114853672, + "grad_norm": 0.9818905591964722, + "learning_rate": 1.0586117031076254e-06, + "loss": 0.5446, + "step": 5068 + }, + { + "epoch": 4.196300386526781, + "grad_norm": 0.9506664872169495, + "learning_rate": 1.0577133324421524e-06, + "loss": 0.558, + "step": 5069 + }, + { + "epoch": 4.1971286581998895, + "grad_norm": 0.9724001884460449, + "learning_rate": 1.056815240845704e-06, + "loss": 0.5575, + "step": 5070 + }, + { + "epoch": 4.197956929872999, + "grad_norm": 0.9375852346420288, + "learning_rate": 1.0559174284920531e-06, + "loss": 0.5326, + "step": 5071 + }, + { + "epoch": 4.198785201546107, + "grad_norm": 0.9587619304656982, + "learning_rate": 1.0550198955549194e-06, + "loss": 0.5861, + "step": 5072 + }, + { + "epoch": 4.199613473219216, + "grad_norm": 0.9693782329559326, + "learning_rate": 1.0541226422079648e-06, + "loss": 0.5612, + "step": 5073 + }, + { + "epoch": 4.200441744892324, + "grad_norm": 0.980484664440155, + "learning_rate": 1.0532256686248008e-06, + "loss": 0.5604, + "step": 5074 + }, + { + "epoch": 4.2012700165654335, + "grad_norm": 0.9361084699630737, + "learning_rate": 1.052328974978984e-06, + "loss": 0.5673, + "step": 5075 + }, + { + "epoch": 4.202098288238542, + "grad_norm": 0.992881715297699, + "learning_rate": 1.051432561444015e-06, + "loss": 0.5581, + "step": 5076 + }, + { + "epoch": 4.202926559911651, + "grad_norm": 0.955773651599884, + "learning_rate": 1.0505364281933418e-06, + "loss": 0.5315, + "step": 5077 + }, + { + "epoch": 4.20375483158476, + "grad_norm": 0.9578268527984619, + "learning_rate": 1.0496405754003588e-06, + "loss": 0.5598, + "step": 5078 + }, + { + "epoch": 4.204583103257868, + "grad_norm": 0.9477030634880066, + "learning_rate": 1.0487450032384045e-06, + "loss": 0.5414, + "step": 5079 + }, + { + "epoch": 4.205411374930978, + "grad_norm": 0.932655930519104, + "learning_rate": 1.0478497118807626e-06, + "loss": 0.535, + "step": 5080 + }, + { + "epoch": 4.206239646604086, + "grad_norm": 0.9418008327484131, + "learning_rate": 1.0469547015006649e-06, + "loss": 0.5484, + "step": 5081 + }, + { + "epoch": 4.207067918277195, + "grad_norm": 0.9843934774398804, + "learning_rate": 1.046059972271288e-06, + "loss": 0.5437, + "step": 5082 + }, + { + "epoch": 4.207896189950303, + "grad_norm": 0.9433165192604065, + "learning_rate": 1.045165524365752e-06, + "loss": 0.5372, + "step": 5083 + }, + { + "epoch": 4.2087244616234125, + "grad_norm": 0.9636999368667603, + "learning_rate": 1.0442713579571257e-06, + "loss": 0.5382, + "step": 5084 + }, + { + "epoch": 4.209552733296522, + "grad_norm": 0.9424169063568115, + "learning_rate": 1.043377473218422e-06, + "loss": 0.5359, + "step": 5085 + }, + { + "epoch": 4.21038100496963, + "grad_norm": 0.9796680212020874, + "learning_rate": 1.0424838703225981e-06, + "loss": 0.5489, + "step": 5086 + }, + { + "epoch": 4.211209276642739, + "grad_norm": 1.0004616975784302, + "learning_rate": 1.0415905494425584e-06, + "loss": 0.5472, + "step": 5087 + }, + { + "epoch": 4.212037548315847, + "grad_norm": 0.9699249267578125, + "learning_rate": 1.0406975107511532e-06, + "loss": 0.5553, + "step": 5088 + }, + { + "epoch": 4.212865819988957, + "grad_norm": 0.956947922706604, + "learning_rate": 1.0398047544211756e-06, + "loss": 0.5533, + "step": 5089 + }, + { + "epoch": 4.213694091662065, + "grad_norm": 1.0034435987472534, + "learning_rate": 1.0389122806253667e-06, + "loss": 0.5597, + "step": 5090 + }, + { + "epoch": 4.214522363335174, + "grad_norm": 0.9732010364532471, + "learning_rate": 1.0380200895364118e-06, + "loss": 0.5387, + "step": 5091 + }, + { + "epoch": 4.215350635008282, + "grad_norm": 0.9744183421134949, + "learning_rate": 1.03712818132694e-06, + "loss": 0.5531, + "step": 5092 + }, + { + "epoch": 4.2161789066813915, + "grad_norm": 1.0084949731826782, + "learning_rate": 1.0362365561695283e-06, + "loss": 0.5601, + "step": 5093 + }, + { + "epoch": 4.217007178354501, + "grad_norm": 0.9781489372253418, + "learning_rate": 1.0353452142366993e-06, + "loss": 0.5537, + "step": 5094 + }, + { + "epoch": 4.217835450027609, + "grad_norm": 0.9882619976997375, + "learning_rate": 1.0344541557009166e-06, + "loss": 0.5543, + "step": 5095 + }, + { + "epoch": 4.218663721700718, + "grad_norm": 0.9402450323104858, + "learning_rate": 1.0335633807345934e-06, + "loss": 0.5586, + "step": 5096 + }, + { + "epoch": 4.219491993373826, + "grad_norm": 0.9744458794593811, + "learning_rate": 1.0326728895100872e-06, + "loss": 0.5716, + "step": 5097 + }, + { + "epoch": 4.220320265046936, + "grad_norm": 0.9703624248504639, + "learning_rate": 1.0317826821996974e-06, + "loss": 0.5534, + "step": 5098 + }, + { + "epoch": 4.221148536720044, + "grad_norm": 0.9642719030380249, + "learning_rate": 1.030892758975672e-06, + "loss": 0.531, + "step": 5099 + }, + { + "epoch": 4.221976808393153, + "grad_norm": 0.9380668997764587, + "learning_rate": 1.030003120010204e-06, + "loss": 0.5602, + "step": 5100 + }, + { + "epoch": 4.222805080066261, + "grad_norm": 0.9409314393997192, + "learning_rate": 1.0291137654754289e-06, + "loss": 0.5551, + "step": 5101 + }, + { + "epoch": 4.2236333517393705, + "grad_norm": 0.9610958695411682, + "learning_rate": 1.028224695543428e-06, + "loss": 0.5671, + "step": 5102 + }, + { + "epoch": 4.22446162341248, + "grad_norm": 0.9575729966163635, + "learning_rate": 1.0273359103862287e-06, + "loss": 0.5667, + "step": 5103 + }, + { + "epoch": 4.225289895085588, + "grad_norm": 0.9521274566650391, + "learning_rate": 1.0264474101758037e-06, + "loss": 0.5526, + "step": 5104 + }, + { + "epoch": 4.226118166758697, + "grad_norm": 0.9705082178115845, + "learning_rate": 1.0255591950840679e-06, + "loss": 0.5455, + "step": 5105 + }, + { + "epoch": 4.226946438431805, + "grad_norm": 0.9718014001846313, + "learning_rate": 1.0246712652828838e-06, + "loss": 0.5555, + "step": 5106 + }, + { + "epoch": 4.227774710104915, + "grad_norm": 0.9703999757766724, + "learning_rate": 1.0237836209440563e-06, + "loss": 0.5347, + "step": 5107 + }, + { + "epoch": 4.228602981778023, + "grad_norm": 0.9935028553009033, + "learning_rate": 1.0228962622393372e-06, + "loss": 0.5711, + "step": 5108 + }, + { + "epoch": 4.229431253451132, + "grad_norm": 0.9573892951011658, + "learning_rate": 1.0220091893404226e-06, + "loss": 0.5231, + "step": 5109 + }, + { + "epoch": 4.230259525124241, + "grad_norm": 0.9764808416366577, + "learning_rate": 1.0211224024189514e-06, + "loss": 0.5383, + "step": 5110 + }, + { + "epoch": 4.2310877967973495, + "grad_norm": 0.985131025314331, + "learning_rate": 1.0202359016465096e-06, + "loss": 0.5625, + "step": 5111 + }, + { + "epoch": 4.231916068470459, + "grad_norm": 0.9500586986541748, + "learning_rate": 1.0193496871946273e-06, + "loss": 0.5497, + "step": 5112 + }, + { + "epoch": 4.232744340143567, + "grad_norm": 0.9569104313850403, + "learning_rate": 1.0184637592347783e-06, + "loss": 0.5657, + "step": 5113 + }, + { + "epoch": 4.233572611816676, + "grad_norm": 0.9736594557762146, + "learning_rate": 1.01757811793838e-06, + "loss": 0.5345, + "step": 5114 + }, + { + "epoch": 4.234400883489784, + "grad_norm": 0.9908674359321594, + "learning_rate": 1.016692763476797e-06, + "loss": 0.5506, + "step": 5115 + }, + { + "epoch": 4.2352291551628936, + "grad_norm": 0.9769543409347534, + "learning_rate": 1.015807696021338e-06, + "loss": 0.5441, + "step": 5116 + }, + { + "epoch": 4.236057426836002, + "grad_norm": 0.9631603360176086, + "learning_rate": 1.0149229157432536e-06, + "loss": 0.549, + "step": 5117 + }, + { + "epoch": 4.236885698509111, + "grad_norm": 0.9673128128051758, + "learning_rate": 1.014038422813741e-06, + "loss": 0.5687, + "step": 5118 + }, + { + "epoch": 4.23771397018222, + "grad_norm": 1.2093791961669922, + "learning_rate": 1.0131542174039427e-06, + "loss": 0.5588, + "step": 5119 + }, + { + "epoch": 4.2385422418553285, + "grad_norm": 0.9489418268203735, + "learning_rate": 1.0122702996849424e-06, + "loss": 0.5711, + "step": 5120 + }, + { + "epoch": 4.239370513528438, + "grad_norm": 1.0532208681106567, + "learning_rate": 1.0113866698277701e-06, + "loss": 0.5489, + "step": 5121 + }, + { + "epoch": 4.240198785201546, + "grad_norm": 0.9641385078430176, + "learning_rate": 1.0105033280034022e-06, + "loss": 0.553, + "step": 5122 + }, + { + "epoch": 4.241027056874655, + "grad_norm": 0.9720795750617981, + "learning_rate": 1.009620274382754e-06, + "loss": 0.5698, + "step": 5123 + }, + { + "epoch": 4.241855328547763, + "grad_norm": 0.9714332818984985, + "learning_rate": 1.0087375091366906e-06, + "loss": 0.5689, + "step": 5124 + }, + { + "epoch": 4.2426836002208725, + "grad_norm": 0.9626659154891968, + "learning_rate": 1.0078550324360179e-06, + "loss": 0.5699, + "step": 5125 + }, + { + "epoch": 4.243511871893981, + "grad_norm": 0.9580199718475342, + "learning_rate": 1.0069728444514857e-06, + "loss": 0.5559, + "step": 5126 + }, + { + "epoch": 4.24434014356709, + "grad_norm": 0.9758951663970947, + "learning_rate": 1.0060909453537903e-06, + "loss": 0.5513, + "step": 5127 + }, + { + "epoch": 4.245168415240199, + "grad_norm": 0.9805962443351746, + "learning_rate": 1.005209335313572e-06, + "loss": 0.5448, + "step": 5128 + }, + { + "epoch": 4.2459966869133074, + "grad_norm": 1.2901999950408936, + "learning_rate": 1.004328014501412e-06, + "loss": 0.5505, + "step": 5129 + }, + { + "epoch": 4.246824958586417, + "grad_norm": 0.976852297782898, + "learning_rate": 1.0034469830878389e-06, + "loss": 0.531, + "step": 5130 + }, + { + "epoch": 4.247653230259525, + "grad_norm": 0.9891388416290283, + "learning_rate": 1.0025662412433246e-06, + "loss": 0.5493, + "step": 5131 + }, + { + "epoch": 4.248481501932634, + "grad_norm": 1.014784336090088, + "learning_rate": 1.001685789138283e-06, + "loss": 0.553, + "step": 5132 + }, + { + "epoch": 4.249309773605742, + "grad_norm": 0.958575427532196, + "learning_rate": 1.0008056269430739e-06, + "loss": 0.5661, + "step": 5133 + }, + { + "epoch": 4.2501380452788515, + "grad_norm": 0.9698933959007263, + "learning_rate": 9.999257548280018e-07, + "loss": 0.557, + "step": 5134 + }, + { + "epoch": 4.25096631695196, + "grad_norm": 0.9629998803138733, + "learning_rate": 9.990461729633124e-07, + "loss": 0.5512, + "step": 5135 + }, + { + "epoch": 4.251794588625069, + "grad_norm": 1.1001232862472534, + "learning_rate": 9.981668815191961e-07, + "loss": 0.5737, + "step": 5136 + }, + { + "epoch": 4.252622860298178, + "grad_norm": 1.025240421295166, + "learning_rate": 9.972878806657883e-07, + "loss": 0.5531, + "step": 5137 + }, + { + "epoch": 4.253451131971286, + "grad_norm": 0.9492990374565125, + "learning_rate": 9.964091705731683e-07, + "loss": 0.563, + "step": 5138 + }, + { + "epoch": 4.254279403644396, + "grad_norm": 0.9647439122200012, + "learning_rate": 9.955307514113566e-07, + "loss": 0.5729, + "step": 5139 + }, + { + "epoch": 4.255107675317504, + "grad_norm": 0.9514984488487244, + "learning_rate": 9.946526233503199e-07, + "loss": 0.5589, + "step": 5140 + }, + { + "epoch": 4.255935946990613, + "grad_norm": 0.9453758001327515, + "learning_rate": 9.93774786559969e-07, + "loss": 0.537, + "step": 5141 + }, + { + "epoch": 4.256764218663721, + "grad_norm": 0.9403778314590454, + "learning_rate": 9.928972412101551e-07, + "loss": 0.5726, + "step": 5142 + }, + { + "epoch": 4.2575924903368305, + "grad_norm": 0.9746240973472595, + "learning_rate": 9.92019987470676e-07, + "loss": 0.5709, + "step": 5143 + }, + { + "epoch": 4.258420762009939, + "grad_norm": 0.9754739999771118, + "learning_rate": 9.911430255112731e-07, + "loss": 0.5392, + "step": 5144 + }, + { + "epoch": 4.259249033683048, + "grad_norm": 0.9354885816574097, + "learning_rate": 9.90266355501628e-07, + "loss": 0.5256, + "step": 5145 + }, + { + "epoch": 4.260077305356157, + "grad_norm": 0.9805651903152466, + "learning_rate": 9.89389977611371e-07, + "loss": 0.571, + "step": 5146 + }, + { + "epoch": 4.260905577029265, + "grad_norm": 0.9707584977149963, + "learning_rate": 9.885138920100716e-07, + "loss": 0.5717, + "step": 5147 + }, + { + "epoch": 4.261733848702375, + "grad_norm": 0.9590045809745789, + "learning_rate": 9.87638098867243e-07, + "loss": 0.5568, + "step": 5148 + }, + { + "epoch": 4.262562120375483, + "grad_norm": 0.9531744122505188, + "learning_rate": 9.867625983523444e-07, + "loss": 0.5572, + "step": 5149 + }, + { + "epoch": 4.263390392048592, + "grad_norm": 0.9503223896026611, + "learning_rate": 9.858873906347777e-07, + "loss": 0.5503, + "step": 5150 + }, + { + "epoch": 4.2642186637217, + "grad_norm": 0.971651017665863, + "learning_rate": 9.850124758838856e-07, + "loss": 0.5329, + "step": 5151 + }, + { + "epoch": 4.2650469353948095, + "grad_norm": 0.9676386117935181, + "learning_rate": 9.841378542689566e-07, + "loss": 0.5445, + "step": 5152 + }, + { + "epoch": 4.265875207067919, + "grad_norm": 0.9490211606025696, + "learning_rate": 9.832635259592236e-07, + "loss": 0.5605, + "step": 5153 + }, + { + "epoch": 4.266703478741027, + "grad_norm": 0.9679016470909119, + "learning_rate": 9.823894911238582e-07, + "loss": 0.5503, + "step": 5154 + }, + { + "epoch": 4.267531750414136, + "grad_norm": 0.9777132272720337, + "learning_rate": 9.815157499319794e-07, + "loss": 0.556, + "step": 5155 + }, + { + "epoch": 4.268360022087244, + "grad_norm": 0.9584012031555176, + "learning_rate": 9.806423025526487e-07, + "loss": 0.5385, + "step": 5156 + }, + { + "epoch": 4.269188293760354, + "grad_norm": 0.9709147810935974, + "learning_rate": 9.797691491548682e-07, + "loss": 0.5553, + "step": 5157 + }, + { + "epoch": 4.270016565433462, + "grad_norm": 0.9639657735824585, + "learning_rate": 9.788962899075867e-07, + "loss": 0.5429, + "step": 5158 + }, + { + "epoch": 4.270844837106571, + "grad_norm": 0.9870532155036926, + "learning_rate": 9.780237249796932e-07, + "loss": 0.5601, + "step": 5159 + }, + { + "epoch": 4.27167310877968, + "grad_norm": 0.9454158544540405, + "learning_rate": 9.771514545400219e-07, + "loss": 0.5371, + "step": 5160 + }, + { + "epoch": 4.2725013804527885, + "grad_norm": 0.9763281941413879, + "learning_rate": 9.762794787573472e-07, + "loss": 0.5404, + "step": 5161 + }, + { + "epoch": 4.273329652125898, + "grad_norm": 1.1138204336166382, + "learning_rate": 9.754077978003896e-07, + "loss": 0.5464, + "step": 5162 + }, + { + "epoch": 4.274157923799006, + "grad_norm": 0.9644905924797058, + "learning_rate": 9.745364118378117e-07, + "loss": 0.5496, + "step": 5163 + }, + { + "epoch": 4.274986195472115, + "grad_norm": 1.0430723428726196, + "learning_rate": 9.736653210382175e-07, + "loss": 0.5508, + "step": 5164 + }, + { + "epoch": 4.275814467145223, + "grad_norm": 0.9571973085403442, + "learning_rate": 9.727945255701552e-07, + "loss": 0.5409, + "step": 5165 + }, + { + "epoch": 4.276642738818333, + "grad_norm": 0.9848223328590393, + "learning_rate": 9.719240256021165e-07, + "loss": 0.5779, + "step": 5166 + }, + { + "epoch": 4.277471010491441, + "grad_norm": 0.9693152904510498, + "learning_rate": 9.710538213025333e-07, + "loss": 0.5393, + "step": 5167 + }, + { + "epoch": 4.27829928216455, + "grad_norm": 0.9640884399414062, + "learning_rate": 9.701839128397842e-07, + "loss": 0.5684, + "step": 5168 + }, + { + "epoch": 4.279127553837659, + "grad_norm": 0.9764716625213623, + "learning_rate": 9.69314300382187e-07, + "loss": 0.5506, + "step": 5169 + }, + { + "epoch": 4.2799558255107675, + "grad_norm": 0.9657016396522522, + "learning_rate": 9.684449840980032e-07, + "loss": 0.5467, + "step": 5170 + }, + { + "epoch": 4.280784097183877, + "grad_norm": 0.9599623680114746, + "learning_rate": 9.675759641554376e-07, + "loss": 0.5491, + "step": 5171 + }, + { + "epoch": 4.281612368856985, + "grad_norm": 0.9276195764541626, + "learning_rate": 9.667072407226392e-07, + "loss": 0.5452, + "step": 5172 + }, + { + "epoch": 4.282440640530094, + "grad_norm": 0.9440439343452454, + "learning_rate": 9.658388139676953e-07, + "loss": 0.5467, + "step": 5173 + }, + { + "epoch": 4.283268912203202, + "grad_norm": 1.0000046491622925, + "learning_rate": 9.649706840586397e-07, + "loss": 0.5472, + "step": 5174 + }, + { + "epoch": 4.2840971838763116, + "grad_norm": 0.9782945513725281, + "learning_rate": 9.641028511634484e-07, + "loss": 0.5846, + "step": 5175 + }, + { + "epoch": 4.28492545554942, + "grad_norm": 0.9756081700325012, + "learning_rate": 9.632353154500371e-07, + "loss": 0.5539, + "step": 5176 + }, + { + "epoch": 4.285753727222529, + "grad_norm": 0.9515950083732605, + "learning_rate": 9.623680770862665e-07, + "loss": 0.5766, + "step": 5177 + }, + { + "epoch": 4.286581998895638, + "grad_norm": 0.9987873435020447, + "learning_rate": 9.615011362399404e-07, + "loss": 0.5608, + "step": 5178 + }, + { + "epoch": 4.2874102705687465, + "grad_norm": 0.988653838634491, + "learning_rate": 9.606344930788016e-07, + "loss": 0.5565, + "step": 5179 + }, + { + "epoch": 4.288238542241856, + "grad_norm": 0.9355522394180298, + "learning_rate": 9.597681477705398e-07, + "loss": 0.5653, + "step": 5180 + }, + { + "epoch": 4.289066813914964, + "grad_norm": 0.9759084582328796, + "learning_rate": 9.589021004827834e-07, + "loss": 0.5759, + "step": 5181 + }, + { + "epoch": 4.289895085588073, + "grad_norm": 0.9520936012268066, + "learning_rate": 9.58036351383104e-07, + "loss": 0.566, + "step": 5182 + }, + { + "epoch": 4.290723357261181, + "grad_norm": 0.9801774024963379, + "learning_rate": 9.571709006390163e-07, + "loss": 0.547, + "step": 5183 + }, + { + "epoch": 4.2915516289342905, + "grad_norm": 0.9738385081291199, + "learning_rate": 9.56305748417978e-07, + "loss": 0.5476, + "step": 5184 + }, + { + "epoch": 4.292379900607399, + "grad_norm": 1.0364208221435547, + "learning_rate": 9.554408948873868e-07, + "loss": 0.5596, + "step": 5185 + }, + { + "epoch": 4.293208172280508, + "grad_norm": 1.0218520164489746, + "learning_rate": 9.545763402145838e-07, + "loss": 0.5593, + "step": 5186 + }, + { + "epoch": 4.294036443953617, + "grad_norm": 0.9752948880195618, + "learning_rate": 9.53712084566854e-07, + "loss": 0.5315, + "step": 5187 + }, + { + "epoch": 4.2948647156267254, + "grad_norm": 0.969005286693573, + "learning_rate": 9.528481281114205e-07, + "loss": 0.5601, + "step": 5188 + }, + { + "epoch": 4.295692987299835, + "grad_norm": 0.9498629570007324, + "learning_rate": 9.519844710154518e-07, + "loss": 0.554, + "step": 5189 + }, + { + "epoch": 4.296521258972943, + "grad_norm": 0.9451913237571716, + "learning_rate": 9.511211134460585e-07, + "loss": 0.5548, + "step": 5190 + }, + { + "epoch": 4.297349530646052, + "grad_norm": 0.9663949608802795, + "learning_rate": 9.502580555702911e-07, + "loss": 0.5296, + "step": 5191 + }, + { + "epoch": 4.29817780231916, + "grad_norm": 0.9395544528961182, + "learning_rate": 9.493952975551427e-07, + "loss": 0.544, + "step": 5192 + }, + { + "epoch": 4.2990060739922695, + "grad_norm": 0.9444777965545654, + "learning_rate": 9.485328395675497e-07, + "loss": 0.5438, + "step": 5193 + }, + { + "epoch": 4.299834345665378, + "grad_norm": 0.9567984938621521, + "learning_rate": 9.476706817743906e-07, + "loss": 0.5497, + "step": 5194 + }, + { + "epoch": 4.300662617338487, + "grad_norm": 0.967971682548523, + "learning_rate": 9.468088243424828e-07, + "loss": 0.5374, + "step": 5195 + }, + { + "epoch": 4.301490889011596, + "grad_norm": 0.9735593199729919, + "learning_rate": 9.459472674385889e-07, + "loss": 0.5743, + "step": 5196 + }, + { + "epoch": 4.302319160684704, + "grad_norm": 0.9545936584472656, + "learning_rate": 9.45086011229413e-07, + "loss": 0.563, + "step": 5197 + }, + { + "epoch": 4.303147432357814, + "grad_norm": 0.9573048949241638, + "learning_rate": 9.442250558815983e-07, + "loss": 0.5456, + "step": 5198 + }, + { + "epoch": 4.303975704030922, + "grad_norm": 0.9941751956939697, + "learning_rate": 9.433644015617327e-07, + "loss": 0.5541, + "step": 5199 + }, + { + "epoch": 4.304803975704031, + "grad_norm": 0.9913632869720459, + "learning_rate": 9.425040484363454e-07, + "loss": 0.5809, + "step": 5200 + }, + { + "epoch": 4.305632247377139, + "grad_norm": 1.0065845251083374, + "learning_rate": 9.416439966719049e-07, + "loss": 0.5579, + "step": 5201 + }, + { + "epoch": 4.3064605190502485, + "grad_norm": 0.9607630968093872, + "learning_rate": 9.407842464348243e-07, + "loss": 0.5579, + "step": 5202 + }, + { + "epoch": 4.307288790723357, + "grad_norm": 0.9504464864730835, + "learning_rate": 9.399247978914591e-07, + "loss": 0.5479, + "step": 5203 + }, + { + "epoch": 4.308117062396466, + "grad_norm": 0.9630054235458374, + "learning_rate": 9.390656512081009e-07, + "loss": 0.5416, + "step": 5204 + }, + { + "epoch": 4.308945334069575, + "grad_norm": 0.9475155472755432, + "learning_rate": 9.382068065509884e-07, + "loss": 0.541, + "step": 5205 + }, + { + "epoch": 4.309773605742683, + "grad_norm": 0.9980082511901855, + "learning_rate": 9.373482640863004e-07, + "loss": 0.5369, + "step": 5206 + }, + { + "epoch": 4.310601877415793, + "grad_norm": 0.9459148049354553, + "learning_rate": 9.364900239801561e-07, + "loss": 0.545, + "step": 5207 + }, + { + "epoch": 4.311430149088901, + "grad_norm": 0.9700430631637573, + "learning_rate": 9.356320863986171e-07, + "loss": 0.5628, + "step": 5208 + }, + { + "epoch": 4.31225842076201, + "grad_norm": 0.9763196110725403, + "learning_rate": 9.347744515076873e-07, + "loss": 0.5602, + "step": 5209 + }, + { + "epoch": 4.313086692435118, + "grad_norm": 0.97758948802948, + "learning_rate": 9.339171194733096e-07, + "loss": 0.5627, + "step": 5210 + }, + { + "epoch": 4.3139149641082275, + "grad_norm": 0.9904363751411438, + "learning_rate": 9.330600904613705e-07, + "loss": 0.5514, + "step": 5211 + }, + { + "epoch": 4.314743235781337, + "grad_norm": 0.9707289338111877, + "learning_rate": 9.322033646376977e-07, + "loss": 0.5391, + "step": 5212 + }, + { + "epoch": 4.315571507454445, + "grad_norm": 1.0019526481628418, + "learning_rate": 9.313469421680582e-07, + "loss": 0.5563, + "step": 5213 + }, + { + "epoch": 4.316399779127554, + "grad_norm": 0.9674393534660339, + "learning_rate": 9.304908232181636e-07, + "loss": 0.5495, + "step": 5214 + }, + { + "epoch": 4.317228050800662, + "grad_norm": 0.9623247385025024, + "learning_rate": 9.296350079536629e-07, + "loss": 0.5639, + "step": 5215 + }, + { + "epoch": 4.318056322473772, + "grad_norm": 0.9826862215995789, + "learning_rate": 9.287794965401503e-07, + "loss": 0.5714, + "step": 5216 + }, + { + "epoch": 4.31888459414688, + "grad_norm": 0.9826351404190063, + "learning_rate": 9.279242891431575e-07, + "loss": 0.5342, + "step": 5217 + }, + { + "epoch": 4.319712865819989, + "grad_norm": 0.9831173419952393, + "learning_rate": 9.270693859281602e-07, + "loss": 0.5801, + "step": 5218 + }, + { + "epoch": 4.320541137493098, + "grad_norm": 1.0009872913360596, + "learning_rate": 9.262147870605748e-07, + "loss": 0.5817, + "step": 5219 + }, + { + "epoch": 4.3213694091662065, + "grad_norm": 0.962291419506073, + "learning_rate": 9.253604927057566e-07, + "loss": 0.53, + "step": 5220 + }, + { + "epoch": 4.322197680839316, + "grad_norm": 0.9815107583999634, + "learning_rate": 9.245065030290046e-07, + "loss": 0.5706, + "step": 5221 + }, + { + "epoch": 4.323025952512424, + "grad_norm": 0.9885838031768799, + "learning_rate": 9.236528181955584e-07, + "loss": 0.5503, + "step": 5222 + }, + { + "epoch": 4.323854224185533, + "grad_norm": 0.9858911633491516, + "learning_rate": 9.227994383705966e-07, + "loss": 0.5493, + "step": 5223 + }, + { + "epoch": 4.324682495858641, + "grad_norm": 1.027745008468628, + "learning_rate": 9.21946363719241e-07, + "loss": 0.5489, + "step": 5224 + }, + { + "epoch": 4.325510767531751, + "grad_norm": 0.9571676850318909, + "learning_rate": 9.210935944065552e-07, + "loss": 0.5333, + "step": 5225 + }, + { + "epoch": 4.326339039204859, + "grad_norm": 0.9350682497024536, + "learning_rate": 9.20241130597539e-07, + "loss": 0.5457, + "step": 5226 + }, + { + "epoch": 4.327167310877968, + "grad_norm": 0.9624955058097839, + "learning_rate": 9.193889724571375e-07, + "loss": 0.5396, + "step": 5227 + }, + { + "epoch": 4.327995582551077, + "grad_norm": 0.9392304420471191, + "learning_rate": 9.185371201502366e-07, + "loss": 0.5403, + "step": 5228 + }, + { + "epoch": 4.3288238542241855, + "grad_norm": 0.9604929089546204, + "learning_rate": 9.176855738416596e-07, + "loss": 0.5437, + "step": 5229 + }, + { + "epoch": 4.329652125897295, + "grad_norm": 0.9560455679893494, + "learning_rate": 9.168343336961741e-07, + "loss": 0.5484, + "step": 5230 + }, + { + "epoch": 4.330480397570403, + "grad_norm": 0.9542271494865417, + "learning_rate": 9.159833998784879e-07, + "loss": 0.5657, + "step": 5231 + }, + { + "epoch": 4.331308669243512, + "grad_norm": 0.944553554058075, + "learning_rate": 9.151327725532469e-07, + "loss": 0.5576, + "step": 5232 + }, + { + "epoch": 4.33213694091662, + "grad_norm": 0.9710668325424194, + "learning_rate": 9.142824518850402e-07, + "loss": 0.5459, + "step": 5233 + }, + { + "epoch": 4.3329652125897296, + "grad_norm": 0.9870629906654358, + "learning_rate": 9.134324380383983e-07, + "loss": 0.5521, + "step": 5234 + }, + { + "epoch": 4.333793484262838, + "grad_norm": 0.9679974913597107, + "learning_rate": 9.12582731177789e-07, + "loss": 0.5673, + "step": 5235 + }, + { + "epoch": 4.334621755935947, + "grad_norm": 0.9391168355941772, + "learning_rate": 9.117333314676233e-07, + "loss": 0.5638, + "step": 5236 + }, + { + "epoch": 4.335450027609056, + "grad_norm": 0.9927806854248047, + "learning_rate": 9.108842390722533e-07, + "loss": 0.5576, + "step": 5237 + }, + { + "epoch": 4.3362782992821645, + "grad_norm": 0.9426097273826599, + "learning_rate": 9.100354541559697e-07, + "loss": 0.5255, + "step": 5238 + }, + { + "epoch": 4.337106570955274, + "grad_norm": 0.9421987533569336, + "learning_rate": 9.09186976883003e-07, + "loss": 0.5463, + "step": 5239 + }, + { + "epoch": 4.337934842628382, + "grad_norm": 0.9485943913459778, + "learning_rate": 9.083388074175273e-07, + "loss": 0.5414, + "step": 5240 + }, + { + "epoch": 4.338763114301491, + "grad_norm": 0.9566770195960999, + "learning_rate": 9.074909459236558e-07, + "loss": 0.5695, + "step": 5241 + }, + { + "epoch": 4.339591385974599, + "grad_norm": 0.9460729360580444, + "learning_rate": 9.066433925654402e-07, + "loss": 0.5618, + "step": 5242 + }, + { + "epoch": 4.3404196576477085, + "grad_norm": 0.9430888295173645, + "learning_rate": 9.057961475068758e-07, + "loss": 0.5538, + "step": 5243 + }, + { + "epoch": 4.341247929320817, + "grad_norm": 0.9726897478103638, + "learning_rate": 9.049492109118951e-07, + "loss": 0.5724, + "step": 5244 + }, + { + "epoch": 4.342076200993926, + "grad_norm": 0.9237701296806335, + "learning_rate": 9.041025829443733e-07, + "loss": 0.5217, + "step": 5245 + }, + { + "epoch": 4.342904472667035, + "grad_norm": 0.9615205526351929, + "learning_rate": 9.032562637681258e-07, + "loss": 0.5401, + "step": 5246 + }, + { + "epoch": 4.3437327443401434, + "grad_norm": 0.9656977653503418, + "learning_rate": 9.024102535469056e-07, + "loss": 0.5558, + "step": 5247 + }, + { + "epoch": 4.344561016013253, + "grad_norm": 0.9590947031974792, + "learning_rate": 9.015645524444097e-07, + "loss": 0.5606, + "step": 5248 + }, + { + "epoch": 4.345389287686361, + "grad_norm": 0.9751269817352295, + "learning_rate": 9.007191606242718e-07, + "loss": 0.5401, + "step": 5249 + }, + { + "epoch": 4.34621755935947, + "grad_norm": 1.0034822225570679, + "learning_rate": 8.998740782500686e-07, + "loss": 0.5426, + "step": 5250 + }, + { + "epoch": 4.347045831032578, + "grad_norm": 0.96883225440979, + "learning_rate": 8.990293054853144e-07, + "loss": 0.544, + "step": 5251 + }, + { + "epoch": 4.3478741027056875, + "grad_norm": 0.998399555683136, + "learning_rate": 8.981848424934655e-07, + "loss": 0.5309, + "step": 5252 + }, + { + "epoch": 4.348702374378796, + "grad_norm": 0.963323712348938, + "learning_rate": 8.973406894379183e-07, + "loss": 0.5821, + "step": 5253 + }, + { + "epoch": 4.349530646051905, + "grad_norm": 0.9739964008331299, + "learning_rate": 8.964968464820073e-07, + "loss": 0.5624, + "step": 5254 + }, + { + "epoch": 4.350358917725014, + "grad_norm": 0.9921715259552002, + "learning_rate": 8.956533137890089e-07, + "loss": 0.5368, + "step": 5255 + }, + { + "epoch": 4.351187189398122, + "grad_norm": 0.9644178748130798, + "learning_rate": 8.948100915221394e-07, + "loss": 0.5481, + "step": 5256 + }, + { + "epoch": 4.352015461071232, + "grad_norm": 0.9639922380447388, + "learning_rate": 8.939671798445529e-07, + "loss": 0.555, + "step": 5257 + }, + { + "epoch": 4.35284373274434, + "grad_norm": 1.0016719102859497, + "learning_rate": 8.93124578919346e-07, + "loss": 0.5438, + "step": 5258 + }, + { + "epoch": 4.353672004417449, + "grad_norm": 0.9898110032081604, + "learning_rate": 8.922822889095556e-07, + "loss": 0.54, + "step": 5259 + }, + { + "epoch": 4.354500276090557, + "grad_norm": 0.9675293564796448, + "learning_rate": 8.914403099781538e-07, + "loss": 0.5664, + "step": 5260 + }, + { + "epoch": 4.3553285477636665, + "grad_norm": 0.9578870534896851, + "learning_rate": 8.905986422880569e-07, + "loss": 0.5553, + "step": 5261 + }, + { + "epoch": 4.356156819436775, + "grad_norm": 0.9734019041061401, + "learning_rate": 8.897572860021212e-07, + "loss": 0.5503, + "step": 5262 + }, + { + "epoch": 4.356985091109884, + "grad_norm": 0.9673689007759094, + "learning_rate": 8.889162412831392e-07, + "loss": 0.5735, + "step": 5263 + }, + { + "epoch": 4.357813362782993, + "grad_norm": 0.9523547887802124, + "learning_rate": 8.880755082938463e-07, + "loss": 0.5446, + "step": 5264 + }, + { + "epoch": 4.358641634456101, + "grad_norm": 0.9520670175552368, + "learning_rate": 8.872350871969171e-07, + "loss": 0.5503, + "step": 5265 + }, + { + "epoch": 4.359469906129211, + "grad_norm": 0.9825593829154968, + "learning_rate": 8.863949781549638e-07, + "loss": 0.5372, + "step": 5266 + }, + { + "epoch": 4.360298177802319, + "grad_norm": 0.9473084211349487, + "learning_rate": 8.855551813305402e-07, + "loss": 0.5435, + "step": 5267 + }, + { + "epoch": 4.361126449475428, + "grad_norm": 0.9561483263969421, + "learning_rate": 8.847156968861403e-07, + "loss": 0.5577, + "step": 5268 + }, + { + "epoch": 4.361954721148536, + "grad_norm": 0.9839791059494019, + "learning_rate": 8.838765249841949e-07, + "loss": 0.5625, + "step": 5269 + }, + { + "epoch": 4.3627829928216455, + "grad_norm": 0.9682125449180603, + "learning_rate": 8.830376657870762e-07, + "loss": 0.5412, + "step": 5270 + }, + { + "epoch": 4.363611264494755, + "grad_norm": 0.9834880232810974, + "learning_rate": 8.821991194570973e-07, + "loss": 0.5572, + "step": 5271 + }, + { + "epoch": 4.364439536167863, + "grad_norm": 0.9266020059585571, + "learning_rate": 8.813608861565076e-07, + "loss": 0.5649, + "step": 5272 + }, + { + "epoch": 4.365267807840972, + "grad_norm": 0.9482194185256958, + "learning_rate": 8.805229660474968e-07, + "loss": 0.5486, + "step": 5273 + }, + { + "epoch": 4.36609607951408, + "grad_norm": 0.9416828751564026, + "learning_rate": 8.796853592921955e-07, + "loss": 0.5481, + "step": 5274 + }, + { + "epoch": 4.36692435118719, + "grad_norm": 0.9814063906669617, + "learning_rate": 8.788480660526738e-07, + "loss": 0.5424, + "step": 5275 + }, + { + "epoch": 4.367752622860298, + "grad_norm": 0.9609609842300415, + "learning_rate": 8.780110864909383e-07, + "loss": 0.5492, + "step": 5276 + }, + { + "epoch": 4.368580894533407, + "grad_norm": 1.005044937133789, + "learning_rate": 8.771744207689373e-07, + "loss": 0.5459, + "step": 5277 + }, + { + "epoch": 4.369409166206516, + "grad_norm": 0.9385774731636047, + "learning_rate": 8.763380690485592e-07, + "loss": 0.5755, + "step": 5278 + }, + { + "epoch": 4.3702374378796245, + "grad_norm": 0.972374439239502, + "learning_rate": 8.755020314916282e-07, + "loss": 0.5451, + "step": 5279 + }, + { + "epoch": 4.371065709552734, + "grad_norm": 0.9493237137794495, + "learning_rate": 8.746663082599108e-07, + "loss": 0.5385, + "step": 5280 + }, + { + "epoch": 4.371893981225842, + "grad_norm": 0.9911744594573975, + "learning_rate": 8.738308995151124e-07, + "loss": 0.5692, + "step": 5281 + }, + { + "epoch": 4.372722252898951, + "grad_norm": 0.9547559022903442, + "learning_rate": 8.729958054188759e-07, + "loss": 0.5427, + "step": 5282 + }, + { + "epoch": 4.373550524572059, + "grad_norm": 0.974426805973053, + "learning_rate": 8.721610261327834e-07, + "loss": 0.5306, + "step": 5283 + }, + { + "epoch": 4.374378796245169, + "grad_norm": 0.9919657707214355, + "learning_rate": 8.713265618183589e-07, + "loss": 0.5438, + "step": 5284 + }, + { + "epoch": 4.375207067918277, + "grad_norm": 0.9478253126144409, + "learning_rate": 8.704924126370615e-07, + "loss": 0.5341, + "step": 5285 + }, + { + "epoch": 4.376035339591386, + "grad_norm": 0.9950957298278809, + "learning_rate": 8.69658578750292e-07, + "loss": 0.568, + "step": 5286 + }, + { + "epoch": 4.376863611264495, + "grad_norm": 0.9636523723602295, + "learning_rate": 8.688250603193907e-07, + "loss": 0.5487, + "step": 5287 + }, + { + "epoch": 4.3776918829376035, + "grad_norm": 0.974831759929657, + "learning_rate": 8.679918575056337e-07, + "loss": 0.5324, + "step": 5288 + }, + { + "epoch": 4.378520154610713, + "grad_norm": 0.9769512414932251, + "learning_rate": 8.671589704702388e-07, + "loss": 0.558, + "step": 5289 + }, + { + "epoch": 4.379348426283821, + "grad_norm": 0.9584938287734985, + "learning_rate": 8.663263993743632e-07, + "loss": 0.5529, + "step": 5290 + }, + { + "epoch": 4.38017669795693, + "grad_norm": 0.9711350798606873, + "learning_rate": 8.65494144379099e-07, + "loss": 0.5558, + "step": 5291 + }, + { + "epoch": 4.381004969630038, + "grad_norm": 0.9583411812782288, + "learning_rate": 8.646622056454815e-07, + "loss": 0.5585, + "step": 5292 + }, + { + "epoch": 4.3818332413031476, + "grad_norm": 0.9572791457176208, + "learning_rate": 8.638305833344834e-07, + "loss": 0.5559, + "step": 5293 + }, + { + "epoch": 4.382661512976256, + "grad_norm": 0.9529664516448975, + "learning_rate": 8.629992776070156e-07, + "loss": 0.5568, + "step": 5294 + }, + { + "epoch": 4.383489784649365, + "grad_norm": 0.9727757573127747, + "learning_rate": 8.621682886239266e-07, + "loss": 0.5604, + "step": 5295 + }, + { + "epoch": 4.384318056322474, + "grad_norm": 0.9406315684318542, + "learning_rate": 8.613376165460061e-07, + "loss": 0.5294, + "step": 5296 + }, + { + "epoch": 4.3851463279955825, + "grad_norm": 0.9377478361129761, + "learning_rate": 8.605072615339824e-07, + "loss": 0.531, + "step": 5297 + }, + { + "epoch": 4.385974599668692, + "grad_norm": 0.9759235382080078, + "learning_rate": 8.596772237485193e-07, + "loss": 0.5461, + "step": 5298 + }, + { + "epoch": 4.3868028713418, + "grad_norm": 0.9580682516098022, + "learning_rate": 8.588475033502228e-07, + "loss": 0.5513, + "step": 5299 + }, + { + "epoch": 4.387631143014909, + "grad_norm": 0.9866207242012024, + "learning_rate": 8.580181004996362e-07, + "loss": 0.5647, + "step": 5300 + }, + { + "epoch": 4.388459414688017, + "grad_norm": 0.9822670817375183, + "learning_rate": 8.571890153572404e-07, + "loss": 0.5608, + "step": 5301 + }, + { + "epoch": 4.3892876863611265, + "grad_norm": 0.9508621096611023, + "learning_rate": 8.563602480834557e-07, + "loss": 0.5297, + "step": 5302 + }, + { + "epoch": 4.390115958034235, + "grad_norm": 0.9849876761436462, + "learning_rate": 8.555317988386422e-07, + "loss": 0.5836, + "step": 5303 + }, + { + "epoch": 4.390944229707344, + "grad_norm": 0.9595887064933777, + "learning_rate": 8.54703667783096e-07, + "loss": 0.5492, + "step": 5304 + }, + { + "epoch": 4.391772501380453, + "grad_norm": 0.9659999012947083, + "learning_rate": 8.53875855077052e-07, + "loss": 0.5245, + "step": 5305 + }, + { + "epoch": 4.3926007730535614, + "grad_norm": 0.9543110132217407, + "learning_rate": 8.530483608806861e-07, + "loss": 0.5379, + "step": 5306 + }, + { + "epoch": 4.393429044726671, + "grad_norm": 0.9603022933006287, + "learning_rate": 8.522211853541087e-07, + "loss": 0.5629, + "step": 5307 + }, + { + "epoch": 4.394257316399779, + "grad_norm": 0.9761829972267151, + "learning_rate": 8.513943286573714e-07, + "loss": 0.5398, + "step": 5308 + }, + { + "epoch": 4.395085588072888, + "grad_norm": 0.9566043615341187, + "learning_rate": 8.505677909504645e-07, + "loss": 0.5671, + "step": 5309 + }, + { + "epoch": 4.395913859745996, + "grad_norm": 0.9672698378562927, + "learning_rate": 8.497415723933133e-07, + "loss": 0.5369, + "step": 5310 + }, + { + "epoch": 4.3967421314191055, + "grad_norm": 0.945254921913147, + "learning_rate": 8.489156731457843e-07, + "loss": 0.5372, + "step": 5311 + }, + { + "epoch": 4.397570403092214, + "grad_norm": 0.9574915170669556, + "learning_rate": 8.480900933676822e-07, + "loss": 0.5583, + "step": 5312 + }, + { + "epoch": 4.398398674765323, + "grad_norm": 0.9455167651176453, + "learning_rate": 8.472648332187471e-07, + "loss": 0.5542, + "step": 5313 + }, + { + "epoch": 4.399226946438432, + "grad_norm": 0.9736841320991516, + "learning_rate": 8.464398928586603e-07, + "loss": 0.5501, + "step": 5314 + }, + { + "epoch": 4.40005521811154, + "grad_norm": 0.9645372629165649, + "learning_rate": 8.456152724470409e-07, + "loss": 0.5785, + "step": 5315 + }, + { + "epoch": 4.40088348978465, + "grad_norm": 0.9789561033248901, + "learning_rate": 8.447909721434441e-07, + "loss": 0.5487, + "step": 5316 + }, + { + "epoch": 4.401711761457758, + "grad_norm": 0.9858975410461426, + "learning_rate": 8.43966992107364e-07, + "loss": 0.5627, + "step": 5317 + }, + { + "epoch": 4.402540033130867, + "grad_norm": 0.9592116475105286, + "learning_rate": 8.431433324982332e-07, + "loss": 0.5741, + "step": 5318 + }, + { + "epoch": 4.403368304803975, + "grad_norm": 0.9792628884315491, + "learning_rate": 8.423199934754237e-07, + "loss": 0.5385, + "step": 5319 + }, + { + "epoch": 4.4041965764770845, + "grad_norm": 0.9497966170310974, + "learning_rate": 8.414969751982419e-07, + "loss": 0.5569, + "step": 5320 + }, + { + "epoch": 4.405024848150193, + "grad_norm": 0.9575057029724121, + "learning_rate": 8.406742778259355e-07, + "loss": 0.534, + "step": 5321 + }, + { + "epoch": 4.405853119823302, + "grad_norm": 0.9563487768173218, + "learning_rate": 8.398519015176879e-07, + "loss": 0.5638, + "step": 5322 + }, + { + "epoch": 4.406681391496411, + "grad_norm": 0.9754065275192261, + "learning_rate": 8.390298464326213e-07, + "loss": 0.5562, + "step": 5323 + }, + { + "epoch": 4.407509663169519, + "grad_norm": 0.963133692741394, + "learning_rate": 8.382081127297972e-07, + "loss": 0.5514, + "step": 5324 + }, + { + "epoch": 4.408337934842629, + "grad_norm": 0.9998530745506287, + "learning_rate": 8.373867005682113e-07, + "loss": 0.5709, + "step": 5325 + }, + { + "epoch": 4.409166206515737, + "grad_norm": 0.9418386220932007, + "learning_rate": 8.365656101068001e-07, + "loss": 0.5449, + "step": 5326 + }, + { + "epoch": 4.409994478188846, + "grad_norm": 0.9713196158409119, + "learning_rate": 8.357448415044378e-07, + "loss": 0.5683, + "step": 5327 + }, + { + "epoch": 4.410822749861954, + "grad_norm": 0.9514254331588745, + "learning_rate": 8.349243949199349e-07, + "loss": 0.5539, + "step": 5328 + }, + { + "epoch": 4.4116510215350635, + "grad_norm": 0.9651909470558167, + "learning_rate": 8.34104270512039e-07, + "loss": 0.5794, + "step": 5329 + }, + { + "epoch": 4.412479293208173, + "grad_norm": 0.9516454339027405, + "learning_rate": 8.332844684394373e-07, + "loss": 0.5617, + "step": 5330 + }, + { + "epoch": 4.413307564881281, + "grad_norm": 1.0173704624176025, + "learning_rate": 8.32464988860755e-07, + "loss": 0.5466, + "step": 5331 + }, + { + "epoch": 4.41413583655439, + "grad_norm": 0.9605720043182373, + "learning_rate": 8.316458319345521e-07, + "loss": 0.5467, + "step": 5332 + }, + { + "epoch": 4.414964108227498, + "grad_norm": 0.9562596678733826, + "learning_rate": 8.308269978193285e-07, + "loss": 0.557, + "step": 5333 + }, + { + "epoch": 4.415792379900608, + "grad_norm": 0.951440691947937, + "learning_rate": 8.300084866735217e-07, + "loss": 0.5452, + "step": 5334 + }, + { + "epoch": 4.416620651573716, + "grad_norm": 0.9654122591018677, + "learning_rate": 8.291902986555045e-07, + "loss": 0.5328, + "step": 5335 + }, + { + "epoch": 4.417448923246825, + "grad_norm": 0.9830499887466431, + "learning_rate": 8.283724339235893e-07, + "loss": 0.5533, + "step": 5336 + }, + { + "epoch": 4.418277194919934, + "grad_norm": 0.9823290109634399, + "learning_rate": 8.275548926360263e-07, + "loss": 0.5422, + "step": 5337 + }, + { + "epoch": 4.4191054665930425, + "grad_norm": 0.9932782053947449, + "learning_rate": 8.267376749510012e-07, + "loss": 0.5666, + "step": 5338 + }, + { + "epoch": 4.419933738266152, + "grad_norm": 0.947321891784668, + "learning_rate": 8.259207810266373e-07, + "loss": 0.5511, + "step": 5339 + }, + { + "epoch": 4.42076200993926, + "grad_norm": 0.9844286441802979, + "learning_rate": 8.251042110209975e-07, + "loss": 0.541, + "step": 5340 + }, + { + "epoch": 4.421590281612369, + "grad_norm": 0.9536518454551697, + "learning_rate": 8.242879650920788e-07, + "loss": 0.561, + "step": 5341 + }, + { + "epoch": 4.422418553285477, + "grad_norm": 0.9690708518028259, + "learning_rate": 8.234720433978183e-07, + "loss": 0.5526, + "step": 5342 + }, + { + "epoch": 4.423246824958587, + "grad_norm": 0.9607888460159302, + "learning_rate": 8.226564460960895e-07, + "loss": 0.5695, + "step": 5343 + }, + { + "epoch": 4.424075096631695, + "grad_norm": 0.9637815356254578, + "learning_rate": 8.218411733447018e-07, + "loss": 0.575, + "step": 5344 + }, + { + "epoch": 4.424903368304804, + "grad_norm": 0.982789158821106, + "learning_rate": 8.210262253014034e-07, + "loss": 0.5437, + "step": 5345 + }, + { + "epoch": 4.425731639977913, + "grad_norm": 0.9953805804252625, + "learning_rate": 8.202116021238801e-07, + "loss": 0.5812, + "step": 5346 + }, + { + "epoch": 4.4265599116510215, + "grad_norm": 0.9684412479400635, + "learning_rate": 8.193973039697519e-07, + "loss": 0.5381, + "step": 5347 + }, + { + "epoch": 4.427388183324131, + "grad_norm": 0.9479103088378906, + "learning_rate": 8.185833309965793e-07, + "loss": 0.5646, + "step": 5348 + }, + { + "epoch": 4.428216454997239, + "grad_norm": 0.9611683487892151, + "learning_rate": 8.17769683361859e-07, + "loss": 0.5568, + "step": 5349 + }, + { + "epoch": 4.429044726670348, + "grad_norm": 0.9750727415084839, + "learning_rate": 8.169563612230233e-07, + "loss": 0.5629, + "step": 5350 + }, + { + "epoch": 4.429872998343456, + "grad_norm": 0.9549238681793213, + "learning_rate": 8.161433647374417e-07, + "loss": 0.5277, + "step": 5351 + }, + { + "epoch": 4.4307012700165656, + "grad_norm": 0.9485230445861816, + "learning_rate": 8.153306940624225e-07, + "loss": 0.5406, + "step": 5352 + }, + { + "epoch": 4.431529541689674, + "grad_norm": 0.9689464569091797, + "learning_rate": 8.145183493552103e-07, + "loss": 0.5563, + "step": 5353 + }, + { + "epoch": 4.432357813362783, + "grad_norm": 0.9542263746261597, + "learning_rate": 8.13706330772985e-07, + "loss": 0.5491, + "step": 5354 + }, + { + "epoch": 4.433186085035892, + "grad_norm": 0.9412059187889099, + "learning_rate": 8.128946384728654e-07, + "loss": 0.5744, + "step": 5355 + }, + { + "epoch": 4.4340143567090005, + "grad_norm": 0.9872552156448364, + "learning_rate": 8.120832726119069e-07, + "loss": 0.5504, + "step": 5356 + }, + { + "epoch": 4.43484262838211, + "grad_norm": 0.9560122489929199, + "learning_rate": 8.112722333470996e-07, + "loss": 0.5451, + "step": 5357 + }, + { + "epoch": 4.435670900055218, + "grad_norm": 0.9743005633354187, + "learning_rate": 8.104615208353731e-07, + "loss": 0.5633, + "step": 5358 + }, + { + "epoch": 4.436499171728327, + "grad_norm": 0.9652897119522095, + "learning_rate": 8.096511352335937e-07, + "loss": 0.5474, + "step": 5359 + }, + { + "epoch": 4.437327443401435, + "grad_norm": 0.9695815443992615, + "learning_rate": 8.088410766985613e-07, + "loss": 0.5522, + "step": 5360 + }, + { + "epoch": 4.4381557150745445, + "grad_norm": 0.9816685318946838, + "learning_rate": 8.080313453870167e-07, + "loss": 0.5472, + "step": 5361 + }, + { + "epoch": 4.438983986747653, + "grad_norm": 0.9512736201286316, + "learning_rate": 8.072219414556345e-07, + "loss": 0.5437, + "step": 5362 + }, + { + "epoch": 4.439812258420762, + "grad_norm": 0.9622536301612854, + "learning_rate": 8.064128650610259e-07, + "loss": 0.5383, + "step": 5363 + }, + { + "epoch": 4.440640530093871, + "grad_norm": 0.9731073379516602, + "learning_rate": 8.056041163597406e-07, + "loss": 0.5777, + "step": 5364 + }, + { + "epoch": 4.4414688017669794, + "grad_norm": 1.0841182470321655, + "learning_rate": 8.047956955082645e-07, + "loss": 0.5542, + "step": 5365 + }, + { + "epoch": 4.442297073440089, + "grad_norm": 0.9848759174346924, + "learning_rate": 8.039876026630181e-07, + "loss": 0.5456, + "step": 5366 + }, + { + "epoch": 4.443125345113197, + "grad_norm": 0.9822127819061279, + "learning_rate": 8.031798379803604e-07, + "loss": 0.5753, + "step": 5367 + }, + { + "epoch": 4.443953616786306, + "grad_norm": 1.094104528427124, + "learning_rate": 8.023724016165873e-07, + "loss": 0.558, + "step": 5368 + }, + { + "epoch": 4.444781888459414, + "grad_norm": 0.9637600779533386, + "learning_rate": 8.015652937279289e-07, + "loss": 0.533, + "step": 5369 + }, + { + "epoch": 4.4456101601325235, + "grad_norm": 0.9773861169815063, + "learning_rate": 8.007585144705532e-07, + "loss": 0.5798, + "step": 5370 + }, + { + "epoch": 4.446438431805632, + "grad_norm": 0.9739058613777161, + "learning_rate": 7.99952064000566e-07, + "loss": 0.563, + "step": 5371 + }, + { + "epoch": 4.447266703478741, + "grad_norm": 0.9480531811714172, + "learning_rate": 7.991459424740064e-07, + "loss": 0.5452, + "step": 5372 + }, + { + "epoch": 4.44809497515185, + "grad_norm": 0.9757310152053833, + "learning_rate": 7.98340150046851e-07, + "loss": 0.5758, + "step": 5373 + }, + { + "epoch": 4.448923246824958, + "grad_norm": 0.9749124646186829, + "learning_rate": 7.975346868750142e-07, + "loss": 0.5413, + "step": 5374 + }, + { + "epoch": 4.449751518498068, + "grad_norm": 0.961652934551239, + "learning_rate": 7.967295531143459e-07, + "loss": 0.5617, + "step": 5375 + }, + { + "epoch": 4.450579790171176, + "grad_norm": 0.9718902111053467, + "learning_rate": 7.959247489206306e-07, + "loss": 0.5496, + "step": 5376 + }, + { + "epoch": 4.451408061844285, + "grad_norm": 0.959939181804657, + "learning_rate": 7.951202744495912e-07, + "loss": 0.5535, + "step": 5377 + }, + { + "epoch": 4.452236333517393, + "grad_norm": 0.9867039918899536, + "learning_rate": 7.943161298568869e-07, + "loss": 0.5542, + "step": 5378 + }, + { + "epoch": 4.4530646051905025, + "grad_norm": 0.9625898003578186, + "learning_rate": 7.935123152981103e-07, + "loss": 0.5565, + "step": 5379 + }, + { + "epoch": 4.453892876863611, + "grad_norm": 0.9669535160064697, + "learning_rate": 7.927088309287931e-07, + "loss": 0.5514, + "step": 5380 + }, + { + "epoch": 4.45472114853672, + "grad_norm": 0.9600343108177185, + "learning_rate": 7.919056769044028e-07, + "loss": 0.5736, + "step": 5381 + }, + { + "epoch": 4.455549420209829, + "grad_norm": 0.9746002554893494, + "learning_rate": 7.911028533803406e-07, + "loss": 0.5438, + "step": 5382 + }, + { + "epoch": 4.456377691882937, + "grad_norm": 0.9626099467277527, + "learning_rate": 7.903003605119469e-07, + "loss": 0.5631, + "step": 5383 + }, + { + "epoch": 4.457205963556047, + "grad_norm": 0.958444356918335, + "learning_rate": 7.894981984544958e-07, + "loss": 0.5527, + "step": 5384 + }, + { + "epoch": 4.458034235229155, + "grad_norm": 1.017724871635437, + "learning_rate": 7.886963673631973e-07, + "loss": 0.5459, + "step": 5385 + }, + { + "epoch": 4.458862506902264, + "grad_norm": 0.9509232044219971, + "learning_rate": 7.878948673931994e-07, + "loss": 0.5478, + "step": 5386 + }, + { + "epoch": 4.459690778575372, + "grad_norm": 1.008229374885559, + "learning_rate": 7.870936986995853e-07, + "loss": 0.5565, + "step": 5387 + }, + { + "epoch": 4.4605190502484815, + "grad_norm": 0.9678364396095276, + "learning_rate": 7.862928614373722e-07, + "loss": 0.5625, + "step": 5388 + }, + { + "epoch": 4.461347321921591, + "grad_norm": 0.9678050875663757, + "learning_rate": 7.854923557615152e-07, + "loss": 0.5556, + "step": 5389 + }, + { + "epoch": 4.462175593594699, + "grad_norm": 0.9574840664863586, + "learning_rate": 7.84692181826906e-07, + "loss": 0.5437, + "step": 5390 + }, + { + "epoch": 4.463003865267808, + "grad_norm": 0.9662474393844604, + "learning_rate": 7.838923397883688e-07, + "loss": 0.5486, + "step": 5391 + }, + { + "epoch": 4.463832136940916, + "grad_norm": 0.9890984296798706, + "learning_rate": 7.830928298006666e-07, + "loss": 0.5411, + "step": 5392 + }, + { + "epoch": 4.464660408614026, + "grad_norm": 0.975991427898407, + "learning_rate": 7.822936520184979e-07, + "loss": 0.5619, + "step": 5393 + }, + { + "epoch": 4.465488680287134, + "grad_norm": 0.9573751091957092, + "learning_rate": 7.814948065964942e-07, + "loss": 0.5609, + "step": 5394 + }, + { + "epoch": 4.466316951960243, + "grad_norm": 0.9584991931915283, + "learning_rate": 7.806962936892268e-07, + "loss": 0.5611, + "step": 5395 + }, + { + "epoch": 4.467145223633352, + "grad_norm": 0.9686765074729919, + "learning_rate": 7.798981134511993e-07, + "loss": 0.5673, + "step": 5396 + }, + { + "epoch": 4.4679734953064605, + "grad_norm": 0.959581732749939, + "learning_rate": 7.791002660368513e-07, + "loss": 0.5536, + "step": 5397 + }, + { + "epoch": 4.46880176697957, + "grad_norm": 0.9748058319091797, + "learning_rate": 7.783027516005601e-07, + "loss": 0.5541, + "step": 5398 + }, + { + "epoch": 4.469630038652678, + "grad_norm": 0.9732323884963989, + "learning_rate": 7.775055702966375e-07, + "loss": 0.5507, + "step": 5399 + }, + { + "epoch": 4.470458310325787, + "grad_norm": 0.9876232147216797, + "learning_rate": 7.767087222793295e-07, + "loss": 0.5486, + "step": 5400 + }, + { + "epoch": 4.471286581998895, + "grad_norm": 0.963202714920044, + "learning_rate": 7.759122077028194e-07, + "loss": 0.5633, + "step": 5401 + }, + { + "epoch": 4.472114853672005, + "grad_norm": 0.9766404628753662, + "learning_rate": 7.751160267212262e-07, + "loss": 0.5662, + "step": 5402 + }, + { + "epoch": 4.472943125345113, + "grad_norm": 0.9662322402000427, + "learning_rate": 7.743201794886021e-07, + "loss": 0.5502, + "step": 5403 + }, + { + "epoch": 4.473771397018222, + "grad_norm": 0.9390942454338074, + "learning_rate": 7.735246661589363e-07, + "loss": 0.5415, + "step": 5404 + }, + { + "epoch": 4.474599668691331, + "grad_norm": 0.9523019194602966, + "learning_rate": 7.727294868861546e-07, + "loss": 0.5482, + "step": 5405 + }, + { + "epoch": 4.4754279403644395, + "grad_norm": 0.9844672083854675, + "learning_rate": 7.719346418241158e-07, + "loss": 0.5576, + "step": 5406 + }, + { + "epoch": 4.476256212037549, + "grad_norm": 0.9560808539390564, + "learning_rate": 7.711401311266143e-07, + "loss": 0.5595, + "step": 5407 + }, + { + "epoch": 4.477084483710657, + "grad_norm": 1.1302635669708252, + "learning_rate": 7.70345954947381e-07, + "loss": 0.5524, + "step": 5408 + }, + { + "epoch": 4.477912755383766, + "grad_norm": 1.0062146186828613, + "learning_rate": 7.695521134400824e-07, + "loss": 0.5599, + "step": 5409 + }, + { + "epoch": 4.478741027056874, + "grad_norm": 0.9748486876487732, + "learning_rate": 7.687586067583184e-07, + "loss": 0.5543, + "step": 5410 + }, + { + "epoch": 4.4795692987299836, + "grad_norm": 0.9799855351448059, + "learning_rate": 7.679654350556254e-07, + "loss": 0.5529, + "step": 5411 + }, + { + "epoch": 4.480397570403092, + "grad_norm": 1.004991888999939, + "learning_rate": 7.671725984854755e-07, + "loss": 0.5644, + "step": 5412 + }, + { + "epoch": 4.481225842076201, + "grad_norm": 0.9852890968322754, + "learning_rate": 7.663800972012742e-07, + "loss": 0.5555, + "step": 5413 + }, + { + "epoch": 4.48205411374931, + "grad_norm": 0.9725239872932434, + "learning_rate": 7.655879313563633e-07, + "loss": 0.5464, + "step": 5414 + }, + { + "epoch": 4.4828823854224185, + "grad_norm": 1.0972882509231567, + "learning_rate": 7.647961011040203e-07, + "loss": 0.5434, + "step": 5415 + }, + { + "epoch": 4.483710657095528, + "grad_norm": 0.9756212830543518, + "learning_rate": 7.640046065974557e-07, + "loss": 0.5448, + "step": 5416 + }, + { + "epoch": 4.484538928768636, + "grad_norm": 0.9724202156066895, + "learning_rate": 7.632134479898178e-07, + "loss": 0.5635, + "step": 5417 + }, + { + "epoch": 4.485367200441745, + "grad_norm": 0.9435184001922607, + "learning_rate": 7.624226254341879e-07, + "loss": 0.546, + "step": 5418 + }, + { + "epoch": 4.486195472114853, + "grad_norm": 0.9800364375114441, + "learning_rate": 7.616321390835812e-07, + "loss": 0.5463, + "step": 5419 + }, + { + "epoch": 4.4870237437879625, + "grad_norm": 0.9564050436019897, + "learning_rate": 7.608419890909511e-07, + "loss": 0.5391, + "step": 5420 + }, + { + "epoch": 4.487852015461071, + "grad_norm": 0.9649475812911987, + "learning_rate": 7.600521756091844e-07, + "loss": 0.5421, + "step": 5421 + }, + { + "epoch": 4.48868028713418, + "grad_norm": 0.9621299505233765, + "learning_rate": 7.592626987911014e-07, + "loss": 0.5669, + "step": 5422 + }, + { + "epoch": 4.489508558807289, + "grad_norm": 0.9605808854103088, + "learning_rate": 7.584735587894595e-07, + "loss": 0.5658, + "step": 5423 + }, + { + "epoch": 4.4903368304803974, + "grad_norm": 0.9706793427467346, + "learning_rate": 7.576847557569503e-07, + "loss": 0.5629, + "step": 5424 + }, + { + "epoch": 4.491165102153507, + "grad_norm": 0.9814993143081665, + "learning_rate": 7.568962898461982e-07, + "loss": 0.534, + "step": 5425 + }, + { + "epoch": 4.491993373826615, + "grad_norm": 0.9362326264381409, + "learning_rate": 7.561081612097654e-07, + "loss": 0.5428, + "step": 5426 + }, + { + "epoch": 4.492821645499724, + "grad_norm": 0.9745808243751526, + "learning_rate": 7.553203700001474e-07, + "loss": 0.5524, + "step": 5427 + }, + { + "epoch": 4.493649917172832, + "grad_norm": 0.9991053938865662, + "learning_rate": 7.545329163697737e-07, + "loss": 0.5646, + "step": 5428 + }, + { + "epoch": 4.4944781888459415, + "grad_norm": 1.0213414430618286, + "learning_rate": 7.537458004710102e-07, + "loss": 0.5481, + "step": 5429 + }, + { + "epoch": 4.49530646051905, + "grad_norm": 0.9790488481521606, + "learning_rate": 7.529590224561553e-07, + "loss": 0.5779, + "step": 5430 + }, + { + "epoch": 4.496134732192159, + "grad_norm": 0.9708014130592346, + "learning_rate": 7.521725824774446e-07, + "loss": 0.5319, + "step": 5431 + }, + { + "epoch": 4.496963003865268, + "grad_norm": 0.9768827557563782, + "learning_rate": 7.513864806870453e-07, + "loss": 0.5468, + "step": 5432 + }, + { + "epoch": 4.497791275538376, + "grad_norm": 0.9755817651748657, + "learning_rate": 7.506007172370617e-07, + "loss": 0.5634, + "step": 5433 + }, + { + "epoch": 4.498619547211486, + "grad_norm": 0.9696100354194641, + "learning_rate": 7.498152922795324e-07, + "loss": 0.5582, + "step": 5434 + }, + { + "epoch": 4.499447818884594, + "grad_norm": 0.9456085562705994, + "learning_rate": 7.490302059664281e-07, + "loss": 0.5484, + "step": 5435 + }, + { + "epoch": 4.500276090557703, + "grad_norm": 0.9718632102012634, + "learning_rate": 7.482454584496568e-07, + "loss": 0.5486, + "step": 5436 + }, + { + "epoch": 4.501104362230811, + "grad_norm": 0.9685472249984741, + "learning_rate": 7.474610498810603e-07, + "loss": 0.5474, + "step": 5437 + }, + { + "epoch": 4.5019326339039205, + "grad_norm": 0.9661766886711121, + "learning_rate": 7.466769804124127e-07, + "loss": 0.5565, + "step": 5438 + }, + { + "epoch": 4.502760905577029, + "grad_norm": 0.9711228013038635, + "learning_rate": 7.458932501954255e-07, + "loss": 0.5381, + "step": 5439 + }, + { + "epoch": 4.503589177250138, + "grad_norm": 0.9743756651878357, + "learning_rate": 7.451098593817444e-07, + "loss": 0.5748, + "step": 5440 + }, + { + "epoch": 4.504417448923247, + "grad_norm": 0.9695295691490173, + "learning_rate": 7.443268081229449e-07, + "loss": 0.5378, + "step": 5441 + }, + { + "epoch": 4.505245720596355, + "grad_norm": 0.9656387567520142, + "learning_rate": 7.435440965705417e-07, + "loss": 0.5512, + "step": 5442 + }, + { + "epoch": 4.506073992269465, + "grad_norm": 0.9670572280883789, + "learning_rate": 7.427617248759833e-07, + "loss": 0.556, + "step": 5443 + }, + { + "epoch": 4.506902263942573, + "grad_norm": 0.9564972519874573, + "learning_rate": 7.419796931906495e-07, + "loss": 0.5607, + "step": 5444 + }, + { + "epoch": 4.507730535615682, + "grad_norm": 0.9703754186630249, + "learning_rate": 7.411980016658571e-07, + "loss": 0.5441, + "step": 5445 + }, + { + "epoch": 4.508558807288791, + "grad_norm": 0.9711794853210449, + "learning_rate": 7.404166504528567e-07, + "loss": 0.5814, + "step": 5446 + }, + { + "epoch": 4.5093870789618995, + "grad_norm": 0.9656071066856384, + "learning_rate": 7.396356397028312e-07, + "loss": 0.5602, + "step": 5447 + }, + { + "epoch": 4.510215350635009, + "grad_norm": 1.0016793012619019, + "learning_rate": 7.388549695668995e-07, + "loss": 0.5596, + "step": 5448 + }, + { + "epoch": 4.511043622308117, + "grad_norm": 1.0137665271759033, + "learning_rate": 7.380746401961148e-07, + "loss": 0.5636, + "step": 5449 + }, + { + "epoch": 4.511871893981226, + "grad_norm": 0.9805740118026733, + "learning_rate": 7.372946517414617e-07, + "loss": 0.5566, + "step": 5450 + }, + { + "epoch": 4.512700165654334, + "grad_norm": 0.9614541530609131, + "learning_rate": 7.36515004353863e-07, + "loss": 0.5648, + "step": 5451 + }, + { + "epoch": 4.513528437327444, + "grad_norm": 0.9508614540100098, + "learning_rate": 7.357356981841707e-07, + "loss": 0.5227, + "step": 5452 + }, + { + "epoch": 4.514356709000552, + "grad_norm": 0.9625178575515747, + "learning_rate": 7.349567333831756e-07, + "loss": 0.5543, + "step": 5453 + }, + { + "epoch": 4.515184980673661, + "grad_norm": 0.9476574063301086, + "learning_rate": 7.341781101015982e-07, + "loss": 0.5722, + "step": 5454 + }, + { + "epoch": 4.51601325234677, + "grad_norm": 0.9598817229270935, + "learning_rate": 7.333998284900959e-07, + "loss": 0.5511, + "step": 5455 + }, + { + "epoch": 4.5168415240198785, + "grad_norm": 0.9833248257637024, + "learning_rate": 7.326218886992592e-07, + "loss": 0.5399, + "step": 5456 + }, + { + "epoch": 4.517669795692988, + "grad_norm": 0.9664196372032166, + "learning_rate": 7.318442908796111e-07, + "loss": 0.5452, + "step": 5457 + }, + { + "epoch": 4.518498067366096, + "grad_norm": 0.9677484035491943, + "learning_rate": 7.31067035181611e-07, + "loss": 0.5477, + "step": 5458 + }, + { + "epoch": 4.519326339039205, + "grad_norm": 0.9622910618782043, + "learning_rate": 7.302901217556491e-07, + "loss": 0.5396, + "step": 5459 + }, + { + "epoch": 4.520154610712313, + "grad_norm": 0.9736341238021851, + "learning_rate": 7.295135507520515e-07, + "loss": 0.564, + "step": 5460 + }, + { + "epoch": 4.520982882385423, + "grad_norm": 0.9707585573196411, + "learning_rate": 7.287373223210784e-07, + "loss": 0.571, + "step": 5461 + }, + { + "epoch": 4.521811154058531, + "grad_norm": 0.9776362776756287, + "learning_rate": 7.279614366129212e-07, + "loss": 0.5601, + "step": 5462 + }, + { + "epoch": 4.52263942573164, + "grad_norm": 0.9470260143280029, + "learning_rate": 7.271858937777079e-07, + "loss": 0.5559, + "step": 5463 + }, + { + "epoch": 4.523467697404749, + "grad_norm": 0.9702407121658325, + "learning_rate": 7.264106939654975e-07, + "loss": 0.556, + "step": 5464 + }, + { + "epoch": 4.5242959690778575, + "grad_norm": 0.9597034454345703, + "learning_rate": 7.256358373262856e-07, + "loss": 0.5383, + "step": 5465 + }, + { + "epoch": 4.525124240750967, + "grad_norm": 0.9662848114967346, + "learning_rate": 7.248613240099978e-07, + "loss": 0.5414, + "step": 5466 + }, + { + "epoch": 4.525952512424075, + "grad_norm": 0.993077278137207, + "learning_rate": 7.240871541664959e-07, + "loss": 0.554, + "step": 5467 + }, + { + "epoch": 4.526780784097184, + "grad_norm": 0.9801212549209595, + "learning_rate": 7.233133279455762e-07, + "loss": 0.5546, + "step": 5468 + }, + { + "epoch": 4.527609055770292, + "grad_norm": 0.9710840582847595, + "learning_rate": 7.225398454969643e-07, + "loss": 0.5506, + "step": 5469 + }, + { + "epoch": 4.5284373274434015, + "grad_norm": 1.0106468200683594, + "learning_rate": 7.21766706970323e-07, + "loss": 0.5908, + "step": 5470 + }, + { + "epoch": 4.52926559911651, + "grad_norm": 0.9547988176345825, + "learning_rate": 7.209939125152482e-07, + "loss": 0.5497, + "step": 5471 + }, + { + "epoch": 4.530093870789619, + "grad_norm": 0.9706175923347473, + "learning_rate": 7.20221462281267e-07, + "loss": 0.5431, + "step": 5472 + }, + { + "epoch": 4.530922142462728, + "grad_norm": 0.9547169804573059, + "learning_rate": 7.19449356417842e-07, + "loss": 0.5584, + "step": 5473 + }, + { + "epoch": 4.5317504141358365, + "grad_norm": 0.9862712621688843, + "learning_rate": 7.1867759507437e-07, + "loss": 0.5695, + "step": 5474 + }, + { + "epoch": 4.532578685808946, + "grad_norm": 0.9707930088043213, + "learning_rate": 7.179061784001765e-07, + "loss": 0.5775, + "step": 5475 + }, + { + "epoch": 4.533406957482054, + "grad_norm": 0.957129716873169, + "learning_rate": 7.171351065445248e-07, + "loss": 0.5423, + "step": 5476 + }, + { + "epoch": 4.534235229155163, + "grad_norm": 0.965663492679596, + "learning_rate": 7.163643796566114e-07, + "loss": 0.5407, + "step": 5477 + }, + { + "epoch": 4.535063500828271, + "grad_norm": 0.940671980381012, + "learning_rate": 7.155939978855628e-07, + "loss": 0.5352, + "step": 5478 + }, + { + "epoch": 4.5358917725013805, + "grad_norm": 0.9513495564460754, + "learning_rate": 7.148239613804414e-07, + "loss": 0.5542, + "step": 5479 + }, + { + "epoch": 4.536720044174489, + "grad_norm": 0.9566417932510376, + "learning_rate": 7.140542702902434e-07, + "loss": 0.5448, + "step": 5480 + }, + { + "epoch": 4.537548315847598, + "grad_norm": 0.9857562780380249, + "learning_rate": 7.132849247638948e-07, + "loss": 0.5536, + "step": 5481 + }, + { + "epoch": 4.538376587520707, + "grad_norm": 0.9420795440673828, + "learning_rate": 7.125159249502575e-07, + "loss": 0.5536, + "step": 5482 + }, + { + "epoch": 4.5392048591938154, + "grad_norm": 0.9628307223320007, + "learning_rate": 7.117472709981272e-07, + "loss": 0.5537, + "step": 5483 + }, + { + "epoch": 4.540033130866925, + "grad_norm": 0.9649658799171448, + "learning_rate": 7.109789630562289e-07, + "loss": 0.544, + "step": 5484 + }, + { + "epoch": 4.540861402540033, + "grad_norm": 0.9789855480194092, + "learning_rate": 7.102110012732253e-07, + "loss": 0.5374, + "step": 5485 + }, + { + "epoch": 4.541689674213142, + "grad_norm": 0.9994000196456909, + "learning_rate": 7.094433857977079e-07, + "loss": 0.5763, + "step": 5486 + }, + { + "epoch": 4.54251794588625, + "grad_norm": 0.944028377532959, + "learning_rate": 7.086761167782047e-07, + "loss": 0.5378, + "step": 5487 + }, + { + "epoch": 4.5433462175593595, + "grad_norm": 0.9605673551559448, + "learning_rate": 7.079091943631736e-07, + "loss": 0.5587, + "step": 5488 + }, + { + "epoch": 4.544174489232468, + "grad_norm": 0.9778985977172852, + "learning_rate": 7.071426187010077e-07, + "loss": 0.5614, + "step": 5489 + }, + { + "epoch": 4.545002760905577, + "grad_norm": 0.9588051438331604, + "learning_rate": 7.063763899400331e-07, + "loss": 0.5262, + "step": 5490 + }, + { + "epoch": 4.545831032578686, + "grad_norm": 0.9682207703590393, + "learning_rate": 7.056105082285064e-07, + "loss": 0.5702, + "step": 5491 + }, + { + "epoch": 4.546659304251794, + "grad_norm": 0.9890053868293762, + "learning_rate": 7.048449737146193e-07, + "loss": 0.5693, + "step": 5492 + }, + { + "epoch": 4.547487575924904, + "grad_norm": 0.9506924748420715, + "learning_rate": 7.040797865464962e-07, + "loss": 0.5465, + "step": 5493 + }, + { + "epoch": 4.548315847598012, + "grad_norm": 0.9579111337661743, + "learning_rate": 7.033149468721923e-07, + "loss": 0.5505, + "step": 5494 + }, + { + "epoch": 4.549144119271121, + "grad_norm": 0.964435875415802, + "learning_rate": 7.025504548396978e-07, + "loss": 0.5543, + "step": 5495 + }, + { + "epoch": 4.549972390944229, + "grad_norm": 0.992845892906189, + "learning_rate": 7.017863105969352e-07, + "loss": 0.552, + "step": 5496 + }, + { + "epoch": 4.5508006626173385, + "grad_norm": 0.9823659062385559, + "learning_rate": 7.01022514291759e-07, + "loss": 0.5497, + "step": 5497 + }, + { + "epoch": 4.551628934290447, + "grad_norm": 0.9746332168579102, + "learning_rate": 7.002590660719555e-07, + "loss": 0.5413, + "step": 5498 + }, + { + "epoch": 4.552457205963556, + "grad_norm": 0.9899871945381165, + "learning_rate": 6.994959660852468e-07, + "loss": 0.5539, + "step": 5499 + }, + { + "epoch": 4.553285477636665, + "grad_norm": 0.9836088418960571, + "learning_rate": 6.987332144792835e-07, + "loss": 0.5674, + "step": 5500 + }, + { + "epoch": 4.554113749309773, + "grad_norm": 0.9729539155960083, + "learning_rate": 6.979708114016522e-07, + "loss": 0.5408, + "step": 5501 + }, + { + "epoch": 4.554942020982883, + "grad_norm": 0.978412389755249, + "learning_rate": 6.972087569998712e-07, + "loss": 0.564, + "step": 5502 + }, + { + "epoch": 4.555770292655991, + "grad_norm": 0.9849222898483276, + "learning_rate": 6.964470514213897e-07, + "loss": 0.5481, + "step": 5503 + }, + { + "epoch": 4.5565985643291, + "grad_norm": 0.9477882981300354, + "learning_rate": 6.956856948135912e-07, + "loss": 0.5522, + "step": 5504 + }, + { + "epoch": 4.557426836002209, + "grad_norm": 0.9613835215568542, + "learning_rate": 6.949246873237919e-07, + "loss": 0.5729, + "step": 5505 + }, + { + "epoch": 4.5582551076753175, + "grad_norm": 0.9619837999343872, + "learning_rate": 6.94164029099238e-07, + "loss": 0.5546, + "step": 5506 + }, + { + "epoch": 4.559083379348427, + "grad_norm": 1.0136449337005615, + "learning_rate": 6.934037202871108e-07, + "loss": 0.5553, + "step": 5507 + }, + { + "epoch": 4.559911651021535, + "grad_norm": 0.9942994713783264, + "learning_rate": 6.926437610345238e-07, + "loss": 0.5715, + "step": 5508 + }, + { + "epoch": 4.560739922694644, + "grad_norm": 0.9909749627113342, + "learning_rate": 6.918841514885207e-07, + "loss": 0.587, + "step": 5509 + }, + { + "epoch": 4.561568194367752, + "grad_norm": 0.9991593956947327, + "learning_rate": 6.911248917960789e-07, + "loss": 0.5667, + "step": 5510 + }, + { + "epoch": 4.562396466040862, + "grad_norm": 0.985620379447937, + "learning_rate": 6.903659821041081e-07, + "loss": 0.5611, + "step": 5511 + }, + { + "epoch": 4.56322473771397, + "grad_norm": 1.0072076320648193, + "learning_rate": 6.896074225594516e-07, + "loss": 0.5507, + "step": 5512 + }, + { + "epoch": 4.564053009387079, + "grad_norm": 0.9806503653526306, + "learning_rate": 6.888492133088815e-07, + "loss": 0.5385, + "step": 5513 + }, + { + "epoch": 4.564881281060188, + "grad_norm": 0.9733349084854126, + "learning_rate": 6.880913544991053e-07, + "loss": 0.5541, + "step": 5514 + }, + { + "epoch": 4.5657095527332965, + "grad_norm": 0.9819521307945251, + "learning_rate": 6.873338462767625e-07, + "loss": 0.5493, + "step": 5515 + }, + { + "epoch": 4.566537824406406, + "grad_norm": 0.9954928755760193, + "learning_rate": 6.865766887884223e-07, + "loss": 0.5562, + "step": 5516 + }, + { + "epoch": 4.567366096079514, + "grad_norm": 0.9706395268440247, + "learning_rate": 6.858198821805881e-07, + "loss": 0.5497, + "step": 5517 + }, + { + "epoch": 4.568194367752623, + "grad_norm": 0.9409603476524353, + "learning_rate": 6.85063426599696e-07, + "loss": 0.5417, + "step": 5518 + }, + { + "epoch": 4.569022639425731, + "grad_norm": 0.9698299169540405, + "learning_rate": 6.843073221921123e-07, + "loss": 0.5625, + "step": 5519 + }, + { + "epoch": 4.569850911098841, + "grad_norm": 0.9876602292060852, + "learning_rate": 6.835515691041353e-07, + "loss": 0.5581, + "step": 5520 + }, + { + "epoch": 4.570679182771949, + "grad_norm": 0.9799392223358154, + "learning_rate": 6.827961674819978e-07, + "loss": 0.5645, + "step": 5521 + }, + { + "epoch": 4.571507454445058, + "grad_norm": 0.9740294814109802, + "learning_rate": 6.820411174718614e-07, + "loss": 0.5512, + "step": 5522 + }, + { + "epoch": 4.572335726118167, + "grad_norm": 0.9723674058914185, + "learning_rate": 6.812864192198215e-07, + "loss": 0.5708, + "step": 5523 + }, + { + "epoch": 4.5731639977912755, + "grad_norm": 0.9980106353759766, + "learning_rate": 6.80532072871907e-07, + "loss": 0.5579, + "step": 5524 + }, + { + "epoch": 4.573992269464385, + "grad_norm": 0.9684682488441467, + "learning_rate": 6.797780785740743e-07, + "loss": 0.5629, + "step": 5525 + }, + { + "epoch": 4.574820541137493, + "grad_norm": 0.9702059030532837, + "learning_rate": 6.790244364722159e-07, + "loss": 0.5804, + "step": 5526 + }, + { + "epoch": 4.575648812810602, + "grad_norm": 0.9612674117088318, + "learning_rate": 6.782711467121547e-07, + "loss": 0.5518, + "step": 5527 + }, + { + "epoch": 4.57647708448371, + "grad_norm": 1.0216538906097412, + "learning_rate": 6.775182094396437e-07, + "loss": 0.5791, + "step": 5528 + }, + { + "epoch": 4.5773053561568195, + "grad_norm": 0.9898926615715027, + "learning_rate": 6.767656248003707e-07, + "loss": 0.5666, + "step": 5529 + }, + { + "epoch": 4.578133627829928, + "grad_norm": 0.9707671999931335, + "learning_rate": 6.760133929399537e-07, + "loss": 0.5506, + "step": 5530 + }, + { + "epoch": 4.578961899503037, + "grad_norm": 0.9658564329147339, + "learning_rate": 6.752615140039423e-07, + "loss": 0.5451, + "step": 5531 + }, + { + "epoch": 4.579790171176146, + "grad_norm": 0.9721407890319824, + "learning_rate": 6.745099881378172e-07, + "loss": 0.5634, + "step": 5532 + }, + { + "epoch": 4.5806184428492545, + "grad_norm": 0.9759280681610107, + "learning_rate": 6.737588154869931e-07, + "loss": 0.5537, + "step": 5533 + }, + { + "epoch": 4.581446714522364, + "grad_norm": 0.9694047570228577, + "learning_rate": 6.730079961968137e-07, + "loss": 0.5396, + "step": 5534 + }, + { + "epoch": 4.582274986195472, + "grad_norm": 0.9477902054786682, + "learning_rate": 6.722575304125555e-07, + "loss": 0.5487, + "step": 5535 + }, + { + "epoch": 4.583103257868581, + "grad_norm": 0.9674749970436096, + "learning_rate": 6.715074182794284e-07, + "loss": 0.5562, + "step": 5536 + }, + { + "epoch": 4.583931529541689, + "grad_norm": 0.9692550897598267, + "learning_rate": 6.707576599425699e-07, + "loss": 0.5463, + "step": 5537 + }, + { + "epoch": 4.5847598012147985, + "grad_norm": 0.9627196192741394, + "learning_rate": 6.700082555470524e-07, + "loss": 0.5505, + "step": 5538 + }, + { + "epoch": 4.585588072887907, + "grad_norm": 0.9593527317047119, + "learning_rate": 6.692592052378791e-07, + "loss": 0.5494, + "step": 5539 + }, + { + "epoch": 4.586416344561016, + "grad_norm": 0.9435111284255981, + "learning_rate": 6.685105091599828e-07, + "loss": 0.5467, + "step": 5540 + }, + { + "epoch": 4.587244616234125, + "grad_norm": 0.9593883752822876, + "learning_rate": 6.677621674582302e-07, + "loss": 0.5393, + "step": 5541 + }, + { + "epoch": 4.588072887907233, + "grad_norm": 0.9862588047981262, + "learning_rate": 6.670141802774191e-07, + "loss": 0.5478, + "step": 5542 + }, + { + "epoch": 4.588901159580343, + "grad_norm": 0.9542217254638672, + "learning_rate": 6.662665477622771e-07, + "loss": 0.5518, + "step": 5543 + }, + { + "epoch": 4.589729431253451, + "grad_norm": 0.9312442541122437, + "learning_rate": 6.655192700574634e-07, + "loss": 0.554, + "step": 5544 + }, + { + "epoch": 4.59055770292656, + "grad_norm": 0.9759663939476013, + "learning_rate": 6.647723473075704e-07, + "loss": 0.5635, + "step": 5545 + }, + { + "epoch": 4.591385974599668, + "grad_norm": 0.9525055885314941, + "learning_rate": 6.640257796571209e-07, + "loss": 0.5448, + "step": 5546 + }, + { + "epoch": 4.5922142462727775, + "grad_norm": 0.9484573006629944, + "learning_rate": 6.632795672505679e-07, + "loss": 0.5516, + "step": 5547 + }, + { + "epoch": 4.593042517945886, + "grad_norm": 0.9842866659164429, + "learning_rate": 6.625337102322968e-07, + "loss": 0.532, + "step": 5548 + }, + { + "epoch": 4.593870789618995, + "grad_norm": 0.9838387966156006, + "learning_rate": 6.617882087466246e-07, + "loss": 0.5666, + "step": 5549 + }, + { + "epoch": 4.594699061292104, + "grad_norm": 0.9601597785949707, + "learning_rate": 6.610430629377979e-07, + "loss": 0.5707, + "step": 5550 + }, + { + "epoch": 4.595527332965212, + "grad_norm": 0.9840213656425476, + "learning_rate": 6.60298272949996e-07, + "loss": 0.5284, + "step": 5551 + }, + { + "epoch": 4.596355604638322, + "grad_norm": 0.9841850399971008, + "learning_rate": 6.595538389273295e-07, + "loss": 0.5475, + "step": 5552 + }, + { + "epoch": 4.59718387631143, + "grad_norm": 0.9589581489562988, + "learning_rate": 6.588097610138389e-07, + "loss": 0.5595, + "step": 5553 + }, + { + "epoch": 4.598012147984539, + "grad_norm": 0.9751944541931152, + "learning_rate": 6.580660393534952e-07, + "loss": 0.5515, + "step": 5554 + }, + { + "epoch": 4.598840419657647, + "grad_norm": 0.9598148465156555, + "learning_rate": 6.573226740902036e-07, + "loss": 0.542, + "step": 5555 + }, + { + "epoch": 4.5996686913307565, + "grad_norm": 0.9537079334259033, + "learning_rate": 6.565796653677963e-07, + "loss": 0.5457, + "step": 5556 + }, + { + "epoch": 4.600496963003865, + "grad_norm": 0.9721077084541321, + "learning_rate": 6.558370133300399e-07, + "loss": 0.5399, + "step": 5557 + }, + { + "epoch": 4.601325234676974, + "grad_norm": 0.9512110352516174, + "learning_rate": 6.55094718120631e-07, + "loss": 0.5554, + "step": 5558 + }, + { + "epoch": 4.602153506350083, + "grad_norm": 0.9435778260231018, + "learning_rate": 6.543527798831953e-07, + "loss": 0.5322, + "step": 5559 + }, + { + "epoch": 4.602981778023191, + "grad_norm": 0.9703747630119324, + "learning_rate": 6.536111987612917e-07, + "loss": 0.547, + "step": 5560 + }, + { + "epoch": 4.603810049696301, + "grad_norm": 0.9578705430030823, + "learning_rate": 6.528699748984104e-07, + "loss": 0.5513, + "step": 5561 + }, + { + "epoch": 4.604638321369409, + "grad_norm": 0.9871172904968262, + "learning_rate": 6.521291084379694e-07, + "loss": 0.5675, + "step": 5562 + }, + { + "epoch": 4.605466593042518, + "grad_norm": 0.9615850448608398, + "learning_rate": 6.513885995233202e-07, + "loss": 0.54, + "step": 5563 + }, + { + "epoch": 4.606294864715627, + "grad_norm": 0.9759035706520081, + "learning_rate": 6.506484482977452e-07, + "loss": 0.5438, + "step": 5564 + }, + { + "epoch": 4.6071231363887355, + "grad_norm": 0.935772180557251, + "learning_rate": 6.499086549044559e-07, + "loss": 0.5596, + "step": 5565 + }, + { + "epoch": 4.607951408061845, + "grad_norm": 0.977527379989624, + "learning_rate": 6.491692194865948e-07, + "loss": 0.5715, + "step": 5566 + }, + { + "epoch": 4.608779679734953, + "grad_norm": 0.9544543027877808, + "learning_rate": 6.484301421872366e-07, + "loss": 0.547, + "step": 5567 + }, + { + "epoch": 4.609607951408062, + "grad_norm": 0.9892758727073669, + "learning_rate": 6.476914231493866e-07, + "loss": 0.5672, + "step": 5568 + }, + { + "epoch": 4.61043622308117, + "grad_norm": 1.0188134908676147, + "learning_rate": 6.469530625159784e-07, + "loss": 0.5587, + "step": 5569 + }, + { + "epoch": 4.61126449475428, + "grad_norm": 0.9610130786895752, + "learning_rate": 6.462150604298786e-07, + "loss": 0.5603, + "step": 5570 + }, + { + "epoch": 4.612092766427388, + "grad_norm": 1.0074714422225952, + "learning_rate": 6.454774170338848e-07, + "loss": 0.5585, + "step": 5571 + }, + { + "epoch": 4.612921038100497, + "grad_norm": 0.9753648042678833, + "learning_rate": 6.447401324707225e-07, + "loss": 0.5524, + "step": 5572 + }, + { + "epoch": 4.613749309773606, + "grad_norm": 0.9903519749641418, + "learning_rate": 6.440032068830498e-07, + "loss": 0.5504, + "step": 5573 + }, + { + "epoch": 4.6145775814467145, + "grad_norm": 0.9634335041046143, + "learning_rate": 6.43266640413456e-07, + "loss": 0.5393, + "step": 5574 + }, + { + "epoch": 4.615405853119824, + "grad_norm": 0.9743342399597168, + "learning_rate": 6.425304332044582e-07, + "loss": 0.5501, + "step": 5575 + }, + { + "epoch": 4.616234124792932, + "grad_norm": 0.9500157833099365, + "learning_rate": 6.417945853985075e-07, + "loss": 0.5477, + "step": 5576 + }, + { + "epoch": 4.617062396466041, + "grad_norm": 0.9560253024101257, + "learning_rate": 6.410590971379824e-07, + "loss": 0.5294, + "step": 5577 + }, + { + "epoch": 4.617890668139149, + "grad_norm": 0.9748409986495972, + "learning_rate": 6.403239685651924e-07, + "loss": 0.5491, + "step": 5578 + }, + { + "epoch": 4.618718939812259, + "grad_norm": 0.9598991870880127, + "learning_rate": 6.395891998223789e-07, + "loss": 0.5512, + "step": 5579 + }, + { + "epoch": 4.619547211485367, + "grad_norm": 0.9680323600769043, + "learning_rate": 6.388547910517132e-07, + "loss": 0.5562, + "step": 5580 + }, + { + "epoch": 4.620375483158476, + "grad_norm": 0.9817914962768555, + "learning_rate": 6.381207423952954e-07, + "loss": 0.5548, + "step": 5581 + }, + { + "epoch": 4.621203754831585, + "grad_norm": 0.9678810238838196, + "learning_rate": 6.373870539951579e-07, + "loss": 0.5406, + "step": 5582 + }, + { + "epoch": 4.6220320265046935, + "grad_norm": 0.9937217831611633, + "learning_rate": 6.366537259932632e-07, + "loss": 0.5747, + "step": 5583 + }, + { + "epoch": 4.622860298177803, + "grad_norm": 1.0302647352218628, + "learning_rate": 6.359207585315017e-07, + "loss": 0.5531, + "step": 5584 + }, + { + "epoch": 4.623688569850911, + "grad_norm": 0.9746495485305786, + "learning_rate": 6.351881517516967e-07, + "loss": 0.5631, + "step": 5585 + }, + { + "epoch": 4.62451684152402, + "grad_norm": 0.9883660674095154, + "learning_rate": 6.344559057956015e-07, + "loss": 0.5582, + "step": 5586 + }, + { + "epoch": 4.625345113197128, + "grad_norm": 0.9649250507354736, + "learning_rate": 6.337240208048981e-07, + "loss": 0.5473, + "step": 5587 + }, + { + "epoch": 4.6261733848702375, + "grad_norm": 0.9564082026481628, + "learning_rate": 6.32992496921199e-07, + "loss": 0.5494, + "step": 5588 + }, + { + "epoch": 4.627001656543346, + "grad_norm": 0.9802419543266296, + "learning_rate": 6.322613342860473e-07, + "loss": 0.5321, + "step": 5589 + }, + { + "epoch": 4.627829928216455, + "grad_norm": 0.9529916048049927, + "learning_rate": 6.315305330409177e-07, + "loss": 0.5621, + "step": 5590 + }, + { + "epoch": 4.628658199889564, + "grad_norm": 0.9516423940658569, + "learning_rate": 6.308000933272116e-07, + "loss": 0.5643, + "step": 5591 + }, + { + "epoch": 4.6294864715626725, + "grad_norm": 0.9464498162269592, + "learning_rate": 6.300700152862627e-07, + "loss": 0.5325, + "step": 5592 + }, + { + "epoch": 4.630314743235782, + "grad_norm": 0.9634289145469666, + "learning_rate": 6.293402990593353e-07, + "loss": 0.5707, + "step": 5593 + }, + { + "epoch": 4.63114301490889, + "grad_norm": 0.9790555238723755, + "learning_rate": 6.286109447876215e-07, + "loss": 0.5601, + "step": 5594 + }, + { + "epoch": 4.631971286581999, + "grad_norm": 0.9650810956954956, + "learning_rate": 6.278819526122454e-07, + "loss": 0.546, + "step": 5595 + }, + { + "epoch": 4.632799558255107, + "grad_norm": 0.9842014908790588, + "learning_rate": 6.271533226742593e-07, + "loss": 0.5522, + "step": 5596 + }, + { + "epoch": 4.6336278299282165, + "grad_norm": 0.9452126026153564, + "learning_rate": 6.26425055114647e-07, + "loss": 0.5368, + "step": 5597 + }, + { + "epoch": 4.634456101601325, + "grad_norm": 0.9557516574859619, + "learning_rate": 6.256971500743217e-07, + "loss": 0.5514, + "step": 5598 + }, + { + "epoch": 4.635284373274434, + "grad_norm": 0.9769173860549927, + "learning_rate": 6.249696076941258e-07, + "loss": 0.5548, + "step": 5599 + }, + { + "epoch": 4.636112644947543, + "grad_norm": 0.9568111300468445, + "learning_rate": 6.242424281148316e-07, + "loss": 0.5624, + "step": 5600 + }, + { + "epoch": 4.636940916620651, + "grad_norm": 0.955300509929657, + "learning_rate": 6.235156114771418e-07, + "loss": 0.5501, + "step": 5601 + }, + { + "epoch": 4.637769188293761, + "grad_norm": 0.9530934691429138, + "learning_rate": 6.227891579216894e-07, + "loss": 0.5439, + "step": 5602 + }, + { + "epoch": 4.638597459966869, + "grad_norm": 0.9716370105743408, + "learning_rate": 6.220630675890352e-07, + "loss": 0.5705, + "step": 5603 + }, + { + "epoch": 4.639425731639978, + "grad_norm": 0.9733977913856506, + "learning_rate": 6.213373406196715e-07, + "loss": 0.5738, + "step": 5604 + }, + { + "epoch": 4.640254003313086, + "grad_norm": 0.9514371156692505, + "learning_rate": 6.206119771540206e-07, + "loss": 0.5748, + "step": 5605 + }, + { + "epoch": 4.6410822749861955, + "grad_norm": 0.9908952116966248, + "learning_rate": 6.198869773324317e-07, + "loss": 0.5474, + "step": 5606 + }, + { + "epoch": 4.641910546659304, + "grad_norm": 0.9772365093231201, + "learning_rate": 6.191623412951867e-07, + "loss": 0.5749, + "step": 5607 + }, + { + "epoch": 4.642738818332413, + "grad_norm": 0.9424340724945068, + "learning_rate": 6.184380691824965e-07, + "loss": 0.572, + "step": 5608 + }, + { + "epoch": 4.643567090005522, + "grad_norm": 0.9462697505950928, + "learning_rate": 6.177141611344991e-07, + "loss": 0.5551, + "step": 5609 + }, + { + "epoch": 4.64439536167863, + "grad_norm": 0.9748279452323914, + "learning_rate": 6.169906172912663e-07, + "loss": 0.5593, + "step": 5610 + }, + { + "epoch": 4.64522363335174, + "grad_norm": 0.9854201674461365, + "learning_rate": 6.162674377927955e-07, + "loss": 0.5698, + "step": 5611 + }, + { + "epoch": 4.646051905024848, + "grad_norm": 0.9688649773597717, + "learning_rate": 6.155446227790149e-07, + "loss": 0.5599, + "step": 5612 + }, + { + "epoch": 4.646880176697957, + "grad_norm": 0.9476209282875061, + "learning_rate": 6.148221723897832e-07, + "loss": 0.552, + "step": 5613 + }, + { + "epoch": 4.647708448371065, + "grad_norm": 0.93233722448349, + "learning_rate": 6.141000867648886e-07, + "loss": 0.5412, + "step": 5614 + }, + { + "epoch": 4.6485367200441745, + "grad_norm": 0.9635641574859619, + "learning_rate": 6.133783660440462e-07, + "loss": 0.5609, + "step": 5615 + }, + { + "epoch": 4.649364991717283, + "grad_norm": 0.9933269619941711, + "learning_rate": 6.12657010366903e-07, + "loss": 0.5547, + "step": 5616 + }, + { + "epoch": 4.650193263390392, + "grad_norm": 0.9645413160324097, + "learning_rate": 6.119360198730356e-07, + "loss": 0.5609, + "step": 5617 + }, + { + "epoch": 4.651021535063501, + "grad_norm": 0.9551970958709717, + "learning_rate": 6.112153947019472e-07, + "loss": 0.5404, + "step": 5618 + }, + { + "epoch": 4.651849806736609, + "grad_norm": 0.9647742509841919, + "learning_rate": 6.104951349930729e-07, + "loss": 0.5276, + "step": 5619 + }, + { + "epoch": 4.652678078409719, + "grad_norm": 0.9530763626098633, + "learning_rate": 6.097752408857771e-07, + "loss": 0.5302, + "step": 5620 + }, + { + "epoch": 4.653506350082827, + "grad_norm": 0.9838370084762573, + "learning_rate": 6.090557125193517e-07, + "loss": 0.5747, + "step": 5621 + }, + { + "epoch": 4.654334621755936, + "grad_norm": 0.9567121267318726, + "learning_rate": 6.08336550033018e-07, + "loss": 0.5533, + "step": 5622 + }, + { + "epoch": 4.655162893429045, + "grad_norm": 0.9746866226196289, + "learning_rate": 6.076177535659277e-07, + "loss": 0.5673, + "step": 5623 + }, + { + "epoch": 4.6559911651021535, + "grad_norm": 0.9545463919639587, + "learning_rate": 6.068993232571626e-07, + "loss": 0.5521, + "step": 5624 + }, + { + "epoch": 4.656819436775263, + "grad_norm": 0.9727249145507812, + "learning_rate": 6.061812592457306e-07, + "loss": 0.5767, + "step": 5625 + }, + { + "epoch": 4.657647708448371, + "grad_norm": 0.9679794311523438, + "learning_rate": 6.054635616705709e-07, + "loss": 0.5522, + "step": 5626 + }, + { + "epoch": 4.65847598012148, + "grad_norm": 0.9779345989227295, + "learning_rate": 6.047462306705517e-07, + "loss": 0.551, + "step": 5627 + }, + { + "epoch": 4.659304251794588, + "grad_norm": 0.9675984978675842, + "learning_rate": 6.040292663844693e-07, + "loss": 0.5495, + "step": 5628 + }, + { + "epoch": 4.660132523467698, + "grad_norm": 0.9447121024131775, + "learning_rate": 6.033126689510497e-07, + "loss": 0.5644, + "step": 5629 + }, + { + "epoch": 4.660960795140806, + "grad_norm": 0.9525550603866577, + "learning_rate": 6.025964385089486e-07, + "loss": 0.5621, + "step": 5630 + }, + { + "epoch": 4.661789066813915, + "grad_norm": 0.9803587794303894, + "learning_rate": 6.018805751967486e-07, + "loss": 0.5582, + "step": 5631 + }, + { + "epoch": 4.662617338487024, + "grad_norm": 0.9682264924049377, + "learning_rate": 6.011650791529639e-07, + "loss": 0.5561, + "step": 5632 + }, + { + "epoch": 4.6634456101601325, + "grad_norm": 0.9648387432098389, + "learning_rate": 6.004499505160358e-07, + "loss": 0.5404, + "step": 5633 + }, + { + "epoch": 4.664273881833242, + "grad_norm": 0.9990679621696472, + "learning_rate": 5.997351894243342e-07, + "loss": 0.5422, + "step": 5634 + }, + { + "epoch": 4.66510215350635, + "grad_norm": 0.9632382392883301, + "learning_rate": 5.990207960161592e-07, + "loss": 0.5237, + "step": 5635 + }, + { + "epoch": 4.665930425179459, + "grad_norm": 0.9712291359901428, + "learning_rate": 5.983067704297402e-07, + "loss": 0.5606, + "step": 5636 + }, + { + "epoch": 4.666758696852567, + "grad_norm": 0.977373480796814, + "learning_rate": 5.975931128032333e-07, + "loss": 0.5453, + "step": 5637 + }, + { + "epoch": 4.667586968525677, + "grad_norm": 1.0031154155731201, + "learning_rate": 5.968798232747247e-07, + "loss": 0.5286, + "step": 5638 + }, + { + "epoch": 4.668415240198785, + "grad_norm": 0.9799690842628479, + "learning_rate": 5.961669019822308e-07, + "loss": 0.5579, + "step": 5639 + }, + { + "epoch": 4.669243511871894, + "grad_norm": 1.0083366632461548, + "learning_rate": 5.95454349063693e-07, + "loss": 0.563, + "step": 5640 + }, + { + "epoch": 4.670071783545003, + "grad_norm": 0.9788227677345276, + "learning_rate": 5.947421646569849e-07, + "loss": 0.5449, + "step": 5641 + }, + { + "epoch": 4.6709000552181115, + "grad_norm": 0.9708746075630188, + "learning_rate": 5.94030348899908e-07, + "loss": 0.545, + "step": 5642 + }, + { + "epoch": 4.671728326891221, + "grad_norm": 0.9622349739074707, + "learning_rate": 5.933189019301913e-07, + "loss": 0.533, + "step": 5643 + }, + { + "epoch": 4.672556598564329, + "grad_norm": 0.9786980748176575, + "learning_rate": 5.926078238854923e-07, + "loss": 0.545, + "step": 5644 + }, + { + "epoch": 4.673384870237438, + "grad_norm": 0.9637975692749023, + "learning_rate": 5.918971149033995e-07, + "loss": 0.5411, + "step": 5645 + }, + { + "epoch": 4.674213141910546, + "grad_norm": 0.9689448475837708, + "learning_rate": 5.911867751214281e-07, + "loss": 0.5511, + "step": 5646 + }, + { + "epoch": 4.6750414135836555, + "grad_norm": 0.9778165817260742, + "learning_rate": 5.904768046770212e-07, + "loss": 0.5557, + "step": 5647 + }, + { + "epoch": 4.675869685256764, + "grad_norm": 0.9706706404685974, + "learning_rate": 5.897672037075524e-07, + "loss": 0.5687, + "step": 5648 + }, + { + "epoch": 4.676697956929873, + "grad_norm": 0.976809024810791, + "learning_rate": 5.890579723503237e-07, + "loss": 0.5519, + "step": 5649 + }, + { + "epoch": 4.677526228602982, + "grad_norm": 0.9677285552024841, + "learning_rate": 5.883491107425627e-07, + "loss": 0.5425, + "step": 5650 + }, + { + "epoch": 4.6783545002760905, + "grad_norm": 0.9553424715995789, + "learning_rate": 5.876406190214287e-07, + "loss": 0.5632, + "step": 5651 + }, + { + "epoch": 4.6791827719492, + "grad_norm": 0.9804576635360718, + "learning_rate": 5.869324973240089e-07, + "loss": 0.5499, + "step": 5652 + }, + { + "epoch": 4.680011043622308, + "grad_norm": 0.9667906165122986, + "learning_rate": 5.862247457873169e-07, + "loss": 0.5482, + "step": 5653 + }, + { + "epoch": 4.680839315295417, + "grad_norm": 0.960593581199646, + "learning_rate": 5.855173645482964e-07, + "loss": 0.5531, + "step": 5654 + }, + { + "epoch": 4.681667586968525, + "grad_norm": 0.9586433172225952, + "learning_rate": 5.848103537438207e-07, + "loss": 0.5774, + "step": 5655 + }, + { + "epoch": 4.6824958586416345, + "grad_norm": 0.9638173580169678, + "learning_rate": 5.841037135106869e-07, + "loss": 0.565, + "step": 5656 + }, + { + "epoch": 4.683324130314743, + "grad_norm": 1.0111193656921387, + "learning_rate": 5.833974439856249e-07, + "loss": 0.5263, + "step": 5657 + }, + { + "epoch": 4.684152401987852, + "grad_norm": 0.9522552490234375, + "learning_rate": 5.826915453052917e-07, + "loss": 0.553, + "step": 5658 + }, + { + "epoch": 4.684980673660961, + "grad_norm": 0.9515351057052612, + "learning_rate": 5.819860176062711e-07, + "loss": 0.571, + "step": 5659 + }, + { + "epoch": 4.685808945334069, + "grad_norm": 1.0340934991836548, + "learning_rate": 5.812808610250764e-07, + "loss": 0.5719, + "step": 5660 + }, + { + "epoch": 4.686637217007179, + "grad_norm": 0.9896745085716248, + "learning_rate": 5.805760756981499e-07, + "loss": 0.544, + "step": 5661 + }, + { + "epoch": 4.687465488680287, + "grad_norm": 0.9758293032646179, + "learning_rate": 5.798716617618594e-07, + "loss": 0.5498, + "step": 5662 + }, + { + "epoch": 4.688293760353396, + "grad_norm": 0.9765741229057312, + "learning_rate": 5.791676193525028e-07, + "loss": 0.5572, + "step": 5663 + }, + { + "epoch": 4.689122032026504, + "grad_norm": 0.9680903553962708, + "learning_rate": 5.784639486063074e-07, + "loss": 0.5389, + "step": 5664 + }, + { + "epoch": 4.6899503036996135, + "grad_norm": 0.9511458277702332, + "learning_rate": 5.777606496594243e-07, + "loss": 0.5507, + "step": 5665 + }, + { + "epoch": 4.690778575372722, + "grad_norm": 0.9745240807533264, + "learning_rate": 5.770577226479376e-07, + "loss": 0.5655, + "step": 5666 + }, + { + "epoch": 4.691606847045831, + "grad_norm": 0.9594129323959351, + "learning_rate": 5.763551677078555e-07, + "loss": 0.5649, + "step": 5667 + }, + { + "epoch": 4.69243511871894, + "grad_norm": 0.9827647805213928, + "learning_rate": 5.756529849751169e-07, + "loss": 0.5441, + "step": 5668 + }, + { + "epoch": 4.693263390392048, + "grad_norm": 0.9480031728744507, + "learning_rate": 5.749511745855865e-07, + "loss": 0.5646, + "step": 5669 + }, + { + "epoch": 4.694091662065158, + "grad_norm": 0.9795644879341125, + "learning_rate": 5.74249736675059e-07, + "loss": 0.5422, + "step": 5670 + }, + { + "epoch": 4.694919933738266, + "grad_norm": 0.9643570184707642, + "learning_rate": 5.735486713792565e-07, + "loss": 0.5587, + "step": 5671 + }, + { + "epoch": 4.695748205411375, + "grad_norm": 0.9786714911460876, + "learning_rate": 5.72847978833827e-07, + "loss": 0.5518, + "step": 5672 + }, + { + "epoch": 4.696576477084483, + "grad_norm": 0.964820384979248, + "learning_rate": 5.721476591743496e-07, + "loss": 0.552, + "step": 5673 + }, + { + "epoch": 4.6974047487575925, + "grad_norm": 0.9419858455657959, + "learning_rate": 5.714477125363286e-07, + "loss": 0.535, + "step": 5674 + }, + { + "epoch": 4.698233020430701, + "grad_norm": 0.9434426426887512, + "learning_rate": 5.707481390551972e-07, + "loss": 0.5422, + "step": 5675 + }, + { + "epoch": 4.69906129210381, + "grad_norm": 0.9646450281143188, + "learning_rate": 5.700489388663175e-07, + "loss": 0.5538, + "step": 5676 + }, + { + "epoch": 4.699889563776919, + "grad_norm": 0.9789668917655945, + "learning_rate": 5.693501121049771e-07, + "loss": 0.546, + "step": 5677 + }, + { + "epoch": 4.700717835450027, + "grad_norm": 0.9620424509048462, + "learning_rate": 5.686516589063923e-07, + "loss": 0.5382, + "step": 5678 + }, + { + "epoch": 4.701546107123137, + "grad_norm": 0.9448021650314331, + "learning_rate": 5.679535794057076e-07, + "loss": 0.5388, + "step": 5679 + }, + { + "epoch": 4.702374378796245, + "grad_norm": 1.0032944679260254, + "learning_rate": 5.672558737379954e-07, + "loss": 0.5493, + "step": 5680 + }, + { + "epoch": 4.703202650469354, + "grad_norm": 0.9643270969390869, + "learning_rate": 5.665585420382546e-07, + "loss": 0.5355, + "step": 5681 + }, + { + "epoch": 4.704030922142463, + "grad_norm": 0.9836111068725586, + "learning_rate": 5.658615844414123e-07, + "loss": 0.5765, + "step": 5682 + }, + { + "epoch": 4.7048591938155715, + "grad_norm": 0.9687516093254089, + "learning_rate": 5.651650010823245e-07, + "loss": 0.5542, + "step": 5683 + }, + { + "epoch": 4.705687465488681, + "grad_norm": 0.9528695344924927, + "learning_rate": 5.644687920957717e-07, + "loss": 0.56, + "step": 5684 + }, + { + "epoch": 4.706515737161789, + "grad_norm": 0.9862529635429382, + "learning_rate": 5.637729576164652e-07, + "loss": 0.5448, + "step": 5685 + }, + { + "epoch": 4.707344008834898, + "grad_norm": 0.9494380950927734, + "learning_rate": 5.630774977790426e-07, + "loss": 0.543, + "step": 5686 + }, + { + "epoch": 4.708172280508006, + "grad_norm": 0.9758495688438416, + "learning_rate": 5.623824127180677e-07, + "loss": 0.5686, + "step": 5687 + }, + { + "epoch": 4.709000552181116, + "grad_norm": 0.9538042545318604, + "learning_rate": 5.616877025680337e-07, + "loss": 0.5763, + "step": 5688 + }, + { + "epoch": 4.709828823854224, + "grad_norm": 0.961929440498352, + "learning_rate": 5.609933674633619e-07, + "loss": 0.5375, + "step": 5689 + }, + { + "epoch": 4.710657095527333, + "grad_norm": 0.9387208819389343, + "learning_rate": 5.602994075383971e-07, + "loss": 0.5475, + "step": 5690 + }, + { + "epoch": 4.711485367200442, + "grad_norm": 0.9944952726364136, + "learning_rate": 5.596058229274152e-07, + "loss": 0.5577, + "step": 5691 + }, + { + "epoch": 4.7123136388735505, + "grad_norm": 0.9691250920295715, + "learning_rate": 5.589126137646192e-07, + "loss": 0.5538, + "step": 5692 + }, + { + "epoch": 4.71314191054666, + "grad_norm": 0.9680503010749817, + "learning_rate": 5.582197801841369e-07, + "loss": 0.5859, + "step": 5693 + }, + { + "epoch": 4.713970182219768, + "grad_norm": 0.9708214402198792, + "learning_rate": 5.575273223200265e-07, + "loss": 0.556, + "step": 5694 + }, + { + "epoch": 4.714798453892877, + "grad_norm": 0.989654541015625, + "learning_rate": 5.568352403062721e-07, + "loss": 0.5754, + "step": 5695 + }, + { + "epoch": 4.715626725565985, + "grad_norm": 0.9700330495834351, + "learning_rate": 5.56143534276784e-07, + "loss": 0.5514, + "step": 5696 + }, + { + "epoch": 4.716454997239095, + "grad_norm": 0.9729028940200806, + "learning_rate": 5.554522043654017e-07, + "loss": 0.562, + "step": 5697 + }, + { + "epoch": 4.717283268912203, + "grad_norm": 0.9634988307952881, + "learning_rate": 5.54761250705892e-07, + "loss": 0.5461, + "step": 5698 + }, + { + "epoch": 4.718111540585312, + "grad_norm": 1.1048837900161743, + "learning_rate": 5.54070673431946e-07, + "loss": 0.5833, + "step": 5699 + }, + { + "epoch": 4.718939812258421, + "grad_norm": 0.9602149724960327, + "learning_rate": 5.533804726771858e-07, + "loss": 0.5383, + "step": 5700 + }, + { + "epoch": 4.7197680839315295, + "grad_norm": 0.9645217061042786, + "learning_rate": 5.52690648575157e-07, + "loss": 0.5617, + "step": 5701 + }, + { + "epoch": 4.720596355604639, + "grad_norm": 0.9855457544326782, + "learning_rate": 5.520012012593362e-07, + "loss": 0.5646, + "step": 5702 + }, + { + "epoch": 4.721424627277747, + "grad_norm": 0.9794975519180298, + "learning_rate": 5.513121308631231e-07, + "loss": 0.5651, + "step": 5703 + }, + { + "epoch": 4.722252898950856, + "grad_norm": 0.9841457605361938, + "learning_rate": 5.506234375198472e-07, + "loss": 0.5477, + "step": 5704 + }, + { + "epoch": 4.723081170623964, + "grad_norm": 0.9675942063331604, + "learning_rate": 5.499351213627649e-07, + "loss": 0.5559, + "step": 5705 + }, + { + "epoch": 4.7239094422970735, + "grad_norm": 0.9632468223571777, + "learning_rate": 5.492471825250579e-07, + "loss": 0.562, + "step": 5706 + }, + { + "epoch": 4.724737713970182, + "grad_norm": 0.9672477841377258, + "learning_rate": 5.485596211398361e-07, + "loss": 0.5453, + "step": 5707 + }, + { + "epoch": 4.725565985643291, + "grad_norm": 0.9824864268302917, + "learning_rate": 5.478724373401372e-07, + "loss": 0.5599, + "step": 5708 + }, + { + "epoch": 4.7263942573164, + "grad_norm": 0.9423843026161194, + "learning_rate": 5.471856312589238e-07, + "loss": 0.5567, + "step": 5709 + }, + { + "epoch": 4.7272225289895085, + "grad_norm": 0.9732121229171753, + "learning_rate": 5.464992030290866e-07, + "loss": 0.5575, + "step": 5710 + }, + { + "epoch": 4.728050800662618, + "grad_norm": 0.9717659950256348, + "learning_rate": 5.458131527834448e-07, + "loss": 0.5816, + "step": 5711 + }, + { + "epoch": 4.728879072335726, + "grad_norm": 0.9714568853378296, + "learning_rate": 5.451274806547399e-07, + "loss": 0.5222, + "step": 5712 + }, + { + "epoch": 4.729707344008835, + "grad_norm": 0.9828087091445923, + "learning_rate": 5.444421867756441e-07, + "loss": 0.5726, + "step": 5713 + }, + { + "epoch": 4.730535615681943, + "grad_norm": 0.9719827771186829, + "learning_rate": 5.437572712787567e-07, + "loss": 0.5437, + "step": 5714 + }, + { + "epoch": 4.7313638873550525, + "grad_norm": 1.0056298971176147, + "learning_rate": 5.430727342966008e-07, + "loss": 0.5396, + "step": 5715 + }, + { + "epoch": 4.732192159028161, + "grad_norm": 0.9678478837013245, + "learning_rate": 5.423885759616282e-07, + "loss": 0.5491, + "step": 5716 + }, + { + "epoch": 4.73302043070127, + "grad_norm": 0.9862358570098877, + "learning_rate": 5.417047964062185e-07, + "loss": 0.5332, + "step": 5717 + }, + { + "epoch": 4.733848702374379, + "grad_norm": 0.9755412936210632, + "learning_rate": 5.41021395762675e-07, + "loss": 0.5643, + "step": 5718 + }, + { + "epoch": 4.734676974047487, + "grad_norm": 0.9771009683609009, + "learning_rate": 5.403383741632301e-07, + "loss": 0.5684, + "step": 5719 + }, + { + "epoch": 4.735505245720597, + "grad_norm": 0.9530584812164307, + "learning_rate": 5.396557317400424e-07, + "loss": 0.5446, + "step": 5720 + }, + { + "epoch": 4.736333517393705, + "grad_norm": 0.9681975841522217, + "learning_rate": 5.389734686251962e-07, + "loss": 0.5377, + "step": 5721 + }, + { + "epoch": 4.737161789066814, + "grad_norm": 0.9460469484329224, + "learning_rate": 5.382915849507034e-07, + "loss": 0.5613, + "step": 5722 + }, + { + "epoch": 4.737990060739922, + "grad_norm": 0.9434804320335388, + "learning_rate": 5.376100808485027e-07, + "loss": 0.5524, + "step": 5723 + }, + { + "epoch": 4.7388183324130315, + "grad_norm": 0.9815676212310791, + "learning_rate": 5.36928956450458e-07, + "loss": 0.5479, + "step": 5724 + }, + { + "epoch": 4.73964660408614, + "grad_norm": 0.9858399629592896, + "learning_rate": 5.362482118883602e-07, + "loss": 0.5633, + "step": 5725 + }, + { + "epoch": 4.740474875759249, + "grad_norm": 0.9802773594856262, + "learning_rate": 5.355678472939274e-07, + "loss": 0.5624, + "step": 5726 + }, + { + "epoch": 4.741303147432358, + "grad_norm": 0.9784073829650879, + "learning_rate": 5.348878627988047e-07, + "loss": 0.539, + "step": 5727 + }, + { + "epoch": 4.742131419105466, + "grad_norm": 0.9714169502258301, + "learning_rate": 5.342082585345615e-07, + "loss": 0.558, + "step": 5728 + }, + { + "epoch": 4.742959690778576, + "grad_norm": 0.9876227378845215, + "learning_rate": 5.335290346326952e-07, + "loss": 0.5471, + "step": 5729 + }, + { + "epoch": 4.743787962451684, + "grad_norm": 0.9695437550544739, + "learning_rate": 5.3285019122463e-07, + "loss": 0.5268, + "step": 5730 + }, + { + "epoch": 4.744616234124793, + "grad_norm": 0.9693663120269775, + "learning_rate": 5.32171728441715e-07, + "loss": 0.561, + "step": 5731 + }, + { + "epoch": 4.745444505797901, + "grad_norm": 0.9851357340812683, + "learning_rate": 5.314936464152265e-07, + "loss": 0.5424, + "step": 5732 + }, + { + "epoch": 4.7462727774710105, + "grad_norm": 0.9912320375442505, + "learning_rate": 5.308159452763678e-07, + "loss": 0.563, + "step": 5733 + }, + { + "epoch": 4.747101049144119, + "grad_norm": 0.9956056475639343, + "learning_rate": 5.301386251562674e-07, + "loss": 0.541, + "step": 5734 + }, + { + "epoch": 4.747929320817228, + "grad_norm": 0.9793626666069031, + "learning_rate": 5.294616861859794e-07, + "loss": 0.5488, + "step": 5735 + }, + { + "epoch": 4.748757592490337, + "grad_norm": 0.9587200880050659, + "learning_rate": 5.287851284964865e-07, + "loss": 0.5235, + "step": 5736 + }, + { + "epoch": 4.749585864163445, + "grad_norm": 0.9694571495056152, + "learning_rate": 5.281089522186955e-07, + "loss": 0.5439, + "step": 5737 + }, + { + "epoch": 4.750414135836555, + "grad_norm": 0.9679348468780518, + "learning_rate": 5.274331574834402e-07, + "loss": 0.5468, + "step": 5738 + }, + { + "epoch": 4.751242407509663, + "grad_norm": 0.9581620097160339, + "learning_rate": 5.267577444214819e-07, + "loss": 0.5477, + "step": 5739 + }, + { + "epoch": 4.752070679182772, + "grad_norm": 0.9385791420936584, + "learning_rate": 5.260827131635046e-07, + "loss": 0.5375, + "step": 5740 + }, + { + "epoch": 4.752898950855881, + "grad_norm": 0.9797219634056091, + "learning_rate": 5.254080638401221e-07, + "loss": 0.5565, + "step": 5741 + }, + { + "epoch": 4.7537272225289895, + "grad_norm": 0.9667260646820068, + "learning_rate": 5.247337965818727e-07, + "loss": 0.532, + "step": 5742 + }, + { + "epoch": 4.754555494202099, + "grad_norm": 0.9527150392532349, + "learning_rate": 5.240599115192199e-07, + "loss": 0.5384, + "step": 5743 + }, + { + "epoch": 4.755383765875207, + "grad_norm": 0.9563965201377869, + "learning_rate": 5.23386408782555e-07, + "loss": 0.573, + "step": 5744 + }, + { + "epoch": 4.756212037548316, + "grad_norm": 0.9735510349273682, + "learning_rate": 5.227132885021946e-07, + "loss": 0.5746, + "step": 5745 + }, + { + "epoch": 4.757040309221424, + "grad_norm": 0.9680667519569397, + "learning_rate": 5.220405508083809e-07, + "loss": 0.5615, + "step": 5746 + }, + { + "epoch": 4.757868580894534, + "grad_norm": 0.9715480804443359, + "learning_rate": 5.213681958312817e-07, + "loss": 0.5411, + "step": 5747 + }, + { + "epoch": 4.758696852567642, + "grad_norm": 0.9722656011581421, + "learning_rate": 5.206962237009927e-07, + "loss": 0.5612, + "step": 5748 + }, + { + "epoch": 4.759525124240751, + "grad_norm": 0.9685169458389282, + "learning_rate": 5.200246345475329e-07, + "loss": 0.5602, + "step": 5749 + }, + { + "epoch": 4.76035339591386, + "grad_norm": 0.9914273023605347, + "learning_rate": 5.193534285008492e-07, + "loss": 0.5775, + "step": 5750 + }, + { + "epoch": 4.7611816675869685, + "grad_norm": 1.0119380950927734, + "learning_rate": 5.186826056908146e-07, + "loss": 0.5682, + "step": 5751 + }, + { + "epoch": 4.762009939260078, + "grad_norm": 0.9605881571769714, + "learning_rate": 5.180121662472251e-07, + "loss": 0.5348, + "step": 5752 + }, + { + "epoch": 4.762838210933186, + "grad_norm": 0.9948599934577942, + "learning_rate": 5.173421102998058e-07, + "loss": 0.5665, + "step": 5753 + }, + { + "epoch": 4.763666482606295, + "grad_norm": 0.9638116955757141, + "learning_rate": 5.166724379782067e-07, + "loss": 0.5482, + "step": 5754 + }, + { + "epoch": 4.764494754279403, + "grad_norm": 0.9764114022254944, + "learning_rate": 5.160031494120021e-07, + "loss": 0.57, + "step": 5755 + }, + { + "epoch": 4.765323025952513, + "grad_norm": 0.9698774218559265, + "learning_rate": 5.153342447306939e-07, + "loss": 0.5474, + "step": 5756 + }, + { + "epoch": 4.766151297625621, + "grad_norm": 0.9637375473976135, + "learning_rate": 5.146657240637079e-07, + "loss": 0.5461, + "step": 5757 + }, + { + "epoch": 4.76697956929873, + "grad_norm": 0.9572364091873169, + "learning_rate": 5.139975875403979e-07, + "loss": 0.5437, + "step": 5758 + }, + { + "epoch": 4.767807840971839, + "grad_norm": 0.9601276516914368, + "learning_rate": 5.13329835290041e-07, + "loss": 0.5599, + "step": 5759 + }, + { + "epoch": 4.7686361126449475, + "grad_norm": 0.9581212997436523, + "learning_rate": 5.12662467441841e-07, + "loss": 0.5467, + "step": 5760 + }, + { + "epoch": 4.769464384318057, + "grad_norm": 0.977631688117981, + "learning_rate": 5.119954841249288e-07, + "loss": 0.5216, + "step": 5761 + }, + { + "epoch": 4.770292655991165, + "grad_norm": 0.9808089733123779, + "learning_rate": 5.113288854683579e-07, + "loss": 0.5711, + "step": 5762 + }, + { + "epoch": 4.771120927664274, + "grad_norm": 0.9823766946792603, + "learning_rate": 5.106626716011093e-07, + "loss": 0.5605, + "step": 5763 + }, + { + "epoch": 4.771949199337382, + "grad_norm": 0.962883710861206, + "learning_rate": 5.099968426520902e-07, + "loss": 0.532, + "step": 5764 + }, + { + "epoch": 4.7727774710104915, + "grad_norm": 0.9629702568054199, + "learning_rate": 5.093313987501306e-07, + "loss": 0.5343, + "step": 5765 + }, + { + "epoch": 4.7736057426836, + "grad_norm": 0.9748992323875427, + "learning_rate": 5.086663400239886e-07, + "loss": 0.5577, + "step": 5766 + }, + { + "epoch": 4.774434014356709, + "grad_norm": 0.9772101640701294, + "learning_rate": 5.080016666023476e-07, + "loss": 0.5626, + "step": 5767 + }, + { + "epoch": 4.775262286029818, + "grad_norm": 0.9644051790237427, + "learning_rate": 5.07337378613815e-07, + "loss": 0.5598, + "step": 5768 + }, + { + "epoch": 4.7760905577029265, + "grad_norm": 0.9621604681015015, + "learning_rate": 5.066734761869235e-07, + "loss": 0.5408, + "step": 5769 + }, + { + "epoch": 4.776918829376036, + "grad_norm": 0.9538135528564453, + "learning_rate": 5.060099594501336e-07, + "loss": 0.555, + "step": 5770 + }, + { + "epoch": 4.777747101049144, + "grad_norm": 0.9480177760124207, + "learning_rate": 5.05346828531828e-07, + "loss": 0.5527, + "step": 5771 + }, + { + "epoch": 4.778575372722253, + "grad_norm": 0.9961148500442505, + "learning_rate": 5.04684083560317e-07, + "loss": 0.5682, + "step": 5772 + }, + { + "epoch": 4.779403644395361, + "grad_norm": 0.962980329990387, + "learning_rate": 5.040217246638368e-07, + "loss": 0.5622, + "step": 5773 + }, + { + "epoch": 4.7802319160684705, + "grad_norm": 0.9808009266853333, + "learning_rate": 5.033597519705457e-07, + "loss": 0.563, + "step": 5774 + }, + { + "epoch": 4.781060187741579, + "grad_norm": 0.9771943688392639, + "learning_rate": 5.0269816560853e-07, + "loss": 0.5694, + "step": 5775 + }, + { + "epoch": 4.781888459414688, + "grad_norm": 0.9685245752334595, + "learning_rate": 5.020369657058016e-07, + "loss": 0.5542, + "step": 5776 + }, + { + "epoch": 4.782716731087797, + "grad_norm": 0.9657905697822571, + "learning_rate": 5.013761523902949e-07, + "loss": 0.5695, + "step": 5777 + }, + { + "epoch": 4.783545002760905, + "grad_norm": 0.9332074522972107, + "learning_rate": 5.007157257898715e-07, + "loss": 0.5392, + "step": 5778 + }, + { + "epoch": 4.784373274434015, + "grad_norm": 0.9893274307250977, + "learning_rate": 5.000556860323191e-07, + "loss": 0.5357, + "step": 5779 + }, + { + "epoch": 4.785201546107123, + "grad_norm": 0.9868603348731995, + "learning_rate": 4.993960332453479e-07, + "loss": 0.5525, + "step": 5780 + }, + { + "epoch": 4.786029817780232, + "grad_norm": 0.9733468294143677, + "learning_rate": 4.987367675565943e-07, + "loss": 0.5552, + "step": 5781 + }, + { + "epoch": 4.78685808945334, + "grad_norm": 0.9812783598899841, + "learning_rate": 4.980778890936208e-07, + "loss": 0.5407, + "step": 5782 + }, + { + "epoch": 4.7876863611264495, + "grad_norm": 0.9803352355957031, + "learning_rate": 4.974193979839146e-07, + "loss": 0.5321, + "step": 5783 + }, + { + "epoch": 4.788514632799558, + "grad_norm": 0.9762402176856995, + "learning_rate": 4.967612943548863e-07, + "loss": 0.5537, + "step": 5784 + }, + { + "epoch": 4.789342904472667, + "grad_norm": 0.955674409866333, + "learning_rate": 4.961035783338739e-07, + "loss": 0.5286, + "step": 5785 + }, + { + "epoch": 4.790171176145776, + "grad_norm": 0.9545698761940002, + "learning_rate": 4.954462500481394e-07, + "loss": 0.5869, + "step": 5786 + }, + { + "epoch": 4.790999447818884, + "grad_norm": 0.9490817189216614, + "learning_rate": 4.947893096248688e-07, + "loss": 0.5462, + "step": 5787 + }, + { + "epoch": 4.791827719491994, + "grad_norm": 0.971339225769043, + "learning_rate": 4.941327571911742e-07, + "loss": 0.5651, + "step": 5788 + }, + { + "epoch": 4.792655991165102, + "grad_norm": 0.9914401173591614, + "learning_rate": 4.934765928740933e-07, + "loss": 0.5396, + "step": 5789 + }, + { + "epoch": 4.793484262838211, + "grad_norm": 1.0018508434295654, + "learning_rate": 4.928208168005874e-07, + "loss": 0.5297, + "step": 5790 + }, + { + "epoch": 4.794312534511319, + "grad_norm": 0.9785000085830688, + "learning_rate": 4.92165429097542e-07, + "loss": 0.56, + "step": 5791 + }, + { + "epoch": 4.7951408061844285, + "grad_norm": 0.9545918703079224, + "learning_rate": 4.9151042989177e-07, + "loss": 0.549, + "step": 5792 + }, + { + "epoch": 4.795969077857537, + "grad_norm": 0.9696705341339111, + "learning_rate": 4.908558193100061e-07, + "loss": 0.5548, + "step": 5793 + }, + { + "epoch": 4.796797349530646, + "grad_norm": 0.9759218692779541, + "learning_rate": 4.902015974789121e-07, + "loss": 0.5634, + "step": 5794 + }, + { + "epoch": 4.797625621203755, + "grad_norm": 0.9847009181976318, + "learning_rate": 4.895477645250749e-07, + "loss": 0.5636, + "step": 5795 + }, + { + "epoch": 4.798453892876863, + "grad_norm": 1.0215561389923096, + "learning_rate": 4.88894320575003e-07, + "loss": 0.5572, + "step": 5796 + }, + { + "epoch": 4.799282164549973, + "grad_norm": 0.9767036437988281, + "learning_rate": 4.882412657551331e-07, + "loss": 0.5488, + "step": 5797 + }, + { + "epoch": 4.800110436223081, + "grad_norm": 0.979062557220459, + "learning_rate": 4.875886001918254e-07, + "loss": 0.5522, + "step": 5798 + }, + { + "epoch": 4.80093870789619, + "grad_norm": 0.9697179794311523, + "learning_rate": 4.869363240113634e-07, + "loss": 0.553, + "step": 5799 + }, + { + "epoch": 4.801766979569299, + "grad_norm": 0.977159321308136, + "learning_rate": 4.862844373399572e-07, + "loss": 0.5451, + "step": 5800 + }, + { + "epoch": 4.8025952512424075, + "grad_norm": 0.9752582907676697, + "learning_rate": 4.856329403037413e-07, + "loss": 0.5296, + "step": 5801 + }, + { + "epoch": 4.803423522915517, + "grad_norm": 0.9821684956550598, + "learning_rate": 4.849818330287739e-07, + "loss": 0.5437, + "step": 5802 + }, + { + "epoch": 4.804251794588625, + "grad_norm": 0.9562684297561646, + "learning_rate": 4.843311156410374e-07, + "loss": 0.577, + "step": 5803 + }, + { + "epoch": 4.805080066261734, + "grad_norm": 0.9621249437332153, + "learning_rate": 4.836807882664401e-07, + "loss": 0.5409, + "step": 5804 + }, + { + "epoch": 4.805908337934842, + "grad_norm": 0.9560291171073914, + "learning_rate": 4.830308510308151e-07, + "loss": 0.5667, + "step": 5805 + }, + { + "epoch": 4.806736609607952, + "grad_norm": 0.9753386378288269, + "learning_rate": 4.823813040599179e-07, + "loss": 0.5615, + "step": 5806 + }, + { + "epoch": 4.80756488128106, + "grad_norm": 0.9691498875617981, + "learning_rate": 4.817321474794304e-07, + "loss": 0.5649, + "step": 5807 + }, + { + "epoch": 4.808393152954169, + "grad_norm": 0.9705379605293274, + "learning_rate": 4.810833814149588e-07, + "loss": 0.5417, + "step": 5808 + }, + { + "epoch": 4.809221424627278, + "grad_norm": 0.9624449014663696, + "learning_rate": 4.804350059920324e-07, + "loss": 0.5458, + "step": 5809 + }, + { + "epoch": 4.8100496963003865, + "grad_norm": 0.9795193672180176, + "learning_rate": 4.797870213361064e-07, + "loss": 0.556, + "step": 5810 + }, + { + "epoch": 4.810877967973496, + "grad_norm": 0.9939634203910828, + "learning_rate": 4.791394275725592e-07, + "loss": 0.5828, + "step": 5811 + }, + { + "epoch": 4.811706239646604, + "grad_norm": 0.9787962436676025, + "learning_rate": 4.784922248266946e-07, + "loss": 0.5475, + "step": 5812 + }, + { + "epoch": 4.812534511319713, + "grad_norm": 1.0147124528884888, + "learning_rate": 4.778454132237409e-07, + "loss": 0.5595, + "step": 5813 + }, + { + "epoch": 4.813362782992821, + "grad_norm": 0.9653933048248291, + "learning_rate": 4.771989928888493e-07, + "loss": 0.552, + "step": 5814 + }, + { + "epoch": 4.814191054665931, + "grad_norm": 0.9682862162590027, + "learning_rate": 4.765529639470956e-07, + "loss": 0.5472, + "step": 5815 + }, + { + "epoch": 4.815019326339039, + "grad_norm": 0.9534293413162231, + "learning_rate": 4.759073265234812e-07, + "loss": 0.5328, + "step": 5816 + }, + { + "epoch": 4.815847598012148, + "grad_norm": 0.9567888379096985, + "learning_rate": 4.7526208074293144e-07, + "loss": 0.53, + "step": 5817 + }, + { + "epoch": 4.816675869685257, + "grad_norm": 0.9659878015518188, + "learning_rate": 4.7461722673029396e-07, + "loss": 0.5409, + "step": 5818 + }, + { + "epoch": 4.8175041413583655, + "grad_norm": 0.979027271270752, + "learning_rate": 4.739727646103426e-07, + "loss": 0.5393, + "step": 5819 + }, + { + "epoch": 4.818332413031475, + "grad_norm": 0.9583582282066345, + "learning_rate": 4.7332869450777586e-07, + "loss": 0.5438, + "step": 5820 + }, + { + "epoch": 4.819160684704583, + "grad_norm": 0.9865652322769165, + "learning_rate": 4.7268501654721357e-07, + "loss": 0.5675, + "step": 5821 + }, + { + "epoch": 4.819988956377692, + "grad_norm": 0.9573209881782532, + "learning_rate": 4.7204173085320185e-07, + "loss": 0.545, + "step": 5822 + }, + { + "epoch": 4.8208172280508, + "grad_norm": 0.9714416861534119, + "learning_rate": 4.713988375502118e-07, + "loss": 0.5579, + "step": 5823 + }, + { + "epoch": 4.8216454997239095, + "grad_norm": 1.0049816370010376, + "learning_rate": 4.707563367626364e-07, + "loss": 0.5413, + "step": 5824 + }, + { + "epoch": 4.822473771397018, + "grad_norm": 0.9911128878593445, + "learning_rate": 4.701142286147925e-07, + "loss": 0.5612, + "step": 5825 + }, + { + "epoch": 4.823302043070127, + "grad_norm": 0.9881139397621155, + "learning_rate": 4.6947251323092383e-07, + "loss": 0.5621, + "step": 5826 + }, + { + "epoch": 4.824130314743236, + "grad_norm": 0.9554377794265747, + "learning_rate": 4.6883119073519503e-07, + "loss": 0.5626, + "step": 5827 + }, + { + "epoch": 4.8249585864163445, + "grad_norm": 0.9953138828277588, + "learning_rate": 4.6819026125169643e-07, + "loss": 0.5553, + "step": 5828 + }, + { + "epoch": 4.825786858089454, + "grad_norm": 0.9436317682266235, + "learning_rate": 4.675497249044425e-07, + "loss": 0.539, + "step": 5829 + }, + { + "epoch": 4.826615129762562, + "grad_norm": 0.9757593870162964, + "learning_rate": 4.6690958181736996e-07, + "loss": 0.5421, + "step": 5830 + }, + { + "epoch": 4.827443401435671, + "grad_norm": 0.9557772278785706, + "learning_rate": 4.6626983211434106e-07, + "loss": 0.5442, + "step": 5831 + }, + { + "epoch": 4.828271673108779, + "grad_norm": 0.9633134007453918, + "learning_rate": 4.6563047591914213e-07, + "loss": 0.5711, + "step": 5832 + }, + { + "epoch": 4.8290999447818885, + "grad_norm": 0.964967668056488, + "learning_rate": 4.649915133554814e-07, + "loss": 0.5645, + "step": 5833 + }, + { + "epoch": 4.829928216454997, + "grad_norm": 0.9668264985084534, + "learning_rate": 4.643529445469924e-07, + "loss": 0.584, + "step": 5834 + }, + { + "epoch": 4.830756488128106, + "grad_norm": 0.9745258688926697, + "learning_rate": 4.6371476961723365e-07, + "loss": 0.5695, + "step": 5835 + }, + { + "epoch": 4.831584759801215, + "grad_norm": 0.9560638070106506, + "learning_rate": 4.6307698868968463e-07, + "loss": 0.5413, + "step": 5836 + }, + { + "epoch": 4.832413031474323, + "grad_norm": 0.9573860764503479, + "learning_rate": 4.6243960188775e-07, + "loss": 0.5645, + "step": 5837 + }, + { + "epoch": 4.833241303147433, + "grad_norm": 0.9806252121925354, + "learning_rate": 4.618026093347583e-07, + "loss": 0.5685, + "step": 5838 + }, + { + "epoch": 4.834069574820541, + "grad_norm": 0.974324107170105, + "learning_rate": 4.611660111539626e-07, + "loss": 0.5593, + "step": 5839 + }, + { + "epoch": 4.83489784649365, + "grad_norm": 0.9738810658454895, + "learning_rate": 4.6052980746853717e-07, + "loss": 0.5492, + "step": 5840 + }, + { + "epoch": 4.835726118166758, + "grad_norm": 0.9708226323127747, + "learning_rate": 4.5989399840158236e-07, + "loss": 0.5081, + "step": 5841 + }, + { + "epoch": 4.8365543898398675, + "grad_norm": 0.984955906867981, + "learning_rate": 4.592585840761221e-07, + "loss": 0.5581, + "step": 5842 + }, + { + "epoch": 4.837382661512976, + "grad_norm": 1.006397008895874, + "learning_rate": 4.586235646151013e-07, + "loss": 0.5536, + "step": 5843 + }, + { + "epoch": 4.838210933186085, + "grad_norm": 0.9361645579338074, + "learning_rate": 4.5798894014139146e-07, + "loss": 0.5606, + "step": 5844 + }, + { + "epoch": 4.839039204859194, + "grad_norm": 0.9568001627922058, + "learning_rate": 4.5735471077778704e-07, + "loss": 0.5242, + "step": 5845 + }, + { + "epoch": 4.839867476532302, + "grad_norm": 0.9710729122161865, + "learning_rate": 4.567208766470041e-07, + "loss": 0.5617, + "step": 5846 + }, + { + "epoch": 4.840695748205412, + "grad_norm": 0.9681148529052734, + "learning_rate": 4.560874378716848e-07, + "loss": 0.5502, + "step": 5847 + }, + { + "epoch": 4.84152401987852, + "grad_norm": 0.9655634164810181, + "learning_rate": 4.5545439457439334e-07, + "loss": 0.5729, + "step": 5848 + }, + { + "epoch": 4.842352291551629, + "grad_norm": 0.9855836033821106, + "learning_rate": 4.548217468776167e-07, + "loss": 0.5413, + "step": 5849 + }, + { + "epoch": 4.843180563224737, + "grad_norm": 0.9914452433586121, + "learning_rate": 4.541894949037673e-07, + "loss": 0.5392, + "step": 5850 + }, + { + "epoch": 4.8440088348978465, + "grad_norm": 0.968498945236206, + "learning_rate": 4.5355763877518057e-07, + "loss": 0.5378, + "step": 5851 + }, + { + "epoch": 4.844837106570955, + "grad_norm": 0.9624379873275757, + "learning_rate": 4.5292617861411315e-07, + "loss": 0.5592, + "step": 5852 + }, + { + "epoch": 4.845665378244064, + "grad_norm": 0.9620005488395691, + "learning_rate": 4.522951145427476e-07, + "loss": 0.5406, + "step": 5853 + }, + { + "epoch": 4.846493649917173, + "grad_norm": 0.9529075026512146, + "learning_rate": 4.5166444668318987e-07, + "loss": 0.5506, + "step": 5854 + }, + { + "epoch": 4.847321921590281, + "grad_norm": 0.9744150638580322, + "learning_rate": 4.510341751574665e-07, + "loss": 0.5573, + "step": 5855 + }, + { + "epoch": 4.848150193263391, + "grad_norm": 0.9540391564369202, + "learning_rate": 4.5040430008752994e-07, + "loss": 0.5433, + "step": 5856 + }, + { + "epoch": 4.848978464936499, + "grad_norm": 0.9721823334693909, + "learning_rate": 4.4977482159525606e-07, + "loss": 0.5534, + "step": 5857 + }, + { + "epoch": 4.849806736609608, + "grad_norm": 0.9671435952186584, + "learning_rate": 4.491457398024421e-07, + "loss": 0.5709, + "step": 5858 + }, + { + "epoch": 4.850635008282717, + "grad_norm": 0.9428475499153137, + "learning_rate": 4.4851705483080904e-07, + "loss": 0.5636, + "step": 5859 + }, + { + "epoch": 4.8514632799558255, + "grad_norm": 0.9652672410011292, + "learning_rate": 4.4788876680200213e-07, + "loss": 0.5506, + "step": 5860 + }, + { + "epoch": 4.852291551628935, + "grad_norm": 0.9987141489982605, + "learning_rate": 4.4726087583759016e-07, + "loss": 0.5303, + "step": 5861 + }, + { + "epoch": 4.853119823302043, + "grad_norm": 0.9547122716903687, + "learning_rate": 4.4663338205906235e-07, + "loss": 0.5568, + "step": 5862 + }, + { + "epoch": 4.853948094975152, + "grad_norm": 0.9808024168014526, + "learning_rate": 4.460062855878342e-07, + "loss": 0.5442, + "step": 5863 + }, + { + "epoch": 4.85477636664826, + "grad_norm": 0.9605748057365417, + "learning_rate": 4.4537958654524317e-07, + "loss": 0.5487, + "step": 5864 + }, + { + "epoch": 4.85560463832137, + "grad_norm": 0.9812973737716675, + "learning_rate": 4.4475328505254877e-07, + "loss": 0.5632, + "step": 5865 + }, + { + "epoch": 4.856432909994478, + "grad_norm": 0.9739463329315186, + "learning_rate": 4.441273812309346e-07, + "loss": 0.5485, + "step": 5866 + }, + { + "epoch": 4.857261181667587, + "grad_norm": 0.9558693170547485, + "learning_rate": 4.4350187520150855e-07, + "loss": 0.5565, + "step": 5867 + }, + { + "epoch": 4.858089453340696, + "grad_norm": 0.9696657061576843, + "learning_rate": 4.4287676708529847e-07, + "loss": 0.5425, + "step": 5868 + }, + { + "epoch": 4.8589177250138045, + "grad_norm": 0.9728769659996033, + "learning_rate": 4.422520570032582e-07, + "loss": 0.5544, + "step": 5869 + }, + { + "epoch": 4.859745996686914, + "grad_norm": 0.9446035623550415, + "learning_rate": 4.416277450762629e-07, + "loss": 0.5422, + "step": 5870 + }, + { + "epoch": 4.860574268360022, + "grad_norm": 0.9597954154014587, + "learning_rate": 4.4100383142511056e-07, + "loss": 0.5414, + "step": 5871 + }, + { + "epoch": 4.861402540033131, + "grad_norm": 0.9638952016830444, + "learning_rate": 4.403803161705228e-07, + "loss": 0.5288, + "step": 5872 + }, + { + "epoch": 4.862230811706239, + "grad_norm": 0.9554084539413452, + "learning_rate": 4.3975719943314535e-07, + "loss": 0.5423, + "step": 5873 + }, + { + "epoch": 4.863059083379349, + "grad_norm": 0.9913690686225891, + "learning_rate": 4.391344813335435e-07, + "loss": 0.5735, + "step": 5874 + }, + { + "epoch": 4.863887355052457, + "grad_norm": 0.9613746404647827, + "learning_rate": 4.3851216199220865e-07, + "loss": 0.5544, + "step": 5875 + }, + { + "epoch": 4.864715626725566, + "grad_norm": 0.9566593766212463, + "learning_rate": 4.378902415295538e-07, + "loss": 0.5506, + "step": 5876 + }, + { + "epoch": 4.865543898398675, + "grad_norm": 0.9866978526115417, + "learning_rate": 4.37268720065914e-07, + "loss": 0.5424, + "step": 5877 + }, + { + "epoch": 4.8663721700717835, + "grad_norm": 0.9751846790313721, + "learning_rate": 4.366475977215484e-07, + "loss": 0.5538, + "step": 5878 + }, + { + "epoch": 4.867200441744893, + "grad_norm": 0.9757003784179688, + "learning_rate": 4.3602687461663904e-07, + "loss": 0.5668, + "step": 5879 + }, + { + "epoch": 4.868028713418001, + "grad_norm": 0.9434172511100769, + "learning_rate": 4.3540655087128846e-07, + "loss": 0.5335, + "step": 5880 + }, + { + "epoch": 4.86885698509111, + "grad_norm": 0.9726729989051819, + "learning_rate": 4.3478662660552504e-07, + "loss": 0.5458, + "step": 5881 + }, + { + "epoch": 4.869685256764218, + "grad_norm": 0.9750825762748718, + "learning_rate": 4.3416710193929733e-07, + "loss": 0.5437, + "step": 5882 + }, + { + "epoch": 4.8705135284373275, + "grad_norm": 0.966927170753479, + "learning_rate": 4.3354797699247833e-07, + "loss": 0.5263, + "step": 5883 + }, + { + "epoch": 4.871341800110436, + "grad_norm": 1.003137230873108, + "learning_rate": 4.329292518848621e-07, + "loss": 0.5589, + "step": 5884 + }, + { + "epoch": 4.872170071783545, + "grad_norm": 0.9725276827812195, + "learning_rate": 4.3231092673616635e-07, + "loss": 0.555, + "step": 5885 + }, + { + "epoch": 4.872998343456654, + "grad_norm": 0.964232861995697, + "learning_rate": 4.3169300166603254e-07, + "loss": 0.5486, + "step": 5886 + }, + { + "epoch": 4.8738266151297625, + "grad_norm": 0.963379442691803, + "learning_rate": 4.3107547679402146e-07, + "loss": 0.5606, + "step": 5887 + }, + { + "epoch": 4.874654886802872, + "grad_norm": 0.9966505765914917, + "learning_rate": 4.3045835223961996e-07, + "loss": 0.5534, + "step": 5888 + }, + { + "epoch": 4.87548315847598, + "grad_norm": 0.9489006400108337, + "learning_rate": 4.298416281222348e-07, + "loss": 0.5332, + "step": 5889 + }, + { + "epoch": 4.876311430149089, + "grad_norm": 0.9371119737625122, + "learning_rate": 4.2922530456119686e-07, + "loss": 0.5489, + "step": 5890 + }, + { + "epoch": 4.877139701822197, + "grad_norm": 0.9697653651237488, + "learning_rate": 4.286093816757597e-07, + "loss": 0.5413, + "step": 5891 + }, + { + "epoch": 4.8779679734953065, + "grad_norm": 0.9692448973655701, + "learning_rate": 4.2799385958509773e-07, + "loss": 0.5739, + "step": 5892 + }, + { + "epoch": 4.878796245168415, + "grad_norm": 0.9299832582473755, + "learning_rate": 4.2737873840830835e-07, + "loss": 0.5394, + "step": 5893 + }, + { + "epoch": 4.879624516841524, + "grad_norm": 0.9768351912498474, + "learning_rate": 4.2676401826441225e-07, + "loss": 0.557, + "step": 5894 + }, + { + "epoch": 4.880452788514633, + "grad_norm": 1.0022796392440796, + "learning_rate": 4.2614969927235297e-07, + "loss": 0.5657, + "step": 5895 + }, + { + "epoch": 4.881281060187741, + "grad_norm": 0.9902967810630798, + "learning_rate": 4.255357815509939e-07, + "loss": 0.5514, + "step": 5896 + }, + { + "epoch": 4.882109331860851, + "grad_norm": 0.9812206029891968, + "learning_rate": 4.2492226521912343e-07, + "loss": 0.5581, + "step": 5897 + }, + { + "epoch": 4.882937603533959, + "grad_norm": 0.9587867856025696, + "learning_rate": 4.2430915039545137e-07, + "loss": 0.5603, + "step": 5898 + }, + { + "epoch": 4.883765875207068, + "grad_norm": 0.9800667762756348, + "learning_rate": 4.236964371986088e-07, + "loss": 0.5491, + "step": 5899 + }, + { + "epoch": 4.884594146880176, + "grad_norm": 0.9664431810379028, + "learning_rate": 4.2308412574715075e-07, + "loss": 0.548, + "step": 5900 + }, + { + "epoch": 4.8854224185532855, + "grad_norm": 1.0057932138442993, + "learning_rate": 4.2247221615955405e-07, + "loss": 0.5391, + "step": 5901 + }, + { + "epoch": 4.886250690226394, + "grad_norm": 0.9711607098579407, + "learning_rate": 4.2186070855421675e-07, + "loss": 0.5397, + "step": 5902 + }, + { + "epoch": 4.887078961899503, + "grad_norm": 0.9427178502082825, + "learning_rate": 4.2124960304946035e-07, + "loss": 0.5356, + "step": 5903 + }, + { + "epoch": 4.887907233572612, + "grad_norm": 0.9526419639587402, + "learning_rate": 4.20638899763528e-07, + "loss": 0.5345, + "step": 5904 + }, + { + "epoch": 4.88873550524572, + "grad_norm": 0.9885551333427429, + "learning_rate": 4.200285988145847e-07, + "loss": 0.5338, + "step": 5905 + }, + { + "epoch": 4.88956377691883, + "grad_norm": 0.9751784205436707, + "learning_rate": 4.1941870032071783e-07, + "loss": 0.5315, + "step": 5906 + }, + { + "epoch": 4.890392048591938, + "grad_norm": 0.9707131385803223, + "learning_rate": 4.1880920439993836e-07, + "loss": 0.5608, + "step": 5907 + }, + { + "epoch": 4.891220320265047, + "grad_norm": 0.9756649136543274, + "learning_rate": 4.1820011117017665e-07, + "loss": 0.5664, + "step": 5908 + }, + { + "epoch": 4.892048591938155, + "grad_norm": 0.977478563785553, + "learning_rate": 4.17591420749287e-07, + "loss": 0.5597, + "step": 5909 + }, + { + "epoch": 4.8928768636112645, + "grad_norm": 0.9484930634498596, + "learning_rate": 4.169831332550464e-07, + "loss": 0.5419, + "step": 5910 + }, + { + "epoch": 4.893705135284373, + "grad_norm": 1.025490403175354, + "learning_rate": 4.163752488051509e-07, + "loss": 0.5608, + "step": 5911 + }, + { + "epoch": 4.894533406957482, + "grad_norm": 0.9627198576927185, + "learning_rate": 4.157677675172217e-07, + "loss": 0.5455, + "step": 5912 + }, + { + "epoch": 4.895361678630591, + "grad_norm": 0.9808639287948608, + "learning_rate": 4.151606895088009e-07, + "loss": 0.5497, + "step": 5913 + }, + { + "epoch": 4.896189950303699, + "grad_norm": 0.993675172328949, + "learning_rate": 4.1455401489735175e-07, + "loss": 0.5649, + "step": 5914 + }, + { + "epoch": 4.897018221976809, + "grad_norm": 0.9799006581306458, + "learning_rate": 4.139477438002606e-07, + "loss": 0.5444, + "step": 5915 + }, + { + "epoch": 4.897846493649917, + "grad_norm": 1.0184911489486694, + "learning_rate": 4.1334187633483487e-07, + "loss": 0.5565, + "step": 5916 + }, + { + "epoch": 4.898674765323026, + "grad_norm": 0.9810110330581665, + "learning_rate": 4.12736412618305e-07, + "loss": 0.5557, + "step": 5917 + }, + { + "epoch": 4.899503036996135, + "grad_norm": 0.9775555729866028, + "learning_rate": 4.121313527678214e-07, + "loss": 0.5748, + "step": 5918 + }, + { + "epoch": 4.9003313086692435, + "grad_norm": 0.9573676586151123, + "learning_rate": 4.1152669690045834e-07, + "loss": 0.5613, + "step": 5919 + }, + { + "epoch": 4.901159580342353, + "grad_norm": 0.9636746048927307, + "learning_rate": 4.109224451332114e-07, + "loss": 0.5407, + "step": 5920 + }, + { + "epoch": 4.901987852015461, + "grad_norm": 0.9665216207504272, + "learning_rate": 4.103185975829965e-07, + "loss": 0.5494, + "step": 5921 + }, + { + "epoch": 4.90281612368857, + "grad_norm": 0.9665354490280151, + "learning_rate": 4.0971515436665334e-07, + "loss": 0.5619, + "step": 5922 + }, + { + "epoch": 4.903644395361678, + "grad_norm": 0.9628219604492188, + "learning_rate": 4.091121156009428e-07, + "loss": 0.5632, + "step": 5923 + }, + { + "epoch": 4.904472667034788, + "grad_norm": 0.9701488614082336, + "learning_rate": 4.085094814025467e-07, + "loss": 0.5579, + "step": 5924 + }, + { + "epoch": 4.905300938707896, + "grad_norm": 0.9564261436462402, + "learning_rate": 4.079072518880689e-07, + "loss": 0.5508, + "step": 5925 + }, + { + "epoch": 4.906129210381005, + "grad_norm": 0.9657530188560486, + "learning_rate": 4.0730542717403705e-07, + "loss": 0.551, + "step": 5926 + }, + { + "epoch": 4.906957482054114, + "grad_norm": 0.9382011294364929, + "learning_rate": 4.0670400737689603e-07, + "loss": 0.5348, + "step": 5927 + }, + { + "epoch": 4.9077857537272225, + "grad_norm": 0.9636839032173157, + "learning_rate": 4.0610299261301613e-07, + "loss": 0.5472, + "step": 5928 + }, + { + "epoch": 4.908614025400332, + "grad_norm": 0.9654035568237305, + "learning_rate": 4.0550238299868846e-07, + "loss": 0.5599, + "step": 5929 + }, + { + "epoch": 4.90944229707344, + "grad_norm": 0.9554421305656433, + "learning_rate": 4.049021786501245e-07, + "loss": 0.5814, + "step": 5930 + }, + { + "epoch": 4.910270568746549, + "grad_norm": 0.9651699662208557, + "learning_rate": 4.0430237968345886e-07, + "loss": 0.5623, + "step": 5931 + }, + { + "epoch": 4.911098840419657, + "grad_norm": 1.0203036069869995, + "learning_rate": 4.0370298621474734e-07, + "loss": 0.5425, + "step": 5932 + }, + { + "epoch": 4.911927112092767, + "grad_norm": 0.9871775507926941, + "learning_rate": 4.0310399835996603e-07, + "loss": 0.5274, + "step": 5933 + }, + { + "epoch": 4.912755383765875, + "grad_norm": 0.9446064233779907, + "learning_rate": 4.0250541623501423e-07, + "loss": 0.5299, + "step": 5934 + }, + { + "epoch": 4.913583655438984, + "grad_norm": 0.9630985260009766, + "learning_rate": 4.019072399557122e-07, + "loss": 0.5482, + "step": 5935 + }, + { + "epoch": 4.914411927112093, + "grad_norm": 0.9475951194763184, + "learning_rate": 4.013094696378006e-07, + "loss": 0.5531, + "step": 5936 + }, + { + "epoch": 4.9152401987852015, + "grad_norm": 0.9971574544906616, + "learning_rate": 4.0071210539694315e-07, + "loss": 0.5509, + "step": 5937 + }, + { + "epoch": 4.916068470458311, + "grad_norm": 0.949856698513031, + "learning_rate": 4.001151473487236e-07, + "loss": 0.5396, + "step": 5938 + }, + { + "epoch": 4.916896742131419, + "grad_norm": 0.9867436289787292, + "learning_rate": 3.995185956086489e-07, + "loss": 0.5769, + "step": 5939 + }, + { + "epoch": 4.917725013804528, + "grad_norm": 0.9721899628639221, + "learning_rate": 3.989224502921446e-07, + "loss": 0.5421, + "step": 5940 + }, + { + "epoch": 4.918553285477636, + "grad_norm": 0.9489734768867493, + "learning_rate": 3.983267115145603e-07, + "loss": 0.5601, + "step": 5941 + }, + { + "epoch": 4.9193815571507455, + "grad_norm": 0.9731948971748352, + "learning_rate": 3.977313793911666e-07, + "loss": 0.5656, + "step": 5942 + }, + { + "epoch": 4.920209828823854, + "grad_norm": 1.0035384893417358, + "learning_rate": 3.9713645403715306e-07, + "loss": 0.5492, + "step": 5943 + }, + { + "epoch": 4.921038100496963, + "grad_norm": 0.9753653407096863, + "learning_rate": 3.965419355676331e-07, + "loss": 0.5454, + "step": 5944 + }, + { + "epoch": 4.921866372170072, + "grad_norm": 0.9717726111412048, + "learning_rate": 3.959478240976411e-07, + "loss": 0.5512, + "step": 5945 + }, + { + "epoch": 4.9226946438431805, + "grad_norm": 0.9888126254081726, + "learning_rate": 3.953541197421307e-07, + "loss": 0.5485, + "step": 5946 + }, + { + "epoch": 4.92352291551629, + "grad_norm": 0.9688169956207275, + "learning_rate": 3.947608226159788e-07, + "loss": 0.5566, + "step": 5947 + }, + { + "epoch": 4.924351187189398, + "grad_norm": 0.9560792446136475, + "learning_rate": 3.9416793283398367e-07, + "loss": 0.5592, + "step": 5948 + }, + { + "epoch": 4.925179458862507, + "grad_norm": 0.9747999310493469, + "learning_rate": 3.93575450510863e-07, + "loss": 0.5569, + "step": 5949 + }, + { + "epoch": 4.926007730535615, + "grad_norm": 0.9653720259666443, + "learning_rate": 3.929833757612564e-07, + "loss": 0.5546, + "step": 5950 + }, + { + "epoch": 4.9268360022087245, + "grad_norm": 0.950875461101532, + "learning_rate": 3.9239170869972585e-07, + "loss": 0.5583, + "step": 5951 + }, + { + "epoch": 4.927664273881833, + "grad_norm": 0.9724762439727783, + "learning_rate": 3.9180044944075205e-07, + "loss": 0.5683, + "step": 5952 + }, + { + "epoch": 4.928492545554942, + "grad_norm": 0.9594883322715759, + "learning_rate": 3.9120959809873905e-07, + "loss": 0.5444, + "step": 5953 + }, + { + "epoch": 4.929320817228051, + "grad_norm": 0.9946019053459167, + "learning_rate": 3.906191547880114e-07, + "loss": 0.5701, + "step": 5954 + }, + { + "epoch": 4.930149088901159, + "grad_norm": 0.9909894466400146, + "learning_rate": 3.900291196228134e-07, + "loss": 0.5607, + "step": 5955 + }, + { + "epoch": 4.930977360574269, + "grad_norm": 0.9909331798553467, + "learning_rate": 3.8943949271731174e-07, + "loss": 0.5438, + "step": 5956 + }, + { + "epoch": 4.931805632247377, + "grad_norm": 0.9559661149978638, + "learning_rate": 3.888502741855946e-07, + "loss": 0.5707, + "step": 5957 + }, + { + "epoch": 4.932633903920486, + "grad_norm": 0.9508602619171143, + "learning_rate": 3.8826146414166904e-07, + "loss": 0.5349, + "step": 5958 + }, + { + "epoch": 4.933462175593594, + "grad_norm": 0.9860195517539978, + "learning_rate": 3.876730626994646e-07, + "loss": 0.5555, + "step": 5959 + }, + { + "epoch": 4.9342904472667035, + "grad_norm": 0.9599242210388184, + "learning_rate": 3.8708506997283207e-07, + "loss": 0.5595, + "step": 5960 + }, + { + "epoch": 4.935118718939812, + "grad_norm": 0.9686924815177917, + "learning_rate": 3.8649748607554243e-07, + "loss": 0.5649, + "step": 5961 + }, + { + "epoch": 4.935946990612921, + "grad_norm": 1.034990906715393, + "learning_rate": 3.859103111212867e-07, + "loss": 0.5693, + "step": 5962 + }, + { + "epoch": 4.93677526228603, + "grad_norm": 1.064138412475586, + "learning_rate": 3.8532354522367923e-07, + "loss": 0.5433, + "step": 5963 + }, + { + "epoch": 4.937603533959138, + "grad_norm": 0.9807191491127014, + "learning_rate": 3.8473718849625226e-07, + "loss": 0.5467, + "step": 5964 + }, + { + "epoch": 4.938431805632248, + "grad_norm": 0.9689276814460754, + "learning_rate": 3.8415124105246115e-07, + "loss": 0.5456, + "step": 5965 + }, + { + "epoch": 4.939260077305356, + "grad_norm": 0.9509407877922058, + "learning_rate": 3.8356570300568165e-07, + "loss": 0.5439, + "step": 5966 + }, + { + "epoch": 4.940088348978465, + "grad_norm": 0.9513766169548035, + "learning_rate": 3.829805744692089e-07, + "loss": 0.5442, + "step": 5967 + }, + { + "epoch": 4.940916620651573, + "grad_norm": 0.971613883972168, + "learning_rate": 3.823958555562607e-07, + "loss": 0.5458, + "step": 5968 + }, + { + "epoch": 4.9417448923246825, + "grad_norm": 0.9698891043663025, + "learning_rate": 3.818115463799746e-07, + "loss": 0.5406, + "step": 5969 + }, + { + "epoch": 4.942573163997791, + "grad_norm": 0.9531323313713074, + "learning_rate": 3.8122764705340847e-07, + "loss": 0.5614, + "step": 5970 + }, + { + "epoch": 4.9434014356709, + "grad_norm": 0.9560294151306152, + "learning_rate": 3.80644157689542e-07, + "loss": 0.546, + "step": 5971 + }, + { + "epoch": 4.944229707344009, + "grad_norm": 0.9792264699935913, + "learning_rate": 3.8006107840127426e-07, + "loss": 0.5458, + "step": 5972 + }, + { + "epoch": 4.945057979017117, + "grad_norm": 0.955904483795166, + "learning_rate": 3.794784093014264e-07, + "loss": 0.5389, + "step": 5973 + }, + { + "epoch": 4.945886250690227, + "grad_norm": 0.9566978812217712, + "learning_rate": 3.788961505027389e-07, + "loss": 0.5482, + "step": 5974 + }, + { + "epoch": 4.946714522363335, + "grad_norm": 0.9445267915725708, + "learning_rate": 3.783143021178731e-07, + "loss": 0.5467, + "step": 5975 + }, + { + "epoch": 4.947542794036444, + "grad_norm": 0.9731202125549316, + "learning_rate": 3.7773286425941273e-07, + "loss": 0.5376, + "step": 5976 + }, + { + "epoch": 4.948371065709553, + "grad_norm": 0.9669923186302185, + "learning_rate": 3.7715183703985883e-07, + "loss": 0.5353, + "step": 5977 + }, + { + "epoch": 4.9491993373826615, + "grad_norm": 0.955851137638092, + "learning_rate": 3.765712205716357e-07, + "loss": 0.5221, + "step": 5978 + }, + { + "epoch": 4.950027609055771, + "grad_norm": 0.9659578800201416, + "learning_rate": 3.7599101496708726e-07, + "loss": 0.5743, + "step": 5979 + }, + { + "epoch": 4.950855880728879, + "grad_norm": 0.965240478515625, + "learning_rate": 3.754112203384774e-07, + "loss": 0.5762, + "step": 5980 + }, + { + "epoch": 4.951684152401988, + "grad_norm": 2.1109437942504883, + "learning_rate": 3.7483183679799113e-07, + "loss": 0.5611, + "step": 5981 + }, + { + "epoch": 4.952512424075096, + "grad_norm": 0.9661439061164856, + "learning_rate": 3.742528644577351e-07, + "loss": 0.5452, + "step": 5982 + }, + { + "epoch": 4.953340695748206, + "grad_norm": 0.965039074420929, + "learning_rate": 3.736743034297327e-07, + "loss": 0.5508, + "step": 5983 + }, + { + "epoch": 4.954168967421314, + "grad_norm": 0.9632126688957214, + "learning_rate": 3.730961538259312e-07, + "loss": 0.5651, + "step": 5984 + }, + { + "epoch": 4.954997239094423, + "grad_norm": 0.9725423455238342, + "learning_rate": 3.7251841575819764e-07, + "loss": 0.5571, + "step": 5985 + }, + { + "epoch": 4.955825510767532, + "grad_norm": 0.9642723202705383, + "learning_rate": 3.7194108933831794e-07, + "loss": 0.5604, + "step": 5986 + }, + { + "epoch": 4.9566537824406405, + "grad_norm": 0.9578214287757874, + "learning_rate": 3.71364174678e-07, + "loss": 0.5249, + "step": 5987 + }, + { + "epoch": 4.95748205411375, + "grad_norm": 0.9682602286338806, + "learning_rate": 3.70787671888872e-07, + "loss": 0.5356, + "step": 5988 + }, + { + "epoch": 4.958310325786858, + "grad_norm": 0.9686048626899719, + "learning_rate": 3.702115810824805e-07, + "loss": 0.5436, + "step": 5989 + }, + { + "epoch": 4.959138597459967, + "grad_norm": 1.0237302780151367, + "learning_rate": 3.696359023702947e-07, + "loss": 0.5369, + "step": 5990 + }, + { + "epoch": 4.959966869133075, + "grad_norm": 0.9748989939689636, + "learning_rate": 3.6906063586370306e-07, + "loss": 0.5727, + "step": 5991 + }, + { + "epoch": 4.9607951408061846, + "grad_norm": 0.972263514995575, + "learning_rate": 3.684857816740137e-07, + "loss": 0.5969, + "step": 5992 + }, + { + "epoch": 4.961623412479293, + "grad_norm": 0.979120671749115, + "learning_rate": 3.6791133991245566e-07, + "loss": 0.5463, + "step": 5993 + }, + { + "epoch": 4.962451684152402, + "grad_norm": 0.9615027904510498, + "learning_rate": 3.6733731069017895e-07, + "loss": 0.5722, + "step": 5994 + }, + { + "epoch": 4.963279955825511, + "grad_norm": 0.9743287563323975, + "learning_rate": 3.6676369411825204e-07, + "loss": 0.5388, + "step": 5995 + }, + { + "epoch": 4.9641082274986195, + "grad_norm": 0.9608129858970642, + "learning_rate": 3.661904903076641e-07, + "loss": 0.5329, + "step": 5996 + }, + { + "epoch": 4.964936499171729, + "grad_norm": 0.9721583724021912, + "learning_rate": 3.656176993693253e-07, + "loss": 0.5473, + "step": 5997 + }, + { + "epoch": 4.965764770844837, + "grad_norm": 0.9767399430274963, + "learning_rate": 3.650453214140656e-07, + "loss": 0.5293, + "step": 5998 + }, + { + "epoch": 4.966593042517946, + "grad_norm": 0.9559426307678223, + "learning_rate": 3.6447335655263397e-07, + "loss": 0.5484, + "step": 5999 + }, + { + "epoch": 4.967421314191054, + "grad_norm": 0.9602168798446655, + "learning_rate": 3.639018048957005e-07, + "loss": 0.5579, + "step": 6000 + }, + { + "epoch": 4.9682495858641635, + "grad_norm": 0.9681665897369385, + "learning_rate": 3.633306665538561e-07, + "loss": 0.5628, + "step": 6001 + }, + { + "epoch": 4.969077857537272, + "grad_norm": 0.9520401954650879, + "learning_rate": 3.627599416376093e-07, + "loss": 0.5504, + "step": 6002 + }, + { + "epoch": 4.969906129210381, + "grad_norm": 0.9589652419090271, + "learning_rate": 3.6218963025739073e-07, + "loss": 0.5507, + "step": 6003 + }, + { + "epoch": 4.97073440088349, + "grad_norm": 0.977666974067688, + "learning_rate": 3.616197325235507e-07, + "loss": 0.5504, + "step": 6004 + }, + { + "epoch": 4.9715626725565985, + "grad_norm": 0.98526930809021, + "learning_rate": 3.6105024854635903e-07, + "loss": 0.5393, + "step": 6005 + }, + { + "epoch": 4.972390944229708, + "grad_norm": 0.9593404531478882, + "learning_rate": 3.6048117843600433e-07, + "loss": 0.5707, + "step": 6006 + }, + { + "epoch": 4.973219215902816, + "grad_norm": 0.9677101373672485, + "learning_rate": 3.5991252230259817e-07, + "loss": 0.5559, + "step": 6007 + }, + { + "epoch": 4.974047487575925, + "grad_norm": 0.9906893372535706, + "learning_rate": 3.5934428025616833e-07, + "loss": 0.5562, + "step": 6008 + }, + { + "epoch": 4.974875759249033, + "grad_norm": 0.9470625519752502, + "learning_rate": 3.587764524066656e-07, + "loss": 0.5308, + "step": 6009 + }, + { + "epoch": 4.9757040309221425, + "grad_norm": 0.9528307318687439, + "learning_rate": 3.5820903886395945e-07, + "loss": 0.537, + "step": 6010 + }, + { + "epoch": 4.976532302595251, + "grad_norm": 0.9625613689422607, + "learning_rate": 3.5764203973783854e-07, + "loss": 0.5555, + "step": 6011 + }, + { + "epoch": 4.97736057426836, + "grad_norm": 0.967427134513855, + "learning_rate": 3.570754551380118e-07, + "loss": 0.5514, + "step": 6012 + }, + { + "epoch": 4.978188845941469, + "grad_norm": 0.9919961094856262, + "learning_rate": 3.565092851741092e-07, + "loss": 0.5294, + "step": 6013 + }, + { + "epoch": 4.979017117614577, + "grad_norm": 0.9725098013877869, + "learning_rate": 3.5594352995567793e-07, + "loss": 0.5561, + "step": 6014 + }, + { + "epoch": 4.979845389287687, + "grad_norm": 0.9572508931159973, + "learning_rate": 3.5537818959218677e-07, + "loss": 0.5545, + "step": 6015 + }, + { + "epoch": 4.980673660960795, + "grad_norm": 0.9887001514434814, + "learning_rate": 3.548132641930249e-07, + "loss": 0.5601, + "step": 6016 + }, + { + "epoch": 4.981501932633904, + "grad_norm": 0.9554132223129272, + "learning_rate": 3.542487538674991e-07, + "loss": 0.5558, + "step": 6017 + }, + { + "epoch": 4.982330204307012, + "grad_norm": 0.981720507144928, + "learning_rate": 3.536846587248363e-07, + "loss": 0.5776, + "step": 6018 + }, + { + "epoch": 4.9831584759801215, + "grad_norm": 0.9739376306533813, + "learning_rate": 3.5312097887418435e-07, + "loss": 0.5627, + "step": 6019 + }, + { + "epoch": 4.98398674765323, + "grad_norm": 0.9445295929908752, + "learning_rate": 3.525577144246106e-07, + "loss": 0.5442, + "step": 6020 + }, + { + "epoch": 4.984815019326339, + "grad_norm": 0.9757813811302185, + "learning_rate": 3.5199486548510046e-07, + "loss": 0.5331, + "step": 6021 + }, + { + "epoch": 4.985643290999448, + "grad_norm": 0.9708597660064697, + "learning_rate": 3.514324321645604e-07, + "loss": 0.5758, + "step": 6022 + }, + { + "epoch": 4.986471562672556, + "grad_norm": 0.9770578742027283, + "learning_rate": 3.508704145718167e-07, + "loss": 0.5569, + "step": 6023 + }, + { + "epoch": 4.987299834345666, + "grad_norm": 0.9736425280570984, + "learning_rate": 3.503088128156129e-07, + "loss": 0.5718, + "step": 6024 + }, + { + "epoch": 4.988128106018774, + "grad_norm": 0.9564449787139893, + "learning_rate": 3.4974762700461573e-07, + "loss": 0.5496, + "step": 6025 + }, + { + "epoch": 4.988956377691883, + "grad_norm": 0.9484078288078308, + "learning_rate": 3.4918685724740747e-07, + "loss": 0.5466, + "step": 6026 + }, + { + "epoch": 4.989784649364991, + "grad_norm": 0.9592824578285217, + "learning_rate": 3.486265036524927e-07, + "loss": 0.556, + "step": 6027 + }, + { + "epoch": 4.9906129210381005, + "grad_norm": 0.9514402747154236, + "learning_rate": 3.4806656632829504e-07, + "loss": 0.5421, + "step": 6028 + }, + { + "epoch": 4.991441192711209, + "grad_norm": 0.9617545008659363, + "learning_rate": 3.4750704538315694e-07, + "loss": 0.576, + "step": 6029 + }, + { + "epoch": 4.992269464384318, + "grad_norm": 0.9632974863052368, + "learning_rate": 3.4694794092533983e-07, + "loss": 0.5495, + "step": 6030 + }, + { + "epoch": 4.993097736057427, + "grad_norm": 0.9788113832473755, + "learning_rate": 3.4638925306302556e-07, + "loss": 0.5757, + "step": 6031 + }, + { + "epoch": 4.993926007730535, + "grad_norm": 1.002131700515747, + "learning_rate": 3.4583098190431544e-07, + "loss": 0.5287, + "step": 6032 + }, + { + "epoch": 4.994754279403645, + "grad_norm": 0.9638229012489319, + "learning_rate": 3.452731275572291e-07, + "loss": 0.5421, + "step": 6033 + }, + { + "epoch": 4.995582551076753, + "grad_norm": 1.010887622833252, + "learning_rate": 3.4471569012970633e-07, + "loss": 0.5613, + "step": 6034 + }, + { + "epoch": 4.996410822749862, + "grad_norm": 0.9634838104248047, + "learning_rate": 3.4415866972960704e-07, + "loss": 0.5388, + "step": 6035 + } + ], + "logging_steps": 1, + "max_steps": 7242, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 1207, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1.5237895367090504e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-6035/training_args.bin b/checkpoint-6035/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..7017eeafdf1f43b26668f0e0e5710ca52345a005 --- /dev/null +++ b/checkpoint-6035/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fdc878d2541c09353dd68f4f8a2464140b28fd84e1bc48b59914a6e84328e85e +size 8056 diff --git a/checkpoint-6035/zero_to_fp32.py b/checkpoint-6035/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-6035/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/checkpoint-7242/README.md b/checkpoint-7242/README.md new file mode 100644 index 0000000000000000000000000000000000000000..049d467664ca6172b7ffbe6ba60b3eac7479cac4 --- /dev/null +++ b/checkpoint-7242/README.md @@ -0,0 +1,202 @@ +--- +base_model: meta-llama/Llama-3.1-8B +library_name: peft +--- + +# Model Card for Model ID + + + + + +## Model Details + +### Model Description + + + + + +- **Developed by:** [More Information Needed] +- **Funded by [optional]:** [More Information Needed] +- **Shared by [optional]:** [More Information Needed] +- **Model type:** [More Information Needed] +- **Language(s) (NLP):** [More Information Needed] +- **License:** [More Information Needed] +- **Finetuned from model [optional]:** [More Information Needed] + +### Model Sources [optional] + + + +- **Repository:** [More Information Needed] +- **Paper [optional]:** [More Information Needed] +- **Demo [optional]:** [More Information Needed] + +## Uses + + + +### Direct Use + + + +[More Information Needed] + +### Downstream Use [optional] + + + +[More Information Needed] + +### Out-of-Scope Use + + + +[More Information Needed] + +## Bias, Risks, and Limitations + + + +[More Information Needed] + +### Recommendations + + + +Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. + +## How to Get Started with the Model + +Use the code below to get started with the model. + +[More Information Needed] + +## Training Details + +### Training Data + + + +[More Information Needed] + +### Training Procedure + + + +#### Preprocessing [optional] + +[More Information Needed] + + +#### Training Hyperparameters + +- **Training regime:** [More Information Needed] + +#### Speeds, Sizes, Times [optional] + + + +[More Information Needed] + +## Evaluation + + + +### Testing Data, Factors & Metrics + +#### Testing Data + + + +[More Information Needed] + +#### Factors + + + +[More Information Needed] + +#### Metrics + + + +[More Information Needed] + +### Results + +[More Information Needed] + +#### Summary + + + +## Model Examination [optional] + + + +[More Information Needed] + +## Environmental Impact + + + +Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). + +- **Hardware Type:** [More Information Needed] +- **Hours used:** [More Information Needed] +- **Cloud Provider:** [More Information Needed] +- **Compute Region:** [More Information Needed] +- **Carbon Emitted:** [More Information Needed] + +## Technical Specifications [optional] + +### Model Architecture and Objective + +[More Information Needed] + +### Compute Infrastructure + +[More Information Needed] + +#### Hardware + +[More Information Needed] + +#### Software + +[More Information Needed] + +## Citation [optional] + + + +**BibTeX:** + +[More Information Needed] + +**APA:** + +[More Information Needed] + +## Glossary [optional] + + + +[More Information Needed] + +## More Information [optional] + +[More Information Needed] + +## Model Card Authors [optional] + +[More Information Needed] + +## Model Card Contact + +[More Information Needed] +### Framework versions + +- PEFT 0.14.0 \ No newline at end of file diff --git a/checkpoint-7242/adapter_config.json b/checkpoint-7242/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..d6fca8a067dce19d599e7f1e24bea8019968fe0f --- /dev/null +++ b/checkpoint-7242/adapter_config.json @@ -0,0 +1,40 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.1-8B", + "bias": "none", + "eva_config": null, + "exclude_modules": null, + "fan_in_fan_out": null, + "inference_mode": true, + "init_lora_weights": true, + "layer_replication": null, + "layers_pattern": null, + "layers_to_transform": null, + "loftq_config": {}, + "lora_alpha": 512, + "lora_bias": false, + "lora_dropout": 0.05, + "megatron_config": null, + "megatron_core": "megatron.core", + "modules_to_save": [ + "embed_tokens", + "lm_head" + ], + "peft_type": "LORA", + "r": 256, + "rank_pattern": {}, + "revision": null, + "target_modules": [ + "gate_proj", + "up_proj", + "q_proj", + "k_proj", + "o_proj", + "v_proj", + "down_proj" + ], + "task_type": "CAUSAL_LM", + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-7242/adapter_model.safetensors b/checkpoint-7242/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8df017a42ac766c44c8af48fbe94759ea5b038ae --- /dev/null +++ b/checkpoint-7242/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:07547a0137768f7add8b98278e833c88c14aa448fb45fe6b306fccdc96439647 +size 3443586272 diff --git a/checkpoint-7242/global_step7238/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-7242/global_step7238/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..b8833f21e65cce80e96f210fdfdd9f18b19952d1 --- /dev/null +++ b/checkpoint-7242/global_step7238/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a7125cf6ffbf524c10d6d33cedce381cc1b8845dafac9a04e1871cd9054d6815 +size 20661195036 diff --git a/checkpoint-7242/global_step7238/mp_rank_00_model_states.pt b/checkpoint-7242/global_step7238/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..6f029ec22d8878e6b6c726b84a6eb215f5b8920e --- /dev/null +++ b/checkpoint-7242/global_step7238/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:397702449e5ac94e0090a2a24f1f62956c2e7a0d06c306381fbed79ccc86a87b +size 3555326841 diff --git a/checkpoint-7242/latest b/checkpoint-7242/latest new file mode 100644 index 0000000000000000000000000000000000000000..7fb860c7d2de186cfab1e84f61bef0a3da848cf7 --- /dev/null +++ b/checkpoint-7242/latest @@ -0,0 +1 @@ +global_step7238 \ No newline at end of file diff --git a/checkpoint-7242/rng_state.pth b/checkpoint-7242/rng_state.pth new file mode 100644 index 0000000000000000000000000000000000000000..5baeb7e0e178e4d282265887a223cf4fbe55008f --- /dev/null +++ b/checkpoint-7242/rng_state.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4f1e7fe86a7a819ecdb3b489672592b82cc60d5aa97b0ef518915dcb1b119b3f +size 14244 diff --git a/checkpoint-7242/scheduler.pt b/checkpoint-7242/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..81fed780a38d3d6ff3d1825cbc73bda18c6bd891 --- /dev/null +++ b/checkpoint-7242/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:986fc804cb523daed856e2f83a2e2096527d813fb4739ec16d3e260f970f1405 +size 1064 diff --git a/checkpoint-7242/special_tokens_map.json b/checkpoint-7242/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..e5b39b6305d89284b04934011c68dbb26bf588ca --- /dev/null +++ b/checkpoint-7242/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/checkpoint-7242/tokenizer.json b/checkpoint-7242/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-7242/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-7242/tokenizer_config.json b/checkpoint-7242/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..81dd14db6632ad5b35b9d447732e37ac074873a5 --- /dev/null +++ b/checkpoint-7242/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +} diff --git a/checkpoint-7242/trainer_state.json b/checkpoint-7242/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..00a157afbd2d742af323f57883be9b1019eb66d4 --- /dev/null +++ b/checkpoint-7242/trainer_state.json @@ -0,0 +1,50727 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 5.995582551076753, + "eval_steps": 500, + "global_step": 7242, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.0008282716731087797, + "grad_norm": 2.4795358180999756, + "learning_rate": 5.0000000000000004e-08, + "loss": 1.1215, + "step": 1 + }, + { + "epoch": 0.0016565433462175593, + "grad_norm": 2.6060028076171875, + "learning_rate": 1.0000000000000001e-07, + "loss": 1.1219, + "step": 2 + }, + { + "epoch": 0.002484815019326339, + "grad_norm": 2.750868082046509, + "learning_rate": 1.5000000000000002e-07, + "loss": 1.125, + "step": 3 + }, + { + "epoch": 0.0033130866924351186, + "grad_norm": 2.7321369647979736, + "learning_rate": 2.0000000000000002e-07, + "loss": 1.1164, + "step": 4 + }, + { + "epoch": 0.0041413583655438985, + "grad_norm": 2.8193929195404053, + "learning_rate": 2.5000000000000004e-07, + "loss": 1.1263, + "step": 5 + }, + { + "epoch": 0.004969630038652678, + "grad_norm": 2.648301839828491, + "learning_rate": 3.0000000000000004e-07, + "loss": 1.1092, + "step": 6 + }, + { + "epoch": 0.005797901711761457, + "grad_norm": 2.751993179321289, + "learning_rate": 3.5000000000000004e-07, + "loss": 1.0997, + "step": 7 + }, + { + "epoch": 0.006626173384870237, + "grad_norm": 2.592381000518799, + "learning_rate": 4.0000000000000003e-07, + "loss": 1.112, + "step": 8 + }, + { + "epoch": 0.007454445057979017, + "grad_norm": 2.3135671615600586, + "learning_rate": 4.5000000000000003e-07, + "loss": 1.1045, + "step": 9 + }, + { + "epoch": 0.008282716731087797, + "grad_norm": 2.60491943359375, + "learning_rate": 5.000000000000001e-07, + "loss": 1.0956, + "step": 10 + }, + { + "epoch": 0.009110988404196576, + "grad_norm": 2.2435271739959717, + "learning_rate": 5.5e-07, + "loss": 1.1241, + "step": 11 + }, + { + "epoch": 0.009939260077305357, + "grad_norm": 2.4412319660186768, + "learning_rate": 6.000000000000001e-07, + "loss": 1.11, + "step": 12 + }, + { + "epoch": 0.010767531750414136, + "grad_norm": 2.2075185775756836, + "learning_rate": 6.5e-07, + "loss": 1.0878, + "step": 13 + }, + { + "epoch": 0.011595803423522915, + "grad_norm": 2.101902723312378, + "learning_rate": 7.000000000000001e-07, + "loss": 1.0863, + "step": 14 + }, + { + "epoch": 0.012424075096631695, + "grad_norm": 2.134535074234009, + "learning_rate": 7.5e-07, + "loss": 1.0735, + "step": 15 + }, + { + "epoch": 0.013252346769740474, + "grad_norm": 2.000504493713379, + "learning_rate": 8.000000000000001e-07, + "loss": 1.0875, + "step": 16 + }, + { + "epoch": 0.014080618442849255, + "grad_norm": 1.9191882610321045, + "learning_rate": 8.500000000000001e-07, + "loss": 1.1039, + "step": 17 + }, + { + "epoch": 0.014908890115958034, + "grad_norm": 1.7732031345367432, + "learning_rate": 9.000000000000001e-07, + "loss": 1.0819, + "step": 18 + }, + { + "epoch": 0.015737161789066815, + "grad_norm": 1.7775251865386963, + "learning_rate": 9.500000000000001e-07, + "loss": 1.0936, + "step": 19 + }, + { + "epoch": 0.016565433462175594, + "grad_norm": 1.6648598909378052, + "learning_rate": 1.0000000000000002e-06, + "loss": 1.0881, + "step": 20 + }, + { + "epoch": 0.017393705135284373, + "grad_norm": 1.5726720094680786, + "learning_rate": 1.0500000000000001e-06, + "loss": 1.07, + "step": 21 + }, + { + "epoch": 0.018221976808393152, + "grad_norm": 1.561417818069458, + "learning_rate": 1.1e-06, + "loss": 1.0668, + "step": 22 + }, + { + "epoch": 0.019050248481501934, + "grad_norm": 1.4953842163085938, + "learning_rate": 1.1500000000000002e-06, + "loss": 1.054, + "step": 23 + }, + { + "epoch": 0.019878520154610713, + "grad_norm": 1.4247829914093018, + "learning_rate": 1.2000000000000002e-06, + "loss": 1.0363, + "step": 24 + }, + { + "epoch": 0.020706791827719492, + "grad_norm": 1.3483426570892334, + "learning_rate": 1.25e-06, + "loss": 1.0143, + "step": 25 + }, + { + "epoch": 0.02153506350082827, + "grad_norm": 1.3412944078445435, + "learning_rate": 1.3e-06, + "loss": 1.025, + "step": 26 + }, + { + "epoch": 0.02236333517393705, + "grad_norm": 1.2834237813949585, + "learning_rate": 1.3500000000000002e-06, + "loss": 1.0227, + "step": 27 + }, + { + "epoch": 0.02319160684704583, + "grad_norm": 1.2128241062164307, + "learning_rate": 1.4000000000000001e-06, + "loss": 1.0161, + "step": 28 + }, + { + "epoch": 0.024019878520154612, + "grad_norm": 1.1665573120117188, + "learning_rate": 1.45e-06, + "loss": 1.0034, + "step": 29 + }, + { + "epoch": 0.02484815019326339, + "grad_norm": 1.1575899124145508, + "learning_rate": 1.5e-06, + "loss": 0.9968, + "step": 30 + }, + { + "epoch": 0.02567642186637217, + "grad_norm": 1.195304274559021, + "learning_rate": 1.5500000000000002e-06, + "loss": 1.0176, + "step": 31 + }, + { + "epoch": 0.02650469353948095, + "grad_norm": 1.1321678161621094, + "learning_rate": 1.6000000000000001e-06, + "loss": 1.0113, + "step": 32 + }, + { + "epoch": 0.027332965212589728, + "grad_norm": 1.1076338291168213, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.9971, + "step": 33 + }, + { + "epoch": 0.02816123688569851, + "grad_norm": 1.0770610570907593, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.9994, + "step": 34 + }, + { + "epoch": 0.02898950855880729, + "grad_norm": 1.0767767429351807, + "learning_rate": 1.75e-06, + "loss": 0.978, + "step": 35 + }, + { + "epoch": 0.02981778023191607, + "grad_norm": 1.054800033569336, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.9829, + "step": 36 + }, + { + "epoch": 0.030646051905024847, + "grad_norm": 1.031015157699585, + "learning_rate": 1.85e-06, + "loss": 0.97, + "step": 37 + }, + { + "epoch": 0.03147432357813363, + "grad_norm": 1.0182578563690186, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.9549, + "step": 38 + }, + { + "epoch": 0.032302595251242405, + "grad_norm": 0.9796578288078308, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.9762, + "step": 39 + }, + { + "epoch": 0.03313086692435119, + "grad_norm": 0.9956578016281128, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.9693, + "step": 40 + }, + { + "epoch": 0.03395913859745997, + "grad_norm": 0.9383260011672974, + "learning_rate": 2.05e-06, + "loss": 0.9556, + "step": 41 + }, + { + "epoch": 0.034787410270568746, + "grad_norm": 0.9372819066047668, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.9419, + "step": 42 + }, + { + "epoch": 0.03561568194367753, + "grad_norm": 0.9334004521369934, + "learning_rate": 2.15e-06, + "loss": 0.9587, + "step": 43 + }, + { + "epoch": 0.036443953616786304, + "grad_norm": 0.8916004300117493, + "learning_rate": 2.2e-06, + "loss": 0.9227, + "step": 44 + }, + { + "epoch": 0.037272225289895086, + "grad_norm": 0.8698756694793701, + "learning_rate": 2.25e-06, + "loss": 0.9275, + "step": 45 + }, + { + "epoch": 0.03810049696300387, + "grad_norm": 0.8775100708007812, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.941, + "step": 46 + }, + { + "epoch": 0.038928768636112644, + "grad_norm": 0.8763384222984314, + "learning_rate": 2.35e-06, + "loss": 0.9253, + "step": 47 + }, + { + "epoch": 0.03975704030922143, + "grad_norm": 0.8534179925918579, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.9327, + "step": 48 + }, + { + "epoch": 0.0405853119823302, + "grad_norm": 0.8432146906852722, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.9236, + "step": 49 + }, + { + "epoch": 0.041413583655438985, + "grad_norm": 0.9899975061416626, + "learning_rate": 2.5e-06, + "loss": 0.8839, + "step": 50 + }, + { + "epoch": 0.04224185532854776, + "grad_norm": 0.8370104432106018, + "learning_rate": 2.55e-06, + "loss": 0.906, + "step": 51 + }, + { + "epoch": 0.04307012700165654, + "grad_norm": 0.8207446336746216, + "learning_rate": 2.6e-06, + "loss": 0.8755, + "step": 52 + }, + { + "epoch": 0.043898398674765325, + "grad_norm": 0.805584728717804, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.9049, + "step": 53 + }, + { + "epoch": 0.0447266703478741, + "grad_norm": 0.8162860870361328, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.8993, + "step": 54 + }, + { + "epoch": 0.04555494202098288, + "grad_norm": 0.8151668906211853, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.9282, + "step": 55 + }, + { + "epoch": 0.04638321369409166, + "grad_norm": 0.7996147274971008, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.8946, + "step": 56 + }, + { + "epoch": 0.04721148536720044, + "grad_norm": 0.8223204016685486, + "learning_rate": 2.85e-06, + "loss": 0.897, + "step": 57 + }, + { + "epoch": 0.048039757040309224, + "grad_norm": 0.8117988109588623, + "learning_rate": 2.9e-06, + "loss": 0.8845, + "step": 58 + }, + { + "epoch": 0.048868028713418, + "grad_norm": 0.7902786731719971, + "learning_rate": 2.95e-06, + "loss": 0.8846, + "step": 59 + }, + { + "epoch": 0.04969630038652678, + "grad_norm": 0.804192304611206, + "learning_rate": 3e-06, + "loss": 0.8722, + "step": 60 + }, + { + "epoch": 0.05052457205963556, + "grad_norm": 0.7852040529251099, + "learning_rate": 3.05e-06, + "loss": 0.8821, + "step": 61 + }, + { + "epoch": 0.05135284373274434, + "grad_norm": 0.8045080304145813, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.8661, + "step": 62 + }, + { + "epoch": 0.05218111540585312, + "grad_norm": 0.7920243144035339, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.8712, + "step": 63 + }, + { + "epoch": 0.0530093870789619, + "grad_norm": 0.7601844072341919, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.8792, + "step": 64 + }, + { + "epoch": 0.05383765875207068, + "grad_norm": 0.7833033800125122, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.8811, + "step": 65 + }, + { + "epoch": 0.054665930425179456, + "grad_norm": 0.7691664099693298, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.8701, + "step": 66 + }, + { + "epoch": 0.05549420209828824, + "grad_norm": 0.7845340967178345, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.8929, + "step": 67 + }, + { + "epoch": 0.05632247377139702, + "grad_norm": 0.9076205492019653, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.8352, + "step": 68 + }, + { + "epoch": 0.057150745444505796, + "grad_norm": 0.7614927887916565, + "learning_rate": 3.45e-06, + "loss": 0.8766, + "step": 69 + }, + { + "epoch": 0.05797901711761458, + "grad_norm": 0.7722934484481812, + "learning_rate": 3.5e-06, + "loss": 0.8666, + "step": 70 + }, + { + "epoch": 0.058807288790723354, + "grad_norm": 0.7603748440742493, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.84, + "step": 71 + }, + { + "epoch": 0.05963556046383214, + "grad_norm": 0.7563591003417969, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.8784, + "step": 72 + }, + { + "epoch": 0.06046383213694092, + "grad_norm": 0.7835114002227783, + "learning_rate": 3.65e-06, + "loss": 0.8516, + "step": 73 + }, + { + "epoch": 0.061292103810049695, + "grad_norm": 0.7760982513427734, + "learning_rate": 3.7e-06, + "loss": 0.8265, + "step": 74 + }, + { + "epoch": 0.06212037548315848, + "grad_norm": 0.7978456020355225, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.8564, + "step": 75 + }, + { + "epoch": 0.06294864715626726, + "grad_norm": 0.7940747141838074, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.8635, + "step": 76 + }, + { + "epoch": 0.06377691882937604, + "grad_norm": 0.8055519461631775, + "learning_rate": 3.85e-06, + "loss": 0.8445, + "step": 77 + }, + { + "epoch": 0.06460519050248481, + "grad_norm": 0.7532406449317932, + "learning_rate": 3.900000000000001e-06, + "loss": 0.8772, + "step": 78 + }, + { + "epoch": 0.06543346217559359, + "grad_norm": 0.763933539390564, + "learning_rate": 3.95e-06, + "loss": 0.8574, + "step": 79 + }, + { + "epoch": 0.06626173384870238, + "grad_norm": 0.7833206057548523, + "learning_rate": 4.000000000000001e-06, + "loss": 0.8643, + "step": 80 + }, + { + "epoch": 0.06709000552181116, + "grad_norm": 0.7825527787208557, + "learning_rate": 4.05e-06, + "loss": 0.8518, + "step": 81 + }, + { + "epoch": 0.06791827719491994, + "grad_norm": 0.7749581336975098, + "learning_rate": 4.1e-06, + "loss": 0.8146, + "step": 82 + }, + { + "epoch": 0.06874654886802871, + "grad_norm": 0.8117144703865051, + "learning_rate": 4.15e-06, + "loss": 0.8719, + "step": 83 + }, + { + "epoch": 0.06957482054113749, + "grad_norm": 0.8440507054328918, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.8665, + "step": 84 + }, + { + "epoch": 0.07040309221424627, + "grad_norm": 0.7941984534263611, + "learning_rate": 4.25e-06, + "loss": 0.8492, + "step": 85 + }, + { + "epoch": 0.07123136388735506, + "grad_norm": 0.8301989436149597, + "learning_rate": 4.3e-06, + "loss": 0.8544, + "step": 86 + }, + { + "epoch": 0.07205963556046384, + "grad_norm": 0.757798969745636, + "learning_rate": 4.350000000000001e-06, + "loss": 0.8426, + "step": 87 + }, + { + "epoch": 0.07288790723357261, + "grad_norm": 0.7817673683166504, + "learning_rate": 4.4e-06, + "loss": 0.8554, + "step": 88 + }, + { + "epoch": 0.07371617890668139, + "grad_norm": 0.7609818577766418, + "learning_rate": 4.450000000000001e-06, + "loss": 0.8473, + "step": 89 + }, + { + "epoch": 0.07454445057979017, + "grad_norm": 0.8146091103553772, + "learning_rate": 4.5e-06, + "loss": 0.8317, + "step": 90 + }, + { + "epoch": 0.07537272225289895, + "grad_norm": 0.8058239817619324, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.8411, + "step": 91 + }, + { + "epoch": 0.07620099392600774, + "grad_norm": 0.7700346112251282, + "learning_rate": 4.600000000000001e-06, + "loss": 0.815, + "step": 92 + }, + { + "epoch": 0.0770292655991165, + "grad_norm": 0.7949067950248718, + "learning_rate": 4.65e-06, + "loss": 0.8512, + "step": 93 + }, + { + "epoch": 0.07785753727222529, + "grad_norm": 0.8112176060676575, + "learning_rate": 4.7e-06, + "loss": 0.8398, + "step": 94 + }, + { + "epoch": 0.07868580894533407, + "grad_norm": 0.7990434765815735, + "learning_rate": 4.75e-06, + "loss": 0.8276, + "step": 95 + }, + { + "epoch": 0.07951408061844285, + "grad_norm": 0.7942920923233032, + "learning_rate": 4.800000000000001e-06, + "loss": 0.8333, + "step": 96 + }, + { + "epoch": 0.08034235229155164, + "grad_norm": 0.8005534410476685, + "learning_rate": 4.85e-06, + "loss": 0.8453, + "step": 97 + }, + { + "epoch": 0.0811706239646604, + "grad_norm": 0.8103216290473938, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.8303, + "step": 98 + }, + { + "epoch": 0.08199889563776919, + "grad_norm": 0.8217238187789917, + "learning_rate": 4.95e-06, + "loss": 0.8204, + "step": 99 + }, + { + "epoch": 0.08282716731087797, + "grad_norm": 0.7958654761314392, + "learning_rate": 5e-06, + "loss": 0.8237, + "step": 100 + }, + { + "epoch": 0.08365543898398675, + "grad_norm": 0.8631463050842285, + "learning_rate": 4.999999758136652e-06, + "loss": 0.821, + "step": 101 + }, + { + "epoch": 0.08448371065709552, + "grad_norm": 0.7666735053062439, + "learning_rate": 4.999999032546657e-06, + "loss": 0.8307, + "step": 102 + }, + { + "epoch": 0.0853119823302043, + "grad_norm": 0.7709226012229919, + "learning_rate": 4.999997823230153e-06, + "loss": 0.8299, + "step": 103 + }, + { + "epoch": 0.08614025400331309, + "grad_norm": 0.8020651340484619, + "learning_rate": 4.9999961301873736e-06, + "loss": 0.8075, + "step": 104 + }, + { + "epoch": 0.08696852567642187, + "grad_norm": 0.8019537329673767, + "learning_rate": 4.999993953418649e-06, + "loss": 0.8615, + "step": 105 + }, + { + "epoch": 0.08779679734953065, + "grad_norm": 0.7904282808303833, + "learning_rate": 4.999991292924399e-06, + "loss": 0.8353, + "step": 106 + }, + { + "epoch": 0.08862506902263942, + "grad_norm": 0.7854570150375366, + "learning_rate": 4.999988148705138e-06, + "loss": 0.8154, + "step": 107 + }, + { + "epoch": 0.0894533406957482, + "grad_norm": 0.8044222593307495, + "learning_rate": 4.999984520761475e-06, + "loss": 0.8117, + "step": 108 + }, + { + "epoch": 0.09028161236885698, + "grad_norm": 0.7986987829208374, + "learning_rate": 4.999980409094112e-06, + "loss": 0.8407, + "step": 109 + }, + { + "epoch": 0.09110988404196577, + "grad_norm": 0.7976786494255066, + "learning_rate": 4.999975813703844e-06, + "loss": 0.7871, + "step": 110 + }, + { + "epoch": 0.09193815571507455, + "grad_norm": 0.8330315351486206, + "learning_rate": 4.9999707345915605e-06, + "loss": 0.8129, + "step": 111 + }, + { + "epoch": 0.09276642738818332, + "grad_norm": 0.8350826501846313, + "learning_rate": 4.999965171758245e-06, + "loss": 0.8144, + "step": 112 + }, + { + "epoch": 0.0935946990612921, + "grad_norm": 0.8054768443107605, + "learning_rate": 4.999959125204973e-06, + "loss": 0.809, + "step": 113 + }, + { + "epoch": 0.09442297073440088, + "grad_norm": 0.8101645708084106, + "learning_rate": 4.999952594932914e-06, + "loss": 0.8081, + "step": 114 + }, + { + "epoch": 0.09525124240750966, + "grad_norm": 0.905028760433197, + "learning_rate": 4.999945580943332e-06, + "loss": 0.8251, + "step": 115 + }, + { + "epoch": 0.09607951408061845, + "grad_norm": 0.8529418706893921, + "learning_rate": 4.999938083237585e-06, + "loss": 0.7984, + "step": 116 + }, + { + "epoch": 0.09690778575372722, + "grad_norm": 0.8047451376914978, + "learning_rate": 4.9999301018171225e-06, + "loss": 0.8222, + "step": 117 + }, + { + "epoch": 0.097736057426836, + "grad_norm": 0.8073400259017944, + "learning_rate": 4.99992163668349e-06, + "loss": 0.8251, + "step": 118 + }, + { + "epoch": 0.09856432909994478, + "grad_norm": 0.8470693826675415, + "learning_rate": 4.999912687838324e-06, + "loss": 0.8291, + "step": 119 + }, + { + "epoch": 0.09939260077305356, + "grad_norm": 0.8261650800704956, + "learning_rate": 4.999903255283357e-06, + "loss": 0.808, + "step": 120 + }, + { + "epoch": 0.10022087244616235, + "grad_norm": 0.8421704769134521, + "learning_rate": 4.999893339020414e-06, + "loss": 0.8295, + "step": 121 + }, + { + "epoch": 0.10104914411927111, + "grad_norm": 0.8244299292564392, + "learning_rate": 4.9998829390514134e-06, + "loss": 0.8215, + "step": 122 + }, + { + "epoch": 0.1018774157923799, + "grad_norm": 0.8365658521652222, + "learning_rate": 4.999872055378368e-06, + "loss": 0.8245, + "step": 123 + }, + { + "epoch": 0.10270568746548868, + "grad_norm": 0.9089561104774475, + "learning_rate": 4.999860688003384e-06, + "loss": 0.8125, + "step": 124 + }, + { + "epoch": 0.10353395913859746, + "grad_norm": 0.8784728646278381, + "learning_rate": 4.9998488369286595e-06, + "loss": 0.8321, + "step": 125 + }, + { + "epoch": 0.10436223081170624, + "grad_norm": 0.8596073389053345, + "learning_rate": 4.999836502156487e-06, + "loss": 0.8333, + "step": 126 + }, + { + "epoch": 0.10519050248481501, + "grad_norm": 0.8472688794136047, + "learning_rate": 4.9998236836892566e-06, + "loss": 0.8068, + "step": 127 + }, + { + "epoch": 0.1060187741579238, + "grad_norm": 0.8716503381729126, + "learning_rate": 4.999810381529447e-06, + "loss": 0.8005, + "step": 128 + }, + { + "epoch": 0.10684704583103258, + "grad_norm": 0.8416318893432617, + "learning_rate": 4.999796595679629e-06, + "loss": 0.7914, + "step": 129 + }, + { + "epoch": 0.10767531750414136, + "grad_norm": 0.838101863861084, + "learning_rate": 4.999782326142474e-06, + "loss": 0.7957, + "step": 130 + }, + { + "epoch": 0.10850358917725014, + "grad_norm": 0.8640785217285156, + "learning_rate": 4.999767572920741e-06, + "loss": 0.8077, + "step": 131 + }, + { + "epoch": 0.10933186085035891, + "grad_norm": 0.8207422494888306, + "learning_rate": 4.999752336017286e-06, + "loss": 0.7973, + "step": 132 + }, + { + "epoch": 0.1101601325234677, + "grad_norm": 0.8640758395195007, + "learning_rate": 4.9997366154350566e-06, + "loss": 0.8165, + "step": 133 + }, + { + "epoch": 0.11098840419657648, + "grad_norm": 0.7965807914733887, + "learning_rate": 4.999720411177094e-06, + "loss": 0.8059, + "step": 134 + }, + { + "epoch": 0.11181667586968526, + "grad_norm": 0.8463795185089111, + "learning_rate": 4.999703723246533e-06, + "loss": 0.8106, + "step": 135 + }, + { + "epoch": 0.11264494754279404, + "grad_norm": 0.8118059635162354, + "learning_rate": 4.999686551646604e-06, + "loss": 0.8099, + "step": 136 + }, + { + "epoch": 0.11347321921590281, + "grad_norm": 0.8256771564483643, + "learning_rate": 4.999668896380629e-06, + "loss": 0.8043, + "step": 137 + }, + { + "epoch": 0.11430149088901159, + "grad_norm": 0.8203126192092896, + "learning_rate": 4.999650757452025e-06, + "loss": 0.8029, + "step": 138 + }, + { + "epoch": 0.11512976256212037, + "grad_norm": 0.8442854881286621, + "learning_rate": 4.9996321348643e-06, + "loss": 0.8004, + "step": 139 + }, + { + "epoch": 0.11595803423522916, + "grad_norm": 0.8523190021514893, + "learning_rate": 4.999613028621059e-06, + "loss": 0.8092, + "step": 140 + }, + { + "epoch": 0.11678630590833794, + "grad_norm": 0.821428656578064, + "learning_rate": 4.999593438725997e-06, + "loss": 0.8105, + "step": 141 + }, + { + "epoch": 0.11761457758144671, + "grad_norm": 0.8406105041503906, + "learning_rate": 4.999573365182906e-06, + "loss": 0.7757, + "step": 142 + }, + { + "epoch": 0.11844284925455549, + "grad_norm": 0.8620600700378418, + "learning_rate": 4.999552807995669e-06, + "loss": 0.781, + "step": 143 + }, + { + "epoch": 0.11927112092766427, + "grad_norm": 0.920437216758728, + "learning_rate": 4.9995317671682655e-06, + "loss": 0.7579, + "step": 144 + }, + { + "epoch": 0.12009939260077306, + "grad_norm": 0.8414936661720276, + "learning_rate": 4.999510242704765e-06, + "loss": 0.8031, + "step": 145 + }, + { + "epoch": 0.12092766427388184, + "grad_norm": 0.8372176289558411, + "learning_rate": 4.999488234609332e-06, + "loss": 0.7931, + "step": 146 + }, + { + "epoch": 0.1217559359469906, + "grad_norm": 0.8785244226455688, + "learning_rate": 4.999465742886226e-06, + "loss": 0.8189, + "step": 147 + }, + { + "epoch": 0.12258420762009939, + "grad_norm": 0.8404353260993958, + "learning_rate": 4.999442767539799e-06, + "loss": 0.7899, + "step": 148 + }, + { + "epoch": 0.12341247929320817, + "grad_norm": 0.8541820049285889, + "learning_rate": 4.999419308574494e-06, + "loss": 0.7764, + "step": 149 + }, + { + "epoch": 0.12424075096631695, + "grad_norm": 0.8220177292823792, + "learning_rate": 4.999395365994854e-06, + "loss": 0.7942, + "step": 150 + }, + { + "epoch": 0.12506902263942574, + "grad_norm": 0.8823731541633606, + "learning_rate": 4.999370939805509e-06, + "loss": 0.793, + "step": 151 + }, + { + "epoch": 0.12589729431253452, + "grad_norm": 0.838860034942627, + "learning_rate": 4.999346030011186e-06, + "loss": 0.7741, + "step": 152 + }, + { + "epoch": 0.1267255659856433, + "grad_norm": 0.8162702322006226, + "learning_rate": 4.999320636616705e-06, + "loss": 0.7989, + "step": 153 + }, + { + "epoch": 0.12755383765875208, + "grad_norm": 0.8328465819358826, + "learning_rate": 4.99929475962698e-06, + "loss": 0.8055, + "step": 154 + }, + { + "epoch": 0.12838210933186084, + "grad_norm": 0.8373605608940125, + "learning_rate": 4.999268399047016e-06, + "loss": 0.801, + "step": 155 + }, + { + "epoch": 0.12921038100496962, + "grad_norm": 0.8408295512199402, + "learning_rate": 4.999241554881915e-06, + "loss": 0.7749, + "step": 156 + }, + { + "epoch": 0.1300386526780784, + "grad_norm": 0.8326342701911926, + "learning_rate": 4.999214227136871e-06, + "loss": 0.7818, + "step": 157 + }, + { + "epoch": 0.13086692435118719, + "grad_norm": 0.8064060211181641, + "learning_rate": 4.99918641581717e-06, + "loss": 0.7755, + "step": 158 + }, + { + "epoch": 0.13169519602429597, + "grad_norm": 0.8274610042572021, + "learning_rate": 4.999158120928196e-06, + "loss": 0.8008, + "step": 159 + }, + { + "epoch": 0.13252346769740475, + "grad_norm": 0.8121576309204102, + "learning_rate": 4.999129342475422e-06, + "loss": 0.7713, + "step": 160 + }, + { + "epoch": 0.13335173937051353, + "grad_norm": 0.8227710723876953, + "learning_rate": 4.9991000804644176e-06, + "loss": 0.7902, + "step": 161 + }, + { + "epoch": 0.13418001104362232, + "grad_norm": 0.8437026739120483, + "learning_rate": 4.999070334900843e-06, + "loss": 0.8176, + "step": 162 + }, + { + "epoch": 0.1350082827167311, + "grad_norm": 0.8379645943641663, + "learning_rate": 4.999040105790455e-06, + "loss": 0.7973, + "step": 163 + }, + { + "epoch": 0.13583655438983988, + "grad_norm": 0.822542130947113, + "learning_rate": 4.999009393139103e-06, + "loss": 0.7955, + "step": 164 + }, + { + "epoch": 0.13666482606294864, + "grad_norm": 0.8141928315162659, + "learning_rate": 4.9989781969527274e-06, + "loss": 0.7766, + "step": 165 + }, + { + "epoch": 0.13749309773605742, + "grad_norm": 0.8325410485267639, + "learning_rate": 4.998946517237367e-06, + "loss": 0.7737, + "step": 166 + }, + { + "epoch": 0.1383213694091662, + "grad_norm": 0.8399263024330139, + "learning_rate": 4.99891435399915e-06, + "loss": 0.7746, + "step": 167 + }, + { + "epoch": 0.13914964108227498, + "grad_norm": 0.8694303035736084, + "learning_rate": 4.9988817072443e-06, + "loss": 0.776, + "step": 168 + }, + { + "epoch": 0.13997791275538377, + "grad_norm": 0.8650498390197754, + "learning_rate": 4.998848576979135e-06, + "loss": 0.7941, + "step": 169 + }, + { + "epoch": 0.14080618442849255, + "grad_norm": 0.8246620893478394, + "learning_rate": 4.998814963210062e-06, + "loss": 0.7604, + "step": 170 + }, + { + "epoch": 0.14163445610160133, + "grad_norm": 0.8137409090995789, + "learning_rate": 4.99878086594359e-06, + "loss": 0.7681, + "step": 171 + }, + { + "epoch": 0.1424627277747101, + "grad_norm": 0.9755353927612305, + "learning_rate": 4.998746285186312e-06, + "loss": 0.7887, + "step": 172 + }, + { + "epoch": 0.1432909994478189, + "grad_norm": 0.8673408031463623, + "learning_rate": 4.998711220944921e-06, + "loss": 0.7803, + "step": 173 + }, + { + "epoch": 0.14411927112092768, + "grad_norm": 0.8312270045280457, + "learning_rate": 4.998675673226202e-06, + "loss": 0.7709, + "step": 174 + }, + { + "epoch": 0.14494754279403643, + "grad_norm": 0.8126094341278076, + "learning_rate": 4.998639642037032e-06, + "loss": 0.7822, + "step": 175 + }, + { + "epoch": 0.14577581446714521, + "grad_norm": 0.8561585545539856, + "learning_rate": 4.9986031273843834e-06, + "loss": 0.7829, + "step": 176 + }, + { + "epoch": 0.146604086140254, + "grad_norm": 0.8787612915039062, + "learning_rate": 4.9985661292753214e-06, + "loss": 0.7913, + "step": 177 + }, + { + "epoch": 0.14743235781336278, + "grad_norm": 0.8798345327377319, + "learning_rate": 4.998528647717006e-06, + "loss": 0.797, + "step": 178 + }, + { + "epoch": 0.14826062948647156, + "grad_norm": 0.7956589460372925, + "learning_rate": 4.998490682716687e-06, + "loss": 0.7571, + "step": 179 + }, + { + "epoch": 0.14908890115958034, + "grad_norm": 0.8505003452301025, + "learning_rate": 4.998452234281712e-06, + "loss": 0.7736, + "step": 180 + }, + { + "epoch": 0.14991717283268913, + "grad_norm": 0.8857074975967407, + "learning_rate": 4.998413302419519e-06, + "loss": 0.7863, + "step": 181 + }, + { + "epoch": 0.1507454445057979, + "grad_norm": 0.8762485980987549, + "learning_rate": 4.998373887137642e-06, + "loss": 0.7588, + "step": 182 + }, + { + "epoch": 0.1515737161789067, + "grad_norm": 0.8497205972671509, + "learning_rate": 4.998333988443709e-06, + "loss": 0.7773, + "step": 183 + }, + { + "epoch": 0.15240198785201547, + "grad_norm": 0.8720596432685852, + "learning_rate": 4.998293606345437e-06, + "loss": 0.7834, + "step": 184 + }, + { + "epoch": 0.15323025952512423, + "grad_norm": 0.8567108511924744, + "learning_rate": 4.998252740850641e-06, + "loss": 0.7794, + "step": 185 + }, + { + "epoch": 0.154058531198233, + "grad_norm": 0.8227219581604004, + "learning_rate": 4.998211391967228e-06, + "loss": 0.7688, + "step": 186 + }, + { + "epoch": 0.1548868028713418, + "grad_norm": 0.8321112394332886, + "learning_rate": 4.998169559703199e-06, + "loss": 0.7747, + "step": 187 + }, + { + "epoch": 0.15571507454445058, + "grad_norm": 0.8556115627288818, + "learning_rate": 4.9981272440666486e-06, + "loss": 0.7916, + "step": 188 + }, + { + "epoch": 0.15654334621755936, + "grad_norm": 0.8473951816558838, + "learning_rate": 4.998084445065763e-06, + "loss": 0.7823, + "step": 189 + }, + { + "epoch": 0.15737161789066814, + "grad_norm": 0.8551077246665955, + "learning_rate": 4.998041162708823e-06, + "loss": 0.7816, + "step": 190 + }, + { + "epoch": 0.15819988956377692, + "grad_norm": 0.945099413394928, + "learning_rate": 4.997997397004206e-06, + "loss": 0.7828, + "step": 191 + }, + { + "epoch": 0.1590281612368857, + "grad_norm": 0.8747237920761108, + "learning_rate": 4.9979531479603785e-06, + "loss": 0.7559, + "step": 192 + }, + { + "epoch": 0.1598564329099945, + "grad_norm": 0.8479958176612854, + "learning_rate": 4.997908415585902e-06, + "loss": 0.7694, + "step": 193 + }, + { + "epoch": 0.16068470458310327, + "grad_norm": 0.8659047484397888, + "learning_rate": 4.997863199889433e-06, + "loss": 0.7534, + "step": 194 + }, + { + "epoch": 0.16151297625621203, + "grad_norm": 0.8744727969169617, + "learning_rate": 4.997817500879719e-06, + "loss": 0.7499, + "step": 195 + }, + { + "epoch": 0.1623412479293208, + "grad_norm": 0.896468997001648, + "learning_rate": 4.997771318565603e-06, + "loss": 0.7974, + "step": 196 + }, + { + "epoch": 0.1631695196024296, + "grad_norm": 0.9058924913406372, + "learning_rate": 4.997724652956021e-06, + "loss": 0.7753, + "step": 197 + }, + { + "epoch": 0.16399779127553837, + "grad_norm": 0.8904663920402527, + "learning_rate": 4.9976775040600014e-06, + "loss": 0.7569, + "step": 198 + }, + { + "epoch": 0.16482606294864716, + "grad_norm": 0.8709238767623901, + "learning_rate": 4.997629871886669e-06, + "loss": 0.763, + "step": 199 + }, + { + "epoch": 0.16565433462175594, + "grad_norm": 0.8400478959083557, + "learning_rate": 4.997581756445238e-06, + "loss": 0.7795, + "step": 200 + }, + { + "epoch": 0.16648260629486472, + "grad_norm": 0.8458966612815857, + "learning_rate": 4.9975331577450204e-06, + "loss": 0.7597, + "step": 201 + }, + { + "epoch": 0.1673108779679735, + "grad_norm": 0.8707828521728516, + "learning_rate": 4.997484075795417e-06, + "loss": 0.759, + "step": 202 + }, + { + "epoch": 0.16813914964108229, + "grad_norm": 0.8774807453155518, + "learning_rate": 4.997434510605927e-06, + "loss": 0.7706, + "step": 203 + }, + { + "epoch": 0.16896742131419104, + "grad_norm": 0.8987929821014404, + "learning_rate": 4.99738446218614e-06, + "loss": 0.7319, + "step": 204 + }, + { + "epoch": 0.16979569298729982, + "grad_norm": 0.8869186639785767, + "learning_rate": 4.99733393054574e-06, + "loss": 0.7812, + "step": 205 + }, + { + "epoch": 0.1706239646604086, + "grad_norm": 0.860407829284668, + "learning_rate": 4.997282915694505e-06, + "loss": 0.7524, + "step": 206 + }, + { + "epoch": 0.1714522363335174, + "grad_norm": 0.8719788789749146, + "learning_rate": 4.9972314176423035e-06, + "loss": 0.7796, + "step": 207 + }, + { + "epoch": 0.17228050800662617, + "grad_norm": 0.8596137762069702, + "learning_rate": 4.997179436399103e-06, + "loss": 0.7554, + "step": 208 + }, + { + "epoch": 0.17310877967973495, + "grad_norm": 0.8505175709724426, + "learning_rate": 4.997126971974959e-06, + "loss": 0.7607, + "step": 209 + }, + { + "epoch": 0.17393705135284374, + "grad_norm": 0.7858585119247437, + "learning_rate": 4.9970740243800245e-06, + "loss": 0.7622, + "step": 210 + }, + { + "epoch": 0.17476532302595252, + "grad_norm": 0.8634621500968933, + "learning_rate": 4.997020593624543e-06, + "loss": 0.7659, + "step": 211 + }, + { + "epoch": 0.1755935946990613, + "grad_norm": 0.8650951385498047, + "learning_rate": 4.9969666797188545e-06, + "loss": 0.7689, + "step": 212 + }, + { + "epoch": 0.17642186637217008, + "grad_norm": 0.8602325916290283, + "learning_rate": 4.996912282673389e-06, + "loss": 0.7695, + "step": 213 + }, + { + "epoch": 0.17725013804527884, + "grad_norm": 0.8453742265701294, + "learning_rate": 4.9968574024986735e-06, + "loss": 0.7434, + "step": 214 + }, + { + "epoch": 0.17807840971838762, + "grad_norm": 0.8542691469192505, + "learning_rate": 4.9968020392053255e-06, + "loss": 0.7298, + "step": 215 + }, + { + "epoch": 0.1789066813914964, + "grad_norm": 0.8806673288345337, + "learning_rate": 4.996746192804058e-06, + "loss": 0.7649, + "step": 216 + }, + { + "epoch": 0.17973495306460519, + "grad_norm": 0.8451183438301086, + "learning_rate": 4.9966898633056765e-06, + "loss": 0.7681, + "step": 217 + }, + { + "epoch": 0.18056322473771397, + "grad_norm": 0.8664626479148865, + "learning_rate": 4.99663305072108e-06, + "loss": 0.7753, + "step": 218 + }, + { + "epoch": 0.18139149641082275, + "grad_norm": 0.8828144669532776, + "learning_rate": 4.996575755061262e-06, + "loss": 0.7514, + "step": 219 + }, + { + "epoch": 0.18221976808393153, + "grad_norm": 0.8493948578834534, + "learning_rate": 4.996517976337308e-06, + "loss": 0.7684, + "step": 220 + }, + { + "epoch": 0.18304803975704032, + "grad_norm": 0.8561757802963257, + "learning_rate": 4.9964597145603975e-06, + "loss": 0.7723, + "step": 221 + }, + { + "epoch": 0.1838763114301491, + "grad_norm": 0.8863773941993713, + "learning_rate": 4.996400969741805e-06, + "loss": 0.7732, + "step": 222 + }, + { + "epoch": 0.18470458310325788, + "grad_norm": 0.8571644425392151, + "learning_rate": 4.996341741892895e-06, + "loss": 0.744, + "step": 223 + }, + { + "epoch": 0.18553285477636663, + "grad_norm": 0.8622035980224609, + "learning_rate": 4.9962820310251286e-06, + "loss": 0.7607, + "step": 224 + }, + { + "epoch": 0.18636112644947542, + "grad_norm": 0.8904410600662231, + "learning_rate": 4.9962218371500595e-06, + "loss": 0.7648, + "step": 225 + }, + { + "epoch": 0.1871893981225842, + "grad_norm": 0.8975846171379089, + "learning_rate": 4.996161160279335e-06, + "loss": 0.7469, + "step": 226 + }, + { + "epoch": 0.18801766979569298, + "grad_norm": 0.8788724541664124, + "learning_rate": 4.996100000424694e-06, + "loss": 0.7473, + "step": 227 + }, + { + "epoch": 0.18884594146880176, + "grad_norm": 0.9446991086006165, + "learning_rate": 4.996038357597971e-06, + "loss": 0.7635, + "step": 228 + }, + { + "epoch": 0.18967421314191055, + "grad_norm": 0.86910480260849, + "learning_rate": 4.995976231811094e-06, + "loss": 0.7539, + "step": 229 + }, + { + "epoch": 0.19050248481501933, + "grad_norm": 0.862197756767273, + "learning_rate": 4.995913623076084e-06, + "loss": 0.7546, + "step": 230 + }, + { + "epoch": 0.1913307564881281, + "grad_norm": 0.906878650188446, + "learning_rate": 4.995850531405054e-06, + "loss": 0.7737, + "step": 231 + }, + { + "epoch": 0.1921590281612369, + "grad_norm": 0.8557183742523193, + "learning_rate": 4.995786956810212e-06, + "loss": 0.7663, + "step": 232 + }, + { + "epoch": 0.19298729983434568, + "grad_norm": 0.8125383853912354, + "learning_rate": 4.995722899303859e-06, + "loss": 0.7647, + "step": 233 + }, + { + "epoch": 0.19381557150745443, + "grad_norm": 0.8610332012176514, + "learning_rate": 4.995658358898391e-06, + "loss": 0.7541, + "step": 234 + }, + { + "epoch": 0.19464384318056321, + "grad_norm": 0.8620829582214355, + "learning_rate": 4.995593335606294e-06, + "loss": 0.7625, + "step": 235 + }, + { + "epoch": 0.195472114853672, + "grad_norm": 0.8296957612037659, + "learning_rate": 4.9955278294401496e-06, + "loss": 0.7536, + "step": 236 + }, + { + "epoch": 0.19630038652678078, + "grad_norm": 0.8974814414978027, + "learning_rate": 4.995461840412634e-06, + "loss": 0.7651, + "step": 237 + }, + { + "epoch": 0.19712865819988956, + "grad_norm": 0.8701095581054688, + "learning_rate": 4.995395368536515e-06, + "loss": 0.788, + "step": 238 + }, + { + "epoch": 0.19795692987299834, + "grad_norm": 0.8556636571884155, + "learning_rate": 4.995328413824653e-06, + "loss": 0.7649, + "step": 239 + }, + { + "epoch": 0.19878520154610713, + "grad_norm": 0.836686909198761, + "learning_rate": 4.995260976290005e-06, + "loss": 0.7583, + "step": 240 + }, + { + "epoch": 0.1996134732192159, + "grad_norm": 0.8807066082954407, + "learning_rate": 4.995193055945618e-06, + "loss": 0.7766, + "step": 241 + }, + { + "epoch": 0.2004417448923247, + "grad_norm": 0.8363582491874695, + "learning_rate": 4.995124652804635e-06, + "loss": 0.7655, + "step": 242 + }, + { + "epoch": 0.20127001656543347, + "grad_norm": 0.8699461817741394, + "learning_rate": 4.9950557668802905e-06, + "loss": 0.743, + "step": 243 + }, + { + "epoch": 0.20209828823854223, + "grad_norm": 0.8571060299873352, + "learning_rate": 4.994986398185914e-06, + "loss": 0.7512, + "step": 244 + }, + { + "epoch": 0.202926559911651, + "grad_norm": 0.8429675102233887, + "learning_rate": 4.994916546734927e-06, + "loss": 0.7589, + "step": 245 + }, + { + "epoch": 0.2037548315847598, + "grad_norm": 0.874822199344635, + "learning_rate": 4.994846212540846e-06, + "loss": 0.7534, + "step": 246 + }, + { + "epoch": 0.20458310325786858, + "grad_norm": 0.878341555595398, + "learning_rate": 4.994775395617279e-06, + "loss": 0.7709, + "step": 247 + }, + { + "epoch": 0.20541137493097736, + "grad_norm": 0.897961437702179, + "learning_rate": 4.99470409597793e-06, + "loss": 0.7768, + "step": 248 + }, + { + "epoch": 0.20623964660408614, + "grad_norm": 0.8439584374427795, + "learning_rate": 4.994632313636593e-06, + "loss": 0.7558, + "step": 249 + }, + { + "epoch": 0.20706791827719492, + "grad_norm": 0.8394468426704407, + "learning_rate": 4.994560048607159e-06, + "loss": 0.7492, + "step": 250 + }, + { + "epoch": 0.2078961899503037, + "grad_norm": 0.8489627242088318, + "learning_rate": 4.9944873009036074e-06, + "loss": 0.7365, + "step": 251 + }, + { + "epoch": 0.2087244616234125, + "grad_norm": 0.8675844669342041, + "learning_rate": 4.994414070540018e-06, + "loss": 0.762, + "step": 252 + }, + { + "epoch": 0.20955273329652127, + "grad_norm": 0.8228297829627991, + "learning_rate": 4.994340357530558e-06, + "loss": 0.7475, + "step": 253 + }, + { + "epoch": 0.21038100496963003, + "grad_norm": 0.8794129490852356, + "learning_rate": 4.9942661618894915e-06, + "loss": 0.7631, + "step": 254 + }, + { + "epoch": 0.2112092766427388, + "grad_norm": 0.8616135716438293, + "learning_rate": 4.9941914836311735e-06, + "loss": 0.7749, + "step": 255 + }, + { + "epoch": 0.2120375483158476, + "grad_norm": 0.8470495939254761, + "learning_rate": 4.994116322770054e-06, + "loss": 0.7484, + "step": 256 + }, + { + "epoch": 0.21286581998895637, + "grad_norm": 0.9346716403961182, + "learning_rate": 4.9940406793206755e-06, + "loss": 0.7557, + "step": 257 + }, + { + "epoch": 0.21369409166206516, + "grad_norm": 0.8993160724639893, + "learning_rate": 4.993964553297676e-06, + "loss": 0.7422, + "step": 258 + }, + { + "epoch": 0.21452236333517394, + "grad_norm": 0.8257086277008057, + "learning_rate": 4.993887944715783e-06, + "loss": 0.7783, + "step": 259 + }, + { + "epoch": 0.21535063500828272, + "grad_norm": 0.8950703144073486, + "learning_rate": 4.993810853589819e-06, + "loss": 0.7311, + "step": 260 + }, + { + "epoch": 0.2161789066813915, + "grad_norm": 0.8733702898025513, + "learning_rate": 4.993733279934704e-06, + "loss": 0.7572, + "step": 261 + }, + { + "epoch": 0.21700717835450029, + "grad_norm": 0.8322939276695251, + "learning_rate": 4.993655223765444e-06, + "loss": 0.7632, + "step": 262 + }, + { + "epoch": 0.21783545002760907, + "grad_norm": 0.8666137456893921, + "learning_rate": 4.993576685097145e-06, + "loss": 0.7619, + "step": 263 + }, + { + "epoch": 0.21866372170071782, + "grad_norm": 0.8770918250083923, + "learning_rate": 4.993497663945002e-06, + "loss": 0.7538, + "step": 264 + }, + { + "epoch": 0.2194919933738266, + "grad_norm": 0.8940248489379883, + "learning_rate": 4.9934181603243045e-06, + "loss": 0.7512, + "step": 265 + }, + { + "epoch": 0.2203202650469354, + "grad_norm": 0.8979714512825012, + "learning_rate": 4.993338174250437e-06, + "loss": 0.7504, + "step": 266 + }, + { + "epoch": 0.22114853672004417, + "grad_norm": 0.8802001476287842, + "learning_rate": 4.993257705738876e-06, + "loss": 0.7501, + "step": 267 + }, + { + "epoch": 0.22197680839315295, + "grad_norm": 0.8734432458877563, + "learning_rate": 4.993176754805189e-06, + "loss": 0.7452, + "step": 268 + }, + { + "epoch": 0.22280508006626173, + "grad_norm": 0.9127711057662964, + "learning_rate": 4.993095321465042e-06, + "loss": 0.7404, + "step": 269 + }, + { + "epoch": 0.22363335173937052, + "grad_norm": 0.8533952832221985, + "learning_rate": 4.993013405734191e-06, + "loss": 0.7382, + "step": 270 + }, + { + "epoch": 0.2244616234124793, + "grad_norm": 0.8463877439498901, + "learning_rate": 4.9929310076284845e-06, + "loss": 0.7574, + "step": 271 + }, + { + "epoch": 0.22528989508558808, + "grad_norm": 0.8598008155822754, + "learning_rate": 4.992848127163868e-06, + "loss": 0.7624, + "step": 272 + }, + { + "epoch": 0.22611816675869686, + "grad_norm": 0.8759154081344604, + "learning_rate": 4.992764764356376e-06, + "loss": 0.7637, + "step": 273 + }, + { + "epoch": 0.22694643843180562, + "grad_norm": 0.8880444765090942, + "learning_rate": 4.992680919222141e-06, + "loss": 0.7514, + "step": 274 + }, + { + "epoch": 0.2277747101049144, + "grad_norm": 0.871366024017334, + "learning_rate": 4.9925965917773826e-06, + "loss": 0.7569, + "step": 275 + }, + { + "epoch": 0.22860298177802318, + "grad_norm": 0.862654983997345, + "learning_rate": 4.99251178203842e-06, + "loss": 0.7349, + "step": 276 + }, + { + "epoch": 0.22943125345113197, + "grad_norm": 0.8615298867225647, + "learning_rate": 4.992426490021662e-06, + "loss": 0.7438, + "step": 277 + }, + { + "epoch": 0.23025952512424075, + "grad_norm": 0.8735869526863098, + "learning_rate": 4.992340715743614e-06, + "loss": 0.7755, + "step": 278 + }, + { + "epoch": 0.23108779679734953, + "grad_norm": 0.9082722067832947, + "learning_rate": 4.9922544592208695e-06, + "loss": 0.7561, + "step": 279 + }, + { + "epoch": 0.23191606847045831, + "grad_norm": 0.8639047145843506, + "learning_rate": 4.992167720470119e-06, + "loss": 0.7493, + "step": 280 + }, + { + "epoch": 0.2327443401435671, + "grad_norm": 0.891640305519104, + "learning_rate": 4.9920804995081475e-06, + "loss": 0.75, + "step": 281 + }, + { + "epoch": 0.23357261181667588, + "grad_norm": 0.8880137205123901, + "learning_rate": 4.99199279635183e-06, + "loss": 0.7801, + "step": 282 + }, + { + "epoch": 0.23440088348978466, + "grad_norm": 0.8600473403930664, + "learning_rate": 4.991904611018137e-06, + "loss": 0.7477, + "step": 283 + }, + { + "epoch": 0.23522915516289342, + "grad_norm": 0.8685399889945984, + "learning_rate": 4.99181594352413e-06, + "loss": 0.741, + "step": 284 + }, + { + "epoch": 0.2360574268360022, + "grad_norm": 0.9006181955337524, + "learning_rate": 4.991726793886966e-06, + "loss": 0.7288, + "step": 285 + }, + { + "epoch": 0.23688569850911098, + "grad_norm": 0.863895833492279, + "learning_rate": 4.991637162123896e-06, + "loss": 0.7375, + "step": 286 + }, + { + "epoch": 0.23771397018221976, + "grad_norm": 0.8620381951332092, + "learning_rate": 4.9915470482522625e-06, + "loss": 0.7249, + "step": 287 + }, + { + "epoch": 0.23854224185532855, + "grad_norm": 0.9634964466094971, + "learning_rate": 4.991456452289499e-06, + "loss": 0.7313, + "step": 288 + }, + { + "epoch": 0.23937051352843733, + "grad_norm": 0.9154514670372009, + "learning_rate": 4.9913653742531385e-06, + "loss": 0.782, + "step": 289 + }, + { + "epoch": 0.2401987852015461, + "grad_norm": 0.9179158806800842, + "learning_rate": 4.991273814160802e-06, + "loss": 0.7388, + "step": 290 + }, + { + "epoch": 0.2410270568746549, + "grad_norm": 0.8609046936035156, + "learning_rate": 4.991181772030206e-06, + "loss": 0.7393, + "step": 291 + }, + { + "epoch": 0.24185532854776368, + "grad_norm": 0.8689826130867004, + "learning_rate": 4.991089247879159e-06, + "loss": 0.7302, + "step": 292 + }, + { + "epoch": 0.24268360022087246, + "grad_norm": 0.8414047956466675, + "learning_rate": 4.990996241725565e-06, + "loss": 0.7496, + "step": 293 + }, + { + "epoch": 0.2435118718939812, + "grad_norm": 0.8621289134025574, + "learning_rate": 4.990902753587418e-06, + "loss": 0.7317, + "step": 294 + }, + { + "epoch": 0.24434014356709, + "grad_norm": 0.856566309928894, + "learning_rate": 4.990808783482809e-06, + "loss": 0.7169, + "step": 295 + }, + { + "epoch": 0.24516841524019878, + "grad_norm": 0.8549672961235046, + "learning_rate": 4.990714331429919e-06, + "loss": 0.7357, + "step": 296 + }, + { + "epoch": 0.24599668691330756, + "grad_norm": 0.8813024759292603, + "learning_rate": 4.990619397447024e-06, + "loss": 0.7437, + "step": 297 + }, + { + "epoch": 0.24682495858641634, + "grad_norm": 0.9008268713951111, + "learning_rate": 4.990523981552493e-06, + "loss": 0.7652, + "step": 298 + }, + { + "epoch": 0.24765323025952513, + "grad_norm": 0.910277783870697, + "learning_rate": 4.990428083764788e-06, + "loss": 0.7598, + "step": 299 + }, + { + "epoch": 0.2484815019326339, + "grad_norm": 0.9261599183082581, + "learning_rate": 4.990331704102464e-06, + "loss": 0.7199, + "step": 300 + }, + { + "epoch": 0.2493097736057427, + "grad_norm": 0.9166674613952637, + "learning_rate": 4.99023484258417e-06, + "loss": 0.7414, + "step": 301 + }, + { + "epoch": 0.2501380452788515, + "grad_norm": 0.8701214790344238, + "learning_rate": 4.990137499228648e-06, + "loss": 0.7326, + "step": 302 + }, + { + "epoch": 0.25096631695196026, + "grad_norm": 0.9120466113090515, + "learning_rate": 4.990039674054732e-06, + "loss": 0.7574, + "step": 303 + }, + { + "epoch": 0.25179458862506904, + "grad_norm": 0.9449286460876465, + "learning_rate": 4.989941367081351e-06, + "loss": 0.725, + "step": 304 + }, + { + "epoch": 0.2526228602981778, + "grad_norm": 0.8989110589027405, + "learning_rate": 4.989842578327526e-06, + "loss": 0.7569, + "step": 305 + }, + { + "epoch": 0.2534511319712866, + "grad_norm": 0.891677975654602, + "learning_rate": 4.989743307812373e-06, + "loss": 0.7135, + "step": 306 + }, + { + "epoch": 0.2542794036443954, + "grad_norm": 0.8689535856246948, + "learning_rate": 4.989643555555099e-06, + "loss": 0.7439, + "step": 307 + }, + { + "epoch": 0.25510767531750417, + "grad_norm": 0.8779630064964294, + "learning_rate": 4.989543321575004e-06, + "loss": 0.7322, + "step": 308 + }, + { + "epoch": 0.2559359469906129, + "grad_norm": 0.8977622985839844, + "learning_rate": 4.9894426058914856e-06, + "loss": 0.7544, + "step": 309 + }, + { + "epoch": 0.2567642186637217, + "grad_norm": 0.8216757774353027, + "learning_rate": 4.989341408524028e-06, + "loss": 0.7344, + "step": 310 + }, + { + "epoch": 0.25759249033683046, + "grad_norm": 0.8494759798049927, + "learning_rate": 4.9892397294922126e-06, + "loss": 0.7389, + "step": 311 + }, + { + "epoch": 0.25842076200993924, + "grad_norm": 0.881244957447052, + "learning_rate": 4.989137568815715e-06, + "loss": 0.7121, + "step": 312 + }, + { + "epoch": 0.259249033683048, + "grad_norm": 0.8679410219192505, + "learning_rate": 4.9890349265143005e-06, + "loss": 0.7469, + "step": 313 + }, + { + "epoch": 0.2600773053561568, + "grad_norm": 0.9746971130371094, + "learning_rate": 4.988931802607831e-06, + "loss": 0.7585, + "step": 314 + }, + { + "epoch": 0.2609055770292656, + "grad_norm": 0.9222471117973328, + "learning_rate": 4.988828197116259e-06, + "loss": 0.761, + "step": 315 + }, + { + "epoch": 0.26173384870237437, + "grad_norm": 0.9104124903678894, + "learning_rate": 4.988724110059631e-06, + "loss": 0.7652, + "step": 316 + }, + { + "epoch": 0.26256212037548315, + "grad_norm": 0.8876475095748901, + "learning_rate": 4.988619541458088e-06, + "loss": 0.7529, + "step": 317 + }, + { + "epoch": 0.26339039204859194, + "grad_norm": 0.8705421686172485, + "learning_rate": 4.988514491331861e-06, + "loss": 0.7437, + "step": 318 + }, + { + "epoch": 0.2642186637217007, + "grad_norm": 0.8594399690628052, + "learning_rate": 4.9884089597012784e-06, + "loss": 0.7406, + "step": 319 + }, + { + "epoch": 0.2650469353948095, + "grad_norm": 0.8807039856910706, + "learning_rate": 4.988302946586759e-06, + "loss": 0.7429, + "step": 320 + }, + { + "epoch": 0.2658752070679183, + "grad_norm": 0.8881412148475647, + "learning_rate": 4.9881964520088144e-06, + "loss": 0.7234, + "step": 321 + }, + { + "epoch": 0.26670347874102707, + "grad_norm": 0.8764941096305847, + "learning_rate": 4.988089475988052e-06, + "loss": 0.7514, + "step": 322 + }, + { + "epoch": 0.26753175041413585, + "grad_norm": 0.8992927670478821, + "learning_rate": 4.987982018545169e-06, + "loss": 0.7143, + "step": 323 + }, + { + "epoch": 0.26836002208724463, + "grad_norm": 0.8961546421051025, + "learning_rate": 4.9878740797009585e-06, + "loss": 0.7396, + "step": 324 + }, + { + "epoch": 0.2691882937603534, + "grad_norm": 0.9031578302383423, + "learning_rate": 4.987765659476304e-06, + "loss": 0.7407, + "step": 325 + }, + { + "epoch": 0.2700165654334622, + "grad_norm": 0.8693655729293823, + "learning_rate": 4.987656757892186e-06, + "loss": 0.7215, + "step": 326 + }, + { + "epoch": 0.270844837106571, + "grad_norm": 0.8739856481552124, + "learning_rate": 4.987547374969676e-06, + "loss": 0.7471, + "step": 327 + }, + { + "epoch": 0.27167310877967976, + "grad_norm": 0.8916322588920593, + "learning_rate": 4.987437510729936e-06, + "loss": 0.7459, + "step": 328 + }, + { + "epoch": 0.2725013804527885, + "grad_norm": 0.890178382396698, + "learning_rate": 4.987327165194225e-06, + "loss": 0.7322, + "step": 329 + }, + { + "epoch": 0.27332965212589727, + "grad_norm": 0.8611598014831543, + "learning_rate": 4.987216338383896e-06, + "loss": 0.733, + "step": 330 + }, + { + "epoch": 0.27415792379900605, + "grad_norm": 0.87791907787323, + "learning_rate": 4.98710503032039e-06, + "loss": 0.7445, + "step": 331 + }, + { + "epoch": 0.27498619547211484, + "grad_norm": 0.963479220867157, + "learning_rate": 4.986993241025245e-06, + "loss": 0.7376, + "step": 332 + }, + { + "epoch": 0.2758144671452236, + "grad_norm": 0.9083762764930725, + "learning_rate": 4.986880970520092e-06, + "loss": 0.7258, + "step": 333 + }, + { + "epoch": 0.2766427388183324, + "grad_norm": 1.048991084098816, + "learning_rate": 4.9867682188266524e-06, + "loss": 0.7399, + "step": 334 + }, + { + "epoch": 0.2774710104914412, + "grad_norm": 0.9404491782188416, + "learning_rate": 4.986654985966745e-06, + "loss": 0.7133, + "step": 335 + }, + { + "epoch": 0.27829928216454997, + "grad_norm": 0.9168346524238586, + "learning_rate": 4.986541271962278e-06, + "loss": 0.74, + "step": 336 + }, + { + "epoch": 0.27912755383765875, + "grad_norm": 0.8716689944267273, + "learning_rate": 4.986427076835253e-06, + "loss": 0.7249, + "step": 337 + }, + { + "epoch": 0.27995582551076753, + "grad_norm": 0.9097291827201843, + "learning_rate": 4.986312400607767e-06, + "loss": 0.7151, + "step": 338 + }, + { + "epoch": 0.2807840971838763, + "grad_norm": 0.8621960282325745, + "learning_rate": 4.98619724330201e-06, + "loss": 0.7125, + "step": 339 + }, + { + "epoch": 0.2816123688569851, + "grad_norm": 0.9823791980743408, + "learning_rate": 4.986081604940261e-06, + "loss": 0.7191, + "step": 340 + }, + { + "epoch": 0.2824406405300939, + "grad_norm": 0.8999047875404358, + "learning_rate": 4.9859654855448966e-06, + "loss": 0.7175, + "step": 341 + }, + { + "epoch": 0.28326891220320266, + "grad_norm": 0.8690859079360962, + "learning_rate": 4.985848885138386e-06, + "loss": 0.7309, + "step": 342 + }, + { + "epoch": 0.28409718387631144, + "grad_norm": 0.889500617980957, + "learning_rate": 4.985731803743287e-06, + "loss": 0.7391, + "step": 343 + }, + { + "epoch": 0.2849254555494202, + "grad_norm": 0.8992016911506653, + "learning_rate": 4.985614241382257e-06, + "loss": 0.7437, + "step": 344 + }, + { + "epoch": 0.285753727222529, + "grad_norm": 0.9100068211555481, + "learning_rate": 4.985496198078041e-06, + "loss": 0.722, + "step": 345 + }, + { + "epoch": 0.2865819988956378, + "grad_norm": 0.9013107419013977, + "learning_rate": 4.985377673853481e-06, + "loss": 0.7434, + "step": 346 + }, + { + "epoch": 0.2874102705687466, + "grad_norm": 0.8794704675674438, + "learning_rate": 4.9852586687315094e-06, + "loss": 0.7387, + "step": 347 + }, + { + "epoch": 0.28823854224185536, + "grad_norm": 0.9288201928138733, + "learning_rate": 4.9851391827351526e-06, + "loss": 0.7165, + "step": 348 + }, + { + "epoch": 0.2890668139149641, + "grad_norm": 0.8735847473144531, + "learning_rate": 4.98501921588753e-06, + "loss": 0.7479, + "step": 349 + }, + { + "epoch": 0.28989508558807286, + "grad_norm": 0.890616238117218, + "learning_rate": 4.984898768211853e-06, + "loss": 0.76, + "step": 350 + }, + { + "epoch": 0.29072335726118165, + "grad_norm": 0.9181028604507446, + "learning_rate": 4.98477783973143e-06, + "loss": 0.7366, + "step": 351 + }, + { + "epoch": 0.29155162893429043, + "grad_norm": 0.9947429299354553, + "learning_rate": 4.984656430469657e-06, + "loss": 0.7199, + "step": 352 + }, + { + "epoch": 0.2923799006073992, + "grad_norm": 0.8730769157409668, + "learning_rate": 4.984534540450027e-06, + "loss": 0.7153, + "step": 353 + }, + { + "epoch": 0.293208172280508, + "grad_norm": 0.8675997853279114, + "learning_rate": 4.9844121696961235e-06, + "loss": 0.7196, + "step": 354 + }, + { + "epoch": 0.2940364439536168, + "grad_norm": 0.9558542966842651, + "learning_rate": 4.984289318231624e-06, + "loss": 0.7311, + "step": 355 + }, + { + "epoch": 0.29486471562672556, + "grad_norm": 0.8817659616470337, + "learning_rate": 4.9841659860803e-06, + "loss": 0.7367, + "step": 356 + }, + { + "epoch": 0.29569298729983434, + "grad_norm": 0.8679526448249817, + "learning_rate": 4.984042173266016e-06, + "loss": 0.719, + "step": 357 + }, + { + "epoch": 0.2965212589729431, + "grad_norm": 0.9102020859718323, + "learning_rate": 4.983917879812726e-06, + "loss": 0.7301, + "step": 358 + }, + { + "epoch": 0.2973495306460519, + "grad_norm": 0.8987215757369995, + "learning_rate": 4.983793105744482e-06, + "loss": 0.7557, + "step": 359 + }, + { + "epoch": 0.2981778023191607, + "grad_norm": 0.8779832720756531, + "learning_rate": 4.983667851085426e-06, + "loss": 0.7471, + "step": 360 + }, + { + "epoch": 0.29900607399226947, + "grad_norm": 0.8762708306312561, + "learning_rate": 4.983542115859792e-06, + "loss": 0.7676, + "step": 361 + }, + { + "epoch": 0.29983434566537825, + "grad_norm": 0.910353422164917, + "learning_rate": 4.98341590009191e-06, + "loss": 0.7285, + "step": 362 + }, + { + "epoch": 0.30066261733848704, + "grad_norm": 0.9385405778884888, + "learning_rate": 4.983289203806202e-06, + "loss": 0.7142, + "step": 363 + }, + { + "epoch": 0.3014908890115958, + "grad_norm": 1.016781210899353, + "learning_rate": 4.983162027027182e-06, + "loss": 0.7408, + "step": 364 + }, + { + "epoch": 0.3023191606847046, + "grad_norm": 0.8642508387565613, + "learning_rate": 4.983034369779457e-06, + "loss": 0.7397, + "step": 365 + }, + { + "epoch": 0.3031474323578134, + "grad_norm": 0.9469170570373535, + "learning_rate": 4.982906232087728e-06, + "loss": 0.7158, + "step": 366 + }, + { + "epoch": 0.30397570403092217, + "grad_norm": 0.8818212151527405, + "learning_rate": 4.982777613976789e-06, + "loss": 0.7408, + "step": 367 + }, + { + "epoch": 0.30480397570403095, + "grad_norm": 0.8639838099479675, + "learning_rate": 4.982648515471526e-06, + "loss": 0.7217, + "step": 368 + }, + { + "epoch": 0.3056322473771397, + "grad_norm": 0.923780083656311, + "learning_rate": 4.982518936596917e-06, + "loss": 0.7284, + "step": 369 + }, + { + "epoch": 0.30646051905024846, + "grad_norm": 0.8992834687232971, + "learning_rate": 4.982388877378037e-06, + "loss": 0.7395, + "step": 370 + }, + { + "epoch": 0.30728879072335724, + "grad_norm": 0.8671404123306274, + "learning_rate": 4.982258337840049e-06, + "loss": 0.7157, + "step": 371 + }, + { + "epoch": 0.308117062396466, + "grad_norm": 0.8689168691635132, + "learning_rate": 4.982127318008212e-06, + "loss": 0.7398, + "step": 372 + }, + { + "epoch": 0.3089453340695748, + "grad_norm": 0.8964621424674988, + "learning_rate": 4.9819958179078765e-06, + "loss": 0.7385, + "step": 373 + }, + { + "epoch": 0.3097736057426836, + "grad_norm": 0.8984093070030212, + "learning_rate": 4.981863837564488e-06, + "loss": 0.7253, + "step": 374 + }, + { + "epoch": 0.31060187741579237, + "grad_norm": 0.8776745200157166, + "learning_rate": 4.981731377003583e-06, + "loss": 0.7126, + "step": 375 + }, + { + "epoch": 0.31143014908890115, + "grad_norm": 0.8782477378845215, + "learning_rate": 4.98159843625079e-06, + "loss": 0.7164, + "step": 376 + }, + { + "epoch": 0.31225842076200994, + "grad_norm": 0.8993703722953796, + "learning_rate": 4.981465015331832e-06, + "loss": 0.7391, + "step": 377 + }, + { + "epoch": 0.3130866924351187, + "grad_norm": 0.8975916504859924, + "learning_rate": 4.981331114272527e-06, + "loss": 0.7385, + "step": 378 + }, + { + "epoch": 0.3139149641082275, + "grad_norm": 0.8726824522018433, + "learning_rate": 4.9811967330987816e-06, + "loss": 0.7426, + "step": 379 + }, + { + "epoch": 0.3147432357813363, + "grad_norm": 0.8741514682769775, + "learning_rate": 4.981061871836597e-06, + "loss": 0.7246, + "step": 380 + }, + { + "epoch": 0.31557150745444507, + "grad_norm": 0.8807122707366943, + "learning_rate": 4.98092653051207e-06, + "loss": 0.7364, + "step": 381 + }, + { + "epoch": 0.31639977912755385, + "grad_norm": 0.835006058216095, + "learning_rate": 4.980790709151385e-06, + "loss": 0.7248, + "step": 382 + }, + { + "epoch": 0.31722805080066263, + "grad_norm": 0.9473997354507446, + "learning_rate": 4.980654407780824e-06, + "loss": 0.7331, + "step": 383 + }, + { + "epoch": 0.3180563224737714, + "grad_norm": 0.8332815170288086, + "learning_rate": 4.980517626426758e-06, + "loss": 0.7299, + "step": 384 + }, + { + "epoch": 0.3188845941468802, + "grad_norm": 0.8536360859870911, + "learning_rate": 4.980380365115656e-06, + "loss": 0.7095, + "step": 385 + }, + { + "epoch": 0.319712865819989, + "grad_norm": 0.9488157629966736, + "learning_rate": 4.980242623874075e-06, + "loss": 0.7382, + "step": 386 + }, + { + "epoch": 0.32054113749309776, + "grad_norm": 0.8623067736625671, + "learning_rate": 4.980104402728666e-06, + "loss": 0.7187, + "step": 387 + }, + { + "epoch": 0.32136940916620654, + "grad_norm": 0.9110252857208252, + "learning_rate": 4.979965701706175e-06, + "loss": 0.7354, + "step": 388 + }, + { + "epoch": 0.32219768083931527, + "grad_norm": 0.8966642618179321, + "learning_rate": 4.979826520833439e-06, + "loss": 0.7251, + "step": 389 + }, + { + "epoch": 0.32302595251242405, + "grad_norm": 0.856674075126648, + "learning_rate": 4.979686860137387e-06, + "loss": 0.723, + "step": 390 + }, + { + "epoch": 0.32385422418553284, + "grad_norm": 0.8623953461647034, + "learning_rate": 4.979546719645043e-06, + "loss": 0.7224, + "step": 391 + }, + { + "epoch": 0.3246824958586416, + "grad_norm": 0.8685624599456787, + "learning_rate": 4.979406099383523e-06, + "loss": 0.7283, + "step": 392 + }, + { + "epoch": 0.3255107675317504, + "grad_norm": 0.8610525727272034, + "learning_rate": 4.979264999380035e-06, + "loss": 0.7385, + "step": 393 + }, + { + "epoch": 0.3263390392048592, + "grad_norm": 0.8860796093940735, + "learning_rate": 4.979123419661882e-06, + "loss": 0.743, + "step": 394 + }, + { + "epoch": 0.32716731087796797, + "grad_norm": 0.8748682737350464, + "learning_rate": 4.978981360256457e-06, + "loss": 0.7447, + "step": 395 + }, + { + "epoch": 0.32799558255107675, + "grad_norm": 0.8750934600830078, + "learning_rate": 4.978838821191247e-06, + "loss": 0.7259, + "step": 396 + }, + { + "epoch": 0.32882385422418553, + "grad_norm": 0.9335777163505554, + "learning_rate": 4.978695802493834e-06, + "loss": 0.7271, + "step": 397 + }, + { + "epoch": 0.3296521258972943, + "grad_norm": 0.8823036551475525, + "learning_rate": 4.978552304191888e-06, + "loss": 0.723, + "step": 398 + }, + { + "epoch": 0.3304803975704031, + "grad_norm": 0.9503542184829712, + "learning_rate": 4.9784083263131764e-06, + "loss": 0.752, + "step": 399 + }, + { + "epoch": 0.3313086692435119, + "grad_norm": 0.8779938220977783, + "learning_rate": 4.978263868885557e-06, + "loss": 0.7098, + "step": 400 + }, + { + "epoch": 0.33213694091662066, + "grad_norm": 0.873857855796814, + "learning_rate": 4.978118931936981e-06, + "loss": 0.7143, + "step": 401 + }, + { + "epoch": 0.33296521258972944, + "grad_norm": 0.8437919020652771, + "learning_rate": 4.9779735154954924e-06, + "loss": 0.7275, + "step": 402 + }, + { + "epoch": 0.3337934842628382, + "grad_norm": 0.8914555311203003, + "learning_rate": 4.977827619589229e-06, + "loss": 0.722, + "step": 403 + }, + { + "epoch": 0.334621755935947, + "grad_norm": 0.9062597751617432, + "learning_rate": 4.977681244246417e-06, + "loss": 0.7067, + "step": 404 + }, + { + "epoch": 0.3354500276090558, + "grad_norm": 0.8555092811584473, + "learning_rate": 4.977534389495382e-06, + "loss": 0.7318, + "step": 405 + }, + { + "epoch": 0.33627829928216457, + "grad_norm": 0.8717684745788574, + "learning_rate": 4.9773870553645386e-06, + "loss": 0.7398, + "step": 406 + }, + { + "epoch": 0.33710657095527335, + "grad_norm": 0.8746877908706665, + "learning_rate": 4.977239241882393e-06, + "loss": 0.7154, + "step": 407 + }, + { + "epoch": 0.3379348426283821, + "grad_norm": 0.8905707597732544, + "learning_rate": 4.977090949077547e-06, + "loss": 0.7345, + "step": 408 + }, + { + "epoch": 0.33876311430149086, + "grad_norm": 0.9037347435951233, + "learning_rate": 4.9769421769786935e-06, + "loss": 0.7271, + "step": 409 + }, + { + "epoch": 0.33959138597459965, + "grad_norm": 0.8588722944259644, + "learning_rate": 4.976792925614619e-06, + "loss": 0.7335, + "step": 410 + }, + { + "epoch": 0.34041965764770843, + "grad_norm": 0.8786173462867737, + "learning_rate": 4.9766431950142005e-06, + "loss": 0.7254, + "step": 411 + }, + { + "epoch": 0.3412479293208172, + "grad_norm": 0.9019079804420471, + "learning_rate": 4.976492985206412e-06, + "loss": 0.715, + "step": 412 + }, + { + "epoch": 0.342076200993926, + "grad_norm": 0.878132700920105, + "learning_rate": 4.976342296220316e-06, + "loss": 0.7063, + "step": 413 + }, + { + "epoch": 0.3429044726670348, + "grad_norm": 0.8722684979438782, + "learning_rate": 4.976191128085069e-06, + "loss": 0.7481, + "step": 414 + }, + { + "epoch": 0.34373274434014356, + "grad_norm": 0.9076220393180847, + "learning_rate": 4.976039480829922e-06, + "loss": 0.7197, + "step": 415 + }, + { + "epoch": 0.34456101601325234, + "grad_norm": 0.8674967288970947, + "learning_rate": 4.975887354484217e-06, + "loss": 0.7282, + "step": 416 + }, + { + "epoch": 0.3453892876863611, + "grad_norm": 0.9406816959381104, + "learning_rate": 4.9757347490773885e-06, + "loss": 0.7222, + "step": 417 + }, + { + "epoch": 0.3462175593594699, + "grad_norm": 0.8747051358222961, + "learning_rate": 4.975581664638965e-06, + "loss": 0.7283, + "step": 418 + }, + { + "epoch": 0.3470458310325787, + "grad_norm": 1.0587687492370605, + "learning_rate": 4.9754281011985666e-06, + "loss": 0.7272, + "step": 419 + }, + { + "epoch": 0.34787410270568747, + "grad_norm": 0.8596946597099304, + "learning_rate": 4.975274058785906e-06, + "loss": 0.7407, + "step": 420 + }, + { + "epoch": 0.34870237437879625, + "grad_norm": 0.9230688214302063, + "learning_rate": 4.975119537430789e-06, + "loss": 0.7303, + "step": 421 + }, + { + "epoch": 0.34953064605190504, + "grad_norm": 0.9276477694511414, + "learning_rate": 4.974964537163115e-06, + "loss": 0.7296, + "step": 422 + }, + { + "epoch": 0.3503589177250138, + "grad_norm": 0.8969418406486511, + "learning_rate": 4.974809058012874e-06, + "loss": 0.7275, + "step": 423 + }, + { + "epoch": 0.3511871893981226, + "grad_norm": 0.8454467058181763, + "learning_rate": 4.97465310001015e-06, + "loss": 0.7144, + "step": 424 + }, + { + "epoch": 0.3520154610712314, + "grad_norm": 0.8833656907081604, + "learning_rate": 4.97449666318512e-06, + "loss": 0.7073, + "step": 425 + }, + { + "epoch": 0.35284373274434017, + "grad_norm": 0.9251653552055359, + "learning_rate": 4.974339747568052e-06, + "loss": 0.7351, + "step": 426 + }, + { + "epoch": 0.35367200441744895, + "grad_norm": 0.8938575983047485, + "learning_rate": 4.974182353189309e-06, + "loss": 0.7031, + "step": 427 + }, + { + "epoch": 0.3545002760905577, + "grad_norm": 0.8487107753753662, + "learning_rate": 4.974024480079343e-06, + "loss": 0.704, + "step": 428 + }, + { + "epoch": 0.35532854776366646, + "grad_norm": 0.8515727519989014, + "learning_rate": 4.973866128268705e-06, + "loss": 0.7092, + "step": 429 + }, + { + "epoch": 0.35615681943677524, + "grad_norm": 0.8869542479515076, + "learning_rate": 4.97370729778803e-06, + "loss": 0.7099, + "step": 430 + }, + { + "epoch": 0.356985091109884, + "grad_norm": 0.8830757141113281, + "learning_rate": 4.973547988668055e-06, + "loss": 0.7294, + "step": 431 + }, + { + "epoch": 0.3578133627829928, + "grad_norm": 0.9321587681770325, + "learning_rate": 4.973388200939601e-06, + "loss": 0.7276, + "step": 432 + }, + { + "epoch": 0.3586416344561016, + "grad_norm": 0.9408143758773804, + "learning_rate": 4.9732279346335866e-06, + "loss": 0.7216, + "step": 433 + }, + { + "epoch": 0.35946990612921037, + "grad_norm": 0.9189308285713196, + "learning_rate": 4.973067189781022e-06, + "loss": 0.7079, + "step": 434 + }, + { + "epoch": 0.36029817780231915, + "grad_norm": 0.9198120832443237, + "learning_rate": 4.97290596641301e-06, + "loss": 0.7188, + "step": 435 + }, + { + "epoch": 0.36112644947542794, + "grad_norm": 0.9123285412788391, + "learning_rate": 4.972744264560746e-06, + "loss": 0.7328, + "step": 436 + }, + { + "epoch": 0.3619547211485367, + "grad_norm": 0.8722008466720581, + "learning_rate": 4.972582084255517e-06, + "loss": 0.7228, + "step": 437 + }, + { + "epoch": 0.3627829928216455, + "grad_norm": 1.035325288772583, + "learning_rate": 4.972419425528704e-06, + "loss": 0.72, + "step": 438 + }, + { + "epoch": 0.3636112644947543, + "grad_norm": 0.9577400088310242, + "learning_rate": 4.97225628841178e-06, + "loss": 0.7382, + "step": 439 + }, + { + "epoch": 0.36443953616786307, + "grad_norm": 0.9206854104995728, + "learning_rate": 4.972092672936309e-06, + "loss": 0.7294, + "step": 440 + }, + { + "epoch": 0.36526780784097185, + "grad_norm": 0.9281194806098938, + "learning_rate": 4.971928579133952e-06, + "loss": 0.7137, + "step": 441 + }, + { + "epoch": 0.36609607951408063, + "grad_norm": 0.951049268245697, + "learning_rate": 4.971764007036457e-06, + "loss": 0.7354, + "step": 442 + }, + { + "epoch": 0.3669243511871894, + "grad_norm": 0.8779566287994385, + "learning_rate": 4.971598956675668e-06, + "loss": 0.6858, + "step": 443 + }, + { + "epoch": 0.3677526228602982, + "grad_norm": 0.9221998453140259, + "learning_rate": 4.971433428083522e-06, + "loss": 0.7116, + "step": 444 + }, + { + "epoch": 0.368580894533407, + "grad_norm": 0.9608840346336365, + "learning_rate": 4.971267421292045e-06, + "loss": 0.7129, + "step": 445 + }, + { + "epoch": 0.36940916620651576, + "grad_norm": 0.8761363625526428, + "learning_rate": 4.971100936333359e-06, + "loss": 0.718, + "step": 446 + }, + { + "epoch": 0.37023743787962454, + "grad_norm": 0.8808798789978027, + "learning_rate": 4.970933973239677e-06, + "loss": 0.6986, + "step": 447 + }, + { + "epoch": 0.37106570955273327, + "grad_norm": 0.8818857669830322, + "learning_rate": 4.9707665320433055e-06, + "loss": 0.7465, + "step": 448 + }, + { + "epoch": 0.37189398122584205, + "grad_norm": 0.9144098162651062, + "learning_rate": 4.970598612776641e-06, + "loss": 0.7187, + "step": 449 + }, + { + "epoch": 0.37272225289895083, + "grad_norm": 0.8439158797264099, + "learning_rate": 4.9704302154721776e-06, + "loss": 0.7195, + "step": 450 + }, + { + "epoch": 0.3735505245720596, + "grad_norm": 0.8661074638366699, + "learning_rate": 4.970261340162495e-06, + "loss": 0.712, + "step": 451 + }, + { + "epoch": 0.3743787962451684, + "grad_norm": 0.8818789124488831, + "learning_rate": 4.970091986880271e-06, + "loss": 0.7199, + "step": 452 + }, + { + "epoch": 0.3752070679182772, + "grad_norm": 0.9560503363609314, + "learning_rate": 4.969922155658273e-06, + "loss": 0.7145, + "step": 453 + }, + { + "epoch": 0.37603533959138596, + "grad_norm": 0.8723775744438171, + "learning_rate": 4.969751846529363e-06, + "loss": 0.7152, + "step": 454 + }, + { + "epoch": 0.37686361126449475, + "grad_norm": 0.8612253069877625, + "learning_rate": 4.9695810595264935e-06, + "loss": 0.728, + "step": 455 + }, + { + "epoch": 0.37769188293760353, + "grad_norm": 0.8995444178581238, + "learning_rate": 4.969409794682709e-06, + "loss": 0.678, + "step": 456 + }, + { + "epoch": 0.3785201546107123, + "grad_norm": 0.8770595788955688, + "learning_rate": 4.96923805203115e-06, + "loss": 0.7354, + "step": 457 + }, + { + "epoch": 0.3793484262838211, + "grad_norm": 0.893329918384552, + "learning_rate": 4.969065831605045e-06, + "loss": 0.7102, + "step": 458 + }, + { + "epoch": 0.3801766979569299, + "grad_norm": 0.954058051109314, + "learning_rate": 4.968893133437718e-06, + "loss": 0.7144, + "step": 459 + }, + { + "epoch": 0.38100496963003866, + "grad_norm": 0.9711002707481384, + "learning_rate": 4.968719957562585e-06, + "loss": 0.7452, + "step": 460 + }, + { + "epoch": 0.38183324130314744, + "grad_norm": 0.9036396741867065, + "learning_rate": 4.968546304013153e-06, + "loss": 0.7535, + "step": 461 + }, + { + "epoch": 0.3826615129762562, + "grad_norm": 0.8948055505752563, + "learning_rate": 4.968372172823023e-06, + "loss": 0.7284, + "step": 462 + }, + { + "epoch": 0.383489784649365, + "grad_norm": 0.9134189486503601, + "learning_rate": 4.968197564025886e-06, + "loss": 0.723, + "step": 463 + }, + { + "epoch": 0.3843180563224738, + "grad_norm": 0.9191608428955078, + "learning_rate": 4.9680224776555305e-06, + "loss": 0.7078, + "step": 464 + }, + { + "epoch": 0.38514632799558257, + "grad_norm": 0.879683792591095, + "learning_rate": 4.9678469137458315e-06, + "loss": 0.6805, + "step": 465 + }, + { + "epoch": 0.38597459966869135, + "grad_norm": 0.8887293338775635, + "learning_rate": 4.96767087233076e-06, + "loss": 0.7296, + "step": 466 + }, + { + "epoch": 0.38680287134180014, + "grad_norm": 0.889991044998169, + "learning_rate": 4.967494353444377e-06, + "loss": 0.702, + "step": 467 + }, + { + "epoch": 0.38763114301490886, + "grad_norm": 0.9667506814002991, + "learning_rate": 4.967317357120839e-06, + "loss": 0.7305, + "step": 468 + }, + { + "epoch": 0.38845941468801765, + "grad_norm": 0.9297274351119995, + "learning_rate": 4.967139883394392e-06, + "loss": 0.7344, + "step": 469 + }, + { + "epoch": 0.38928768636112643, + "grad_norm": 0.9167365431785583, + "learning_rate": 4.966961932299377e-06, + "loss": 0.7016, + "step": 470 + }, + { + "epoch": 0.3901159580342352, + "grad_norm": 0.877378523349762, + "learning_rate": 4.966783503870224e-06, + "loss": 0.7313, + "step": 471 + }, + { + "epoch": 0.390944229707344, + "grad_norm": 0.9625933766365051, + "learning_rate": 4.966604598141458e-06, + "loss": 0.7245, + "step": 472 + }, + { + "epoch": 0.3917725013804528, + "grad_norm": 0.8859156966209412, + "learning_rate": 4.9664252151476955e-06, + "loss": 0.7094, + "step": 473 + }, + { + "epoch": 0.39260077305356156, + "grad_norm": 0.9172984957695007, + "learning_rate": 4.966245354923645e-06, + "loss": 0.6966, + "step": 474 + }, + { + "epoch": 0.39342904472667034, + "grad_norm": 0.9040864109992981, + "learning_rate": 4.96606501750411e-06, + "loss": 0.722, + "step": 475 + }, + { + "epoch": 0.3942573163997791, + "grad_norm": 0.8900951147079468, + "learning_rate": 4.965884202923982e-06, + "loss": 0.7263, + "step": 476 + }, + { + "epoch": 0.3950855880728879, + "grad_norm": 0.9096625447273254, + "learning_rate": 4.965702911218247e-06, + "loss": 0.702, + "step": 477 + }, + { + "epoch": 0.3959138597459967, + "grad_norm": 0.8801544904708862, + "learning_rate": 4.965521142421984e-06, + "loss": 0.7113, + "step": 478 + }, + { + "epoch": 0.39674213141910547, + "grad_norm": 0.9054175615310669, + "learning_rate": 4.965338896570363e-06, + "loss": 0.7135, + "step": 479 + }, + { + "epoch": 0.39757040309221425, + "grad_norm": 0.8734445571899414, + "learning_rate": 4.965156173698648e-06, + "loss": 0.7181, + "step": 480 + }, + { + "epoch": 0.39839867476532304, + "grad_norm": 0.9113215208053589, + "learning_rate": 4.964972973842191e-06, + "loss": 0.7008, + "step": 481 + }, + { + "epoch": 0.3992269464384318, + "grad_norm": 0.8941736221313477, + "learning_rate": 4.964789297036444e-06, + "loss": 0.7201, + "step": 482 + }, + { + "epoch": 0.4000552181115406, + "grad_norm": 0.8555464148521423, + "learning_rate": 4.964605143316944e-06, + "loss": 0.7173, + "step": 483 + }, + { + "epoch": 0.4008834897846494, + "grad_norm": 0.8719691634178162, + "learning_rate": 4.9644205127193225e-06, + "loss": 0.7211, + "step": 484 + }, + { + "epoch": 0.40171176145775817, + "grad_norm": 0.901626467704773, + "learning_rate": 4.964235405279306e-06, + "loss": 0.7223, + "step": 485 + }, + { + "epoch": 0.40254003313086695, + "grad_norm": 0.856769859790802, + "learning_rate": 4.9640498210327095e-06, + "loss": 0.7103, + "step": 486 + }, + { + "epoch": 0.40336830480397573, + "grad_norm": 0.9053338766098022, + "learning_rate": 4.963863760015442e-06, + "loss": 0.7301, + "step": 487 + }, + { + "epoch": 0.40419657647708446, + "grad_norm": 0.8989829421043396, + "learning_rate": 4.963677222263505e-06, + "loss": 0.7391, + "step": 488 + }, + { + "epoch": 0.40502484815019324, + "grad_norm": 0.9187886118888855, + "learning_rate": 4.963490207812991e-06, + "loss": 0.7179, + "step": 489 + }, + { + "epoch": 0.405853119823302, + "grad_norm": 0.8712778687477112, + "learning_rate": 4.963302716700087e-06, + "loss": 0.7088, + "step": 490 + }, + { + "epoch": 0.4066813914964108, + "grad_norm": 1.0485447645187378, + "learning_rate": 4.9631147489610695e-06, + "loss": 0.7068, + "step": 491 + }, + { + "epoch": 0.4075096631695196, + "grad_norm": 0.9332602024078369, + "learning_rate": 4.962926304632309e-06, + "loss": 0.7201, + "step": 492 + }, + { + "epoch": 0.40833793484262837, + "grad_norm": 0.9264400601387024, + "learning_rate": 4.962737383750268e-06, + "loss": 0.7136, + "step": 493 + }, + { + "epoch": 0.40916620651573715, + "grad_norm": 0.8812651634216309, + "learning_rate": 4.962547986351501e-06, + "loss": 0.7315, + "step": 494 + }, + { + "epoch": 0.40999447818884593, + "grad_norm": 0.8905262351036072, + "learning_rate": 4.962358112472653e-06, + "loss": 0.7127, + "step": 495 + }, + { + "epoch": 0.4108227498619547, + "grad_norm": 0.887514591217041, + "learning_rate": 4.962167762150465e-06, + "loss": 0.7602, + "step": 496 + }, + { + "epoch": 0.4116510215350635, + "grad_norm": 0.8683431148529053, + "learning_rate": 4.9619769354217665e-06, + "loss": 0.7091, + "step": 497 + }, + { + "epoch": 0.4124792932081723, + "grad_norm": 0.944595217704773, + "learning_rate": 4.961785632323481e-06, + "loss": 0.708, + "step": 498 + }, + { + "epoch": 0.41330756488128106, + "grad_norm": 0.8834611773490906, + "learning_rate": 4.961593852892625e-06, + "loss": 0.7308, + "step": 499 + }, + { + "epoch": 0.41413583655438985, + "grad_norm": 0.8479524850845337, + "learning_rate": 4.961401597166304e-06, + "loss": 0.7061, + "step": 500 + }, + { + "epoch": 0.41496410822749863, + "grad_norm": 0.9092981219291687, + "learning_rate": 4.96120886518172e-06, + "loss": 0.7176, + "step": 501 + }, + { + "epoch": 0.4157923799006074, + "grad_norm": 0.9024932384490967, + "learning_rate": 4.961015656976163e-06, + "loss": 0.6944, + "step": 502 + }, + { + "epoch": 0.4166206515737162, + "grad_norm": 0.8945576548576355, + "learning_rate": 4.960821972587018e-06, + "loss": 0.6958, + "step": 503 + }, + { + "epoch": 0.417448923246825, + "grad_norm": 0.8668336868286133, + "learning_rate": 4.9606278120517606e-06, + "loss": 0.7219, + "step": 504 + }, + { + "epoch": 0.41827719491993376, + "grad_norm": 0.8831227421760559, + "learning_rate": 4.96043317540796e-06, + "loss": 0.7124, + "step": 505 + }, + { + "epoch": 0.41910546659304254, + "grad_norm": 0.9637073874473572, + "learning_rate": 4.960238062693274e-06, + "loss": 0.7337, + "step": 506 + }, + { + "epoch": 0.4199337382661513, + "grad_norm": 0.8895899653434753, + "learning_rate": 4.960042473945459e-06, + "loss": 0.7459, + "step": 507 + }, + { + "epoch": 0.42076200993926005, + "grad_norm": 0.8762213587760925, + "learning_rate": 4.959846409202356e-06, + "loss": 0.7319, + "step": 508 + }, + { + "epoch": 0.42159028161236883, + "grad_norm": 0.9177393317222595, + "learning_rate": 4.959649868501903e-06, + "loss": 0.7114, + "step": 509 + }, + { + "epoch": 0.4224185532854776, + "grad_norm": 0.88541179895401, + "learning_rate": 4.9594528518821305e-06, + "loss": 0.716, + "step": 510 + }, + { + "epoch": 0.4232468249585864, + "grad_norm": 0.865566611289978, + "learning_rate": 4.959255359381158e-06, + "loss": 0.7098, + "step": 511 + }, + { + "epoch": 0.4240750966316952, + "grad_norm": 0.8629707098007202, + "learning_rate": 4.959057391037197e-06, + "loss": 0.7146, + "step": 512 + }, + { + "epoch": 0.42490336830480396, + "grad_norm": 0.8642273545265198, + "learning_rate": 4.958858946888554e-06, + "loss": 0.7345, + "step": 513 + }, + { + "epoch": 0.42573163997791275, + "grad_norm": 0.9157605171203613, + "learning_rate": 4.958660026973626e-06, + "loss": 0.7177, + "step": 514 + }, + { + "epoch": 0.42655991165102153, + "grad_norm": 0.9108191728591919, + "learning_rate": 4.958460631330903e-06, + "loss": 0.7317, + "step": 515 + }, + { + "epoch": 0.4273881833241303, + "grad_norm": 0.8524913191795349, + "learning_rate": 4.958260759998965e-06, + "loss": 0.7193, + "step": 516 + }, + { + "epoch": 0.4282164549972391, + "grad_norm": 0.9197820425033569, + "learning_rate": 4.958060413016485e-06, + "loss": 0.7127, + "step": 517 + }, + { + "epoch": 0.4290447266703479, + "grad_norm": 0.8744186758995056, + "learning_rate": 4.9578595904222296e-06, + "loss": 0.6968, + "step": 518 + }, + { + "epoch": 0.42987299834345666, + "grad_norm": 0.9071037769317627, + "learning_rate": 4.957658292255055e-06, + "loss": 0.6963, + "step": 519 + }, + { + "epoch": 0.43070127001656544, + "grad_norm": 0.937796950340271, + "learning_rate": 4.95745651855391e-06, + "loss": 0.7193, + "step": 520 + }, + { + "epoch": 0.4315295416896742, + "grad_norm": 0.883471667766571, + "learning_rate": 4.9572542693578375e-06, + "loss": 0.6886, + "step": 521 + }, + { + "epoch": 0.432357813362783, + "grad_norm": 0.8738589882850647, + "learning_rate": 4.9570515447059705e-06, + "loss": 0.7072, + "step": 522 + }, + { + "epoch": 0.4331860850358918, + "grad_norm": 0.9039497971534729, + "learning_rate": 4.9568483446375335e-06, + "loss": 0.7251, + "step": 523 + }, + { + "epoch": 0.43401435670900057, + "grad_norm": 0.896789014339447, + "learning_rate": 4.956644669191845e-06, + "loss": 0.7021, + "step": 524 + }, + { + "epoch": 0.43484262838210935, + "grad_norm": 0.8883570432662964, + "learning_rate": 4.9564405184083126e-06, + "loss": 0.7153, + "step": 525 + }, + { + "epoch": 0.43567090005521814, + "grad_norm": 0.8869197368621826, + "learning_rate": 4.956235892326438e-06, + "loss": 0.7061, + "step": 526 + }, + { + "epoch": 0.4364991717283269, + "grad_norm": 0.9237920641899109, + "learning_rate": 4.9560307909858165e-06, + "loss": 0.6988, + "step": 527 + }, + { + "epoch": 0.43732744340143564, + "grad_norm": 0.9359283447265625, + "learning_rate": 4.955825214426131e-06, + "loss": 0.738, + "step": 528 + }, + { + "epoch": 0.4381557150745444, + "grad_norm": 0.8602650761604309, + "learning_rate": 4.95561916268716e-06, + "loss": 0.709, + "step": 529 + }, + { + "epoch": 0.4389839867476532, + "grad_norm": 0.9013513922691345, + "learning_rate": 4.955412635808771e-06, + "loss": 0.6836, + "step": 530 + }, + { + "epoch": 0.439812258420762, + "grad_norm": 0.9136670827865601, + "learning_rate": 4.955205633830927e-06, + "loss": 0.7263, + "step": 531 + }, + { + "epoch": 0.4406405300938708, + "grad_norm": 0.9013793468475342, + "learning_rate": 4.95499815679368e-06, + "loss": 0.6937, + "step": 532 + }, + { + "epoch": 0.44146880176697956, + "grad_norm": 0.9017112851142883, + "learning_rate": 4.954790204737174e-06, + "loss": 0.6913, + "step": 533 + }, + { + "epoch": 0.44229707344008834, + "grad_norm": 0.8901436924934387, + "learning_rate": 4.954581777701647e-06, + "loss": 0.7241, + "step": 534 + }, + { + "epoch": 0.4431253451131971, + "grad_norm": 0.8882759809494019, + "learning_rate": 4.954372875727427e-06, + "loss": 0.7115, + "step": 535 + }, + { + "epoch": 0.4439536167863059, + "grad_norm": 0.8909885883331299, + "learning_rate": 4.954163498854936e-06, + "loss": 0.7146, + "step": 536 + }, + { + "epoch": 0.4447818884594147, + "grad_norm": 0.8951786756515503, + "learning_rate": 4.953953647124684e-06, + "loss": 0.699, + "step": 537 + }, + { + "epoch": 0.44561016013252347, + "grad_norm": 0.8890933990478516, + "learning_rate": 4.953743320577278e-06, + "loss": 0.7018, + "step": 538 + }, + { + "epoch": 0.44643843180563225, + "grad_norm": 0.910393238067627, + "learning_rate": 4.953532519253412e-06, + "loss": 0.7299, + "step": 539 + }, + { + "epoch": 0.44726670347874103, + "grad_norm": 0.9152857065200806, + "learning_rate": 4.953321243193876e-06, + "loss": 0.6984, + "step": 540 + }, + { + "epoch": 0.4480949751518498, + "grad_norm": 0.9001837372779846, + "learning_rate": 4.9531094924395485e-06, + "loss": 0.727, + "step": 541 + }, + { + "epoch": 0.4489232468249586, + "grad_norm": 0.882060170173645, + "learning_rate": 4.952897267031401e-06, + "loss": 0.7016, + "step": 542 + }, + { + "epoch": 0.4497515184980674, + "grad_norm": 0.8665159344673157, + "learning_rate": 4.9526845670104995e-06, + "loss": 0.7139, + "step": 543 + }, + { + "epoch": 0.45057979017117616, + "grad_norm": 0.9200844764709473, + "learning_rate": 4.952471392417997e-06, + "loss": 0.7016, + "step": 544 + }, + { + "epoch": 0.45140806184428495, + "grad_norm": 0.8852037191390991, + "learning_rate": 4.9522577432951416e-06, + "loss": 0.7048, + "step": 545 + }, + { + "epoch": 0.45223633351739373, + "grad_norm": 0.9003865122795105, + "learning_rate": 4.952043619683274e-06, + "loss": 0.7106, + "step": 546 + }, + { + "epoch": 0.4530646051905025, + "grad_norm": 0.8888446092605591, + "learning_rate": 4.951829021623822e-06, + "loss": 0.7202, + "step": 547 + }, + { + "epoch": 0.45389287686361124, + "grad_norm": 0.8940277099609375, + "learning_rate": 4.9516139491583094e-06, + "loss": 0.7156, + "step": 548 + }, + { + "epoch": 0.45472114853672, + "grad_norm": 0.9066211581230164, + "learning_rate": 4.951398402328353e-06, + "loss": 0.6865, + "step": 549 + }, + { + "epoch": 0.4555494202098288, + "grad_norm": 0.9352865815162659, + "learning_rate": 4.9511823811756565e-06, + "loss": 0.7117, + "step": 550 + }, + { + "epoch": 0.4563776918829376, + "grad_norm": 0.8672269582748413, + "learning_rate": 4.95096588574202e-06, + "loss": 0.7166, + "step": 551 + }, + { + "epoch": 0.45720596355604637, + "grad_norm": 0.8730685114860535, + "learning_rate": 4.95074891606933e-06, + "loss": 0.6998, + "step": 552 + }, + { + "epoch": 0.45803423522915515, + "grad_norm": 0.896821916103363, + "learning_rate": 4.950531472199572e-06, + "loss": 0.6839, + "step": 553 + }, + { + "epoch": 0.45886250690226393, + "grad_norm": 0.918572187423706, + "learning_rate": 4.950313554174817e-06, + "loss": 0.7155, + "step": 554 + }, + { + "epoch": 0.4596907785753727, + "grad_norm": 0.8670752048492432, + "learning_rate": 4.950095162037232e-06, + "loss": 0.708, + "step": 555 + }, + { + "epoch": 0.4605190502484815, + "grad_norm": 0.875618040561676, + "learning_rate": 4.9498762958290715e-06, + "loss": 0.6956, + "step": 556 + }, + { + "epoch": 0.4613473219215903, + "grad_norm": 0.9120676517486572, + "learning_rate": 4.949656955592685e-06, + "loss": 0.7327, + "step": 557 + }, + { + "epoch": 0.46217559359469906, + "grad_norm": 0.9670370817184448, + "learning_rate": 4.949437141370514e-06, + "loss": 0.705, + "step": 558 + }, + { + "epoch": 0.46300386526780785, + "grad_norm": 0.9155322313308716, + "learning_rate": 4.949216853205089e-06, + "loss": 0.7126, + "step": 559 + }, + { + "epoch": 0.46383213694091663, + "grad_norm": 0.8945392370223999, + "learning_rate": 4.948996091139034e-06, + "loss": 0.7083, + "step": 560 + }, + { + "epoch": 0.4646604086140254, + "grad_norm": 0.8922162055969238, + "learning_rate": 4.948774855215065e-06, + "loss": 0.7189, + "step": 561 + }, + { + "epoch": 0.4654886802871342, + "grad_norm": 0.8940657377243042, + "learning_rate": 4.948553145475989e-06, + "loss": 0.6934, + "step": 562 + }, + { + "epoch": 0.466316951960243, + "grad_norm": 0.8946855068206787, + "learning_rate": 4.948330961964703e-06, + "loss": 0.6835, + "step": 563 + }, + { + "epoch": 0.46714522363335176, + "grad_norm": 0.9002820253372192, + "learning_rate": 4.948108304724201e-06, + "loss": 0.712, + "step": 564 + }, + { + "epoch": 0.46797349530646054, + "grad_norm": 0.8715711832046509, + "learning_rate": 4.947885173797561e-06, + "loss": 0.7197, + "step": 565 + }, + { + "epoch": 0.4688017669795693, + "grad_norm": 0.9269874095916748, + "learning_rate": 4.94766156922796e-06, + "loss": 0.7311, + "step": 566 + }, + { + "epoch": 0.4696300386526781, + "grad_norm": 0.8999423980712891, + "learning_rate": 4.947437491058662e-06, + "loss": 0.7144, + "step": 567 + }, + { + "epoch": 0.47045831032578683, + "grad_norm": 0.8704255223274231, + "learning_rate": 4.947212939333024e-06, + "loss": 0.7156, + "step": 568 + }, + { + "epoch": 0.4712865819988956, + "grad_norm": 0.8770347237586975, + "learning_rate": 4.946987914094495e-06, + "loss": 0.6969, + "step": 569 + }, + { + "epoch": 0.4721148536720044, + "grad_norm": 0.8865682482719421, + "learning_rate": 4.946762415386615e-06, + "loss": 0.7089, + "step": 570 + }, + { + "epoch": 0.4729431253451132, + "grad_norm": 0.8756279349327087, + "learning_rate": 4.946536443253016e-06, + "loss": 0.7159, + "step": 571 + }, + { + "epoch": 0.47377139701822196, + "grad_norm": 0.8844839930534363, + "learning_rate": 4.946309997737422e-06, + "loss": 0.7234, + "step": 572 + }, + { + "epoch": 0.47459966869133074, + "grad_norm": 0.8549885153770447, + "learning_rate": 4.946083078883647e-06, + "loss": 0.6927, + "step": 573 + }, + { + "epoch": 0.4754279403644395, + "grad_norm": 0.8967456817626953, + "learning_rate": 4.945855686735599e-06, + "loss": 0.7082, + "step": 574 + }, + { + "epoch": 0.4762562120375483, + "grad_norm": 0.9120200276374817, + "learning_rate": 4.9456278213372755e-06, + "loss": 0.7135, + "step": 575 + }, + { + "epoch": 0.4770844837106571, + "grad_norm": 0.8780766725540161, + "learning_rate": 4.945399482732765e-06, + "loss": 0.7139, + "step": 576 + }, + { + "epoch": 0.4779127553837659, + "grad_norm": 0.8900999426841736, + "learning_rate": 4.945170670966253e-06, + "loss": 0.7132, + "step": 577 + }, + { + "epoch": 0.47874102705687466, + "grad_norm": 0.9402151107788086, + "learning_rate": 4.944941386082008e-06, + "loss": 0.7139, + "step": 578 + }, + { + "epoch": 0.47956929872998344, + "grad_norm": 0.8769465088844299, + "learning_rate": 4.944711628124396e-06, + "loss": 0.7236, + "step": 579 + }, + { + "epoch": 0.4803975704030922, + "grad_norm": 0.9185771346092224, + "learning_rate": 4.944481397137874e-06, + "loss": 0.6884, + "step": 580 + }, + { + "epoch": 0.481225842076201, + "grad_norm": 0.8775897026062012, + "learning_rate": 4.944250693166988e-06, + "loss": 0.731, + "step": 581 + }, + { + "epoch": 0.4820541137493098, + "grad_norm": 0.8751668930053711, + "learning_rate": 4.94401951625638e-06, + "loss": 0.7007, + "step": 582 + }, + { + "epoch": 0.48288238542241857, + "grad_norm": 0.9042515754699707, + "learning_rate": 4.943787866450776e-06, + "loss": 0.7064, + "step": 583 + }, + { + "epoch": 0.48371065709552735, + "grad_norm": 0.8845478892326355, + "learning_rate": 4.943555743795002e-06, + "loss": 0.7069, + "step": 584 + }, + { + "epoch": 0.48453892876863613, + "grad_norm": 0.8959096670150757, + "learning_rate": 4.943323148333969e-06, + "loss": 0.6871, + "step": 585 + }, + { + "epoch": 0.4853672004417449, + "grad_norm": 0.9021669626235962, + "learning_rate": 4.943090080112684e-06, + "loss": 0.7156, + "step": 586 + }, + { + "epoch": 0.4861954721148537, + "grad_norm": 0.8768396973609924, + "learning_rate": 4.942856539176243e-06, + "loss": 0.681, + "step": 587 + }, + { + "epoch": 0.4870237437879624, + "grad_norm": 0.8734706044197083, + "learning_rate": 4.942622525569832e-06, + "loss": 0.7371, + "step": 588 + }, + { + "epoch": 0.4878520154610712, + "grad_norm": 0.9155406355857849, + "learning_rate": 4.942388039338733e-06, + "loss": 0.7012, + "step": 589 + }, + { + "epoch": 0.48868028713418, + "grad_norm": 0.8778123259544373, + "learning_rate": 4.942153080528317e-06, + "loss": 0.7201, + "step": 590 + }, + { + "epoch": 0.4895085588072888, + "grad_norm": 0.9102092981338501, + "learning_rate": 4.941917649184045e-06, + "loss": 0.6951, + "step": 591 + }, + { + "epoch": 0.49033683048039756, + "grad_norm": 0.8855768442153931, + "learning_rate": 4.941681745351471e-06, + "loss": 0.6997, + "step": 592 + }, + { + "epoch": 0.49116510215350634, + "grad_norm": 0.8770707845687866, + "learning_rate": 4.94144536907624e-06, + "loss": 0.7092, + "step": 593 + }, + { + "epoch": 0.4919933738266151, + "grad_norm": 0.900458574295044, + "learning_rate": 4.9412085204040896e-06, + "loss": 0.6909, + "step": 594 + }, + { + "epoch": 0.4928216454997239, + "grad_norm": 0.903489887714386, + "learning_rate": 4.940971199380846e-06, + "loss": 0.7066, + "step": 595 + }, + { + "epoch": 0.4936499171728327, + "grad_norm": 0.9331871271133423, + "learning_rate": 4.940733406052431e-06, + "loss": 0.6924, + "step": 596 + }, + { + "epoch": 0.49447818884594147, + "grad_norm": 0.8855593800544739, + "learning_rate": 4.9404951404648535e-06, + "loss": 0.6991, + "step": 597 + }, + { + "epoch": 0.49530646051905025, + "grad_norm": 0.9075731039047241, + "learning_rate": 4.940256402664216e-06, + "loss": 0.6973, + "step": 598 + }, + { + "epoch": 0.49613473219215903, + "grad_norm": 0.8970743417739868, + "learning_rate": 4.940017192696713e-06, + "loss": 0.6767, + "step": 599 + }, + { + "epoch": 0.4969630038652678, + "grad_norm": 0.8644358515739441, + "learning_rate": 4.939777510608628e-06, + "loss": 0.7284, + "step": 600 + }, + { + "epoch": 0.4977912755383766, + "grad_norm": 0.8807653784751892, + "learning_rate": 4.939537356446339e-06, + "loss": 0.7143, + "step": 601 + }, + { + "epoch": 0.4986195472114854, + "grad_norm": 0.892946183681488, + "learning_rate": 4.939296730256312e-06, + "loss": 0.7021, + "step": 602 + }, + { + "epoch": 0.49944781888459416, + "grad_norm": 0.9661110639572144, + "learning_rate": 4.939055632085107e-06, + "loss": 0.7198, + "step": 603 + }, + { + "epoch": 0.500276090557703, + "grad_norm": 0.8638496398925781, + "learning_rate": 4.9388140619793735e-06, + "loss": 0.6943, + "step": 604 + }, + { + "epoch": 0.5011043622308117, + "grad_norm": 0.9025457501411438, + "learning_rate": 4.938572019985853e-06, + "loss": 0.7114, + "step": 605 + }, + { + "epoch": 0.5019326339039205, + "grad_norm": 0.9622734785079956, + "learning_rate": 4.93832950615138e-06, + "loss": 0.7086, + "step": 606 + }, + { + "epoch": 0.5027609055770292, + "grad_norm": 0.9565468430519104, + "learning_rate": 4.938086520522876e-06, + "loss": 0.7125, + "step": 607 + }, + { + "epoch": 0.5035891772501381, + "grad_norm": 0.8970094323158264, + "learning_rate": 4.937843063147359e-06, + "loss": 0.7023, + "step": 608 + }, + { + "epoch": 0.5044174489232468, + "grad_norm": 0.8920996785163879, + "learning_rate": 4.937599134071933e-06, + "loss": 0.7041, + "step": 609 + }, + { + "epoch": 0.5052457205963556, + "grad_norm": 0.9304755926132202, + "learning_rate": 4.937354733343799e-06, + "loss": 0.7165, + "step": 610 + }, + { + "epoch": 0.5060739922694644, + "grad_norm": 0.8995933532714844, + "learning_rate": 4.937109861010244e-06, + "loss": 0.6907, + "step": 611 + }, + { + "epoch": 0.5069022639425732, + "grad_norm": 0.8648850321769714, + "learning_rate": 4.936864517118649e-06, + "loss": 0.6994, + "step": 612 + }, + { + "epoch": 0.5077305356156819, + "grad_norm": 0.8414506316184998, + "learning_rate": 4.936618701716487e-06, + "loss": 0.7081, + "step": 613 + }, + { + "epoch": 0.5085588072887908, + "grad_norm": 0.9282650947570801, + "learning_rate": 4.936372414851321e-06, + "loss": 0.7303, + "step": 614 + }, + { + "epoch": 0.5093870789618995, + "grad_norm": 0.877757728099823, + "learning_rate": 4.936125656570802e-06, + "loss": 0.7132, + "step": 615 + }, + { + "epoch": 0.5102153506350083, + "grad_norm": 0.8645135164260864, + "learning_rate": 4.935878426922679e-06, + "loss": 0.723, + "step": 616 + }, + { + "epoch": 0.5110436223081171, + "grad_norm": 0.9014747738838196, + "learning_rate": 4.935630725954787e-06, + "loss": 0.7167, + "step": 617 + }, + { + "epoch": 0.5118718939812258, + "grad_norm": 0.9295386672019958, + "learning_rate": 4.935382553715055e-06, + "loss": 0.705, + "step": 618 + }, + { + "epoch": 0.5127001656543346, + "grad_norm": 0.8910616040229797, + "learning_rate": 4.935133910251499e-06, + "loss": 0.6781, + "step": 619 + }, + { + "epoch": 0.5135284373274434, + "grad_norm": 0.8804931640625, + "learning_rate": 4.934884795612234e-06, + "loss": 0.6984, + "step": 620 + }, + { + "epoch": 0.5143567090005522, + "grad_norm": 0.8223545551300049, + "learning_rate": 4.934635209845458e-06, + "loss": 0.6892, + "step": 621 + }, + { + "epoch": 0.5151849806736609, + "grad_norm": 0.8509450554847717, + "learning_rate": 4.934385152999463e-06, + "loss": 0.7139, + "step": 622 + }, + { + "epoch": 0.5160132523467698, + "grad_norm": 0.8944748044013977, + "learning_rate": 4.934134625122636e-06, + "loss": 0.6976, + "step": 623 + }, + { + "epoch": 0.5168415240198785, + "grad_norm": 0.9021086692810059, + "learning_rate": 4.933883626263448e-06, + "loss": 0.7189, + "step": 624 + }, + { + "epoch": 0.5176697956929873, + "grad_norm": 0.8939547538757324, + "learning_rate": 4.933632156470467e-06, + "loss": 0.704, + "step": 625 + }, + { + "epoch": 0.518498067366096, + "grad_norm": 0.8793498277664185, + "learning_rate": 4.93338021579235e-06, + "loss": 0.7046, + "step": 626 + }, + { + "epoch": 0.5193263390392049, + "grad_norm": 0.8728756308555603, + "learning_rate": 4.933127804277845e-06, + "loss": 0.6826, + "step": 627 + }, + { + "epoch": 0.5201546107123136, + "grad_norm": 0.8989412188529968, + "learning_rate": 4.932874921975792e-06, + "loss": 0.6842, + "step": 628 + }, + { + "epoch": 0.5209828823854225, + "grad_norm": 0.8836191296577454, + "learning_rate": 4.93262156893512e-06, + "loss": 0.7061, + "step": 629 + }, + { + "epoch": 0.5218111540585312, + "grad_norm": 0.9429040551185608, + "learning_rate": 4.9323677452048505e-06, + "loss": 0.7053, + "step": 630 + }, + { + "epoch": 0.52263942573164, + "grad_norm": 0.879463255405426, + "learning_rate": 4.932113450834097e-06, + "loss": 0.6926, + "step": 631 + }, + { + "epoch": 0.5234676974047487, + "grad_norm": 0.8566175699234009, + "learning_rate": 4.931858685872062e-06, + "loss": 0.7126, + "step": 632 + }, + { + "epoch": 0.5242959690778576, + "grad_norm": 0.8694538474082947, + "learning_rate": 4.931603450368042e-06, + "loss": 0.7153, + "step": 633 + }, + { + "epoch": 0.5251242407509663, + "grad_norm": 0.8754484057426453, + "learning_rate": 4.931347744371421e-06, + "loss": 0.711, + "step": 634 + }, + { + "epoch": 0.5259525124240751, + "grad_norm": 0.8982204794883728, + "learning_rate": 4.931091567931675e-06, + "loss": 0.7168, + "step": 635 + }, + { + "epoch": 0.5267807840971839, + "grad_norm": 0.9262725710868835, + "learning_rate": 4.930834921098374e-06, + "loss": 0.7315, + "step": 636 + }, + { + "epoch": 0.5276090557702926, + "grad_norm": 0.9271771311759949, + "learning_rate": 4.930577803921176e-06, + "loss": 0.6968, + "step": 637 + }, + { + "epoch": 0.5284373274434014, + "grad_norm": 0.8855197429656982, + "learning_rate": 4.93032021644983e-06, + "loss": 0.6986, + "step": 638 + }, + { + "epoch": 0.5292655991165102, + "grad_norm": 0.8615745902061462, + "learning_rate": 4.930062158734178e-06, + "loss": 0.7014, + "step": 639 + }, + { + "epoch": 0.530093870789619, + "grad_norm": 0.9004185795783997, + "learning_rate": 4.92980363082415e-06, + "loss": 0.7045, + "step": 640 + }, + { + "epoch": 0.5309221424627277, + "grad_norm": 0.8533529043197632, + "learning_rate": 4.92954463276977e-06, + "loss": 0.6911, + "step": 641 + }, + { + "epoch": 0.5317504141358366, + "grad_norm": 0.881901204586029, + "learning_rate": 4.929285164621152e-06, + "loss": 0.7148, + "step": 642 + }, + { + "epoch": 0.5325786858089453, + "grad_norm": 0.8778982162475586, + "learning_rate": 4.9290252264285e-06, + "loss": 0.6879, + "step": 643 + }, + { + "epoch": 0.5334069574820541, + "grad_norm": 0.9455735087394714, + "learning_rate": 4.928764818242109e-06, + "loss": 0.7088, + "step": 644 + }, + { + "epoch": 0.5342352291551629, + "grad_norm": 0.9255897998809814, + "learning_rate": 4.928503940112367e-06, + "loss": 0.6817, + "step": 645 + }, + { + "epoch": 0.5350635008282717, + "grad_norm": 0.8854040503501892, + "learning_rate": 4.928242592089752e-06, + "loss": 0.7074, + "step": 646 + }, + { + "epoch": 0.5358917725013804, + "grad_norm": 0.8824400901794434, + "learning_rate": 4.927980774224829e-06, + "loss": 0.7092, + "step": 647 + }, + { + "epoch": 0.5367200441744893, + "grad_norm": 0.8873158693313599, + "learning_rate": 4.927718486568261e-06, + "loss": 0.7214, + "step": 648 + }, + { + "epoch": 0.537548315847598, + "grad_norm": 0.9013757109642029, + "learning_rate": 4.927455729170796e-06, + "loss": 0.6955, + "step": 649 + }, + { + "epoch": 0.5383765875207068, + "grad_norm": 0.9100024104118347, + "learning_rate": 4.927192502083276e-06, + "loss": 0.7089, + "step": 650 + }, + { + "epoch": 0.5392048591938156, + "grad_norm": 0.8935851454734802, + "learning_rate": 4.9269288053566335e-06, + "loss": 0.7018, + "step": 651 + }, + { + "epoch": 0.5400331308669244, + "grad_norm": 0.8726267218589783, + "learning_rate": 4.92666463904189e-06, + "loss": 0.6829, + "step": 652 + }, + { + "epoch": 0.5408614025400331, + "grad_norm": 0.9320974349975586, + "learning_rate": 4.926400003190161e-06, + "loss": 0.6766, + "step": 653 + }, + { + "epoch": 0.541689674213142, + "grad_norm": 0.9257067441940308, + "learning_rate": 4.926134897852648e-06, + "loss": 0.6913, + "step": 654 + }, + { + "epoch": 0.5425179458862507, + "grad_norm": 0.8642577528953552, + "learning_rate": 4.92586932308065e-06, + "loss": 0.7155, + "step": 655 + }, + { + "epoch": 0.5433462175593595, + "grad_norm": 0.8861536383628845, + "learning_rate": 4.9256032789255505e-06, + "loss": 0.6695, + "step": 656 + }, + { + "epoch": 0.5441744892324683, + "grad_norm": 0.8901777267456055, + "learning_rate": 4.925336765438829e-06, + "loss": 0.6847, + "step": 657 + }, + { + "epoch": 0.545002760905577, + "grad_norm": 0.8810168504714966, + "learning_rate": 4.9250697826720515e-06, + "loss": 0.7038, + "step": 658 + }, + { + "epoch": 0.5458310325786858, + "grad_norm": 0.8655433654785156, + "learning_rate": 4.924802330676877e-06, + "loss": 0.6908, + "step": 659 + }, + { + "epoch": 0.5466593042517945, + "grad_norm": 0.8786777853965759, + "learning_rate": 4.924534409505055e-06, + "loss": 0.663, + "step": 660 + }, + { + "epoch": 0.5474875759249034, + "grad_norm": 0.8837904334068298, + "learning_rate": 4.924266019208427e-06, + "loss": 0.6843, + "step": 661 + }, + { + "epoch": 0.5483158475980121, + "grad_norm": 0.8715303540229797, + "learning_rate": 4.923997159838923e-06, + "loss": 0.6841, + "step": 662 + }, + { + "epoch": 0.549144119271121, + "grad_norm": 0.8825753927230835, + "learning_rate": 4.923727831448564e-06, + "loss": 0.7053, + "step": 663 + }, + { + "epoch": 0.5499723909442297, + "grad_norm": 0.877668559551239, + "learning_rate": 4.923458034089463e-06, + "loss": 0.6624, + "step": 664 + }, + { + "epoch": 0.5508006626173385, + "grad_norm": 0.8721411824226379, + "learning_rate": 4.923187767813825e-06, + "loss": 0.7339, + "step": 665 + }, + { + "epoch": 0.5516289342904472, + "grad_norm": 0.8607308268547058, + "learning_rate": 4.922917032673941e-06, + "loss": 0.7184, + "step": 666 + }, + { + "epoch": 0.5524572059635561, + "grad_norm": 0.8456798791885376, + "learning_rate": 4.922645828722199e-06, + "loss": 0.7056, + "step": 667 + }, + { + "epoch": 0.5532854776366648, + "grad_norm": 0.8557775020599365, + "learning_rate": 4.922374156011071e-06, + "loss": 0.6942, + "step": 668 + }, + { + "epoch": 0.5541137493097736, + "grad_norm": 0.8877674341201782, + "learning_rate": 4.922102014593126e-06, + "loss": 0.7173, + "step": 669 + }, + { + "epoch": 0.5549420209828824, + "grad_norm": 0.8662481904029846, + "learning_rate": 4.921829404521019e-06, + "loss": 0.7113, + "step": 670 + }, + { + "epoch": 0.5557702926559912, + "grad_norm": 0.8887649774551392, + "learning_rate": 4.921556325847499e-06, + "loss": 0.7028, + "step": 671 + }, + { + "epoch": 0.5565985643290999, + "grad_norm": 0.9837973713874817, + "learning_rate": 4.921282778625403e-06, + "loss": 0.6975, + "step": 672 + }, + { + "epoch": 0.5574268360022088, + "grad_norm": 0.8754701018333435, + "learning_rate": 4.921008762907661e-06, + "loss": 0.6983, + "step": 673 + }, + { + "epoch": 0.5582551076753175, + "grad_norm": 0.9125426411628723, + "learning_rate": 4.920734278747291e-06, + "loss": 0.6982, + "step": 674 + }, + { + "epoch": 0.5590833793484263, + "grad_norm": 0.9424439072608948, + "learning_rate": 4.920459326197403e-06, + "loss": 0.7304, + "step": 675 + }, + { + "epoch": 0.5599116510215351, + "grad_norm": 0.8718393445014954, + "learning_rate": 4.9201839053112e-06, + "loss": 0.6644, + "step": 676 + }, + { + "epoch": 0.5607399226946438, + "grad_norm": 0.9002220630645752, + "learning_rate": 4.919908016141972e-06, + "loss": 0.6966, + "step": 677 + }, + { + "epoch": 0.5615681943677526, + "grad_norm": 0.8925678133964539, + "learning_rate": 4.9196316587431e-06, + "loss": 0.713, + "step": 678 + }, + { + "epoch": 0.5623964660408614, + "grad_norm": 0.8854464292526245, + "learning_rate": 4.919354833168059e-06, + "loss": 0.7075, + "step": 679 + }, + { + "epoch": 0.5632247377139702, + "grad_norm": 0.8867958188056946, + "learning_rate": 4.919077539470409e-06, + "loss": 0.7166, + "step": 680 + }, + { + "epoch": 0.5640530093870789, + "grad_norm": 0.862531840801239, + "learning_rate": 4.918799777703806e-06, + "loss": 0.7119, + "step": 681 + }, + { + "epoch": 0.5648812810601878, + "grad_norm": 0.8623220920562744, + "learning_rate": 4.9185215479219944e-06, + "loss": 0.6758, + "step": 682 + }, + { + "epoch": 0.5657095527332965, + "grad_norm": 0.9171786904335022, + "learning_rate": 4.918242850178808e-06, + "loss": 0.7049, + "step": 683 + }, + { + "epoch": 0.5665378244064053, + "grad_norm": 0.8947017192840576, + "learning_rate": 4.9179636845281735e-06, + "loss": 0.7144, + "step": 684 + }, + { + "epoch": 0.567366096079514, + "grad_norm": 0.9343388080596924, + "learning_rate": 4.917684051024104e-06, + "loss": 0.7024, + "step": 685 + }, + { + "epoch": 0.5681943677526229, + "grad_norm": 0.8929025530815125, + "learning_rate": 4.917403949720709e-06, + "loss": 0.7162, + "step": 686 + }, + { + "epoch": 0.5690226394257316, + "grad_norm": 0.9000421762466431, + "learning_rate": 4.9171233806721854e-06, + "loss": 0.7085, + "step": 687 + }, + { + "epoch": 0.5698509110988405, + "grad_norm": 0.9404364228248596, + "learning_rate": 4.91684234393282e-06, + "loss": 0.6996, + "step": 688 + }, + { + "epoch": 0.5706791827719492, + "grad_norm": 0.9566411375999451, + "learning_rate": 4.916560839556989e-06, + "loss": 0.7241, + "step": 689 + }, + { + "epoch": 0.571507454445058, + "grad_norm": 0.9066181182861328, + "learning_rate": 4.916278867599163e-06, + "loss": 0.724, + "step": 690 + }, + { + "epoch": 0.5723357261181667, + "grad_norm": 0.8914163708686829, + "learning_rate": 4.915996428113901e-06, + "loss": 0.7046, + "step": 691 + }, + { + "epoch": 0.5731639977912756, + "grad_norm": 0.8977548480033875, + "learning_rate": 4.91571352115585e-06, + "loss": 0.6823, + "step": 692 + }, + { + "epoch": 0.5739922694643843, + "grad_norm": 0.8902036547660828, + "learning_rate": 4.915430146779754e-06, + "loss": 0.702, + "step": 693 + }, + { + "epoch": 0.5748205411374931, + "grad_norm": 0.8859509229660034, + "learning_rate": 4.915146305040439e-06, + "loss": 0.6996, + "step": 694 + }, + { + "epoch": 0.5756488128106019, + "grad_norm": 0.8916864991188049, + "learning_rate": 4.914861995992829e-06, + "loss": 0.6978, + "step": 695 + }, + { + "epoch": 0.5764770844837107, + "grad_norm": 0.8961358070373535, + "learning_rate": 4.9145772196919336e-06, + "loss": 0.6803, + "step": 696 + }, + { + "epoch": 0.5773053561568194, + "grad_norm": 0.8817684650421143, + "learning_rate": 4.914291976192854e-06, + "loss": 0.6891, + "step": 697 + }, + { + "epoch": 0.5781336278299282, + "grad_norm": 0.8821113705635071, + "learning_rate": 4.914006265550783e-06, + "loss": 0.7009, + "step": 698 + }, + { + "epoch": 0.578961899503037, + "grad_norm": 0.9553062319755554, + "learning_rate": 4.913720087821003e-06, + "loss": 0.7118, + "step": 699 + }, + { + "epoch": 0.5797901711761457, + "grad_norm": 0.8997507691383362, + "learning_rate": 4.913433443058887e-06, + "loss": 0.7036, + "step": 700 + }, + { + "epoch": 0.5806184428492546, + "grad_norm": 0.8659473657608032, + "learning_rate": 4.9131463313198964e-06, + "loss": 0.6766, + "step": 701 + }, + { + "epoch": 0.5814467145223633, + "grad_norm": 0.8916030526161194, + "learning_rate": 4.912858752659586e-06, + "loss": 0.6976, + "step": 702 + }, + { + "epoch": 0.5822749861954721, + "grad_norm": 0.8726498484611511, + "learning_rate": 4.912570707133599e-06, + "loss": 0.6989, + "step": 703 + }, + { + "epoch": 0.5831032578685809, + "grad_norm": 0.8725529909133911, + "learning_rate": 4.91228219479767e-06, + "loss": 0.6862, + "step": 704 + }, + { + "epoch": 0.5839315295416897, + "grad_norm": 0.8807168006896973, + "learning_rate": 4.911993215707624e-06, + "loss": 0.6964, + "step": 705 + }, + { + "epoch": 0.5847598012147984, + "grad_norm": 0.9244167804718018, + "learning_rate": 4.911703769919375e-06, + "loss": 0.6742, + "step": 706 + }, + { + "epoch": 0.5855880728879073, + "grad_norm": 0.9174355268478394, + "learning_rate": 4.9114138574889285e-06, + "loss": 0.6852, + "step": 707 + }, + { + "epoch": 0.586416344561016, + "grad_norm": 0.8894474506378174, + "learning_rate": 4.9111234784723785e-06, + "loss": 0.689, + "step": 708 + }, + { + "epoch": 0.5872446162341248, + "grad_norm": 0.8856050968170166, + "learning_rate": 4.910832632925912e-06, + "loss": 0.6742, + "step": 709 + }, + { + "epoch": 0.5880728879072336, + "grad_norm": 0.9120550751686096, + "learning_rate": 4.910541320905804e-06, + "loss": 0.7016, + "step": 710 + }, + { + "epoch": 0.5889011595803424, + "grad_norm": 0.8925642967224121, + "learning_rate": 4.910249542468422e-06, + "loss": 0.7165, + "step": 711 + }, + { + "epoch": 0.5897294312534511, + "grad_norm": 0.8786419630050659, + "learning_rate": 4.909957297670222e-06, + "loss": 0.7207, + "step": 712 + }, + { + "epoch": 0.59055770292656, + "grad_norm": 0.89600670337677, + "learning_rate": 4.909664586567749e-06, + "loss": 0.672, + "step": 713 + }, + { + "epoch": 0.5913859745996687, + "grad_norm": 0.889835000038147, + "learning_rate": 4.909371409217642e-06, + "loss": 0.7089, + "step": 714 + }, + { + "epoch": 0.5922142462727775, + "grad_norm": 0.9029288291931152, + "learning_rate": 4.909077765676627e-06, + "loss": 0.7065, + "step": 715 + }, + { + "epoch": 0.5930425179458862, + "grad_norm": 0.9212722778320312, + "learning_rate": 4.908783656001521e-06, + "loss": 0.6787, + "step": 716 + }, + { + "epoch": 0.593870789618995, + "grad_norm": 0.9079681038856506, + "learning_rate": 4.9084890802492325e-06, + "loss": 0.6996, + "step": 717 + }, + { + "epoch": 0.5946990612921038, + "grad_norm": 0.9110735654830933, + "learning_rate": 4.908194038476758e-06, + "loss": 0.6962, + "step": 718 + }, + { + "epoch": 0.5955273329652125, + "grad_norm": 0.8945547938346863, + "learning_rate": 4.9078985307411865e-06, + "loss": 0.6648, + "step": 719 + }, + { + "epoch": 0.5963556046383214, + "grad_norm": 0.8831511735916138, + "learning_rate": 4.907602557099696e-06, + "loss": 0.7077, + "step": 720 + }, + { + "epoch": 0.5971838763114301, + "grad_norm": 0.9393531680107117, + "learning_rate": 4.9073061176095525e-06, + "loss": 0.6722, + "step": 721 + }, + { + "epoch": 0.5980121479845389, + "grad_norm": 0.8807207942008972, + "learning_rate": 4.907009212328117e-06, + "loss": 0.69, + "step": 722 + }, + { + "epoch": 0.5988404196576477, + "grad_norm": 0.8811002969741821, + "learning_rate": 4.906711841312836e-06, + "loss": 0.6902, + "step": 723 + }, + { + "epoch": 0.5996686913307565, + "grad_norm": 0.8953952789306641, + "learning_rate": 4.906414004621249e-06, + "loss": 0.6747, + "step": 724 + }, + { + "epoch": 0.6004969630038652, + "grad_norm": 0.8661350607872009, + "learning_rate": 4.9061157023109855e-06, + "loss": 0.7047, + "step": 725 + }, + { + "epoch": 0.6013252346769741, + "grad_norm": 0.8722146153450012, + "learning_rate": 4.905816934439762e-06, + "loss": 0.7231, + "step": 726 + }, + { + "epoch": 0.6021535063500828, + "grad_norm": 0.8933189511299133, + "learning_rate": 4.905517701065389e-06, + "loss": 0.7029, + "step": 727 + }, + { + "epoch": 0.6029817780231916, + "grad_norm": 0.8863442540168762, + "learning_rate": 4.905218002245764e-06, + "loss": 0.6886, + "step": 728 + }, + { + "epoch": 0.6038100496963004, + "grad_norm": 0.8672969341278076, + "learning_rate": 4.904917838038877e-06, + "loss": 0.7048, + "step": 729 + }, + { + "epoch": 0.6046383213694092, + "grad_norm": 0.8892956972122192, + "learning_rate": 4.904617208502807e-06, + "loss": 0.7009, + "step": 730 + }, + { + "epoch": 0.6054665930425179, + "grad_norm": 0.9083442687988281, + "learning_rate": 4.904316113695723e-06, + "loss": 0.7169, + "step": 731 + }, + { + "epoch": 0.6062948647156268, + "grad_norm": 0.8885907530784607, + "learning_rate": 4.904014553675883e-06, + "loss": 0.6939, + "step": 732 + }, + { + "epoch": 0.6071231363887355, + "grad_norm": 0.8859131336212158, + "learning_rate": 4.903712528501637e-06, + "loss": 0.6881, + "step": 733 + }, + { + "epoch": 0.6079514080618443, + "grad_norm": 0.874146580696106, + "learning_rate": 4.903410038231425e-06, + "loss": 0.6857, + "step": 734 + }, + { + "epoch": 0.6087796797349531, + "grad_norm": 0.8719701170921326, + "learning_rate": 4.903107082923774e-06, + "loss": 0.6834, + "step": 735 + }, + { + "epoch": 0.6096079514080619, + "grad_norm": 0.9656700491905212, + "learning_rate": 4.902803662637304e-06, + "loss": 0.6933, + "step": 736 + }, + { + "epoch": 0.6104362230811706, + "grad_norm": 0.9022018313407898, + "learning_rate": 4.9024997774307235e-06, + "loss": 0.683, + "step": 737 + }, + { + "epoch": 0.6112644947542794, + "grad_norm": 0.8694130182266235, + "learning_rate": 4.902195427362832e-06, + "loss": 0.642, + "step": 738 + }, + { + "epoch": 0.6120927664273882, + "grad_norm": 0.9153549075126648, + "learning_rate": 4.901890612492519e-06, + "loss": 0.6674, + "step": 739 + }, + { + "epoch": 0.6129210381004969, + "grad_norm": 0.9431132674217224, + "learning_rate": 4.901585332878762e-06, + "loss": 0.6918, + "step": 740 + }, + { + "epoch": 0.6137493097736058, + "grad_norm": 0.881587028503418, + "learning_rate": 4.901279588580631e-06, + "loss": 0.687, + "step": 741 + }, + { + "epoch": 0.6145775814467145, + "grad_norm": 0.9289383888244629, + "learning_rate": 4.9009733796572834e-06, + "loss": 0.6789, + "step": 742 + }, + { + "epoch": 0.6154058531198233, + "grad_norm": 0.9058337211608887, + "learning_rate": 4.900666706167968e-06, + "loss": 0.6856, + "step": 743 + }, + { + "epoch": 0.616234124792932, + "grad_norm": 0.8999397158622742, + "learning_rate": 4.900359568172024e-06, + "loss": 0.7159, + "step": 744 + }, + { + "epoch": 0.6170623964660409, + "grad_norm": 0.895321249961853, + "learning_rate": 4.90005196572888e-06, + "loss": 0.7071, + "step": 745 + }, + { + "epoch": 0.6178906681391496, + "grad_norm": 0.9142102599143982, + "learning_rate": 4.899743898898054e-06, + "loss": 0.6918, + "step": 746 + }, + { + "epoch": 0.6187189398122585, + "grad_norm": 0.8867097496986389, + "learning_rate": 4.899435367739152e-06, + "loss": 0.6991, + "step": 747 + }, + { + "epoch": 0.6195472114853672, + "grad_norm": 0.8594791293144226, + "learning_rate": 4.899126372311876e-06, + "loss": 0.7153, + "step": 748 + }, + { + "epoch": 0.620375483158476, + "grad_norm": 0.8740563988685608, + "learning_rate": 4.8988169126760085e-06, + "loss": 0.6786, + "step": 749 + }, + { + "epoch": 0.6212037548315847, + "grad_norm": 0.8818353414535522, + "learning_rate": 4.898506988891432e-06, + "loss": 0.6965, + "step": 750 + }, + { + "epoch": 0.6220320265046936, + "grad_norm": 0.9459843635559082, + "learning_rate": 4.898196601018111e-06, + "loss": 0.6821, + "step": 751 + }, + { + "epoch": 0.6228602981778023, + "grad_norm": 0.923457682132721, + "learning_rate": 4.897885749116104e-06, + "loss": 0.6822, + "step": 752 + }, + { + "epoch": 0.6236885698509111, + "grad_norm": 0.9342297315597534, + "learning_rate": 4.897574433245557e-06, + "loss": 0.7001, + "step": 753 + }, + { + "epoch": 0.6245168415240199, + "grad_norm": 0.9327735900878906, + "learning_rate": 4.897262653466706e-06, + "loss": 0.6921, + "step": 754 + }, + { + "epoch": 0.6253451131971287, + "grad_norm": 0.8876991868019104, + "learning_rate": 4.8969504098398805e-06, + "loss": 0.7087, + "step": 755 + }, + { + "epoch": 0.6261733848702374, + "grad_norm": 0.8983528017997742, + "learning_rate": 4.896637702425493e-06, + "loss": 0.6824, + "step": 756 + }, + { + "epoch": 0.6270016565433462, + "grad_norm": 0.9074965715408325, + "learning_rate": 4.896324531284053e-06, + "loss": 0.6533, + "step": 757 + }, + { + "epoch": 0.627829928216455, + "grad_norm": 0.8938930630683899, + "learning_rate": 4.896010896476153e-06, + "loss": 0.6859, + "step": 758 + }, + { + "epoch": 0.6286581998895637, + "grad_norm": 0.9036867022514343, + "learning_rate": 4.8956967980624806e-06, + "loss": 0.6904, + "step": 759 + }, + { + "epoch": 0.6294864715626726, + "grad_norm": 0.9162903428077698, + "learning_rate": 4.895382236103809e-06, + "loss": 0.6995, + "step": 760 + }, + { + "epoch": 0.6303147432357813, + "grad_norm": 1.0000487565994263, + "learning_rate": 4.895067210661005e-06, + "loss": 0.671, + "step": 761 + }, + { + "epoch": 0.6311430149088901, + "grad_norm": 0.895163893699646, + "learning_rate": 4.894751721795022e-06, + "loss": 0.6871, + "step": 762 + }, + { + "epoch": 0.6319712865819989, + "grad_norm": 0.9401512742042542, + "learning_rate": 4.894435769566905e-06, + "loss": 0.6811, + "step": 763 + }, + { + "epoch": 0.6327995582551077, + "grad_norm": 0.924578845500946, + "learning_rate": 4.894119354037787e-06, + "loss": 0.692, + "step": 764 + }, + { + "epoch": 0.6336278299282164, + "grad_norm": 0.9114750623703003, + "learning_rate": 4.893802475268892e-06, + "loss": 0.6977, + "step": 765 + }, + { + "epoch": 0.6344561016013253, + "grad_norm": 1.1472550630569458, + "learning_rate": 4.893485133321533e-06, + "loss": 0.679, + "step": 766 + }, + { + "epoch": 0.635284373274434, + "grad_norm": 0.9460884928703308, + "learning_rate": 4.893167328257111e-06, + "loss": 0.6988, + "step": 767 + }, + { + "epoch": 0.6361126449475428, + "grad_norm": 0.9040783643722534, + "learning_rate": 4.8928490601371216e-06, + "loss": 0.6706, + "step": 768 + }, + { + "epoch": 0.6369409166206516, + "grad_norm": 0.9785428047180176, + "learning_rate": 4.892530329023144e-06, + "loss": 0.7185, + "step": 769 + }, + { + "epoch": 0.6377691882937604, + "grad_norm": 0.9795847535133362, + "learning_rate": 4.892211134976851e-06, + "loss": 0.703, + "step": 770 + }, + { + "epoch": 0.6385974599668691, + "grad_norm": 0.9539455771446228, + "learning_rate": 4.8918914780600025e-06, + "loss": 0.6985, + "step": 771 + }, + { + "epoch": 0.639425731639978, + "grad_norm": 0.9304025173187256, + "learning_rate": 4.8915713583344505e-06, + "loss": 0.6731, + "step": 772 + }, + { + "epoch": 0.6402540033130867, + "grad_norm": 0.9466667771339417, + "learning_rate": 4.891250775862134e-06, + "loss": 0.6972, + "step": 773 + }, + { + "epoch": 0.6410822749861955, + "grad_norm": 0.9522298574447632, + "learning_rate": 4.890929730705084e-06, + "loss": 0.6793, + "step": 774 + }, + { + "epoch": 0.6419105466593042, + "grad_norm": 0.9077293276786804, + "learning_rate": 4.890608222925419e-06, + "loss": 0.6892, + "step": 775 + }, + { + "epoch": 0.6427388183324131, + "grad_norm": 0.8763419389724731, + "learning_rate": 4.890286252585348e-06, + "loss": 0.6985, + "step": 776 + }, + { + "epoch": 0.6435670900055218, + "grad_norm": 0.9089516401290894, + "learning_rate": 4.8899638197471685e-06, + "loss": 0.6778, + "step": 777 + }, + { + "epoch": 0.6443953616786305, + "grad_norm": 0.8733325600624084, + "learning_rate": 4.88964092447327e-06, + "loss": 0.6915, + "step": 778 + }, + { + "epoch": 0.6452236333517394, + "grad_norm": 0.9472198486328125, + "learning_rate": 4.889317566826128e-06, + "loss": 0.6984, + "step": 779 + }, + { + "epoch": 0.6460519050248481, + "grad_norm": 0.8729666471481323, + "learning_rate": 4.888993746868309e-06, + "loss": 0.6724, + "step": 780 + }, + { + "epoch": 0.6468801766979569, + "grad_norm": 0.9395233988761902, + "learning_rate": 4.88866946466247e-06, + "loss": 0.6697, + "step": 781 + }, + { + "epoch": 0.6477084483710657, + "grad_norm": 0.9158744812011719, + "learning_rate": 4.888344720271357e-06, + "loss": 0.6838, + "step": 782 + }, + { + "epoch": 0.6485367200441745, + "grad_norm": 0.8961113095283508, + "learning_rate": 4.888019513757804e-06, + "loss": 0.6488, + "step": 783 + }, + { + "epoch": 0.6493649917172832, + "grad_norm": 0.8911786079406738, + "learning_rate": 4.8876938451847364e-06, + "loss": 0.7089, + "step": 784 + }, + { + "epoch": 0.6501932633903921, + "grad_norm": 0.9214697480201721, + "learning_rate": 4.887367714615167e-06, + "loss": 0.6929, + "step": 785 + }, + { + "epoch": 0.6510215350635008, + "grad_norm": 0.9161441326141357, + "learning_rate": 4.8870411221122e-06, + "loss": 0.6761, + "step": 786 + }, + { + "epoch": 0.6518498067366096, + "grad_norm": 0.9161947965621948, + "learning_rate": 4.886714067739028e-06, + "loss": 0.6804, + "step": 787 + }, + { + "epoch": 0.6526780784097184, + "grad_norm": 0.8706585168838501, + "learning_rate": 4.886386551558933e-06, + "loss": 0.6836, + "step": 788 + }, + { + "epoch": 0.6535063500828272, + "grad_norm": 0.9037661552429199, + "learning_rate": 4.886058573635284e-06, + "loss": 0.6786, + "step": 789 + }, + { + "epoch": 0.6543346217559359, + "grad_norm": 0.8977832794189453, + "learning_rate": 4.885730134031545e-06, + "loss": 0.6938, + "step": 790 + }, + { + "epoch": 0.6551628934290448, + "grad_norm": 0.9106955528259277, + "learning_rate": 4.8854012328112645e-06, + "loss": 0.678, + "step": 791 + }, + { + "epoch": 0.6559911651021535, + "grad_norm": 0.8769396543502808, + "learning_rate": 4.885071870038082e-06, + "loss": 0.6969, + "step": 792 + }, + { + "epoch": 0.6568194367752623, + "grad_norm": 0.9946926832199097, + "learning_rate": 4.884742045775725e-06, + "loss": 0.6799, + "step": 793 + }, + { + "epoch": 0.6576477084483711, + "grad_norm": 0.9399042725563049, + "learning_rate": 4.884411760088012e-06, + "loss": 0.7038, + "step": 794 + }, + { + "epoch": 0.6584759801214799, + "grad_norm": 0.8914628624916077, + "learning_rate": 4.8840810130388526e-06, + "loss": 0.6867, + "step": 795 + }, + { + "epoch": 0.6593042517945886, + "grad_norm": 0.8900558352470398, + "learning_rate": 4.88374980469224e-06, + "loss": 0.6823, + "step": 796 + }, + { + "epoch": 0.6601325234676974, + "grad_norm": 0.8608913421630859, + "learning_rate": 4.8834181351122615e-06, + "loss": 0.6705, + "step": 797 + }, + { + "epoch": 0.6609607951408062, + "grad_norm": 0.900460422039032, + "learning_rate": 4.883086004363092e-06, + "loss": 0.7029, + "step": 798 + }, + { + "epoch": 0.6617890668139149, + "grad_norm": 0.8932703733444214, + "learning_rate": 4.882753412508996e-06, + "loss": 0.6845, + "step": 799 + }, + { + "epoch": 0.6626173384870238, + "grad_norm": 0.8737874627113342, + "learning_rate": 4.882420359614325e-06, + "loss": 0.6979, + "step": 800 + }, + { + "epoch": 0.6634456101601325, + "grad_norm": 0.8805396556854248, + "learning_rate": 4.882086845743524e-06, + "loss": 0.6935, + "step": 801 + }, + { + "epoch": 0.6642738818332413, + "grad_norm": 0.8952041268348694, + "learning_rate": 4.881752870961123e-06, + "loss": 0.7137, + "step": 802 + }, + { + "epoch": 0.66510215350635, + "grad_norm": 0.889869213104248, + "learning_rate": 4.881418435331745e-06, + "loss": 0.6643, + "step": 803 + }, + { + "epoch": 0.6659304251794589, + "grad_norm": 0.8912904858589172, + "learning_rate": 4.881083538920098e-06, + "loss": 0.6712, + "step": 804 + }, + { + "epoch": 0.6667586968525676, + "grad_norm": 0.8943006992340088, + "learning_rate": 4.880748181790984e-06, + "loss": 0.7001, + "step": 805 + }, + { + "epoch": 0.6675869685256764, + "grad_norm": 0.9042482376098633, + "learning_rate": 4.880412364009289e-06, + "loss": 0.6744, + "step": 806 + }, + { + "epoch": 0.6684152401987852, + "grad_norm": 0.9052988886833191, + "learning_rate": 4.880076085639991e-06, + "loss": 0.72, + "step": 807 + }, + { + "epoch": 0.669243511871894, + "grad_norm": 0.9702905416488647, + "learning_rate": 4.879739346748158e-06, + "loss": 0.6871, + "step": 808 + }, + { + "epoch": 0.6700717835450027, + "grad_norm": 0.8741766810417175, + "learning_rate": 4.879402147398944e-06, + "loss": 0.7174, + "step": 809 + }, + { + "epoch": 0.6709000552181116, + "grad_norm": 0.9226665496826172, + "learning_rate": 4.879064487657596e-06, + "loss": 0.6991, + "step": 810 + }, + { + "epoch": 0.6717283268912203, + "grad_norm": 0.878304123878479, + "learning_rate": 4.878726367589446e-06, + "loss": 0.6925, + "step": 811 + }, + { + "epoch": 0.6725565985643291, + "grad_norm": 0.893649160861969, + "learning_rate": 4.878387787259919e-06, + "loss": 0.6689, + "step": 812 + }, + { + "epoch": 0.6733848702374379, + "grad_norm": 0.917861819267273, + "learning_rate": 4.878048746734526e-06, + "loss": 0.7171, + "step": 813 + }, + { + "epoch": 0.6742131419105467, + "grad_norm": 0.879375696182251, + "learning_rate": 4.877709246078869e-06, + "loss": 0.6662, + "step": 814 + }, + { + "epoch": 0.6750414135836554, + "grad_norm": 0.9098542928695679, + "learning_rate": 4.877369285358637e-06, + "loss": 0.6793, + "step": 815 + }, + { + "epoch": 0.6758696852567642, + "grad_norm": 0.9465262293815613, + "learning_rate": 4.877028864639609e-06, + "loss": 0.7065, + "step": 816 + }, + { + "epoch": 0.676697956929873, + "grad_norm": 0.9029957056045532, + "learning_rate": 4.876687983987654e-06, + "loss": 0.7008, + "step": 817 + }, + { + "epoch": 0.6775262286029817, + "grad_norm": 0.9073920845985413, + "learning_rate": 4.876346643468731e-06, + "loss": 0.6836, + "step": 818 + }, + { + "epoch": 0.6783545002760906, + "grad_norm": 0.9028745889663696, + "learning_rate": 4.876004843148884e-06, + "loss": 0.6965, + "step": 819 + }, + { + "epoch": 0.6791827719491993, + "grad_norm": 0.8745572566986084, + "learning_rate": 4.875662583094249e-06, + "loss": 0.6974, + "step": 820 + }, + { + "epoch": 0.6800110436223081, + "grad_norm": 0.9203384518623352, + "learning_rate": 4.875319863371048e-06, + "loss": 0.686, + "step": 821 + }, + { + "epoch": 0.6808393152954169, + "grad_norm": 0.9087005853652954, + "learning_rate": 4.874976684045597e-06, + "loss": 0.683, + "step": 822 + }, + { + "epoch": 0.6816675869685257, + "grad_norm": 0.9007741808891296, + "learning_rate": 4.874633045184297e-06, + "loss": 0.6938, + "step": 823 + }, + { + "epoch": 0.6824958586416344, + "grad_norm": 0.9854089617729187, + "learning_rate": 4.874288946853639e-06, + "loss": 0.6945, + "step": 824 + }, + { + "epoch": 0.6833241303147433, + "grad_norm": 0.9044802784919739, + "learning_rate": 4.8739443891202025e-06, + "loss": 0.714, + "step": 825 + }, + { + "epoch": 0.684152401987852, + "grad_norm": 0.9388754963874817, + "learning_rate": 4.873599372050656e-06, + "loss": 0.6838, + "step": 826 + }, + { + "epoch": 0.6849806736609608, + "grad_norm": 0.9215866327285767, + "learning_rate": 4.873253895711757e-06, + "loss": 0.6994, + "step": 827 + }, + { + "epoch": 0.6858089453340696, + "grad_norm": 0.8786854147911072, + "learning_rate": 4.872907960170353e-06, + "loss": 0.6883, + "step": 828 + }, + { + "epoch": 0.6866372170071784, + "grad_norm": 0.9032605290412903, + "learning_rate": 4.872561565493379e-06, + "loss": 0.6847, + "step": 829 + }, + { + "epoch": 0.6874654886802871, + "grad_norm": 0.8940309882164001, + "learning_rate": 4.872214711747858e-06, + "loss": 0.6877, + "step": 830 + }, + { + "epoch": 0.688293760353396, + "grad_norm": 0.8703469634056091, + "learning_rate": 4.8718673990009045e-06, + "loss": 0.6829, + "step": 831 + }, + { + "epoch": 0.6891220320265047, + "grad_norm": 0.8618217706680298, + "learning_rate": 4.871519627319719e-06, + "loss": 0.6987, + "step": 832 + }, + { + "epoch": 0.6899503036996135, + "grad_norm": 0.8871967196464539, + "learning_rate": 4.871171396771594e-06, + "loss": 0.6824, + "step": 833 + }, + { + "epoch": 0.6907785753727222, + "grad_norm": 0.9162507057189941, + "learning_rate": 4.8708227074239064e-06, + "loss": 0.7158, + "step": 834 + }, + { + "epoch": 0.6916068470458311, + "grad_norm": 0.8829584717750549, + "learning_rate": 4.8704735593441255e-06, + "loss": 0.6714, + "step": 835 + }, + { + "epoch": 0.6924351187189398, + "grad_norm": 0.9130700826644897, + "learning_rate": 4.8701239525998076e-06, + "loss": 0.6841, + "step": 836 + }, + { + "epoch": 0.6932633903920485, + "grad_norm": 0.8706396222114563, + "learning_rate": 4.8697738872586e-06, + "loss": 0.7162, + "step": 837 + }, + { + "epoch": 0.6940916620651574, + "grad_norm": 0.8906163573265076, + "learning_rate": 4.869423363388235e-06, + "loss": 0.6586, + "step": 838 + }, + { + "epoch": 0.6949199337382661, + "grad_norm": 0.9134309887886047, + "learning_rate": 4.869072381056537e-06, + "loss": 0.6842, + "step": 839 + }, + { + "epoch": 0.6957482054113749, + "grad_norm": 0.869674563407898, + "learning_rate": 4.8687209403314175e-06, + "loss": 0.7049, + "step": 840 + }, + { + "epoch": 0.6965764770844837, + "grad_norm": 0.8405192494392395, + "learning_rate": 4.8683690412808775e-06, + "loss": 0.6677, + "step": 841 + }, + { + "epoch": 0.6974047487575925, + "grad_norm": 0.9173605442047119, + "learning_rate": 4.868016683973005e-06, + "loss": 0.6784, + "step": 842 + }, + { + "epoch": 0.6982330204307012, + "grad_norm": 0.9220597743988037, + "learning_rate": 4.867663868475978e-06, + "loss": 0.6774, + "step": 843 + }, + { + "epoch": 0.6990612921038101, + "grad_norm": 0.8687174320220947, + "learning_rate": 4.867310594858065e-06, + "loss": 0.7094, + "step": 844 + }, + { + "epoch": 0.6998895637769188, + "grad_norm": 0.9161131381988525, + "learning_rate": 4.866956863187619e-06, + "loss": 0.6867, + "step": 845 + }, + { + "epoch": 0.7007178354500276, + "grad_norm": 0.9338458776473999, + "learning_rate": 4.866602673533084e-06, + "loss": 0.6783, + "step": 846 + }, + { + "epoch": 0.7015461071231364, + "grad_norm": 0.9268187284469604, + "learning_rate": 4.866248025962993e-06, + "loss": 0.6681, + "step": 847 + }, + { + "epoch": 0.7023743787962452, + "grad_norm": 0.8843579292297363, + "learning_rate": 4.865892920545967e-06, + "loss": 0.6854, + "step": 848 + }, + { + "epoch": 0.7032026504693539, + "grad_norm": 0.889299213886261, + "learning_rate": 4.865537357350716e-06, + "loss": 0.7055, + "step": 849 + }, + { + "epoch": 0.7040309221424628, + "grad_norm": 0.9062727093696594, + "learning_rate": 4.865181336446036e-06, + "loss": 0.6806, + "step": 850 + }, + { + "epoch": 0.7048591938155715, + "grad_norm": 0.9007996320724487, + "learning_rate": 4.864824857900817e-06, + "loss": 0.6937, + "step": 851 + }, + { + "epoch": 0.7056874654886803, + "grad_norm": 0.8883542418479919, + "learning_rate": 4.864467921784032e-06, + "loss": 0.6957, + "step": 852 + }, + { + "epoch": 0.7065157371617891, + "grad_norm": 0.9422078132629395, + "learning_rate": 4.864110528164746e-06, + "loss": 0.7025, + "step": 853 + }, + { + "epoch": 0.7073440088348979, + "grad_norm": 0.8961433172225952, + "learning_rate": 4.86375267711211e-06, + "loss": 0.6795, + "step": 854 + }, + { + "epoch": 0.7081722805080066, + "grad_norm": 0.925815761089325, + "learning_rate": 4.863394368695366e-06, + "loss": 0.6824, + "step": 855 + }, + { + "epoch": 0.7090005521811154, + "grad_norm": 0.9326804280281067, + "learning_rate": 4.863035602983843e-06, + "loss": 0.6829, + "step": 856 + }, + { + "epoch": 0.7098288238542242, + "grad_norm": 0.8784539699554443, + "learning_rate": 4.862676380046959e-06, + "loss": 0.6776, + "step": 857 + }, + { + "epoch": 0.7106570955273329, + "grad_norm": 0.8577290177345276, + "learning_rate": 4.86231669995422e-06, + "loss": 0.6573, + "step": 858 + }, + { + "epoch": 0.7114853672004418, + "grad_norm": 0.9261177778244019, + "learning_rate": 4.86195656277522e-06, + "loss": 0.6891, + "step": 859 + }, + { + "epoch": 0.7123136388735505, + "grad_norm": 0.9008646607398987, + "learning_rate": 4.861595968579643e-06, + "loss": 0.6714, + "step": 860 + }, + { + "epoch": 0.7131419105466593, + "grad_norm": 0.8729333877563477, + "learning_rate": 4.861234917437261e-06, + "loss": 0.6819, + "step": 861 + }, + { + "epoch": 0.713970182219768, + "grad_norm": 0.9006103873252869, + "learning_rate": 4.860873409417934e-06, + "loss": 0.6702, + "step": 862 + }, + { + "epoch": 0.7147984538928769, + "grad_norm": 0.8699921369552612, + "learning_rate": 4.860511444591609e-06, + "loss": 0.6681, + "step": 863 + }, + { + "epoch": 0.7156267255659856, + "grad_norm": 0.9359689950942993, + "learning_rate": 4.860149023028325e-06, + "loss": 0.6834, + "step": 864 + }, + { + "epoch": 0.7164549972390944, + "grad_norm": 0.874977707862854, + "learning_rate": 4.859786144798205e-06, + "loss": 0.6692, + "step": 865 + }, + { + "epoch": 0.7172832689122032, + "grad_norm": 0.8821915984153748, + "learning_rate": 4.859422809971464e-06, + "loss": 0.6865, + "step": 866 + }, + { + "epoch": 0.718111540585312, + "grad_norm": 0.905248761177063, + "learning_rate": 4.859059018618405e-06, + "loss": 0.6705, + "step": 867 + }, + { + "epoch": 0.7189398122584207, + "grad_norm": 0.9134542346000671, + "learning_rate": 4.858694770809415e-06, + "loss": 0.7039, + "step": 868 + }, + { + "epoch": 0.7197680839315296, + "grad_norm": 0.9048055410385132, + "learning_rate": 4.858330066614975e-06, + "loss": 0.6848, + "step": 869 + }, + { + "epoch": 0.7205963556046383, + "grad_norm": 0.8780005574226379, + "learning_rate": 4.857964906105651e-06, + "loss": 0.689, + "step": 870 + }, + { + "epoch": 0.7214246272777471, + "grad_norm": 0.8826956152915955, + "learning_rate": 4.857599289352098e-06, + "loss": 0.6799, + "step": 871 + }, + { + "epoch": 0.7222528989508559, + "grad_norm": 0.919905960559845, + "learning_rate": 4.857233216425061e-06, + "loss": 0.6587, + "step": 872 + }, + { + "epoch": 0.7230811706239647, + "grad_norm": 0.8711065053939819, + "learning_rate": 4.8568666873953696e-06, + "loss": 0.6754, + "step": 873 + }, + { + "epoch": 0.7239094422970734, + "grad_norm": 0.8740426898002625, + "learning_rate": 4.856499702333944e-06, + "loss": 0.6837, + "step": 874 + }, + { + "epoch": 0.7247377139701823, + "grad_norm": 0.921416699886322, + "learning_rate": 4.8561322613117945e-06, + "loss": 0.6709, + "step": 875 + }, + { + "epoch": 0.725565985643291, + "grad_norm": 0.8988221287727356, + "learning_rate": 4.855764364400016e-06, + "loss": 0.7111, + "step": 876 + }, + { + "epoch": 0.7263942573163997, + "grad_norm": 0.8838291168212891, + "learning_rate": 4.8553960116697915e-06, + "loss": 0.6821, + "step": 877 + }, + { + "epoch": 0.7272225289895086, + "grad_norm": 0.9051623940467834, + "learning_rate": 4.8550272031923965e-06, + "loss": 0.6848, + "step": 878 + }, + { + "epoch": 0.7280508006626173, + "grad_norm": 0.9308082461357117, + "learning_rate": 4.85465793903919e-06, + "loss": 0.7087, + "step": 879 + }, + { + "epoch": 0.7288790723357261, + "grad_norm": 0.873214066028595, + "learning_rate": 4.854288219281623e-06, + "loss": 0.6877, + "step": 880 + }, + { + "epoch": 0.7297073440088349, + "grad_norm": 0.8890054225921631, + "learning_rate": 4.853918043991232e-06, + "loss": 0.6719, + "step": 881 + }, + { + "epoch": 0.7305356156819437, + "grad_norm": 0.9580398201942444, + "learning_rate": 4.853547413239642e-06, + "loss": 0.6439, + "step": 882 + }, + { + "epoch": 0.7313638873550524, + "grad_norm": 0.875035285949707, + "learning_rate": 4.853176327098567e-06, + "loss": 0.6645, + "step": 883 + }, + { + "epoch": 0.7321921590281613, + "grad_norm": 0.9299217462539673, + "learning_rate": 4.85280478563981e-06, + "loss": 0.707, + "step": 884 + }, + { + "epoch": 0.73302043070127, + "grad_norm": 0.8840609192848206, + "learning_rate": 4.8524327889352585e-06, + "loss": 0.6789, + "step": 885 + }, + { + "epoch": 0.7338487023743788, + "grad_norm": 0.8983805179595947, + "learning_rate": 4.852060337056891e-06, + "loss": 0.6814, + "step": 886 + }, + { + "epoch": 0.7346769740474876, + "grad_norm": 0.8730597496032715, + "learning_rate": 4.851687430076775e-06, + "loss": 0.7006, + "step": 887 + }, + { + "epoch": 0.7355052457205964, + "grad_norm": 0.8980515003204346, + "learning_rate": 4.851314068067063e-06, + "loss": 0.6729, + "step": 888 + }, + { + "epoch": 0.7363335173937051, + "grad_norm": 0.8919468522071838, + "learning_rate": 4.850940251099997e-06, + "loss": 0.6764, + "step": 889 + }, + { + "epoch": 0.737161789066814, + "grad_norm": 0.8987070918083191, + "learning_rate": 4.850565979247908e-06, + "loss": 0.6997, + "step": 890 + }, + { + "epoch": 0.7379900607399227, + "grad_norm": 0.9056184887886047, + "learning_rate": 4.850191252583214e-06, + "loss": 0.6887, + "step": 891 + }, + { + "epoch": 0.7388183324130315, + "grad_norm": 0.8714621067047119, + "learning_rate": 4.84981607117842e-06, + "loss": 0.6946, + "step": 892 + }, + { + "epoch": 0.7396466040861402, + "grad_norm": 0.8845617175102234, + "learning_rate": 4.849440435106122e-06, + "loss": 0.6845, + "step": 893 + }, + { + "epoch": 0.7404748757592491, + "grad_norm": 0.8926048278808594, + "learning_rate": 4.849064344439e-06, + "loss": 0.6922, + "step": 894 + }, + { + "epoch": 0.7413031474323578, + "grad_norm": 0.8731569647789001, + "learning_rate": 4.848687799249826e-06, + "loss": 0.6837, + "step": 895 + }, + { + "epoch": 0.7421314191054665, + "grad_norm": 0.8801183104515076, + "learning_rate": 4.848310799611457e-06, + "loss": 0.6932, + "step": 896 + }, + { + "epoch": 0.7429596907785754, + "grad_norm": 0.8818139433860779, + "learning_rate": 4.847933345596838e-06, + "loss": 0.6758, + "step": 897 + }, + { + "epoch": 0.7437879624516841, + "grad_norm": 0.9117923974990845, + "learning_rate": 4.847555437279004e-06, + "loss": 0.6628, + "step": 898 + }, + { + "epoch": 0.7446162341247929, + "grad_norm": 0.891360342502594, + "learning_rate": 4.847177074731078e-06, + "loss": 0.6876, + "step": 899 + }, + { + "epoch": 0.7454445057979017, + "grad_norm": 0.8905263543128967, + "learning_rate": 4.846798258026267e-06, + "loss": 0.6862, + "step": 900 + }, + { + "epoch": 0.7462727774710105, + "grad_norm": 0.8853707909584045, + "learning_rate": 4.846418987237868e-06, + "loss": 0.654, + "step": 901 + }, + { + "epoch": 0.7471010491441192, + "grad_norm": 0.8782089948654175, + "learning_rate": 4.846039262439271e-06, + "loss": 0.7005, + "step": 902 + }, + { + "epoch": 0.7479293208172281, + "grad_norm": 0.8774383068084717, + "learning_rate": 4.8456590837039445e-06, + "loss": 0.7044, + "step": 903 + }, + { + "epoch": 0.7487575924903368, + "grad_norm": 0.9385423064231873, + "learning_rate": 4.845278451105451e-06, + "loss": 0.676, + "step": 904 + }, + { + "epoch": 0.7495858641634456, + "grad_norm": 0.9211432337760925, + "learning_rate": 4.8448973647174405e-06, + "loss": 0.697, + "step": 905 + }, + { + "epoch": 0.7504141358365544, + "grad_norm": 0.887610912322998, + "learning_rate": 4.844515824613649e-06, + "loss": 0.6965, + "step": 906 + }, + { + "epoch": 0.7512424075096632, + "grad_norm": 0.885761559009552, + "learning_rate": 4.8441338308679e-06, + "loss": 0.6851, + "step": 907 + }, + { + "epoch": 0.7520706791827719, + "grad_norm": 0.9209930300712585, + "learning_rate": 4.843751383554106e-06, + "loss": 0.6718, + "step": 908 + }, + { + "epoch": 0.7528989508558808, + "grad_norm": 0.9308252334594727, + "learning_rate": 4.843368482746269e-06, + "loss": 0.6802, + "step": 909 + }, + { + "epoch": 0.7537272225289895, + "grad_norm": 0.8931897282600403, + "learning_rate": 4.842985128518473e-06, + "loss": 0.7076, + "step": 910 + }, + { + "epoch": 0.7545554942020983, + "grad_norm": 0.9102733731269836, + "learning_rate": 4.842601320944898e-06, + "loss": 0.6898, + "step": 911 + }, + { + "epoch": 0.7553837658752071, + "grad_norm": 0.9336893558502197, + "learning_rate": 4.842217060099804e-06, + "loss": 0.6709, + "step": 912 + }, + { + "epoch": 0.7562120375483159, + "grad_norm": 0.9366991519927979, + "learning_rate": 4.8418323460575425e-06, + "loss": 0.6665, + "step": 913 + }, + { + "epoch": 0.7570403092214246, + "grad_norm": 0.9371415376663208, + "learning_rate": 4.841447178892552e-06, + "loss": 0.7013, + "step": 914 + }, + { + "epoch": 0.7578685808945335, + "grad_norm": 1.031123399734497, + "learning_rate": 4.84106155867936e-06, + "loss": 0.687, + "step": 915 + }, + { + "epoch": 0.7586968525676422, + "grad_norm": 0.9319218397140503, + "learning_rate": 4.8406754854925785e-06, + "loss": 0.669, + "step": 916 + }, + { + "epoch": 0.7595251242407509, + "grad_norm": 0.8563560843467712, + "learning_rate": 4.840288959406911e-06, + "loss": 0.6905, + "step": 917 + }, + { + "epoch": 0.7603533959138598, + "grad_norm": 0.9424378275871277, + "learning_rate": 4.839901980497145e-06, + "loss": 0.6685, + "step": 918 + }, + { + "epoch": 0.7611816675869685, + "grad_norm": 0.9202269315719604, + "learning_rate": 4.839514548838158e-06, + "loss": 0.6728, + "step": 919 + }, + { + "epoch": 0.7620099392600773, + "grad_norm": 0.9152582287788391, + "learning_rate": 4.839126664504915e-06, + "loss": 0.6842, + "step": 920 + }, + { + "epoch": 0.762838210933186, + "grad_norm": 0.9457129836082458, + "learning_rate": 4.838738327572468e-06, + "loss": 0.6822, + "step": 921 + }, + { + "epoch": 0.7636664826062949, + "grad_norm": 0.93124920129776, + "learning_rate": 4.838349538115955e-06, + "loss": 0.691, + "step": 922 + }, + { + "epoch": 0.7644947542794036, + "grad_norm": 0.9249301552772522, + "learning_rate": 4.837960296210605e-06, + "loss": 0.6574, + "step": 923 + }, + { + "epoch": 0.7653230259525124, + "grad_norm": 1.0551042556762695, + "learning_rate": 4.837570601931731e-06, + "loss": 0.6782, + "step": 924 + }, + { + "epoch": 0.7661512976256212, + "grad_norm": 0.9404964447021484, + "learning_rate": 4.8371804553547365e-06, + "loss": 0.6703, + "step": 925 + }, + { + "epoch": 0.76697956929873, + "grad_norm": 0.909015417098999, + "learning_rate": 4.836789856555111e-06, + "loss": 0.6812, + "step": 926 + }, + { + "epoch": 0.7678078409718387, + "grad_norm": 0.8572261929512024, + "learning_rate": 4.836398805608431e-06, + "loss": 0.6831, + "step": 927 + }, + { + "epoch": 0.7686361126449476, + "grad_norm": 0.951753556728363, + "learning_rate": 4.836007302590362e-06, + "loss": 0.6672, + "step": 928 + }, + { + "epoch": 0.7694643843180563, + "grad_norm": 0.8875932693481445, + "learning_rate": 4.8356153475766564e-06, + "loss": 0.6819, + "step": 929 + }, + { + "epoch": 0.7702926559911651, + "grad_norm": 0.9405681490898132, + "learning_rate": 4.835222940643153e-06, + "loss": 0.6917, + "step": 930 + }, + { + "epoch": 0.7711209276642739, + "grad_norm": 0.942520022392273, + "learning_rate": 4.834830081865779e-06, + "loss": 0.6657, + "step": 931 + }, + { + "epoch": 0.7719491993373827, + "grad_norm": 0.9063259363174438, + "learning_rate": 4.83443677132055e-06, + "loss": 0.6857, + "step": 932 + }, + { + "epoch": 0.7727774710104914, + "grad_norm": 0.8778199553489685, + "learning_rate": 4.834043009083566e-06, + "loss": 0.7028, + "step": 933 + }, + { + "epoch": 0.7736057426836003, + "grad_norm": 0.9215186834335327, + "learning_rate": 4.833648795231018e-06, + "loss": 0.6712, + "step": 934 + }, + { + "epoch": 0.774434014356709, + "grad_norm": 0.8992031812667847, + "learning_rate": 4.8332541298391825e-06, + "loss": 0.6827, + "step": 935 + }, + { + "epoch": 0.7752622860298177, + "grad_norm": 0.9203833341598511, + "learning_rate": 4.8328590129844224e-06, + "loss": 0.6835, + "step": 936 + }, + { + "epoch": 0.7760905577029266, + "grad_norm": 0.9109310507774353, + "learning_rate": 4.832463444743191e-06, + "loss": 0.673, + "step": 937 + }, + { + "epoch": 0.7769188293760353, + "grad_norm": 0.978349506855011, + "learning_rate": 4.832067425192024e-06, + "loss": 0.6638, + "step": 938 + }, + { + "epoch": 0.7777471010491441, + "grad_norm": 0.9552261829376221, + "learning_rate": 4.831670954407551e-06, + "loss": 0.6919, + "step": 939 + }, + { + "epoch": 0.7785753727222529, + "grad_norm": 0.9106539487838745, + "learning_rate": 4.831274032466484e-06, + "loss": 0.6964, + "step": 940 + }, + { + "epoch": 0.7794036443953617, + "grad_norm": 0.9195299744606018, + "learning_rate": 4.830876659445623e-06, + "loss": 0.6793, + "step": 941 + }, + { + "epoch": 0.7802319160684704, + "grad_norm": 0.9075512290000916, + "learning_rate": 4.8304788354218564e-06, + "loss": 0.6649, + "step": 942 + }, + { + "epoch": 0.7810601877415793, + "grad_norm": 0.9142526984214783, + "learning_rate": 4.83008056047216e-06, + "loss": 0.6811, + "step": 943 + }, + { + "epoch": 0.781888459414688, + "grad_norm": 0.8898900151252747, + "learning_rate": 4.829681834673597e-06, + "loss": 0.6906, + "step": 944 + }, + { + "epoch": 0.7827167310877968, + "grad_norm": 0.8884983062744141, + "learning_rate": 4.8292826581033135e-06, + "loss": 0.6903, + "step": 945 + }, + { + "epoch": 0.7835450027609056, + "grad_norm": 0.9075689315795898, + "learning_rate": 4.828883030838551e-06, + "loss": 0.6794, + "step": 946 + }, + { + "epoch": 0.7843732744340144, + "grad_norm": 0.8970300555229187, + "learning_rate": 4.828482952956631e-06, + "loss": 0.6715, + "step": 947 + }, + { + "epoch": 0.7852015461071231, + "grad_norm": 0.8906447887420654, + "learning_rate": 4.828082424534965e-06, + "loss": 0.6878, + "step": 948 + }, + { + "epoch": 0.786029817780232, + "grad_norm": 0.9083748459815979, + "learning_rate": 4.827681445651052e-06, + "loss": 0.6787, + "step": 949 + }, + { + "epoch": 0.7868580894533407, + "grad_norm": 0.8975276947021484, + "learning_rate": 4.8272800163824784e-06, + "loss": 0.687, + "step": 950 + }, + { + "epoch": 0.7876863611264495, + "grad_norm": 0.9573839902877808, + "learning_rate": 4.8268781368069156e-06, + "loss": 0.6996, + "step": 951 + }, + { + "epoch": 0.7885146327995582, + "grad_norm": 0.913536787033081, + "learning_rate": 4.826475807002125e-06, + "loss": 0.6844, + "step": 952 + }, + { + "epoch": 0.7893429044726671, + "grad_norm": 0.9202503561973572, + "learning_rate": 4.826073027045952e-06, + "loss": 0.6702, + "step": 953 + }, + { + "epoch": 0.7901711761457758, + "grad_norm": 0.8800237774848938, + "learning_rate": 4.8256697970163315e-06, + "loss": 0.685, + "step": 954 + }, + { + "epoch": 0.7909994478188846, + "grad_norm": 0.8923048377037048, + "learning_rate": 4.825266116991285e-06, + "loss": 0.6839, + "step": 955 + }, + { + "epoch": 0.7918277194919934, + "grad_norm": 0.879743754863739, + "learning_rate": 4.824861987048922e-06, + "loss": 0.7112, + "step": 956 + }, + { + "epoch": 0.7926559911651021, + "grad_norm": 0.8575169444084167, + "learning_rate": 4.824457407267436e-06, + "loss": 0.671, + "step": 957 + }, + { + "epoch": 0.7934842628382109, + "grad_norm": 0.8851709961891174, + "learning_rate": 4.824052377725109e-06, + "loss": 0.669, + "step": 958 + }, + { + "epoch": 0.7943125345113197, + "grad_norm": 0.9084329605102539, + "learning_rate": 4.823646898500311e-06, + "loss": 0.6731, + "step": 959 + }, + { + "epoch": 0.7951408061844285, + "grad_norm": 0.908256471157074, + "learning_rate": 4.823240969671501e-06, + "loss": 0.688, + "step": 960 + }, + { + "epoch": 0.7959690778575372, + "grad_norm": 0.8993384838104248, + "learning_rate": 4.822834591317219e-06, + "loss": 0.6736, + "step": 961 + }, + { + "epoch": 0.7967973495306461, + "grad_norm": 0.893530011177063, + "learning_rate": 4.822427763516098e-06, + "loss": 0.6787, + "step": 962 + }, + { + "epoch": 0.7976256212037548, + "grad_norm": 0.9479735493659973, + "learning_rate": 4.822020486346852e-06, + "loss": 0.6842, + "step": 963 + }, + { + "epoch": 0.7984538928768636, + "grad_norm": 0.8716689348220825, + "learning_rate": 4.821612759888289e-06, + "loss": 0.6726, + "step": 964 + }, + { + "epoch": 0.7992821645499724, + "grad_norm": 0.8902738690376282, + "learning_rate": 4.821204584219299e-06, + "loss": 0.6672, + "step": 965 + }, + { + "epoch": 0.8001104362230812, + "grad_norm": 0.8713969588279724, + "learning_rate": 4.820795959418859e-06, + "loss": 0.6783, + "step": 966 + }, + { + "epoch": 0.8009387078961899, + "grad_norm": 0.891788899898529, + "learning_rate": 4.820386885566036e-06, + "loss": 0.6691, + "step": 967 + }, + { + "epoch": 0.8017669795692988, + "grad_norm": 0.8484700918197632, + "learning_rate": 4.81997736273998e-06, + "loss": 0.6766, + "step": 968 + }, + { + "epoch": 0.8025952512424075, + "grad_norm": 0.8931223154067993, + "learning_rate": 4.819567391019931e-06, + "loss": 0.6944, + "step": 969 + }, + { + "epoch": 0.8034235229155163, + "grad_norm": 0.9707916975021362, + "learning_rate": 4.819156970485216e-06, + "loss": 0.7076, + "step": 970 + }, + { + "epoch": 0.8042517945886251, + "grad_norm": 0.9172869920730591, + "learning_rate": 4.818746101215245e-06, + "loss": 0.6864, + "step": 971 + }, + { + "epoch": 0.8050800662617339, + "grad_norm": 0.8982006311416626, + "learning_rate": 4.818334783289518e-06, + "loss": 0.6685, + "step": 972 + }, + { + "epoch": 0.8059083379348426, + "grad_norm": 0.8960995078086853, + "learning_rate": 4.817923016787623e-06, + "loss": 0.6959, + "step": 973 + }, + { + "epoch": 0.8067366096079515, + "grad_norm": 0.8806949257850647, + "learning_rate": 4.81751080178923e-06, + "loss": 0.7028, + "step": 974 + }, + { + "epoch": 0.8075648812810602, + "grad_norm": 0.8798978924751282, + "learning_rate": 4.817098138374102e-06, + "loss": 0.6718, + "step": 975 + }, + { + "epoch": 0.8083931529541689, + "grad_norm": 0.9217897653579712, + "learning_rate": 4.8166850266220835e-06, + "loss": 0.6782, + "step": 976 + }, + { + "epoch": 0.8092214246272778, + "grad_norm": 0.8915667533874512, + "learning_rate": 4.8162714666131074e-06, + "loss": 0.6917, + "step": 977 + }, + { + "epoch": 0.8100496963003865, + "grad_norm": 0.8895093202590942, + "learning_rate": 4.815857458427195e-06, + "loss": 0.698, + "step": 978 + }, + { + "epoch": 0.8108779679734953, + "grad_norm": 0.9167448282241821, + "learning_rate": 4.815443002144453e-06, + "loss": 0.6796, + "step": 979 + }, + { + "epoch": 0.811706239646604, + "grad_norm": 0.913827657699585, + "learning_rate": 4.815028097845074e-06, + "loss": 0.6731, + "step": 980 + }, + { + "epoch": 0.8125345113197129, + "grad_norm": 0.9183399081230164, + "learning_rate": 4.814612745609338e-06, + "loss": 0.7076, + "step": 981 + }, + { + "epoch": 0.8133627829928216, + "grad_norm": 0.8812996745109558, + "learning_rate": 4.814196945517613e-06, + "loss": 0.6416, + "step": 982 + }, + { + "epoch": 0.8141910546659304, + "grad_norm": 0.9032955169677734, + "learning_rate": 4.813780697650351e-06, + "loss": 0.6799, + "step": 983 + }, + { + "epoch": 0.8150193263390392, + "grad_norm": 0.9380100965499878, + "learning_rate": 4.813364002088093e-06, + "loss": 0.6771, + "step": 984 + }, + { + "epoch": 0.815847598012148, + "grad_norm": 0.9470316171646118, + "learning_rate": 4.8129468589114655e-06, + "loss": 0.6851, + "step": 985 + }, + { + "epoch": 0.8166758696852567, + "grad_norm": 0.9401133060455322, + "learning_rate": 4.812529268201183e-06, + "loss": 0.676, + "step": 986 + }, + { + "epoch": 0.8175041413583656, + "grad_norm": 0.9379092454910278, + "learning_rate": 4.812111230038043e-06, + "loss": 0.672, + "step": 987 + }, + { + "epoch": 0.8183324130314743, + "grad_norm": 0.9591637253761292, + "learning_rate": 4.811692744502933e-06, + "loss": 0.6684, + "step": 988 + }, + { + "epoch": 0.8191606847045831, + "grad_norm": 0.8911624550819397, + "learning_rate": 4.8112738116768276e-06, + "loss": 0.6448, + "step": 989 + }, + { + "epoch": 0.8199889563776919, + "grad_norm": 0.9036282896995544, + "learning_rate": 4.810854431640784e-06, + "loss": 0.703, + "step": 990 + }, + { + "epoch": 0.8208172280508007, + "grad_norm": 0.9197635650634766, + "learning_rate": 4.81043460447595e-06, + "loss": 0.6655, + "step": 991 + }, + { + "epoch": 0.8216454997239094, + "grad_norm": 0.9548643827438354, + "learning_rate": 4.8100143302635575e-06, + "loss": 0.6892, + "step": 992 + }, + { + "epoch": 0.8224737713970183, + "grad_norm": 0.9139766097068787, + "learning_rate": 4.809593609084925e-06, + "loss": 0.6723, + "step": 993 + }, + { + "epoch": 0.823302043070127, + "grad_norm": 0.9067809581756592, + "learning_rate": 4.8091724410214595e-06, + "loss": 0.6649, + "step": 994 + }, + { + "epoch": 0.8241303147432358, + "grad_norm": 0.9705583453178406, + "learning_rate": 4.808750826154652e-06, + "loss": 0.6889, + "step": 995 + }, + { + "epoch": 0.8249585864163446, + "grad_norm": 0.9664806127548218, + "learning_rate": 4.808328764566082e-06, + "loss": 0.6758, + "step": 996 + }, + { + "epoch": 0.8257868580894533, + "grad_norm": 0.8840530514717102, + "learning_rate": 4.807906256337414e-06, + "loss": 0.6719, + "step": 997 + }, + { + "epoch": 0.8266151297625621, + "grad_norm": 0.8803462386131287, + "learning_rate": 4.8074833015504e-06, + "loss": 0.6903, + "step": 998 + }, + { + "epoch": 0.8274434014356709, + "grad_norm": 0.9404571652412415, + "learning_rate": 4.8070599002868756e-06, + "loss": 0.6917, + "step": 999 + }, + { + "epoch": 0.8282716731087797, + "grad_norm": 1.0206997394561768, + "learning_rate": 4.806636052628768e-06, + "loss": 0.6854, + "step": 1000 + }, + { + "epoch": 0.8290999447818884, + "grad_norm": 0.9294790029525757, + "learning_rate": 4.806211758658086e-06, + "loss": 0.6791, + "step": 1001 + }, + { + "epoch": 0.8299282164549973, + "grad_norm": 0.874011754989624, + "learning_rate": 4.805787018456927e-06, + "loss": 0.6915, + "step": 1002 + }, + { + "epoch": 0.830756488128106, + "grad_norm": 0.9098061323165894, + "learning_rate": 4.805361832107474e-06, + "loss": 0.6643, + "step": 1003 + }, + { + "epoch": 0.8315847598012148, + "grad_norm": 0.9783986210823059, + "learning_rate": 4.804936199691997e-06, + "loss": 0.6853, + "step": 1004 + }, + { + "epoch": 0.8324130314743236, + "grad_norm": 0.9316959381103516, + "learning_rate": 4.804510121292852e-06, + "loss": 0.6696, + "step": 1005 + }, + { + "epoch": 0.8332413031474324, + "grad_norm": 0.8930666446685791, + "learning_rate": 4.80408359699248e-06, + "loss": 0.6805, + "step": 1006 + }, + { + "epoch": 0.8340695748205411, + "grad_norm": 0.9417555332183838, + "learning_rate": 4.803656626873412e-06, + "loss": 0.66, + "step": 1007 + }, + { + "epoch": 0.83489784649365, + "grad_norm": 0.9435075521469116, + "learning_rate": 4.80322921101826e-06, + "loss": 0.6664, + "step": 1008 + }, + { + "epoch": 0.8357261181667587, + "grad_norm": 0.9090076684951782, + "learning_rate": 4.8028013495097255e-06, + "loss": 0.6595, + "step": 1009 + }, + { + "epoch": 0.8365543898398675, + "grad_norm": 0.9024788737297058, + "learning_rate": 4.802373042430597e-06, + "loss": 0.6672, + "step": 1010 + }, + { + "epoch": 0.8373826615129762, + "grad_norm": 0.899813711643219, + "learning_rate": 4.801944289863747e-06, + "loss": 0.6743, + "step": 1011 + }, + { + "epoch": 0.8382109331860851, + "grad_norm": 0.887106716632843, + "learning_rate": 4.801515091892136e-06, + "loss": 0.6823, + "step": 1012 + }, + { + "epoch": 0.8390392048591938, + "grad_norm": 0.8690978288650513, + "learning_rate": 4.801085448598809e-06, + "loss": 0.6642, + "step": 1013 + }, + { + "epoch": 0.8398674765323026, + "grad_norm": 0.9131479859352112, + "learning_rate": 4.800655360066898e-06, + "loss": 0.6841, + "step": 1014 + }, + { + "epoch": 0.8406957482054114, + "grad_norm": 0.9031343460083008, + "learning_rate": 4.800224826379622e-06, + "loss": 0.6879, + "step": 1015 + }, + { + "epoch": 0.8415240198785201, + "grad_norm": 0.8914244174957275, + "learning_rate": 4.799793847620283e-06, + "loss": 0.6562, + "step": 1016 + }, + { + "epoch": 0.8423522915516289, + "grad_norm": 0.8692408800125122, + "learning_rate": 4.799362423872274e-06, + "loss": 0.6759, + "step": 1017 + }, + { + "epoch": 0.8431805632247377, + "grad_norm": 0.9511806964874268, + "learning_rate": 4.798930555219069e-06, + "loss": 0.6637, + "step": 1018 + }, + { + "epoch": 0.8440088348978465, + "grad_norm": 0.9556595683097839, + "learning_rate": 4.7984982417442325e-06, + "loss": 0.6925, + "step": 1019 + }, + { + "epoch": 0.8448371065709552, + "grad_norm": 0.9107236862182617, + "learning_rate": 4.798065483531413e-06, + "loss": 0.67, + "step": 1020 + }, + { + "epoch": 0.8456653782440641, + "grad_norm": 0.875293493270874, + "learning_rate": 4.797632280664345e-06, + "loss": 0.6755, + "step": 1021 + }, + { + "epoch": 0.8464936499171728, + "grad_norm": 0.8972271084785461, + "learning_rate": 4.797198633226849e-06, + "loss": 0.6637, + "step": 1022 + }, + { + "epoch": 0.8473219215902816, + "grad_norm": 0.8903595805168152, + "learning_rate": 4.796764541302831e-06, + "loss": 0.6997, + "step": 1023 + }, + { + "epoch": 0.8481501932633904, + "grad_norm": 0.8948636651039124, + "learning_rate": 4.796330004976285e-06, + "loss": 0.6579, + "step": 1024 + }, + { + "epoch": 0.8489784649364992, + "grad_norm": 0.8590925335884094, + "learning_rate": 4.795895024331289e-06, + "loss": 0.6651, + "step": 1025 + }, + { + "epoch": 0.8498067366096079, + "grad_norm": 0.870811402797699, + "learning_rate": 4.795459599452007e-06, + "loss": 0.6707, + "step": 1026 + }, + { + "epoch": 0.8506350082827168, + "grad_norm": 0.8807471394538879, + "learning_rate": 4.795023730422692e-06, + "loss": 0.6787, + "step": 1027 + }, + { + "epoch": 0.8514632799558255, + "grad_norm": 0.8817671537399292, + "learning_rate": 4.794587417327678e-06, + "loss": 0.66, + "step": 1028 + }, + { + "epoch": 0.8522915516289343, + "grad_norm": 0.8924570679664612, + "learning_rate": 4.794150660251389e-06, + "loss": 0.6842, + "step": 1029 + }, + { + "epoch": 0.8531198233020431, + "grad_norm": 0.8746559619903564, + "learning_rate": 4.793713459278333e-06, + "loss": 0.6582, + "step": 1030 + }, + { + "epoch": 0.8539480949751519, + "grad_norm": 0.8885277509689331, + "learning_rate": 4.793275814493104e-06, + "loss": 0.6784, + "step": 1031 + }, + { + "epoch": 0.8547763666482606, + "grad_norm": 0.8797710537910461, + "learning_rate": 4.792837725980383e-06, + "loss": 0.6638, + "step": 1032 + }, + { + "epoch": 0.8556046383213695, + "grad_norm": 0.8932235240936279, + "learning_rate": 4.7923991938249355e-06, + "loss": 0.6918, + "step": 1033 + }, + { + "epoch": 0.8564329099944782, + "grad_norm": 0.9317487478256226, + "learning_rate": 4.791960218111613e-06, + "loss": 0.6905, + "step": 1034 + }, + { + "epoch": 0.8572611816675869, + "grad_norm": 0.8760019540786743, + "learning_rate": 4.791520798925354e-06, + "loss": 0.673, + "step": 1035 + }, + { + "epoch": 0.8580894533406958, + "grad_norm": 0.8644204139709473, + "learning_rate": 4.7910809363511825e-06, + "loss": 0.6755, + "step": 1036 + }, + { + "epoch": 0.8589177250138045, + "grad_norm": 0.8761067390441895, + "learning_rate": 4.790640630474206e-06, + "loss": 0.699, + "step": 1037 + }, + { + "epoch": 0.8597459966869133, + "grad_norm": 0.9135014414787292, + "learning_rate": 4.790199881379621e-06, + "loss": 0.6659, + "step": 1038 + }, + { + "epoch": 0.860574268360022, + "grad_norm": 0.9065434336662292, + "learning_rate": 4.789758689152708e-06, + "loss": 0.6664, + "step": 1039 + }, + { + "epoch": 0.8614025400331309, + "grad_norm": 0.8814099431037903, + "learning_rate": 4.789317053878833e-06, + "loss": 0.681, + "step": 1040 + }, + { + "epoch": 0.8622308117062396, + "grad_norm": 0.872502863407135, + "learning_rate": 4.788874975643449e-06, + "loss": 0.7044, + "step": 1041 + }, + { + "epoch": 0.8630590833793484, + "grad_norm": 0.9532367587089539, + "learning_rate": 4.788432454532093e-06, + "loss": 0.655, + "step": 1042 + }, + { + "epoch": 0.8638873550524572, + "grad_norm": 0.9009373784065247, + "learning_rate": 4.787989490630391e-06, + "loss": 0.6946, + "step": 1043 + }, + { + "epoch": 0.864715626725566, + "grad_norm": 0.889933168888092, + "learning_rate": 4.787546084024051e-06, + "loss": 0.6595, + "step": 1044 + }, + { + "epoch": 0.8655438983986747, + "grad_norm": 0.9179873466491699, + "learning_rate": 4.787102234798867e-06, + "loss": 0.658, + "step": 1045 + }, + { + "epoch": 0.8663721700717836, + "grad_norm": 0.8873292207717896, + "learning_rate": 4.786657943040721e-06, + "loss": 0.7023, + "step": 1046 + }, + { + "epoch": 0.8672004417448923, + "grad_norm": 0.8388610482215881, + "learning_rate": 4.78621320883558e-06, + "loss": 0.6644, + "step": 1047 + }, + { + "epoch": 0.8680287134180011, + "grad_norm": 0.8971349596977234, + "learning_rate": 4.785768032269493e-06, + "loss": 0.6552, + "step": 1048 + }, + { + "epoch": 0.8688569850911099, + "grad_norm": 0.8789131045341492, + "learning_rate": 4.785322413428601e-06, + "loss": 0.6681, + "step": 1049 + }, + { + "epoch": 0.8696852567642187, + "grad_norm": 0.8867958784103394, + "learning_rate": 4.784876352399125e-06, + "loss": 0.6859, + "step": 1050 + }, + { + "epoch": 0.8705135284373274, + "grad_norm": 0.8584237694740295, + "learning_rate": 4.784429849267373e-06, + "loss": 0.6655, + "step": 1051 + }, + { + "epoch": 0.8713418001104363, + "grad_norm": 0.9230811595916748, + "learning_rate": 4.783982904119743e-06, + "loss": 0.6648, + "step": 1052 + }, + { + "epoch": 0.872170071783545, + "grad_norm": 0.9217789769172668, + "learning_rate": 4.783535517042709e-06, + "loss": 0.6867, + "step": 1053 + }, + { + "epoch": 0.8729983434566538, + "grad_norm": 0.9209116697311401, + "learning_rate": 4.783087688122842e-06, + "loss": 0.6972, + "step": 1054 + }, + { + "epoch": 0.8738266151297626, + "grad_norm": 0.9493632316589355, + "learning_rate": 4.782639417446789e-06, + "loss": 0.6785, + "step": 1055 + }, + { + "epoch": 0.8746548868028713, + "grad_norm": 0.8790786862373352, + "learning_rate": 4.782190705101286e-06, + "loss": 0.6925, + "step": 1056 + }, + { + "epoch": 0.8754831584759801, + "grad_norm": 0.8662691712379456, + "learning_rate": 4.781741551173157e-06, + "loss": 0.6686, + "step": 1057 + }, + { + "epoch": 0.8763114301490889, + "grad_norm": 0.919658899307251, + "learning_rate": 4.781291955749307e-06, + "loss": 0.6814, + "step": 1058 + }, + { + "epoch": 0.8771397018221977, + "grad_norm": 0.9639708399772644, + "learning_rate": 4.780841918916731e-06, + "loss": 0.6629, + "step": 1059 + }, + { + "epoch": 0.8779679734953064, + "grad_norm": 0.8694306015968323, + "learning_rate": 4.780391440762505e-06, + "loss": 0.6893, + "step": 1060 + }, + { + "epoch": 0.8787962451684153, + "grad_norm": 0.9725603461265564, + "learning_rate": 4.779940521373793e-06, + "loss": 0.67, + "step": 1061 + }, + { + "epoch": 0.879624516841524, + "grad_norm": 0.93159419298172, + "learning_rate": 4.779489160837842e-06, + "loss": 0.686, + "step": 1062 + }, + { + "epoch": 0.8804527885146328, + "grad_norm": 0.9366991519927979, + "learning_rate": 4.779037359241989e-06, + "loss": 0.649, + "step": 1063 + }, + { + "epoch": 0.8812810601877415, + "grad_norm": 0.9052289724349976, + "learning_rate": 4.7785851166736505e-06, + "loss": 0.6677, + "step": 1064 + }, + { + "epoch": 0.8821093318608504, + "grad_norm": 0.882251501083374, + "learning_rate": 4.778132433220334e-06, + "loss": 0.6886, + "step": 1065 + }, + { + "epoch": 0.8829376035339591, + "grad_norm": 0.8867698907852173, + "learning_rate": 4.777679308969627e-06, + "loss": 0.6645, + "step": 1066 + }, + { + "epoch": 0.883765875207068, + "grad_norm": 0.890396773815155, + "learning_rate": 4.777225744009207e-06, + "loss": 0.6684, + "step": 1067 + }, + { + "epoch": 0.8845941468801767, + "grad_norm": 0.8827376365661621, + "learning_rate": 4.776771738426833e-06, + "loss": 0.6738, + "step": 1068 + }, + { + "epoch": 0.8854224185532855, + "grad_norm": 0.8953343033790588, + "learning_rate": 4.776317292310352e-06, + "loss": 0.684, + "step": 1069 + }, + { + "epoch": 0.8862506902263942, + "grad_norm": 0.8751050233840942, + "learning_rate": 4.775862405747694e-06, + "loss": 0.6779, + "step": 1070 + }, + { + "epoch": 0.8870789618995031, + "grad_norm": 0.9136618971824646, + "learning_rate": 4.775407078826876e-06, + "loss": 0.6939, + "step": 1071 + }, + { + "epoch": 0.8879072335726118, + "grad_norm": 0.8887941837310791, + "learning_rate": 4.774951311635999e-06, + "loss": 0.6547, + "step": 1072 + }, + { + "epoch": 0.8887355052457206, + "grad_norm": 0.8873274326324463, + "learning_rate": 4.77449510426325e-06, + "loss": 0.6769, + "step": 1073 + }, + { + "epoch": 0.8895637769188294, + "grad_norm": 0.9307074546813965, + "learning_rate": 4.774038456796901e-06, + "loss": 0.6984, + "step": 1074 + }, + { + "epoch": 0.8903920485919381, + "grad_norm": 0.9022747874259949, + "learning_rate": 4.77358136932531e-06, + "loss": 0.6733, + "step": 1075 + }, + { + "epoch": 0.8912203202650469, + "grad_norm": 0.8779094219207764, + "learning_rate": 4.773123841936916e-06, + "loss": 0.6434, + "step": 1076 + }, + { + "epoch": 0.8920485919381557, + "grad_norm": 0.9465747475624084, + "learning_rate": 4.7726658747202505e-06, + "loss": 0.6761, + "step": 1077 + }, + { + "epoch": 0.8928768636112645, + "grad_norm": 0.8694362044334412, + "learning_rate": 4.772207467763923e-06, + "loss": 0.6756, + "step": 1078 + }, + { + "epoch": 0.8937051352843732, + "grad_norm": 0.8839150071144104, + "learning_rate": 4.771748621156631e-06, + "loss": 0.6714, + "step": 1079 + }, + { + "epoch": 0.8945334069574821, + "grad_norm": 0.9220975041389465, + "learning_rate": 4.7712893349871585e-06, + "loss": 0.6724, + "step": 1080 + }, + { + "epoch": 0.8953616786305908, + "grad_norm": 0.9234760999679565, + "learning_rate": 4.770829609344372e-06, + "loss": 0.67, + "step": 1081 + }, + { + "epoch": 0.8961899503036996, + "grad_norm": 0.9262274503707886, + "learning_rate": 4.770369444317225e-06, + "loss": 0.6619, + "step": 1082 + }, + { + "epoch": 0.8970182219768084, + "grad_norm": 0.8889190554618835, + "learning_rate": 4.769908839994753e-06, + "loss": 0.688, + "step": 1083 + }, + { + "epoch": 0.8978464936499172, + "grad_norm": 0.9028982520103455, + "learning_rate": 4.769447796466082e-06, + "loss": 0.6633, + "step": 1084 + }, + { + "epoch": 0.8986747653230259, + "grad_norm": 0.9155063033103943, + "learning_rate": 4.7689863138204185e-06, + "loss": 0.6709, + "step": 1085 + }, + { + "epoch": 0.8995030369961348, + "grad_norm": 0.9066855311393738, + "learning_rate": 4.768524392147052e-06, + "loss": 0.6713, + "step": 1086 + }, + { + "epoch": 0.9003313086692435, + "grad_norm": 0.9365159273147583, + "learning_rate": 4.768062031535365e-06, + "loss": 0.6675, + "step": 1087 + }, + { + "epoch": 0.9011595803423523, + "grad_norm": 0.9106162190437317, + "learning_rate": 4.767599232074816e-06, + "loss": 0.6804, + "step": 1088 + }, + { + "epoch": 0.9019878520154611, + "grad_norm": 0.9183253645896912, + "learning_rate": 4.767135993854955e-06, + "loss": 0.658, + "step": 1089 + }, + { + "epoch": 0.9028161236885699, + "grad_norm": 0.9543409943580627, + "learning_rate": 4.766672316965413e-06, + "loss": 0.696, + "step": 1090 + }, + { + "epoch": 0.9036443953616786, + "grad_norm": 0.8894531726837158, + "learning_rate": 4.766208201495906e-06, + "loss": 0.6781, + "step": 1091 + }, + { + "epoch": 0.9044726670347875, + "grad_norm": 0.915542721748352, + "learning_rate": 4.765743647536238e-06, + "loss": 0.6804, + "step": 1092 + }, + { + "epoch": 0.9053009387078962, + "grad_norm": 0.8959763646125793, + "learning_rate": 4.765278655176296e-06, + "loss": 0.6873, + "step": 1093 + }, + { + "epoch": 0.906129210381005, + "grad_norm": 0.881188690662384, + "learning_rate": 4.76481322450605e-06, + "loss": 0.6582, + "step": 1094 + }, + { + "epoch": 0.9069574820541138, + "grad_norm": 0.9096754193305969, + "learning_rate": 4.764347355615558e-06, + "loss": 0.693, + "step": 1095 + }, + { + "epoch": 0.9077857537272225, + "grad_norm": 0.891132652759552, + "learning_rate": 4.76388104859496e-06, + "loss": 0.6781, + "step": 1096 + }, + { + "epoch": 0.9086140254003313, + "grad_norm": 0.893624484539032, + "learning_rate": 4.763414303534484e-06, + "loss": 0.6636, + "step": 1097 + }, + { + "epoch": 0.90944229707344, + "grad_norm": 0.8795678019523621, + "learning_rate": 4.762947120524438e-06, + "loss": 0.6604, + "step": 1098 + }, + { + "epoch": 0.9102705687465489, + "grad_norm": 0.9177265763282776, + "learning_rate": 4.76247949965522e-06, + "loss": 0.6557, + "step": 1099 + }, + { + "epoch": 0.9110988404196576, + "grad_norm": 0.9416985511779785, + "learning_rate": 4.762011441017309e-06, + "loss": 0.7008, + "step": 1100 + }, + { + "epoch": 0.9119271120927664, + "grad_norm": 0.9255479574203491, + "learning_rate": 4.7615429447012704e-06, + "loss": 0.6665, + "step": 1101 + }, + { + "epoch": 0.9127553837658752, + "grad_norm": 0.8948953151702881, + "learning_rate": 4.761074010797754e-06, + "loss": 0.6699, + "step": 1102 + }, + { + "epoch": 0.913583655438984, + "grad_norm": 0.9027442336082458, + "learning_rate": 4.760604639397493e-06, + "loss": 0.6649, + "step": 1103 + }, + { + "epoch": 0.9144119271120927, + "grad_norm": 0.9094238877296448, + "learning_rate": 4.7601348305913085e-06, + "loss": 0.6753, + "step": 1104 + }, + { + "epoch": 0.9152401987852016, + "grad_norm": 0.9098204374313354, + "learning_rate": 4.759664584470103e-06, + "loss": 0.6845, + "step": 1105 + }, + { + "epoch": 0.9160684704583103, + "grad_norm": 0.9339573383331299, + "learning_rate": 4.759193901124864e-06, + "loss": 0.6285, + "step": 1106 + }, + { + "epoch": 0.9168967421314191, + "grad_norm": 0.8812451958656311, + "learning_rate": 4.758722780646666e-06, + "loss": 0.6764, + "step": 1107 + }, + { + "epoch": 0.9177250138045279, + "grad_norm": 0.8832173347473145, + "learning_rate": 4.758251223126665e-06, + "loss": 0.6908, + "step": 1108 + }, + { + "epoch": 0.9185532854776367, + "grad_norm": 0.9124787449836731, + "learning_rate": 4.757779228656103e-06, + "loss": 0.6855, + "step": 1109 + }, + { + "epoch": 0.9193815571507454, + "grad_norm": 0.8778328895568848, + "learning_rate": 4.757306797326309e-06, + "loss": 0.6404, + "step": 1110 + }, + { + "epoch": 0.9202098288238543, + "grad_norm": 0.8831816911697388, + "learning_rate": 4.75683392922869e-06, + "loss": 0.6425, + "step": 1111 + }, + { + "epoch": 0.921038100496963, + "grad_norm": 0.9060006141662598, + "learning_rate": 4.756360624454744e-06, + "loss": 0.6738, + "step": 1112 + }, + { + "epoch": 0.9218663721700718, + "grad_norm": 0.8746777176856995, + "learning_rate": 4.755886883096051e-06, + "loss": 0.6665, + "step": 1113 + }, + { + "epoch": 0.9226946438431806, + "grad_norm": 0.9185945987701416, + "learning_rate": 4.755412705244274e-06, + "loss": 0.6564, + "step": 1114 + }, + { + "epoch": 0.9235229155162893, + "grad_norm": 0.9648581743240356, + "learning_rate": 4.754938090991165e-06, + "loss": 0.6597, + "step": 1115 + }, + { + "epoch": 0.9243511871893981, + "grad_norm": 0.8800092935562134, + "learning_rate": 4.754463040428555e-06, + "loss": 0.6884, + "step": 1116 + }, + { + "epoch": 0.9251794588625069, + "grad_norm": 0.8659892082214355, + "learning_rate": 4.753987553648363e-06, + "loss": 0.68, + "step": 1117 + }, + { + "epoch": 0.9260077305356157, + "grad_norm": 0.88991379737854, + "learning_rate": 4.75351163074259e-06, + "loss": 0.668, + "step": 1118 + }, + { + "epoch": 0.9268360022087244, + "grad_norm": 0.9683054089546204, + "learning_rate": 4.753035271803323e-06, + "loss": 0.6789, + "step": 1119 + }, + { + "epoch": 0.9276642738818333, + "grad_norm": 0.928001880645752, + "learning_rate": 4.752558476922735e-06, + "loss": 0.6786, + "step": 1120 + }, + { + "epoch": 0.928492545554942, + "grad_norm": 0.8760973215103149, + "learning_rate": 4.752081246193078e-06, + "loss": 0.6716, + "step": 1121 + }, + { + "epoch": 0.9293208172280508, + "grad_norm": 0.9171329140663147, + "learning_rate": 4.751603579706695e-06, + "loss": 0.662, + "step": 1122 + }, + { + "epoch": 0.9301490889011595, + "grad_norm": 0.9380231499671936, + "learning_rate": 4.751125477556008e-06, + "loss": 0.6836, + "step": 1123 + }, + { + "epoch": 0.9309773605742684, + "grad_norm": 0.9471802711486816, + "learning_rate": 4.750646939833526e-06, + "loss": 0.668, + "step": 1124 + }, + { + "epoch": 0.9318056322473771, + "grad_norm": 0.8860809206962585, + "learning_rate": 4.750167966631842e-06, + "loss": 0.6635, + "step": 1125 + }, + { + "epoch": 0.932633903920486, + "grad_norm": 0.8694644570350647, + "learning_rate": 4.7496885580436305e-06, + "loss": 0.6505, + "step": 1126 + }, + { + "epoch": 0.9334621755935947, + "grad_norm": 0.9602686166763306, + "learning_rate": 4.749208714161655e-06, + "loss": 0.6874, + "step": 1127 + }, + { + "epoch": 0.9342904472667035, + "grad_norm": 0.9198856353759766, + "learning_rate": 4.748728435078761e-06, + "loss": 0.697, + "step": 1128 + }, + { + "epoch": 0.9351187189398122, + "grad_norm": 0.9162499904632568, + "learning_rate": 4.748247720887876e-06, + "loss": 0.6521, + "step": 1129 + }, + { + "epoch": 0.9359469906129211, + "grad_norm": 0.9064576029777527, + "learning_rate": 4.747766571682016e-06, + "loss": 0.6525, + "step": 1130 + }, + { + "epoch": 0.9367752622860298, + "grad_norm": 0.9518057703971863, + "learning_rate": 4.747284987554277e-06, + "loss": 0.6745, + "step": 1131 + }, + { + "epoch": 0.9376035339591386, + "grad_norm": 0.8963000178337097, + "learning_rate": 4.746802968597842e-06, + "loss": 0.6999, + "step": 1132 + }, + { + "epoch": 0.9384318056322474, + "grad_norm": 0.8687212467193604, + "learning_rate": 4.746320514905977e-06, + "loss": 0.6353, + "step": 1133 + }, + { + "epoch": 0.9392600773053562, + "grad_norm": 0.8598489165306091, + "learning_rate": 4.745837626572033e-06, + "loss": 0.6897, + "step": 1134 + }, + { + "epoch": 0.9400883489784649, + "grad_norm": 0.9425498843193054, + "learning_rate": 4.745354303689444e-06, + "loss": 0.667, + "step": 1135 + }, + { + "epoch": 0.9409166206515737, + "grad_norm": 0.9026349782943726, + "learning_rate": 4.744870546351727e-06, + "loss": 0.676, + "step": 1136 + }, + { + "epoch": 0.9417448923246825, + "grad_norm": 0.8824150562286377, + "learning_rate": 4.744386354652486e-06, + "loss": 0.6941, + "step": 1137 + }, + { + "epoch": 0.9425731639977912, + "grad_norm": 0.8910468816757202, + "learning_rate": 4.743901728685406e-06, + "loss": 0.6701, + "step": 1138 + }, + { + "epoch": 0.9434014356709001, + "grad_norm": 0.9114013314247131, + "learning_rate": 4.74341666854426e-06, + "loss": 0.6735, + "step": 1139 + }, + { + "epoch": 0.9442297073440088, + "grad_norm": 0.9107307195663452, + "learning_rate": 4.742931174322902e-06, + "loss": 0.6505, + "step": 1140 + }, + { + "epoch": 0.9450579790171176, + "grad_norm": 0.8906923532485962, + "learning_rate": 4.742445246115271e-06, + "loss": 0.6648, + "step": 1141 + }, + { + "epoch": 0.9458862506902264, + "grad_norm": 0.902644157409668, + "learning_rate": 4.741958884015387e-06, + "loss": 0.6674, + "step": 1142 + }, + { + "epoch": 0.9467145223633352, + "grad_norm": 0.9522087574005127, + "learning_rate": 4.741472088117358e-06, + "loss": 0.6825, + "step": 1143 + }, + { + "epoch": 0.9475427940364439, + "grad_norm": 0.9152255654335022, + "learning_rate": 4.740984858515376e-06, + "loss": 0.6708, + "step": 1144 + }, + { + "epoch": 0.9483710657095528, + "grad_norm": 0.8779835104942322, + "learning_rate": 4.740497195303713e-06, + "loss": 0.6562, + "step": 1145 + }, + { + "epoch": 0.9491993373826615, + "grad_norm": 0.8842780590057373, + "learning_rate": 4.740009098576729e-06, + "loss": 0.6769, + "step": 1146 + }, + { + "epoch": 0.9500276090557703, + "grad_norm": 0.9137415289878845, + "learning_rate": 4.739520568428866e-06, + "loss": 0.6705, + "step": 1147 + }, + { + "epoch": 0.950855880728879, + "grad_norm": 0.8867765665054321, + "learning_rate": 4.739031604954648e-06, + "loss": 0.659, + "step": 1148 + }, + { + "epoch": 0.9516841524019879, + "grad_norm": 0.8965260982513428, + "learning_rate": 4.7385422082486874e-06, + "loss": 0.6744, + "step": 1149 + }, + { + "epoch": 0.9525124240750966, + "grad_norm": 0.8861697912216187, + "learning_rate": 4.738052378405677e-06, + "loss": 0.6575, + "step": 1150 + }, + { + "epoch": 0.9533406957482055, + "grad_norm": 0.9442310333251953, + "learning_rate": 4.737562115520394e-06, + "loss": 0.7164, + "step": 1151 + }, + { + "epoch": 0.9541689674213142, + "grad_norm": 0.9340727925300598, + "learning_rate": 4.7370714196877e-06, + "loss": 0.6577, + "step": 1152 + }, + { + "epoch": 0.954997239094423, + "grad_norm": 0.9337487816810608, + "learning_rate": 4.7365802910025405e-06, + "loss": 0.6664, + "step": 1153 + }, + { + "epoch": 0.9558255107675317, + "grad_norm": 0.8849001526832581, + "learning_rate": 4.736088729559943e-06, + "loss": 0.6721, + "step": 1154 + }, + { + "epoch": 0.9566537824406405, + "grad_norm": 0.917323648929596, + "learning_rate": 4.735596735455022e-06, + "loss": 0.6715, + "step": 1155 + }, + { + "epoch": 0.9574820541137493, + "grad_norm": 0.9480959177017212, + "learning_rate": 4.735104308782972e-06, + "loss": 0.6784, + "step": 1156 + }, + { + "epoch": 0.958310325786858, + "grad_norm": 0.9519251585006714, + "learning_rate": 4.734611449639073e-06, + "loss": 0.6617, + "step": 1157 + }, + { + "epoch": 0.9591385974599669, + "grad_norm": 0.8652079105377197, + "learning_rate": 4.734118158118689e-06, + "loss": 0.6846, + "step": 1158 + }, + { + "epoch": 0.9599668691330756, + "grad_norm": 0.8860392570495605, + "learning_rate": 4.733624434317269e-06, + "loss": 0.6729, + "step": 1159 + }, + { + "epoch": 0.9607951408061844, + "grad_norm": 0.9467882513999939, + "learning_rate": 4.7331302783303416e-06, + "loss": 0.6362, + "step": 1160 + }, + { + "epoch": 0.9616234124792932, + "grad_norm": 0.9138529896736145, + "learning_rate": 4.732635690253523e-06, + "loss": 0.6613, + "step": 1161 + }, + { + "epoch": 0.962451684152402, + "grad_norm": 0.9401054978370667, + "learning_rate": 4.7321406701825105e-06, + "loss": 0.6845, + "step": 1162 + }, + { + "epoch": 0.9632799558255107, + "grad_norm": 0.9049288630485535, + "learning_rate": 4.731645218213086e-06, + "loss": 0.6857, + "step": 1163 + }, + { + "epoch": 0.9641082274986196, + "grad_norm": 0.8972534537315369, + "learning_rate": 4.731149334441114e-06, + "loss": 0.6745, + "step": 1164 + }, + { + "epoch": 0.9649364991717283, + "grad_norm": 1.0799483060836792, + "learning_rate": 4.730653018962545e-06, + "loss": 0.6754, + "step": 1165 + }, + { + "epoch": 0.9657647708448371, + "grad_norm": 0.9841833710670471, + "learning_rate": 4.730156271873412e-06, + "loss": 0.6718, + "step": 1166 + }, + { + "epoch": 0.9665930425179459, + "grad_norm": 0.9388216733932495, + "learning_rate": 4.729659093269828e-06, + "loss": 0.6622, + "step": 1167 + }, + { + "epoch": 0.9674213141910547, + "grad_norm": 0.8879573345184326, + "learning_rate": 4.729161483247994e-06, + "loss": 0.6909, + "step": 1168 + }, + { + "epoch": 0.9682495858641634, + "grad_norm": 0.9713517427444458, + "learning_rate": 4.728663441904194e-06, + "loss": 0.6956, + "step": 1169 + }, + { + "epoch": 0.9690778575372723, + "grad_norm": 0.9200823903083801, + "learning_rate": 4.7281649693347925e-06, + "loss": 0.6766, + "step": 1170 + }, + { + "epoch": 0.969906129210381, + "grad_norm": 0.8983714580535889, + "learning_rate": 4.727666065636241e-06, + "loss": 0.6626, + "step": 1171 + }, + { + "epoch": 0.9707344008834898, + "grad_norm": 0.8792792558670044, + "learning_rate": 4.72716673090507e-06, + "loss": 0.666, + "step": 1172 + }, + { + "epoch": 0.9715626725565986, + "grad_norm": 1.0116846561431885, + "learning_rate": 4.7266669652379005e-06, + "loss": 0.6574, + "step": 1173 + }, + { + "epoch": 0.9723909442297074, + "grad_norm": 0.9433228373527527, + "learning_rate": 4.7261667687314285e-06, + "loss": 0.6972, + "step": 1174 + }, + { + "epoch": 0.9732192159028161, + "grad_norm": 0.8979770541191101, + "learning_rate": 4.725666141482439e-06, + "loss": 0.6523, + "step": 1175 + }, + { + "epoch": 0.9740474875759249, + "grad_norm": 0.8990604281425476, + "learning_rate": 4.725165083587798e-06, + "loss": 0.6535, + "step": 1176 + }, + { + "epoch": 0.9748757592490337, + "grad_norm": 0.9411956667900085, + "learning_rate": 4.724663595144457e-06, + "loss": 0.6947, + "step": 1177 + }, + { + "epoch": 0.9757040309221424, + "grad_norm": 1.0203242301940918, + "learning_rate": 4.724161676249448e-06, + "loss": 0.6894, + "step": 1178 + }, + { + "epoch": 0.9765323025952513, + "grad_norm": 0.8938170075416565, + "learning_rate": 4.723659326999888e-06, + "loss": 0.6648, + "step": 1179 + }, + { + "epoch": 0.97736057426836, + "grad_norm": 0.8810412883758545, + "learning_rate": 4.7231565474929765e-06, + "loss": 0.6761, + "step": 1180 + }, + { + "epoch": 0.9781888459414688, + "grad_norm": 0.8903076648712158, + "learning_rate": 4.722653337825998e-06, + "loss": 0.6738, + "step": 1181 + }, + { + "epoch": 0.9790171176145775, + "grad_norm": 0.905060350894928, + "learning_rate": 4.7221496980963175e-06, + "loss": 0.6625, + "step": 1182 + }, + { + "epoch": 0.9798453892876864, + "grad_norm": 0.8791100978851318, + "learning_rate": 4.721645628401385e-06, + "loss": 0.6884, + "step": 1183 + }, + { + "epoch": 0.9806736609607951, + "grad_norm": 0.897101104259491, + "learning_rate": 4.721141128838733e-06, + "loss": 0.6693, + "step": 1184 + }, + { + "epoch": 0.981501932633904, + "grad_norm": 0.9102833867073059, + "learning_rate": 4.7206361995059775e-06, + "loss": 0.6682, + "step": 1185 + }, + { + "epoch": 0.9823302043070127, + "grad_norm": 0.9492197632789612, + "learning_rate": 4.720130840500819e-06, + "loss": 0.6501, + "step": 1186 + }, + { + "epoch": 0.9831584759801215, + "grad_norm": 0.9783265590667725, + "learning_rate": 4.719625051921038e-06, + "loss": 0.6454, + "step": 1187 + }, + { + "epoch": 0.9839867476532302, + "grad_norm": 0.8938109874725342, + "learning_rate": 4.719118833864501e-06, + "loss": 0.6671, + "step": 1188 + }, + { + "epoch": 0.9848150193263391, + "grad_norm": 0.897484302520752, + "learning_rate": 4.718612186429156e-06, + "loss": 0.6664, + "step": 1189 + }, + { + "epoch": 0.9856432909994478, + "grad_norm": 0.9640376567840576, + "learning_rate": 4.718105109713034e-06, + "loss": 0.6845, + "step": 1190 + }, + { + "epoch": 0.9864715626725566, + "grad_norm": 0.9495493769645691, + "learning_rate": 4.7175976038142505e-06, + "loss": 0.6864, + "step": 1191 + }, + { + "epoch": 0.9872998343456654, + "grad_norm": 0.9390130639076233, + "learning_rate": 4.717089668831002e-06, + "loss": 0.6734, + "step": 1192 + }, + { + "epoch": 0.9881281060187742, + "grad_norm": 0.8854318261146545, + "learning_rate": 4.716581304861571e-06, + "loss": 0.6546, + "step": 1193 + }, + { + "epoch": 0.9889563776918829, + "grad_norm": 0.9242726564407349, + "learning_rate": 4.71607251200432e-06, + "loss": 0.6663, + "step": 1194 + }, + { + "epoch": 0.9897846493649917, + "grad_norm": 0.9302003383636475, + "learning_rate": 4.715563290357696e-06, + "loss": 0.6844, + "step": 1195 + }, + { + "epoch": 0.9906129210381005, + "grad_norm": 0.9279932975769043, + "learning_rate": 4.715053640020228e-06, + "loss": 0.6731, + "step": 1196 + }, + { + "epoch": 0.9914411927112092, + "grad_norm": 0.8753612041473389, + "learning_rate": 4.71454356109053e-06, + "loss": 0.6738, + "step": 1197 + }, + { + "epoch": 0.9922694643843181, + "grad_norm": 0.890133261680603, + "learning_rate": 4.714033053667296e-06, + "loss": 0.6638, + "step": 1198 + }, + { + "epoch": 0.9930977360574268, + "grad_norm": 0.9343729615211487, + "learning_rate": 4.713522117849305e-06, + "loss": 0.6826, + "step": 1199 + }, + { + "epoch": 0.9939260077305356, + "grad_norm": 0.9052755832672119, + "learning_rate": 4.7130107537354185e-06, + "loss": 0.6897, + "step": 1200 + }, + { + "epoch": 0.9947542794036444, + "grad_norm": 0.8683924674987793, + "learning_rate": 4.7124989614245805e-06, + "loss": 0.6687, + "step": 1201 + }, + { + "epoch": 0.9955825510767532, + "grad_norm": 0.8677132725715637, + "learning_rate": 4.711986741015818e-06, + "loss": 0.6917, + "step": 1202 + }, + { + "epoch": 0.9964108227498619, + "grad_norm": 0.8840690851211548, + "learning_rate": 4.711474092608243e-06, + "loss": 0.6841, + "step": 1203 + }, + { + "epoch": 0.9972390944229708, + "grad_norm": 0.9750391840934753, + "learning_rate": 4.7109610163010435e-06, + "loss": 0.6574, + "step": 1204 + }, + { + "epoch": 0.9980673660960795, + "grad_norm": 0.8752943873405457, + "learning_rate": 4.710447512193499e-06, + "loss": 0.636, + "step": 1205 + }, + { + "epoch": 0.9988956377691883, + "grad_norm": 0.9100313782691956, + "learning_rate": 4.709933580384965e-06, + "loss": 0.6712, + "step": 1206 + }, + { + "epoch": 0.999723909442297, + "grad_norm": 0.9202356338500977, + "learning_rate": 4.709419220974886e-06, + "loss": 0.6641, + "step": 1207 + }, + { + "epoch": 1.0, + "grad_norm": 0.9202356338500977, + "learning_rate": 4.708904434062783e-06, + "loss": 0.6199, + "step": 1208 + }, + { + "epoch": 1.0008282716731087, + "grad_norm": 1.6839245557785034, + "learning_rate": 4.708389219748263e-06, + "loss": 0.643, + "step": 1209 + }, + { + "epoch": 1.0016565433462175, + "grad_norm": 0.8636482357978821, + "learning_rate": 4.707873578131015e-06, + "loss": 0.6467, + "step": 1210 + }, + { + "epoch": 1.0024848150193264, + "grad_norm": 0.8548988103866577, + "learning_rate": 4.707357509310811e-06, + "loss": 0.6574, + "step": 1211 + }, + { + "epoch": 1.0033130866924351, + "grad_norm": 0.8961352705955505, + "learning_rate": 4.7068410133875066e-06, + "loss": 0.6367, + "step": 1212 + }, + { + "epoch": 1.0041413583655439, + "grad_norm": 0.8914269208908081, + "learning_rate": 4.706324090461037e-06, + "loss": 0.6294, + "step": 1213 + }, + { + "epoch": 1.0049696300386526, + "grad_norm": 0.879915177822113, + "learning_rate": 4.7058067406314235e-06, + "loss": 0.6346, + "step": 1214 + }, + { + "epoch": 1.0057979017117615, + "grad_norm": 0.8798143863677979, + "learning_rate": 4.7052889639987674e-06, + "loss": 0.6559, + "step": 1215 + }, + { + "epoch": 1.0066261733848703, + "grad_norm": 0.8888561129570007, + "learning_rate": 4.704770760663255e-06, + "loss": 0.6471, + "step": 1216 + }, + { + "epoch": 1.007454445057979, + "grad_norm": 0.8707575798034668, + "learning_rate": 4.704252130725152e-06, + "loss": 0.6485, + "step": 1217 + }, + { + "epoch": 1.0082827167310877, + "grad_norm": 0.8609346151351929, + "learning_rate": 4.7037330742848105e-06, + "loss": 0.6544, + "step": 1218 + }, + { + "epoch": 1.0091109884041967, + "grad_norm": 0.8588053584098816, + "learning_rate": 4.703213591442663e-06, + "loss": 0.6421, + "step": 1219 + }, + { + "epoch": 1.0099392600773054, + "grad_norm": 0.8932695388793945, + "learning_rate": 4.702693682299222e-06, + "loss": 0.6333, + "step": 1220 + }, + { + "epoch": 1.0107675317504141, + "grad_norm": 0.8615652322769165, + "learning_rate": 4.7021733469550865e-06, + "loss": 0.6371, + "step": 1221 + }, + { + "epoch": 1.0115958034235228, + "grad_norm": 0.8876634240150452, + "learning_rate": 4.7016525855109376e-06, + "loss": 0.6368, + "step": 1222 + }, + { + "epoch": 1.0124240750966318, + "grad_norm": 0.895575225353241, + "learning_rate": 4.701131398067537e-06, + "loss": 0.6226, + "step": 1223 + }, + { + "epoch": 1.0132523467697405, + "grad_norm": 0.8836749196052551, + "learning_rate": 4.70060978472573e-06, + "loss": 0.6516, + "step": 1224 + }, + { + "epoch": 1.0140806184428492, + "grad_norm": 0.9471280574798584, + "learning_rate": 4.700087745586442e-06, + "loss": 0.633, + "step": 1225 + }, + { + "epoch": 1.014908890115958, + "grad_norm": 0.8721561431884766, + "learning_rate": 4.699565280750685e-06, + "loss": 0.6568, + "step": 1226 + }, + { + "epoch": 1.015737161789067, + "grad_norm": 0.8946447372436523, + "learning_rate": 4.69904239031955e-06, + "loss": 0.6503, + "step": 1227 + }, + { + "epoch": 1.0165654334621756, + "grad_norm": 0.8854631185531616, + "learning_rate": 4.698519074394212e-06, + "loss": 0.6341, + "step": 1228 + }, + { + "epoch": 1.0173937051352844, + "grad_norm": 0.9348397850990295, + "learning_rate": 4.697995333075927e-06, + "loss": 0.6317, + "step": 1229 + }, + { + "epoch": 1.018221976808393, + "grad_norm": 0.9037336111068726, + "learning_rate": 4.697471166466035e-06, + "loss": 0.6618, + "step": 1230 + }, + { + "epoch": 1.0190502484815018, + "grad_norm": 0.8866400718688965, + "learning_rate": 4.696946574665957e-06, + "loss": 0.6421, + "step": 1231 + }, + { + "epoch": 1.0198785201546108, + "grad_norm": 0.8897701501846313, + "learning_rate": 4.696421557777196e-06, + "loss": 0.6608, + "step": 1232 + }, + { + "epoch": 1.0207067918277195, + "grad_norm": 0.9046173095703125, + "learning_rate": 4.6958961159013386e-06, + "loss": 0.6464, + "step": 1233 + }, + { + "epoch": 1.0215350635008282, + "grad_norm": 0.9492906332015991, + "learning_rate": 4.695370249140052e-06, + "loss": 0.6578, + "step": 1234 + }, + { + "epoch": 1.022363335173937, + "grad_norm": 0.8906951546669006, + "learning_rate": 4.6948439575950885e-06, + "loss": 0.6276, + "step": 1235 + }, + { + "epoch": 1.023191606847046, + "grad_norm": 0.9218266606330872, + "learning_rate": 4.694317241368278e-06, + "loss": 0.627, + "step": 1236 + }, + { + "epoch": 1.0240198785201546, + "grad_norm": 0.901157557964325, + "learning_rate": 4.693790100561537e-06, + "loss": 0.6379, + "step": 1237 + }, + { + "epoch": 1.0248481501932634, + "grad_norm": 0.8992452025413513, + "learning_rate": 4.693262535276863e-06, + "loss": 0.6585, + "step": 1238 + }, + { + "epoch": 1.025676421866372, + "grad_norm": 0.9525597095489502, + "learning_rate": 4.692734545616331e-06, + "loss": 0.6625, + "step": 1239 + }, + { + "epoch": 1.026504693539481, + "grad_norm": 0.9229065775871277, + "learning_rate": 4.692206131682106e-06, + "loss": 0.6454, + "step": 1240 + }, + { + "epoch": 1.0273329652125898, + "grad_norm": 1.012373685836792, + "learning_rate": 4.691677293576431e-06, + "loss": 0.6492, + "step": 1241 + }, + { + "epoch": 1.0281612368856985, + "grad_norm": 0.8650060892105103, + "learning_rate": 4.69114803140163e-06, + "loss": 0.6739, + "step": 1242 + }, + { + "epoch": 1.0289895085588072, + "grad_norm": 0.9396429657936096, + "learning_rate": 4.69061834526011e-06, + "loss": 0.6729, + "step": 1243 + }, + { + "epoch": 1.0298177802319162, + "grad_norm": 0.9584783911705017, + "learning_rate": 4.6900882352543614e-06, + "loss": 0.6678, + "step": 1244 + }, + { + "epoch": 1.030646051905025, + "grad_norm": 0.985394299030304, + "learning_rate": 4.689557701486954e-06, + "loss": 0.6478, + "step": 1245 + }, + { + "epoch": 1.0314743235781336, + "grad_norm": 0.9014914631843567, + "learning_rate": 4.689026744060543e-06, + "loss": 0.6393, + "step": 1246 + }, + { + "epoch": 1.0323025952512423, + "grad_norm": 0.9904847145080566, + "learning_rate": 4.688495363077863e-06, + "loss": 0.6752, + "step": 1247 + }, + { + "epoch": 1.0331308669243513, + "grad_norm": 0.9439241886138916, + "learning_rate": 4.687963558641731e-06, + "loss": 0.6417, + "step": 1248 + }, + { + "epoch": 1.03395913859746, + "grad_norm": 0.9277704954147339, + "learning_rate": 4.687431330855046e-06, + "loss": 0.6493, + "step": 1249 + }, + { + "epoch": 1.0347874102705688, + "grad_norm": 0.9456571340560913, + "learning_rate": 4.686898679820789e-06, + "loss": 0.6774, + "step": 1250 + }, + { + "epoch": 1.0356156819436775, + "grad_norm": 0.9043509364128113, + "learning_rate": 4.686365605642025e-06, + "loss": 0.6198, + "step": 1251 + }, + { + "epoch": 1.0364439536167862, + "grad_norm": 0.9246267676353455, + "learning_rate": 4.685832108421896e-06, + "loss": 0.6513, + "step": 1252 + }, + { + "epoch": 1.0372722252898952, + "grad_norm": 0.9227612614631653, + "learning_rate": 4.685298188263631e-06, + "loss": 0.6597, + "step": 1253 + }, + { + "epoch": 1.0381004969630039, + "grad_norm": 0.8927509188652039, + "learning_rate": 4.684763845270537e-06, + "loss": 0.641, + "step": 1254 + }, + { + "epoch": 1.0389287686361126, + "grad_norm": 0.8812937140464783, + "learning_rate": 4.684229079546005e-06, + "loss": 0.6292, + "step": 1255 + }, + { + "epoch": 1.0397570403092213, + "grad_norm": 0.9186881184577942, + "learning_rate": 4.683693891193508e-06, + "loss": 0.6485, + "step": 1256 + }, + { + "epoch": 1.0405853119823303, + "grad_norm": 0.8928134441375732, + "learning_rate": 4.683158280316599e-06, + "loss": 0.6435, + "step": 1257 + }, + { + "epoch": 1.041413583655439, + "grad_norm": 0.8919920325279236, + "learning_rate": 4.682622247018913e-06, + "loss": 0.622, + "step": 1258 + }, + { + "epoch": 1.0422418553285477, + "grad_norm": 0.885259747505188, + "learning_rate": 4.682085791404169e-06, + "loss": 0.6521, + "step": 1259 + }, + { + "epoch": 1.0430701270016565, + "grad_norm": 0.9034169316291809, + "learning_rate": 4.681548913576165e-06, + "loss": 0.6437, + "step": 1260 + }, + { + "epoch": 1.0438983986747654, + "grad_norm": 0.884848952293396, + "learning_rate": 4.6810116136387825e-06, + "loss": 0.6733, + "step": 1261 + }, + { + "epoch": 1.0447266703478741, + "grad_norm": 0.9137836694717407, + "learning_rate": 4.680473891695985e-06, + "loss": 0.6415, + "step": 1262 + }, + { + "epoch": 1.0455549420209829, + "grad_norm": 0.9044780135154724, + "learning_rate": 4.6799357478518145e-06, + "loss": 0.6453, + "step": 1263 + }, + { + "epoch": 1.0463832136940916, + "grad_norm": 0.9111152291297913, + "learning_rate": 4.6793971822103985e-06, + "loss": 0.6644, + "step": 1264 + }, + { + "epoch": 1.0472114853672005, + "grad_norm": 0.9158084988594055, + "learning_rate": 4.678858194875944e-06, + "loss": 0.6434, + "step": 1265 + }, + { + "epoch": 1.0480397570403093, + "grad_norm": 0.8970723152160645, + "learning_rate": 4.678318785952739e-06, + "loss": 0.6391, + "step": 1266 + }, + { + "epoch": 1.048868028713418, + "grad_norm": 0.8958215117454529, + "learning_rate": 4.677778955545155e-06, + "loss": 0.6562, + "step": 1267 + }, + { + "epoch": 1.0496963003865267, + "grad_norm": 0.9229645729064941, + "learning_rate": 4.677238703757644e-06, + "loss": 0.646, + "step": 1268 + }, + { + "epoch": 1.0505245720596355, + "grad_norm": 0.8986689448356628, + "learning_rate": 4.676698030694741e-06, + "loss": 0.6241, + "step": 1269 + }, + { + "epoch": 1.0513528437327444, + "grad_norm": 0.8943328261375427, + "learning_rate": 4.6761569364610585e-06, + "loss": 0.6455, + "step": 1270 + }, + { + "epoch": 1.0521811154058531, + "grad_norm": 0.9147390127182007, + "learning_rate": 4.675615421161295e-06, + "loss": 0.6299, + "step": 1271 + }, + { + "epoch": 1.0530093870789619, + "grad_norm": 0.9124075174331665, + "learning_rate": 4.675073484900229e-06, + "loss": 0.6379, + "step": 1272 + }, + { + "epoch": 1.0538376587520706, + "grad_norm": 0.9267525672912598, + "learning_rate": 4.674531127782718e-06, + "loss": 0.6478, + "step": 1273 + }, + { + "epoch": 1.0546659304251795, + "grad_norm": 0.8710078001022339, + "learning_rate": 4.673988349913707e-06, + "loss": 0.6314, + "step": 1274 + }, + { + "epoch": 1.0554942020982883, + "grad_norm": 0.8905670642852783, + "learning_rate": 4.673445151398214e-06, + "loss": 0.6456, + "step": 1275 + }, + { + "epoch": 1.056322473771397, + "grad_norm": 0.9125292897224426, + "learning_rate": 4.672901532341346e-06, + "loss": 0.6439, + "step": 1276 + }, + { + "epoch": 1.0571507454445057, + "grad_norm": 0.9003625512123108, + "learning_rate": 4.6723574928482865e-06, + "loss": 0.6561, + "step": 1277 + }, + { + "epoch": 1.0579790171176147, + "grad_norm": 0.96099454164505, + "learning_rate": 4.6718130330243014e-06, + "loss": 0.6515, + "step": 1278 + }, + { + "epoch": 1.0588072887907234, + "grad_norm": 0.8841069340705872, + "learning_rate": 4.671268152974742e-06, + "loss": 0.6389, + "step": 1279 + }, + { + "epoch": 1.0596355604638321, + "grad_norm": 0.8906645178794861, + "learning_rate": 4.670722852805033e-06, + "loss": 0.6448, + "step": 1280 + }, + { + "epoch": 1.0604638321369408, + "grad_norm": 0.9538317918777466, + "learning_rate": 4.670177132620689e-06, + "loss": 0.6544, + "step": 1281 + }, + { + "epoch": 1.0612921038100498, + "grad_norm": 0.9158053994178772, + "learning_rate": 4.669630992527299e-06, + "loss": 0.6494, + "step": 1282 + }, + { + "epoch": 1.0621203754831585, + "grad_norm": 0.9756888151168823, + "learning_rate": 4.6690844326305374e-06, + "loss": 0.6419, + "step": 1283 + }, + { + "epoch": 1.0629486471562672, + "grad_norm": 0.905055820941925, + "learning_rate": 4.668537453036159e-06, + "loss": 0.6463, + "step": 1284 + }, + { + "epoch": 1.063776918829376, + "grad_norm": 0.9011061191558838, + "learning_rate": 4.667990053849997e-06, + "loss": 0.6353, + "step": 1285 + }, + { + "epoch": 1.0646051905024847, + "grad_norm": 0.955047070980072, + "learning_rate": 4.667442235177969e-06, + "loss": 0.6485, + "step": 1286 + }, + { + "epoch": 1.0654334621755936, + "grad_norm": 0.9253432750701904, + "learning_rate": 4.666893997126074e-06, + "loss": 0.6391, + "step": 1287 + }, + { + "epoch": 1.0662617338487024, + "grad_norm": 0.8840277194976807, + "learning_rate": 4.66634533980039e-06, + "loss": 0.6372, + "step": 1288 + }, + { + "epoch": 1.067090005521811, + "grad_norm": 1.0881837606430054, + "learning_rate": 4.6657962633070765e-06, + "loss": 0.6419, + "step": 1289 + }, + { + "epoch": 1.06791827719492, + "grad_norm": 0.9263116717338562, + "learning_rate": 4.665246767752376e-06, + "loss": 0.6226, + "step": 1290 + }, + { + "epoch": 1.0687465488680288, + "grad_norm": 0.9110594987869263, + "learning_rate": 4.66469685324261e-06, + "loss": 0.6495, + "step": 1291 + }, + { + "epoch": 1.0695748205411375, + "grad_norm": 0.9086888432502747, + "learning_rate": 4.664146519884182e-06, + "loss": 0.6408, + "step": 1292 + }, + { + "epoch": 1.0704030922142462, + "grad_norm": 0.8991205096244812, + "learning_rate": 4.663595767783575e-06, + "loss": 0.6424, + "step": 1293 + }, + { + "epoch": 1.071231363887355, + "grad_norm": 0.926933228969574, + "learning_rate": 4.6630445970473576e-06, + "loss": 0.6379, + "step": 1294 + }, + { + "epoch": 1.072059635560464, + "grad_norm": 0.8793080449104309, + "learning_rate": 4.662493007782174e-06, + "loss": 0.6254, + "step": 1295 + }, + { + "epoch": 1.0728879072335726, + "grad_norm": 0.9588155150413513, + "learning_rate": 4.661941000094751e-06, + "loss": 0.6307, + "step": 1296 + }, + { + "epoch": 1.0737161789066814, + "grad_norm": 0.9468860030174255, + "learning_rate": 4.661388574091898e-06, + "loss": 0.6308, + "step": 1297 + }, + { + "epoch": 1.07454445057979, + "grad_norm": 0.8930124640464783, + "learning_rate": 4.660835729880505e-06, + "loss": 0.6371, + "step": 1298 + }, + { + "epoch": 1.075372722252899, + "grad_norm": 0.9233878254890442, + "learning_rate": 4.660282467567541e-06, + "loss": 0.6327, + "step": 1299 + }, + { + "epoch": 1.0762009939260078, + "grad_norm": 0.9714951515197754, + "learning_rate": 4.659728787260057e-06, + "loss": 0.6218, + "step": 1300 + }, + { + "epoch": 1.0770292655991165, + "grad_norm": 0.9464428424835205, + "learning_rate": 4.659174689065185e-06, + "loss": 0.6406, + "step": 1301 + }, + { + "epoch": 1.0778575372722252, + "grad_norm": 0.8984559774398804, + "learning_rate": 4.6586201730901395e-06, + "loss": 0.6483, + "step": 1302 + }, + { + "epoch": 1.0786858089453342, + "grad_norm": 0.9007774591445923, + "learning_rate": 4.658065239442212e-06, + "loss": 0.6288, + "step": 1303 + }, + { + "epoch": 1.079514080618443, + "grad_norm": 0.907171368598938, + "learning_rate": 4.657509888228779e-06, + "loss": 0.6445, + "step": 1304 + }, + { + "epoch": 1.0803423522915516, + "grad_norm": 0.9207282066345215, + "learning_rate": 4.656954119557293e-06, + "loss": 0.6148, + "step": 1305 + }, + { + "epoch": 1.0811706239646603, + "grad_norm": 0.9467616677284241, + "learning_rate": 4.656397933535293e-06, + "loss": 0.6374, + "step": 1306 + }, + { + "epoch": 1.0819988956377693, + "grad_norm": 0.9310930967330933, + "learning_rate": 4.655841330270393e-06, + "loss": 0.6362, + "step": 1307 + }, + { + "epoch": 1.082827167310878, + "grad_norm": 0.890816330909729, + "learning_rate": 4.655284309870294e-06, + "loss": 0.6372, + "step": 1308 + }, + { + "epoch": 1.0836554389839868, + "grad_norm": 0.9206095933914185, + "learning_rate": 4.654726872442771e-06, + "loss": 0.6297, + "step": 1309 + }, + { + "epoch": 1.0844837106570955, + "grad_norm": 0.9015856385231018, + "learning_rate": 4.654169018095685e-06, + "loss": 0.658, + "step": 1310 + }, + { + "epoch": 1.0853119823302042, + "grad_norm": 0.8909441828727722, + "learning_rate": 4.653610746936975e-06, + "loss": 0.6382, + "step": 1311 + }, + { + "epoch": 1.0861402540033132, + "grad_norm": 0.9035969972610474, + "learning_rate": 4.653052059074661e-06, + "loss": 0.6441, + "step": 1312 + }, + { + "epoch": 1.0869685256764219, + "grad_norm": 0.8988869786262512, + "learning_rate": 4.652492954616844e-06, + "loss": 0.6359, + "step": 1313 + }, + { + "epoch": 1.0877967973495306, + "grad_norm": 0.881885826587677, + "learning_rate": 4.6519334336717055e-06, + "loss": 0.6397, + "step": 1314 + }, + { + "epoch": 1.0886250690226393, + "grad_norm": 0.980633020401001, + "learning_rate": 4.651373496347508e-06, + "loss": 0.6223, + "step": 1315 + }, + { + "epoch": 1.0894533406957483, + "grad_norm": 0.9027420878410339, + "learning_rate": 4.650813142752593e-06, + "loss": 0.6211, + "step": 1316 + }, + { + "epoch": 1.090281612368857, + "grad_norm": 0.9011778831481934, + "learning_rate": 4.6502523729953855e-06, + "loss": 0.6442, + "step": 1317 + }, + { + "epoch": 1.0911098840419657, + "grad_norm": 0.926692545413971, + "learning_rate": 4.649691187184387e-06, + "loss": 0.6661, + "step": 1318 + }, + { + "epoch": 1.0919381557150745, + "grad_norm": 0.9758741855621338, + "learning_rate": 4.649129585428184e-06, + "loss": 0.6461, + "step": 1319 + }, + { + "epoch": 1.0927664273881834, + "grad_norm": 0.967698335647583, + "learning_rate": 4.64856756783544e-06, + "loss": 0.6461, + "step": 1320 + }, + { + "epoch": 1.0935946990612921, + "grad_norm": 0.9372200965881348, + "learning_rate": 4.6480051345149e-06, + "loss": 0.6348, + "step": 1321 + }, + { + "epoch": 1.0944229707344009, + "grad_norm": 0.8976414799690247, + "learning_rate": 4.64744228557539e-06, + "loss": 0.6297, + "step": 1322 + }, + { + "epoch": 1.0952512424075096, + "grad_norm": 0.9163550734519958, + "learning_rate": 4.646879021125816e-06, + "loss": 0.6746, + "step": 1323 + }, + { + "epoch": 1.0960795140806185, + "grad_norm": 0.9064576625823975, + "learning_rate": 4.646315341275164e-06, + "loss": 0.6242, + "step": 1324 + }, + { + "epoch": 1.0969077857537273, + "grad_norm": 0.9352027773857117, + "learning_rate": 4.645751246132501e-06, + "loss": 0.6568, + "step": 1325 + }, + { + "epoch": 1.097736057426836, + "grad_norm": 0.8680023550987244, + "learning_rate": 4.645186735806976e-06, + "loss": 0.6484, + "step": 1326 + }, + { + "epoch": 1.0985643290999447, + "grad_norm": 0.8782095909118652, + "learning_rate": 4.644621810407813e-06, + "loss": 0.646, + "step": 1327 + }, + { + "epoch": 1.0993926007730535, + "grad_norm": 0.8931819796562195, + "learning_rate": 4.644056470044323e-06, + "loss": 0.6708, + "step": 1328 + }, + { + "epoch": 1.1002208724461624, + "grad_norm": 0.9228624105453491, + "learning_rate": 4.6434907148258915e-06, + "loss": 0.6255, + "step": 1329 + }, + { + "epoch": 1.1010491441192711, + "grad_norm": 0.8960729837417603, + "learning_rate": 4.642924544861988e-06, + "loss": 0.6523, + "step": 1330 + }, + { + "epoch": 1.1018774157923799, + "grad_norm": 0.9092081785202026, + "learning_rate": 4.642357960262162e-06, + "loss": 0.6534, + "step": 1331 + }, + { + "epoch": 1.1027056874654886, + "grad_norm": 0.9024838209152222, + "learning_rate": 4.641790961136041e-06, + "loss": 0.6238, + "step": 1332 + }, + { + "epoch": 1.1035339591385975, + "grad_norm": 0.9016388058662415, + "learning_rate": 4.641223547593334e-06, + "loss": 0.6643, + "step": 1333 + }, + { + "epoch": 1.1043622308117063, + "grad_norm": 0.8877143859863281, + "learning_rate": 4.6406557197438324e-06, + "loss": 0.6222, + "step": 1334 + }, + { + "epoch": 1.105190502484815, + "grad_norm": 0.9066653251647949, + "learning_rate": 4.640087477697403e-06, + "loss": 0.6205, + "step": 1335 + }, + { + "epoch": 1.1060187741579237, + "grad_norm": 0.903465747833252, + "learning_rate": 4.639518821563997e-06, + "loss": 0.6299, + "step": 1336 + }, + { + "epoch": 1.1068470458310327, + "grad_norm": 0.8908408880233765, + "learning_rate": 4.6389497514536415e-06, + "loss": 0.6247, + "step": 1337 + }, + { + "epoch": 1.1076753175041414, + "grad_norm": 0.8957250714302063, + "learning_rate": 4.63838026747645e-06, + "loss": 0.632, + "step": 1338 + }, + { + "epoch": 1.1085035891772501, + "grad_norm": 0.938494086265564, + "learning_rate": 4.637810369742609e-06, + "loss": 0.6309, + "step": 1339 + }, + { + "epoch": 1.1093318608503588, + "grad_norm": 0.9913409948348999, + "learning_rate": 4.637240058362391e-06, + "loss": 0.6571, + "step": 1340 + }, + { + "epoch": 1.1101601325234678, + "grad_norm": 0.9037449359893799, + "learning_rate": 4.636669333446145e-06, + "loss": 0.6467, + "step": 1341 + }, + { + "epoch": 1.1109884041965765, + "grad_norm": 0.911696195602417, + "learning_rate": 4.636098195104299e-06, + "loss": 0.6361, + "step": 1342 + }, + { + "epoch": 1.1118166758696852, + "grad_norm": 0.9105570316314697, + "learning_rate": 4.6355266434473665e-06, + "loss": 0.6418, + "step": 1343 + }, + { + "epoch": 1.112644947542794, + "grad_norm": 0.9101701378822327, + "learning_rate": 4.634954678585935e-06, + "loss": 0.6454, + "step": 1344 + }, + { + "epoch": 1.1134732192159027, + "grad_norm": 0.8980433344841003, + "learning_rate": 4.634382300630675e-06, + "loss": 0.6159, + "step": 1345 + }, + { + "epoch": 1.1143014908890116, + "grad_norm": 0.9253267049789429, + "learning_rate": 4.633809509692336e-06, + "loss": 0.6465, + "step": 1346 + }, + { + "epoch": 1.1151297625621204, + "grad_norm": 0.9150062203407288, + "learning_rate": 4.6332363058817484e-06, + "loss": 0.639, + "step": 1347 + }, + { + "epoch": 1.115958034235229, + "grad_norm": 0.9257630705833435, + "learning_rate": 4.632662689309821e-06, + "loss": 0.6571, + "step": 1348 + }, + { + "epoch": 1.116786305908338, + "grad_norm": 0.9183084964752197, + "learning_rate": 4.632088660087545e-06, + "loss": 0.6291, + "step": 1349 + }, + { + "epoch": 1.1176145775814468, + "grad_norm": 0.9229403138160706, + "learning_rate": 4.631514218325987e-06, + "loss": 0.6487, + "step": 1350 + }, + { + "epoch": 1.1184428492545555, + "grad_norm": 0.9379578828811646, + "learning_rate": 4.630939364136298e-06, + "loss": 0.6564, + "step": 1351 + }, + { + "epoch": 1.1192711209276642, + "grad_norm": 0.9094021320343018, + "learning_rate": 4.630364097629706e-06, + "loss": 0.6354, + "step": 1352 + }, + { + "epoch": 1.120099392600773, + "grad_norm": 0.9528741836547852, + "learning_rate": 4.62978841891752e-06, + "loss": 0.6259, + "step": 1353 + }, + { + "epoch": 1.120927664273882, + "grad_norm": 0.8859900832176208, + "learning_rate": 4.629212328111129e-06, + "loss": 0.6783, + "step": 1354 + }, + { + "epoch": 1.1217559359469906, + "grad_norm": 0.8739688396453857, + "learning_rate": 4.628635825322e-06, + "loss": 0.6518, + "step": 1355 + }, + { + "epoch": 1.1225842076200994, + "grad_norm": 0.8773418068885803, + "learning_rate": 4.628058910661683e-06, + "loss": 0.6477, + "step": 1356 + }, + { + "epoch": 1.123412479293208, + "grad_norm": 0.9025301933288574, + "learning_rate": 4.627481584241803e-06, + "loss": 0.6471, + "step": 1357 + }, + { + "epoch": 1.124240750966317, + "grad_norm": 0.89899080991745, + "learning_rate": 4.62690384617407e-06, + "loss": 0.6196, + "step": 1358 + }, + { + "epoch": 1.1250690226394258, + "grad_norm": 0.9514591693878174, + "learning_rate": 4.6263256965702684e-06, + "loss": 0.6495, + "step": 1359 + }, + { + "epoch": 1.1258972943125345, + "grad_norm": 0.908789336681366, + "learning_rate": 4.6257471355422654e-06, + "loss": 0.6606, + "step": 1360 + }, + { + "epoch": 1.1267255659856432, + "grad_norm": 0.9262552857398987, + "learning_rate": 4.625168163202009e-06, + "loss": 0.6308, + "step": 1361 + }, + { + "epoch": 1.1275538376587522, + "grad_norm": 0.9395532608032227, + "learning_rate": 4.624588779661523e-06, + "loss": 0.6196, + "step": 1362 + }, + { + "epoch": 1.128382109331861, + "grad_norm": 0.9126531481742859, + "learning_rate": 4.624008985032913e-06, + "loss": 0.6474, + "step": 1363 + }, + { + "epoch": 1.1292103810049696, + "grad_norm": 0.9105198383331299, + "learning_rate": 4.623428779428365e-06, + "loss": 0.6455, + "step": 1364 + }, + { + "epoch": 1.1300386526780783, + "grad_norm": 0.9048190712928772, + "learning_rate": 4.6228481629601415e-06, + "loss": 0.6647, + "step": 1365 + }, + { + "epoch": 1.1308669243511873, + "grad_norm": 0.9415149092674255, + "learning_rate": 4.622267135740588e-06, + "loss": 0.6403, + "step": 1366 + }, + { + "epoch": 1.131695196024296, + "grad_norm": 0.954728901386261, + "learning_rate": 4.6216856978821275e-06, + "loss": 0.6351, + "step": 1367 + }, + { + "epoch": 1.1325234676974048, + "grad_norm": 0.9392480850219727, + "learning_rate": 4.621103849497261e-06, + "loss": 0.6481, + "step": 1368 + }, + { + "epoch": 1.1333517393705135, + "grad_norm": 0.9199283719062805, + "learning_rate": 4.620521590698574e-06, + "loss": 0.6578, + "step": 1369 + }, + { + "epoch": 1.1341800110436222, + "grad_norm": 0.9175401926040649, + "learning_rate": 4.6199389215987266e-06, + "loss": 0.6508, + "step": 1370 + }, + { + "epoch": 1.1350082827167312, + "grad_norm": 0.9140470027923584, + "learning_rate": 4.619355842310459e-06, + "loss": 0.6536, + "step": 1371 + }, + { + "epoch": 1.1358365543898399, + "grad_norm": 0.8863652348518372, + "learning_rate": 4.618772352946593e-06, + "loss": 0.6376, + "step": 1372 + }, + { + "epoch": 1.1366648260629486, + "grad_norm": 0.9074656963348389, + "learning_rate": 4.618188453620026e-06, + "loss": 0.6253, + "step": 1373 + }, + { + "epoch": 1.1374930977360573, + "grad_norm": 0.891055703163147, + "learning_rate": 4.61760414444374e-06, + "loss": 0.6507, + "step": 1374 + }, + { + "epoch": 1.1383213694091663, + "grad_norm": 0.9153634905815125, + "learning_rate": 4.617019425530791e-06, + "loss": 0.6699, + "step": 1375 + }, + { + "epoch": 1.139149641082275, + "grad_norm": 0.9172940850257874, + "learning_rate": 4.6164342969943196e-06, + "loss": 0.6385, + "step": 1376 + }, + { + "epoch": 1.1399779127553837, + "grad_norm": 0.8978968262672424, + "learning_rate": 4.615848758947539e-06, + "loss": 0.6394, + "step": 1377 + }, + { + "epoch": 1.1408061844284925, + "grad_norm": 0.9372277855873108, + "learning_rate": 4.615262811503749e-06, + "loss": 0.6418, + "step": 1378 + }, + { + "epoch": 1.1416344561016014, + "grad_norm": 0.9596112370491028, + "learning_rate": 4.614676454776321e-06, + "loss": 0.6573, + "step": 1379 + }, + { + "epoch": 1.1424627277747101, + "grad_norm": 0.9243675470352173, + "learning_rate": 4.614089688878713e-06, + "loss": 0.6546, + "step": 1380 + }, + { + "epoch": 1.1432909994478189, + "grad_norm": 0.8839207291603088, + "learning_rate": 4.613502513924459e-06, + "loss": 0.654, + "step": 1381 + }, + { + "epoch": 1.1441192711209276, + "grad_norm": 0.8940262198448181, + "learning_rate": 4.612914930027168e-06, + "loss": 0.6223, + "step": 1382 + }, + { + "epoch": 1.1449475427940365, + "grad_norm": 0.9091162085533142, + "learning_rate": 4.612326937300535e-06, + "loss": 0.6512, + "step": 1383 + }, + { + "epoch": 1.1457758144671453, + "grad_norm": 0.9332976937294006, + "learning_rate": 4.6117385358583315e-06, + "loss": 0.6373, + "step": 1384 + }, + { + "epoch": 1.146604086140254, + "grad_norm": 0.9103875756263733, + "learning_rate": 4.611149725814406e-06, + "loss": 0.6502, + "step": 1385 + }, + { + "epoch": 1.1474323578133627, + "grad_norm": 0.9004862308502197, + "learning_rate": 4.610560507282688e-06, + "loss": 0.6361, + "step": 1386 + }, + { + "epoch": 1.1482606294864715, + "grad_norm": 0.880293071269989, + "learning_rate": 4.609970880377187e-06, + "loss": 0.655, + "step": 1387 + }, + { + "epoch": 1.1490889011595804, + "grad_norm": 0.9099947214126587, + "learning_rate": 4.60938084521199e-06, + "loss": 0.6517, + "step": 1388 + }, + { + "epoch": 1.1499171728326891, + "grad_norm": 0.8913806080818176, + "learning_rate": 4.608790401901262e-06, + "loss": 0.6429, + "step": 1389 + }, + { + "epoch": 1.1507454445057979, + "grad_norm": 0.9198290109634399, + "learning_rate": 4.608199550559249e-06, + "loss": 0.6537, + "step": 1390 + }, + { + "epoch": 1.1515737161789068, + "grad_norm": 0.9192278981208801, + "learning_rate": 4.607608291300275e-06, + "loss": 0.6531, + "step": 1391 + }, + { + "epoch": 1.1524019878520155, + "grad_norm": 0.8977665305137634, + "learning_rate": 4.607016624238744e-06, + "loss": 0.6358, + "step": 1392 + }, + { + "epoch": 1.1532302595251243, + "grad_norm": 0.9576854705810547, + "learning_rate": 4.606424549489138e-06, + "loss": 0.6493, + "step": 1393 + }, + { + "epoch": 1.154058531198233, + "grad_norm": 0.905611515045166, + "learning_rate": 4.605832067166017e-06, + "loss": 0.63, + "step": 1394 + }, + { + "epoch": 1.1548868028713417, + "grad_norm": 0.8840015530586243, + "learning_rate": 4.605239177384021e-06, + "loss": 0.6504, + "step": 1395 + }, + { + "epoch": 1.1557150745444507, + "grad_norm": 0.9247309565544128, + "learning_rate": 4.604645880257869e-06, + "loss": 0.6477, + "step": 1396 + }, + { + "epoch": 1.1565433462175594, + "grad_norm": 0.9473649263381958, + "learning_rate": 4.604052175902359e-06, + "loss": 0.6374, + "step": 1397 + }, + { + "epoch": 1.1573716178906681, + "grad_norm": 0.9354954361915588, + "learning_rate": 4.603458064432367e-06, + "loss": 0.6491, + "step": 1398 + }, + { + "epoch": 1.1581998895637768, + "grad_norm": 0.8788221478462219, + "learning_rate": 4.6028635459628476e-06, + "loss": 0.6187, + "step": 1399 + }, + { + "epoch": 1.1590281612368858, + "grad_norm": 0.8836075067520142, + "learning_rate": 4.602268620608834e-06, + "loss": 0.665, + "step": 1400 + }, + { + "epoch": 1.1598564329099945, + "grad_norm": 0.9195554256439209, + "learning_rate": 4.60167328848544e-06, + "loss": 0.6291, + "step": 1401 + }, + { + "epoch": 1.1606847045831032, + "grad_norm": 0.8984520435333252, + "learning_rate": 4.601077549707856e-06, + "loss": 0.6361, + "step": 1402 + }, + { + "epoch": 1.161512976256212, + "grad_norm": 0.8667109608650208, + "learning_rate": 4.600481404391352e-06, + "loss": 0.6345, + "step": 1403 + }, + { + "epoch": 1.1623412479293207, + "grad_norm": 0.9312148094177246, + "learning_rate": 4.599884852651277e-06, + "loss": 0.6388, + "step": 1404 + }, + { + "epoch": 1.1631695196024296, + "grad_norm": 0.9096866846084595, + "learning_rate": 4.5992878946030575e-06, + "loss": 0.6383, + "step": 1405 + }, + { + "epoch": 1.1639977912755384, + "grad_norm": 0.9010055661201477, + "learning_rate": 4.5986905303622e-06, + "loss": 0.6524, + "step": 1406 + }, + { + "epoch": 1.164826062948647, + "grad_norm": 0.9414717555046082, + "learning_rate": 4.598092760044288e-06, + "loss": 0.6825, + "step": 1407 + }, + { + "epoch": 1.165654334621756, + "grad_norm": 0.9162404537200928, + "learning_rate": 4.597494583764987e-06, + "loss": 0.6467, + "step": 1408 + }, + { + "epoch": 1.1664826062948648, + "grad_norm": 0.8935617208480835, + "learning_rate": 4.596896001640034e-06, + "loss": 0.6375, + "step": 1409 + }, + { + "epoch": 1.1673108779679735, + "grad_norm": 0.8976510167121887, + "learning_rate": 4.596297013785253e-06, + "loss": 0.6135, + "step": 1410 + }, + { + "epoch": 1.1681391496410822, + "grad_norm": 0.9503909349441528, + "learning_rate": 4.5956976203165414e-06, + "loss": 0.6305, + "step": 1411 + }, + { + "epoch": 1.168967421314191, + "grad_norm": 0.9147273302078247, + "learning_rate": 4.595097821349876e-06, + "loss": 0.6548, + "step": 1412 + }, + { + "epoch": 1.1697956929873, + "grad_norm": 0.9171034693717957, + "learning_rate": 4.594497617001312e-06, + "loss": 0.6253, + "step": 1413 + }, + { + "epoch": 1.1706239646604086, + "grad_norm": 0.8970120549201965, + "learning_rate": 4.593897007386985e-06, + "loss": 0.6414, + "step": 1414 + }, + { + "epoch": 1.1714522363335174, + "grad_norm": 0.9032952189445496, + "learning_rate": 4.593295992623105e-06, + "loss": 0.6595, + "step": 1415 + }, + { + "epoch": 1.172280508006626, + "grad_norm": 0.9066448211669922, + "learning_rate": 4.592694572825964e-06, + "loss": 0.6654, + "step": 1416 + }, + { + "epoch": 1.173108779679735, + "grad_norm": 0.8747309446334839, + "learning_rate": 4.592092748111931e-06, + "loss": 0.6619, + "step": 1417 + }, + { + "epoch": 1.1739370513528438, + "grad_norm": 0.8836600184440613, + "learning_rate": 4.5914905185974535e-06, + "loss": 0.6443, + "step": 1418 + }, + { + "epoch": 1.1747653230259525, + "grad_norm": 0.9340765476226807, + "learning_rate": 4.590887884399058e-06, + "loss": 0.6633, + "step": 1419 + }, + { + "epoch": 1.1755935946990612, + "grad_norm": 0.8889044523239136, + "learning_rate": 4.590284845633347e-06, + "loss": 0.6489, + "step": 1420 + }, + { + "epoch": 1.1764218663721702, + "grad_norm": 0.9043524265289307, + "learning_rate": 4.589681402417005e-06, + "loss": 0.6392, + "step": 1421 + }, + { + "epoch": 1.177250138045279, + "grad_norm": 0.9210046529769897, + "learning_rate": 4.5890775548667895e-06, + "loss": 0.6105, + "step": 1422 + }, + { + "epoch": 1.1780784097183876, + "grad_norm": 0.8954651951789856, + "learning_rate": 4.588473303099542e-06, + "loss": 0.6425, + "step": 1423 + }, + { + "epoch": 1.1789066813914963, + "grad_norm": 0.945517897605896, + "learning_rate": 4.587868647232179e-06, + "loss": 0.6404, + "step": 1424 + }, + { + "epoch": 1.1797349530646053, + "grad_norm": 0.9059463739395142, + "learning_rate": 4.587263587381696e-06, + "loss": 0.6521, + "step": 1425 + }, + { + "epoch": 1.180563224737714, + "grad_norm": 0.8876296281814575, + "learning_rate": 4.586658123665165e-06, + "loss": 0.6447, + "step": 1426 + }, + { + "epoch": 1.1813914964108228, + "grad_norm": 0.8862487077713013, + "learning_rate": 4.58605225619974e-06, + "loss": 0.6445, + "step": 1427 + }, + { + "epoch": 1.1822197680839315, + "grad_norm": 0.9066351056098938, + "learning_rate": 4.585445985102649e-06, + "loss": 0.6563, + "step": 1428 + }, + { + "epoch": 1.1830480397570402, + "grad_norm": 0.913724958896637, + "learning_rate": 4.584839310491199e-06, + "loss": 0.6507, + "step": 1429 + }, + { + "epoch": 1.1838763114301492, + "grad_norm": 0.8942010402679443, + "learning_rate": 4.584232232482778e-06, + "loss": 0.6424, + "step": 1430 + }, + { + "epoch": 1.1847045831032579, + "grad_norm": 0.8986000418663025, + "learning_rate": 4.58362475119485e-06, + "loss": 0.6491, + "step": 1431 + }, + { + "epoch": 1.1855328547763666, + "grad_norm": 0.8879220485687256, + "learning_rate": 4.5830168667449544e-06, + "loss": 0.6412, + "step": 1432 + }, + { + "epoch": 1.1863611264494753, + "grad_norm": 0.9497302770614624, + "learning_rate": 4.5824085792507135e-06, + "loss": 0.6453, + "step": 1433 + }, + { + "epoch": 1.1871893981225843, + "grad_norm": 0.8933898210525513, + "learning_rate": 4.581799888829824e-06, + "loss": 0.6381, + "step": 1434 + }, + { + "epoch": 1.188017669795693, + "grad_norm": 0.9034610986709595, + "learning_rate": 4.581190795600062e-06, + "loss": 0.6551, + "step": 1435 + }, + { + "epoch": 1.1888459414688017, + "grad_norm": 0.8968634605407715, + "learning_rate": 4.580581299679282e-06, + "loss": 0.637, + "step": 1436 + }, + { + "epoch": 1.1896742131419105, + "grad_norm": 0.8974384069442749, + "learning_rate": 4.579971401185416e-06, + "loss": 0.6301, + "step": 1437 + }, + { + "epoch": 1.1905024848150194, + "grad_norm": 0.8726125359535217, + "learning_rate": 4.579361100236473e-06, + "loss": 0.6487, + "step": 1438 + }, + { + "epoch": 1.1913307564881281, + "grad_norm": 0.9131139516830444, + "learning_rate": 4.5787503969505405e-06, + "loss": 0.6211, + "step": 1439 + }, + { + "epoch": 1.1921590281612369, + "grad_norm": 0.8924953937530518, + "learning_rate": 4.578139291445783e-06, + "loss": 0.6339, + "step": 1440 + }, + { + "epoch": 1.1929872998343456, + "grad_norm": 0.890883207321167, + "learning_rate": 4.577527783840447e-06, + "loss": 0.66, + "step": 1441 + }, + { + "epoch": 1.1938155715074545, + "grad_norm": 0.9326318502426147, + "learning_rate": 4.5769158742528494e-06, + "loss": 0.6611, + "step": 1442 + }, + { + "epoch": 1.1946438431805633, + "grad_norm": 0.9354713559150696, + "learning_rate": 4.576303562801392e-06, + "loss": 0.6445, + "step": 1443 + }, + { + "epoch": 1.195472114853672, + "grad_norm": 0.8868383169174194, + "learning_rate": 4.57569084960455e-06, + "loss": 0.6344, + "step": 1444 + }, + { + "epoch": 1.1963003865267807, + "grad_norm": 0.893924355506897, + "learning_rate": 4.5750777347808775e-06, + "loss": 0.602, + "step": 1445 + }, + { + "epoch": 1.1971286581998895, + "grad_norm": 0.9299339652061462, + "learning_rate": 4.574464218449006e-06, + "loss": 0.6206, + "step": 1446 + }, + { + "epoch": 1.1979569298729984, + "grad_norm": 0.9058142304420471, + "learning_rate": 4.573850300727648e-06, + "loss": 0.6073, + "step": 1447 + }, + { + "epoch": 1.1987852015461071, + "grad_norm": 0.8893733620643616, + "learning_rate": 4.5732359817355885e-06, + "loss": 0.6336, + "step": 1448 + }, + { + "epoch": 1.1996134732192159, + "grad_norm": 0.9018000960350037, + "learning_rate": 4.572621261591692e-06, + "loss": 0.6268, + "step": 1449 + }, + { + "epoch": 1.2004417448923248, + "grad_norm": 0.9087574481964111, + "learning_rate": 4.572006140414903e-06, + "loss": 0.6511, + "step": 1450 + }, + { + "epoch": 1.2012700165654335, + "grad_norm": 0.9092984795570374, + "learning_rate": 4.5713906183242404e-06, + "loss": 0.6424, + "step": 1451 + }, + { + "epoch": 1.2020982882385423, + "grad_norm": 0.9040535092353821, + "learning_rate": 4.5707746954388034e-06, + "loss": 0.6242, + "step": 1452 + }, + { + "epoch": 1.202926559911651, + "grad_norm": 0.891659677028656, + "learning_rate": 4.570158371877766e-06, + "loss": 0.643, + "step": 1453 + }, + { + "epoch": 1.2037548315847597, + "grad_norm": 0.9254544973373413, + "learning_rate": 4.569541647760381e-06, + "loss": 0.6191, + "step": 1454 + }, + { + "epoch": 1.2045831032578687, + "grad_norm": 0.8978536128997803, + "learning_rate": 4.568924523205979e-06, + "loss": 0.6632, + "step": 1455 + }, + { + "epoch": 1.2054113749309774, + "grad_norm": 0.9179201126098633, + "learning_rate": 4.568306998333968e-06, + "loss": 0.6485, + "step": 1456 + }, + { + "epoch": 1.2062396466040861, + "grad_norm": 0.9061882495880127, + "learning_rate": 4.567689073263834e-06, + "loss": 0.6704, + "step": 1457 + }, + { + "epoch": 1.2070679182771948, + "grad_norm": 0.9143427610397339, + "learning_rate": 4.567070748115139e-06, + "loss": 0.6453, + "step": 1458 + }, + { + "epoch": 1.2078961899503038, + "grad_norm": 0.8740119338035583, + "learning_rate": 4.566452023007523e-06, + "loss": 0.6453, + "step": 1459 + }, + { + "epoch": 1.2087244616234125, + "grad_norm": 0.8977463841438293, + "learning_rate": 4.565832898060703e-06, + "loss": 0.6451, + "step": 1460 + }, + { + "epoch": 1.2095527332965212, + "grad_norm": 0.9416611194610596, + "learning_rate": 4.565213373394476e-06, + "loss": 0.6265, + "step": 1461 + }, + { + "epoch": 1.21038100496963, + "grad_norm": 0.9132000207901001, + "learning_rate": 4.5645934491287115e-06, + "loss": 0.6412, + "step": 1462 + }, + { + "epoch": 1.2112092766427387, + "grad_norm": 0.9480498433113098, + "learning_rate": 4.563973125383362e-06, + "loss": 0.6363, + "step": 1463 + }, + { + "epoch": 1.2120375483158476, + "grad_norm": 0.9170709848403931, + "learning_rate": 4.563352402278451e-06, + "loss": 0.6201, + "step": 1464 + }, + { + "epoch": 1.2128658199889564, + "grad_norm": 0.9655918478965759, + "learning_rate": 4.562731279934086e-06, + "loss": 0.6497, + "step": 1465 + }, + { + "epoch": 1.213694091662065, + "grad_norm": 0.8885562419891357, + "learning_rate": 4.562109758470446e-06, + "loss": 0.6265, + "step": 1466 + }, + { + "epoch": 1.214522363335174, + "grad_norm": 0.9072791337966919, + "learning_rate": 4.561487838007792e-06, + "loss": 0.6257, + "step": 1467 + }, + { + "epoch": 1.2153506350082828, + "grad_norm": 0.8992362022399902, + "learning_rate": 4.560865518666458e-06, + "loss": 0.6425, + "step": 1468 + }, + { + "epoch": 1.2161789066813915, + "grad_norm": 0.8934991955757141, + "learning_rate": 4.560242800566855e-06, + "loss": 0.6583, + "step": 1469 + }, + { + "epoch": 1.2170071783545002, + "grad_norm": 0.9292325973510742, + "learning_rate": 4.559619683829477e-06, + "loss": 0.6403, + "step": 1470 + }, + { + "epoch": 1.217835450027609, + "grad_norm": 0.8756814002990723, + "learning_rate": 4.55899616857489e-06, + "loss": 0.6261, + "step": 1471 + }, + { + "epoch": 1.218663721700718, + "grad_norm": 0.9050565361976624, + "learning_rate": 4.558372254923738e-06, + "loss": 0.6325, + "step": 1472 + }, + { + "epoch": 1.2194919933738266, + "grad_norm": 0.9230931401252747, + "learning_rate": 4.557747942996742e-06, + "loss": 0.6275, + "step": 1473 + }, + { + "epoch": 1.2203202650469354, + "grad_norm": 0.894026517868042, + "learning_rate": 4.557123232914702e-06, + "loss": 0.6348, + "step": 1474 + }, + { + "epoch": 1.221148536720044, + "grad_norm": 0.8882637023925781, + "learning_rate": 4.556498124798492e-06, + "loss": 0.6123, + "step": 1475 + }, + { + "epoch": 1.221976808393153, + "grad_norm": 0.9191063642501831, + "learning_rate": 4.555872618769065e-06, + "loss": 0.6531, + "step": 1476 + }, + { + "epoch": 1.2228050800662618, + "grad_norm": 0.9398623704910278, + "learning_rate": 4.555246714947452e-06, + "loss": 0.6553, + "step": 1477 + }, + { + "epoch": 1.2236333517393705, + "grad_norm": 0.9399452805519104, + "learning_rate": 4.554620413454757e-06, + "loss": 0.6485, + "step": 1478 + }, + { + "epoch": 1.2244616234124792, + "grad_norm": 0.9001572728157043, + "learning_rate": 4.553993714412166e-06, + "loss": 0.6467, + "step": 1479 + }, + { + "epoch": 1.2252898950855882, + "grad_norm": 0.9266572594642639, + "learning_rate": 4.553366617940938e-06, + "loss": 0.6325, + "step": 1480 + }, + { + "epoch": 1.226118166758697, + "grad_norm": 0.928301215171814, + "learning_rate": 4.552739124162411e-06, + "loss": 0.6398, + "step": 1481 + }, + { + "epoch": 1.2269464384318056, + "grad_norm": 0.880954921245575, + "learning_rate": 4.552111233197999e-06, + "loss": 0.669, + "step": 1482 + }, + { + "epoch": 1.2277747101049143, + "grad_norm": 0.9089685082435608, + "learning_rate": 4.551482945169191e-06, + "loss": 0.62, + "step": 1483 + }, + { + "epoch": 1.2286029817780233, + "grad_norm": 0.9187189936637878, + "learning_rate": 4.550854260197559e-06, + "loss": 0.6381, + "step": 1484 + }, + { + "epoch": 1.229431253451132, + "grad_norm": 0.8782045245170593, + "learning_rate": 4.550225178404744e-06, + "loss": 0.6397, + "step": 1485 + }, + { + "epoch": 1.2302595251242407, + "grad_norm": 0.8798948526382446, + "learning_rate": 4.54959569991247e-06, + "loss": 0.6522, + "step": 1486 + }, + { + "epoch": 1.2310877967973495, + "grad_norm": 0.9257029294967651, + "learning_rate": 4.548965824842534e-06, + "loss": 0.6522, + "step": 1487 + }, + { + "epoch": 1.2319160684704582, + "grad_norm": 0.8923602104187012, + "learning_rate": 4.548335553316811e-06, + "loss": 0.6538, + "step": 1488 + }, + { + "epoch": 1.2327443401435672, + "grad_norm": 0.9071294069290161, + "learning_rate": 4.5477048854572524e-06, + "loss": 0.6305, + "step": 1489 + }, + { + "epoch": 1.2335726118166759, + "grad_norm": 0.9495804905891418, + "learning_rate": 4.547073821385888e-06, + "loss": 0.6398, + "step": 1490 + }, + { + "epoch": 1.2344008834897846, + "grad_norm": 0.9163299202919006, + "learning_rate": 4.54644236122482e-06, + "loss": 0.6454, + "step": 1491 + }, + { + "epoch": 1.2352291551628933, + "grad_norm": 0.9286475777626038, + "learning_rate": 4.545810505096233e-06, + "loss": 0.6345, + "step": 1492 + }, + { + "epoch": 1.2360574268360023, + "grad_norm": 0.9116879105567932, + "learning_rate": 4.545178253122384e-06, + "loss": 0.6451, + "step": 1493 + }, + { + "epoch": 1.236885698509111, + "grad_norm": 0.917826771736145, + "learning_rate": 4.544545605425607e-06, + "loss": 0.6374, + "step": 1494 + }, + { + "epoch": 1.2377139701822197, + "grad_norm": 0.9433108568191528, + "learning_rate": 4.543912562128316e-06, + "loss": 0.6321, + "step": 1495 + }, + { + "epoch": 1.2385422418553285, + "grad_norm": 0.9017132520675659, + "learning_rate": 4.543279123352996e-06, + "loss": 0.6341, + "step": 1496 + }, + { + "epoch": 1.2393705135284374, + "grad_norm": 0.9557158350944519, + "learning_rate": 4.542645289222214e-06, + "loss": 0.6364, + "step": 1497 + }, + { + "epoch": 1.2401987852015461, + "grad_norm": 0.9258891344070435, + "learning_rate": 4.542011059858609e-06, + "loss": 0.6598, + "step": 1498 + }, + { + "epoch": 1.2410270568746549, + "grad_norm": 0.8988100290298462, + "learning_rate": 4.541376435384899e-06, + "loss": 0.6496, + "step": 1499 + }, + { + "epoch": 1.2418553285477636, + "grad_norm": 0.898154079914093, + "learning_rate": 4.540741415923878e-06, + "loss": 0.6486, + "step": 1500 + }, + { + "epoch": 1.2426836002208725, + "grad_norm": 0.9216037392616272, + "learning_rate": 4.540106001598418e-06, + "loss": 0.6436, + "step": 1501 + }, + { + "epoch": 1.2435118718939813, + "grad_norm": 0.9086095690727234, + "learning_rate": 4.539470192531463e-06, + "loss": 0.6287, + "step": 1502 + }, + { + "epoch": 1.24434014356709, + "grad_norm": 0.921291172504425, + "learning_rate": 4.538833988846039e-06, + "loss": 0.629, + "step": 1503 + }, + { + "epoch": 1.2451684152401987, + "grad_norm": 0.8621019721031189, + "learning_rate": 4.538197390665242e-06, + "loss": 0.6543, + "step": 1504 + }, + { + "epoch": 1.2459966869133074, + "grad_norm": 0.9523105025291443, + "learning_rate": 4.537560398112251e-06, + "loss": 0.6482, + "step": 1505 + }, + { + "epoch": 1.2468249585864164, + "grad_norm": 0.971931517124176, + "learning_rate": 4.536923011310317e-06, + "loss": 0.6463, + "step": 1506 + }, + { + "epoch": 1.2476532302595251, + "grad_norm": 0.927219033241272, + "learning_rate": 4.536285230382767e-06, + "loss": 0.6443, + "step": 1507 + }, + { + "epoch": 1.2484815019326339, + "grad_norm": 0.9433056116104126, + "learning_rate": 4.535647055453007e-06, + "loss": 0.6301, + "step": 1508 + }, + { + "epoch": 1.2493097736057428, + "grad_norm": 0.8992047905921936, + "learning_rate": 4.5350084866445195e-06, + "loss": 0.6177, + "step": 1509 + }, + { + "epoch": 1.2501380452788515, + "grad_norm": 0.8946895003318787, + "learning_rate": 4.534369524080858e-06, + "loss": 0.6275, + "step": 1510 + }, + { + "epoch": 1.2509663169519603, + "grad_norm": 0.8732156157493591, + "learning_rate": 4.5337301678856595e-06, + "loss": 0.6572, + "step": 1511 + }, + { + "epoch": 1.251794588625069, + "grad_norm": 0.9505333304405212, + "learning_rate": 4.533090418182631e-06, + "loss": 0.6352, + "step": 1512 + }, + { + "epoch": 1.2526228602981777, + "grad_norm": 0.8931122422218323, + "learning_rate": 4.532450275095558e-06, + "loss": 0.649, + "step": 1513 + }, + { + "epoch": 1.2534511319712867, + "grad_norm": 0.8781880140304565, + "learning_rate": 4.531809738748304e-06, + "loss": 0.651, + "step": 1514 + }, + { + "epoch": 1.2542794036443954, + "grad_norm": 0.9201871156692505, + "learning_rate": 4.531168809264805e-06, + "loss": 0.6451, + "step": 1515 + }, + { + "epoch": 1.2551076753175041, + "grad_norm": 0.9318148493766785, + "learning_rate": 4.530527486769077e-06, + "loss": 0.6458, + "step": 1516 + }, + { + "epoch": 1.2559359469906128, + "grad_norm": 0.9131407141685486, + "learning_rate": 4.529885771385208e-06, + "loss": 0.6679, + "step": 1517 + }, + { + "epoch": 1.2567642186637218, + "grad_norm": 0.8942316770553589, + "learning_rate": 4.529243663237365e-06, + "loss": 0.6262, + "step": 1518 + }, + { + "epoch": 1.2575924903368305, + "grad_norm": 0.9058274030685425, + "learning_rate": 4.528601162449788e-06, + "loss": 0.635, + "step": 1519 + }, + { + "epoch": 1.2584207620099392, + "grad_norm": 0.9129453897476196, + "learning_rate": 4.527958269146798e-06, + "loss": 0.6375, + "step": 1520 + }, + { + "epoch": 1.259249033683048, + "grad_norm": 0.8737504482269287, + "learning_rate": 4.527314983452787e-06, + "loss": 0.6437, + "step": 1521 + }, + { + "epoch": 1.2600773053561567, + "grad_norm": 0.9680130481719971, + "learning_rate": 4.526671305492225e-06, + "loss": 0.6546, + "step": 1522 + }, + { + "epoch": 1.2609055770292656, + "grad_norm": 0.9186084270477295, + "learning_rate": 4.526027235389658e-06, + "loss": 0.6505, + "step": 1523 + }, + { + "epoch": 1.2617338487023744, + "grad_norm": 0.9164871573448181, + "learning_rate": 4.525382773269706e-06, + "loss": 0.6541, + "step": 1524 + }, + { + "epoch": 1.262562120375483, + "grad_norm": 0.8994123935699463, + "learning_rate": 4.5247379192570695e-06, + "loss": 0.6313, + "step": 1525 + }, + { + "epoch": 1.263390392048592, + "grad_norm": 0.8780331015586853, + "learning_rate": 4.524092673476519e-06, + "loss": 0.6496, + "step": 1526 + }, + { + "epoch": 1.2642186637217008, + "grad_norm": 0.8670293092727661, + "learning_rate": 4.5234470360529045e-06, + "loss": 0.6422, + "step": 1527 + }, + { + "epoch": 1.2650469353948095, + "grad_norm": 0.9305918216705322, + "learning_rate": 4.522801007111152e-06, + "loss": 0.6352, + "step": 1528 + }, + { + "epoch": 1.2658752070679182, + "grad_norm": 0.9444066286087036, + "learning_rate": 4.52215458677626e-06, + "loss": 0.6357, + "step": 1529 + }, + { + "epoch": 1.266703478741027, + "grad_norm": 0.8987369537353516, + "learning_rate": 4.521507775173305e-06, + "loss": 0.6407, + "step": 1530 + }, + { + "epoch": 1.267531750414136, + "grad_norm": 0.9200116395950317, + "learning_rate": 4.5208605724274415e-06, + "loss": 0.6366, + "step": 1531 + }, + { + "epoch": 1.2683600220872446, + "grad_norm": 0.8680561780929565, + "learning_rate": 4.520212978663894e-06, + "loss": 0.6389, + "step": 1532 + }, + { + "epoch": 1.2691882937603534, + "grad_norm": 0.8928949236869812, + "learning_rate": 4.519564994007968e-06, + "loss": 0.6552, + "step": 1533 + }, + { + "epoch": 1.2700165654334623, + "grad_norm": 0.8986184000968933, + "learning_rate": 4.518916618585042e-06, + "loss": 0.6554, + "step": 1534 + }, + { + "epoch": 1.270844837106571, + "grad_norm": 0.9239423274993896, + "learning_rate": 4.51826785252057e-06, + "loss": 0.6448, + "step": 1535 + }, + { + "epoch": 1.2716731087796798, + "grad_norm": 0.9534674286842346, + "learning_rate": 4.517618695940082e-06, + "loss": 0.6303, + "step": 1536 + }, + { + "epoch": 1.2725013804527885, + "grad_norm": 0.9341598749160767, + "learning_rate": 4.516969148969186e-06, + "loss": 0.6335, + "step": 1537 + }, + { + "epoch": 1.2733296521258972, + "grad_norm": 0.9264217615127563, + "learning_rate": 4.51631921173356e-06, + "loss": 0.6543, + "step": 1538 + }, + { + "epoch": 1.274157923799006, + "grad_norm": 0.9460370540618896, + "learning_rate": 4.515668884358963e-06, + "loss": 0.6531, + "step": 1539 + }, + { + "epoch": 1.274986195472115, + "grad_norm": 0.9690954089164734, + "learning_rate": 4.515018166971227e-06, + "loss": 0.6495, + "step": 1540 + }, + { + "epoch": 1.2758144671452236, + "grad_norm": 0.8746740818023682, + "learning_rate": 4.514367059696259e-06, + "loss": 0.6169, + "step": 1541 + }, + { + "epoch": 1.2766427388183323, + "grad_norm": 0.9288576245307922, + "learning_rate": 4.513715562660043e-06, + "loss": 0.6459, + "step": 1542 + }, + { + "epoch": 1.2774710104914413, + "grad_norm": 0.9935420751571655, + "learning_rate": 4.513063675988637e-06, + "loss": 0.6471, + "step": 1543 + }, + { + "epoch": 1.27829928216455, + "grad_norm": 0.9584968686103821, + "learning_rate": 4.512411399808175e-06, + "loss": 0.6473, + "step": 1544 + }, + { + "epoch": 1.2791275538376587, + "grad_norm": 0.9049419164657593, + "learning_rate": 4.511758734244867e-06, + "loss": 0.6667, + "step": 1545 + }, + { + "epoch": 1.2799558255107675, + "grad_norm": 0.9456910490989685, + "learning_rate": 4.511105679424998e-06, + "loss": 0.625, + "step": 1546 + }, + { + "epoch": 1.2807840971838762, + "grad_norm": 0.9267060160636902, + "learning_rate": 4.510452235474926e-06, + "loss": 0.6239, + "step": 1547 + }, + { + "epoch": 1.2816123688569852, + "grad_norm": 0.9363065958023071, + "learning_rate": 4.509798402521088e-06, + "loss": 0.6416, + "step": 1548 + }, + { + "epoch": 1.2824406405300939, + "grad_norm": 0.85664302110672, + "learning_rate": 4.509144180689995e-06, + "loss": 0.6297, + "step": 1549 + }, + { + "epoch": 1.2832689122032026, + "grad_norm": 0.9298888444900513, + "learning_rate": 4.508489570108231e-06, + "loss": 0.6305, + "step": 1550 + }, + { + "epoch": 1.2840971838763116, + "grad_norm": 0.9485051035881042, + "learning_rate": 4.507834570902459e-06, + "loss": 0.6409, + "step": 1551 + }, + { + "epoch": 1.2849254555494203, + "grad_norm": 0.9129849076271057, + "learning_rate": 4.507179183199414e-06, + "loss": 0.6215, + "step": 1552 + }, + { + "epoch": 1.285753727222529, + "grad_norm": 0.9630280137062073, + "learning_rate": 4.506523407125907e-06, + "loss": 0.6229, + "step": 1553 + }, + { + "epoch": 1.2865819988956377, + "grad_norm": 0.9047848582267761, + "learning_rate": 4.505867242808826e-06, + "loss": 0.6434, + "step": 1554 + }, + { + "epoch": 1.2874102705687465, + "grad_norm": 0.9219448566436768, + "learning_rate": 4.5052106903751315e-06, + "loss": 0.6449, + "step": 1555 + }, + { + "epoch": 1.2882385422418554, + "grad_norm": 0.9347684383392334, + "learning_rate": 4.504553749951861e-06, + "loss": 0.6272, + "step": 1556 + }, + { + "epoch": 1.2890668139149641, + "grad_norm": 0.8930262923240662, + "learning_rate": 4.503896421666126e-06, + "loss": 0.6364, + "step": 1557 + }, + { + "epoch": 1.2898950855880729, + "grad_norm": 0.9067763090133667, + "learning_rate": 4.503238705645114e-06, + "loss": 0.6317, + "step": 1558 + }, + { + "epoch": 1.2907233572611816, + "grad_norm": 0.9524715542793274, + "learning_rate": 4.5025806020160865e-06, + "loss": 0.6525, + "step": 1559 + }, + { + "epoch": 1.2915516289342905, + "grad_norm": 0.927255392074585, + "learning_rate": 4.5019221109063795e-06, + "loss": 0.6343, + "step": 1560 + }, + { + "epoch": 1.2923799006073993, + "grad_norm": 0.904295802116394, + "learning_rate": 4.5012632324434065e-06, + "loss": 0.6484, + "step": 1561 + }, + { + "epoch": 1.293208172280508, + "grad_norm": 0.9138978123664856, + "learning_rate": 4.500603966754653e-06, + "loss": 0.6374, + "step": 1562 + }, + { + "epoch": 1.2940364439536167, + "grad_norm": 0.9038779139518738, + "learning_rate": 4.499944313967681e-06, + "loss": 0.6395, + "step": 1563 + }, + { + "epoch": 1.2948647156267254, + "grad_norm": 0.9218693971633911, + "learning_rate": 4.499284274210128e-06, + "loss": 0.6325, + "step": 1564 + }, + { + "epoch": 1.2956929872998344, + "grad_norm": 0.9210799932479858, + "learning_rate": 4.4986238476097055e-06, + "loss": 0.6476, + "step": 1565 + }, + { + "epoch": 1.2965212589729431, + "grad_norm": 0.8977149724960327, + "learning_rate": 4.4979630342941994e-06, + "loss": 0.6489, + "step": 1566 + }, + { + "epoch": 1.2973495306460519, + "grad_norm": 0.9423681497573853, + "learning_rate": 4.49730183439147e-06, + "loss": 0.6391, + "step": 1567 + }, + { + "epoch": 1.2981778023191608, + "grad_norm": 0.9404032826423645, + "learning_rate": 4.4966402480294545e-06, + "loss": 0.6233, + "step": 1568 + }, + { + "epoch": 1.2990060739922695, + "grad_norm": 0.9104741215705872, + "learning_rate": 4.495978275336164e-06, + "loss": 0.6369, + "step": 1569 + }, + { + "epoch": 1.2998343456653783, + "grad_norm": 0.8869790434837341, + "learning_rate": 4.4953159164396835e-06, + "loss": 0.6331, + "step": 1570 + }, + { + "epoch": 1.300662617338487, + "grad_norm": 0.9269986152648926, + "learning_rate": 4.494653171468173e-06, + "loss": 0.6361, + "step": 1571 + }, + { + "epoch": 1.3014908890115957, + "grad_norm": 0.9210803508758545, + "learning_rate": 4.4939900405498675e-06, + "loss": 0.6314, + "step": 1572 + }, + { + "epoch": 1.3023191606847047, + "grad_norm": 0.9426820874214172, + "learning_rate": 4.4933265238130775e-06, + "loss": 0.6625, + "step": 1573 + }, + { + "epoch": 1.3031474323578134, + "grad_norm": 0.9198150634765625, + "learning_rate": 4.492662621386186e-06, + "loss": 0.627, + "step": 1574 + }, + { + "epoch": 1.3039757040309221, + "grad_norm": 0.9312534928321838, + "learning_rate": 4.4919983333976525e-06, + "loss": 0.6386, + "step": 1575 + }, + { + "epoch": 1.304803975704031, + "grad_norm": 0.8926329016685486, + "learning_rate": 4.491333659976011e-06, + "loss": 0.6559, + "step": 1576 + }, + { + "epoch": 1.3056322473771398, + "grad_norm": 0.9172425270080566, + "learning_rate": 4.4906686012498705e-06, + "loss": 0.6254, + "step": 1577 + }, + { + "epoch": 1.3064605190502485, + "grad_norm": 0.9896116256713867, + "learning_rate": 4.490003157347911e-06, + "loss": 0.6331, + "step": 1578 + }, + { + "epoch": 1.3072887907233572, + "grad_norm": 0.9267039895057678, + "learning_rate": 4.489337328398891e-06, + "loss": 0.6296, + "step": 1579 + }, + { + "epoch": 1.308117062396466, + "grad_norm": 0.9033195376396179, + "learning_rate": 4.488671114531643e-06, + "loss": 0.6354, + "step": 1580 + }, + { + "epoch": 1.3089453340695747, + "grad_norm": 0.9047731161117554, + "learning_rate": 4.488004515875072e-06, + "loss": 0.6351, + "step": 1581 + }, + { + "epoch": 1.3097736057426836, + "grad_norm": 0.9096142649650574, + "learning_rate": 4.4873375325581596e-06, + "loss": 0.6504, + "step": 1582 + }, + { + "epoch": 1.3106018774157924, + "grad_norm": 0.8848153352737427, + "learning_rate": 4.48667016470996e-06, + "loss": 0.6354, + "step": 1583 + }, + { + "epoch": 1.311430149088901, + "grad_norm": 0.9036188721656799, + "learning_rate": 4.486002412459603e-06, + "loss": 0.6431, + "step": 1584 + }, + { + "epoch": 1.31225842076201, + "grad_norm": 0.8881173133850098, + "learning_rate": 4.4853342759362924e-06, + "loss": 0.6509, + "step": 1585 + }, + { + "epoch": 1.3130866924351188, + "grad_norm": 0.9195591807365417, + "learning_rate": 4.484665755269307e-06, + "loss": 0.6522, + "step": 1586 + }, + { + "epoch": 1.3139149641082275, + "grad_norm": 0.9021509885787964, + "learning_rate": 4.483996850587998e-06, + "loss": 0.6532, + "step": 1587 + }, + { + "epoch": 1.3147432357813362, + "grad_norm": 0.8902403116226196, + "learning_rate": 4.483327562021794e-06, + "loss": 0.5942, + "step": 1588 + }, + { + "epoch": 1.315571507454445, + "grad_norm": 0.8933504819869995, + "learning_rate": 4.482657889700194e-06, + "loss": 0.6305, + "step": 1589 + }, + { + "epoch": 1.316399779127554, + "grad_norm": 0.9233856201171875, + "learning_rate": 4.481987833752776e-06, + "loss": 0.6275, + "step": 1590 + }, + { + "epoch": 1.3172280508006626, + "grad_norm": 0.9482141733169556, + "learning_rate": 4.4813173943091865e-06, + "loss": 0.6232, + "step": 1591 + }, + { + "epoch": 1.3180563224737714, + "grad_norm": 0.94732266664505, + "learning_rate": 4.480646571499151e-06, + "loss": 0.6297, + "step": 1592 + }, + { + "epoch": 1.3188845941468803, + "grad_norm": 0.9066787958145142, + "learning_rate": 4.479975365452468e-06, + "loss": 0.6334, + "step": 1593 + }, + { + "epoch": 1.319712865819989, + "grad_norm": 0.9063938856124878, + "learning_rate": 4.479303776299008e-06, + "loss": 0.6415, + "step": 1594 + }, + { + "epoch": 1.3205411374930978, + "grad_norm": 0.9286819696426392, + "learning_rate": 4.478631804168719e-06, + "loss": 0.6534, + "step": 1595 + }, + { + "epoch": 1.3213694091662065, + "grad_norm": 0.9161240458488464, + "learning_rate": 4.47795944919162e-06, + "loss": 0.6401, + "step": 1596 + }, + { + "epoch": 1.3221976808393152, + "grad_norm": 0.956538200378418, + "learning_rate": 4.4772867114978056e-06, + "loss": 0.6649, + "step": 1597 + }, + { + "epoch": 1.323025952512424, + "grad_norm": 0.910504162311554, + "learning_rate": 4.4766135912174455e-06, + "loss": 0.6456, + "step": 1598 + }, + { + "epoch": 1.323854224185533, + "grad_norm": 0.9054026007652283, + "learning_rate": 4.4759400884807805e-06, + "loss": 0.6458, + "step": 1599 + }, + { + "epoch": 1.3246824958586416, + "grad_norm": 0.933506429195404, + "learning_rate": 4.475266203418127e-06, + "loss": 0.626, + "step": 1600 + }, + { + "epoch": 1.3255107675317503, + "grad_norm": 0.9042014479637146, + "learning_rate": 4.474591936159878e-06, + "loss": 0.6376, + "step": 1601 + }, + { + "epoch": 1.3263390392048593, + "grad_norm": 0.9005138278007507, + "learning_rate": 4.473917286836496e-06, + "loss": 0.6093, + "step": 1602 + }, + { + "epoch": 1.327167310877968, + "grad_norm": 0.9096264839172363, + "learning_rate": 4.4732422555785196e-06, + "loss": 0.6344, + "step": 1603 + }, + { + "epoch": 1.3279955825510767, + "grad_norm": 0.9171631336212158, + "learning_rate": 4.4725668425165605e-06, + "loss": 0.6454, + "step": 1604 + }, + { + "epoch": 1.3288238542241855, + "grad_norm": 0.8821735382080078, + "learning_rate": 4.4718910477813055e-06, + "loss": 0.6357, + "step": 1605 + }, + { + "epoch": 1.3296521258972942, + "grad_norm": 0.918499231338501, + "learning_rate": 4.471214871503514e-06, + "loss": 0.6361, + "step": 1606 + }, + { + "epoch": 1.3304803975704032, + "grad_norm": 0.9187904596328735, + "learning_rate": 4.470538313814021e-06, + "loss": 0.6362, + "step": 1607 + }, + { + "epoch": 1.3313086692435119, + "grad_norm": 0.9272632598876953, + "learning_rate": 4.469861374843734e-06, + "loss": 0.6659, + "step": 1608 + }, + { + "epoch": 1.3321369409166206, + "grad_norm": 0.8950392603874207, + "learning_rate": 4.469184054723632e-06, + "loss": 0.6551, + "step": 1609 + }, + { + "epoch": 1.3329652125897296, + "grad_norm": 0.9604961276054382, + "learning_rate": 4.468506353584773e-06, + "loss": 0.643, + "step": 1610 + }, + { + "epoch": 1.3337934842628383, + "grad_norm": 0.933073103427887, + "learning_rate": 4.467828271558286e-06, + "loss": 0.6193, + "step": 1611 + }, + { + "epoch": 1.334621755935947, + "grad_norm": 0.890964686870575, + "learning_rate": 4.467149808775371e-06, + "loss": 0.6423, + "step": 1612 + }, + { + "epoch": 1.3354500276090557, + "grad_norm": 0.8916820883750916, + "learning_rate": 4.466470965367306e-06, + "loss": 0.6399, + "step": 1613 + }, + { + "epoch": 1.3362782992821645, + "grad_norm": 0.9151459336280823, + "learning_rate": 4.465791741465439e-06, + "loss": 0.6439, + "step": 1614 + }, + { + "epoch": 1.3371065709552734, + "grad_norm": 0.9155884385108948, + "learning_rate": 4.465112137201197e-06, + "loss": 0.6516, + "step": 1615 + }, + { + "epoch": 1.3379348426283821, + "grad_norm": 0.9187855124473572, + "learning_rate": 4.464432152706073e-06, + "loss": 0.6263, + "step": 1616 + }, + { + "epoch": 1.3387631143014909, + "grad_norm": 0.9506946206092834, + "learning_rate": 4.46375178811164e-06, + "loss": 0.6586, + "step": 1617 + }, + { + "epoch": 1.3395913859745996, + "grad_norm": 0.9056413769721985, + "learning_rate": 4.463071043549543e-06, + "loss": 0.6301, + "step": 1618 + }, + { + "epoch": 1.3404196576477085, + "grad_norm": 0.9056457877159119, + "learning_rate": 4.462389919151498e-06, + "loss": 0.6352, + "step": 1619 + }, + { + "epoch": 1.3412479293208173, + "grad_norm": 0.9204022288322449, + "learning_rate": 4.461708415049297e-06, + "loss": 0.6446, + "step": 1620 + }, + { + "epoch": 1.342076200993926, + "grad_norm": 0.9411336183547974, + "learning_rate": 4.461026531374804e-06, + "loss": 0.6594, + "step": 1621 + }, + { + "epoch": 1.3429044726670347, + "grad_norm": 0.8956605792045593, + "learning_rate": 4.460344268259958e-06, + "loss": 0.6492, + "step": 1622 + }, + { + "epoch": 1.3437327443401434, + "grad_norm": 0.9010135531425476, + "learning_rate": 4.45966162583677e-06, + "loss": 0.6233, + "step": 1623 + }, + { + "epoch": 1.3445610160132524, + "grad_norm": 0.8948563933372498, + "learning_rate": 4.458978604237325e-06, + "loss": 0.6447, + "step": 1624 + }, + { + "epoch": 1.3453892876863611, + "grad_norm": 0.9227688312530518, + "learning_rate": 4.458295203593782e-06, + "loss": 0.6459, + "step": 1625 + }, + { + "epoch": 1.3462175593594699, + "grad_norm": 0.9532939195632935, + "learning_rate": 4.4576114240383725e-06, + "loss": 0.6724, + "step": 1626 + }, + { + "epoch": 1.3470458310325788, + "grad_norm": 0.9089633226394653, + "learning_rate": 4.4569272657034005e-06, + "loss": 0.6303, + "step": 1627 + }, + { + "epoch": 1.3478741027056875, + "grad_norm": 0.8903496861457825, + "learning_rate": 4.456242728721244e-06, + "loss": 0.6655, + "step": 1628 + }, + { + "epoch": 1.3487023743787963, + "grad_norm": 0.9009859561920166, + "learning_rate": 4.455557813224356e-06, + "loss": 0.6268, + "step": 1629 + }, + { + "epoch": 1.349530646051905, + "grad_norm": 0.9265439510345459, + "learning_rate": 4.454872519345261e-06, + "loss": 0.6586, + "step": 1630 + }, + { + "epoch": 1.3503589177250137, + "grad_norm": 0.8945439457893372, + "learning_rate": 4.454186847216556e-06, + "loss": 0.62, + "step": 1631 + }, + { + "epoch": 1.3511871893981227, + "grad_norm": 0.9313004016876221, + "learning_rate": 4.453500796970913e-06, + "loss": 0.6452, + "step": 1632 + }, + { + "epoch": 1.3520154610712314, + "grad_norm": 0.9005258083343506, + "learning_rate": 4.452814368741076e-06, + "loss": 0.6436, + "step": 1633 + }, + { + "epoch": 1.35284373274434, + "grad_norm": 0.8860014081001282, + "learning_rate": 4.452127562659864e-06, + "loss": 0.6423, + "step": 1634 + }, + { + "epoch": 1.353672004417449, + "grad_norm": 0.9049930572509766, + "learning_rate": 4.4514403788601645e-06, + "loss": 0.6343, + "step": 1635 + }, + { + "epoch": 1.3545002760905578, + "grad_norm": 0.8849244117736816, + "learning_rate": 4.4507528174749425e-06, + "loss": 0.6481, + "step": 1636 + }, + { + "epoch": 1.3553285477636665, + "grad_norm": 0.9214544892311096, + "learning_rate": 4.450064878637235e-06, + "loss": 0.6166, + "step": 1637 + }, + { + "epoch": 1.3561568194367752, + "grad_norm": 0.8948760628700256, + "learning_rate": 4.449376562480153e-06, + "loss": 0.6195, + "step": 1638 + }, + { + "epoch": 1.356985091109884, + "grad_norm": 0.9000245928764343, + "learning_rate": 4.448687869136878e-06, + "loss": 0.6378, + "step": 1639 + }, + { + "epoch": 1.3578133627829927, + "grad_norm": 0.8963801264762878, + "learning_rate": 4.447998798740665e-06, + "loss": 0.6118, + "step": 1640 + }, + { + "epoch": 1.3586416344561016, + "grad_norm": 0.8774341344833374, + "learning_rate": 4.447309351424843e-06, + "loss": 0.6184, + "step": 1641 + }, + { + "epoch": 1.3594699061292104, + "grad_norm": 0.8901870846748352, + "learning_rate": 4.446619527322815e-06, + "loss": 0.6499, + "step": 1642 + }, + { + "epoch": 1.360298177802319, + "grad_norm": 0.9108647108078003, + "learning_rate": 4.445929326568055e-06, + "loss": 0.6243, + "step": 1643 + }, + { + "epoch": 1.361126449475428, + "grad_norm": 0.9178787469863892, + "learning_rate": 4.445238749294109e-06, + "loss": 0.6458, + "step": 1644 + }, + { + "epoch": 1.3619547211485368, + "grad_norm": 0.9278320670127869, + "learning_rate": 4.444547795634598e-06, + "loss": 0.6281, + "step": 1645 + }, + { + "epoch": 1.3627829928216455, + "grad_norm": 0.8769298195838928, + "learning_rate": 4.443856465723216e-06, + "loss": 0.6387, + "step": 1646 + }, + { + "epoch": 1.3636112644947542, + "grad_norm": 0.9313247203826904, + "learning_rate": 4.443164759693729e-06, + "loss": 0.6304, + "step": 1647 + }, + { + "epoch": 1.364439536167863, + "grad_norm": 1.0037778615951538, + "learning_rate": 4.4424726776799745e-06, + "loss": 0.6378, + "step": 1648 + }, + { + "epoch": 1.365267807840972, + "grad_norm": 0.9090721011161804, + "learning_rate": 4.441780219815863e-06, + "loss": 0.6456, + "step": 1649 + }, + { + "epoch": 1.3660960795140806, + "grad_norm": 0.8968444466590881, + "learning_rate": 4.441087386235382e-06, + "loss": 0.6377, + "step": 1650 + }, + { + "epoch": 1.3669243511871894, + "grad_norm": 0.9638763070106506, + "learning_rate": 4.440394177072586e-06, + "loss": 0.651, + "step": 1651 + }, + { + "epoch": 1.3677526228602983, + "grad_norm": 0.9224078059196472, + "learning_rate": 4.439700592461604e-06, + "loss": 0.6492, + "step": 1652 + }, + { + "epoch": 1.368580894533407, + "grad_norm": 0.891180157661438, + "learning_rate": 4.439006632536639e-06, + "loss": 0.6402, + "step": 1653 + }, + { + "epoch": 1.3694091662065158, + "grad_norm": 0.8817652463912964, + "learning_rate": 4.438312297431967e-06, + "loss": 0.6095, + "step": 1654 + }, + { + "epoch": 1.3702374378796245, + "grad_norm": 0.8814468383789062, + "learning_rate": 4.437617587281932e-06, + "loss": 0.6334, + "step": 1655 + }, + { + "epoch": 1.3710657095527332, + "grad_norm": 0.8895065784454346, + "learning_rate": 4.436922502220958e-06, + "loss": 0.6163, + "step": 1656 + }, + { + "epoch": 1.371893981225842, + "grad_norm": 0.9043355584144592, + "learning_rate": 4.436227042383535e-06, + "loss": 0.6625, + "step": 1657 + }, + { + "epoch": 1.372722252898951, + "grad_norm": 0.942014753818512, + "learning_rate": 4.435531207904229e-06, + "loss": 0.6298, + "step": 1658 + }, + { + "epoch": 1.3735505245720596, + "grad_norm": 0.8830228447914124, + "learning_rate": 4.434834998917677e-06, + "loss": 0.6215, + "step": 1659 + }, + { + "epoch": 1.3743787962451683, + "grad_norm": 0.9138136506080627, + "learning_rate": 4.434138415558588e-06, + "loss": 0.63, + "step": 1660 + }, + { + "epoch": 1.3752070679182773, + "grad_norm": 0.8880951404571533, + "learning_rate": 4.433441457961746e-06, + "loss": 0.6411, + "step": 1661 + }, + { + "epoch": 1.376035339591386, + "grad_norm": 0.9435725808143616, + "learning_rate": 4.432744126262005e-06, + "loss": 0.6279, + "step": 1662 + }, + { + "epoch": 1.3768636112644947, + "grad_norm": 0.894296407699585, + "learning_rate": 4.432046420594292e-06, + "loss": 0.6298, + "step": 1663 + }, + { + "epoch": 1.3776918829376035, + "grad_norm": 0.8850218653678894, + "learning_rate": 4.431348341093608e-06, + "loss": 0.627, + "step": 1664 + }, + { + "epoch": 1.3785201546107122, + "grad_norm": 0.8956788182258606, + "learning_rate": 4.430649887895025e-06, + "loss": 0.6412, + "step": 1665 + }, + { + "epoch": 1.3793484262838211, + "grad_norm": 0.9128378033638, + "learning_rate": 4.429951061133683e-06, + "loss": 0.6208, + "step": 1666 + }, + { + "epoch": 1.3801766979569299, + "grad_norm": 0.8763598203659058, + "learning_rate": 4.429251860944803e-06, + "loss": 0.6317, + "step": 1667 + }, + { + "epoch": 1.3810049696300386, + "grad_norm": 0.9210475087165833, + "learning_rate": 4.428552287463672e-06, + "loss": 0.6321, + "step": 1668 + }, + { + "epoch": 1.3818332413031476, + "grad_norm": 0.9203745126724243, + "learning_rate": 4.427852340825651e-06, + "loss": 0.6442, + "step": 1669 + }, + { + "epoch": 1.3826615129762563, + "grad_norm": 0.9229859709739685, + "learning_rate": 4.4271520211661735e-06, + "loss": 0.6402, + "step": 1670 + }, + { + "epoch": 1.383489784649365, + "grad_norm": 0.9455426335334778, + "learning_rate": 4.426451328620744e-06, + "loss": 0.6402, + "step": 1671 + }, + { + "epoch": 1.3843180563224737, + "grad_norm": 0.8894345760345459, + "learning_rate": 4.425750263324942e-06, + "loss": 0.6429, + "step": 1672 + }, + { + "epoch": 1.3851463279955825, + "grad_norm": 0.871449887752533, + "learning_rate": 4.425048825414414e-06, + "loss": 0.6449, + "step": 1673 + }, + { + "epoch": 1.3859745996686914, + "grad_norm": 0.9102624654769897, + "learning_rate": 4.424347015024884e-06, + "loss": 0.634, + "step": 1674 + }, + { + "epoch": 1.3868028713418001, + "grad_norm": 0.881172776222229, + "learning_rate": 4.4236448322921445e-06, + "loss": 0.621, + "step": 1675 + }, + { + "epoch": 1.3876311430149089, + "grad_norm": 0.9143398404121399, + "learning_rate": 4.422942277352063e-06, + "loss": 0.6379, + "step": 1676 + }, + { + "epoch": 1.3884594146880176, + "grad_norm": 0.9155046343803406, + "learning_rate": 4.4222393503405754e-06, + "loss": 0.6168, + "step": 1677 + }, + { + "epoch": 1.3892876863611265, + "grad_norm": 0.9501062631607056, + "learning_rate": 4.421536051393694e-06, + "loss": 0.6238, + "step": 1678 + }, + { + "epoch": 1.3901159580342353, + "grad_norm": 0.9055907130241394, + "learning_rate": 4.420832380647498e-06, + "loss": 0.634, + "step": 1679 + }, + { + "epoch": 1.390944229707344, + "grad_norm": 0.9199142456054688, + "learning_rate": 4.420128338238141e-06, + "loss": 0.6283, + "step": 1680 + }, + { + "epoch": 1.3917725013804527, + "grad_norm": 0.8932124376296997, + "learning_rate": 4.419423924301851e-06, + "loss": 0.6146, + "step": 1681 + }, + { + "epoch": 1.3926007730535614, + "grad_norm": 0.9010130167007446, + "learning_rate": 4.418719138974924e-06, + "loss": 0.6409, + "step": 1682 + }, + { + "epoch": 1.3934290447266704, + "grad_norm": 0.9154212474822998, + "learning_rate": 4.41801398239373e-06, + "loss": 0.6115, + "step": 1683 + }, + { + "epoch": 1.3942573163997791, + "grad_norm": 0.9208158850669861, + "learning_rate": 4.417308454694709e-06, + "loss": 0.6242, + "step": 1684 + }, + { + "epoch": 1.3950855880728878, + "grad_norm": 0.932996928691864, + "learning_rate": 4.4166025560143756e-06, + "loss": 0.6277, + "step": 1685 + }, + { + "epoch": 1.3959138597459968, + "grad_norm": 0.8990349769592285, + "learning_rate": 4.415896286489314e-06, + "loss": 0.6128, + "step": 1686 + }, + { + "epoch": 1.3967421314191055, + "grad_norm": 0.926906168460846, + "learning_rate": 4.4151896462561805e-06, + "loss": 0.6442, + "step": 1687 + }, + { + "epoch": 1.3975704030922143, + "grad_norm": 0.9330859184265137, + "learning_rate": 4.414482635451704e-06, + "loss": 0.5995, + "step": 1688 + }, + { + "epoch": 1.398398674765323, + "grad_norm": 0.9039576649665833, + "learning_rate": 4.413775254212683e-06, + "loss": 0.6363, + "step": 1689 + }, + { + "epoch": 1.3992269464384317, + "grad_norm": 0.9228506684303284, + "learning_rate": 4.413067502675992e-06, + "loss": 0.6294, + "step": 1690 + }, + { + "epoch": 1.4000552181115407, + "grad_norm": 0.921620786190033, + "learning_rate": 4.412359380978571e-06, + "loss": 0.64, + "step": 1691 + }, + { + "epoch": 1.4008834897846494, + "grad_norm": 0.9379149079322815, + "learning_rate": 4.4116508892574375e-06, + "loss": 0.6207, + "step": 1692 + }, + { + "epoch": 1.401711761457758, + "grad_norm": 1.0528016090393066, + "learning_rate": 4.410942027649677e-06, + "loss": 0.6204, + "step": 1693 + }, + { + "epoch": 1.402540033130867, + "grad_norm": 0.9510579705238342, + "learning_rate": 4.410232796292447e-06, + "loss": 0.6314, + "step": 1694 + }, + { + "epoch": 1.4033683048039758, + "grad_norm": 0.8922870755195618, + "learning_rate": 4.409523195322979e-06, + "loss": 0.6056, + "step": 1695 + }, + { + "epoch": 1.4041965764770845, + "grad_norm": 0.9239864349365234, + "learning_rate": 4.408813224878573e-06, + "loss": 0.6404, + "step": 1696 + }, + { + "epoch": 1.4050248481501932, + "grad_norm": 0.8696756362915039, + "learning_rate": 4.408102885096601e-06, + "loss": 0.6553, + "step": 1697 + }, + { + "epoch": 1.405853119823302, + "grad_norm": 0.9176343679428101, + "learning_rate": 4.407392176114509e-06, + "loss": 0.6286, + "step": 1698 + }, + { + "epoch": 1.4066813914964107, + "grad_norm": 0.9825496077537537, + "learning_rate": 4.40668109806981e-06, + "loss": 0.6421, + "step": 1699 + }, + { + "epoch": 1.4075096631695196, + "grad_norm": 0.9334873557090759, + "learning_rate": 4.405969651100093e-06, + "loss": 0.6347, + "step": 1700 + }, + { + "epoch": 1.4083379348426284, + "grad_norm": 0.8881192803382874, + "learning_rate": 4.405257835343015e-06, + "loss": 0.6377, + "step": 1701 + }, + { + "epoch": 1.409166206515737, + "grad_norm": 0.8896775245666504, + "learning_rate": 4.404545650936307e-06, + "loss": 0.624, + "step": 1702 + }, + { + "epoch": 1.409994478188846, + "grad_norm": 0.9361454844474792, + "learning_rate": 4.40383309801777e-06, + "loss": 0.6561, + "step": 1703 + }, + { + "epoch": 1.4108227498619548, + "grad_norm": 0.9082345962524414, + "learning_rate": 4.403120176725276e-06, + "loss": 0.6415, + "step": 1704 + }, + { + "epoch": 1.4116510215350635, + "grad_norm": 0.9453962445259094, + "learning_rate": 4.402406887196767e-06, + "loss": 0.6443, + "step": 1705 + }, + { + "epoch": 1.4124792932081722, + "grad_norm": 0.9420861005783081, + "learning_rate": 4.401693229570261e-06, + "loss": 0.6544, + "step": 1706 + }, + { + "epoch": 1.413307564881281, + "grad_norm": 0.869742214679718, + "learning_rate": 4.400979203983841e-06, + "loss": 0.6072, + "step": 1707 + }, + { + "epoch": 1.41413583655439, + "grad_norm": 0.915322482585907, + "learning_rate": 4.4002648105756665e-06, + "loss": 0.6657, + "step": 1708 + }, + { + "epoch": 1.4149641082274986, + "grad_norm": 0.9029238224029541, + "learning_rate": 4.3995500494839656e-06, + "loss": 0.6157, + "step": 1709 + }, + { + "epoch": 1.4157923799006074, + "grad_norm": 0.8992076516151428, + "learning_rate": 4.398834920847037e-06, + "loss": 0.6321, + "step": 1710 + }, + { + "epoch": 1.4166206515737163, + "grad_norm": 0.9099612236022949, + "learning_rate": 4.398119424803252e-06, + "loss": 0.6123, + "step": 1711 + }, + { + "epoch": 1.417448923246825, + "grad_norm": 0.9081803560256958, + "learning_rate": 4.397403561491052e-06, + "loss": 0.6369, + "step": 1712 + }, + { + "epoch": 1.4182771949199338, + "grad_norm": 0.8873599767684937, + "learning_rate": 4.3966873310489505e-06, + "loss": 0.6503, + "step": 1713 + }, + { + "epoch": 1.4191054665930425, + "grad_norm": 0.8841976523399353, + "learning_rate": 4.395970733615531e-06, + "loss": 0.6303, + "step": 1714 + }, + { + "epoch": 1.4199337382661512, + "grad_norm": 0.8922518491744995, + "learning_rate": 4.395253769329449e-06, + "loss": 0.6105, + "step": 1715 + }, + { + "epoch": 1.42076200993926, + "grad_norm": 0.9023404717445374, + "learning_rate": 4.39453643832943e-06, + "loss": 0.6292, + "step": 1716 + }, + { + "epoch": 1.421590281612369, + "grad_norm": 0.9042630791664124, + "learning_rate": 4.39381874075427e-06, + "loss": 0.6675, + "step": 1717 + }, + { + "epoch": 1.4224185532854776, + "grad_norm": 0.8933060169219971, + "learning_rate": 4.393100676742838e-06, + "loss": 0.6125, + "step": 1718 + }, + { + "epoch": 1.4232468249585863, + "grad_norm": 0.8821312189102173, + "learning_rate": 4.392382246434073e-06, + "loss": 0.6266, + "step": 1719 + }, + { + "epoch": 1.4240750966316953, + "grad_norm": 0.904460608959198, + "learning_rate": 4.391663449966983e-06, + "loss": 0.5984, + "step": 1720 + }, + { + "epoch": 1.424903368304804, + "grad_norm": 0.8999350666999817, + "learning_rate": 4.39094428748065e-06, + "loss": 0.6653, + "step": 1721 + }, + { + "epoch": 1.4257316399779127, + "grad_norm": 0.9061364531517029, + "learning_rate": 4.390224759114224e-06, + "loss": 0.6306, + "step": 1722 + }, + { + "epoch": 1.4265599116510215, + "grad_norm": 0.8995294570922852, + "learning_rate": 4.389504865006927e-06, + "loss": 0.6656, + "step": 1723 + }, + { + "epoch": 1.4273881833241302, + "grad_norm": 0.904737114906311, + "learning_rate": 4.388784605298053e-06, + "loss": 0.6234, + "step": 1724 + }, + { + "epoch": 1.4282164549972391, + "grad_norm": 0.8827239871025085, + "learning_rate": 4.388063980126966e-06, + "loss": 0.6418, + "step": 1725 + }, + { + "epoch": 1.4290447266703479, + "grad_norm": 0.8791747689247131, + "learning_rate": 4.387342989633097e-06, + "loss": 0.6174, + "step": 1726 + }, + { + "epoch": 1.4298729983434566, + "grad_norm": 0.8831204175949097, + "learning_rate": 4.386621633955955e-06, + "loss": 0.6222, + "step": 1727 + }, + { + "epoch": 1.4307012700165656, + "grad_norm": 0.908511221408844, + "learning_rate": 4.3858999132351124e-06, + "loss": 0.6373, + "step": 1728 + }, + { + "epoch": 1.4315295416896743, + "grad_norm": 0.900497555732727, + "learning_rate": 4.385177827610218e-06, + "loss": 0.6291, + "step": 1729 + }, + { + "epoch": 1.432357813362783, + "grad_norm": 0.930695652961731, + "learning_rate": 4.384455377220986e-06, + "loss": 0.6408, + "step": 1730 + }, + { + "epoch": 1.4331860850358917, + "grad_norm": 0.9583060145378113, + "learning_rate": 4.383732562207206e-06, + "loss": 0.6452, + "step": 1731 + }, + { + "epoch": 1.4340143567090005, + "grad_norm": 0.9173291921615601, + "learning_rate": 4.383009382708734e-06, + "loss": 0.6491, + "step": 1732 + }, + { + "epoch": 1.4348426283821094, + "grad_norm": 0.9054985642433167, + "learning_rate": 4.382285838865501e-06, + "loss": 0.6179, + "step": 1733 + }, + { + "epoch": 1.4356709000552181, + "grad_norm": 0.9162827730178833, + "learning_rate": 4.381561930817504e-06, + "loss": 0.6428, + "step": 1734 + }, + { + "epoch": 1.4364991717283269, + "grad_norm": 0.9013632535934448, + "learning_rate": 4.380837658704813e-06, + "loss": 0.6385, + "step": 1735 + }, + { + "epoch": 1.4373274434014356, + "grad_norm": 0.9274044632911682, + "learning_rate": 4.3801130226675685e-06, + "loss": 0.6508, + "step": 1736 + }, + { + "epoch": 1.4381557150745445, + "grad_norm": 0.8884710669517517, + "learning_rate": 4.37938802284598e-06, + "loss": 0.6675, + "step": 1737 + }, + { + "epoch": 1.4389839867476533, + "grad_norm": 0.9156724214553833, + "learning_rate": 4.378662659380329e-06, + "loss": 0.6424, + "step": 1738 + }, + { + "epoch": 1.439812258420762, + "grad_norm": 0.9267123341560364, + "learning_rate": 4.377936932410966e-06, + "loss": 0.6333, + "step": 1739 + }, + { + "epoch": 1.4406405300938707, + "grad_norm": 0.8859686851501465, + "learning_rate": 4.377210842078311e-06, + "loss": 0.6343, + "step": 1740 + }, + { + "epoch": 1.4414688017669794, + "grad_norm": 0.92928147315979, + "learning_rate": 4.376484388522858e-06, + "loss": 0.6124, + "step": 1741 + }, + { + "epoch": 1.4422970734400884, + "grad_norm": 0.8908365964889526, + "learning_rate": 4.375757571885169e-06, + "loss": 0.631, + "step": 1742 + }, + { + "epoch": 1.4431253451131971, + "grad_norm": 0.8956666588783264, + "learning_rate": 4.375030392305875e-06, + "loss": 0.6314, + "step": 1743 + }, + { + "epoch": 1.4439536167863058, + "grad_norm": 0.8836198449134827, + "learning_rate": 4.374302849925679e-06, + "loss": 0.6179, + "step": 1744 + }, + { + "epoch": 1.4447818884594148, + "grad_norm": 0.87999027967453, + "learning_rate": 4.373574944885354e-06, + "loss": 0.6226, + "step": 1745 + }, + { + "epoch": 1.4456101601325235, + "grad_norm": 0.9118010997772217, + "learning_rate": 4.3728466773257414e-06, + "loss": 0.6319, + "step": 1746 + }, + { + "epoch": 1.4464384318056323, + "grad_norm": 0.8989576101303101, + "learning_rate": 4.3721180473877555e-06, + "loss": 0.6049, + "step": 1747 + }, + { + "epoch": 1.447266703478741, + "grad_norm": 0.8896577954292297, + "learning_rate": 4.371389055212379e-06, + "loss": 0.6242, + "step": 1748 + }, + { + "epoch": 1.4480949751518497, + "grad_norm": 0.8993588089942932, + "learning_rate": 4.370659700940665e-06, + "loss": 0.6255, + "step": 1749 + }, + { + "epoch": 1.4489232468249587, + "grad_norm": 0.896486759185791, + "learning_rate": 4.369929984713738e-06, + "loss": 0.6354, + "step": 1750 + }, + { + "epoch": 1.4497515184980674, + "grad_norm": 0.9003145694732666, + "learning_rate": 4.369199906672789e-06, + "loss": 0.6464, + "step": 1751 + }, + { + "epoch": 1.450579790171176, + "grad_norm": 0.8885106444358826, + "learning_rate": 4.368469466959083e-06, + "loss": 0.6567, + "step": 1752 + }, + { + "epoch": 1.451408061844285, + "grad_norm": 0.8914943933486938, + "learning_rate": 4.367738665713953e-06, + "loss": 0.6784, + "step": 1753 + }, + { + "epoch": 1.4522363335173938, + "grad_norm": 0.933029055595398, + "learning_rate": 4.367007503078802e-06, + "loss": 0.6495, + "step": 1754 + }, + { + "epoch": 1.4530646051905025, + "grad_norm": 0.8788006901741028, + "learning_rate": 4.366275979195102e-06, + "loss": 0.6352, + "step": 1755 + }, + { + "epoch": 1.4538928768636112, + "grad_norm": 0.8768455982208252, + "learning_rate": 4.365544094204399e-06, + "loss": 0.6329, + "step": 1756 + }, + { + "epoch": 1.45472114853672, + "grad_norm": 0.9021796584129333, + "learning_rate": 4.364811848248303e-06, + "loss": 0.6616, + "step": 1757 + }, + { + "epoch": 1.4555494202098287, + "grad_norm": 0.9197986125946045, + "learning_rate": 4.364079241468499e-06, + "loss": 0.6248, + "step": 1758 + }, + { + "epoch": 1.4563776918829376, + "grad_norm": 0.9095743894577026, + "learning_rate": 4.3633462740067375e-06, + "loss": 0.6722, + "step": 1759 + }, + { + "epoch": 1.4572059635560464, + "grad_norm": 0.9114678502082825, + "learning_rate": 4.362612946004843e-06, + "loss": 0.6278, + "step": 1760 + }, + { + "epoch": 1.458034235229155, + "grad_norm": 0.8978061676025391, + "learning_rate": 4.3618792576047054e-06, + "loss": 0.633, + "step": 1761 + }, + { + "epoch": 1.458862506902264, + "grad_norm": 0.8578571677207947, + "learning_rate": 4.361145208948287e-06, + "loss": 0.6263, + "step": 1762 + }, + { + "epoch": 1.4596907785753728, + "grad_norm": 0.9333001375198364, + "learning_rate": 4.360410800177622e-06, + "loss": 0.6383, + "step": 1763 + }, + { + "epoch": 1.4605190502484815, + "grad_norm": 0.9448907971382141, + "learning_rate": 4.359676031434808e-06, + "loss": 0.6324, + "step": 1764 + }, + { + "epoch": 1.4613473219215902, + "grad_norm": 0.8806003928184509, + "learning_rate": 4.358940902862019e-06, + "loss": 0.6365, + "step": 1765 + }, + { + "epoch": 1.462175593594699, + "grad_norm": 0.9975590109825134, + "learning_rate": 4.358205414601493e-06, + "loss": 0.6436, + "step": 1766 + }, + { + "epoch": 1.463003865267808, + "grad_norm": 1.0076019763946533, + "learning_rate": 4.357469566795542e-06, + "loss": 0.6234, + "step": 1767 + }, + { + "epoch": 1.4638321369409166, + "grad_norm": 0.9254529476165771, + "learning_rate": 4.356733359586545e-06, + "loss": 0.6244, + "step": 1768 + }, + { + "epoch": 1.4646604086140254, + "grad_norm": 0.9129828810691833, + "learning_rate": 4.35599679311695e-06, + "loss": 0.6519, + "step": 1769 + }, + { + "epoch": 1.4654886802871343, + "grad_norm": 0.9141498804092407, + "learning_rate": 4.355259867529278e-06, + "loss": 0.6343, + "step": 1770 + }, + { + "epoch": 1.466316951960243, + "grad_norm": 0.920790433883667, + "learning_rate": 4.354522582966116e-06, + "loss": 0.6505, + "step": 1771 + }, + { + "epoch": 1.4671452236333518, + "grad_norm": 0.9683033227920532, + "learning_rate": 4.353784939570121e-06, + "loss": 0.6238, + "step": 1772 + }, + { + "epoch": 1.4679734953064605, + "grad_norm": 0.882957935333252, + "learning_rate": 4.3530469374840226e-06, + "loss": 0.6358, + "step": 1773 + }, + { + "epoch": 1.4688017669795692, + "grad_norm": 0.943648099899292, + "learning_rate": 4.352308576850615e-06, + "loss": 0.6274, + "step": 1774 + }, + { + "epoch": 1.4696300386526782, + "grad_norm": 0.9609524607658386, + "learning_rate": 4.351569857812764e-06, + "loss": 0.6368, + "step": 1775 + }, + { + "epoch": 1.4704583103257869, + "grad_norm": 0.9170458912849426, + "learning_rate": 4.3508307805134055e-06, + "loss": 0.6515, + "step": 1776 + }, + { + "epoch": 1.4712865819988956, + "grad_norm": 0.8956430554389954, + "learning_rate": 4.350091345095545e-06, + "loss": 0.6394, + "step": 1777 + }, + { + "epoch": 1.4721148536720043, + "grad_norm": 0.9145321846008301, + "learning_rate": 4.349351551702256e-06, + "loss": 0.6572, + "step": 1778 + }, + { + "epoch": 1.4729431253451133, + "grad_norm": 0.8807089328765869, + "learning_rate": 4.34861140047668e-06, + "loss": 0.6391, + "step": 1779 + }, + { + "epoch": 1.473771397018222, + "grad_norm": 0.8828510642051697, + "learning_rate": 4.347870891562032e-06, + "loss": 0.6583, + "step": 1780 + }, + { + "epoch": 1.4745996686913307, + "grad_norm": 0.9017962217330933, + "learning_rate": 4.34713002510159e-06, + "loss": 0.6337, + "step": 1781 + }, + { + "epoch": 1.4754279403644395, + "grad_norm": 0.8738356232643127, + "learning_rate": 4.346388801238708e-06, + "loss": 0.637, + "step": 1782 + }, + { + "epoch": 1.4762562120375482, + "grad_norm": 0.8805137872695923, + "learning_rate": 4.345647220116805e-06, + "loss": 0.6306, + "step": 1783 + }, + { + "epoch": 1.4770844837106571, + "grad_norm": 0.923599362373352, + "learning_rate": 4.34490528187937e-06, + "loss": 0.6254, + "step": 1784 + }, + { + "epoch": 1.4779127553837659, + "grad_norm": 0.9181063175201416, + "learning_rate": 4.344162986669961e-06, + "loss": 0.6257, + "step": 1785 + }, + { + "epoch": 1.4787410270568746, + "grad_norm": 0.9060764312744141, + "learning_rate": 4.343420334632204e-06, + "loss": 0.6326, + "step": 1786 + }, + { + "epoch": 1.4795692987299836, + "grad_norm": 0.9101755023002625, + "learning_rate": 4.342677325909798e-06, + "loss": 0.6319, + "step": 1787 + }, + { + "epoch": 1.4803975704030923, + "grad_norm": 0.9417787790298462, + "learning_rate": 4.341933960646505e-06, + "loss": 0.6147, + "step": 1788 + }, + { + "epoch": 1.481225842076201, + "grad_norm": 0.9138630628585815, + "learning_rate": 4.341190238986162e-06, + "loss": 0.6256, + "step": 1789 + }, + { + "epoch": 1.4820541137493097, + "grad_norm": 0.9004842638969421, + "learning_rate": 4.340446161072671e-06, + "loss": 0.6443, + "step": 1790 + }, + { + "epoch": 1.4828823854224185, + "grad_norm": 0.9181780219078064, + "learning_rate": 4.339701727050004e-06, + "loss": 0.6408, + "step": 1791 + }, + { + "epoch": 1.4837106570955274, + "grad_norm": 0.880736768245697, + "learning_rate": 4.338956937062203e-06, + "loss": 0.6436, + "step": 1792 + }, + { + "epoch": 1.4845389287686361, + "grad_norm": 0.9168913960456848, + "learning_rate": 4.338211791253376e-06, + "loss": 0.6391, + "step": 1793 + }, + { + "epoch": 1.4853672004417449, + "grad_norm": 0.942925751209259, + "learning_rate": 4.337466289767704e-06, + "loss": 0.6303, + "step": 1794 + }, + { + "epoch": 1.4861954721148538, + "grad_norm": 0.8768817186355591, + "learning_rate": 4.336720432749433e-06, + "loss": 0.6239, + "step": 1795 + }, + { + "epoch": 1.4870237437879625, + "grad_norm": 0.9388988614082336, + "learning_rate": 4.33597422034288e-06, + "loss": 0.6441, + "step": 1796 + }, + { + "epoch": 1.4878520154610713, + "grad_norm": 0.9257978200912476, + "learning_rate": 4.3352276526924295e-06, + "loss": 0.6373, + "step": 1797 + }, + { + "epoch": 1.48868028713418, + "grad_norm": 0.9225389957427979, + "learning_rate": 4.334480729942537e-06, + "loss": 0.6555, + "step": 1798 + }, + { + "epoch": 1.4895085588072887, + "grad_norm": 0.8992138504981995, + "learning_rate": 4.3337334522377235e-06, + "loss": 0.6375, + "step": 1799 + }, + { + "epoch": 1.4903368304803974, + "grad_norm": 0.9207738041877747, + "learning_rate": 4.332985819722582e-06, + "loss": 0.6439, + "step": 1800 + }, + { + "epoch": 1.4911651021535064, + "grad_norm": 0.8866941332817078, + "learning_rate": 4.3322378325417705e-06, + "loss": 0.6326, + "step": 1801 + }, + { + "epoch": 1.4919933738266151, + "grad_norm": 0.9069377779960632, + "learning_rate": 4.3314894908400175e-06, + "loss": 0.6193, + "step": 1802 + }, + { + "epoch": 1.4928216454997238, + "grad_norm": 0.8918569087982178, + "learning_rate": 4.3307407947621216e-06, + "loss": 0.6374, + "step": 1803 + }, + { + "epoch": 1.4936499171728328, + "grad_norm": 0.8993483781814575, + "learning_rate": 4.329991744452948e-06, + "loss": 0.6251, + "step": 1804 + }, + { + "epoch": 1.4944781888459415, + "grad_norm": 0.918445885181427, + "learning_rate": 4.32924234005743e-06, + "loss": 0.6494, + "step": 1805 + }, + { + "epoch": 1.4953064605190503, + "grad_norm": 0.9245757460594177, + "learning_rate": 4.328492581720573e-06, + "loss": 0.6172, + "step": 1806 + }, + { + "epoch": 1.496134732192159, + "grad_norm": 0.9166370034217834, + "learning_rate": 4.327742469587445e-06, + "loss": 0.622, + "step": 1807 + }, + { + "epoch": 1.4969630038652677, + "grad_norm": 0.9083957076072693, + "learning_rate": 4.326992003803187e-06, + "loss": 0.6205, + "step": 1808 + }, + { + "epoch": 1.4977912755383767, + "grad_norm": 0.8895595073699951, + "learning_rate": 4.326241184513008e-06, + "loss": 0.6245, + "step": 1809 + }, + { + "epoch": 1.4986195472114854, + "grad_norm": 0.9845629334449768, + "learning_rate": 4.325490011862184e-06, + "loss": 0.6163, + "step": 1810 + }, + { + "epoch": 1.499447818884594, + "grad_norm": 0.9501281976699829, + "learning_rate": 4.324738485996058e-06, + "loss": 0.6218, + "step": 1811 + }, + { + "epoch": 1.500276090557703, + "grad_norm": 0.9685792922973633, + "learning_rate": 4.323986607060047e-06, + "loss": 0.634, + "step": 1812 + }, + { + "epoch": 1.5011043622308118, + "grad_norm": 0.8959722518920898, + "learning_rate": 4.32323437519963e-06, + "loss": 0.6316, + "step": 1813 + }, + { + "epoch": 1.5019326339039205, + "grad_norm": 0.914463222026825, + "learning_rate": 4.322481790560356e-06, + "loss": 0.6258, + "step": 1814 + }, + { + "epoch": 1.5027609055770292, + "grad_norm": 0.9052320718765259, + "learning_rate": 4.321728853287846e-06, + "loss": 0.6345, + "step": 1815 + }, + { + "epoch": 1.503589177250138, + "grad_norm": 0.9306650161743164, + "learning_rate": 4.320975563527785e-06, + "loss": 0.5976, + "step": 1816 + }, + { + "epoch": 1.5044174489232467, + "grad_norm": 0.8930769562721252, + "learning_rate": 4.320221921425926e-06, + "loss": 0.6427, + "step": 1817 + }, + { + "epoch": 1.5052457205963556, + "grad_norm": 0.9205043911933899, + "learning_rate": 4.3194679271280935e-06, + "loss": 0.617, + "step": 1818 + }, + { + "epoch": 1.5060739922694644, + "grad_norm": 1.0371367931365967, + "learning_rate": 4.318713580780179e-06, + "loss": 0.6301, + "step": 1819 + }, + { + "epoch": 1.5069022639425733, + "grad_norm": 0.9442588686943054, + "learning_rate": 4.317958882528139e-06, + "loss": 0.6252, + "step": 1820 + }, + { + "epoch": 1.507730535615682, + "grad_norm": 0.9214493036270142, + "learning_rate": 4.317203832518003e-06, + "loss": 0.6326, + "step": 1821 + }, + { + "epoch": 1.5085588072887908, + "grad_norm": 0.9270440340042114, + "learning_rate": 4.316448430895866e-06, + "loss": 0.6377, + "step": 1822 + }, + { + "epoch": 1.5093870789618995, + "grad_norm": 0.8822277188301086, + "learning_rate": 4.315692677807889e-06, + "loss": 0.6277, + "step": 1823 + }, + { + "epoch": 1.5102153506350082, + "grad_norm": 0.8926153779029846, + "learning_rate": 4.314936573400305e-06, + "loss": 0.6439, + "step": 1824 + }, + { + "epoch": 1.511043622308117, + "grad_norm": 0.9214072227478027, + "learning_rate": 4.314180117819412e-06, + "loss": 0.6519, + "step": 1825 + }, + { + "epoch": 1.5118718939812257, + "grad_norm": 0.881522536277771, + "learning_rate": 4.313423311211578e-06, + "loss": 0.6213, + "step": 1826 + }, + { + "epoch": 1.5127001656543346, + "grad_norm": 0.8960263729095459, + "learning_rate": 4.312666153723238e-06, + "loss": 0.6592, + "step": 1827 + }, + { + "epoch": 1.5135284373274434, + "grad_norm": 0.9176173210144043, + "learning_rate": 4.3119086455008956e-06, + "loss": 0.6569, + "step": 1828 + }, + { + "epoch": 1.5143567090005523, + "grad_norm": 0.9312153458595276, + "learning_rate": 4.311150786691119e-06, + "loss": 0.6324, + "step": 1829 + }, + { + "epoch": 1.515184980673661, + "grad_norm": 0.9202299118041992, + "learning_rate": 4.310392577440549e-06, + "loss": 0.6381, + "step": 1830 + }, + { + "epoch": 1.5160132523467698, + "grad_norm": 0.8895224928855896, + "learning_rate": 4.309634017895893e-06, + "loss": 0.6491, + "step": 1831 + }, + { + "epoch": 1.5168415240198785, + "grad_norm": 0.9424769878387451, + "learning_rate": 4.308875108203922e-06, + "loss": 0.6342, + "step": 1832 + }, + { + "epoch": 1.5176697956929872, + "grad_norm": 0.8676947951316833, + "learning_rate": 4.30811584851148e-06, + "loss": 0.6287, + "step": 1833 + }, + { + "epoch": 1.518498067366096, + "grad_norm": 0.8764673471450806, + "learning_rate": 4.3073562389654764e-06, + "loss": 0.6123, + "step": 1834 + }, + { + "epoch": 1.5193263390392049, + "grad_norm": 0.9222663640975952, + "learning_rate": 4.306596279712889e-06, + "loss": 0.6433, + "step": 1835 + }, + { + "epoch": 1.5201546107123136, + "grad_norm": 0.8839907050132751, + "learning_rate": 4.305835970900762e-06, + "loss": 0.6416, + "step": 1836 + }, + { + "epoch": 1.5209828823854226, + "grad_norm": 0.9095149636268616, + "learning_rate": 4.305075312676209e-06, + "loss": 0.6417, + "step": 1837 + }, + { + "epoch": 1.5218111540585313, + "grad_norm": 0.8718581199645996, + "learning_rate": 4.304314305186409e-06, + "loss": 0.6174, + "step": 1838 + }, + { + "epoch": 1.52263942573164, + "grad_norm": 0.9424081444740295, + "learning_rate": 4.303552948578611e-06, + "loss": 0.6064, + "step": 1839 + }, + { + "epoch": 1.5234676974047487, + "grad_norm": 0.9170297384262085, + "learning_rate": 4.30279124300013e-06, + "loss": 0.6078, + "step": 1840 + }, + { + "epoch": 1.5242959690778575, + "grad_norm": 0.8814254999160767, + "learning_rate": 4.302029188598348e-06, + "loss": 0.6194, + "step": 1841 + }, + { + "epoch": 1.5251242407509662, + "grad_norm": 0.9248694181442261, + "learning_rate": 4.301266785520718e-06, + "loss": 0.6452, + "step": 1842 + }, + { + "epoch": 1.5259525124240751, + "grad_norm": 0.8757627010345459, + "learning_rate": 4.300504033914754e-06, + "loss": 0.6486, + "step": 1843 + }, + { + "epoch": 1.5267807840971839, + "grad_norm": 0.9229605793952942, + "learning_rate": 4.299740933928045e-06, + "loss": 0.6363, + "step": 1844 + }, + { + "epoch": 1.5276090557702926, + "grad_norm": 0.9192973971366882, + "learning_rate": 4.298977485708243e-06, + "loss": 0.6383, + "step": 1845 + }, + { + "epoch": 1.5284373274434015, + "grad_norm": 0.9071361422538757, + "learning_rate": 4.2982136894030655e-06, + "loss": 0.6541, + "step": 1846 + }, + { + "epoch": 1.5292655991165103, + "grad_norm": 0.9067169427871704, + "learning_rate": 4.297449545160303e-06, + "loss": 0.6317, + "step": 1847 + }, + { + "epoch": 1.530093870789619, + "grad_norm": 0.9047722816467285, + "learning_rate": 4.2966850531278086e-06, + "loss": 0.6365, + "step": 1848 + }, + { + "epoch": 1.5309221424627277, + "grad_norm": 0.8948541879653931, + "learning_rate": 4.295920213453504e-06, + "loss": 0.6268, + "step": 1849 + }, + { + "epoch": 1.5317504141358365, + "grad_norm": 0.9061904549598694, + "learning_rate": 4.295155026285382e-06, + "loss": 0.6271, + "step": 1850 + }, + { + "epoch": 1.5325786858089452, + "grad_norm": 0.9092041850090027, + "learning_rate": 4.2943894917714946e-06, + "loss": 0.6148, + "step": 1851 + }, + { + "epoch": 1.5334069574820541, + "grad_norm": 0.932331919670105, + "learning_rate": 4.293623610059968e-06, + "loss": 0.647, + "step": 1852 + }, + { + "epoch": 1.5342352291551629, + "grad_norm": 0.9136486649513245, + "learning_rate": 4.292857381298993e-06, + "loss": 0.6326, + "step": 1853 + }, + { + "epoch": 1.5350635008282718, + "grad_norm": 0.9760140180587769, + "learning_rate": 4.292090805636827e-06, + "loss": 0.6324, + "step": 1854 + }, + { + "epoch": 1.5358917725013805, + "grad_norm": 0.9075965881347656, + "learning_rate": 4.291323883221797e-06, + "loss": 0.6194, + "step": 1855 + }, + { + "epoch": 1.5367200441744893, + "grad_norm": 0.8804707527160645, + "learning_rate": 4.290556614202292e-06, + "loss": 0.6246, + "step": 1856 + }, + { + "epoch": 1.537548315847598, + "grad_norm": 0.9123021960258484, + "learning_rate": 4.2897889987267755e-06, + "loss": 0.6156, + "step": 1857 + }, + { + "epoch": 1.5383765875207067, + "grad_norm": 0.9396944642066956, + "learning_rate": 4.289021036943771e-06, + "loss": 0.6268, + "step": 1858 + }, + { + "epoch": 1.5392048591938154, + "grad_norm": 0.858566164970398, + "learning_rate": 4.288252729001874e-06, + "loss": 0.6237, + "step": 1859 + }, + { + "epoch": 1.5400331308669244, + "grad_norm": 0.899341344833374, + "learning_rate": 4.287484075049743e-06, + "loss": 0.5973, + "step": 1860 + }, + { + "epoch": 1.5408614025400331, + "grad_norm": 0.9021979570388794, + "learning_rate": 4.286715075236106e-06, + "loss": 0.6164, + "step": 1861 + }, + { + "epoch": 1.541689674213142, + "grad_norm": 0.9693799018859863, + "learning_rate": 4.285945729709757e-06, + "loss": 0.6409, + "step": 1862 + }, + { + "epoch": 1.5425179458862508, + "grad_norm": 0.9177312254905701, + "learning_rate": 4.285176038619559e-06, + "loss": 0.6276, + "step": 1863 + }, + { + "epoch": 1.5433462175593595, + "grad_norm": 0.9382041692733765, + "learning_rate": 4.284406002114438e-06, + "loss": 0.643, + "step": 1864 + }, + { + "epoch": 1.5441744892324683, + "grad_norm": 0.8778985142707825, + "learning_rate": 4.28363562034339e-06, + "loss": 0.6119, + "step": 1865 + }, + { + "epoch": 1.545002760905577, + "grad_norm": 0.8890271782875061, + "learning_rate": 4.282864893455476e-06, + "loss": 0.6388, + "step": 1866 + }, + { + "epoch": 1.5458310325786857, + "grad_norm": 0.9081838130950928, + "learning_rate": 4.282093821599824e-06, + "loss": 0.6464, + "step": 1867 + }, + { + "epoch": 1.5466593042517944, + "grad_norm": 0.8805630803108215, + "learning_rate": 4.281322404925631e-06, + "loss": 0.6533, + "step": 1868 + }, + { + "epoch": 1.5474875759249034, + "grad_norm": 0.8817034363746643, + "learning_rate": 4.280550643582158e-06, + "loss": 0.643, + "step": 1869 + }, + { + "epoch": 1.548315847598012, + "grad_norm": 0.9003176689147949, + "learning_rate": 4.279778537718733e-06, + "loss": 0.6322, + "step": 1870 + }, + { + "epoch": 1.549144119271121, + "grad_norm": 0.8982899785041809, + "learning_rate": 4.2790060874847525e-06, + "loss": 0.6456, + "step": 1871 + }, + { + "epoch": 1.5499723909442298, + "grad_norm": 0.8730716109275818, + "learning_rate": 4.278233293029677e-06, + "loss": 0.6428, + "step": 1872 + }, + { + "epoch": 1.5508006626173385, + "grad_norm": 0.9135062098503113, + "learning_rate": 4.277460154503036e-06, + "loss": 0.645, + "step": 1873 + }, + { + "epoch": 1.5516289342904472, + "grad_norm": 0.9091652631759644, + "learning_rate": 4.276686672054424e-06, + "loss": 0.6256, + "step": 1874 + }, + { + "epoch": 1.552457205963556, + "grad_norm": 0.9090970754623413, + "learning_rate": 4.275912845833504e-06, + "loss": 0.6404, + "step": 1875 + }, + { + "epoch": 1.5532854776366647, + "grad_norm": 0.91478031873703, + "learning_rate": 4.275138675990003e-06, + "loss": 0.6498, + "step": 1876 + }, + { + "epoch": 1.5541137493097736, + "grad_norm": 0.8979254961013794, + "learning_rate": 4.274364162673715e-06, + "loss": 0.6275, + "step": 1877 + }, + { + "epoch": 1.5549420209828824, + "grad_norm": 0.92198646068573, + "learning_rate": 4.273589306034503e-06, + "loss": 0.6585, + "step": 1878 + }, + { + "epoch": 1.5557702926559913, + "grad_norm": 0.9540958404541016, + "learning_rate": 4.272814106222293e-06, + "loss": 0.6111, + "step": 1879 + }, + { + "epoch": 1.5565985643291, + "grad_norm": 0.8989667892456055, + "learning_rate": 4.272038563387079e-06, + "loss": 0.6255, + "step": 1880 + }, + { + "epoch": 1.5574268360022088, + "grad_norm": 0.9551357626914978, + "learning_rate": 4.2712626776789224e-06, + "loss": 0.6164, + "step": 1881 + }, + { + "epoch": 1.5582551076753175, + "grad_norm": 0.9278994798660278, + "learning_rate": 4.270486449247949e-06, + "loss": 0.6376, + "step": 1882 + }, + { + "epoch": 1.5590833793484262, + "grad_norm": 0.927180826663971, + "learning_rate": 4.269709878244352e-06, + "loss": 0.6602, + "step": 1883 + }, + { + "epoch": 1.559911651021535, + "grad_norm": 0.932837724685669, + "learning_rate": 4.268932964818389e-06, + "loss": 0.6229, + "step": 1884 + }, + { + "epoch": 1.5607399226946437, + "grad_norm": 0.895219087600708, + "learning_rate": 4.26815570912039e-06, + "loss": 0.6192, + "step": 1885 + }, + { + "epoch": 1.5615681943677526, + "grad_norm": 0.9075414538383484, + "learning_rate": 4.2673781113007415e-06, + "loss": 0.6418, + "step": 1886 + }, + { + "epoch": 1.5623964660408614, + "grad_norm": 0.9130772948265076, + "learning_rate": 4.266600171509905e-06, + "loss": 0.6402, + "step": 1887 + }, + { + "epoch": 1.5632247377139703, + "grad_norm": 0.8892185688018799, + "learning_rate": 4.265821889898402e-06, + "loss": 0.6213, + "step": 1888 + }, + { + "epoch": 1.564053009387079, + "grad_norm": 0.8853991031646729, + "learning_rate": 4.265043266616825e-06, + "loss": 0.6357, + "step": 1889 + }, + { + "epoch": 1.5648812810601878, + "grad_norm": 0.9273191690444946, + "learning_rate": 4.26426430181583e-06, + "loss": 0.6311, + "step": 1890 + }, + { + "epoch": 1.5657095527332965, + "grad_norm": 0.8890771865844727, + "learning_rate": 4.263484995646138e-06, + "loss": 0.6053, + "step": 1891 + }, + { + "epoch": 1.5665378244064052, + "grad_norm": 0.9022220373153687, + "learning_rate": 4.262705348258538e-06, + "loss": 0.6256, + "step": 1892 + }, + { + "epoch": 1.567366096079514, + "grad_norm": 0.9201255440711975, + "learning_rate": 4.261925359803886e-06, + "loss": 0.6302, + "step": 1893 + }, + { + "epoch": 1.5681943677526229, + "grad_norm": 0.9675379991531372, + "learning_rate": 4.261145030433101e-06, + "loss": 0.602, + "step": 1894 + }, + { + "epoch": 1.5690226394257316, + "grad_norm": 0.9039328098297119, + "learning_rate": 4.26036436029717e-06, + "loss": 0.6477, + "step": 1895 + }, + { + "epoch": 1.5698509110988406, + "grad_norm": 0.911361813545227, + "learning_rate": 4.259583349547144e-06, + "loss": 0.6211, + "step": 1896 + }, + { + "epoch": 1.5706791827719493, + "grad_norm": 0.8901238441467285, + "learning_rate": 4.258801998334143e-06, + "loss": 0.6442, + "step": 1897 + }, + { + "epoch": 1.571507454445058, + "grad_norm": 0.9290407299995422, + "learning_rate": 4.258020306809351e-06, + "loss": 0.6382, + "step": 1898 + }, + { + "epoch": 1.5723357261181667, + "grad_norm": 0.9179307222366333, + "learning_rate": 4.257238275124017e-06, + "loss": 0.6513, + "step": 1899 + }, + { + "epoch": 1.5731639977912755, + "grad_norm": 0.9212455153465271, + "learning_rate": 4.256455903429459e-06, + "loss": 0.6257, + "step": 1900 + }, + { + "epoch": 1.5739922694643842, + "grad_norm": 0.9126132130622864, + "learning_rate": 4.2556731918770565e-06, + "loss": 0.6384, + "step": 1901 + }, + { + "epoch": 1.5748205411374931, + "grad_norm": 0.8735355734825134, + "learning_rate": 4.254890140618257e-06, + "loss": 0.6031, + "step": 1902 + }, + { + "epoch": 1.5756488128106019, + "grad_norm": 0.9375724196434021, + "learning_rate": 4.254106749804575e-06, + "loss": 0.6333, + "step": 1903 + }, + { + "epoch": 1.5764770844837108, + "grad_norm": 0.9451910853385925, + "learning_rate": 4.253323019587588e-06, + "loss": 0.6186, + "step": 1904 + }, + { + "epoch": 1.5773053561568195, + "grad_norm": 0.9468234777450562, + "learning_rate": 4.25253895011894e-06, + "loss": 0.6023, + "step": 1905 + }, + { + "epoch": 1.5781336278299283, + "grad_norm": 0.8941681385040283, + "learning_rate": 4.251754541550343e-06, + "loss": 0.634, + "step": 1906 + }, + { + "epoch": 1.578961899503037, + "grad_norm": 0.9508136510848999, + "learning_rate": 4.250969794033573e-06, + "loss": 0.6346, + "step": 1907 + }, + { + "epoch": 1.5797901711761457, + "grad_norm": 0.8964453935623169, + "learning_rate": 4.2501847077204685e-06, + "loss": 0.6298, + "step": 1908 + }, + { + "epoch": 1.5806184428492545, + "grad_norm": 0.8885358572006226, + "learning_rate": 4.249399282762938e-06, + "loss": 0.6319, + "step": 1909 + }, + { + "epoch": 1.5814467145223632, + "grad_norm": 0.8929858207702637, + "learning_rate": 4.248613519312955e-06, + "loss": 0.6272, + "step": 1910 + }, + { + "epoch": 1.5822749861954721, + "grad_norm": 0.9664214849472046, + "learning_rate": 4.247827417522557e-06, + "loss": 0.6336, + "step": 1911 + }, + { + "epoch": 1.5831032578685809, + "grad_norm": 0.890846312046051, + "learning_rate": 4.247040977543845e-06, + "loss": 0.6242, + "step": 1912 + }, + { + "epoch": 1.5839315295416898, + "grad_norm": 0.8973182439804077, + "learning_rate": 4.246254199528991e-06, + "loss": 0.6336, + "step": 1913 + }, + { + "epoch": 1.5847598012147985, + "grad_norm": 0.9789074659347534, + "learning_rate": 4.245467083630227e-06, + "loss": 0.6191, + "step": 1914 + }, + { + "epoch": 1.5855880728879073, + "grad_norm": 0.9195513129234314, + "learning_rate": 4.244679629999853e-06, + "loss": 0.6333, + "step": 1915 + }, + { + "epoch": 1.586416344561016, + "grad_norm": 0.950316846370697, + "learning_rate": 4.243891838790235e-06, + "loss": 0.635, + "step": 1916 + }, + { + "epoch": 1.5872446162341247, + "grad_norm": 0.8971677422523499, + "learning_rate": 4.243103710153802e-06, + "loss": 0.6352, + "step": 1917 + }, + { + "epoch": 1.5880728879072334, + "grad_norm": 0.9198834896087646, + "learning_rate": 4.24231524424305e-06, + "loss": 0.6252, + "step": 1918 + }, + { + "epoch": 1.5889011595803424, + "grad_norm": 0.9535729885101318, + "learning_rate": 4.241526441210541e-06, + "loss": 0.6177, + "step": 1919 + }, + { + "epoch": 1.5897294312534511, + "grad_norm": 0.8910471200942993, + "learning_rate": 4.240737301208899e-06, + "loss": 0.6408, + "step": 1920 + }, + { + "epoch": 1.59055770292656, + "grad_norm": 0.862991452217102, + "learning_rate": 4.2399478243908164e-06, + "loss": 0.6221, + "step": 1921 + }, + { + "epoch": 1.5913859745996688, + "grad_norm": 0.8935141563415527, + "learning_rate": 4.239158010909049e-06, + "loss": 0.608, + "step": 1922 + }, + { + "epoch": 1.5922142462727775, + "grad_norm": 0.9177968502044678, + "learning_rate": 4.23836786091642e-06, + "loss": 0.6442, + "step": 1923 + }, + { + "epoch": 1.5930425179458862, + "grad_norm": 0.9157758951187134, + "learning_rate": 4.237577374565813e-06, + "loss": 0.6419, + "step": 1924 + }, + { + "epoch": 1.593870789618995, + "grad_norm": 0.9484630227088928, + "learning_rate": 4.236786552010182e-06, + "loss": 0.6733, + "step": 1925 + }, + { + "epoch": 1.5946990612921037, + "grad_norm": 0.9401251077651978, + "learning_rate": 4.235995393402544e-06, + "loss": 0.6325, + "step": 1926 + }, + { + "epoch": 1.5955273329652124, + "grad_norm": 0.9104607105255127, + "learning_rate": 4.23520389889598e-06, + "loss": 0.6313, + "step": 1927 + }, + { + "epoch": 1.5963556046383214, + "grad_norm": 0.9173511862754822, + "learning_rate": 4.234412068643637e-06, + "loss": 0.6044, + "step": 1928 + }, + { + "epoch": 1.59718387631143, + "grad_norm": 0.9189788699150085, + "learning_rate": 4.233619902798727e-06, + "loss": 0.6182, + "step": 1929 + }, + { + "epoch": 1.598012147984539, + "grad_norm": 0.910804808139801, + "learning_rate": 4.232827401514525e-06, + "loss": 0.6435, + "step": 1930 + }, + { + "epoch": 1.5988404196576478, + "grad_norm": 0.8986726403236389, + "learning_rate": 4.232034564944375e-06, + "loss": 0.6268, + "step": 1931 + }, + { + "epoch": 1.5996686913307565, + "grad_norm": 0.8782910108566284, + "learning_rate": 4.231241393241682e-06, + "loss": 0.6381, + "step": 1932 + }, + { + "epoch": 1.6004969630038652, + "grad_norm": 0.9176427721977234, + "learning_rate": 4.230447886559919e-06, + "loss": 0.6131, + "step": 1933 + }, + { + "epoch": 1.601325234676974, + "grad_norm": 0.8755510449409485, + "learning_rate": 4.22965404505262e-06, + "loss": 0.654, + "step": 1934 + }, + { + "epoch": 1.6021535063500827, + "grad_norm": 0.8644832968711853, + "learning_rate": 4.228859868873387e-06, + "loss": 0.627, + "step": 1935 + }, + { + "epoch": 1.6029817780231916, + "grad_norm": 0.9181885719299316, + "learning_rate": 4.228065358175886e-06, + "loss": 0.6062, + "step": 1936 + }, + { + "epoch": 1.6038100496963004, + "grad_norm": 0.9515942335128784, + "learning_rate": 4.227270513113846e-06, + "loss": 0.6262, + "step": 1937 + }, + { + "epoch": 1.6046383213694093, + "grad_norm": 0.9062270522117615, + "learning_rate": 4.226475333841063e-06, + "loss": 0.6139, + "step": 1938 + }, + { + "epoch": 1.605466593042518, + "grad_norm": 0.9040980339050293, + "learning_rate": 4.225679820511398e-06, + "loss": 0.64, + "step": 1939 + }, + { + "epoch": 1.6062948647156268, + "grad_norm": 0.8910935521125793, + "learning_rate": 4.224883973278775e-06, + "loss": 0.6239, + "step": 1940 + }, + { + "epoch": 1.6071231363887355, + "grad_norm": 0.9150440692901611, + "learning_rate": 4.224087792297181e-06, + "loss": 0.6283, + "step": 1941 + }, + { + "epoch": 1.6079514080618442, + "grad_norm": 0.9175113439559937, + "learning_rate": 4.223291277720671e-06, + "loss": 0.6498, + "step": 1942 + }, + { + "epoch": 1.608779679734953, + "grad_norm": 0.8720189332962036, + "learning_rate": 4.222494429703363e-06, + "loss": 0.623, + "step": 1943 + }, + { + "epoch": 1.609607951408062, + "grad_norm": 0.8868989944458008, + "learning_rate": 4.22169724839944e-06, + "loss": 0.6357, + "step": 1944 + }, + { + "epoch": 1.6104362230811706, + "grad_norm": 0.9084488153457642, + "learning_rate": 4.22089973396315e-06, + "loss": 0.6607, + "step": 1945 + }, + { + "epoch": 1.6112644947542794, + "grad_norm": 0.9061319231987, + "learning_rate": 4.220101886548802e-06, + "loss": 0.6309, + "step": 1946 + }, + { + "epoch": 1.6120927664273883, + "grad_norm": 0.895646870136261, + "learning_rate": 4.219303706310775e-06, + "loss": 0.6276, + "step": 1947 + }, + { + "epoch": 1.612921038100497, + "grad_norm": 0.8863298892974854, + "learning_rate": 4.218505193403506e-06, + "loss": 0.6444, + "step": 1948 + }, + { + "epoch": 1.6137493097736058, + "grad_norm": 0.9005767107009888, + "learning_rate": 4.2177063479815025e-06, + "loss": 0.6356, + "step": 1949 + }, + { + "epoch": 1.6145775814467145, + "grad_norm": 0.9094089865684509, + "learning_rate": 4.216907170199334e-06, + "loss": 0.6053, + "step": 1950 + }, + { + "epoch": 1.6154058531198232, + "grad_norm": 0.8869677782058716, + "learning_rate": 4.216107660211632e-06, + "loss": 0.6308, + "step": 1951 + }, + { + "epoch": 1.616234124792932, + "grad_norm": 0.9285000562667847, + "learning_rate": 4.215307818173095e-06, + "loss": 0.6188, + "step": 1952 + }, + { + "epoch": 1.6170623964660409, + "grad_norm": 0.9039749503135681, + "learning_rate": 4.214507644238485e-06, + "loss": 0.63, + "step": 1953 + }, + { + "epoch": 1.6178906681391496, + "grad_norm": 0.9202860593795776, + "learning_rate": 4.213707138562628e-06, + "loss": 0.632, + "step": 1954 + }, + { + "epoch": 1.6187189398122586, + "grad_norm": 0.8759778141975403, + "learning_rate": 4.212906301300415e-06, + "loss": 0.6253, + "step": 1955 + }, + { + "epoch": 1.6195472114853673, + "grad_norm": 0.8918818831443787, + "learning_rate": 4.212105132606801e-06, + "loss": 0.625, + "step": 1956 + }, + { + "epoch": 1.620375483158476, + "grad_norm": 0.9184549450874329, + "learning_rate": 4.211303632636804e-06, + "loss": 0.6293, + "step": 1957 + }, + { + "epoch": 1.6212037548315847, + "grad_norm": 0.8887206315994263, + "learning_rate": 4.210501801545506e-06, + "loss": 0.6118, + "step": 1958 + }, + { + "epoch": 1.6220320265046935, + "grad_norm": 0.9010420441627502, + "learning_rate": 4.209699639488054e-06, + "loss": 0.6209, + "step": 1959 + }, + { + "epoch": 1.6228602981778022, + "grad_norm": 0.8888309001922607, + "learning_rate": 4.20889714661966e-06, + "loss": 0.6301, + "step": 1960 + }, + { + "epoch": 1.6236885698509111, + "grad_norm": 0.9068322777748108, + "learning_rate": 4.208094323095597e-06, + "loss": 0.6234, + "step": 1961 + }, + { + "epoch": 1.6245168415240199, + "grad_norm": 0.8940340876579285, + "learning_rate": 4.207291169071207e-06, + "loss": 0.6076, + "step": 1962 + }, + { + "epoch": 1.6253451131971288, + "grad_norm": 0.907363772392273, + "learning_rate": 4.20648768470189e-06, + "loss": 0.6522, + "step": 1963 + }, + { + "epoch": 1.6261733848702375, + "grad_norm": 0.9104205965995789, + "learning_rate": 4.205683870143114e-06, + "loss": 0.6177, + "step": 1964 + }, + { + "epoch": 1.6270016565433463, + "grad_norm": 0.9051296710968018, + "learning_rate": 4.204879725550409e-06, + "loss": 0.6069, + "step": 1965 + }, + { + "epoch": 1.627829928216455, + "grad_norm": 0.9034865498542786, + "learning_rate": 4.20407525107937e-06, + "loss": 0.6099, + "step": 1966 + }, + { + "epoch": 1.6286581998895637, + "grad_norm": 0.9055590033531189, + "learning_rate": 4.203270446885655e-06, + "loss": 0.6226, + "step": 1967 + }, + { + "epoch": 1.6294864715626725, + "grad_norm": 0.897927463054657, + "learning_rate": 4.202465313124986e-06, + "loss": 0.6342, + "step": 1968 + }, + { + "epoch": 1.6303147432357812, + "grad_norm": 0.8995800018310547, + "learning_rate": 4.2016598499531495e-06, + "loss": 0.6408, + "step": 1969 + }, + { + "epoch": 1.6311430149088901, + "grad_norm": 0.8727264404296875, + "learning_rate": 4.200854057525995e-06, + "loss": 0.6263, + "step": 1970 + }, + { + "epoch": 1.6319712865819989, + "grad_norm": 0.9289891719818115, + "learning_rate": 4.200047935999435e-06, + "loss": 0.6351, + "step": 1971 + }, + { + "epoch": 1.6327995582551078, + "grad_norm": 0.8936645984649658, + "learning_rate": 4.199241485529446e-06, + "loss": 0.6205, + "step": 1972 + }, + { + "epoch": 1.6336278299282165, + "grad_norm": 0.9054183959960938, + "learning_rate": 4.198434706272072e-06, + "loss": 0.6101, + "step": 1973 + }, + { + "epoch": 1.6344561016013253, + "grad_norm": 0.9181169271469116, + "learning_rate": 4.1976275983834135e-06, + "loss": 0.6437, + "step": 1974 + }, + { + "epoch": 1.635284373274434, + "grad_norm": 0.9325777888298035, + "learning_rate": 4.1968201620196404e-06, + "loss": 0.6343, + "step": 1975 + }, + { + "epoch": 1.6361126449475427, + "grad_norm": 0.878094494342804, + "learning_rate": 4.196012397336983e-06, + "loss": 0.6298, + "step": 1976 + }, + { + "epoch": 1.6369409166206514, + "grad_norm": 0.9336429238319397, + "learning_rate": 4.195204304491737e-06, + "loss": 0.6232, + "step": 1977 + }, + { + "epoch": 1.6377691882937604, + "grad_norm": 0.9159499406814575, + "learning_rate": 4.19439588364026e-06, + "loss": 0.621, + "step": 1978 + }, + { + "epoch": 1.6385974599668691, + "grad_norm": 0.9075011610984802, + "learning_rate": 4.193587134938975e-06, + "loss": 0.6375, + "step": 1979 + }, + { + "epoch": 1.639425731639978, + "grad_norm": 0.8912638425827026, + "learning_rate": 4.192778058544367e-06, + "loss": 0.6494, + "step": 1980 + }, + { + "epoch": 1.6402540033130868, + "grad_norm": 0.9125227928161621, + "learning_rate": 4.191968654612984e-06, + "loss": 0.6238, + "step": 1981 + }, + { + "epoch": 1.6410822749861955, + "grad_norm": 0.906326174736023, + "learning_rate": 4.191158923301439e-06, + "loss": 0.5904, + "step": 1982 + }, + { + "epoch": 1.6419105466593042, + "grad_norm": 0.8858581781387329, + "learning_rate": 4.190348864766407e-06, + "loss": 0.6222, + "step": 1983 + }, + { + "epoch": 1.642738818332413, + "grad_norm": 0.864389955997467, + "learning_rate": 4.189538479164627e-06, + "loss": 0.6218, + "step": 1984 + }, + { + "epoch": 1.6435670900055217, + "grad_norm": 0.8983911871910095, + "learning_rate": 4.188727766652901e-06, + "loss": 0.6511, + "step": 1985 + }, + { + "epoch": 1.6443953616786304, + "grad_norm": 0.9633772373199463, + "learning_rate": 4.187916727388094e-06, + "loss": 0.623, + "step": 1986 + }, + { + "epoch": 1.6452236333517394, + "grad_norm": 0.9191408753395081, + "learning_rate": 4.187105361527135e-06, + "loss": 0.6245, + "step": 1987 + }, + { + "epoch": 1.646051905024848, + "grad_norm": 0.907792866230011, + "learning_rate": 4.186293669227016e-06, + "loss": 0.5994, + "step": 1988 + }, + { + "epoch": 1.646880176697957, + "grad_norm": 0.906218409538269, + "learning_rate": 4.18548165064479e-06, + "loss": 0.653, + "step": 1989 + }, + { + "epoch": 1.6477084483710658, + "grad_norm": 0.9408275485038757, + "learning_rate": 4.184669305937577e-06, + "loss": 0.5978, + "step": 1990 + }, + { + "epoch": 1.6485367200441745, + "grad_norm": 0.9278435111045837, + "learning_rate": 4.183856635262558e-06, + "loss": 0.6167, + "step": 1991 + }, + { + "epoch": 1.6493649917172832, + "grad_norm": 0.91252601146698, + "learning_rate": 4.183043638776978e-06, + "loss": 0.6229, + "step": 1992 + }, + { + "epoch": 1.650193263390392, + "grad_norm": 0.9178313612937927, + "learning_rate": 4.182230316638142e-06, + "loss": 0.6318, + "step": 1993 + }, + { + "epoch": 1.6510215350635007, + "grad_norm": 0.9056826233863831, + "learning_rate": 4.181416669003421e-06, + "loss": 0.5932, + "step": 1994 + }, + { + "epoch": 1.6518498067366096, + "grad_norm": 0.9139112830162048, + "learning_rate": 4.180602696030248e-06, + "loss": 0.6407, + "step": 1995 + }, + { + "epoch": 1.6526780784097184, + "grad_norm": 0.9195094108581543, + "learning_rate": 4.179788397876121e-06, + "loss": 0.6371, + "step": 1996 + }, + { + "epoch": 1.6535063500828273, + "grad_norm": 0.9221140146255493, + "learning_rate": 4.178973774698597e-06, + "loss": 0.611, + "step": 1997 + }, + { + "epoch": 1.654334621755936, + "grad_norm": 0.9180639386177063, + "learning_rate": 4.178158826655299e-06, + "loss": 0.6138, + "step": 1998 + }, + { + "epoch": 1.6551628934290448, + "grad_norm": 0.9255500435829163, + "learning_rate": 4.177343553903912e-06, + "loss": 0.6425, + "step": 1999 + }, + { + "epoch": 1.6559911651021535, + "grad_norm": 0.9378511905670166, + "learning_rate": 4.176527956602182e-06, + "loss": 0.6357, + "step": 2000 + }, + { + "epoch": 1.6568194367752622, + "grad_norm": 0.9212007522583008, + "learning_rate": 4.175712034907922e-06, + "loss": 0.6307, + "step": 2001 + }, + { + "epoch": 1.657647708448371, + "grad_norm": 0.9185839891433716, + "learning_rate": 4.174895788979004e-06, + "loss": 0.6173, + "step": 2002 + }, + { + "epoch": 1.65847598012148, + "grad_norm": 0.9658777117729187, + "learning_rate": 4.174079218973364e-06, + "loss": 0.622, + "step": 2003 + }, + { + "epoch": 1.6593042517945886, + "grad_norm": 0.9310017824172974, + "learning_rate": 4.173262325049e-06, + "loss": 0.6407, + "step": 2004 + }, + { + "epoch": 1.6601325234676974, + "grad_norm": 0.9048784375190735, + "learning_rate": 4.172445107363974e-06, + "loss": 0.6557, + "step": 2005 + }, + { + "epoch": 1.6609607951408063, + "grad_norm": 0.8820594549179077, + "learning_rate": 4.17162756607641e-06, + "loss": 0.6077, + "step": 2006 + }, + { + "epoch": 1.661789066813915, + "grad_norm": 0.8961437344551086, + "learning_rate": 4.170809701344496e-06, + "loss": 0.6151, + "step": 2007 + }, + { + "epoch": 1.6626173384870238, + "grad_norm": 0.9773615598678589, + "learning_rate": 4.169991513326479e-06, + "loss": 0.6339, + "step": 2008 + }, + { + "epoch": 1.6634456101601325, + "grad_norm": 0.9309290647506714, + "learning_rate": 4.169173002180672e-06, + "loss": 0.6235, + "step": 2009 + }, + { + "epoch": 1.6642738818332412, + "grad_norm": 1.0016266107559204, + "learning_rate": 4.168354168065448e-06, + "loss": 0.6355, + "step": 2010 + }, + { + "epoch": 1.66510215350635, + "grad_norm": 0.8820549249649048, + "learning_rate": 4.167535011139246e-06, + "loss": 0.6258, + "step": 2011 + }, + { + "epoch": 1.6659304251794589, + "grad_norm": 0.9030908942222595, + "learning_rate": 4.166715531560564e-06, + "loss": 0.6217, + "step": 2012 + }, + { + "epoch": 1.6667586968525676, + "grad_norm": 0.983755350112915, + "learning_rate": 4.165895729487962e-06, + "loss": 0.6281, + "step": 2013 + }, + { + "epoch": 1.6675869685256766, + "grad_norm": 0.948001503944397, + "learning_rate": 4.165075605080066e-06, + "loss": 0.6242, + "step": 2014 + }, + { + "epoch": 1.6684152401987853, + "grad_norm": 0.9468982219696045, + "learning_rate": 4.164255158495563e-06, + "loss": 0.6563, + "step": 2015 + }, + { + "epoch": 1.669243511871894, + "grad_norm": 0.9012095332145691, + "learning_rate": 4.1634343898932e-06, + "loss": 0.618, + "step": 2016 + }, + { + "epoch": 1.6700717835450027, + "grad_norm": 0.9554924964904785, + "learning_rate": 4.162613299431789e-06, + "loss": 0.6306, + "step": 2017 + }, + { + "epoch": 1.6709000552181115, + "grad_norm": 0.9587515592575073, + "learning_rate": 4.161791887270203e-06, + "loss": 0.6305, + "step": 2018 + }, + { + "epoch": 1.6717283268912202, + "grad_norm": 0.8792864084243774, + "learning_rate": 4.160970153567379e-06, + "loss": 0.6181, + "step": 2019 + }, + { + "epoch": 1.6725565985643291, + "grad_norm": 0.8958392143249512, + "learning_rate": 4.160148098482313e-06, + "loss": 0.623, + "step": 2020 + }, + { + "epoch": 1.6733848702374379, + "grad_norm": 0.9501842260360718, + "learning_rate": 4.159325722174065e-06, + "loss": 0.6454, + "step": 2021 + }, + { + "epoch": 1.6742131419105468, + "grad_norm": 0.9107666015625, + "learning_rate": 4.158503024801759e-06, + "loss": 0.6237, + "step": 2022 + }, + { + "epoch": 1.6750414135836555, + "grad_norm": 0.9405720233917236, + "learning_rate": 4.157680006524577e-06, + "loss": 0.6423, + "step": 2023 + }, + { + "epoch": 1.6758696852567643, + "grad_norm": 0.9008631706237793, + "learning_rate": 4.1568566675017675e-06, + "loss": 0.6385, + "step": 2024 + }, + { + "epoch": 1.676697956929873, + "grad_norm": 0.9056677222251892, + "learning_rate": 4.156033007892637e-06, + "loss": 0.6225, + "step": 2025 + }, + { + "epoch": 1.6775262286029817, + "grad_norm": 0.899982750415802, + "learning_rate": 4.155209027856557e-06, + "loss": 0.6268, + "step": 2026 + }, + { + "epoch": 1.6783545002760905, + "grad_norm": 0.928506076335907, + "learning_rate": 4.15438472755296e-06, + "loss": 0.6343, + "step": 2027 + }, + { + "epoch": 1.6791827719491992, + "grad_norm": 0.8831804394721985, + "learning_rate": 4.15356010714134e-06, + "loss": 0.6404, + "step": 2028 + }, + { + "epoch": 1.6800110436223081, + "grad_norm": 0.8637033700942993, + "learning_rate": 4.152735166781253e-06, + "loss": 0.6305, + "step": 2029 + }, + { + "epoch": 1.6808393152954169, + "grad_norm": 0.9164902567863464, + "learning_rate": 4.151909906632319e-06, + "loss": 0.6171, + "step": 2030 + }, + { + "epoch": 1.6816675869685258, + "grad_norm": 0.9124466776847839, + "learning_rate": 4.151084326854216e-06, + "loss": 0.6446, + "step": 2031 + }, + { + "epoch": 1.6824958586416345, + "grad_norm": 0.9212199449539185, + "learning_rate": 4.150258427606687e-06, + "loss": 0.6074, + "step": 2032 + }, + { + "epoch": 1.6833241303147433, + "grad_norm": 0.9536449313163757, + "learning_rate": 4.149432209049536e-06, + "loss": 0.6149, + "step": 2033 + }, + { + "epoch": 1.684152401987852, + "grad_norm": 0.9232591390609741, + "learning_rate": 4.148605671342629e-06, + "loss": 0.6238, + "step": 2034 + }, + { + "epoch": 1.6849806736609607, + "grad_norm": 0.8844353556632996, + "learning_rate": 4.147778814645892e-06, + "loss": 0.6434, + "step": 2035 + }, + { + "epoch": 1.6858089453340694, + "grad_norm": 0.8500795960426331, + "learning_rate": 4.1469516391193155e-06, + "loss": 0.6304, + "step": 2036 + }, + { + "epoch": 1.6866372170071784, + "grad_norm": 0.8676083087921143, + "learning_rate": 4.146124144922949e-06, + "loss": 0.6158, + "step": 2037 + }, + { + "epoch": 1.6874654886802871, + "grad_norm": 0.8914940357208252, + "learning_rate": 4.1452963322169046e-06, + "loss": 0.6336, + "step": 2038 + }, + { + "epoch": 1.688293760353396, + "grad_norm": 0.9245527386665344, + "learning_rate": 4.144468201161358e-06, + "loss": 0.6385, + "step": 2039 + }, + { + "epoch": 1.6891220320265048, + "grad_norm": 0.8946509957313538, + "learning_rate": 4.143639751916544e-06, + "loss": 0.6243, + "step": 2040 + }, + { + "epoch": 1.6899503036996135, + "grad_norm": 0.882095217704773, + "learning_rate": 4.14281098464276e-06, + "loss": 0.6171, + "step": 2041 + }, + { + "epoch": 1.6907785753727222, + "grad_norm": 0.8981113433837891, + "learning_rate": 4.141981899500364e-06, + "loss": 0.6389, + "step": 2042 + }, + { + "epoch": 1.691606847045831, + "grad_norm": 0.928701639175415, + "learning_rate": 4.1411524966497775e-06, + "loss": 0.6488, + "step": 2043 + }, + { + "epoch": 1.6924351187189397, + "grad_norm": 0.8937084674835205, + "learning_rate": 4.140322776251481e-06, + "loss": 0.6415, + "step": 2044 + }, + { + "epoch": 1.6932633903920484, + "grad_norm": 0.8940730094909668, + "learning_rate": 4.1394927384660185e-06, + "loss": 0.6419, + "step": 2045 + }, + { + "epoch": 1.6940916620651574, + "grad_norm": 0.9067767262458801, + "learning_rate": 4.138662383453994e-06, + "loss": 0.608, + "step": 2046 + }, + { + "epoch": 1.694919933738266, + "grad_norm": 0.9123063087463379, + "learning_rate": 4.137831711376074e-06, + "loss": 0.633, + "step": 2047 + }, + { + "epoch": 1.695748205411375, + "grad_norm": 0.89399653673172, + "learning_rate": 4.137000722392985e-06, + "loss": 0.6338, + "step": 2048 + }, + { + "epoch": 1.6965764770844838, + "grad_norm": 0.9146568775177002, + "learning_rate": 4.136169416665517e-06, + "loss": 0.6214, + "step": 2049 + }, + { + "epoch": 1.6974047487575925, + "grad_norm": 0.9402440786361694, + "learning_rate": 4.135337794354518e-06, + "loss": 0.6266, + "step": 2050 + }, + { + "epoch": 1.6982330204307012, + "grad_norm": 0.8790574669837952, + "learning_rate": 4.134505855620902e-06, + "loss": 0.5989, + "step": 2051 + }, + { + "epoch": 1.69906129210381, + "grad_norm": 0.9421536326408386, + "learning_rate": 4.133673600625637e-06, + "loss": 0.611, + "step": 2052 + }, + { + "epoch": 1.6998895637769187, + "grad_norm": 0.9365546703338623, + "learning_rate": 4.132841029529762e-06, + "loss": 0.6201, + "step": 2053 + }, + { + "epoch": 1.7007178354500276, + "grad_norm": 0.9213638305664062, + "learning_rate": 4.132008142494367e-06, + "loss": 0.632, + "step": 2054 + }, + { + "epoch": 1.7015461071231364, + "grad_norm": 0.9204041957855225, + "learning_rate": 4.13117493968061e-06, + "loss": 0.6086, + "step": 2055 + }, + { + "epoch": 1.7023743787962453, + "grad_norm": 0.8972902297973633, + "learning_rate": 4.130341421249708e-06, + "loss": 0.656, + "step": 2056 + }, + { + "epoch": 1.703202650469354, + "grad_norm": 0.9163963794708252, + "learning_rate": 4.129507587362939e-06, + "loss": 0.623, + "step": 2057 + }, + { + "epoch": 1.7040309221424628, + "grad_norm": 0.9178527593612671, + "learning_rate": 4.128673438181642e-06, + "loss": 0.6327, + "step": 2058 + }, + { + "epoch": 1.7048591938155715, + "grad_norm": 0.8982755541801453, + "learning_rate": 4.127838973867217e-06, + "loss": 0.6165, + "step": 2059 + }, + { + "epoch": 1.7056874654886802, + "grad_norm": 0.8995897769927979, + "learning_rate": 4.127004194581125e-06, + "loss": 0.6204, + "step": 2060 + }, + { + "epoch": 1.706515737161789, + "grad_norm": 0.9148748517036438, + "learning_rate": 4.126169100484888e-06, + "loss": 0.6177, + "step": 2061 + }, + { + "epoch": 1.707344008834898, + "grad_norm": 0.9249931573867798, + "learning_rate": 4.12533369174009e-06, + "loss": 0.6113, + "step": 2062 + }, + { + "epoch": 1.7081722805080066, + "grad_norm": 0.9162423610687256, + "learning_rate": 4.124497968508372e-06, + "loss": 0.6119, + "step": 2063 + }, + { + "epoch": 1.7090005521811154, + "grad_norm": 0.8919503688812256, + "learning_rate": 4.123661930951441e-06, + "loss": 0.6194, + "step": 2064 + }, + { + "epoch": 1.7098288238542243, + "grad_norm": 0.8884986639022827, + "learning_rate": 4.122825579231063e-06, + "loss": 0.656, + "step": 2065 + }, + { + "epoch": 1.710657095527333, + "grad_norm": 0.9072293043136597, + "learning_rate": 4.121988913509062e-06, + "loss": 0.6351, + "step": 2066 + }, + { + "epoch": 1.7114853672004418, + "grad_norm": 0.9255547523498535, + "learning_rate": 4.121151933947327e-06, + "loss": 0.601, + "step": 2067 + }, + { + "epoch": 1.7123136388735505, + "grad_norm": 0.9141303300857544, + "learning_rate": 4.120314640707804e-06, + "loss": 0.6257, + "step": 2068 + }, + { + "epoch": 1.7131419105466592, + "grad_norm": 0.8791571855545044, + "learning_rate": 4.119477033952504e-06, + "loss": 0.6355, + "step": 2069 + }, + { + "epoch": 1.713970182219768, + "grad_norm": 0.8910380005836487, + "learning_rate": 4.118639113843493e-06, + "loss": 0.6405, + "step": 2070 + }, + { + "epoch": 1.7147984538928769, + "grad_norm": 0.9583408236503601, + "learning_rate": 4.117800880542904e-06, + "loss": 0.6408, + "step": 2071 + }, + { + "epoch": 1.7156267255659856, + "grad_norm": 0.9684842824935913, + "learning_rate": 4.116962334212924e-06, + "loss": 0.6155, + "step": 2072 + }, + { + "epoch": 1.7164549972390946, + "grad_norm": 0.8882633447647095, + "learning_rate": 4.116123475015806e-06, + "loss": 0.6332, + "step": 2073 + }, + { + "epoch": 1.7172832689122033, + "grad_norm": 0.8599907755851746, + "learning_rate": 4.1152843031138604e-06, + "loss": 0.6416, + "step": 2074 + }, + { + "epoch": 1.718111540585312, + "grad_norm": 0.9216133952140808, + "learning_rate": 4.114444818669461e-06, + "loss": 0.61, + "step": 2075 + }, + { + "epoch": 1.7189398122584207, + "grad_norm": 0.943729817867279, + "learning_rate": 4.113605021845037e-06, + "loss": 0.643, + "step": 2076 + }, + { + "epoch": 1.7197680839315295, + "grad_norm": 0.9158332943916321, + "learning_rate": 4.112764912803084e-06, + "loss": 0.6353, + "step": 2077 + }, + { + "epoch": 1.7205963556046382, + "grad_norm": 0.9036465287208557, + "learning_rate": 4.111924491706154e-06, + "loss": 0.6143, + "step": 2078 + }, + { + "epoch": 1.7214246272777471, + "grad_norm": 0.9264538884162903, + "learning_rate": 4.111083758716862e-06, + "loss": 0.6302, + "step": 2079 + }, + { + "epoch": 1.7222528989508559, + "grad_norm": 0.9309650659561157, + "learning_rate": 4.11024271399788e-06, + "loss": 0.6394, + "step": 2080 + }, + { + "epoch": 1.7230811706239648, + "grad_norm": 0.9261552095413208, + "learning_rate": 4.109401357711944e-06, + "loss": 0.6338, + "step": 2081 + }, + { + "epoch": 1.7239094422970735, + "grad_norm": 0.9027671217918396, + "learning_rate": 4.1085596900218476e-06, + "loss": 0.633, + "step": 2082 + }, + { + "epoch": 1.7247377139701823, + "grad_norm": 0.8899479508399963, + "learning_rate": 4.107717711090446e-06, + "loss": 0.6372, + "step": 2083 + }, + { + "epoch": 1.725565985643291, + "grad_norm": 0.9513740539550781, + "learning_rate": 4.106875421080654e-06, + "loss": 0.6057, + "step": 2084 + }, + { + "epoch": 1.7263942573163997, + "grad_norm": 0.8926359415054321, + "learning_rate": 4.106032820155447e-06, + "loss": 0.6219, + "step": 2085 + }, + { + "epoch": 1.7272225289895085, + "grad_norm": 0.8961042165756226, + "learning_rate": 4.105189908477862e-06, + "loss": 0.5926, + "step": 2086 + }, + { + "epoch": 1.7280508006626172, + "grad_norm": 0.8870651125907898, + "learning_rate": 4.104346686210992e-06, + "loss": 0.6298, + "step": 2087 + }, + { + "epoch": 1.7288790723357261, + "grad_norm": 0.9134538173675537, + "learning_rate": 4.103503153517993e-06, + "loss": 0.6544, + "step": 2088 + }, + { + "epoch": 1.7297073440088349, + "grad_norm": 0.900499165058136, + "learning_rate": 4.102659310562082e-06, + "loss": 0.6182, + "step": 2089 + }, + { + "epoch": 1.7305356156819438, + "grad_norm": 0.9146167039871216, + "learning_rate": 4.101815157506534e-06, + "loss": 0.6297, + "step": 2090 + }, + { + "epoch": 1.7313638873550525, + "grad_norm": 0.909670352935791, + "learning_rate": 4.100970694514687e-06, + "loss": 0.6272, + "step": 2091 + }, + { + "epoch": 1.7321921590281613, + "grad_norm": 1.127607822418213, + "learning_rate": 4.100125921749932e-06, + "loss": 0.6416, + "step": 2092 + }, + { + "epoch": 1.73302043070127, + "grad_norm": 0.8950872421264648, + "learning_rate": 4.099280839375729e-06, + "loss": 0.6411, + "step": 2093 + }, + { + "epoch": 1.7338487023743787, + "grad_norm": 0.9612913727760315, + "learning_rate": 4.098435447555591e-06, + "loss": 0.626, + "step": 2094 + }, + { + "epoch": 1.7346769740474874, + "grad_norm": 0.927638053894043, + "learning_rate": 4.097589746453094e-06, + "loss": 0.6197, + "step": 2095 + }, + { + "epoch": 1.7355052457205964, + "grad_norm": 0.9428266286849976, + "learning_rate": 4.096743736231875e-06, + "loss": 0.6407, + "step": 2096 + }, + { + "epoch": 1.7363335173937051, + "grad_norm": 0.8619174361228943, + "learning_rate": 4.095897417055627e-06, + "loss": 0.6386, + "step": 2097 + }, + { + "epoch": 1.737161789066814, + "grad_norm": 0.8943668007850647, + "learning_rate": 4.095050789088106e-06, + "loss": 0.6206, + "step": 2098 + }, + { + "epoch": 1.7379900607399228, + "grad_norm": 0.9190096259117126, + "learning_rate": 4.0942038524931245e-06, + "loss": 0.6395, + "step": 2099 + }, + { + "epoch": 1.7388183324130315, + "grad_norm": 0.9055671095848083, + "learning_rate": 4.09335660743456e-06, + "loss": 0.6384, + "step": 2100 + }, + { + "epoch": 1.7396466040861402, + "grad_norm": 0.9204139709472656, + "learning_rate": 4.092509054076345e-06, + "loss": 0.6438, + "step": 2101 + }, + { + "epoch": 1.740474875759249, + "grad_norm": 0.9572716355323792, + "learning_rate": 4.091661192582473e-06, + "loss": 0.6233, + "step": 2102 + }, + { + "epoch": 1.7413031474323577, + "grad_norm": 0.9479691982269287, + "learning_rate": 4.0908130231169975e-06, + "loss": 0.6284, + "step": 2103 + }, + { + "epoch": 1.7421314191054664, + "grad_norm": 0.9081270694732666, + "learning_rate": 4.089964545844032e-06, + "loss": 0.6284, + "step": 2104 + }, + { + "epoch": 1.7429596907785754, + "grad_norm": 0.9227046966552734, + "learning_rate": 4.0891157609277475e-06, + "loss": 0.6611, + "step": 2105 + }, + { + "epoch": 1.743787962451684, + "grad_norm": 0.9033496975898743, + "learning_rate": 4.088266668532377e-06, + "loss": 0.6122, + "step": 2106 + }, + { + "epoch": 1.744616234124793, + "grad_norm": 0.933539628982544, + "learning_rate": 4.087417268822211e-06, + "loss": 0.6167, + "step": 2107 + }, + { + "epoch": 1.7454445057979018, + "grad_norm": 0.9211486577987671, + "learning_rate": 4.086567561961602e-06, + "loss": 0.6364, + "step": 2108 + }, + { + "epoch": 1.7462727774710105, + "grad_norm": 0.8782284259796143, + "learning_rate": 4.08571754811496e-06, + "loss": 0.6164, + "step": 2109 + }, + { + "epoch": 1.7471010491441192, + "grad_norm": 0.8936499357223511, + "learning_rate": 4.084867227446753e-06, + "loss": 0.6155, + "step": 2110 + }, + { + "epoch": 1.747929320817228, + "grad_norm": 0.8931904435157776, + "learning_rate": 4.084016600121513e-06, + "loss": 0.6512, + "step": 2111 + }, + { + "epoch": 1.7487575924903367, + "grad_norm": 0.9070253372192383, + "learning_rate": 4.083165666303826e-06, + "loss": 0.6197, + "step": 2112 + }, + { + "epoch": 1.7495858641634456, + "grad_norm": 0.9232584238052368, + "learning_rate": 4.08231442615834e-06, + "loss": 0.6332, + "step": 2113 + }, + { + "epoch": 1.7504141358365544, + "grad_norm": 0.92015540599823, + "learning_rate": 4.081462879849765e-06, + "loss": 0.6351, + "step": 2114 + }, + { + "epoch": 1.7512424075096633, + "grad_norm": 0.9353263974189758, + "learning_rate": 4.080611027542863e-06, + "loss": 0.6479, + "step": 2115 + }, + { + "epoch": 1.752070679182772, + "grad_norm": 0.8943057060241699, + "learning_rate": 4.079758869402462e-06, + "loss": 0.6179, + "step": 2116 + }, + { + "epoch": 1.7528989508558808, + "grad_norm": 0.9062668681144714, + "learning_rate": 4.078906405593446e-06, + "loss": 0.6166, + "step": 2117 + }, + { + "epoch": 1.7537272225289895, + "grad_norm": 0.9368826150894165, + "learning_rate": 4.078053636280759e-06, + "loss": 0.6206, + "step": 2118 + }, + { + "epoch": 1.7545554942020982, + "grad_norm": 0.9012559652328491, + "learning_rate": 4.0772005616294034e-06, + "loss": 0.6445, + "step": 2119 + }, + { + "epoch": 1.755383765875207, + "grad_norm": 0.9061569571495056, + "learning_rate": 4.076347181804442e-06, + "loss": 0.6312, + "step": 2120 + }, + { + "epoch": 1.756212037548316, + "grad_norm": 0.892900288105011, + "learning_rate": 4.075493496970996e-06, + "loss": 0.64, + "step": 2121 + }, + { + "epoch": 1.7570403092214246, + "grad_norm": 0.9200263023376465, + "learning_rate": 4.074639507294243e-06, + "loss": 0.6435, + "step": 2122 + }, + { + "epoch": 1.7578685808945336, + "grad_norm": 0.921215295791626, + "learning_rate": 4.0737852129394265e-06, + "loss": 0.6135, + "step": 2123 + }, + { + "epoch": 1.7586968525676423, + "grad_norm": 0.8980262279510498, + "learning_rate": 4.07293061407184e-06, + "loss": 0.635, + "step": 2124 + }, + { + "epoch": 1.759525124240751, + "grad_norm": 0.9192932844161987, + "learning_rate": 4.072075710856843e-06, + "loss": 0.6317, + "step": 2125 + }, + { + "epoch": 1.7603533959138598, + "grad_norm": 0.9387120008468628, + "learning_rate": 4.07122050345985e-06, + "loss": 0.6324, + "step": 2126 + }, + { + "epoch": 1.7611816675869685, + "grad_norm": 0.8787287473678589, + "learning_rate": 4.070364992046338e-06, + "loss": 0.6177, + "step": 2127 + }, + { + "epoch": 1.7620099392600772, + "grad_norm": 0.8963474631309509, + "learning_rate": 4.069509176781838e-06, + "loss": 0.6158, + "step": 2128 + }, + { + "epoch": 1.762838210933186, + "grad_norm": 0.8926964402198792, + "learning_rate": 4.068653057831942e-06, + "loss": 0.6528, + "step": 2129 + }, + { + "epoch": 1.7636664826062949, + "grad_norm": 0.9290505647659302, + "learning_rate": 4.067796635362303e-06, + "loss": 0.6274, + "step": 2130 + }, + { + "epoch": 1.7644947542794036, + "grad_norm": 0.8848921060562134, + "learning_rate": 4.06693990953863e-06, + "loss": 0.6507, + "step": 2131 + }, + { + "epoch": 1.7653230259525126, + "grad_norm": 0.8977338075637817, + "learning_rate": 4.0660828805266915e-06, + "loss": 0.6437, + "step": 2132 + }, + { + "epoch": 1.7661512976256213, + "grad_norm": 0.9125422239303589, + "learning_rate": 4.065225548492314e-06, + "loss": 0.6221, + "step": 2133 + }, + { + "epoch": 1.76697956929873, + "grad_norm": 0.9186370968818665, + "learning_rate": 4.064367913601384e-06, + "loss": 0.6174, + "step": 2134 + }, + { + "epoch": 1.7678078409718387, + "grad_norm": 0.9084877967834473, + "learning_rate": 4.063509976019845e-06, + "loss": 0.6228, + "step": 2135 + }, + { + "epoch": 1.7686361126449475, + "grad_norm": 0.9180804491043091, + "learning_rate": 4.0626517359137e-06, + "loss": 0.6126, + "step": 2136 + }, + { + "epoch": 1.7694643843180562, + "grad_norm": 0.9099741578102112, + "learning_rate": 4.061793193449013e-06, + "loss": 0.6027, + "step": 2137 + }, + { + "epoch": 1.7702926559911651, + "grad_norm": 0.9060433506965637, + "learning_rate": 4.0609343487919005e-06, + "loss": 0.632, + "step": 2138 + }, + { + "epoch": 1.7711209276642739, + "grad_norm": 0.9404996037483215, + "learning_rate": 4.060075202108542e-06, + "loss": 0.6405, + "step": 2139 + }, + { + "epoch": 1.7719491993373828, + "grad_norm": 0.9031956791877747, + "learning_rate": 4.059215753565176e-06, + "loss": 0.6137, + "step": 2140 + }, + { + "epoch": 1.7727774710104915, + "grad_norm": 0.8913756012916565, + "learning_rate": 4.058356003328095e-06, + "loss": 0.6387, + "step": 2141 + }, + { + "epoch": 1.7736057426836003, + "grad_norm": 0.8986163139343262, + "learning_rate": 4.057495951563655e-06, + "loss": 0.6151, + "step": 2142 + }, + { + "epoch": 1.774434014356709, + "grad_norm": 0.9017037153244019, + "learning_rate": 4.0566355984382676e-06, + "loss": 0.6466, + "step": 2143 + }, + { + "epoch": 1.7752622860298177, + "grad_norm": 0.9211136698722839, + "learning_rate": 4.0557749441184024e-06, + "loss": 0.638, + "step": 2144 + }, + { + "epoch": 1.7760905577029265, + "grad_norm": 0.9097506999969482, + "learning_rate": 4.054913988770588e-06, + "loss": 0.6529, + "step": 2145 + }, + { + "epoch": 1.7769188293760352, + "grad_norm": 0.8985908627510071, + "learning_rate": 4.054052732561411e-06, + "loss": 0.6388, + "step": 2146 + }, + { + "epoch": 1.7777471010491441, + "grad_norm": 0.9066522121429443, + "learning_rate": 4.053191175657518e-06, + "loss": 0.6429, + "step": 2147 + }, + { + "epoch": 1.7785753727222529, + "grad_norm": 0.9080630540847778, + "learning_rate": 4.052329318225611e-06, + "loss": 0.619, + "step": 2148 + }, + { + "epoch": 1.7794036443953618, + "grad_norm": 0.8701719641685486, + "learning_rate": 4.0514671604324505e-06, + "loss": 0.6245, + "step": 2149 + }, + { + "epoch": 1.7802319160684705, + "grad_norm": 0.9059606194496155, + "learning_rate": 4.0506047024448585e-06, + "loss": 0.6124, + "step": 2150 + }, + { + "epoch": 1.7810601877415793, + "grad_norm": 0.8912079334259033, + "learning_rate": 4.04974194442971e-06, + "loss": 0.6295, + "step": 2151 + }, + { + "epoch": 1.781888459414688, + "grad_norm": 0.9063137769699097, + "learning_rate": 4.048878886553942e-06, + "loss": 0.6299, + "step": 2152 + }, + { + "epoch": 1.7827167310877967, + "grad_norm": 0.8850398659706116, + "learning_rate": 4.048015528984548e-06, + "loss": 0.6251, + "step": 2153 + }, + { + "epoch": 1.7835450027609054, + "grad_norm": 0.9059123992919922, + "learning_rate": 4.04715187188858e-06, + "loss": 0.6278, + "step": 2154 + }, + { + "epoch": 1.7843732744340144, + "grad_norm": 0.8777562379837036, + "learning_rate": 4.046287915433147e-06, + "loss": 0.6003, + "step": 2155 + }, + { + "epoch": 1.7852015461071231, + "grad_norm": 0.9048421382904053, + "learning_rate": 4.045423659785417e-06, + "loss": 0.6143, + "step": 2156 + }, + { + "epoch": 1.786029817780232, + "grad_norm": 0.9388517737388611, + "learning_rate": 4.044559105112614e-06, + "loss": 0.6314, + "step": 2157 + }, + { + "epoch": 1.7868580894533408, + "grad_norm": 0.9405019283294678, + "learning_rate": 4.043694251582023e-06, + "loss": 0.6381, + "step": 2158 + }, + { + "epoch": 1.7876863611264495, + "grad_norm": 0.8942162990570068, + "learning_rate": 4.0428290993609845e-06, + "loss": 0.6318, + "step": 2159 + }, + { + "epoch": 1.7885146327995582, + "grad_norm": 0.9065560698509216, + "learning_rate": 4.041963648616897e-06, + "loss": 0.6258, + "step": 2160 + }, + { + "epoch": 1.789342904472667, + "grad_norm": 0.8766914010047913, + "learning_rate": 4.041097899517218e-06, + "loss": 0.6292, + "step": 2161 + }, + { + "epoch": 1.7901711761457757, + "grad_norm": 0.8942935466766357, + "learning_rate": 4.040231852229462e-06, + "loss": 0.662, + "step": 2162 + }, + { + "epoch": 1.7909994478188846, + "grad_norm": 0.8892539143562317, + "learning_rate": 4.039365506921198e-06, + "loss": 0.6251, + "step": 2163 + }, + { + "epoch": 1.7918277194919934, + "grad_norm": 0.9359427690505981, + "learning_rate": 4.038498863760061e-06, + "loss": 0.6222, + "step": 2164 + }, + { + "epoch": 1.792655991165102, + "grad_norm": 0.8899942636489868, + "learning_rate": 4.037631922913733e-06, + "loss": 0.6318, + "step": 2165 + }, + { + "epoch": 1.793484262838211, + "grad_norm": 0.9233553409576416, + "learning_rate": 4.0367646845499635e-06, + "loss": 0.6188, + "step": 2166 + }, + { + "epoch": 1.7943125345113198, + "grad_norm": 0.8946877121925354, + "learning_rate": 4.035897148836553e-06, + "loss": 0.6247, + "step": 2167 + }, + { + "epoch": 1.7951408061844285, + "grad_norm": 0.9200116991996765, + "learning_rate": 4.03502931594136e-06, + "loss": 0.5981, + "step": 2168 + }, + { + "epoch": 1.7959690778575372, + "grad_norm": 0.9193991422653198, + "learning_rate": 4.034161186032306e-06, + "loss": 0.6144, + "step": 2169 + }, + { + "epoch": 1.796797349530646, + "grad_norm": 0.9637252688407898, + "learning_rate": 4.0332927592773615e-06, + "loss": 0.6094, + "step": 2170 + }, + { + "epoch": 1.7976256212037547, + "grad_norm": 0.8958486318588257, + "learning_rate": 4.032424035844564e-06, + "loss": 0.6304, + "step": 2171 + }, + { + "epoch": 1.7984538928768636, + "grad_norm": 0.9411646127700806, + "learning_rate": 4.031555015901998e-06, + "loss": 0.5984, + "step": 2172 + }, + { + "epoch": 1.7992821645499724, + "grad_norm": 0.8594862222671509, + "learning_rate": 4.030685699617815e-06, + "loss": 0.6176, + "step": 2173 + }, + { + "epoch": 1.8001104362230813, + "grad_norm": 0.8969242572784424, + "learning_rate": 4.029816087160217e-06, + "loss": 0.6294, + "step": 2174 + }, + { + "epoch": 1.80093870789619, + "grad_norm": 0.8950173854827881, + "learning_rate": 4.028946178697467e-06, + "loss": 0.6441, + "step": 2175 + }, + { + "epoch": 1.8017669795692988, + "grad_norm": 0.9084720611572266, + "learning_rate": 4.0280759743978835e-06, + "loss": 0.6245, + "step": 2176 + }, + { + "epoch": 1.8025952512424075, + "grad_norm": 0.9174644947052002, + "learning_rate": 4.027205474429845e-06, + "loss": 0.6136, + "step": 2177 + }, + { + "epoch": 1.8034235229155162, + "grad_norm": 0.9103249907493591, + "learning_rate": 4.026334678961783e-06, + "loss": 0.6341, + "step": 2178 + }, + { + "epoch": 1.804251794588625, + "grad_norm": 0.8916360139846802, + "learning_rate": 4.025463588162189e-06, + "loss": 0.6424, + "step": 2179 + }, + { + "epoch": 1.805080066261734, + "grad_norm": 0.9347325563430786, + "learning_rate": 4.024592202199611e-06, + "loss": 0.6425, + "step": 2180 + }, + { + "epoch": 1.8059083379348426, + "grad_norm": 0.9135183691978455, + "learning_rate": 4.023720521242653e-06, + "loss": 0.6265, + "step": 2181 + }, + { + "epoch": 1.8067366096079516, + "grad_norm": 0.8996950387954712, + "learning_rate": 4.022848545459979e-06, + "loss": 0.6083, + "step": 2182 + }, + { + "epoch": 1.8075648812810603, + "grad_norm": 0.9104750752449036, + "learning_rate": 4.021976275020307e-06, + "loss": 0.6087, + "step": 2183 + }, + { + "epoch": 1.808393152954169, + "grad_norm": 0.9272193908691406, + "learning_rate": 4.0211037100924135e-06, + "loss": 0.6254, + "step": 2184 + }, + { + "epoch": 1.8092214246272778, + "grad_norm": 0.8605275750160217, + "learning_rate": 4.020230850845132e-06, + "loss": 0.6244, + "step": 2185 + }, + { + "epoch": 1.8100496963003865, + "grad_norm": 0.8817747831344604, + "learning_rate": 4.019357697447352e-06, + "loss": 0.6226, + "step": 2186 + }, + { + "epoch": 1.8108779679734952, + "grad_norm": 0.8978784084320068, + "learning_rate": 4.018484250068021e-06, + "loss": 0.615, + "step": 2187 + }, + { + "epoch": 1.811706239646604, + "grad_norm": 0.8996117115020752, + "learning_rate": 4.017610508876143e-06, + "loss": 0.6248, + "step": 2188 + }, + { + "epoch": 1.8125345113197129, + "grad_norm": 0.9133120179176331, + "learning_rate": 4.016736474040778e-06, + "loss": 0.6364, + "step": 2189 + }, + { + "epoch": 1.8133627829928216, + "grad_norm": 0.903982400894165, + "learning_rate": 4.015862145731044e-06, + "loss": 0.6078, + "step": 2190 + }, + { + "epoch": 1.8141910546659306, + "grad_norm": 0.9343582987785339, + "learning_rate": 4.014987524116115e-06, + "loss": 0.6254, + "step": 2191 + }, + { + "epoch": 1.8150193263390393, + "grad_norm": 0.883305549621582, + "learning_rate": 4.014112609365224e-06, + "loss": 0.6306, + "step": 2192 + }, + { + "epoch": 1.815847598012148, + "grad_norm": 0.9044662714004517, + "learning_rate": 4.013237401647656e-06, + "loss": 0.6288, + "step": 2193 + }, + { + "epoch": 1.8166758696852567, + "grad_norm": 0.9016374349594116, + "learning_rate": 4.012361901132758e-06, + "loss": 0.6254, + "step": 2194 + }, + { + "epoch": 1.8175041413583655, + "grad_norm": 0.9198315739631653, + "learning_rate": 4.01148610798993e-06, + "loss": 0.621, + "step": 2195 + }, + { + "epoch": 1.8183324130314742, + "grad_norm": 0.8884013891220093, + "learning_rate": 4.01061002238863e-06, + "loss": 0.6195, + "step": 2196 + }, + { + "epoch": 1.8191606847045831, + "grad_norm": 0.901606023311615, + "learning_rate": 4.009733644498372e-06, + "loss": 0.646, + "step": 2197 + }, + { + "epoch": 1.8199889563776919, + "grad_norm": 0.9003238677978516, + "learning_rate": 4.008856974488728e-06, + "loss": 0.6276, + "step": 2198 + }, + { + "epoch": 1.8208172280508008, + "grad_norm": 0.8835832476615906, + "learning_rate": 4.007980012529324e-06, + "loss": 0.6232, + "step": 2199 + }, + { + "epoch": 1.8216454997239095, + "grad_norm": 0.8751473426818848, + "learning_rate": 4.007102758789846e-06, + "loss": 0.635, + "step": 2200 + }, + { + "epoch": 1.8224737713970183, + "grad_norm": 0.8990851640701294, + "learning_rate": 4.006225213440032e-06, + "loss": 0.605, + "step": 2201 + }, + { + "epoch": 1.823302043070127, + "grad_norm": 0.8766496181488037, + "learning_rate": 4.0053473766496805e-06, + "loss": 0.6396, + "step": 2202 + }, + { + "epoch": 1.8241303147432357, + "grad_norm": 0.914273202419281, + "learning_rate": 4.004469248588644e-06, + "loss": 0.6435, + "step": 2203 + }, + { + "epoch": 1.8249585864163445, + "grad_norm": 0.9063234925270081, + "learning_rate": 4.003590829426832e-06, + "loss": 0.6144, + "step": 2204 + }, + { + "epoch": 1.8257868580894532, + "grad_norm": 0.9479252696037292, + "learning_rate": 4.002712119334212e-06, + "loss": 0.6584, + "step": 2205 + }, + { + "epoch": 1.8266151297625621, + "grad_norm": 0.9253218770027161, + "learning_rate": 4.001833118480804e-06, + "loss": 0.6203, + "step": 2206 + }, + { + "epoch": 1.8274434014356709, + "grad_norm": 0.930874764919281, + "learning_rate": 4.000953827036689e-06, + "loss": 0.6254, + "step": 2207 + }, + { + "epoch": 1.8282716731087798, + "grad_norm": 0.9202263951301575, + "learning_rate": 4.000074245171999e-06, + "loss": 0.6351, + "step": 2208 + }, + { + "epoch": 1.8290999447818885, + "grad_norm": 0.8993806838989258, + "learning_rate": 3.999194373056926e-06, + "loss": 0.6439, + "step": 2209 + }, + { + "epoch": 1.8299282164549973, + "grad_norm": 0.9138032793998718, + "learning_rate": 3.998314210861718e-06, + "loss": 0.6011, + "step": 2210 + }, + { + "epoch": 1.830756488128106, + "grad_norm": 0.8995500206947327, + "learning_rate": 3.9974337587566765e-06, + "loss": 0.6285, + "step": 2211 + }, + { + "epoch": 1.8315847598012147, + "grad_norm": 0.9092652797698975, + "learning_rate": 3.996553016912161e-06, + "loss": 0.6364, + "step": 2212 + }, + { + "epoch": 1.8324130314743234, + "grad_norm": 0.9085352420806885, + "learning_rate": 3.995671985498589e-06, + "loss": 0.6245, + "step": 2213 + }, + { + "epoch": 1.8332413031474324, + "grad_norm": 0.9546533226966858, + "learning_rate": 3.994790664686429e-06, + "loss": 0.6512, + "step": 2214 + }, + { + "epoch": 1.8340695748205411, + "grad_norm": 0.8606089353561401, + "learning_rate": 3.99390905464621e-06, + "loss": 0.6135, + "step": 2215 + }, + { + "epoch": 1.83489784649365, + "grad_norm": 0.9450663328170776, + "learning_rate": 3.993027155548515e-06, + "loss": 0.6193, + "step": 2216 + }, + { + "epoch": 1.8357261181667588, + "grad_norm": 0.9496508836746216, + "learning_rate": 3.992144967563984e-06, + "loss": 0.5978, + "step": 2217 + }, + { + "epoch": 1.8365543898398675, + "grad_norm": 0.8837412595748901, + "learning_rate": 3.99126249086331e-06, + "loss": 0.6104, + "step": 2218 + }, + { + "epoch": 1.8373826615129762, + "grad_norm": 0.8774031400680542, + "learning_rate": 3.9903797256172464e-06, + "loss": 0.6302, + "step": 2219 + }, + { + "epoch": 1.838210933186085, + "grad_norm": 0.8978205919265747, + "learning_rate": 3.989496671996599e-06, + "loss": 0.6411, + "step": 2220 + }, + { + "epoch": 1.8390392048591937, + "grad_norm": 0.9087855815887451, + "learning_rate": 3.98861333017223e-06, + "loss": 0.592, + "step": 2221 + }, + { + "epoch": 1.8398674765323026, + "grad_norm": 0.9040684700012207, + "learning_rate": 3.987729700315059e-06, + "loss": 0.6121, + "step": 2222 + }, + { + "epoch": 1.8406957482054114, + "grad_norm": 0.8980532288551331, + "learning_rate": 3.986845782596058e-06, + "loss": 0.6231, + "step": 2223 + }, + { + "epoch": 1.84152401987852, + "grad_norm": 0.9104854464530945, + "learning_rate": 3.9859615771862595e-06, + "loss": 0.6228, + "step": 2224 + }, + { + "epoch": 1.842352291551629, + "grad_norm": 0.9314625263214111, + "learning_rate": 3.985077084256747e-06, + "loss": 0.6136, + "step": 2225 + }, + { + "epoch": 1.8431805632247378, + "grad_norm": 0.9033634662628174, + "learning_rate": 3.984192303978663e-06, + "loss": 0.6408, + "step": 2226 + }, + { + "epoch": 1.8440088348978465, + "grad_norm": 0.9022733569145203, + "learning_rate": 3.983307236523203e-06, + "loss": 0.5954, + "step": 2227 + }, + { + "epoch": 1.8448371065709552, + "grad_norm": 0.8900256156921387, + "learning_rate": 3.982421882061621e-06, + "loss": 0.631, + "step": 2228 + }, + { + "epoch": 1.845665378244064, + "grad_norm": 0.8882872462272644, + "learning_rate": 3.981536240765223e-06, + "loss": 0.6204, + "step": 2229 + }, + { + "epoch": 1.8464936499171727, + "grad_norm": 0.894923985004425, + "learning_rate": 3.980650312805373e-06, + "loss": 0.6013, + "step": 2230 + }, + { + "epoch": 1.8473219215902816, + "grad_norm": 0.8910638093948364, + "learning_rate": 3.979764098353491e-06, + "loss": 0.6373, + "step": 2231 + }, + { + "epoch": 1.8481501932633904, + "grad_norm": 0.8952053189277649, + "learning_rate": 3.978877597581049e-06, + "loss": 0.6374, + "step": 2232 + }, + { + "epoch": 1.8489784649364993, + "grad_norm": 0.9100720882415771, + "learning_rate": 3.977990810659579e-06, + "loss": 0.6417, + "step": 2233 + }, + { + "epoch": 1.849806736609608, + "grad_norm": 0.8954150676727295, + "learning_rate": 3.9771037377606634e-06, + "loss": 0.6027, + "step": 2234 + }, + { + "epoch": 1.8506350082827168, + "grad_norm": 0.8989905118942261, + "learning_rate": 3.976216379055944e-06, + "loss": 0.6228, + "step": 2235 + }, + { + "epoch": 1.8514632799558255, + "grad_norm": 0.9060820937156677, + "learning_rate": 3.975328734717117e-06, + "loss": 0.6291, + "step": 2236 + }, + { + "epoch": 1.8522915516289342, + "grad_norm": 0.9164367318153381, + "learning_rate": 3.974440804915932e-06, + "loss": 0.6249, + "step": 2237 + }, + { + "epoch": 1.853119823302043, + "grad_norm": 0.9227840304374695, + "learning_rate": 3.973552589824196e-06, + "loss": 0.6328, + "step": 2238 + }, + { + "epoch": 1.853948094975152, + "grad_norm": 0.8612478971481323, + "learning_rate": 3.972664089613772e-06, + "loss": 0.5794, + "step": 2239 + }, + { + "epoch": 1.8547763666482606, + "grad_norm": 0.8819253444671631, + "learning_rate": 3.971775304456573e-06, + "loss": 0.6145, + "step": 2240 + }, + { + "epoch": 1.8556046383213696, + "grad_norm": 0.9216543436050415, + "learning_rate": 3.970886234524573e-06, + "loss": 0.609, + "step": 2241 + }, + { + "epoch": 1.8564329099944783, + "grad_norm": 0.9114699959754944, + "learning_rate": 3.969996879989797e-06, + "loss": 0.6105, + "step": 2242 + }, + { + "epoch": 1.857261181667587, + "grad_norm": 0.9464704990386963, + "learning_rate": 3.969107241024328e-06, + "loss": 0.6389, + "step": 2243 + }, + { + "epoch": 1.8580894533406958, + "grad_norm": 0.8921800255775452, + "learning_rate": 3.968217317800303e-06, + "loss": 0.6338, + "step": 2244 + }, + { + "epoch": 1.8589177250138045, + "grad_norm": 0.897452175617218, + "learning_rate": 3.967327110489913e-06, + "loss": 0.6248, + "step": 2245 + }, + { + "epoch": 1.8597459966869132, + "grad_norm": 0.9093567132949829, + "learning_rate": 3.9664366192654066e-06, + "loss": 0.6258, + "step": 2246 + }, + { + "epoch": 1.860574268360022, + "grad_norm": 0.894188642501831, + "learning_rate": 3.965545844299084e-06, + "loss": 0.6095, + "step": 2247 + }, + { + "epoch": 1.8614025400331309, + "grad_norm": 0.9238452911376953, + "learning_rate": 3.964654785763301e-06, + "loss": 0.6237, + "step": 2248 + }, + { + "epoch": 1.8622308117062396, + "grad_norm": 0.8715214133262634, + "learning_rate": 3.9637634438304715e-06, + "loss": 0.6153, + "step": 2249 + }, + { + "epoch": 1.8630590833793486, + "grad_norm": 0.8790848255157471, + "learning_rate": 3.962871818673061e-06, + "loss": 0.6389, + "step": 2250 + }, + { + "epoch": 1.8638873550524573, + "grad_norm": 0.9859224557876587, + "learning_rate": 3.961979910463589e-06, + "loss": 0.6075, + "step": 2251 + }, + { + "epoch": 1.864715626725566, + "grad_norm": 0.9127727150917053, + "learning_rate": 3.9610877193746335e-06, + "loss": 0.6332, + "step": 2252 + }, + { + "epoch": 1.8655438983986747, + "grad_norm": 0.9248382449150085, + "learning_rate": 3.960195245578825e-06, + "loss": 0.622, + "step": 2253 + }, + { + "epoch": 1.8663721700717835, + "grad_norm": 0.9030913710594177, + "learning_rate": 3.959302489248848e-06, + "loss": 0.6326, + "step": 2254 + }, + { + "epoch": 1.8672004417448922, + "grad_norm": 0.8922805190086365, + "learning_rate": 3.958409450557441e-06, + "loss": 0.6099, + "step": 2255 + }, + { + "epoch": 1.8680287134180011, + "grad_norm": 0.9165307879447937, + "learning_rate": 3.957516129677402e-06, + "loss": 0.6245, + "step": 2256 + }, + { + "epoch": 1.8688569850911099, + "grad_norm": 0.9445623159408569, + "learning_rate": 3.956622526781579e-06, + "loss": 0.6317, + "step": 2257 + }, + { + "epoch": 1.8696852567642188, + "grad_norm": 0.8833973407745361, + "learning_rate": 3.955728642042874e-06, + "loss": 0.6264, + "step": 2258 + }, + { + "epoch": 1.8705135284373275, + "grad_norm": 0.9512470364570618, + "learning_rate": 3.9548344756342486e-06, + "loss": 0.6137, + "step": 2259 + }, + { + "epoch": 1.8713418001104363, + "grad_norm": 0.8834348320960999, + "learning_rate": 3.9539400277287134e-06, + "loss": 0.6597, + "step": 2260 + }, + { + "epoch": 1.872170071783545, + "grad_norm": 0.8881180882453918, + "learning_rate": 3.953045298499335e-06, + "loss": 0.6502, + "step": 2261 + }, + { + "epoch": 1.8729983434566537, + "grad_norm": 0.8921699523925781, + "learning_rate": 3.952150288119239e-06, + "loss": 0.6216, + "step": 2262 + }, + { + "epoch": 1.8738266151297625, + "grad_norm": 0.91098552942276, + "learning_rate": 3.9512549967615965e-06, + "loss": 0.6286, + "step": 2263 + }, + { + "epoch": 1.8746548868028712, + "grad_norm": 0.9343264698982239, + "learning_rate": 3.950359424599642e-06, + "loss": 0.6368, + "step": 2264 + }, + { + "epoch": 1.8754831584759801, + "grad_norm": 0.9168746471405029, + "learning_rate": 3.949463571806658e-06, + "loss": 0.6355, + "step": 2265 + }, + { + "epoch": 1.8763114301490889, + "grad_norm": 0.9206068515777588, + "learning_rate": 3.948567438555986e-06, + "loss": 0.607, + "step": 2266 + }, + { + "epoch": 1.8771397018221978, + "grad_norm": 0.9262584447860718, + "learning_rate": 3.947671025021017e-06, + "loss": 0.6197, + "step": 2267 + }, + { + "epoch": 1.8779679734953065, + "grad_norm": 0.8699168562889099, + "learning_rate": 3.946774331375199e-06, + "loss": 0.6181, + "step": 2268 + }, + { + "epoch": 1.8787962451684153, + "grad_norm": 0.8892862200737, + "learning_rate": 3.9458773577920365e-06, + "loss": 0.6245, + "step": 2269 + }, + { + "epoch": 1.879624516841524, + "grad_norm": 0.9038919806480408, + "learning_rate": 3.944980104445082e-06, + "loss": 0.6209, + "step": 2270 + }, + { + "epoch": 1.8804527885146327, + "grad_norm": 0.8780872821807861, + "learning_rate": 3.9440825715079475e-06, + "loss": 0.6355, + "step": 2271 + }, + { + "epoch": 1.8812810601877414, + "grad_norm": 0.9379895925521851, + "learning_rate": 3.943184759154297e-06, + "loss": 0.6381, + "step": 2272 + }, + { + "epoch": 1.8821093318608504, + "grad_norm": 0.9112582802772522, + "learning_rate": 3.942286667557849e-06, + "loss": 0.6273, + "step": 2273 + }, + { + "epoch": 1.8829376035339591, + "grad_norm": 0.9367797374725342, + "learning_rate": 3.9413882968923754e-06, + "loss": 0.6111, + "step": 2274 + }, + { + "epoch": 1.883765875207068, + "grad_norm": 0.8923424482345581, + "learning_rate": 3.940489647331703e-06, + "loss": 0.6187, + "step": 2275 + }, + { + "epoch": 1.8845941468801768, + "grad_norm": 0.8797968029975891, + "learning_rate": 3.939590719049712e-06, + "loss": 0.603, + "step": 2276 + }, + { + "epoch": 1.8854224185532855, + "grad_norm": 0.8977354168891907, + "learning_rate": 3.938691512220336e-06, + "loss": 0.614, + "step": 2277 + }, + { + "epoch": 1.8862506902263942, + "grad_norm": 0.9207090139389038, + "learning_rate": 3.937792027017564e-06, + "loss": 0.6162, + "step": 2278 + }, + { + "epoch": 1.887078961899503, + "grad_norm": 0.8815562725067139, + "learning_rate": 3.936892263615438e-06, + "loss": 0.6315, + "step": 2279 + }, + { + "epoch": 1.8879072335726117, + "grad_norm": 0.9983104467391968, + "learning_rate": 3.935992222188054e-06, + "loss": 0.6337, + "step": 2280 + }, + { + "epoch": 1.8887355052457206, + "grad_norm": 0.8987579941749573, + "learning_rate": 3.93509190290956e-06, + "loss": 0.644, + "step": 2281 + }, + { + "epoch": 1.8895637769188294, + "grad_norm": 0.9132242798805237, + "learning_rate": 3.934191305954161e-06, + "loss": 0.6063, + "step": 2282 + }, + { + "epoch": 1.890392048591938, + "grad_norm": 0.8940858840942383, + "learning_rate": 3.933290431496114e-06, + "loss": 0.6204, + "step": 2283 + }, + { + "epoch": 1.891220320265047, + "grad_norm": 0.8862229585647583, + "learning_rate": 3.932389279709729e-06, + "loss": 0.6257, + "step": 2284 + }, + { + "epoch": 1.8920485919381558, + "grad_norm": 0.8841557502746582, + "learning_rate": 3.931487850769371e-06, + "loss": 0.6329, + "step": 2285 + }, + { + "epoch": 1.8928768636112645, + "grad_norm": 0.9075404405593872, + "learning_rate": 3.930586144849459e-06, + "loss": 0.6195, + "step": 2286 + }, + { + "epoch": 1.8937051352843732, + "grad_norm": 0.9349889755249023, + "learning_rate": 3.929684162124462e-06, + "loss": 0.6159, + "step": 2287 + }, + { + "epoch": 1.894533406957482, + "grad_norm": 0.9378470182418823, + "learning_rate": 3.928781902768908e-06, + "loss": 0.6221, + "step": 2288 + }, + { + "epoch": 1.8953616786305907, + "grad_norm": 0.8877297043800354, + "learning_rate": 3.927879366957374e-06, + "loss": 0.6359, + "step": 2289 + }, + { + "epoch": 1.8961899503036996, + "grad_norm": 0.9335882067680359, + "learning_rate": 3.926976554864495e-06, + "loss": 0.6397, + "step": 2290 + }, + { + "epoch": 1.8970182219768084, + "grad_norm": 0.8982004523277283, + "learning_rate": 3.926073466664953e-06, + "loss": 0.6303, + "step": 2291 + }, + { + "epoch": 1.8978464936499173, + "grad_norm": 0.8909676671028137, + "learning_rate": 3.92517010253349e-06, + "loss": 0.6476, + "step": 2292 + }, + { + "epoch": 1.898674765323026, + "grad_norm": 0.9127813577651978, + "learning_rate": 3.924266462644896e-06, + "loss": 0.6078, + "step": 2293 + }, + { + "epoch": 1.8995030369961348, + "grad_norm": 0.8923580050468445, + "learning_rate": 3.923362547174019e-06, + "loss": 0.6371, + "step": 2294 + }, + { + "epoch": 1.9003313086692435, + "grad_norm": 0.912458062171936, + "learning_rate": 3.922458356295757e-06, + "loss": 0.6162, + "step": 2295 + }, + { + "epoch": 1.9011595803423522, + "grad_norm": 0.8866751790046692, + "learning_rate": 3.921553890185064e-06, + "loss": 0.6011, + "step": 2296 + }, + { + "epoch": 1.901987852015461, + "grad_norm": 0.9114490151405334, + "learning_rate": 3.920649149016944e-06, + "loss": 0.6322, + "step": 2297 + }, + { + "epoch": 1.90281612368857, + "grad_norm": 0.912030816078186, + "learning_rate": 3.919744132966457e-06, + "loss": 0.6293, + "step": 2298 + }, + { + "epoch": 1.9036443953616786, + "grad_norm": 0.887301504611969, + "learning_rate": 3.918838842208715e-06, + "loss": 0.6288, + "step": 2299 + }, + { + "epoch": 1.9044726670347876, + "grad_norm": 0.9039818048477173, + "learning_rate": 3.917933276918884e-06, + "loss": 0.6154, + "step": 2300 + }, + { + "epoch": 1.9053009387078963, + "grad_norm": 0.9260784983634949, + "learning_rate": 3.917027437272182e-06, + "loss": 0.6111, + "step": 2301 + }, + { + "epoch": 1.906129210381005, + "grad_norm": 0.8966069221496582, + "learning_rate": 3.916121323443879e-06, + "loss": 0.6334, + "step": 2302 + }, + { + "epoch": 1.9069574820541138, + "grad_norm": 0.8983698487281799, + "learning_rate": 3.9152149356093e-06, + "loss": 0.6235, + "step": 2303 + }, + { + "epoch": 1.9077857537272225, + "grad_norm": 0.9150595664978027, + "learning_rate": 3.914308273943826e-06, + "loss": 0.6118, + "step": 2304 + }, + { + "epoch": 1.9086140254003312, + "grad_norm": 0.8768419027328491, + "learning_rate": 3.913401338622883e-06, + "loss": 0.5935, + "step": 2305 + }, + { + "epoch": 1.90944229707344, + "grad_norm": 0.887562096118927, + "learning_rate": 3.912494129821958e-06, + "loss": 0.6232, + "step": 2306 + }, + { + "epoch": 1.9102705687465489, + "grad_norm": 0.8916515111923218, + "learning_rate": 3.911586647716585e-06, + "loss": 0.6409, + "step": 2307 + }, + { + "epoch": 1.9110988404196576, + "grad_norm": 0.9207882881164551, + "learning_rate": 3.910678892482354e-06, + "loss": 0.6155, + "step": 2308 + }, + { + "epoch": 1.9119271120927666, + "grad_norm": 0.9300670027732849, + "learning_rate": 3.9097708642949085e-06, + "loss": 0.6389, + "step": 2309 + }, + { + "epoch": 1.9127553837658753, + "grad_norm": 0.8685980439186096, + "learning_rate": 3.908862563329941e-06, + "loss": 0.6392, + "step": 2310 + }, + { + "epoch": 1.913583655438984, + "grad_norm": 0.9150967001914978, + "learning_rate": 3.907953989763202e-06, + "loss": 0.6387, + "step": 2311 + }, + { + "epoch": 1.9144119271120927, + "grad_norm": 0.9094190001487732, + "learning_rate": 3.90704514377049e-06, + "loss": 0.6055, + "step": 2312 + }, + { + "epoch": 1.9152401987852015, + "grad_norm": 0.8863343596458435, + "learning_rate": 3.90613602552766e-06, + "loss": 0.6073, + "step": 2313 + }, + { + "epoch": 1.9160684704583102, + "grad_norm": 0.9878994822502136, + "learning_rate": 3.905226635210616e-06, + "loss": 0.613, + "step": 2314 + }, + { + "epoch": 1.9168967421314191, + "grad_norm": 0.8863592147827148, + "learning_rate": 3.904316972995318e-06, + "loss": 0.6102, + "step": 2315 + }, + { + "epoch": 1.9177250138045279, + "grad_norm": 0.8859980702400208, + "learning_rate": 3.9034070390577765e-06, + "loss": 0.6182, + "step": 2316 + }, + { + "epoch": 1.9185532854776368, + "grad_norm": 0.8953589200973511, + "learning_rate": 3.9024968335740556e-06, + "loss": 0.6212, + "step": 2317 + }, + { + "epoch": 1.9193815571507455, + "grad_norm": 0.873466968536377, + "learning_rate": 3.90158635672027e-06, + "loss": 0.616, + "step": 2318 + }, + { + "epoch": 1.9202098288238543, + "grad_norm": 0.8589543700218201, + "learning_rate": 3.900675608672592e-06, + "loss": 0.6141, + "step": 2319 + }, + { + "epoch": 1.921038100496963, + "grad_norm": 0.8717133402824402, + "learning_rate": 3.899764589607239e-06, + "loss": 0.6281, + "step": 2320 + }, + { + "epoch": 1.9218663721700717, + "grad_norm": 0.893227219581604, + "learning_rate": 3.898853299700486e-06, + "loss": 0.6139, + "step": 2321 + }, + { + "epoch": 1.9226946438431805, + "grad_norm": 0.8736181259155273, + "learning_rate": 3.89794173912866e-06, + "loss": 0.5987, + "step": 2322 + }, + { + "epoch": 1.9235229155162892, + "grad_norm": 0.8975626826286316, + "learning_rate": 3.897029908068139e-06, + "loss": 0.6417, + "step": 2323 + }, + { + "epoch": 1.9243511871893981, + "grad_norm": 0.8916004300117493, + "learning_rate": 3.896117806695355e-06, + "loss": 0.6259, + "step": 2324 + }, + { + "epoch": 1.9251794588625069, + "grad_norm": 0.8739155530929565, + "learning_rate": 3.8952054351867885e-06, + "loss": 0.6252, + "step": 2325 + }, + { + "epoch": 1.9260077305356158, + "grad_norm": 0.8667280673980713, + "learning_rate": 3.894292793718976e-06, + "loss": 0.6083, + "step": 2326 + }, + { + "epoch": 1.9268360022087245, + "grad_norm": 0.8749732375144958, + "learning_rate": 3.893379882468506e-06, + "loss": 0.6294, + "step": 2327 + }, + { + "epoch": 1.9276642738818333, + "grad_norm": 0.9085330963134766, + "learning_rate": 3.8924667016120175e-06, + "loss": 0.6101, + "step": 2328 + }, + { + "epoch": 1.928492545554942, + "grad_norm": 0.9394292235374451, + "learning_rate": 3.891553251326203e-06, + "loss": 0.6182, + "step": 2329 + }, + { + "epoch": 1.9293208172280507, + "grad_norm": 0.8756951093673706, + "learning_rate": 3.890639531787805e-06, + "loss": 0.6159, + "step": 2330 + }, + { + "epoch": 1.9301490889011594, + "grad_norm": 0.8765501379966736, + "learning_rate": 3.8897255431736225e-06, + "loss": 0.6182, + "step": 2331 + }, + { + "epoch": 1.9309773605742684, + "grad_norm": 0.907835841178894, + "learning_rate": 3.888811285660501e-06, + "loss": 0.6336, + "step": 2332 + }, + { + "epoch": 1.9318056322473771, + "grad_norm": 1.0617926120758057, + "learning_rate": 3.887896759425342e-06, + "loss": 0.6336, + "step": 2333 + }, + { + "epoch": 1.932633903920486, + "grad_norm": 0.8745515942573547, + "learning_rate": 3.886981964645099e-06, + "loss": 0.6244, + "step": 2334 + }, + { + "epoch": 1.9334621755935948, + "grad_norm": 0.8918715119361877, + "learning_rate": 3.886066901496775e-06, + "loss": 0.6289, + "step": 2335 + }, + { + "epoch": 1.9342904472667035, + "grad_norm": 0.8779492378234863, + "learning_rate": 3.885151570157426e-06, + "loss": 0.6335, + "step": 2336 + }, + { + "epoch": 1.9351187189398122, + "grad_norm": 0.8955790400505066, + "learning_rate": 3.88423597080416e-06, + "loss": 0.6125, + "step": 2337 + }, + { + "epoch": 1.935946990612921, + "grad_norm": 0.8905174136161804, + "learning_rate": 3.883320103614137e-06, + "loss": 0.5913, + "step": 2338 + }, + { + "epoch": 1.9367752622860297, + "grad_norm": 0.9078717231750488, + "learning_rate": 3.88240396876457e-06, + "loss": 0.6356, + "step": 2339 + }, + { + "epoch": 1.9376035339591386, + "grad_norm": 0.9211397171020508, + "learning_rate": 3.881487566432722e-06, + "loss": 0.6305, + "step": 2340 + }, + { + "epoch": 1.9384318056322474, + "grad_norm": 0.9131335020065308, + "learning_rate": 3.880570896795908e-06, + "loss": 0.6184, + "step": 2341 + }, + { + "epoch": 1.9392600773053563, + "grad_norm": 0.8874300718307495, + "learning_rate": 3.879653960031494e-06, + "loss": 0.6191, + "step": 2342 + }, + { + "epoch": 1.940088348978465, + "grad_norm": 0.9244474768638611, + "learning_rate": 3.8787367563169e-06, + "loss": 0.5994, + "step": 2343 + }, + { + "epoch": 1.9409166206515738, + "grad_norm": 0.8986138701438904, + "learning_rate": 3.877819285829596e-06, + "loss": 0.6544, + "step": 2344 + }, + { + "epoch": 1.9417448923246825, + "grad_norm": 0.9084672927856445, + "learning_rate": 3.876901548747104e-06, + "loss": 0.6248, + "step": 2345 + }, + { + "epoch": 1.9425731639977912, + "grad_norm": 0.9052600264549255, + "learning_rate": 3.875983545246997e-06, + "loss": 0.6112, + "step": 2346 + }, + { + "epoch": 1.9434014356709, + "grad_norm": 0.9029438495635986, + "learning_rate": 3.8750652755069025e-06, + "loss": 0.6157, + "step": 2347 + }, + { + "epoch": 1.9442297073440087, + "grad_norm": 0.882449746131897, + "learning_rate": 3.874146739704494e-06, + "loss": 0.5886, + "step": 2348 + }, + { + "epoch": 1.9450579790171176, + "grad_norm": 0.8628272414207458, + "learning_rate": 3.873227938017502e-06, + "loss": 0.608, + "step": 2349 + }, + { + "epoch": 1.9458862506902264, + "grad_norm": 0.8977845907211304, + "learning_rate": 3.872308870623704e-06, + "loss": 0.6475, + "step": 2350 + }, + { + "epoch": 1.9467145223633353, + "grad_norm": 0.8801501393318176, + "learning_rate": 3.871389537700933e-06, + "loss": 0.6246, + "step": 2351 + }, + { + "epoch": 1.947542794036444, + "grad_norm": 0.9245358109474182, + "learning_rate": 3.87046993942707e-06, + "loss": 0.6317, + "step": 2352 + }, + { + "epoch": 1.9483710657095528, + "grad_norm": 0.8959249258041382, + "learning_rate": 3.869550075980049e-06, + "loss": 0.6157, + "step": 2353 + }, + { + "epoch": 1.9491993373826615, + "grad_norm": 0.861315906047821, + "learning_rate": 3.868629947537855e-06, + "loss": 0.6251, + "step": 2354 + }, + { + "epoch": 1.9500276090557702, + "grad_norm": 0.8839966654777527, + "learning_rate": 3.867709554278524e-06, + "loss": 0.6289, + "step": 2355 + }, + { + "epoch": 1.950855880728879, + "grad_norm": 0.8878759741783142, + "learning_rate": 3.8667888963801445e-06, + "loss": 0.6113, + "step": 2356 + }, + { + "epoch": 1.951684152401988, + "grad_norm": 0.8725037574768066, + "learning_rate": 3.865867974020854e-06, + "loss": 0.6324, + "step": 2357 + }, + { + "epoch": 1.9525124240750966, + "grad_norm": 0.9059397578239441, + "learning_rate": 3.8649467873788436e-06, + "loss": 0.5981, + "step": 2358 + }, + { + "epoch": 1.9533406957482056, + "grad_norm": 0.8883296847343445, + "learning_rate": 3.8640253366323535e-06, + "loss": 0.598, + "step": 2359 + }, + { + "epoch": 1.9541689674213143, + "grad_norm": 0.8987225890159607, + "learning_rate": 3.863103621959676e-06, + "loss": 0.6253, + "step": 2360 + }, + { + "epoch": 1.954997239094423, + "grad_norm": 0.8977320194244385, + "learning_rate": 3.862181643539155e-06, + "loss": 0.6424, + "step": 2361 + }, + { + "epoch": 1.9558255107675317, + "grad_norm": 0.9267589449882507, + "learning_rate": 3.861259401549183e-06, + "loss": 0.647, + "step": 2362 + }, + { + "epoch": 1.9566537824406405, + "grad_norm": 0.9297959804534912, + "learning_rate": 3.860336896168207e-06, + "loss": 0.6448, + "step": 2363 + }, + { + "epoch": 1.9574820541137492, + "grad_norm": 0.9627453088760376, + "learning_rate": 3.859414127574722e-06, + "loss": 0.6154, + "step": 2364 + }, + { + "epoch": 1.958310325786858, + "grad_norm": 0.8945866227149963, + "learning_rate": 3.858491095947275e-06, + "loss": 0.6201, + "step": 2365 + }, + { + "epoch": 1.9591385974599669, + "grad_norm": 0.8859817981719971, + "learning_rate": 3.857567801464466e-06, + "loss": 0.6171, + "step": 2366 + }, + { + "epoch": 1.9599668691330756, + "grad_norm": 0.9104642271995544, + "learning_rate": 3.8566442443049425e-06, + "loss": 0.6273, + "step": 2367 + }, + { + "epoch": 1.9607951408061846, + "grad_norm": 0.8821743726730347, + "learning_rate": 3.855720424647403e-06, + "loss": 0.5993, + "step": 2368 + }, + { + "epoch": 1.9616234124792933, + "grad_norm": 0.903322160243988, + "learning_rate": 3.854796342670601e-06, + "loss": 0.6471, + "step": 2369 + }, + { + "epoch": 1.962451684152402, + "grad_norm": 0.8744645714759827, + "learning_rate": 3.853871998553334e-06, + "loss": 0.6444, + "step": 2370 + }, + { + "epoch": 1.9632799558255107, + "grad_norm": 0.9538072347640991, + "learning_rate": 3.852947392474457e-06, + "loss": 0.656, + "step": 2371 + }, + { + "epoch": 1.9641082274986195, + "grad_norm": 0.875407874584198, + "learning_rate": 3.852022524612872e-06, + "loss": 0.6479, + "step": 2372 + }, + { + "epoch": 1.9649364991717282, + "grad_norm": 0.8908282518386841, + "learning_rate": 3.8510973951475305e-06, + "loss": 0.6262, + "step": 2373 + }, + { + "epoch": 1.9657647708448371, + "grad_norm": 0.8698679208755493, + "learning_rate": 3.850172004257438e-06, + "loss": 0.6431, + "step": 2374 + }, + { + "epoch": 1.9665930425179459, + "grad_norm": 0.8997004628181458, + "learning_rate": 3.84924635212165e-06, + "loss": 0.6349, + "step": 2375 + }, + { + "epoch": 1.9674213141910548, + "grad_norm": 0.8815768957138062, + "learning_rate": 3.8483204389192695e-06, + "loss": 0.6449, + "step": 2376 + }, + { + "epoch": 1.9682495858641635, + "grad_norm": 0.9148927927017212, + "learning_rate": 3.847394264829453e-06, + "loss": 0.6106, + "step": 2377 + }, + { + "epoch": 1.9690778575372723, + "grad_norm": 0.8647779226303101, + "learning_rate": 3.846467830031407e-06, + "loss": 0.5954, + "step": 2378 + }, + { + "epoch": 1.969906129210381, + "grad_norm": 0.8877570033073425, + "learning_rate": 3.845541134704387e-06, + "loss": 0.6006, + "step": 2379 + }, + { + "epoch": 1.9707344008834897, + "grad_norm": 0.899742603302002, + "learning_rate": 3.844614179027701e-06, + "loss": 0.6069, + "step": 2380 + }, + { + "epoch": 1.9715626725565985, + "grad_norm": 0.9085001349449158, + "learning_rate": 3.843686963180705e-06, + "loss": 0.6332, + "step": 2381 + }, + { + "epoch": 1.9723909442297074, + "grad_norm": 0.8954899311065674, + "learning_rate": 3.842759487342808e-06, + "loss": 0.6186, + "step": 2382 + }, + { + "epoch": 1.9732192159028161, + "grad_norm": 0.9011021256446838, + "learning_rate": 3.841831751693466e-06, + "loss": 0.6177, + "step": 2383 + }, + { + "epoch": 1.9740474875759249, + "grad_norm": 0.8824598789215088, + "learning_rate": 3.8409037564121895e-06, + "loss": 0.6392, + "step": 2384 + }, + { + "epoch": 1.9748757592490338, + "grad_norm": 0.8936701416969299, + "learning_rate": 3.839975501678536e-06, + "loss": 0.6508, + "step": 2385 + }, + { + "epoch": 1.9757040309221425, + "grad_norm": 0.9009436368942261, + "learning_rate": 3.839046987672113e-06, + "loss": 0.6169, + "step": 2386 + }, + { + "epoch": 1.9765323025952513, + "grad_norm": 0.8839378356933594, + "learning_rate": 3.8381182145725815e-06, + "loss": 0.6107, + "step": 2387 + }, + { + "epoch": 1.97736057426836, + "grad_norm": 0.9240637421607971, + "learning_rate": 3.837189182559649e-06, + "loss": 0.6083, + "step": 2388 + }, + { + "epoch": 1.9781888459414687, + "grad_norm": 0.881908118724823, + "learning_rate": 3.836259891813075e-06, + "loss": 0.5995, + "step": 2389 + }, + { + "epoch": 1.9790171176145774, + "grad_norm": 0.8754116296768188, + "learning_rate": 3.835330342512667e-06, + "loss": 0.6111, + "step": 2390 + }, + { + "epoch": 1.9798453892876864, + "grad_norm": 0.9713384509086609, + "learning_rate": 3.834400534838287e-06, + "loss": 0.641, + "step": 2391 + }, + { + "epoch": 1.9806736609607951, + "grad_norm": 0.9263403415679932, + "learning_rate": 3.833470468969842e-06, + "loss": 0.6254, + "step": 2392 + }, + { + "epoch": 1.981501932633904, + "grad_norm": 0.9033388495445251, + "learning_rate": 3.832540145087291e-06, + "loss": 0.6049, + "step": 2393 + }, + { + "epoch": 1.9823302043070128, + "grad_norm": 0.8710628151893616, + "learning_rate": 3.8316095633706455e-06, + "loss": 0.5999, + "step": 2394 + }, + { + "epoch": 1.9831584759801215, + "grad_norm": 0.9053885340690613, + "learning_rate": 3.830678723999961e-06, + "loss": 0.6393, + "step": 2395 + }, + { + "epoch": 1.9839867476532302, + "grad_norm": 0.9075314998626709, + "learning_rate": 3.829747627155348e-06, + "loss": 0.6188, + "step": 2396 + }, + { + "epoch": 1.984815019326339, + "grad_norm": 0.9151870012283325, + "learning_rate": 3.828816273016966e-06, + "loss": 0.625, + "step": 2397 + }, + { + "epoch": 1.9856432909994477, + "grad_norm": 0.903843104839325, + "learning_rate": 3.827884661765022e-06, + "loss": 0.618, + "step": 2398 + }, + { + "epoch": 1.9864715626725566, + "grad_norm": 0.875468373298645, + "learning_rate": 3.826952793579774e-06, + "loss": 0.6121, + "step": 2399 + }, + { + "epoch": 1.9872998343456654, + "grad_norm": 0.8908163905143738, + "learning_rate": 3.8260206686415295e-06, + "loss": 0.6292, + "step": 2400 + }, + { + "epoch": 1.9881281060187743, + "grad_norm": 0.8985518217086792, + "learning_rate": 3.825088287130647e-06, + "loss": 0.6356, + "step": 2401 + }, + { + "epoch": 1.988956377691883, + "grad_norm": 0.9277573227882385, + "learning_rate": 3.8241556492275335e-06, + "loss": 0.6424, + "step": 2402 + }, + { + "epoch": 1.9897846493649918, + "grad_norm": 0.9075362682342529, + "learning_rate": 3.823222755112646e-06, + "loss": 0.623, + "step": 2403 + }, + { + "epoch": 1.9906129210381005, + "grad_norm": 0.9038227796554565, + "learning_rate": 3.822289604966489e-06, + "loss": 0.6195, + "step": 2404 + }, + { + "epoch": 1.9914411927112092, + "grad_norm": 0.8869004845619202, + "learning_rate": 3.82135619896962e-06, + "loss": 0.6034, + "step": 2405 + }, + { + "epoch": 1.992269464384318, + "grad_norm": 0.9017077684402466, + "learning_rate": 3.820422537302645e-06, + "loss": 0.6469, + "step": 2406 + }, + { + "epoch": 1.9930977360574267, + "grad_norm": 0.8785225749015808, + "learning_rate": 3.819488620146218e-06, + "loss": 0.6327, + "step": 2407 + }, + { + "epoch": 1.9939260077305356, + "grad_norm": 1.0367780923843384, + "learning_rate": 3.818554447681041e-06, + "loss": 0.6259, + "step": 2408 + }, + { + "epoch": 1.9947542794036444, + "grad_norm": 0.9392277002334595, + "learning_rate": 3.817620020087872e-06, + "loss": 0.6393, + "step": 2409 + }, + { + "epoch": 1.9955825510767533, + "grad_norm": 0.8935515880584717, + "learning_rate": 3.816685337547511e-06, + "loss": 0.6153, + "step": 2410 + }, + { + "epoch": 1.996410822749862, + "grad_norm": 0.8895331025123596, + "learning_rate": 3.815750400240811e-06, + "loss": 0.5978, + "step": 2411 + }, + { + "epoch": 1.9972390944229708, + "grad_norm": 0.8733546733856201, + "learning_rate": 3.814815208348673e-06, + "loss": 0.6254, + "step": 2412 + }, + { + "epoch": 1.9980673660960795, + "grad_norm": 0.9177356958389282, + "learning_rate": 3.8138797620520507e-06, + "loss": 0.631, + "step": 2413 + }, + { + "epoch": 1.9988956377691882, + "grad_norm": 0.9259604811668396, + "learning_rate": 3.8129440615319414e-06, + "loss": 0.6138, + "step": 2414 + }, + { + "epoch": 1.999723909442297, + "grad_norm": 0.9226608872413635, + "learning_rate": 3.812008106969395e-06, + "loss": 0.607, + "step": 2415 + }, + { + "epoch": 2.0, + "grad_norm": 0.9226608872413635, + "learning_rate": 3.8110718985455097e-06, + "loss": 0.5993, + "step": 2416 + }, + { + "epoch": 2.0008282716731087, + "grad_norm": 1.62429940700531, + "learning_rate": 3.810135436441435e-06, + "loss": 0.5954, + "step": 2417 + }, + { + "epoch": 2.0016565433462175, + "grad_norm": 0.8664761781692505, + "learning_rate": 3.8091987208383656e-06, + "loss": 0.5947, + "step": 2418 + }, + { + "epoch": 2.002484815019326, + "grad_norm": 0.9208955764770508, + "learning_rate": 3.808261751917548e-06, + "loss": 0.6028, + "step": 2419 + }, + { + "epoch": 2.003313086692435, + "grad_norm": 0.8842632174491882, + "learning_rate": 3.8073245298602775e-06, + "loss": 0.5913, + "step": 2420 + }, + { + "epoch": 2.004141358365544, + "grad_norm": 0.9232655763626099, + "learning_rate": 3.8063870548478964e-06, + "loss": 0.5977, + "step": 2421 + }, + { + "epoch": 2.004969630038653, + "grad_norm": 0.8898128271102905, + "learning_rate": 3.805449327061798e-06, + "loss": 0.6335, + "step": 2422 + }, + { + "epoch": 2.0057979017117615, + "grad_norm": 0.9356569051742554, + "learning_rate": 3.8045113466834248e-06, + "loss": 0.616, + "step": 2423 + }, + { + "epoch": 2.0066261733848703, + "grad_norm": 0.890315055847168, + "learning_rate": 3.803573113894266e-06, + "loss": 0.5874, + "step": 2424 + }, + { + "epoch": 2.007454445057979, + "grad_norm": 0.8921117186546326, + "learning_rate": 3.802634628875861e-06, + "loss": 0.5679, + "step": 2425 + }, + { + "epoch": 2.0082827167310877, + "grad_norm": 0.8871824145317078, + "learning_rate": 3.8016958918097995e-06, + "loss": 0.6066, + "step": 2426 + }, + { + "epoch": 2.0091109884041964, + "grad_norm": 0.902680516242981, + "learning_rate": 3.800756902877717e-06, + "loss": 0.5936, + "step": 2427 + }, + { + "epoch": 2.009939260077305, + "grad_norm": 0.9199647307395935, + "learning_rate": 3.799817662261298e-06, + "loss": 0.5836, + "step": 2428 + }, + { + "epoch": 2.0107675317504143, + "grad_norm": 0.9029080271720886, + "learning_rate": 3.7988781701422795e-06, + "loss": 0.5858, + "step": 2429 + }, + { + "epoch": 2.011595803423523, + "grad_norm": 0.8834925889968872, + "learning_rate": 3.7979384267024422e-06, + "loss": 0.6061, + "step": 2430 + }, + { + "epoch": 2.012424075096632, + "grad_norm": 0.9098751544952393, + "learning_rate": 3.796998432123619e-06, + "loss": 0.6238, + "step": 2431 + }, + { + "epoch": 2.0132523467697405, + "grad_norm": 0.9316835999488831, + "learning_rate": 3.7960581865876898e-06, + "loss": 0.6139, + "step": 2432 + }, + { + "epoch": 2.0140806184428492, + "grad_norm": 0.9198770523071289, + "learning_rate": 3.795117690276583e-06, + "loss": 0.6113, + "step": 2433 + }, + { + "epoch": 2.014908890115958, + "grad_norm": 0.8970876336097717, + "learning_rate": 3.794176943372275e-06, + "loss": 0.6033, + "step": 2434 + }, + { + "epoch": 2.0157371617890667, + "grad_norm": 0.9025728702545166, + "learning_rate": 3.7932359460567945e-06, + "loss": 0.586, + "step": 2435 + }, + { + "epoch": 2.0165654334621754, + "grad_norm": 0.891909122467041, + "learning_rate": 3.7922946985122124e-06, + "loss": 0.6013, + "step": 2436 + }, + { + "epoch": 2.017393705135284, + "grad_norm": 0.9115222096443176, + "learning_rate": 3.791353200920654e-06, + "loss": 0.5783, + "step": 2437 + }, + { + "epoch": 2.0182219768083933, + "grad_norm": 0.8973708152770996, + "learning_rate": 3.790411453464288e-06, + "loss": 0.5862, + "step": 2438 + }, + { + "epoch": 2.019050248481502, + "grad_norm": 0.910248875617981, + "learning_rate": 3.7894694563253352e-06, + "loss": 0.5884, + "step": 2439 + }, + { + "epoch": 2.019878520154611, + "grad_norm": 0.9227123260498047, + "learning_rate": 3.7885272096860625e-06, + "loss": 0.6185, + "step": 2440 + }, + { + "epoch": 2.0207067918277195, + "grad_norm": 0.917061448097229, + "learning_rate": 3.787584713728787e-06, + "loss": 0.5767, + "step": 2441 + }, + { + "epoch": 2.0215350635008282, + "grad_norm": 0.9035825133323669, + "learning_rate": 3.786641968635871e-06, + "loss": 0.6064, + "step": 2442 + }, + { + "epoch": 2.022363335173937, + "grad_norm": 0.8991772532463074, + "learning_rate": 3.78569897458973e-06, + "loss": 0.5912, + "step": 2443 + }, + { + "epoch": 2.0231916068470457, + "grad_norm": 0.9237397909164429, + "learning_rate": 3.78475573177282e-06, + "loss": 0.5818, + "step": 2444 + }, + { + "epoch": 2.0240198785201544, + "grad_norm": 0.8913205862045288, + "learning_rate": 3.7838122403676543e-06, + "loss": 0.5972, + "step": 2445 + }, + { + "epoch": 2.0248481501932636, + "grad_norm": 0.8960773944854736, + "learning_rate": 3.782868500556787e-06, + "loss": 0.6118, + "step": 2446 + }, + { + "epoch": 2.0256764218663723, + "grad_norm": 0.9198427200317383, + "learning_rate": 3.7819245125228228e-06, + "loss": 0.6058, + "step": 2447 + }, + { + "epoch": 2.026504693539481, + "grad_norm": 0.9039538502693176, + "learning_rate": 3.780980276448416e-06, + "loss": 0.6122, + "step": 2448 + }, + { + "epoch": 2.0273329652125898, + "grad_norm": 0.9103277325630188, + "learning_rate": 3.7800357925162675e-06, + "loss": 0.584, + "step": 2449 + }, + { + "epoch": 2.0281612368856985, + "grad_norm": 0.9114090204238892, + "learning_rate": 3.779091060909124e-06, + "loss": 0.6071, + "step": 2450 + }, + { + "epoch": 2.028989508558807, + "grad_norm": 0.9064953327178955, + "learning_rate": 3.7781460818097847e-06, + "loss": 0.6097, + "step": 2451 + }, + { + "epoch": 2.029817780231916, + "grad_norm": 0.893298864364624, + "learning_rate": 3.7772008554010937e-06, + "loss": 0.5959, + "step": 2452 + }, + { + "epoch": 2.0306460519050247, + "grad_norm": 0.8867486119270325, + "learning_rate": 3.776255381865942e-06, + "loss": 0.5788, + "step": 2453 + }, + { + "epoch": 2.031474323578134, + "grad_norm": 0.9018534421920776, + "learning_rate": 3.775309661387272e-06, + "loss": 0.6075, + "step": 2454 + }, + { + "epoch": 2.0323025952512426, + "grad_norm": 0.9353018999099731, + "learning_rate": 3.77436369414807e-06, + "loss": 0.5894, + "step": 2455 + }, + { + "epoch": 2.0331308669243513, + "grad_norm": 0.8828651309013367, + "learning_rate": 3.7734174803313733e-06, + "loss": 0.608, + "step": 2456 + }, + { + "epoch": 2.03395913859746, + "grad_norm": 0.8971118330955505, + "learning_rate": 3.7724710201202645e-06, + "loss": 0.597, + "step": 2457 + }, + { + "epoch": 2.0347874102705688, + "grad_norm": 0.9064204096794128, + "learning_rate": 3.7715243136978753e-06, + "loss": 0.61, + "step": 2458 + }, + { + "epoch": 2.0356156819436775, + "grad_norm": 0.9204864501953125, + "learning_rate": 3.770577361247384e-06, + "loss": 0.5908, + "step": 2459 + }, + { + "epoch": 2.036443953616786, + "grad_norm": 0.905817449092865, + "learning_rate": 3.7696301629520183e-06, + "loss": 0.5867, + "step": 2460 + }, + { + "epoch": 2.037272225289895, + "grad_norm": 0.8872947096824646, + "learning_rate": 3.7686827189950505e-06, + "loss": 0.5805, + "step": 2461 + }, + { + "epoch": 2.0381004969630037, + "grad_norm": 0.8944478631019592, + "learning_rate": 3.7677350295598035e-06, + "loss": 0.5835, + "step": 2462 + }, + { + "epoch": 2.038928768636113, + "grad_norm": 0.884630560874939, + "learning_rate": 3.766787094829646e-06, + "loss": 0.5909, + "step": 2463 + }, + { + "epoch": 2.0397570403092216, + "grad_norm": 0.8915899395942688, + "learning_rate": 3.7658389149879936e-06, + "loss": 0.5923, + "step": 2464 + }, + { + "epoch": 2.0405853119823303, + "grad_norm": 0.8971114158630371, + "learning_rate": 3.7648904902183128e-06, + "loss": 0.6097, + "step": 2465 + }, + { + "epoch": 2.041413583655439, + "grad_norm": 0.9223624467849731, + "learning_rate": 3.7639418207041124e-06, + "loss": 0.6048, + "step": 2466 + }, + { + "epoch": 2.0422418553285477, + "grad_norm": 0.915450930595398, + "learning_rate": 3.7629929066289517e-06, + "loss": 0.626, + "step": 2467 + }, + { + "epoch": 2.0430701270016565, + "grad_norm": 0.919333279132843, + "learning_rate": 3.762043748176437e-06, + "loss": 0.6234, + "step": 2468 + }, + { + "epoch": 2.043898398674765, + "grad_norm": 0.9282277822494507, + "learning_rate": 3.7610943455302223e-06, + "loss": 0.6008, + "step": 2469 + }, + { + "epoch": 2.044726670347874, + "grad_norm": 0.9419016242027283, + "learning_rate": 3.760144698874007e-06, + "loss": 0.5836, + "step": 2470 + }, + { + "epoch": 2.045554942020983, + "grad_norm": 0.91156405210495, + "learning_rate": 3.7591948083915403e-06, + "loss": 0.6035, + "step": 2471 + }, + { + "epoch": 2.046383213694092, + "grad_norm": 0.9019245505332947, + "learning_rate": 3.758244674266615e-06, + "loss": 0.5949, + "step": 2472 + }, + { + "epoch": 2.0472114853672005, + "grad_norm": 0.8976905345916748, + "learning_rate": 3.757294296683075e-06, + "loss": 0.5818, + "step": 2473 + }, + { + "epoch": 2.0480397570403093, + "grad_norm": 0.8748987913131714, + "learning_rate": 3.7563436758248094e-06, + "loss": 0.5816, + "step": 2474 + }, + { + "epoch": 2.048868028713418, + "grad_norm": 0.8929162621498108, + "learning_rate": 3.7553928118757532e-06, + "loss": 0.5869, + "step": 2475 + }, + { + "epoch": 2.0496963003865267, + "grad_norm": 0.9968146681785583, + "learning_rate": 3.7544417050198913e-06, + "loss": 0.6065, + "step": 2476 + }, + { + "epoch": 2.0505245720596355, + "grad_norm": 0.9206544756889343, + "learning_rate": 3.753490355441253e-06, + "loss": 0.5919, + "step": 2477 + }, + { + "epoch": 2.051352843732744, + "grad_norm": 0.8804112672805786, + "learning_rate": 3.7525387633239153e-06, + "loss": 0.5918, + "step": 2478 + }, + { + "epoch": 2.052181115405853, + "grad_norm": 0.922532320022583, + "learning_rate": 3.7515869288520027e-06, + "loss": 0.6212, + "step": 2479 + }, + { + "epoch": 2.053009387078962, + "grad_norm": 0.9353113770484924, + "learning_rate": 3.7506348522096876e-06, + "loss": 0.5695, + "step": 2480 + }, + { + "epoch": 2.053837658752071, + "grad_norm": 0.9110690951347351, + "learning_rate": 3.749682533581186e-06, + "loss": 0.6156, + "step": 2481 + }, + { + "epoch": 2.0546659304251795, + "grad_norm": 0.9076771140098572, + "learning_rate": 3.7487299731507642e-06, + "loss": 0.5987, + "step": 2482 + }, + { + "epoch": 2.0554942020982883, + "grad_norm": 0.8937394618988037, + "learning_rate": 3.747777171102732e-06, + "loss": 0.6057, + "step": 2483 + }, + { + "epoch": 2.056322473771397, + "grad_norm": 0.9488244652748108, + "learning_rate": 3.74682412762145e-06, + "loss": 0.5983, + "step": 2484 + }, + { + "epoch": 2.0571507454445057, + "grad_norm": 0.897693395614624, + "learning_rate": 3.7458708428913217e-06, + "loss": 0.606, + "step": 2485 + }, + { + "epoch": 2.0579790171176144, + "grad_norm": 0.8989591598510742, + "learning_rate": 3.744917317096799e-06, + "loss": 0.6138, + "step": 2486 + }, + { + "epoch": 2.058807288790723, + "grad_norm": 0.9028336405754089, + "learning_rate": 3.7439635504223804e-06, + "loss": 0.5714, + "step": 2487 + }, + { + "epoch": 2.0596355604638323, + "grad_norm": 0.8895571231842041, + "learning_rate": 3.7430095430526113e-06, + "loss": 0.5903, + "step": 2488 + }, + { + "epoch": 2.060463832136941, + "grad_norm": 0.926414430141449, + "learning_rate": 3.7420552951720822e-06, + "loss": 0.5841, + "step": 2489 + }, + { + "epoch": 2.06129210381005, + "grad_norm": 0.9279632568359375, + "learning_rate": 3.741100806965432e-06, + "loss": 0.6062, + "step": 2490 + }, + { + "epoch": 2.0621203754831585, + "grad_norm": 0.9196921586990356, + "learning_rate": 3.7401460786173456e-06, + "loss": 0.579, + "step": 2491 + }, + { + "epoch": 2.0629486471562672, + "grad_norm": 0.8959674835205078, + "learning_rate": 3.739191110312552e-06, + "loss": 0.5888, + "step": 2492 + }, + { + "epoch": 2.063776918829376, + "grad_norm": 0.9123788475990295, + "learning_rate": 3.7382359022358322e-06, + "loss": 0.5891, + "step": 2493 + }, + { + "epoch": 2.0646051905024847, + "grad_norm": 0.9637120366096497, + "learning_rate": 3.7372804545720067e-06, + "loss": 0.6124, + "step": 2494 + }, + { + "epoch": 2.0654334621755934, + "grad_norm": 0.9360100626945496, + "learning_rate": 3.736324767505947e-06, + "loss": 0.5884, + "step": 2495 + }, + { + "epoch": 2.0662617338487026, + "grad_norm": 0.9066753387451172, + "learning_rate": 3.7353688412225704e-06, + "loss": 0.6013, + "step": 2496 + }, + { + "epoch": 2.0670900055218113, + "grad_norm": 0.8961504101753235, + "learning_rate": 3.7344126759068382e-06, + "loss": 0.6303, + "step": 2497 + }, + { + "epoch": 2.06791827719492, + "grad_norm": 0.8995484709739685, + "learning_rate": 3.7334562717437608e-06, + "loss": 0.6031, + "step": 2498 + }, + { + "epoch": 2.068746548868029, + "grad_norm": 0.9279561042785645, + "learning_rate": 3.7324996289183927e-06, + "loss": 0.5848, + "step": 2499 + }, + { + "epoch": 2.0695748205411375, + "grad_norm": 0.9250274300575256, + "learning_rate": 3.7315427476158357e-06, + "loss": 0.5949, + "step": 2500 + }, + { + "epoch": 2.0704030922142462, + "grad_norm": 0.9003238677978516, + "learning_rate": 3.730585628021237e-06, + "loss": 0.6124, + "step": 2501 + }, + { + "epoch": 2.071231363887355, + "grad_norm": 0.9069858193397522, + "learning_rate": 3.729628270319791e-06, + "loss": 0.6051, + "step": 2502 + }, + { + "epoch": 2.0720596355604637, + "grad_norm": 0.9361990094184875, + "learning_rate": 3.7286706746967372e-06, + "loss": 0.5863, + "step": 2503 + }, + { + "epoch": 2.0728879072335724, + "grad_norm": 0.9443079233169556, + "learning_rate": 3.7277128413373607e-06, + "loss": 0.5959, + "step": 2504 + }, + { + "epoch": 2.0737161789066816, + "grad_norm": 0.9413166046142578, + "learning_rate": 3.7267547704269957e-06, + "loss": 0.5882, + "step": 2505 + }, + { + "epoch": 2.0745444505797903, + "grad_norm": 0.8937835097312927, + "learning_rate": 3.7257964621510167e-06, + "loss": 0.6108, + "step": 2506 + }, + { + "epoch": 2.075372722252899, + "grad_norm": 0.9188135862350464, + "learning_rate": 3.7248379166948497e-06, + "loss": 0.5812, + "step": 2507 + }, + { + "epoch": 2.0762009939260078, + "grad_norm": 0.916559636592865, + "learning_rate": 3.723879134243963e-06, + "loss": 0.6169, + "step": 2508 + }, + { + "epoch": 2.0770292655991165, + "grad_norm": 0.9284812211990356, + "learning_rate": 3.7229201149838727e-06, + "loss": 0.6224, + "step": 2509 + }, + { + "epoch": 2.077857537272225, + "grad_norm": 0.918048083782196, + "learning_rate": 3.721960859100141e-06, + "loss": 0.6043, + "step": 2510 + }, + { + "epoch": 2.078685808945334, + "grad_norm": 0.9250833988189697, + "learning_rate": 3.721001366778373e-06, + "loss": 0.5882, + "step": 2511 + }, + { + "epoch": 2.0795140806184427, + "grad_norm": 0.8975113034248352, + "learning_rate": 3.720041638204223e-06, + "loss": 0.6049, + "step": 2512 + }, + { + "epoch": 2.080342352291552, + "grad_norm": 0.9108197093009949, + "learning_rate": 3.7190816735633896e-06, + "loss": 0.6174, + "step": 2513 + }, + { + "epoch": 2.0811706239646606, + "grad_norm": 0.91886967420578, + "learning_rate": 3.718121473041615e-06, + "loss": 0.5854, + "step": 2514 + }, + { + "epoch": 2.0819988956377693, + "grad_norm": 0.9149959683418274, + "learning_rate": 3.717161036824692e-06, + "loss": 0.5808, + "step": 2515 + }, + { + "epoch": 2.082827167310878, + "grad_norm": 0.9402881860733032, + "learning_rate": 3.7162003650984545e-06, + "loss": 0.5936, + "step": 2516 + }, + { + "epoch": 2.0836554389839868, + "grad_norm": 0.8952768445014954, + "learning_rate": 3.7152394580487825e-06, + "loss": 0.5977, + "step": 2517 + }, + { + "epoch": 2.0844837106570955, + "grad_norm": 0.9138445258140564, + "learning_rate": 3.714278315861605e-06, + "loss": 0.5822, + "step": 2518 + }, + { + "epoch": 2.085311982330204, + "grad_norm": 0.9174036979675293, + "learning_rate": 3.713316938722892e-06, + "loss": 0.6061, + "step": 2519 + }, + { + "epoch": 2.086140254003313, + "grad_norm": 0.9291302561759949, + "learning_rate": 3.7123553268186618e-06, + "loss": 0.614, + "step": 2520 + }, + { + "epoch": 2.0869685256764217, + "grad_norm": 0.9058101773262024, + "learning_rate": 3.711393480334977e-06, + "loss": 0.6053, + "step": 2521 + }, + { + "epoch": 2.087796797349531, + "grad_norm": 0.926360011100769, + "learning_rate": 3.7104313994579473e-06, + "loss": 0.5781, + "step": 2522 + }, + { + "epoch": 2.0886250690226396, + "grad_norm": 0.898629367351532, + "learning_rate": 3.709469084373725e-06, + "loss": 0.5995, + "step": 2523 + }, + { + "epoch": 2.0894533406957483, + "grad_norm": 0.9013628363609314, + "learning_rate": 3.708506535268509e-06, + "loss": 0.5929, + "step": 2524 + }, + { + "epoch": 2.090281612368857, + "grad_norm": 0.8909673094749451, + "learning_rate": 3.7075437523285433e-06, + "loss": 0.5495, + "step": 2525 + }, + { + "epoch": 2.0911098840419657, + "grad_norm": 0.9176347255706787, + "learning_rate": 3.706580735740119e-06, + "loss": 0.6134, + "step": 2526 + }, + { + "epoch": 2.0919381557150745, + "grad_norm": 0.9174275398254395, + "learning_rate": 3.7056174856895703e-06, + "loss": 0.601, + "step": 2527 + }, + { + "epoch": 2.092766427388183, + "grad_norm": 0.9134400486946106, + "learning_rate": 3.704654002363276e-06, + "loss": 0.5952, + "step": 2528 + }, + { + "epoch": 2.093594699061292, + "grad_norm": 0.9206627607345581, + "learning_rate": 3.703690285947662e-06, + "loss": 0.6016, + "step": 2529 + }, + { + "epoch": 2.094422970734401, + "grad_norm": 0.9271345734596252, + "learning_rate": 3.702726336629199e-06, + "loss": 0.5937, + "step": 2530 + }, + { + "epoch": 2.09525124240751, + "grad_norm": 0.9390851855278015, + "learning_rate": 3.7017621545944003e-06, + "loss": 0.5858, + "step": 2531 + }, + { + "epoch": 2.0960795140806185, + "grad_norm": 0.9351701736450195, + "learning_rate": 3.700797740029828e-06, + "loss": 0.5978, + "step": 2532 + }, + { + "epoch": 2.0969077857537273, + "grad_norm": 0.9432377815246582, + "learning_rate": 3.6998330931220873e-06, + "loss": 0.5884, + "step": 2533 + }, + { + "epoch": 2.097736057426836, + "grad_norm": 0.9192067384719849, + "learning_rate": 3.698868214057827e-06, + "loss": 0.5989, + "step": 2534 + }, + { + "epoch": 2.0985643290999447, + "grad_norm": 0.9810166954994202, + "learning_rate": 3.6979031030237433e-06, + "loss": 0.6078, + "step": 2535 + }, + { + "epoch": 2.0993926007730535, + "grad_norm": 0.9065185785293579, + "learning_rate": 3.696937760206576e-06, + "loss": 0.5947, + "step": 2536 + }, + { + "epoch": 2.100220872446162, + "grad_norm": 0.9212245345115662, + "learning_rate": 3.6959721857931088e-06, + "loss": 0.5951, + "step": 2537 + }, + { + "epoch": 2.101049144119271, + "grad_norm": 0.9291266798973083, + "learning_rate": 3.695006379970173e-06, + "loss": 0.6033, + "step": 2538 + }, + { + "epoch": 2.10187741579238, + "grad_norm": 0.9154751896858215, + "learning_rate": 3.694040342924643e-06, + "loss": 0.612, + "step": 2539 + }, + { + "epoch": 2.102705687465489, + "grad_norm": 0.9028742909431458, + "learning_rate": 3.693074074843437e-06, + "loss": 0.5948, + "step": 2540 + }, + { + "epoch": 2.1035339591385975, + "grad_norm": 0.9132126569747925, + "learning_rate": 3.692107575913519e-06, + "loss": 0.6198, + "step": 2541 + }, + { + "epoch": 2.1043622308117063, + "grad_norm": 0.9116196036338806, + "learning_rate": 3.6911408463218974e-06, + "loss": 0.5927, + "step": 2542 + }, + { + "epoch": 2.105190502484815, + "grad_norm": 0.9670912027359009, + "learning_rate": 3.6901738862556263e-06, + "loss": 0.5813, + "step": 2543 + }, + { + "epoch": 2.1060187741579237, + "grad_norm": 0.9534422755241394, + "learning_rate": 3.6892066959018034e-06, + "loss": 0.5892, + "step": 2544 + }, + { + "epoch": 2.1068470458310324, + "grad_norm": 0.9175525903701782, + "learning_rate": 3.688239275447569e-06, + "loss": 0.5946, + "step": 2545 + }, + { + "epoch": 2.107675317504141, + "grad_norm": 0.8945409655570984, + "learning_rate": 3.6872716250801125e-06, + "loss": 0.6013, + "step": 2546 + }, + { + "epoch": 2.1085035891772503, + "grad_norm": 0.9498820900917053, + "learning_rate": 3.686303744986664e-06, + "loss": 0.5785, + "step": 2547 + }, + { + "epoch": 2.109331860850359, + "grad_norm": 0.9338169693946838, + "learning_rate": 3.685335635354499e-06, + "loss": 0.5959, + "step": 2548 + }, + { + "epoch": 2.110160132523468, + "grad_norm": 0.9100049734115601, + "learning_rate": 3.684367296370939e-06, + "loss": 0.5829, + "step": 2549 + }, + { + "epoch": 2.1109884041965765, + "grad_norm": 0.904279887676239, + "learning_rate": 3.6833987282233467e-06, + "loss": 0.6015, + "step": 2550 + }, + { + "epoch": 2.1118166758696852, + "grad_norm": 0.8693552613258362, + "learning_rate": 3.6824299310991324e-06, + "loss": 0.5825, + "step": 2551 + }, + { + "epoch": 2.112644947542794, + "grad_norm": 0.9180129766464233, + "learning_rate": 3.681460905185749e-06, + "loss": 0.6076, + "step": 2552 + }, + { + "epoch": 2.1134732192159027, + "grad_norm": 0.9554781913757324, + "learning_rate": 3.6804916506706936e-06, + "loss": 0.5755, + "step": 2553 + }, + { + "epoch": 2.1143014908890114, + "grad_norm": 0.9498181939125061, + "learning_rate": 3.6795221677415085e-06, + "loss": 0.6086, + "step": 2554 + }, + { + "epoch": 2.1151297625621206, + "grad_norm": 0.9110347032546997, + "learning_rate": 3.678552456585779e-06, + "loss": 0.591, + "step": 2555 + }, + { + "epoch": 2.1159580342352293, + "grad_norm": 0.8883147835731506, + "learning_rate": 3.6775825173911363e-06, + "loss": 0.5753, + "step": 2556 + }, + { + "epoch": 2.116786305908338, + "grad_norm": 0.9354976415634155, + "learning_rate": 3.676612350345253e-06, + "loss": 0.6063, + "step": 2557 + }, + { + "epoch": 2.1176145775814468, + "grad_norm": 0.9463969469070435, + "learning_rate": 3.675641955635848e-06, + "loss": 0.5983, + "step": 2558 + }, + { + "epoch": 2.1184428492545555, + "grad_norm": 0.9220679998397827, + "learning_rate": 3.674671333450684e-06, + "loss": 0.6026, + "step": 2559 + }, + { + "epoch": 2.1192711209276642, + "grad_norm": 0.9050368666648865, + "learning_rate": 3.673700483977567e-06, + "loss": 0.6075, + "step": 2560 + }, + { + "epoch": 2.120099392600773, + "grad_norm": 0.9299139976501465, + "learning_rate": 3.6727294074043478e-06, + "loss": 0.5971, + "step": 2561 + }, + { + "epoch": 2.1209276642738817, + "grad_norm": 0.9060289263725281, + "learning_rate": 3.6717581039189205e-06, + "loss": 0.5794, + "step": 2562 + }, + { + "epoch": 2.1217559359469904, + "grad_norm": 0.9253103137016296, + "learning_rate": 3.6707865737092228e-06, + "loss": 0.593, + "step": 2563 + }, + { + "epoch": 2.1225842076200996, + "grad_norm": 0.9475048780441284, + "learning_rate": 3.669814816963237e-06, + "loss": 0.5867, + "step": 2564 + }, + { + "epoch": 2.1234124792932083, + "grad_norm": 0.9546397924423218, + "learning_rate": 3.6688428338689887e-06, + "loss": 0.5766, + "step": 2565 + }, + { + "epoch": 2.124240750966317, + "grad_norm": 0.9377962350845337, + "learning_rate": 3.667870624614548e-06, + "loss": 0.5853, + "step": 2566 + }, + { + "epoch": 2.1250690226394258, + "grad_norm": 0.9562273621559143, + "learning_rate": 3.6668981893880284e-06, + "loss": 0.5992, + "step": 2567 + }, + { + "epoch": 2.1258972943125345, + "grad_norm": 0.9207432866096497, + "learning_rate": 3.6659255283775867e-06, + "loss": 0.6029, + "step": 2568 + }, + { + "epoch": 2.126725565985643, + "grad_norm": 0.9359074234962463, + "learning_rate": 3.6649526417714244e-06, + "loss": 0.6068, + "step": 2569 + }, + { + "epoch": 2.127553837658752, + "grad_norm": 0.9253597259521484, + "learning_rate": 3.663979529757784e-06, + "loss": 0.581, + "step": 2570 + }, + { + "epoch": 2.1283821093318607, + "grad_norm": 0.9551291465759277, + "learning_rate": 3.6630061925249566e-06, + "loss": 0.607, + "step": 2571 + }, + { + "epoch": 2.1292103810049694, + "grad_norm": 0.91109699010849, + "learning_rate": 3.662032630261272e-06, + "loss": 0.5724, + "step": 2572 + }, + { + "epoch": 2.1300386526780786, + "grad_norm": 0.9486557245254517, + "learning_rate": 3.661058843155105e-06, + "loss": 0.5904, + "step": 2573 + }, + { + "epoch": 2.1308669243511873, + "grad_norm": 0.9550185203552246, + "learning_rate": 3.6600848313948762e-06, + "loss": 0.5967, + "step": 2574 + }, + { + "epoch": 2.131695196024296, + "grad_norm": 0.9457482695579529, + "learning_rate": 3.6591105951690464e-06, + "loss": 0.6253, + "step": 2575 + }, + { + "epoch": 2.1325234676974048, + "grad_norm": 0.9686241149902344, + "learning_rate": 3.6581361346661204e-06, + "loss": 0.5928, + "step": 2576 + }, + { + "epoch": 2.1333517393705135, + "grad_norm": 0.9125609993934631, + "learning_rate": 3.65716145007465e-06, + "loss": 0.6039, + "step": 2577 + }, + { + "epoch": 2.134180011043622, + "grad_norm": 0.927651047706604, + "learning_rate": 3.656186541583225e-06, + "loss": 0.5881, + "step": 2578 + }, + { + "epoch": 2.135008282716731, + "grad_norm": 0.9446879029273987, + "learning_rate": 3.6552114093804825e-06, + "loss": 0.5938, + "step": 2579 + }, + { + "epoch": 2.13583655438984, + "grad_norm": 0.944601833820343, + "learning_rate": 3.6542360536551003e-06, + "loss": 0.5804, + "step": 2580 + }, + { + "epoch": 2.136664826062949, + "grad_norm": 0.9230138063430786, + "learning_rate": 3.6532604745958015e-06, + "loss": 0.6002, + "step": 2581 + }, + { + "epoch": 2.1374930977360576, + "grad_norm": 0.9163529276847839, + "learning_rate": 3.6522846723913517e-06, + "loss": 0.5811, + "step": 2582 + }, + { + "epoch": 2.1383213694091663, + "grad_norm": 0.9153618812561035, + "learning_rate": 3.6513086472305596e-06, + "loss": 0.6018, + "step": 2583 + }, + { + "epoch": 2.139149641082275, + "grad_norm": 0.9299359917640686, + "learning_rate": 3.6503323993022767e-06, + "loss": 0.5944, + "step": 2584 + }, + { + "epoch": 2.1399779127553837, + "grad_norm": 0.9075869917869568, + "learning_rate": 3.6493559287953974e-06, + "loss": 0.595, + "step": 2585 + }, + { + "epoch": 2.1408061844284925, + "grad_norm": 0.9388751983642578, + "learning_rate": 3.64837923589886e-06, + "loss": 0.6006, + "step": 2586 + }, + { + "epoch": 2.141634456101601, + "grad_norm": 0.91156405210495, + "learning_rate": 3.647402320801645e-06, + "loss": 0.5955, + "step": 2587 + }, + { + "epoch": 2.14246272777471, + "grad_norm": 0.9187628626823425, + "learning_rate": 3.6464251836927776e-06, + "loss": 0.6119, + "step": 2588 + }, + { + "epoch": 2.143290999447819, + "grad_norm": 0.9578025341033936, + "learning_rate": 3.645447824761324e-06, + "loss": 0.6076, + "step": 2589 + }, + { + "epoch": 2.144119271120928, + "grad_norm": 0.9159366488456726, + "learning_rate": 3.644470244196394e-06, + "loss": 0.6015, + "step": 2590 + }, + { + "epoch": 2.1449475427940365, + "grad_norm": 0.924027681350708, + "learning_rate": 3.6434924421871403e-06, + "loss": 0.6082, + "step": 2591 + }, + { + "epoch": 2.1457758144671453, + "grad_norm": 0.8900635242462158, + "learning_rate": 3.642514418922758e-06, + "loss": 0.611, + "step": 2592 + }, + { + "epoch": 2.146604086140254, + "grad_norm": 0.9609982371330261, + "learning_rate": 3.641536174592487e-06, + "loss": 0.5974, + "step": 2593 + }, + { + "epoch": 2.1474323578133627, + "grad_norm": 0.901737630367279, + "learning_rate": 3.6405577093856072e-06, + "loss": 0.5869, + "step": 2594 + }, + { + "epoch": 2.1482606294864715, + "grad_norm": 0.9342123866081238, + "learning_rate": 3.6395790234914423e-06, + "loss": 0.6104, + "step": 2595 + }, + { + "epoch": 2.14908890115958, + "grad_norm": 0.9063602685928345, + "learning_rate": 3.6386001170993596e-06, + "loss": 0.5801, + "step": 2596 + }, + { + "epoch": 2.149917172832689, + "grad_norm": 0.9269636869430542, + "learning_rate": 3.637620990398768e-06, + "loss": 0.5953, + "step": 2597 + }, + { + "epoch": 2.150745444505798, + "grad_norm": 0.9278765320777893, + "learning_rate": 3.63664164357912e-06, + "loss": 0.6151, + "step": 2598 + }, + { + "epoch": 2.151573716178907, + "grad_norm": 0.9786942601203918, + "learning_rate": 3.6356620768299087e-06, + "loss": 0.5773, + "step": 2599 + }, + { + "epoch": 2.1524019878520155, + "grad_norm": 0.9069851636886597, + "learning_rate": 3.6346822903406733e-06, + "loss": 0.6087, + "step": 2600 + }, + { + "epoch": 2.1532302595251243, + "grad_norm": 0.8867887854576111, + "learning_rate": 3.6337022843009913e-06, + "loss": 0.5894, + "step": 2601 + }, + { + "epoch": 2.154058531198233, + "grad_norm": 0.9444507360458374, + "learning_rate": 3.632722058900486e-06, + "loss": 0.6125, + "step": 2602 + }, + { + "epoch": 2.1548868028713417, + "grad_norm": 0.9410359859466553, + "learning_rate": 3.631741614328821e-06, + "loss": 0.5845, + "step": 2603 + }, + { + "epoch": 2.1557150745444504, + "grad_norm": 0.9473987817764282, + "learning_rate": 3.630760950775703e-06, + "loss": 0.5944, + "step": 2604 + }, + { + "epoch": 2.156543346217559, + "grad_norm": 0.913758397102356, + "learning_rate": 3.6297800684308828e-06, + "loss": 0.5998, + "step": 2605 + }, + { + "epoch": 2.1573716178906683, + "grad_norm": 0.89509516954422, + "learning_rate": 3.6287989674841505e-06, + "loss": 0.6013, + "step": 2606 + }, + { + "epoch": 2.158199889563777, + "grad_norm": 0.9042367339134216, + "learning_rate": 3.6278176481253413e-06, + "loss": 0.5915, + "step": 2607 + }, + { + "epoch": 2.159028161236886, + "grad_norm": 0.9108945727348328, + "learning_rate": 3.6268361105443297e-06, + "loss": 0.5803, + "step": 2608 + }, + { + "epoch": 2.1598564329099945, + "grad_norm": 0.8903773427009583, + "learning_rate": 3.625854354931035e-06, + "loss": 0.6025, + "step": 2609 + }, + { + "epoch": 2.1606847045831032, + "grad_norm": 0.9396197199821472, + "learning_rate": 3.6248723814754183e-06, + "loss": 0.6271, + "step": 2610 + }, + { + "epoch": 2.161512976256212, + "grad_norm": 0.9247238039970398, + "learning_rate": 3.623890190367481e-06, + "loss": 0.5945, + "step": 2611 + }, + { + "epoch": 2.1623412479293207, + "grad_norm": 0.9240396022796631, + "learning_rate": 3.622907781797269e-06, + "loss": 0.6148, + "step": 2612 + }, + { + "epoch": 2.1631695196024294, + "grad_norm": 0.9257000684738159, + "learning_rate": 3.621925155954869e-06, + "loss": 0.5964, + "step": 2613 + }, + { + "epoch": 2.1639977912755386, + "grad_norm": 0.8854860663414001, + "learning_rate": 3.62094231303041e-06, + "loss": 0.6012, + "step": 2614 + }, + { + "epoch": 2.1648260629486473, + "grad_norm": 0.9251745343208313, + "learning_rate": 3.619959253214062e-06, + "loss": 0.6001, + "step": 2615 + }, + { + "epoch": 2.165654334621756, + "grad_norm": 0.9081252813339233, + "learning_rate": 3.618975976696039e-06, + "loss": 0.5927, + "step": 2616 + }, + { + "epoch": 2.1664826062948648, + "grad_norm": 0.9065699577331543, + "learning_rate": 3.6179924836665953e-06, + "loss": 0.6079, + "step": 2617 + }, + { + "epoch": 2.1673108779679735, + "grad_norm": 0.9421206712722778, + "learning_rate": 3.617008774316028e-06, + "loss": 0.6041, + "step": 2618 + }, + { + "epoch": 2.1681391496410822, + "grad_norm": 0.8863933682441711, + "learning_rate": 3.6160248488346757e-06, + "loss": 0.5826, + "step": 2619 + }, + { + "epoch": 2.168967421314191, + "grad_norm": 0.915887713432312, + "learning_rate": 3.615040707412918e-06, + "loss": 0.5972, + "step": 2620 + }, + { + "epoch": 2.1697956929872997, + "grad_norm": 0.9150416851043701, + "learning_rate": 3.614056350241178e-06, + "loss": 0.606, + "step": 2621 + }, + { + "epoch": 2.1706239646604084, + "grad_norm": 0.9050796627998352, + "learning_rate": 3.6130717775099195e-06, + "loss": 0.5955, + "step": 2622 + }, + { + "epoch": 2.1714522363335176, + "grad_norm": 0.9249832630157471, + "learning_rate": 3.612086989409648e-06, + "loss": 0.5968, + "step": 2623 + }, + { + "epoch": 2.1722805080066263, + "grad_norm": 0.8855478763580322, + "learning_rate": 3.6111019861309105e-06, + "loss": 0.5657, + "step": 2624 + }, + { + "epoch": 2.173108779679735, + "grad_norm": 0.9182875156402588, + "learning_rate": 3.6101167678642967e-06, + "loss": 0.6096, + "step": 2625 + }, + { + "epoch": 2.1739370513528438, + "grad_norm": 0.9722150564193726, + "learning_rate": 3.609131334800436e-06, + "loss": 0.6029, + "step": 2626 + }, + { + "epoch": 2.1747653230259525, + "grad_norm": 0.8925914168357849, + "learning_rate": 3.6081456871300014e-06, + "loss": 0.5928, + "step": 2627 + }, + { + "epoch": 2.175593594699061, + "grad_norm": 0.8883031010627747, + "learning_rate": 3.6071598250437067e-06, + "loss": 0.5763, + "step": 2628 + }, + { + "epoch": 2.17642186637217, + "grad_norm": 0.8948857188224792, + "learning_rate": 3.606173748732307e-06, + "loss": 0.6078, + "step": 2629 + }, + { + "epoch": 2.1772501380452787, + "grad_norm": 0.9444127082824707, + "learning_rate": 3.6051874583865975e-06, + "loss": 0.6005, + "step": 2630 + }, + { + "epoch": 2.1780784097183874, + "grad_norm": 0.9361352920532227, + "learning_rate": 3.6042009541974176e-06, + "loss": 0.5878, + "step": 2631 + }, + { + "epoch": 2.1789066813914966, + "grad_norm": 0.9124311804771423, + "learning_rate": 3.6032142363556467e-06, + "loss": 0.6115, + "step": 2632 + }, + { + "epoch": 2.1797349530646053, + "grad_norm": 0.9168087840080261, + "learning_rate": 3.602227305052205e-06, + "loss": 0.5928, + "step": 2633 + }, + { + "epoch": 2.180563224737714, + "grad_norm": 0.9189275503158569, + "learning_rate": 3.6012401604780543e-06, + "loss": 0.6081, + "step": 2634 + }, + { + "epoch": 2.1813914964108228, + "grad_norm": 0.937911868095398, + "learning_rate": 3.6002528028241984e-06, + "loss": 0.6062, + "step": 2635 + }, + { + "epoch": 2.1822197680839315, + "grad_norm": 0.9139302968978882, + "learning_rate": 3.5992652322816817e-06, + "loss": 0.6111, + "step": 2636 + }, + { + "epoch": 2.18304803975704, + "grad_norm": 0.951528012752533, + "learning_rate": 3.598277449041589e-06, + "loss": 0.6007, + "step": 2637 + }, + { + "epoch": 2.183876311430149, + "grad_norm": 0.9146255850791931, + "learning_rate": 3.5972894532950487e-06, + "loss": 0.6094, + "step": 2638 + }, + { + "epoch": 2.184704583103258, + "grad_norm": 0.9217225313186646, + "learning_rate": 3.596301245233228e-06, + "loss": 0.6082, + "step": 2639 + }, + { + "epoch": 2.185532854776367, + "grad_norm": 0.9742298126220703, + "learning_rate": 3.5953128250473355e-06, + "loss": 0.6269, + "step": 2640 + }, + { + "epoch": 2.1863611264494756, + "grad_norm": 0.9038170576095581, + "learning_rate": 3.5943241929286223e-06, + "loss": 0.6062, + "step": 2641 + }, + { + "epoch": 2.1871893981225843, + "grad_norm": 0.8960873484611511, + "learning_rate": 3.5933353490683786e-06, + "loss": 0.5748, + "step": 2642 + }, + { + "epoch": 2.188017669795693, + "grad_norm": 0.8801732659339905, + "learning_rate": 3.5923462936579362e-06, + "loss": 0.5911, + "step": 2643 + }, + { + "epoch": 2.1888459414688017, + "grad_norm": 0.8918222188949585, + "learning_rate": 3.5913570268886695e-06, + "loss": 0.5931, + "step": 2644 + }, + { + "epoch": 2.1896742131419105, + "grad_norm": 0.9046133160591125, + "learning_rate": 3.5903675489519917e-06, + "loss": 0.5985, + "step": 2645 + }, + { + "epoch": 2.190502484815019, + "grad_norm": 0.9003219604492188, + "learning_rate": 3.589377860039357e-06, + "loss": 0.608, + "step": 2646 + }, + { + "epoch": 2.191330756488128, + "grad_norm": 0.9139525890350342, + "learning_rate": 3.5883879603422616e-06, + "loss": 0.6186, + "step": 2647 + }, + { + "epoch": 2.192159028161237, + "grad_norm": 0.9125071167945862, + "learning_rate": 3.587397850052241e-06, + "loss": 0.6129, + "step": 2648 + }, + { + "epoch": 2.192987299834346, + "grad_norm": 0.922412097454071, + "learning_rate": 3.586407529360874e-06, + "loss": 0.6008, + "step": 2649 + }, + { + "epoch": 2.1938155715074545, + "grad_norm": 0.9294847249984741, + "learning_rate": 3.5854169984597776e-06, + "loss": 0.5984, + "step": 2650 + }, + { + "epoch": 2.1946438431805633, + "grad_norm": 0.9262758493423462, + "learning_rate": 3.5844262575406084e-06, + "loss": 0.5982, + "step": 2651 + }, + { + "epoch": 2.195472114853672, + "grad_norm": 0.9311552047729492, + "learning_rate": 3.5834353067950694e-06, + "loss": 0.5965, + "step": 2652 + }, + { + "epoch": 2.1963003865267807, + "grad_norm": 0.9087333679199219, + "learning_rate": 3.5824441464148975e-06, + "loss": 0.5972, + "step": 2653 + }, + { + "epoch": 2.1971286581998895, + "grad_norm": 0.9212601184844971, + "learning_rate": 3.5814527765918733e-06, + "loss": 0.5821, + "step": 2654 + }, + { + "epoch": 2.197956929872998, + "grad_norm": 0.9208887219429016, + "learning_rate": 3.5804611975178184e-06, + "loss": 0.592, + "step": 2655 + }, + { + "epoch": 2.198785201546107, + "grad_norm": 0.923478364944458, + "learning_rate": 3.5794694093845933e-06, + "loss": 0.599, + "step": 2656 + }, + { + "epoch": 2.199613473219216, + "grad_norm": 0.9030081033706665, + "learning_rate": 3.5784774123841006e-06, + "loss": 0.5954, + "step": 2657 + }, + { + "epoch": 2.200441744892325, + "grad_norm": 0.9108036160469055, + "learning_rate": 3.5774852067082817e-06, + "loss": 0.6119, + "step": 2658 + }, + { + "epoch": 2.2012700165654335, + "grad_norm": 0.9323570728302002, + "learning_rate": 3.576492792549119e-06, + "loss": 0.6061, + "step": 2659 + }, + { + "epoch": 2.2020982882385423, + "grad_norm": 0.9002283811569214, + "learning_rate": 3.575500170098637e-06, + "loss": 0.6183, + "step": 2660 + }, + { + "epoch": 2.202926559911651, + "grad_norm": 0.9354086518287659, + "learning_rate": 3.574507339548897e-06, + "loss": 0.5989, + "step": 2661 + }, + { + "epoch": 2.2037548315847597, + "grad_norm": 0.9035785794258118, + "learning_rate": 3.5735143010920033e-06, + "loss": 0.592, + "step": 2662 + }, + { + "epoch": 2.2045831032578684, + "grad_norm": 0.9337678551673889, + "learning_rate": 3.5725210549201e-06, + "loss": 0.5881, + "step": 2663 + }, + { + "epoch": 2.205411374930977, + "grad_norm": 0.9226170182228088, + "learning_rate": 3.57152760122537e-06, + "loss": 0.5745, + "step": 2664 + }, + { + "epoch": 2.2062396466040863, + "grad_norm": 0.9272558093070984, + "learning_rate": 3.5705339402000377e-06, + "loss": 0.607, + "step": 2665 + }, + { + "epoch": 2.207067918277195, + "grad_norm": 0.930325448513031, + "learning_rate": 3.569540072036368e-06, + "loss": 0.5947, + "step": 2666 + }, + { + "epoch": 2.207896189950304, + "grad_norm": 0.9052232503890991, + "learning_rate": 3.568545996926664e-06, + "loss": 0.5883, + "step": 2667 + }, + { + "epoch": 2.2087244616234125, + "grad_norm": 0.8831207752227783, + "learning_rate": 3.5675517150632706e-06, + "loss": 0.5962, + "step": 2668 + }, + { + "epoch": 2.2095527332965212, + "grad_norm": 0.8999590873718262, + "learning_rate": 3.566557226638572e-06, + "loss": 0.6257, + "step": 2669 + }, + { + "epoch": 2.21038100496963, + "grad_norm": 0.9310323596000671, + "learning_rate": 3.565562531844992e-06, + "loss": 0.585, + "step": 2670 + }, + { + "epoch": 2.2112092766427387, + "grad_norm": 0.9469419121742249, + "learning_rate": 3.5645676308749954e-06, + "loss": 0.6163, + "step": 2671 + }, + { + "epoch": 2.2120375483158474, + "grad_norm": 0.9391891956329346, + "learning_rate": 3.563572523921086e-06, + "loss": 0.5863, + "step": 2672 + }, + { + "epoch": 2.2128658199889566, + "grad_norm": 0.9035549163818359, + "learning_rate": 3.5625772111758085e-06, + "loss": 0.6074, + "step": 2673 + }, + { + "epoch": 2.2136940916620653, + "grad_norm": 0.9008681178092957, + "learning_rate": 3.561581692831745e-06, + "loss": 0.5965, + "step": 2674 + }, + { + "epoch": 2.214522363335174, + "grad_norm": 0.9130175113677979, + "learning_rate": 3.56058596908152e-06, + "loss": 0.5913, + "step": 2675 + }, + { + "epoch": 2.2153506350082828, + "grad_norm": 0.9367324113845825, + "learning_rate": 3.5595900401177957e-06, + "loss": 0.617, + "step": 2676 + }, + { + "epoch": 2.2161789066813915, + "grad_norm": 0.9082489609718323, + "learning_rate": 3.5585939061332775e-06, + "loss": 0.6052, + "step": 2677 + }, + { + "epoch": 2.2170071783545002, + "grad_norm": 0.921264111995697, + "learning_rate": 3.557597567320706e-06, + "loss": 0.5832, + "step": 2678 + }, + { + "epoch": 2.217835450027609, + "grad_norm": 0.9040131568908691, + "learning_rate": 3.556601023872863e-06, + "loss": 0.5994, + "step": 2679 + }, + { + "epoch": 2.2186637217007177, + "grad_norm": 0.9144189357757568, + "learning_rate": 3.5556042759825733e-06, + "loss": 0.6068, + "step": 2680 + }, + { + "epoch": 2.2194919933738264, + "grad_norm": 0.91495680809021, + "learning_rate": 3.5546073238426955e-06, + "loss": 0.6154, + "step": 2681 + }, + { + "epoch": 2.2203202650469356, + "grad_norm": 0.925397515296936, + "learning_rate": 3.553610167646131e-06, + "loss": 0.6013, + "step": 2682 + }, + { + "epoch": 2.2211485367200443, + "grad_norm": 0.9633582234382629, + "learning_rate": 3.552612807585821e-06, + "loss": 0.6063, + "step": 2683 + }, + { + "epoch": 2.221976808393153, + "grad_norm": 0.9403156042098999, + "learning_rate": 3.551615243854746e-06, + "loss": 0.6141, + "step": 2684 + }, + { + "epoch": 2.2228050800662618, + "grad_norm": 0.9272189140319824, + "learning_rate": 3.550617476645923e-06, + "loss": 0.6013, + "step": 2685 + }, + { + "epoch": 2.2236333517393705, + "grad_norm": 0.92893385887146, + "learning_rate": 3.549619506152413e-06, + "loss": 0.5909, + "step": 2686 + }, + { + "epoch": 2.224461623412479, + "grad_norm": 0.9130013585090637, + "learning_rate": 3.548621332567312e-06, + "loss": 0.6005, + "step": 2687 + }, + { + "epoch": 2.225289895085588, + "grad_norm": 0.9111050367355347, + "learning_rate": 3.547622956083759e-06, + "loss": 0.5904, + "step": 2688 + }, + { + "epoch": 2.2261181667586967, + "grad_norm": 0.9379095435142517, + "learning_rate": 3.5466243768949292e-06, + "loss": 0.605, + "step": 2689 + }, + { + "epoch": 2.2269464384318054, + "grad_norm": 0.898931622505188, + "learning_rate": 3.5456255951940393e-06, + "loss": 0.5978, + "step": 2690 + }, + { + "epoch": 2.2277747101049146, + "grad_norm": 0.9569118022918701, + "learning_rate": 3.5446266111743444e-06, + "loss": 0.6162, + "step": 2691 + }, + { + "epoch": 2.2286029817780233, + "grad_norm": 0.9167311191558838, + "learning_rate": 3.543627425029138e-06, + "loss": 0.5876, + "step": 2692 + }, + { + "epoch": 2.229431253451132, + "grad_norm": 0.9584367275238037, + "learning_rate": 3.542628036951753e-06, + "loss": 0.5704, + "step": 2693 + }, + { + "epoch": 2.2302595251242407, + "grad_norm": 0.8896269202232361, + "learning_rate": 3.541628447135562e-06, + "loss": 0.5806, + "step": 2694 + }, + { + "epoch": 2.2310877967973495, + "grad_norm": 0.9340798854827881, + "learning_rate": 3.5406286557739767e-06, + "loss": 0.6218, + "step": 2695 + }, + { + "epoch": 2.231916068470458, + "grad_norm": 0.9442514777183533, + "learning_rate": 3.539628663060447e-06, + "loss": 0.5899, + "step": 2696 + }, + { + "epoch": 2.232744340143567, + "grad_norm": 0.9123642444610596, + "learning_rate": 3.5386284691884633e-06, + "loss": 0.5825, + "step": 2697 + }, + { + "epoch": 2.233572611816676, + "grad_norm": 0.9076257348060608, + "learning_rate": 3.537628074351551e-06, + "loss": 0.6001, + "step": 2698 + }, + { + "epoch": 2.234400883489785, + "grad_norm": 0.9335387945175171, + "learning_rate": 3.5366274787432804e-06, + "loss": 0.6048, + "step": 2699 + }, + { + "epoch": 2.2352291551628936, + "grad_norm": 0.9016919732093811, + "learning_rate": 3.535626682557255e-06, + "loss": 0.6175, + "step": 2700 + }, + { + "epoch": 2.2360574268360023, + "grad_norm": 0.9110099673271179, + "learning_rate": 3.5346256859871207e-06, + "loss": 0.6027, + "step": 2701 + }, + { + "epoch": 2.236885698509111, + "grad_norm": 0.9513245224952698, + "learning_rate": 3.5336244892265616e-06, + "loss": 0.6015, + "step": 2702 + }, + { + "epoch": 2.2377139701822197, + "grad_norm": 0.8986294865608215, + "learning_rate": 3.532623092469299e-06, + "loss": 0.61, + "step": 2703 + }, + { + "epoch": 2.2385422418553285, + "grad_norm": 0.8991696834564209, + "learning_rate": 3.5316214959090933e-06, + "loss": 0.6013, + "step": 2704 + }, + { + "epoch": 2.239370513528437, + "grad_norm": 0.9667947292327881, + "learning_rate": 3.5306196997397453e-06, + "loss": 0.6125, + "step": 2705 + }, + { + "epoch": 2.240198785201546, + "grad_norm": 0.9183294177055359, + "learning_rate": 3.529617704155092e-06, + "loss": 0.5901, + "step": 2706 + }, + { + "epoch": 2.241027056874655, + "grad_norm": 0.9139719605445862, + "learning_rate": 3.5286155093490117e-06, + "loss": 0.5798, + "step": 2707 + }, + { + "epoch": 2.241855328547764, + "grad_norm": 0.9529849290847778, + "learning_rate": 3.5276131155154193e-06, + "loss": 0.5763, + "step": 2708 + }, + { + "epoch": 2.2426836002208725, + "grad_norm": 0.9126972556114197, + "learning_rate": 3.526610522848267e-06, + "loss": 0.5804, + "step": 2709 + }, + { + "epoch": 2.2435118718939813, + "grad_norm": 0.9101532101631165, + "learning_rate": 3.525607731541549e-06, + "loss": 0.5815, + "step": 2710 + }, + { + "epoch": 2.24434014356709, + "grad_norm": 0.9072697162628174, + "learning_rate": 3.524604741789296e-06, + "loss": 0.6238, + "step": 2711 + }, + { + "epoch": 2.2451684152401987, + "grad_norm": 0.9228153228759766, + "learning_rate": 3.5236015537855765e-06, + "loss": 0.6011, + "step": 2712 + }, + { + "epoch": 2.2459966869133074, + "grad_norm": 0.9280546307563782, + "learning_rate": 3.522598167724498e-06, + "loss": 0.6018, + "step": 2713 + }, + { + "epoch": 2.246824958586416, + "grad_norm": 0.9077439904212952, + "learning_rate": 3.5215945838002075e-06, + "loss": 0.6231, + "step": 2714 + }, + { + "epoch": 2.247653230259525, + "grad_norm": 0.9201393127441406, + "learning_rate": 3.520590802206887e-06, + "loss": 0.591, + "step": 2715 + }, + { + "epoch": 2.248481501932634, + "grad_norm": 0.8888259530067444, + "learning_rate": 3.519586823138761e-06, + "loss": 0.585, + "step": 2716 + }, + { + "epoch": 2.249309773605743, + "grad_norm": 0.918370246887207, + "learning_rate": 3.5185826467900886e-06, + "loss": 0.6151, + "step": 2717 + }, + { + "epoch": 2.2501380452788515, + "grad_norm": 0.9071224331855774, + "learning_rate": 3.517578273355169e-06, + "loss": 0.6053, + "step": 2718 + }, + { + "epoch": 2.2509663169519603, + "grad_norm": 0.9325213432312012, + "learning_rate": 3.5165737030283404e-06, + "loss": 0.5959, + "step": 2719 + }, + { + "epoch": 2.251794588625069, + "grad_norm": 0.9014924168586731, + "learning_rate": 3.515568936003976e-06, + "loss": 0.6092, + "step": 2720 + }, + { + "epoch": 2.2526228602981777, + "grad_norm": 0.923115611076355, + "learning_rate": 3.5145639724764886e-06, + "loss": 0.6031, + "step": 2721 + }, + { + "epoch": 2.2534511319712864, + "grad_norm": 0.9452555179595947, + "learning_rate": 3.513558812640331e-06, + "loss": 0.5957, + "step": 2722 + }, + { + "epoch": 2.2542794036443956, + "grad_norm": 0.960884153842926, + "learning_rate": 3.5125534566899915e-06, + "loss": 0.5862, + "step": 2723 + }, + { + "epoch": 2.2551076753175043, + "grad_norm": 0.9164205193519592, + "learning_rate": 3.5115479048199964e-06, + "loss": 0.5992, + "step": 2724 + }, + { + "epoch": 2.255935946990613, + "grad_norm": 0.943760097026825, + "learning_rate": 3.5105421572249123e-06, + "loss": 0.5842, + "step": 2725 + }, + { + "epoch": 2.256764218663722, + "grad_norm": 0.9578623175621033, + "learning_rate": 3.50953621409934e-06, + "loss": 0.6094, + "step": 2726 + }, + { + "epoch": 2.2575924903368305, + "grad_norm": 0.9307887554168701, + "learning_rate": 3.5085300756379216e-06, + "loss": 0.5905, + "step": 2727 + }, + { + "epoch": 2.2584207620099392, + "grad_norm": 0.9150220155715942, + "learning_rate": 3.5075237420353346e-06, + "loss": 0.6035, + "step": 2728 + }, + { + "epoch": 2.259249033683048, + "grad_norm": 0.8887060880661011, + "learning_rate": 3.506517213486296e-06, + "loss": 0.6032, + "step": 2729 + }, + { + "epoch": 2.2600773053561567, + "grad_norm": 0.9477519989013672, + "learning_rate": 3.5055104901855585e-06, + "loss": 0.576, + "step": 2730 + }, + { + "epoch": 2.2609055770292654, + "grad_norm": 0.9668344855308533, + "learning_rate": 3.504503572327915e-06, + "loss": 0.5973, + "step": 2731 + }, + { + "epoch": 2.2617338487023746, + "grad_norm": 0.905135989189148, + "learning_rate": 3.503496460108194e-06, + "loss": 0.5999, + "step": 2732 + }, + { + "epoch": 2.2625621203754833, + "grad_norm": 0.933996856212616, + "learning_rate": 3.5024891537212624e-06, + "loss": 0.5803, + "step": 2733 + }, + { + "epoch": 2.263390392048592, + "grad_norm": 0.9417551159858704, + "learning_rate": 3.501481653362025e-06, + "loss": 0.5918, + "step": 2734 + }, + { + "epoch": 2.2642186637217008, + "grad_norm": 0.9239657521247864, + "learning_rate": 3.500473959225422e-06, + "loss": 0.5828, + "step": 2735 + }, + { + "epoch": 2.2650469353948095, + "grad_norm": 0.9021532535552979, + "learning_rate": 3.499466071506436e-06, + "loss": 0.5958, + "step": 2736 + }, + { + "epoch": 2.2658752070679182, + "grad_norm": 0.9159536957740784, + "learning_rate": 3.498457990400081e-06, + "loss": 0.5867, + "step": 2737 + }, + { + "epoch": 2.266703478741027, + "grad_norm": 0.9347800612449646, + "learning_rate": 3.4974497161014124e-06, + "loss": 0.6058, + "step": 2738 + }, + { + "epoch": 2.2675317504141357, + "grad_norm": 0.9229647517204285, + "learning_rate": 3.496441248805522e-06, + "loss": 0.5714, + "step": 2739 + }, + { + "epoch": 2.2683600220872444, + "grad_norm": 0.9330629706382751, + "learning_rate": 3.4954325887075386e-06, + "loss": 0.5808, + "step": 2740 + }, + { + "epoch": 2.2691882937603536, + "grad_norm": 0.8958399891853333, + "learning_rate": 3.494423736002629e-06, + "loss": 0.5793, + "step": 2741 + }, + { + "epoch": 2.2700165654334623, + "grad_norm": 0.8952097296714783, + "learning_rate": 3.493414690885997e-06, + "loss": 0.6007, + "step": 2742 + }, + { + "epoch": 2.270844837106571, + "grad_norm": 0.9001036882400513, + "learning_rate": 3.4924054535528813e-06, + "loss": 0.5821, + "step": 2743 + }, + { + "epoch": 2.2716731087796798, + "grad_norm": 0.9016004204750061, + "learning_rate": 3.4913960241985616e-06, + "loss": 0.5819, + "step": 2744 + }, + { + "epoch": 2.2725013804527885, + "grad_norm": 0.9043833017349243, + "learning_rate": 3.490386403018354e-06, + "loss": 0.6144, + "step": 2745 + }, + { + "epoch": 2.273329652125897, + "grad_norm": 0.9396303296089172, + "learning_rate": 3.489376590207608e-06, + "loss": 0.602, + "step": 2746 + }, + { + "epoch": 2.274157923799006, + "grad_norm": 0.9685373306274414, + "learning_rate": 3.488366585961716e-06, + "loss": 0.6247, + "step": 2747 + }, + { + "epoch": 2.2749861954721147, + "grad_norm": 0.9028962254524231, + "learning_rate": 3.487356390476103e-06, + "loss": 0.5959, + "step": 2748 + }, + { + "epoch": 2.2758144671452234, + "grad_norm": 0.9150850772857666, + "learning_rate": 3.486346003946232e-06, + "loss": 0.5632, + "step": 2749 + }, + { + "epoch": 2.2766427388183326, + "grad_norm": 0.902158796787262, + "learning_rate": 3.4853354265676043e-06, + "loss": 0.5901, + "step": 2750 + }, + { + "epoch": 2.2774710104914413, + "grad_norm": 0.8942626118659973, + "learning_rate": 3.4843246585357565e-06, + "loss": 0.6082, + "step": 2751 + }, + { + "epoch": 2.27829928216455, + "grad_norm": 0.912050187587738, + "learning_rate": 3.4833137000462623e-06, + "loss": 0.5812, + "step": 2752 + }, + { + "epoch": 2.2791275538376587, + "grad_norm": 0.9262410998344421, + "learning_rate": 3.482302551294735e-06, + "loss": 0.6199, + "step": 2753 + }, + { + "epoch": 2.2799558255107675, + "grad_norm": 0.9281290173530579, + "learning_rate": 3.4812912124768193e-06, + "loss": 0.591, + "step": 2754 + }, + { + "epoch": 2.280784097183876, + "grad_norm": 0.8852386474609375, + "learning_rate": 3.4802796837882023e-06, + "loss": 0.6035, + "step": 2755 + }, + { + "epoch": 2.281612368856985, + "grad_norm": 0.8982499241828918, + "learning_rate": 3.479267965424605e-06, + "loss": 0.6023, + "step": 2756 + }, + { + "epoch": 2.282440640530094, + "grad_norm": 0.9022878408432007, + "learning_rate": 3.4782560575817837e-06, + "loss": 0.6004, + "step": 2757 + }, + { + "epoch": 2.283268912203203, + "grad_norm": 0.9256078600883484, + "learning_rate": 3.4772439604555353e-06, + "loss": 0.5811, + "step": 2758 + }, + { + "epoch": 2.2840971838763116, + "grad_norm": 0.9009407758712769, + "learning_rate": 3.4762316742416913e-06, + "loss": 0.6053, + "step": 2759 + }, + { + "epoch": 2.2849254555494203, + "grad_norm": 0.9257068037986755, + "learning_rate": 3.4752191991361174e-06, + "loss": 0.5664, + "step": 2760 + }, + { + "epoch": 2.285753727222529, + "grad_norm": 0.882075309753418, + "learning_rate": 3.4742065353347193e-06, + "loss": 0.6099, + "step": 2761 + }, + { + "epoch": 2.2865819988956377, + "grad_norm": 0.9553154706954956, + "learning_rate": 3.4731936830334384e-06, + "loss": 0.6072, + "step": 2762 + }, + { + "epoch": 2.2874102705687465, + "grad_norm": 0.9329802989959717, + "learning_rate": 3.4721806424282512e-06, + "loss": 0.601, + "step": 2763 + }, + { + "epoch": 2.288238542241855, + "grad_norm": 0.9146580696105957, + "learning_rate": 3.471167413715173e-06, + "loss": 0.599, + "step": 2764 + }, + { + "epoch": 2.289066813914964, + "grad_norm": 0.9170219302177429, + "learning_rate": 3.470153997090253e-06, + "loss": 0.5969, + "step": 2765 + }, + { + "epoch": 2.289895085588073, + "grad_norm": 0.8903328776359558, + "learning_rate": 3.469140392749578e-06, + "loss": 0.6063, + "step": 2766 + }, + { + "epoch": 2.290723357261182, + "grad_norm": 0.9068611264228821, + "learning_rate": 3.468126600889272e-06, + "loss": 0.5807, + "step": 2767 + }, + { + "epoch": 2.2915516289342905, + "grad_norm": 0.9464553594589233, + "learning_rate": 3.467112621705493e-06, + "loss": 0.5955, + "step": 2768 + }, + { + "epoch": 2.2923799006073993, + "grad_norm": 0.9302126169204712, + "learning_rate": 3.4660984553944367e-06, + "loss": 0.5956, + "step": 2769 + }, + { + "epoch": 2.293208172280508, + "grad_norm": 0.8797576427459717, + "learning_rate": 3.465084102152336e-06, + "loss": 0.5964, + "step": 2770 + }, + { + "epoch": 2.2940364439536167, + "grad_norm": 0.9544422030448914, + "learning_rate": 3.4640695621754573e-06, + "loss": 0.6225, + "step": 2771 + }, + { + "epoch": 2.2948647156267254, + "grad_norm": 0.919297993183136, + "learning_rate": 3.463054835660105e-06, + "loss": 0.5901, + "step": 2772 + }, + { + "epoch": 2.295692987299834, + "grad_norm": 0.9211829900741577, + "learning_rate": 3.4620399228026203e-06, + "loss": 0.5949, + "step": 2773 + }, + { + "epoch": 2.296521258972943, + "grad_norm": 0.929501473903656, + "learning_rate": 3.4610248237993775e-06, + "loss": 0.5654, + "step": 2774 + }, + { + "epoch": 2.297349530646052, + "grad_norm": 0.9604958891868591, + "learning_rate": 3.4600095388467907e-06, + "loss": 0.6059, + "step": 2775 + }, + { + "epoch": 2.298177802319161, + "grad_norm": 0.9042990207672119, + "learning_rate": 3.458994068141307e-06, + "loss": 0.5901, + "step": 2776 + }, + { + "epoch": 2.2990060739922695, + "grad_norm": 0.8958423733711243, + "learning_rate": 3.4579784118794107e-06, + "loss": 0.5938, + "step": 2777 + }, + { + "epoch": 2.2998343456653783, + "grad_norm": 0.9388688206672668, + "learning_rate": 3.456962570257622e-06, + "loss": 0.6024, + "step": 2778 + }, + { + "epoch": 2.300662617338487, + "grad_norm": 0.9006328582763672, + "learning_rate": 3.455946543472497e-06, + "loss": 0.5929, + "step": 2779 + }, + { + "epoch": 2.3014908890115957, + "grad_norm": 0.887873113155365, + "learning_rate": 3.454930331720626e-06, + "loss": 0.599, + "step": 2780 + }, + { + "epoch": 2.3023191606847044, + "grad_norm": 0.8919411301612854, + "learning_rate": 3.4539139351986385e-06, + "loss": 0.6173, + "step": 2781 + }, + { + "epoch": 2.3031474323578136, + "grad_norm": 0.9478135704994202, + "learning_rate": 3.452897354103196e-06, + "loss": 0.6025, + "step": 2782 + }, + { + "epoch": 2.3039757040309223, + "grad_norm": 0.9009284377098083, + "learning_rate": 3.451880588630999e-06, + "loss": 0.602, + "step": 2783 + }, + { + "epoch": 2.304803975704031, + "grad_norm": 0.9043265581130981, + "learning_rate": 3.4508636389787816e-06, + "loss": 0.6002, + "step": 2784 + }, + { + "epoch": 2.30563224737714, + "grad_norm": 0.9005764126777649, + "learning_rate": 3.4498465053433126e-06, + "loss": 0.618, + "step": 2785 + }, + { + "epoch": 2.3064605190502485, + "grad_norm": 0.9352818727493286, + "learning_rate": 3.4488291879214003e-06, + "loss": 0.6069, + "step": 2786 + }, + { + "epoch": 2.3072887907233572, + "grad_norm": 0.9092200994491577, + "learning_rate": 3.4478116869098854e-06, + "loss": 0.581, + "step": 2787 + }, + { + "epoch": 2.308117062396466, + "grad_norm": 0.9055072665214539, + "learning_rate": 3.4467940025056433e-06, + "loss": 0.598, + "step": 2788 + }, + { + "epoch": 2.3089453340695747, + "grad_norm": 1.0053009986877441, + "learning_rate": 3.4457761349055885e-06, + "loss": 0.5929, + "step": 2789 + }, + { + "epoch": 2.3097736057426834, + "grad_norm": 0.9241059422492981, + "learning_rate": 3.4447580843066675e-06, + "loss": 0.5962, + "step": 2790 + }, + { + "epoch": 2.3106018774157926, + "grad_norm": 0.927590548992157, + "learning_rate": 3.443739850905864e-06, + "loss": 0.5863, + "step": 2791 + }, + { + "epoch": 2.3114301490889013, + "grad_norm": 0.9227457046508789, + "learning_rate": 3.442721434900197e-06, + "loss": 0.5928, + "step": 2792 + }, + { + "epoch": 2.31225842076201, + "grad_norm": 0.9112377762794495, + "learning_rate": 3.4417028364867205e-06, + "loss": 0.6113, + "step": 2793 + }, + { + "epoch": 2.3130866924351188, + "grad_norm": 0.9256125092506409, + "learning_rate": 3.4406840558625237e-06, + "loss": 0.6163, + "step": 2794 + }, + { + "epoch": 2.3139149641082275, + "grad_norm": 0.933461606502533, + "learning_rate": 3.4396650932247303e-06, + "loss": 0.6159, + "step": 2795 + }, + { + "epoch": 2.3147432357813362, + "grad_norm": 0.9031276702880859, + "learning_rate": 3.4386459487705016e-06, + "loss": 0.5981, + "step": 2796 + }, + { + "epoch": 2.315571507454445, + "grad_norm": 0.9465809464454651, + "learning_rate": 3.437626622697031e-06, + "loss": 0.6029, + "step": 2797 + }, + { + "epoch": 2.3163997791275537, + "grad_norm": 0.9535313248634338, + "learning_rate": 3.43660711520155e-06, + "loss": 0.5998, + "step": 2798 + }, + { + "epoch": 2.3172280508006624, + "grad_norm": 0.9048331379890442, + "learning_rate": 3.4355874264813228e-06, + "loss": 0.5991, + "step": 2799 + }, + { + "epoch": 2.3180563224737716, + "grad_norm": 0.9119572639465332, + "learning_rate": 3.4345675567336506e-06, + "loss": 0.6133, + "step": 2800 + }, + { + "epoch": 2.3188845941468803, + "grad_norm": 0.9217002391815186, + "learning_rate": 3.4335475061558676e-06, + "loss": 0.6012, + "step": 2801 + }, + { + "epoch": 2.319712865819989, + "grad_norm": 0.9501274824142456, + "learning_rate": 3.432527274945344e-06, + "loss": 0.5874, + "step": 2802 + }, + { + "epoch": 2.3205411374930978, + "grad_norm": 0.9126407504081726, + "learning_rate": 3.4315068632994864e-06, + "loss": 0.5985, + "step": 2803 + }, + { + "epoch": 2.3213694091662065, + "grad_norm": 0.9414592981338501, + "learning_rate": 3.430486271415734e-06, + "loss": 0.5989, + "step": 2804 + }, + { + "epoch": 2.322197680839315, + "grad_norm": 0.9485170841217041, + "learning_rate": 3.4294654994915623e-06, + "loss": 0.5795, + "step": 2805 + }, + { + "epoch": 2.323025952512424, + "grad_norm": 0.9311578273773193, + "learning_rate": 3.4284445477244804e-06, + "loss": 0.5606, + "step": 2806 + }, + { + "epoch": 2.3238542241855327, + "grad_norm": 0.9476608037948608, + "learning_rate": 3.4274234163120328e-06, + "loss": 0.593, + "step": 2807 + }, + { + "epoch": 2.3246824958586414, + "grad_norm": 0.9367009401321411, + "learning_rate": 3.4264021054518004e-06, + "loss": 0.6104, + "step": 2808 + }, + { + "epoch": 2.3255107675317506, + "grad_norm": 0.9663116335868835, + "learning_rate": 3.425380615341396e-06, + "loss": 0.6091, + "step": 2809 + }, + { + "epoch": 2.3263390392048593, + "grad_norm": 0.9176538586616516, + "learning_rate": 3.4243589461784687e-06, + "loss": 0.585, + "step": 2810 + }, + { + "epoch": 2.327167310877968, + "grad_norm": 0.9261791110038757, + "learning_rate": 3.423337098160703e-06, + "loss": 0.6189, + "step": 2811 + }, + { + "epoch": 2.3279955825510767, + "grad_norm": 0.9573442339897156, + "learning_rate": 3.422315071485815e-06, + "loss": 0.5925, + "step": 2812 + }, + { + "epoch": 2.3288238542241855, + "grad_norm": 0.9381136894226074, + "learning_rate": 3.4212928663515584e-06, + "loss": 0.588, + "step": 2813 + }, + { + "epoch": 2.329652125897294, + "grad_norm": 0.9276189804077148, + "learning_rate": 3.420270482955721e-06, + "loss": 0.5885, + "step": 2814 + }, + { + "epoch": 2.330480397570403, + "grad_norm": 0.9124462604522705, + "learning_rate": 3.419247921496123e-06, + "loss": 0.6004, + "step": 2815 + }, + { + "epoch": 2.331308669243512, + "grad_norm": 0.9146862626075745, + "learning_rate": 3.418225182170622e-06, + "loss": 0.6036, + "step": 2816 + }, + { + "epoch": 2.332136940916621, + "grad_norm": 0.9196732044219971, + "learning_rate": 3.4172022651771075e-06, + "loss": 0.6054, + "step": 2817 + }, + { + "epoch": 2.3329652125897296, + "grad_norm": 0.9340062737464905, + "learning_rate": 3.416179170713505e-06, + "loss": 0.6037, + "step": 2818 + }, + { + "epoch": 2.3337934842628383, + "grad_norm": 0.9004904627799988, + "learning_rate": 3.415155898977773e-06, + "loss": 0.6087, + "step": 2819 + }, + { + "epoch": 2.334621755935947, + "grad_norm": 0.8993189334869385, + "learning_rate": 3.414132450167906e-06, + "loss": 0.5972, + "step": 2820 + }, + { + "epoch": 2.3354500276090557, + "grad_norm": 0.9195947051048279, + "learning_rate": 3.413108824481931e-06, + "loss": 0.5784, + "step": 2821 + }, + { + "epoch": 2.3362782992821645, + "grad_norm": 0.9149767756462097, + "learning_rate": 3.412085022117911e-06, + "loss": 0.5838, + "step": 2822 + }, + { + "epoch": 2.337106570955273, + "grad_norm": 0.8880875110626221, + "learning_rate": 3.411061043273941e-06, + "loss": 0.6008, + "step": 2823 + }, + { + "epoch": 2.337934842628382, + "grad_norm": 0.9110493659973145, + "learning_rate": 3.4100368881481515e-06, + "loss": 0.6333, + "step": 2824 + }, + { + "epoch": 2.338763114301491, + "grad_norm": 0.9558914303779602, + "learning_rate": 3.409012556938708e-06, + "loss": 0.602, + "step": 2825 + }, + { + "epoch": 2.3395913859746, + "grad_norm": 0.9668992161750793, + "learning_rate": 3.4079880498438083e-06, + "loss": 0.5952, + "step": 2826 + }, + { + "epoch": 2.3404196576477085, + "grad_norm": 0.9352533221244812, + "learning_rate": 3.406963367061685e-06, + "loss": 0.6093, + "step": 2827 + }, + { + "epoch": 2.3412479293208173, + "grad_norm": 0.9072561264038086, + "learning_rate": 3.4059385087906055e-06, + "loss": 0.603, + "step": 2828 + }, + { + "epoch": 2.342076200993926, + "grad_norm": 0.9563443064689636, + "learning_rate": 3.4049134752288692e-06, + "loss": 0.585, + "step": 2829 + }, + { + "epoch": 2.3429044726670347, + "grad_norm": 0.9061548709869385, + "learning_rate": 3.4038882665748102e-06, + "loss": 0.6111, + "step": 2830 + }, + { + "epoch": 2.3437327443401434, + "grad_norm": 0.9177330136299133, + "learning_rate": 3.4028628830267986e-06, + "loss": 0.5872, + "step": 2831 + }, + { + "epoch": 2.344561016013252, + "grad_norm": 0.9479212164878845, + "learning_rate": 3.401837324783235e-06, + "loss": 0.5911, + "step": 2832 + }, + { + "epoch": 2.345389287686361, + "grad_norm": 0.9118002653121948, + "learning_rate": 3.4008115920425556e-06, + "loss": 0.6153, + "step": 2833 + }, + { + "epoch": 2.34621755935947, + "grad_norm": 0.9086247682571411, + "learning_rate": 3.399785685003231e-06, + "loss": 0.5913, + "step": 2834 + }, + { + "epoch": 2.347045831032579, + "grad_norm": 0.937494158744812, + "learning_rate": 3.398759603863763e-06, + "loss": 0.601, + "step": 2835 + }, + { + "epoch": 2.3478741027056875, + "grad_norm": 0.9359503388404846, + "learning_rate": 3.397733348822691e-06, + "loss": 0.5894, + "step": 2836 + }, + { + "epoch": 2.3487023743787963, + "grad_norm": 0.9384423494338989, + "learning_rate": 3.396706920078584e-06, + "loss": 0.5974, + "step": 2837 + }, + { + "epoch": 2.349530646051905, + "grad_norm": 0.9438448548316956, + "learning_rate": 3.3956803178300474e-06, + "loss": 0.6079, + "step": 2838 + }, + { + "epoch": 2.3503589177250137, + "grad_norm": 0.9250973463058472, + "learning_rate": 3.3946535422757187e-06, + "loss": 0.6011, + "step": 2839 + }, + { + "epoch": 2.3511871893981224, + "grad_norm": 0.9577994346618652, + "learning_rate": 3.3936265936142685e-06, + "loss": 0.6052, + "step": 2840 + }, + { + "epoch": 2.3520154610712316, + "grad_norm": 0.9307487607002258, + "learning_rate": 3.3925994720444033e-06, + "loss": 0.5925, + "step": 2841 + }, + { + "epoch": 2.3528437327443403, + "grad_norm": 0.9049618244171143, + "learning_rate": 3.3915721777648615e-06, + "loss": 0.6096, + "step": 2842 + }, + { + "epoch": 2.353672004417449, + "grad_norm": 0.928248405456543, + "learning_rate": 3.3905447109744137e-06, + "loss": 0.5938, + "step": 2843 + }, + { + "epoch": 2.354500276090558, + "grad_norm": 0.9107298254966736, + "learning_rate": 3.3895170718718664e-06, + "loss": 0.5947, + "step": 2844 + }, + { + "epoch": 2.3553285477636665, + "grad_norm": 0.8896966576576233, + "learning_rate": 3.388489260656057e-06, + "loss": 0.5827, + "step": 2845 + }, + { + "epoch": 2.3561568194367752, + "grad_norm": 0.9374940991401672, + "learning_rate": 3.3874612775258586e-06, + "loss": 0.6038, + "step": 2846 + }, + { + "epoch": 2.356985091109884, + "grad_norm": 0.9086610078811646, + "learning_rate": 3.3864331226801755e-06, + "loss": 0.5758, + "step": 2847 + }, + { + "epoch": 2.3578133627829927, + "grad_norm": 0.9486793875694275, + "learning_rate": 3.3854047963179464e-06, + "loss": 0.5656, + "step": 2848 + }, + { + "epoch": 2.3586416344561014, + "grad_norm": 0.9411890506744385, + "learning_rate": 3.3843762986381436e-06, + "loss": 0.605, + "step": 2849 + }, + { + "epoch": 2.3594699061292106, + "grad_norm": 0.9491676688194275, + "learning_rate": 3.3833476298397706e-06, + "loss": 0.5992, + "step": 2850 + }, + { + "epoch": 2.3602981778023193, + "grad_norm": 0.9281409382820129, + "learning_rate": 3.382318790121865e-06, + "loss": 0.5856, + "step": 2851 + }, + { + "epoch": 2.361126449475428, + "grad_norm": 0.9243955016136169, + "learning_rate": 3.381289779683499e-06, + "loss": 0.5966, + "step": 2852 + }, + { + "epoch": 2.3619547211485368, + "grad_norm": 0.8931578993797302, + "learning_rate": 3.3802605987237757e-06, + "loss": 0.6095, + "step": 2853 + }, + { + "epoch": 2.3627829928216455, + "grad_norm": 0.9200635552406311, + "learning_rate": 3.3792312474418324e-06, + "loss": 0.6014, + "step": 2854 + }, + { + "epoch": 2.3636112644947542, + "grad_norm": 0.9007267951965332, + "learning_rate": 3.378201726036839e-06, + "loss": 0.5964, + "step": 2855 + }, + { + "epoch": 2.364439536167863, + "grad_norm": 0.9420115947723389, + "learning_rate": 3.377172034707998e-06, + "loss": 0.6075, + "step": 2856 + }, + { + "epoch": 2.3652678078409717, + "grad_norm": 0.8935693502426147, + "learning_rate": 3.3761421736545453e-06, + "loss": 0.6184, + "step": 2857 + }, + { + "epoch": 2.3660960795140804, + "grad_norm": 0.9142755270004272, + "learning_rate": 3.3751121430757482e-06, + "loss": 0.6114, + "step": 2858 + }, + { + "epoch": 2.3669243511871896, + "grad_norm": 0.9280282855033875, + "learning_rate": 3.3740819431709103e-06, + "loss": 0.5922, + "step": 2859 + }, + { + "epoch": 2.3677526228602983, + "grad_norm": 0.9347119927406311, + "learning_rate": 3.3730515741393643e-06, + "loss": 0.5733, + "step": 2860 + }, + { + "epoch": 2.368580894533407, + "grad_norm": 0.9286894202232361, + "learning_rate": 3.3720210361804775e-06, + "loss": 0.597, + "step": 2861 + }, + { + "epoch": 2.3694091662065158, + "grad_norm": 0.9092636704444885, + "learning_rate": 3.3709903294936482e-06, + "loss": 0.5842, + "step": 2862 + }, + { + "epoch": 2.3702374378796245, + "grad_norm": 0.9205367565155029, + "learning_rate": 3.369959454278309e-06, + "loss": 0.6166, + "step": 2863 + }, + { + "epoch": 2.371065709552733, + "grad_norm": 0.9211122393608093, + "learning_rate": 3.3689284107339265e-06, + "loss": 0.5746, + "step": 2864 + }, + { + "epoch": 2.371893981225842, + "grad_norm": 0.908984899520874, + "learning_rate": 3.367897199059996e-06, + "loss": 0.5883, + "step": 2865 + }, + { + "epoch": 2.3727222528989507, + "grad_norm": 0.9270497560501099, + "learning_rate": 3.3668658194560477e-06, + "loss": 0.5985, + "step": 2866 + }, + { + "epoch": 2.3735505245720594, + "grad_norm": 0.9320128560066223, + "learning_rate": 3.3658342721216448e-06, + "loss": 0.5907, + "step": 2867 + }, + { + "epoch": 2.3743787962451686, + "grad_norm": 0.9428799152374268, + "learning_rate": 3.364802557256381e-06, + "loss": 0.6041, + "step": 2868 + }, + { + "epoch": 2.3752070679182773, + "grad_norm": 0.915873646736145, + "learning_rate": 3.363770675059883e-06, + "loss": 0.615, + "step": 2869 + }, + { + "epoch": 2.376035339591386, + "grad_norm": 0.9327977895736694, + "learning_rate": 3.362738625731813e-06, + "loss": 0.6147, + "step": 2870 + }, + { + "epoch": 2.3768636112644947, + "grad_norm": 0.9171217679977417, + "learning_rate": 3.3617064094718602e-06, + "loss": 0.6057, + "step": 2871 + }, + { + "epoch": 2.3776918829376035, + "grad_norm": 0.9192221760749817, + "learning_rate": 3.3606740264797504e-06, + "loss": 0.6092, + "step": 2872 + }, + { + "epoch": 2.378520154610712, + "grad_norm": 0.9024589657783508, + "learning_rate": 3.3596414769552394e-06, + "loss": 0.6047, + "step": 2873 + }, + { + "epoch": 2.379348426283821, + "grad_norm": 0.910923182964325, + "learning_rate": 3.3586087610981155e-06, + "loss": 0.5741, + "step": 2874 + }, + { + "epoch": 2.38017669795693, + "grad_norm": 0.929965078830719, + "learning_rate": 3.3575758791082014e-06, + "loss": 0.6114, + "step": 2875 + }, + { + "epoch": 2.381004969630039, + "grad_norm": 0.910308837890625, + "learning_rate": 3.356542831185348e-06, + "loss": 0.5938, + "step": 2876 + }, + { + "epoch": 2.3818332413031476, + "grad_norm": 0.9371251463890076, + "learning_rate": 3.355509617529441e-06, + "loss": 0.5883, + "step": 2877 + }, + { + "epoch": 2.3826615129762563, + "grad_norm": 0.9064118266105652, + "learning_rate": 3.3544762383403995e-06, + "loss": 0.5879, + "step": 2878 + }, + { + "epoch": 2.383489784649365, + "grad_norm": 0.954288899898529, + "learning_rate": 3.3534426938181704e-06, + "loss": 0.5981, + "step": 2879 + }, + { + "epoch": 2.3843180563224737, + "grad_norm": 0.9465821981430054, + "learning_rate": 3.3524089841627348e-06, + "loss": 0.6129, + "step": 2880 + }, + { + "epoch": 2.3851463279955825, + "grad_norm": 0.9406200051307678, + "learning_rate": 3.3513751095741084e-06, + "loss": 0.6035, + "step": 2881 + }, + { + "epoch": 2.385974599668691, + "grad_norm": 0.9627492427825928, + "learning_rate": 3.3503410702523337e-06, + "loss": 0.5701, + "step": 2882 + }, + { + "epoch": 2.3868028713418, + "grad_norm": 0.8899156451225281, + "learning_rate": 3.3493068663974892e-06, + "loss": 0.6055, + "step": 2883 + }, + { + "epoch": 2.387631143014909, + "grad_norm": 0.9304690957069397, + "learning_rate": 3.348272498209683e-06, + "loss": 0.6078, + "step": 2884 + }, + { + "epoch": 2.388459414688018, + "grad_norm": 0.9416152834892273, + "learning_rate": 3.3472379658890554e-06, + "loss": 0.609, + "step": 2885 + }, + { + "epoch": 2.3892876863611265, + "grad_norm": 0.9296873211860657, + "learning_rate": 3.34620326963578e-06, + "loss": 0.608, + "step": 2886 + }, + { + "epoch": 2.3901159580342353, + "grad_norm": 0.9163168668746948, + "learning_rate": 3.34516840965006e-06, + "loss": 0.5725, + "step": 2887 + }, + { + "epoch": 2.390944229707344, + "grad_norm": 0.9371008276939392, + "learning_rate": 3.344133386132131e-06, + "loss": 0.6223, + "step": 2888 + }, + { + "epoch": 2.3917725013804527, + "grad_norm": 0.9486922025680542, + "learning_rate": 3.3430981992822608e-06, + "loss": 0.5878, + "step": 2889 + }, + { + "epoch": 2.3926007730535614, + "grad_norm": 0.9124024510383606, + "learning_rate": 3.3420628493007483e-06, + "loss": 0.5816, + "step": 2890 + }, + { + "epoch": 2.39342904472667, + "grad_norm": 0.9193150401115417, + "learning_rate": 3.3410273363879236e-06, + "loss": 0.5909, + "step": 2891 + }, + { + "epoch": 2.394257316399779, + "grad_norm": 0.9287682175636292, + "learning_rate": 3.3399916607441502e-06, + "loss": 0.6095, + "step": 2892 + }, + { + "epoch": 2.395085588072888, + "grad_norm": 0.9130192995071411, + "learning_rate": 3.3389558225698204e-06, + "loss": 0.5782, + "step": 2893 + }, + { + "epoch": 2.395913859745997, + "grad_norm": 0.9145562052726746, + "learning_rate": 3.3379198220653587e-06, + "loss": 0.5919, + "step": 2894 + }, + { + "epoch": 2.3967421314191055, + "grad_norm": 0.9094864726066589, + "learning_rate": 3.3368836594312235e-06, + "loss": 0.6165, + "step": 2895 + }, + { + "epoch": 2.3975704030922143, + "grad_norm": 0.9446457028388977, + "learning_rate": 3.3358473348679016e-06, + "loss": 0.5967, + "step": 2896 + }, + { + "epoch": 2.398398674765323, + "grad_norm": 0.9345584511756897, + "learning_rate": 3.334810848575911e-06, + "loss": 0.6006, + "step": 2897 + }, + { + "epoch": 2.3992269464384317, + "grad_norm": 0.8865429759025574, + "learning_rate": 3.3337742007558043e-06, + "loss": 0.5766, + "step": 2898 + }, + { + "epoch": 2.4000552181115404, + "grad_norm": 0.9188973903656006, + "learning_rate": 3.3327373916081617e-06, + "loss": 0.6107, + "step": 2899 + }, + { + "epoch": 2.4008834897846496, + "grad_norm": 0.895247220993042, + "learning_rate": 3.3317004213335967e-06, + "loss": 0.6037, + "step": 2900 + }, + { + "epoch": 2.4017117614577583, + "grad_norm": 0.9131169319152832, + "learning_rate": 3.330663290132753e-06, + "loss": 0.5825, + "step": 2901 + }, + { + "epoch": 2.402540033130867, + "grad_norm": 0.9109821915626526, + "learning_rate": 3.329625998206306e-06, + "loss": 0.6194, + "step": 2902 + }, + { + "epoch": 2.403368304803976, + "grad_norm": 0.90902179479599, + "learning_rate": 3.3285885457549618e-06, + "loss": 0.5921, + "step": 2903 + }, + { + "epoch": 2.4041965764770845, + "grad_norm": 0.9419400095939636, + "learning_rate": 3.3275509329794585e-06, + "loss": 0.5973, + "step": 2904 + }, + { + "epoch": 2.4050248481501932, + "grad_norm": 0.905164897441864, + "learning_rate": 3.3265131600805636e-06, + "loss": 0.5828, + "step": 2905 + }, + { + "epoch": 2.405853119823302, + "grad_norm": 0.9232257008552551, + "learning_rate": 3.3254752272590775e-06, + "loss": 0.5972, + "step": 2906 + }, + { + "epoch": 2.4066813914964107, + "grad_norm": 0.9152960181236267, + "learning_rate": 3.3244371347158294e-06, + "loss": 0.5968, + "step": 2907 + }, + { + "epoch": 2.4075096631695194, + "grad_norm": 0.9337223172187805, + "learning_rate": 3.3233988826516807e-06, + "loss": 0.5907, + "step": 2908 + }, + { + "epoch": 2.4083379348426286, + "grad_norm": 0.9352282881736755, + "learning_rate": 3.322360471267524e-06, + "loss": 0.573, + "step": 2909 + }, + { + "epoch": 2.4091662065157373, + "grad_norm": 0.928322970867157, + "learning_rate": 3.321321900764282e-06, + "loss": 0.5836, + "step": 2910 + }, + { + "epoch": 2.409994478188846, + "grad_norm": 0.9335048794746399, + "learning_rate": 3.3202831713429084e-06, + "loss": 0.607, + "step": 2911 + }, + { + "epoch": 2.4108227498619548, + "grad_norm": 0.9128180742263794, + "learning_rate": 3.319244283204389e-06, + "loss": 0.5956, + "step": 2912 + }, + { + "epoch": 2.4116510215350635, + "grad_norm": 0.9457159042358398, + "learning_rate": 3.3182052365497364e-06, + "loss": 0.602, + "step": 2913 + }, + { + "epoch": 2.4124792932081722, + "grad_norm": 0.9508596658706665, + "learning_rate": 3.317166031579998e-06, + "loss": 0.5972, + "step": 2914 + }, + { + "epoch": 2.413307564881281, + "grad_norm": 0.9274892807006836, + "learning_rate": 3.31612666849625e-06, + "loss": 0.6015, + "step": 2915 + }, + { + "epoch": 2.4141358365543897, + "grad_norm": 0.9340214729309082, + "learning_rate": 3.3150871474995994e-06, + "loss": 0.5984, + "step": 2916 + }, + { + "epoch": 2.4149641082274984, + "grad_norm": 0.9376348853111267, + "learning_rate": 3.3140474687911844e-06, + "loss": 0.6205, + "step": 2917 + }, + { + "epoch": 2.4157923799006076, + "grad_norm": 0.9363590478897095, + "learning_rate": 3.313007632572173e-06, + "loss": 0.5926, + "step": 2918 + }, + { + "epoch": 2.4166206515737163, + "grad_norm": 0.8942010402679443, + "learning_rate": 3.3119676390437618e-06, + "loss": 0.5966, + "step": 2919 + }, + { + "epoch": 2.417448923246825, + "grad_norm": 0.9249299168586731, + "learning_rate": 3.310927488407183e-06, + "loss": 0.5951, + "step": 2920 + }, + { + "epoch": 2.4182771949199338, + "grad_norm": 0.9157326817512512, + "learning_rate": 3.3098871808636936e-06, + "loss": 0.5693, + "step": 2921 + }, + { + "epoch": 2.4191054665930425, + "grad_norm": 0.9088039994239807, + "learning_rate": 3.3088467166145843e-06, + "loss": 0.6084, + "step": 2922 + }, + { + "epoch": 2.419933738266151, + "grad_norm": 0.9258041381835938, + "learning_rate": 3.307806095861177e-06, + "loss": 0.6036, + "step": 2923 + }, + { + "epoch": 2.42076200993926, + "grad_norm": 0.905963659286499, + "learning_rate": 3.306765318804819e-06, + "loss": 0.5918, + "step": 2924 + }, + { + "epoch": 2.4215902816123687, + "grad_norm": 0.9063541293144226, + "learning_rate": 3.3057243856468924e-06, + "loss": 0.5906, + "step": 2925 + }, + { + "epoch": 2.4224185532854774, + "grad_norm": 0.9308085441589355, + "learning_rate": 3.304683296588808e-06, + "loss": 0.59, + "step": 2926 + }, + { + "epoch": 2.4232468249585866, + "grad_norm": 0.8989183902740479, + "learning_rate": 3.303642051832007e-06, + "loss": 0.5915, + "step": 2927 + }, + { + "epoch": 2.4240750966316953, + "grad_norm": 0.9413655996322632, + "learning_rate": 3.3026006515779614e-06, + "loss": 0.5952, + "step": 2928 + }, + { + "epoch": 2.424903368304804, + "grad_norm": 0.9431564211845398, + "learning_rate": 3.3015590960281708e-06, + "loss": 0.6007, + "step": 2929 + }, + { + "epoch": 2.4257316399779127, + "grad_norm": 0.8911551237106323, + "learning_rate": 3.3005173853841666e-06, + "loss": 0.573, + "step": 2930 + }, + { + "epoch": 2.4265599116510215, + "grad_norm": 0.9177538156509399, + "learning_rate": 3.299475519847511e-06, + "loss": 0.6116, + "step": 2931 + }, + { + "epoch": 2.42738818332413, + "grad_norm": 0.9592459797859192, + "learning_rate": 3.298433499619795e-06, + "loss": 0.6039, + "step": 2932 + }, + { + "epoch": 2.428216454997239, + "grad_norm": 0.9287276864051819, + "learning_rate": 3.2973913249026392e-06, + "loss": 0.5837, + "step": 2933 + }, + { + "epoch": 2.429044726670348, + "grad_norm": 0.9205804467201233, + "learning_rate": 3.296348995897696e-06, + "loss": 0.5904, + "step": 2934 + }, + { + "epoch": 2.429872998343457, + "grad_norm": 0.9075837731361389, + "learning_rate": 3.295306512806646e-06, + "loss": 0.5719, + "step": 2935 + }, + { + "epoch": 2.4307012700165656, + "grad_norm": 0.9203813672065735, + "learning_rate": 3.294263875831198e-06, + "loss": 0.604, + "step": 2936 + }, + { + "epoch": 2.4315295416896743, + "grad_norm": 0.8868334293365479, + "learning_rate": 3.2932210851730952e-06, + "loss": 0.6113, + "step": 2937 + }, + { + "epoch": 2.432357813362783, + "grad_norm": 0.9620448350906372, + "learning_rate": 3.2921781410341064e-06, + "loss": 0.5876, + "step": 2938 + }, + { + "epoch": 2.4331860850358917, + "grad_norm": 0.8991201519966125, + "learning_rate": 3.2911350436160317e-06, + "loss": 0.5994, + "step": 2939 + }, + { + "epoch": 2.4340143567090005, + "grad_norm": 0.9383120536804199, + "learning_rate": 3.2900917931207016e-06, + "loss": 0.6098, + "step": 2940 + }, + { + "epoch": 2.434842628382109, + "grad_norm": 0.9405770897865295, + "learning_rate": 3.289048389749974e-06, + "loss": 0.5978, + "step": 2941 + }, + { + "epoch": 2.435670900055218, + "grad_norm": 0.9153317213058472, + "learning_rate": 3.2880048337057387e-06, + "loss": 0.6091, + "step": 2942 + }, + { + "epoch": 2.436499171728327, + "grad_norm": 0.9396141171455383, + "learning_rate": 3.2869611251899136e-06, + "loss": 0.6143, + "step": 2943 + }, + { + "epoch": 2.437327443401436, + "grad_norm": 0.9370556473731995, + "learning_rate": 3.285917264404446e-06, + "loss": 0.61, + "step": 2944 + }, + { + "epoch": 2.4381557150745445, + "grad_norm": 0.9142118096351624, + "learning_rate": 3.2848732515513153e-06, + "loss": 0.5875, + "step": 2945 + }, + { + "epoch": 2.4389839867476533, + "grad_norm": 0.9196925759315491, + "learning_rate": 3.2838290868325267e-06, + "loss": 0.6058, + "step": 2946 + }, + { + "epoch": 2.439812258420762, + "grad_norm": 0.892217755317688, + "learning_rate": 3.282784770450116e-06, + "loss": 0.6005, + "step": 2947 + }, + { + "epoch": 2.4406405300938707, + "grad_norm": 0.9278191924095154, + "learning_rate": 3.2817403026061496e-06, + "loss": 0.6206, + "step": 2948 + }, + { + "epoch": 2.4414688017669794, + "grad_norm": 0.9343185424804688, + "learning_rate": 3.2806956835027216e-06, + "loss": 0.604, + "step": 2949 + }, + { + "epoch": 2.442297073440088, + "grad_norm": 0.9401823878288269, + "learning_rate": 3.2796509133419564e-06, + "loss": 0.6039, + "step": 2950 + }, + { + "epoch": 2.443125345113197, + "grad_norm": 0.9102358818054199, + "learning_rate": 3.278605992326008e-06, + "loss": 0.5892, + "step": 2951 + }, + { + "epoch": 2.443953616786306, + "grad_norm": 0.9241907000541687, + "learning_rate": 3.2775609206570573e-06, + "loss": 0.6054, + "step": 2952 + }, + { + "epoch": 2.444781888459415, + "grad_norm": 0.9492688775062561, + "learning_rate": 3.276515698537317e-06, + "loss": 0.5922, + "step": 2953 + }, + { + "epoch": 2.4456101601325235, + "grad_norm": 0.9492383003234863, + "learning_rate": 3.275470326169028e-06, + "loss": 0.5735, + "step": 2954 + }, + { + "epoch": 2.4464384318056323, + "grad_norm": 0.8857248425483704, + "learning_rate": 3.2744248037544597e-06, + "loss": 0.584, + "step": 2955 + }, + { + "epoch": 2.447266703478741, + "grad_norm": 0.9185032248497009, + "learning_rate": 3.2733791314959108e-06, + "loss": 0.5906, + "step": 2956 + }, + { + "epoch": 2.4480949751518497, + "grad_norm": 0.9257567524909973, + "learning_rate": 3.2723333095957094e-06, + "loss": 0.587, + "step": 2957 + }, + { + "epoch": 2.4489232468249584, + "grad_norm": 0.9427464008331299, + "learning_rate": 3.2712873382562117e-06, + "loss": 0.5932, + "step": 2958 + }, + { + "epoch": 2.4497515184980676, + "grad_norm": 0.9002125263214111, + "learning_rate": 3.2702412176798045e-06, + "loss": 0.58, + "step": 2959 + }, + { + "epoch": 2.4505797901711763, + "grad_norm": 0.906929075717926, + "learning_rate": 3.2691949480689018e-06, + "loss": 0.5878, + "step": 2960 + }, + { + "epoch": 2.451408061844285, + "grad_norm": 0.9162207841873169, + "learning_rate": 3.268148529625946e-06, + "loss": 0.5975, + "step": 2961 + }, + { + "epoch": 2.452236333517394, + "grad_norm": 0.9268636107444763, + "learning_rate": 3.2671019625534117e-06, + "loss": 0.6028, + "step": 2962 + }, + { + "epoch": 2.4530646051905025, + "grad_norm": 0.9216850996017456, + "learning_rate": 3.2660552470537975e-06, + "loss": 0.5938, + "step": 2963 + }, + { + "epoch": 2.4538928768636112, + "grad_norm": 0.9353698492050171, + "learning_rate": 3.2650083833296344e-06, + "loss": 0.6069, + "step": 2964 + }, + { + "epoch": 2.45472114853672, + "grad_norm": 0.9393898248672485, + "learning_rate": 3.2639613715834805e-06, + "loss": 0.5874, + "step": 2965 + }, + { + "epoch": 2.4555494202098287, + "grad_norm": 0.9410581588745117, + "learning_rate": 3.2629142120179224e-06, + "loss": 0.5918, + "step": 2966 + }, + { + "epoch": 2.4563776918829374, + "grad_norm": 0.9445161819458008, + "learning_rate": 3.2618669048355763e-06, + "loss": 0.5743, + "step": 2967 + }, + { + "epoch": 2.4572059635560466, + "grad_norm": 0.9591469764709473, + "learning_rate": 3.2608194502390867e-06, + "loss": 0.5781, + "step": 2968 + }, + { + "epoch": 2.4580342352291553, + "grad_norm": 0.9237606525421143, + "learning_rate": 3.2597718484311246e-06, + "loss": 0.6163, + "step": 2969 + }, + { + "epoch": 2.458862506902264, + "grad_norm": 0.9216769337654114, + "learning_rate": 3.2587240996143933e-06, + "loss": 0.6157, + "step": 2970 + }, + { + "epoch": 2.4596907785753728, + "grad_norm": 0.9081043004989624, + "learning_rate": 3.257676203991621e-06, + "loss": 0.5821, + "step": 2971 + }, + { + "epoch": 2.4605190502484815, + "grad_norm": 0.9316040873527527, + "learning_rate": 3.256628161765566e-06, + "loss": 0.5911, + "step": 2972 + }, + { + "epoch": 2.4613473219215902, + "grad_norm": 0.9326239824295044, + "learning_rate": 3.2555799731390158e-06, + "loss": 0.5963, + "step": 2973 + }, + { + "epoch": 2.462175593594699, + "grad_norm": 0.9429934024810791, + "learning_rate": 3.254531638314784e-06, + "loss": 0.5651, + "step": 2974 + }, + { + "epoch": 2.4630038652678077, + "grad_norm": 0.9090480804443359, + "learning_rate": 3.2534831574957134e-06, + "loss": 0.5779, + "step": 2975 + }, + { + "epoch": 2.4638321369409164, + "grad_norm": 0.9582802653312683, + "learning_rate": 3.2524345308846756e-06, + "loss": 0.5887, + "step": 2976 + }, + { + "epoch": 2.4646604086140256, + "grad_norm": 0.9248860478401184, + "learning_rate": 3.2513857586845705e-06, + "loss": 0.5936, + "step": 2977 + }, + { + "epoch": 2.4654886802871343, + "grad_norm": 0.9345134496688843, + "learning_rate": 3.250336841098325e-06, + "loss": 0.5985, + "step": 2978 + }, + { + "epoch": 2.466316951960243, + "grad_norm": 0.9092442393302917, + "learning_rate": 3.249287778328896e-06, + "loss": 0.5926, + "step": 2979 + }, + { + "epoch": 2.4671452236333518, + "grad_norm": 0.932470977306366, + "learning_rate": 3.2482385705792655e-06, + "loss": 0.5797, + "step": 2980 + }, + { + "epoch": 2.4679734953064605, + "grad_norm": 0.9335013031959534, + "learning_rate": 3.247189218052448e-06, + "loss": 0.5933, + "step": 2981 + }, + { + "epoch": 2.468801766979569, + "grad_norm": 0.9283298254013062, + "learning_rate": 3.2461397209514812e-06, + "loss": 0.6003, + "step": 2982 + }, + { + "epoch": 2.469630038652678, + "grad_norm": 0.9429066181182861, + "learning_rate": 3.245090079479434e-06, + "loss": 0.5944, + "step": 2983 + }, + { + "epoch": 2.4704583103257867, + "grad_norm": 0.9284510612487793, + "learning_rate": 3.2440402938394017e-06, + "loss": 0.5782, + "step": 2984 + }, + { + "epoch": 2.4712865819988954, + "grad_norm": 0.9150790572166443, + "learning_rate": 3.2429903642345094e-06, + "loss": 0.5862, + "step": 2985 + }, + { + "epoch": 2.4721148536720046, + "grad_norm": 0.9177762866020203, + "learning_rate": 3.2419402908679065e-06, + "loss": 0.5931, + "step": 2986 + }, + { + "epoch": 2.4729431253451133, + "grad_norm": 0.9098034501075745, + "learning_rate": 3.2408900739427744e-06, + "loss": 0.5864, + "step": 2987 + }, + { + "epoch": 2.473771397018222, + "grad_norm": 0.9237478971481323, + "learning_rate": 3.2398397136623195e-06, + "loss": 0.5873, + "step": 2988 + }, + { + "epoch": 2.4745996686913307, + "grad_norm": 0.9212510585784912, + "learning_rate": 3.2387892102297757e-06, + "loss": 0.5884, + "step": 2989 + }, + { + "epoch": 2.4754279403644395, + "grad_norm": 0.8987523913383484, + "learning_rate": 3.237738563848408e-06, + "loss": 0.5806, + "step": 2990 + }, + { + "epoch": 2.476256212037548, + "grad_norm": 0.9036835432052612, + "learning_rate": 3.236687774721504e-06, + "loss": 0.5993, + "step": 2991 + }, + { + "epoch": 2.477084483710657, + "grad_norm": 0.9291606545448303, + "learning_rate": 3.235636843052384e-06, + "loss": 0.5784, + "step": 2992 + }, + { + "epoch": 2.477912755383766, + "grad_norm": 0.9339813590049744, + "learning_rate": 3.234585769044392e-06, + "loss": 0.5951, + "step": 2993 + }, + { + "epoch": 2.478741027056875, + "grad_norm": 0.9638957977294922, + "learning_rate": 3.2335345529009017e-06, + "loss": 0.591, + "step": 2994 + }, + { + "epoch": 2.4795692987299836, + "grad_norm": 0.9618088603019714, + "learning_rate": 3.2324831948253123e-06, + "loss": 0.6102, + "step": 2995 + }, + { + "epoch": 2.4803975704030923, + "grad_norm": 0.9491007924079895, + "learning_rate": 3.2314316950210547e-06, + "loss": 0.599, + "step": 2996 + }, + { + "epoch": 2.481225842076201, + "grad_norm": 0.9213217496871948, + "learning_rate": 3.2303800536915804e-06, + "loss": 0.5955, + "step": 2997 + }, + { + "epoch": 2.4820541137493097, + "grad_norm": 0.923991322517395, + "learning_rate": 3.229328271040375e-06, + "loss": 0.5903, + "step": 2998 + }, + { + "epoch": 2.4828823854224185, + "grad_norm": 0.9173145890235901, + "learning_rate": 3.2282763472709484e-06, + "loss": 0.5899, + "step": 2999 + }, + { + "epoch": 2.483710657095527, + "grad_norm": 0.9357689023017883, + "learning_rate": 3.2272242825868363e-06, + "loss": 0.5885, + "step": 3000 + }, + { + "epoch": 2.484538928768636, + "grad_norm": 0.8902420401573181, + "learning_rate": 3.2261720771916056e-06, + "loss": 0.5899, + "step": 3001 + }, + { + "epoch": 2.485367200441745, + "grad_norm": 0.9079567790031433, + "learning_rate": 3.2251197312888475e-06, + "loss": 0.5966, + "step": 3002 + }, + { + "epoch": 2.486195472114854, + "grad_norm": 0.9068807363510132, + "learning_rate": 3.22406724508218e-06, + "loss": 0.6202, + "step": 3003 + }, + { + "epoch": 2.4870237437879625, + "grad_norm": 0.9458021521568298, + "learning_rate": 3.223014618775251e-06, + "loss": 0.5812, + "step": 3004 + }, + { + "epoch": 2.4878520154610713, + "grad_norm": 0.9506248235702515, + "learning_rate": 3.2219618525717328e-06, + "loss": 0.5819, + "step": 3005 + }, + { + "epoch": 2.48868028713418, + "grad_norm": 0.9573289155960083, + "learning_rate": 3.220908946675326e-06, + "loss": 0.581, + "step": 3006 + }, + { + "epoch": 2.4895085588072887, + "grad_norm": 0.9035286903381348, + "learning_rate": 3.219855901289758e-06, + "loss": 0.6182, + "step": 3007 + }, + { + "epoch": 2.4903368304803974, + "grad_norm": 0.9300901293754578, + "learning_rate": 3.218802716618784e-06, + "loss": 0.6079, + "step": 3008 + }, + { + "epoch": 2.491165102153506, + "grad_norm": 0.9116837382316589, + "learning_rate": 3.2177493928661854e-06, + "loss": 0.6034, + "step": 3009 + }, + { + "epoch": 2.491993373826615, + "grad_norm": 0.9291449189186096, + "learning_rate": 3.2166959302357694e-06, + "loss": 0.5889, + "step": 3010 + }, + { + "epoch": 2.492821645499724, + "grad_norm": 0.9002729654312134, + "learning_rate": 3.215642328931371e-06, + "loss": 0.5969, + "step": 3011 + }, + { + "epoch": 2.493649917172833, + "grad_norm": 0.8961556553840637, + "learning_rate": 3.214588589156854e-06, + "loss": 0.5899, + "step": 3012 + }, + { + "epoch": 2.4944781888459415, + "grad_norm": 0.9115450382232666, + "learning_rate": 3.2135347111161075e-06, + "loss": 0.5727, + "step": 3013 + }, + { + "epoch": 2.4953064605190503, + "grad_norm": 0.9060878157615662, + "learning_rate": 3.212480695013044e-06, + "loss": 0.5765, + "step": 3014 + }, + { + "epoch": 2.496134732192159, + "grad_norm": 0.902313768863678, + "learning_rate": 3.2114265410516088e-06, + "loss": 0.5913, + "step": 3015 + }, + { + "epoch": 2.4969630038652677, + "grad_norm": 0.931974470615387, + "learning_rate": 3.210372249435769e-06, + "loss": 0.5891, + "step": 3016 + }, + { + "epoch": 2.4977912755383764, + "grad_norm": 1.1711225509643555, + "learning_rate": 3.209317820369521e-06, + "loss": 0.6246, + "step": 3017 + }, + { + "epoch": 2.4986195472114856, + "grad_norm": 0.9561757445335388, + "learning_rate": 3.208263254056887e-06, + "loss": 0.5847, + "step": 3018 + }, + { + "epoch": 2.4994478188845943, + "grad_norm": 0.9621199369430542, + "learning_rate": 3.207208550701916e-06, + "loss": 0.5971, + "step": 3019 + }, + { + "epoch": 2.500276090557703, + "grad_norm": 0.9152470827102661, + "learning_rate": 3.206153710508683e-06, + "loss": 0.59, + "step": 3020 + }, + { + "epoch": 2.501104362230812, + "grad_norm": 0.9315138459205627, + "learning_rate": 3.20509873368129e-06, + "loss": 0.5841, + "step": 3021 + }, + { + "epoch": 2.5019326339039205, + "grad_norm": 0.9124136567115784, + "learning_rate": 3.204043620423864e-06, + "loss": 0.6007, + "step": 3022 + }, + { + "epoch": 2.5027609055770292, + "grad_norm": 0.8976690173149109, + "learning_rate": 3.2029883709405607e-06, + "loss": 0.5969, + "step": 3023 + }, + { + "epoch": 2.503589177250138, + "grad_norm": 0.8946467041969299, + "learning_rate": 3.201932985435561e-06, + "loss": 0.6069, + "step": 3024 + }, + { + "epoch": 2.5044174489232467, + "grad_norm": 0.9290626049041748, + "learning_rate": 3.2008774641130722e-06, + "loss": 0.591, + "step": 3025 + }, + { + "epoch": 2.5052457205963554, + "grad_norm": 0.8982367515563965, + "learning_rate": 3.1998218071773275e-06, + "loss": 0.5874, + "step": 3026 + }, + { + "epoch": 2.5060739922694646, + "grad_norm": 0.9062193036079407, + "learning_rate": 3.1987660148325867e-06, + "loss": 0.6107, + "step": 3027 + }, + { + "epoch": 2.5069022639425733, + "grad_norm": 0.9240522980690002, + "learning_rate": 3.1977100872831353e-06, + "loss": 0.5806, + "step": 3028 + }, + { + "epoch": 2.507730535615682, + "grad_norm": 0.9382046461105347, + "learning_rate": 3.1966540247332865e-06, + "loss": 0.5944, + "step": 3029 + }, + { + "epoch": 2.5085588072887908, + "grad_norm": 0.9194355010986328, + "learning_rate": 3.1955978273873776e-06, + "loss": 0.6085, + "step": 3030 + }, + { + "epoch": 2.5093870789618995, + "grad_norm": 0.9219276309013367, + "learning_rate": 3.1945414954497744e-06, + "loss": 0.5829, + "step": 3031 + }, + { + "epoch": 2.5102153506350082, + "grad_norm": 0.8998081088066101, + "learning_rate": 3.193485029124865e-06, + "loss": 0.6095, + "step": 3032 + }, + { + "epoch": 2.511043622308117, + "grad_norm": 0.9184247851371765, + "learning_rate": 3.192428428617068e-06, + "loss": 0.6129, + "step": 3033 + }, + { + "epoch": 2.5118718939812257, + "grad_norm": 0.9389240741729736, + "learning_rate": 3.1913716941308243e-06, + "loss": 0.607, + "step": 3034 + }, + { + "epoch": 2.5127001656543344, + "grad_norm": 0.9263836741447449, + "learning_rate": 3.190314825870603e-06, + "loss": 0.5761, + "step": 3035 + }, + { + "epoch": 2.5135284373274436, + "grad_norm": 0.9472658038139343, + "learning_rate": 3.1892578240408977e-06, + "loss": 0.5906, + "step": 3036 + }, + { + "epoch": 2.5143567090005523, + "grad_norm": 0.9045345783233643, + "learning_rate": 3.1882006888462285e-06, + "loss": 0.5993, + "step": 3037 + }, + { + "epoch": 2.515184980673661, + "grad_norm": 0.8930817246437073, + "learning_rate": 3.187143420491142e-06, + "loss": 0.5947, + "step": 3038 + }, + { + "epoch": 2.5160132523467698, + "grad_norm": 0.9130663871765137, + "learning_rate": 3.186086019180208e-06, + "loss": 0.5892, + "step": 3039 + }, + { + "epoch": 2.5168415240198785, + "grad_norm": 0.9008771777153015, + "learning_rate": 3.185028485118026e-06, + "loss": 0.5787, + "step": 3040 + }, + { + "epoch": 2.517669795692987, + "grad_norm": 0.9192292094230652, + "learning_rate": 3.1839708185092177e-06, + "loss": 0.5848, + "step": 3041 + }, + { + "epoch": 2.518498067366096, + "grad_norm": 0.8970836997032166, + "learning_rate": 3.1829130195584314e-06, + "loss": 0.5939, + "step": 3042 + }, + { + "epoch": 2.519326339039205, + "grad_norm": 0.9393385052680969, + "learning_rate": 3.181855088470342e-06, + "loss": 0.6096, + "step": 3043 + }, + { + "epoch": 2.5201546107123134, + "grad_norm": 0.9437024593353271, + "learning_rate": 3.180797025449649e-06, + "loss": 0.5998, + "step": 3044 + }, + { + "epoch": 2.5209828823854226, + "grad_norm": 0.9014863967895508, + "learning_rate": 3.179738830701078e-06, + "loss": 0.6015, + "step": 3045 + }, + { + "epoch": 2.5218111540585313, + "grad_norm": 0.9243669509887695, + "learning_rate": 3.1786805044293802e-06, + "loss": 0.5971, + "step": 3046 + }, + { + "epoch": 2.52263942573164, + "grad_norm": 0.9175970554351807, + "learning_rate": 3.1776220468393305e-06, + "loss": 0.6084, + "step": 3047 + }, + { + "epoch": 2.5234676974047487, + "grad_norm": 0.8962888121604919, + "learning_rate": 3.176563458135732e-06, + "loss": 0.5979, + "step": 3048 + }, + { + "epoch": 2.5242959690778575, + "grad_norm": 0.9321010708808899, + "learning_rate": 3.1755047385234117e-06, + "loss": 0.5941, + "step": 3049 + }, + { + "epoch": 2.525124240750966, + "grad_norm": 0.9147899746894836, + "learning_rate": 3.17444588820722e-06, + "loss": 0.5987, + "step": 3050 + }, + { + "epoch": 2.525952512424075, + "grad_norm": 0.9318193793296814, + "learning_rate": 3.173386907392037e-06, + "loss": 0.5658, + "step": 3051 + }, + { + "epoch": 2.526780784097184, + "grad_norm": 0.9039092659950256, + "learning_rate": 3.1723277962827647e-06, + "loss": 0.5956, + "step": 3052 + }, + { + "epoch": 2.5276090557702924, + "grad_norm": 0.9080818295478821, + "learning_rate": 3.1712685550843314e-06, + "loss": 0.5957, + "step": 3053 + }, + { + "epoch": 2.5284373274434015, + "grad_norm": 0.9120905995368958, + "learning_rate": 3.17020918400169e-06, + "loss": 0.5917, + "step": 3054 + }, + { + "epoch": 2.5292655991165103, + "grad_norm": 0.8983855843544006, + "learning_rate": 3.169149683239819e-06, + "loss": 0.6045, + "step": 3055 + }, + { + "epoch": 2.530093870789619, + "grad_norm": 0.9203280210494995, + "learning_rate": 3.168090053003721e-06, + "loss": 0.6093, + "step": 3056 + }, + { + "epoch": 2.5309221424627277, + "grad_norm": 0.9343932867050171, + "learning_rate": 3.1670302934984272e-06, + "loss": 0.5654, + "step": 3057 + }, + { + "epoch": 2.5317504141358365, + "grad_norm": 0.9054449200630188, + "learning_rate": 3.1659704049289885e-06, + "loss": 0.5878, + "step": 3058 + }, + { + "epoch": 2.532578685808945, + "grad_norm": 0.9011321067810059, + "learning_rate": 3.1649103875004856e-06, + "loss": 0.5831, + "step": 3059 + }, + { + "epoch": 2.533406957482054, + "grad_norm": 0.8886737823486328, + "learning_rate": 3.1638502414180206e-06, + "loss": 0.5915, + "step": 3060 + }, + { + "epoch": 2.534235229155163, + "grad_norm": 0.9289557337760925, + "learning_rate": 3.1627899668867214e-06, + "loss": 0.5944, + "step": 3061 + }, + { + "epoch": 2.535063500828272, + "grad_norm": 0.9269216060638428, + "learning_rate": 3.1617295641117425e-06, + "loss": 0.605, + "step": 3062 + }, + { + "epoch": 2.5358917725013805, + "grad_norm": 0.8877415657043457, + "learning_rate": 3.1606690332982613e-06, + "loss": 0.5954, + "step": 3063 + }, + { + "epoch": 2.5367200441744893, + "grad_norm": 0.9080086350440979, + "learning_rate": 3.1596083746514817e-06, + "loss": 0.5896, + "step": 3064 + }, + { + "epoch": 2.537548315847598, + "grad_norm": 0.9625695943832397, + "learning_rate": 3.15854758837663e-06, + "loss": 0.5666, + "step": 3065 + }, + { + "epoch": 2.5383765875207067, + "grad_norm": 0.9442515969276428, + "learning_rate": 3.1574866746789583e-06, + "loss": 0.5837, + "step": 3066 + }, + { + "epoch": 2.5392048591938154, + "grad_norm": 0.9054381847381592, + "learning_rate": 3.1564256337637443e-06, + "loss": 0.5799, + "step": 3067 + }, + { + "epoch": 2.5400331308669246, + "grad_norm": 0.9435595273971558, + "learning_rate": 3.1553644658362897e-06, + "loss": 0.6135, + "step": 3068 + }, + { + "epoch": 2.540861402540033, + "grad_norm": 0.9356298446655273, + "learning_rate": 3.1543031711019203e-06, + "loss": 0.5853, + "step": 3069 + }, + { + "epoch": 2.541689674213142, + "grad_norm": 0.917759895324707, + "learning_rate": 3.153241749765986e-06, + "loss": 0.5902, + "step": 3070 + }, + { + "epoch": 2.542517945886251, + "grad_norm": 0.9233857989311218, + "learning_rate": 3.1521802020338627e-06, + "loss": 0.6066, + "step": 3071 + }, + { + "epoch": 2.5433462175593595, + "grad_norm": 0.9497340321540833, + "learning_rate": 3.1511185281109504e-06, + "loss": 0.6003, + "step": 3072 + }, + { + "epoch": 2.5441744892324683, + "grad_norm": 0.9373629689216614, + "learning_rate": 3.1500567282026716e-06, + "loss": 0.5827, + "step": 3073 + }, + { + "epoch": 2.545002760905577, + "grad_norm": 0.8916833996772766, + "learning_rate": 3.148994802514476e-06, + "loss": 0.5996, + "step": 3074 + }, + { + "epoch": 2.5458310325786857, + "grad_norm": 0.9327612519264221, + "learning_rate": 3.1479327512518363e-06, + "loss": 0.6159, + "step": 3075 + }, + { + "epoch": 2.5466593042517944, + "grad_norm": 0.9380506873130798, + "learning_rate": 3.1468705746202493e-06, + "loss": 0.5924, + "step": 3076 + }, + { + "epoch": 2.5474875759249036, + "grad_norm": 0.9138150215148926, + "learning_rate": 3.1458082728252358e-06, + "loss": 0.5734, + "step": 3077 + }, + { + "epoch": 2.548315847598012, + "grad_norm": 0.8954431414604187, + "learning_rate": 3.144745846072341e-06, + "loss": 0.5902, + "step": 3078 + }, + { + "epoch": 2.549144119271121, + "grad_norm": 0.9259476661682129, + "learning_rate": 3.1436832945671362e-06, + "loss": 0.6057, + "step": 3079 + }, + { + "epoch": 2.54997239094423, + "grad_norm": 0.9295196533203125, + "learning_rate": 3.142620618515214e-06, + "loss": 0.5971, + "step": 3080 + }, + { + "epoch": 2.5508006626173385, + "grad_norm": 0.9318671822547913, + "learning_rate": 3.1415578181221928e-06, + "loss": 0.5852, + "step": 3081 + }, + { + "epoch": 2.5516289342904472, + "grad_norm": 0.9349870085716248, + "learning_rate": 3.1404948935937145e-06, + "loss": 0.5736, + "step": 3082 + }, + { + "epoch": 2.552457205963556, + "grad_norm": 0.9280831217765808, + "learning_rate": 3.1394318451354446e-06, + "loss": 0.5898, + "step": 3083 + }, + { + "epoch": 2.5532854776366647, + "grad_norm": 0.8969367742538452, + "learning_rate": 3.1383686729530727e-06, + "loss": 0.5983, + "step": 3084 + }, + { + "epoch": 2.5541137493097734, + "grad_norm": 0.9363370537757874, + "learning_rate": 3.1373053772523145e-06, + "loss": 0.6052, + "step": 3085 + }, + { + "epoch": 2.5549420209828826, + "grad_norm": 0.8944993615150452, + "learning_rate": 3.1362419582389057e-06, + "loss": 0.599, + "step": 3086 + }, + { + "epoch": 2.5557702926559913, + "grad_norm": 0.891412615776062, + "learning_rate": 3.1351784161186093e-06, + "loss": 0.5916, + "step": 3087 + }, + { + "epoch": 2.5565985643291, + "grad_norm": 0.930220365524292, + "learning_rate": 3.134114751097211e-06, + "loss": 0.5826, + "step": 3088 + }, + { + "epoch": 2.5574268360022088, + "grad_norm": 0.9193615317344666, + "learning_rate": 3.1330509633805183e-06, + "loss": 0.5896, + "step": 3089 + }, + { + "epoch": 2.5582551076753175, + "grad_norm": 0.9178160429000854, + "learning_rate": 3.131987053174367e-06, + "loss": 0.6104, + "step": 3090 + }, + { + "epoch": 2.5590833793484262, + "grad_norm": 0.9119648933410645, + "learning_rate": 3.130923020684611e-06, + "loss": 0.6198, + "step": 3091 + }, + { + "epoch": 2.559911651021535, + "grad_norm": 0.957272469997406, + "learning_rate": 3.1298588661171323e-06, + "loss": 0.5716, + "step": 3092 + }, + { + "epoch": 2.5607399226946437, + "grad_norm": 0.9073137044906616, + "learning_rate": 3.128794589677835e-06, + "loss": 0.5959, + "step": 3093 + }, + { + "epoch": 2.5615681943677524, + "grad_norm": 0.9372225403785706, + "learning_rate": 3.1277301915726456e-06, + "loss": 0.6146, + "step": 3094 + }, + { + "epoch": 2.5623964660408616, + "grad_norm": 0.9377845525741577, + "learning_rate": 3.126665672007515e-06, + "loss": 0.6055, + "step": 3095 + }, + { + "epoch": 2.5632247377139703, + "grad_norm": 0.9114481210708618, + "learning_rate": 3.1256010311884193e-06, + "loss": 0.6038, + "step": 3096 + }, + { + "epoch": 2.564053009387079, + "grad_norm": 0.9000943303108215, + "learning_rate": 3.124536269321356e-06, + "loss": 0.5806, + "step": 3097 + }, + { + "epoch": 2.5648812810601878, + "grad_norm": 0.9771347641944885, + "learning_rate": 3.123471386612347e-06, + "loss": 0.6029, + "step": 3098 + }, + { + "epoch": 2.5657095527332965, + "grad_norm": 0.9150393605232239, + "learning_rate": 3.1224063832674358e-06, + "loss": 0.6033, + "step": 3099 + }, + { + "epoch": 2.566537824406405, + "grad_norm": 0.8910076022148132, + "learning_rate": 3.1213412594926906e-06, + "loss": 0.6029, + "step": 3100 + }, + { + "epoch": 2.567366096079514, + "grad_norm": 0.9558795094490051, + "learning_rate": 3.1202760154942047e-06, + "loss": 0.5792, + "step": 3101 + }, + { + "epoch": 2.568194367752623, + "grad_norm": 0.958629310131073, + "learning_rate": 3.1192106514780924e-06, + "loss": 0.5933, + "step": 3102 + }, + { + "epoch": 2.5690226394257314, + "grad_norm": 0.9466328620910645, + "learning_rate": 3.1181451676504907e-06, + "loss": 0.6027, + "step": 3103 + }, + { + "epoch": 2.5698509110988406, + "grad_norm": 0.9397157430648804, + "learning_rate": 3.1170795642175614e-06, + "loss": 0.5894, + "step": 3104 + }, + { + "epoch": 2.5706791827719493, + "grad_norm": 0.9229122996330261, + "learning_rate": 3.116013841385489e-06, + "loss": 0.5812, + "step": 3105 + }, + { + "epoch": 2.571507454445058, + "grad_norm": 0.9332060217857361, + "learning_rate": 3.1149479993604805e-06, + "loss": 0.5738, + "step": 3106 + }, + { + "epoch": 2.5723357261181667, + "grad_norm": 0.9117311239242554, + "learning_rate": 3.113882038348766e-06, + "loss": 0.5923, + "step": 3107 + }, + { + "epoch": 2.5731639977912755, + "grad_norm": 0.9390400052070618, + "learning_rate": 3.1128159585566003e-06, + "loss": 0.6029, + "step": 3108 + }, + { + "epoch": 2.573992269464384, + "grad_norm": 0.9229103326797485, + "learning_rate": 3.1117497601902596e-06, + "loss": 0.6118, + "step": 3109 + }, + { + "epoch": 2.574820541137493, + "grad_norm": 0.8975940346717834, + "learning_rate": 3.1106834434560424e-06, + "loss": 0.6005, + "step": 3110 + }, + { + "epoch": 2.575648812810602, + "grad_norm": 0.9088030457496643, + "learning_rate": 3.1096170085602718e-06, + "loss": 0.5896, + "step": 3111 + }, + { + "epoch": 2.576477084483711, + "grad_norm": 0.982429027557373, + "learning_rate": 3.108550455709292e-06, + "loss": 0.5947, + "step": 3112 + }, + { + "epoch": 2.5773053561568195, + "grad_norm": 0.9404535293579102, + "learning_rate": 3.107483785109473e-06, + "loss": 0.6154, + "step": 3113 + }, + { + "epoch": 2.5781336278299283, + "grad_norm": 0.9074321389198303, + "learning_rate": 3.1064169969672043e-06, + "loss": 0.5889, + "step": 3114 + }, + { + "epoch": 2.578961899503037, + "grad_norm": 0.9000370502471924, + "learning_rate": 3.105350091488899e-06, + "loss": 0.5725, + "step": 3115 + }, + { + "epoch": 2.5797901711761457, + "grad_norm": 0.9296747446060181, + "learning_rate": 3.1042830688809944e-06, + "loss": 0.5983, + "step": 3116 + }, + { + "epoch": 2.5806184428492545, + "grad_norm": 0.9164032936096191, + "learning_rate": 3.1032159293499486e-06, + "loss": 0.5832, + "step": 3117 + }, + { + "epoch": 2.581446714522363, + "grad_norm": 0.9070228338241577, + "learning_rate": 3.1021486731022443e-06, + "loss": 0.5977, + "step": 3118 + }, + { + "epoch": 2.582274986195472, + "grad_norm": 0.9384707808494568, + "learning_rate": 3.1010813003443847e-06, + "loss": 0.5926, + "step": 3119 + }, + { + "epoch": 2.583103257868581, + "grad_norm": 0.937782347202301, + "learning_rate": 3.100013811282896e-06, + "loss": 0.5859, + "step": 3120 + }, + { + "epoch": 2.58393152954169, + "grad_norm": 0.9293384552001953, + "learning_rate": 3.098946206124329e-06, + "loss": 0.5826, + "step": 3121 + }, + { + "epoch": 2.5847598012147985, + "grad_norm": 0.935950756072998, + "learning_rate": 3.097878485075255e-06, + "loss": 0.589, + "step": 3122 + }, + { + "epoch": 2.5855880728879073, + "grad_norm": 0.9377515316009521, + "learning_rate": 3.0968106483422655e-06, + "loss": 0.5985, + "step": 3123 + }, + { + "epoch": 2.586416344561016, + "grad_norm": 0.9256250858306885, + "learning_rate": 3.09574269613198e-06, + "loss": 0.5944, + "step": 3124 + }, + { + "epoch": 2.5872446162341247, + "grad_norm": 0.873123288154602, + "learning_rate": 3.0946746286510364e-06, + "loss": 0.5918, + "step": 3125 + }, + { + "epoch": 2.5880728879072334, + "grad_norm": 0.9320067167282104, + "learning_rate": 3.093606446106096e-06, + "loss": 0.6365, + "step": 3126 + }, + { + "epoch": 2.5889011595803426, + "grad_norm": 0.9514285326004028, + "learning_rate": 3.0925381487038415e-06, + "loss": 0.585, + "step": 3127 + }, + { + "epoch": 2.589729431253451, + "grad_norm": 0.9028947353363037, + "learning_rate": 3.0914697366509777e-06, + "loss": 0.6063, + "step": 3128 + }, + { + "epoch": 2.59055770292656, + "grad_norm": 0.8989999294281006, + "learning_rate": 3.0904012101542346e-06, + "loss": 0.6012, + "step": 3129 + }, + { + "epoch": 2.591385974599669, + "grad_norm": 0.9131087064743042, + "learning_rate": 3.089332569420361e-06, + "loss": 0.5742, + "step": 3130 + }, + { + "epoch": 2.5922142462727775, + "grad_norm": 0.916441023349762, + "learning_rate": 3.088263814656128e-06, + "loss": 0.6011, + "step": 3131 + }, + { + "epoch": 2.5930425179458862, + "grad_norm": 0.8957144618034363, + "learning_rate": 3.087194946068332e-06, + "loss": 0.6008, + "step": 3132 + }, + { + "epoch": 2.593870789618995, + "grad_norm": 0.90632563829422, + "learning_rate": 3.086125963863787e-06, + "loss": 0.6062, + "step": 3133 + }, + { + "epoch": 2.5946990612921037, + "grad_norm": 0.94493168592453, + "learning_rate": 3.0850568682493315e-06, + "loss": 0.5667, + "step": 3134 + }, + { + "epoch": 2.5955273329652124, + "grad_norm": 0.9155673980712891, + "learning_rate": 3.0839876594318257e-06, + "loss": 0.5755, + "step": 3135 + }, + { + "epoch": 2.5963556046383216, + "grad_norm": 0.9013570547103882, + "learning_rate": 3.0829183376181524e-06, + "loss": 0.6053, + "step": 3136 + }, + { + "epoch": 2.59718387631143, + "grad_norm": 0.9294494390487671, + "learning_rate": 3.081848903015214e-06, + "loss": 0.5881, + "step": 3137 + }, + { + "epoch": 2.598012147984539, + "grad_norm": 0.8956479430198669, + "learning_rate": 3.0807793558299377e-06, + "loss": 0.5976, + "step": 3138 + }, + { + "epoch": 2.598840419657648, + "grad_norm": 0.9205741286277771, + "learning_rate": 3.079709696269269e-06, + "loss": 0.5776, + "step": 3139 + }, + { + "epoch": 2.5996686913307565, + "grad_norm": 0.901066780090332, + "learning_rate": 3.0786399245401784e-06, + "loss": 0.6101, + "step": 3140 + }, + { + "epoch": 2.6004969630038652, + "grad_norm": 0.9003482460975647, + "learning_rate": 3.0775700408496568e-06, + "loss": 0.5983, + "step": 3141 + }, + { + "epoch": 2.601325234676974, + "grad_norm": 0.9000759124755859, + "learning_rate": 3.0765000454047155e-06, + "loss": 0.605, + "step": 3142 + }, + { + "epoch": 2.6021535063500827, + "grad_norm": 0.8797931671142578, + "learning_rate": 3.0754299384123905e-06, + "loss": 0.5874, + "step": 3143 + }, + { + "epoch": 2.6029817780231914, + "grad_norm": 0.9307072758674622, + "learning_rate": 3.074359720079736e-06, + "loss": 0.6035, + "step": 3144 + }, + { + "epoch": 2.6038100496963006, + "grad_norm": 0.8928053379058838, + "learning_rate": 3.073289390613829e-06, + "loss": 0.5759, + "step": 3145 + }, + { + "epoch": 2.6046383213694093, + "grad_norm": 0.8879983425140381, + "learning_rate": 3.0722189502217705e-06, + "loss": 0.5947, + "step": 3146 + }, + { + "epoch": 2.605466593042518, + "grad_norm": 0.8889674544334412, + "learning_rate": 3.0711483991106784e-06, + "loss": 0.5939, + "step": 3147 + }, + { + "epoch": 2.6062948647156268, + "grad_norm": 0.9318514466285706, + "learning_rate": 3.070077737487695e-06, + "loss": 0.599, + "step": 3148 + }, + { + "epoch": 2.6071231363887355, + "grad_norm": 0.9251010417938232, + "learning_rate": 3.069006965559985e-06, + "loss": 0.5956, + "step": 3149 + }, + { + "epoch": 2.6079514080618442, + "grad_norm": 0.9122523069381714, + "learning_rate": 3.06793608353473e-06, + "loss": 0.6011, + "step": 3150 + }, + { + "epoch": 2.608779679734953, + "grad_norm": 0.9270116090774536, + "learning_rate": 3.0668650916191365e-06, + "loss": 0.5923, + "step": 3151 + }, + { + "epoch": 2.609607951408062, + "grad_norm": 0.939582884311676, + "learning_rate": 3.0657939900204338e-06, + "loss": 0.6053, + "step": 3152 + }, + { + "epoch": 2.6104362230811704, + "grad_norm": 0.9206486940383911, + "learning_rate": 3.064722778945867e-06, + "loss": 0.6197, + "step": 3153 + }, + { + "epoch": 2.6112644947542796, + "grad_norm": 0.9047932624816895, + "learning_rate": 3.063651458602708e-06, + "loss": 0.5774, + "step": 3154 + }, + { + "epoch": 2.6120927664273883, + "grad_norm": 0.9285469055175781, + "learning_rate": 3.062580029198245e-06, + "loss": 0.6107, + "step": 3155 + }, + { + "epoch": 2.612921038100497, + "grad_norm": 0.9187634587287903, + "learning_rate": 3.0615084909397906e-06, + "loss": 0.5919, + "step": 3156 + }, + { + "epoch": 2.6137493097736058, + "grad_norm": 0.8950710296630859, + "learning_rate": 3.060436844034678e-06, + "loss": 0.6029, + "step": 3157 + }, + { + "epoch": 2.6145775814467145, + "grad_norm": 0.9206116795539856, + "learning_rate": 3.0593650886902606e-06, + "loss": 0.5632, + "step": 3158 + }, + { + "epoch": 2.615405853119823, + "grad_norm": 0.9081333875656128, + "learning_rate": 3.058293225113912e-06, + "loss": 0.6127, + "step": 3159 + }, + { + "epoch": 2.616234124792932, + "grad_norm": 0.9028277397155762, + "learning_rate": 3.0572212535130297e-06, + "loss": 0.6089, + "step": 3160 + }, + { + "epoch": 2.617062396466041, + "grad_norm": 0.9449213147163391, + "learning_rate": 3.0561491740950288e-06, + "loss": 0.5841, + "step": 3161 + }, + { + "epoch": 2.6178906681391494, + "grad_norm": 0.9926438927650452, + "learning_rate": 3.055076987067347e-06, + "loss": 0.564, + "step": 3162 + }, + { + "epoch": 2.6187189398122586, + "grad_norm": 0.9098719954490662, + "learning_rate": 3.0540046926374422e-06, + "loss": 0.6268, + "step": 3163 + }, + { + "epoch": 2.6195472114853673, + "grad_norm": 0.8997418880462646, + "learning_rate": 3.0529322910127944e-06, + "loss": 0.5972, + "step": 3164 + }, + { + "epoch": 2.620375483158476, + "grad_norm": 0.9076839685440063, + "learning_rate": 3.0518597824009015e-06, + "loss": 0.5921, + "step": 3165 + }, + { + "epoch": 2.6212037548315847, + "grad_norm": 0.9348479509353638, + "learning_rate": 3.050787167009287e-06, + "loss": 0.5802, + "step": 3166 + }, + { + "epoch": 2.6220320265046935, + "grad_norm": 0.9381133913993835, + "learning_rate": 3.0497144450454886e-06, + "loss": 0.5943, + "step": 3167 + }, + { + "epoch": 2.622860298177802, + "grad_norm": 0.9327574968338013, + "learning_rate": 3.0486416167170694e-06, + "loss": 0.5832, + "step": 3168 + }, + { + "epoch": 2.623688569850911, + "grad_norm": 0.9142323732376099, + "learning_rate": 3.0475686822316126e-06, + "loss": 0.6048, + "step": 3169 + }, + { + "epoch": 2.62451684152402, + "grad_norm": 0.9799601435661316, + "learning_rate": 3.04649564179672e-06, + "loss": 0.5783, + "step": 3170 + }, + { + "epoch": 2.625345113197129, + "grad_norm": 0.9203731417655945, + "learning_rate": 3.045422495620015e-06, + "loss": 0.6028, + "step": 3171 + }, + { + "epoch": 2.6261733848702375, + "grad_norm": 0.9614686369895935, + "learning_rate": 3.0443492439091415e-06, + "loss": 0.5895, + "step": 3172 + }, + { + "epoch": 2.6270016565433463, + "grad_norm": 0.980112612247467, + "learning_rate": 3.0432758868717626e-06, + "loss": 0.5887, + "step": 3173 + }, + { + "epoch": 2.627829928216455, + "grad_norm": 0.9174870252609253, + "learning_rate": 3.042202424715565e-06, + "loss": 0.5831, + "step": 3174 + }, + { + "epoch": 2.6286581998895637, + "grad_norm": 0.9396457672119141, + "learning_rate": 3.0411288576482525e-06, + "loss": 0.5816, + "step": 3175 + }, + { + "epoch": 2.6294864715626725, + "grad_norm": 0.9067835211753845, + "learning_rate": 3.04005518587755e-06, + "loss": 0.5748, + "step": 3176 + }, + { + "epoch": 2.630314743235781, + "grad_norm": 0.9329981207847595, + "learning_rate": 3.038981409611204e-06, + "loss": 0.5901, + "step": 3177 + }, + { + "epoch": 2.63114301490889, + "grad_norm": 0.9418731927871704, + "learning_rate": 3.0379075290569793e-06, + "loss": 0.5981, + "step": 3178 + }, + { + "epoch": 2.631971286581999, + "grad_norm": 0.9180083274841309, + "learning_rate": 3.0368335444226614e-06, + "loss": 0.5653, + "step": 3179 + }, + { + "epoch": 2.632799558255108, + "grad_norm": 0.9566802978515625, + "learning_rate": 3.0357594559160575e-06, + "loss": 0.5968, + "step": 3180 + }, + { + "epoch": 2.6336278299282165, + "grad_norm": 0.9177946448326111, + "learning_rate": 3.034685263744994e-06, + "loss": 0.6076, + "step": 3181 + }, + { + "epoch": 2.6344561016013253, + "grad_norm": 0.905975341796875, + "learning_rate": 3.033610968117314e-06, + "loss": 0.6064, + "step": 3182 + }, + { + "epoch": 2.635284373274434, + "grad_norm": 0.9021571278572083, + "learning_rate": 3.0325365692408883e-06, + "loss": 0.5818, + "step": 3183 + }, + { + "epoch": 2.6361126449475427, + "grad_norm": 0.9211055040359497, + "learning_rate": 3.0314620673235995e-06, + "loss": 0.5811, + "step": 3184 + }, + { + "epoch": 2.6369409166206514, + "grad_norm": 0.929892897605896, + "learning_rate": 3.0303874625733548e-06, + "loss": 0.6039, + "step": 3185 + }, + { + "epoch": 2.6377691882937606, + "grad_norm": 0.9520539045333862, + "learning_rate": 3.0293127551980805e-06, + "loss": 0.6156, + "step": 3186 + }, + { + "epoch": 2.638597459966869, + "grad_norm": 0.9111108779907227, + "learning_rate": 3.028237945405722e-06, + "loss": 0.6081, + "step": 3187 + }, + { + "epoch": 2.639425731639978, + "grad_norm": 0.9411360025405884, + "learning_rate": 3.027163033404246e-06, + "loss": 0.5897, + "step": 3188 + }, + { + "epoch": 2.640254003313087, + "grad_norm": 0.9253203868865967, + "learning_rate": 3.026088019401637e-06, + "loss": 0.5957, + "step": 3189 + }, + { + "epoch": 2.6410822749861955, + "grad_norm": 1.010617971420288, + "learning_rate": 3.0250129036059e-06, + "loss": 0.602, + "step": 3190 + }, + { + "epoch": 2.6419105466593042, + "grad_norm": 0.9021222591400146, + "learning_rate": 3.02393768622506e-06, + "loss": 0.5796, + "step": 3191 + }, + { + "epoch": 2.642738818332413, + "grad_norm": 0.9229567050933838, + "learning_rate": 3.0228623674671627e-06, + "loss": 0.5827, + "step": 3192 + }, + { + "epoch": 2.6435670900055217, + "grad_norm": 0.9603396654129028, + "learning_rate": 3.0217869475402704e-06, + "loss": 0.5886, + "step": 3193 + }, + { + "epoch": 2.6443953616786304, + "grad_norm": 0.9483562707901001, + "learning_rate": 3.020711426652469e-06, + "loss": 0.6089, + "step": 3194 + }, + { + "epoch": 2.6452236333517396, + "grad_norm": 0.9240924715995789, + "learning_rate": 3.019635805011859e-06, + "loss": 0.6182, + "step": 3195 + }, + { + "epoch": 2.646051905024848, + "grad_norm": 0.9176941514015198, + "learning_rate": 3.018560082826566e-06, + "loss": 0.593, + "step": 3196 + }, + { + "epoch": 2.646880176697957, + "grad_norm": 0.9058155417442322, + "learning_rate": 3.0174842603047307e-06, + "loss": 0.5853, + "step": 3197 + }, + { + "epoch": 2.647708448371066, + "grad_norm": 0.9544078707695007, + "learning_rate": 3.016408337654514e-06, + "loss": 0.5743, + "step": 3198 + }, + { + "epoch": 2.6485367200441745, + "grad_norm": 0.944963812828064, + "learning_rate": 3.0153323150840987e-06, + "loss": 0.6007, + "step": 3199 + }, + { + "epoch": 2.6493649917172832, + "grad_norm": 0.9196237921714783, + "learning_rate": 3.014256192801685e-06, + "loss": 0.5818, + "step": 3200 + }, + { + "epoch": 2.650193263390392, + "grad_norm": 0.9315558075904846, + "learning_rate": 3.01317997101549e-06, + "loss": 0.5861, + "step": 3201 + }, + { + "epoch": 2.6510215350635007, + "grad_norm": 0.9272212982177734, + "learning_rate": 3.012103649933756e-06, + "loss": 0.597, + "step": 3202 + }, + { + "epoch": 2.6518498067366094, + "grad_norm": 0.9128772616386414, + "learning_rate": 3.011027229764739e-06, + "loss": 0.5863, + "step": 3203 + }, + { + "epoch": 2.6526780784097186, + "grad_norm": 0.9234123826026917, + "learning_rate": 3.0099507107167158e-06, + "loss": 0.589, + "step": 3204 + }, + { + "epoch": 2.6535063500828273, + "grad_norm": 0.9162564873695374, + "learning_rate": 3.0088740929979854e-06, + "loss": 0.5855, + "step": 3205 + }, + { + "epoch": 2.654334621755936, + "grad_norm": 0.8842553496360779, + "learning_rate": 3.0077973768168604e-06, + "loss": 0.5947, + "step": 3206 + }, + { + "epoch": 2.6551628934290448, + "grad_norm": 0.9413550496101379, + "learning_rate": 3.0067205623816775e-06, + "loss": 0.5882, + "step": 3207 + }, + { + "epoch": 2.6559911651021535, + "grad_norm": 0.9217298626899719, + "learning_rate": 3.005643649900789e-06, + "loss": 0.5806, + "step": 3208 + }, + { + "epoch": 2.656819436775262, + "grad_norm": 0.9214861392974854, + "learning_rate": 3.004566639582568e-06, + "loss": 0.5656, + "step": 3209 + }, + { + "epoch": 2.657647708448371, + "grad_norm": 0.9789690375328064, + "learning_rate": 3.003489531635405e-06, + "loss": 0.5897, + "step": 3210 + }, + { + "epoch": 2.65847598012148, + "grad_norm": 0.9465369582176208, + "learning_rate": 3.0024123262677123e-06, + "loss": 0.5996, + "step": 3211 + }, + { + "epoch": 2.6593042517945884, + "grad_norm": 0.9011247754096985, + "learning_rate": 3.001335023687917e-06, + "loss": 0.5849, + "step": 3212 + }, + { + "epoch": 2.6601325234676976, + "grad_norm": 0.9762759804725647, + "learning_rate": 3.0002576241044683e-06, + "loss": 0.6161, + "step": 3213 + }, + { + "epoch": 2.6609607951408063, + "grad_norm": 0.9362512230873108, + "learning_rate": 2.9991801277258327e-06, + "loss": 0.6186, + "step": 3214 + }, + { + "epoch": 2.661789066813915, + "grad_norm": 0.9391586780548096, + "learning_rate": 2.998102534760495e-06, + "loss": 0.5787, + "step": 3215 + }, + { + "epoch": 2.6626173384870238, + "grad_norm": 0.9361854195594788, + "learning_rate": 2.997024845416961e-06, + "loss": 0.5996, + "step": 3216 + }, + { + "epoch": 2.6634456101601325, + "grad_norm": 0.9153965711593628, + "learning_rate": 2.9959470599037525e-06, + "loss": 0.5777, + "step": 3217 + }, + { + "epoch": 2.664273881833241, + "grad_norm": 0.9155426621437073, + "learning_rate": 2.994869178429411e-06, + "loss": 0.5906, + "step": 3218 + }, + { + "epoch": 2.66510215350635, + "grad_norm": 0.8978738784790039, + "learning_rate": 2.993791201202496e-06, + "loss": 0.5707, + "step": 3219 + }, + { + "epoch": 2.665930425179459, + "grad_norm": 0.9147471785545349, + "learning_rate": 2.9927131284315874e-06, + "loss": 0.6003, + "step": 3220 + }, + { + "epoch": 2.6667586968525674, + "grad_norm": 0.9578046202659607, + "learning_rate": 2.991634960325281e-06, + "loss": 0.6068, + "step": 3221 + }, + { + "epoch": 2.6675869685256766, + "grad_norm": 0.9182425141334534, + "learning_rate": 2.9905566970921933e-06, + "loss": 0.5939, + "step": 3222 + }, + { + "epoch": 2.6684152401987853, + "grad_norm": 0.9168176651000977, + "learning_rate": 2.989478338940957e-06, + "loss": 0.5967, + "step": 3223 + }, + { + "epoch": 2.669243511871894, + "grad_norm": 0.913521409034729, + "learning_rate": 2.9883998860802253e-06, + "loss": 0.6081, + "step": 3224 + }, + { + "epoch": 2.6700717835450027, + "grad_norm": 0.9223765134811401, + "learning_rate": 2.9873213387186684e-06, + "loss": 0.5901, + "step": 3225 + }, + { + "epoch": 2.6709000552181115, + "grad_norm": 0.9228116273880005, + "learning_rate": 2.9862426970649743e-06, + "loss": 0.5959, + "step": 3226 + }, + { + "epoch": 2.67172832689122, + "grad_norm": 0.9379467368125916, + "learning_rate": 2.985163961327852e-06, + "loss": 0.5986, + "step": 3227 + }, + { + "epoch": 2.672556598564329, + "grad_norm": 0.921449601650238, + "learning_rate": 2.984085131716027e-06, + "loss": 0.5587, + "step": 3228 + }, + { + "epoch": 2.673384870237438, + "grad_norm": 0.9038673043251038, + "learning_rate": 2.9830062084382393e-06, + "loss": 0.5908, + "step": 3229 + }, + { + "epoch": 2.674213141910547, + "grad_norm": 0.9072180986404419, + "learning_rate": 2.981927191703254e-06, + "loss": 0.6064, + "step": 3230 + }, + { + "epoch": 2.6750414135836555, + "grad_norm": 0.9178011417388916, + "learning_rate": 2.9808480817198494e-06, + "loss": 0.5969, + "step": 3231 + }, + { + "epoch": 2.6758696852567643, + "grad_norm": 0.9064451456069946, + "learning_rate": 2.9797688786968233e-06, + "loss": 0.571, + "step": 3232 + }, + { + "epoch": 2.676697956929873, + "grad_norm": 0.9392306804656982, + "learning_rate": 2.9786895828429914e-06, + "loss": 0.5903, + "step": 3233 + }, + { + "epoch": 2.6775262286029817, + "grad_norm": 0.9398954510688782, + "learning_rate": 2.9776101943671875e-06, + "loss": 0.6163, + "step": 3234 + }, + { + "epoch": 2.6783545002760905, + "grad_norm": 0.9082908630371094, + "learning_rate": 2.9765307134782637e-06, + "loss": 0.5838, + "step": 3235 + }, + { + "epoch": 2.679182771949199, + "grad_norm": 0.926916778087616, + "learning_rate": 2.975451140385089e-06, + "loss": 0.5599, + "step": 3236 + }, + { + "epoch": 2.680011043622308, + "grad_norm": 0.9106405377388, + "learning_rate": 2.97437147529655e-06, + "loss": 0.6012, + "step": 3237 + }, + { + "epoch": 2.680839315295417, + "grad_norm": 0.9428172707557678, + "learning_rate": 2.9732917184215524e-06, + "loss": 0.5985, + "step": 3238 + }, + { + "epoch": 2.681667586968526, + "grad_norm": 0.9456861019134521, + "learning_rate": 2.9722118699690212e-06, + "loss": 0.6206, + "step": 3239 + }, + { + "epoch": 2.6824958586416345, + "grad_norm": 0.9116294384002686, + "learning_rate": 2.971131930147893e-06, + "loss": 0.6166, + "step": 3240 + }, + { + "epoch": 2.6833241303147433, + "grad_norm": 0.9061588644981384, + "learning_rate": 2.970051899167129e-06, + "loss": 0.5792, + "step": 3241 + }, + { + "epoch": 2.684152401987852, + "grad_norm": 0.947784423828125, + "learning_rate": 2.968971777235704e-06, + "loss": 0.5813, + "step": 3242 + }, + { + "epoch": 2.6849806736609607, + "grad_norm": 0.9296494722366333, + "learning_rate": 2.9678915645626115e-06, + "loss": 0.5916, + "step": 3243 + }, + { + "epoch": 2.6858089453340694, + "grad_norm": 0.9151764512062073, + "learning_rate": 2.9668112613568632e-06, + "loss": 0.6097, + "step": 3244 + }, + { + "epoch": 2.6866372170071786, + "grad_norm": 0.9887247681617737, + "learning_rate": 2.965730867827488e-06, + "loss": 0.5951, + "step": 3245 + }, + { + "epoch": 2.687465488680287, + "grad_norm": 0.9406042695045471, + "learning_rate": 2.964650384183531e-06, + "loss": 0.5918, + "step": 3246 + }, + { + "epoch": 2.688293760353396, + "grad_norm": 0.9440297484397888, + "learning_rate": 2.963569810634056e-06, + "loss": 0.6098, + "step": 3247 + }, + { + "epoch": 2.689122032026505, + "grad_norm": 0.93993079662323, + "learning_rate": 2.962489147388144e-06, + "loss": 0.6242, + "step": 3248 + }, + { + "epoch": 2.6899503036996135, + "grad_norm": 0.9704070091247559, + "learning_rate": 2.9614083946548927e-06, + "loss": 0.5764, + "step": 3249 + }, + { + "epoch": 2.6907785753727222, + "grad_norm": 0.9809379577636719, + "learning_rate": 2.9603275526434188e-06, + "loss": 0.5997, + "step": 3250 + }, + { + "epoch": 2.691606847045831, + "grad_norm": 0.9246951937675476, + "learning_rate": 2.9592466215628547e-06, + "loss": 0.577, + "step": 3251 + }, + { + "epoch": 2.6924351187189397, + "grad_norm": 0.9204201698303223, + "learning_rate": 2.95816560162235e-06, + "loss": 0.6036, + "step": 3252 + }, + { + "epoch": 2.6932633903920484, + "grad_norm": 0.9180527925491333, + "learning_rate": 2.9570844930310734e-06, + "loss": 0.5919, + "step": 3253 + }, + { + "epoch": 2.6940916620651576, + "grad_norm": 0.9165668487548828, + "learning_rate": 2.9560032959982067e-06, + "loss": 0.5646, + "step": 3254 + }, + { + "epoch": 2.694919933738266, + "grad_norm": 0.9150571823120117, + "learning_rate": 2.9549220107329536e-06, + "loss": 0.581, + "step": 3255 + }, + { + "epoch": 2.695748205411375, + "grad_norm": 0.9429931044578552, + "learning_rate": 2.9538406374445334e-06, + "loss": 0.595, + "step": 3256 + }, + { + "epoch": 2.696576477084484, + "grad_norm": 0.939118504524231, + "learning_rate": 2.9527591763421785e-06, + "loss": 0.5772, + "step": 3257 + }, + { + "epoch": 2.6974047487575925, + "grad_norm": 0.9130142331123352, + "learning_rate": 2.9516776276351446e-06, + "loss": 0.6031, + "step": 3258 + }, + { + "epoch": 2.6982330204307012, + "grad_norm": 0.9334836006164551, + "learning_rate": 2.9505959915327004e-06, + "loss": 0.5899, + "step": 3259 + }, + { + "epoch": 2.69906129210381, + "grad_norm": 0.9407964944839478, + "learning_rate": 2.9495142682441314e-06, + "loss": 0.6232, + "step": 3260 + }, + { + "epoch": 2.6998895637769187, + "grad_norm": 0.9573985934257507, + "learning_rate": 2.9484324579787427e-06, + "loss": 0.5932, + "step": 3261 + }, + { + "epoch": 2.7007178354500274, + "grad_norm": 0.9223983287811279, + "learning_rate": 2.947350560945853e-06, + "loss": 0.5861, + "step": 3262 + }, + { + "epoch": 2.7015461071231366, + "grad_norm": 0.9206790328025818, + "learning_rate": 2.9462685773548004e-06, + "loss": 0.5867, + "step": 3263 + }, + { + "epoch": 2.7023743787962453, + "grad_norm": 0.9197454452514648, + "learning_rate": 2.9451865074149378e-06, + "loss": 0.5867, + "step": 3264 + }, + { + "epoch": 2.703202650469354, + "grad_norm": 0.9365606904029846, + "learning_rate": 2.944104351335636e-06, + "loss": 0.6106, + "step": 3265 + }, + { + "epoch": 2.7040309221424628, + "grad_norm": 0.9116243720054626, + "learning_rate": 2.9430221093262823e-06, + "loss": 0.5904, + "step": 3266 + }, + { + "epoch": 2.7048591938155715, + "grad_norm": 0.9128034114837646, + "learning_rate": 2.9419397815962802e-06, + "loss": 0.6055, + "step": 3267 + }, + { + "epoch": 2.70568746548868, + "grad_norm": 0.9147705435752869, + "learning_rate": 2.94085736835505e-06, + "loss": 0.5941, + "step": 3268 + }, + { + "epoch": 2.706515737161789, + "grad_norm": 0.9419110417366028, + "learning_rate": 2.939774869812029e-06, + "loss": 0.5595, + "step": 3269 + }, + { + "epoch": 2.707344008834898, + "grad_norm": 0.9202789664268494, + "learning_rate": 2.93869228617667e-06, + "loss": 0.5788, + "step": 3270 + }, + { + "epoch": 2.7081722805080064, + "grad_norm": 0.9068663716316223, + "learning_rate": 2.9376096176584424e-06, + "loss": 0.5749, + "step": 3271 + }, + { + "epoch": 2.7090005521811156, + "grad_norm": 0.923497200012207, + "learning_rate": 2.936526864466833e-06, + "loss": 0.6029, + "step": 3272 + }, + { + "epoch": 2.7098288238542243, + "grad_norm": 0.9812871217727661, + "learning_rate": 2.935444026811346e-06, + "loss": 0.6193, + "step": 3273 + }, + { + "epoch": 2.710657095527333, + "grad_norm": 0.9538823366165161, + "learning_rate": 2.934361104901498e-06, + "loss": 0.5809, + "step": 3274 + }, + { + "epoch": 2.7114853672004418, + "grad_norm": 0.9111268520355225, + "learning_rate": 2.9332780989468256e-06, + "loss": 0.6181, + "step": 3275 + }, + { + "epoch": 2.7123136388735505, + "grad_norm": 0.9485889673233032, + "learning_rate": 2.9321950091568795e-06, + "loss": 0.5823, + "step": 3276 + }, + { + "epoch": 2.713141910546659, + "grad_norm": 0.9233952760696411, + "learning_rate": 2.931111835741229e-06, + "loss": 0.6008, + "step": 3277 + }, + { + "epoch": 2.713970182219768, + "grad_norm": 0.9075472950935364, + "learning_rate": 2.9300285789094563e-06, + "loss": 0.5808, + "step": 3278 + }, + { + "epoch": 2.714798453892877, + "grad_norm": 0.9238584041595459, + "learning_rate": 2.9289452388711624e-06, + "loss": 0.6095, + "step": 3279 + }, + { + "epoch": 2.7156267255659854, + "grad_norm": 0.889460027217865, + "learning_rate": 2.9278618158359635e-06, + "loss": 0.6015, + "step": 3280 + }, + { + "epoch": 2.7164549972390946, + "grad_norm": 0.9549591541290283, + "learning_rate": 2.9267783100134915e-06, + "loss": 0.6029, + "step": 3281 + }, + { + "epoch": 2.7172832689122033, + "grad_norm": 0.9522138237953186, + "learning_rate": 2.9256947216133946e-06, + "loss": 0.5998, + "step": 3282 + }, + { + "epoch": 2.718111540585312, + "grad_norm": 0.8992515802383423, + "learning_rate": 2.924611050845338e-06, + "loss": 0.6012, + "step": 3283 + }, + { + "epoch": 2.7189398122584207, + "grad_norm": 0.9058646559715271, + "learning_rate": 2.9235272979190014e-06, + "loss": 0.6092, + "step": 3284 + }, + { + "epoch": 2.7197680839315295, + "grad_norm": 0.916130542755127, + "learning_rate": 2.92244346304408e-06, + "loss": 0.6054, + "step": 3285 + }, + { + "epoch": 2.720596355604638, + "grad_norm": 0.9203442335128784, + "learning_rate": 2.921359546430287e-06, + "loss": 0.596, + "step": 3286 + }, + { + "epoch": 2.721424627277747, + "grad_norm": 0.9603051543235779, + "learning_rate": 2.920275548287349e-06, + "loss": 0.5976, + "step": 3287 + }, + { + "epoch": 2.722252898950856, + "grad_norm": 0.9188669919967651, + "learning_rate": 2.91919146882501e-06, + "loss": 0.5812, + "step": 3288 + }, + { + "epoch": 2.723081170623965, + "grad_norm": 0.9154419898986816, + "learning_rate": 2.9181073082530302e-06, + "loss": 0.6127, + "step": 3289 + }, + { + "epoch": 2.7239094422970735, + "grad_norm": 0.9228849411010742, + "learning_rate": 2.917023066781184e-06, + "loss": 0.6223, + "step": 3290 + }, + { + "epoch": 2.7247377139701823, + "grad_norm": 0.9034006595611572, + "learning_rate": 2.915938744619261e-06, + "loss": 0.5981, + "step": 3291 + }, + { + "epoch": 2.725565985643291, + "grad_norm": 0.9119758009910583, + "learning_rate": 2.914854341977069e-06, + "loss": 0.5974, + "step": 3292 + }, + { + "epoch": 2.7263942573163997, + "grad_norm": 0.9350270628929138, + "learning_rate": 2.9137698590644283e-06, + "loss": 0.5958, + "step": 3293 + }, + { + "epoch": 2.7272225289895085, + "grad_norm": 0.9229425191879272, + "learning_rate": 2.9126852960911773e-06, + "loss": 0.5873, + "step": 3294 + }, + { + "epoch": 2.728050800662617, + "grad_norm": 0.9276396632194519, + "learning_rate": 2.9116006532671685e-06, + "loss": 0.6069, + "step": 3295 + }, + { + "epoch": 2.728879072335726, + "grad_norm": 0.9154210090637207, + "learning_rate": 2.9105159308022705e-06, + "loss": 0.5741, + "step": 3296 + }, + { + "epoch": 2.729707344008835, + "grad_norm": 0.9545791149139404, + "learning_rate": 2.909431128906367e-06, + "loss": 0.5995, + "step": 3297 + }, + { + "epoch": 2.730535615681944, + "grad_norm": 0.9200988411903381, + "learning_rate": 2.908346247789356e-06, + "loss": 0.6152, + "step": 3298 + }, + { + "epoch": 2.7313638873550525, + "grad_norm": 0.9288673996925354, + "learning_rate": 2.9072612876611524e-06, + "loss": 0.5904, + "step": 3299 + }, + { + "epoch": 2.7321921590281613, + "grad_norm": 0.918323278427124, + "learning_rate": 2.906176248731686e-06, + "loss": 0.5999, + "step": 3300 + }, + { + "epoch": 2.73302043070127, + "grad_norm": 0.936652421951294, + "learning_rate": 2.905091131210903e-06, + "loss": 0.5928, + "step": 3301 + }, + { + "epoch": 2.7338487023743787, + "grad_norm": 0.8942639827728271, + "learning_rate": 2.9040059353087614e-06, + "loss": 0.5739, + "step": 3302 + }, + { + "epoch": 2.7346769740474874, + "grad_norm": 0.9197317957878113, + "learning_rate": 2.902920661235238e-06, + "loss": 0.5776, + "step": 3303 + }, + { + "epoch": 2.7355052457205966, + "grad_norm": 0.895505428314209, + "learning_rate": 2.9018353092003216e-06, + "loss": 0.6067, + "step": 3304 + }, + { + "epoch": 2.736333517393705, + "grad_norm": 0.947441816329956, + "learning_rate": 2.9007498794140193e-06, + "loss": 0.6072, + "step": 3305 + }, + { + "epoch": 2.737161789066814, + "grad_norm": 0.9008322358131409, + "learning_rate": 2.8996643720863515e-06, + "loss": 0.5908, + "step": 3306 + }, + { + "epoch": 2.737990060739923, + "grad_norm": 0.9137061238288879, + "learning_rate": 2.898578787427353e-06, + "loss": 0.607, + "step": 3307 + }, + { + "epoch": 2.7388183324130315, + "grad_norm": 0.9126008749008179, + "learning_rate": 2.8974931256470745e-06, + "loss": 0.5809, + "step": 3308 + }, + { + "epoch": 2.7396466040861402, + "grad_norm": 0.9215322136878967, + "learning_rate": 2.896407386955582e-06, + "loss": 0.5807, + "step": 3309 + }, + { + "epoch": 2.740474875759249, + "grad_norm": 0.8903998136520386, + "learning_rate": 2.8953215715629547e-06, + "loss": 0.5968, + "step": 3310 + }, + { + "epoch": 2.7413031474323577, + "grad_norm": 0.9156417846679688, + "learning_rate": 2.8942356796792888e-06, + "loss": 0.5809, + "step": 3311 + }, + { + "epoch": 2.7421314191054664, + "grad_norm": 0.952264666557312, + "learning_rate": 2.893149711514694e-06, + "loss": 0.5818, + "step": 3312 + }, + { + "epoch": 2.7429596907785756, + "grad_norm": 0.9201667904853821, + "learning_rate": 2.8920636672792945e-06, + "loss": 0.5789, + "step": 3313 + }, + { + "epoch": 2.743787962451684, + "grad_norm": 0.9581197500228882, + "learning_rate": 2.890977547183231e-06, + "loss": 0.6007, + "step": 3314 + }, + { + "epoch": 2.744616234124793, + "grad_norm": 0.9360357522964478, + "learning_rate": 2.889891351436656e-06, + "loss": 0.6096, + "step": 3315 + }, + { + "epoch": 2.745444505797902, + "grad_norm": 0.9209451079368591, + "learning_rate": 2.8888050802497388e-06, + "loss": 0.6015, + "step": 3316 + }, + { + "epoch": 2.7462727774710105, + "grad_norm": 0.9496802091598511, + "learning_rate": 2.8877187338326634e-06, + "loss": 0.5802, + "step": 3317 + }, + { + "epoch": 2.7471010491441192, + "grad_norm": 0.9407662153244019, + "learning_rate": 2.8866323123956274e-06, + "loss": 0.588, + "step": 3318 + }, + { + "epoch": 2.747929320817228, + "grad_norm": 0.9603942036628723, + "learning_rate": 2.885545816148843e-06, + "loss": 0.5759, + "step": 3319 + }, + { + "epoch": 2.7487575924903367, + "grad_norm": 0.930496335029602, + "learning_rate": 2.8844592453025372e-06, + "loss": 0.5916, + "step": 3320 + }, + { + "epoch": 2.7495858641634454, + "grad_norm": 0.8831289410591125, + "learning_rate": 2.883372600066951e-06, + "loss": 0.5889, + "step": 3321 + }, + { + "epoch": 2.7504141358365546, + "grad_norm": 0.9076021909713745, + "learning_rate": 2.8822858806523407e-06, + "loss": 0.5842, + "step": 3322 + }, + { + "epoch": 2.7512424075096633, + "grad_norm": 0.9093723297119141, + "learning_rate": 2.8811990872689764e-06, + "loss": 0.584, + "step": 3323 + }, + { + "epoch": 2.752070679182772, + "grad_norm": 0.9458228349685669, + "learning_rate": 2.880112220127142e-06, + "loss": 0.5812, + "step": 3324 + }, + { + "epoch": 2.7528989508558808, + "grad_norm": 0.9406691789627075, + "learning_rate": 2.8790252794371365e-06, + "loss": 0.5889, + "step": 3325 + }, + { + "epoch": 2.7537272225289895, + "grad_norm": 0.9094856381416321, + "learning_rate": 2.8779382654092726e-06, + "loss": 0.5919, + "step": 3326 + }, + { + "epoch": 2.754555494202098, + "grad_norm": 0.9350448250770569, + "learning_rate": 2.8768511782538766e-06, + "loss": 0.5738, + "step": 3327 + }, + { + "epoch": 2.755383765875207, + "grad_norm": 0.9427900314331055, + "learning_rate": 2.875764018181291e-06, + "loss": 0.5962, + "step": 3328 + }, + { + "epoch": 2.756212037548316, + "grad_norm": 0.9439454674720764, + "learning_rate": 2.8746767854018715e-06, + "loss": 0.5994, + "step": 3329 + }, + { + "epoch": 2.7570403092214244, + "grad_norm": 0.9291850328445435, + "learning_rate": 2.8735894801259863e-06, + "loss": 0.6055, + "step": 3330 + }, + { + "epoch": 2.7578685808945336, + "grad_norm": 0.9287533164024353, + "learning_rate": 2.872502102564019e-06, + "loss": 0.5798, + "step": 3331 + }, + { + "epoch": 2.7586968525676423, + "grad_norm": 0.8941923379898071, + "learning_rate": 2.871414652926367e-06, + "loss": 0.6066, + "step": 3332 + }, + { + "epoch": 2.759525124240751, + "grad_norm": 0.924333930015564, + "learning_rate": 2.8703271314234425e-06, + "loss": 0.5872, + "step": 3333 + }, + { + "epoch": 2.7603533959138598, + "grad_norm": 0.8994147777557373, + "learning_rate": 2.8692395382656702e-06, + "loss": 0.5906, + "step": 3334 + }, + { + "epoch": 2.7611816675869685, + "grad_norm": 0.9152485132217407, + "learning_rate": 2.8681518736634895e-06, + "loss": 0.5774, + "step": 3335 + }, + { + "epoch": 2.762009939260077, + "grad_norm": 0.9269137978553772, + "learning_rate": 2.867064137827353e-06, + "loss": 0.5759, + "step": 3336 + }, + { + "epoch": 2.762838210933186, + "grad_norm": 0.916441023349762, + "learning_rate": 2.8659763309677267e-06, + "loss": 0.5927, + "step": 3337 + }, + { + "epoch": 2.763666482606295, + "grad_norm": 0.9280357956886292, + "learning_rate": 2.8648884532950925e-06, + "loss": 0.6004, + "step": 3338 + }, + { + "epoch": 2.7644947542794034, + "grad_norm": 0.9075552821159363, + "learning_rate": 2.8638005050199443e-06, + "loss": 0.6185, + "step": 3339 + }, + { + "epoch": 2.7653230259525126, + "grad_norm": 0.9340760707855225, + "learning_rate": 2.8627124863527895e-06, + "loss": 0.6119, + "step": 3340 + }, + { + "epoch": 2.7661512976256213, + "grad_norm": 0.9393087029457092, + "learning_rate": 2.8616243975041495e-06, + "loss": 0.6058, + "step": 3341 + }, + { + "epoch": 2.76697956929873, + "grad_norm": 0.9380460381507874, + "learning_rate": 2.8605362386845605e-06, + "loss": 0.5955, + "step": 3342 + }, + { + "epoch": 2.7678078409718387, + "grad_norm": 0.9393378496170044, + "learning_rate": 2.8594480101045683e-06, + "loss": 0.5841, + "step": 3343 + }, + { + "epoch": 2.7686361126449475, + "grad_norm": 0.9186044335365295, + "learning_rate": 2.8583597119747385e-06, + "loss": 0.5977, + "step": 3344 + }, + { + "epoch": 2.769464384318056, + "grad_norm": 0.9224127531051636, + "learning_rate": 2.8572713445056448e-06, + "loss": 0.5863, + "step": 3345 + }, + { + "epoch": 2.770292655991165, + "grad_norm": 0.9517519474029541, + "learning_rate": 2.8561829079078764e-06, + "loss": 0.5748, + "step": 3346 + }, + { + "epoch": 2.771120927664274, + "grad_norm": 0.9624918103218079, + "learning_rate": 2.855094402392036e-06, + "loss": 0.588, + "step": 3347 + }, + { + "epoch": 2.771949199337383, + "grad_norm": 0.9524730443954468, + "learning_rate": 2.8540058281687375e-06, + "loss": 0.6032, + "step": 3348 + }, + { + "epoch": 2.7727774710104915, + "grad_norm": 0.9574846029281616, + "learning_rate": 2.852917185448612e-06, + "loss": 0.5915, + "step": 3349 + }, + { + "epoch": 2.7736057426836003, + "grad_norm": 0.9290011525154114, + "learning_rate": 2.851828474442301e-06, + "loss": 0.5715, + "step": 3350 + }, + { + "epoch": 2.774434014356709, + "grad_norm": 0.9393135905265808, + "learning_rate": 2.8507396953604597e-06, + "loss": 0.5855, + "step": 3351 + }, + { + "epoch": 2.7752622860298177, + "grad_norm": 0.9188078045845032, + "learning_rate": 2.8496508484137574e-06, + "loss": 0.598, + "step": 3352 + }, + { + "epoch": 2.7760905577029265, + "grad_norm": 0.9129231572151184, + "learning_rate": 2.8485619338128755e-06, + "loss": 0.6076, + "step": 3353 + }, + { + "epoch": 2.776918829376035, + "grad_norm": 0.941085934638977, + "learning_rate": 2.847472951768508e-06, + "loss": 0.5787, + "step": 3354 + }, + { + "epoch": 2.777747101049144, + "grad_norm": 1.0103200674057007, + "learning_rate": 2.846383902491363e-06, + "loss": 0.606, + "step": 3355 + }, + { + "epoch": 2.778575372722253, + "grad_norm": 0.9307713508605957, + "learning_rate": 2.845294786192163e-06, + "loss": 0.5969, + "step": 3356 + }, + { + "epoch": 2.779403644395362, + "grad_norm": 0.890837550163269, + "learning_rate": 2.84420560308164e-06, + "loss": 0.5826, + "step": 3357 + }, + { + "epoch": 2.7802319160684705, + "grad_norm": 0.8857460618019104, + "learning_rate": 2.843116353370542e-06, + "loss": 0.6005, + "step": 3358 + }, + { + "epoch": 2.7810601877415793, + "grad_norm": 0.8894275426864624, + "learning_rate": 2.8420270372696273e-06, + "loss": 0.6017, + "step": 3359 + }, + { + "epoch": 2.781888459414688, + "grad_norm": 0.9014497995376587, + "learning_rate": 2.8409376549896693e-06, + "loss": 0.5845, + "step": 3360 + }, + { + "epoch": 2.7827167310877967, + "grad_norm": 0.9019520282745361, + "learning_rate": 2.839848206741454e-06, + "loss": 0.5993, + "step": 3361 + }, + { + "epoch": 2.7835450027609054, + "grad_norm": 0.9594439268112183, + "learning_rate": 2.838758692735778e-06, + "loss": 0.5905, + "step": 3362 + }, + { + "epoch": 2.7843732744340146, + "grad_norm": 0.9144065976142883, + "learning_rate": 2.8376691131834525e-06, + "loss": 0.6006, + "step": 3363 + }, + { + "epoch": 2.785201546107123, + "grad_norm": 0.9487935900688171, + "learning_rate": 2.836579468295303e-06, + "loss": 0.5832, + "step": 3364 + }, + { + "epoch": 2.786029817780232, + "grad_norm": 0.8950932025909424, + "learning_rate": 2.835489758282163e-06, + "loss": 0.5792, + "step": 3365 + }, + { + "epoch": 2.786858089453341, + "grad_norm": 0.9522806406021118, + "learning_rate": 2.8343999833548814e-06, + "loss": 0.5891, + "step": 3366 + }, + { + "epoch": 2.7876863611264495, + "grad_norm": 0.964898407459259, + "learning_rate": 2.833310143724321e-06, + "loss": 0.572, + "step": 3367 + }, + { + "epoch": 2.7885146327995582, + "grad_norm": 0.9154083728790283, + "learning_rate": 2.832220239601355e-06, + "loss": 0.5982, + "step": 3368 + }, + { + "epoch": 2.789342904472667, + "grad_norm": 0.9093793034553528, + "learning_rate": 2.831130271196869e-06, + "loss": 0.5592, + "step": 3369 + }, + { + "epoch": 2.7901711761457757, + "grad_norm": 0.9114746451377869, + "learning_rate": 2.8300402387217628e-06, + "loss": 0.5531, + "step": 3370 + }, + { + "epoch": 2.7909994478188844, + "grad_norm": 0.959976077079773, + "learning_rate": 2.8289501423869464e-06, + "loss": 0.5932, + "step": 3371 + }, + { + "epoch": 2.7918277194919936, + "grad_norm": 0.9415104389190674, + "learning_rate": 2.8278599824033443e-06, + "loss": 0.5516, + "step": 3372 + }, + { + "epoch": 2.792655991165102, + "grad_norm": 0.9304796457290649, + "learning_rate": 2.826769758981892e-06, + "loss": 0.5786, + "step": 3373 + }, + { + "epoch": 2.793484262838211, + "grad_norm": 0.9369680285453796, + "learning_rate": 2.825679472333537e-06, + "loss": 0.6077, + "step": 3374 + }, + { + "epoch": 2.79431253451132, + "grad_norm": 0.9102757573127747, + "learning_rate": 2.8245891226692405e-06, + "loss": 0.606, + "step": 3375 + }, + { + "epoch": 2.7951408061844285, + "grad_norm": 0.921359121799469, + "learning_rate": 2.8234987101999747e-06, + "loss": 0.613, + "step": 3376 + }, + { + "epoch": 2.7959690778575372, + "grad_norm": 0.9242355227470398, + "learning_rate": 2.822408235136723e-06, + "loss": 0.5758, + "step": 3377 + }, + { + "epoch": 2.796797349530646, + "grad_norm": 0.9175317287445068, + "learning_rate": 2.8213176976904844e-06, + "loss": 0.5744, + "step": 3378 + }, + { + "epoch": 2.7976256212037547, + "grad_norm": 0.9504632949829102, + "learning_rate": 2.8202270980722663e-06, + "loss": 0.5871, + "step": 3379 + }, + { + "epoch": 2.7984538928768634, + "grad_norm": 0.9071774482727051, + "learning_rate": 2.8191364364930895e-06, + "loss": 0.6003, + "step": 3380 + }, + { + "epoch": 2.7992821645499726, + "grad_norm": 0.913815975189209, + "learning_rate": 2.8180457131639878e-06, + "loss": 0.596, + "step": 3381 + }, + { + "epoch": 2.8001104362230813, + "grad_norm": 0.9608022570610046, + "learning_rate": 2.8169549282960044e-06, + "loss": 0.6055, + "step": 3382 + }, + { + "epoch": 2.80093870789619, + "grad_norm": 2.0446290969848633, + "learning_rate": 2.815864082100198e-06, + "loss": 0.565, + "step": 3383 + }, + { + "epoch": 2.8017669795692988, + "grad_norm": 0.9219574928283691, + "learning_rate": 2.814773174787636e-06, + "loss": 0.5808, + "step": 3384 + }, + { + "epoch": 2.8025952512424075, + "grad_norm": 0.9542830586433411, + "learning_rate": 2.8136822065693987e-06, + "loss": 0.5992, + "step": 3385 + }, + { + "epoch": 2.803423522915516, + "grad_norm": 0.9478740096092224, + "learning_rate": 2.8125911776565785e-06, + "loss": 0.5705, + "step": 3386 + }, + { + "epoch": 2.804251794588625, + "grad_norm": 0.9299196600914001, + "learning_rate": 2.8115000882602792e-06, + "loss": 0.5773, + "step": 3387 + }, + { + "epoch": 2.805080066261734, + "grad_norm": 0.9167691469192505, + "learning_rate": 2.8104089385916168e-06, + "loss": 0.5789, + "step": 3388 + }, + { + "epoch": 2.8059083379348424, + "grad_norm": 0.9034630656242371, + "learning_rate": 2.8093177288617187e-06, + "loss": 0.5837, + "step": 3389 + }, + { + "epoch": 2.8067366096079516, + "grad_norm": 0.9147744178771973, + "learning_rate": 2.808226459281723e-06, + "loss": 0.5826, + "step": 3390 + }, + { + "epoch": 2.8075648812810603, + "grad_norm": 0.9159218072891235, + "learning_rate": 2.8071351300627803e-06, + "loss": 0.5931, + "step": 3391 + }, + { + "epoch": 2.808393152954169, + "grad_norm": 0.9260812401771545, + "learning_rate": 2.8060437414160546e-06, + "loss": 0.6182, + "step": 3392 + }, + { + "epoch": 2.8092214246272778, + "grad_norm": 0.8975473642349243, + "learning_rate": 2.804952293552717e-06, + "loss": 0.5786, + "step": 3393 + }, + { + "epoch": 2.8100496963003865, + "grad_norm": 0.9130522608757019, + "learning_rate": 2.8038607866839522e-06, + "loss": 0.572, + "step": 3394 + }, + { + "epoch": 2.810877967973495, + "grad_norm": 0.9170857667922974, + "learning_rate": 2.802769221020959e-06, + "loss": 0.569, + "step": 3395 + }, + { + "epoch": 2.811706239646604, + "grad_norm": 0.9313328266143799, + "learning_rate": 2.801677596774944e-06, + "loss": 0.568, + "step": 3396 + }, + { + "epoch": 2.812534511319713, + "grad_norm": 0.941814124584198, + "learning_rate": 2.8005859141571257e-06, + "loss": 0.5852, + "step": 3397 + }, + { + "epoch": 2.8133627829928214, + "grad_norm": 0.9179267883300781, + "learning_rate": 2.799494173378735e-06, + "loss": 0.5887, + "step": 3398 + }, + { + "epoch": 2.8141910546659306, + "grad_norm": 0.9193063378334045, + "learning_rate": 2.7984023746510135e-06, + "loss": 0.588, + "step": 3399 + }, + { + "epoch": 2.8150193263390393, + "grad_norm": 0.9381787776947021, + "learning_rate": 2.7973105181852145e-06, + "loss": 0.5938, + "step": 3400 + }, + { + "epoch": 2.815847598012148, + "grad_norm": 0.9631656408309937, + "learning_rate": 2.7962186041926015e-06, + "loss": 0.5828, + "step": 3401 + }, + { + "epoch": 2.8166758696852567, + "grad_norm": 0.9512215852737427, + "learning_rate": 2.7951266328844495e-06, + "loss": 0.5836, + "step": 3402 + }, + { + "epoch": 2.8175041413583655, + "grad_norm": 0.9171256422996521, + "learning_rate": 2.794034604472047e-06, + "loss": 0.574, + "step": 3403 + }, + { + "epoch": 2.818332413031474, + "grad_norm": 0.9122748374938965, + "learning_rate": 2.792942519166688e-06, + "loss": 0.6024, + "step": 3404 + }, + { + "epoch": 2.819160684704583, + "grad_norm": 0.9503622651100159, + "learning_rate": 2.7918503771796817e-06, + "loss": 0.5743, + "step": 3405 + }, + { + "epoch": 2.819988956377692, + "grad_norm": 0.9348991513252258, + "learning_rate": 2.7907581787223494e-06, + "loss": 0.5824, + "step": 3406 + }, + { + "epoch": 2.820817228050801, + "grad_norm": 1.0846208333969116, + "learning_rate": 2.7896659240060192e-06, + "loss": 0.5995, + "step": 3407 + }, + { + "epoch": 2.8216454997239095, + "grad_norm": 0.9554671049118042, + "learning_rate": 2.7885736132420326e-06, + "loss": 0.5903, + "step": 3408 + }, + { + "epoch": 2.8224737713970183, + "grad_norm": 0.9836533665657043, + "learning_rate": 2.7874812466417434e-06, + "loss": 0.5982, + "step": 3409 + }, + { + "epoch": 2.823302043070127, + "grad_norm": 0.9414785504341125, + "learning_rate": 2.786388824416511e-06, + "loss": 0.5743, + "step": 3410 + }, + { + "epoch": 2.8241303147432357, + "grad_norm": 0.9228699207305908, + "learning_rate": 2.7852963467777117e-06, + "loss": 0.5879, + "step": 3411 + }, + { + "epoch": 2.8249585864163445, + "grad_norm": 0.919826090335846, + "learning_rate": 2.7842038139367288e-06, + "loss": 0.6023, + "step": 3412 + }, + { + "epoch": 2.825786858089453, + "grad_norm": 0.9112126231193542, + "learning_rate": 2.783111226104957e-06, + "loss": 0.578, + "step": 3413 + }, + { + "epoch": 2.826615129762562, + "grad_norm": 0.9262592792510986, + "learning_rate": 2.7820185834938024e-06, + "loss": 0.6063, + "step": 3414 + }, + { + "epoch": 2.827443401435671, + "grad_norm": 0.9020029902458191, + "learning_rate": 2.780925886314681e-06, + "loss": 0.5837, + "step": 3415 + }, + { + "epoch": 2.82827167310878, + "grad_norm": 0.9186374545097351, + "learning_rate": 2.7798331347790185e-06, + "loss": 0.5801, + "step": 3416 + }, + { + "epoch": 2.8290999447818885, + "grad_norm": 0.9116926193237305, + "learning_rate": 2.7787403290982535e-06, + "loss": 0.5735, + "step": 3417 + }, + { + "epoch": 2.8299282164549973, + "grad_norm": 0.9124463796615601, + "learning_rate": 2.7776474694838333e-06, + "loss": 0.6064, + "step": 3418 + }, + { + "epoch": 2.830756488128106, + "grad_norm": 0.9450061321258545, + "learning_rate": 2.7765545561472153e-06, + "loss": 0.5885, + "step": 3419 + }, + { + "epoch": 2.8315847598012147, + "grad_norm": 0.8993310928344727, + "learning_rate": 2.77546158929987e-06, + "loss": 0.5767, + "step": 3420 + }, + { + "epoch": 2.8324130314743234, + "grad_norm": 0.9315094947814941, + "learning_rate": 2.7743685691532734e-06, + "loss": 0.595, + "step": 3421 + }, + { + "epoch": 2.8332413031474326, + "grad_norm": 0.9254942536354065, + "learning_rate": 2.773275495918917e-06, + "loss": 0.6071, + "step": 3422 + }, + { + "epoch": 2.834069574820541, + "grad_norm": 0.9551193118095398, + "learning_rate": 2.7721823698083e-06, + "loss": 0.588, + "step": 3423 + }, + { + "epoch": 2.83489784649365, + "grad_norm": 0.9356088042259216, + "learning_rate": 2.7710891910329317e-06, + "loss": 0.5917, + "step": 3424 + }, + { + "epoch": 2.835726118166759, + "grad_norm": 0.9254491925239563, + "learning_rate": 2.7699959598043315e-06, + "loss": 0.5754, + "step": 3425 + }, + { + "epoch": 2.8365543898398675, + "grad_norm": 0.928747832775116, + "learning_rate": 2.7689026763340302e-06, + "loss": 0.5925, + "step": 3426 + }, + { + "epoch": 2.8373826615129762, + "grad_norm": 0.9232720732688904, + "learning_rate": 2.7678093408335677e-06, + "loss": 0.594, + "step": 3427 + }, + { + "epoch": 2.838210933186085, + "grad_norm": 0.9319408535957336, + "learning_rate": 2.766715953514494e-06, + "loss": 0.607, + "step": 3428 + }, + { + "epoch": 2.8390392048591937, + "grad_norm": 0.9321261644363403, + "learning_rate": 2.76562251458837e-06, + "loss": 0.5839, + "step": 3429 + }, + { + "epoch": 2.8398674765323024, + "grad_norm": 0.8851343393325806, + "learning_rate": 2.7645290242667648e-06, + "loss": 0.5929, + "step": 3430 + }, + { + "epoch": 2.8406957482054116, + "grad_norm": 0.9318198561668396, + "learning_rate": 2.76343548276126e-06, + "loss": 0.5758, + "step": 3431 + }, + { + "epoch": 2.84152401987852, + "grad_norm": 0.9212138056755066, + "learning_rate": 2.7623418902834455e-06, + "loss": 0.5689, + "step": 3432 + }, + { + "epoch": 2.842352291551629, + "grad_norm": 0.8917051553726196, + "learning_rate": 2.761248247044919e-06, + "loss": 0.5915, + "step": 3433 + }, + { + "epoch": 2.843180563224738, + "grad_norm": 0.9266699552536011, + "learning_rate": 2.760154553257294e-06, + "loss": 0.5643, + "step": 3434 + }, + { + "epoch": 2.8440088348978465, + "grad_norm": 0.9837039709091187, + "learning_rate": 2.7590608091321867e-06, + "loss": 0.6135, + "step": 3435 + }, + { + "epoch": 2.8448371065709552, + "grad_norm": 0.9872124195098877, + "learning_rate": 2.7579670148812278e-06, + "loss": 0.5649, + "step": 3436 + }, + { + "epoch": 2.845665378244064, + "grad_norm": 0.9581724405288696, + "learning_rate": 2.756873170716057e-06, + "loss": 0.5833, + "step": 3437 + }, + { + "epoch": 2.8464936499171727, + "grad_norm": 0.9363225698471069, + "learning_rate": 2.7557792768483223e-06, + "loss": 0.5852, + "step": 3438 + }, + { + "epoch": 2.8473219215902814, + "grad_norm": 0.9177983403205872, + "learning_rate": 2.7546853334896813e-06, + "loss": 0.5959, + "step": 3439 + }, + { + "epoch": 2.8481501932633906, + "grad_norm": 0.9081945419311523, + "learning_rate": 2.7535913408518023e-06, + "loss": 0.5862, + "step": 3440 + }, + { + "epoch": 2.8489784649364993, + "grad_norm": 0.928897500038147, + "learning_rate": 2.7524972991463626e-06, + "loss": 0.5978, + "step": 3441 + }, + { + "epoch": 2.849806736609608, + "grad_norm": 0.9136974811553955, + "learning_rate": 2.75140320858505e-06, + "loss": 0.5926, + "step": 3442 + }, + { + "epoch": 2.8506350082827168, + "grad_norm": 0.9227224588394165, + "learning_rate": 2.7503090693795605e-06, + "loss": 0.6073, + "step": 3443 + }, + { + "epoch": 2.8514632799558255, + "grad_norm": 0.9390876889228821, + "learning_rate": 2.749214881741598e-06, + "loss": 0.5757, + "step": 3444 + }, + { + "epoch": 2.852291551628934, + "grad_norm": 0.9460882544517517, + "learning_rate": 2.74812064588288e-06, + "loss": 0.6172, + "step": 3445 + }, + { + "epoch": 2.853119823302043, + "grad_norm": 0.9087659120559692, + "learning_rate": 2.74702636201513e-06, + "loss": 0.5937, + "step": 3446 + }, + { + "epoch": 2.853948094975152, + "grad_norm": 0.8995270133018494, + "learning_rate": 2.745932030350081e-06, + "loss": 0.5798, + "step": 3447 + }, + { + "epoch": 2.8547763666482604, + "grad_norm": 0.9307460188865662, + "learning_rate": 2.744837651099478e-06, + "loss": 0.6009, + "step": 3448 + }, + { + "epoch": 2.8556046383213696, + "grad_norm": 0.9634310007095337, + "learning_rate": 2.7437432244750705e-06, + "loss": 0.5985, + "step": 3449 + }, + { + "epoch": 2.8564329099944783, + "grad_norm": 0.9647858142852783, + "learning_rate": 2.742648750688622e-06, + "loss": 0.5718, + "step": 3450 + }, + { + "epoch": 2.857261181667587, + "grad_norm": 0.9150539040565491, + "learning_rate": 2.7415542299519016e-06, + "loss": 0.5964, + "step": 3451 + }, + { + "epoch": 2.8580894533406958, + "grad_norm": 0.9129363298416138, + "learning_rate": 2.7404596624766892e-06, + "loss": 0.5514, + "step": 3452 + }, + { + "epoch": 2.8589177250138045, + "grad_norm": 0.9430226683616638, + "learning_rate": 2.739365048474774e-06, + "loss": 0.5795, + "step": 3453 + }, + { + "epoch": 2.859745996686913, + "grad_norm": 0.9402962327003479, + "learning_rate": 2.738270388157954e-06, + "loss": 0.5728, + "step": 3454 + }, + { + "epoch": 2.860574268360022, + "grad_norm": 0.9169720411300659, + "learning_rate": 2.7371756817380346e-06, + "loss": 0.6171, + "step": 3455 + }, + { + "epoch": 2.861402540033131, + "grad_norm": 0.9437370896339417, + "learning_rate": 2.7360809294268315e-06, + "loss": 0.5827, + "step": 3456 + }, + { + "epoch": 2.8622308117062394, + "grad_norm": 0.8999599814414978, + "learning_rate": 2.7349861314361693e-06, + "loss": 0.5847, + "step": 3457 + }, + { + "epoch": 2.8630590833793486, + "grad_norm": 0.9228375554084778, + "learning_rate": 2.7338912879778814e-06, + "loss": 0.6069, + "step": 3458 + }, + { + "epoch": 2.8638873550524573, + "grad_norm": 0.9543429017066956, + "learning_rate": 2.732796399263809e-06, + "loss": 0.6128, + "step": 3459 + }, + { + "epoch": 2.864715626725566, + "grad_norm": 0.923698902130127, + "learning_rate": 2.7317014655058038e-06, + "loss": 0.5714, + "step": 3460 + }, + { + "epoch": 2.8655438983986747, + "grad_norm": 0.95343017578125, + "learning_rate": 2.730606486915725e-06, + "loss": 0.5821, + "step": 3461 + }, + { + "epoch": 2.8663721700717835, + "grad_norm": 0.9078912734985352, + "learning_rate": 2.729511463705441e-06, + "loss": 0.5833, + "step": 3462 + }, + { + "epoch": 2.867200441744892, + "grad_norm": 0.9045196771621704, + "learning_rate": 2.7284163960868274e-06, + "loss": 0.5647, + "step": 3463 + }, + { + "epoch": 2.868028713418001, + "grad_norm": 0.9145541787147522, + "learning_rate": 2.727321284271771e-06, + "loss": 0.5731, + "step": 3464 + }, + { + "epoch": 2.86885698509111, + "grad_norm": 0.9013593792915344, + "learning_rate": 2.726226128472166e-06, + "loss": 0.5901, + "step": 3465 + }, + { + "epoch": 2.869685256764219, + "grad_norm": 0.9561366438865662, + "learning_rate": 2.7251309288999123e-06, + "loss": 0.585, + "step": 3466 + }, + { + "epoch": 2.8705135284373275, + "grad_norm": 0.9313666224479675, + "learning_rate": 2.724035685766923e-06, + "loss": 0.5755, + "step": 3467 + }, + { + "epoch": 2.8713418001104363, + "grad_norm": 0.9003223776817322, + "learning_rate": 2.722940399285118e-06, + "loss": 0.5961, + "step": 3468 + }, + { + "epoch": 2.872170071783545, + "grad_norm": 0.9185585975646973, + "learning_rate": 2.721845069666422e-06, + "loss": 0.5741, + "step": 3469 + }, + { + "epoch": 2.8729983434566537, + "grad_norm": 0.9295309782028198, + "learning_rate": 2.7207496971227743e-06, + "loss": 0.588, + "step": 3470 + }, + { + "epoch": 2.8738266151297625, + "grad_norm": 0.9500821232795715, + "learning_rate": 2.7196542818661188e-06, + "loss": 0.6026, + "step": 3471 + }, + { + "epoch": 2.874654886802871, + "grad_norm": 0.9265774488449097, + "learning_rate": 2.718558824108406e-06, + "loss": 0.5857, + "step": 3472 + }, + { + "epoch": 2.87548315847598, + "grad_norm": 0.9159520864486694, + "learning_rate": 2.717463324061599e-06, + "loss": 0.6023, + "step": 3473 + }, + { + "epoch": 2.876311430149089, + "grad_norm": 0.9114859700202942, + "learning_rate": 2.7163677819376653e-06, + "loss": 0.6144, + "step": 3474 + }, + { + "epoch": 2.877139701822198, + "grad_norm": 0.9181182980537415, + "learning_rate": 2.7152721979485824e-06, + "loss": 0.5853, + "step": 3475 + }, + { + "epoch": 2.8779679734953065, + "grad_norm": 0.9219772219657898, + "learning_rate": 2.7141765723063366e-06, + "loss": 0.5824, + "step": 3476 + }, + { + "epoch": 2.8787962451684153, + "grad_norm": 0.925654411315918, + "learning_rate": 2.713080905222921e-06, + "loss": 0.5859, + "step": 3477 + }, + { + "epoch": 2.879624516841524, + "grad_norm": 0.927891731262207, + "learning_rate": 2.711985196910336e-06, + "loss": 0.6047, + "step": 3478 + }, + { + "epoch": 2.8804527885146327, + "grad_norm": 0.9178606271743774, + "learning_rate": 2.7108894475805912e-06, + "loss": 0.5863, + "step": 3479 + }, + { + "epoch": 2.8812810601877414, + "grad_norm": 0.9341709613800049, + "learning_rate": 2.7097936574457046e-06, + "loss": 0.5718, + "step": 3480 + }, + { + "epoch": 2.8821093318608506, + "grad_norm": 0.9023662805557251, + "learning_rate": 2.7086978267177007e-06, + "loss": 0.593, + "step": 3481 + }, + { + "epoch": 2.882937603533959, + "grad_norm": 0.9269591569900513, + "learning_rate": 2.7076019556086136e-06, + "loss": 0.5955, + "step": 3482 + }, + { + "epoch": 2.883765875207068, + "grad_norm": 0.9572223424911499, + "learning_rate": 2.7065060443304823e-06, + "loss": 0.5698, + "step": 3483 + }, + { + "epoch": 2.884594146880177, + "grad_norm": 0.9251832365989685, + "learning_rate": 2.705410093095357e-06, + "loss": 0.5903, + "step": 3484 + }, + { + "epoch": 2.8854224185532855, + "grad_norm": 0.959985077381134, + "learning_rate": 2.7043141021152937e-06, + "loss": 0.5922, + "step": 3485 + }, + { + "epoch": 2.8862506902263942, + "grad_norm": 0.9356866478919983, + "learning_rate": 2.703218071602355e-06, + "loss": 0.576, + "step": 3486 + }, + { + "epoch": 2.887078961899503, + "grad_norm": 0.9041283130645752, + "learning_rate": 2.702122001768615e-06, + "loss": 0.5748, + "step": 3487 + }, + { + "epoch": 2.8879072335726117, + "grad_norm": 0.9056842923164368, + "learning_rate": 2.7010258928261514e-06, + "loss": 0.5795, + "step": 3488 + }, + { + "epoch": 2.8887355052457204, + "grad_norm": 0.9829550385475159, + "learning_rate": 2.6999297449870522e-06, + "loss": 0.6017, + "step": 3489 + }, + { + "epoch": 2.8895637769188296, + "grad_norm": 0.9483397603034973, + "learning_rate": 2.6988335584634103e-06, + "loss": 0.5849, + "step": 3490 + }, + { + "epoch": 2.890392048591938, + "grad_norm": 0.9553128480911255, + "learning_rate": 2.6977373334673286e-06, + "loss": 0.5518, + "step": 3491 + }, + { + "epoch": 2.891220320265047, + "grad_norm": 0.9298998713493347, + "learning_rate": 2.6966410702109153e-06, + "loss": 0.5947, + "step": 3492 + }, + { + "epoch": 2.8920485919381558, + "grad_norm": 0.9143223166465759, + "learning_rate": 2.6955447689062893e-06, + "loss": 0.5757, + "step": 3493 + }, + { + "epoch": 2.8928768636112645, + "grad_norm": 0.9143601059913635, + "learning_rate": 2.6944484297655727e-06, + "loss": 0.5867, + "step": 3494 + }, + { + "epoch": 2.8937051352843732, + "grad_norm": 0.9283237457275391, + "learning_rate": 2.693352053000898e-06, + "loss": 0.5928, + "step": 3495 + }, + { + "epoch": 2.894533406957482, + "grad_norm": 0.8984352946281433, + "learning_rate": 2.692255638824403e-06, + "loss": 0.5904, + "step": 3496 + }, + { + "epoch": 2.8953616786305907, + "grad_norm": 0.9056443572044373, + "learning_rate": 2.6911591874482334e-06, + "loss": 0.5847, + "step": 3497 + }, + { + "epoch": 2.8961899503036994, + "grad_norm": 0.9254133701324463, + "learning_rate": 2.6900626990845442e-06, + "loss": 0.608, + "step": 3498 + }, + { + "epoch": 2.8970182219768086, + "grad_norm": 0.9658727049827576, + "learning_rate": 2.6889661739454946e-06, + "loss": 0.6063, + "step": 3499 + }, + { + "epoch": 2.8978464936499173, + "grad_norm": 0.9350060820579529, + "learning_rate": 2.6878696122432505e-06, + "loss": 0.5902, + "step": 3500 + }, + { + "epoch": 2.898674765323026, + "grad_norm": 0.9070295691490173, + "learning_rate": 2.686773014189989e-06, + "loss": 0.5883, + "step": 3501 + }, + { + "epoch": 2.8995030369961348, + "grad_norm": 0.9471203088760376, + "learning_rate": 2.6856763799978895e-06, + "loss": 0.6138, + "step": 3502 + }, + { + "epoch": 2.9003313086692435, + "grad_norm": 0.901057243347168, + "learning_rate": 2.6845797098791415e-06, + "loss": 0.5925, + "step": 3503 + }, + { + "epoch": 2.901159580342352, + "grad_norm": 0.9290406107902527, + "learning_rate": 2.683483004045941e-06, + "loss": 0.5728, + "step": 3504 + }, + { + "epoch": 2.901987852015461, + "grad_norm": 0.9371906518936157, + "learning_rate": 2.682386262710489e-06, + "loss": 0.5894, + "step": 3505 + }, + { + "epoch": 2.90281612368857, + "grad_norm": 0.9183947443962097, + "learning_rate": 2.6812894860849953e-06, + "loss": 0.5959, + "step": 3506 + }, + { + "epoch": 2.9036443953616784, + "grad_norm": 0.9351047277450562, + "learning_rate": 2.6801926743816765e-06, + "loss": 0.5863, + "step": 3507 + }, + { + "epoch": 2.9044726670347876, + "grad_norm": 0.93602454662323, + "learning_rate": 2.6790958278127536e-06, + "loss": 0.5845, + "step": 3508 + }, + { + "epoch": 2.9053009387078963, + "grad_norm": 0.9618241786956787, + "learning_rate": 2.677998946590459e-06, + "loss": 0.6068, + "step": 3509 + }, + { + "epoch": 2.906129210381005, + "grad_norm": 0.9157661199569702, + "learning_rate": 2.6769020309270277e-06, + "loss": 0.5608, + "step": 3510 + }, + { + "epoch": 2.9069574820541138, + "grad_norm": 0.9157334566116333, + "learning_rate": 2.6758050810347012e-06, + "loss": 0.5883, + "step": 3511 + }, + { + "epoch": 2.9077857537272225, + "grad_norm": 0.9259895086288452, + "learning_rate": 2.674708097125731e-06, + "loss": 0.578, + "step": 3512 + }, + { + "epoch": 2.908614025400331, + "grad_norm": 0.9327818155288696, + "learning_rate": 2.6736110794123727e-06, + "loss": 0.6157, + "step": 3513 + }, + { + "epoch": 2.90944229707344, + "grad_norm": 0.9186248779296875, + "learning_rate": 2.6725140281068882e-06, + "loss": 0.5901, + "step": 3514 + }, + { + "epoch": 2.910270568746549, + "grad_norm": 0.9106549620628357, + "learning_rate": 2.671416943421548e-06, + "loss": 0.5983, + "step": 3515 + }, + { + "epoch": 2.9110988404196574, + "grad_norm": 0.9274003505706787, + "learning_rate": 2.6703198255686275e-06, + "loss": 0.5886, + "step": 3516 + }, + { + "epoch": 2.9119271120927666, + "grad_norm": 0.9015113115310669, + "learning_rate": 2.6692226747604082e-06, + "loss": 0.5979, + "step": 3517 + }, + { + "epoch": 2.9127553837658753, + "grad_norm": 0.9261288046836853, + "learning_rate": 2.6681254912091788e-06, + "loss": 0.5741, + "step": 3518 + }, + { + "epoch": 2.913583655438984, + "grad_norm": 0.8937399983406067, + "learning_rate": 2.6670282751272335e-06, + "loss": 0.5726, + "step": 3519 + }, + { + "epoch": 2.9144119271120927, + "grad_norm": 0.9202954769134521, + "learning_rate": 2.6659310267268744e-06, + "loss": 0.5788, + "step": 3520 + }, + { + "epoch": 2.9152401987852015, + "grad_norm": 0.9400494694709778, + "learning_rate": 2.6648337462204087e-06, + "loss": 0.568, + "step": 3521 + }, + { + "epoch": 2.91606847045831, + "grad_norm": 0.9291467070579529, + "learning_rate": 2.6637364338201497e-06, + "loss": 0.589, + "step": 3522 + }, + { + "epoch": 2.916896742131419, + "grad_norm": 0.9186866879463196, + "learning_rate": 2.662639089738417e-06, + "loss": 0.59, + "step": 3523 + }, + { + "epoch": 2.917725013804528, + "grad_norm": 0.9596429467201233, + "learning_rate": 2.6615417141875367e-06, + "loss": 0.5759, + "step": 3524 + }, + { + "epoch": 2.918553285477637, + "grad_norm": 0.9279852509498596, + "learning_rate": 2.6604443073798395e-06, + "loss": 0.5969, + "step": 3525 + }, + { + "epoch": 2.9193815571507455, + "grad_norm": 0.9540870189666748, + "learning_rate": 2.6593468695276657e-06, + "loss": 0.5696, + "step": 3526 + }, + { + "epoch": 2.9202098288238543, + "grad_norm": 0.9415677785873413, + "learning_rate": 2.658249400843358e-06, + "loss": 0.5973, + "step": 3527 + }, + { + "epoch": 2.921038100496963, + "grad_norm": 0.8950624465942383, + "learning_rate": 2.657151901539266e-06, + "loss": 0.5952, + "step": 3528 + }, + { + "epoch": 2.9218663721700717, + "grad_norm": 0.8927432298660278, + "learning_rate": 2.656054371827746e-06, + "loss": 0.59, + "step": 3529 + }, + { + "epoch": 2.9226946438431805, + "grad_norm": 0.9516159892082214, + "learning_rate": 2.65495681192116e-06, + "loss": 0.6024, + "step": 3530 + }, + { + "epoch": 2.923522915516289, + "grad_norm": 1.0094369649887085, + "learning_rate": 2.653859222031875e-06, + "loss": 0.596, + "step": 3531 + }, + { + "epoch": 2.924351187189398, + "grad_norm": 0.9162117838859558, + "learning_rate": 2.6527616023722647e-06, + "loss": 0.5739, + "step": 3532 + }, + { + "epoch": 2.925179458862507, + "grad_norm": 0.9844574332237244, + "learning_rate": 2.651663953154709e-06, + "loss": 0.6028, + "step": 3533 + }, + { + "epoch": 2.926007730535616, + "grad_norm": 0.9091664552688599, + "learning_rate": 2.6505662745915915e-06, + "loss": 0.5867, + "step": 3534 + }, + { + "epoch": 2.9268360022087245, + "grad_norm": 0.9107911586761475, + "learning_rate": 2.649468566895304e-06, + "loss": 0.5985, + "step": 3535 + }, + { + "epoch": 2.9276642738818333, + "grad_norm": 0.9392855167388916, + "learning_rate": 2.648370830278241e-06, + "loss": 0.5995, + "step": 3536 + }, + { + "epoch": 2.928492545554942, + "grad_norm": 0.9228532910346985, + "learning_rate": 2.6472730649528068e-06, + "loss": 0.5928, + "step": 3537 + }, + { + "epoch": 2.9293208172280507, + "grad_norm": 0.9555805325508118, + "learning_rate": 2.646175271131407e-06, + "loss": 0.606, + "step": 3538 + }, + { + "epoch": 2.9301490889011594, + "grad_norm": 0.9101611971855164, + "learning_rate": 2.6450774490264547e-06, + "loss": 0.5881, + "step": 3539 + }, + { + "epoch": 2.9309773605742686, + "grad_norm": 0.9572548866271973, + "learning_rate": 2.6439795988503685e-06, + "loss": 0.5861, + "step": 3540 + }, + { + "epoch": 2.931805632247377, + "grad_norm": 0.9441561102867126, + "learning_rate": 2.6428817208155728e-06, + "loss": 0.594, + "step": 3541 + }, + { + "epoch": 2.932633903920486, + "grad_norm": 0.9282614588737488, + "learning_rate": 2.6417838151344945e-06, + "loss": 0.5991, + "step": 3542 + }, + { + "epoch": 2.933462175593595, + "grad_norm": 0.9630649089813232, + "learning_rate": 2.640685882019571e-06, + "loss": 0.6147, + "step": 3543 + }, + { + "epoch": 2.9342904472667035, + "grad_norm": 0.9284330606460571, + "learning_rate": 2.63958792168324e-06, + "loss": 0.6021, + "step": 3544 + }, + { + "epoch": 2.9351187189398122, + "grad_norm": 0.9337431788444519, + "learning_rate": 2.638489934337948e-06, + "loss": 0.602, + "step": 3545 + }, + { + "epoch": 2.935946990612921, + "grad_norm": 0.8977872729301453, + "learning_rate": 2.637391920196144e-06, + "loss": 0.5901, + "step": 3546 + }, + { + "epoch": 2.9367752622860297, + "grad_norm": 0.962690532207489, + "learning_rate": 2.636293879470284e-06, + "loss": 0.5793, + "step": 3547 + }, + { + "epoch": 2.9376035339591384, + "grad_norm": 0.8999794721603394, + "learning_rate": 2.6351958123728294e-06, + "loss": 0.6027, + "step": 3548 + }, + { + "epoch": 2.9384318056322476, + "grad_norm": 0.9200299382209778, + "learning_rate": 2.6340977191162457e-06, + "loss": 0.5985, + "step": 3549 + }, + { + "epoch": 2.9392600773053563, + "grad_norm": 0.9161362648010254, + "learning_rate": 2.6329995999130032e-06, + "loss": 0.595, + "step": 3550 + }, + { + "epoch": 2.940088348978465, + "grad_norm": 0.9192216992378235, + "learning_rate": 2.631901454975578e-06, + "loss": 0.5844, + "step": 3551 + }, + { + "epoch": 2.9409166206515738, + "grad_norm": 0.9136098623275757, + "learning_rate": 2.6308032845164503e-06, + "loss": 0.5932, + "step": 3552 + }, + { + "epoch": 2.9417448923246825, + "grad_norm": 0.9625688791275024, + "learning_rate": 2.629705088748106e-06, + "loss": 0.5811, + "step": 3553 + }, + { + "epoch": 2.9425731639977912, + "grad_norm": 0.9300727248191833, + "learning_rate": 2.6286068678830375e-06, + "loss": 0.6035, + "step": 3554 + }, + { + "epoch": 2.9434014356709, + "grad_norm": 0.9947618246078491, + "learning_rate": 2.627508622133738e-06, + "loss": 0.5822, + "step": 3555 + }, + { + "epoch": 2.9442297073440087, + "grad_norm": 0.9255596995353699, + "learning_rate": 2.626410351712709e-06, + "loss": 0.5824, + "step": 3556 + }, + { + "epoch": 2.9450579790171174, + "grad_norm": 0.9395889043807983, + "learning_rate": 2.6253120568324557e-06, + "loss": 0.6121, + "step": 3557 + }, + { + "epoch": 2.9458862506902266, + "grad_norm": 0.9438552856445312, + "learning_rate": 2.6242137377054865e-06, + "loss": 0.5843, + "step": 3558 + }, + { + "epoch": 2.9467145223633353, + "grad_norm": 0.9153048992156982, + "learning_rate": 2.623115394544318e-06, + "loss": 0.5893, + "step": 3559 + }, + { + "epoch": 2.947542794036444, + "grad_norm": 0.9119593501091003, + "learning_rate": 2.6220170275614677e-06, + "loss": 0.6047, + "step": 3560 + }, + { + "epoch": 2.9483710657095528, + "grad_norm": 0.9174615144729614, + "learning_rate": 2.6209186369694606e-06, + "loss": 0.5949, + "step": 3561 + }, + { + "epoch": 2.9491993373826615, + "grad_norm": 0.9030957818031311, + "learning_rate": 2.619820222980824e-06, + "loss": 0.6146, + "step": 3562 + }, + { + "epoch": 2.95002760905577, + "grad_norm": 0.9036881923675537, + "learning_rate": 2.618721785808091e-06, + "loss": 0.5946, + "step": 3563 + }, + { + "epoch": 2.950855880728879, + "grad_norm": 0.8928536772727966, + "learning_rate": 2.617623325663799e-06, + "loss": 0.5679, + "step": 3564 + }, + { + "epoch": 2.951684152401988, + "grad_norm": 0.9225850701332092, + "learning_rate": 2.6165248427604907e-06, + "loss": 0.5938, + "step": 3565 + }, + { + "epoch": 2.9525124240750964, + "grad_norm": 0.9023733139038086, + "learning_rate": 2.6154263373107113e-06, + "loss": 0.5896, + "step": 3566 + }, + { + "epoch": 2.9533406957482056, + "grad_norm": 0.9055750370025635, + "learning_rate": 2.6143278095270113e-06, + "loss": 0.6007, + "step": 3567 + }, + { + "epoch": 2.9541689674213143, + "grad_norm": 0.8988812565803528, + "learning_rate": 2.613229259621947e-06, + "loss": 0.6097, + "step": 3568 + }, + { + "epoch": 2.954997239094423, + "grad_norm": 0.9200608730316162, + "learning_rate": 2.6121306878080753e-06, + "loss": 0.58, + "step": 3569 + }, + { + "epoch": 2.9558255107675317, + "grad_norm": 0.9378446936607361, + "learning_rate": 2.6110320942979606e-06, + "loss": 0.6043, + "step": 3570 + }, + { + "epoch": 2.9566537824406405, + "grad_norm": 0.9543853402137756, + "learning_rate": 2.6099334793041714e-06, + "loss": 0.58, + "step": 3571 + }, + { + "epoch": 2.957482054113749, + "grad_norm": 0.9590248465538025, + "learning_rate": 2.6088348430392784e-06, + "loss": 0.6004, + "step": 3572 + }, + { + "epoch": 2.958310325786858, + "grad_norm": 0.980459988117218, + "learning_rate": 2.607736185715858e-06, + "loss": 0.6118, + "step": 3573 + }, + { + "epoch": 2.959138597459967, + "grad_norm": 0.9050843715667725, + "learning_rate": 2.60663750754649e-06, + "loss": 0.5913, + "step": 3574 + }, + { + "epoch": 2.9599668691330754, + "grad_norm": 0.9189395904541016, + "learning_rate": 2.6055388087437576e-06, + "loss": 0.5997, + "step": 3575 + }, + { + "epoch": 2.9607951408061846, + "grad_norm": 0.9273754954338074, + "learning_rate": 2.6044400895202502e-06, + "loss": 0.586, + "step": 3576 + }, + { + "epoch": 2.9616234124792933, + "grad_norm": 0.90342116355896, + "learning_rate": 2.6033413500885594e-06, + "loss": 0.5846, + "step": 3577 + }, + { + "epoch": 2.962451684152402, + "grad_norm": 0.9073745012283325, + "learning_rate": 2.6022425906612803e-06, + "loss": 0.5899, + "step": 3578 + }, + { + "epoch": 2.9632799558255107, + "grad_norm": 0.9221594333648682, + "learning_rate": 2.6011438114510135e-06, + "loss": 0.5837, + "step": 3579 + }, + { + "epoch": 2.9641082274986195, + "grad_norm": 0.9228225350379944, + "learning_rate": 2.6000450126703613e-06, + "loss": 0.5761, + "step": 3580 + }, + { + "epoch": 2.964936499171728, + "grad_norm": 0.9136195182800293, + "learning_rate": 2.5989461945319315e-06, + "loss": 0.5977, + "step": 3581 + }, + { + "epoch": 2.965764770844837, + "grad_norm": 0.9152894616127014, + "learning_rate": 2.5978473572483366e-06, + "loss": 0.5829, + "step": 3582 + }, + { + "epoch": 2.966593042517946, + "grad_norm": 0.9124417304992676, + "learning_rate": 2.596748501032189e-06, + "loss": 0.5718, + "step": 3583 + }, + { + "epoch": 2.967421314191055, + "grad_norm": 0.9212865829467773, + "learning_rate": 2.5956496260961094e-06, + "loss": 0.6056, + "step": 3584 + }, + { + "epoch": 2.9682495858641635, + "grad_norm": 0.9101969003677368, + "learning_rate": 2.594550732652718e-06, + "loss": 0.6075, + "step": 3585 + }, + { + "epoch": 2.9690778575372723, + "grad_norm": 0.9225336313247681, + "learning_rate": 2.5934518209146414e-06, + "loss": 0.5894, + "step": 3586 + }, + { + "epoch": 2.969906129210381, + "grad_norm": 0.9271754026412964, + "learning_rate": 2.592352891094509e-06, + "loss": 0.5939, + "step": 3587 + }, + { + "epoch": 2.9707344008834897, + "grad_norm": 0.9217578768730164, + "learning_rate": 2.591253943404953e-06, + "loss": 0.5858, + "step": 3588 + }, + { + "epoch": 2.9715626725565985, + "grad_norm": 0.8871975541114807, + "learning_rate": 2.5901549780586085e-06, + "loss": 0.6052, + "step": 3589 + }, + { + "epoch": 2.9723909442297076, + "grad_norm": 0.9371136426925659, + "learning_rate": 2.5890559952681177e-06, + "loss": 0.6069, + "step": 3590 + }, + { + "epoch": 2.973219215902816, + "grad_norm": 0.9185692667961121, + "learning_rate": 2.5879569952461215e-06, + "loss": 0.5913, + "step": 3591 + }, + { + "epoch": 2.974047487575925, + "grad_norm": 0.9043979048728943, + "learning_rate": 2.586857978205266e-06, + "loss": 0.5747, + "step": 3592 + }, + { + "epoch": 2.974875759249034, + "grad_norm": 0.8951820731163025, + "learning_rate": 2.5857589443582014e-06, + "loss": 0.5904, + "step": 3593 + }, + { + "epoch": 2.9757040309221425, + "grad_norm": 0.9490121006965637, + "learning_rate": 2.5846598939175816e-06, + "loss": 0.583, + "step": 3594 + }, + { + "epoch": 2.9765323025952513, + "grad_norm": 0.9266814589500427, + "learning_rate": 2.583560827096061e-06, + "loss": 0.6039, + "step": 3595 + }, + { + "epoch": 2.97736057426836, + "grad_norm": 0.9517508149147034, + "learning_rate": 2.582461744106299e-06, + "loss": 0.572, + "step": 3596 + }, + { + "epoch": 2.9781888459414687, + "grad_norm": 0.9260841608047485, + "learning_rate": 2.581362645160958e-06, + "loss": 0.5898, + "step": 3597 + }, + { + "epoch": 2.9790171176145774, + "grad_norm": 0.9134445190429688, + "learning_rate": 2.5802635304727043e-06, + "loss": 0.6037, + "step": 3598 + }, + { + "epoch": 2.9798453892876866, + "grad_norm": 0.9307652115821838, + "learning_rate": 2.5791644002542054e-06, + "loss": 0.5941, + "step": 3599 + }, + { + "epoch": 2.980673660960795, + "grad_norm": 0.9665980935096741, + "learning_rate": 2.578065254718133e-06, + "loss": 0.5816, + "step": 3600 + }, + { + "epoch": 2.981501932633904, + "grad_norm": 0.9137181639671326, + "learning_rate": 2.5769660940771614e-06, + "loss": 0.5962, + "step": 3601 + }, + { + "epoch": 2.982330204307013, + "grad_norm": 0.9031209349632263, + "learning_rate": 2.575866918543968e-06, + "loss": 0.5853, + "step": 3602 + }, + { + "epoch": 2.9831584759801215, + "grad_norm": 0.9292592406272888, + "learning_rate": 2.5747677283312332e-06, + "loss": 0.5639, + "step": 3603 + }, + { + "epoch": 2.9839867476532302, + "grad_norm": 0.9193013906478882, + "learning_rate": 2.57366852365164e-06, + "loss": 0.5942, + "step": 3604 + }, + { + "epoch": 2.984815019326339, + "grad_norm": 0.8945804238319397, + "learning_rate": 2.572569304717874e-06, + "loss": 0.5882, + "step": 3605 + }, + { + "epoch": 2.9856432909994477, + "grad_norm": 0.9231047034263611, + "learning_rate": 2.5714700717426237e-06, + "loss": 0.5768, + "step": 3606 + }, + { + "epoch": 2.9864715626725564, + "grad_norm": 0.9362046718597412, + "learning_rate": 2.5703708249385824e-06, + "loss": 0.6145, + "step": 3607 + }, + { + "epoch": 2.9872998343456656, + "grad_norm": 0.9291548728942871, + "learning_rate": 2.5692715645184412e-06, + "loss": 0.5959, + "step": 3608 + }, + { + "epoch": 2.9881281060187743, + "grad_norm": 0.9203015565872192, + "learning_rate": 2.5681722906948977e-06, + "loss": 0.5906, + "step": 3609 + }, + { + "epoch": 2.988956377691883, + "grad_norm": 0.9317998886108398, + "learning_rate": 2.567073003680652e-06, + "loss": 0.5845, + "step": 3610 + }, + { + "epoch": 2.9897846493649918, + "grad_norm": 0.9188578128814697, + "learning_rate": 2.5659737036884054e-06, + "loss": 0.5901, + "step": 3611 + }, + { + "epoch": 2.9906129210381005, + "grad_norm": 0.9100846648216248, + "learning_rate": 2.564874390930862e-06, + "loss": 0.5952, + "step": 3612 + }, + { + "epoch": 2.9914411927112092, + "grad_norm": 0.9292994737625122, + "learning_rate": 2.563775065620729e-06, + "loss": 0.6112, + "step": 3613 + }, + { + "epoch": 2.992269464384318, + "grad_norm": 0.9025140404701233, + "learning_rate": 2.5626757279707143e-06, + "loss": 0.5691, + "step": 3614 + }, + { + "epoch": 2.9930977360574267, + "grad_norm": 0.9413566589355469, + "learning_rate": 2.561576378193531e-06, + "loss": 0.574, + "step": 3615 + }, + { + "epoch": 2.9939260077305354, + "grad_norm": 0.9244924783706665, + "learning_rate": 2.560477016501893e-06, + "loss": 0.606, + "step": 3616 + }, + { + "epoch": 2.9947542794036446, + "grad_norm": 0.8985790014266968, + "learning_rate": 2.5593776431085154e-06, + "loss": 0.5823, + "step": 3617 + }, + { + "epoch": 2.9955825510767533, + "grad_norm": 0.8941566348075867, + "learning_rate": 2.5582782582261182e-06, + "loss": 0.5643, + "step": 3618 + }, + { + "epoch": 2.996410822749862, + "grad_norm": 0.8992639183998108, + "learning_rate": 2.557178862067421e-06, + "loss": 0.6073, + "step": 3619 + }, + { + "epoch": 2.9972390944229708, + "grad_norm": 0.9369136691093445, + "learning_rate": 2.5560794548451464e-06, + "loss": 0.6031, + "step": 3620 + }, + { + "epoch": 2.9980673660960795, + "grad_norm": 0.8975582718849182, + "learning_rate": 2.5549800367720207e-06, + "loss": 0.6034, + "step": 3621 + }, + { + "epoch": 2.998895637769188, + "grad_norm": 0.8998291492462158, + "learning_rate": 2.5538806080607697e-06, + "loss": 0.5782, + "step": 3622 + }, + { + "epoch": 2.999723909442297, + "grad_norm": 0.9796467423439026, + "learning_rate": 2.5527811689241246e-06, + "loss": 0.581, + "step": 3623 + }, + { + "epoch": 3.0, + "grad_norm": 1.6821987628936768, + "learning_rate": 2.5516817195748143e-06, + "loss": 0.5955, + "step": 3624 + }, + { + "epoch": 3.0008282716731087, + "grad_norm": 0.9084760546684265, + "learning_rate": 2.5505822602255727e-06, + "loss": 0.5821, + "step": 3625 + }, + { + "epoch": 3.0016565433462175, + "grad_norm": 0.9643696546554565, + "learning_rate": 2.5494827910891357e-06, + "loss": 0.5627, + "step": 3626 + }, + { + "epoch": 3.002484815019326, + "grad_norm": 0.8933544754981995, + "learning_rate": 2.5483833123782405e-06, + "loss": 0.5573, + "step": 3627 + }, + { + "epoch": 3.003313086692435, + "grad_norm": 0.939877450466156, + "learning_rate": 2.5472838243056242e-06, + "loss": 0.5641, + "step": 3628 + }, + { + "epoch": 3.004141358365544, + "grad_norm": 0.9558863043785095, + "learning_rate": 2.54618432708403e-06, + "loss": 0.5753, + "step": 3629 + }, + { + "epoch": 3.004969630038653, + "grad_norm": 0.9423503875732422, + "learning_rate": 2.5450848209261987e-06, + "loss": 0.5786, + "step": 3630 + }, + { + "epoch": 3.0057979017117615, + "grad_norm": 0.9265878796577454, + "learning_rate": 2.5439853060448737e-06, + "loss": 0.5791, + "step": 3631 + }, + { + "epoch": 3.0066261733848703, + "grad_norm": 0.9355331063270569, + "learning_rate": 2.542885782652803e-06, + "loss": 0.5535, + "step": 3632 + }, + { + "epoch": 3.007454445057979, + "grad_norm": 0.9215316772460938, + "learning_rate": 2.5417862509627333e-06, + "loss": 0.5775, + "step": 3633 + }, + { + "epoch": 3.0082827167310877, + "grad_norm": 0.9157372713088989, + "learning_rate": 2.5406867111874127e-06, + "loss": 0.5547, + "step": 3634 + }, + { + "epoch": 3.0091109884041964, + "grad_norm": 0.9143363237380981, + "learning_rate": 2.539587163539594e-06, + "loss": 0.562, + "step": 3635 + }, + { + "epoch": 3.009939260077305, + "grad_norm": 0.942833423614502, + "learning_rate": 2.538487608232027e-06, + "loss": 0.5852, + "step": 3636 + }, + { + "epoch": 3.0107675317504143, + "grad_norm": 0.9196774363517761, + "learning_rate": 2.5373880454774667e-06, + "loss": 0.573, + "step": 3637 + }, + { + "epoch": 3.011595803423523, + "grad_norm": 0.9290075302124023, + "learning_rate": 2.536288475488669e-06, + "loss": 0.5672, + "step": 3638 + }, + { + "epoch": 3.012424075096632, + "grad_norm": 0.9856467843055725, + "learning_rate": 2.535188898478389e-06, + "loss": 0.5685, + "step": 3639 + }, + { + "epoch": 3.0132523467697405, + "grad_norm": 0.9258086085319519, + "learning_rate": 2.5340893146593855e-06, + "loss": 0.5705, + "step": 3640 + }, + { + "epoch": 3.0140806184428492, + "grad_norm": 0.8912391662597656, + "learning_rate": 2.5329897242444174e-06, + "loss": 0.5653, + "step": 3641 + }, + { + "epoch": 3.014908890115958, + "grad_norm": 0.9336050152778625, + "learning_rate": 2.531890127446244e-06, + "loss": 0.5515, + "step": 3642 + }, + { + "epoch": 3.0157371617890667, + "grad_norm": 0.9292400479316711, + "learning_rate": 2.5307905244776295e-06, + "loss": 0.5715, + "step": 3643 + }, + { + "epoch": 3.0165654334621754, + "grad_norm": 0.9212039709091187, + "learning_rate": 2.529690915551335e-06, + "loss": 0.5842, + "step": 3644 + }, + { + "epoch": 3.017393705135284, + "grad_norm": 0.9198867082595825, + "learning_rate": 2.528591300880125e-06, + "loss": 0.5498, + "step": 3645 + }, + { + "epoch": 3.0182219768083933, + "grad_norm": 0.9194673299789429, + "learning_rate": 2.5274916806767656e-06, + "loss": 0.5696, + "step": 3646 + }, + { + "epoch": 3.019050248481502, + "grad_norm": 0.9359262585639954, + "learning_rate": 2.5263920551540217e-06, + "loss": 0.5635, + "step": 3647 + }, + { + "epoch": 3.019878520154611, + "grad_norm": 0.9225771427154541, + "learning_rate": 2.525292424524661e-06, + "loss": 0.5489, + "step": 3648 + }, + { + "epoch": 3.0207067918277195, + "grad_norm": 0.9490647315979004, + "learning_rate": 2.524192789001452e-06, + "loss": 0.5692, + "step": 3649 + }, + { + "epoch": 3.0215350635008282, + "grad_norm": 0.8977577686309814, + "learning_rate": 2.5230931487971645e-06, + "loss": 0.5652, + "step": 3650 + }, + { + "epoch": 3.022363335173937, + "grad_norm": 0.963139533996582, + "learning_rate": 2.521993504124567e-06, + "loss": 0.5373, + "step": 3651 + }, + { + "epoch": 3.0231916068470457, + "grad_norm": 0.9195258617401123, + "learning_rate": 2.5208938551964326e-06, + "loss": 0.574, + "step": 3652 + }, + { + "epoch": 3.0240198785201544, + "grad_norm": 0.9074851274490356, + "learning_rate": 2.5197942022255305e-06, + "loss": 0.557, + "step": 3653 + }, + { + "epoch": 3.0248481501932636, + "grad_norm": 0.9083223342895508, + "learning_rate": 2.518694545424636e-06, + "loss": 0.5641, + "step": 3654 + }, + { + "epoch": 3.0256764218663723, + "grad_norm": 0.9047815203666687, + "learning_rate": 2.5175948850065207e-06, + "loss": 0.563, + "step": 3655 + }, + { + "epoch": 3.026504693539481, + "grad_norm": 0.918021559715271, + "learning_rate": 2.516495221183959e-06, + "loss": 0.5889, + "step": 3656 + }, + { + "epoch": 3.0273329652125898, + "grad_norm": 0.935458242893219, + "learning_rate": 2.5153955541697266e-06, + "loss": 0.5486, + "step": 3657 + }, + { + "epoch": 3.0281612368856985, + "grad_norm": 0.9182383418083191, + "learning_rate": 2.5142958841765976e-06, + "loss": 0.563, + "step": 3658 + }, + { + "epoch": 3.028989508558807, + "grad_norm": 0.9635967016220093, + "learning_rate": 2.5131962114173474e-06, + "loss": 0.557, + "step": 3659 + }, + { + "epoch": 3.029817780231916, + "grad_norm": 0.9150283932685852, + "learning_rate": 2.512096536104754e-06, + "loss": 0.5582, + "step": 3660 + }, + { + "epoch": 3.0306460519050247, + "grad_norm": 0.9288967847824097, + "learning_rate": 2.5109968584515935e-06, + "loss": 0.5628, + "step": 3661 + }, + { + "epoch": 3.031474323578134, + "grad_norm": 0.936221718788147, + "learning_rate": 2.5098971786706434e-06, + "loss": 0.564, + "step": 3662 + }, + { + "epoch": 3.0323025952512426, + "grad_norm": 0.9719551801681519, + "learning_rate": 2.5087974969746828e-06, + "loss": 0.5509, + "step": 3663 + }, + { + "epoch": 3.0331308669243513, + "grad_norm": 0.9522861838340759, + "learning_rate": 2.507697813576487e-06, + "loss": 0.5919, + "step": 3664 + }, + { + "epoch": 3.03395913859746, + "grad_norm": 0.9098208546638489, + "learning_rate": 2.506598128688836e-06, + "loss": 0.5661, + "step": 3665 + }, + { + "epoch": 3.0347874102705688, + "grad_norm": 0.9221910238265991, + "learning_rate": 2.5054984425245095e-06, + "loss": 0.5849, + "step": 3666 + }, + { + "epoch": 3.0356156819436775, + "grad_norm": 0.9386799335479736, + "learning_rate": 2.5043987552962844e-06, + "loss": 0.5765, + "step": 3667 + }, + { + "epoch": 3.036443953616786, + "grad_norm": 0.9208933711051941, + "learning_rate": 2.503299067216942e-06, + "loss": 0.5689, + "step": 3668 + }, + { + "epoch": 3.037272225289895, + "grad_norm": 0.9429235458374023, + "learning_rate": 2.5021993784992616e-06, + "loss": 0.555, + "step": 3669 + }, + { + "epoch": 3.0381004969630037, + "grad_norm": 0.9263848066329956, + "learning_rate": 2.5010996893560203e-06, + "loss": 0.562, + "step": 3670 + }, + { + "epoch": 3.038928768636113, + "grad_norm": 0.9523965120315552, + "learning_rate": 2.5e-06, + "loss": 0.5677, + "step": 3671 + }, + { + "epoch": 3.0397570403092216, + "grad_norm": 0.9585251212120056, + "learning_rate": 2.49890031064398e-06, + "loss": 0.5821, + "step": 3672 + }, + { + "epoch": 3.0405853119823303, + "grad_norm": 0.9311655163764954, + "learning_rate": 2.4978006215007393e-06, + "loss": 0.5752, + "step": 3673 + }, + { + "epoch": 3.041413583655439, + "grad_norm": 0.9548931121826172, + "learning_rate": 2.4967009327830584e-06, + "loss": 0.5719, + "step": 3674 + }, + { + "epoch": 3.0422418553285477, + "grad_norm": 0.9751641154289246, + "learning_rate": 2.495601244703716e-06, + "loss": 0.5869, + "step": 3675 + }, + { + "epoch": 3.0430701270016565, + "grad_norm": 0.9528283476829529, + "learning_rate": 2.494501557475491e-06, + "loss": 0.571, + "step": 3676 + }, + { + "epoch": 3.043898398674765, + "grad_norm": 0.9225226044654846, + "learning_rate": 2.493401871311165e-06, + "loss": 0.5635, + "step": 3677 + }, + { + "epoch": 3.044726670347874, + "grad_norm": 0.9444342851638794, + "learning_rate": 2.4923021864235143e-06, + "loss": 0.5593, + "step": 3678 + }, + { + "epoch": 3.045554942020983, + "grad_norm": 0.9299778938293457, + "learning_rate": 2.491202503025319e-06, + "loss": 0.5787, + "step": 3679 + }, + { + "epoch": 3.046383213694092, + "grad_norm": 0.913502037525177, + "learning_rate": 2.490102821329357e-06, + "loss": 0.5826, + "step": 3680 + }, + { + "epoch": 3.0472114853672005, + "grad_norm": 0.9337341785430908, + "learning_rate": 2.489003141548407e-06, + "loss": 0.5575, + "step": 3681 + }, + { + "epoch": 3.0480397570403093, + "grad_norm": 0.9271352887153625, + "learning_rate": 2.4879034638952464e-06, + "loss": 0.5759, + "step": 3682 + }, + { + "epoch": 3.048868028713418, + "grad_norm": 0.967474102973938, + "learning_rate": 2.486803788582653e-06, + "loss": 0.5542, + "step": 3683 + }, + { + "epoch": 3.0496963003865267, + "grad_norm": 0.9283983111381531, + "learning_rate": 2.4857041158234032e-06, + "loss": 0.5932, + "step": 3684 + }, + { + "epoch": 3.0505245720596355, + "grad_norm": 0.9075651168823242, + "learning_rate": 2.4846044458302742e-06, + "loss": 0.5704, + "step": 3685 + }, + { + "epoch": 3.051352843732744, + "grad_norm": 0.9511797428131104, + "learning_rate": 2.4835047788160415e-06, + "loss": 0.5788, + "step": 3686 + }, + { + "epoch": 3.052181115405853, + "grad_norm": 0.9016287922859192, + "learning_rate": 2.4824051149934797e-06, + "loss": 0.5829, + "step": 3687 + }, + { + "epoch": 3.053009387078962, + "grad_norm": 0.9479110836982727, + "learning_rate": 2.4813054545753657e-06, + "loss": 0.5721, + "step": 3688 + }, + { + "epoch": 3.053837658752071, + "grad_norm": 0.906857430934906, + "learning_rate": 2.4802057977744703e-06, + "loss": 0.5497, + "step": 3689 + }, + { + "epoch": 3.0546659304251795, + "grad_norm": 0.9445266723632812, + "learning_rate": 2.479106144803569e-06, + "loss": 0.5837, + "step": 3690 + }, + { + "epoch": 3.0554942020982883, + "grad_norm": 0.9599757194519043, + "learning_rate": 2.478006495875434e-06, + "loss": 0.5571, + "step": 3691 + }, + { + "epoch": 3.056322473771397, + "grad_norm": 0.924673318862915, + "learning_rate": 2.4769068512028368e-06, + "loss": 0.5644, + "step": 3692 + }, + { + "epoch": 3.0571507454445057, + "grad_norm": 0.9168586730957031, + "learning_rate": 2.4758072109985487e-06, + "loss": 0.5682, + "step": 3693 + }, + { + "epoch": 3.0579790171176144, + "grad_norm": 0.9143668413162231, + "learning_rate": 2.4747075754753395e-06, + "loss": 0.5521, + "step": 3694 + }, + { + "epoch": 3.058807288790723, + "grad_norm": 0.9329923391342163, + "learning_rate": 2.4736079448459787e-06, + "loss": 0.5534, + "step": 3695 + }, + { + "epoch": 3.0596355604638323, + "grad_norm": 0.9511533975601196, + "learning_rate": 2.472508319323235e-06, + "loss": 0.5795, + "step": 3696 + }, + { + "epoch": 3.060463832136941, + "grad_norm": 0.9574876427650452, + "learning_rate": 2.471408699119875e-06, + "loss": 0.5723, + "step": 3697 + }, + { + "epoch": 3.06129210381005, + "grad_norm": 0.9656964540481567, + "learning_rate": 2.470309084448665e-06, + "loss": 0.5876, + "step": 3698 + }, + { + "epoch": 3.0621203754831585, + "grad_norm": 0.9273092150688171, + "learning_rate": 2.469209475522371e-06, + "loss": 0.5591, + "step": 3699 + }, + { + "epoch": 3.0629486471562672, + "grad_norm": 0.9115599393844604, + "learning_rate": 2.4681098725537566e-06, + "loss": 0.5766, + "step": 3700 + }, + { + "epoch": 3.063776918829376, + "grad_norm": 0.9338927268981934, + "learning_rate": 2.4670102757555843e-06, + "loss": 0.5584, + "step": 3701 + }, + { + "epoch": 3.0646051905024847, + "grad_norm": 0.9274653792381287, + "learning_rate": 2.4659106853406158e-06, + "loss": 0.5545, + "step": 3702 + }, + { + "epoch": 3.0654334621755934, + "grad_norm": 0.9300533533096313, + "learning_rate": 2.464811101521612e-06, + "loss": 0.5751, + "step": 3703 + }, + { + "epoch": 3.0662617338487026, + "grad_norm": 0.9616261124610901, + "learning_rate": 2.463711524511332e-06, + "loss": 0.5635, + "step": 3704 + }, + { + "epoch": 3.0670900055218113, + "grad_norm": 1.0845082998275757, + "learning_rate": 2.4626119545225337e-06, + "loss": 0.5712, + "step": 3705 + }, + { + "epoch": 3.06791827719492, + "grad_norm": 0.9370817542076111, + "learning_rate": 2.4615123917679737e-06, + "loss": 0.5737, + "step": 3706 + }, + { + "epoch": 3.068746548868029, + "grad_norm": 0.984732985496521, + "learning_rate": 2.460412836460407e-06, + "loss": 0.5631, + "step": 3707 + }, + { + "epoch": 3.0695748205411375, + "grad_norm": 0.90977942943573, + "learning_rate": 2.4593132888125877e-06, + "loss": 0.5755, + "step": 3708 + }, + { + "epoch": 3.0704030922142462, + "grad_norm": 0.9592271447181702, + "learning_rate": 2.4582137490372676e-06, + "loss": 0.552, + "step": 3709 + }, + { + "epoch": 3.071231363887355, + "grad_norm": 0.9560198187828064, + "learning_rate": 2.457114217347197e-06, + "loss": 0.5744, + "step": 3710 + }, + { + "epoch": 3.0720596355604637, + "grad_norm": 0.9570426940917969, + "learning_rate": 2.4560146939551268e-06, + "loss": 0.5769, + "step": 3711 + }, + { + "epoch": 3.0728879072335724, + "grad_norm": 0.9208659529685974, + "learning_rate": 2.4549151790738025e-06, + "loss": 0.5697, + "step": 3712 + }, + { + "epoch": 3.0737161789066816, + "grad_norm": 0.9417058229446411, + "learning_rate": 2.453815672915971e-06, + "loss": 0.5697, + "step": 3713 + }, + { + "epoch": 3.0745444505797903, + "grad_norm": 0.9602729082107544, + "learning_rate": 2.4527161756943766e-06, + "loss": 0.5737, + "step": 3714 + }, + { + "epoch": 3.075372722252899, + "grad_norm": 0.931280255317688, + "learning_rate": 2.4516166876217603e-06, + "loss": 0.5704, + "step": 3715 + }, + { + "epoch": 3.0762009939260078, + "grad_norm": 0.920081377029419, + "learning_rate": 2.4505172089108648e-06, + "loss": 0.57, + "step": 3716 + }, + { + "epoch": 3.0770292655991165, + "grad_norm": 0.929806649684906, + "learning_rate": 2.449417739774428e-06, + "loss": 0.5648, + "step": 3717 + }, + { + "epoch": 3.077857537272225, + "grad_norm": 0.9344229698181152, + "learning_rate": 2.4483182804251866e-06, + "loss": 0.5666, + "step": 3718 + }, + { + "epoch": 3.078685808945334, + "grad_norm": 0.9535424113273621, + "learning_rate": 2.4472188310758766e-06, + "loss": 0.5777, + "step": 3719 + }, + { + "epoch": 3.0795140806184427, + "grad_norm": 0.9594505429267883, + "learning_rate": 2.4461193919392307e-06, + "loss": 0.5614, + "step": 3720 + }, + { + "epoch": 3.080342352291552, + "grad_norm": 0.9511420726776123, + "learning_rate": 2.4450199632279797e-06, + "loss": 0.5925, + "step": 3721 + }, + { + "epoch": 3.0811706239646606, + "grad_norm": 0.9244424104690552, + "learning_rate": 2.443920545154855e-06, + "loss": 0.5694, + "step": 3722 + }, + { + "epoch": 3.0819988956377693, + "grad_norm": 0.9384777545928955, + "learning_rate": 2.4428211379325804e-06, + "loss": 0.5443, + "step": 3723 + }, + { + "epoch": 3.082827167310878, + "grad_norm": 0.9536921381950378, + "learning_rate": 2.441721741773883e-06, + "loss": 0.5817, + "step": 3724 + }, + { + "epoch": 3.0836554389839868, + "grad_norm": 0.9547799825668335, + "learning_rate": 2.4406223568914854e-06, + "loss": 0.5565, + "step": 3725 + }, + { + "epoch": 3.0844837106570955, + "grad_norm": 0.9509586691856384, + "learning_rate": 2.4395229834981076e-06, + "loss": 0.5712, + "step": 3726 + }, + { + "epoch": 3.085311982330204, + "grad_norm": 0.9227497577667236, + "learning_rate": 2.4384236218064692e-06, + "loss": 0.5602, + "step": 3727 + }, + { + "epoch": 3.086140254003313, + "grad_norm": 0.9162245392799377, + "learning_rate": 2.4373242720292865e-06, + "loss": 0.555, + "step": 3728 + }, + { + "epoch": 3.0869685256764217, + "grad_norm": 0.9121986627578735, + "learning_rate": 2.4362249343792722e-06, + "loss": 0.5748, + "step": 3729 + }, + { + "epoch": 3.087796797349531, + "grad_norm": 0.9310787320137024, + "learning_rate": 2.435125609069139e-06, + "loss": 0.5671, + "step": 3730 + }, + { + "epoch": 3.0886250690226396, + "grad_norm": 0.9523546695709229, + "learning_rate": 2.4340262963115955e-06, + "loss": 0.595, + "step": 3731 + }, + { + "epoch": 3.0894533406957483, + "grad_norm": 0.9429517388343811, + "learning_rate": 2.432926996319348e-06, + "loss": 0.5381, + "step": 3732 + }, + { + "epoch": 3.090281612368857, + "grad_norm": 0.9443867802619934, + "learning_rate": 2.4318277093051023e-06, + "loss": 0.5588, + "step": 3733 + }, + { + "epoch": 3.0911098840419657, + "grad_norm": 0.9320855140686035, + "learning_rate": 2.43072843548156e-06, + "loss": 0.5765, + "step": 3734 + }, + { + "epoch": 3.0919381557150745, + "grad_norm": 0.9335711002349854, + "learning_rate": 2.429629175061419e-06, + "loss": 0.5734, + "step": 3735 + }, + { + "epoch": 3.092766427388183, + "grad_norm": 0.9539975523948669, + "learning_rate": 2.4285299282573767e-06, + "loss": 0.5733, + "step": 3736 + }, + { + "epoch": 3.093594699061292, + "grad_norm": 0.9393835067749023, + "learning_rate": 2.4274306952821262e-06, + "loss": 0.5808, + "step": 3737 + }, + { + "epoch": 3.094422970734401, + "grad_norm": 0.9707219004631042, + "learning_rate": 2.4263314763483607e-06, + "loss": 0.5755, + "step": 3738 + }, + { + "epoch": 3.09525124240751, + "grad_norm": 0.9541774988174438, + "learning_rate": 2.4252322716687676e-06, + "loss": 0.5472, + "step": 3739 + }, + { + "epoch": 3.0960795140806185, + "grad_norm": 0.9300408959388733, + "learning_rate": 2.424133081456032e-06, + "loss": 0.5771, + "step": 3740 + }, + { + "epoch": 3.0969077857537273, + "grad_norm": 0.9487392902374268, + "learning_rate": 2.423033905922839e-06, + "loss": 0.5805, + "step": 3741 + }, + { + "epoch": 3.097736057426836, + "grad_norm": 0.9021924734115601, + "learning_rate": 2.4219347452818678e-06, + "loss": 0.5648, + "step": 3742 + }, + { + "epoch": 3.0985643290999447, + "grad_norm": 0.9229711294174194, + "learning_rate": 2.420835599745795e-06, + "loss": 0.5569, + "step": 3743 + }, + { + "epoch": 3.0993926007730535, + "grad_norm": 0.9335191249847412, + "learning_rate": 2.419736469527296e-06, + "loss": 0.5728, + "step": 3744 + }, + { + "epoch": 3.100220872446162, + "grad_norm": 0.9294394254684448, + "learning_rate": 2.4186373548390427e-06, + "loss": 0.5522, + "step": 3745 + }, + { + "epoch": 3.101049144119271, + "grad_norm": 0.931654691696167, + "learning_rate": 2.417538255893702e-06, + "loss": 0.5747, + "step": 3746 + }, + { + "epoch": 3.10187741579238, + "grad_norm": 0.9018174409866333, + "learning_rate": 2.4164391729039405e-06, + "loss": 0.5697, + "step": 3747 + }, + { + "epoch": 3.102705687465489, + "grad_norm": 0.9389755129814148, + "learning_rate": 2.4153401060824193e-06, + "loss": 0.5559, + "step": 3748 + }, + { + "epoch": 3.1035339591385975, + "grad_norm": 0.9191637635231018, + "learning_rate": 2.414241055641799e-06, + "loss": 0.5678, + "step": 3749 + }, + { + "epoch": 3.1043622308117063, + "grad_norm": 0.9245947599411011, + "learning_rate": 2.4131420217947347e-06, + "loss": 0.5677, + "step": 3750 + }, + { + "epoch": 3.105190502484815, + "grad_norm": 0.9380972981452942, + "learning_rate": 2.4120430047538797e-06, + "loss": 0.5611, + "step": 3751 + }, + { + "epoch": 3.1060187741579237, + "grad_norm": 0.9427217841148376, + "learning_rate": 2.410944004731883e-06, + "loss": 0.592, + "step": 3752 + }, + { + "epoch": 3.1068470458310324, + "grad_norm": 0.9142959117889404, + "learning_rate": 2.4098450219413915e-06, + "loss": 0.5588, + "step": 3753 + }, + { + "epoch": 3.107675317504141, + "grad_norm": 0.9501208066940308, + "learning_rate": 2.4087460565950475e-06, + "loss": 0.5735, + "step": 3754 + }, + { + "epoch": 3.1085035891772503, + "grad_norm": 0.9072937369346619, + "learning_rate": 2.4076471089054916e-06, + "loss": 0.557, + "step": 3755 + }, + { + "epoch": 3.109331860850359, + "grad_norm": 0.9310604333877563, + "learning_rate": 2.4065481790853595e-06, + "loss": 0.5616, + "step": 3756 + }, + { + "epoch": 3.110160132523468, + "grad_norm": 0.9229986071586609, + "learning_rate": 2.405449267347283e-06, + "loss": 0.5806, + "step": 3757 + }, + { + "epoch": 3.1109884041965765, + "grad_norm": 0.9491189122200012, + "learning_rate": 2.4043503739038922e-06, + "loss": 0.571, + "step": 3758 + }, + { + "epoch": 3.1118166758696852, + "grad_norm": 0.9373705983161926, + "learning_rate": 2.4032514989678113e-06, + "loss": 0.557, + "step": 3759 + }, + { + "epoch": 3.112644947542794, + "grad_norm": 0.9476242661476135, + "learning_rate": 2.4021526427516647e-06, + "loss": 0.5749, + "step": 3760 + }, + { + "epoch": 3.1134732192159027, + "grad_norm": 0.9147936701774597, + "learning_rate": 2.401053805468069e-06, + "loss": 0.5579, + "step": 3761 + }, + { + "epoch": 3.1143014908890114, + "grad_norm": 0.9299486875534058, + "learning_rate": 2.399954987329639e-06, + "loss": 0.5707, + "step": 3762 + }, + { + "epoch": 3.1151297625621206, + "grad_norm": 0.9515376091003418, + "learning_rate": 2.3988561885489874e-06, + "loss": 0.5923, + "step": 3763 + }, + { + "epoch": 3.1159580342352293, + "grad_norm": 0.9455316662788391, + "learning_rate": 2.39775740933872e-06, + "loss": 0.5686, + "step": 3764 + }, + { + "epoch": 3.116786305908338, + "grad_norm": 0.952534019947052, + "learning_rate": 2.3966586499114406e-06, + "loss": 0.581, + "step": 3765 + }, + { + "epoch": 3.1176145775814468, + "grad_norm": 0.9540290832519531, + "learning_rate": 2.3955599104797493e-06, + "loss": 0.5612, + "step": 3766 + }, + { + "epoch": 3.1184428492545555, + "grad_norm": 0.9238110780715942, + "learning_rate": 2.3944611912562423e-06, + "loss": 0.5608, + "step": 3767 + }, + { + "epoch": 3.1192711209276642, + "grad_norm": 0.9312852025032043, + "learning_rate": 2.3933624924535113e-06, + "loss": 0.5752, + "step": 3768 + }, + { + "epoch": 3.120099392600773, + "grad_norm": 0.9356149435043335, + "learning_rate": 2.3922638142841433e-06, + "loss": 0.5719, + "step": 3769 + }, + { + "epoch": 3.1209276642738817, + "grad_norm": 0.9629190564155579, + "learning_rate": 2.391165156960722e-06, + "loss": 0.5791, + "step": 3770 + }, + { + "epoch": 3.1217559359469904, + "grad_norm": 0.9337695240974426, + "learning_rate": 2.3900665206958294e-06, + "loss": 0.5636, + "step": 3771 + }, + { + "epoch": 3.1225842076200996, + "grad_norm": 0.9723961353302002, + "learning_rate": 2.3889679057020402e-06, + "loss": 0.5897, + "step": 3772 + }, + { + "epoch": 3.1234124792932083, + "grad_norm": 0.9874154925346375, + "learning_rate": 2.3878693121919255e-06, + "loss": 0.5846, + "step": 3773 + }, + { + "epoch": 3.124240750966317, + "grad_norm": 0.9433327913284302, + "learning_rate": 2.386770740378054e-06, + "loss": 0.5631, + "step": 3774 + }, + { + "epoch": 3.1250690226394258, + "grad_norm": 0.9409958124160767, + "learning_rate": 2.385672190472989e-06, + "loss": 0.548, + "step": 3775 + }, + { + "epoch": 3.1258972943125345, + "grad_norm": 0.9370530247688293, + "learning_rate": 2.3845736626892895e-06, + "loss": 0.5277, + "step": 3776 + }, + { + "epoch": 3.126725565985643, + "grad_norm": 0.9603546857833862, + "learning_rate": 2.3834751572395097e-06, + "loss": 0.5635, + "step": 3777 + }, + { + "epoch": 3.127553837658752, + "grad_norm": 0.9234820008277893, + "learning_rate": 2.382376674336201e-06, + "loss": 0.5743, + "step": 3778 + }, + { + "epoch": 3.1283821093318607, + "grad_norm": 0.9674520492553711, + "learning_rate": 2.38127821419191e-06, + "loss": 0.5694, + "step": 3779 + }, + { + "epoch": 3.1292103810049694, + "grad_norm": 0.9508289098739624, + "learning_rate": 2.380179777019177e-06, + "loss": 0.5625, + "step": 3780 + }, + { + "epoch": 3.1300386526780786, + "grad_norm": 0.9463353753089905, + "learning_rate": 2.3790813630305403e-06, + "loss": 0.5859, + "step": 3781 + }, + { + "epoch": 3.1308669243511873, + "grad_norm": 0.9280157089233398, + "learning_rate": 2.3779829724385327e-06, + "loss": 0.5632, + "step": 3782 + }, + { + "epoch": 3.131695196024296, + "grad_norm": 0.931016206741333, + "learning_rate": 2.376884605455683e-06, + "loss": 0.556, + "step": 3783 + }, + { + "epoch": 3.1325234676974048, + "grad_norm": 0.9319890141487122, + "learning_rate": 2.375786262294514e-06, + "loss": 0.563, + "step": 3784 + }, + { + "epoch": 3.1333517393705135, + "grad_norm": 0.9274983406066895, + "learning_rate": 2.374687943167545e-06, + "loss": 0.5713, + "step": 3785 + }, + { + "epoch": 3.134180011043622, + "grad_norm": 0.9298895001411438, + "learning_rate": 2.3735896482872917e-06, + "loss": 0.5705, + "step": 3786 + }, + { + "epoch": 3.135008282716731, + "grad_norm": 0.9218611121177673, + "learning_rate": 2.3724913778662624e-06, + "loss": 0.5738, + "step": 3787 + }, + { + "epoch": 3.13583655438984, + "grad_norm": 0.9677892327308655, + "learning_rate": 2.371393132116963e-06, + "loss": 0.5675, + "step": 3788 + }, + { + "epoch": 3.136664826062949, + "grad_norm": 0.9483129978179932, + "learning_rate": 2.3702949112518935e-06, + "loss": 0.5525, + "step": 3789 + }, + { + "epoch": 3.1374930977360576, + "grad_norm": 0.9588272571563721, + "learning_rate": 2.3691967154835505e-06, + "loss": 0.5819, + "step": 3790 + }, + { + "epoch": 3.1383213694091663, + "grad_norm": 0.9432960748672485, + "learning_rate": 2.3680985450244233e-06, + "loss": 0.5757, + "step": 3791 + }, + { + "epoch": 3.139149641082275, + "grad_norm": 0.952733039855957, + "learning_rate": 2.367000400086998e-06, + "loss": 0.5807, + "step": 3792 + }, + { + "epoch": 3.1399779127553837, + "grad_norm": 0.9517266154289246, + "learning_rate": 2.365902280883755e-06, + "loss": 0.5503, + "step": 3793 + }, + { + "epoch": 3.1408061844284925, + "grad_norm": 0.9321509599685669, + "learning_rate": 2.364804187627171e-06, + "loss": 0.5534, + "step": 3794 + }, + { + "epoch": 3.141634456101601, + "grad_norm": 0.9490827918052673, + "learning_rate": 2.3637061205297167e-06, + "loss": 0.582, + "step": 3795 + }, + { + "epoch": 3.14246272777471, + "grad_norm": 0.9612736701965332, + "learning_rate": 2.3626080798038565e-06, + "loss": 0.5673, + "step": 3796 + }, + { + "epoch": 3.143290999447819, + "grad_norm": 0.9634802937507629, + "learning_rate": 2.361510065662053e-06, + "loss": 0.5784, + "step": 3797 + }, + { + "epoch": 3.144119271120928, + "grad_norm": 0.9495031237602234, + "learning_rate": 2.360412078316761e-06, + "loss": 0.5661, + "step": 3798 + }, + { + "epoch": 3.1449475427940365, + "grad_norm": 0.9234485030174255, + "learning_rate": 2.3593141179804295e-06, + "loss": 0.5718, + "step": 3799 + }, + { + "epoch": 3.1457758144671453, + "grad_norm": 0.9367272257804871, + "learning_rate": 2.3582161848655054e-06, + "loss": 0.5332, + "step": 3800 + }, + { + "epoch": 3.146604086140254, + "grad_norm": 0.9217962622642517, + "learning_rate": 2.357118279184429e-06, + "loss": 0.586, + "step": 3801 + }, + { + "epoch": 3.1474323578133627, + "grad_norm": 0.943759560585022, + "learning_rate": 2.3560204011496324e-06, + "loss": 0.5606, + "step": 3802 + }, + { + "epoch": 3.1482606294864715, + "grad_norm": 0.959769070148468, + "learning_rate": 2.354922550973546e-06, + "loss": 0.5858, + "step": 3803 + }, + { + "epoch": 3.14908890115958, + "grad_norm": 0.9223443865776062, + "learning_rate": 2.3538247288685935e-06, + "loss": 0.5924, + "step": 3804 + }, + { + "epoch": 3.149917172832689, + "grad_norm": 0.958400547504425, + "learning_rate": 2.3527269350471936e-06, + "loss": 0.5669, + "step": 3805 + }, + { + "epoch": 3.150745444505798, + "grad_norm": 0.9326207041740417, + "learning_rate": 2.351629169721759e-06, + "loss": 0.5691, + "step": 3806 + }, + { + "epoch": 3.151573716178907, + "grad_norm": 0.9581301808357239, + "learning_rate": 2.3505314331046964e-06, + "loss": 0.5767, + "step": 3807 + }, + { + "epoch": 3.1524019878520155, + "grad_norm": 0.921093225479126, + "learning_rate": 2.349433725408409e-06, + "loss": 0.5773, + "step": 3808 + }, + { + "epoch": 3.1532302595251243, + "grad_norm": 0.9264110326766968, + "learning_rate": 2.348336046845292e-06, + "loss": 0.5388, + "step": 3809 + }, + { + "epoch": 3.154058531198233, + "grad_norm": 0.9559600949287415, + "learning_rate": 2.3472383976277353e-06, + "loss": 0.5601, + "step": 3810 + }, + { + "epoch": 3.1548868028713417, + "grad_norm": 0.8962564468383789, + "learning_rate": 2.3461407779681255e-06, + "loss": 0.5718, + "step": 3811 + }, + { + "epoch": 3.1557150745444504, + "grad_norm": 0.9371088743209839, + "learning_rate": 2.345043188078841e-06, + "loss": 0.5841, + "step": 3812 + }, + { + "epoch": 3.156543346217559, + "grad_norm": 0.952690839767456, + "learning_rate": 2.343945628172255e-06, + "loss": 0.5517, + "step": 3813 + }, + { + "epoch": 3.1573716178906683, + "grad_norm": 0.9315347075462341, + "learning_rate": 2.3428480984607356e-06, + "loss": 0.5773, + "step": 3814 + }, + { + "epoch": 3.158199889563777, + "grad_norm": 0.9349079728126526, + "learning_rate": 2.341750599156643e-06, + "loss": 0.568, + "step": 3815 + }, + { + "epoch": 3.159028161236886, + "grad_norm": 0.9342600703239441, + "learning_rate": 2.340653130472335e-06, + "loss": 0.5495, + "step": 3816 + }, + { + "epoch": 3.1598564329099945, + "grad_norm": 0.9394509792327881, + "learning_rate": 2.339555692620161e-06, + "loss": 0.6008, + "step": 3817 + }, + { + "epoch": 3.1606847045831032, + "grad_norm": 0.9620388150215149, + "learning_rate": 2.338458285812464e-06, + "loss": 0.5652, + "step": 3818 + }, + { + "epoch": 3.161512976256212, + "grad_norm": 0.9644536375999451, + "learning_rate": 2.3373609102615834e-06, + "loss": 0.5556, + "step": 3819 + }, + { + "epoch": 3.1623412479293207, + "grad_norm": 0.9285350441932678, + "learning_rate": 2.3362635661798507e-06, + "loss": 0.5758, + "step": 3820 + }, + { + "epoch": 3.1631695196024294, + "grad_norm": 0.9304336905479431, + "learning_rate": 2.335166253779591e-06, + "loss": 0.6108, + "step": 3821 + }, + { + "epoch": 3.1639977912755386, + "grad_norm": 0.9232069253921509, + "learning_rate": 2.334068973273125e-06, + "loss": 0.5835, + "step": 3822 + }, + { + "epoch": 3.1648260629486473, + "grad_norm": 0.9440987706184387, + "learning_rate": 2.3329717248727665e-06, + "loss": 0.5778, + "step": 3823 + }, + { + "epoch": 3.165654334621756, + "grad_norm": 0.9345200657844543, + "learning_rate": 2.3318745087908225e-06, + "loss": 0.56, + "step": 3824 + }, + { + "epoch": 3.1664826062948648, + "grad_norm": 0.9822887778282166, + "learning_rate": 2.330777325239593e-06, + "loss": 0.556, + "step": 3825 + }, + { + "epoch": 3.1673108779679735, + "grad_norm": 0.9316609501838684, + "learning_rate": 2.3296801744313733e-06, + "loss": 0.5749, + "step": 3826 + }, + { + "epoch": 3.1681391496410822, + "grad_norm": 0.933875322341919, + "learning_rate": 2.3285830565784528e-06, + "loss": 0.5686, + "step": 3827 + }, + { + "epoch": 3.168967421314191, + "grad_norm": 0.9843499660491943, + "learning_rate": 2.327485971893112e-06, + "loss": 0.6005, + "step": 3828 + }, + { + "epoch": 3.1697956929872997, + "grad_norm": 0.9594619870185852, + "learning_rate": 2.326388920587628e-06, + "loss": 0.5754, + "step": 3829 + }, + { + "epoch": 3.1706239646604084, + "grad_norm": 0.959398090839386, + "learning_rate": 2.3252919028742698e-06, + "loss": 0.5762, + "step": 3830 + }, + { + "epoch": 3.1714522363335176, + "grad_norm": 0.9445636868476868, + "learning_rate": 2.3241949189652996e-06, + "loss": 0.5757, + "step": 3831 + }, + { + "epoch": 3.1722805080066263, + "grad_norm": 0.9507540464401245, + "learning_rate": 2.323097969072973e-06, + "loss": 0.5695, + "step": 3832 + }, + { + "epoch": 3.173108779679735, + "grad_norm": 0.9695854187011719, + "learning_rate": 2.322001053409541e-06, + "loss": 0.5665, + "step": 3833 + }, + { + "epoch": 3.1739370513528438, + "grad_norm": 0.9487887620925903, + "learning_rate": 2.3209041721872464e-06, + "loss": 0.5651, + "step": 3834 + }, + { + "epoch": 3.1747653230259525, + "grad_norm": 0.9431695938110352, + "learning_rate": 2.319807325618325e-06, + "loss": 0.5516, + "step": 3835 + }, + { + "epoch": 3.175593594699061, + "grad_norm": 0.9398961067199707, + "learning_rate": 2.318710513915006e-06, + "loss": 0.571, + "step": 3836 + }, + { + "epoch": 3.17642186637217, + "grad_norm": 0.9242047667503357, + "learning_rate": 2.317613737289512e-06, + "loss": 0.5557, + "step": 3837 + }, + { + "epoch": 3.1772501380452787, + "grad_norm": 0.9494120478630066, + "learning_rate": 2.31651699595406e-06, + "loss": 0.568, + "step": 3838 + }, + { + "epoch": 3.1780784097183874, + "grad_norm": 0.9196777939796448, + "learning_rate": 2.3154202901208593e-06, + "loss": 0.5656, + "step": 3839 + }, + { + "epoch": 3.1789066813914966, + "grad_norm": 0.9382702112197876, + "learning_rate": 2.314323620002111e-06, + "loss": 0.5792, + "step": 3840 + }, + { + "epoch": 3.1797349530646053, + "grad_norm": 1.018182635307312, + "learning_rate": 2.3132269858100118e-06, + "loss": 0.5804, + "step": 3841 + }, + { + "epoch": 3.180563224737714, + "grad_norm": 0.9466580748558044, + "learning_rate": 2.31213038775675e-06, + "loss": 0.565, + "step": 3842 + }, + { + "epoch": 3.1813914964108228, + "grad_norm": 0.9668580889701843, + "learning_rate": 2.3110338260545062e-06, + "loss": 0.5724, + "step": 3843 + }, + { + "epoch": 3.1822197680839315, + "grad_norm": 0.9272748231887817, + "learning_rate": 2.309937300915456e-06, + "loss": 0.579, + "step": 3844 + }, + { + "epoch": 3.18304803975704, + "grad_norm": 0.9182299375534058, + "learning_rate": 2.3088408125517666e-06, + "loss": 0.5906, + "step": 3845 + }, + { + "epoch": 3.183876311430149, + "grad_norm": 1.034818410873413, + "learning_rate": 2.307744361175598e-06, + "loss": 0.5709, + "step": 3846 + }, + { + "epoch": 3.184704583103258, + "grad_norm": 0.9287042617797852, + "learning_rate": 2.306647946999104e-06, + "loss": 0.5855, + "step": 3847 + }, + { + "epoch": 3.185532854776367, + "grad_norm": 0.9378973841667175, + "learning_rate": 2.305551570234428e-06, + "loss": 0.5822, + "step": 3848 + }, + { + "epoch": 3.1863611264494756, + "grad_norm": 0.9447490572929382, + "learning_rate": 2.304455231093712e-06, + "loss": 0.5781, + "step": 3849 + }, + { + "epoch": 3.1871893981225843, + "grad_norm": 0.9659326076507568, + "learning_rate": 2.303358929789085e-06, + "loss": 0.6217, + "step": 3850 + }, + { + "epoch": 3.188017669795693, + "grad_norm": 0.9089621305465698, + "learning_rate": 2.302262666532672e-06, + "loss": 0.5733, + "step": 3851 + }, + { + "epoch": 3.1888459414688017, + "grad_norm": 0.9480805993080139, + "learning_rate": 2.3011664415365905e-06, + "loss": 0.5615, + "step": 3852 + }, + { + "epoch": 3.1896742131419105, + "grad_norm": 0.9534367322921753, + "learning_rate": 2.300070255012949e-06, + "loss": 0.5776, + "step": 3853 + }, + { + "epoch": 3.190502484815019, + "grad_norm": 0.9526930451393127, + "learning_rate": 2.298974107173849e-06, + "loss": 0.5541, + "step": 3854 + }, + { + "epoch": 3.191330756488128, + "grad_norm": 0.9261385202407837, + "learning_rate": 2.2978779982313853e-06, + "loss": 0.5797, + "step": 3855 + }, + { + "epoch": 3.192159028161237, + "grad_norm": 0.9275192618370056, + "learning_rate": 2.296781928397645e-06, + "loss": 0.559, + "step": 3856 + }, + { + "epoch": 3.192987299834346, + "grad_norm": 0.9341071248054504, + "learning_rate": 2.2956858978847076e-06, + "loss": 0.5523, + "step": 3857 + }, + { + "epoch": 3.1938155715074545, + "grad_norm": 0.9231052398681641, + "learning_rate": 2.2945899069046444e-06, + "loss": 0.5663, + "step": 3858 + }, + { + "epoch": 3.1946438431805633, + "grad_norm": 0.9468841552734375, + "learning_rate": 2.293493955669519e-06, + "loss": 0.5815, + "step": 3859 + }, + { + "epoch": 3.195472114853672, + "grad_norm": 0.9482542276382446, + "learning_rate": 2.2923980443913876e-06, + "loss": 0.5679, + "step": 3860 + }, + { + "epoch": 3.1963003865267807, + "grad_norm": 0.9444701075553894, + "learning_rate": 2.2913021732822997e-06, + "loss": 0.5796, + "step": 3861 + }, + { + "epoch": 3.1971286581998895, + "grad_norm": 0.9483172297477722, + "learning_rate": 2.2902063425542963e-06, + "loss": 0.5391, + "step": 3862 + }, + { + "epoch": 3.197956929872998, + "grad_norm": 0.8964424729347229, + "learning_rate": 2.289110552419409e-06, + "loss": 0.5719, + "step": 3863 + }, + { + "epoch": 3.198785201546107, + "grad_norm": 0.9366036653518677, + "learning_rate": 2.2880148030896648e-06, + "loss": 0.5729, + "step": 3864 + }, + { + "epoch": 3.199613473219216, + "grad_norm": 0.9433067440986633, + "learning_rate": 2.28691909477708e-06, + "loss": 0.5811, + "step": 3865 + }, + { + "epoch": 3.200441744892325, + "grad_norm": 1.0146312713623047, + "learning_rate": 2.285823427693664e-06, + "loss": 0.5524, + "step": 3866 + }, + { + "epoch": 3.2012700165654335, + "grad_norm": 0.9418644905090332, + "learning_rate": 2.2847278020514176e-06, + "loss": 0.5687, + "step": 3867 + }, + { + "epoch": 3.2020982882385423, + "grad_norm": 0.9421892166137695, + "learning_rate": 2.2836322180623355e-06, + "loss": 0.559, + "step": 3868 + }, + { + "epoch": 3.202926559911651, + "grad_norm": 0.9407430291175842, + "learning_rate": 2.2825366759384027e-06, + "loss": 0.5876, + "step": 3869 + }, + { + "epoch": 3.2037548315847597, + "grad_norm": 0.9176812767982483, + "learning_rate": 2.281441175891595e-06, + "loss": 0.5514, + "step": 3870 + }, + { + "epoch": 3.2045831032578684, + "grad_norm": 0.9171411991119385, + "learning_rate": 2.2803457181338825e-06, + "loss": 0.5778, + "step": 3871 + }, + { + "epoch": 3.205411374930977, + "grad_norm": 0.9477943778038025, + "learning_rate": 2.279250302877226e-06, + "loss": 0.5806, + "step": 3872 + }, + { + "epoch": 3.2062396466040863, + "grad_norm": 0.9568420052528381, + "learning_rate": 2.2781549303335783e-06, + "loss": 0.6025, + "step": 3873 + }, + { + "epoch": 3.207067918277195, + "grad_norm": 0.9442486763000488, + "learning_rate": 2.2770596007148833e-06, + "loss": 0.5678, + "step": 3874 + }, + { + "epoch": 3.207896189950304, + "grad_norm": 0.9460006356239319, + "learning_rate": 2.2759643142330775e-06, + "loss": 0.5651, + "step": 3875 + }, + { + "epoch": 3.2087244616234125, + "grad_norm": 0.9530642032623291, + "learning_rate": 2.2748690711000885e-06, + "loss": 0.5704, + "step": 3876 + }, + { + "epoch": 3.2095527332965212, + "grad_norm": 0.9378021955490112, + "learning_rate": 2.2737738715278353e-06, + "loss": 0.5605, + "step": 3877 + }, + { + "epoch": 3.21038100496963, + "grad_norm": 0.9239064455032349, + "learning_rate": 2.272678715728229e-06, + "loss": 0.5514, + "step": 3878 + }, + { + "epoch": 3.2112092766427387, + "grad_norm": 0.9366111755371094, + "learning_rate": 2.271583603913173e-06, + "loss": 0.5746, + "step": 3879 + }, + { + "epoch": 3.2120375483158474, + "grad_norm": 0.9362907409667969, + "learning_rate": 2.270488536294559e-06, + "loss": 0.5622, + "step": 3880 + }, + { + "epoch": 3.2128658199889566, + "grad_norm": 0.9395036697387695, + "learning_rate": 2.2693935130842763e-06, + "loss": 0.5765, + "step": 3881 + }, + { + "epoch": 3.2136940916620653, + "grad_norm": 0.9549781084060669, + "learning_rate": 2.2682985344941966e-06, + "loss": 0.5716, + "step": 3882 + }, + { + "epoch": 3.214522363335174, + "grad_norm": 0.9677711129188538, + "learning_rate": 2.2672036007361915e-06, + "loss": 0.5922, + "step": 3883 + }, + { + "epoch": 3.2153506350082828, + "grad_norm": 0.9358668923377991, + "learning_rate": 2.26610871202212e-06, + "loss": 0.5794, + "step": 3884 + }, + { + "epoch": 3.2161789066813915, + "grad_norm": 0.938670814037323, + "learning_rate": 2.265013868563831e-06, + "loss": 0.5809, + "step": 3885 + }, + { + "epoch": 3.2170071783545002, + "grad_norm": 0.9247921109199524, + "learning_rate": 2.2639190705731694e-06, + "loss": 0.5714, + "step": 3886 + }, + { + "epoch": 3.217835450027609, + "grad_norm": 0.9363803863525391, + "learning_rate": 2.262824318261966e-06, + "loss": 0.5714, + "step": 3887 + }, + { + "epoch": 3.2186637217007177, + "grad_norm": 0.9042513370513916, + "learning_rate": 2.2617296118420463e-06, + "loss": 0.5612, + "step": 3888 + }, + { + "epoch": 3.2194919933738264, + "grad_norm": 0.9226694703102112, + "learning_rate": 2.260634951525226e-06, + "loss": 0.5579, + "step": 3889 + }, + { + "epoch": 3.2203202650469356, + "grad_norm": 0.9465897679328918, + "learning_rate": 2.2595403375233107e-06, + "loss": 0.5507, + "step": 3890 + }, + { + "epoch": 3.2211485367200443, + "grad_norm": 0.9200103282928467, + "learning_rate": 2.2584457700480984e-06, + "loss": 0.568, + "step": 3891 + }, + { + "epoch": 3.221976808393153, + "grad_norm": 0.9169338345527649, + "learning_rate": 2.2573512493113797e-06, + "loss": 0.5588, + "step": 3892 + }, + { + "epoch": 3.2228050800662618, + "grad_norm": 0.9455548524856567, + "learning_rate": 2.2562567755249304e-06, + "loss": 0.5725, + "step": 3893 + }, + { + "epoch": 3.2236333517393705, + "grad_norm": 0.9347724318504333, + "learning_rate": 2.2551623489005233e-06, + "loss": 0.5659, + "step": 3894 + }, + { + "epoch": 3.224461623412479, + "grad_norm": 0.9664618372917175, + "learning_rate": 2.2540679696499194e-06, + "loss": 0.5584, + "step": 3895 + }, + { + "epoch": 3.225289895085588, + "grad_norm": 1.0180222988128662, + "learning_rate": 2.252973637984871e-06, + "loss": 0.5586, + "step": 3896 + }, + { + "epoch": 3.2261181667586967, + "grad_norm": 0.9532806277275085, + "learning_rate": 2.2518793541171204e-06, + "loss": 0.5583, + "step": 3897 + }, + { + "epoch": 3.2269464384318054, + "grad_norm": 0.9502059817314148, + "learning_rate": 2.2507851182584024e-06, + "loss": 0.5676, + "step": 3898 + }, + { + "epoch": 3.2277747101049146, + "grad_norm": 0.9380828738212585, + "learning_rate": 2.24969093062044e-06, + "loss": 0.5644, + "step": 3899 + }, + { + "epoch": 3.2286029817780233, + "grad_norm": 0.9528313875198364, + "learning_rate": 2.2485967914149503e-06, + "loss": 0.5704, + "step": 3900 + }, + { + "epoch": 3.229431253451132, + "grad_norm": 0.9411875009536743, + "learning_rate": 2.2475027008536374e-06, + "loss": 0.5694, + "step": 3901 + }, + { + "epoch": 3.2302595251242407, + "grad_norm": 0.9563047289848328, + "learning_rate": 2.2464086591481977e-06, + "loss": 0.5654, + "step": 3902 + }, + { + "epoch": 3.2310877967973495, + "grad_norm": 0.9029218554496765, + "learning_rate": 2.24531466651032e-06, + "loss": 0.5818, + "step": 3903 + }, + { + "epoch": 3.231916068470458, + "grad_norm": 0.9468775391578674, + "learning_rate": 2.244220723151679e-06, + "loss": 0.548, + "step": 3904 + }, + { + "epoch": 3.232744340143567, + "grad_norm": 0.9399493932723999, + "learning_rate": 2.243126829283944e-06, + "loss": 0.5746, + "step": 3905 + }, + { + "epoch": 3.233572611816676, + "grad_norm": 0.9424317479133606, + "learning_rate": 2.2420329851187726e-06, + "loss": 0.5775, + "step": 3906 + }, + { + "epoch": 3.234400883489785, + "grad_norm": 0.9486375451087952, + "learning_rate": 2.2409391908678137e-06, + "loss": 0.5815, + "step": 3907 + }, + { + "epoch": 3.2352291551628936, + "grad_norm": 0.9591338634490967, + "learning_rate": 2.2398454467427073e-06, + "loss": 0.5826, + "step": 3908 + }, + { + "epoch": 3.2360574268360023, + "grad_norm": 0.9704396724700928, + "learning_rate": 2.2387517529550813e-06, + "loss": 0.5912, + "step": 3909 + }, + { + "epoch": 3.236885698509111, + "grad_norm": 0.9416005611419678, + "learning_rate": 2.2376581097165553e-06, + "loss": 0.5614, + "step": 3910 + }, + { + "epoch": 3.2377139701822197, + "grad_norm": 1.0255080461502075, + "learning_rate": 2.2365645172387403e-06, + "loss": 0.5688, + "step": 3911 + }, + { + "epoch": 3.2385422418553285, + "grad_norm": 0.9364955425262451, + "learning_rate": 2.2354709757332356e-06, + "loss": 0.572, + "step": 3912 + }, + { + "epoch": 3.239370513528437, + "grad_norm": 0.9503747224807739, + "learning_rate": 2.2343774854116306e-06, + "loss": 0.5893, + "step": 3913 + }, + { + "epoch": 3.240198785201546, + "grad_norm": 0.9294464588165283, + "learning_rate": 2.2332840464855072e-06, + "loss": 0.5671, + "step": 3914 + }, + { + "epoch": 3.241027056874655, + "grad_norm": 0.9659671783447266, + "learning_rate": 2.232190659166433e-06, + "loss": 0.5741, + "step": 3915 + }, + { + "epoch": 3.241855328547764, + "grad_norm": 0.9359234571456909, + "learning_rate": 2.231097323665971e-06, + "loss": 0.5608, + "step": 3916 + }, + { + "epoch": 3.2426836002208725, + "grad_norm": 0.9326402544975281, + "learning_rate": 2.2300040401956693e-06, + "loss": 0.5696, + "step": 3917 + }, + { + "epoch": 3.2435118718939813, + "grad_norm": 0.9257575869560242, + "learning_rate": 2.2289108089670687e-06, + "loss": 0.5709, + "step": 3918 + }, + { + "epoch": 3.24434014356709, + "grad_norm": 0.9669248461723328, + "learning_rate": 2.2278176301917004e-06, + "loss": 0.5666, + "step": 3919 + }, + { + "epoch": 3.2451684152401987, + "grad_norm": 0.9328755736351013, + "learning_rate": 2.2267245040810832e-06, + "loss": 0.573, + "step": 3920 + }, + { + "epoch": 3.2459966869133074, + "grad_norm": 0.942805826663971, + "learning_rate": 2.225631430846727e-06, + "loss": 0.5712, + "step": 3921 + }, + { + "epoch": 3.246824958586416, + "grad_norm": 0.9514304995536804, + "learning_rate": 2.224538410700131e-06, + "loss": 0.5775, + "step": 3922 + }, + { + "epoch": 3.247653230259525, + "grad_norm": 0.9471868872642517, + "learning_rate": 2.2234454438527847e-06, + "loss": 0.5794, + "step": 3923 + }, + { + "epoch": 3.248481501932634, + "grad_norm": 0.9507884979248047, + "learning_rate": 2.2223525305161675e-06, + "loss": 0.5726, + "step": 3924 + }, + { + "epoch": 3.249309773605743, + "grad_norm": 0.9477060437202454, + "learning_rate": 2.221259670901747e-06, + "loss": 0.5895, + "step": 3925 + }, + { + "epoch": 3.2501380452788515, + "grad_norm": 0.9437663555145264, + "learning_rate": 2.2201668652209823e-06, + "loss": 0.5538, + "step": 3926 + }, + { + "epoch": 3.2509663169519603, + "grad_norm": 0.9790158271789551, + "learning_rate": 2.2190741136853205e-06, + "loss": 0.5729, + "step": 3927 + }, + { + "epoch": 3.251794588625069, + "grad_norm": 0.9320138692855835, + "learning_rate": 2.2179814165061984e-06, + "loss": 0.5807, + "step": 3928 + }, + { + "epoch": 3.2526228602981777, + "grad_norm": 0.9543532729148865, + "learning_rate": 2.2168887738950438e-06, + "loss": 0.5645, + "step": 3929 + }, + { + "epoch": 3.2534511319712864, + "grad_norm": 0.9545353055000305, + "learning_rate": 2.2157961860632716e-06, + "loss": 0.5762, + "step": 3930 + }, + { + "epoch": 3.2542794036443956, + "grad_norm": 0.9640288352966309, + "learning_rate": 2.2147036532222887e-06, + "loss": 0.5738, + "step": 3931 + }, + { + "epoch": 3.2551076753175043, + "grad_norm": 0.9393999576568604, + "learning_rate": 2.2136111755834895e-06, + "loss": 0.5692, + "step": 3932 + }, + { + "epoch": 3.255935946990613, + "grad_norm": 0.9366894364356995, + "learning_rate": 2.212518753358258e-06, + "loss": 0.5957, + "step": 3933 + }, + { + "epoch": 3.256764218663722, + "grad_norm": 0.9408980011940002, + "learning_rate": 2.2114263867579673e-06, + "loss": 0.5644, + "step": 3934 + }, + { + "epoch": 3.2575924903368305, + "grad_norm": 0.9842433333396912, + "learning_rate": 2.2103340759939816e-06, + "loss": 0.5772, + "step": 3935 + }, + { + "epoch": 3.2584207620099392, + "grad_norm": 0.9410622119903564, + "learning_rate": 2.209241821277651e-06, + "loss": 0.5637, + "step": 3936 + }, + { + "epoch": 3.259249033683048, + "grad_norm": 0.9571241736412048, + "learning_rate": 2.208149622820319e-06, + "loss": 0.5754, + "step": 3937 + }, + { + "epoch": 3.2600773053561567, + "grad_norm": 0.9451292157173157, + "learning_rate": 2.2070574808333135e-06, + "loss": 0.557, + "step": 3938 + }, + { + "epoch": 3.2609055770292654, + "grad_norm": 0.987385094165802, + "learning_rate": 2.205965395527955e-06, + "loss": 0.5458, + "step": 3939 + }, + { + "epoch": 3.2617338487023746, + "grad_norm": 0.9282477498054504, + "learning_rate": 2.204873367115551e-06, + "loss": 0.5726, + "step": 3940 + }, + { + "epoch": 3.2625621203754833, + "grad_norm": 0.9609903693199158, + "learning_rate": 2.2037813958073994e-06, + "loss": 0.5608, + "step": 3941 + }, + { + "epoch": 3.263390392048592, + "grad_norm": 0.9469774961471558, + "learning_rate": 2.2026894818147864e-06, + "loss": 0.5663, + "step": 3942 + }, + { + "epoch": 3.2642186637217008, + "grad_norm": 0.9910560846328735, + "learning_rate": 2.2015976253489873e-06, + "loss": 0.5604, + "step": 3943 + }, + { + "epoch": 3.2650469353948095, + "grad_norm": 0.949170708656311, + "learning_rate": 2.2005058266212657e-06, + "loss": 0.5795, + "step": 3944 + }, + { + "epoch": 3.2658752070679182, + "grad_norm": 0.9276071786880493, + "learning_rate": 2.199414085842875e-06, + "loss": 0.5821, + "step": 3945 + }, + { + "epoch": 3.266703478741027, + "grad_norm": 0.9555459022521973, + "learning_rate": 2.1983224032250573e-06, + "loss": 0.5832, + "step": 3946 + }, + { + "epoch": 3.2675317504141357, + "grad_norm": 0.972984790802002, + "learning_rate": 2.1972307789790413e-06, + "loss": 0.5868, + "step": 3947 + }, + { + "epoch": 3.2683600220872444, + "grad_norm": 0.9383103251457214, + "learning_rate": 2.196139213316049e-06, + "loss": 0.5624, + "step": 3948 + }, + { + "epoch": 3.2691882937603536, + "grad_norm": 0.9539969563484192, + "learning_rate": 2.195047706447285e-06, + "loss": 0.566, + "step": 3949 + }, + { + "epoch": 3.2700165654334623, + "grad_norm": 0.955715537071228, + "learning_rate": 2.193956258583947e-06, + "loss": 0.5548, + "step": 3950 + }, + { + "epoch": 3.270844837106571, + "grad_norm": 0.9660782217979431, + "learning_rate": 2.19286486993722e-06, + "loss": 0.5726, + "step": 3951 + }, + { + "epoch": 3.2716731087796798, + "grad_norm": 0.9660666584968567, + "learning_rate": 2.1917735407182777e-06, + "loss": 0.5728, + "step": 3952 + }, + { + "epoch": 3.2725013804527885, + "grad_norm": 0.9096773862838745, + "learning_rate": 2.190682271138282e-06, + "loss": 0.5549, + "step": 3953 + }, + { + "epoch": 3.273329652125897, + "grad_norm": 0.9581109285354614, + "learning_rate": 2.189591061408384e-06, + "loss": 0.5834, + "step": 3954 + }, + { + "epoch": 3.274157923799006, + "grad_norm": 0.9483600854873657, + "learning_rate": 2.188499911739721e-06, + "loss": 0.5749, + "step": 3955 + }, + { + "epoch": 3.2749861954721147, + "grad_norm": 0.9495676159858704, + "learning_rate": 2.187408822343422e-06, + "loss": 0.5907, + "step": 3956 + }, + { + "epoch": 3.2758144671452234, + "grad_norm": 0.9316544532775879, + "learning_rate": 2.186317793430602e-06, + "loss": 0.555, + "step": 3957 + }, + { + "epoch": 3.2766427388183326, + "grad_norm": 0.9391309022903442, + "learning_rate": 2.1852268252123644e-06, + "loss": 0.5751, + "step": 3958 + }, + { + "epoch": 3.2774710104914413, + "grad_norm": 0.91743403673172, + "learning_rate": 2.1841359178998024e-06, + "loss": 0.5504, + "step": 3959 + }, + { + "epoch": 3.27829928216455, + "grad_norm": 0.9177739024162292, + "learning_rate": 2.183045071703996e-06, + "loss": 0.5452, + "step": 3960 + }, + { + "epoch": 3.2791275538376587, + "grad_norm": 0.964051365852356, + "learning_rate": 2.1819542868360135e-06, + "loss": 0.5731, + "step": 3961 + }, + { + "epoch": 3.2799558255107675, + "grad_norm": 0.9402073621749878, + "learning_rate": 2.1808635635069113e-06, + "loss": 0.5655, + "step": 3962 + }, + { + "epoch": 3.280784097183876, + "grad_norm": 0.9163627624511719, + "learning_rate": 2.1797729019277346e-06, + "loss": 0.5772, + "step": 3963 + }, + { + "epoch": 3.281612368856985, + "grad_norm": 0.9189632534980774, + "learning_rate": 2.1786823023095164e-06, + "loss": 0.5738, + "step": 3964 + }, + { + "epoch": 3.282440640530094, + "grad_norm": 0.9654256701469421, + "learning_rate": 2.1775917648632772e-06, + "loss": 0.5752, + "step": 3965 + }, + { + "epoch": 3.283268912203203, + "grad_norm": 1.0078915357589722, + "learning_rate": 2.1765012898000257e-06, + "loss": 0.5847, + "step": 3966 + }, + { + "epoch": 3.2840971838763116, + "grad_norm": 0.9313945174217224, + "learning_rate": 2.17541087733076e-06, + "loss": 0.5671, + "step": 3967 + }, + { + "epoch": 3.2849254555494203, + "grad_norm": 0.9611067771911621, + "learning_rate": 2.1743205276664634e-06, + "loss": 0.5501, + "step": 3968 + }, + { + "epoch": 3.285753727222529, + "grad_norm": 0.9783805012702942, + "learning_rate": 2.1732302410181084e-06, + "loss": 0.5969, + "step": 3969 + }, + { + "epoch": 3.2865819988956377, + "grad_norm": 0.9382510781288147, + "learning_rate": 2.172140017596656e-06, + "loss": 0.5716, + "step": 3970 + }, + { + "epoch": 3.2874102705687465, + "grad_norm": 0.9219527840614319, + "learning_rate": 2.1710498576130545e-06, + "loss": 0.5916, + "step": 3971 + }, + { + "epoch": 3.288238542241855, + "grad_norm": 0.9681289792060852, + "learning_rate": 2.1699597612782385e-06, + "loss": 0.5725, + "step": 3972 + }, + { + "epoch": 3.289066813914964, + "grad_norm": 0.946723997592926, + "learning_rate": 2.168869728803132e-06, + "loss": 0.5664, + "step": 3973 + }, + { + "epoch": 3.289895085588073, + "grad_norm": 0.9302477836608887, + "learning_rate": 2.167779760398646e-06, + "loss": 0.5777, + "step": 3974 + }, + { + "epoch": 3.290723357261182, + "grad_norm": 0.9480409026145935, + "learning_rate": 2.16668985627568e-06, + "loss": 0.5696, + "step": 3975 + }, + { + "epoch": 3.2915516289342905, + "grad_norm": 0.9363623261451721, + "learning_rate": 2.1656000166451195e-06, + "loss": 0.5798, + "step": 3976 + }, + { + "epoch": 3.2923799006073993, + "grad_norm": 0.9291804432868958, + "learning_rate": 2.164510241717838e-06, + "loss": 0.5658, + "step": 3977 + }, + { + "epoch": 3.293208172280508, + "grad_norm": 0.9314830899238586, + "learning_rate": 2.163420531704698e-06, + "loss": 0.5698, + "step": 3978 + }, + { + "epoch": 3.2940364439536167, + "grad_norm": 0.974621057510376, + "learning_rate": 2.1623308868165474e-06, + "loss": 0.5644, + "step": 3979 + }, + { + "epoch": 3.2948647156267254, + "grad_norm": 0.9464275240898132, + "learning_rate": 2.161241307264222e-06, + "loss": 0.5704, + "step": 3980 + }, + { + "epoch": 3.295692987299834, + "grad_norm": 0.9375281929969788, + "learning_rate": 2.1601517932585462e-06, + "loss": 0.5714, + "step": 3981 + }, + { + "epoch": 3.296521258972943, + "grad_norm": 0.910499095916748, + "learning_rate": 2.159062345010331e-06, + "loss": 0.5639, + "step": 3982 + }, + { + "epoch": 3.297349530646052, + "grad_norm": 0.9525747895240784, + "learning_rate": 2.1579729627303735e-06, + "loss": 0.5703, + "step": 3983 + }, + { + "epoch": 3.298177802319161, + "grad_norm": 0.9756594896316528, + "learning_rate": 2.1568836466294596e-06, + "loss": 0.5862, + "step": 3984 + }, + { + "epoch": 3.2990060739922695, + "grad_norm": 0.9418894648551941, + "learning_rate": 2.1557943969183605e-06, + "loss": 0.5554, + "step": 3985 + }, + { + "epoch": 3.2998343456653783, + "grad_norm": 0.9190452694892883, + "learning_rate": 2.154705213807838e-06, + "loss": 0.5519, + "step": 3986 + }, + { + "epoch": 3.300662617338487, + "grad_norm": 0.9576460719108582, + "learning_rate": 2.153616097508637e-06, + "loss": 0.5723, + "step": 3987 + }, + { + "epoch": 3.3014908890115957, + "grad_norm": 0.9196321964263916, + "learning_rate": 2.1525270482314924e-06, + "loss": 0.6006, + "step": 3988 + }, + { + "epoch": 3.3023191606847044, + "grad_norm": 0.9210584163665771, + "learning_rate": 2.1514380661871253e-06, + "loss": 0.5569, + "step": 3989 + }, + { + "epoch": 3.3031474323578136, + "grad_norm": 0.9365322589874268, + "learning_rate": 2.150349151586243e-06, + "loss": 0.5747, + "step": 3990 + }, + { + "epoch": 3.3039757040309223, + "grad_norm": 0.948229193687439, + "learning_rate": 2.1492603046395407e-06, + "loss": 0.556, + "step": 3991 + }, + { + "epoch": 3.304803975704031, + "grad_norm": 0.9285109043121338, + "learning_rate": 2.148171525557699e-06, + "loss": 0.5747, + "step": 3992 + }, + { + "epoch": 3.30563224737714, + "grad_norm": 0.9224300384521484, + "learning_rate": 2.1470828145513882e-06, + "loss": 0.5883, + "step": 3993 + }, + { + "epoch": 3.3064605190502485, + "grad_norm": 0.9375912547111511, + "learning_rate": 2.1459941718312633e-06, + "loss": 0.5663, + "step": 3994 + }, + { + "epoch": 3.3072887907233572, + "grad_norm": 0.9629141688346863, + "learning_rate": 2.144905597607966e-06, + "loss": 0.5593, + "step": 3995 + }, + { + "epoch": 3.308117062396466, + "grad_norm": 0.9608798623085022, + "learning_rate": 2.143817092092124e-06, + "loss": 0.573, + "step": 3996 + }, + { + "epoch": 3.3089453340695747, + "grad_norm": 0.9235004186630249, + "learning_rate": 2.1427286554943557e-06, + "loss": 0.5818, + "step": 3997 + }, + { + "epoch": 3.3097736057426834, + "grad_norm": 0.9361316561698914, + "learning_rate": 2.141640288025262e-06, + "loss": 0.556, + "step": 3998 + }, + { + "epoch": 3.3106018774157926, + "grad_norm": 0.9709519743919373, + "learning_rate": 2.140551989895432e-06, + "loss": 0.5575, + "step": 3999 + }, + { + "epoch": 3.3114301490889013, + "grad_norm": 0.9461862444877625, + "learning_rate": 2.1394637613154403e-06, + "loss": 0.5676, + "step": 4000 + }, + { + "epoch": 3.31225842076201, + "grad_norm": 0.9697636961936951, + "learning_rate": 2.138375602495851e-06, + "loss": 0.5678, + "step": 4001 + }, + { + "epoch": 3.3130866924351188, + "grad_norm": 0.9442050457000732, + "learning_rate": 2.1372875136472113e-06, + "loss": 0.573, + "step": 4002 + }, + { + "epoch": 3.3139149641082275, + "grad_norm": 0.9325514435768127, + "learning_rate": 2.136199494980056e-06, + "loss": 0.5711, + "step": 4003 + }, + { + "epoch": 3.3147432357813362, + "grad_norm": 0.934421181678772, + "learning_rate": 2.135111546704908e-06, + "loss": 0.5634, + "step": 4004 + }, + { + "epoch": 3.315571507454445, + "grad_norm": 0.9201632738113403, + "learning_rate": 2.134023669032274e-06, + "loss": 0.5627, + "step": 4005 + }, + { + "epoch": 3.3163997791275537, + "grad_norm": 0.9563625454902649, + "learning_rate": 2.1329358621726484e-06, + "loss": 0.5668, + "step": 4006 + }, + { + "epoch": 3.3172280508006624, + "grad_norm": 0.9408397078514099, + "learning_rate": 2.1318481263365117e-06, + "loss": 0.5712, + "step": 4007 + }, + { + "epoch": 3.3180563224737716, + "grad_norm": 0.9219162464141846, + "learning_rate": 2.13076046173433e-06, + "loss": 0.5694, + "step": 4008 + }, + { + "epoch": 3.3188845941468803, + "grad_norm": 0.9441806077957153, + "learning_rate": 2.129672868576558e-06, + "loss": 0.5901, + "step": 4009 + }, + { + "epoch": 3.319712865819989, + "grad_norm": 0.9196688532829285, + "learning_rate": 2.1285853470736336e-06, + "loss": 0.5768, + "step": 4010 + }, + { + "epoch": 3.3205411374930978, + "grad_norm": 0.9800437688827515, + "learning_rate": 2.1274978974359813e-06, + "loss": 0.5728, + "step": 4011 + }, + { + "epoch": 3.3213694091662065, + "grad_norm": 0.9329850077629089, + "learning_rate": 2.126410519874014e-06, + "loss": 0.5806, + "step": 4012 + }, + { + "epoch": 3.322197680839315, + "grad_norm": 0.9533185362815857, + "learning_rate": 2.1253232145981293e-06, + "loss": 0.5665, + "step": 4013 + }, + { + "epoch": 3.323025952512424, + "grad_norm": 0.9292604327201843, + "learning_rate": 2.1242359818187085e-06, + "loss": 0.5552, + "step": 4014 + }, + { + "epoch": 3.3238542241855327, + "grad_norm": 0.9430820941925049, + "learning_rate": 2.1231488217461234e-06, + "loss": 0.5849, + "step": 4015 + }, + { + "epoch": 3.3246824958586414, + "grad_norm": 0.9454953670501709, + "learning_rate": 2.1220617345907287e-06, + "loss": 0.5724, + "step": 4016 + }, + { + "epoch": 3.3255107675317506, + "grad_norm": 0.9445587396621704, + "learning_rate": 2.120974720562865e-06, + "loss": 0.5908, + "step": 4017 + }, + { + "epoch": 3.3263390392048593, + "grad_norm": 0.9306840896606445, + "learning_rate": 2.1198877798728595e-06, + "loss": 0.5938, + "step": 4018 + }, + { + "epoch": 3.327167310877968, + "grad_norm": 0.9418679475784302, + "learning_rate": 2.1188009127310244e-06, + "loss": 0.5692, + "step": 4019 + }, + { + "epoch": 3.3279955825510767, + "grad_norm": 0.9448477625846863, + "learning_rate": 2.11771411934766e-06, + "loss": 0.578, + "step": 4020 + }, + { + "epoch": 3.3288238542241855, + "grad_norm": 0.9564743638038635, + "learning_rate": 2.1166273999330497e-06, + "loss": 0.5577, + "step": 4021 + }, + { + "epoch": 3.329652125897294, + "grad_norm": 0.9450514316558838, + "learning_rate": 2.115540754697463e-06, + "loss": 0.5812, + "step": 4022 + }, + { + "epoch": 3.330480397570403, + "grad_norm": 0.932601809501648, + "learning_rate": 2.1144541838511578e-06, + "loss": 0.5534, + "step": 4023 + }, + { + "epoch": 3.331308669243512, + "grad_norm": 0.9476706385612488, + "learning_rate": 2.1133676876043734e-06, + "loss": 0.5684, + "step": 4024 + }, + { + "epoch": 3.332136940916621, + "grad_norm": 0.9359769225120544, + "learning_rate": 2.1122812661673365e-06, + "loss": 0.5624, + "step": 4025 + }, + { + "epoch": 3.3329652125897296, + "grad_norm": 0.9415692687034607, + "learning_rate": 2.1111949197502612e-06, + "loss": 0.5627, + "step": 4026 + }, + { + "epoch": 3.3337934842628383, + "grad_norm": 0.9686550498008728, + "learning_rate": 2.1101086485633455e-06, + "loss": 0.5401, + "step": 4027 + }, + { + "epoch": 3.334621755935947, + "grad_norm": 0.9065554738044739, + "learning_rate": 2.1090224528167703e-06, + "loss": 0.5785, + "step": 4028 + }, + { + "epoch": 3.3354500276090557, + "grad_norm": 0.9449801445007324, + "learning_rate": 2.1079363327207067e-06, + "loss": 0.5695, + "step": 4029 + }, + { + "epoch": 3.3362782992821645, + "grad_norm": 0.9398133754730225, + "learning_rate": 2.106850288485307e-06, + "loss": 0.5605, + "step": 4030 + }, + { + "epoch": 3.337106570955273, + "grad_norm": 0.9175443649291992, + "learning_rate": 2.1057643203207125e-06, + "loss": 0.5675, + "step": 4031 + }, + { + "epoch": 3.337934842628382, + "grad_norm": 0.9433302879333496, + "learning_rate": 2.1046784284370466e-06, + "loss": 0.5501, + "step": 4032 + }, + { + "epoch": 3.338763114301491, + "grad_norm": 0.9375688433647156, + "learning_rate": 2.103592613044419e-06, + "loss": 0.579, + "step": 4033 + }, + { + "epoch": 3.3395913859746, + "grad_norm": 0.9344912171363831, + "learning_rate": 2.102506874352926e-06, + "loss": 0.5698, + "step": 4034 + }, + { + "epoch": 3.3404196576477085, + "grad_norm": 0.9581173062324524, + "learning_rate": 2.101421212572648e-06, + "loss": 0.5774, + "step": 4035 + }, + { + "epoch": 3.3412479293208173, + "grad_norm": 0.9601131081581116, + "learning_rate": 2.100335627913649e-06, + "loss": 0.5769, + "step": 4036 + }, + { + "epoch": 3.342076200993926, + "grad_norm": 0.9634785652160645, + "learning_rate": 2.0992501205859806e-06, + "loss": 0.5883, + "step": 4037 + }, + { + "epoch": 3.3429044726670347, + "grad_norm": 1.0006943941116333, + "learning_rate": 2.0981646907996784e-06, + "loss": 0.5744, + "step": 4038 + }, + { + "epoch": 3.3437327443401434, + "grad_norm": 0.9540338516235352, + "learning_rate": 2.0970793387647636e-06, + "loss": 0.5775, + "step": 4039 + }, + { + "epoch": 3.344561016013252, + "grad_norm": 0.9318099617958069, + "learning_rate": 2.09599406469124e-06, + "loss": 0.5765, + "step": 4040 + }, + { + "epoch": 3.345389287686361, + "grad_norm": 0.9622073173522949, + "learning_rate": 2.0949088687890976e-06, + "loss": 0.5532, + "step": 4041 + }, + { + "epoch": 3.34621755935947, + "grad_norm": 0.9623757600784302, + "learning_rate": 2.0938237512683142e-06, + "loss": 0.5518, + "step": 4042 + }, + { + "epoch": 3.347045831032579, + "grad_norm": 0.9352701306343079, + "learning_rate": 2.0927387123388485e-06, + "loss": 0.5488, + "step": 4043 + }, + { + "epoch": 3.3478741027056875, + "grad_norm": 0.9591230154037476, + "learning_rate": 2.0916537522106446e-06, + "loss": 0.5887, + "step": 4044 + }, + { + "epoch": 3.3487023743787963, + "grad_norm": 0.9517300724983215, + "learning_rate": 2.0905688710936338e-06, + "loss": 0.5619, + "step": 4045 + }, + { + "epoch": 3.349530646051905, + "grad_norm": 0.9537909626960754, + "learning_rate": 2.08948406919773e-06, + "loss": 0.5949, + "step": 4046 + }, + { + "epoch": 3.3503589177250137, + "grad_norm": 0.946260929107666, + "learning_rate": 2.088399346732831e-06, + "loss": 0.5989, + "step": 4047 + }, + { + "epoch": 3.3511871893981224, + "grad_norm": 0.9061339497566223, + "learning_rate": 2.0873147039088227e-06, + "loss": 0.5855, + "step": 4048 + }, + { + "epoch": 3.3520154610712316, + "grad_norm": 0.9527447819709778, + "learning_rate": 2.0862301409355717e-06, + "loss": 0.5703, + "step": 4049 + }, + { + "epoch": 3.3528437327443403, + "grad_norm": 0.9357138872146606, + "learning_rate": 2.0851456580229325e-06, + "loss": 0.551, + "step": 4050 + }, + { + "epoch": 3.353672004417449, + "grad_norm": 0.9259575009346008, + "learning_rate": 2.08406125538074e-06, + "loss": 0.5777, + "step": 4051 + }, + { + "epoch": 3.354500276090558, + "grad_norm": 0.944807231426239, + "learning_rate": 2.082976933218817e-06, + "loss": 0.5741, + "step": 4052 + }, + { + "epoch": 3.3553285477636665, + "grad_norm": 0.9251223206520081, + "learning_rate": 2.08189269174697e-06, + "loss": 0.5829, + "step": 4053 + }, + { + "epoch": 3.3561568194367752, + "grad_norm": 0.9122853875160217, + "learning_rate": 2.0808085311749904e-06, + "loss": 0.5602, + "step": 4054 + }, + { + "epoch": 3.356985091109884, + "grad_norm": 0.9430744647979736, + "learning_rate": 2.0797244517126512e-06, + "loss": 0.5633, + "step": 4055 + }, + { + "epoch": 3.3578133627829927, + "grad_norm": 0.9500707983970642, + "learning_rate": 2.078640453569714e-06, + "loss": 0.5791, + "step": 4056 + }, + { + "epoch": 3.3586416344561014, + "grad_norm": 0.970301628112793, + "learning_rate": 2.077556536955921e-06, + "loss": 0.5668, + "step": 4057 + }, + { + "epoch": 3.3594699061292106, + "grad_norm": 0.9528861045837402, + "learning_rate": 2.0764727020809995e-06, + "loss": 0.5616, + "step": 4058 + }, + { + "epoch": 3.3602981778023193, + "grad_norm": 0.9321720004081726, + "learning_rate": 2.0753889491546624e-06, + "loss": 0.5841, + "step": 4059 + }, + { + "epoch": 3.361126449475428, + "grad_norm": 0.9030178189277649, + "learning_rate": 2.0743052783866054e-06, + "loss": 0.5757, + "step": 4060 + }, + { + "epoch": 3.3619547211485368, + "grad_norm": 0.9501296877861023, + "learning_rate": 2.0732216899865098e-06, + "loss": 0.5477, + "step": 4061 + }, + { + "epoch": 3.3627829928216455, + "grad_norm": 0.9475130438804626, + "learning_rate": 2.0721381841640377e-06, + "loss": 0.5798, + "step": 4062 + }, + { + "epoch": 3.3636112644947542, + "grad_norm": 0.954949676990509, + "learning_rate": 2.0710547611288384e-06, + "loss": 0.579, + "step": 4063 + }, + { + "epoch": 3.364439536167863, + "grad_norm": 0.9534462690353394, + "learning_rate": 2.069971421090545e-06, + "loss": 0.5448, + "step": 4064 + }, + { + "epoch": 3.3652678078409717, + "grad_norm": 0.9504910111427307, + "learning_rate": 2.068888164258772e-06, + "loss": 0.5698, + "step": 4065 + }, + { + "epoch": 3.3660960795140804, + "grad_norm": 0.934983491897583, + "learning_rate": 2.067804990843121e-06, + "loss": 0.5643, + "step": 4066 + }, + { + "epoch": 3.3669243511871896, + "grad_norm": 0.9342026114463806, + "learning_rate": 2.066721901053175e-06, + "loss": 0.5938, + "step": 4067 + }, + { + "epoch": 3.3677526228602983, + "grad_norm": 0.9376976490020752, + "learning_rate": 2.0656388950985027e-06, + "loss": 0.5897, + "step": 4068 + }, + { + "epoch": 3.368580894533407, + "grad_norm": 0.9228143095970154, + "learning_rate": 2.064555973188655e-06, + "loss": 0.5789, + "step": 4069 + }, + { + "epoch": 3.3694091662065158, + "grad_norm": 0.9547181129455566, + "learning_rate": 2.0634731355331665e-06, + "loss": 0.5707, + "step": 4070 + }, + { + "epoch": 3.3702374378796245, + "grad_norm": 0.9388718605041504, + "learning_rate": 2.062390382341558e-06, + "loss": 0.5719, + "step": 4071 + }, + { + "epoch": 3.371065709552733, + "grad_norm": 0.9529668092727661, + "learning_rate": 2.0613077138233313e-06, + "loss": 0.5634, + "step": 4072 + }, + { + "epoch": 3.371893981225842, + "grad_norm": 0.9569711089134216, + "learning_rate": 2.060225130187973e-06, + "loss": 0.5565, + "step": 4073 + }, + { + "epoch": 3.3727222528989507, + "grad_norm": 0.936896026134491, + "learning_rate": 2.0591426316449513e-06, + "loss": 0.5543, + "step": 4074 + }, + { + "epoch": 3.3735505245720594, + "grad_norm": 0.9453547596931458, + "learning_rate": 2.0580602184037206e-06, + "loss": 0.5652, + "step": 4075 + }, + { + "epoch": 3.3743787962451686, + "grad_norm": 0.9260196089744568, + "learning_rate": 2.0569778906737186e-06, + "loss": 0.5765, + "step": 4076 + }, + { + "epoch": 3.3752070679182773, + "grad_norm": 0.9467774629592896, + "learning_rate": 2.055895648664365e-06, + "loss": 0.5637, + "step": 4077 + }, + { + "epoch": 3.376035339591386, + "grad_norm": 0.9502113461494446, + "learning_rate": 2.0548134925850626e-06, + "loss": 0.5725, + "step": 4078 + }, + { + "epoch": 3.3768636112644947, + "grad_norm": 0.9475006461143494, + "learning_rate": 2.0537314226452004e-06, + "loss": 0.5936, + "step": 4079 + }, + { + "epoch": 3.3776918829376035, + "grad_norm": 0.9401916861534119, + "learning_rate": 2.0526494390541478e-06, + "loss": 0.5797, + "step": 4080 + }, + { + "epoch": 3.378520154610712, + "grad_norm": 0.9407974481582642, + "learning_rate": 2.0515675420212577e-06, + "loss": 0.5653, + "step": 4081 + }, + { + "epoch": 3.379348426283821, + "grad_norm": 0.9521286487579346, + "learning_rate": 2.0504857317558686e-06, + "loss": 0.5794, + "step": 4082 + }, + { + "epoch": 3.38017669795693, + "grad_norm": 0.9705640077590942, + "learning_rate": 2.0494040084673004e-06, + "loss": 0.5821, + "step": 4083 + }, + { + "epoch": 3.381004969630039, + "grad_norm": 0.9214118719100952, + "learning_rate": 2.0483223723648567e-06, + "loss": 0.5757, + "step": 4084 + }, + { + "epoch": 3.3818332413031476, + "grad_norm": 0.9589064717292786, + "learning_rate": 2.0472408236578228e-06, + "loss": 0.5551, + "step": 4085 + }, + { + "epoch": 3.3826615129762563, + "grad_norm": 0.9503619074821472, + "learning_rate": 2.0461593625554683e-06, + "loss": 0.5805, + "step": 4086 + }, + { + "epoch": 3.383489784649365, + "grad_norm": 0.9328595995903015, + "learning_rate": 2.045077989267047e-06, + "loss": 0.5736, + "step": 4087 + }, + { + "epoch": 3.3843180563224737, + "grad_norm": 0.9475532174110413, + "learning_rate": 2.043996704001794e-06, + "loss": 0.5652, + "step": 4088 + }, + { + "epoch": 3.3851463279955825, + "grad_norm": 0.9307344555854797, + "learning_rate": 2.042915506968928e-06, + "loss": 0.5801, + "step": 4089 + }, + { + "epoch": 3.385974599668691, + "grad_norm": 0.9430568218231201, + "learning_rate": 2.04183439837765e-06, + "loss": 0.5864, + "step": 4090 + }, + { + "epoch": 3.3868028713418, + "grad_norm": 0.9192523956298828, + "learning_rate": 2.0407533784371457e-06, + "loss": 0.5486, + "step": 4091 + }, + { + "epoch": 3.387631143014909, + "grad_norm": 0.9655841588973999, + "learning_rate": 2.039672447356581e-06, + "loss": 0.5602, + "step": 4092 + }, + { + "epoch": 3.388459414688018, + "grad_norm": 0.9553002119064331, + "learning_rate": 2.0385916053451073e-06, + "loss": 0.568, + "step": 4093 + }, + { + "epoch": 3.3892876863611265, + "grad_norm": 0.9539245963096619, + "learning_rate": 2.0375108526118563e-06, + "loss": 0.5898, + "step": 4094 + }, + { + "epoch": 3.3901159580342353, + "grad_norm": 0.9733588695526123, + "learning_rate": 2.036430189365945e-06, + "loss": 0.5817, + "step": 4095 + }, + { + "epoch": 3.390944229707344, + "grad_norm": 0.9487296938896179, + "learning_rate": 2.03534961581647e-06, + "loss": 0.5841, + "step": 4096 + }, + { + "epoch": 3.3917725013804527, + "grad_norm": 0.9788404107093811, + "learning_rate": 2.0342691321725127e-06, + "loss": 0.5951, + "step": 4097 + }, + { + "epoch": 3.3926007730535614, + "grad_norm": 0.9286518692970276, + "learning_rate": 2.033188738643137e-06, + "loss": 0.5555, + "step": 4098 + }, + { + "epoch": 3.39342904472667, + "grad_norm": 0.9736733436584473, + "learning_rate": 2.032108435437389e-06, + "loss": 0.5773, + "step": 4099 + }, + { + "epoch": 3.394257316399779, + "grad_norm": 0.9288291335105896, + "learning_rate": 2.0310282227642963e-06, + "loss": 0.5562, + "step": 4100 + }, + { + "epoch": 3.395085588072888, + "grad_norm": 0.9414420127868652, + "learning_rate": 2.0299481008328716e-06, + "loss": 0.5923, + "step": 4101 + }, + { + "epoch": 3.395913859745997, + "grad_norm": 0.9327258467674255, + "learning_rate": 2.028868069852108e-06, + "loss": 0.5636, + "step": 4102 + }, + { + "epoch": 3.3967421314191055, + "grad_norm": 0.9412753582000732, + "learning_rate": 2.02778813003098e-06, + "loss": 0.5896, + "step": 4103 + }, + { + "epoch": 3.3975704030922143, + "grad_norm": 0.973572313785553, + "learning_rate": 2.026708281578447e-06, + "loss": 0.5581, + "step": 4104 + }, + { + "epoch": 3.398398674765323, + "grad_norm": 1.1189706325531006, + "learning_rate": 2.0256285247034505e-06, + "loss": 0.5541, + "step": 4105 + }, + { + "epoch": 3.3992269464384317, + "grad_norm": 0.9494182467460632, + "learning_rate": 2.0245488596149113e-06, + "loss": 0.5686, + "step": 4106 + }, + { + "epoch": 3.4000552181115404, + "grad_norm": 0.9740265011787415, + "learning_rate": 2.0234692865217375e-06, + "loss": 0.5744, + "step": 4107 + }, + { + "epoch": 3.4008834897846496, + "grad_norm": 0.9565280079841614, + "learning_rate": 2.0223898056328133e-06, + "loss": 0.589, + "step": 4108 + }, + { + "epoch": 3.4017117614577583, + "grad_norm": 0.9839944839477539, + "learning_rate": 2.0213104171570094e-06, + "loss": 0.5704, + "step": 4109 + }, + { + "epoch": 3.402540033130867, + "grad_norm": 0.9117640256881714, + "learning_rate": 2.020231121303178e-06, + "loss": 0.5591, + "step": 4110 + }, + { + "epoch": 3.403368304803976, + "grad_norm": 0.9455560445785522, + "learning_rate": 2.019151918280151e-06, + "loss": 0.592, + "step": 4111 + }, + { + "epoch": 3.4041965764770845, + "grad_norm": 0.9271209836006165, + "learning_rate": 2.0180728082967464e-06, + "loss": 0.5832, + "step": 4112 + }, + { + "epoch": 3.4050248481501932, + "grad_norm": 0.9203625321388245, + "learning_rate": 2.016993791561761e-06, + "loss": 0.576, + "step": 4113 + }, + { + "epoch": 3.405853119823302, + "grad_norm": 0.9395371675491333, + "learning_rate": 2.015914868283974e-06, + "loss": 0.5823, + "step": 4114 + }, + { + "epoch": 3.4066813914964107, + "grad_norm": 0.9351215958595276, + "learning_rate": 2.014836038672148e-06, + "loss": 0.5903, + "step": 4115 + }, + { + "epoch": 3.4075096631695194, + "grad_norm": 0.9305378198623657, + "learning_rate": 2.0137573029350253e-06, + "loss": 0.5607, + "step": 4116 + }, + { + "epoch": 3.4083379348426286, + "grad_norm": 0.9378963708877563, + "learning_rate": 2.0126786612813316e-06, + "loss": 0.5648, + "step": 4117 + }, + { + "epoch": 3.4091662065157373, + "grad_norm": 1.2515968084335327, + "learning_rate": 2.011600113919776e-06, + "loss": 0.5915, + "step": 4118 + }, + { + "epoch": 3.409994478188846, + "grad_norm": 0.9324381947517395, + "learning_rate": 2.010521661059044e-06, + "loss": 0.5573, + "step": 4119 + }, + { + "epoch": 3.4108227498619548, + "grad_norm": 0.9508990049362183, + "learning_rate": 2.009443302907808e-06, + "loss": 0.5612, + "step": 4120 + }, + { + "epoch": 3.4116510215350635, + "grad_norm": 0.9441195130348206, + "learning_rate": 2.00836503967472e-06, + "loss": 0.5561, + "step": 4121 + }, + { + "epoch": 3.4124792932081722, + "grad_norm": 0.940579354763031, + "learning_rate": 2.007286871568413e-06, + "loss": 0.5787, + "step": 4122 + }, + { + "epoch": 3.413307564881281, + "grad_norm": 0.9296087026596069, + "learning_rate": 2.0062087987975043e-06, + "loss": 0.5638, + "step": 4123 + }, + { + "epoch": 3.4141358365543897, + "grad_norm": 0.9607263803482056, + "learning_rate": 2.00513082157059e-06, + "loss": 0.5903, + "step": 4124 + }, + { + "epoch": 3.4149641082274984, + "grad_norm": 0.9708670377731323, + "learning_rate": 2.004052940096248e-06, + "loss": 0.567, + "step": 4125 + }, + { + "epoch": 3.4157923799006076, + "grad_norm": 0.9483271241188049, + "learning_rate": 2.002975154583039e-06, + "loss": 0.5687, + "step": 4126 + }, + { + "epoch": 3.4166206515737163, + "grad_norm": 0.9517681002616882, + "learning_rate": 2.001897465239505e-06, + "loss": 0.5656, + "step": 4127 + }, + { + "epoch": 3.417448923246825, + "grad_norm": 0.9440080523490906, + "learning_rate": 2.0008198722741677e-06, + "loss": 0.5751, + "step": 4128 + }, + { + "epoch": 3.4182771949199338, + "grad_norm": 0.9267042279243469, + "learning_rate": 1.999742375895533e-06, + "loss": 0.5708, + "step": 4129 + }, + { + "epoch": 3.4191054665930425, + "grad_norm": 0.9693537950515747, + "learning_rate": 1.998664976312084e-06, + "loss": 0.5859, + "step": 4130 + }, + { + "epoch": 3.419933738266151, + "grad_norm": 0.9269915223121643, + "learning_rate": 1.997587673732289e-06, + "loss": 0.5804, + "step": 4131 + }, + { + "epoch": 3.42076200993926, + "grad_norm": 0.9441049695014954, + "learning_rate": 1.9965104683645957e-06, + "loss": 0.5547, + "step": 4132 + }, + { + "epoch": 3.4215902816123687, + "grad_norm": 0.9141139984130859, + "learning_rate": 1.9954333604174325e-06, + "loss": 0.5852, + "step": 4133 + }, + { + "epoch": 3.4224185532854774, + "grad_norm": 0.9308720827102661, + "learning_rate": 1.9943563500992115e-06, + "loss": 0.5446, + "step": 4134 + }, + { + "epoch": 3.4232468249585866, + "grad_norm": 0.9180085062980652, + "learning_rate": 1.9932794376183233e-06, + "loss": 0.5837, + "step": 4135 + }, + { + "epoch": 3.4240750966316953, + "grad_norm": 0.9463955163955688, + "learning_rate": 1.99220262318314e-06, + "loss": 0.5622, + "step": 4136 + }, + { + "epoch": 3.424903368304804, + "grad_norm": 0.9471439719200134, + "learning_rate": 1.991125907002015e-06, + "loss": 0.5682, + "step": 4137 + }, + { + "epoch": 3.4257316399779127, + "grad_norm": 0.9394784569740295, + "learning_rate": 1.990049289283284e-06, + "loss": 0.5702, + "step": 4138 + }, + { + "epoch": 3.4265599116510215, + "grad_norm": 1.005246639251709, + "learning_rate": 1.9889727702352618e-06, + "loss": 0.5772, + "step": 4139 + }, + { + "epoch": 3.42738818332413, + "grad_norm": 0.9534215927124023, + "learning_rate": 1.9878963500662453e-06, + "loss": 0.5626, + "step": 4140 + }, + { + "epoch": 3.428216454997239, + "grad_norm": 1.010836124420166, + "learning_rate": 1.98682002898451e-06, + "loss": 0.5748, + "step": 4141 + }, + { + "epoch": 3.429044726670348, + "grad_norm": 0.9219326376914978, + "learning_rate": 1.9857438071983163e-06, + "loss": 0.5893, + "step": 4142 + }, + { + "epoch": 3.429872998343457, + "grad_norm": 0.9372469186782837, + "learning_rate": 1.9846676849159017e-06, + "loss": 0.5555, + "step": 4143 + }, + { + "epoch": 3.4307012700165656, + "grad_norm": 0.9133549928665161, + "learning_rate": 1.9835916623454865e-06, + "loss": 0.5536, + "step": 4144 + }, + { + "epoch": 3.4315295416896743, + "grad_norm": 0.9400215148925781, + "learning_rate": 1.98251573969527e-06, + "loss": 0.5489, + "step": 4145 + }, + { + "epoch": 3.432357813362783, + "grad_norm": 0.9408988356590271, + "learning_rate": 1.9814399171734347e-06, + "loss": 0.5533, + "step": 4146 + }, + { + "epoch": 3.4331860850358917, + "grad_norm": 0.9352306723594666, + "learning_rate": 1.9803641949881415e-06, + "loss": 0.5645, + "step": 4147 + }, + { + "epoch": 3.4340143567090005, + "grad_norm": 0.9516900181770325, + "learning_rate": 1.979288573347532e-06, + "loss": 0.5764, + "step": 4148 + }, + { + "epoch": 3.434842628382109, + "grad_norm": 0.945179283618927, + "learning_rate": 1.97821305245973e-06, + "loss": 0.5792, + "step": 4149 + }, + { + "epoch": 3.435670900055218, + "grad_norm": 0.9419721961021423, + "learning_rate": 1.9771376325328377e-06, + "loss": 0.5598, + "step": 4150 + }, + { + "epoch": 3.436499171728327, + "grad_norm": 0.9680830836296082, + "learning_rate": 1.9760623137749394e-06, + "loss": 0.5675, + "step": 4151 + }, + { + "epoch": 3.437327443401436, + "grad_norm": 0.9370323419570923, + "learning_rate": 1.974987096394101e-06, + "loss": 0.5657, + "step": 4152 + }, + { + "epoch": 3.4381557150745445, + "grad_norm": 0.9793617725372314, + "learning_rate": 1.973911980598364e-06, + "loss": 0.5692, + "step": 4153 + }, + { + "epoch": 3.4389839867476533, + "grad_norm": 0.9549874067306519, + "learning_rate": 1.9728369665957547e-06, + "loss": 0.5798, + "step": 4154 + }, + { + "epoch": 3.439812258420762, + "grad_norm": 0.9228386878967285, + "learning_rate": 1.9717620545942784e-06, + "loss": 0.5759, + "step": 4155 + }, + { + "epoch": 3.4406405300938707, + "grad_norm": 0.9429793357849121, + "learning_rate": 1.97068724480192e-06, + "loss": 0.5618, + "step": 4156 + }, + { + "epoch": 3.4414688017669794, + "grad_norm": 0.9321681261062622, + "learning_rate": 1.969612537426646e-06, + "loss": 0.5602, + "step": 4157 + }, + { + "epoch": 3.442297073440088, + "grad_norm": 0.9411936402320862, + "learning_rate": 1.9685379326764014e-06, + "loss": 0.5787, + "step": 4158 + }, + { + "epoch": 3.443125345113197, + "grad_norm": 0.9223849177360535, + "learning_rate": 1.967463430759112e-06, + "loss": 0.5565, + "step": 4159 + }, + { + "epoch": 3.443953616786306, + "grad_norm": 0.9482093453407288, + "learning_rate": 1.966389031882686e-06, + "loss": 0.5729, + "step": 4160 + }, + { + "epoch": 3.444781888459415, + "grad_norm": 0.9146328568458557, + "learning_rate": 1.9653147362550075e-06, + "loss": 0.5661, + "step": 4161 + }, + { + "epoch": 3.4456101601325235, + "grad_norm": 0.9412453770637512, + "learning_rate": 1.9642405440839425e-06, + "loss": 0.5622, + "step": 4162 + }, + { + "epoch": 3.4464384318056323, + "grad_norm": 0.9408630132675171, + "learning_rate": 1.96316645557734e-06, + "loss": 0.5542, + "step": 4163 + }, + { + "epoch": 3.447266703478741, + "grad_norm": 0.9337356090545654, + "learning_rate": 1.962092470943022e-06, + "loss": 0.5634, + "step": 4164 + }, + { + "epoch": 3.4480949751518497, + "grad_norm": 0.9458538293838501, + "learning_rate": 1.9610185903887973e-06, + "loss": 0.5796, + "step": 4165 + }, + { + "epoch": 3.4489232468249584, + "grad_norm": 0.9226371049880981, + "learning_rate": 1.9599448141224507e-06, + "loss": 0.5596, + "step": 4166 + }, + { + "epoch": 3.4497515184980676, + "grad_norm": 0.947045087814331, + "learning_rate": 1.958871142351748e-06, + "loss": 0.554, + "step": 4167 + }, + { + "epoch": 3.4505797901711763, + "grad_norm": 0.9368892908096313, + "learning_rate": 1.9577975752844357e-06, + "loss": 0.5708, + "step": 4168 + }, + { + "epoch": 3.451408061844285, + "grad_norm": 0.9360406398773193, + "learning_rate": 1.9567241131282378e-06, + "loss": 0.5664, + "step": 4169 + }, + { + "epoch": 3.452236333517394, + "grad_norm": 0.9660393595695496, + "learning_rate": 1.9556507560908594e-06, + "loss": 0.5597, + "step": 4170 + }, + { + "epoch": 3.4530646051905025, + "grad_norm": 0.9472611546516418, + "learning_rate": 1.9545775043799858e-06, + "loss": 0.5704, + "step": 4171 + }, + { + "epoch": 3.4538928768636112, + "grad_norm": 0.941719651222229, + "learning_rate": 1.953504358203281e-06, + "loss": 0.5713, + "step": 4172 + }, + { + "epoch": 3.45472114853672, + "grad_norm": 0.9945985674858093, + "learning_rate": 1.9524313177683873e-06, + "loss": 0.563, + "step": 4173 + }, + { + "epoch": 3.4555494202098287, + "grad_norm": 0.9355506300926208, + "learning_rate": 1.9513583832829314e-06, + "loss": 0.5677, + "step": 4174 + }, + { + "epoch": 3.4563776918829374, + "grad_norm": 0.9413464665412903, + "learning_rate": 1.950285554954512e-06, + "loss": 0.5416, + "step": 4175 + }, + { + "epoch": 3.4572059635560466, + "grad_norm": 0.9668782353401184, + "learning_rate": 1.949212832990714e-06, + "loss": 0.5832, + "step": 4176 + }, + { + "epoch": 3.4580342352291553, + "grad_norm": 0.9598191976547241, + "learning_rate": 1.948140217599099e-06, + "loss": 0.597, + "step": 4177 + }, + { + "epoch": 3.458862506902264, + "grad_norm": 0.9517234563827515, + "learning_rate": 1.9470677089872064e-06, + "loss": 0.5812, + "step": 4178 + }, + { + "epoch": 3.4596907785753728, + "grad_norm": 0.9452085494995117, + "learning_rate": 1.9459953073625586e-06, + "loss": 0.5593, + "step": 4179 + }, + { + "epoch": 3.4605190502484815, + "grad_norm": 0.9393478035926819, + "learning_rate": 1.9449230129326543e-06, + "loss": 0.5828, + "step": 4180 + }, + { + "epoch": 3.4613473219215902, + "grad_norm": 0.9487680792808533, + "learning_rate": 1.9438508259049716e-06, + "loss": 0.5884, + "step": 4181 + }, + { + "epoch": 3.462175593594699, + "grad_norm": 0.9236283898353577, + "learning_rate": 1.942778746486971e-06, + "loss": 0.5534, + "step": 4182 + }, + { + "epoch": 3.4630038652678077, + "grad_norm": 0.9297188520431519, + "learning_rate": 1.941706774886088e-06, + "loss": 0.5702, + "step": 4183 + }, + { + "epoch": 3.4638321369409164, + "grad_norm": 0.9298705458641052, + "learning_rate": 1.94063491130974e-06, + "loss": 0.5645, + "step": 4184 + }, + { + "epoch": 3.4646604086140256, + "grad_norm": 0.93392014503479, + "learning_rate": 1.9395631559653223e-06, + "loss": 0.5805, + "step": 4185 + }, + { + "epoch": 3.4654886802871343, + "grad_norm": 0.9221041798591614, + "learning_rate": 1.93849150906021e-06, + "loss": 0.5746, + "step": 4186 + }, + { + "epoch": 3.466316951960243, + "grad_norm": 0.9339867830276489, + "learning_rate": 1.937419970801756e-06, + "loss": 0.5557, + "step": 4187 + }, + { + "epoch": 3.4671452236333518, + "grad_norm": 0.9240173697471619, + "learning_rate": 1.9363485413972934e-06, + "loss": 0.5616, + "step": 4188 + }, + { + "epoch": 3.4679734953064605, + "grad_norm": 0.9132168889045715, + "learning_rate": 1.9352772210541332e-06, + "loss": 0.5665, + "step": 4189 + }, + { + "epoch": 3.468801766979569, + "grad_norm": 0.9252684116363525, + "learning_rate": 1.934206009979567e-06, + "loss": 0.5534, + "step": 4190 + }, + { + "epoch": 3.469630038652678, + "grad_norm": 0.940039873123169, + "learning_rate": 1.933134908380864e-06, + "loss": 0.5686, + "step": 4191 + }, + { + "epoch": 3.4704583103257867, + "grad_norm": 0.9629718065261841, + "learning_rate": 1.9320639164652706e-06, + "loss": 0.6051, + "step": 4192 + }, + { + "epoch": 3.4712865819988954, + "grad_norm": 0.9440659880638123, + "learning_rate": 1.9309930344400164e-06, + "loss": 0.5612, + "step": 4193 + }, + { + "epoch": 3.4721148536720046, + "grad_norm": 0.9437550902366638, + "learning_rate": 1.9299222625123054e-06, + "loss": 0.5848, + "step": 4194 + }, + { + "epoch": 3.4729431253451133, + "grad_norm": 1.0566316843032837, + "learning_rate": 1.928851600889322e-06, + "loss": 0.5655, + "step": 4195 + }, + { + "epoch": 3.473771397018222, + "grad_norm": 0.9688880443572998, + "learning_rate": 1.92778104977823e-06, + "loss": 0.5792, + "step": 4196 + }, + { + "epoch": 3.4745996686913307, + "grad_norm": 0.9409403204917908, + "learning_rate": 1.9267106093861716e-06, + "loss": 0.5659, + "step": 4197 + }, + { + "epoch": 3.4754279403644395, + "grad_norm": 0.948984682559967, + "learning_rate": 1.9256402799202657e-06, + "loss": 0.5565, + "step": 4198 + }, + { + "epoch": 3.476256212037548, + "grad_norm": 0.9317809343338013, + "learning_rate": 1.924570061587611e-06, + "loss": 0.5739, + "step": 4199 + }, + { + "epoch": 3.477084483710657, + "grad_norm": 0.9296276569366455, + "learning_rate": 1.923499954595285e-06, + "loss": 0.5599, + "step": 4200 + }, + { + "epoch": 3.477912755383766, + "grad_norm": 0.9425923228263855, + "learning_rate": 1.922429959150344e-06, + "loss": 0.5686, + "step": 4201 + }, + { + "epoch": 3.478741027056875, + "grad_norm": 0.935650110244751, + "learning_rate": 1.9213600754598225e-06, + "loss": 0.5637, + "step": 4202 + }, + { + "epoch": 3.4795692987299836, + "grad_norm": 0.9613479375839233, + "learning_rate": 1.9202903037307312e-06, + "loss": 0.5766, + "step": 4203 + }, + { + "epoch": 3.4803975704030923, + "grad_norm": 0.9408090710639954, + "learning_rate": 1.9192206441700636e-06, + "loss": 0.5751, + "step": 4204 + }, + { + "epoch": 3.481225842076201, + "grad_norm": 0.9520177245140076, + "learning_rate": 1.9181510969847865e-06, + "loss": 0.5736, + "step": 4205 + }, + { + "epoch": 3.4820541137493097, + "grad_norm": 0.963941752910614, + "learning_rate": 1.9170816623818484e-06, + "loss": 0.5662, + "step": 4206 + }, + { + "epoch": 3.4828823854224185, + "grad_norm": 0.950707733631134, + "learning_rate": 1.9160123405681743e-06, + "loss": 0.576, + "step": 4207 + }, + { + "epoch": 3.483710657095527, + "grad_norm": 0.9403618574142456, + "learning_rate": 1.9149431317506693e-06, + "loss": 0.6119, + "step": 4208 + }, + { + "epoch": 3.484538928768636, + "grad_norm": 0.9552726745605469, + "learning_rate": 1.913874036136214e-06, + "loss": 0.5616, + "step": 4209 + }, + { + "epoch": 3.485367200441745, + "grad_norm": 0.9421485662460327, + "learning_rate": 1.912805053931669e-06, + "loss": 0.5708, + "step": 4210 + }, + { + "epoch": 3.486195472114854, + "grad_norm": 0.9480198621749878, + "learning_rate": 1.911736185343872e-06, + "loss": 0.5814, + "step": 4211 + }, + { + "epoch": 3.4870237437879625, + "grad_norm": 0.911723792552948, + "learning_rate": 1.91066743057964e-06, + "loss": 0.57, + "step": 4212 + }, + { + "epoch": 3.4878520154610713, + "grad_norm": 0.9331412315368652, + "learning_rate": 1.909598789845766e-06, + "loss": 0.5624, + "step": 4213 + }, + { + "epoch": 3.48868028713418, + "grad_norm": 1.0301762819290161, + "learning_rate": 1.9085302633490227e-06, + "loss": 0.5433, + "step": 4214 + }, + { + "epoch": 3.4895085588072887, + "grad_norm": 0.9518493413925171, + "learning_rate": 1.9074618512961597e-06, + "loss": 0.5702, + "step": 4215 + }, + { + "epoch": 3.4903368304803974, + "grad_norm": 0.9399821162223816, + "learning_rate": 1.9063935538939049e-06, + "loss": 0.5584, + "step": 4216 + }, + { + "epoch": 3.491165102153506, + "grad_norm": 0.9549731016159058, + "learning_rate": 1.9053253713489643e-06, + "loss": 0.5664, + "step": 4217 + }, + { + "epoch": 3.491993373826615, + "grad_norm": 0.9505523443222046, + "learning_rate": 1.9042573038680198e-06, + "loss": 0.5763, + "step": 4218 + }, + { + "epoch": 3.492821645499724, + "grad_norm": 0.9283603429794312, + "learning_rate": 1.9031893516577343e-06, + "loss": 0.5902, + "step": 4219 + }, + { + "epoch": 3.493649917172833, + "grad_norm": 0.9389188885688782, + "learning_rate": 1.9021215149247468e-06, + "loss": 0.5535, + "step": 4220 + }, + { + "epoch": 3.4944781888459415, + "grad_norm": 0.9431526064872742, + "learning_rate": 1.9010537938756719e-06, + "loss": 0.5698, + "step": 4221 + }, + { + "epoch": 3.4953064605190503, + "grad_norm": 0.9633996486663818, + "learning_rate": 1.8999861887171045e-06, + "loss": 0.573, + "step": 4222 + }, + { + "epoch": 3.496134732192159, + "grad_norm": 0.9711135625839233, + "learning_rate": 1.8989186996556161e-06, + "loss": 0.5603, + "step": 4223 + }, + { + "epoch": 3.4969630038652677, + "grad_norm": 0.9533023238182068, + "learning_rate": 1.8978513268977564e-06, + "loss": 0.5543, + "step": 4224 + }, + { + "epoch": 3.4977912755383764, + "grad_norm": 0.9520569443702698, + "learning_rate": 1.8967840706500519e-06, + "loss": 0.5859, + "step": 4225 + }, + { + "epoch": 3.4986195472114856, + "grad_norm": 0.952743411064148, + "learning_rate": 1.895716931119006e-06, + "loss": 0.5687, + "step": 4226 + }, + { + "epoch": 3.4994478188845943, + "grad_norm": 0.9404147267341614, + "learning_rate": 1.8946499085111014e-06, + "loss": 0.5669, + "step": 4227 + }, + { + "epoch": 3.500276090557703, + "grad_norm": 0.962302029132843, + "learning_rate": 1.8935830030327967e-06, + "loss": 0.5721, + "step": 4228 + }, + { + "epoch": 3.501104362230812, + "grad_norm": 0.9707161784172058, + "learning_rate": 1.8925162148905275e-06, + "loss": 0.5858, + "step": 4229 + }, + { + "epoch": 3.5019326339039205, + "grad_norm": 0.9720497727394104, + "learning_rate": 1.891449544290708e-06, + "loss": 0.5892, + "step": 4230 + }, + { + "epoch": 3.5027609055770292, + "grad_norm": 0.9394986629486084, + "learning_rate": 1.8903829914397295e-06, + "loss": 0.547, + "step": 4231 + }, + { + "epoch": 3.503589177250138, + "grad_norm": 0.9768742322921753, + "learning_rate": 1.8893165565439588e-06, + "loss": 0.5647, + "step": 4232 + }, + { + "epoch": 3.5044174489232467, + "grad_norm": 0.9559404253959656, + "learning_rate": 1.888250239809742e-06, + "loss": 0.556, + "step": 4233 + }, + { + "epoch": 3.5052457205963554, + "grad_norm": 0.9628231525421143, + "learning_rate": 1.8871840414434e-06, + "loss": 0.5633, + "step": 4234 + }, + { + "epoch": 3.5060739922694646, + "grad_norm": 0.9246355891227722, + "learning_rate": 1.8861179616512346e-06, + "loss": 0.5678, + "step": 4235 + }, + { + "epoch": 3.5069022639425733, + "grad_norm": 0.9590763449668884, + "learning_rate": 1.8850520006395206e-06, + "loss": 0.5612, + "step": 4236 + }, + { + "epoch": 3.507730535615682, + "grad_norm": 0.9386741518974304, + "learning_rate": 1.8839861586145117e-06, + "loss": 0.5737, + "step": 4237 + }, + { + "epoch": 3.5085588072887908, + "grad_norm": 0.9735057353973389, + "learning_rate": 1.8829204357824394e-06, + "loss": 0.5725, + "step": 4238 + }, + { + "epoch": 3.5093870789618995, + "grad_norm": 0.9464735984802246, + "learning_rate": 1.8818548323495101e-06, + "loss": 0.561, + "step": 4239 + }, + { + "epoch": 3.5102153506350082, + "grad_norm": 0.9822002053260803, + "learning_rate": 1.880789348521908e-06, + "loss": 0.562, + "step": 4240 + }, + { + "epoch": 3.511043622308117, + "grad_norm": 0.9448258876800537, + "learning_rate": 1.879723984505795e-06, + "loss": 0.5526, + "step": 4241 + }, + { + "epoch": 3.5118718939812257, + "grad_norm": 0.9565535187721252, + "learning_rate": 1.87865874050731e-06, + "loss": 0.5631, + "step": 4242 + }, + { + "epoch": 3.5127001656543344, + "grad_norm": 0.9824585318565369, + "learning_rate": 1.877593616732566e-06, + "loss": 0.5924, + "step": 4243 + }, + { + "epoch": 3.5135284373274436, + "grad_norm": 0.9289377927780151, + "learning_rate": 1.8765286133876549e-06, + "loss": 0.5667, + "step": 4244 + }, + { + "epoch": 3.5143567090005523, + "grad_norm": 0.9740702509880066, + "learning_rate": 1.8754637306786447e-06, + "loss": 0.567, + "step": 4245 + }, + { + "epoch": 3.515184980673661, + "grad_norm": 0.9304373264312744, + "learning_rate": 1.8743989688115813e-06, + "loss": 0.5973, + "step": 4246 + }, + { + "epoch": 3.5160132523467698, + "grad_norm": 0.9417708516120911, + "learning_rate": 1.8733343279924857e-06, + "loss": 0.5663, + "step": 4247 + }, + { + "epoch": 3.5168415240198785, + "grad_norm": 0.9550634622573853, + "learning_rate": 1.8722698084273555e-06, + "loss": 0.5829, + "step": 4248 + }, + { + "epoch": 3.517669795692987, + "grad_norm": 0.9428004622459412, + "learning_rate": 1.8712054103221661e-06, + "loss": 0.5595, + "step": 4249 + }, + { + "epoch": 3.518498067366096, + "grad_norm": 0.9728983640670776, + "learning_rate": 1.8701411338828685e-06, + "loss": 0.5611, + "step": 4250 + }, + { + "epoch": 3.519326339039205, + "grad_norm": 0.9784600138664246, + "learning_rate": 1.869076979315389e-06, + "loss": 0.5699, + "step": 4251 + }, + { + "epoch": 3.5201546107123134, + "grad_norm": 0.9751117825508118, + "learning_rate": 1.8680129468256336e-06, + "loss": 0.5622, + "step": 4252 + }, + { + "epoch": 3.5209828823854226, + "grad_norm": 0.9501684904098511, + "learning_rate": 1.866949036619482e-06, + "loss": 0.5594, + "step": 4253 + }, + { + "epoch": 3.5218111540585313, + "grad_norm": 0.9380241632461548, + "learning_rate": 1.86588524890279e-06, + "loss": 0.5759, + "step": 4254 + }, + { + "epoch": 3.52263942573164, + "grad_norm": 0.9474608898162842, + "learning_rate": 1.8648215838813915e-06, + "loss": 0.592, + "step": 4255 + }, + { + "epoch": 3.5234676974047487, + "grad_norm": 0.9680253863334656, + "learning_rate": 1.8637580417610949e-06, + "loss": 0.5369, + "step": 4256 + }, + { + "epoch": 3.5242959690778575, + "grad_norm": 0.9809057712554932, + "learning_rate": 1.8626946227476867e-06, + "loss": 0.5678, + "step": 4257 + }, + { + "epoch": 3.525124240750966, + "grad_norm": 0.9424263834953308, + "learning_rate": 1.8616313270469277e-06, + "loss": 0.5735, + "step": 4258 + }, + { + "epoch": 3.525952512424075, + "grad_norm": 0.9373759627342224, + "learning_rate": 1.860568154864556e-06, + "loss": 0.5572, + "step": 4259 + }, + { + "epoch": 3.526780784097184, + "grad_norm": 0.9195491075515747, + "learning_rate": 1.8595051064062864e-06, + "loss": 0.5552, + "step": 4260 + }, + { + "epoch": 3.5276090557702924, + "grad_norm": 0.9787479043006897, + "learning_rate": 1.8584421818778076e-06, + "loss": 0.5486, + "step": 4261 + }, + { + "epoch": 3.5284373274434015, + "grad_norm": 0.9167375564575195, + "learning_rate": 1.8573793814847856e-06, + "loss": 0.561, + "step": 4262 + }, + { + "epoch": 3.5292655991165103, + "grad_norm": 0.9589730501174927, + "learning_rate": 1.8563167054328635e-06, + "loss": 0.5792, + "step": 4263 + }, + { + "epoch": 3.530093870789619, + "grad_norm": 0.9304484724998474, + "learning_rate": 1.8552541539276587e-06, + "loss": 0.5628, + "step": 4264 + }, + { + "epoch": 3.5309221424627277, + "grad_norm": 0.9468231201171875, + "learning_rate": 1.8541917271747653e-06, + "loss": 0.5705, + "step": 4265 + }, + { + "epoch": 3.5317504141358365, + "grad_norm": 0.9367941617965698, + "learning_rate": 1.8531294253797522e-06, + "loss": 0.5769, + "step": 4266 + }, + { + "epoch": 3.532578685808945, + "grad_norm": 0.9852622151374817, + "learning_rate": 1.8520672487481645e-06, + "loss": 0.572, + "step": 4267 + }, + { + "epoch": 3.533406957482054, + "grad_norm": 0.9441319108009338, + "learning_rate": 1.8510051974855244e-06, + "loss": 0.5587, + "step": 4268 + }, + { + "epoch": 3.534235229155163, + "grad_norm": 0.9801215529441833, + "learning_rate": 1.8499432717973293e-06, + "loss": 0.6019, + "step": 4269 + }, + { + "epoch": 3.535063500828272, + "grad_norm": 0.9506904482841492, + "learning_rate": 1.8488814718890505e-06, + "loss": 0.569, + "step": 4270 + }, + { + "epoch": 3.5358917725013805, + "grad_norm": 0.9413098096847534, + "learning_rate": 1.8478197979661377e-06, + "loss": 0.5682, + "step": 4271 + }, + { + "epoch": 3.5367200441744893, + "grad_norm": 0.9410218596458435, + "learning_rate": 1.8467582502340149e-06, + "loss": 0.5843, + "step": 4272 + }, + { + "epoch": 3.537548315847598, + "grad_norm": 0.9326966404914856, + "learning_rate": 1.845696828898081e-06, + "loss": 0.5853, + "step": 4273 + }, + { + "epoch": 3.5383765875207067, + "grad_norm": 0.9390862584114075, + "learning_rate": 1.8446355341637107e-06, + "loss": 0.5767, + "step": 4274 + }, + { + "epoch": 3.5392048591938154, + "grad_norm": 0.9381791949272156, + "learning_rate": 1.843574366236256e-06, + "loss": 0.5534, + "step": 4275 + }, + { + "epoch": 3.5400331308669246, + "grad_norm": 1.0620906352996826, + "learning_rate": 1.8425133253210425e-06, + "loss": 0.5663, + "step": 4276 + }, + { + "epoch": 3.540861402540033, + "grad_norm": 0.945441484451294, + "learning_rate": 1.8414524116233715e-06, + "loss": 0.5496, + "step": 4277 + }, + { + "epoch": 3.541689674213142, + "grad_norm": 0.9375214576721191, + "learning_rate": 1.8403916253485193e-06, + "loss": 0.5556, + "step": 4278 + }, + { + "epoch": 3.542517945886251, + "grad_norm": 0.9695723056793213, + "learning_rate": 1.8393309667017388e-06, + "loss": 0.5457, + "step": 4279 + }, + { + "epoch": 3.5433462175593595, + "grad_norm": 0.9985876679420471, + "learning_rate": 1.8382704358882583e-06, + "loss": 0.5749, + "step": 4280 + }, + { + "epoch": 3.5441744892324683, + "grad_norm": 0.9784291386604309, + "learning_rate": 1.8372100331132797e-06, + "loss": 0.5671, + "step": 4281 + }, + { + "epoch": 3.545002760905577, + "grad_norm": 0.9699963331222534, + "learning_rate": 1.8361497585819804e-06, + "loss": 0.5742, + "step": 4282 + }, + { + "epoch": 3.5458310325786857, + "grad_norm": 0.951996386051178, + "learning_rate": 1.8350896124995155e-06, + "loss": 0.5729, + "step": 4283 + }, + { + "epoch": 3.5466593042517944, + "grad_norm": 0.9519326686859131, + "learning_rate": 1.8340295950710119e-06, + "loss": 0.5746, + "step": 4284 + }, + { + "epoch": 3.5474875759249036, + "grad_norm": 0.9227757453918457, + "learning_rate": 1.8329697065015734e-06, + "loss": 0.5441, + "step": 4285 + }, + { + "epoch": 3.548315847598012, + "grad_norm": 0.9924660325050354, + "learning_rate": 1.8319099469962787e-06, + "loss": 0.5695, + "step": 4286 + }, + { + "epoch": 3.549144119271121, + "grad_norm": 0.9582194089889526, + "learning_rate": 1.8308503167601827e-06, + "loss": 0.5757, + "step": 4287 + }, + { + "epoch": 3.54997239094423, + "grad_norm": 0.924054741859436, + "learning_rate": 1.8297908159983114e-06, + "loss": 0.5739, + "step": 4288 + }, + { + "epoch": 3.5508006626173385, + "grad_norm": 0.9424325227737427, + "learning_rate": 1.82873144491567e-06, + "loss": 0.6041, + "step": 4289 + }, + { + "epoch": 3.5516289342904472, + "grad_norm": 0.9287232160568237, + "learning_rate": 1.827672203717236e-06, + "loss": 0.5494, + "step": 4290 + }, + { + "epoch": 3.552457205963556, + "grad_norm": 0.94130539894104, + "learning_rate": 1.8266130926079637e-06, + "loss": 0.5642, + "step": 4291 + }, + { + "epoch": 3.5532854776366647, + "grad_norm": 0.933063805103302, + "learning_rate": 1.8255541117927805e-06, + "loss": 0.5759, + "step": 4292 + }, + { + "epoch": 3.5541137493097734, + "grad_norm": 0.9383630156517029, + "learning_rate": 1.8244952614765894e-06, + "loss": 0.5584, + "step": 4293 + }, + { + "epoch": 3.5549420209828826, + "grad_norm": 0.9373190999031067, + "learning_rate": 1.8234365418642685e-06, + "loss": 0.5597, + "step": 4294 + }, + { + "epoch": 3.5557702926559913, + "grad_norm": 0.9637420773506165, + "learning_rate": 1.82237795316067e-06, + "loss": 0.5736, + "step": 4295 + }, + { + "epoch": 3.5565985643291, + "grad_norm": 0.9421122670173645, + "learning_rate": 1.8213194955706204e-06, + "loss": 0.5424, + "step": 4296 + }, + { + "epoch": 3.5574268360022088, + "grad_norm": 0.9612517952919006, + "learning_rate": 1.8202611692989224e-06, + "loss": 0.5704, + "step": 4297 + }, + { + "epoch": 3.5582551076753175, + "grad_norm": 0.9415401220321655, + "learning_rate": 1.8192029745503515e-06, + "loss": 0.5774, + "step": 4298 + }, + { + "epoch": 3.5590833793484262, + "grad_norm": 0.9469128251075745, + "learning_rate": 1.8181449115296596e-06, + "loss": 0.5721, + "step": 4299 + }, + { + "epoch": 3.559911651021535, + "grad_norm": 0.9917946457862854, + "learning_rate": 1.8170869804415703e-06, + "loss": 0.5767, + "step": 4300 + }, + { + "epoch": 3.5607399226946437, + "grad_norm": 0.9557772874832153, + "learning_rate": 1.8160291814907835e-06, + "loss": 0.5677, + "step": 4301 + }, + { + "epoch": 3.5615681943677524, + "grad_norm": 0.9430398344993591, + "learning_rate": 1.8149715148819752e-06, + "loss": 0.5679, + "step": 4302 + }, + { + "epoch": 3.5623964660408616, + "grad_norm": 0.9426287412643433, + "learning_rate": 1.8139139808197926e-06, + "loss": 0.563, + "step": 4303 + }, + { + "epoch": 3.5632247377139703, + "grad_norm": 0.9437983632087708, + "learning_rate": 1.8128565795088588e-06, + "loss": 0.5703, + "step": 4304 + }, + { + "epoch": 3.564053009387079, + "grad_norm": 0.9386832118034363, + "learning_rate": 1.8117993111537717e-06, + "loss": 0.5842, + "step": 4305 + }, + { + "epoch": 3.5648812810601878, + "grad_norm": 0.954119086265564, + "learning_rate": 1.8107421759591031e-06, + "loss": 0.5515, + "step": 4306 + }, + { + "epoch": 3.5657095527332965, + "grad_norm": 0.945758581161499, + "learning_rate": 1.8096851741293974e-06, + "loss": 0.5455, + "step": 4307 + }, + { + "epoch": 3.566537824406405, + "grad_norm": 0.9291911721229553, + "learning_rate": 1.8086283058691759e-06, + "loss": 0.5843, + "step": 4308 + }, + { + "epoch": 3.567366096079514, + "grad_norm": 0.9620952606201172, + "learning_rate": 1.8075715713829323e-06, + "loss": 0.568, + "step": 4309 + }, + { + "epoch": 3.568194367752623, + "grad_norm": 0.9722322821617126, + "learning_rate": 1.8065149708751357e-06, + "loss": 0.5816, + "step": 4310 + }, + { + "epoch": 3.5690226394257314, + "grad_norm": 0.9354492425918579, + "learning_rate": 1.805458504550227e-06, + "loss": 0.5856, + "step": 4311 + }, + { + "epoch": 3.5698509110988406, + "grad_norm": 0.956453263759613, + "learning_rate": 1.8044021726126226e-06, + "loss": 0.5582, + "step": 4312 + }, + { + "epoch": 3.5706791827719493, + "grad_norm": 0.926895797252655, + "learning_rate": 1.8033459752667143e-06, + "loss": 0.5837, + "step": 4313 + }, + { + "epoch": 3.571507454445058, + "grad_norm": 0.9227427840232849, + "learning_rate": 1.8022899127168655e-06, + "loss": 0.5567, + "step": 4314 + }, + { + "epoch": 3.5723357261181667, + "grad_norm": 0.9473888278007507, + "learning_rate": 1.801233985167414e-06, + "loss": 0.5672, + "step": 4315 + }, + { + "epoch": 3.5731639977912755, + "grad_norm": 0.9397003054618835, + "learning_rate": 1.8001781928226734e-06, + "loss": 0.5663, + "step": 4316 + }, + { + "epoch": 3.573992269464384, + "grad_norm": 0.9553276300430298, + "learning_rate": 1.7991225358869286e-06, + "loss": 0.5716, + "step": 4317 + }, + { + "epoch": 3.574820541137493, + "grad_norm": 0.9149162769317627, + "learning_rate": 1.798067014564439e-06, + "loss": 0.5409, + "step": 4318 + }, + { + "epoch": 3.575648812810602, + "grad_norm": 1.0992645025253296, + "learning_rate": 1.7970116290594391e-06, + "loss": 0.5542, + "step": 4319 + }, + { + "epoch": 3.576477084483711, + "grad_norm": 0.9605867862701416, + "learning_rate": 1.7959563795761364e-06, + "loss": 0.5426, + "step": 4320 + }, + { + "epoch": 3.5773053561568195, + "grad_norm": 0.944625198841095, + "learning_rate": 1.7949012663187115e-06, + "loss": 0.5695, + "step": 4321 + }, + { + "epoch": 3.5781336278299283, + "grad_norm": 0.9884684085845947, + "learning_rate": 1.793846289491318e-06, + "loss": 0.5897, + "step": 4322 + }, + { + "epoch": 3.578961899503037, + "grad_norm": 0.9456211924552917, + "learning_rate": 1.7927914492980845e-06, + "loss": 0.5969, + "step": 4323 + }, + { + "epoch": 3.5797901711761457, + "grad_norm": 0.9710149168968201, + "learning_rate": 1.7917367459431135e-06, + "loss": 0.5797, + "step": 4324 + }, + { + "epoch": 3.5806184428492545, + "grad_norm": 0.9509037733078003, + "learning_rate": 1.7906821796304796e-06, + "loss": 0.5655, + "step": 4325 + }, + { + "epoch": 3.581446714522363, + "grad_norm": 0.9468769431114197, + "learning_rate": 1.7896277505642316e-06, + "loss": 0.5519, + "step": 4326 + }, + { + "epoch": 3.582274986195472, + "grad_norm": 0.9285255670547485, + "learning_rate": 1.7885734589483923e-06, + "loss": 0.5765, + "step": 4327 + }, + { + "epoch": 3.583103257868581, + "grad_norm": 0.9228923320770264, + "learning_rate": 1.7875193049869566e-06, + "loss": 0.5742, + "step": 4328 + }, + { + "epoch": 3.58393152954169, + "grad_norm": 0.9343339204788208, + "learning_rate": 1.7864652888838935e-06, + "loss": 0.5601, + "step": 4329 + }, + { + "epoch": 3.5847598012147985, + "grad_norm": 0.937486469745636, + "learning_rate": 1.7854114108431459e-06, + "loss": 0.5392, + "step": 4330 + }, + { + "epoch": 3.5855880728879073, + "grad_norm": 0.9428637027740479, + "learning_rate": 1.7843576710686288e-06, + "loss": 0.5502, + "step": 4331 + }, + { + "epoch": 3.586416344561016, + "grad_norm": 0.9170352220535278, + "learning_rate": 1.7833040697642312e-06, + "loss": 0.5707, + "step": 4332 + }, + { + "epoch": 3.5872446162341247, + "grad_norm": 0.9696611166000366, + "learning_rate": 1.7822506071338163e-06, + "loss": 0.5714, + "step": 4333 + }, + { + "epoch": 3.5880728879072334, + "grad_norm": 0.912272572517395, + "learning_rate": 1.7811972833812167e-06, + "loss": 0.5633, + "step": 4334 + }, + { + "epoch": 3.5889011595803426, + "grad_norm": 0.9608253836631775, + "learning_rate": 1.7801440987102424e-06, + "loss": 0.572, + "step": 4335 + }, + { + "epoch": 3.589729431253451, + "grad_norm": 0.9655832648277283, + "learning_rate": 1.7790910533246752e-06, + "loss": 0.5727, + "step": 4336 + }, + { + "epoch": 3.59055770292656, + "grad_norm": 0.9179590940475464, + "learning_rate": 1.7780381474282683e-06, + "loss": 0.5701, + "step": 4337 + }, + { + "epoch": 3.591385974599669, + "grad_norm": 0.9642294049263, + "learning_rate": 1.7769853812247501e-06, + "loss": 0.5659, + "step": 4338 + }, + { + "epoch": 3.5922142462727775, + "grad_norm": 0.9484195709228516, + "learning_rate": 1.7759327549178208e-06, + "loss": 0.5572, + "step": 4339 + }, + { + "epoch": 3.5930425179458862, + "grad_norm": 0.9366327524185181, + "learning_rate": 1.7748802687111533e-06, + "loss": 0.5596, + "step": 4340 + }, + { + "epoch": 3.593870789618995, + "grad_norm": 0.9256154894828796, + "learning_rate": 1.7738279228083946e-06, + "loss": 0.553, + "step": 4341 + }, + { + "epoch": 3.5946990612921037, + "grad_norm": 0.9243422746658325, + "learning_rate": 1.7727757174131637e-06, + "loss": 0.5703, + "step": 4342 + }, + { + "epoch": 3.5955273329652124, + "grad_norm": 0.9723832607269287, + "learning_rate": 1.7717236527290526e-06, + "loss": 0.5812, + "step": 4343 + }, + { + "epoch": 3.5963556046383216, + "grad_norm": 0.9459588527679443, + "learning_rate": 1.7706717289596263e-06, + "loss": 0.5871, + "step": 4344 + }, + { + "epoch": 3.59718387631143, + "grad_norm": 0.9725896716117859, + "learning_rate": 1.7696199463084205e-06, + "loss": 0.5594, + "step": 4345 + }, + { + "epoch": 3.598012147984539, + "grad_norm": 0.9436221122741699, + "learning_rate": 1.768568304978947e-06, + "loss": 0.5587, + "step": 4346 + }, + { + "epoch": 3.598840419657648, + "grad_norm": 0.9552313685417175, + "learning_rate": 1.7675168051746884e-06, + "loss": 0.572, + "step": 4347 + }, + { + "epoch": 3.5996686913307565, + "grad_norm": 0.9535266160964966, + "learning_rate": 1.7664654470990993e-06, + "loss": 0.5695, + "step": 4348 + }, + { + "epoch": 3.6004969630038652, + "grad_norm": 0.9626250863075256, + "learning_rate": 1.7654142309556088e-06, + "loss": 0.5695, + "step": 4349 + }, + { + "epoch": 3.601325234676974, + "grad_norm": 0.9406429529190063, + "learning_rate": 1.7643631569476166e-06, + "loss": 0.5525, + "step": 4350 + }, + { + "epoch": 3.6021535063500827, + "grad_norm": 0.946359395980835, + "learning_rate": 1.7633122252784963e-06, + "loss": 0.5626, + "step": 4351 + }, + { + "epoch": 3.6029817780231914, + "grad_norm": 0.9508920907974243, + "learning_rate": 1.7622614361515928e-06, + "loss": 0.5569, + "step": 4352 + }, + { + "epoch": 3.6038100496963006, + "grad_norm": 0.9463246464729309, + "learning_rate": 1.7612107897702245e-06, + "loss": 0.5608, + "step": 4353 + }, + { + "epoch": 3.6046383213694093, + "grad_norm": 0.9517480731010437, + "learning_rate": 1.7601602863376816e-06, + "loss": 0.5753, + "step": 4354 + }, + { + "epoch": 3.605466593042518, + "grad_norm": 0.9344833493232727, + "learning_rate": 1.759109926057227e-06, + "loss": 0.5633, + "step": 4355 + }, + { + "epoch": 3.6062948647156268, + "grad_norm": 0.9408402442932129, + "learning_rate": 1.758059709132094e-06, + "loss": 0.5829, + "step": 4356 + }, + { + "epoch": 3.6071231363887355, + "grad_norm": 0.9551196098327637, + "learning_rate": 1.7570096357654918e-06, + "loss": 0.5553, + "step": 4357 + }, + { + "epoch": 3.6079514080618442, + "grad_norm": 0.9403008222579956, + "learning_rate": 1.7559597061605987e-06, + "loss": 0.576, + "step": 4358 + }, + { + "epoch": 3.608779679734953, + "grad_norm": 0.9454553127288818, + "learning_rate": 1.7549099205205673e-06, + "loss": 0.5643, + "step": 4359 + }, + { + "epoch": 3.609607951408062, + "grad_norm": 0.9868146777153015, + "learning_rate": 1.7538602790485194e-06, + "loss": 0.591, + "step": 4360 + }, + { + "epoch": 3.6104362230811704, + "grad_norm": 0.9502206444740295, + "learning_rate": 1.752810781947553e-06, + "loss": 0.5431, + "step": 4361 + }, + { + "epoch": 3.6112644947542796, + "grad_norm": 0.9451961517333984, + "learning_rate": 1.7517614294207347e-06, + "loss": 0.5762, + "step": 4362 + }, + { + "epoch": 3.6120927664273883, + "grad_norm": 0.9424105882644653, + "learning_rate": 1.7507122216711048e-06, + "loss": 0.6107, + "step": 4363 + }, + { + "epoch": 3.612921038100497, + "grad_norm": 0.967083215713501, + "learning_rate": 1.7496631589016754e-06, + "loss": 0.5486, + "step": 4364 + }, + { + "epoch": 3.6137493097736058, + "grad_norm": 0.9303386211395264, + "learning_rate": 1.7486142413154304e-06, + "loss": 0.5345, + "step": 4365 + }, + { + "epoch": 3.6145775814467145, + "grad_norm": 0.9635606408119202, + "learning_rate": 1.7475654691153257e-06, + "loss": 0.563, + "step": 4366 + }, + { + "epoch": 3.615405853119823, + "grad_norm": 0.9364803433418274, + "learning_rate": 1.7465168425042883e-06, + "loss": 0.561, + "step": 4367 + }, + { + "epoch": 3.616234124792932, + "grad_norm": 0.9707603454589844, + "learning_rate": 1.7454683616852172e-06, + "loss": 0.585, + "step": 4368 + }, + { + "epoch": 3.617062396466041, + "grad_norm": 0.9474611282348633, + "learning_rate": 1.744420026860985e-06, + "loss": 0.5754, + "step": 4369 + }, + { + "epoch": 3.6178906681391494, + "grad_norm": 0.9372676014900208, + "learning_rate": 1.7433718382344344e-06, + "loss": 0.5624, + "step": 4370 + }, + { + "epoch": 3.6187189398122586, + "grad_norm": 0.9487974643707275, + "learning_rate": 1.7423237960083794e-06, + "loss": 0.5406, + "step": 4371 + }, + { + "epoch": 3.6195472114853673, + "grad_norm": 0.9595673084259033, + "learning_rate": 1.7412759003856073e-06, + "loss": 0.5977, + "step": 4372 + }, + { + "epoch": 3.620375483158476, + "grad_norm": 0.954731285572052, + "learning_rate": 1.7402281515688756e-06, + "loss": 0.56, + "step": 4373 + }, + { + "epoch": 3.6212037548315847, + "grad_norm": 0.9508479237556458, + "learning_rate": 1.7391805497609137e-06, + "loss": 0.5607, + "step": 4374 + }, + { + "epoch": 3.6220320265046935, + "grad_norm": 0.9357259273529053, + "learning_rate": 1.7381330951644237e-06, + "loss": 0.5545, + "step": 4375 + }, + { + "epoch": 3.622860298177802, + "grad_norm": 0.9336346983909607, + "learning_rate": 1.737085787982078e-06, + "loss": 0.5636, + "step": 4376 + }, + { + "epoch": 3.623688569850911, + "grad_norm": 0.9238201379776001, + "learning_rate": 1.73603862841652e-06, + "loss": 0.5497, + "step": 4377 + }, + { + "epoch": 3.62451684152402, + "grad_norm": 0.9505547881126404, + "learning_rate": 1.7349916166703668e-06, + "loss": 0.5872, + "step": 4378 + }, + { + "epoch": 3.625345113197129, + "grad_norm": 0.945364773273468, + "learning_rate": 1.7339447529462033e-06, + "loss": 0.539, + "step": 4379 + }, + { + "epoch": 3.6261733848702375, + "grad_norm": 0.9546788930892944, + "learning_rate": 1.7328980374465896e-06, + "loss": 0.5489, + "step": 4380 + }, + { + "epoch": 3.6270016565433463, + "grad_norm": 0.9455381035804749, + "learning_rate": 1.7318514703740546e-06, + "loss": 0.5624, + "step": 4381 + }, + { + "epoch": 3.627829928216455, + "grad_norm": 0.9461554288864136, + "learning_rate": 1.7308050519310993e-06, + "loss": 0.5651, + "step": 4382 + }, + { + "epoch": 3.6286581998895637, + "grad_norm": 0.9569746255874634, + "learning_rate": 1.7297587823201964e-06, + "loss": 0.5792, + "step": 4383 + }, + { + "epoch": 3.6294864715626725, + "grad_norm": 0.9243269562721252, + "learning_rate": 1.7287126617437889e-06, + "loss": 0.5493, + "step": 4384 + }, + { + "epoch": 3.630314743235781, + "grad_norm": 0.9244057536125183, + "learning_rate": 1.727666690404291e-06, + "loss": 0.5487, + "step": 4385 + }, + { + "epoch": 3.63114301490889, + "grad_norm": 0.9674320220947266, + "learning_rate": 1.7266208685040896e-06, + "loss": 0.5572, + "step": 4386 + }, + { + "epoch": 3.631971286581999, + "grad_norm": 0.9546018838882446, + "learning_rate": 1.7255751962455407e-06, + "loss": 0.581, + "step": 4387 + }, + { + "epoch": 3.632799558255108, + "grad_norm": 0.9394852519035339, + "learning_rate": 1.7245296738309719e-06, + "loss": 0.5742, + "step": 4388 + }, + { + "epoch": 3.6336278299282165, + "grad_norm": 0.9570075273513794, + "learning_rate": 1.723484301462684e-06, + "loss": 0.5631, + "step": 4389 + }, + { + "epoch": 3.6344561016013253, + "grad_norm": 0.9604787230491638, + "learning_rate": 1.7224390793429435e-06, + "loss": 0.5616, + "step": 4390 + }, + { + "epoch": 3.635284373274434, + "grad_norm": 0.9690744280815125, + "learning_rate": 1.7213940076739931e-06, + "loss": 0.5912, + "step": 4391 + }, + { + "epoch": 3.6361126449475427, + "grad_norm": 0.9428882598876953, + "learning_rate": 1.720349086658044e-06, + "loss": 0.5763, + "step": 4392 + }, + { + "epoch": 3.6369409166206514, + "grad_norm": 0.9487605094909668, + "learning_rate": 1.7193043164972788e-06, + "loss": 0.5652, + "step": 4393 + }, + { + "epoch": 3.6377691882937606, + "grad_norm": 0.9717780351638794, + "learning_rate": 1.7182596973938512e-06, + "loss": 0.5897, + "step": 4394 + }, + { + "epoch": 3.638597459966869, + "grad_norm": 0.9824546575546265, + "learning_rate": 1.7172152295498848e-06, + "loss": 0.5647, + "step": 4395 + }, + { + "epoch": 3.639425731639978, + "grad_norm": 0.9322198629379272, + "learning_rate": 1.716170913167474e-06, + "loss": 0.5532, + "step": 4396 + }, + { + "epoch": 3.640254003313087, + "grad_norm": 0.9403592944145203, + "learning_rate": 1.715126748448685e-06, + "loss": 0.5768, + "step": 4397 + }, + { + "epoch": 3.6410822749861955, + "grad_norm": 0.9282305836677551, + "learning_rate": 1.7140827355955541e-06, + "loss": 0.5354, + "step": 4398 + }, + { + "epoch": 3.6419105466593042, + "grad_norm": 1.0102676153182983, + "learning_rate": 1.7130388748100868e-06, + "loss": 0.5792, + "step": 4399 + }, + { + "epoch": 3.642738818332413, + "grad_norm": 0.9756748676300049, + "learning_rate": 1.711995166294263e-06, + "loss": 0.5691, + "step": 4400 + }, + { + "epoch": 3.6435670900055217, + "grad_norm": 0.9297899603843689, + "learning_rate": 1.710951610250027e-06, + "loss": 0.548, + "step": 4401 + }, + { + "epoch": 3.6443953616786304, + "grad_norm": 0.9422688484191895, + "learning_rate": 1.7099082068792999e-06, + "loss": 0.5344, + "step": 4402 + }, + { + "epoch": 3.6452236333517396, + "grad_norm": 0.9583215713500977, + "learning_rate": 1.7088649563839693e-06, + "loss": 0.5652, + "step": 4403 + }, + { + "epoch": 3.646051905024848, + "grad_norm": 0.9639995098114014, + "learning_rate": 1.7078218589658946e-06, + "loss": 0.5649, + "step": 4404 + }, + { + "epoch": 3.646880176697957, + "grad_norm": 0.9570342898368835, + "learning_rate": 1.7067789148269056e-06, + "loss": 0.5513, + "step": 4405 + }, + { + "epoch": 3.647708448371066, + "grad_norm": 0.9547348618507385, + "learning_rate": 1.7057361241688024e-06, + "loss": 0.5659, + "step": 4406 + }, + { + "epoch": 3.6485367200441745, + "grad_norm": 0.9749144911766052, + "learning_rate": 1.7046934871933551e-06, + "loss": 0.5672, + "step": 4407 + }, + { + "epoch": 3.6493649917172832, + "grad_norm": 0.9632282853126526, + "learning_rate": 1.7036510041023045e-06, + "loss": 0.5761, + "step": 4408 + }, + { + "epoch": 3.650193263390392, + "grad_norm": 0.9284268617630005, + "learning_rate": 1.7026086750973612e-06, + "loss": 0.5758, + "step": 4409 + }, + { + "epoch": 3.6510215350635007, + "grad_norm": 0.9373230934143066, + "learning_rate": 1.7015665003802053e-06, + "loss": 0.5537, + "step": 4410 + }, + { + "epoch": 3.6518498067366094, + "grad_norm": 0.9570644497871399, + "learning_rate": 1.7005244801524896e-06, + "loss": 0.5922, + "step": 4411 + }, + { + "epoch": 3.6526780784097186, + "grad_norm": 0.9559333920478821, + "learning_rate": 1.6994826146158345e-06, + "loss": 0.5854, + "step": 4412 + }, + { + "epoch": 3.6535063500828273, + "grad_norm": 0.943041205406189, + "learning_rate": 1.698440903971831e-06, + "loss": 0.5475, + "step": 4413 + }, + { + "epoch": 3.654334621755936, + "grad_norm": 0.9560774564743042, + "learning_rate": 1.6973993484220403e-06, + "loss": 0.5863, + "step": 4414 + }, + { + "epoch": 3.6551628934290448, + "grad_norm": 0.9528188109397888, + "learning_rate": 1.6963579481679932e-06, + "loss": 0.5536, + "step": 4415 + }, + { + "epoch": 3.6559911651021535, + "grad_norm": 0.9496455192565918, + "learning_rate": 1.6953167034111926e-06, + "loss": 0.5896, + "step": 4416 + }, + { + "epoch": 3.656819436775262, + "grad_norm": 0.9072014689445496, + "learning_rate": 1.6942756143531086e-06, + "loss": 0.5665, + "step": 4417 + }, + { + "epoch": 3.657647708448371, + "grad_norm": 0.9448863863945007, + "learning_rate": 1.6932346811951817e-06, + "loss": 0.5765, + "step": 4418 + }, + { + "epoch": 3.65847598012148, + "grad_norm": 0.9441279768943787, + "learning_rate": 1.692193904138824e-06, + "loss": 0.5721, + "step": 4419 + }, + { + "epoch": 3.6593042517945884, + "grad_norm": 0.9575514793395996, + "learning_rate": 1.6911532833854155e-06, + "loss": 0.5657, + "step": 4420 + }, + { + "epoch": 3.6601325234676976, + "grad_norm": 1.0165443420410156, + "learning_rate": 1.690112819136307e-06, + "loss": 0.5853, + "step": 4421 + }, + { + "epoch": 3.6609607951408063, + "grad_norm": 1.0065933465957642, + "learning_rate": 1.6890725115928176e-06, + "loss": 0.559, + "step": 4422 + }, + { + "epoch": 3.661789066813915, + "grad_norm": 0.9567911028862, + "learning_rate": 1.6880323609562388e-06, + "loss": 0.5734, + "step": 4423 + }, + { + "epoch": 3.6626173384870238, + "grad_norm": 0.9482328295707703, + "learning_rate": 1.6869923674278288e-06, + "loss": 0.5704, + "step": 4424 + }, + { + "epoch": 3.6634456101601325, + "grad_norm": 0.9326344728469849, + "learning_rate": 1.6859525312088166e-06, + "loss": 0.5858, + "step": 4425 + }, + { + "epoch": 3.664273881833241, + "grad_norm": 0.9377856254577637, + "learning_rate": 1.6849128525004016e-06, + "loss": 0.5769, + "step": 4426 + }, + { + "epoch": 3.66510215350635, + "grad_norm": 0.939058244228363, + "learning_rate": 1.6838733315037506e-06, + "loss": 0.5574, + "step": 4427 + }, + { + "epoch": 3.665930425179459, + "grad_norm": 0.9490857720375061, + "learning_rate": 1.682833968420003e-06, + "loss": 0.5636, + "step": 4428 + }, + { + "epoch": 3.6667586968525674, + "grad_norm": 0.902042806148529, + "learning_rate": 1.6817947634502646e-06, + "loss": 0.5698, + "step": 4429 + }, + { + "epoch": 3.6675869685256766, + "grad_norm": 0.9587634205818176, + "learning_rate": 1.6807557167956118e-06, + "loss": 0.5509, + "step": 4430 + }, + { + "epoch": 3.6684152401987853, + "grad_norm": 0.9205626845359802, + "learning_rate": 1.6797168286570914e-06, + "loss": 0.5712, + "step": 4431 + }, + { + "epoch": 3.669243511871894, + "grad_norm": 0.9732058048248291, + "learning_rate": 1.6786780992357182e-06, + "loss": 0.5991, + "step": 4432 + }, + { + "epoch": 3.6700717835450027, + "grad_norm": 0.9664617776870728, + "learning_rate": 1.6776395287324763e-06, + "loss": 0.5626, + "step": 4433 + }, + { + "epoch": 3.6709000552181115, + "grad_norm": 0.943077802658081, + "learning_rate": 1.676601117348321e-06, + "loss": 0.5699, + "step": 4434 + }, + { + "epoch": 3.67172832689122, + "grad_norm": 0.9534650444984436, + "learning_rate": 1.6755628652841721e-06, + "loss": 0.5783, + "step": 4435 + }, + { + "epoch": 3.672556598564329, + "grad_norm": 0.9525501132011414, + "learning_rate": 1.6745247727409242e-06, + "loss": 0.5793, + "step": 4436 + }, + { + "epoch": 3.673384870237438, + "grad_norm": 0.9676088690757751, + "learning_rate": 1.6734868399194377e-06, + "loss": 0.5638, + "step": 4437 + }, + { + "epoch": 3.674213141910547, + "grad_norm": 0.9717739820480347, + "learning_rate": 1.6724490670205423e-06, + "loss": 0.585, + "step": 4438 + }, + { + "epoch": 3.6750414135836555, + "grad_norm": 0.9457798600196838, + "learning_rate": 1.6714114542450388e-06, + "loss": 0.5558, + "step": 4439 + }, + { + "epoch": 3.6758696852567643, + "grad_norm": 0.940519392490387, + "learning_rate": 1.6703740017936948e-06, + "loss": 0.5815, + "step": 4440 + }, + { + "epoch": 3.676697956929873, + "grad_norm": 0.9216038584709167, + "learning_rate": 1.6693367098672475e-06, + "loss": 0.5629, + "step": 4441 + }, + { + "epoch": 3.6775262286029817, + "grad_norm": 0.927438497543335, + "learning_rate": 1.668299578666404e-06, + "loss": 0.5742, + "step": 4442 + }, + { + "epoch": 3.6783545002760905, + "grad_norm": 0.9554510712623596, + "learning_rate": 1.6672626083918387e-06, + "loss": 0.5709, + "step": 4443 + }, + { + "epoch": 3.679182771949199, + "grad_norm": 0.976613461971283, + "learning_rate": 1.6662257992441957e-06, + "loss": 0.5867, + "step": 4444 + }, + { + "epoch": 3.680011043622308, + "grad_norm": 0.9392770528793335, + "learning_rate": 1.6651891514240887e-06, + "loss": 0.563, + "step": 4445 + }, + { + "epoch": 3.680839315295417, + "grad_norm": 0.925324022769928, + "learning_rate": 1.6641526651320995e-06, + "loss": 0.5641, + "step": 4446 + }, + { + "epoch": 3.681667586968526, + "grad_norm": 0.9984707832336426, + "learning_rate": 1.6631163405687773e-06, + "loss": 0.5811, + "step": 4447 + }, + { + "epoch": 3.6824958586416345, + "grad_norm": 0.9349227547645569, + "learning_rate": 1.662080177934642e-06, + "loss": 0.5696, + "step": 4448 + }, + { + "epoch": 3.6833241303147433, + "grad_norm": 0.9371916651725769, + "learning_rate": 1.6610441774301806e-06, + "loss": 0.5787, + "step": 4449 + }, + { + "epoch": 3.684152401987852, + "grad_norm": 0.9504124522209167, + "learning_rate": 1.6600083392558508e-06, + "loss": 0.5706, + "step": 4450 + }, + { + "epoch": 3.6849806736609607, + "grad_norm": 0.9354562759399414, + "learning_rate": 1.6589726636120768e-06, + "loss": 0.5826, + "step": 4451 + }, + { + "epoch": 3.6858089453340694, + "grad_norm": 0.9202892780303955, + "learning_rate": 1.6579371506992523e-06, + "loss": 0.5807, + "step": 4452 + }, + { + "epoch": 3.6866372170071786, + "grad_norm": 0.9467330574989319, + "learning_rate": 1.65690180071774e-06, + "loss": 0.582, + "step": 4453 + }, + { + "epoch": 3.687465488680287, + "grad_norm": 0.9724621772766113, + "learning_rate": 1.6558666138678698e-06, + "loss": 0.5847, + "step": 4454 + }, + { + "epoch": 3.688293760353396, + "grad_norm": 0.9587774276733398, + "learning_rate": 1.6548315903499407e-06, + "loss": 0.5711, + "step": 4455 + }, + { + "epoch": 3.689122032026505, + "grad_norm": 0.9332905411720276, + "learning_rate": 1.6537967303642205e-06, + "loss": 0.5783, + "step": 4456 + }, + { + "epoch": 3.6899503036996135, + "grad_norm": 0.9627639055252075, + "learning_rate": 1.6527620341109452e-06, + "loss": 0.5741, + "step": 4457 + }, + { + "epoch": 3.6907785753727222, + "grad_norm": 0.9341310262680054, + "learning_rate": 1.6517275017903183e-06, + "loss": 0.5605, + "step": 4458 + }, + { + "epoch": 3.691606847045831, + "grad_norm": 0.9802857637405396, + "learning_rate": 1.6506931336025122e-06, + "loss": 0.5723, + "step": 4459 + }, + { + "epoch": 3.6924351187189397, + "grad_norm": 0.9871952533721924, + "learning_rate": 1.649658929747667e-06, + "loss": 0.5768, + "step": 4460 + }, + { + "epoch": 3.6932633903920484, + "grad_norm": 0.9286062717437744, + "learning_rate": 1.6486248904258929e-06, + "loss": 0.566, + "step": 4461 + }, + { + "epoch": 3.6940916620651576, + "grad_norm": 0.9631725549697876, + "learning_rate": 1.6475910158372654e-06, + "loss": 0.571, + "step": 4462 + }, + { + "epoch": 3.694919933738266, + "grad_norm": 0.9414538741111755, + "learning_rate": 1.6465573061818302e-06, + "loss": 0.5526, + "step": 4463 + }, + { + "epoch": 3.695748205411375, + "grad_norm": 0.9554463028907776, + "learning_rate": 1.6455237616596011e-06, + "loss": 0.5846, + "step": 4464 + }, + { + "epoch": 3.696576477084484, + "grad_norm": 0.9494985938072205, + "learning_rate": 1.6444903824705587e-06, + "loss": 0.5667, + "step": 4465 + }, + { + "epoch": 3.6974047487575925, + "grad_norm": 0.9633604288101196, + "learning_rate": 1.6434571688146522e-06, + "loss": 0.5862, + "step": 4466 + }, + { + "epoch": 3.6982330204307012, + "grad_norm": 0.955493152141571, + "learning_rate": 1.642424120891799e-06, + "loss": 0.5591, + "step": 4467 + }, + { + "epoch": 3.69906129210381, + "grad_norm": 0.9700822830200195, + "learning_rate": 1.641391238901885e-06, + "loss": 0.5733, + "step": 4468 + }, + { + "epoch": 3.6998895637769187, + "grad_norm": 0.9600526094436646, + "learning_rate": 1.6403585230447617e-06, + "loss": 0.5819, + "step": 4469 + }, + { + "epoch": 3.7007178354500274, + "grad_norm": 0.9373101592063904, + "learning_rate": 1.639325973520251e-06, + "loss": 0.5749, + "step": 4470 + }, + { + "epoch": 3.7015461071231366, + "grad_norm": 0.9339377284049988, + "learning_rate": 1.6382935905281406e-06, + "loss": 0.5812, + "step": 4471 + }, + { + "epoch": 3.7023743787962453, + "grad_norm": 0.9409409165382385, + "learning_rate": 1.637261374268188e-06, + "loss": 0.561, + "step": 4472 + }, + { + "epoch": 3.703202650469354, + "grad_norm": 0.9259868264198303, + "learning_rate": 1.6362293249401174e-06, + "loss": 0.5636, + "step": 4473 + }, + { + "epoch": 3.7040309221424628, + "grad_norm": 0.9868617057800293, + "learning_rate": 1.6351974427436198e-06, + "loss": 0.5753, + "step": 4474 + }, + { + "epoch": 3.7048591938155715, + "grad_norm": 1.001633882522583, + "learning_rate": 1.6341657278783563e-06, + "loss": 0.5834, + "step": 4475 + }, + { + "epoch": 3.70568746548868, + "grad_norm": 0.9641216993331909, + "learning_rate": 1.6331341805439527e-06, + "loss": 0.6006, + "step": 4476 + }, + { + "epoch": 3.706515737161789, + "grad_norm": 0.9380208253860474, + "learning_rate": 1.6321028009400041e-06, + "loss": 0.5827, + "step": 4477 + }, + { + "epoch": 3.707344008834898, + "grad_norm": 0.9479442238807678, + "learning_rate": 1.631071589266074e-06, + "loss": 0.5537, + "step": 4478 + }, + { + "epoch": 3.7081722805080064, + "grad_norm": 0.9482294321060181, + "learning_rate": 1.6300405457216912e-06, + "loss": 0.5811, + "step": 4479 + }, + { + "epoch": 3.7090005521811156, + "grad_norm": 0.9395027756690979, + "learning_rate": 1.629009670506353e-06, + "loss": 0.5683, + "step": 4480 + }, + { + "epoch": 3.7098288238542243, + "grad_norm": 0.9259159564971924, + "learning_rate": 1.6279789638195242e-06, + "loss": 0.5659, + "step": 4481 + }, + { + "epoch": 3.710657095527333, + "grad_norm": 0.9183517694473267, + "learning_rate": 1.6269484258606366e-06, + "loss": 0.5657, + "step": 4482 + }, + { + "epoch": 3.7114853672004418, + "grad_norm": 0.9426186680793762, + "learning_rate": 1.6259180568290905e-06, + "loss": 0.5749, + "step": 4483 + }, + { + "epoch": 3.7123136388735505, + "grad_norm": 0.9540229439735413, + "learning_rate": 1.6248878569242522e-06, + "loss": 0.5578, + "step": 4484 + }, + { + "epoch": 3.713141910546659, + "grad_norm": 0.9624539613723755, + "learning_rate": 1.6238578263454558e-06, + "loss": 0.5855, + "step": 4485 + }, + { + "epoch": 3.713970182219768, + "grad_norm": 0.9445024728775024, + "learning_rate": 1.6228279652920026e-06, + "loss": 0.5464, + "step": 4486 + }, + { + "epoch": 3.714798453892877, + "grad_norm": 0.9346494078636169, + "learning_rate": 1.6217982739631617e-06, + "loss": 0.5523, + "step": 4487 + }, + { + "epoch": 3.7156267255659854, + "grad_norm": 1.0044512748718262, + "learning_rate": 1.6207687525581684e-06, + "loss": 0.5586, + "step": 4488 + }, + { + "epoch": 3.7164549972390946, + "grad_norm": 0.9639768600463867, + "learning_rate": 1.6197394012762245e-06, + "loss": 0.5676, + "step": 4489 + }, + { + "epoch": 3.7172832689122033, + "grad_norm": 0.9930342435836792, + "learning_rate": 1.6187102203165011e-06, + "loss": 0.5463, + "step": 4490 + }, + { + "epoch": 3.718111540585312, + "grad_norm": 0.9646494388580322, + "learning_rate": 1.617681209878136e-06, + "loss": 0.5516, + "step": 4491 + }, + { + "epoch": 3.7189398122584207, + "grad_norm": 0.9502953886985779, + "learning_rate": 1.616652370160231e-06, + "loss": 0.5792, + "step": 4492 + }, + { + "epoch": 3.7197680839315295, + "grad_norm": 0.9354153871536255, + "learning_rate": 1.6156237013618573e-06, + "loss": 0.549, + "step": 4493 + }, + { + "epoch": 3.720596355604638, + "grad_norm": 0.9361369609832764, + "learning_rate": 1.6145952036820538e-06, + "loss": 0.566, + "step": 4494 + }, + { + "epoch": 3.721424627277747, + "grad_norm": 0.9847058653831482, + "learning_rate": 1.6135668773198252e-06, + "loss": 0.561, + "step": 4495 + }, + { + "epoch": 3.722252898950856, + "grad_norm": 0.9457981586456299, + "learning_rate": 1.6125387224741422e-06, + "loss": 0.5689, + "step": 4496 + }, + { + "epoch": 3.723081170623965, + "grad_norm": 0.9350807666778564, + "learning_rate": 1.6115107393439433e-06, + "loss": 0.5603, + "step": 4497 + }, + { + "epoch": 3.7239094422970735, + "grad_norm": 0.9243747591972351, + "learning_rate": 1.6104829281281343e-06, + "loss": 0.578, + "step": 4498 + }, + { + "epoch": 3.7247377139701823, + "grad_norm": 0.9140326976776123, + "learning_rate": 1.6094552890255869e-06, + "loss": 0.5831, + "step": 4499 + }, + { + "epoch": 3.725565985643291, + "grad_norm": 0.9408622980117798, + "learning_rate": 1.6084278222351391e-06, + "loss": 0.5684, + "step": 4500 + }, + { + "epoch": 3.7263942573163997, + "grad_norm": 0.9529958367347717, + "learning_rate": 1.6074005279555965e-06, + "loss": 0.5876, + "step": 4501 + }, + { + "epoch": 3.7272225289895085, + "grad_norm": 0.9634376764297485, + "learning_rate": 1.6063734063857323e-06, + "loss": 0.5809, + "step": 4502 + }, + { + "epoch": 3.728050800662617, + "grad_norm": 0.9465759992599487, + "learning_rate": 1.6053464577242828e-06, + "loss": 0.564, + "step": 4503 + }, + { + "epoch": 3.728879072335726, + "grad_norm": 0.943699836730957, + "learning_rate": 1.6043196821699536e-06, + "loss": 0.5797, + "step": 4504 + }, + { + "epoch": 3.729707344008835, + "grad_norm": 0.9360603094100952, + "learning_rate": 1.6032930799214164e-06, + "loss": 0.561, + "step": 4505 + }, + { + "epoch": 3.730535615681944, + "grad_norm": 0.9656838178634644, + "learning_rate": 1.6022666511773098e-06, + "loss": 0.5726, + "step": 4506 + }, + { + "epoch": 3.7313638873550525, + "grad_norm": 0.9636357426643372, + "learning_rate": 1.6012403961362372e-06, + "loss": 0.5859, + "step": 4507 + }, + { + "epoch": 3.7321921590281613, + "grad_norm": 0.9537240862846375, + "learning_rate": 1.6002143149967697e-06, + "loss": 0.5798, + "step": 4508 + }, + { + "epoch": 3.73302043070127, + "grad_norm": 0.9626145362854004, + "learning_rate": 1.5991884079574448e-06, + "loss": 0.5778, + "step": 4509 + }, + { + "epoch": 3.7338487023743787, + "grad_norm": 0.9493876099586487, + "learning_rate": 1.598162675216766e-06, + "loss": 0.5688, + "step": 4510 + }, + { + "epoch": 3.7346769740474874, + "grad_norm": 0.9417327046394348, + "learning_rate": 1.597137116973202e-06, + "loss": 0.5628, + "step": 4511 + }, + { + "epoch": 3.7355052457205966, + "grad_norm": 0.9836271405220032, + "learning_rate": 1.59611173342519e-06, + "loss": 0.5616, + "step": 4512 + }, + { + "epoch": 3.736333517393705, + "grad_norm": 0.9088386297225952, + "learning_rate": 1.5950865247711322e-06, + "loss": 0.5635, + "step": 4513 + }, + { + "epoch": 3.737161789066814, + "grad_norm": 0.9311172366142273, + "learning_rate": 1.5940614912093957e-06, + "loss": 0.5795, + "step": 4514 + }, + { + "epoch": 3.737990060739923, + "grad_norm": 0.9570069909095764, + "learning_rate": 1.5930366329383157e-06, + "loss": 0.5753, + "step": 4515 + }, + { + "epoch": 3.7388183324130315, + "grad_norm": 1.1909716129302979, + "learning_rate": 1.592011950156192e-06, + "loss": 0.5637, + "step": 4516 + }, + { + "epoch": 3.7396466040861402, + "grad_norm": 1.0059658288955688, + "learning_rate": 1.5909874430612925e-06, + "loss": 0.5894, + "step": 4517 + }, + { + "epoch": 3.740474875759249, + "grad_norm": 0.9572764039039612, + "learning_rate": 1.5899631118518491e-06, + "loss": 0.5762, + "step": 4518 + }, + { + "epoch": 3.7413031474323577, + "grad_norm": 0.9477603435516357, + "learning_rate": 1.58893895672606e-06, + "loss": 0.5602, + "step": 4519 + }, + { + "epoch": 3.7421314191054664, + "grad_norm": 0.9420804381370544, + "learning_rate": 1.5879149778820898e-06, + "loss": 0.5717, + "step": 4520 + }, + { + "epoch": 3.7429596907785756, + "grad_norm": 0.9785239100456238, + "learning_rate": 1.5868911755180694e-06, + "loss": 0.571, + "step": 4521 + }, + { + "epoch": 3.743787962451684, + "grad_norm": 0.9285955429077148, + "learning_rate": 1.585867549832094e-06, + "loss": 0.5603, + "step": 4522 + }, + { + "epoch": 3.744616234124793, + "grad_norm": 0.971987247467041, + "learning_rate": 1.584844101022227e-06, + "loss": 0.5829, + "step": 4523 + }, + { + "epoch": 3.745444505797902, + "grad_norm": 0.9516214728355408, + "learning_rate": 1.5838208292864954e-06, + "loss": 0.583, + "step": 4524 + }, + { + "epoch": 3.7462727774710105, + "grad_norm": 0.938643753528595, + "learning_rate": 1.5827977348228935e-06, + "loss": 0.5427, + "step": 4525 + }, + { + "epoch": 3.7471010491441192, + "grad_norm": 0.9617331027984619, + "learning_rate": 1.581774817829379e-06, + "loss": 0.5663, + "step": 4526 + }, + { + "epoch": 3.747929320817228, + "grad_norm": 0.9507145881652832, + "learning_rate": 1.5807520785038776e-06, + "loss": 0.5855, + "step": 4527 + }, + { + "epoch": 3.7487575924903367, + "grad_norm": 0.955471932888031, + "learning_rate": 1.57972951704428e-06, + "loss": 0.5676, + "step": 4528 + }, + { + "epoch": 3.7495858641634454, + "grad_norm": 0.9277394413948059, + "learning_rate": 1.5787071336484422e-06, + "loss": 0.5452, + "step": 4529 + }, + { + "epoch": 3.7504141358365546, + "grad_norm": 0.9697600603103638, + "learning_rate": 1.5776849285141858e-06, + "loss": 0.5598, + "step": 4530 + }, + { + "epoch": 3.7512424075096633, + "grad_norm": 0.9462587833404541, + "learning_rate": 1.5766629018392981e-06, + "loss": 0.5766, + "step": 4531 + }, + { + "epoch": 3.752070679182772, + "grad_norm": 0.9286374449729919, + "learning_rate": 1.5756410538215317e-06, + "loss": 0.5747, + "step": 4532 + }, + { + "epoch": 3.7528989508558808, + "grad_norm": 0.9680078625679016, + "learning_rate": 1.5746193846586045e-06, + "loss": 0.5789, + "step": 4533 + }, + { + "epoch": 3.7537272225289895, + "grad_norm": 0.9353684782981873, + "learning_rate": 1.5735978945482e-06, + "loss": 0.5763, + "step": 4534 + }, + { + "epoch": 3.754555494202098, + "grad_norm": 0.9365376830101013, + "learning_rate": 1.5725765836879674e-06, + "loss": 0.5504, + "step": 4535 + }, + { + "epoch": 3.755383765875207, + "grad_norm": 0.9342913627624512, + "learning_rate": 1.571555452275521e-06, + "loss": 0.558, + "step": 4536 + }, + { + "epoch": 3.756212037548316, + "grad_norm": 0.9252297878265381, + "learning_rate": 1.5705345005084394e-06, + "loss": 0.5666, + "step": 4537 + }, + { + "epoch": 3.7570403092214244, + "grad_norm": 1.0099254846572876, + "learning_rate": 1.5695137285842666e-06, + "loss": 0.5627, + "step": 4538 + }, + { + "epoch": 3.7578685808945336, + "grad_norm": 0.9464092254638672, + "learning_rate": 1.5684931367005144e-06, + "loss": 0.5642, + "step": 4539 + }, + { + "epoch": 3.7586968525676423, + "grad_norm": 0.93072509765625, + "learning_rate": 1.5674727250546568e-06, + "loss": 0.5821, + "step": 4540 + }, + { + "epoch": 3.759525124240751, + "grad_norm": 0.9577572345733643, + "learning_rate": 1.5664524938441334e-06, + "loss": 0.56, + "step": 4541 + }, + { + "epoch": 3.7603533959138598, + "grad_norm": 0.9372087121009827, + "learning_rate": 1.5654324432663503e-06, + "loss": 0.5583, + "step": 4542 + }, + { + "epoch": 3.7611816675869685, + "grad_norm": 0.9543371796607971, + "learning_rate": 1.5644125735186776e-06, + "loss": 0.5611, + "step": 4543 + }, + { + "epoch": 3.762009939260077, + "grad_norm": 0.9668126702308655, + "learning_rate": 1.5633928847984503e-06, + "loss": 0.5637, + "step": 4544 + }, + { + "epoch": 3.762838210933186, + "grad_norm": 0.9686347246170044, + "learning_rate": 1.5623733773029692e-06, + "loss": 0.5737, + "step": 4545 + }, + { + "epoch": 3.763666482606295, + "grad_norm": 0.9634156227111816, + "learning_rate": 1.561354051229499e-06, + "loss": 0.5602, + "step": 4546 + }, + { + "epoch": 3.7644947542794034, + "grad_norm": 0.9353019595146179, + "learning_rate": 1.5603349067752705e-06, + "loss": 0.5518, + "step": 4547 + }, + { + "epoch": 3.7653230259525126, + "grad_norm": 0.9436436295509338, + "learning_rate": 1.5593159441374778e-06, + "loss": 0.5652, + "step": 4548 + }, + { + "epoch": 3.7661512976256213, + "grad_norm": 1.0265226364135742, + "learning_rate": 1.55829716351328e-06, + "loss": 0.5624, + "step": 4549 + }, + { + "epoch": 3.76697956929873, + "grad_norm": 0.9422723650932312, + "learning_rate": 1.5572785650998035e-06, + "loss": 0.5649, + "step": 4550 + }, + { + "epoch": 3.7678078409718387, + "grad_norm": 0.9558239579200745, + "learning_rate": 1.5562601490941365e-06, + "loss": 0.5553, + "step": 4551 + }, + { + "epoch": 3.7686361126449475, + "grad_norm": 0.9281789660453796, + "learning_rate": 1.555241915693333e-06, + "loss": 0.5582, + "step": 4552 + }, + { + "epoch": 3.769464384318056, + "grad_norm": 0.9265138506889343, + "learning_rate": 1.5542238650944121e-06, + "loss": 0.5803, + "step": 4553 + }, + { + "epoch": 3.770292655991165, + "grad_norm": 0.9384198784828186, + "learning_rate": 1.5532059974943569e-06, + "loss": 0.5821, + "step": 4554 + }, + { + "epoch": 3.771120927664274, + "grad_norm": 0.9323135018348694, + "learning_rate": 1.552188313090115e-06, + "loss": 0.56, + "step": 4555 + }, + { + "epoch": 3.771949199337383, + "grad_norm": 0.9448620080947876, + "learning_rate": 1.5511708120785994e-06, + "loss": 0.5521, + "step": 4556 + }, + { + "epoch": 3.7727774710104915, + "grad_norm": 0.9806753993034363, + "learning_rate": 1.5501534946566871e-06, + "loss": 0.5842, + "step": 4557 + }, + { + "epoch": 3.7736057426836003, + "grad_norm": 0.9242349863052368, + "learning_rate": 1.549136361021219e-06, + "loss": 0.5764, + "step": 4558 + }, + { + "epoch": 3.774434014356709, + "grad_norm": 0.956167459487915, + "learning_rate": 1.5481194113690017e-06, + "loss": 0.5891, + "step": 4559 + }, + { + "epoch": 3.7752622860298177, + "grad_norm": 0.9456217288970947, + "learning_rate": 1.5471026458968041e-06, + "loss": 0.5821, + "step": 4560 + }, + { + "epoch": 3.7760905577029265, + "grad_norm": 0.9384570717811584, + "learning_rate": 1.5460860648013626e-06, + "loss": 0.5662, + "step": 4561 + }, + { + "epoch": 3.776918829376035, + "grad_norm": 0.9760017395019531, + "learning_rate": 1.5450696682793747e-06, + "loss": 0.5862, + "step": 4562 + }, + { + "epoch": 3.777747101049144, + "grad_norm": 0.9794814586639404, + "learning_rate": 1.5440534565275039e-06, + "loss": 0.5852, + "step": 4563 + }, + { + "epoch": 3.778575372722253, + "grad_norm": 0.9752492904663086, + "learning_rate": 1.5430374297423784e-06, + "loss": 0.5909, + "step": 4564 + }, + { + "epoch": 3.779403644395362, + "grad_norm": 0.969682514667511, + "learning_rate": 1.5420215881205897e-06, + "loss": 0.5505, + "step": 4565 + }, + { + "epoch": 3.7802319160684705, + "grad_norm": 0.9354566335678101, + "learning_rate": 1.5410059318586934e-06, + "loss": 0.5643, + "step": 4566 + }, + { + "epoch": 3.7810601877415793, + "grad_norm": 0.9543263912200928, + "learning_rate": 1.5399904611532095e-06, + "loss": 0.579, + "step": 4567 + }, + { + "epoch": 3.781888459414688, + "grad_norm": 0.9573609232902527, + "learning_rate": 1.5389751762006225e-06, + "loss": 0.5815, + "step": 4568 + }, + { + "epoch": 3.7827167310877967, + "grad_norm": 0.9919794201850891, + "learning_rate": 1.5379600771973805e-06, + "loss": 0.5643, + "step": 4569 + }, + { + "epoch": 3.7835450027609054, + "grad_norm": 0.966391921043396, + "learning_rate": 1.5369451643398959e-06, + "loss": 0.5702, + "step": 4570 + }, + { + "epoch": 3.7843732744340146, + "grad_norm": 0.9727235436439514, + "learning_rate": 1.5359304378245435e-06, + "loss": 0.5607, + "step": 4571 + }, + { + "epoch": 3.785201546107123, + "grad_norm": 0.9183128476142883, + "learning_rate": 1.5349158978476653e-06, + "loss": 0.539, + "step": 4572 + }, + { + "epoch": 3.786029817780232, + "grad_norm": 0.980755090713501, + "learning_rate": 1.533901544605564e-06, + "loss": 0.5761, + "step": 4573 + }, + { + "epoch": 3.786858089453341, + "grad_norm": 0.96429842710495, + "learning_rate": 1.532887378294508e-06, + "loss": 0.5797, + "step": 4574 + }, + { + "epoch": 3.7876863611264495, + "grad_norm": 0.9631879329681396, + "learning_rate": 1.5318733991107287e-06, + "loss": 0.5851, + "step": 4575 + }, + { + "epoch": 3.7885146327995582, + "grad_norm": 0.9594787359237671, + "learning_rate": 1.5308596072504223e-06, + "loss": 0.5656, + "step": 4576 + }, + { + "epoch": 3.789342904472667, + "grad_norm": 0.9394322633743286, + "learning_rate": 1.5298460029097479e-06, + "loss": 0.5554, + "step": 4577 + }, + { + "epoch": 3.7901711761457757, + "grad_norm": 0.9391270875930786, + "learning_rate": 1.5288325862848275e-06, + "loss": 0.5733, + "step": 4578 + }, + { + "epoch": 3.7909994478188844, + "grad_norm": 0.9480547904968262, + "learning_rate": 1.5278193575717492e-06, + "loss": 0.5629, + "step": 4579 + }, + { + "epoch": 3.7918277194919936, + "grad_norm": 0.9235183000564575, + "learning_rate": 1.5268063169665626e-06, + "loss": 0.5626, + "step": 4580 + }, + { + "epoch": 3.792655991165102, + "grad_norm": 0.9616346955299377, + "learning_rate": 1.5257934646652822e-06, + "loss": 0.557, + "step": 4581 + }, + { + "epoch": 3.793484262838211, + "grad_norm": 0.9767917394638062, + "learning_rate": 1.5247808008638843e-06, + "loss": 0.5693, + "step": 4582 + }, + { + "epoch": 3.79431253451132, + "grad_norm": 0.9262551665306091, + "learning_rate": 1.5237683257583102e-06, + "loss": 0.5638, + "step": 4583 + }, + { + "epoch": 3.7951408061844285, + "grad_norm": 0.9267538189888, + "learning_rate": 1.522756039544465e-06, + "loss": 0.5801, + "step": 4584 + }, + { + "epoch": 3.7959690778575372, + "grad_norm": 0.9917179346084595, + "learning_rate": 1.5217439424182167e-06, + "loss": 0.5715, + "step": 4585 + }, + { + "epoch": 3.796797349530646, + "grad_norm": 0.9482493996620178, + "learning_rate": 1.5207320345753957e-06, + "loss": 0.5618, + "step": 4586 + }, + { + "epoch": 3.7976256212037547, + "grad_norm": 0.9342687726020813, + "learning_rate": 1.5197203162117979e-06, + "loss": 0.5863, + "step": 4587 + }, + { + "epoch": 3.7984538928768634, + "grad_norm": 0.9577085375785828, + "learning_rate": 1.5187087875231809e-06, + "loss": 0.5501, + "step": 4588 + }, + { + "epoch": 3.7992821645499726, + "grad_norm": 0.9561404585838318, + "learning_rate": 1.5176974487052658e-06, + "loss": 0.576, + "step": 4589 + }, + { + "epoch": 3.8001104362230813, + "grad_norm": 0.9302974343299866, + "learning_rate": 1.5166862999537377e-06, + "loss": 0.5551, + "step": 4590 + }, + { + "epoch": 3.80093870789619, + "grad_norm": 0.9463371634483337, + "learning_rate": 1.5156753414642445e-06, + "loss": 0.5531, + "step": 4591 + }, + { + "epoch": 3.8017669795692988, + "grad_norm": 0.9446096420288086, + "learning_rate": 1.5146645734323961e-06, + "loss": 0.5652, + "step": 4592 + }, + { + "epoch": 3.8025952512424075, + "grad_norm": 0.9582951068878174, + "learning_rate": 1.513653996053769e-06, + "loss": 0.5709, + "step": 4593 + }, + { + "epoch": 3.803423522915516, + "grad_norm": 0.961769163608551, + "learning_rate": 1.5126436095238978e-06, + "loss": 0.576, + "step": 4594 + }, + { + "epoch": 3.804251794588625, + "grad_norm": 0.97950279712677, + "learning_rate": 1.5116334140382846e-06, + "loss": 0.562, + "step": 4595 + }, + { + "epoch": 3.805080066261734, + "grad_norm": 0.9386860132217407, + "learning_rate": 1.5106234097923922e-06, + "loss": 0.5723, + "step": 4596 + }, + { + "epoch": 3.8059083379348424, + "grad_norm": 0.9200562238693237, + "learning_rate": 1.509613596981647e-06, + "loss": 0.5708, + "step": 4597 + }, + { + "epoch": 3.8067366096079516, + "grad_norm": 0.9222211837768555, + "learning_rate": 1.5086039758014386e-06, + "loss": 0.5711, + "step": 4598 + }, + { + "epoch": 3.8075648812810603, + "grad_norm": 0.9545297026634216, + "learning_rate": 1.5075945464471198e-06, + "loss": 0.5628, + "step": 4599 + }, + { + "epoch": 3.808393152954169, + "grad_norm": 0.918067455291748, + "learning_rate": 1.5065853091140041e-06, + "loss": 0.5429, + "step": 4600 + }, + { + "epoch": 3.8092214246272778, + "grad_norm": 0.9256977438926697, + "learning_rate": 1.5055762639973714e-06, + "loss": 0.5593, + "step": 4601 + }, + { + "epoch": 3.8100496963003865, + "grad_norm": 0.9570629000663757, + "learning_rate": 1.5045674112924612e-06, + "loss": 0.5545, + "step": 4602 + }, + { + "epoch": 3.810877967973495, + "grad_norm": 0.9828989505767822, + "learning_rate": 1.5035587511944778e-06, + "loss": 0.5776, + "step": 4603 + }, + { + "epoch": 3.811706239646604, + "grad_norm": 0.9521099925041199, + "learning_rate": 1.5025502838985884e-06, + "loss": 0.5645, + "step": 4604 + }, + { + "epoch": 3.812534511319713, + "grad_norm": 0.9580879211425781, + "learning_rate": 1.5015420095999197e-06, + "loss": 0.5754, + "step": 4605 + }, + { + "epoch": 3.8133627829928214, + "grad_norm": 0.9599860906600952, + "learning_rate": 1.500533928493565e-06, + "loss": 0.5623, + "step": 4606 + }, + { + "epoch": 3.8141910546659306, + "grad_norm": 0.9527266621589661, + "learning_rate": 1.4995260407745782e-06, + "loss": 0.5931, + "step": 4607 + }, + { + "epoch": 3.8150193263390393, + "grad_norm": 0.9410575032234192, + "learning_rate": 1.498518346637976e-06, + "loss": 0.569, + "step": 4608 + }, + { + "epoch": 3.815847598012148, + "grad_norm": 0.9750106930732727, + "learning_rate": 1.4975108462787382e-06, + "loss": 0.5838, + "step": 4609 + }, + { + "epoch": 3.8166758696852567, + "grad_norm": 0.9242303967475891, + "learning_rate": 1.4965035398918065e-06, + "loss": 0.5595, + "step": 4610 + }, + { + "epoch": 3.8175041413583655, + "grad_norm": 0.9424212574958801, + "learning_rate": 1.495496427672085e-06, + "loss": 0.5949, + "step": 4611 + }, + { + "epoch": 3.818332413031474, + "grad_norm": 0.9485701322555542, + "learning_rate": 1.4944895098144419e-06, + "loss": 0.5799, + "step": 4612 + }, + { + "epoch": 3.819160684704583, + "grad_norm": 0.9476267695426941, + "learning_rate": 1.4934827865137047e-06, + "loss": 0.5584, + "step": 4613 + }, + { + "epoch": 3.819988956377692, + "grad_norm": 0.9722000360488892, + "learning_rate": 1.4924762579646653e-06, + "loss": 0.5665, + "step": 4614 + }, + { + "epoch": 3.820817228050801, + "grad_norm": 0.9728968739509583, + "learning_rate": 1.4914699243620795e-06, + "loss": 0.5786, + "step": 4615 + }, + { + "epoch": 3.8216454997239095, + "grad_norm": 0.96755051612854, + "learning_rate": 1.4904637859006605e-06, + "loss": 0.5529, + "step": 4616 + }, + { + "epoch": 3.8224737713970183, + "grad_norm": 0.9613732695579529, + "learning_rate": 1.4894578427750888e-06, + "loss": 0.5847, + "step": 4617 + }, + { + "epoch": 3.823302043070127, + "grad_norm": 0.9679767489433289, + "learning_rate": 1.488452095180004e-06, + "loss": 0.5623, + "step": 4618 + }, + { + "epoch": 3.8241303147432357, + "grad_norm": 0.9799363613128662, + "learning_rate": 1.4874465433100093e-06, + "loss": 0.5712, + "step": 4619 + }, + { + "epoch": 3.8249585864163445, + "grad_norm": 0.962107241153717, + "learning_rate": 1.4864411873596695e-06, + "loss": 0.5596, + "step": 4620 + }, + { + "epoch": 3.825786858089453, + "grad_norm": 0.9862436056137085, + "learning_rate": 1.4854360275235118e-06, + "loss": 0.5801, + "step": 4621 + }, + { + "epoch": 3.826615129762562, + "grad_norm": 0.9500226974487305, + "learning_rate": 1.484431063996025e-06, + "loss": 0.5467, + "step": 4622 + }, + { + "epoch": 3.827443401435671, + "grad_norm": 0.9733493328094482, + "learning_rate": 1.4834262969716607e-06, + "loss": 0.549, + "step": 4623 + }, + { + "epoch": 3.82827167310878, + "grad_norm": 0.9474700093269348, + "learning_rate": 1.4824217266448312e-06, + "loss": 0.5497, + "step": 4624 + }, + { + "epoch": 3.8290999447818885, + "grad_norm": 0.9376130104064941, + "learning_rate": 1.4814173532099116e-06, + "loss": 0.5681, + "step": 4625 + }, + { + "epoch": 3.8299282164549973, + "grad_norm": 0.9183100461959839, + "learning_rate": 1.4804131768612406e-06, + "loss": 0.5719, + "step": 4626 + }, + { + "epoch": 3.830756488128106, + "grad_norm": 0.9759798645973206, + "learning_rate": 1.4794091977931136e-06, + "loss": 0.5773, + "step": 4627 + }, + { + "epoch": 3.8315847598012147, + "grad_norm": 0.9638318419456482, + "learning_rate": 1.4784054161997937e-06, + "loss": 0.5681, + "step": 4628 + }, + { + "epoch": 3.8324130314743234, + "grad_norm": 0.9652249217033386, + "learning_rate": 1.4774018322755027e-06, + "loss": 0.5866, + "step": 4629 + }, + { + "epoch": 3.8332413031474326, + "grad_norm": 0.9402887225151062, + "learning_rate": 1.4763984462144237e-06, + "loss": 0.569, + "step": 4630 + }, + { + "epoch": 3.834069574820541, + "grad_norm": 0.925959587097168, + "learning_rate": 1.4753952582107046e-06, + "loss": 0.5945, + "step": 4631 + }, + { + "epoch": 3.83489784649365, + "grad_norm": 0.9718639254570007, + "learning_rate": 1.4743922684584514e-06, + "loss": 0.5523, + "step": 4632 + }, + { + "epoch": 3.835726118166759, + "grad_norm": 0.9743525981903076, + "learning_rate": 1.4733894771517337e-06, + "loss": 0.58, + "step": 4633 + }, + { + "epoch": 3.8365543898398675, + "grad_norm": 0.9634748697280884, + "learning_rate": 1.4723868844845818e-06, + "loss": 0.5715, + "step": 4634 + }, + { + "epoch": 3.8373826615129762, + "grad_norm": 0.9775907397270203, + "learning_rate": 1.471384490650989e-06, + "loss": 0.5697, + "step": 4635 + }, + { + "epoch": 3.838210933186085, + "grad_norm": 0.9504139423370361, + "learning_rate": 1.4703822958449081e-06, + "loss": 0.5797, + "step": 4636 + }, + { + "epoch": 3.8390392048591937, + "grad_norm": 0.9717209339141846, + "learning_rate": 1.4693803002602553e-06, + "loss": 0.5692, + "step": 4637 + }, + { + "epoch": 3.8398674765323024, + "grad_norm": 0.9564605951309204, + "learning_rate": 1.4683785040909078e-06, + "loss": 0.5719, + "step": 4638 + }, + { + "epoch": 3.8406957482054116, + "grad_norm": 0.951598584651947, + "learning_rate": 1.4673769075307023e-06, + "loss": 0.5481, + "step": 4639 + }, + { + "epoch": 3.84152401987852, + "grad_norm": 0.9717726111412048, + "learning_rate": 1.4663755107734395e-06, + "loss": 0.537, + "step": 4640 + }, + { + "epoch": 3.842352291551629, + "grad_norm": 0.9653568267822266, + "learning_rate": 1.4653743140128802e-06, + "loss": 0.5729, + "step": 4641 + }, + { + "epoch": 3.843180563224738, + "grad_norm": 0.9710245728492737, + "learning_rate": 1.4643733174427454e-06, + "loss": 0.5816, + "step": 4642 + }, + { + "epoch": 3.8440088348978465, + "grad_norm": 0.9537909626960754, + "learning_rate": 1.4633725212567206e-06, + "loss": 0.5585, + "step": 4643 + }, + { + "epoch": 3.8448371065709552, + "grad_norm": 0.9765188694000244, + "learning_rate": 1.4623719256484495e-06, + "loss": 0.5542, + "step": 4644 + }, + { + "epoch": 3.845665378244064, + "grad_norm": 0.9446712136268616, + "learning_rate": 1.4613715308115378e-06, + "loss": 0.5696, + "step": 4645 + }, + { + "epoch": 3.8464936499171727, + "grad_norm": 0.9561092853546143, + "learning_rate": 1.4603713369395533e-06, + "loss": 0.555, + "step": 4646 + }, + { + "epoch": 3.8473219215902814, + "grad_norm": 0.9431309700012207, + "learning_rate": 1.4593713442260233e-06, + "loss": 0.5589, + "step": 4647 + }, + { + "epoch": 3.8481501932633906, + "grad_norm": 0.9409856796264648, + "learning_rate": 1.4583715528644384e-06, + "loss": 0.5574, + "step": 4648 + }, + { + "epoch": 3.8489784649364993, + "grad_norm": 0.9805465936660767, + "learning_rate": 1.4573719630482485e-06, + "loss": 0.5555, + "step": 4649 + }, + { + "epoch": 3.849806736609608, + "grad_norm": 0.9668418765068054, + "learning_rate": 1.456372574970863e-06, + "loss": 0.5721, + "step": 4650 + }, + { + "epoch": 3.8506350082827168, + "grad_norm": 0.937393844127655, + "learning_rate": 1.455373388825656e-06, + "loss": 0.5619, + "step": 4651 + }, + { + "epoch": 3.8514632799558255, + "grad_norm": 0.956749439239502, + "learning_rate": 1.4543744048059613e-06, + "loss": 0.5882, + "step": 4652 + }, + { + "epoch": 3.852291551628934, + "grad_norm": 0.95976722240448, + "learning_rate": 1.4533756231050712e-06, + "loss": 0.5573, + "step": 4653 + }, + { + "epoch": 3.853119823302043, + "grad_norm": 0.9402530789375305, + "learning_rate": 1.4523770439162414e-06, + "loss": 0.549, + "step": 4654 + }, + { + "epoch": 3.853948094975152, + "grad_norm": 0.960904061794281, + "learning_rate": 1.4513786674326888e-06, + "loss": 0.5671, + "step": 4655 + }, + { + "epoch": 3.8547763666482604, + "grad_norm": 0.929169774055481, + "learning_rate": 1.450380493847588e-06, + "loss": 0.5482, + "step": 4656 + }, + { + "epoch": 3.8556046383213696, + "grad_norm": 0.9257175922393799, + "learning_rate": 1.449382523354077e-06, + "loss": 0.5526, + "step": 4657 + }, + { + "epoch": 3.8564329099944783, + "grad_norm": 0.9734342098236084, + "learning_rate": 1.4483847561452553e-06, + "loss": 0.5601, + "step": 4658 + }, + { + "epoch": 3.857261181667587, + "grad_norm": 0.9698025584220886, + "learning_rate": 1.4473871924141788e-06, + "loss": 0.568, + "step": 4659 + }, + { + "epoch": 3.8580894533406958, + "grad_norm": 0.9453846216201782, + "learning_rate": 1.4463898323538694e-06, + "loss": 0.5618, + "step": 4660 + }, + { + "epoch": 3.8589177250138045, + "grad_norm": 0.9435855150222778, + "learning_rate": 1.4453926761573062e-06, + "loss": 0.559, + "step": 4661 + }, + { + "epoch": 3.859745996686913, + "grad_norm": 0.9480921030044556, + "learning_rate": 1.4443957240174277e-06, + "loss": 0.5593, + "step": 4662 + }, + { + "epoch": 3.860574268360022, + "grad_norm": 0.9516395926475525, + "learning_rate": 1.443398976127137e-06, + "loss": 0.552, + "step": 4663 + }, + { + "epoch": 3.861402540033131, + "grad_norm": 0.9439289569854736, + "learning_rate": 1.4424024326792955e-06, + "loss": 0.5564, + "step": 4664 + }, + { + "epoch": 3.8622308117062394, + "grad_norm": 0.9423355460166931, + "learning_rate": 1.4414060938667236e-06, + "loss": 0.5428, + "step": 4665 + }, + { + "epoch": 3.8630590833793486, + "grad_norm": 0.9807745814323425, + "learning_rate": 1.4404099598822041e-06, + "loss": 0.5695, + "step": 4666 + }, + { + "epoch": 3.8638873550524573, + "grad_norm": 0.9256470203399658, + "learning_rate": 1.4394140309184815e-06, + "loss": 0.5766, + "step": 4667 + }, + { + "epoch": 3.864715626725566, + "grad_norm": 0.9468652009963989, + "learning_rate": 1.4384183071682562e-06, + "loss": 0.5762, + "step": 4668 + }, + { + "epoch": 3.8655438983986747, + "grad_norm": 0.9390266537666321, + "learning_rate": 1.4374227888241921e-06, + "loss": 0.5779, + "step": 4669 + }, + { + "epoch": 3.8663721700717835, + "grad_norm": 0.9410127997398376, + "learning_rate": 1.4364274760789143e-06, + "loss": 0.5448, + "step": 4670 + }, + { + "epoch": 3.867200441744892, + "grad_norm": 0.9276284575462341, + "learning_rate": 1.4354323691250048e-06, + "loss": 0.5764, + "step": 4671 + }, + { + "epoch": 3.868028713418001, + "grad_norm": 0.9957787394523621, + "learning_rate": 1.4344374681550089e-06, + "loss": 0.5819, + "step": 4672 + }, + { + "epoch": 3.86885698509111, + "grad_norm": 0.9688808917999268, + "learning_rate": 1.4334427733614287e-06, + "loss": 0.5565, + "step": 4673 + }, + { + "epoch": 3.869685256764219, + "grad_norm": 0.9403797388076782, + "learning_rate": 1.4324482849367304e-06, + "loss": 0.5538, + "step": 4674 + }, + { + "epoch": 3.8705135284373275, + "grad_norm": 0.9486374855041504, + "learning_rate": 1.4314540030733364e-06, + "loss": 0.5529, + "step": 4675 + }, + { + "epoch": 3.8713418001104363, + "grad_norm": 0.9329063296318054, + "learning_rate": 1.4304599279636333e-06, + "loss": 0.5656, + "step": 4676 + }, + { + "epoch": 3.872170071783545, + "grad_norm": 0.9217836856842041, + "learning_rate": 1.4294660597999627e-06, + "loss": 0.5557, + "step": 4677 + }, + { + "epoch": 3.8729983434566537, + "grad_norm": 0.926406741142273, + "learning_rate": 1.4284723987746305e-06, + "loss": 0.5532, + "step": 4678 + }, + { + "epoch": 3.8738266151297625, + "grad_norm": 0.9605270624160767, + "learning_rate": 1.427478945079901e-06, + "loss": 0.5652, + "step": 4679 + }, + { + "epoch": 3.874654886802871, + "grad_norm": 0.9510946273803711, + "learning_rate": 1.4264856989079973e-06, + "loss": 0.5618, + "step": 4680 + }, + { + "epoch": 3.87548315847598, + "grad_norm": 0.9574974775314331, + "learning_rate": 1.4254926604511033e-06, + "loss": 0.5822, + "step": 4681 + }, + { + "epoch": 3.876311430149089, + "grad_norm": 0.9598250985145569, + "learning_rate": 1.424499829901364e-06, + "loss": 0.5537, + "step": 4682 + }, + { + "epoch": 3.877139701822198, + "grad_norm": 0.9691941738128662, + "learning_rate": 1.4235072074508822e-06, + "loss": 0.568, + "step": 4683 + }, + { + "epoch": 3.8779679734953065, + "grad_norm": 0.9728984236717224, + "learning_rate": 1.4225147932917193e-06, + "loss": 0.567, + "step": 4684 + }, + { + "epoch": 3.8787962451684153, + "grad_norm": 0.921758234500885, + "learning_rate": 1.4215225876159003e-06, + "loss": 0.5587, + "step": 4685 + }, + { + "epoch": 3.879624516841524, + "grad_norm": 0.9491950869560242, + "learning_rate": 1.420530590615408e-06, + "loss": 0.5479, + "step": 4686 + }, + { + "epoch": 3.8804527885146327, + "grad_norm": 0.9212395548820496, + "learning_rate": 1.4195388024821827e-06, + "loss": 0.5623, + "step": 4687 + }, + { + "epoch": 3.8812810601877414, + "grad_norm": 0.970206081867218, + "learning_rate": 1.418547223408127e-06, + "loss": 0.571, + "step": 4688 + }, + { + "epoch": 3.8821093318608506, + "grad_norm": 0.9378070831298828, + "learning_rate": 1.4175558535851037e-06, + "loss": 0.5823, + "step": 4689 + }, + { + "epoch": 3.882937603533959, + "grad_norm": 0.9203671813011169, + "learning_rate": 1.416564693204931e-06, + "loss": 0.5705, + "step": 4690 + }, + { + "epoch": 3.883765875207068, + "grad_norm": 0.9873926043510437, + "learning_rate": 1.4155737424593907e-06, + "loss": 0.5612, + "step": 4691 + }, + { + "epoch": 3.884594146880177, + "grad_norm": 0.9374533295631409, + "learning_rate": 1.4145830015402235e-06, + "loss": 0.5808, + "step": 4692 + }, + { + "epoch": 3.8854224185532855, + "grad_norm": 0.9469799399375916, + "learning_rate": 1.413592470639126e-06, + "loss": 0.5673, + "step": 4693 + }, + { + "epoch": 3.8862506902263942, + "grad_norm": 0.9671788811683655, + "learning_rate": 1.4126021499477592e-06, + "loss": 0.5755, + "step": 4694 + }, + { + "epoch": 3.887078961899503, + "grad_norm": 0.9435954689979553, + "learning_rate": 1.4116120396577398e-06, + "loss": 0.5686, + "step": 4695 + }, + { + "epoch": 3.8879072335726117, + "grad_norm": 0.9866729378700256, + "learning_rate": 1.410622139960644e-06, + "loss": 0.5456, + "step": 4696 + }, + { + "epoch": 3.8887355052457204, + "grad_norm": 0.9504481554031372, + "learning_rate": 1.409632451048009e-06, + "loss": 0.5779, + "step": 4697 + }, + { + "epoch": 3.8895637769188296, + "grad_norm": 0.9329410195350647, + "learning_rate": 1.4086429731113316e-06, + "loss": 0.5625, + "step": 4698 + }, + { + "epoch": 3.890392048591938, + "grad_norm": 0.9381130933761597, + "learning_rate": 1.4076537063420642e-06, + "loss": 0.5905, + "step": 4699 + }, + { + "epoch": 3.891220320265047, + "grad_norm": 0.9857736229896545, + "learning_rate": 1.4066646509316222e-06, + "loss": 0.5887, + "step": 4700 + }, + { + "epoch": 3.8920485919381558, + "grad_norm": 0.9994931817054749, + "learning_rate": 1.4056758070713788e-06, + "loss": 0.5691, + "step": 4701 + }, + { + "epoch": 3.8928768636112645, + "grad_norm": 0.9347181916236877, + "learning_rate": 1.404687174952665e-06, + "loss": 0.5806, + "step": 4702 + }, + { + "epoch": 3.8937051352843732, + "grad_norm": 0.9398673176765442, + "learning_rate": 1.4036987547667725e-06, + "loss": 0.5626, + "step": 4703 + }, + { + "epoch": 3.894533406957482, + "grad_norm": 0.9418836832046509, + "learning_rate": 1.4027105467049521e-06, + "loss": 0.5514, + "step": 4704 + }, + { + "epoch": 3.8953616786305907, + "grad_norm": 0.9223387241363525, + "learning_rate": 1.401722550958411e-06, + "loss": 0.5374, + "step": 4705 + }, + { + "epoch": 3.8961899503036994, + "grad_norm": 0.9222207069396973, + "learning_rate": 1.4007347677183197e-06, + "loss": 0.5535, + "step": 4706 + }, + { + "epoch": 3.8970182219768086, + "grad_norm": 0.9431470036506653, + "learning_rate": 1.3997471971758022e-06, + "loss": 0.5727, + "step": 4707 + }, + { + "epoch": 3.8978464936499173, + "grad_norm": 0.9395588040351868, + "learning_rate": 1.398759839521947e-06, + "loss": 0.5484, + "step": 4708 + }, + { + "epoch": 3.898674765323026, + "grad_norm": 0.9754805564880371, + "learning_rate": 1.397772694947796e-06, + "loss": 0.5617, + "step": 4709 + }, + { + "epoch": 3.8995030369961348, + "grad_norm": 0.9596531987190247, + "learning_rate": 1.3967857636443539e-06, + "loss": 0.5472, + "step": 4710 + }, + { + "epoch": 3.9003313086692435, + "grad_norm": 0.961883008480072, + "learning_rate": 1.3957990458025833e-06, + "loss": 0.5697, + "step": 4711 + }, + { + "epoch": 3.901159580342352, + "grad_norm": 0.9568914175033569, + "learning_rate": 1.3948125416134029e-06, + "loss": 0.56, + "step": 4712 + }, + { + "epoch": 3.901987852015461, + "grad_norm": 0.9424316883087158, + "learning_rate": 1.3938262512676937e-06, + "loss": 0.5485, + "step": 4713 + }, + { + "epoch": 3.90281612368857, + "grad_norm": 0.9500327706336975, + "learning_rate": 1.392840174956294e-06, + "loss": 0.5748, + "step": 4714 + }, + { + "epoch": 3.9036443953616784, + "grad_norm": 0.9448386430740356, + "learning_rate": 1.3918543128699986e-06, + "loss": 0.5407, + "step": 4715 + }, + { + "epoch": 3.9044726670347876, + "grad_norm": 0.9605383276939392, + "learning_rate": 1.390868665199564e-06, + "loss": 0.5728, + "step": 4716 + }, + { + "epoch": 3.9053009387078963, + "grad_norm": 0.9455352425575256, + "learning_rate": 1.389883232135705e-06, + "loss": 0.5506, + "step": 4717 + }, + { + "epoch": 3.906129210381005, + "grad_norm": 0.9913429617881775, + "learning_rate": 1.3888980138690906e-06, + "loss": 0.5536, + "step": 4718 + }, + { + "epoch": 3.9069574820541138, + "grad_norm": 0.9688431024551392, + "learning_rate": 1.3879130105903526e-06, + "loss": 0.555, + "step": 4719 + }, + { + "epoch": 3.9077857537272225, + "grad_norm": 0.9419496655464172, + "learning_rate": 1.3869282224900815e-06, + "loss": 0.5727, + "step": 4720 + }, + { + "epoch": 3.908614025400331, + "grad_norm": 0.9332079291343689, + "learning_rate": 1.3859436497588225e-06, + "loss": 0.572, + "step": 4721 + }, + { + "epoch": 3.90944229707344, + "grad_norm": 0.952717125415802, + "learning_rate": 1.3849592925870821e-06, + "loss": 0.56, + "step": 4722 + }, + { + "epoch": 3.910270568746549, + "grad_norm": 0.944283127784729, + "learning_rate": 1.3839751511653253e-06, + "loss": 0.5803, + "step": 4723 + }, + { + "epoch": 3.9110988404196574, + "grad_norm": 0.9370842576026917, + "learning_rate": 1.3829912256839725e-06, + "loss": 0.5802, + "step": 4724 + }, + { + "epoch": 3.9119271120927666, + "grad_norm": 0.9332932829856873, + "learning_rate": 1.3820075163334047e-06, + "loss": 0.5648, + "step": 4725 + }, + { + "epoch": 3.9127553837658753, + "grad_norm": 0.9598782658576965, + "learning_rate": 1.3810240233039618e-06, + "loss": 0.5565, + "step": 4726 + }, + { + "epoch": 3.913583655438984, + "grad_norm": 0.9564656019210815, + "learning_rate": 1.3800407467859383e-06, + "loss": 0.5594, + "step": 4727 + }, + { + "epoch": 3.9144119271120927, + "grad_norm": 0.9756644368171692, + "learning_rate": 1.3790576869695913e-06, + "loss": 0.5425, + "step": 4728 + }, + { + "epoch": 3.9152401987852015, + "grad_norm": 0.9616433382034302, + "learning_rate": 1.3780748440451314e-06, + "loss": 0.5802, + "step": 4729 + }, + { + "epoch": 3.91606847045831, + "grad_norm": 0.9436020851135254, + "learning_rate": 1.3770922182027319e-06, + "loss": 0.5554, + "step": 4730 + }, + { + "epoch": 3.916896742131419, + "grad_norm": 0.9598777890205383, + "learning_rate": 1.3761098096325195e-06, + "loss": 0.5552, + "step": 4731 + }, + { + "epoch": 3.917725013804528, + "grad_norm": 0.9497062563896179, + "learning_rate": 1.3751276185245821e-06, + "loss": 0.5612, + "step": 4732 + }, + { + "epoch": 3.918553285477637, + "grad_norm": 0.9690037369728088, + "learning_rate": 1.3741456450689656e-06, + "loss": 0.5682, + "step": 4733 + }, + { + "epoch": 3.9193815571507455, + "grad_norm": 0.9524545073509216, + "learning_rate": 1.373163889455671e-06, + "loss": 0.5795, + "step": 4734 + }, + { + "epoch": 3.9202098288238543, + "grad_norm": 0.9572216272354126, + "learning_rate": 1.3721823518746591e-06, + "loss": 0.558, + "step": 4735 + }, + { + "epoch": 3.921038100496963, + "grad_norm": 0.9508054256439209, + "learning_rate": 1.3712010325158499e-06, + "loss": 0.5578, + "step": 4736 + }, + { + "epoch": 3.9218663721700717, + "grad_norm": 0.9361369013786316, + "learning_rate": 1.3702199315691176e-06, + "loss": 0.5661, + "step": 4737 + }, + { + "epoch": 3.9226946438431805, + "grad_norm": 0.9344385266304016, + "learning_rate": 1.3692390492242973e-06, + "loss": 0.571, + "step": 4738 + }, + { + "epoch": 3.923522915516289, + "grad_norm": 0.9559999704360962, + "learning_rate": 1.3682583856711807e-06, + "loss": 0.5441, + "step": 4739 + }, + { + "epoch": 3.924351187189398, + "grad_norm": 0.9483859539031982, + "learning_rate": 1.3672779410995155e-06, + "loss": 0.5645, + "step": 4740 + }, + { + "epoch": 3.925179458862507, + "grad_norm": 0.9504157304763794, + "learning_rate": 1.3662977156990093e-06, + "loss": 0.5512, + "step": 4741 + }, + { + "epoch": 3.926007730535616, + "grad_norm": 0.9570707678794861, + "learning_rate": 1.3653177096593282e-06, + "loss": 0.5722, + "step": 4742 + }, + { + "epoch": 3.9268360022087245, + "grad_norm": 0.9542663097381592, + "learning_rate": 1.3643379231700915e-06, + "loss": 0.5385, + "step": 4743 + }, + { + "epoch": 3.9276642738818333, + "grad_norm": 0.9425368309020996, + "learning_rate": 1.3633583564208808e-06, + "loss": 0.547, + "step": 4744 + }, + { + "epoch": 3.928492545554942, + "grad_norm": 0.9227926731109619, + "learning_rate": 1.362379009601233e-06, + "loss": 0.5664, + "step": 4745 + }, + { + "epoch": 3.9293208172280507, + "grad_norm": 0.9677916765213013, + "learning_rate": 1.3613998829006408e-06, + "loss": 0.5622, + "step": 4746 + }, + { + "epoch": 3.9301490889011594, + "grad_norm": 0.9727103114128113, + "learning_rate": 1.360420976508558e-06, + "loss": 0.5794, + "step": 4747 + }, + { + "epoch": 3.9309773605742686, + "grad_norm": 0.930270254611969, + "learning_rate": 1.359442290614394e-06, + "loss": 0.573, + "step": 4748 + }, + { + "epoch": 3.931805632247377, + "grad_norm": 1.0135208368301392, + "learning_rate": 1.3584638254075132e-06, + "loss": 0.5741, + "step": 4749 + }, + { + "epoch": 3.932633903920486, + "grad_norm": 0.992636501789093, + "learning_rate": 1.3574855810772414e-06, + "loss": 0.5592, + "step": 4750 + }, + { + "epoch": 3.933462175593595, + "grad_norm": 0.9497395157814026, + "learning_rate": 1.3565075578128603e-06, + "loss": 0.5579, + "step": 4751 + }, + { + "epoch": 3.9342904472667035, + "grad_norm": 0.9188456535339355, + "learning_rate": 1.355529755803607e-06, + "loss": 0.573, + "step": 4752 + }, + { + "epoch": 3.9351187189398122, + "grad_norm": 0.9138023257255554, + "learning_rate": 1.3545521752386765e-06, + "loss": 0.5895, + "step": 4753 + }, + { + "epoch": 3.935946990612921, + "grad_norm": 0.9659999012947083, + "learning_rate": 1.3535748163072233e-06, + "loss": 0.5506, + "step": 4754 + }, + { + "epoch": 3.9367752622860297, + "grad_norm": 0.9442874789237976, + "learning_rate": 1.3525976791983553e-06, + "loss": 0.5679, + "step": 4755 + }, + { + "epoch": 3.9376035339591384, + "grad_norm": 0.9444726705551147, + "learning_rate": 1.3516207641011407e-06, + "loss": 0.5795, + "step": 4756 + }, + { + "epoch": 3.9384318056322476, + "grad_norm": 0.9704516530036926, + "learning_rate": 1.3506440712046037e-06, + "loss": 0.568, + "step": 4757 + }, + { + "epoch": 3.9392600773053563, + "grad_norm": 0.9629958271980286, + "learning_rate": 1.349667600697724e-06, + "loss": 0.5967, + "step": 4758 + }, + { + "epoch": 3.940088348978465, + "grad_norm": 0.9573612213134766, + "learning_rate": 1.3486913527694406e-06, + "loss": 0.5553, + "step": 4759 + }, + { + "epoch": 3.9409166206515738, + "grad_norm": 0.9458691477775574, + "learning_rate": 1.3477153276086485e-06, + "loss": 0.5768, + "step": 4760 + }, + { + "epoch": 3.9417448923246825, + "grad_norm": 0.938666820526123, + "learning_rate": 1.3467395254041985e-06, + "loss": 0.5465, + "step": 4761 + }, + { + "epoch": 3.9425731639977912, + "grad_norm": 0.9396777153015137, + "learning_rate": 1.3457639463449008e-06, + "loss": 0.5699, + "step": 4762 + }, + { + "epoch": 3.9434014356709, + "grad_norm": 0.9707586765289307, + "learning_rate": 1.3447885906195186e-06, + "loss": 0.5472, + "step": 4763 + }, + { + "epoch": 3.9442297073440087, + "grad_norm": 0.9531437754631042, + "learning_rate": 1.343813458416776e-06, + "loss": 0.5712, + "step": 4764 + }, + { + "epoch": 3.9450579790171174, + "grad_norm": 0.942042350769043, + "learning_rate": 1.342838549925351e-06, + "loss": 0.5494, + "step": 4765 + }, + { + "epoch": 3.9458862506902266, + "grad_norm": 0.9602234959602356, + "learning_rate": 1.3418638653338795e-06, + "loss": 0.5627, + "step": 4766 + }, + { + "epoch": 3.9467145223633353, + "grad_norm": 0.9987154006958008, + "learning_rate": 1.340889404830955e-06, + "loss": 0.5824, + "step": 4767 + }, + { + "epoch": 3.947542794036444, + "grad_norm": 0.9674735069274902, + "learning_rate": 1.3399151686051246e-06, + "loss": 0.5772, + "step": 4768 + }, + { + "epoch": 3.9483710657095528, + "grad_norm": 0.9081838130950928, + "learning_rate": 1.3389411568448948e-06, + "loss": 0.555, + "step": 4769 + }, + { + "epoch": 3.9491993373826615, + "grad_norm": 0.9668666124343872, + "learning_rate": 1.337967369738729e-06, + "loss": 0.5655, + "step": 4770 + }, + { + "epoch": 3.95002760905577, + "grad_norm": 0.9260414838790894, + "learning_rate": 1.3369938074750438e-06, + "loss": 0.5685, + "step": 4771 + }, + { + "epoch": 3.950855880728879, + "grad_norm": 0.9354709386825562, + "learning_rate": 1.3360204702422154e-06, + "loss": 0.5623, + "step": 4772 + }, + { + "epoch": 3.951684152401988, + "grad_norm": 0.965032696723938, + "learning_rate": 1.3350473582285777e-06, + "loss": 0.5846, + "step": 4773 + }, + { + "epoch": 3.9525124240750964, + "grad_norm": 0.9616831541061401, + "learning_rate": 1.3340744716224143e-06, + "loss": 0.5576, + "step": 4774 + }, + { + "epoch": 3.9533406957482056, + "grad_norm": 0.9441544413566589, + "learning_rate": 1.3331018106119726e-06, + "loss": 0.5918, + "step": 4775 + }, + { + "epoch": 3.9541689674213143, + "grad_norm": 0.9979905486106873, + "learning_rate": 1.332129375385453e-06, + "loss": 0.5539, + "step": 4776 + }, + { + "epoch": 3.954997239094423, + "grad_norm": 0.950507402420044, + "learning_rate": 1.3311571661310124e-06, + "loss": 0.5463, + "step": 4777 + }, + { + "epoch": 3.9558255107675317, + "grad_norm": 0.9438148140907288, + "learning_rate": 1.3301851830367637e-06, + "loss": 0.5645, + "step": 4778 + }, + { + "epoch": 3.9566537824406405, + "grad_norm": 0.9625552296638489, + "learning_rate": 1.3292134262907785e-06, + "loss": 0.5782, + "step": 4779 + }, + { + "epoch": 3.957482054113749, + "grad_norm": 0.958839476108551, + "learning_rate": 1.3282418960810803e-06, + "loss": 0.562, + "step": 4780 + }, + { + "epoch": 3.958310325786858, + "grad_norm": 0.9578886032104492, + "learning_rate": 1.3272705925956522e-06, + "loss": 0.5609, + "step": 4781 + }, + { + "epoch": 3.959138597459967, + "grad_norm": 0.9848271608352661, + "learning_rate": 1.3262995160224335e-06, + "loss": 0.5627, + "step": 4782 + }, + { + "epoch": 3.9599668691330754, + "grad_norm": 0.9630492925643921, + "learning_rate": 1.3253286665493164e-06, + "loss": 0.5356, + "step": 4783 + }, + { + "epoch": 3.9607951408061846, + "grad_norm": 0.945763111114502, + "learning_rate": 1.3243580443641518e-06, + "loss": 0.5688, + "step": 4784 + }, + { + "epoch": 3.9616234124792933, + "grad_norm": 0.9506339430809021, + "learning_rate": 1.3233876496547477e-06, + "loss": 0.5725, + "step": 4785 + }, + { + "epoch": 3.962451684152402, + "grad_norm": 0.9367161393165588, + "learning_rate": 1.3224174826088654e-06, + "loss": 0.5605, + "step": 4786 + }, + { + "epoch": 3.9632799558255107, + "grad_norm": 0.9559187889099121, + "learning_rate": 1.3214475434142215e-06, + "loss": 0.5606, + "step": 4787 + }, + { + "epoch": 3.9641082274986195, + "grad_norm": 0.9559148550033569, + "learning_rate": 1.320477832258492e-06, + "loss": 0.5516, + "step": 4788 + }, + { + "epoch": 3.964936499171728, + "grad_norm": 0.9415640234947205, + "learning_rate": 1.319508349329307e-06, + "loss": 0.5607, + "step": 4789 + }, + { + "epoch": 3.965764770844837, + "grad_norm": 0.961673378944397, + "learning_rate": 1.3185390948142516e-06, + "loss": 0.571, + "step": 4790 + }, + { + "epoch": 3.966593042517946, + "grad_norm": 0.9346276521682739, + "learning_rate": 1.3175700689008678e-06, + "loss": 0.5716, + "step": 4791 + }, + { + "epoch": 3.967421314191055, + "grad_norm": 0.9483547806739807, + "learning_rate": 1.3166012717766541e-06, + "loss": 0.5654, + "step": 4792 + }, + { + "epoch": 3.9682495858641635, + "grad_norm": 0.9249748587608337, + "learning_rate": 1.3156327036290616e-06, + "loss": 0.5669, + "step": 4793 + }, + { + "epoch": 3.9690778575372723, + "grad_norm": 0.965886116027832, + "learning_rate": 1.3146643646455007e-06, + "loss": 0.5739, + "step": 4794 + }, + { + "epoch": 3.969906129210381, + "grad_norm": 0.9492125511169434, + "learning_rate": 1.3136962550133364e-06, + "loss": 0.5521, + "step": 4795 + }, + { + "epoch": 3.9707344008834897, + "grad_norm": 0.9473196864128113, + "learning_rate": 1.3127283749198885e-06, + "loss": 0.5412, + "step": 4796 + }, + { + "epoch": 3.9715626725565985, + "grad_norm": 0.9501288533210754, + "learning_rate": 1.3117607245524314e-06, + "loss": 0.5451, + "step": 4797 + }, + { + "epoch": 3.9723909442297076, + "grad_norm": 0.9439449310302734, + "learning_rate": 1.3107933040981985e-06, + "loss": 0.5702, + "step": 4798 + }, + { + "epoch": 3.973219215902816, + "grad_norm": 0.9604007005691528, + "learning_rate": 1.3098261137443744e-06, + "loss": 0.5674, + "step": 4799 + }, + { + "epoch": 3.974047487575925, + "grad_norm": 0.9559398889541626, + "learning_rate": 1.3088591536781026e-06, + "loss": 0.5618, + "step": 4800 + }, + { + "epoch": 3.974875759249034, + "grad_norm": 0.93898606300354, + "learning_rate": 1.3078924240864821e-06, + "loss": 0.5539, + "step": 4801 + }, + { + "epoch": 3.9757040309221425, + "grad_norm": 0.985381007194519, + "learning_rate": 1.306925925156564e-06, + "loss": 0.5748, + "step": 4802 + }, + { + "epoch": 3.9765323025952513, + "grad_norm": 0.9823545217514038, + "learning_rate": 1.3059596570753575e-06, + "loss": 0.569, + "step": 4803 + }, + { + "epoch": 3.97736057426836, + "grad_norm": 0.9789100289344788, + "learning_rate": 1.3049936200298274e-06, + "loss": 0.5784, + "step": 4804 + }, + { + "epoch": 3.9781888459414687, + "grad_norm": 0.9664195775985718, + "learning_rate": 1.3040278142068912e-06, + "loss": 0.574, + "step": 4805 + }, + { + "epoch": 3.9790171176145774, + "grad_norm": 0.9816479682922363, + "learning_rate": 1.3030622397934244e-06, + "loss": 0.607, + "step": 4806 + }, + { + "epoch": 3.9798453892876866, + "grad_norm": 0.9329001903533936, + "learning_rate": 1.3020968969762577e-06, + "loss": 0.5666, + "step": 4807 + }, + { + "epoch": 3.980673660960795, + "grad_norm": 0.9252684116363525, + "learning_rate": 1.3011317859421744e-06, + "loss": 0.5528, + "step": 4808 + }, + { + "epoch": 3.981501932633904, + "grad_norm": 0.9480351209640503, + "learning_rate": 1.3001669068779137e-06, + "loss": 0.5685, + "step": 4809 + }, + { + "epoch": 3.982330204307013, + "grad_norm": 0.9316738247871399, + "learning_rate": 1.2992022599701723e-06, + "loss": 0.569, + "step": 4810 + }, + { + "epoch": 3.9831584759801215, + "grad_norm": 0.9480258822441101, + "learning_rate": 1.2982378454056005e-06, + "loss": 0.5519, + "step": 4811 + }, + { + "epoch": 3.9839867476532302, + "grad_norm": 0.9391124248504639, + "learning_rate": 1.2972736633708021e-06, + "loss": 0.5657, + "step": 4812 + }, + { + "epoch": 3.984815019326339, + "grad_norm": 0.9289792776107788, + "learning_rate": 1.2963097140523389e-06, + "loss": 0.5258, + "step": 4813 + }, + { + "epoch": 3.9856432909994477, + "grad_norm": 0.9436705708503723, + "learning_rate": 1.295345997636725e-06, + "loss": 0.5672, + "step": 4814 + }, + { + "epoch": 3.9864715626725564, + "grad_norm": 0.9601346850395203, + "learning_rate": 1.2943825143104305e-06, + "loss": 0.5628, + "step": 4815 + }, + { + "epoch": 3.9872998343456656, + "grad_norm": 0.9829810261726379, + "learning_rate": 1.2934192642598818e-06, + "loss": 0.5772, + "step": 4816 + }, + { + "epoch": 3.9881281060187743, + "grad_norm": 0.9281584620475769, + "learning_rate": 1.292456247671457e-06, + "loss": 0.5763, + "step": 4817 + }, + { + "epoch": 3.988956377691883, + "grad_norm": 0.9712902903556824, + "learning_rate": 1.2914934647314914e-06, + "loss": 0.5553, + "step": 4818 + }, + { + "epoch": 3.9897846493649918, + "grad_norm": 0.9826544523239136, + "learning_rate": 1.290530915626276e-06, + "loss": 0.5802, + "step": 4819 + }, + { + "epoch": 3.9906129210381005, + "grad_norm": 0.9495535492897034, + "learning_rate": 1.289568600542054e-06, + "loss": 0.5746, + "step": 4820 + }, + { + "epoch": 3.9914411927112092, + "grad_norm": 0.9266762137413025, + "learning_rate": 1.2886065196650232e-06, + "loss": 0.5825, + "step": 4821 + }, + { + "epoch": 3.992269464384318, + "grad_norm": 0.946962833404541, + "learning_rate": 1.2876446731813387e-06, + "loss": 0.5825, + "step": 4822 + }, + { + "epoch": 3.9930977360574267, + "grad_norm": 0.9361686110496521, + "learning_rate": 1.2866830612771091e-06, + "loss": 0.5473, + "step": 4823 + }, + { + "epoch": 3.9939260077305354, + "grad_norm": 0.9517462849617004, + "learning_rate": 1.2857216841383958e-06, + "loss": 0.5628, + "step": 4824 + }, + { + "epoch": 3.9947542794036446, + "grad_norm": 0.957137405872345, + "learning_rate": 1.2847605419512174e-06, + "loss": 0.5516, + "step": 4825 + }, + { + "epoch": 3.9955825510767533, + "grad_norm": 0.9692016243934631, + "learning_rate": 1.2837996349015465e-06, + "loss": 0.5738, + "step": 4826 + }, + { + "epoch": 3.996410822749862, + "grad_norm": 0.9570938348770142, + "learning_rate": 1.2828389631753085e-06, + "loss": 0.5971, + "step": 4827 + }, + { + "epoch": 3.9972390944229708, + "grad_norm": 0.9742267727851868, + "learning_rate": 1.2818785269583845e-06, + "loss": 0.5721, + "step": 4828 + }, + { + "epoch": 3.9980673660960795, + "grad_norm": 0.9525173902511597, + "learning_rate": 1.2809183264366115e-06, + "loss": 0.5548, + "step": 4829 + }, + { + "epoch": 3.998895637769188, + "grad_norm": 0.9499869346618652, + "learning_rate": 1.279958361795778e-06, + "loss": 0.5825, + "step": 4830 + }, + { + "epoch": 3.999723909442297, + "grad_norm": 0.9081862568855286, + "learning_rate": 1.2789986332216276e-06, + "loss": 0.5835, + "step": 4831 + }, + { + "epoch": 4.0, + "grad_norm": 0.9081862568855286, + "learning_rate": 1.2780391408998607e-06, + "loss": 0.5353, + "step": 4832 + }, + { + "epoch": 4.000828271673109, + "grad_norm": 1.8408721685409546, + "learning_rate": 1.2770798850161281e-06, + "loss": 0.544, + "step": 4833 + }, + { + "epoch": 4.0016565433462175, + "grad_norm": 0.950079083442688, + "learning_rate": 1.2761208657560376e-06, + "loss": 0.5485, + "step": 4834 + }, + { + "epoch": 4.002484815019327, + "grad_norm": 0.9583503603935242, + "learning_rate": 1.275162083305152e-06, + "loss": 0.5657, + "step": 4835 + }, + { + "epoch": 4.003313086692435, + "grad_norm": 0.9331668615341187, + "learning_rate": 1.2742035378489841e-06, + "loss": 0.5673, + "step": 4836 + }, + { + "epoch": 4.004141358365544, + "grad_norm": 0.9338731169700623, + "learning_rate": 1.2732452295730053e-06, + "loss": 0.5333, + "step": 4837 + }, + { + "epoch": 4.004969630038652, + "grad_norm": 0.9427415132522583, + "learning_rate": 1.2722871586626395e-06, + "loss": 0.5584, + "step": 4838 + }, + { + "epoch": 4.0057979017117615, + "grad_norm": 0.9849832653999329, + "learning_rate": 1.2713293253032632e-06, + "loss": 0.5486, + "step": 4839 + }, + { + "epoch": 4.00662617338487, + "grad_norm": 0.9528194069862366, + "learning_rate": 1.270371729680209e-06, + "loss": 0.5574, + "step": 4840 + }, + { + "epoch": 4.007454445057979, + "grad_norm": 0.9479176998138428, + "learning_rate": 1.2694143719787633e-06, + "loss": 0.5601, + "step": 4841 + }, + { + "epoch": 4.008282716731088, + "grad_norm": 0.9426152110099792, + "learning_rate": 1.2684572523841654e-06, + "loss": 0.5599, + "step": 4842 + }, + { + "epoch": 4.009110988404196, + "grad_norm": 0.9592022895812988, + "learning_rate": 1.2675003710816081e-06, + "loss": 0.5579, + "step": 4843 + }, + { + "epoch": 4.009939260077306, + "grad_norm": 0.9946890473365784, + "learning_rate": 1.2665437282562398e-06, + "loss": 0.5443, + "step": 4844 + }, + { + "epoch": 4.010767531750414, + "grad_norm": 0.9415493011474609, + "learning_rate": 1.2655873240931626e-06, + "loss": 0.5685, + "step": 4845 + }, + { + "epoch": 4.011595803423523, + "grad_norm": 0.9373499155044556, + "learning_rate": 1.2646311587774304e-06, + "loss": 0.5609, + "step": 4846 + }, + { + "epoch": 4.012424075096631, + "grad_norm": 0.9726147651672363, + "learning_rate": 1.263675232494053e-06, + "loss": 0.571, + "step": 4847 + }, + { + "epoch": 4.0132523467697405, + "grad_norm": 0.9378729462623596, + "learning_rate": 1.262719545427994e-06, + "loss": 0.5653, + "step": 4848 + }, + { + "epoch": 4.01408061844285, + "grad_norm": 0.9564341902732849, + "learning_rate": 1.2617640977641684e-06, + "loss": 0.5502, + "step": 4849 + }, + { + "epoch": 4.014908890115958, + "grad_norm": 0.9591152667999268, + "learning_rate": 1.260808889687447e-06, + "loss": 0.551, + "step": 4850 + }, + { + "epoch": 4.015737161789067, + "grad_norm": 0.9530390501022339, + "learning_rate": 1.259853921382655e-06, + "loss": 0.5297, + "step": 4851 + }, + { + "epoch": 4.016565433462175, + "grad_norm": 0.9545491933822632, + "learning_rate": 1.258899193034569e-06, + "loss": 0.5399, + "step": 4852 + }, + { + "epoch": 4.017393705135285, + "grad_norm": 0.9423142671585083, + "learning_rate": 1.2579447048279184e-06, + "loss": 0.584, + "step": 4853 + }, + { + "epoch": 4.018221976808393, + "grad_norm": 0.933307945728302, + "learning_rate": 1.2569904569473901e-06, + "loss": 0.5463, + "step": 4854 + }, + { + "epoch": 4.019050248481502, + "grad_norm": 0.9511076211929321, + "learning_rate": 1.2560364495776202e-06, + "loss": 0.5652, + "step": 4855 + }, + { + "epoch": 4.01987852015461, + "grad_norm": 0.9643457531929016, + "learning_rate": 1.2550826829032015e-06, + "loss": 0.5329, + "step": 4856 + }, + { + "epoch": 4.0207067918277195, + "grad_norm": 0.9959326386451721, + "learning_rate": 1.2541291571086794e-06, + "loss": 0.5545, + "step": 4857 + }, + { + "epoch": 4.021535063500829, + "grad_norm": 0.9828082323074341, + "learning_rate": 1.2531758723785505e-06, + "loss": 0.5396, + "step": 4858 + }, + { + "epoch": 4.022363335173937, + "grad_norm": 0.9460323452949524, + "learning_rate": 1.2522228288972676e-06, + "loss": 0.5334, + "step": 4859 + }, + { + "epoch": 4.023191606847046, + "grad_norm": 0.9510787725448608, + "learning_rate": 1.251270026849237e-06, + "loss": 0.5765, + "step": 4860 + }, + { + "epoch": 4.024019878520154, + "grad_norm": 0.9397110342979431, + "learning_rate": 1.2503174664188145e-06, + "loss": 0.52, + "step": 4861 + }, + { + "epoch": 4.024848150193264, + "grad_norm": 0.9416722655296326, + "learning_rate": 1.2493651477903126e-06, + "loss": 0.5497, + "step": 4862 + }, + { + "epoch": 4.025676421866372, + "grad_norm": 0.9606363773345947, + "learning_rate": 1.2484130711479977e-06, + "loss": 0.5442, + "step": 4863 + }, + { + "epoch": 4.026504693539481, + "grad_norm": 0.9509909749031067, + "learning_rate": 1.2474612366760861e-06, + "loss": 0.5768, + "step": 4864 + }, + { + "epoch": 4.027332965212589, + "grad_norm": 0.9398588538169861, + "learning_rate": 1.2465096445587486e-06, + "loss": 0.5337, + "step": 4865 + }, + { + "epoch": 4.0281612368856985, + "grad_norm": 1.0803090333938599, + "learning_rate": 1.2455582949801095e-06, + "loss": 0.5437, + "step": 4866 + }, + { + "epoch": 4.028989508558808, + "grad_norm": 0.9432228207588196, + "learning_rate": 1.2446071881242478e-06, + "loss": 0.5439, + "step": 4867 + }, + { + "epoch": 4.029817780231916, + "grad_norm": 0.962143063545227, + "learning_rate": 1.2436563241751914e-06, + "loss": 0.5444, + "step": 4868 + }, + { + "epoch": 4.030646051905025, + "grad_norm": 0.9307393431663513, + "learning_rate": 1.2427057033169252e-06, + "loss": 0.5684, + "step": 4869 + }, + { + "epoch": 4.031474323578133, + "grad_norm": 0.9873812198638916, + "learning_rate": 1.2417553257333856e-06, + "loss": 0.5602, + "step": 4870 + }, + { + "epoch": 4.032302595251243, + "grad_norm": 0.9589493870735168, + "learning_rate": 1.2408051916084603e-06, + "loss": 0.5257, + "step": 4871 + }, + { + "epoch": 4.033130866924351, + "grad_norm": 0.9406089186668396, + "learning_rate": 1.239855301125993e-06, + "loss": 0.5417, + "step": 4872 + }, + { + "epoch": 4.03395913859746, + "grad_norm": 0.9455104470252991, + "learning_rate": 1.2389056544697781e-06, + "loss": 0.52, + "step": 4873 + }, + { + "epoch": 4.034787410270568, + "grad_norm": 0.9494171142578125, + "learning_rate": 1.2379562518235628e-06, + "loss": 0.5469, + "step": 4874 + }, + { + "epoch": 4.0356156819436775, + "grad_norm": 0.9548773765563965, + "learning_rate": 1.2370070933710491e-06, + "loss": 0.5548, + "step": 4875 + }, + { + "epoch": 4.036443953616787, + "grad_norm": 0.9773174524307251, + "learning_rate": 1.2360581792958893e-06, + "loss": 0.5484, + "step": 4876 + }, + { + "epoch": 4.037272225289895, + "grad_norm": 0.9558678269386292, + "learning_rate": 1.2351095097816885e-06, + "loss": 0.5578, + "step": 4877 + }, + { + "epoch": 4.038100496963004, + "grad_norm": 1.1254467964172363, + "learning_rate": 1.2341610850120064e-06, + "loss": 0.5588, + "step": 4878 + }, + { + "epoch": 4.038928768636112, + "grad_norm": 0.9522780776023865, + "learning_rate": 1.2332129051703553e-06, + "loss": 0.5518, + "step": 4879 + }, + { + "epoch": 4.039757040309222, + "grad_norm": 0.9933323860168457, + "learning_rate": 1.2322649704401973e-06, + "loss": 0.556, + "step": 4880 + }, + { + "epoch": 4.04058531198233, + "grad_norm": 0.9555357098579407, + "learning_rate": 1.2313172810049497e-06, + "loss": 0.5678, + "step": 4881 + }, + { + "epoch": 4.041413583655439, + "grad_norm": 0.9392233490943909, + "learning_rate": 1.230369837047983e-06, + "loss": 0.5307, + "step": 4882 + }, + { + "epoch": 4.042241855328548, + "grad_norm": 0.9643675088882446, + "learning_rate": 1.2294226387526165e-06, + "loss": 0.5391, + "step": 4883 + }, + { + "epoch": 4.0430701270016565, + "grad_norm": 0.9979854226112366, + "learning_rate": 1.2284756863021249e-06, + "loss": 0.567, + "step": 4884 + }, + { + "epoch": 4.043898398674766, + "grad_norm": 0.9585314393043518, + "learning_rate": 1.227528979879736e-06, + "loss": 0.5315, + "step": 4885 + }, + { + "epoch": 4.044726670347874, + "grad_norm": 0.9538570642471313, + "learning_rate": 1.226582519668628e-06, + "loss": 0.5525, + "step": 4886 + }, + { + "epoch": 4.045554942020983, + "grad_norm": 0.9781275987625122, + "learning_rate": 1.2256363058519305e-06, + "loss": 0.5745, + "step": 4887 + }, + { + "epoch": 4.046383213694091, + "grad_norm": 0.9645776152610779, + "learning_rate": 1.2246903386127284e-06, + "loss": 0.5631, + "step": 4888 + }, + { + "epoch": 4.0472114853672005, + "grad_norm": 0.9528216123580933, + "learning_rate": 1.2237446181340587e-06, + "loss": 0.5697, + "step": 4889 + }, + { + "epoch": 4.048039757040309, + "grad_norm": 0.9601432085037231, + "learning_rate": 1.2227991445989072e-06, + "loss": 0.5333, + "step": 4890 + }, + { + "epoch": 4.048868028713418, + "grad_norm": 0.9265902638435364, + "learning_rate": 1.221853918190216e-06, + "loss": 0.5535, + "step": 4891 + }, + { + "epoch": 4.049696300386527, + "grad_norm": 0.940017819404602, + "learning_rate": 1.2209089390908762e-06, + "loss": 0.5493, + "step": 4892 + }, + { + "epoch": 4.0505245720596355, + "grad_norm": 0.935411274433136, + "learning_rate": 1.2199642074837334e-06, + "loss": 0.5763, + "step": 4893 + }, + { + "epoch": 4.051352843732745, + "grad_norm": 0.9772312045097351, + "learning_rate": 1.2190197235515846e-06, + "loss": 0.5326, + "step": 4894 + }, + { + "epoch": 4.052181115405853, + "grad_norm": 0.9477226138114929, + "learning_rate": 1.2180754874771774e-06, + "loss": 0.548, + "step": 4895 + }, + { + "epoch": 4.053009387078962, + "grad_norm": 0.9361460208892822, + "learning_rate": 1.2171314994432134e-06, + "loss": 0.5692, + "step": 4896 + }, + { + "epoch": 4.05383765875207, + "grad_norm": 0.9590089321136475, + "learning_rate": 1.2161877596323466e-06, + "loss": 0.5494, + "step": 4897 + }, + { + "epoch": 4.0546659304251795, + "grad_norm": 0.9448971152305603, + "learning_rate": 1.2152442682271807e-06, + "loss": 0.5623, + "step": 4898 + }, + { + "epoch": 4.055494202098288, + "grad_norm": 0.9581391215324402, + "learning_rate": 1.2143010254102716e-06, + "loss": 0.5525, + "step": 4899 + }, + { + "epoch": 4.056322473771397, + "grad_norm": 0.9311408996582031, + "learning_rate": 1.2133580313641289e-06, + "loss": 0.5317, + "step": 4900 + }, + { + "epoch": 4.057150745444506, + "grad_norm": 0.9715316295623779, + "learning_rate": 1.212415286271214e-06, + "loss": 0.5626, + "step": 4901 + }, + { + "epoch": 4.057979017117614, + "grad_norm": 0.9454675316810608, + "learning_rate": 1.2114727903139379e-06, + "loss": 0.567, + "step": 4902 + }, + { + "epoch": 4.058807288790724, + "grad_norm": 0.9326028227806091, + "learning_rate": 1.210530543674665e-06, + "loss": 0.5372, + "step": 4903 + }, + { + "epoch": 4.059635560463832, + "grad_norm": 0.9475375413894653, + "learning_rate": 1.2095885465357128e-06, + "loss": 0.5521, + "step": 4904 + }, + { + "epoch": 4.060463832136941, + "grad_norm": 0.9660394787788391, + "learning_rate": 1.2086467990793469e-06, + "loss": 0.5478, + "step": 4905 + }, + { + "epoch": 4.061292103810049, + "grad_norm": 0.9396241903305054, + "learning_rate": 1.2077053014877874e-06, + "loss": 0.5489, + "step": 4906 + }, + { + "epoch": 4.0621203754831585, + "grad_norm": 0.9579562544822693, + "learning_rate": 1.2067640539432063e-06, + "loss": 0.5455, + "step": 4907 + }, + { + "epoch": 4.062948647156268, + "grad_norm": 0.9574836492538452, + "learning_rate": 1.2058230566277246e-06, + "loss": 0.5425, + "step": 4908 + }, + { + "epoch": 4.063776918829376, + "grad_norm": 0.9654014706611633, + "learning_rate": 1.204882309723418e-06, + "loss": 0.5399, + "step": 4909 + }, + { + "epoch": 4.064605190502485, + "grad_norm": 0.9588503241539001, + "learning_rate": 1.203941813412312e-06, + "loss": 0.5633, + "step": 4910 + }, + { + "epoch": 4.065433462175593, + "grad_norm": 0.9639830589294434, + "learning_rate": 1.203001567876382e-06, + "loss": 0.5445, + "step": 4911 + }, + { + "epoch": 4.066261733848703, + "grad_norm": 0.9721859693527222, + "learning_rate": 1.2020615732975582e-06, + "loss": 0.5378, + "step": 4912 + }, + { + "epoch": 4.067090005521811, + "grad_norm": 0.9711841940879822, + "learning_rate": 1.2011218298577218e-06, + "loss": 0.5462, + "step": 4913 + }, + { + "epoch": 4.06791827719492, + "grad_norm": 0.9579488039016724, + "learning_rate": 1.2001823377387022e-06, + "loss": 0.5468, + "step": 4914 + }, + { + "epoch": 4.068746548868028, + "grad_norm": 0.9761105179786682, + "learning_rate": 1.1992430971222839e-06, + "loss": 0.5534, + "step": 4915 + }, + { + "epoch": 4.0695748205411375, + "grad_norm": 0.9403853416442871, + "learning_rate": 1.1983041081902016e-06, + "loss": 0.5701, + "step": 4916 + }, + { + "epoch": 4.070403092214247, + "grad_norm": 0.9711827039718628, + "learning_rate": 1.197365371124139e-06, + "loss": 0.5674, + "step": 4917 + }, + { + "epoch": 4.071231363887355, + "grad_norm": 0.9828927516937256, + "learning_rate": 1.1964268861057343e-06, + "loss": 0.5562, + "step": 4918 + }, + { + "epoch": 4.072059635560464, + "grad_norm": 0.9552168250083923, + "learning_rate": 1.195488653316576e-06, + "loss": 0.5549, + "step": 4919 + }, + { + "epoch": 4.072887907233572, + "grad_norm": 0.9774886965751648, + "learning_rate": 1.194550672938203e-06, + "loss": 0.5602, + "step": 4920 + }, + { + "epoch": 4.073716178906682, + "grad_norm": 0.9632622599601746, + "learning_rate": 1.1936129451521048e-06, + "loss": 0.562, + "step": 4921 + }, + { + "epoch": 4.07454445057979, + "grad_norm": 0.9444198608398438, + "learning_rate": 1.1926754701397234e-06, + "loss": 0.5557, + "step": 4922 + }, + { + "epoch": 4.075372722252899, + "grad_norm": 0.9769295454025269, + "learning_rate": 1.191738248082453e-06, + "loss": 0.5604, + "step": 4923 + }, + { + "epoch": 4.076200993926007, + "grad_norm": 0.9791744947433472, + "learning_rate": 1.190801279161635e-06, + "loss": 0.5539, + "step": 4924 + }, + { + "epoch": 4.0770292655991165, + "grad_norm": 0.9577887654304504, + "learning_rate": 1.1898645635585652e-06, + "loss": 0.5288, + "step": 4925 + }, + { + "epoch": 4.077857537272226, + "grad_norm": 0.9718350768089294, + "learning_rate": 1.1889281014544907e-06, + "loss": 0.5587, + "step": 4926 + }, + { + "epoch": 4.078685808945334, + "grad_norm": 0.9948698878288269, + "learning_rate": 1.1879918930306057e-06, + "loss": 0.5454, + "step": 4927 + }, + { + "epoch": 4.079514080618443, + "grad_norm": 0.9597614407539368, + "learning_rate": 1.187055938468059e-06, + "loss": 0.5506, + "step": 4928 + }, + { + "epoch": 4.080342352291551, + "grad_norm": 1.0018571615219116, + "learning_rate": 1.1861202379479504e-06, + "loss": 0.5659, + "step": 4929 + }, + { + "epoch": 4.081170623964661, + "grad_norm": 0.9607214331626892, + "learning_rate": 1.1851847916513265e-06, + "loss": 0.5459, + "step": 4930 + }, + { + "epoch": 4.081998895637769, + "grad_norm": 0.958617091178894, + "learning_rate": 1.1842495997591891e-06, + "loss": 0.5569, + "step": 4931 + }, + { + "epoch": 4.082827167310878, + "grad_norm": 0.9821828007698059, + "learning_rate": 1.1833146624524907e-06, + "loss": 0.5503, + "step": 4932 + }, + { + "epoch": 4.083655438983987, + "grad_norm": 0.9603815078735352, + "learning_rate": 1.1823799799121293e-06, + "loss": 0.5537, + "step": 4933 + }, + { + "epoch": 4.0844837106570955, + "grad_norm": 0.9634932279586792, + "learning_rate": 1.181445552318959e-06, + "loss": 0.5436, + "step": 4934 + }, + { + "epoch": 4.085311982330205, + "grad_norm": 0.9784684777259827, + "learning_rate": 1.180511379853784e-06, + "loss": 0.5454, + "step": 4935 + }, + { + "epoch": 4.086140254003313, + "grad_norm": 0.9718360304832458, + "learning_rate": 1.1795774626973558e-06, + "loss": 0.5568, + "step": 4936 + }, + { + "epoch": 4.086968525676422, + "grad_norm": 0.962182879447937, + "learning_rate": 1.1786438010303799e-06, + "loss": 0.5388, + "step": 4937 + }, + { + "epoch": 4.08779679734953, + "grad_norm": 0.9819624423980713, + "learning_rate": 1.1777103950335119e-06, + "loss": 0.5678, + "step": 4938 + }, + { + "epoch": 4.08862506902264, + "grad_norm": 0.9512423872947693, + "learning_rate": 1.1767772448873553e-06, + "loss": 0.54, + "step": 4939 + }, + { + "epoch": 4.089453340695748, + "grad_norm": 0.9459388256072998, + "learning_rate": 1.1758443507724667e-06, + "loss": 0.5427, + "step": 4940 + }, + { + "epoch": 4.090281612368857, + "grad_norm": 0.9718794822692871, + "learning_rate": 1.1749117128693538e-06, + "loss": 0.542, + "step": 4941 + }, + { + "epoch": 4.091109884041966, + "grad_norm": 0.9619993567466736, + "learning_rate": 1.1739793313584711e-06, + "loss": 0.556, + "step": 4942 + }, + { + "epoch": 4.0919381557150745, + "grad_norm": 0.9623501300811768, + "learning_rate": 1.1730472064202276e-06, + "loss": 0.5503, + "step": 4943 + }, + { + "epoch": 4.092766427388184, + "grad_norm": 0.9375072717666626, + "learning_rate": 1.172115338234979e-06, + "loss": 0.5542, + "step": 4944 + }, + { + "epoch": 4.093594699061292, + "grad_norm": 0.9621657133102417, + "learning_rate": 1.171183726983035e-06, + "loss": 0.5379, + "step": 4945 + }, + { + "epoch": 4.094422970734401, + "grad_norm": 0.9603687524795532, + "learning_rate": 1.1702523728446522e-06, + "loss": 0.535, + "step": 4946 + }, + { + "epoch": 4.095251242407509, + "grad_norm": 0.9662330150604248, + "learning_rate": 1.1693212760000392e-06, + "loss": 0.5437, + "step": 4947 + }, + { + "epoch": 4.0960795140806185, + "grad_norm": 0.9543896913528442, + "learning_rate": 1.1683904366293558e-06, + "loss": 0.5531, + "step": 4948 + }, + { + "epoch": 4.096907785753727, + "grad_norm": 0.9495199918746948, + "learning_rate": 1.1674598549127088e-06, + "loss": 0.5416, + "step": 4949 + }, + { + "epoch": 4.097736057426836, + "grad_norm": 1.0251307487487793, + "learning_rate": 1.1665295310301582e-06, + "loss": 0.5424, + "step": 4950 + }, + { + "epoch": 4.098564329099945, + "grad_norm": 0.9514643549919128, + "learning_rate": 1.1655994651617138e-06, + "loss": 0.5618, + "step": 4951 + }, + { + "epoch": 4.0993926007730535, + "grad_norm": 0.9422398209571838, + "learning_rate": 1.1646696574873332e-06, + "loss": 0.5412, + "step": 4952 + }, + { + "epoch": 4.100220872446163, + "grad_norm": 0.9614496231079102, + "learning_rate": 1.1637401081869266e-06, + "loss": 0.5473, + "step": 4953 + }, + { + "epoch": 4.101049144119271, + "grad_norm": 0.953124463558197, + "learning_rate": 1.1628108174403526e-06, + "loss": 0.5448, + "step": 4954 + }, + { + "epoch": 4.10187741579238, + "grad_norm": 0.9754185080528259, + "learning_rate": 1.1618817854274195e-06, + "loss": 0.5265, + "step": 4955 + }, + { + "epoch": 4.102705687465488, + "grad_norm": 0.9708113074302673, + "learning_rate": 1.160953012327887e-06, + "loss": 0.5399, + "step": 4956 + }, + { + "epoch": 4.1035339591385975, + "grad_norm": 0.9540616869926453, + "learning_rate": 1.1600244983214654e-06, + "loss": 0.5396, + "step": 4957 + }, + { + "epoch": 4.104362230811706, + "grad_norm": 0.9439674019813538, + "learning_rate": 1.159096243587811e-06, + "loss": 0.5688, + "step": 4958 + }, + { + "epoch": 4.105190502484815, + "grad_norm": 0.9611501097679138, + "learning_rate": 1.158168248306534e-06, + "loss": 0.5548, + "step": 4959 + }, + { + "epoch": 4.106018774157924, + "grad_norm": 0.9381349682807922, + "learning_rate": 1.1572405126571935e-06, + "loss": 0.5291, + "step": 4960 + }, + { + "epoch": 4.106847045831032, + "grad_norm": 0.9494814276695251, + "learning_rate": 1.1563130368192955e-06, + "loss": 0.5478, + "step": 4961 + }, + { + "epoch": 4.107675317504142, + "grad_norm": 0.9880107641220093, + "learning_rate": 1.1553858209722996e-06, + "loss": 0.5592, + "step": 4962 + }, + { + "epoch": 4.10850358917725, + "grad_norm": 0.9652872681617737, + "learning_rate": 1.1544588652956135e-06, + "loss": 0.5697, + "step": 4963 + }, + { + "epoch": 4.109331860850359, + "grad_norm": 0.9371438026428223, + "learning_rate": 1.1535321699685934e-06, + "loss": 0.5363, + "step": 4964 + }, + { + "epoch": 4.110160132523467, + "grad_norm": 0.9895375967025757, + "learning_rate": 1.1526057351705475e-06, + "loss": 0.5427, + "step": 4965 + }, + { + "epoch": 4.1109884041965765, + "grad_norm": 0.9242417812347412, + "learning_rate": 1.1516795610807318e-06, + "loss": 0.5222, + "step": 4966 + }, + { + "epoch": 4.111816675869686, + "grad_norm": 0.9818521738052368, + "learning_rate": 1.1507536478783509e-06, + "loss": 0.5729, + "step": 4967 + }, + { + "epoch": 4.112644947542794, + "grad_norm": 0.9748761653900146, + "learning_rate": 1.149827995742562e-06, + "loss": 0.5553, + "step": 4968 + }, + { + "epoch": 4.113473219215903, + "grad_norm": 0.9590509533882141, + "learning_rate": 1.1489026048524707e-06, + "loss": 0.5397, + "step": 4969 + }, + { + "epoch": 4.114301490889011, + "grad_norm": 0.956331729888916, + "learning_rate": 1.1479774753871294e-06, + "loss": 0.5818, + "step": 4970 + }, + { + "epoch": 4.115129762562121, + "grad_norm": 0.9534357786178589, + "learning_rate": 1.1470526075255434e-06, + "loss": 0.5665, + "step": 4971 + }, + { + "epoch": 4.115958034235229, + "grad_norm": 0.9746752381324768, + "learning_rate": 1.1461280014466667e-06, + "loss": 0.5518, + "step": 4972 + }, + { + "epoch": 4.116786305908338, + "grad_norm": 0.9731650948524475, + "learning_rate": 1.1452036573294e-06, + "loss": 0.5683, + "step": 4973 + }, + { + "epoch": 4.117614577581446, + "grad_norm": 0.9399099349975586, + "learning_rate": 1.1442795753525969e-06, + "loss": 0.5614, + "step": 4974 + }, + { + "epoch": 4.1184428492545555, + "grad_norm": 0.9529602527618408, + "learning_rate": 1.1433557556950583e-06, + "loss": 0.5456, + "step": 4975 + }, + { + "epoch": 4.119271120927665, + "grad_norm": 0.9645585417747498, + "learning_rate": 1.142432198535534e-06, + "loss": 0.5386, + "step": 4976 + }, + { + "epoch": 4.120099392600773, + "grad_norm": 0.9642344117164612, + "learning_rate": 1.141508904052725e-06, + "loss": 0.5716, + "step": 4977 + }, + { + "epoch": 4.120927664273882, + "grad_norm": 0.9634251594543457, + "learning_rate": 1.1405858724252787e-06, + "loss": 0.5613, + "step": 4978 + }, + { + "epoch": 4.12175593594699, + "grad_norm": 0.9380096793174744, + "learning_rate": 1.1396631038317944e-06, + "loss": 0.5439, + "step": 4979 + }, + { + "epoch": 4.1225842076201, + "grad_norm": 0.963697075843811, + "learning_rate": 1.1387405984508177e-06, + "loss": 0.5605, + "step": 4980 + }, + { + "epoch": 4.123412479293208, + "grad_norm": 0.9700136780738831, + "learning_rate": 1.1378183564608459e-06, + "loss": 0.5671, + "step": 4981 + }, + { + "epoch": 4.124240750966317, + "grad_norm": 0.9514760971069336, + "learning_rate": 1.1368963780403248e-06, + "loss": 0.5446, + "step": 4982 + }, + { + "epoch": 4.125069022639425, + "grad_norm": 0.9677363634109497, + "learning_rate": 1.135974663367647e-06, + "loss": 0.5387, + "step": 4983 + }, + { + "epoch": 4.1258972943125345, + "grad_norm": 0.9543762803077698, + "learning_rate": 1.1350532126211564e-06, + "loss": 0.5859, + "step": 4984 + }, + { + "epoch": 4.126725565985644, + "grad_norm": 0.9666879773139954, + "learning_rate": 1.1341320259791463e-06, + "loss": 0.5626, + "step": 4985 + }, + { + "epoch": 4.127553837658752, + "grad_norm": 0.9627404808998108, + "learning_rate": 1.133211103619856e-06, + "loss": 0.5593, + "step": 4986 + }, + { + "epoch": 4.128382109331861, + "grad_norm": 0.9710012674331665, + "learning_rate": 1.1322904457214759e-06, + "loss": 0.5665, + "step": 4987 + }, + { + "epoch": 4.129210381004969, + "grad_norm": 0.9189611077308655, + "learning_rate": 1.1313700524621468e-06, + "loss": 0.5435, + "step": 4988 + }, + { + "epoch": 4.130038652678079, + "grad_norm": 0.9373254776000977, + "learning_rate": 1.1304499240199523e-06, + "loss": 0.5832, + "step": 4989 + }, + { + "epoch": 4.130866924351187, + "grad_norm": 0.967099130153656, + "learning_rate": 1.1295300605729311e-06, + "loss": 0.5398, + "step": 4990 + }, + { + "epoch": 4.131695196024296, + "grad_norm": 0.9356493353843689, + "learning_rate": 1.1286104622990684e-06, + "loss": 0.5659, + "step": 4991 + }, + { + "epoch": 4.132523467697405, + "grad_norm": 0.970832347869873, + "learning_rate": 1.1276911293762968e-06, + "loss": 0.557, + "step": 4992 + }, + { + "epoch": 4.1333517393705135, + "grad_norm": 0.9779406189918518, + "learning_rate": 1.1267720619824986e-06, + "loss": 0.5918, + "step": 4993 + }, + { + "epoch": 4.134180011043623, + "grad_norm": 0.9740846157073975, + "learning_rate": 1.1258532602955069e-06, + "loss": 0.55, + "step": 4994 + }, + { + "epoch": 4.135008282716731, + "grad_norm": 0.959419846534729, + "learning_rate": 1.1249347244930983e-06, + "loss": 0.5383, + "step": 4995 + }, + { + "epoch": 4.13583655438984, + "grad_norm": 0.9859365224838257, + "learning_rate": 1.1240164547530025e-06, + "loss": 0.5651, + "step": 4996 + }, + { + "epoch": 4.136664826062948, + "grad_norm": 0.9637149572372437, + "learning_rate": 1.123098451252897e-06, + "loss": 0.5352, + "step": 4997 + }, + { + "epoch": 4.137493097736058, + "grad_norm": 0.9376708269119263, + "learning_rate": 1.1221807141704048e-06, + "loss": 0.5651, + "step": 4998 + }, + { + "epoch": 4.138321369409166, + "grad_norm": 0.971984326839447, + "learning_rate": 1.1212632436831015e-06, + "loss": 0.5429, + "step": 4999 + }, + { + "epoch": 4.139149641082275, + "grad_norm": 0.9628618955612183, + "learning_rate": 1.1203460399685071e-06, + "loss": 0.534, + "step": 5000 + }, + { + "epoch": 4.139977912755384, + "grad_norm": 0.9970594048500061, + "learning_rate": 1.119429103204094e-06, + "loss": 0.5468, + "step": 5001 + }, + { + "epoch": 4.1408061844284925, + "grad_norm": 0.9739755988121033, + "learning_rate": 1.118512433567279e-06, + "loss": 0.5457, + "step": 5002 + }, + { + "epoch": 4.141634456101602, + "grad_norm": 0.9746037125587463, + "learning_rate": 1.11759603123543e-06, + "loss": 0.5637, + "step": 5003 + }, + { + "epoch": 4.14246272777471, + "grad_norm": 0.9784834980964661, + "learning_rate": 1.1166798963858633e-06, + "loss": 0.5673, + "step": 5004 + }, + { + "epoch": 4.143290999447819, + "grad_norm": 0.9432827830314636, + "learning_rate": 1.1157640291958404e-06, + "loss": 0.5435, + "step": 5005 + }, + { + "epoch": 4.144119271120927, + "grad_norm": 0.9606834053993225, + "learning_rate": 1.1148484298425743e-06, + "loss": 0.5407, + "step": 5006 + }, + { + "epoch": 4.1449475427940365, + "grad_norm": 0.9408406019210815, + "learning_rate": 1.113933098503226e-06, + "loss": 0.5439, + "step": 5007 + }, + { + "epoch": 4.145775814467145, + "grad_norm": 0.9570668339729309, + "learning_rate": 1.1130180353549012e-06, + "loss": 0.5335, + "step": 5008 + }, + { + "epoch": 4.146604086140254, + "grad_norm": 0.9712949991226196, + "learning_rate": 1.1121032405746574e-06, + "loss": 0.5564, + "step": 5009 + }, + { + "epoch": 4.147432357813363, + "grad_norm": 0.9611795544624329, + "learning_rate": 1.1111887143394995e-06, + "loss": 0.556, + "step": 5010 + }, + { + "epoch": 4.1482606294864715, + "grad_norm": 1.120573878288269, + "learning_rate": 1.110274456826379e-06, + "loss": 0.5556, + "step": 5011 + }, + { + "epoch": 4.149088901159581, + "grad_norm": 0.979116678237915, + "learning_rate": 1.1093604682121956e-06, + "loss": 0.5448, + "step": 5012 + }, + { + "epoch": 4.149917172832689, + "grad_norm": 0.964521050453186, + "learning_rate": 1.1084467486737988e-06, + "loss": 0.5537, + "step": 5013 + }, + { + "epoch": 4.150745444505798, + "grad_norm": 0.9442644715309143, + "learning_rate": 1.1075332983879835e-06, + "loss": 0.5662, + "step": 5014 + }, + { + "epoch": 4.151573716178906, + "grad_norm": 0.9522957801818848, + "learning_rate": 1.1066201175314945e-06, + "loss": 0.5296, + "step": 5015 + }, + { + "epoch": 4.1524019878520155, + "grad_norm": 1.3315223455429077, + "learning_rate": 1.1057072062810247e-06, + "loss": 0.583, + "step": 5016 + }, + { + "epoch": 4.153230259525124, + "grad_norm": 0.9419928789138794, + "learning_rate": 1.1047945648132124e-06, + "loss": 0.5538, + "step": 5017 + }, + { + "epoch": 4.154058531198233, + "grad_norm": 0.9866177439689636, + "learning_rate": 1.1038821933046456e-06, + "loss": 0.5587, + "step": 5018 + }, + { + "epoch": 4.154886802871342, + "grad_norm": 0.9400057196617126, + "learning_rate": 1.102970091931861e-06, + "loss": 0.5498, + "step": 5019 + }, + { + "epoch": 4.15571507454445, + "grad_norm": 0.9604130387306213, + "learning_rate": 1.1020582608713398e-06, + "loss": 0.5431, + "step": 5020 + }, + { + "epoch": 4.15654334621756, + "grad_norm": 0.9357389807701111, + "learning_rate": 1.1011467002995135e-06, + "loss": 0.5611, + "step": 5021 + }, + { + "epoch": 4.157371617890668, + "grad_norm": 0.9960111379623413, + "learning_rate": 1.100235410392762e-06, + "loss": 0.5611, + "step": 5022 + }, + { + "epoch": 4.158199889563777, + "grad_norm": 0.987533450126648, + "learning_rate": 1.0993243913274099e-06, + "loss": 0.5598, + "step": 5023 + }, + { + "epoch": 4.159028161236885, + "grad_norm": 0.9703227877616882, + "learning_rate": 1.09841364327973e-06, + "loss": 0.5574, + "step": 5024 + }, + { + "epoch": 4.1598564329099945, + "grad_norm": 0.9649814963340759, + "learning_rate": 1.097503166425945e-06, + "loss": 0.5696, + "step": 5025 + }, + { + "epoch": 4.160684704583104, + "grad_norm": 0.9445302486419678, + "learning_rate": 1.0965929609422243e-06, + "loss": 0.5529, + "step": 5026 + }, + { + "epoch": 4.161512976256212, + "grad_norm": 0.9760971069335938, + "learning_rate": 1.0956830270046823e-06, + "loss": 0.5412, + "step": 5027 + }, + { + "epoch": 4.162341247929321, + "grad_norm": 0.9667978882789612, + "learning_rate": 1.0947733647893845e-06, + "loss": 0.5407, + "step": 5028 + }, + { + "epoch": 4.163169519602429, + "grad_norm": 0.9345208406448364, + "learning_rate": 1.0938639744723405e-06, + "loss": 0.5673, + "step": 5029 + }, + { + "epoch": 4.163997791275539, + "grad_norm": 0.9527465105056763, + "learning_rate": 1.0929548562295098e-06, + "loss": 0.5441, + "step": 5030 + }, + { + "epoch": 4.164826062948647, + "grad_norm": 0.9564769268035889, + "learning_rate": 1.0920460102367988e-06, + "loss": 0.5487, + "step": 5031 + }, + { + "epoch": 4.165654334621756, + "grad_norm": 0.9697523713111877, + "learning_rate": 1.0911374366700593e-06, + "loss": 0.5309, + "step": 5032 + }, + { + "epoch": 4.166482606294864, + "grad_norm": 0.9563925266265869, + "learning_rate": 1.0902291357050932e-06, + "loss": 0.5432, + "step": 5033 + }, + { + "epoch": 4.1673108779679735, + "grad_norm": 0.9529309272766113, + "learning_rate": 1.0893211075176465e-06, + "loss": 0.5528, + "step": 5034 + }, + { + "epoch": 4.168139149641083, + "grad_norm": 0.9415030479431152, + "learning_rate": 1.0884133522834168e-06, + "loss": 0.5491, + "step": 5035 + }, + { + "epoch": 4.168967421314191, + "grad_norm": 0.9647647142410278, + "learning_rate": 1.0875058701780432e-06, + "loss": 0.5509, + "step": 5036 + }, + { + "epoch": 4.1697956929873, + "grad_norm": 0.9571723937988281, + "learning_rate": 1.0865986613771173e-06, + "loss": 0.5494, + "step": 5037 + }, + { + "epoch": 4.170623964660408, + "grad_norm": 0.9916718602180481, + "learning_rate": 1.0856917260561755e-06, + "loss": 0.5801, + "step": 5038 + }, + { + "epoch": 4.171452236333518, + "grad_norm": 0.9753178358078003, + "learning_rate": 1.0847850643906997e-06, + "loss": 0.5663, + "step": 5039 + }, + { + "epoch": 4.172280508006626, + "grad_norm": 0.9760473966598511, + "learning_rate": 1.0838786765561217e-06, + "loss": 0.5528, + "step": 5040 + }, + { + "epoch": 4.173108779679735, + "grad_norm": 0.9597117304801941, + "learning_rate": 1.0829725627278195e-06, + "loss": 0.5523, + "step": 5041 + }, + { + "epoch": 4.173937051352843, + "grad_norm": 0.9663171768188477, + "learning_rate": 1.0820667230811164e-06, + "loss": 0.5468, + "step": 5042 + }, + { + "epoch": 4.1747653230259525, + "grad_norm": 0.9724160432815552, + "learning_rate": 1.0811611577912841e-06, + "loss": 0.5627, + "step": 5043 + }, + { + "epoch": 4.175593594699062, + "grad_norm": 0.9543465375900269, + "learning_rate": 1.080255867033543e-06, + "loss": 0.5518, + "step": 5044 + }, + { + "epoch": 4.17642186637217, + "grad_norm": 0.9619806408882141, + "learning_rate": 1.0793508509830567e-06, + "loss": 0.5268, + "step": 5045 + }, + { + "epoch": 4.177250138045279, + "grad_norm": 0.9508680105209351, + "learning_rate": 1.0784461098149367e-06, + "loss": 0.5328, + "step": 5046 + }, + { + "epoch": 4.178078409718387, + "grad_norm": 0.944930911064148, + "learning_rate": 1.0775416437042438e-06, + "loss": 0.5393, + "step": 5047 + }, + { + "epoch": 4.178906681391497, + "grad_norm": 0.987951397895813, + "learning_rate": 1.0766374528259822e-06, + "loss": 0.5556, + "step": 5048 + }, + { + "epoch": 4.179734953064605, + "grad_norm": 0.960111677646637, + "learning_rate": 1.0757335373551047e-06, + "loss": 0.5473, + "step": 5049 + }, + { + "epoch": 4.180563224737714, + "grad_norm": 0.9849942922592163, + "learning_rate": 1.0748298974665119e-06, + "loss": 0.5559, + "step": 5050 + }, + { + "epoch": 4.181391496410823, + "grad_norm": 0.9856019020080566, + "learning_rate": 1.073926533335048e-06, + "loss": 0.5547, + "step": 5051 + }, + { + "epoch": 4.1822197680839315, + "grad_norm": 0.9418588280677795, + "learning_rate": 1.073023445135506e-06, + "loss": 0.5377, + "step": 5052 + }, + { + "epoch": 4.183048039757041, + "grad_norm": 0.9605504274368286, + "learning_rate": 1.072120633042626e-06, + "loss": 0.5513, + "step": 5053 + }, + { + "epoch": 4.183876311430149, + "grad_norm": 0.9515420198440552, + "learning_rate": 1.0712180972310923e-06, + "loss": 0.5331, + "step": 5054 + }, + { + "epoch": 4.184704583103258, + "grad_norm": 0.9456093907356262, + "learning_rate": 1.0703158378755376e-06, + "loss": 0.5318, + "step": 5055 + }, + { + "epoch": 4.185532854776366, + "grad_norm": 0.986414909362793, + "learning_rate": 1.0694138551505424e-06, + "loss": 0.553, + "step": 5056 + }, + { + "epoch": 4.186361126449476, + "grad_norm": 0.9519484639167786, + "learning_rate": 1.0685121492306299e-06, + "loss": 0.5364, + "step": 5057 + }, + { + "epoch": 4.187189398122584, + "grad_norm": 0.9635508060455322, + "learning_rate": 1.0676107202902717e-06, + "loss": 0.5414, + "step": 5058 + }, + { + "epoch": 4.188017669795693, + "grad_norm": 0.948654294013977, + "learning_rate": 1.0667095685038864e-06, + "loss": 0.5756, + "step": 5059 + }, + { + "epoch": 4.188845941468802, + "grad_norm": 0.960182249546051, + "learning_rate": 1.0658086940458397e-06, + "loss": 0.5499, + "step": 5060 + }, + { + "epoch": 4.1896742131419105, + "grad_norm": 0.9578927755355835, + "learning_rate": 1.0649080970904405e-06, + "loss": 0.5625, + "step": 5061 + }, + { + "epoch": 4.19050248481502, + "grad_norm": 0.9450972676277161, + "learning_rate": 1.0640077778119466e-06, + "loss": 0.5544, + "step": 5062 + }, + { + "epoch": 4.191330756488128, + "grad_norm": 0.9294493794441223, + "learning_rate": 1.0631077363845624e-06, + "loss": 0.547, + "step": 5063 + }, + { + "epoch": 4.192159028161237, + "grad_norm": 0.9553670287132263, + "learning_rate": 1.062207972982436e-06, + "loss": 0.5548, + "step": 5064 + }, + { + "epoch": 4.192987299834345, + "grad_norm": 0.9614834189414978, + "learning_rate": 1.0613084877796636e-06, + "loss": 0.5564, + "step": 5065 + }, + { + "epoch": 4.1938155715074545, + "grad_norm": 0.9543241262435913, + "learning_rate": 1.0604092809502886e-06, + "loss": 0.5657, + "step": 5066 + }, + { + "epoch": 4.194643843180563, + "grad_norm": 0.9624195694923401, + "learning_rate": 1.0595103526682982e-06, + "loss": 0.5594, + "step": 5067 + }, + { + "epoch": 4.195472114853672, + "grad_norm": 0.9818905591964722, + "learning_rate": 1.0586117031076254e-06, + "loss": 0.5446, + "step": 5068 + }, + { + "epoch": 4.196300386526781, + "grad_norm": 0.9506664872169495, + "learning_rate": 1.0577133324421524e-06, + "loss": 0.558, + "step": 5069 + }, + { + "epoch": 4.1971286581998895, + "grad_norm": 0.9724001884460449, + "learning_rate": 1.056815240845704e-06, + "loss": 0.5575, + "step": 5070 + }, + { + "epoch": 4.197956929872999, + "grad_norm": 0.9375852346420288, + "learning_rate": 1.0559174284920531e-06, + "loss": 0.5326, + "step": 5071 + }, + { + "epoch": 4.198785201546107, + "grad_norm": 0.9587619304656982, + "learning_rate": 1.0550198955549194e-06, + "loss": 0.5861, + "step": 5072 + }, + { + "epoch": 4.199613473219216, + "grad_norm": 0.9693782329559326, + "learning_rate": 1.0541226422079648e-06, + "loss": 0.5612, + "step": 5073 + }, + { + "epoch": 4.200441744892324, + "grad_norm": 0.980484664440155, + "learning_rate": 1.0532256686248008e-06, + "loss": 0.5604, + "step": 5074 + }, + { + "epoch": 4.2012700165654335, + "grad_norm": 0.9361084699630737, + "learning_rate": 1.052328974978984e-06, + "loss": 0.5673, + "step": 5075 + }, + { + "epoch": 4.202098288238542, + "grad_norm": 0.992881715297699, + "learning_rate": 1.051432561444015e-06, + "loss": 0.5581, + "step": 5076 + }, + { + "epoch": 4.202926559911651, + "grad_norm": 0.955773651599884, + "learning_rate": 1.0505364281933418e-06, + "loss": 0.5315, + "step": 5077 + }, + { + "epoch": 4.20375483158476, + "grad_norm": 0.9578268527984619, + "learning_rate": 1.0496405754003588e-06, + "loss": 0.5598, + "step": 5078 + }, + { + "epoch": 4.204583103257868, + "grad_norm": 0.9477030634880066, + "learning_rate": 1.0487450032384045e-06, + "loss": 0.5414, + "step": 5079 + }, + { + "epoch": 4.205411374930978, + "grad_norm": 0.932655930519104, + "learning_rate": 1.0478497118807626e-06, + "loss": 0.535, + "step": 5080 + }, + { + "epoch": 4.206239646604086, + "grad_norm": 0.9418008327484131, + "learning_rate": 1.0469547015006649e-06, + "loss": 0.5484, + "step": 5081 + }, + { + "epoch": 4.207067918277195, + "grad_norm": 0.9843934774398804, + "learning_rate": 1.046059972271288e-06, + "loss": 0.5437, + "step": 5082 + }, + { + "epoch": 4.207896189950303, + "grad_norm": 0.9433165192604065, + "learning_rate": 1.045165524365752e-06, + "loss": 0.5372, + "step": 5083 + }, + { + "epoch": 4.2087244616234125, + "grad_norm": 0.9636999368667603, + "learning_rate": 1.0442713579571257e-06, + "loss": 0.5382, + "step": 5084 + }, + { + "epoch": 4.209552733296522, + "grad_norm": 0.9424169063568115, + "learning_rate": 1.043377473218422e-06, + "loss": 0.5359, + "step": 5085 + }, + { + "epoch": 4.21038100496963, + "grad_norm": 0.9796680212020874, + "learning_rate": 1.0424838703225981e-06, + "loss": 0.5489, + "step": 5086 + }, + { + "epoch": 4.211209276642739, + "grad_norm": 1.0004616975784302, + "learning_rate": 1.0415905494425584e-06, + "loss": 0.5472, + "step": 5087 + }, + { + "epoch": 4.212037548315847, + "grad_norm": 0.9699249267578125, + "learning_rate": 1.0406975107511532e-06, + "loss": 0.5553, + "step": 5088 + }, + { + "epoch": 4.212865819988957, + "grad_norm": 0.956947922706604, + "learning_rate": 1.0398047544211756e-06, + "loss": 0.5533, + "step": 5089 + }, + { + "epoch": 4.213694091662065, + "grad_norm": 1.0034435987472534, + "learning_rate": 1.0389122806253667e-06, + "loss": 0.5597, + "step": 5090 + }, + { + "epoch": 4.214522363335174, + "grad_norm": 0.9732010364532471, + "learning_rate": 1.0380200895364118e-06, + "loss": 0.5387, + "step": 5091 + }, + { + "epoch": 4.215350635008282, + "grad_norm": 0.9744183421134949, + "learning_rate": 1.03712818132694e-06, + "loss": 0.5531, + "step": 5092 + }, + { + "epoch": 4.2161789066813915, + "grad_norm": 1.0084949731826782, + "learning_rate": 1.0362365561695283e-06, + "loss": 0.5601, + "step": 5093 + }, + { + "epoch": 4.217007178354501, + "grad_norm": 0.9781489372253418, + "learning_rate": 1.0353452142366993e-06, + "loss": 0.5537, + "step": 5094 + }, + { + "epoch": 4.217835450027609, + "grad_norm": 0.9882619976997375, + "learning_rate": 1.0344541557009166e-06, + "loss": 0.5543, + "step": 5095 + }, + { + "epoch": 4.218663721700718, + "grad_norm": 0.9402450323104858, + "learning_rate": 1.0335633807345934e-06, + "loss": 0.5586, + "step": 5096 + }, + { + "epoch": 4.219491993373826, + "grad_norm": 0.9744458794593811, + "learning_rate": 1.0326728895100872e-06, + "loss": 0.5716, + "step": 5097 + }, + { + "epoch": 4.220320265046936, + "grad_norm": 0.9703624248504639, + "learning_rate": 1.0317826821996974e-06, + "loss": 0.5534, + "step": 5098 + }, + { + "epoch": 4.221148536720044, + "grad_norm": 0.9642719030380249, + "learning_rate": 1.030892758975672e-06, + "loss": 0.531, + "step": 5099 + }, + { + "epoch": 4.221976808393153, + "grad_norm": 0.9380668997764587, + "learning_rate": 1.030003120010204e-06, + "loss": 0.5602, + "step": 5100 + }, + { + "epoch": 4.222805080066261, + "grad_norm": 0.9409314393997192, + "learning_rate": 1.0291137654754289e-06, + "loss": 0.5551, + "step": 5101 + }, + { + "epoch": 4.2236333517393705, + "grad_norm": 0.9610958695411682, + "learning_rate": 1.028224695543428e-06, + "loss": 0.5671, + "step": 5102 + }, + { + "epoch": 4.22446162341248, + "grad_norm": 0.9575729966163635, + "learning_rate": 1.0273359103862287e-06, + "loss": 0.5667, + "step": 5103 + }, + { + "epoch": 4.225289895085588, + "grad_norm": 0.9521274566650391, + "learning_rate": 1.0264474101758037e-06, + "loss": 0.5526, + "step": 5104 + }, + { + "epoch": 4.226118166758697, + "grad_norm": 0.9705082178115845, + "learning_rate": 1.0255591950840679e-06, + "loss": 0.5455, + "step": 5105 + }, + { + "epoch": 4.226946438431805, + "grad_norm": 0.9718014001846313, + "learning_rate": 1.0246712652828838e-06, + "loss": 0.5555, + "step": 5106 + }, + { + "epoch": 4.227774710104915, + "grad_norm": 0.9703999757766724, + "learning_rate": 1.0237836209440563e-06, + "loss": 0.5347, + "step": 5107 + }, + { + "epoch": 4.228602981778023, + "grad_norm": 0.9935028553009033, + "learning_rate": 1.0228962622393372e-06, + "loss": 0.5711, + "step": 5108 + }, + { + "epoch": 4.229431253451132, + "grad_norm": 0.9573892951011658, + "learning_rate": 1.0220091893404226e-06, + "loss": 0.5231, + "step": 5109 + }, + { + "epoch": 4.230259525124241, + "grad_norm": 0.9764808416366577, + "learning_rate": 1.0211224024189514e-06, + "loss": 0.5383, + "step": 5110 + }, + { + "epoch": 4.2310877967973495, + "grad_norm": 0.985131025314331, + "learning_rate": 1.0202359016465096e-06, + "loss": 0.5625, + "step": 5111 + }, + { + "epoch": 4.231916068470459, + "grad_norm": 0.9500586986541748, + "learning_rate": 1.0193496871946273e-06, + "loss": 0.5497, + "step": 5112 + }, + { + "epoch": 4.232744340143567, + "grad_norm": 0.9569104313850403, + "learning_rate": 1.0184637592347783e-06, + "loss": 0.5657, + "step": 5113 + }, + { + "epoch": 4.233572611816676, + "grad_norm": 0.9736594557762146, + "learning_rate": 1.01757811793838e-06, + "loss": 0.5345, + "step": 5114 + }, + { + "epoch": 4.234400883489784, + "grad_norm": 0.9908674359321594, + "learning_rate": 1.016692763476797e-06, + "loss": 0.5506, + "step": 5115 + }, + { + "epoch": 4.2352291551628936, + "grad_norm": 0.9769543409347534, + "learning_rate": 1.015807696021338e-06, + "loss": 0.5441, + "step": 5116 + }, + { + "epoch": 4.236057426836002, + "grad_norm": 0.9631603360176086, + "learning_rate": 1.0149229157432536e-06, + "loss": 0.549, + "step": 5117 + }, + { + "epoch": 4.236885698509111, + "grad_norm": 0.9673128128051758, + "learning_rate": 1.014038422813741e-06, + "loss": 0.5687, + "step": 5118 + }, + { + "epoch": 4.23771397018222, + "grad_norm": 1.2093791961669922, + "learning_rate": 1.0131542174039427e-06, + "loss": 0.5588, + "step": 5119 + }, + { + "epoch": 4.2385422418553285, + "grad_norm": 0.9489418268203735, + "learning_rate": 1.0122702996849424e-06, + "loss": 0.5711, + "step": 5120 + }, + { + "epoch": 4.239370513528438, + "grad_norm": 1.0532208681106567, + "learning_rate": 1.0113866698277701e-06, + "loss": 0.5489, + "step": 5121 + }, + { + "epoch": 4.240198785201546, + "grad_norm": 0.9641385078430176, + "learning_rate": 1.0105033280034022e-06, + "loss": 0.553, + "step": 5122 + }, + { + "epoch": 4.241027056874655, + "grad_norm": 0.9720795750617981, + "learning_rate": 1.009620274382754e-06, + "loss": 0.5698, + "step": 5123 + }, + { + "epoch": 4.241855328547763, + "grad_norm": 0.9714332818984985, + "learning_rate": 1.0087375091366906e-06, + "loss": 0.5689, + "step": 5124 + }, + { + "epoch": 4.2426836002208725, + "grad_norm": 0.9626659154891968, + "learning_rate": 1.0078550324360179e-06, + "loss": 0.5699, + "step": 5125 + }, + { + "epoch": 4.243511871893981, + "grad_norm": 0.9580199718475342, + "learning_rate": 1.0069728444514857e-06, + "loss": 0.5559, + "step": 5126 + }, + { + "epoch": 4.24434014356709, + "grad_norm": 0.9758951663970947, + "learning_rate": 1.0060909453537903e-06, + "loss": 0.5513, + "step": 5127 + }, + { + "epoch": 4.245168415240199, + "grad_norm": 0.9805962443351746, + "learning_rate": 1.005209335313572e-06, + "loss": 0.5448, + "step": 5128 + }, + { + "epoch": 4.2459966869133074, + "grad_norm": 1.2901999950408936, + "learning_rate": 1.004328014501412e-06, + "loss": 0.5505, + "step": 5129 + }, + { + "epoch": 4.246824958586417, + "grad_norm": 0.976852297782898, + "learning_rate": 1.0034469830878389e-06, + "loss": 0.531, + "step": 5130 + }, + { + "epoch": 4.247653230259525, + "grad_norm": 0.9891388416290283, + "learning_rate": 1.0025662412433246e-06, + "loss": 0.5493, + "step": 5131 + }, + { + "epoch": 4.248481501932634, + "grad_norm": 1.014784336090088, + "learning_rate": 1.001685789138283e-06, + "loss": 0.553, + "step": 5132 + }, + { + "epoch": 4.249309773605742, + "grad_norm": 0.958575427532196, + "learning_rate": 1.0008056269430739e-06, + "loss": 0.5661, + "step": 5133 + }, + { + "epoch": 4.2501380452788515, + "grad_norm": 0.9698933959007263, + "learning_rate": 9.999257548280018e-07, + "loss": 0.557, + "step": 5134 + }, + { + "epoch": 4.25096631695196, + "grad_norm": 0.9629998803138733, + "learning_rate": 9.990461729633124e-07, + "loss": 0.5512, + "step": 5135 + }, + { + "epoch": 4.251794588625069, + "grad_norm": 1.1001232862472534, + "learning_rate": 9.981668815191961e-07, + "loss": 0.5737, + "step": 5136 + }, + { + "epoch": 4.252622860298178, + "grad_norm": 1.025240421295166, + "learning_rate": 9.972878806657883e-07, + "loss": 0.5531, + "step": 5137 + }, + { + "epoch": 4.253451131971286, + "grad_norm": 0.9492990374565125, + "learning_rate": 9.964091705731683e-07, + "loss": 0.563, + "step": 5138 + }, + { + "epoch": 4.254279403644396, + "grad_norm": 0.9647439122200012, + "learning_rate": 9.955307514113566e-07, + "loss": 0.5729, + "step": 5139 + }, + { + "epoch": 4.255107675317504, + "grad_norm": 0.9514984488487244, + "learning_rate": 9.946526233503199e-07, + "loss": 0.5589, + "step": 5140 + }, + { + "epoch": 4.255935946990613, + "grad_norm": 0.9453758001327515, + "learning_rate": 9.93774786559969e-07, + "loss": 0.537, + "step": 5141 + }, + { + "epoch": 4.256764218663721, + "grad_norm": 0.9403778314590454, + "learning_rate": 9.928972412101551e-07, + "loss": 0.5726, + "step": 5142 + }, + { + "epoch": 4.2575924903368305, + "grad_norm": 0.9746240973472595, + "learning_rate": 9.92019987470676e-07, + "loss": 0.5709, + "step": 5143 + }, + { + "epoch": 4.258420762009939, + "grad_norm": 0.9754739999771118, + "learning_rate": 9.911430255112731e-07, + "loss": 0.5392, + "step": 5144 + }, + { + "epoch": 4.259249033683048, + "grad_norm": 0.9354885816574097, + "learning_rate": 9.90266355501628e-07, + "loss": 0.5256, + "step": 5145 + }, + { + "epoch": 4.260077305356157, + "grad_norm": 0.9805651903152466, + "learning_rate": 9.89389977611371e-07, + "loss": 0.571, + "step": 5146 + }, + { + "epoch": 4.260905577029265, + "grad_norm": 0.9707584977149963, + "learning_rate": 9.885138920100716e-07, + "loss": 0.5717, + "step": 5147 + }, + { + "epoch": 4.261733848702375, + "grad_norm": 0.9590045809745789, + "learning_rate": 9.87638098867243e-07, + "loss": 0.5568, + "step": 5148 + }, + { + "epoch": 4.262562120375483, + "grad_norm": 0.9531744122505188, + "learning_rate": 9.867625983523444e-07, + "loss": 0.5572, + "step": 5149 + }, + { + "epoch": 4.263390392048592, + "grad_norm": 0.9503223896026611, + "learning_rate": 9.858873906347777e-07, + "loss": 0.5503, + "step": 5150 + }, + { + "epoch": 4.2642186637217, + "grad_norm": 0.971651017665863, + "learning_rate": 9.850124758838856e-07, + "loss": 0.5329, + "step": 5151 + }, + { + "epoch": 4.2650469353948095, + "grad_norm": 0.9676386117935181, + "learning_rate": 9.841378542689566e-07, + "loss": 0.5445, + "step": 5152 + }, + { + "epoch": 4.265875207067919, + "grad_norm": 0.9490211606025696, + "learning_rate": 9.832635259592236e-07, + "loss": 0.5605, + "step": 5153 + }, + { + "epoch": 4.266703478741027, + "grad_norm": 0.9679016470909119, + "learning_rate": 9.823894911238582e-07, + "loss": 0.5503, + "step": 5154 + }, + { + "epoch": 4.267531750414136, + "grad_norm": 0.9777132272720337, + "learning_rate": 9.815157499319794e-07, + "loss": 0.556, + "step": 5155 + }, + { + "epoch": 4.268360022087244, + "grad_norm": 0.9584012031555176, + "learning_rate": 9.806423025526487e-07, + "loss": 0.5385, + "step": 5156 + }, + { + "epoch": 4.269188293760354, + "grad_norm": 0.9709147810935974, + "learning_rate": 9.797691491548682e-07, + "loss": 0.5553, + "step": 5157 + }, + { + "epoch": 4.270016565433462, + "grad_norm": 0.9639657735824585, + "learning_rate": 9.788962899075867e-07, + "loss": 0.5429, + "step": 5158 + }, + { + "epoch": 4.270844837106571, + "grad_norm": 0.9870532155036926, + "learning_rate": 9.780237249796932e-07, + "loss": 0.5601, + "step": 5159 + }, + { + "epoch": 4.27167310877968, + "grad_norm": 0.9454158544540405, + "learning_rate": 9.771514545400219e-07, + "loss": 0.5371, + "step": 5160 + }, + { + "epoch": 4.2725013804527885, + "grad_norm": 0.9763281941413879, + "learning_rate": 9.762794787573472e-07, + "loss": 0.5404, + "step": 5161 + }, + { + "epoch": 4.273329652125898, + "grad_norm": 1.1138204336166382, + "learning_rate": 9.754077978003896e-07, + "loss": 0.5464, + "step": 5162 + }, + { + "epoch": 4.274157923799006, + "grad_norm": 0.9644905924797058, + "learning_rate": 9.745364118378117e-07, + "loss": 0.5496, + "step": 5163 + }, + { + "epoch": 4.274986195472115, + "grad_norm": 1.0430723428726196, + "learning_rate": 9.736653210382175e-07, + "loss": 0.5508, + "step": 5164 + }, + { + "epoch": 4.275814467145223, + "grad_norm": 0.9571973085403442, + "learning_rate": 9.727945255701552e-07, + "loss": 0.5409, + "step": 5165 + }, + { + "epoch": 4.276642738818333, + "grad_norm": 0.9848223328590393, + "learning_rate": 9.719240256021165e-07, + "loss": 0.5779, + "step": 5166 + }, + { + "epoch": 4.277471010491441, + "grad_norm": 0.9693152904510498, + "learning_rate": 9.710538213025333e-07, + "loss": 0.5393, + "step": 5167 + }, + { + "epoch": 4.27829928216455, + "grad_norm": 0.9640884399414062, + "learning_rate": 9.701839128397842e-07, + "loss": 0.5684, + "step": 5168 + }, + { + "epoch": 4.279127553837659, + "grad_norm": 0.9764716625213623, + "learning_rate": 9.69314300382187e-07, + "loss": 0.5506, + "step": 5169 + }, + { + "epoch": 4.2799558255107675, + "grad_norm": 0.9657016396522522, + "learning_rate": 9.684449840980032e-07, + "loss": 0.5467, + "step": 5170 + }, + { + "epoch": 4.280784097183877, + "grad_norm": 0.9599623680114746, + "learning_rate": 9.675759641554376e-07, + "loss": 0.5491, + "step": 5171 + }, + { + "epoch": 4.281612368856985, + "grad_norm": 0.9276195764541626, + "learning_rate": 9.667072407226392e-07, + "loss": 0.5452, + "step": 5172 + }, + { + "epoch": 4.282440640530094, + "grad_norm": 0.9440439343452454, + "learning_rate": 9.658388139676953e-07, + "loss": 0.5467, + "step": 5173 + }, + { + "epoch": 4.283268912203202, + "grad_norm": 1.0000046491622925, + "learning_rate": 9.649706840586397e-07, + "loss": 0.5472, + "step": 5174 + }, + { + "epoch": 4.2840971838763116, + "grad_norm": 0.9782945513725281, + "learning_rate": 9.641028511634484e-07, + "loss": 0.5846, + "step": 5175 + }, + { + "epoch": 4.28492545554942, + "grad_norm": 0.9756081700325012, + "learning_rate": 9.632353154500371e-07, + "loss": 0.5539, + "step": 5176 + }, + { + "epoch": 4.285753727222529, + "grad_norm": 0.9515950083732605, + "learning_rate": 9.623680770862665e-07, + "loss": 0.5766, + "step": 5177 + }, + { + "epoch": 4.286581998895638, + "grad_norm": 0.9987873435020447, + "learning_rate": 9.615011362399404e-07, + "loss": 0.5608, + "step": 5178 + }, + { + "epoch": 4.2874102705687465, + "grad_norm": 0.988653838634491, + "learning_rate": 9.606344930788016e-07, + "loss": 0.5565, + "step": 5179 + }, + { + "epoch": 4.288238542241856, + "grad_norm": 0.9355522394180298, + "learning_rate": 9.597681477705398e-07, + "loss": 0.5653, + "step": 5180 + }, + { + "epoch": 4.289066813914964, + "grad_norm": 0.9759084582328796, + "learning_rate": 9.589021004827834e-07, + "loss": 0.5759, + "step": 5181 + }, + { + "epoch": 4.289895085588073, + "grad_norm": 0.9520936012268066, + "learning_rate": 9.58036351383104e-07, + "loss": 0.566, + "step": 5182 + }, + { + "epoch": 4.290723357261181, + "grad_norm": 0.9801774024963379, + "learning_rate": 9.571709006390163e-07, + "loss": 0.547, + "step": 5183 + }, + { + "epoch": 4.2915516289342905, + "grad_norm": 0.9738385081291199, + "learning_rate": 9.56305748417978e-07, + "loss": 0.5476, + "step": 5184 + }, + { + "epoch": 4.292379900607399, + "grad_norm": 1.0364208221435547, + "learning_rate": 9.554408948873868e-07, + "loss": 0.5596, + "step": 5185 + }, + { + "epoch": 4.293208172280508, + "grad_norm": 1.0218520164489746, + "learning_rate": 9.545763402145838e-07, + "loss": 0.5593, + "step": 5186 + }, + { + "epoch": 4.294036443953617, + "grad_norm": 0.9752948880195618, + "learning_rate": 9.53712084566854e-07, + "loss": 0.5315, + "step": 5187 + }, + { + "epoch": 4.2948647156267254, + "grad_norm": 0.969005286693573, + "learning_rate": 9.528481281114205e-07, + "loss": 0.5601, + "step": 5188 + }, + { + "epoch": 4.295692987299835, + "grad_norm": 0.9498629570007324, + "learning_rate": 9.519844710154518e-07, + "loss": 0.554, + "step": 5189 + }, + { + "epoch": 4.296521258972943, + "grad_norm": 0.9451913237571716, + "learning_rate": 9.511211134460585e-07, + "loss": 0.5548, + "step": 5190 + }, + { + "epoch": 4.297349530646052, + "grad_norm": 0.9663949608802795, + "learning_rate": 9.502580555702911e-07, + "loss": 0.5296, + "step": 5191 + }, + { + "epoch": 4.29817780231916, + "grad_norm": 0.9395544528961182, + "learning_rate": 9.493952975551427e-07, + "loss": 0.544, + "step": 5192 + }, + { + "epoch": 4.2990060739922695, + "grad_norm": 0.9444777965545654, + "learning_rate": 9.485328395675497e-07, + "loss": 0.5438, + "step": 5193 + }, + { + "epoch": 4.299834345665378, + "grad_norm": 0.9567984938621521, + "learning_rate": 9.476706817743906e-07, + "loss": 0.5497, + "step": 5194 + }, + { + "epoch": 4.300662617338487, + "grad_norm": 0.967971682548523, + "learning_rate": 9.468088243424828e-07, + "loss": 0.5374, + "step": 5195 + }, + { + "epoch": 4.301490889011596, + "grad_norm": 0.9735593199729919, + "learning_rate": 9.459472674385889e-07, + "loss": 0.5743, + "step": 5196 + }, + { + "epoch": 4.302319160684704, + "grad_norm": 0.9545936584472656, + "learning_rate": 9.45086011229413e-07, + "loss": 0.563, + "step": 5197 + }, + { + "epoch": 4.303147432357814, + "grad_norm": 0.9573048949241638, + "learning_rate": 9.442250558815983e-07, + "loss": 0.5456, + "step": 5198 + }, + { + "epoch": 4.303975704030922, + "grad_norm": 0.9941751956939697, + "learning_rate": 9.433644015617327e-07, + "loss": 0.5541, + "step": 5199 + }, + { + "epoch": 4.304803975704031, + "grad_norm": 0.9913632869720459, + "learning_rate": 9.425040484363454e-07, + "loss": 0.5809, + "step": 5200 + }, + { + "epoch": 4.305632247377139, + "grad_norm": 1.0065845251083374, + "learning_rate": 9.416439966719049e-07, + "loss": 0.5579, + "step": 5201 + }, + { + "epoch": 4.3064605190502485, + "grad_norm": 0.9607630968093872, + "learning_rate": 9.407842464348243e-07, + "loss": 0.5579, + "step": 5202 + }, + { + "epoch": 4.307288790723357, + "grad_norm": 0.9504464864730835, + "learning_rate": 9.399247978914591e-07, + "loss": 0.5479, + "step": 5203 + }, + { + "epoch": 4.308117062396466, + "grad_norm": 0.9630054235458374, + "learning_rate": 9.390656512081009e-07, + "loss": 0.5416, + "step": 5204 + }, + { + "epoch": 4.308945334069575, + "grad_norm": 0.9475155472755432, + "learning_rate": 9.382068065509884e-07, + "loss": 0.541, + "step": 5205 + }, + { + "epoch": 4.309773605742683, + "grad_norm": 0.9980082511901855, + "learning_rate": 9.373482640863004e-07, + "loss": 0.5369, + "step": 5206 + }, + { + "epoch": 4.310601877415793, + "grad_norm": 0.9459148049354553, + "learning_rate": 9.364900239801561e-07, + "loss": 0.545, + "step": 5207 + }, + { + "epoch": 4.311430149088901, + "grad_norm": 0.9700430631637573, + "learning_rate": 9.356320863986171e-07, + "loss": 0.5628, + "step": 5208 + }, + { + "epoch": 4.31225842076201, + "grad_norm": 0.9763196110725403, + "learning_rate": 9.347744515076873e-07, + "loss": 0.5602, + "step": 5209 + }, + { + "epoch": 4.313086692435118, + "grad_norm": 0.97758948802948, + "learning_rate": 9.339171194733096e-07, + "loss": 0.5627, + "step": 5210 + }, + { + "epoch": 4.3139149641082275, + "grad_norm": 0.9904363751411438, + "learning_rate": 9.330600904613705e-07, + "loss": 0.5514, + "step": 5211 + }, + { + "epoch": 4.314743235781337, + "grad_norm": 0.9707289338111877, + "learning_rate": 9.322033646376977e-07, + "loss": 0.5391, + "step": 5212 + }, + { + "epoch": 4.315571507454445, + "grad_norm": 1.0019526481628418, + "learning_rate": 9.313469421680582e-07, + "loss": 0.5563, + "step": 5213 + }, + { + "epoch": 4.316399779127554, + "grad_norm": 0.9674393534660339, + "learning_rate": 9.304908232181636e-07, + "loss": 0.5495, + "step": 5214 + }, + { + "epoch": 4.317228050800662, + "grad_norm": 0.9623247385025024, + "learning_rate": 9.296350079536629e-07, + "loss": 0.5639, + "step": 5215 + }, + { + "epoch": 4.318056322473772, + "grad_norm": 0.9826862215995789, + "learning_rate": 9.287794965401503e-07, + "loss": 0.5714, + "step": 5216 + }, + { + "epoch": 4.31888459414688, + "grad_norm": 0.9826351404190063, + "learning_rate": 9.279242891431575e-07, + "loss": 0.5342, + "step": 5217 + }, + { + "epoch": 4.319712865819989, + "grad_norm": 0.9831173419952393, + "learning_rate": 9.270693859281602e-07, + "loss": 0.5801, + "step": 5218 + }, + { + "epoch": 4.320541137493098, + "grad_norm": 1.0009872913360596, + "learning_rate": 9.262147870605748e-07, + "loss": 0.5817, + "step": 5219 + }, + { + "epoch": 4.3213694091662065, + "grad_norm": 0.962291419506073, + "learning_rate": 9.253604927057566e-07, + "loss": 0.53, + "step": 5220 + }, + { + "epoch": 4.322197680839316, + "grad_norm": 0.9815107583999634, + "learning_rate": 9.245065030290046e-07, + "loss": 0.5706, + "step": 5221 + }, + { + "epoch": 4.323025952512424, + "grad_norm": 0.9885838031768799, + "learning_rate": 9.236528181955584e-07, + "loss": 0.5503, + "step": 5222 + }, + { + "epoch": 4.323854224185533, + "grad_norm": 0.9858911633491516, + "learning_rate": 9.227994383705966e-07, + "loss": 0.5493, + "step": 5223 + }, + { + "epoch": 4.324682495858641, + "grad_norm": 1.027745008468628, + "learning_rate": 9.21946363719241e-07, + "loss": 0.5489, + "step": 5224 + }, + { + "epoch": 4.325510767531751, + "grad_norm": 0.9571676850318909, + "learning_rate": 9.210935944065552e-07, + "loss": 0.5333, + "step": 5225 + }, + { + "epoch": 4.326339039204859, + "grad_norm": 0.9350682497024536, + "learning_rate": 9.20241130597539e-07, + "loss": 0.5457, + "step": 5226 + }, + { + "epoch": 4.327167310877968, + "grad_norm": 0.9624955058097839, + "learning_rate": 9.193889724571375e-07, + "loss": 0.5396, + "step": 5227 + }, + { + "epoch": 4.327995582551077, + "grad_norm": 0.9392304420471191, + "learning_rate": 9.185371201502366e-07, + "loss": 0.5403, + "step": 5228 + }, + { + "epoch": 4.3288238542241855, + "grad_norm": 0.9604929089546204, + "learning_rate": 9.176855738416596e-07, + "loss": 0.5437, + "step": 5229 + }, + { + "epoch": 4.329652125897295, + "grad_norm": 0.9560455679893494, + "learning_rate": 9.168343336961741e-07, + "loss": 0.5484, + "step": 5230 + }, + { + "epoch": 4.330480397570403, + "grad_norm": 0.9542271494865417, + "learning_rate": 9.159833998784879e-07, + "loss": 0.5657, + "step": 5231 + }, + { + "epoch": 4.331308669243512, + "grad_norm": 0.944553554058075, + "learning_rate": 9.151327725532469e-07, + "loss": 0.5576, + "step": 5232 + }, + { + "epoch": 4.33213694091662, + "grad_norm": 0.9710668325424194, + "learning_rate": 9.142824518850402e-07, + "loss": 0.5459, + "step": 5233 + }, + { + "epoch": 4.3329652125897296, + "grad_norm": 0.9870629906654358, + "learning_rate": 9.134324380383983e-07, + "loss": 0.5521, + "step": 5234 + }, + { + "epoch": 4.333793484262838, + "grad_norm": 0.9679974913597107, + "learning_rate": 9.12582731177789e-07, + "loss": 0.5673, + "step": 5235 + }, + { + "epoch": 4.334621755935947, + "grad_norm": 0.9391168355941772, + "learning_rate": 9.117333314676233e-07, + "loss": 0.5638, + "step": 5236 + }, + { + "epoch": 4.335450027609056, + "grad_norm": 0.9927806854248047, + "learning_rate": 9.108842390722533e-07, + "loss": 0.5576, + "step": 5237 + }, + { + "epoch": 4.3362782992821645, + "grad_norm": 0.9426097273826599, + "learning_rate": 9.100354541559697e-07, + "loss": 0.5255, + "step": 5238 + }, + { + "epoch": 4.337106570955274, + "grad_norm": 0.9421987533569336, + "learning_rate": 9.09186976883003e-07, + "loss": 0.5463, + "step": 5239 + }, + { + "epoch": 4.337934842628382, + "grad_norm": 0.9485943913459778, + "learning_rate": 9.083388074175273e-07, + "loss": 0.5414, + "step": 5240 + }, + { + "epoch": 4.338763114301491, + "grad_norm": 0.9566770195960999, + "learning_rate": 9.074909459236558e-07, + "loss": 0.5695, + "step": 5241 + }, + { + "epoch": 4.339591385974599, + "grad_norm": 0.9460729360580444, + "learning_rate": 9.066433925654402e-07, + "loss": 0.5618, + "step": 5242 + }, + { + "epoch": 4.3404196576477085, + "grad_norm": 0.9430888295173645, + "learning_rate": 9.057961475068758e-07, + "loss": 0.5538, + "step": 5243 + }, + { + "epoch": 4.341247929320817, + "grad_norm": 0.9726897478103638, + "learning_rate": 9.049492109118951e-07, + "loss": 0.5724, + "step": 5244 + }, + { + "epoch": 4.342076200993926, + "grad_norm": 0.9237701296806335, + "learning_rate": 9.041025829443733e-07, + "loss": 0.5217, + "step": 5245 + }, + { + "epoch": 4.342904472667035, + "grad_norm": 0.9615205526351929, + "learning_rate": 9.032562637681258e-07, + "loss": 0.5401, + "step": 5246 + }, + { + "epoch": 4.3437327443401434, + "grad_norm": 0.9656977653503418, + "learning_rate": 9.024102535469056e-07, + "loss": 0.5558, + "step": 5247 + }, + { + "epoch": 4.344561016013253, + "grad_norm": 0.9590947031974792, + "learning_rate": 9.015645524444097e-07, + "loss": 0.5606, + "step": 5248 + }, + { + "epoch": 4.345389287686361, + "grad_norm": 0.9751269817352295, + "learning_rate": 9.007191606242718e-07, + "loss": 0.5401, + "step": 5249 + }, + { + "epoch": 4.34621755935947, + "grad_norm": 1.0034822225570679, + "learning_rate": 8.998740782500686e-07, + "loss": 0.5426, + "step": 5250 + }, + { + "epoch": 4.347045831032578, + "grad_norm": 0.96883225440979, + "learning_rate": 8.990293054853144e-07, + "loss": 0.544, + "step": 5251 + }, + { + "epoch": 4.3478741027056875, + "grad_norm": 0.998399555683136, + "learning_rate": 8.981848424934655e-07, + "loss": 0.5309, + "step": 5252 + }, + { + "epoch": 4.348702374378796, + "grad_norm": 0.963323712348938, + "learning_rate": 8.973406894379183e-07, + "loss": 0.5821, + "step": 5253 + }, + { + "epoch": 4.349530646051905, + "grad_norm": 0.9739964008331299, + "learning_rate": 8.964968464820073e-07, + "loss": 0.5624, + "step": 5254 + }, + { + "epoch": 4.350358917725014, + "grad_norm": 0.9921715259552002, + "learning_rate": 8.956533137890089e-07, + "loss": 0.5368, + "step": 5255 + }, + { + "epoch": 4.351187189398122, + "grad_norm": 0.9644178748130798, + "learning_rate": 8.948100915221394e-07, + "loss": 0.5481, + "step": 5256 + }, + { + "epoch": 4.352015461071232, + "grad_norm": 0.9639922380447388, + "learning_rate": 8.939671798445529e-07, + "loss": 0.555, + "step": 5257 + }, + { + "epoch": 4.35284373274434, + "grad_norm": 1.0016719102859497, + "learning_rate": 8.93124578919346e-07, + "loss": 0.5438, + "step": 5258 + }, + { + "epoch": 4.353672004417449, + "grad_norm": 0.9898110032081604, + "learning_rate": 8.922822889095556e-07, + "loss": 0.54, + "step": 5259 + }, + { + "epoch": 4.354500276090557, + "grad_norm": 0.9675293564796448, + "learning_rate": 8.914403099781538e-07, + "loss": 0.5664, + "step": 5260 + }, + { + "epoch": 4.3553285477636665, + "grad_norm": 0.9578870534896851, + "learning_rate": 8.905986422880569e-07, + "loss": 0.5553, + "step": 5261 + }, + { + "epoch": 4.356156819436775, + "grad_norm": 0.9734019041061401, + "learning_rate": 8.897572860021212e-07, + "loss": 0.5503, + "step": 5262 + }, + { + "epoch": 4.356985091109884, + "grad_norm": 0.9673689007759094, + "learning_rate": 8.889162412831392e-07, + "loss": 0.5735, + "step": 5263 + }, + { + "epoch": 4.357813362782993, + "grad_norm": 0.9523547887802124, + "learning_rate": 8.880755082938463e-07, + "loss": 0.5446, + "step": 5264 + }, + { + "epoch": 4.358641634456101, + "grad_norm": 0.9520670175552368, + "learning_rate": 8.872350871969171e-07, + "loss": 0.5503, + "step": 5265 + }, + { + "epoch": 4.359469906129211, + "grad_norm": 0.9825593829154968, + "learning_rate": 8.863949781549638e-07, + "loss": 0.5372, + "step": 5266 + }, + { + "epoch": 4.360298177802319, + "grad_norm": 0.9473084211349487, + "learning_rate": 8.855551813305402e-07, + "loss": 0.5435, + "step": 5267 + }, + { + "epoch": 4.361126449475428, + "grad_norm": 0.9561483263969421, + "learning_rate": 8.847156968861403e-07, + "loss": 0.5577, + "step": 5268 + }, + { + "epoch": 4.361954721148536, + "grad_norm": 0.9839791059494019, + "learning_rate": 8.838765249841949e-07, + "loss": 0.5625, + "step": 5269 + }, + { + "epoch": 4.3627829928216455, + "grad_norm": 0.9682125449180603, + "learning_rate": 8.830376657870762e-07, + "loss": 0.5412, + "step": 5270 + }, + { + "epoch": 4.363611264494755, + "grad_norm": 0.9834880232810974, + "learning_rate": 8.821991194570973e-07, + "loss": 0.5572, + "step": 5271 + }, + { + "epoch": 4.364439536167863, + "grad_norm": 0.9266020059585571, + "learning_rate": 8.813608861565076e-07, + "loss": 0.5649, + "step": 5272 + }, + { + "epoch": 4.365267807840972, + "grad_norm": 0.9482194185256958, + "learning_rate": 8.805229660474968e-07, + "loss": 0.5486, + "step": 5273 + }, + { + "epoch": 4.36609607951408, + "grad_norm": 0.9416828751564026, + "learning_rate": 8.796853592921955e-07, + "loss": 0.5481, + "step": 5274 + }, + { + "epoch": 4.36692435118719, + "grad_norm": 0.9814063906669617, + "learning_rate": 8.788480660526738e-07, + "loss": 0.5424, + "step": 5275 + }, + { + "epoch": 4.367752622860298, + "grad_norm": 0.9609609842300415, + "learning_rate": 8.780110864909383e-07, + "loss": 0.5492, + "step": 5276 + }, + { + "epoch": 4.368580894533407, + "grad_norm": 1.005044937133789, + "learning_rate": 8.771744207689373e-07, + "loss": 0.5459, + "step": 5277 + }, + { + "epoch": 4.369409166206516, + "grad_norm": 0.9385774731636047, + "learning_rate": 8.763380690485592e-07, + "loss": 0.5755, + "step": 5278 + }, + { + "epoch": 4.3702374378796245, + "grad_norm": 0.972374439239502, + "learning_rate": 8.755020314916282e-07, + "loss": 0.5451, + "step": 5279 + }, + { + "epoch": 4.371065709552734, + "grad_norm": 0.9493237137794495, + "learning_rate": 8.746663082599108e-07, + "loss": 0.5385, + "step": 5280 + }, + { + "epoch": 4.371893981225842, + "grad_norm": 0.9911744594573975, + "learning_rate": 8.738308995151124e-07, + "loss": 0.5692, + "step": 5281 + }, + { + "epoch": 4.372722252898951, + "grad_norm": 0.9547559022903442, + "learning_rate": 8.729958054188759e-07, + "loss": 0.5427, + "step": 5282 + }, + { + "epoch": 4.373550524572059, + "grad_norm": 0.974426805973053, + "learning_rate": 8.721610261327834e-07, + "loss": 0.5306, + "step": 5283 + }, + { + "epoch": 4.374378796245169, + "grad_norm": 0.9919657707214355, + "learning_rate": 8.713265618183589e-07, + "loss": 0.5438, + "step": 5284 + }, + { + "epoch": 4.375207067918277, + "grad_norm": 0.9478253126144409, + "learning_rate": 8.704924126370615e-07, + "loss": 0.5341, + "step": 5285 + }, + { + "epoch": 4.376035339591386, + "grad_norm": 0.9950957298278809, + "learning_rate": 8.69658578750292e-07, + "loss": 0.568, + "step": 5286 + }, + { + "epoch": 4.376863611264495, + "grad_norm": 0.9636523723602295, + "learning_rate": 8.688250603193907e-07, + "loss": 0.5487, + "step": 5287 + }, + { + "epoch": 4.3776918829376035, + "grad_norm": 0.974831759929657, + "learning_rate": 8.679918575056337e-07, + "loss": 0.5324, + "step": 5288 + }, + { + "epoch": 4.378520154610713, + "grad_norm": 0.9769512414932251, + "learning_rate": 8.671589704702388e-07, + "loss": 0.558, + "step": 5289 + }, + { + "epoch": 4.379348426283821, + "grad_norm": 0.9584938287734985, + "learning_rate": 8.663263993743632e-07, + "loss": 0.5529, + "step": 5290 + }, + { + "epoch": 4.38017669795693, + "grad_norm": 0.9711350798606873, + "learning_rate": 8.65494144379099e-07, + "loss": 0.5558, + "step": 5291 + }, + { + "epoch": 4.381004969630038, + "grad_norm": 0.9583411812782288, + "learning_rate": 8.646622056454815e-07, + "loss": 0.5585, + "step": 5292 + }, + { + "epoch": 4.3818332413031476, + "grad_norm": 0.9572791457176208, + "learning_rate": 8.638305833344834e-07, + "loss": 0.5559, + "step": 5293 + }, + { + "epoch": 4.382661512976256, + "grad_norm": 0.9529664516448975, + "learning_rate": 8.629992776070156e-07, + "loss": 0.5568, + "step": 5294 + }, + { + "epoch": 4.383489784649365, + "grad_norm": 0.9727757573127747, + "learning_rate": 8.621682886239266e-07, + "loss": 0.5604, + "step": 5295 + }, + { + "epoch": 4.384318056322474, + "grad_norm": 0.9406315684318542, + "learning_rate": 8.613376165460061e-07, + "loss": 0.5294, + "step": 5296 + }, + { + "epoch": 4.3851463279955825, + "grad_norm": 0.9377478361129761, + "learning_rate": 8.605072615339824e-07, + "loss": 0.531, + "step": 5297 + }, + { + "epoch": 4.385974599668692, + "grad_norm": 0.9759235382080078, + "learning_rate": 8.596772237485193e-07, + "loss": 0.5461, + "step": 5298 + }, + { + "epoch": 4.3868028713418, + "grad_norm": 0.9580682516098022, + "learning_rate": 8.588475033502228e-07, + "loss": 0.5513, + "step": 5299 + }, + { + "epoch": 4.387631143014909, + "grad_norm": 0.9866207242012024, + "learning_rate": 8.580181004996362e-07, + "loss": 0.5647, + "step": 5300 + }, + { + "epoch": 4.388459414688017, + "grad_norm": 0.9822670817375183, + "learning_rate": 8.571890153572404e-07, + "loss": 0.5608, + "step": 5301 + }, + { + "epoch": 4.3892876863611265, + "grad_norm": 0.9508621096611023, + "learning_rate": 8.563602480834557e-07, + "loss": 0.5297, + "step": 5302 + }, + { + "epoch": 4.390115958034235, + "grad_norm": 0.9849876761436462, + "learning_rate": 8.555317988386422e-07, + "loss": 0.5836, + "step": 5303 + }, + { + "epoch": 4.390944229707344, + "grad_norm": 0.9595887064933777, + "learning_rate": 8.54703667783096e-07, + "loss": 0.5492, + "step": 5304 + }, + { + "epoch": 4.391772501380453, + "grad_norm": 0.9659999012947083, + "learning_rate": 8.53875855077052e-07, + "loss": 0.5245, + "step": 5305 + }, + { + "epoch": 4.3926007730535614, + "grad_norm": 0.9543110132217407, + "learning_rate": 8.530483608806861e-07, + "loss": 0.5379, + "step": 5306 + }, + { + "epoch": 4.393429044726671, + "grad_norm": 0.9603022933006287, + "learning_rate": 8.522211853541087e-07, + "loss": 0.5629, + "step": 5307 + }, + { + "epoch": 4.394257316399779, + "grad_norm": 0.9761829972267151, + "learning_rate": 8.513943286573714e-07, + "loss": 0.5398, + "step": 5308 + }, + { + "epoch": 4.395085588072888, + "grad_norm": 0.9566043615341187, + "learning_rate": 8.505677909504645e-07, + "loss": 0.5671, + "step": 5309 + }, + { + "epoch": 4.395913859745996, + "grad_norm": 0.9672698378562927, + "learning_rate": 8.497415723933133e-07, + "loss": 0.5369, + "step": 5310 + }, + { + "epoch": 4.3967421314191055, + "grad_norm": 0.945254921913147, + "learning_rate": 8.489156731457843e-07, + "loss": 0.5372, + "step": 5311 + }, + { + "epoch": 4.397570403092214, + "grad_norm": 0.9574915170669556, + "learning_rate": 8.480900933676822e-07, + "loss": 0.5583, + "step": 5312 + }, + { + "epoch": 4.398398674765323, + "grad_norm": 0.9455167651176453, + "learning_rate": 8.472648332187471e-07, + "loss": 0.5542, + "step": 5313 + }, + { + "epoch": 4.399226946438432, + "grad_norm": 0.9736841320991516, + "learning_rate": 8.464398928586603e-07, + "loss": 0.5501, + "step": 5314 + }, + { + "epoch": 4.40005521811154, + "grad_norm": 0.9645372629165649, + "learning_rate": 8.456152724470409e-07, + "loss": 0.5785, + "step": 5315 + }, + { + "epoch": 4.40088348978465, + "grad_norm": 0.9789561033248901, + "learning_rate": 8.447909721434441e-07, + "loss": 0.5487, + "step": 5316 + }, + { + "epoch": 4.401711761457758, + "grad_norm": 0.9858975410461426, + "learning_rate": 8.43966992107364e-07, + "loss": 0.5627, + "step": 5317 + }, + { + "epoch": 4.402540033130867, + "grad_norm": 0.9592116475105286, + "learning_rate": 8.431433324982332e-07, + "loss": 0.5741, + "step": 5318 + }, + { + "epoch": 4.403368304803975, + "grad_norm": 0.9792628884315491, + "learning_rate": 8.423199934754237e-07, + "loss": 0.5385, + "step": 5319 + }, + { + "epoch": 4.4041965764770845, + "grad_norm": 0.9497966170310974, + "learning_rate": 8.414969751982419e-07, + "loss": 0.5569, + "step": 5320 + }, + { + "epoch": 4.405024848150193, + "grad_norm": 0.9575057029724121, + "learning_rate": 8.406742778259355e-07, + "loss": 0.534, + "step": 5321 + }, + { + "epoch": 4.405853119823302, + "grad_norm": 0.9563487768173218, + "learning_rate": 8.398519015176879e-07, + "loss": 0.5638, + "step": 5322 + }, + { + "epoch": 4.406681391496411, + "grad_norm": 0.9754065275192261, + "learning_rate": 8.390298464326213e-07, + "loss": 0.5562, + "step": 5323 + }, + { + "epoch": 4.407509663169519, + "grad_norm": 0.963133692741394, + "learning_rate": 8.382081127297972e-07, + "loss": 0.5514, + "step": 5324 + }, + { + "epoch": 4.408337934842629, + "grad_norm": 0.9998530745506287, + "learning_rate": 8.373867005682113e-07, + "loss": 0.5709, + "step": 5325 + }, + { + "epoch": 4.409166206515737, + "grad_norm": 0.9418386220932007, + "learning_rate": 8.365656101068001e-07, + "loss": 0.5449, + "step": 5326 + }, + { + "epoch": 4.409994478188846, + "grad_norm": 0.9713196158409119, + "learning_rate": 8.357448415044378e-07, + "loss": 0.5683, + "step": 5327 + }, + { + "epoch": 4.410822749861954, + "grad_norm": 0.9514254331588745, + "learning_rate": 8.349243949199349e-07, + "loss": 0.5539, + "step": 5328 + }, + { + "epoch": 4.4116510215350635, + "grad_norm": 0.9651909470558167, + "learning_rate": 8.34104270512039e-07, + "loss": 0.5794, + "step": 5329 + }, + { + "epoch": 4.412479293208173, + "grad_norm": 0.9516454339027405, + "learning_rate": 8.332844684394373e-07, + "loss": 0.5617, + "step": 5330 + }, + { + "epoch": 4.413307564881281, + "grad_norm": 1.0173704624176025, + "learning_rate": 8.32464988860755e-07, + "loss": 0.5466, + "step": 5331 + }, + { + "epoch": 4.41413583655439, + "grad_norm": 0.9605720043182373, + "learning_rate": 8.316458319345521e-07, + "loss": 0.5467, + "step": 5332 + }, + { + "epoch": 4.414964108227498, + "grad_norm": 0.9562596678733826, + "learning_rate": 8.308269978193285e-07, + "loss": 0.557, + "step": 5333 + }, + { + "epoch": 4.415792379900608, + "grad_norm": 0.951440691947937, + "learning_rate": 8.300084866735217e-07, + "loss": 0.5452, + "step": 5334 + }, + { + "epoch": 4.416620651573716, + "grad_norm": 0.9654122591018677, + "learning_rate": 8.291902986555045e-07, + "loss": 0.5328, + "step": 5335 + }, + { + "epoch": 4.417448923246825, + "grad_norm": 0.9830499887466431, + "learning_rate": 8.283724339235893e-07, + "loss": 0.5533, + "step": 5336 + }, + { + "epoch": 4.418277194919934, + "grad_norm": 0.9823290109634399, + "learning_rate": 8.275548926360263e-07, + "loss": 0.5422, + "step": 5337 + }, + { + "epoch": 4.4191054665930425, + "grad_norm": 0.9932782053947449, + "learning_rate": 8.267376749510012e-07, + "loss": 0.5666, + "step": 5338 + }, + { + "epoch": 4.419933738266152, + "grad_norm": 0.947321891784668, + "learning_rate": 8.259207810266373e-07, + "loss": 0.5511, + "step": 5339 + }, + { + "epoch": 4.42076200993926, + "grad_norm": 0.9844286441802979, + "learning_rate": 8.251042110209975e-07, + "loss": 0.541, + "step": 5340 + }, + { + "epoch": 4.421590281612369, + "grad_norm": 0.9536518454551697, + "learning_rate": 8.242879650920788e-07, + "loss": 0.561, + "step": 5341 + }, + { + "epoch": 4.422418553285477, + "grad_norm": 0.9690708518028259, + "learning_rate": 8.234720433978183e-07, + "loss": 0.5526, + "step": 5342 + }, + { + "epoch": 4.423246824958587, + "grad_norm": 0.9607888460159302, + "learning_rate": 8.226564460960895e-07, + "loss": 0.5695, + "step": 5343 + }, + { + "epoch": 4.424075096631695, + "grad_norm": 0.9637815356254578, + "learning_rate": 8.218411733447018e-07, + "loss": 0.575, + "step": 5344 + }, + { + "epoch": 4.424903368304804, + "grad_norm": 0.982789158821106, + "learning_rate": 8.210262253014034e-07, + "loss": 0.5437, + "step": 5345 + }, + { + "epoch": 4.425731639977913, + "grad_norm": 0.9953805804252625, + "learning_rate": 8.202116021238801e-07, + "loss": 0.5812, + "step": 5346 + }, + { + "epoch": 4.4265599116510215, + "grad_norm": 0.9684412479400635, + "learning_rate": 8.193973039697519e-07, + "loss": 0.5381, + "step": 5347 + }, + { + "epoch": 4.427388183324131, + "grad_norm": 0.9479103088378906, + "learning_rate": 8.185833309965793e-07, + "loss": 0.5646, + "step": 5348 + }, + { + "epoch": 4.428216454997239, + "grad_norm": 0.9611683487892151, + "learning_rate": 8.17769683361859e-07, + "loss": 0.5568, + "step": 5349 + }, + { + "epoch": 4.429044726670348, + "grad_norm": 0.9750727415084839, + "learning_rate": 8.169563612230233e-07, + "loss": 0.5629, + "step": 5350 + }, + { + "epoch": 4.429872998343456, + "grad_norm": 0.9549238681793213, + "learning_rate": 8.161433647374417e-07, + "loss": 0.5277, + "step": 5351 + }, + { + "epoch": 4.4307012700165656, + "grad_norm": 0.9485230445861816, + "learning_rate": 8.153306940624225e-07, + "loss": 0.5406, + "step": 5352 + }, + { + "epoch": 4.431529541689674, + "grad_norm": 0.9689464569091797, + "learning_rate": 8.145183493552103e-07, + "loss": 0.5563, + "step": 5353 + }, + { + "epoch": 4.432357813362783, + "grad_norm": 0.9542263746261597, + "learning_rate": 8.13706330772985e-07, + "loss": 0.5491, + "step": 5354 + }, + { + "epoch": 4.433186085035892, + "grad_norm": 0.9412059187889099, + "learning_rate": 8.128946384728654e-07, + "loss": 0.5744, + "step": 5355 + }, + { + "epoch": 4.4340143567090005, + "grad_norm": 0.9872552156448364, + "learning_rate": 8.120832726119069e-07, + "loss": 0.5504, + "step": 5356 + }, + { + "epoch": 4.43484262838211, + "grad_norm": 0.9560122489929199, + "learning_rate": 8.112722333470996e-07, + "loss": 0.5451, + "step": 5357 + }, + { + "epoch": 4.435670900055218, + "grad_norm": 0.9743005633354187, + "learning_rate": 8.104615208353731e-07, + "loss": 0.5633, + "step": 5358 + }, + { + "epoch": 4.436499171728327, + "grad_norm": 0.9652897119522095, + "learning_rate": 8.096511352335937e-07, + "loss": 0.5474, + "step": 5359 + }, + { + "epoch": 4.437327443401435, + "grad_norm": 0.9695815443992615, + "learning_rate": 8.088410766985613e-07, + "loss": 0.5522, + "step": 5360 + }, + { + "epoch": 4.4381557150745445, + "grad_norm": 0.9816685318946838, + "learning_rate": 8.080313453870167e-07, + "loss": 0.5472, + "step": 5361 + }, + { + "epoch": 4.438983986747653, + "grad_norm": 0.9512736201286316, + "learning_rate": 8.072219414556345e-07, + "loss": 0.5437, + "step": 5362 + }, + { + "epoch": 4.439812258420762, + "grad_norm": 0.9622536301612854, + "learning_rate": 8.064128650610259e-07, + "loss": 0.5383, + "step": 5363 + }, + { + "epoch": 4.440640530093871, + "grad_norm": 0.9731073379516602, + "learning_rate": 8.056041163597406e-07, + "loss": 0.5777, + "step": 5364 + }, + { + "epoch": 4.4414688017669794, + "grad_norm": 1.0841182470321655, + "learning_rate": 8.047956955082645e-07, + "loss": 0.5542, + "step": 5365 + }, + { + "epoch": 4.442297073440089, + "grad_norm": 0.9848759174346924, + "learning_rate": 8.039876026630181e-07, + "loss": 0.5456, + "step": 5366 + }, + { + "epoch": 4.443125345113197, + "grad_norm": 0.9822127819061279, + "learning_rate": 8.031798379803604e-07, + "loss": 0.5753, + "step": 5367 + }, + { + "epoch": 4.443953616786306, + "grad_norm": 1.094104528427124, + "learning_rate": 8.023724016165873e-07, + "loss": 0.558, + "step": 5368 + }, + { + "epoch": 4.444781888459414, + "grad_norm": 0.9637600779533386, + "learning_rate": 8.015652937279289e-07, + "loss": 0.533, + "step": 5369 + }, + { + "epoch": 4.4456101601325235, + "grad_norm": 0.9773861169815063, + "learning_rate": 8.007585144705532e-07, + "loss": 0.5798, + "step": 5370 + }, + { + "epoch": 4.446438431805632, + "grad_norm": 0.9739058613777161, + "learning_rate": 7.99952064000566e-07, + "loss": 0.563, + "step": 5371 + }, + { + "epoch": 4.447266703478741, + "grad_norm": 0.9480531811714172, + "learning_rate": 7.991459424740064e-07, + "loss": 0.5452, + "step": 5372 + }, + { + "epoch": 4.44809497515185, + "grad_norm": 0.9757310152053833, + "learning_rate": 7.98340150046851e-07, + "loss": 0.5758, + "step": 5373 + }, + { + "epoch": 4.448923246824958, + "grad_norm": 0.9749124646186829, + "learning_rate": 7.975346868750142e-07, + "loss": 0.5413, + "step": 5374 + }, + { + "epoch": 4.449751518498068, + "grad_norm": 0.961652934551239, + "learning_rate": 7.967295531143459e-07, + "loss": 0.5617, + "step": 5375 + }, + { + "epoch": 4.450579790171176, + "grad_norm": 0.9718902111053467, + "learning_rate": 7.959247489206306e-07, + "loss": 0.5496, + "step": 5376 + }, + { + "epoch": 4.451408061844285, + "grad_norm": 0.959939181804657, + "learning_rate": 7.951202744495912e-07, + "loss": 0.5535, + "step": 5377 + }, + { + "epoch": 4.452236333517393, + "grad_norm": 0.9867039918899536, + "learning_rate": 7.943161298568869e-07, + "loss": 0.5542, + "step": 5378 + }, + { + "epoch": 4.4530646051905025, + "grad_norm": 0.9625898003578186, + "learning_rate": 7.935123152981103e-07, + "loss": 0.5565, + "step": 5379 + }, + { + "epoch": 4.453892876863611, + "grad_norm": 0.9669535160064697, + "learning_rate": 7.927088309287931e-07, + "loss": 0.5514, + "step": 5380 + }, + { + "epoch": 4.45472114853672, + "grad_norm": 0.9600343108177185, + "learning_rate": 7.919056769044028e-07, + "loss": 0.5736, + "step": 5381 + }, + { + "epoch": 4.455549420209829, + "grad_norm": 0.9746002554893494, + "learning_rate": 7.911028533803406e-07, + "loss": 0.5438, + "step": 5382 + }, + { + "epoch": 4.456377691882937, + "grad_norm": 0.9626099467277527, + "learning_rate": 7.903003605119469e-07, + "loss": 0.5631, + "step": 5383 + }, + { + "epoch": 4.457205963556047, + "grad_norm": 0.958444356918335, + "learning_rate": 7.894981984544958e-07, + "loss": 0.5527, + "step": 5384 + }, + { + "epoch": 4.458034235229155, + "grad_norm": 1.017724871635437, + "learning_rate": 7.886963673631973e-07, + "loss": 0.5459, + "step": 5385 + }, + { + "epoch": 4.458862506902264, + "grad_norm": 0.9509232044219971, + "learning_rate": 7.878948673931994e-07, + "loss": 0.5478, + "step": 5386 + }, + { + "epoch": 4.459690778575372, + "grad_norm": 1.008229374885559, + "learning_rate": 7.870936986995853e-07, + "loss": 0.5565, + "step": 5387 + }, + { + "epoch": 4.4605190502484815, + "grad_norm": 0.9678364396095276, + "learning_rate": 7.862928614373722e-07, + "loss": 0.5625, + "step": 5388 + }, + { + "epoch": 4.461347321921591, + "grad_norm": 0.9678050875663757, + "learning_rate": 7.854923557615152e-07, + "loss": 0.5556, + "step": 5389 + }, + { + "epoch": 4.462175593594699, + "grad_norm": 0.9574840664863586, + "learning_rate": 7.84692181826906e-07, + "loss": 0.5437, + "step": 5390 + }, + { + "epoch": 4.463003865267808, + "grad_norm": 0.9662474393844604, + "learning_rate": 7.838923397883688e-07, + "loss": 0.5486, + "step": 5391 + }, + { + "epoch": 4.463832136940916, + "grad_norm": 0.9890984296798706, + "learning_rate": 7.830928298006666e-07, + "loss": 0.5411, + "step": 5392 + }, + { + "epoch": 4.464660408614026, + "grad_norm": 0.975991427898407, + "learning_rate": 7.822936520184979e-07, + "loss": 0.5619, + "step": 5393 + }, + { + "epoch": 4.465488680287134, + "grad_norm": 0.9573751091957092, + "learning_rate": 7.814948065964942e-07, + "loss": 0.5609, + "step": 5394 + }, + { + "epoch": 4.466316951960243, + "grad_norm": 0.9584991931915283, + "learning_rate": 7.806962936892268e-07, + "loss": 0.5611, + "step": 5395 + }, + { + "epoch": 4.467145223633352, + "grad_norm": 0.9686765074729919, + "learning_rate": 7.798981134511993e-07, + "loss": 0.5673, + "step": 5396 + }, + { + "epoch": 4.4679734953064605, + "grad_norm": 0.959581732749939, + "learning_rate": 7.791002660368513e-07, + "loss": 0.5536, + "step": 5397 + }, + { + "epoch": 4.46880176697957, + "grad_norm": 0.9748058319091797, + "learning_rate": 7.783027516005601e-07, + "loss": 0.5541, + "step": 5398 + }, + { + "epoch": 4.469630038652678, + "grad_norm": 0.9732323884963989, + "learning_rate": 7.775055702966375e-07, + "loss": 0.5507, + "step": 5399 + }, + { + "epoch": 4.470458310325787, + "grad_norm": 0.9876232147216797, + "learning_rate": 7.767087222793295e-07, + "loss": 0.5486, + "step": 5400 + }, + { + "epoch": 4.471286581998895, + "grad_norm": 0.963202714920044, + "learning_rate": 7.759122077028194e-07, + "loss": 0.5633, + "step": 5401 + }, + { + "epoch": 4.472114853672005, + "grad_norm": 0.9766404628753662, + "learning_rate": 7.751160267212262e-07, + "loss": 0.5662, + "step": 5402 + }, + { + "epoch": 4.472943125345113, + "grad_norm": 0.9662322402000427, + "learning_rate": 7.743201794886021e-07, + "loss": 0.5502, + "step": 5403 + }, + { + "epoch": 4.473771397018222, + "grad_norm": 0.9390942454338074, + "learning_rate": 7.735246661589363e-07, + "loss": 0.5415, + "step": 5404 + }, + { + "epoch": 4.474599668691331, + "grad_norm": 0.9523019194602966, + "learning_rate": 7.727294868861546e-07, + "loss": 0.5482, + "step": 5405 + }, + { + "epoch": 4.4754279403644395, + "grad_norm": 0.9844672083854675, + "learning_rate": 7.719346418241158e-07, + "loss": 0.5576, + "step": 5406 + }, + { + "epoch": 4.476256212037549, + "grad_norm": 0.9560808539390564, + "learning_rate": 7.711401311266143e-07, + "loss": 0.5595, + "step": 5407 + }, + { + "epoch": 4.477084483710657, + "grad_norm": 1.1302635669708252, + "learning_rate": 7.70345954947381e-07, + "loss": 0.5524, + "step": 5408 + }, + { + "epoch": 4.477912755383766, + "grad_norm": 1.0062146186828613, + "learning_rate": 7.695521134400824e-07, + "loss": 0.5599, + "step": 5409 + }, + { + "epoch": 4.478741027056874, + "grad_norm": 0.9748486876487732, + "learning_rate": 7.687586067583184e-07, + "loss": 0.5543, + "step": 5410 + }, + { + "epoch": 4.4795692987299836, + "grad_norm": 0.9799855351448059, + "learning_rate": 7.679654350556254e-07, + "loss": 0.5529, + "step": 5411 + }, + { + "epoch": 4.480397570403092, + "grad_norm": 1.004991888999939, + "learning_rate": 7.671725984854755e-07, + "loss": 0.5644, + "step": 5412 + }, + { + "epoch": 4.481225842076201, + "grad_norm": 0.9852890968322754, + "learning_rate": 7.663800972012742e-07, + "loss": 0.5555, + "step": 5413 + }, + { + "epoch": 4.48205411374931, + "grad_norm": 0.9725239872932434, + "learning_rate": 7.655879313563633e-07, + "loss": 0.5464, + "step": 5414 + }, + { + "epoch": 4.4828823854224185, + "grad_norm": 1.0972882509231567, + "learning_rate": 7.647961011040203e-07, + "loss": 0.5434, + "step": 5415 + }, + { + "epoch": 4.483710657095528, + "grad_norm": 0.9756212830543518, + "learning_rate": 7.640046065974557e-07, + "loss": 0.5448, + "step": 5416 + }, + { + "epoch": 4.484538928768636, + "grad_norm": 0.9724202156066895, + "learning_rate": 7.632134479898178e-07, + "loss": 0.5635, + "step": 5417 + }, + { + "epoch": 4.485367200441745, + "grad_norm": 0.9435184001922607, + "learning_rate": 7.624226254341879e-07, + "loss": 0.546, + "step": 5418 + }, + { + "epoch": 4.486195472114853, + "grad_norm": 0.9800364375114441, + "learning_rate": 7.616321390835812e-07, + "loss": 0.5463, + "step": 5419 + }, + { + "epoch": 4.4870237437879625, + "grad_norm": 0.9564050436019897, + "learning_rate": 7.608419890909511e-07, + "loss": 0.5391, + "step": 5420 + }, + { + "epoch": 4.487852015461071, + "grad_norm": 0.9649475812911987, + "learning_rate": 7.600521756091844e-07, + "loss": 0.5421, + "step": 5421 + }, + { + "epoch": 4.48868028713418, + "grad_norm": 0.9621299505233765, + "learning_rate": 7.592626987911014e-07, + "loss": 0.5669, + "step": 5422 + }, + { + "epoch": 4.489508558807289, + "grad_norm": 0.9605808854103088, + "learning_rate": 7.584735587894595e-07, + "loss": 0.5658, + "step": 5423 + }, + { + "epoch": 4.4903368304803974, + "grad_norm": 0.9706793427467346, + "learning_rate": 7.576847557569503e-07, + "loss": 0.5629, + "step": 5424 + }, + { + "epoch": 4.491165102153507, + "grad_norm": 0.9814993143081665, + "learning_rate": 7.568962898461982e-07, + "loss": 0.534, + "step": 5425 + }, + { + "epoch": 4.491993373826615, + "grad_norm": 0.9362326264381409, + "learning_rate": 7.561081612097654e-07, + "loss": 0.5428, + "step": 5426 + }, + { + "epoch": 4.492821645499724, + "grad_norm": 0.9745808243751526, + "learning_rate": 7.553203700001474e-07, + "loss": 0.5524, + "step": 5427 + }, + { + "epoch": 4.493649917172832, + "grad_norm": 0.9991053938865662, + "learning_rate": 7.545329163697737e-07, + "loss": 0.5646, + "step": 5428 + }, + { + "epoch": 4.4944781888459415, + "grad_norm": 1.0213414430618286, + "learning_rate": 7.537458004710102e-07, + "loss": 0.5481, + "step": 5429 + }, + { + "epoch": 4.49530646051905, + "grad_norm": 0.9790488481521606, + "learning_rate": 7.529590224561553e-07, + "loss": 0.5779, + "step": 5430 + }, + { + "epoch": 4.496134732192159, + "grad_norm": 0.9708014130592346, + "learning_rate": 7.521725824774446e-07, + "loss": 0.5319, + "step": 5431 + }, + { + "epoch": 4.496963003865268, + "grad_norm": 0.9768827557563782, + "learning_rate": 7.513864806870453e-07, + "loss": 0.5468, + "step": 5432 + }, + { + "epoch": 4.497791275538376, + "grad_norm": 0.9755817651748657, + "learning_rate": 7.506007172370617e-07, + "loss": 0.5634, + "step": 5433 + }, + { + "epoch": 4.498619547211486, + "grad_norm": 0.9696100354194641, + "learning_rate": 7.498152922795324e-07, + "loss": 0.5582, + "step": 5434 + }, + { + "epoch": 4.499447818884594, + "grad_norm": 0.9456085562705994, + "learning_rate": 7.490302059664281e-07, + "loss": 0.5484, + "step": 5435 + }, + { + "epoch": 4.500276090557703, + "grad_norm": 0.9718632102012634, + "learning_rate": 7.482454584496568e-07, + "loss": 0.5486, + "step": 5436 + }, + { + "epoch": 4.501104362230811, + "grad_norm": 0.9685472249984741, + "learning_rate": 7.474610498810603e-07, + "loss": 0.5474, + "step": 5437 + }, + { + "epoch": 4.5019326339039205, + "grad_norm": 0.9661766886711121, + "learning_rate": 7.466769804124127e-07, + "loss": 0.5565, + "step": 5438 + }, + { + "epoch": 4.502760905577029, + "grad_norm": 0.9711228013038635, + "learning_rate": 7.458932501954255e-07, + "loss": 0.5381, + "step": 5439 + }, + { + "epoch": 4.503589177250138, + "grad_norm": 0.9743756651878357, + "learning_rate": 7.451098593817444e-07, + "loss": 0.5748, + "step": 5440 + }, + { + "epoch": 4.504417448923247, + "grad_norm": 0.9695295691490173, + "learning_rate": 7.443268081229449e-07, + "loss": 0.5378, + "step": 5441 + }, + { + "epoch": 4.505245720596355, + "grad_norm": 0.9656387567520142, + "learning_rate": 7.435440965705417e-07, + "loss": 0.5512, + "step": 5442 + }, + { + "epoch": 4.506073992269465, + "grad_norm": 0.9670572280883789, + "learning_rate": 7.427617248759833e-07, + "loss": 0.556, + "step": 5443 + }, + { + "epoch": 4.506902263942573, + "grad_norm": 0.9564972519874573, + "learning_rate": 7.419796931906495e-07, + "loss": 0.5607, + "step": 5444 + }, + { + "epoch": 4.507730535615682, + "grad_norm": 0.9703754186630249, + "learning_rate": 7.411980016658571e-07, + "loss": 0.5441, + "step": 5445 + }, + { + "epoch": 4.508558807288791, + "grad_norm": 0.9711794853210449, + "learning_rate": 7.404166504528567e-07, + "loss": 0.5814, + "step": 5446 + }, + { + "epoch": 4.5093870789618995, + "grad_norm": 0.9656071066856384, + "learning_rate": 7.396356397028312e-07, + "loss": 0.5602, + "step": 5447 + }, + { + "epoch": 4.510215350635009, + "grad_norm": 1.0016793012619019, + "learning_rate": 7.388549695668995e-07, + "loss": 0.5596, + "step": 5448 + }, + { + "epoch": 4.511043622308117, + "grad_norm": 1.0137665271759033, + "learning_rate": 7.380746401961148e-07, + "loss": 0.5636, + "step": 5449 + }, + { + "epoch": 4.511871893981226, + "grad_norm": 0.9805740118026733, + "learning_rate": 7.372946517414617e-07, + "loss": 0.5566, + "step": 5450 + }, + { + "epoch": 4.512700165654334, + "grad_norm": 0.9614541530609131, + "learning_rate": 7.36515004353863e-07, + "loss": 0.5648, + "step": 5451 + }, + { + "epoch": 4.513528437327444, + "grad_norm": 0.9508614540100098, + "learning_rate": 7.357356981841707e-07, + "loss": 0.5227, + "step": 5452 + }, + { + "epoch": 4.514356709000552, + "grad_norm": 0.9625178575515747, + "learning_rate": 7.349567333831756e-07, + "loss": 0.5543, + "step": 5453 + }, + { + "epoch": 4.515184980673661, + "grad_norm": 0.9476574063301086, + "learning_rate": 7.341781101015982e-07, + "loss": 0.5722, + "step": 5454 + }, + { + "epoch": 4.51601325234677, + "grad_norm": 0.9598817229270935, + "learning_rate": 7.333998284900959e-07, + "loss": 0.5511, + "step": 5455 + }, + { + "epoch": 4.5168415240198785, + "grad_norm": 0.9833248257637024, + "learning_rate": 7.326218886992592e-07, + "loss": 0.5399, + "step": 5456 + }, + { + "epoch": 4.517669795692988, + "grad_norm": 0.9664196372032166, + "learning_rate": 7.318442908796111e-07, + "loss": 0.5452, + "step": 5457 + }, + { + "epoch": 4.518498067366096, + "grad_norm": 0.9677484035491943, + "learning_rate": 7.31067035181611e-07, + "loss": 0.5477, + "step": 5458 + }, + { + "epoch": 4.519326339039205, + "grad_norm": 0.9622910618782043, + "learning_rate": 7.302901217556491e-07, + "loss": 0.5396, + "step": 5459 + }, + { + "epoch": 4.520154610712313, + "grad_norm": 0.9736341238021851, + "learning_rate": 7.295135507520515e-07, + "loss": 0.564, + "step": 5460 + }, + { + "epoch": 4.520982882385423, + "grad_norm": 0.9707585573196411, + "learning_rate": 7.287373223210784e-07, + "loss": 0.571, + "step": 5461 + }, + { + "epoch": 4.521811154058531, + "grad_norm": 0.9776362776756287, + "learning_rate": 7.279614366129212e-07, + "loss": 0.5601, + "step": 5462 + }, + { + "epoch": 4.52263942573164, + "grad_norm": 0.9470260143280029, + "learning_rate": 7.271858937777079e-07, + "loss": 0.5559, + "step": 5463 + }, + { + "epoch": 4.523467697404749, + "grad_norm": 0.9702407121658325, + "learning_rate": 7.264106939654975e-07, + "loss": 0.556, + "step": 5464 + }, + { + "epoch": 4.5242959690778575, + "grad_norm": 0.9597034454345703, + "learning_rate": 7.256358373262856e-07, + "loss": 0.5383, + "step": 5465 + }, + { + "epoch": 4.525124240750967, + "grad_norm": 0.9662848114967346, + "learning_rate": 7.248613240099978e-07, + "loss": 0.5414, + "step": 5466 + }, + { + "epoch": 4.525952512424075, + "grad_norm": 0.993077278137207, + "learning_rate": 7.240871541664959e-07, + "loss": 0.554, + "step": 5467 + }, + { + "epoch": 4.526780784097184, + "grad_norm": 0.9801212549209595, + "learning_rate": 7.233133279455762e-07, + "loss": 0.5546, + "step": 5468 + }, + { + "epoch": 4.527609055770292, + "grad_norm": 0.9710840582847595, + "learning_rate": 7.225398454969643e-07, + "loss": 0.5506, + "step": 5469 + }, + { + "epoch": 4.5284373274434015, + "grad_norm": 1.0106468200683594, + "learning_rate": 7.21766706970323e-07, + "loss": 0.5908, + "step": 5470 + }, + { + "epoch": 4.52926559911651, + "grad_norm": 0.9547988176345825, + "learning_rate": 7.209939125152482e-07, + "loss": 0.5497, + "step": 5471 + }, + { + "epoch": 4.530093870789619, + "grad_norm": 0.9706175923347473, + "learning_rate": 7.20221462281267e-07, + "loss": 0.5431, + "step": 5472 + }, + { + "epoch": 4.530922142462728, + "grad_norm": 0.9547169804573059, + "learning_rate": 7.19449356417842e-07, + "loss": 0.5584, + "step": 5473 + }, + { + "epoch": 4.5317504141358365, + "grad_norm": 0.9862712621688843, + "learning_rate": 7.1867759507437e-07, + "loss": 0.5695, + "step": 5474 + }, + { + "epoch": 4.532578685808946, + "grad_norm": 0.9707930088043213, + "learning_rate": 7.179061784001765e-07, + "loss": 0.5775, + "step": 5475 + }, + { + "epoch": 4.533406957482054, + "grad_norm": 0.957129716873169, + "learning_rate": 7.171351065445248e-07, + "loss": 0.5423, + "step": 5476 + }, + { + "epoch": 4.534235229155163, + "grad_norm": 0.965663492679596, + "learning_rate": 7.163643796566114e-07, + "loss": 0.5407, + "step": 5477 + }, + { + "epoch": 4.535063500828271, + "grad_norm": 0.940671980381012, + "learning_rate": 7.155939978855628e-07, + "loss": 0.5352, + "step": 5478 + }, + { + "epoch": 4.5358917725013805, + "grad_norm": 0.9513495564460754, + "learning_rate": 7.148239613804414e-07, + "loss": 0.5542, + "step": 5479 + }, + { + "epoch": 4.536720044174489, + "grad_norm": 0.9566417932510376, + "learning_rate": 7.140542702902434e-07, + "loss": 0.5448, + "step": 5480 + }, + { + "epoch": 4.537548315847598, + "grad_norm": 0.9857562780380249, + "learning_rate": 7.132849247638948e-07, + "loss": 0.5536, + "step": 5481 + }, + { + "epoch": 4.538376587520707, + "grad_norm": 0.9420795440673828, + "learning_rate": 7.125159249502575e-07, + "loss": 0.5536, + "step": 5482 + }, + { + "epoch": 4.5392048591938154, + "grad_norm": 0.9628307223320007, + "learning_rate": 7.117472709981272e-07, + "loss": 0.5537, + "step": 5483 + }, + { + "epoch": 4.540033130866925, + "grad_norm": 0.9649658799171448, + "learning_rate": 7.109789630562289e-07, + "loss": 0.544, + "step": 5484 + }, + { + "epoch": 4.540861402540033, + "grad_norm": 0.9789855480194092, + "learning_rate": 7.102110012732253e-07, + "loss": 0.5374, + "step": 5485 + }, + { + "epoch": 4.541689674213142, + "grad_norm": 0.9994000196456909, + "learning_rate": 7.094433857977079e-07, + "loss": 0.5763, + "step": 5486 + }, + { + "epoch": 4.54251794588625, + "grad_norm": 0.944028377532959, + "learning_rate": 7.086761167782047e-07, + "loss": 0.5378, + "step": 5487 + }, + { + "epoch": 4.5433462175593595, + "grad_norm": 0.9605673551559448, + "learning_rate": 7.079091943631736e-07, + "loss": 0.5587, + "step": 5488 + }, + { + "epoch": 4.544174489232468, + "grad_norm": 0.9778985977172852, + "learning_rate": 7.071426187010077e-07, + "loss": 0.5614, + "step": 5489 + }, + { + "epoch": 4.545002760905577, + "grad_norm": 0.9588051438331604, + "learning_rate": 7.063763899400331e-07, + "loss": 0.5262, + "step": 5490 + }, + { + "epoch": 4.545831032578686, + "grad_norm": 0.9682207703590393, + "learning_rate": 7.056105082285064e-07, + "loss": 0.5702, + "step": 5491 + }, + { + "epoch": 4.546659304251794, + "grad_norm": 0.9890053868293762, + "learning_rate": 7.048449737146193e-07, + "loss": 0.5693, + "step": 5492 + }, + { + "epoch": 4.547487575924904, + "grad_norm": 0.9506924748420715, + "learning_rate": 7.040797865464962e-07, + "loss": 0.5465, + "step": 5493 + }, + { + "epoch": 4.548315847598012, + "grad_norm": 0.9579111337661743, + "learning_rate": 7.033149468721923e-07, + "loss": 0.5505, + "step": 5494 + }, + { + "epoch": 4.549144119271121, + "grad_norm": 0.964435875415802, + "learning_rate": 7.025504548396978e-07, + "loss": 0.5543, + "step": 5495 + }, + { + "epoch": 4.549972390944229, + "grad_norm": 0.992845892906189, + "learning_rate": 7.017863105969352e-07, + "loss": 0.552, + "step": 5496 + }, + { + "epoch": 4.5508006626173385, + "grad_norm": 0.9823659062385559, + "learning_rate": 7.01022514291759e-07, + "loss": 0.5497, + "step": 5497 + }, + { + "epoch": 4.551628934290447, + "grad_norm": 0.9746332168579102, + "learning_rate": 7.002590660719555e-07, + "loss": 0.5413, + "step": 5498 + }, + { + "epoch": 4.552457205963556, + "grad_norm": 0.9899871945381165, + "learning_rate": 6.994959660852468e-07, + "loss": 0.5539, + "step": 5499 + }, + { + "epoch": 4.553285477636665, + "grad_norm": 0.9836088418960571, + "learning_rate": 6.987332144792835e-07, + "loss": 0.5674, + "step": 5500 + }, + { + "epoch": 4.554113749309773, + "grad_norm": 0.9729539155960083, + "learning_rate": 6.979708114016522e-07, + "loss": 0.5408, + "step": 5501 + }, + { + "epoch": 4.554942020982883, + "grad_norm": 0.978412389755249, + "learning_rate": 6.972087569998712e-07, + "loss": 0.564, + "step": 5502 + }, + { + "epoch": 4.555770292655991, + "grad_norm": 0.9849222898483276, + "learning_rate": 6.964470514213897e-07, + "loss": 0.5481, + "step": 5503 + }, + { + "epoch": 4.5565985643291, + "grad_norm": 0.9477882981300354, + "learning_rate": 6.956856948135912e-07, + "loss": 0.5522, + "step": 5504 + }, + { + "epoch": 4.557426836002209, + "grad_norm": 0.9613835215568542, + "learning_rate": 6.949246873237919e-07, + "loss": 0.5729, + "step": 5505 + }, + { + "epoch": 4.5582551076753175, + "grad_norm": 0.9619837999343872, + "learning_rate": 6.94164029099238e-07, + "loss": 0.5546, + "step": 5506 + }, + { + "epoch": 4.559083379348427, + "grad_norm": 1.0136449337005615, + "learning_rate": 6.934037202871108e-07, + "loss": 0.5553, + "step": 5507 + }, + { + "epoch": 4.559911651021535, + "grad_norm": 0.9942994713783264, + "learning_rate": 6.926437610345238e-07, + "loss": 0.5715, + "step": 5508 + }, + { + "epoch": 4.560739922694644, + "grad_norm": 0.9909749627113342, + "learning_rate": 6.918841514885207e-07, + "loss": 0.587, + "step": 5509 + }, + { + "epoch": 4.561568194367752, + "grad_norm": 0.9991593956947327, + "learning_rate": 6.911248917960789e-07, + "loss": 0.5667, + "step": 5510 + }, + { + "epoch": 4.562396466040862, + "grad_norm": 0.985620379447937, + "learning_rate": 6.903659821041081e-07, + "loss": 0.5611, + "step": 5511 + }, + { + "epoch": 4.56322473771397, + "grad_norm": 1.0072076320648193, + "learning_rate": 6.896074225594516e-07, + "loss": 0.5507, + "step": 5512 + }, + { + "epoch": 4.564053009387079, + "grad_norm": 0.9806503653526306, + "learning_rate": 6.888492133088815e-07, + "loss": 0.5385, + "step": 5513 + }, + { + "epoch": 4.564881281060188, + "grad_norm": 0.9733349084854126, + "learning_rate": 6.880913544991053e-07, + "loss": 0.5541, + "step": 5514 + }, + { + "epoch": 4.5657095527332965, + "grad_norm": 0.9819521307945251, + "learning_rate": 6.873338462767625e-07, + "loss": 0.5493, + "step": 5515 + }, + { + "epoch": 4.566537824406406, + "grad_norm": 0.9954928755760193, + "learning_rate": 6.865766887884223e-07, + "loss": 0.5562, + "step": 5516 + }, + { + "epoch": 4.567366096079514, + "grad_norm": 0.9706395268440247, + "learning_rate": 6.858198821805881e-07, + "loss": 0.5497, + "step": 5517 + }, + { + "epoch": 4.568194367752623, + "grad_norm": 0.9409603476524353, + "learning_rate": 6.85063426599696e-07, + "loss": 0.5417, + "step": 5518 + }, + { + "epoch": 4.569022639425731, + "grad_norm": 0.9698299169540405, + "learning_rate": 6.843073221921123e-07, + "loss": 0.5625, + "step": 5519 + }, + { + "epoch": 4.569850911098841, + "grad_norm": 0.9876602292060852, + "learning_rate": 6.835515691041353e-07, + "loss": 0.5581, + "step": 5520 + }, + { + "epoch": 4.570679182771949, + "grad_norm": 0.9799392223358154, + "learning_rate": 6.827961674819978e-07, + "loss": 0.5645, + "step": 5521 + }, + { + "epoch": 4.571507454445058, + "grad_norm": 0.9740294814109802, + "learning_rate": 6.820411174718614e-07, + "loss": 0.5512, + "step": 5522 + }, + { + "epoch": 4.572335726118167, + "grad_norm": 0.9723674058914185, + "learning_rate": 6.812864192198215e-07, + "loss": 0.5708, + "step": 5523 + }, + { + "epoch": 4.5731639977912755, + "grad_norm": 0.9980106353759766, + "learning_rate": 6.80532072871907e-07, + "loss": 0.5579, + "step": 5524 + }, + { + "epoch": 4.573992269464385, + "grad_norm": 0.9684682488441467, + "learning_rate": 6.797780785740743e-07, + "loss": 0.5629, + "step": 5525 + }, + { + "epoch": 4.574820541137493, + "grad_norm": 0.9702059030532837, + "learning_rate": 6.790244364722159e-07, + "loss": 0.5804, + "step": 5526 + }, + { + "epoch": 4.575648812810602, + "grad_norm": 0.9612674117088318, + "learning_rate": 6.782711467121547e-07, + "loss": 0.5518, + "step": 5527 + }, + { + "epoch": 4.57647708448371, + "grad_norm": 1.0216538906097412, + "learning_rate": 6.775182094396437e-07, + "loss": 0.5791, + "step": 5528 + }, + { + "epoch": 4.5773053561568195, + "grad_norm": 0.9898926615715027, + "learning_rate": 6.767656248003707e-07, + "loss": 0.5666, + "step": 5529 + }, + { + "epoch": 4.578133627829928, + "grad_norm": 0.9707671999931335, + "learning_rate": 6.760133929399537e-07, + "loss": 0.5506, + "step": 5530 + }, + { + "epoch": 4.578961899503037, + "grad_norm": 0.9658564329147339, + "learning_rate": 6.752615140039423e-07, + "loss": 0.5451, + "step": 5531 + }, + { + "epoch": 4.579790171176146, + "grad_norm": 0.9721407890319824, + "learning_rate": 6.745099881378172e-07, + "loss": 0.5634, + "step": 5532 + }, + { + "epoch": 4.5806184428492545, + "grad_norm": 0.9759280681610107, + "learning_rate": 6.737588154869931e-07, + "loss": 0.5537, + "step": 5533 + }, + { + "epoch": 4.581446714522364, + "grad_norm": 0.9694047570228577, + "learning_rate": 6.730079961968137e-07, + "loss": 0.5396, + "step": 5534 + }, + { + "epoch": 4.582274986195472, + "grad_norm": 0.9477902054786682, + "learning_rate": 6.722575304125555e-07, + "loss": 0.5487, + "step": 5535 + }, + { + "epoch": 4.583103257868581, + "grad_norm": 0.9674749970436096, + "learning_rate": 6.715074182794284e-07, + "loss": 0.5562, + "step": 5536 + }, + { + "epoch": 4.583931529541689, + "grad_norm": 0.9692550897598267, + "learning_rate": 6.707576599425699e-07, + "loss": 0.5463, + "step": 5537 + }, + { + "epoch": 4.5847598012147985, + "grad_norm": 0.9627196192741394, + "learning_rate": 6.700082555470524e-07, + "loss": 0.5505, + "step": 5538 + }, + { + "epoch": 4.585588072887907, + "grad_norm": 0.9593527317047119, + "learning_rate": 6.692592052378791e-07, + "loss": 0.5494, + "step": 5539 + }, + { + "epoch": 4.586416344561016, + "grad_norm": 0.9435111284255981, + "learning_rate": 6.685105091599828e-07, + "loss": 0.5467, + "step": 5540 + }, + { + "epoch": 4.587244616234125, + "grad_norm": 0.9593883752822876, + "learning_rate": 6.677621674582302e-07, + "loss": 0.5393, + "step": 5541 + }, + { + "epoch": 4.588072887907233, + "grad_norm": 0.9862588047981262, + "learning_rate": 6.670141802774191e-07, + "loss": 0.5478, + "step": 5542 + }, + { + "epoch": 4.588901159580343, + "grad_norm": 0.9542217254638672, + "learning_rate": 6.662665477622771e-07, + "loss": 0.5518, + "step": 5543 + }, + { + "epoch": 4.589729431253451, + "grad_norm": 0.9312442541122437, + "learning_rate": 6.655192700574634e-07, + "loss": 0.554, + "step": 5544 + }, + { + "epoch": 4.59055770292656, + "grad_norm": 0.9759663939476013, + "learning_rate": 6.647723473075704e-07, + "loss": 0.5635, + "step": 5545 + }, + { + "epoch": 4.591385974599668, + "grad_norm": 0.9525055885314941, + "learning_rate": 6.640257796571209e-07, + "loss": 0.5448, + "step": 5546 + }, + { + "epoch": 4.5922142462727775, + "grad_norm": 0.9484573006629944, + "learning_rate": 6.632795672505679e-07, + "loss": 0.5516, + "step": 5547 + }, + { + "epoch": 4.593042517945886, + "grad_norm": 0.9842866659164429, + "learning_rate": 6.625337102322968e-07, + "loss": 0.532, + "step": 5548 + }, + { + "epoch": 4.593870789618995, + "grad_norm": 0.9838387966156006, + "learning_rate": 6.617882087466246e-07, + "loss": 0.5666, + "step": 5549 + }, + { + "epoch": 4.594699061292104, + "grad_norm": 0.9601597785949707, + "learning_rate": 6.610430629377979e-07, + "loss": 0.5707, + "step": 5550 + }, + { + "epoch": 4.595527332965212, + "grad_norm": 0.9840213656425476, + "learning_rate": 6.60298272949996e-07, + "loss": 0.5284, + "step": 5551 + }, + { + "epoch": 4.596355604638322, + "grad_norm": 0.9841850399971008, + "learning_rate": 6.595538389273295e-07, + "loss": 0.5475, + "step": 5552 + }, + { + "epoch": 4.59718387631143, + "grad_norm": 0.9589581489562988, + "learning_rate": 6.588097610138389e-07, + "loss": 0.5595, + "step": 5553 + }, + { + "epoch": 4.598012147984539, + "grad_norm": 0.9751944541931152, + "learning_rate": 6.580660393534952e-07, + "loss": 0.5515, + "step": 5554 + }, + { + "epoch": 4.598840419657647, + "grad_norm": 0.9598148465156555, + "learning_rate": 6.573226740902036e-07, + "loss": 0.542, + "step": 5555 + }, + { + "epoch": 4.5996686913307565, + "grad_norm": 0.9537079334259033, + "learning_rate": 6.565796653677963e-07, + "loss": 0.5457, + "step": 5556 + }, + { + "epoch": 4.600496963003865, + "grad_norm": 0.9721077084541321, + "learning_rate": 6.558370133300399e-07, + "loss": 0.5399, + "step": 5557 + }, + { + "epoch": 4.601325234676974, + "grad_norm": 0.9512110352516174, + "learning_rate": 6.55094718120631e-07, + "loss": 0.5554, + "step": 5558 + }, + { + "epoch": 4.602153506350083, + "grad_norm": 0.9435778260231018, + "learning_rate": 6.543527798831953e-07, + "loss": 0.5322, + "step": 5559 + }, + { + "epoch": 4.602981778023191, + "grad_norm": 0.9703747630119324, + "learning_rate": 6.536111987612917e-07, + "loss": 0.547, + "step": 5560 + }, + { + "epoch": 4.603810049696301, + "grad_norm": 0.9578705430030823, + "learning_rate": 6.528699748984104e-07, + "loss": 0.5513, + "step": 5561 + }, + { + "epoch": 4.604638321369409, + "grad_norm": 0.9871172904968262, + "learning_rate": 6.521291084379694e-07, + "loss": 0.5675, + "step": 5562 + }, + { + "epoch": 4.605466593042518, + "grad_norm": 0.9615850448608398, + "learning_rate": 6.513885995233202e-07, + "loss": 0.54, + "step": 5563 + }, + { + "epoch": 4.606294864715627, + "grad_norm": 0.9759035706520081, + "learning_rate": 6.506484482977452e-07, + "loss": 0.5438, + "step": 5564 + }, + { + "epoch": 4.6071231363887355, + "grad_norm": 0.935772180557251, + "learning_rate": 6.499086549044559e-07, + "loss": 0.5596, + "step": 5565 + }, + { + "epoch": 4.607951408061845, + "grad_norm": 0.977527379989624, + "learning_rate": 6.491692194865948e-07, + "loss": 0.5715, + "step": 5566 + }, + { + "epoch": 4.608779679734953, + "grad_norm": 0.9544543027877808, + "learning_rate": 6.484301421872366e-07, + "loss": 0.547, + "step": 5567 + }, + { + "epoch": 4.609607951408062, + "grad_norm": 0.9892758727073669, + "learning_rate": 6.476914231493866e-07, + "loss": 0.5672, + "step": 5568 + }, + { + "epoch": 4.61043622308117, + "grad_norm": 1.0188134908676147, + "learning_rate": 6.469530625159784e-07, + "loss": 0.5587, + "step": 5569 + }, + { + "epoch": 4.61126449475428, + "grad_norm": 0.9610130786895752, + "learning_rate": 6.462150604298786e-07, + "loss": 0.5603, + "step": 5570 + }, + { + "epoch": 4.612092766427388, + "grad_norm": 1.0074714422225952, + "learning_rate": 6.454774170338848e-07, + "loss": 0.5585, + "step": 5571 + }, + { + "epoch": 4.612921038100497, + "grad_norm": 0.9753648042678833, + "learning_rate": 6.447401324707225e-07, + "loss": 0.5524, + "step": 5572 + }, + { + "epoch": 4.613749309773606, + "grad_norm": 0.9903519749641418, + "learning_rate": 6.440032068830498e-07, + "loss": 0.5504, + "step": 5573 + }, + { + "epoch": 4.6145775814467145, + "grad_norm": 0.9634335041046143, + "learning_rate": 6.43266640413456e-07, + "loss": 0.5393, + "step": 5574 + }, + { + "epoch": 4.615405853119824, + "grad_norm": 0.9743342399597168, + "learning_rate": 6.425304332044582e-07, + "loss": 0.5501, + "step": 5575 + }, + { + "epoch": 4.616234124792932, + "grad_norm": 0.9500157833099365, + "learning_rate": 6.417945853985075e-07, + "loss": 0.5477, + "step": 5576 + }, + { + "epoch": 4.617062396466041, + "grad_norm": 0.9560253024101257, + "learning_rate": 6.410590971379824e-07, + "loss": 0.5294, + "step": 5577 + }, + { + "epoch": 4.617890668139149, + "grad_norm": 0.9748409986495972, + "learning_rate": 6.403239685651924e-07, + "loss": 0.5491, + "step": 5578 + }, + { + "epoch": 4.618718939812259, + "grad_norm": 0.9598991870880127, + "learning_rate": 6.395891998223789e-07, + "loss": 0.5512, + "step": 5579 + }, + { + "epoch": 4.619547211485367, + "grad_norm": 0.9680323600769043, + "learning_rate": 6.388547910517132e-07, + "loss": 0.5562, + "step": 5580 + }, + { + "epoch": 4.620375483158476, + "grad_norm": 0.9817914962768555, + "learning_rate": 6.381207423952954e-07, + "loss": 0.5548, + "step": 5581 + }, + { + "epoch": 4.621203754831585, + "grad_norm": 0.9678810238838196, + "learning_rate": 6.373870539951579e-07, + "loss": 0.5406, + "step": 5582 + }, + { + "epoch": 4.6220320265046935, + "grad_norm": 0.9937217831611633, + "learning_rate": 6.366537259932632e-07, + "loss": 0.5747, + "step": 5583 + }, + { + "epoch": 4.622860298177803, + "grad_norm": 1.0302647352218628, + "learning_rate": 6.359207585315017e-07, + "loss": 0.5531, + "step": 5584 + }, + { + "epoch": 4.623688569850911, + "grad_norm": 0.9746495485305786, + "learning_rate": 6.351881517516967e-07, + "loss": 0.5631, + "step": 5585 + }, + { + "epoch": 4.62451684152402, + "grad_norm": 0.9883660674095154, + "learning_rate": 6.344559057956015e-07, + "loss": 0.5582, + "step": 5586 + }, + { + "epoch": 4.625345113197128, + "grad_norm": 0.9649250507354736, + "learning_rate": 6.337240208048981e-07, + "loss": 0.5473, + "step": 5587 + }, + { + "epoch": 4.6261733848702375, + "grad_norm": 0.9564082026481628, + "learning_rate": 6.32992496921199e-07, + "loss": 0.5494, + "step": 5588 + }, + { + "epoch": 4.627001656543346, + "grad_norm": 0.9802419543266296, + "learning_rate": 6.322613342860473e-07, + "loss": 0.5321, + "step": 5589 + }, + { + "epoch": 4.627829928216455, + "grad_norm": 0.9529916048049927, + "learning_rate": 6.315305330409177e-07, + "loss": 0.5621, + "step": 5590 + }, + { + "epoch": 4.628658199889564, + "grad_norm": 0.9516423940658569, + "learning_rate": 6.308000933272116e-07, + "loss": 0.5643, + "step": 5591 + }, + { + "epoch": 4.6294864715626725, + "grad_norm": 0.9464498162269592, + "learning_rate": 6.300700152862627e-07, + "loss": 0.5325, + "step": 5592 + }, + { + "epoch": 4.630314743235782, + "grad_norm": 0.9634289145469666, + "learning_rate": 6.293402990593353e-07, + "loss": 0.5707, + "step": 5593 + }, + { + "epoch": 4.63114301490889, + "grad_norm": 0.9790555238723755, + "learning_rate": 6.286109447876215e-07, + "loss": 0.5601, + "step": 5594 + }, + { + "epoch": 4.631971286581999, + "grad_norm": 0.9650810956954956, + "learning_rate": 6.278819526122454e-07, + "loss": 0.546, + "step": 5595 + }, + { + "epoch": 4.632799558255107, + "grad_norm": 0.9842014908790588, + "learning_rate": 6.271533226742593e-07, + "loss": 0.5522, + "step": 5596 + }, + { + "epoch": 4.6336278299282165, + "grad_norm": 0.9452126026153564, + "learning_rate": 6.26425055114647e-07, + "loss": 0.5368, + "step": 5597 + }, + { + "epoch": 4.634456101601325, + "grad_norm": 0.9557516574859619, + "learning_rate": 6.256971500743217e-07, + "loss": 0.5514, + "step": 5598 + }, + { + "epoch": 4.635284373274434, + "grad_norm": 0.9769173860549927, + "learning_rate": 6.249696076941258e-07, + "loss": 0.5548, + "step": 5599 + }, + { + "epoch": 4.636112644947543, + "grad_norm": 0.9568111300468445, + "learning_rate": 6.242424281148316e-07, + "loss": 0.5624, + "step": 5600 + }, + { + "epoch": 4.636940916620651, + "grad_norm": 0.955300509929657, + "learning_rate": 6.235156114771418e-07, + "loss": 0.5501, + "step": 5601 + }, + { + "epoch": 4.637769188293761, + "grad_norm": 0.9530934691429138, + "learning_rate": 6.227891579216894e-07, + "loss": 0.5439, + "step": 5602 + }, + { + "epoch": 4.638597459966869, + "grad_norm": 0.9716370105743408, + "learning_rate": 6.220630675890352e-07, + "loss": 0.5705, + "step": 5603 + }, + { + "epoch": 4.639425731639978, + "grad_norm": 0.9733977913856506, + "learning_rate": 6.213373406196715e-07, + "loss": 0.5738, + "step": 5604 + }, + { + "epoch": 4.640254003313086, + "grad_norm": 0.9514371156692505, + "learning_rate": 6.206119771540206e-07, + "loss": 0.5748, + "step": 5605 + }, + { + "epoch": 4.6410822749861955, + "grad_norm": 0.9908952116966248, + "learning_rate": 6.198869773324317e-07, + "loss": 0.5474, + "step": 5606 + }, + { + "epoch": 4.641910546659304, + "grad_norm": 0.9772365093231201, + "learning_rate": 6.191623412951867e-07, + "loss": 0.5749, + "step": 5607 + }, + { + "epoch": 4.642738818332413, + "grad_norm": 0.9424340724945068, + "learning_rate": 6.184380691824965e-07, + "loss": 0.572, + "step": 5608 + }, + { + "epoch": 4.643567090005522, + "grad_norm": 0.9462697505950928, + "learning_rate": 6.177141611344991e-07, + "loss": 0.5551, + "step": 5609 + }, + { + "epoch": 4.64439536167863, + "grad_norm": 0.9748279452323914, + "learning_rate": 6.169906172912663e-07, + "loss": 0.5593, + "step": 5610 + }, + { + "epoch": 4.64522363335174, + "grad_norm": 0.9854201674461365, + "learning_rate": 6.162674377927955e-07, + "loss": 0.5698, + "step": 5611 + }, + { + "epoch": 4.646051905024848, + "grad_norm": 0.9688649773597717, + "learning_rate": 6.155446227790149e-07, + "loss": 0.5599, + "step": 5612 + }, + { + "epoch": 4.646880176697957, + "grad_norm": 0.9476209282875061, + "learning_rate": 6.148221723897832e-07, + "loss": 0.552, + "step": 5613 + }, + { + "epoch": 4.647708448371065, + "grad_norm": 0.93233722448349, + "learning_rate": 6.141000867648886e-07, + "loss": 0.5412, + "step": 5614 + }, + { + "epoch": 4.6485367200441745, + "grad_norm": 0.9635641574859619, + "learning_rate": 6.133783660440462e-07, + "loss": 0.5609, + "step": 5615 + }, + { + "epoch": 4.649364991717283, + "grad_norm": 0.9933269619941711, + "learning_rate": 6.12657010366903e-07, + "loss": 0.5547, + "step": 5616 + }, + { + "epoch": 4.650193263390392, + "grad_norm": 0.9645413160324097, + "learning_rate": 6.119360198730356e-07, + "loss": 0.5609, + "step": 5617 + }, + { + "epoch": 4.651021535063501, + "grad_norm": 0.9551970958709717, + "learning_rate": 6.112153947019472e-07, + "loss": 0.5404, + "step": 5618 + }, + { + "epoch": 4.651849806736609, + "grad_norm": 0.9647742509841919, + "learning_rate": 6.104951349930729e-07, + "loss": 0.5276, + "step": 5619 + }, + { + "epoch": 4.652678078409719, + "grad_norm": 0.9530763626098633, + "learning_rate": 6.097752408857771e-07, + "loss": 0.5302, + "step": 5620 + }, + { + "epoch": 4.653506350082827, + "grad_norm": 0.9838370084762573, + "learning_rate": 6.090557125193517e-07, + "loss": 0.5747, + "step": 5621 + }, + { + "epoch": 4.654334621755936, + "grad_norm": 0.9567121267318726, + "learning_rate": 6.08336550033018e-07, + "loss": 0.5533, + "step": 5622 + }, + { + "epoch": 4.655162893429045, + "grad_norm": 0.9746866226196289, + "learning_rate": 6.076177535659277e-07, + "loss": 0.5673, + "step": 5623 + }, + { + "epoch": 4.6559911651021535, + "grad_norm": 0.9545463919639587, + "learning_rate": 6.068993232571626e-07, + "loss": 0.5521, + "step": 5624 + }, + { + "epoch": 4.656819436775263, + "grad_norm": 0.9727249145507812, + "learning_rate": 6.061812592457306e-07, + "loss": 0.5767, + "step": 5625 + }, + { + "epoch": 4.657647708448371, + "grad_norm": 0.9679794311523438, + "learning_rate": 6.054635616705709e-07, + "loss": 0.5522, + "step": 5626 + }, + { + "epoch": 4.65847598012148, + "grad_norm": 0.9779345989227295, + "learning_rate": 6.047462306705517e-07, + "loss": 0.551, + "step": 5627 + }, + { + "epoch": 4.659304251794588, + "grad_norm": 0.9675984978675842, + "learning_rate": 6.040292663844693e-07, + "loss": 0.5495, + "step": 5628 + }, + { + "epoch": 4.660132523467698, + "grad_norm": 0.9447121024131775, + "learning_rate": 6.033126689510497e-07, + "loss": 0.5644, + "step": 5629 + }, + { + "epoch": 4.660960795140806, + "grad_norm": 0.9525550603866577, + "learning_rate": 6.025964385089486e-07, + "loss": 0.5621, + "step": 5630 + }, + { + "epoch": 4.661789066813915, + "grad_norm": 0.9803587794303894, + "learning_rate": 6.018805751967486e-07, + "loss": 0.5582, + "step": 5631 + }, + { + "epoch": 4.662617338487024, + "grad_norm": 0.9682264924049377, + "learning_rate": 6.011650791529639e-07, + "loss": 0.5561, + "step": 5632 + }, + { + "epoch": 4.6634456101601325, + "grad_norm": 0.9648387432098389, + "learning_rate": 6.004499505160358e-07, + "loss": 0.5404, + "step": 5633 + }, + { + "epoch": 4.664273881833242, + "grad_norm": 0.9990679621696472, + "learning_rate": 5.997351894243342e-07, + "loss": 0.5422, + "step": 5634 + }, + { + "epoch": 4.66510215350635, + "grad_norm": 0.9632382392883301, + "learning_rate": 5.990207960161592e-07, + "loss": 0.5237, + "step": 5635 + }, + { + "epoch": 4.665930425179459, + "grad_norm": 0.9712291359901428, + "learning_rate": 5.983067704297402e-07, + "loss": 0.5606, + "step": 5636 + }, + { + "epoch": 4.666758696852567, + "grad_norm": 0.977373480796814, + "learning_rate": 5.975931128032333e-07, + "loss": 0.5453, + "step": 5637 + }, + { + "epoch": 4.667586968525677, + "grad_norm": 1.0031154155731201, + "learning_rate": 5.968798232747247e-07, + "loss": 0.5286, + "step": 5638 + }, + { + "epoch": 4.668415240198785, + "grad_norm": 0.9799690842628479, + "learning_rate": 5.961669019822308e-07, + "loss": 0.5579, + "step": 5639 + }, + { + "epoch": 4.669243511871894, + "grad_norm": 1.0083366632461548, + "learning_rate": 5.95454349063693e-07, + "loss": 0.563, + "step": 5640 + }, + { + "epoch": 4.670071783545003, + "grad_norm": 0.9788227677345276, + "learning_rate": 5.947421646569849e-07, + "loss": 0.5449, + "step": 5641 + }, + { + "epoch": 4.6709000552181115, + "grad_norm": 0.9708746075630188, + "learning_rate": 5.94030348899908e-07, + "loss": 0.545, + "step": 5642 + }, + { + "epoch": 4.671728326891221, + "grad_norm": 0.9622349739074707, + "learning_rate": 5.933189019301913e-07, + "loss": 0.533, + "step": 5643 + }, + { + "epoch": 4.672556598564329, + "grad_norm": 0.9786980748176575, + "learning_rate": 5.926078238854923e-07, + "loss": 0.545, + "step": 5644 + }, + { + "epoch": 4.673384870237438, + "grad_norm": 0.9637975692749023, + "learning_rate": 5.918971149033995e-07, + "loss": 0.5411, + "step": 5645 + }, + { + "epoch": 4.674213141910546, + "grad_norm": 0.9689448475837708, + "learning_rate": 5.911867751214281e-07, + "loss": 0.5511, + "step": 5646 + }, + { + "epoch": 4.6750414135836555, + "grad_norm": 0.9778165817260742, + "learning_rate": 5.904768046770212e-07, + "loss": 0.5557, + "step": 5647 + }, + { + "epoch": 4.675869685256764, + "grad_norm": 0.9706706404685974, + "learning_rate": 5.897672037075524e-07, + "loss": 0.5687, + "step": 5648 + }, + { + "epoch": 4.676697956929873, + "grad_norm": 0.976809024810791, + "learning_rate": 5.890579723503237e-07, + "loss": 0.5519, + "step": 5649 + }, + { + "epoch": 4.677526228602982, + "grad_norm": 0.9677285552024841, + "learning_rate": 5.883491107425627e-07, + "loss": 0.5425, + "step": 5650 + }, + { + "epoch": 4.6783545002760905, + "grad_norm": 0.9553424715995789, + "learning_rate": 5.876406190214287e-07, + "loss": 0.5632, + "step": 5651 + }, + { + "epoch": 4.6791827719492, + "grad_norm": 0.9804576635360718, + "learning_rate": 5.869324973240089e-07, + "loss": 0.5499, + "step": 5652 + }, + { + "epoch": 4.680011043622308, + "grad_norm": 0.9667906165122986, + "learning_rate": 5.862247457873169e-07, + "loss": 0.5482, + "step": 5653 + }, + { + "epoch": 4.680839315295417, + "grad_norm": 0.960593581199646, + "learning_rate": 5.855173645482964e-07, + "loss": 0.5531, + "step": 5654 + }, + { + "epoch": 4.681667586968525, + "grad_norm": 0.9586433172225952, + "learning_rate": 5.848103537438207e-07, + "loss": 0.5774, + "step": 5655 + }, + { + "epoch": 4.6824958586416345, + "grad_norm": 0.9638173580169678, + "learning_rate": 5.841037135106869e-07, + "loss": 0.565, + "step": 5656 + }, + { + "epoch": 4.683324130314743, + "grad_norm": 1.0111193656921387, + "learning_rate": 5.833974439856249e-07, + "loss": 0.5263, + "step": 5657 + }, + { + "epoch": 4.684152401987852, + "grad_norm": 0.9522552490234375, + "learning_rate": 5.826915453052917e-07, + "loss": 0.553, + "step": 5658 + }, + { + "epoch": 4.684980673660961, + "grad_norm": 0.9515351057052612, + "learning_rate": 5.819860176062711e-07, + "loss": 0.571, + "step": 5659 + }, + { + "epoch": 4.685808945334069, + "grad_norm": 1.0340934991836548, + "learning_rate": 5.812808610250764e-07, + "loss": 0.5719, + "step": 5660 + }, + { + "epoch": 4.686637217007179, + "grad_norm": 0.9896745085716248, + "learning_rate": 5.805760756981499e-07, + "loss": 0.544, + "step": 5661 + }, + { + "epoch": 4.687465488680287, + "grad_norm": 0.9758293032646179, + "learning_rate": 5.798716617618594e-07, + "loss": 0.5498, + "step": 5662 + }, + { + "epoch": 4.688293760353396, + "grad_norm": 0.9765741229057312, + "learning_rate": 5.791676193525028e-07, + "loss": 0.5572, + "step": 5663 + }, + { + "epoch": 4.689122032026504, + "grad_norm": 0.9680903553962708, + "learning_rate": 5.784639486063074e-07, + "loss": 0.5389, + "step": 5664 + }, + { + "epoch": 4.6899503036996135, + "grad_norm": 0.9511458277702332, + "learning_rate": 5.777606496594243e-07, + "loss": 0.5507, + "step": 5665 + }, + { + "epoch": 4.690778575372722, + "grad_norm": 0.9745240807533264, + "learning_rate": 5.770577226479376e-07, + "loss": 0.5655, + "step": 5666 + }, + { + "epoch": 4.691606847045831, + "grad_norm": 0.9594129323959351, + "learning_rate": 5.763551677078555e-07, + "loss": 0.5649, + "step": 5667 + }, + { + "epoch": 4.69243511871894, + "grad_norm": 0.9827647805213928, + "learning_rate": 5.756529849751169e-07, + "loss": 0.5441, + "step": 5668 + }, + { + "epoch": 4.693263390392048, + "grad_norm": 0.9480031728744507, + "learning_rate": 5.749511745855865e-07, + "loss": 0.5646, + "step": 5669 + }, + { + "epoch": 4.694091662065158, + "grad_norm": 0.9795644879341125, + "learning_rate": 5.74249736675059e-07, + "loss": 0.5422, + "step": 5670 + }, + { + "epoch": 4.694919933738266, + "grad_norm": 0.9643570184707642, + "learning_rate": 5.735486713792565e-07, + "loss": 0.5587, + "step": 5671 + }, + { + "epoch": 4.695748205411375, + "grad_norm": 0.9786714911460876, + "learning_rate": 5.72847978833827e-07, + "loss": 0.5518, + "step": 5672 + }, + { + "epoch": 4.696576477084483, + "grad_norm": 0.964820384979248, + "learning_rate": 5.721476591743496e-07, + "loss": 0.552, + "step": 5673 + }, + { + "epoch": 4.6974047487575925, + "grad_norm": 0.9419858455657959, + "learning_rate": 5.714477125363286e-07, + "loss": 0.535, + "step": 5674 + }, + { + "epoch": 4.698233020430701, + "grad_norm": 0.9434426426887512, + "learning_rate": 5.707481390551972e-07, + "loss": 0.5422, + "step": 5675 + }, + { + "epoch": 4.69906129210381, + "grad_norm": 0.9646450281143188, + "learning_rate": 5.700489388663175e-07, + "loss": 0.5538, + "step": 5676 + }, + { + "epoch": 4.699889563776919, + "grad_norm": 0.9789668917655945, + "learning_rate": 5.693501121049771e-07, + "loss": 0.546, + "step": 5677 + }, + { + "epoch": 4.700717835450027, + "grad_norm": 0.9620424509048462, + "learning_rate": 5.686516589063923e-07, + "loss": 0.5382, + "step": 5678 + }, + { + "epoch": 4.701546107123137, + "grad_norm": 0.9448021650314331, + "learning_rate": 5.679535794057076e-07, + "loss": 0.5388, + "step": 5679 + }, + { + "epoch": 4.702374378796245, + "grad_norm": 1.0032944679260254, + "learning_rate": 5.672558737379954e-07, + "loss": 0.5493, + "step": 5680 + }, + { + "epoch": 4.703202650469354, + "grad_norm": 0.9643270969390869, + "learning_rate": 5.665585420382546e-07, + "loss": 0.5355, + "step": 5681 + }, + { + "epoch": 4.704030922142463, + "grad_norm": 0.9836111068725586, + "learning_rate": 5.658615844414123e-07, + "loss": 0.5765, + "step": 5682 + }, + { + "epoch": 4.7048591938155715, + "grad_norm": 0.9687516093254089, + "learning_rate": 5.651650010823245e-07, + "loss": 0.5542, + "step": 5683 + }, + { + "epoch": 4.705687465488681, + "grad_norm": 0.9528695344924927, + "learning_rate": 5.644687920957717e-07, + "loss": 0.56, + "step": 5684 + }, + { + "epoch": 4.706515737161789, + "grad_norm": 0.9862529635429382, + "learning_rate": 5.637729576164652e-07, + "loss": 0.5448, + "step": 5685 + }, + { + "epoch": 4.707344008834898, + "grad_norm": 0.9494380950927734, + "learning_rate": 5.630774977790426e-07, + "loss": 0.543, + "step": 5686 + }, + { + "epoch": 4.708172280508006, + "grad_norm": 0.9758495688438416, + "learning_rate": 5.623824127180677e-07, + "loss": 0.5686, + "step": 5687 + }, + { + "epoch": 4.709000552181116, + "grad_norm": 0.9538042545318604, + "learning_rate": 5.616877025680337e-07, + "loss": 0.5763, + "step": 5688 + }, + { + "epoch": 4.709828823854224, + "grad_norm": 0.961929440498352, + "learning_rate": 5.609933674633619e-07, + "loss": 0.5375, + "step": 5689 + }, + { + "epoch": 4.710657095527333, + "grad_norm": 0.9387208819389343, + "learning_rate": 5.602994075383971e-07, + "loss": 0.5475, + "step": 5690 + }, + { + "epoch": 4.711485367200442, + "grad_norm": 0.9944952726364136, + "learning_rate": 5.596058229274152e-07, + "loss": 0.5577, + "step": 5691 + }, + { + "epoch": 4.7123136388735505, + "grad_norm": 0.9691250920295715, + "learning_rate": 5.589126137646192e-07, + "loss": 0.5538, + "step": 5692 + }, + { + "epoch": 4.71314191054666, + "grad_norm": 0.9680503010749817, + "learning_rate": 5.582197801841369e-07, + "loss": 0.5859, + "step": 5693 + }, + { + "epoch": 4.713970182219768, + "grad_norm": 0.9708214402198792, + "learning_rate": 5.575273223200265e-07, + "loss": 0.556, + "step": 5694 + }, + { + "epoch": 4.714798453892877, + "grad_norm": 0.989654541015625, + "learning_rate": 5.568352403062721e-07, + "loss": 0.5754, + "step": 5695 + }, + { + "epoch": 4.715626725565985, + "grad_norm": 0.9700330495834351, + "learning_rate": 5.56143534276784e-07, + "loss": 0.5514, + "step": 5696 + }, + { + "epoch": 4.716454997239095, + "grad_norm": 0.9729028940200806, + "learning_rate": 5.554522043654017e-07, + "loss": 0.562, + "step": 5697 + }, + { + "epoch": 4.717283268912203, + "grad_norm": 0.9634988307952881, + "learning_rate": 5.54761250705892e-07, + "loss": 0.5461, + "step": 5698 + }, + { + "epoch": 4.718111540585312, + "grad_norm": 1.1048837900161743, + "learning_rate": 5.54070673431946e-07, + "loss": 0.5833, + "step": 5699 + }, + { + "epoch": 4.718939812258421, + "grad_norm": 0.9602149724960327, + "learning_rate": 5.533804726771858e-07, + "loss": 0.5383, + "step": 5700 + }, + { + "epoch": 4.7197680839315295, + "grad_norm": 0.9645217061042786, + "learning_rate": 5.52690648575157e-07, + "loss": 0.5617, + "step": 5701 + }, + { + "epoch": 4.720596355604639, + "grad_norm": 0.9855457544326782, + "learning_rate": 5.520012012593362e-07, + "loss": 0.5646, + "step": 5702 + }, + { + "epoch": 4.721424627277747, + "grad_norm": 0.9794975519180298, + "learning_rate": 5.513121308631231e-07, + "loss": 0.5651, + "step": 5703 + }, + { + "epoch": 4.722252898950856, + "grad_norm": 0.9841457605361938, + "learning_rate": 5.506234375198472e-07, + "loss": 0.5477, + "step": 5704 + }, + { + "epoch": 4.723081170623964, + "grad_norm": 0.9675942063331604, + "learning_rate": 5.499351213627649e-07, + "loss": 0.5559, + "step": 5705 + }, + { + "epoch": 4.7239094422970735, + "grad_norm": 0.9632468223571777, + "learning_rate": 5.492471825250579e-07, + "loss": 0.562, + "step": 5706 + }, + { + "epoch": 4.724737713970182, + "grad_norm": 0.9672477841377258, + "learning_rate": 5.485596211398361e-07, + "loss": 0.5453, + "step": 5707 + }, + { + "epoch": 4.725565985643291, + "grad_norm": 0.9824864268302917, + "learning_rate": 5.478724373401372e-07, + "loss": 0.5599, + "step": 5708 + }, + { + "epoch": 4.7263942573164, + "grad_norm": 0.9423843026161194, + "learning_rate": 5.471856312589238e-07, + "loss": 0.5567, + "step": 5709 + }, + { + "epoch": 4.7272225289895085, + "grad_norm": 0.9732121229171753, + "learning_rate": 5.464992030290866e-07, + "loss": 0.5575, + "step": 5710 + }, + { + "epoch": 4.728050800662618, + "grad_norm": 0.9717659950256348, + "learning_rate": 5.458131527834448e-07, + "loss": 0.5816, + "step": 5711 + }, + { + "epoch": 4.728879072335726, + "grad_norm": 0.9714568853378296, + "learning_rate": 5.451274806547399e-07, + "loss": 0.5222, + "step": 5712 + }, + { + "epoch": 4.729707344008835, + "grad_norm": 0.9828087091445923, + "learning_rate": 5.444421867756441e-07, + "loss": 0.5726, + "step": 5713 + }, + { + "epoch": 4.730535615681943, + "grad_norm": 0.9719827771186829, + "learning_rate": 5.437572712787567e-07, + "loss": 0.5437, + "step": 5714 + }, + { + "epoch": 4.7313638873550525, + "grad_norm": 1.0056298971176147, + "learning_rate": 5.430727342966008e-07, + "loss": 0.5396, + "step": 5715 + }, + { + "epoch": 4.732192159028161, + "grad_norm": 0.9678478837013245, + "learning_rate": 5.423885759616282e-07, + "loss": 0.5491, + "step": 5716 + }, + { + "epoch": 4.73302043070127, + "grad_norm": 0.9862358570098877, + "learning_rate": 5.417047964062185e-07, + "loss": 0.5332, + "step": 5717 + }, + { + "epoch": 4.733848702374379, + "grad_norm": 0.9755412936210632, + "learning_rate": 5.41021395762675e-07, + "loss": 0.5643, + "step": 5718 + }, + { + "epoch": 4.734676974047487, + "grad_norm": 0.9771009683609009, + "learning_rate": 5.403383741632301e-07, + "loss": 0.5684, + "step": 5719 + }, + { + "epoch": 4.735505245720597, + "grad_norm": 0.9530584812164307, + "learning_rate": 5.396557317400424e-07, + "loss": 0.5446, + "step": 5720 + }, + { + "epoch": 4.736333517393705, + "grad_norm": 0.9681975841522217, + "learning_rate": 5.389734686251962e-07, + "loss": 0.5377, + "step": 5721 + }, + { + "epoch": 4.737161789066814, + "grad_norm": 0.9460469484329224, + "learning_rate": 5.382915849507034e-07, + "loss": 0.5613, + "step": 5722 + }, + { + "epoch": 4.737990060739922, + "grad_norm": 0.9434804320335388, + "learning_rate": 5.376100808485027e-07, + "loss": 0.5524, + "step": 5723 + }, + { + "epoch": 4.7388183324130315, + "grad_norm": 0.9815676212310791, + "learning_rate": 5.36928956450458e-07, + "loss": 0.5479, + "step": 5724 + }, + { + "epoch": 4.73964660408614, + "grad_norm": 0.9858399629592896, + "learning_rate": 5.362482118883602e-07, + "loss": 0.5633, + "step": 5725 + }, + { + "epoch": 4.740474875759249, + "grad_norm": 0.9802773594856262, + "learning_rate": 5.355678472939274e-07, + "loss": 0.5624, + "step": 5726 + }, + { + "epoch": 4.741303147432358, + "grad_norm": 0.9784073829650879, + "learning_rate": 5.348878627988047e-07, + "loss": 0.539, + "step": 5727 + }, + { + "epoch": 4.742131419105466, + "grad_norm": 0.9714169502258301, + "learning_rate": 5.342082585345615e-07, + "loss": 0.558, + "step": 5728 + }, + { + "epoch": 4.742959690778576, + "grad_norm": 0.9876227378845215, + "learning_rate": 5.335290346326952e-07, + "loss": 0.5471, + "step": 5729 + }, + { + "epoch": 4.743787962451684, + "grad_norm": 0.9695437550544739, + "learning_rate": 5.3285019122463e-07, + "loss": 0.5268, + "step": 5730 + }, + { + "epoch": 4.744616234124793, + "grad_norm": 0.9693663120269775, + "learning_rate": 5.32171728441715e-07, + "loss": 0.561, + "step": 5731 + }, + { + "epoch": 4.745444505797901, + "grad_norm": 0.9851357340812683, + "learning_rate": 5.314936464152265e-07, + "loss": 0.5424, + "step": 5732 + }, + { + "epoch": 4.7462727774710105, + "grad_norm": 0.9912320375442505, + "learning_rate": 5.308159452763678e-07, + "loss": 0.563, + "step": 5733 + }, + { + "epoch": 4.747101049144119, + "grad_norm": 0.9956056475639343, + "learning_rate": 5.301386251562674e-07, + "loss": 0.541, + "step": 5734 + }, + { + "epoch": 4.747929320817228, + "grad_norm": 0.9793626666069031, + "learning_rate": 5.294616861859794e-07, + "loss": 0.5488, + "step": 5735 + }, + { + "epoch": 4.748757592490337, + "grad_norm": 0.9587200880050659, + "learning_rate": 5.287851284964865e-07, + "loss": 0.5235, + "step": 5736 + }, + { + "epoch": 4.749585864163445, + "grad_norm": 0.9694571495056152, + "learning_rate": 5.281089522186955e-07, + "loss": 0.5439, + "step": 5737 + }, + { + "epoch": 4.750414135836555, + "grad_norm": 0.9679348468780518, + "learning_rate": 5.274331574834402e-07, + "loss": 0.5468, + "step": 5738 + }, + { + "epoch": 4.751242407509663, + "grad_norm": 0.9581620097160339, + "learning_rate": 5.267577444214819e-07, + "loss": 0.5477, + "step": 5739 + }, + { + "epoch": 4.752070679182772, + "grad_norm": 0.9385791420936584, + "learning_rate": 5.260827131635046e-07, + "loss": 0.5375, + "step": 5740 + }, + { + "epoch": 4.752898950855881, + "grad_norm": 0.9797219634056091, + "learning_rate": 5.254080638401221e-07, + "loss": 0.5565, + "step": 5741 + }, + { + "epoch": 4.7537272225289895, + "grad_norm": 0.9667260646820068, + "learning_rate": 5.247337965818727e-07, + "loss": 0.532, + "step": 5742 + }, + { + "epoch": 4.754555494202099, + "grad_norm": 0.9527150392532349, + "learning_rate": 5.240599115192199e-07, + "loss": 0.5384, + "step": 5743 + }, + { + "epoch": 4.755383765875207, + "grad_norm": 0.9563965201377869, + "learning_rate": 5.23386408782555e-07, + "loss": 0.573, + "step": 5744 + }, + { + "epoch": 4.756212037548316, + "grad_norm": 0.9735510349273682, + "learning_rate": 5.227132885021946e-07, + "loss": 0.5746, + "step": 5745 + }, + { + "epoch": 4.757040309221424, + "grad_norm": 0.9680667519569397, + "learning_rate": 5.220405508083809e-07, + "loss": 0.5615, + "step": 5746 + }, + { + "epoch": 4.757868580894534, + "grad_norm": 0.9715480804443359, + "learning_rate": 5.213681958312817e-07, + "loss": 0.5411, + "step": 5747 + }, + { + "epoch": 4.758696852567642, + "grad_norm": 0.9722656011581421, + "learning_rate": 5.206962237009927e-07, + "loss": 0.5612, + "step": 5748 + }, + { + "epoch": 4.759525124240751, + "grad_norm": 0.9685169458389282, + "learning_rate": 5.200246345475329e-07, + "loss": 0.5602, + "step": 5749 + }, + { + "epoch": 4.76035339591386, + "grad_norm": 0.9914273023605347, + "learning_rate": 5.193534285008492e-07, + "loss": 0.5775, + "step": 5750 + }, + { + "epoch": 4.7611816675869685, + "grad_norm": 1.0119380950927734, + "learning_rate": 5.186826056908146e-07, + "loss": 0.5682, + "step": 5751 + }, + { + "epoch": 4.762009939260078, + "grad_norm": 0.9605881571769714, + "learning_rate": 5.180121662472251e-07, + "loss": 0.5348, + "step": 5752 + }, + { + "epoch": 4.762838210933186, + "grad_norm": 0.9948599934577942, + "learning_rate": 5.173421102998058e-07, + "loss": 0.5665, + "step": 5753 + }, + { + "epoch": 4.763666482606295, + "grad_norm": 0.9638116955757141, + "learning_rate": 5.166724379782067e-07, + "loss": 0.5482, + "step": 5754 + }, + { + "epoch": 4.764494754279403, + "grad_norm": 0.9764114022254944, + "learning_rate": 5.160031494120021e-07, + "loss": 0.57, + "step": 5755 + }, + { + "epoch": 4.765323025952513, + "grad_norm": 0.9698774218559265, + "learning_rate": 5.153342447306939e-07, + "loss": 0.5474, + "step": 5756 + }, + { + "epoch": 4.766151297625621, + "grad_norm": 0.9637375473976135, + "learning_rate": 5.146657240637079e-07, + "loss": 0.5461, + "step": 5757 + }, + { + "epoch": 4.76697956929873, + "grad_norm": 0.9572364091873169, + "learning_rate": 5.139975875403979e-07, + "loss": 0.5437, + "step": 5758 + }, + { + "epoch": 4.767807840971839, + "grad_norm": 0.9601276516914368, + "learning_rate": 5.13329835290041e-07, + "loss": 0.5599, + "step": 5759 + }, + { + "epoch": 4.7686361126449475, + "grad_norm": 0.9581212997436523, + "learning_rate": 5.12662467441841e-07, + "loss": 0.5467, + "step": 5760 + }, + { + "epoch": 4.769464384318057, + "grad_norm": 0.977631688117981, + "learning_rate": 5.119954841249288e-07, + "loss": 0.5216, + "step": 5761 + }, + { + "epoch": 4.770292655991165, + "grad_norm": 0.9808089733123779, + "learning_rate": 5.113288854683579e-07, + "loss": 0.5711, + "step": 5762 + }, + { + "epoch": 4.771120927664274, + "grad_norm": 0.9823766946792603, + "learning_rate": 5.106626716011093e-07, + "loss": 0.5605, + "step": 5763 + }, + { + "epoch": 4.771949199337382, + "grad_norm": 0.962883710861206, + "learning_rate": 5.099968426520902e-07, + "loss": 0.532, + "step": 5764 + }, + { + "epoch": 4.7727774710104915, + "grad_norm": 0.9629702568054199, + "learning_rate": 5.093313987501306e-07, + "loss": 0.5343, + "step": 5765 + }, + { + "epoch": 4.7736057426836, + "grad_norm": 0.9748992323875427, + "learning_rate": 5.086663400239886e-07, + "loss": 0.5577, + "step": 5766 + }, + { + "epoch": 4.774434014356709, + "grad_norm": 0.9772101640701294, + "learning_rate": 5.080016666023476e-07, + "loss": 0.5626, + "step": 5767 + }, + { + "epoch": 4.775262286029818, + "grad_norm": 0.9644051790237427, + "learning_rate": 5.07337378613815e-07, + "loss": 0.5598, + "step": 5768 + }, + { + "epoch": 4.7760905577029265, + "grad_norm": 0.9621604681015015, + "learning_rate": 5.066734761869235e-07, + "loss": 0.5408, + "step": 5769 + }, + { + "epoch": 4.776918829376036, + "grad_norm": 0.9538135528564453, + "learning_rate": 5.060099594501336e-07, + "loss": 0.555, + "step": 5770 + }, + { + "epoch": 4.777747101049144, + "grad_norm": 0.9480177760124207, + "learning_rate": 5.05346828531828e-07, + "loss": 0.5527, + "step": 5771 + }, + { + "epoch": 4.778575372722253, + "grad_norm": 0.9961148500442505, + "learning_rate": 5.04684083560317e-07, + "loss": 0.5682, + "step": 5772 + }, + { + "epoch": 4.779403644395361, + "grad_norm": 0.962980329990387, + "learning_rate": 5.040217246638368e-07, + "loss": 0.5622, + "step": 5773 + }, + { + "epoch": 4.7802319160684705, + "grad_norm": 0.9808009266853333, + "learning_rate": 5.033597519705457e-07, + "loss": 0.563, + "step": 5774 + }, + { + "epoch": 4.781060187741579, + "grad_norm": 0.9771943688392639, + "learning_rate": 5.0269816560853e-07, + "loss": 0.5694, + "step": 5775 + }, + { + "epoch": 4.781888459414688, + "grad_norm": 0.9685245752334595, + "learning_rate": 5.020369657058016e-07, + "loss": 0.5542, + "step": 5776 + }, + { + "epoch": 4.782716731087797, + "grad_norm": 0.9657905697822571, + "learning_rate": 5.013761523902949e-07, + "loss": 0.5695, + "step": 5777 + }, + { + "epoch": 4.783545002760905, + "grad_norm": 0.9332074522972107, + "learning_rate": 5.007157257898715e-07, + "loss": 0.5392, + "step": 5778 + }, + { + "epoch": 4.784373274434015, + "grad_norm": 0.9893274307250977, + "learning_rate": 5.000556860323191e-07, + "loss": 0.5357, + "step": 5779 + }, + { + "epoch": 4.785201546107123, + "grad_norm": 0.9868603348731995, + "learning_rate": 4.993960332453479e-07, + "loss": 0.5525, + "step": 5780 + }, + { + "epoch": 4.786029817780232, + "grad_norm": 0.9733468294143677, + "learning_rate": 4.987367675565943e-07, + "loss": 0.5552, + "step": 5781 + }, + { + "epoch": 4.78685808945334, + "grad_norm": 0.9812783598899841, + "learning_rate": 4.980778890936208e-07, + "loss": 0.5407, + "step": 5782 + }, + { + "epoch": 4.7876863611264495, + "grad_norm": 0.9803352355957031, + "learning_rate": 4.974193979839146e-07, + "loss": 0.5321, + "step": 5783 + }, + { + "epoch": 4.788514632799558, + "grad_norm": 0.9762402176856995, + "learning_rate": 4.967612943548863e-07, + "loss": 0.5537, + "step": 5784 + }, + { + "epoch": 4.789342904472667, + "grad_norm": 0.955674409866333, + "learning_rate": 4.961035783338739e-07, + "loss": 0.5286, + "step": 5785 + }, + { + "epoch": 4.790171176145776, + "grad_norm": 0.9545698761940002, + "learning_rate": 4.954462500481394e-07, + "loss": 0.5869, + "step": 5786 + }, + { + "epoch": 4.790999447818884, + "grad_norm": 0.9490817189216614, + "learning_rate": 4.947893096248688e-07, + "loss": 0.5462, + "step": 5787 + }, + { + "epoch": 4.791827719491994, + "grad_norm": 0.971339225769043, + "learning_rate": 4.941327571911742e-07, + "loss": 0.5651, + "step": 5788 + }, + { + "epoch": 4.792655991165102, + "grad_norm": 0.9914401173591614, + "learning_rate": 4.934765928740933e-07, + "loss": 0.5396, + "step": 5789 + }, + { + "epoch": 4.793484262838211, + "grad_norm": 1.0018508434295654, + "learning_rate": 4.928208168005874e-07, + "loss": 0.5297, + "step": 5790 + }, + { + "epoch": 4.794312534511319, + "grad_norm": 0.9785000085830688, + "learning_rate": 4.92165429097542e-07, + "loss": 0.56, + "step": 5791 + }, + { + "epoch": 4.7951408061844285, + "grad_norm": 0.9545918703079224, + "learning_rate": 4.9151042989177e-07, + "loss": 0.549, + "step": 5792 + }, + { + "epoch": 4.795969077857537, + "grad_norm": 0.9696705341339111, + "learning_rate": 4.908558193100061e-07, + "loss": 0.5548, + "step": 5793 + }, + { + "epoch": 4.796797349530646, + "grad_norm": 0.9759218692779541, + "learning_rate": 4.902015974789121e-07, + "loss": 0.5634, + "step": 5794 + }, + { + "epoch": 4.797625621203755, + "grad_norm": 0.9847009181976318, + "learning_rate": 4.895477645250749e-07, + "loss": 0.5636, + "step": 5795 + }, + { + "epoch": 4.798453892876863, + "grad_norm": 1.0215561389923096, + "learning_rate": 4.88894320575003e-07, + "loss": 0.5572, + "step": 5796 + }, + { + "epoch": 4.799282164549973, + "grad_norm": 0.9767036437988281, + "learning_rate": 4.882412657551331e-07, + "loss": 0.5488, + "step": 5797 + }, + { + "epoch": 4.800110436223081, + "grad_norm": 0.979062557220459, + "learning_rate": 4.875886001918254e-07, + "loss": 0.5522, + "step": 5798 + }, + { + "epoch": 4.80093870789619, + "grad_norm": 0.9697179794311523, + "learning_rate": 4.869363240113634e-07, + "loss": 0.553, + "step": 5799 + }, + { + "epoch": 4.801766979569299, + "grad_norm": 0.977159321308136, + "learning_rate": 4.862844373399572e-07, + "loss": 0.5451, + "step": 5800 + }, + { + "epoch": 4.8025952512424075, + "grad_norm": 0.9752582907676697, + "learning_rate": 4.856329403037413e-07, + "loss": 0.5296, + "step": 5801 + }, + { + "epoch": 4.803423522915517, + "grad_norm": 0.9821684956550598, + "learning_rate": 4.849818330287739e-07, + "loss": 0.5437, + "step": 5802 + }, + { + "epoch": 4.804251794588625, + "grad_norm": 0.9562684297561646, + "learning_rate": 4.843311156410374e-07, + "loss": 0.577, + "step": 5803 + }, + { + "epoch": 4.805080066261734, + "grad_norm": 0.9621249437332153, + "learning_rate": 4.836807882664401e-07, + "loss": 0.5409, + "step": 5804 + }, + { + "epoch": 4.805908337934842, + "grad_norm": 0.9560291171073914, + "learning_rate": 4.830308510308151e-07, + "loss": 0.5667, + "step": 5805 + }, + { + "epoch": 4.806736609607952, + "grad_norm": 0.9753386378288269, + "learning_rate": 4.823813040599179e-07, + "loss": 0.5615, + "step": 5806 + }, + { + "epoch": 4.80756488128106, + "grad_norm": 0.9691498875617981, + "learning_rate": 4.817321474794304e-07, + "loss": 0.5649, + "step": 5807 + }, + { + "epoch": 4.808393152954169, + "grad_norm": 0.9705379605293274, + "learning_rate": 4.810833814149588e-07, + "loss": 0.5417, + "step": 5808 + }, + { + "epoch": 4.809221424627278, + "grad_norm": 0.9624449014663696, + "learning_rate": 4.804350059920324e-07, + "loss": 0.5458, + "step": 5809 + }, + { + "epoch": 4.8100496963003865, + "grad_norm": 0.9795193672180176, + "learning_rate": 4.797870213361064e-07, + "loss": 0.556, + "step": 5810 + }, + { + "epoch": 4.810877967973496, + "grad_norm": 0.9939634203910828, + "learning_rate": 4.791394275725592e-07, + "loss": 0.5828, + "step": 5811 + }, + { + "epoch": 4.811706239646604, + "grad_norm": 0.9787962436676025, + "learning_rate": 4.784922248266946e-07, + "loss": 0.5475, + "step": 5812 + }, + { + "epoch": 4.812534511319713, + "grad_norm": 1.0147124528884888, + "learning_rate": 4.778454132237409e-07, + "loss": 0.5595, + "step": 5813 + }, + { + "epoch": 4.813362782992821, + "grad_norm": 0.9653933048248291, + "learning_rate": 4.771989928888493e-07, + "loss": 0.552, + "step": 5814 + }, + { + "epoch": 4.814191054665931, + "grad_norm": 0.9682862162590027, + "learning_rate": 4.765529639470956e-07, + "loss": 0.5472, + "step": 5815 + }, + { + "epoch": 4.815019326339039, + "grad_norm": 0.9534293413162231, + "learning_rate": 4.759073265234812e-07, + "loss": 0.5328, + "step": 5816 + }, + { + "epoch": 4.815847598012148, + "grad_norm": 0.9567888379096985, + "learning_rate": 4.7526208074293144e-07, + "loss": 0.53, + "step": 5817 + }, + { + "epoch": 4.816675869685257, + "grad_norm": 0.9659878015518188, + "learning_rate": 4.7461722673029396e-07, + "loss": 0.5409, + "step": 5818 + }, + { + "epoch": 4.8175041413583655, + "grad_norm": 0.979027271270752, + "learning_rate": 4.739727646103426e-07, + "loss": 0.5393, + "step": 5819 + }, + { + "epoch": 4.818332413031475, + "grad_norm": 0.9583582282066345, + "learning_rate": 4.7332869450777586e-07, + "loss": 0.5438, + "step": 5820 + }, + { + "epoch": 4.819160684704583, + "grad_norm": 0.9865652322769165, + "learning_rate": 4.7268501654721357e-07, + "loss": 0.5675, + "step": 5821 + }, + { + "epoch": 4.819988956377692, + "grad_norm": 0.9573209881782532, + "learning_rate": 4.7204173085320185e-07, + "loss": 0.545, + "step": 5822 + }, + { + "epoch": 4.8208172280508, + "grad_norm": 0.9714416861534119, + "learning_rate": 4.713988375502118e-07, + "loss": 0.5579, + "step": 5823 + }, + { + "epoch": 4.8216454997239095, + "grad_norm": 1.0049816370010376, + "learning_rate": 4.707563367626364e-07, + "loss": 0.5413, + "step": 5824 + }, + { + "epoch": 4.822473771397018, + "grad_norm": 0.9911128878593445, + "learning_rate": 4.701142286147925e-07, + "loss": 0.5612, + "step": 5825 + }, + { + "epoch": 4.823302043070127, + "grad_norm": 0.9881139397621155, + "learning_rate": 4.6947251323092383e-07, + "loss": 0.5621, + "step": 5826 + }, + { + "epoch": 4.824130314743236, + "grad_norm": 0.9554377794265747, + "learning_rate": 4.6883119073519503e-07, + "loss": 0.5626, + "step": 5827 + }, + { + "epoch": 4.8249585864163445, + "grad_norm": 0.9953138828277588, + "learning_rate": 4.6819026125169643e-07, + "loss": 0.5553, + "step": 5828 + }, + { + "epoch": 4.825786858089454, + "grad_norm": 0.9436317682266235, + "learning_rate": 4.675497249044425e-07, + "loss": 0.539, + "step": 5829 + }, + { + "epoch": 4.826615129762562, + "grad_norm": 0.9757593870162964, + "learning_rate": 4.6690958181736996e-07, + "loss": 0.5421, + "step": 5830 + }, + { + "epoch": 4.827443401435671, + "grad_norm": 0.9557772278785706, + "learning_rate": 4.6626983211434106e-07, + "loss": 0.5442, + "step": 5831 + }, + { + "epoch": 4.828271673108779, + "grad_norm": 0.9633134007453918, + "learning_rate": 4.6563047591914213e-07, + "loss": 0.5711, + "step": 5832 + }, + { + "epoch": 4.8290999447818885, + "grad_norm": 0.964967668056488, + "learning_rate": 4.649915133554814e-07, + "loss": 0.5645, + "step": 5833 + }, + { + "epoch": 4.829928216454997, + "grad_norm": 0.9668264985084534, + "learning_rate": 4.643529445469924e-07, + "loss": 0.584, + "step": 5834 + }, + { + "epoch": 4.830756488128106, + "grad_norm": 0.9745258688926697, + "learning_rate": 4.6371476961723365e-07, + "loss": 0.5695, + "step": 5835 + }, + { + "epoch": 4.831584759801215, + "grad_norm": 0.9560638070106506, + "learning_rate": 4.6307698868968463e-07, + "loss": 0.5413, + "step": 5836 + }, + { + "epoch": 4.832413031474323, + "grad_norm": 0.9573860764503479, + "learning_rate": 4.6243960188775e-07, + "loss": 0.5645, + "step": 5837 + }, + { + "epoch": 4.833241303147433, + "grad_norm": 0.9806252121925354, + "learning_rate": 4.618026093347583e-07, + "loss": 0.5685, + "step": 5838 + }, + { + "epoch": 4.834069574820541, + "grad_norm": 0.974324107170105, + "learning_rate": 4.611660111539626e-07, + "loss": 0.5593, + "step": 5839 + }, + { + "epoch": 4.83489784649365, + "grad_norm": 0.9738810658454895, + "learning_rate": 4.6052980746853717e-07, + "loss": 0.5492, + "step": 5840 + }, + { + "epoch": 4.835726118166758, + "grad_norm": 0.9708226323127747, + "learning_rate": 4.5989399840158236e-07, + "loss": 0.5081, + "step": 5841 + }, + { + "epoch": 4.8365543898398675, + "grad_norm": 0.984955906867981, + "learning_rate": 4.592585840761221e-07, + "loss": 0.5581, + "step": 5842 + }, + { + "epoch": 4.837382661512976, + "grad_norm": 1.006397008895874, + "learning_rate": 4.586235646151013e-07, + "loss": 0.5536, + "step": 5843 + }, + { + "epoch": 4.838210933186085, + "grad_norm": 0.9361645579338074, + "learning_rate": 4.5798894014139146e-07, + "loss": 0.5606, + "step": 5844 + }, + { + "epoch": 4.839039204859194, + "grad_norm": 0.9568001627922058, + "learning_rate": 4.5735471077778704e-07, + "loss": 0.5242, + "step": 5845 + }, + { + "epoch": 4.839867476532302, + "grad_norm": 0.9710729122161865, + "learning_rate": 4.567208766470041e-07, + "loss": 0.5617, + "step": 5846 + }, + { + "epoch": 4.840695748205412, + "grad_norm": 0.9681148529052734, + "learning_rate": 4.560874378716848e-07, + "loss": 0.5502, + "step": 5847 + }, + { + "epoch": 4.84152401987852, + "grad_norm": 0.9655634164810181, + "learning_rate": 4.5545439457439334e-07, + "loss": 0.5729, + "step": 5848 + }, + { + "epoch": 4.842352291551629, + "grad_norm": 0.9855836033821106, + "learning_rate": 4.548217468776167e-07, + "loss": 0.5413, + "step": 5849 + }, + { + "epoch": 4.843180563224737, + "grad_norm": 0.9914452433586121, + "learning_rate": 4.541894949037673e-07, + "loss": 0.5392, + "step": 5850 + }, + { + "epoch": 4.8440088348978465, + "grad_norm": 0.968498945236206, + "learning_rate": 4.5355763877518057e-07, + "loss": 0.5378, + "step": 5851 + }, + { + "epoch": 4.844837106570955, + "grad_norm": 0.9624379873275757, + "learning_rate": 4.5292617861411315e-07, + "loss": 0.5592, + "step": 5852 + }, + { + "epoch": 4.845665378244064, + "grad_norm": 0.9620005488395691, + "learning_rate": 4.522951145427476e-07, + "loss": 0.5406, + "step": 5853 + }, + { + "epoch": 4.846493649917173, + "grad_norm": 0.9529075026512146, + "learning_rate": 4.5166444668318987e-07, + "loss": 0.5506, + "step": 5854 + }, + { + "epoch": 4.847321921590281, + "grad_norm": 0.9744150638580322, + "learning_rate": 4.510341751574665e-07, + "loss": 0.5573, + "step": 5855 + }, + { + "epoch": 4.848150193263391, + "grad_norm": 0.9540391564369202, + "learning_rate": 4.5040430008752994e-07, + "loss": 0.5433, + "step": 5856 + }, + { + "epoch": 4.848978464936499, + "grad_norm": 0.9721823334693909, + "learning_rate": 4.4977482159525606e-07, + "loss": 0.5534, + "step": 5857 + }, + { + "epoch": 4.849806736609608, + "grad_norm": 0.9671435952186584, + "learning_rate": 4.491457398024421e-07, + "loss": 0.5709, + "step": 5858 + }, + { + "epoch": 4.850635008282717, + "grad_norm": 0.9428475499153137, + "learning_rate": 4.4851705483080904e-07, + "loss": 0.5636, + "step": 5859 + }, + { + "epoch": 4.8514632799558255, + "grad_norm": 0.9652672410011292, + "learning_rate": 4.4788876680200213e-07, + "loss": 0.5506, + "step": 5860 + }, + { + "epoch": 4.852291551628935, + "grad_norm": 0.9987141489982605, + "learning_rate": 4.4726087583759016e-07, + "loss": 0.5303, + "step": 5861 + }, + { + "epoch": 4.853119823302043, + "grad_norm": 0.9547122716903687, + "learning_rate": 4.4663338205906235e-07, + "loss": 0.5568, + "step": 5862 + }, + { + "epoch": 4.853948094975152, + "grad_norm": 0.9808024168014526, + "learning_rate": 4.460062855878342e-07, + "loss": 0.5442, + "step": 5863 + }, + { + "epoch": 4.85477636664826, + "grad_norm": 0.9605748057365417, + "learning_rate": 4.4537958654524317e-07, + "loss": 0.5487, + "step": 5864 + }, + { + "epoch": 4.85560463832137, + "grad_norm": 0.9812973737716675, + "learning_rate": 4.4475328505254877e-07, + "loss": 0.5632, + "step": 5865 + }, + { + "epoch": 4.856432909994478, + "grad_norm": 0.9739463329315186, + "learning_rate": 4.441273812309346e-07, + "loss": 0.5485, + "step": 5866 + }, + { + "epoch": 4.857261181667587, + "grad_norm": 0.9558693170547485, + "learning_rate": 4.4350187520150855e-07, + "loss": 0.5565, + "step": 5867 + }, + { + "epoch": 4.858089453340696, + "grad_norm": 0.9696657061576843, + "learning_rate": 4.4287676708529847e-07, + "loss": 0.5425, + "step": 5868 + }, + { + "epoch": 4.8589177250138045, + "grad_norm": 0.9728769659996033, + "learning_rate": 4.422520570032582e-07, + "loss": 0.5544, + "step": 5869 + }, + { + "epoch": 4.859745996686914, + "grad_norm": 0.9446035623550415, + "learning_rate": 4.416277450762629e-07, + "loss": 0.5422, + "step": 5870 + }, + { + "epoch": 4.860574268360022, + "grad_norm": 0.9597954154014587, + "learning_rate": 4.4100383142511056e-07, + "loss": 0.5414, + "step": 5871 + }, + { + "epoch": 4.861402540033131, + "grad_norm": 0.9638952016830444, + "learning_rate": 4.403803161705228e-07, + "loss": 0.5288, + "step": 5872 + }, + { + "epoch": 4.862230811706239, + "grad_norm": 0.9554084539413452, + "learning_rate": 4.3975719943314535e-07, + "loss": 0.5423, + "step": 5873 + }, + { + "epoch": 4.863059083379349, + "grad_norm": 0.9913690686225891, + "learning_rate": 4.391344813335435e-07, + "loss": 0.5735, + "step": 5874 + }, + { + "epoch": 4.863887355052457, + "grad_norm": 0.9613746404647827, + "learning_rate": 4.3851216199220865e-07, + "loss": 0.5544, + "step": 5875 + }, + { + "epoch": 4.864715626725566, + "grad_norm": 0.9566593766212463, + "learning_rate": 4.378902415295538e-07, + "loss": 0.5506, + "step": 5876 + }, + { + "epoch": 4.865543898398675, + "grad_norm": 0.9866978526115417, + "learning_rate": 4.37268720065914e-07, + "loss": 0.5424, + "step": 5877 + }, + { + "epoch": 4.8663721700717835, + "grad_norm": 0.9751846790313721, + "learning_rate": 4.366475977215484e-07, + "loss": 0.5538, + "step": 5878 + }, + { + "epoch": 4.867200441744893, + "grad_norm": 0.9757003784179688, + "learning_rate": 4.3602687461663904e-07, + "loss": 0.5668, + "step": 5879 + }, + { + "epoch": 4.868028713418001, + "grad_norm": 0.9434172511100769, + "learning_rate": 4.3540655087128846e-07, + "loss": 0.5335, + "step": 5880 + }, + { + "epoch": 4.86885698509111, + "grad_norm": 0.9726729989051819, + "learning_rate": 4.3478662660552504e-07, + "loss": 0.5458, + "step": 5881 + }, + { + "epoch": 4.869685256764218, + "grad_norm": 0.9750825762748718, + "learning_rate": 4.3416710193929733e-07, + "loss": 0.5437, + "step": 5882 + }, + { + "epoch": 4.8705135284373275, + "grad_norm": 0.966927170753479, + "learning_rate": 4.3354797699247833e-07, + "loss": 0.5263, + "step": 5883 + }, + { + "epoch": 4.871341800110436, + "grad_norm": 1.003137230873108, + "learning_rate": 4.329292518848621e-07, + "loss": 0.5589, + "step": 5884 + }, + { + "epoch": 4.872170071783545, + "grad_norm": 0.9725276827812195, + "learning_rate": 4.3231092673616635e-07, + "loss": 0.555, + "step": 5885 + }, + { + "epoch": 4.872998343456654, + "grad_norm": 0.964232861995697, + "learning_rate": 4.3169300166603254e-07, + "loss": 0.5486, + "step": 5886 + }, + { + "epoch": 4.8738266151297625, + "grad_norm": 0.963379442691803, + "learning_rate": 4.3107547679402146e-07, + "loss": 0.5606, + "step": 5887 + }, + { + "epoch": 4.874654886802872, + "grad_norm": 0.9966505765914917, + "learning_rate": 4.3045835223961996e-07, + "loss": 0.5534, + "step": 5888 + }, + { + "epoch": 4.87548315847598, + "grad_norm": 0.9489006400108337, + "learning_rate": 4.298416281222348e-07, + "loss": 0.5332, + "step": 5889 + }, + { + "epoch": 4.876311430149089, + "grad_norm": 0.9371119737625122, + "learning_rate": 4.2922530456119686e-07, + "loss": 0.5489, + "step": 5890 + }, + { + "epoch": 4.877139701822197, + "grad_norm": 0.9697653651237488, + "learning_rate": 4.286093816757597e-07, + "loss": 0.5413, + "step": 5891 + }, + { + "epoch": 4.8779679734953065, + "grad_norm": 0.9692448973655701, + "learning_rate": 4.2799385958509773e-07, + "loss": 0.5739, + "step": 5892 + }, + { + "epoch": 4.878796245168415, + "grad_norm": 0.9299832582473755, + "learning_rate": 4.2737873840830835e-07, + "loss": 0.5394, + "step": 5893 + }, + { + "epoch": 4.879624516841524, + "grad_norm": 0.9768351912498474, + "learning_rate": 4.2676401826441225e-07, + "loss": 0.557, + "step": 5894 + }, + { + "epoch": 4.880452788514633, + "grad_norm": 1.0022796392440796, + "learning_rate": 4.2614969927235297e-07, + "loss": 0.5657, + "step": 5895 + }, + { + "epoch": 4.881281060187741, + "grad_norm": 0.9902967810630798, + "learning_rate": 4.255357815509939e-07, + "loss": 0.5514, + "step": 5896 + }, + { + "epoch": 4.882109331860851, + "grad_norm": 0.9812206029891968, + "learning_rate": 4.2492226521912343e-07, + "loss": 0.5581, + "step": 5897 + }, + { + "epoch": 4.882937603533959, + "grad_norm": 0.9587867856025696, + "learning_rate": 4.2430915039545137e-07, + "loss": 0.5603, + "step": 5898 + }, + { + "epoch": 4.883765875207068, + "grad_norm": 0.9800667762756348, + "learning_rate": 4.236964371986088e-07, + "loss": 0.5491, + "step": 5899 + }, + { + "epoch": 4.884594146880176, + "grad_norm": 0.9664431810379028, + "learning_rate": 4.2308412574715075e-07, + "loss": 0.548, + "step": 5900 + }, + { + "epoch": 4.8854224185532855, + "grad_norm": 1.0057932138442993, + "learning_rate": 4.2247221615955405e-07, + "loss": 0.5391, + "step": 5901 + }, + { + "epoch": 4.886250690226394, + "grad_norm": 0.9711607098579407, + "learning_rate": 4.2186070855421675e-07, + "loss": 0.5397, + "step": 5902 + }, + { + "epoch": 4.887078961899503, + "grad_norm": 0.9427178502082825, + "learning_rate": 4.2124960304946035e-07, + "loss": 0.5356, + "step": 5903 + }, + { + "epoch": 4.887907233572612, + "grad_norm": 0.9526419639587402, + "learning_rate": 4.20638899763528e-07, + "loss": 0.5345, + "step": 5904 + }, + { + "epoch": 4.88873550524572, + "grad_norm": 0.9885551333427429, + "learning_rate": 4.200285988145847e-07, + "loss": 0.5338, + "step": 5905 + }, + { + "epoch": 4.88956377691883, + "grad_norm": 0.9751784205436707, + "learning_rate": 4.1941870032071783e-07, + "loss": 0.5315, + "step": 5906 + }, + { + "epoch": 4.890392048591938, + "grad_norm": 0.9707131385803223, + "learning_rate": 4.1880920439993836e-07, + "loss": 0.5608, + "step": 5907 + }, + { + "epoch": 4.891220320265047, + "grad_norm": 0.9756649136543274, + "learning_rate": 4.1820011117017665e-07, + "loss": 0.5664, + "step": 5908 + }, + { + "epoch": 4.892048591938155, + "grad_norm": 0.977478563785553, + "learning_rate": 4.17591420749287e-07, + "loss": 0.5597, + "step": 5909 + }, + { + "epoch": 4.8928768636112645, + "grad_norm": 0.9484930634498596, + "learning_rate": 4.169831332550464e-07, + "loss": 0.5419, + "step": 5910 + }, + { + "epoch": 4.893705135284373, + "grad_norm": 1.025490403175354, + "learning_rate": 4.163752488051509e-07, + "loss": 0.5608, + "step": 5911 + }, + { + "epoch": 4.894533406957482, + "grad_norm": 0.9627198576927185, + "learning_rate": 4.157677675172217e-07, + "loss": 0.5455, + "step": 5912 + }, + { + "epoch": 4.895361678630591, + "grad_norm": 0.9808639287948608, + "learning_rate": 4.151606895088009e-07, + "loss": 0.5497, + "step": 5913 + }, + { + "epoch": 4.896189950303699, + "grad_norm": 0.993675172328949, + "learning_rate": 4.1455401489735175e-07, + "loss": 0.5649, + "step": 5914 + }, + { + "epoch": 4.897018221976809, + "grad_norm": 0.9799006581306458, + "learning_rate": 4.139477438002606e-07, + "loss": 0.5444, + "step": 5915 + }, + { + "epoch": 4.897846493649917, + "grad_norm": 1.0184911489486694, + "learning_rate": 4.1334187633483487e-07, + "loss": 0.5565, + "step": 5916 + }, + { + "epoch": 4.898674765323026, + "grad_norm": 0.9810110330581665, + "learning_rate": 4.12736412618305e-07, + "loss": 0.5557, + "step": 5917 + }, + { + "epoch": 4.899503036996135, + "grad_norm": 0.9775555729866028, + "learning_rate": 4.121313527678214e-07, + "loss": 0.5748, + "step": 5918 + }, + { + "epoch": 4.9003313086692435, + "grad_norm": 0.9573676586151123, + "learning_rate": 4.1152669690045834e-07, + "loss": 0.5613, + "step": 5919 + }, + { + "epoch": 4.901159580342353, + "grad_norm": 0.9636746048927307, + "learning_rate": 4.109224451332114e-07, + "loss": 0.5407, + "step": 5920 + }, + { + "epoch": 4.901987852015461, + "grad_norm": 0.9665216207504272, + "learning_rate": 4.103185975829965e-07, + "loss": 0.5494, + "step": 5921 + }, + { + "epoch": 4.90281612368857, + "grad_norm": 0.9665354490280151, + "learning_rate": 4.0971515436665334e-07, + "loss": 0.5619, + "step": 5922 + }, + { + "epoch": 4.903644395361678, + "grad_norm": 0.9628219604492188, + "learning_rate": 4.091121156009428e-07, + "loss": 0.5632, + "step": 5923 + }, + { + "epoch": 4.904472667034788, + "grad_norm": 0.9701488614082336, + "learning_rate": 4.085094814025467e-07, + "loss": 0.5579, + "step": 5924 + }, + { + "epoch": 4.905300938707896, + "grad_norm": 0.9564261436462402, + "learning_rate": 4.079072518880689e-07, + "loss": 0.5508, + "step": 5925 + }, + { + "epoch": 4.906129210381005, + "grad_norm": 0.9657530188560486, + "learning_rate": 4.0730542717403705e-07, + "loss": 0.551, + "step": 5926 + }, + { + "epoch": 4.906957482054114, + "grad_norm": 0.9382011294364929, + "learning_rate": 4.0670400737689603e-07, + "loss": 0.5348, + "step": 5927 + }, + { + "epoch": 4.9077857537272225, + "grad_norm": 0.9636839032173157, + "learning_rate": 4.0610299261301613e-07, + "loss": 0.5472, + "step": 5928 + }, + { + "epoch": 4.908614025400332, + "grad_norm": 0.9654035568237305, + "learning_rate": 4.0550238299868846e-07, + "loss": 0.5599, + "step": 5929 + }, + { + "epoch": 4.90944229707344, + "grad_norm": 0.9554421305656433, + "learning_rate": 4.049021786501245e-07, + "loss": 0.5814, + "step": 5930 + }, + { + "epoch": 4.910270568746549, + "grad_norm": 0.9651699662208557, + "learning_rate": 4.0430237968345886e-07, + "loss": 0.5623, + "step": 5931 + }, + { + "epoch": 4.911098840419657, + "grad_norm": 1.0203036069869995, + "learning_rate": 4.0370298621474734e-07, + "loss": 0.5425, + "step": 5932 + }, + { + "epoch": 4.911927112092767, + "grad_norm": 0.9871775507926941, + "learning_rate": 4.0310399835996603e-07, + "loss": 0.5274, + "step": 5933 + }, + { + "epoch": 4.912755383765875, + "grad_norm": 0.9446064233779907, + "learning_rate": 4.0250541623501423e-07, + "loss": 0.5299, + "step": 5934 + }, + { + "epoch": 4.913583655438984, + "grad_norm": 0.9630985260009766, + "learning_rate": 4.019072399557122e-07, + "loss": 0.5482, + "step": 5935 + }, + { + "epoch": 4.914411927112093, + "grad_norm": 0.9475951194763184, + "learning_rate": 4.013094696378006e-07, + "loss": 0.5531, + "step": 5936 + }, + { + "epoch": 4.9152401987852015, + "grad_norm": 0.9971574544906616, + "learning_rate": 4.0071210539694315e-07, + "loss": 0.5509, + "step": 5937 + }, + { + "epoch": 4.916068470458311, + "grad_norm": 0.949856698513031, + "learning_rate": 4.001151473487236e-07, + "loss": 0.5396, + "step": 5938 + }, + { + "epoch": 4.916896742131419, + "grad_norm": 0.9867436289787292, + "learning_rate": 3.995185956086489e-07, + "loss": 0.5769, + "step": 5939 + }, + { + "epoch": 4.917725013804528, + "grad_norm": 0.9721899628639221, + "learning_rate": 3.989224502921446e-07, + "loss": 0.5421, + "step": 5940 + }, + { + "epoch": 4.918553285477636, + "grad_norm": 0.9489734768867493, + "learning_rate": 3.983267115145603e-07, + "loss": 0.5601, + "step": 5941 + }, + { + "epoch": 4.9193815571507455, + "grad_norm": 0.9731948971748352, + "learning_rate": 3.977313793911666e-07, + "loss": 0.5656, + "step": 5942 + }, + { + "epoch": 4.920209828823854, + "grad_norm": 1.0035384893417358, + "learning_rate": 3.9713645403715306e-07, + "loss": 0.5492, + "step": 5943 + }, + { + "epoch": 4.921038100496963, + "grad_norm": 0.9753653407096863, + "learning_rate": 3.965419355676331e-07, + "loss": 0.5454, + "step": 5944 + }, + { + "epoch": 4.921866372170072, + "grad_norm": 0.9717726111412048, + "learning_rate": 3.959478240976411e-07, + "loss": 0.5512, + "step": 5945 + }, + { + "epoch": 4.9226946438431805, + "grad_norm": 0.9888126254081726, + "learning_rate": 3.953541197421307e-07, + "loss": 0.5485, + "step": 5946 + }, + { + "epoch": 4.92352291551629, + "grad_norm": 0.9688169956207275, + "learning_rate": 3.947608226159788e-07, + "loss": 0.5566, + "step": 5947 + }, + { + "epoch": 4.924351187189398, + "grad_norm": 0.9560792446136475, + "learning_rate": 3.9416793283398367e-07, + "loss": 0.5592, + "step": 5948 + }, + { + "epoch": 4.925179458862507, + "grad_norm": 0.9747999310493469, + "learning_rate": 3.93575450510863e-07, + "loss": 0.5569, + "step": 5949 + }, + { + "epoch": 4.926007730535615, + "grad_norm": 0.9653720259666443, + "learning_rate": 3.929833757612564e-07, + "loss": 0.5546, + "step": 5950 + }, + { + "epoch": 4.9268360022087245, + "grad_norm": 0.950875461101532, + "learning_rate": 3.9239170869972585e-07, + "loss": 0.5583, + "step": 5951 + }, + { + "epoch": 4.927664273881833, + "grad_norm": 0.9724762439727783, + "learning_rate": 3.9180044944075205e-07, + "loss": 0.5683, + "step": 5952 + }, + { + "epoch": 4.928492545554942, + "grad_norm": 0.9594883322715759, + "learning_rate": 3.9120959809873905e-07, + "loss": 0.5444, + "step": 5953 + }, + { + "epoch": 4.929320817228051, + "grad_norm": 0.9946019053459167, + "learning_rate": 3.906191547880114e-07, + "loss": 0.5701, + "step": 5954 + }, + { + "epoch": 4.930149088901159, + "grad_norm": 0.9909894466400146, + "learning_rate": 3.900291196228134e-07, + "loss": 0.5607, + "step": 5955 + }, + { + "epoch": 4.930977360574269, + "grad_norm": 0.9909331798553467, + "learning_rate": 3.8943949271731174e-07, + "loss": 0.5438, + "step": 5956 + }, + { + "epoch": 4.931805632247377, + "grad_norm": 0.9559661149978638, + "learning_rate": 3.888502741855946e-07, + "loss": 0.5707, + "step": 5957 + }, + { + "epoch": 4.932633903920486, + "grad_norm": 0.9508602619171143, + "learning_rate": 3.8826146414166904e-07, + "loss": 0.5349, + "step": 5958 + }, + { + "epoch": 4.933462175593594, + "grad_norm": 0.9860195517539978, + "learning_rate": 3.876730626994646e-07, + "loss": 0.5555, + "step": 5959 + }, + { + "epoch": 4.9342904472667035, + "grad_norm": 0.9599242210388184, + "learning_rate": 3.8708506997283207e-07, + "loss": 0.5595, + "step": 5960 + }, + { + "epoch": 4.935118718939812, + "grad_norm": 0.9686924815177917, + "learning_rate": 3.8649748607554243e-07, + "loss": 0.5649, + "step": 5961 + }, + { + "epoch": 4.935946990612921, + "grad_norm": 1.034990906715393, + "learning_rate": 3.859103111212867e-07, + "loss": 0.5693, + "step": 5962 + }, + { + "epoch": 4.93677526228603, + "grad_norm": 1.064138412475586, + "learning_rate": 3.8532354522367923e-07, + "loss": 0.5433, + "step": 5963 + }, + { + "epoch": 4.937603533959138, + "grad_norm": 0.9807191491127014, + "learning_rate": 3.8473718849625226e-07, + "loss": 0.5467, + "step": 5964 + }, + { + "epoch": 4.938431805632248, + "grad_norm": 0.9689276814460754, + "learning_rate": 3.8415124105246115e-07, + "loss": 0.5456, + "step": 5965 + }, + { + "epoch": 4.939260077305356, + "grad_norm": 0.9509407877922058, + "learning_rate": 3.8356570300568165e-07, + "loss": 0.5439, + "step": 5966 + }, + { + "epoch": 4.940088348978465, + "grad_norm": 0.9513766169548035, + "learning_rate": 3.829805744692089e-07, + "loss": 0.5442, + "step": 5967 + }, + { + "epoch": 4.940916620651573, + "grad_norm": 0.971613883972168, + "learning_rate": 3.823958555562607e-07, + "loss": 0.5458, + "step": 5968 + }, + { + "epoch": 4.9417448923246825, + "grad_norm": 0.9698891043663025, + "learning_rate": 3.818115463799746e-07, + "loss": 0.5406, + "step": 5969 + }, + { + "epoch": 4.942573163997791, + "grad_norm": 0.9531323313713074, + "learning_rate": 3.8122764705340847e-07, + "loss": 0.5614, + "step": 5970 + }, + { + "epoch": 4.9434014356709, + "grad_norm": 0.9560294151306152, + "learning_rate": 3.80644157689542e-07, + "loss": 0.546, + "step": 5971 + }, + { + "epoch": 4.944229707344009, + "grad_norm": 0.9792264699935913, + "learning_rate": 3.8006107840127426e-07, + "loss": 0.5458, + "step": 5972 + }, + { + "epoch": 4.945057979017117, + "grad_norm": 0.955904483795166, + "learning_rate": 3.794784093014264e-07, + "loss": 0.5389, + "step": 5973 + }, + { + "epoch": 4.945886250690227, + "grad_norm": 0.9566978812217712, + "learning_rate": 3.788961505027389e-07, + "loss": 0.5482, + "step": 5974 + }, + { + "epoch": 4.946714522363335, + "grad_norm": 0.9445267915725708, + "learning_rate": 3.783143021178731e-07, + "loss": 0.5467, + "step": 5975 + }, + { + "epoch": 4.947542794036444, + "grad_norm": 0.9731202125549316, + "learning_rate": 3.7773286425941273e-07, + "loss": 0.5376, + "step": 5976 + }, + { + "epoch": 4.948371065709553, + "grad_norm": 0.9669923186302185, + "learning_rate": 3.7715183703985883e-07, + "loss": 0.5353, + "step": 5977 + }, + { + "epoch": 4.9491993373826615, + "grad_norm": 0.955851137638092, + "learning_rate": 3.765712205716357e-07, + "loss": 0.5221, + "step": 5978 + }, + { + "epoch": 4.950027609055771, + "grad_norm": 0.9659578800201416, + "learning_rate": 3.7599101496708726e-07, + "loss": 0.5743, + "step": 5979 + }, + { + "epoch": 4.950855880728879, + "grad_norm": 0.965240478515625, + "learning_rate": 3.754112203384774e-07, + "loss": 0.5762, + "step": 5980 + }, + { + "epoch": 4.951684152401988, + "grad_norm": 2.1109437942504883, + "learning_rate": 3.7483183679799113e-07, + "loss": 0.5611, + "step": 5981 + }, + { + "epoch": 4.952512424075096, + "grad_norm": 0.9661439061164856, + "learning_rate": 3.742528644577351e-07, + "loss": 0.5452, + "step": 5982 + }, + { + "epoch": 4.953340695748206, + "grad_norm": 0.965039074420929, + "learning_rate": 3.736743034297327e-07, + "loss": 0.5508, + "step": 5983 + }, + { + "epoch": 4.954168967421314, + "grad_norm": 0.9632126688957214, + "learning_rate": 3.730961538259312e-07, + "loss": 0.5651, + "step": 5984 + }, + { + "epoch": 4.954997239094423, + "grad_norm": 0.9725423455238342, + "learning_rate": 3.7251841575819764e-07, + "loss": 0.5571, + "step": 5985 + }, + { + "epoch": 4.955825510767532, + "grad_norm": 0.9642723202705383, + "learning_rate": 3.7194108933831794e-07, + "loss": 0.5604, + "step": 5986 + }, + { + "epoch": 4.9566537824406405, + "grad_norm": 0.9578214287757874, + "learning_rate": 3.71364174678e-07, + "loss": 0.5249, + "step": 5987 + }, + { + "epoch": 4.95748205411375, + "grad_norm": 0.9682602286338806, + "learning_rate": 3.70787671888872e-07, + "loss": 0.5356, + "step": 5988 + }, + { + "epoch": 4.958310325786858, + "grad_norm": 0.9686048626899719, + "learning_rate": 3.702115810824805e-07, + "loss": 0.5436, + "step": 5989 + }, + { + "epoch": 4.959138597459967, + "grad_norm": 1.0237302780151367, + "learning_rate": 3.696359023702947e-07, + "loss": 0.5369, + "step": 5990 + }, + { + "epoch": 4.959966869133075, + "grad_norm": 0.9748989939689636, + "learning_rate": 3.6906063586370306e-07, + "loss": 0.5727, + "step": 5991 + }, + { + "epoch": 4.9607951408061846, + "grad_norm": 0.972263514995575, + "learning_rate": 3.684857816740137e-07, + "loss": 0.5969, + "step": 5992 + }, + { + "epoch": 4.961623412479293, + "grad_norm": 0.979120671749115, + "learning_rate": 3.6791133991245566e-07, + "loss": 0.5463, + "step": 5993 + }, + { + "epoch": 4.962451684152402, + "grad_norm": 0.9615027904510498, + "learning_rate": 3.6733731069017895e-07, + "loss": 0.5722, + "step": 5994 + }, + { + "epoch": 4.963279955825511, + "grad_norm": 0.9743287563323975, + "learning_rate": 3.6676369411825204e-07, + "loss": 0.5388, + "step": 5995 + }, + { + "epoch": 4.9641082274986195, + "grad_norm": 0.9608129858970642, + "learning_rate": 3.661904903076641e-07, + "loss": 0.5329, + "step": 5996 + }, + { + "epoch": 4.964936499171729, + "grad_norm": 0.9721583724021912, + "learning_rate": 3.656176993693253e-07, + "loss": 0.5473, + "step": 5997 + }, + { + "epoch": 4.965764770844837, + "grad_norm": 0.9767399430274963, + "learning_rate": 3.650453214140656e-07, + "loss": 0.5293, + "step": 5998 + }, + { + "epoch": 4.966593042517946, + "grad_norm": 0.9559426307678223, + "learning_rate": 3.6447335655263397e-07, + "loss": 0.5484, + "step": 5999 + }, + { + "epoch": 4.967421314191054, + "grad_norm": 0.9602168798446655, + "learning_rate": 3.639018048957005e-07, + "loss": 0.5579, + "step": 6000 + }, + { + "epoch": 4.9682495858641635, + "grad_norm": 0.9681665897369385, + "learning_rate": 3.633306665538561e-07, + "loss": 0.5628, + "step": 6001 + }, + { + "epoch": 4.969077857537272, + "grad_norm": 0.9520401954650879, + "learning_rate": 3.627599416376093e-07, + "loss": 0.5504, + "step": 6002 + }, + { + "epoch": 4.969906129210381, + "grad_norm": 0.9589652419090271, + "learning_rate": 3.6218963025739073e-07, + "loss": 0.5507, + "step": 6003 + }, + { + "epoch": 4.97073440088349, + "grad_norm": 0.977666974067688, + "learning_rate": 3.616197325235507e-07, + "loss": 0.5504, + "step": 6004 + }, + { + "epoch": 4.9715626725565985, + "grad_norm": 0.98526930809021, + "learning_rate": 3.6105024854635903e-07, + "loss": 0.5393, + "step": 6005 + }, + { + "epoch": 4.972390944229708, + "grad_norm": 0.9593404531478882, + "learning_rate": 3.6048117843600433e-07, + "loss": 0.5707, + "step": 6006 + }, + { + "epoch": 4.973219215902816, + "grad_norm": 0.9677101373672485, + "learning_rate": 3.5991252230259817e-07, + "loss": 0.5559, + "step": 6007 + }, + { + "epoch": 4.974047487575925, + "grad_norm": 0.9906893372535706, + "learning_rate": 3.5934428025616833e-07, + "loss": 0.5562, + "step": 6008 + }, + { + "epoch": 4.974875759249033, + "grad_norm": 0.9470625519752502, + "learning_rate": 3.587764524066656e-07, + "loss": 0.5308, + "step": 6009 + }, + { + "epoch": 4.9757040309221425, + "grad_norm": 0.9528307318687439, + "learning_rate": 3.5820903886395945e-07, + "loss": 0.537, + "step": 6010 + }, + { + "epoch": 4.976532302595251, + "grad_norm": 0.9625613689422607, + "learning_rate": 3.5764203973783854e-07, + "loss": 0.5555, + "step": 6011 + }, + { + "epoch": 4.97736057426836, + "grad_norm": 0.967427134513855, + "learning_rate": 3.570754551380118e-07, + "loss": 0.5514, + "step": 6012 + }, + { + "epoch": 4.978188845941469, + "grad_norm": 0.9919961094856262, + "learning_rate": 3.565092851741092e-07, + "loss": 0.5294, + "step": 6013 + }, + { + "epoch": 4.979017117614577, + "grad_norm": 0.9725098013877869, + "learning_rate": 3.5594352995567793e-07, + "loss": 0.5561, + "step": 6014 + }, + { + "epoch": 4.979845389287687, + "grad_norm": 0.9572508931159973, + "learning_rate": 3.5537818959218677e-07, + "loss": 0.5545, + "step": 6015 + }, + { + "epoch": 4.980673660960795, + "grad_norm": 0.9887001514434814, + "learning_rate": 3.548132641930249e-07, + "loss": 0.5601, + "step": 6016 + }, + { + "epoch": 4.981501932633904, + "grad_norm": 0.9554132223129272, + "learning_rate": 3.542487538674991e-07, + "loss": 0.5558, + "step": 6017 + }, + { + "epoch": 4.982330204307012, + "grad_norm": 0.981720507144928, + "learning_rate": 3.536846587248363e-07, + "loss": 0.5776, + "step": 6018 + }, + { + "epoch": 4.9831584759801215, + "grad_norm": 0.9739376306533813, + "learning_rate": 3.5312097887418435e-07, + "loss": 0.5627, + "step": 6019 + }, + { + "epoch": 4.98398674765323, + "grad_norm": 0.9445295929908752, + "learning_rate": 3.525577144246106e-07, + "loss": 0.5442, + "step": 6020 + }, + { + "epoch": 4.984815019326339, + "grad_norm": 0.9757813811302185, + "learning_rate": 3.5199486548510046e-07, + "loss": 0.5331, + "step": 6021 + }, + { + "epoch": 4.985643290999448, + "grad_norm": 0.9708597660064697, + "learning_rate": 3.514324321645604e-07, + "loss": 0.5758, + "step": 6022 + }, + { + "epoch": 4.986471562672556, + "grad_norm": 0.9770578742027283, + "learning_rate": 3.508704145718167e-07, + "loss": 0.5569, + "step": 6023 + }, + { + "epoch": 4.987299834345666, + "grad_norm": 0.9736425280570984, + "learning_rate": 3.503088128156129e-07, + "loss": 0.5718, + "step": 6024 + }, + { + "epoch": 4.988128106018774, + "grad_norm": 0.9564449787139893, + "learning_rate": 3.4974762700461573e-07, + "loss": 0.5496, + "step": 6025 + }, + { + "epoch": 4.988956377691883, + "grad_norm": 0.9484078288078308, + "learning_rate": 3.4918685724740747e-07, + "loss": 0.5466, + "step": 6026 + }, + { + "epoch": 4.989784649364991, + "grad_norm": 0.9592824578285217, + "learning_rate": 3.486265036524927e-07, + "loss": 0.556, + "step": 6027 + }, + { + "epoch": 4.9906129210381005, + "grad_norm": 0.9514402747154236, + "learning_rate": 3.4806656632829504e-07, + "loss": 0.5421, + "step": 6028 + }, + { + "epoch": 4.991441192711209, + "grad_norm": 0.9617545008659363, + "learning_rate": 3.4750704538315694e-07, + "loss": 0.576, + "step": 6029 + }, + { + "epoch": 4.992269464384318, + "grad_norm": 0.9632974863052368, + "learning_rate": 3.4694794092533983e-07, + "loss": 0.5495, + "step": 6030 + }, + { + "epoch": 4.993097736057427, + "grad_norm": 0.9788113832473755, + "learning_rate": 3.4638925306302556e-07, + "loss": 0.5757, + "step": 6031 + }, + { + "epoch": 4.993926007730535, + "grad_norm": 1.002131700515747, + "learning_rate": 3.4583098190431544e-07, + "loss": 0.5287, + "step": 6032 + }, + { + "epoch": 4.994754279403645, + "grad_norm": 0.9638229012489319, + "learning_rate": 3.452731275572291e-07, + "loss": 0.5421, + "step": 6033 + }, + { + "epoch": 4.995582551076753, + "grad_norm": 1.010887622833252, + "learning_rate": 3.4471569012970633e-07, + "loss": 0.5613, + "step": 6034 + }, + { + "epoch": 4.996410822749862, + "grad_norm": 0.9634838104248047, + "learning_rate": 3.4415866972960704e-07, + "loss": 0.5388, + "step": 6035 + }, + { + "epoch": 4.997239094422971, + "grad_norm": 1.0258270502090454, + "learning_rate": 3.4360206646470775e-07, + "loss": 0.5654, + "step": 6036 + }, + { + "epoch": 4.9980673660960795, + "grad_norm": 0.979051411151886, + "learning_rate": 3.4304588044270725e-07, + "loss": 0.5501, + "step": 6037 + }, + { + "epoch": 4.998895637769189, + "grad_norm": 0.9722236394882202, + "learning_rate": 3.424901117712223e-07, + "loss": 0.5516, + "step": 6038 + }, + { + "epoch": 4.999723909442297, + "grad_norm": 0.9711024761199951, + "learning_rate": 3.419347605577888e-07, + "loss": 0.5486, + "step": 6039 + }, + { + "epoch": 5.0, + "grad_norm": 0.9711024761199951, + "learning_rate": 3.413798269098614e-07, + "loss": 0.5686, + "step": 6040 + }, + { + "epoch": 5.000828271673109, + "grad_norm": 1.8983371257781982, + "learning_rate": 3.4082531093481557e-07, + "loss": 0.5311, + "step": 6041 + }, + { + "epoch": 5.0016565433462175, + "grad_norm": 0.9680057168006897, + "learning_rate": 3.402712127399438e-07, + "loss": 0.5308, + "step": 6042 + }, + { + "epoch": 5.002484815019327, + "grad_norm": 0.955004870891571, + "learning_rate": 3.3971753243245973e-07, + "loss": 0.5256, + "step": 6043 + }, + { + "epoch": 5.003313086692435, + "grad_norm": 0.9773672819137573, + "learning_rate": 3.391642701194961e-07, + "loss": 0.5438, + "step": 6044 + }, + { + "epoch": 5.004141358365544, + "grad_norm": 0.9682551026344299, + "learning_rate": 3.386114259081022e-07, + "loss": 0.5391, + "step": 6045 + }, + { + "epoch": 5.004969630038652, + "grad_norm": 0.9608836770057678, + "learning_rate": 3.3805899990524944e-07, + "loss": 0.5465, + "step": 6046 + }, + { + "epoch": 5.0057979017117615, + "grad_norm": 0.9543644785881042, + "learning_rate": 3.375069922178273e-07, + "loss": 0.5726, + "step": 6047 + }, + { + "epoch": 5.00662617338487, + "grad_norm": 0.9833614826202393, + "learning_rate": 3.3695540295264283e-07, + "loss": 0.5574, + "step": 6048 + }, + { + "epoch": 5.007454445057979, + "grad_norm": 0.9778210520744324, + "learning_rate": 3.364042322164246e-07, + "loss": 0.5446, + "step": 6049 + }, + { + "epoch": 5.008282716731088, + "grad_norm": 0.9737818837165833, + "learning_rate": 3.358534801158192e-07, + "loss": 0.5383, + "step": 6050 + }, + { + "epoch": 5.009110988404196, + "grad_norm": 0.9598701596260071, + "learning_rate": 3.353031467573914e-07, + "loss": 0.5337, + "step": 6051 + }, + { + "epoch": 5.009939260077306, + "grad_norm": 1.0241256952285767, + "learning_rate": 3.347532322476249e-07, + "loss": 0.5556, + "step": 6052 + }, + { + "epoch": 5.010767531750414, + "grad_norm": 0.969450056552887, + "learning_rate": 3.3420373669292405e-07, + "loss": 0.5277, + "step": 6053 + }, + { + "epoch": 5.011595803423523, + "grad_norm": 0.9688952565193176, + "learning_rate": 3.336546601996113e-07, + "loss": 0.5495, + "step": 6054 + }, + { + "epoch": 5.012424075096631, + "grad_norm": 0.9898849129676819, + "learning_rate": 3.3310600287392693e-07, + "loss": 0.5264, + "step": 6055 + }, + { + "epoch": 5.0132523467697405, + "grad_norm": 0.962328314781189, + "learning_rate": 3.325577648220313e-07, + "loss": 0.5458, + "step": 6056 + }, + { + "epoch": 5.01408061844285, + "grad_norm": 0.9720776677131653, + "learning_rate": 3.320099461500043e-07, + "loss": 0.5553, + "step": 6057 + }, + { + "epoch": 5.014908890115958, + "grad_norm": 0.9657216668128967, + "learning_rate": 3.314625469638422e-07, + "loss": 0.5189, + "step": 6058 + }, + { + "epoch": 5.015737161789067, + "grad_norm": 0.9419111013412476, + "learning_rate": 3.309155673694625e-07, + "loss": 0.5652, + "step": 6059 + }, + { + "epoch": 5.016565433462175, + "grad_norm": 0.9682337641716003, + "learning_rate": 3.303690074727012e-07, + "loss": 0.5495, + "step": 6060 + }, + { + "epoch": 5.017393705135285, + "grad_norm": 0.982504665851593, + "learning_rate": 3.2982286737931135e-07, + "loss": 0.5531, + "step": 6061 + }, + { + "epoch": 5.018221976808393, + "grad_norm": 0.9997338056564331, + "learning_rate": 3.292771471949671e-07, + "loss": 0.525, + "step": 6062 + }, + { + "epoch": 5.019050248481502, + "grad_norm": 0.9822561144828796, + "learning_rate": 3.2873184702525956e-07, + "loss": 0.5507, + "step": 6063 + }, + { + "epoch": 5.01987852015461, + "grad_norm": 0.9881250262260437, + "learning_rate": 3.281869669756987e-07, + "loss": 0.5621, + "step": 6064 + }, + { + "epoch": 5.0207067918277195, + "grad_norm": 0.9926158785820007, + "learning_rate": 3.276425071517145e-07, + "loss": 0.5546, + "step": 6065 + }, + { + "epoch": 5.021535063500829, + "grad_norm": 1.0024833679199219, + "learning_rate": 3.2709846765865526e-07, + "loss": 0.5322, + "step": 6066 + }, + { + "epoch": 5.022363335173937, + "grad_norm": 0.9675854444503784, + "learning_rate": 3.265548486017864e-07, + "loss": 0.5288, + "step": 6067 + }, + { + "epoch": 5.023191606847046, + "grad_norm": 0.9715272188186646, + "learning_rate": 3.2601165008629365e-07, + "loss": 0.5456, + "step": 6068 + }, + { + "epoch": 5.024019878520154, + "grad_norm": 0.9742488861083984, + "learning_rate": 3.2546887221728163e-07, + "loss": 0.5561, + "step": 6069 + }, + { + "epoch": 5.024848150193264, + "grad_norm": 0.9488270878791809, + "learning_rate": 3.249265150997713e-07, + "loss": 0.5471, + "step": 6070 + }, + { + "epoch": 5.025676421866372, + "grad_norm": 0.9789018630981445, + "learning_rate": 3.2438457883870485e-07, + "loss": 0.539, + "step": 6071 + }, + { + "epoch": 5.026504693539481, + "grad_norm": 0.9265579581260681, + "learning_rate": 3.23843063538942e-07, + "loss": 0.5143, + "step": 6072 + }, + { + "epoch": 5.027332965212589, + "grad_norm": 0.9542847871780396, + "learning_rate": 3.2330196930526055e-07, + "loss": 0.5381, + "step": 6073 + }, + { + "epoch": 5.0281612368856985, + "grad_norm": 0.9665813446044922, + "learning_rate": 3.2276129624235625e-07, + "loss": 0.5297, + "step": 6074 + }, + { + "epoch": 5.028989508558808, + "grad_norm": 0.9629183411598206, + "learning_rate": 3.2222104445484543e-07, + "loss": 0.5547, + "step": 6075 + }, + { + "epoch": 5.029817780231916, + "grad_norm": 0.9836278557777405, + "learning_rate": 3.2168121404726223e-07, + "loss": 0.5538, + "step": 6076 + }, + { + "epoch": 5.030646051905025, + "grad_norm": 0.987966001033783, + "learning_rate": 3.2114180512405725e-07, + "loss": 0.5377, + "step": 6077 + }, + { + "epoch": 5.031474323578133, + "grad_norm": 0.9594748616218567, + "learning_rate": 3.20602817789602e-07, + "loss": 0.5449, + "step": 6078 + }, + { + "epoch": 5.032302595251243, + "grad_norm": 0.9832162857055664, + "learning_rate": 3.2006425214818586e-07, + "loss": 0.5423, + "step": 6079 + }, + { + "epoch": 5.033130866924351, + "grad_norm": 0.976398766040802, + "learning_rate": 3.195261083040155e-07, + "loss": 0.5497, + "step": 6080 + }, + { + "epoch": 5.03395913859746, + "grad_norm": 0.97797691822052, + "learning_rate": 3.18988386361217e-07, + "loss": 0.5558, + "step": 6081 + }, + { + "epoch": 5.034787410270568, + "grad_norm": 0.9641791582107544, + "learning_rate": 3.184510864238352e-07, + "loss": 0.5308, + "step": 6082 + }, + { + "epoch": 5.0356156819436775, + "grad_norm": 0.9750100374221802, + "learning_rate": 3.179142085958312e-07, + "loss": 0.5324, + "step": 6083 + }, + { + "epoch": 5.036443953616787, + "grad_norm": 0.9927523136138916, + "learning_rate": 3.173777529810876e-07, + "loss": 0.5517, + "step": 6084 + }, + { + "epoch": 5.037272225289895, + "grad_norm": 0.975371778011322, + "learning_rate": 3.1684171968340255e-07, + "loss": 0.5495, + "step": 6085 + }, + { + "epoch": 5.038100496963004, + "grad_norm": 0.9746067523956299, + "learning_rate": 3.163061088064928e-07, + "loss": 0.5581, + "step": 6086 + }, + { + "epoch": 5.038928768636112, + "grad_norm": 0.9755904674530029, + "learning_rate": 3.1577092045399524e-07, + "loss": 0.5555, + "step": 6087 + }, + { + "epoch": 5.039757040309222, + "grad_norm": 0.9534891843795776, + "learning_rate": 3.1523615472946386e-07, + "loss": 0.5317, + "step": 6088 + }, + { + "epoch": 5.04058531198233, + "grad_norm": 0.9799898266792297, + "learning_rate": 3.147018117363698e-07, + "loss": 0.5519, + "step": 6089 + }, + { + "epoch": 5.041413583655439, + "grad_norm": 0.9911451935768127, + "learning_rate": 3.1416789157810405e-07, + "loss": 0.5569, + "step": 6090 + }, + { + "epoch": 5.042241855328548, + "grad_norm": 0.9995951056480408, + "learning_rate": 3.1363439435797586e-07, + "loss": 0.5591, + "step": 6091 + }, + { + "epoch": 5.0430701270016565, + "grad_norm": 0.9571877121925354, + "learning_rate": 3.131013201792107e-07, + "loss": 0.5565, + "step": 6092 + }, + { + "epoch": 5.043898398674766, + "grad_norm": 0.9767590761184692, + "learning_rate": 3.125686691449542e-07, + "loss": 0.5491, + "step": 6093 + }, + { + "epoch": 5.044726670347874, + "grad_norm": 0.9946463704109192, + "learning_rate": 3.1203644135826967e-07, + "loss": 0.5382, + "step": 6094 + }, + { + "epoch": 5.045554942020983, + "grad_norm": 0.9605308175086975, + "learning_rate": 3.1150463692213776e-07, + "loss": 0.5343, + "step": 6095 + }, + { + "epoch": 5.046383213694091, + "grad_norm": 0.9933757781982422, + "learning_rate": 3.1097325593945736e-07, + "loss": 0.5543, + "step": 6096 + }, + { + "epoch": 5.0472114853672005, + "grad_norm": 0.9608505368232727, + "learning_rate": 3.104422985130459e-07, + "loss": 0.5575, + "step": 6097 + }, + { + "epoch": 5.048039757040309, + "grad_norm": 0.9656239748001099, + "learning_rate": 3.0991176474563955e-07, + "loss": 0.5287, + "step": 6098 + }, + { + "epoch": 5.048868028713418, + "grad_norm": 0.9719324111938477, + "learning_rate": 3.0938165473989037e-07, + "loss": 0.5446, + "step": 6099 + }, + { + "epoch": 5.049696300386527, + "grad_norm": 0.963914692401886, + "learning_rate": 3.088519685983704e-07, + "loss": 0.5409, + "step": 6100 + }, + { + "epoch": 5.0505245720596355, + "grad_norm": 0.9715657830238342, + "learning_rate": 3.0832270642356943e-07, + "loss": 0.5436, + "step": 6101 + }, + { + "epoch": 5.051352843732745, + "grad_norm": 0.9748386144638062, + "learning_rate": 3.0779386831789356e-07, + "loss": 0.5442, + "step": 6102 + }, + { + "epoch": 5.052181115405853, + "grad_norm": 0.9816659688949585, + "learning_rate": 3.0726545438366924e-07, + "loss": 0.546, + "step": 6103 + }, + { + "epoch": 5.053009387078962, + "grad_norm": 0.9797165393829346, + "learning_rate": 3.0673746472313857e-07, + "loss": 0.5369, + "step": 6104 + }, + { + "epoch": 5.05383765875207, + "grad_norm": 0.9616203904151917, + "learning_rate": 3.0620989943846294e-07, + "loss": 0.5467, + "step": 6105 + }, + { + "epoch": 5.0546659304251795, + "grad_norm": 0.9547003507614136, + "learning_rate": 3.056827586317224e-07, + "loss": 0.5673, + "step": 6106 + }, + { + "epoch": 5.055494202098288, + "grad_norm": 0.9676609039306641, + "learning_rate": 3.051560424049127e-07, + "loss": 0.5676, + "step": 6107 + }, + { + "epoch": 5.056322473771397, + "grad_norm": 0.9564599990844727, + "learning_rate": 3.046297508599483e-07, + "loss": 0.5357, + "step": 6108 + }, + { + "epoch": 5.057150745444506, + "grad_norm": 0.9876315593719482, + "learning_rate": 3.041038840986621e-07, + "loss": 0.5472, + "step": 6109 + }, + { + "epoch": 5.057979017117614, + "grad_norm": 0.9621401429176331, + "learning_rate": 3.0357844222280496e-07, + "loss": 0.5376, + "step": 6110 + }, + { + "epoch": 5.058807288790724, + "grad_norm": 0.9491943120956421, + "learning_rate": 3.030534253340439e-07, + "loss": 0.5342, + "step": 6111 + }, + { + "epoch": 5.059635560463832, + "grad_norm": 0.9764898419380188, + "learning_rate": 3.0252883353396544e-07, + "loss": 0.5404, + "step": 6112 + }, + { + "epoch": 5.060463832136941, + "grad_norm": 0.9715072512626648, + "learning_rate": 3.0200466692407363e-07, + "loss": 0.5407, + "step": 6113 + }, + { + "epoch": 5.061292103810049, + "grad_norm": 0.9575538039207458, + "learning_rate": 3.014809256057888e-07, + "loss": 0.559, + "step": 6114 + }, + { + "epoch": 5.0621203754831585, + "grad_norm": 1.0109620094299316, + "learning_rate": 3.0095760968045034e-07, + "loss": 0.5533, + "step": 6115 + }, + { + "epoch": 5.062948647156268, + "grad_norm": 0.9899606108665466, + "learning_rate": 3.0043471924931596e-07, + "loss": 0.5507, + "step": 6116 + }, + { + "epoch": 5.063776918829376, + "grad_norm": 1.0038081407546997, + "learning_rate": 2.9991225441355856e-07, + "loss": 0.5545, + "step": 6117 + }, + { + "epoch": 5.064605190502485, + "grad_norm": 0.9636290073394775, + "learning_rate": 2.993902152742717e-07, + "loss": 0.5755, + "step": 6118 + }, + { + "epoch": 5.065433462175593, + "grad_norm": 0.9941927790641785, + "learning_rate": 2.9886860193246436e-07, + "loss": 0.5413, + "step": 6119 + }, + { + "epoch": 5.066261733848703, + "grad_norm": 0.9713919758796692, + "learning_rate": 2.983474144890633e-07, + "loss": 0.5327, + "step": 6120 + }, + { + "epoch": 5.067090005521811, + "grad_norm": 0.9749616384506226, + "learning_rate": 2.978266530449139e-07, + "loss": 0.5609, + "step": 6121 + }, + { + "epoch": 5.06791827719492, + "grad_norm": 0.9638389945030212, + "learning_rate": 2.973063177007793e-07, + "loss": 0.5537, + "step": 6122 + }, + { + "epoch": 5.068746548868028, + "grad_norm": 0.9680934548377991, + "learning_rate": 2.967864085573385e-07, + "loss": 0.5842, + "step": 6123 + }, + { + "epoch": 5.0695748205411375, + "grad_norm": 0.9608201384544373, + "learning_rate": 2.962669257151896e-07, + "loss": 0.5391, + "step": 6124 + }, + { + "epoch": 5.070403092214247, + "grad_norm": 0.9817207455635071, + "learning_rate": 2.9574786927484804e-07, + "loss": 0.5523, + "step": 6125 + }, + { + "epoch": 5.071231363887355, + "grad_norm": 0.9830054640769958, + "learning_rate": 2.9522923933674565e-07, + "loss": 0.5507, + "step": 6126 + }, + { + "epoch": 5.072059635560464, + "grad_norm": 0.9513229727745056, + "learning_rate": 2.9471103600123256e-07, + "loss": 0.5396, + "step": 6127 + }, + { + "epoch": 5.072887907233572, + "grad_norm": 0.9507657289505005, + "learning_rate": 2.941932593685773e-07, + "loss": 0.511, + "step": 6128 + }, + { + "epoch": 5.073716178906682, + "grad_norm": 0.9561945199966431, + "learning_rate": 2.936759095389641e-07, + "loss": 0.5467, + "step": 6129 + }, + { + "epoch": 5.07454445057979, + "grad_norm": 0.9509428143501282, + "learning_rate": 2.931589866124948e-07, + "loss": 0.5461, + "step": 6130 + }, + { + "epoch": 5.075372722252899, + "grad_norm": 0.9904574155807495, + "learning_rate": 2.926424906891895e-07, + "loss": 0.5443, + "step": 6131 + }, + { + "epoch": 5.076200993926007, + "grad_norm": 0.9647341966629028, + "learning_rate": 2.921264218689862e-07, + "loss": 0.5301, + "step": 6132 + }, + { + "epoch": 5.0770292655991165, + "grad_norm": 0.9626126885414124, + "learning_rate": 2.9161078025173816e-07, + "loss": 0.5525, + "step": 6133 + }, + { + "epoch": 5.077857537272226, + "grad_norm": 0.9646494388580322, + "learning_rate": 2.910955659372178e-07, + "loss": 0.5577, + "step": 6134 + }, + { + "epoch": 5.078685808945334, + "grad_norm": 0.9926782846450806, + "learning_rate": 2.905807790251147e-07, + "loss": 0.5219, + "step": 6135 + }, + { + "epoch": 5.079514080618443, + "grad_norm": 0.9474660158157349, + "learning_rate": 2.9006641961503463e-07, + "loss": 0.5308, + "step": 6136 + }, + { + "epoch": 5.080342352291551, + "grad_norm": 0.961175262928009, + "learning_rate": 2.895524878065015e-07, + "loss": 0.5459, + "step": 6137 + }, + { + "epoch": 5.081170623964661, + "grad_norm": 0.9793338775634766, + "learning_rate": 2.8903898369895683e-07, + "loss": 0.534, + "step": 6138 + }, + { + "epoch": 5.081998895637769, + "grad_norm": 0.9658337235450745, + "learning_rate": 2.8852590739175817e-07, + "loss": 0.5492, + "step": 6139 + }, + { + "epoch": 5.082827167310878, + "grad_norm": 0.9644724726676941, + "learning_rate": 2.8801325898418135e-07, + "loss": 0.5497, + "step": 6140 + }, + { + "epoch": 5.083655438983987, + "grad_norm": 0.9669801592826843, + "learning_rate": 2.875010385754198e-07, + "loss": 0.5392, + "step": 6141 + }, + { + "epoch": 5.0844837106570955, + "grad_norm": 0.9624497294425964, + "learning_rate": 2.869892462645818e-07, + "loss": 0.5107, + "step": 6142 + }, + { + "epoch": 5.085311982330205, + "grad_norm": 0.9657315611839294, + "learning_rate": 2.8647788215069526e-07, + "loss": 0.5373, + "step": 6143 + }, + { + "epoch": 5.086140254003313, + "grad_norm": 0.9853419661521912, + "learning_rate": 2.85966946332705e-07, + "loss": 0.5302, + "step": 6144 + }, + { + "epoch": 5.086968525676422, + "grad_norm": 0.990183413028717, + "learning_rate": 2.854564389094708e-07, + "loss": 0.531, + "step": 6145 + }, + { + "epoch": 5.08779679734953, + "grad_norm": 0.9916974306106567, + "learning_rate": 2.849463599797722e-07, + "loss": 0.5523, + "step": 6146 + }, + { + "epoch": 5.08862506902264, + "grad_norm": 0.9707847237586975, + "learning_rate": 2.844367096423048e-07, + "loss": 0.5258, + "step": 6147 + }, + { + "epoch": 5.089453340695748, + "grad_norm": 0.9810246825218201, + "learning_rate": 2.8392748799568047e-07, + "loss": 0.5483, + "step": 6148 + }, + { + "epoch": 5.090281612368857, + "grad_norm": 0.9569841027259827, + "learning_rate": 2.8341869513842933e-07, + "loss": 0.567, + "step": 6149 + }, + { + "epoch": 5.091109884041966, + "grad_norm": 0.9600836038589478, + "learning_rate": 2.8291033116899845e-07, + "loss": 0.5393, + "step": 6150 + }, + { + "epoch": 5.0919381557150745, + "grad_norm": 0.9942939281463623, + "learning_rate": 2.8240239618575056e-07, + "loss": 0.5712, + "step": 6151 + }, + { + "epoch": 5.092766427388184, + "grad_norm": 0.9982702732086182, + "learning_rate": 2.8189489028696724e-07, + "loss": 0.549, + "step": 6152 + }, + { + "epoch": 5.093594699061292, + "grad_norm": 0.9879527688026428, + "learning_rate": 2.813878135708453e-07, + "loss": 0.5447, + "step": 6153 + }, + { + "epoch": 5.094422970734401, + "grad_norm": 0.9737223386764526, + "learning_rate": 2.8088116613550013e-07, + "loss": 0.5388, + "step": 6154 + }, + { + "epoch": 5.095251242407509, + "grad_norm": 0.9854503870010376, + "learning_rate": 2.803749480789625e-07, + "loss": 0.5242, + "step": 6155 + }, + { + "epoch": 5.0960795140806185, + "grad_norm": 0.979864239692688, + "learning_rate": 2.798691594991815e-07, + "loss": 0.5571, + "step": 6156 + }, + { + "epoch": 5.096907785753727, + "grad_norm": 1.0167231559753418, + "learning_rate": 2.793638004940227e-07, + "loss": 0.5395, + "step": 6157 + }, + { + "epoch": 5.097736057426836, + "grad_norm": 0.9759885668754578, + "learning_rate": 2.788588711612675e-07, + "loss": 0.5463, + "step": 6158 + }, + { + "epoch": 5.098564329099945, + "grad_norm": 0.9751389026641846, + "learning_rate": 2.7835437159861563e-07, + "loss": 0.5292, + "step": 6159 + }, + { + "epoch": 5.0993926007730535, + "grad_norm": 0.9850022196769714, + "learning_rate": 2.7785030190368325e-07, + "loss": 0.5649, + "step": 6160 + }, + { + "epoch": 5.100220872446163, + "grad_norm": 0.9720654487609863, + "learning_rate": 2.773466621740026e-07, + "loss": 0.5667, + "step": 6161 + }, + { + "epoch": 5.101049144119271, + "grad_norm": 0.9541118144989014, + "learning_rate": 2.7684345250702377e-07, + "loss": 0.5482, + "step": 6162 + }, + { + "epoch": 5.10187741579238, + "grad_norm": 1.0875790119171143, + "learning_rate": 2.763406730001131e-07, + "loss": 0.5592, + "step": 6163 + }, + { + "epoch": 5.102705687465488, + "grad_norm": 0.9536864757537842, + "learning_rate": 2.758383237505527e-07, + "loss": 0.5526, + "step": 6164 + }, + { + "epoch": 5.1035339591385975, + "grad_norm": 0.981012761592865, + "learning_rate": 2.7533640485554347e-07, + "loss": 0.542, + "step": 6165 + }, + { + "epoch": 5.104362230811706, + "grad_norm": 0.9592416286468506, + "learning_rate": 2.7483491641220247e-07, + "loss": 0.5539, + "step": 6166 + }, + { + "epoch": 5.105190502484815, + "grad_norm": 0.966204822063446, + "learning_rate": 2.743338585175617e-07, + "loss": 0.5283, + "step": 6167 + }, + { + "epoch": 5.106018774157924, + "grad_norm": 0.9759148955345154, + "learning_rate": 2.738332312685721e-07, + "loss": 0.5555, + "step": 6168 + }, + { + "epoch": 5.106847045831032, + "grad_norm": 0.9391621947288513, + "learning_rate": 2.733330347621005e-07, + "loss": 0.5667, + "step": 6169 + }, + { + "epoch": 5.107675317504142, + "grad_norm": 0.9608196020126343, + "learning_rate": 2.728332690949295e-07, + "loss": 0.5403, + "step": 6170 + }, + { + "epoch": 5.10850358917725, + "grad_norm": 0.9858448505401611, + "learning_rate": 2.723339343637596e-07, + "loss": 0.5345, + "step": 6171 + }, + { + "epoch": 5.109331860850359, + "grad_norm": 0.9921685457229614, + "learning_rate": 2.718350306652079e-07, + "loss": 0.5428, + "step": 6172 + }, + { + "epoch": 5.110160132523467, + "grad_norm": 0.9900442957878113, + "learning_rate": 2.7133655809580663e-07, + "loss": 0.5636, + "step": 6173 + }, + { + "epoch": 5.1109884041965765, + "grad_norm": 0.98037189245224, + "learning_rate": 2.7083851675200583e-07, + "loss": 0.5168, + "step": 6174 + }, + { + "epoch": 5.111816675869686, + "grad_norm": 0.9850472211837769, + "learning_rate": 2.7034090673017266e-07, + "loss": 0.5614, + "step": 6175 + }, + { + "epoch": 5.112644947542794, + "grad_norm": 0.9582345485687256, + "learning_rate": 2.698437281265895e-07, + "loss": 0.5395, + "step": 6176 + }, + { + "epoch": 5.113473219215903, + "grad_norm": 0.9706653952598572, + "learning_rate": 2.6934698103745523e-07, + "loss": 0.5305, + "step": 6177 + }, + { + "epoch": 5.114301490889011, + "grad_norm": 0.9736168384552002, + "learning_rate": 2.688506655588866e-07, + "loss": 0.5338, + "step": 6178 + }, + { + "epoch": 5.115129762562121, + "grad_norm": 0.976378321647644, + "learning_rate": 2.68354781786915e-07, + "loss": 0.5541, + "step": 6179 + }, + { + "epoch": 5.115958034235229, + "grad_norm": 0.9677260518074036, + "learning_rate": 2.678593298174903e-07, + "loss": 0.5341, + "step": 6180 + }, + { + "epoch": 5.116786305908338, + "grad_norm": 0.935033917427063, + "learning_rate": 2.6736430974647785e-07, + "loss": 0.5459, + "step": 6181 + }, + { + "epoch": 5.117614577581446, + "grad_norm": 0.9987776875495911, + "learning_rate": 2.668697216696589e-07, + "loss": 0.5454, + "step": 6182 + }, + { + "epoch": 5.1184428492545555, + "grad_norm": 0.9799374938011169, + "learning_rate": 2.6637556568273157e-07, + "loss": 0.5506, + "step": 6183 + }, + { + "epoch": 5.119271120927665, + "grad_norm": 0.9752628207206726, + "learning_rate": 2.658818418813111e-07, + "loss": 0.564, + "step": 6184 + }, + { + "epoch": 5.120099392600773, + "grad_norm": 0.9776111245155334, + "learning_rate": 2.653885503609277e-07, + "loss": 0.5635, + "step": 6185 + }, + { + "epoch": 5.120927664273882, + "grad_norm": 0.9646061062812805, + "learning_rate": 2.6489569121702946e-07, + "loss": 0.5439, + "step": 6186 + }, + { + "epoch": 5.12175593594699, + "grad_norm": 0.9467222690582275, + "learning_rate": 2.644032645449793e-07, + "loss": 0.533, + "step": 6187 + }, + { + "epoch": 5.1225842076201, + "grad_norm": 0.9790500402450562, + "learning_rate": 2.639112704400579e-07, + "loss": 0.5276, + "step": 6188 + }, + { + "epoch": 5.123412479293208, + "grad_norm": 0.9900994300842285, + "learning_rate": 2.634197089974605e-07, + "loss": 0.543, + "step": 6189 + }, + { + "epoch": 5.124240750966317, + "grad_norm": 0.9717191457748413, + "learning_rate": 2.629285803123002e-07, + "loss": 0.5434, + "step": 6190 + }, + { + "epoch": 5.125069022639425, + "grad_norm": 0.9777794480323792, + "learning_rate": 2.624378844796066e-07, + "loss": 0.5358, + "step": 6191 + }, + { + "epoch": 5.1258972943125345, + "grad_norm": 0.9691375494003296, + "learning_rate": 2.619476215943234e-07, + "loss": 0.546, + "step": 6192 + }, + { + "epoch": 5.126725565985644, + "grad_norm": 1.0080442428588867, + "learning_rate": 2.6145779175131255e-07, + "loss": 0.5479, + "step": 6193 + }, + { + "epoch": 5.127553837658752, + "grad_norm": 0.9976667761802673, + "learning_rate": 2.609683950453523e-07, + "loss": 0.5323, + "step": 6194 + }, + { + "epoch": 5.128382109331861, + "grad_norm": 0.9813701510429382, + "learning_rate": 2.604794315711348e-07, + "loss": 0.5545, + "step": 6195 + }, + { + "epoch": 5.129210381004969, + "grad_norm": 0.9809463024139404, + "learning_rate": 2.5999090142327097e-07, + "loss": 0.5362, + "step": 6196 + }, + { + "epoch": 5.130038652678079, + "grad_norm": 0.9613211750984192, + "learning_rate": 2.5950280469628773e-07, + "loss": 0.5582, + "step": 6197 + }, + { + "epoch": 5.130866924351187, + "grad_norm": 0.9763631820678711, + "learning_rate": 2.590151414846248e-07, + "loss": 0.5375, + "step": 6198 + }, + { + "epoch": 5.131695196024296, + "grad_norm": 1.0060200691223145, + "learning_rate": 2.5852791188264197e-07, + "loss": 0.5249, + "step": 6199 + }, + { + "epoch": 5.132523467697405, + "grad_norm": 0.9742701649665833, + "learning_rate": 2.580411159846141e-07, + "loss": 0.5427, + "step": 6200 + }, + { + "epoch": 5.1333517393705135, + "grad_norm": 0.9736101627349854, + "learning_rate": 2.575547538847303e-07, + "loss": 0.5382, + "step": 6201 + }, + { + "epoch": 5.134180011043623, + "grad_norm": 0.9874021410942078, + "learning_rate": 2.5706882567709785e-07, + "loss": 0.5481, + "step": 6202 + }, + { + "epoch": 5.135008282716731, + "grad_norm": 0.9580921530723572, + "learning_rate": 2.565833314557398e-07, + "loss": 0.5615, + "step": 6203 + }, + { + "epoch": 5.13583655438984, + "grad_norm": 0.964108407497406, + "learning_rate": 2.5609827131459375e-07, + "loss": 0.5185, + "step": 6204 + }, + { + "epoch": 5.136664826062948, + "grad_norm": 0.9922768473625183, + "learning_rate": 2.55613645347515e-07, + "loss": 0.529, + "step": 6205 + }, + { + "epoch": 5.137493097736058, + "grad_norm": 0.9517302513122559, + "learning_rate": 2.551294536482743e-07, + "loss": 0.5534, + "step": 6206 + }, + { + "epoch": 5.138321369409166, + "grad_norm": 0.9788957238197327, + "learning_rate": 2.546456963105573e-07, + "loss": 0.5673, + "step": 6207 + }, + { + "epoch": 5.139149641082275, + "grad_norm": 0.9687431454658508, + "learning_rate": 2.541623734279677e-07, + "loss": 0.5579, + "step": 6208 + }, + { + "epoch": 5.139977912755384, + "grad_norm": 0.9680584669113159, + "learning_rate": 2.53679485094023e-07, + "loss": 0.5477, + "step": 6209 + }, + { + "epoch": 5.1408061844284925, + "grad_norm": 0.9760451912879944, + "learning_rate": 2.531970314021584e-07, + "loss": 0.5489, + "step": 6210 + }, + { + "epoch": 5.141634456101602, + "grad_norm": 0.9779831767082214, + "learning_rate": 2.5271501244572336e-07, + "loss": 0.5266, + "step": 6211 + }, + { + "epoch": 5.14246272777471, + "grad_norm": 1.0007282495498657, + "learning_rate": 2.5223342831798445e-07, + "loss": 0.5404, + "step": 6212 + }, + { + "epoch": 5.143290999447819, + "grad_norm": 0.9632928967475891, + "learning_rate": 2.5175227911212437e-07, + "loss": 0.556, + "step": 6213 + }, + { + "epoch": 5.144119271120927, + "grad_norm": 0.976509153842926, + "learning_rate": 2.512715649212399e-07, + "loss": 0.5547, + "step": 6214 + }, + { + "epoch": 5.1449475427940365, + "grad_norm": 1.0062603950500488, + "learning_rate": 2.507912858383452e-07, + "loss": 0.522, + "step": 6215 + }, + { + "epoch": 5.145775814467145, + "grad_norm": 1.0067973136901855, + "learning_rate": 2.5031144195637014e-07, + "loss": 0.5592, + "step": 6216 + }, + { + "epoch": 5.146604086140254, + "grad_norm": 0.9223517775535583, + "learning_rate": 2.498320333681592e-07, + "loss": 0.5621, + "step": 6217 + }, + { + "epoch": 5.147432357813363, + "grad_norm": 0.9999002814292908, + "learning_rate": 2.4935306016647424e-07, + "loss": 0.5429, + "step": 6218 + }, + { + "epoch": 5.1482606294864715, + "grad_norm": 0.9779905080795288, + "learning_rate": 2.488745224439923e-07, + "loss": 0.5472, + "step": 6219 + }, + { + "epoch": 5.149088901159581, + "grad_norm": 0.9957510232925415, + "learning_rate": 2.483964202933056e-07, + "loss": 0.584, + "step": 6220 + }, + { + "epoch": 5.149917172832689, + "grad_norm": 0.9871817827224731, + "learning_rate": 2.4791875380692196e-07, + "loss": 0.5311, + "step": 6221 + }, + { + "epoch": 5.150745444505798, + "grad_norm": 0.9485102891921997, + "learning_rate": 2.4744152307726615e-07, + "loss": 0.5482, + "step": 6222 + }, + { + "epoch": 5.151573716178906, + "grad_norm": 0.9466943740844727, + "learning_rate": 2.4696472819667696e-07, + "loss": 0.5343, + "step": 6223 + }, + { + "epoch": 5.1524019878520155, + "grad_norm": 0.9697267413139343, + "learning_rate": 2.464883692574105e-07, + "loss": 0.5477, + "step": 6224 + }, + { + "epoch": 5.153230259525124, + "grad_norm": 0.9745460152626038, + "learning_rate": 2.4601244635163805e-07, + "loss": 0.5624, + "step": 6225 + }, + { + "epoch": 5.154058531198233, + "grad_norm": 0.9862279891967773, + "learning_rate": 2.455369595714455e-07, + "loss": 0.5342, + "step": 6226 + }, + { + "epoch": 5.154886802871342, + "grad_norm": 1.0072344541549683, + "learning_rate": 2.450619090088352e-07, + "loss": 0.5536, + "step": 6227 + }, + { + "epoch": 5.15571507454445, + "grad_norm": 0.9548507332801819, + "learning_rate": 2.445872947557257e-07, + "loss": 0.5429, + "step": 6228 + }, + { + "epoch": 5.15654334621756, + "grad_norm": 0.959712028503418, + "learning_rate": 2.4411311690394986e-07, + "loss": 0.5371, + "step": 6229 + }, + { + "epoch": 5.157371617890668, + "grad_norm": 0.9656761884689331, + "learning_rate": 2.4363937554525653e-07, + "loss": 0.5525, + "step": 6230 + }, + { + "epoch": 5.158199889563777, + "grad_norm": 0.9887470006942749, + "learning_rate": 2.4316607077131097e-07, + "loss": 0.5361, + "step": 6231 + }, + { + "epoch": 5.159028161236885, + "grad_norm": 0.9570305943489075, + "learning_rate": 2.4269320267369277e-07, + "loss": 0.5366, + "step": 6232 + }, + { + "epoch": 5.1598564329099945, + "grad_norm": 0.978033721446991, + "learning_rate": 2.4222077134389707e-07, + "loss": 0.5315, + "step": 6233 + }, + { + "epoch": 5.160684704583104, + "grad_norm": 1.0171160697937012, + "learning_rate": 2.4174877687333534e-07, + "loss": 0.5322, + "step": 6234 + }, + { + "epoch": 5.161512976256212, + "grad_norm": 0.984002411365509, + "learning_rate": 2.412772193533347e-07, + "loss": 0.5526, + "step": 6235 + }, + { + "epoch": 5.162341247929321, + "grad_norm": 0.9785957932472229, + "learning_rate": 2.4080609887513594e-07, + "loss": 0.5485, + "step": 6236 + }, + { + "epoch": 5.163169519602429, + "grad_norm": 0.949811577796936, + "learning_rate": 2.403354155298973e-07, + "loss": 0.5312, + "step": 6237 + }, + { + "epoch": 5.163997791275539, + "grad_norm": 0.9676275849342346, + "learning_rate": 2.3986516940869174e-07, + "loss": 0.5466, + "step": 6238 + }, + { + "epoch": 5.164826062948647, + "grad_norm": 0.9725461602210999, + "learning_rate": 2.393953606025068e-07, + "loss": 0.5374, + "step": 6239 + }, + { + "epoch": 5.165654334621756, + "grad_norm": 0.9541648626327515, + "learning_rate": 2.389259892022469e-07, + "loss": 0.5418, + "step": 6240 + }, + { + "epoch": 5.166482606294864, + "grad_norm": 0.9559710025787354, + "learning_rate": 2.384570552987303e-07, + "loss": 0.547, + "step": 6241 + }, + { + "epoch": 5.1673108779679735, + "grad_norm": 0.9778188467025757, + "learning_rate": 2.379885589826919e-07, + "loss": 0.537, + "step": 6242 + }, + { + "epoch": 5.168139149641083, + "grad_norm": 0.9679521322250366, + "learning_rate": 2.3752050034478086e-07, + "loss": 0.5626, + "step": 6243 + }, + { + "epoch": 5.168967421314191, + "grad_norm": 0.9614160656929016, + "learning_rate": 2.3705287947556277e-07, + "loss": 0.5584, + "step": 6244 + }, + { + "epoch": 5.1697956929873, + "grad_norm": 0.9659228920936584, + "learning_rate": 2.3658569646551722e-07, + "loss": 0.5483, + "step": 6245 + }, + { + "epoch": 5.170623964660408, + "grad_norm": 0.9844391345977783, + "learning_rate": 2.361189514050402e-07, + "loss": 0.5442, + "step": 6246 + }, + { + "epoch": 5.171452236333518, + "grad_norm": 0.9415281414985657, + "learning_rate": 2.3565264438444285e-07, + "loss": 0.5311, + "step": 6247 + }, + { + "epoch": 5.172280508006626, + "grad_norm": 0.9558689594268799, + "learning_rate": 2.3518677549395018e-07, + "loss": 0.5556, + "step": 6248 + }, + { + "epoch": 5.173108779679735, + "grad_norm": 0.9737265706062317, + "learning_rate": 2.347213448237043e-07, + "loss": 0.5536, + "step": 6249 + }, + { + "epoch": 5.173937051352843, + "grad_norm": 0.9662517309188843, + "learning_rate": 2.3425635246376198e-07, + "loss": 0.5262, + "step": 6250 + }, + { + "epoch": 5.1747653230259525, + "grad_norm": 0.9569178819656372, + "learning_rate": 2.3379179850409412e-07, + "loss": 0.5676, + "step": 6251 + }, + { + "epoch": 5.175593594699062, + "grad_norm": 0.9627852439880371, + "learning_rate": 2.3332768303458764e-07, + "loss": 0.5385, + "step": 6252 + }, + { + "epoch": 5.17642186637217, + "grad_norm": 0.9722919464111328, + "learning_rate": 2.3286400614504572e-07, + "loss": 0.5307, + "step": 6253 + }, + { + "epoch": 5.177250138045279, + "grad_norm": 0.9715434312820435, + "learning_rate": 2.324007679251844e-07, + "loss": 0.5646, + "step": 6254 + }, + { + "epoch": 5.178078409718387, + "grad_norm": 0.9760845899581909, + "learning_rate": 2.3193796846463557e-07, + "loss": 0.5328, + "step": 6255 + }, + { + "epoch": 5.178906681391497, + "grad_norm": 1.0039637088775635, + "learning_rate": 2.3147560785294786e-07, + "loss": 0.5497, + "step": 6256 + }, + { + "epoch": 5.179734953064605, + "grad_norm": 0.9844917058944702, + "learning_rate": 2.310136861795828e-07, + "loss": 0.5528, + "step": 6257 + }, + { + "epoch": 5.180563224737714, + "grad_norm": 0.9769337177276611, + "learning_rate": 2.3055220353391804e-07, + "loss": 0.5444, + "step": 6258 + }, + { + "epoch": 5.181391496410823, + "grad_norm": 0.9749703407287598, + "learning_rate": 2.3009116000524688e-07, + "loss": 0.5327, + "step": 6259 + }, + { + "epoch": 5.1822197680839315, + "grad_norm": 0.9632902145385742, + "learning_rate": 2.2963055568277603e-07, + "loss": 0.5353, + "step": 6260 + }, + { + "epoch": 5.183048039757041, + "grad_norm": 0.9834912419319153, + "learning_rate": 2.2917039065562835e-07, + "loss": 0.5541, + "step": 6261 + }, + { + "epoch": 5.183876311430149, + "grad_norm": 0.9458034038543701, + "learning_rate": 2.2871066501284233e-07, + "loss": 0.5257, + "step": 6262 + }, + { + "epoch": 5.184704583103258, + "grad_norm": 0.9775984883308411, + "learning_rate": 2.2825137884336934e-07, + "loss": 0.548, + "step": 6263 + }, + { + "epoch": 5.185532854776366, + "grad_norm": 1.1132928133010864, + "learning_rate": 2.2779253223607772e-07, + "loss": 0.5444, + "step": 6264 + }, + { + "epoch": 5.186361126449476, + "grad_norm": 0.9759193658828735, + "learning_rate": 2.2733412527975036e-07, + "loss": 0.5334, + "step": 6265 + }, + { + "epoch": 5.187189398122584, + "grad_norm": 0.9685881733894348, + "learning_rate": 2.2687615806308404e-07, + "loss": 0.5323, + "step": 6266 + }, + { + "epoch": 5.188017669795693, + "grad_norm": 0.991648256778717, + "learning_rate": 2.26418630674691e-07, + "loss": 0.5639, + "step": 6267 + }, + { + "epoch": 5.188845941468802, + "grad_norm": 0.9838926792144775, + "learning_rate": 2.25961543203099e-07, + "loss": 0.5462, + "step": 6268 + }, + { + "epoch": 5.1896742131419105, + "grad_norm": 1.1861064434051514, + "learning_rate": 2.2550489573675034e-07, + "loss": 0.5346, + "step": 6269 + }, + { + "epoch": 5.19050248481502, + "grad_norm": 0.9433376789093018, + "learning_rate": 2.250486883640013e-07, + "loss": 0.5272, + "step": 6270 + }, + { + "epoch": 5.191330756488128, + "grad_norm": 0.980558454990387, + "learning_rate": 2.245929211731243e-07, + "loss": 0.5285, + "step": 6271 + }, + { + "epoch": 5.192159028161237, + "grad_norm": 0.9543962478637695, + "learning_rate": 2.2413759425230663e-07, + "loss": 0.5367, + "step": 6272 + }, + { + "epoch": 5.192987299834345, + "grad_norm": 0.9564104080200195, + "learning_rate": 2.2368270768964835e-07, + "loss": 0.5702, + "step": 6273 + }, + { + "epoch": 5.1938155715074545, + "grad_norm": 0.9902310371398926, + "learning_rate": 2.232282615731668e-07, + "loss": 0.5315, + "step": 6274 + }, + { + "epoch": 5.194643843180563, + "grad_norm": 0.9771626591682434, + "learning_rate": 2.2277425599079334e-07, + "loss": 0.5329, + "step": 6275 + }, + { + "epoch": 5.195472114853672, + "grad_norm": 0.9694120287895203, + "learning_rate": 2.2232069103037324e-07, + "loss": 0.5543, + "step": 6276 + }, + { + "epoch": 5.196300386526781, + "grad_norm": 1.0038840770721436, + "learning_rate": 2.218675667796666e-07, + "loss": 0.5404, + "step": 6277 + }, + { + "epoch": 5.1971286581998895, + "grad_norm": 0.9844067096710205, + "learning_rate": 2.214148833263499e-07, + "loss": 0.5507, + "step": 6278 + }, + { + "epoch": 5.197956929872999, + "grad_norm": 0.9688404202461243, + "learning_rate": 2.2096264075801204e-07, + "loss": 0.5392, + "step": 6279 + }, + { + "epoch": 5.198785201546107, + "grad_norm": 0.9584043025970459, + "learning_rate": 2.2051083916215827e-07, + "loss": 0.5424, + "step": 6280 + }, + { + "epoch": 5.199613473219216, + "grad_norm": 0.977668821811676, + "learning_rate": 2.2005947862620835e-07, + "loss": 0.5412, + "step": 6281 + }, + { + "epoch": 5.200441744892324, + "grad_norm": 0.9770166873931885, + "learning_rate": 2.196085592374958e-07, + "loss": 0.5218, + "step": 6282 + }, + { + "epoch": 5.2012700165654335, + "grad_norm": 1.0270694494247437, + "learning_rate": 2.191580810832694e-07, + "loss": 0.553, + "step": 6283 + }, + { + "epoch": 5.202098288238542, + "grad_norm": 0.9948989748954773, + "learning_rate": 2.187080442506928e-07, + "loss": 0.5446, + "step": 6284 + }, + { + "epoch": 5.202926559911651, + "grad_norm": 0.980826199054718, + "learning_rate": 2.1825844882684355e-07, + "loss": 0.5538, + "step": 6285 + }, + { + "epoch": 5.20375483158476, + "grad_norm": 0.9694545865058899, + "learning_rate": 2.1780929489871426e-07, + "loss": 0.5632, + "step": 6286 + }, + { + "epoch": 5.204583103257868, + "grad_norm": 0.9740391373634338, + "learning_rate": 2.1736058255321241e-07, + "loss": 0.5492, + "step": 6287 + }, + { + "epoch": 5.205411374930978, + "grad_norm": 0.9584625363349915, + "learning_rate": 2.169123118771596e-07, + "loss": 0.5762, + "step": 6288 + }, + { + "epoch": 5.206239646604086, + "grad_norm": 0.9970041513442993, + "learning_rate": 2.1646448295729093e-07, + "loss": 0.5437, + "step": 6289 + }, + { + "epoch": 5.207067918277195, + "grad_norm": 0.9862902760505676, + "learning_rate": 2.160170958802582e-07, + "loss": 0.5483, + "step": 6290 + }, + { + "epoch": 5.207896189950303, + "grad_norm": 0.9648549556732178, + "learning_rate": 2.1557015073262689e-07, + "loss": 0.5464, + "step": 6291 + }, + { + "epoch": 5.2087244616234125, + "grad_norm": 0.9971094727516174, + "learning_rate": 2.1512364760087583e-07, + "loss": 0.56, + "step": 6292 + }, + { + "epoch": 5.209552733296522, + "grad_norm": 0.9579539895057678, + "learning_rate": 2.146775865713993e-07, + "loss": 0.5357, + "step": 6293 + }, + { + "epoch": 5.21038100496963, + "grad_norm": 1.003116488456726, + "learning_rate": 2.1423196773050715e-07, + "loss": 0.5425, + "step": 6294 + }, + { + "epoch": 5.211209276642739, + "grad_norm": 0.9752295613288879, + "learning_rate": 2.1378679116442092e-07, + "loss": 0.5542, + "step": 6295 + }, + { + "epoch": 5.212037548315847, + "grad_norm": 0.9679763913154602, + "learning_rate": 2.1334205695927868e-07, + "loss": 0.554, + "step": 6296 + }, + { + "epoch": 5.212865819988957, + "grad_norm": 0.9780550599098206, + "learning_rate": 2.1289776520113298e-07, + "loss": 0.5136, + "step": 6297 + }, + { + "epoch": 5.213694091662065, + "grad_norm": 0.9568700194358826, + "learning_rate": 2.1245391597594923e-07, + "loss": 0.5532, + "step": 6298 + }, + { + "epoch": 5.214522363335174, + "grad_norm": 0.9589282870292664, + "learning_rate": 2.12010509369609e-07, + "loss": 0.5599, + "step": 6299 + }, + { + "epoch": 5.215350635008282, + "grad_norm": 0.9796894788742065, + "learning_rate": 2.115675454679067e-07, + "loss": 0.5276, + "step": 6300 + }, + { + "epoch": 5.2161789066813915, + "grad_norm": 0.9596220254898071, + "learning_rate": 2.1112502435655157e-07, + "loss": 0.5286, + "step": 6301 + }, + { + "epoch": 5.217007178354501, + "grad_norm": 0.9631205797195435, + "learning_rate": 2.1068294612116735e-07, + "loss": 0.5282, + "step": 6302 + }, + { + "epoch": 5.217835450027609, + "grad_norm": 0.966499924659729, + "learning_rate": 2.102413108472931e-07, + "loss": 0.5672, + "step": 6303 + }, + { + "epoch": 5.218663721700718, + "grad_norm": 1.001829981803894, + "learning_rate": 2.0980011862037963e-07, + "loss": 0.5469, + "step": 6304 + }, + { + "epoch": 5.219491993373826, + "grad_norm": 0.9720007181167603, + "learning_rate": 2.0935936952579422e-07, + "loss": 0.5476, + "step": 6305 + }, + { + "epoch": 5.220320265046936, + "grad_norm": 0.9557669758796692, + "learning_rate": 2.0891906364881835e-07, + "loss": 0.532, + "step": 6306 + }, + { + "epoch": 5.221148536720044, + "grad_norm": 0.9431495070457458, + "learning_rate": 2.0847920107464588e-07, + "loss": 0.5453, + "step": 6307 + }, + { + "epoch": 5.221976808393153, + "grad_norm": 0.9483916163444519, + "learning_rate": 2.0803978188838698e-07, + "loss": 0.5406, + "step": 6308 + }, + { + "epoch": 5.222805080066261, + "grad_norm": 0.9639368057250977, + "learning_rate": 2.0760080617506505e-07, + "loss": 0.5391, + "step": 6309 + }, + { + "epoch": 5.2236333517393705, + "grad_norm": 0.9679994583129883, + "learning_rate": 2.0716227401961776e-07, + "loss": 0.5515, + "step": 6310 + }, + { + "epoch": 5.22446162341248, + "grad_norm": 0.9801360964775085, + "learning_rate": 2.067241855068966e-07, + "loss": 0.5551, + "step": 6311 + }, + { + "epoch": 5.225289895085588, + "grad_norm": 0.9779489636421204, + "learning_rate": 2.062865407216677e-07, + "loss": 0.5516, + "step": 6312 + }, + { + "epoch": 5.226118166758697, + "grad_norm": 0.9863239526748657, + "learning_rate": 2.0584933974861192e-07, + "loss": 0.5168, + "step": 6313 + }, + { + "epoch": 5.226946438431805, + "grad_norm": 0.9666626453399658, + "learning_rate": 2.0541258267232272e-07, + "loss": 0.5538, + "step": 6314 + }, + { + "epoch": 5.227774710104915, + "grad_norm": 1.0013031959533691, + "learning_rate": 2.0497626957730914e-07, + "loss": 0.5599, + "step": 6315 + }, + { + "epoch": 5.228602981778023, + "grad_norm": 0.9619392156600952, + "learning_rate": 2.0454040054799313e-07, + "loss": 0.5491, + "step": 6316 + }, + { + "epoch": 5.229431253451132, + "grad_norm": 0.978722333908081, + "learning_rate": 2.041049756687119e-07, + "loss": 0.562, + "step": 6317 + }, + { + "epoch": 5.230259525124241, + "grad_norm": 0.9707070589065552, + "learning_rate": 2.0366999502371587e-07, + "loss": 0.5536, + "step": 6318 + }, + { + "epoch": 5.2310877967973495, + "grad_norm": 0.9803135991096497, + "learning_rate": 2.0323545869716937e-07, + "loss": 0.5348, + "step": 6319 + }, + { + "epoch": 5.231916068470459, + "grad_norm": 0.983718991279602, + "learning_rate": 2.0280136677315155e-07, + "loss": 0.5408, + "step": 6320 + }, + { + "epoch": 5.232744340143567, + "grad_norm": 0.9826907515525818, + "learning_rate": 2.0236771933565546e-07, + "loss": 0.561, + "step": 6321 + }, + { + "epoch": 5.233572611816676, + "grad_norm": 0.9846853613853455, + "learning_rate": 2.0193451646858737e-07, + "loss": 0.5341, + "step": 6322 + }, + { + "epoch": 5.234400883489784, + "grad_norm": 0.9684202075004578, + "learning_rate": 2.015017582557674e-07, + "loss": 0.539, + "step": 6323 + }, + { + "epoch": 5.2352291551628936, + "grad_norm": 0.9837101101875305, + "learning_rate": 2.010694447809311e-07, + "loss": 0.5248, + "step": 6324 + }, + { + "epoch": 5.236057426836002, + "grad_norm": 0.9677354693412781, + "learning_rate": 2.0063757612772738e-07, + "loss": 0.5344, + "step": 6325 + }, + { + "epoch": 5.236885698509111, + "grad_norm": 0.9670676589012146, + "learning_rate": 2.0020615237971773e-07, + "loss": 0.5591, + "step": 6326 + }, + { + "epoch": 5.23771397018222, + "grad_norm": 0.969111442565918, + "learning_rate": 1.9977517362037928e-07, + "loss": 0.5388, + "step": 6327 + }, + { + "epoch": 5.2385422418553285, + "grad_norm": 0.9784060120582581, + "learning_rate": 1.9934463993310283e-07, + "loss": 0.5295, + "step": 6328 + }, + { + "epoch": 5.239370513528438, + "grad_norm": 0.9658244848251343, + "learning_rate": 1.9891455140119143e-07, + "loss": 0.5498, + "step": 6329 + }, + { + "epoch": 5.240198785201546, + "grad_norm": 0.974251389503479, + "learning_rate": 1.984849081078641e-07, + "loss": 0.5441, + "step": 6330 + }, + { + "epoch": 5.241027056874655, + "grad_norm": 0.9342301487922668, + "learning_rate": 1.9805571013625292e-07, + "loss": 0.5588, + "step": 6331 + }, + { + "epoch": 5.241855328547763, + "grad_norm": 0.9887391328811646, + "learning_rate": 1.9762695756940287e-07, + "loss": 0.5563, + "step": 6332 + }, + { + "epoch": 5.2426836002208725, + "grad_norm": 0.965055525302887, + "learning_rate": 1.971986504902748e-07, + "loss": 0.5488, + "step": 6333 + }, + { + "epoch": 5.243511871893981, + "grad_norm": 0.9694443345069885, + "learning_rate": 1.9677078898174108e-07, + "loss": 0.5356, + "step": 6334 + }, + { + "epoch": 5.24434014356709, + "grad_norm": 0.9638567566871643, + "learning_rate": 1.9634337312658902e-07, + "loss": 0.5291, + "step": 6335 + }, + { + "epoch": 5.245168415240199, + "grad_norm": 0.9594679474830627, + "learning_rate": 1.9591640300751975e-07, + "loss": 0.5364, + "step": 6336 + }, + { + "epoch": 5.2459966869133074, + "grad_norm": 0.9493943452835083, + "learning_rate": 1.954898787071488e-07, + "loss": 0.5391, + "step": 6337 + }, + { + "epoch": 5.246824958586417, + "grad_norm": 0.9794185757637024, + "learning_rate": 1.9506380030800325e-07, + "loss": 0.5504, + "step": 6338 + }, + { + "epoch": 5.247653230259525, + "grad_norm": 0.9784255027770996, + "learning_rate": 1.9463816789252605e-07, + "loss": 0.5471, + "step": 6339 + }, + { + "epoch": 5.248481501932634, + "grad_norm": 0.9476204514503479, + "learning_rate": 1.942129815430735e-07, + "loss": 0.5583, + "step": 6340 + }, + { + "epoch": 5.249309773605742, + "grad_norm": 0.9716140627861023, + "learning_rate": 1.9378824134191426e-07, + "loss": 0.5359, + "step": 6341 + }, + { + "epoch": 5.2501380452788515, + "grad_norm": 0.9689304828643799, + "learning_rate": 1.933639473712323e-07, + "loss": 0.5466, + "step": 6342 + }, + { + "epoch": 5.25096631695196, + "grad_norm": 1.0089303255081177, + "learning_rate": 1.929400997131245e-07, + "loss": 0.5554, + "step": 6343 + }, + { + "epoch": 5.251794588625069, + "grad_norm": 1.00104558467865, + "learning_rate": 1.925166984496013e-07, + "loss": 0.5278, + "step": 6344 + }, + { + "epoch": 5.252622860298178, + "grad_norm": 0.9637569785118103, + "learning_rate": 1.9209374366258665e-07, + "loss": 0.5535, + "step": 6345 + }, + { + "epoch": 5.253451131971286, + "grad_norm": 0.9478625655174255, + "learning_rate": 1.916712354339184e-07, + "loss": 0.51, + "step": 6346 + }, + { + "epoch": 5.254279403644396, + "grad_norm": 0.9619899988174438, + "learning_rate": 1.912491738453487e-07, + "loss": 0.547, + "step": 6347 + }, + { + "epoch": 5.255107675317504, + "grad_norm": 0.9798589944839478, + "learning_rate": 1.9082755897854133e-07, + "loss": 0.5117, + "step": 6348 + }, + { + "epoch": 5.255935946990613, + "grad_norm": 0.9652154445648193, + "learning_rate": 1.904063909150755e-07, + "loss": 0.5671, + "step": 6349 + }, + { + "epoch": 5.256764218663721, + "grad_norm": 0.9821009039878845, + "learning_rate": 1.8998566973644383e-07, + "loss": 0.5142, + "step": 6350 + }, + { + "epoch": 5.2575924903368305, + "grad_norm": 0.9646831154823303, + "learning_rate": 1.895653955240509e-07, + "loss": 0.5453, + "step": 6351 + }, + { + "epoch": 5.258420762009939, + "grad_norm": 0.9639331102371216, + "learning_rate": 1.891455683592164e-07, + "loss": 0.5747, + "step": 6352 + }, + { + "epoch": 5.259249033683048, + "grad_norm": 0.9825933575630188, + "learning_rate": 1.8872618832317336e-07, + "loss": 0.5428, + "step": 6353 + }, + { + "epoch": 5.260077305356157, + "grad_norm": 0.9847123026847839, + "learning_rate": 1.883072554970669e-07, + "loss": 0.5636, + "step": 6354 + }, + { + "epoch": 5.260905577029265, + "grad_norm": 0.9773681163787842, + "learning_rate": 1.8788876996195748e-07, + "loss": 0.562, + "step": 6355 + }, + { + "epoch": 5.261733848702375, + "grad_norm": 0.9792121052742004, + "learning_rate": 1.874707317988181e-07, + "loss": 0.5313, + "step": 6356 + }, + { + "epoch": 5.262562120375483, + "grad_norm": 0.9793497323989868, + "learning_rate": 1.8705314108853457e-07, + "loss": 0.5507, + "step": 6357 + }, + { + "epoch": 5.263390392048592, + "grad_norm": 0.980713963508606, + "learning_rate": 1.86635997911907e-07, + "loss": 0.5952, + "step": 6358 + }, + { + "epoch": 5.2642186637217, + "grad_norm": 0.9744186997413635, + "learning_rate": 1.8621930234964947e-07, + "loss": 0.547, + "step": 6359 + }, + { + "epoch": 5.2650469353948095, + "grad_norm": 0.9798989295959473, + "learning_rate": 1.8580305448238772e-07, + "loss": 0.5427, + "step": 6360 + }, + { + "epoch": 5.265875207067919, + "grad_norm": 0.9932631254196167, + "learning_rate": 1.8538725439066235e-07, + "loss": 0.5597, + "step": 6361 + }, + { + "epoch": 5.266703478741027, + "grad_norm": 0.9676787257194519, + "learning_rate": 1.8497190215492705e-07, + "loss": 0.5374, + "step": 6362 + }, + { + "epoch": 5.267531750414136, + "grad_norm": 0.9780867099761963, + "learning_rate": 1.8455699785554782e-07, + "loss": 0.5515, + "step": 6363 + }, + { + "epoch": 5.268360022087244, + "grad_norm": 0.972269594669342, + "learning_rate": 1.8414254157280514e-07, + "loss": 0.5454, + "step": 6364 + }, + { + "epoch": 5.269188293760354, + "grad_norm": 0.9527044296264648, + "learning_rate": 1.8372853338689322e-07, + "loss": 0.5539, + "step": 6365 + }, + { + "epoch": 5.270016565433462, + "grad_norm": 0.9782390594482422, + "learning_rate": 1.8331497337791744e-07, + "loss": 0.5316, + "step": 6366 + }, + { + "epoch": 5.270844837106571, + "grad_norm": 0.9573436975479126, + "learning_rate": 1.82901861625899e-07, + "loss": 0.5419, + "step": 6367 + }, + { + "epoch": 5.27167310877968, + "grad_norm": 0.9779508709907532, + "learning_rate": 1.8248919821077016e-07, + "loss": 0.53, + "step": 6368 + }, + { + "epoch": 5.2725013804527885, + "grad_norm": 1.0114938020706177, + "learning_rate": 1.820769832123781e-07, + "loss": 0.5544, + "step": 6369 + }, + { + "epoch": 5.273329652125898, + "grad_norm": 0.9842972159385681, + "learning_rate": 1.8166521671048236e-07, + "loss": 0.5711, + "step": 6370 + }, + { + "epoch": 5.274157923799006, + "grad_norm": 0.976719319820404, + "learning_rate": 1.8125389878475563e-07, + "loss": 0.542, + "step": 6371 + }, + { + "epoch": 5.274986195472115, + "grad_norm": 1.0023735761642456, + "learning_rate": 1.8084302951478504e-07, + "loss": 0.5374, + "step": 6372 + }, + { + "epoch": 5.275814467145223, + "grad_norm": 0.9452418088912964, + "learning_rate": 1.804326089800687e-07, + "loss": 0.5507, + "step": 6373 + }, + { + "epoch": 5.276642738818333, + "grad_norm": 0.9909770488739014, + "learning_rate": 1.8002263726002e-07, + "loss": 0.5315, + "step": 6374 + }, + { + "epoch": 5.277471010491441, + "grad_norm": 0.9580590724945068, + "learning_rate": 1.7961311443396468e-07, + "loss": 0.5478, + "step": 6375 + }, + { + "epoch": 5.27829928216455, + "grad_norm": 0.9607858061790466, + "learning_rate": 1.79204040581141e-07, + "loss": 0.5501, + "step": 6376 + }, + { + "epoch": 5.279127553837659, + "grad_norm": 0.9457074403762817, + "learning_rate": 1.7879541578070175e-07, + "loss": 0.5329, + "step": 6377 + }, + { + "epoch": 5.2799558255107675, + "grad_norm": 0.9904559254646301, + "learning_rate": 1.783872401117115e-07, + "loss": 0.5769, + "step": 6378 + }, + { + "epoch": 5.280784097183877, + "grad_norm": 0.9917809367179871, + "learning_rate": 1.7797951365314813e-07, + "loss": 0.5512, + "step": 6379 + }, + { + "epoch": 5.281612368856985, + "grad_norm": 0.9760978817939758, + "learning_rate": 1.7757223648390305e-07, + "loss": 0.549, + "step": 6380 + }, + { + "epoch": 5.282440640530094, + "grad_norm": 0.994718074798584, + "learning_rate": 1.771654086827815e-07, + "loss": 0.5589, + "step": 6381 + }, + { + "epoch": 5.283268912203202, + "grad_norm": 0.96318519115448, + "learning_rate": 1.7675903032849966e-07, + "loss": 0.5422, + "step": 6382 + }, + { + "epoch": 5.2840971838763116, + "grad_norm": 0.9674662351608276, + "learning_rate": 1.763531014996886e-07, + "loss": 0.5471, + "step": 6383 + }, + { + "epoch": 5.28492545554942, + "grad_norm": 0.9378229975700378, + "learning_rate": 1.7594762227489182e-07, + "loss": 0.5522, + "step": 6384 + }, + { + "epoch": 5.285753727222529, + "grad_norm": 0.9545474052429199, + "learning_rate": 1.7554259273256518e-07, + "loss": 0.5505, + "step": 6385 + }, + { + "epoch": 5.286581998895638, + "grad_norm": 0.9547621607780457, + "learning_rate": 1.751380129510788e-07, + "loss": 0.535, + "step": 6386 + }, + { + "epoch": 5.2874102705687465, + "grad_norm": 0.9777008891105652, + "learning_rate": 1.7473388300871502e-07, + "loss": 0.5617, + "step": 6387 + }, + { + "epoch": 5.288238542241856, + "grad_norm": 0.9793879985809326, + "learning_rate": 1.7433020298366854e-07, + "loss": 0.5637, + "step": 6388 + }, + { + "epoch": 5.289066813914964, + "grad_norm": 1.0022300481796265, + "learning_rate": 1.739269729540488e-07, + "loss": 0.5558, + "step": 6389 + }, + { + "epoch": 5.289895085588073, + "grad_norm": 0.9759293794631958, + "learning_rate": 1.7352419299787593e-07, + "loss": 0.545, + "step": 6390 + }, + { + "epoch": 5.290723357261181, + "grad_norm": 0.9910255670547485, + "learning_rate": 1.7312186319308504e-07, + "loss": 0.5587, + "step": 6391 + }, + { + "epoch": 5.2915516289342905, + "grad_norm": 0.9770117402076721, + "learning_rate": 1.7271998361752224e-07, + "loss": 0.551, + "step": 6392 + }, + { + "epoch": 5.292379900607399, + "grad_norm": 0.9753358364105225, + "learning_rate": 1.7231855434894834e-07, + "loss": 0.5657, + "step": 6393 + }, + { + "epoch": 5.293208172280508, + "grad_norm": 0.9842093586921692, + "learning_rate": 1.719175754650354e-07, + "loss": 0.5365, + "step": 6394 + }, + { + "epoch": 5.294036443953617, + "grad_norm": 0.9638484716415405, + "learning_rate": 1.715170470433697e-07, + "loss": 0.5533, + "step": 6395 + }, + { + "epoch": 5.2948647156267254, + "grad_norm": 0.957156777381897, + "learning_rate": 1.7111696916144978e-07, + "loss": 0.5118, + "step": 6396 + }, + { + "epoch": 5.295692987299835, + "grad_norm": 0.9706379771232605, + "learning_rate": 1.707173418966865e-07, + "loss": 0.5593, + "step": 6397 + }, + { + "epoch": 5.296521258972943, + "grad_norm": 0.9784908890724182, + "learning_rate": 1.703181653264041e-07, + "loss": 0.5724, + "step": 6398 + }, + { + "epoch": 5.297349530646052, + "grad_norm": 0.9688151478767395, + "learning_rate": 1.6991943952784029e-07, + "loss": 0.547, + "step": 6399 + }, + { + "epoch": 5.29817780231916, + "grad_norm": 0.9698705077171326, + "learning_rate": 1.695211645781436e-07, + "loss": 0.5458, + "step": 6400 + }, + { + "epoch": 5.2990060739922695, + "grad_norm": 0.9868040680885315, + "learning_rate": 1.6912334055437768e-07, + "loss": 0.5524, + "step": 6401 + }, + { + "epoch": 5.299834345665378, + "grad_norm": 0.9721259474754333, + "learning_rate": 1.687259675335165e-07, + "loss": 0.5641, + "step": 6402 + }, + { + "epoch": 5.300662617338487, + "grad_norm": 0.9706754088401794, + "learning_rate": 1.683290455924494e-07, + "loss": 0.5239, + "step": 6403 + }, + { + "epoch": 5.301490889011596, + "grad_norm": 0.9786049723625183, + "learning_rate": 1.6793257480797604e-07, + "loss": 0.5528, + "step": 6404 + }, + { + "epoch": 5.302319160684704, + "grad_norm": 0.9797729253768921, + "learning_rate": 1.6753655525681007e-07, + "loss": 0.5517, + "step": 6405 + }, + { + "epoch": 5.303147432357814, + "grad_norm": 0.9785734415054321, + "learning_rate": 1.6714098701557824e-07, + "loss": 0.5406, + "step": 6406 + }, + { + "epoch": 5.303975704030922, + "grad_norm": 0.9811814427375793, + "learning_rate": 1.667458701608182e-07, + "loss": 0.5582, + "step": 6407 + }, + { + "epoch": 5.304803975704031, + "grad_norm": 0.9791476130485535, + "learning_rate": 1.663512047689822e-07, + "loss": 0.5399, + "step": 6408 + }, + { + "epoch": 5.305632247377139, + "grad_norm": 0.9707565307617188, + "learning_rate": 1.6595699091643435e-07, + "loss": 0.5509, + "step": 6409 + }, + { + "epoch": 5.3064605190502485, + "grad_norm": 0.9604771733283997, + "learning_rate": 1.655632286794509e-07, + "loss": 0.5271, + "step": 6410 + }, + { + "epoch": 5.307288790723357, + "grad_norm": 0.9599454998970032, + "learning_rate": 1.6516991813422117e-07, + "loss": 0.5748, + "step": 6411 + }, + { + "epoch": 5.308117062396466, + "grad_norm": 0.9757176637649536, + "learning_rate": 1.647770593568479e-07, + "loss": 0.5581, + "step": 6412 + }, + { + "epoch": 5.308945334069575, + "grad_norm": 0.9501317739486694, + "learning_rate": 1.6438465242334445e-07, + "loss": 0.544, + "step": 6413 + }, + { + "epoch": 5.309773605742683, + "grad_norm": 0.9671623110771179, + "learning_rate": 1.6399269740963815e-07, + "loss": 0.5473, + "step": 6414 + }, + { + "epoch": 5.310601877415793, + "grad_norm": 0.9799640774726868, + "learning_rate": 1.6360119439156967e-07, + "loss": 0.5356, + "step": 6415 + }, + { + "epoch": 5.311430149088901, + "grad_norm": 0.9644803404808044, + "learning_rate": 1.632101434448899e-07, + "loss": 0.5386, + "step": 6416 + }, + { + "epoch": 5.31225842076201, + "grad_norm": 0.939059853553772, + "learning_rate": 1.6281954464526407e-07, + "loss": 0.5503, + "step": 6417 + }, + { + "epoch": 5.313086692435118, + "grad_norm": 0.9843038320541382, + "learning_rate": 1.624293980682698e-07, + "loss": 0.5411, + "step": 6418 + }, + { + "epoch": 5.3139149641082275, + "grad_norm": 0.9967756271362305, + "learning_rate": 1.6203970378939587e-07, + "loss": 0.5484, + "step": 6419 + }, + { + "epoch": 5.314743235781337, + "grad_norm": 0.9777352809906006, + "learning_rate": 1.6165046188404532e-07, + "loss": 0.5407, + "step": 6420 + }, + { + "epoch": 5.315571507454445, + "grad_norm": 0.9712015390396118, + "learning_rate": 1.6126167242753287e-07, + "loss": 0.5512, + "step": 6421 + }, + { + "epoch": 5.316399779127554, + "grad_norm": 0.9921703338623047, + "learning_rate": 1.6087333549508505e-07, + "loss": 0.532, + "step": 6422 + }, + { + "epoch": 5.317228050800662, + "grad_norm": 0.9545930624008179, + "learning_rate": 1.6048545116184229e-07, + "loss": 0.566, + "step": 6423 + }, + { + "epoch": 5.318056322473772, + "grad_norm": 0.9978237748146057, + "learning_rate": 1.6009801950285537e-07, + "loss": 0.5617, + "step": 6424 + }, + { + "epoch": 5.31888459414688, + "grad_norm": 0.9477809071540833, + "learning_rate": 1.5971104059308988e-07, + "loss": 0.531, + "step": 6425 + }, + { + "epoch": 5.319712865819989, + "grad_norm": 0.9565157294273376, + "learning_rate": 1.593245145074218e-07, + "loss": 0.5412, + "step": 6426 + }, + { + "epoch": 5.320541137493098, + "grad_norm": 0.9754008054733276, + "learning_rate": 1.5893844132064046e-07, + "loss": 0.5217, + "step": 6427 + }, + { + "epoch": 5.3213694091662065, + "grad_norm": 0.9599457383155823, + "learning_rate": 1.585528211074483e-07, + "loss": 0.5413, + "step": 6428 + }, + { + "epoch": 5.322197680839316, + "grad_norm": 0.99076247215271, + "learning_rate": 1.5816765394245815e-07, + "loss": 0.5683, + "step": 6429 + }, + { + "epoch": 5.323025952512424, + "grad_norm": 0.9798392653465271, + "learning_rate": 1.577829399001965e-07, + "loss": 0.5396, + "step": 6430 + }, + { + "epoch": 5.323854224185533, + "grad_norm": 0.9322737455368042, + "learning_rate": 1.5739867905510242e-07, + "loss": 0.5257, + "step": 6431 + }, + { + "epoch": 5.324682495858641, + "grad_norm": 0.997534453868866, + "learning_rate": 1.5701487148152644e-07, + "loss": 0.5759, + "step": 6432 + }, + { + "epoch": 5.325510767531751, + "grad_norm": 0.9972415566444397, + "learning_rate": 1.5663151725373165e-07, + "loss": 0.5378, + "step": 6433 + }, + { + "epoch": 5.326339039204859, + "grad_norm": 0.9688319563865662, + "learning_rate": 1.5624861644589402e-07, + "loss": 0.5616, + "step": 6434 + }, + { + "epoch": 5.327167310877968, + "grad_norm": 0.9457203149795532, + "learning_rate": 1.558661691321009e-07, + "loss": 0.5507, + "step": 6435 + }, + { + "epoch": 5.327995582551077, + "grad_norm": 0.9818011522293091, + "learning_rate": 1.5548417538635202e-07, + "loss": 0.5594, + "step": 6436 + }, + { + "epoch": 5.3288238542241855, + "grad_norm": 0.981209933757782, + "learning_rate": 1.551026352825602e-07, + "loss": 0.5163, + "step": 6437 + }, + { + "epoch": 5.329652125897295, + "grad_norm": 0.9571893811225891, + "learning_rate": 1.5472154889454917e-07, + "loss": 0.5265, + "step": 6438 + }, + { + "epoch": 5.330480397570403, + "grad_norm": 0.9521164894104004, + "learning_rate": 1.543409162960563e-07, + "loss": 0.5304, + "step": 6439 + }, + { + "epoch": 5.331308669243512, + "grad_norm": 0.9704666137695312, + "learning_rate": 1.539607375607302e-07, + "loss": 0.5612, + "step": 6440 + }, + { + "epoch": 5.33213694091662, + "grad_norm": 0.9678462743759155, + "learning_rate": 1.5358101276213179e-07, + "loss": 0.5217, + "step": 6441 + }, + { + "epoch": 5.3329652125897296, + "grad_norm": 0.9798896908760071, + "learning_rate": 1.5320174197373416e-07, + "loss": 0.5463, + "step": 6442 + }, + { + "epoch": 5.333793484262838, + "grad_norm": 0.976833701133728, + "learning_rate": 1.528229252689234e-07, + "loss": 0.5612, + "step": 6443 + }, + { + "epoch": 5.334621755935947, + "grad_norm": 0.966899037361145, + "learning_rate": 1.5244456272099612e-07, + "loss": 0.5504, + "step": 6444 + }, + { + "epoch": 5.335450027609056, + "grad_norm": 0.968900203704834, + "learning_rate": 1.5206665440316208e-07, + "loss": 0.5165, + "step": 6445 + }, + { + "epoch": 5.3362782992821645, + "grad_norm": 1.0155616998672485, + "learning_rate": 1.5168920038854395e-07, + "loss": 0.5568, + "step": 6446 + }, + { + "epoch": 5.337106570955274, + "grad_norm": 0.9766435027122498, + "learning_rate": 1.5131220075017466e-07, + "loss": 0.5552, + "step": 6447 + }, + { + "epoch": 5.337934842628382, + "grad_norm": 0.9692011475563049, + "learning_rate": 1.5093565556100005e-07, + "loss": 0.5548, + "step": 6448 + }, + { + "epoch": 5.338763114301491, + "grad_norm": 0.9665151238441467, + "learning_rate": 1.5055956489387847e-07, + "loss": 0.5425, + "step": 6449 + }, + { + "epoch": 5.339591385974599, + "grad_norm": 0.9732781052589417, + "learning_rate": 1.5018392882158005e-07, + "loss": 0.5538, + "step": 6450 + }, + { + "epoch": 5.3404196576477085, + "grad_norm": 0.9849833846092224, + "learning_rate": 1.4980874741678642e-07, + "loss": 0.5166, + "step": 6451 + }, + { + "epoch": 5.341247929320817, + "grad_norm": 0.9553214311599731, + "learning_rate": 1.4943402075209197e-07, + "loss": 0.5475, + "step": 6452 + }, + { + "epoch": 5.342076200993926, + "grad_norm": 1.0181409120559692, + "learning_rate": 1.4905974890000313e-07, + "loss": 0.5401, + "step": 6453 + }, + { + "epoch": 5.342904472667035, + "grad_norm": 0.9713455438613892, + "learning_rate": 1.4868593193293757e-07, + "loss": 0.5617, + "step": 6454 + }, + { + "epoch": 5.3437327443401434, + "grad_norm": 0.9756832718849182, + "learning_rate": 1.4831256992322574e-07, + "loss": 0.5534, + "step": 6455 + }, + { + "epoch": 5.344561016013253, + "grad_norm": 0.982082188129425, + "learning_rate": 1.4793966294310901e-07, + "loss": 0.5475, + "step": 6456 + }, + { + "epoch": 5.345389287686361, + "grad_norm": 0.9937077164649963, + "learning_rate": 1.475672110647422e-07, + "loss": 0.5288, + "step": 6457 + }, + { + "epoch": 5.34621755935947, + "grad_norm": 0.935522735118866, + "learning_rate": 1.4719521436019067e-07, + "loss": 0.5347, + "step": 6458 + }, + { + "epoch": 5.347045831032578, + "grad_norm": 0.9514837265014648, + "learning_rate": 1.4682367290143302e-07, + "loss": 0.5505, + "step": 6459 + }, + { + "epoch": 5.3478741027056875, + "grad_norm": 0.9884480834007263, + "learning_rate": 1.4645258676035805e-07, + "loss": 0.5186, + "step": 6460 + }, + { + "epoch": 5.348702374378796, + "grad_norm": 1.0336134433746338, + "learning_rate": 1.460819560087684e-07, + "loss": 0.5425, + "step": 6461 + }, + { + "epoch": 5.349530646051905, + "grad_norm": 0.975571870803833, + "learning_rate": 1.4571178071837748e-07, + "loss": 0.5417, + "step": 6462 + }, + { + "epoch": 5.350358917725014, + "grad_norm": 0.9570359587669373, + "learning_rate": 1.4534206096081026e-07, + "loss": 0.5629, + "step": 6463 + }, + { + "epoch": 5.351187189398122, + "grad_norm": 0.9685426354408264, + "learning_rate": 1.4497279680760424e-07, + "loss": 0.5502, + "step": 6464 + }, + { + "epoch": 5.352015461071232, + "grad_norm": 0.9712944626808167, + "learning_rate": 1.4460398833020944e-07, + "loss": 0.5298, + "step": 6465 + }, + { + "epoch": 5.35284373274434, + "grad_norm": 0.9540491104125977, + "learning_rate": 1.4423563559998549e-07, + "loss": 0.5513, + "step": 6466 + }, + { + "epoch": 5.353672004417449, + "grad_norm": 0.9860627055168152, + "learning_rate": 1.438677386882062e-07, + "loss": 0.5569, + "step": 6467 + }, + { + "epoch": 5.354500276090557, + "grad_norm": 0.9652723073959351, + "learning_rate": 1.4350029766605628e-07, + "loss": 0.5482, + "step": 6468 + }, + { + "epoch": 5.3553285477636665, + "grad_norm": 0.968623697757721, + "learning_rate": 1.4313331260463137e-07, + "loss": 0.5345, + "step": 6469 + }, + { + "epoch": 5.356156819436775, + "grad_norm": 0.9684195518493652, + "learning_rate": 1.4276678357493968e-07, + "loss": 0.5588, + "step": 6470 + }, + { + "epoch": 5.356985091109884, + "grad_norm": 0.9607696533203125, + "learning_rate": 1.4240071064790224e-07, + "loss": 0.5526, + "step": 6471 + }, + { + "epoch": 5.357813362782993, + "grad_norm": 0.9947494864463806, + "learning_rate": 1.4203509389434933e-07, + "loss": 0.5359, + "step": 6472 + }, + { + "epoch": 5.358641634456101, + "grad_norm": 0.9699355363845825, + "learning_rate": 1.4166993338502522e-07, + "loss": 0.5444, + "step": 6473 + }, + { + "epoch": 5.359469906129211, + "grad_norm": 0.964187741279602, + "learning_rate": 1.4130522919058532e-07, + "loss": 0.5377, + "step": 6474 + }, + { + "epoch": 5.360298177802319, + "grad_norm": 0.9660773277282715, + "learning_rate": 1.4094098138159594e-07, + "loss": 0.5465, + "step": 6475 + }, + { + "epoch": 5.361126449475428, + "grad_norm": 0.9664810299873352, + "learning_rate": 1.4057719002853548e-07, + "loss": 0.5482, + "step": 6476 + }, + { + "epoch": 5.361954721148536, + "grad_norm": 1.0010203123092651, + "learning_rate": 1.4021385520179508e-07, + "loss": 0.5242, + "step": 6477 + }, + { + "epoch": 5.3627829928216455, + "grad_norm": 0.9833728075027466, + "learning_rate": 1.3985097697167548e-07, + "loss": 0.5573, + "step": 6478 + }, + { + "epoch": 5.363611264494755, + "grad_norm": 0.9455319046974182, + "learning_rate": 1.3948855540839078e-07, + "loss": 0.5174, + "step": 6479 + }, + { + "epoch": 5.364439536167863, + "grad_norm": 0.9610325694084167, + "learning_rate": 1.3912659058206685e-07, + "loss": 0.5419, + "step": 6480 + }, + { + "epoch": 5.365267807840972, + "grad_norm": 0.9534977674484253, + "learning_rate": 1.3876508256273957e-07, + "loss": 0.5467, + "step": 6481 + }, + { + "epoch": 5.36609607951408, + "grad_norm": 0.9848024249076843, + "learning_rate": 1.3840403142035747e-07, + "loss": 0.5497, + "step": 6482 + }, + { + "epoch": 5.36692435118719, + "grad_norm": 0.9774238467216492, + "learning_rate": 1.3804343722478074e-07, + "loss": 0.5627, + "step": 6483 + }, + { + "epoch": 5.367752622860298, + "grad_norm": 0.9779729247093201, + "learning_rate": 1.376833000457814e-07, + "loss": 0.5558, + "step": 6484 + }, + { + "epoch": 5.368580894533407, + "grad_norm": 0.9725696444511414, + "learning_rate": 1.37323619953042e-07, + "loss": 0.5383, + "step": 6485 + }, + { + "epoch": 5.369409166206516, + "grad_norm": 0.9570741653442383, + "learning_rate": 1.3696439701615744e-07, + "loss": 0.5645, + "step": 6486 + }, + { + "epoch": 5.3702374378796245, + "grad_norm": 0.9903878569602966, + "learning_rate": 1.3660563130463432e-07, + "loss": 0.5331, + "step": 6487 + }, + { + "epoch": 5.371065709552734, + "grad_norm": 0.9721629619598389, + "learning_rate": 1.362473228878902e-07, + "loss": 0.5579, + "step": 6488 + }, + { + "epoch": 5.371893981225842, + "grad_norm": 0.9872861504554749, + "learning_rate": 1.3588947183525437e-07, + "loss": 0.5373, + "step": 6489 + }, + { + "epoch": 5.372722252898951, + "grad_norm": 1.0035595893859863, + "learning_rate": 1.3553207821596803e-07, + "loss": 0.5387, + "step": 6490 + }, + { + "epoch": 5.373550524572059, + "grad_norm": 0.9744288325309753, + "learning_rate": 1.3517514209918314e-07, + "loss": 0.5674, + "step": 6491 + }, + { + "epoch": 5.374378796245169, + "grad_norm": 0.9723195433616638, + "learning_rate": 1.348186635539636e-07, + "loss": 0.5482, + "step": 6492 + }, + { + "epoch": 5.375207067918277, + "grad_norm": 0.9631553888320923, + "learning_rate": 1.3446264264928511e-07, + "loss": 0.5526, + "step": 6493 + }, + { + "epoch": 5.376035339591386, + "grad_norm": 0.9763345122337341, + "learning_rate": 1.3410707945403334e-07, + "loss": 0.55, + "step": 6494 + }, + { + "epoch": 5.376863611264495, + "grad_norm": 0.969550609588623, + "learning_rate": 1.337519740370072e-07, + "loss": 0.5496, + "step": 6495 + }, + { + "epoch": 5.3776918829376035, + "grad_norm": 0.963131844997406, + "learning_rate": 1.3339732646691667e-07, + "loss": 0.5306, + "step": 6496 + }, + { + "epoch": 5.378520154610713, + "grad_norm": 0.9639487862586975, + "learning_rate": 1.3304313681238191e-07, + "loss": 0.5351, + "step": 6497 + }, + { + "epoch": 5.379348426283821, + "grad_norm": 1.0015922784805298, + "learning_rate": 1.3268940514193585e-07, + "loss": 0.5435, + "step": 6498 + }, + { + "epoch": 5.38017669795693, + "grad_norm": 0.981356680393219, + "learning_rate": 1.323361315240221e-07, + "loss": 0.5469, + "step": 6499 + }, + { + "epoch": 5.381004969630038, + "grad_norm": 0.9677076935768127, + "learning_rate": 1.3198331602699566e-07, + "loss": 0.5296, + "step": 6500 + }, + { + "epoch": 5.3818332413031476, + "grad_norm": 0.9949377179145813, + "learning_rate": 1.3163095871912335e-07, + "loss": 0.5514, + "step": 6501 + }, + { + "epoch": 5.382661512976256, + "grad_norm": 0.9958040714263916, + "learning_rate": 1.3127905966858285e-07, + "loss": 0.5419, + "step": 6502 + }, + { + "epoch": 5.383489784649365, + "grad_norm": 0.984910249710083, + "learning_rate": 1.3092761894346356e-07, + "loss": 0.5478, + "step": 6503 + }, + { + "epoch": 5.384318056322474, + "grad_norm": 1.0185965299606323, + "learning_rate": 1.3057663661176557e-07, + "loss": 0.5421, + "step": 6504 + }, + { + "epoch": 5.3851463279955825, + "grad_norm": 0.9846193790435791, + "learning_rate": 1.302261127414009e-07, + "loss": 0.537, + "step": 6505 + }, + { + "epoch": 5.385974599668692, + "grad_norm": 0.9657633304595947, + "learning_rate": 1.2987604740019282e-07, + "loss": 0.5518, + "step": 6506 + }, + { + "epoch": 5.3868028713418, + "grad_norm": 0.9823387265205383, + "learning_rate": 1.295264406558752e-07, + "loss": 0.5755, + "step": 6507 + }, + { + "epoch": 5.387631143014909, + "grad_norm": 0.9645293951034546, + "learning_rate": 1.2917729257609423e-07, + "loss": 0.5538, + "step": 6508 + }, + { + "epoch": 5.388459414688017, + "grad_norm": 0.9698269963264465, + "learning_rate": 1.2882860322840694e-07, + "loss": 0.533, + "step": 6509 + }, + { + "epoch": 5.3892876863611265, + "grad_norm": 0.9648250341415405, + "learning_rate": 1.284803726802808e-07, + "loss": 0.542, + "step": 6510 + }, + { + "epoch": 5.390115958034235, + "grad_norm": 0.9796745777130127, + "learning_rate": 1.2813260099909546e-07, + "loss": 0.5621, + "step": 6511 + }, + { + "epoch": 5.390944229707344, + "grad_norm": 0.9613428711891174, + "learning_rate": 1.2778528825214187e-07, + "loss": 0.532, + "step": 6512 + }, + { + "epoch": 5.391772501380453, + "grad_norm": 0.9411634206771851, + "learning_rate": 1.2743843450662125e-07, + "loss": 0.5405, + "step": 6513 + }, + { + "epoch": 5.3926007730535614, + "grad_norm": 0.9709793925285339, + "learning_rate": 1.2709203982964714e-07, + "loss": 0.5642, + "step": 6514 + }, + { + "epoch": 5.393429044726671, + "grad_norm": 0.9789018034934998, + "learning_rate": 1.2674610428824314e-07, + "loss": 0.5354, + "step": 6515 + }, + { + "epoch": 5.394257316399779, + "grad_norm": 0.9703934192657471, + "learning_rate": 1.264006279493446e-07, + "loss": 0.5246, + "step": 6516 + }, + { + "epoch": 5.395085588072888, + "grad_norm": 0.9630893468856812, + "learning_rate": 1.2605561087979805e-07, + "loss": 0.5496, + "step": 6517 + }, + { + "epoch": 5.395913859745996, + "grad_norm": 0.9630357027053833, + "learning_rate": 1.2571105314636177e-07, + "loss": 0.5505, + "step": 6518 + }, + { + "epoch": 5.3967421314191055, + "grad_norm": 0.9539121389389038, + "learning_rate": 1.2536695481570327e-07, + "loss": 0.5471, + "step": 6519 + }, + { + "epoch": 5.397570403092214, + "grad_norm": 0.9708060026168823, + "learning_rate": 1.2502331595440313e-07, + "loss": 0.5692, + "step": 6520 + }, + { + "epoch": 5.398398674765323, + "grad_norm": 0.9790439009666443, + "learning_rate": 1.246801366289524e-07, + "loss": 0.5531, + "step": 6521 + }, + { + "epoch": 5.399226946438432, + "grad_norm": 0.9796111583709717, + "learning_rate": 1.2433741690575235e-07, + "loss": 0.5425, + "step": 6522 + }, + { + "epoch": 5.40005521811154, + "grad_norm": 0.9549072980880737, + "learning_rate": 1.2399515685111663e-07, + "loss": 0.5408, + "step": 6523 + }, + { + "epoch": 5.40088348978465, + "grad_norm": 0.9418009519577026, + "learning_rate": 1.2365335653126946e-07, + "loss": 0.5242, + "step": 6524 + }, + { + "epoch": 5.401711761457758, + "grad_norm": 0.9835243821144104, + "learning_rate": 1.2331201601234576e-07, + "loss": 0.5636, + "step": 6525 + }, + { + "epoch": 5.402540033130867, + "grad_norm": 0.9590522050857544, + "learning_rate": 1.229711353603913e-07, + "loss": 0.54, + "step": 6526 + }, + { + "epoch": 5.403368304803975, + "grad_norm": 0.9823058843612671, + "learning_rate": 1.226307146413641e-07, + "loss": 0.5376, + "step": 6527 + }, + { + "epoch": 5.4041965764770845, + "grad_norm": 0.9607473611831665, + "learning_rate": 1.222907539211321e-07, + "loss": 0.5372, + "step": 6528 + }, + { + "epoch": 5.405024848150193, + "grad_norm": 0.9673418998718262, + "learning_rate": 1.219512532654743e-07, + "loss": 0.5394, + "step": 6529 + }, + { + "epoch": 5.405853119823302, + "grad_norm": 0.9815991520881653, + "learning_rate": 1.216122127400815e-07, + "loss": 0.5392, + "step": 6530 + }, + { + "epoch": 5.406681391496411, + "grad_norm": 0.9607836008071899, + "learning_rate": 1.21273632410554e-07, + "loss": 0.5597, + "step": 6531 + }, + { + "epoch": 5.407509663169519, + "grad_norm": 0.9671915173530579, + "learning_rate": 1.209355123424044e-07, + "loss": 0.5387, + "step": 6532 + }, + { + "epoch": 5.408337934842629, + "grad_norm": 0.9607694149017334, + "learning_rate": 1.2059785260105622e-07, + "loss": 0.5418, + "step": 6533 + }, + { + "epoch": 5.409166206515737, + "grad_norm": 0.9608633518218994, + "learning_rate": 1.20260653251843e-07, + "loss": 0.5329, + "step": 6534 + }, + { + "epoch": 5.409994478188846, + "grad_norm": 0.9839193224906921, + "learning_rate": 1.1992391436000921e-07, + "loss": 0.549, + "step": 6535 + }, + { + "epoch": 5.410822749861954, + "grad_norm": 0.9589143395423889, + "learning_rate": 1.1958763599071194e-07, + "loss": 0.5422, + "step": 6536 + }, + { + "epoch": 5.4116510215350635, + "grad_norm": 0.9641323089599609, + "learning_rate": 1.1925181820901688e-07, + "loss": 0.5372, + "step": 6537 + }, + { + "epoch": 5.412479293208173, + "grad_norm": 0.9806573390960693, + "learning_rate": 1.1891646107990179e-07, + "loss": 0.5283, + "step": 6538 + }, + { + "epoch": 5.413307564881281, + "grad_norm": 0.9691280126571655, + "learning_rate": 1.1858156466825532e-07, + "loss": 0.5466, + "step": 6539 + }, + { + "epoch": 5.41413583655439, + "grad_norm": 0.9809518456459045, + "learning_rate": 1.1824712903887703e-07, + "loss": 0.5561, + "step": 6540 + }, + { + "epoch": 5.414964108227498, + "grad_norm": 0.9775456190109253, + "learning_rate": 1.1791315425647682e-07, + "loss": 0.5168, + "step": 6541 + }, + { + "epoch": 5.415792379900608, + "grad_norm": 0.9600995779037476, + "learning_rate": 1.1757964038567521e-07, + "loss": 0.5548, + "step": 6542 + }, + { + "epoch": 5.416620651573716, + "grad_norm": 0.9704177379608154, + "learning_rate": 1.1724658749100532e-07, + "loss": 0.5372, + "step": 6543 + }, + { + "epoch": 5.417448923246825, + "grad_norm": 0.9765699505805969, + "learning_rate": 1.1691399563690836e-07, + "loss": 0.5315, + "step": 6544 + }, + { + "epoch": 5.418277194919934, + "grad_norm": 0.9948807954788208, + "learning_rate": 1.165818648877387e-07, + "loss": 0.5497, + "step": 6545 + }, + { + "epoch": 5.4191054665930425, + "grad_norm": 0.9774356484413147, + "learning_rate": 1.1625019530776049e-07, + "loss": 0.5462, + "step": 6546 + }, + { + "epoch": 5.419933738266152, + "grad_norm": 0.9601755142211914, + "learning_rate": 1.1591898696114795e-07, + "loss": 0.5361, + "step": 6547 + }, + { + "epoch": 5.42076200993926, + "grad_norm": 0.9878173470497131, + "learning_rate": 1.1558823991198786e-07, + "loss": 0.542, + "step": 6548 + }, + { + "epoch": 5.421590281612369, + "grad_norm": 0.9504754543304443, + "learning_rate": 1.15257954224276e-07, + "loss": 0.5815, + "step": 6549 + }, + { + "epoch": 5.422418553285477, + "grad_norm": 0.9804036617279053, + "learning_rate": 1.1492812996191927e-07, + "loss": 0.5577, + "step": 6550 + }, + { + "epoch": 5.423246824958587, + "grad_norm": 0.9689943194389343, + "learning_rate": 1.1459876718873637e-07, + "loss": 0.5455, + "step": 6551 + }, + { + "epoch": 5.424075096631695, + "grad_norm": 0.9837014675140381, + "learning_rate": 1.1426986596845547e-07, + "loss": 0.5449, + "step": 6552 + }, + { + "epoch": 5.424903368304804, + "grad_norm": 0.9726439118385315, + "learning_rate": 1.1394142636471595e-07, + "loss": 0.5257, + "step": 6553 + }, + { + "epoch": 5.425731639977913, + "grad_norm": 0.9793080687522888, + "learning_rate": 1.1361344844106809e-07, + "loss": 0.5499, + "step": 6554 + }, + { + "epoch": 5.4265599116510215, + "grad_norm": 0.9626556038856506, + "learning_rate": 1.1328593226097223e-07, + "loss": 0.5285, + "step": 6555 + }, + { + "epoch": 5.427388183324131, + "grad_norm": 0.9585309028625488, + "learning_rate": 1.129588778877999e-07, + "loss": 0.5393, + "step": 6556 + }, + { + "epoch": 5.428216454997239, + "grad_norm": 0.9659401178359985, + "learning_rate": 1.1263228538483273e-07, + "loss": 0.52, + "step": 6557 + }, + { + "epoch": 5.429044726670348, + "grad_norm": 0.9528523087501526, + "learning_rate": 1.1230615481526403e-07, + "loss": 0.5466, + "step": 6558 + }, + { + "epoch": 5.429872998343456, + "grad_norm": 0.9761300086975098, + "learning_rate": 1.1198048624219637e-07, + "loss": 0.55, + "step": 6559 + }, + { + "epoch": 5.4307012700165656, + "grad_norm": 0.9895758628845215, + "learning_rate": 1.1165527972864355e-07, + "loss": 0.57, + "step": 6560 + }, + { + "epoch": 5.431529541689674, + "grad_norm": 0.9610238075256348, + "learning_rate": 1.1133053533753047e-07, + "loss": 0.5466, + "step": 6561 + }, + { + "epoch": 5.432357813362783, + "grad_norm": 0.987244188785553, + "learning_rate": 1.1100625313169189e-07, + "loss": 0.5378, + "step": 6562 + }, + { + "epoch": 5.433186085035892, + "grad_norm": 0.9941465258598328, + "learning_rate": 1.1068243317387345e-07, + "loss": 0.5442, + "step": 6563 + }, + { + "epoch": 5.4340143567090005, + "grad_norm": 0.9611039161682129, + "learning_rate": 1.1035907552673086e-07, + "loss": 0.5417, + "step": 6564 + }, + { + "epoch": 5.43484262838211, + "grad_norm": 0.9905521273612976, + "learning_rate": 1.1003618025283185e-07, + "loss": 0.5445, + "step": 6565 + }, + { + "epoch": 5.435670900055218, + "grad_norm": 0.9693180918693542, + "learning_rate": 1.0971374741465257e-07, + "loss": 0.5519, + "step": 6566 + }, + { + "epoch": 5.436499171728327, + "grad_norm": 0.9613069891929626, + "learning_rate": 1.0939177707458115e-07, + "loss": 0.5424, + "step": 6567 + }, + { + "epoch": 5.437327443401435, + "grad_norm": 0.9792628288269043, + "learning_rate": 1.090702692949161e-07, + "loss": 0.5503, + "step": 6568 + }, + { + "epoch": 5.4381557150745445, + "grad_norm": 0.969942569732666, + "learning_rate": 1.0874922413786599e-07, + "loss": 0.5472, + "step": 6569 + }, + { + "epoch": 5.438983986747653, + "grad_norm": 0.9609864354133606, + "learning_rate": 1.0842864166555028e-07, + "loss": 0.5452, + "step": 6570 + }, + { + "epoch": 5.439812258420762, + "grad_norm": 0.9580171704292297, + "learning_rate": 1.0810852193999827e-07, + "loss": 0.5393, + "step": 6571 + }, + { + "epoch": 5.440640530093871, + "grad_norm": 0.9603419899940491, + "learning_rate": 1.0778886502315011e-07, + "loss": 0.5343, + "step": 6572 + }, + { + "epoch": 5.4414688017669794, + "grad_norm": 0.9555606245994568, + "learning_rate": 1.0746967097685662e-07, + "loss": 0.5243, + "step": 6573 + }, + { + "epoch": 5.442297073440089, + "grad_norm": 0.9738579392433167, + "learning_rate": 1.071509398628795e-07, + "loss": 0.551, + "step": 6574 + }, + { + "epoch": 5.443125345113197, + "grad_norm": 1.0075610876083374, + "learning_rate": 1.0683267174288886e-07, + "loss": 0.5714, + "step": 6575 + }, + { + "epoch": 5.443953616786306, + "grad_norm": 0.9633387923240662, + "learning_rate": 1.0651486667846794e-07, + "loss": 0.5317, + "step": 6576 + }, + { + "epoch": 5.444781888459414, + "grad_norm": 0.9699281454086304, + "learning_rate": 1.0619752473110839e-07, + "loss": 0.5309, + "step": 6577 + }, + { + "epoch": 5.4456101601325235, + "grad_norm": 0.9670154452323914, + "learning_rate": 1.0588064596221303e-07, + "loss": 0.529, + "step": 6578 + }, + { + "epoch": 5.446438431805632, + "grad_norm": 0.9683130979537964, + "learning_rate": 1.0556423043309505e-07, + "loss": 0.5465, + "step": 6579 + }, + { + "epoch": 5.447266703478741, + "grad_norm": 0.9681844711303711, + "learning_rate": 1.0524827820497796e-07, + "loss": 0.5318, + "step": 6580 + }, + { + "epoch": 5.44809497515185, + "grad_norm": 0.9617595672607422, + "learning_rate": 1.0493278933899536e-07, + "loss": 0.5355, + "step": 6581 + }, + { + "epoch": 5.448923246824958, + "grad_norm": 0.9680930376052856, + "learning_rate": 1.0461776389619149e-07, + "loss": 0.5284, + "step": 6582 + }, + { + "epoch": 5.449751518498068, + "grad_norm": 0.9545103907585144, + "learning_rate": 1.0430320193752064e-07, + "loss": 0.5475, + "step": 6583 + }, + { + "epoch": 5.450579790171176, + "grad_norm": 0.993318498134613, + "learning_rate": 1.03989103523848e-07, + "loss": 0.5254, + "step": 6584 + }, + { + "epoch": 5.451408061844285, + "grad_norm": 0.962974488735199, + "learning_rate": 1.036754687159483e-07, + "loss": 0.5657, + "step": 6585 + }, + { + "epoch": 5.452236333517393, + "grad_norm": 0.9528799653053284, + "learning_rate": 1.0336229757450717e-07, + "loss": 0.5366, + "step": 6586 + }, + { + "epoch": 5.4530646051905025, + "grad_norm": 0.9792569875717163, + "learning_rate": 1.0304959016012056e-07, + "loss": 0.5598, + "step": 6587 + }, + { + "epoch": 5.453892876863611, + "grad_norm": 0.9835836291313171, + "learning_rate": 1.0273734653329398e-07, + "loss": 0.5532, + "step": 6588 + }, + { + "epoch": 5.45472114853672, + "grad_norm": 0.9834234118461609, + "learning_rate": 1.0242556675444382e-07, + "loss": 0.5645, + "step": 6589 + }, + { + "epoch": 5.455549420209829, + "grad_norm": 0.981913685798645, + "learning_rate": 1.0211425088389682e-07, + "loss": 0.5479, + "step": 6590 + }, + { + "epoch": 5.456377691882937, + "grad_norm": 1.038151741027832, + "learning_rate": 1.0180339898188924e-07, + "loss": 0.552, + "step": 6591 + }, + { + "epoch": 5.457205963556047, + "grad_norm": 0.9866636991500854, + "learning_rate": 1.0149301110856852e-07, + "loss": 0.5409, + "step": 6592 + }, + { + "epoch": 5.458034235229155, + "grad_norm": 0.986129879951477, + "learning_rate": 1.0118308732399162e-07, + "loss": 0.5599, + "step": 6593 + }, + { + "epoch": 5.458862506902264, + "grad_norm": 0.9705114960670471, + "learning_rate": 1.0087362768812531e-07, + "loss": 0.5653, + "step": 6594 + }, + { + "epoch": 5.459690778575372, + "grad_norm": 1.0074405670166016, + "learning_rate": 1.0056463226084778e-07, + "loss": 0.5556, + "step": 6595 + }, + { + "epoch": 5.4605190502484815, + "grad_norm": 0.9861671328544617, + "learning_rate": 1.0025610110194706e-07, + "loss": 0.5334, + "step": 6596 + }, + { + "epoch": 5.461347321921591, + "grad_norm": 0.9770742654800415, + "learning_rate": 9.994803427112038e-08, + "loss": 0.5218, + "step": 6597 + }, + { + "epoch": 5.462175593594699, + "grad_norm": 0.9769787788391113, + "learning_rate": 9.96404318279759e-08, + "loss": 0.5342, + "step": 6598 + }, + { + "epoch": 5.463003865267808, + "grad_norm": 0.9732609391212463, + "learning_rate": 9.933329383203239e-08, + "loss": 0.5402, + "step": 6599 + }, + { + "epoch": 5.463832136940916, + "grad_norm": 0.9503108263015747, + "learning_rate": 9.902662034271731e-08, + "loss": 0.551, + "step": 6600 + }, + { + "epoch": 5.464660408614026, + "grad_norm": 0.973845899105072, + "learning_rate": 9.872041141936989e-08, + "loss": 0.54, + "step": 6601 + }, + { + "epoch": 5.465488680287134, + "grad_norm": 0.9672338366508484, + "learning_rate": 9.841466712123853e-08, + "loss": 0.5486, + "step": 6602 + }, + { + "epoch": 5.466316951960243, + "grad_norm": 0.9420118927955627, + "learning_rate": 9.81093875074815e-08, + "loss": 0.5698, + "step": 6603 + }, + { + "epoch": 5.467145223633352, + "grad_norm": 0.9618524312973022, + "learning_rate": 9.780457263716819e-08, + "loss": 0.5442, + "step": 6604 + }, + { + "epoch": 5.4679734953064605, + "grad_norm": 0.9696351885795593, + "learning_rate": 9.750022256927699e-08, + "loss": 0.5521, + "step": 6605 + }, + { + "epoch": 5.46880176697957, + "grad_norm": 0.9931103587150574, + "learning_rate": 9.71963373626969e-08, + "loss": 0.5456, + "step": 6606 + }, + { + "epoch": 5.469630038652678, + "grad_norm": 0.9815472960472107, + "learning_rate": 9.689291707622673e-08, + "loss": 0.5448, + "step": 6607 + }, + { + "epoch": 5.470458310325787, + "grad_norm": 0.9499985575675964, + "learning_rate": 9.658996176857588e-08, + "loss": 0.5591, + "step": 6608 + }, + { + "epoch": 5.471286581998895, + "grad_norm": 0.9819093942642212, + "learning_rate": 9.628747149836276e-08, + "loss": 0.5382, + "step": 6609 + }, + { + "epoch": 5.472114853672005, + "grad_norm": 0.9757983088493347, + "learning_rate": 9.598544632411693e-08, + "loss": 0.5384, + "step": 6610 + }, + { + "epoch": 5.472943125345113, + "grad_norm": 0.9771463871002197, + "learning_rate": 9.56838863042775e-08, + "loss": 0.5497, + "step": 6611 + }, + { + "epoch": 5.473771397018222, + "grad_norm": 0.9590486288070679, + "learning_rate": 9.538279149719304e-08, + "loss": 0.5458, + "step": 6612 + }, + { + "epoch": 5.474599668691331, + "grad_norm": 0.975226104259491, + "learning_rate": 9.50821619611228e-08, + "loss": 0.5483, + "step": 6613 + }, + { + "epoch": 5.4754279403644395, + "grad_norm": 0.9690378904342651, + "learning_rate": 9.478199775423636e-08, + "loss": 0.5378, + "step": 6614 + }, + { + "epoch": 5.476256212037549, + "grad_norm": 0.9699209332466125, + "learning_rate": 9.448229893461197e-08, + "loss": 0.5473, + "step": 6615 + }, + { + "epoch": 5.477084483710657, + "grad_norm": 0.9918218851089478, + "learning_rate": 9.418306556023882e-08, + "loss": 0.5348, + "step": 6616 + }, + { + "epoch": 5.477912755383766, + "grad_norm": 0.9537071585655212, + "learning_rate": 9.38842976890153e-08, + "loss": 0.5509, + "step": 6617 + }, + { + "epoch": 5.478741027056874, + "grad_norm": 0.9812236428260803, + "learning_rate": 9.35859953787513e-08, + "loss": 0.5393, + "step": 6618 + }, + { + "epoch": 5.4795692987299836, + "grad_norm": 0.965298593044281, + "learning_rate": 9.32881586871645e-08, + "loss": 0.5441, + "step": 6619 + }, + { + "epoch": 5.480397570403092, + "grad_norm": 0.9991268515586853, + "learning_rate": 9.299078767188386e-08, + "loss": 0.5546, + "step": 6620 + }, + { + "epoch": 5.481225842076201, + "grad_norm": 0.9741199016571045, + "learning_rate": 9.26938823904483e-08, + "loss": 0.524, + "step": 6621 + }, + { + "epoch": 5.48205411374931, + "grad_norm": 0.9905590415000916, + "learning_rate": 9.23974429003055e-08, + "loss": 0.5542, + "step": 6622 + }, + { + "epoch": 5.4828823854224185, + "grad_norm": 0.9653900861740112, + "learning_rate": 9.210146925881402e-08, + "loss": 0.5415, + "step": 6623 + }, + { + "epoch": 5.483710657095528, + "grad_norm": 0.9884553551673889, + "learning_rate": 9.18059615232425e-08, + "loss": 0.5605, + "step": 6624 + }, + { + "epoch": 5.484538928768636, + "grad_norm": 1.006697177886963, + "learning_rate": 9.151091975076825e-08, + "loss": 0.544, + "step": 6625 + }, + { + "epoch": 5.485367200441745, + "grad_norm": 0.9931198954582214, + "learning_rate": 9.12163439984795e-08, + "loss": 0.556, + "step": 6626 + }, + { + "epoch": 5.486195472114853, + "grad_norm": 0.9848148822784424, + "learning_rate": 9.092223432337427e-08, + "loss": 0.5516, + "step": 6627 + }, + { + "epoch": 5.4870237437879625, + "grad_norm": 1.007230281829834, + "learning_rate": 9.062859078235897e-08, + "loss": 0.5316, + "step": 6628 + }, + { + "epoch": 5.487852015461071, + "grad_norm": 0.9918413758277893, + "learning_rate": 9.03354134322515e-08, + "loss": 0.566, + "step": 6629 + }, + { + "epoch": 5.48868028713418, + "grad_norm": 0.9656996130943298, + "learning_rate": 9.004270232977896e-08, + "loss": 0.5422, + "step": 6630 + }, + { + "epoch": 5.489508558807289, + "grad_norm": 0.9777333736419678, + "learning_rate": 8.97504575315783e-08, + "loss": 0.556, + "step": 6631 + }, + { + "epoch": 5.4903368304803974, + "grad_norm": 0.9547977447509766, + "learning_rate": 8.945867909419592e-08, + "loss": 0.551, + "step": 6632 + }, + { + "epoch": 5.491165102153507, + "grad_norm": 0.9571183919906616, + "learning_rate": 8.916736707408862e-08, + "loss": 0.5529, + "step": 6633 + }, + { + "epoch": 5.491993373826615, + "grad_norm": 0.996627926826477, + "learning_rate": 8.887652152762188e-08, + "loss": 0.5464, + "step": 6634 + }, + { + "epoch": 5.492821645499724, + "grad_norm": 0.9693316221237183, + "learning_rate": 8.858614251107205e-08, + "loss": 0.5382, + "step": 6635 + }, + { + "epoch": 5.493649917172832, + "grad_norm": 0.9636437296867371, + "learning_rate": 8.829623008062504e-08, + "loss": 0.5298, + "step": 6636 + }, + { + "epoch": 5.4944781888459415, + "grad_norm": 0.9786088466644287, + "learning_rate": 8.80067842923757e-08, + "loss": 0.5683, + "step": 6637 + }, + { + "epoch": 5.49530646051905, + "grad_norm": 0.9921557307243347, + "learning_rate": 8.771780520232948e-08, + "loss": 0.5352, + "step": 6638 + }, + { + "epoch": 5.496134732192159, + "grad_norm": 0.9634941816329956, + "learning_rate": 8.742929286640084e-08, + "loss": 0.5619, + "step": 6639 + }, + { + "epoch": 5.496963003865268, + "grad_norm": 1.0045983791351318, + "learning_rate": 8.714124734041429e-08, + "loss": 0.5601, + "step": 6640 + }, + { + "epoch": 5.497791275538376, + "grad_norm": 0.9863957762718201, + "learning_rate": 8.685366868010415e-08, + "loss": 0.5366, + "step": 6641 + }, + { + "epoch": 5.498619547211486, + "grad_norm": 0.9702649712562561, + "learning_rate": 8.656655694111365e-08, + "loss": 0.5387, + "step": 6642 + }, + { + "epoch": 5.499447818884594, + "grad_norm": 0.9980773329734802, + "learning_rate": 8.6279912178997e-08, + "loss": 0.5492, + "step": 6643 + }, + { + "epoch": 5.500276090557703, + "grad_norm": 0.9538503885269165, + "learning_rate": 8.599373444921677e-08, + "loss": 0.565, + "step": 6644 + }, + { + "epoch": 5.501104362230811, + "grad_norm": 0.9639506340026855, + "learning_rate": 8.570802380714588e-08, + "loss": 0.5409, + "step": 6645 + }, + { + "epoch": 5.5019326339039205, + "grad_norm": 0.9807303547859192, + "learning_rate": 8.542278030806678e-08, + "loss": 0.5512, + "step": 6646 + }, + { + "epoch": 5.502760905577029, + "grad_norm": 0.9546551704406738, + "learning_rate": 8.513800400717115e-08, + "loss": 0.5223, + "step": 6647 + }, + { + "epoch": 5.503589177250138, + "grad_norm": 0.9704589247703552, + "learning_rate": 8.485369495956075e-08, + "loss": 0.5303, + "step": 6648 + }, + { + "epoch": 5.504417448923247, + "grad_norm": 0.9702087640762329, + "learning_rate": 8.456985322024686e-08, + "loss": 0.5302, + "step": 6649 + }, + { + "epoch": 5.505245720596355, + "grad_norm": 0.9621816277503967, + "learning_rate": 8.42864788441497e-08, + "loss": 0.5309, + "step": 6650 + }, + { + "epoch": 5.506073992269465, + "grad_norm": 0.9710858464241028, + "learning_rate": 8.400357188609986e-08, + "loss": 0.5473, + "step": 6651 + }, + { + "epoch": 5.506902263942573, + "grad_norm": 0.9770440459251404, + "learning_rate": 8.372113240083774e-08, + "loss": 0.5371, + "step": 6652 + }, + { + "epoch": 5.507730535615682, + "grad_norm": 0.9698758721351624, + "learning_rate": 8.343916044301154e-08, + "loss": 0.5405, + "step": 6653 + }, + { + "epoch": 5.508558807288791, + "grad_norm": 0.9831439852714539, + "learning_rate": 8.315765606718123e-08, + "loss": 0.5413, + "step": 6654 + }, + { + "epoch": 5.5093870789618995, + "grad_norm": 0.9693325161933899, + "learning_rate": 8.287661932781521e-08, + "loss": 0.5582, + "step": 6655 + }, + { + "epoch": 5.510215350635009, + "grad_norm": 0.9818661212921143, + "learning_rate": 8.25960502792908e-08, + "loss": 0.5473, + "step": 6656 + }, + { + "epoch": 5.511043622308117, + "grad_norm": 0.9796103835105896, + "learning_rate": 8.231594897589595e-08, + "loss": 0.5744, + "step": 6657 + }, + { + "epoch": 5.511871893981226, + "grad_norm": 0.9636395573616028, + "learning_rate": 8.203631547182788e-08, + "loss": 0.5397, + "step": 6658 + }, + { + "epoch": 5.512700165654334, + "grad_norm": 0.9690009951591492, + "learning_rate": 8.175714982119248e-08, + "loss": 0.5292, + "step": 6659 + }, + { + "epoch": 5.513528437327444, + "grad_norm": 0.9523963928222656, + "learning_rate": 8.147845207800597e-08, + "loss": 0.5378, + "step": 6660 + }, + { + "epoch": 5.514356709000552, + "grad_norm": 0.967872142791748, + "learning_rate": 8.120022229619412e-08, + "loss": 0.5563, + "step": 6661 + }, + { + "epoch": 5.515184980673661, + "grad_norm": 0.9898565411567688, + "learning_rate": 8.092246052959163e-08, + "loss": 0.5494, + "step": 6662 + }, + { + "epoch": 5.51601325234677, + "grad_norm": 0.9669835567474365, + "learning_rate": 8.064516683194218e-08, + "loss": 0.5387, + "step": 6663 + }, + { + "epoch": 5.5168415240198785, + "grad_norm": 0.9909335970878601, + "learning_rate": 8.036834125690036e-08, + "loss": 0.5296, + "step": 6664 + }, + { + "epoch": 5.517669795692988, + "grad_norm": 0.9827402234077454, + "learning_rate": 8.009198385802886e-08, + "loss": 0.5441, + "step": 6665 + }, + { + "epoch": 5.518498067366096, + "grad_norm": 0.9958509206771851, + "learning_rate": 7.981609468880047e-08, + "loss": 0.5416, + "step": 6666 + }, + { + "epoch": 5.519326339039205, + "grad_norm": 0.9610104560852051, + "learning_rate": 7.954067380259667e-08, + "loss": 0.547, + "step": 6667 + }, + { + "epoch": 5.520154610712313, + "grad_norm": 0.9862914681434631, + "learning_rate": 7.926572125270982e-08, + "loss": 0.5551, + "step": 6668 + }, + { + "epoch": 5.520982882385423, + "grad_norm": 0.9723713994026184, + "learning_rate": 7.899123709233986e-08, + "loss": 0.523, + "step": 6669 + }, + { + "epoch": 5.521811154058531, + "grad_norm": 0.952841579914093, + "learning_rate": 7.871722137459737e-08, + "loss": 0.5527, + "step": 6670 + }, + { + "epoch": 5.52263942573164, + "grad_norm": 0.9491703510284424, + "learning_rate": 7.844367415250131e-08, + "loss": 0.5597, + "step": 6671 + }, + { + "epoch": 5.523467697404749, + "grad_norm": 0.9483358263969421, + "learning_rate": 7.817059547898131e-08, + "loss": 0.5394, + "step": 6672 + }, + { + "epoch": 5.5242959690778575, + "grad_norm": 0.9957411289215088, + "learning_rate": 7.789798540687454e-08, + "loss": 0.5501, + "step": 6673 + }, + { + "epoch": 5.525124240750967, + "grad_norm": 0.9639349579811096, + "learning_rate": 7.762584398892936e-08, + "loss": 0.5656, + "step": 6674 + }, + { + "epoch": 5.525952512424075, + "grad_norm": 0.9424917101860046, + "learning_rate": 7.735417127780226e-08, + "loss": 0.5418, + "step": 6675 + }, + { + "epoch": 5.526780784097184, + "grad_norm": 0.958835780620575, + "learning_rate": 7.708296732605925e-08, + "loss": 0.5474, + "step": 6676 + }, + { + "epoch": 5.527609055770292, + "grad_norm": 0.9803102612495422, + "learning_rate": 7.681223218617612e-08, + "loss": 0.5479, + "step": 6677 + }, + { + "epoch": 5.5284373274434015, + "grad_norm": 0.9641696214675903, + "learning_rate": 7.654196591053709e-08, + "loss": 0.5412, + "step": 6678 + }, + { + "epoch": 5.52926559911651, + "grad_norm": 0.9648889899253845, + "learning_rate": 7.627216855143671e-08, + "loss": 0.5443, + "step": 6679 + }, + { + "epoch": 5.530093870789619, + "grad_norm": 0.9735212326049805, + "learning_rate": 7.600284016107795e-08, + "loss": 0.5425, + "step": 6680 + }, + { + "epoch": 5.530922142462728, + "grad_norm": 0.9463320970535278, + "learning_rate": 7.57339807915733e-08, + "loss": 0.5308, + "step": 6681 + }, + { + "epoch": 5.5317504141358365, + "grad_norm": 0.9822217226028442, + "learning_rate": 7.546559049494473e-08, + "loss": 0.5709, + "step": 6682 + }, + { + "epoch": 5.532578685808946, + "grad_norm": 0.9626421332359314, + "learning_rate": 7.519766932312377e-08, + "loss": 0.5642, + "step": 6683 + }, + { + "epoch": 5.533406957482054, + "grad_norm": 0.973578155040741, + "learning_rate": 7.493021732794926e-08, + "loss": 0.5304, + "step": 6684 + }, + { + "epoch": 5.534235229155163, + "grad_norm": 0.9777551889419556, + "learning_rate": 7.466323456117169e-08, + "loss": 0.5549, + "step": 6685 + }, + { + "epoch": 5.535063500828271, + "grad_norm": 0.9757583737373352, + "learning_rate": 7.439672107444978e-08, + "loss": 0.5603, + "step": 6686 + }, + { + "epoch": 5.5358917725013805, + "grad_norm": 0.9676750898361206, + "learning_rate": 7.413067691935088e-08, + "loss": 0.5553, + "step": 6687 + }, + { + "epoch": 5.536720044174489, + "grad_norm": 0.9599202871322632, + "learning_rate": 7.386510214735215e-08, + "loss": 0.5603, + "step": 6688 + }, + { + "epoch": 5.537548315847598, + "grad_norm": 0.9806623458862305, + "learning_rate": 7.359999680984053e-08, + "loss": 0.5467, + "step": 6689 + }, + { + "epoch": 5.538376587520707, + "grad_norm": 0.9614994525909424, + "learning_rate": 7.333536095811056e-08, + "loss": 0.5753, + "step": 6690 + }, + { + "epoch": 5.5392048591938154, + "grad_norm": 0.977724015712738, + "learning_rate": 7.30711946433671e-08, + "loss": 0.5421, + "step": 6691 + }, + { + "epoch": 5.540033130866925, + "grad_norm": 0.9876384139060974, + "learning_rate": 7.280749791672425e-08, + "loss": 0.5585, + "step": 6692 + }, + { + "epoch": 5.540861402540033, + "grad_norm": 1.0018366575241089, + "learning_rate": 7.254427082920428e-08, + "loss": 0.5423, + "step": 6693 + }, + { + "epoch": 5.541689674213142, + "grad_norm": 0.976466715335846, + "learning_rate": 7.228151343173978e-08, + "loss": 0.5486, + "step": 6694 + }, + { + "epoch": 5.54251794588625, + "grad_norm": 0.983713686466217, + "learning_rate": 7.201922577517118e-08, + "loss": 0.5545, + "step": 6695 + }, + { + "epoch": 5.5433462175593595, + "grad_norm": 0.985798716545105, + "learning_rate": 7.175740791024927e-08, + "loss": 0.5511, + "step": 6696 + }, + { + "epoch": 5.544174489232468, + "grad_norm": 0.9781362414360046, + "learning_rate": 7.149605988763298e-08, + "loss": 0.5492, + "step": 6697 + }, + { + "epoch": 5.545002760905577, + "grad_norm": 0.9735119342803955, + "learning_rate": 7.123518175789101e-08, + "loss": 0.5298, + "step": 6698 + }, + { + "epoch": 5.545831032578686, + "grad_norm": 0.9664042592048645, + "learning_rate": 7.097477357150079e-08, + "loss": 0.5256, + "step": 6699 + }, + { + "epoch": 5.546659304251794, + "grad_norm": 0.9921199679374695, + "learning_rate": 7.071483537884838e-08, + "loss": 0.5403, + "step": 6700 + }, + { + "epoch": 5.547487575924904, + "grad_norm": 0.9512025713920593, + "learning_rate": 7.045536723022995e-08, + "loss": 0.5563, + "step": 6701 + }, + { + "epoch": 5.548315847598012, + "grad_norm": 0.9513158202171326, + "learning_rate": 7.019636917585032e-08, + "loss": 0.5608, + "step": 6702 + }, + { + "epoch": 5.549144119271121, + "grad_norm": 0.9949489831924438, + "learning_rate": 6.993784126582275e-08, + "loss": 0.5614, + "step": 6703 + }, + { + "epoch": 5.549972390944229, + "grad_norm": 0.9675762057304382, + "learning_rate": 6.967978355017002e-08, + "loss": 0.5533, + "step": 6704 + }, + { + "epoch": 5.5508006626173385, + "grad_norm": 0.9690775871276855, + "learning_rate": 6.942219607882439e-08, + "loss": 0.5592, + "step": 6705 + }, + { + "epoch": 5.551628934290447, + "grad_norm": 1.01649808883667, + "learning_rate": 6.916507890162627e-08, + "loss": 0.5435, + "step": 6706 + }, + { + "epoch": 5.552457205963556, + "grad_norm": 0.9586731195449829, + "learning_rate": 6.890843206832504e-08, + "loss": 0.5571, + "step": 6707 + }, + { + "epoch": 5.553285477636665, + "grad_norm": 0.9982333183288574, + "learning_rate": 6.865225562858013e-08, + "loss": 0.5551, + "step": 6708 + }, + { + "epoch": 5.554113749309773, + "grad_norm": 1.0094980001449585, + "learning_rate": 6.839654963195885e-08, + "loss": 0.5485, + "step": 6709 + }, + { + "epoch": 5.554942020982883, + "grad_norm": 0.9490573406219482, + "learning_rate": 6.8141314127938e-08, + "loss": 0.5494, + "step": 6710 + }, + { + "epoch": 5.555770292655991, + "grad_norm": 0.9914820194244385, + "learning_rate": 6.788654916590365e-08, + "loss": 0.5516, + "step": 6711 + }, + { + "epoch": 5.5565985643291, + "grad_norm": 0.9605741500854492, + "learning_rate": 6.763225479514996e-08, + "loss": 0.5489, + "step": 6712 + }, + { + "epoch": 5.557426836002209, + "grad_norm": 0.9726978540420532, + "learning_rate": 6.737843106488062e-08, + "loss": 0.5457, + "step": 6713 + }, + { + "epoch": 5.5582551076753175, + "grad_norm": 0.9850461483001709, + "learning_rate": 6.712507802420859e-08, + "loss": 0.5471, + "step": 6714 + }, + { + "epoch": 5.559083379348427, + "grad_norm": 0.9779561758041382, + "learning_rate": 6.687219572215492e-08, + "loss": 0.5162, + "step": 6715 + }, + { + "epoch": 5.559911651021535, + "grad_norm": 0.9608659148216248, + "learning_rate": 6.661978420764964e-08, + "loss": 0.543, + "step": 6716 + }, + { + "epoch": 5.560739922694644, + "grad_norm": 0.9841420650482178, + "learning_rate": 6.636784352953313e-08, + "loss": 0.5492, + "step": 6717 + }, + { + "epoch": 5.561568194367752, + "grad_norm": 0.9621257781982422, + "learning_rate": 6.611637373655249e-08, + "loss": 0.5661, + "step": 6718 + }, + { + "epoch": 5.562396466040862, + "grad_norm": 0.9492973685264587, + "learning_rate": 6.586537487736522e-08, + "loss": 0.5214, + "step": 6719 + }, + { + "epoch": 5.56322473771397, + "grad_norm": 0.9798749089241028, + "learning_rate": 6.561484700053688e-08, + "loss": 0.5536, + "step": 6720 + }, + { + "epoch": 5.564053009387079, + "grad_norm": 0.9854906797409058, + "learning_rate": 6.536479015454289e-08, + "loss": 0.5576, + "step": 6721 + }, + { + "epoch": 5.564881281060188, + "grad_norm": 0.9609218239784241, + "learning_rate": 6.511520438776676e-08, + "loss": 0.5521, + "step": 6722 + }, + { + "epoch": 5.5657095527332965, + "grad_norm": 0.9567620754241943, + "learning_rate": 6.48660897485004e-08, + "loss": 0.5735, + "step": 6723 + }, + { + "epoch": 5.566537824406406, + "grad_norm": 0.982399582862854, + "learning_rate": 6.46174462849461e-08, + "loss": 0.5511, + "step": 6724 + }, + { + "epoch": 5.567366096079514, + "grad_norm": 1.0123873949050903, + "learning_rate": 6.436927404521343e-08, + "loss": 0.5293, + "step": 6725 + }, + { + "epoch": 5.568194367752623, + "grad_norm": 0.9643762707710266, + "learning_rate": 6.412157307732148e-08, + "loss": 0.5413, + "step": 6726 + }, + { + "epoch": 5.569022639425731, + "grad_norm": 0.998612642288208, + "learning_rate": 6.38743434291983e-08, + "loss": 0.5323, + "step": 6727 + }, + { + "epoch": 5.569850911098841, + "grad_norm": 0.9602347612380981, + "learning_rate": 6.362758514868062e-08, + "loss": 0.5663, + "step": 6728 + }, + { + "epoch": 5.570679182771949, + "grad_norm": 0.9574209451675415, + "learning_rate": 6.338129828351331e-08, + "loss": 0.5357, + "step": 6729 + }, + { + "epoch": 5.571507454445058, + "grad_norm": 0.9795578122138977, + "learning_rate": 6.313548288135102e-08, + "loss": 0.5426, + "step": 6730 + }, + { + "epoch": 5.572335726118167, + "grad_norm": 0.9603869915008545, + "learning_rate": 6.289013898975654e-08, + "loss": 0.532, + "step": 6731 + }, + { + "epoch": 5.5731639977912755, + "grad_norm": 0.9846640825271606, + "learning_rate": 6.264526665620163e-08, + "loss": 0.5503, + "step": 6732 + }, + { + "epoch": 5.573992269464385, + "grad_norm": 0.9611700773239136, + "learning_rate": 6.240086592806727e-08, + "loss": 0.5506, + "step": 6733 + }, + { + "epoch": 5.574820541137493, + "grad_norm": 0.9696215391159058, + "learning_rate": 6.215693685264201e-08, + "loss": 0.5537, + "step": 6734 + }, + { + "epoch": 5.575648812810602, + "grad_norm": 0.9591673612594604, + "learning_rate": 6.191347947712423e-08, + "loss": 0.5537, + "step": 6735 + }, + { + "epoch": 5.57647708448371, + "grad_norm": 0.9840232133865356, + "learning_rate": 6.167049384862067e-08, + "loss": 0.5495, + "step": 6736 + }, + { + "epoch": 5.5773053561568195, + "grad_norm": 0.9635297656059265, + "learning_rate": 6.142798001414679e-08, + "loss": 0.5584, + "step": 6737 + }, + { + "epoch": 5.578133627829928, + "grad_norm": 0.9583635330200195, + "learning_rate": 6.118593802062645e-08, + "loss": 0.5411, + "step": 6738 + }, + { + "epoch": 5.578961899503037, + "grad_norm": 0.9491806030273438, + "learning_rate": 6.094436791489327e-08, + "loss": 0.5589, + "step": 6739 + }, + { + "epoch": 5.579790171176146, + "grad_norm": 0.9747976064682007, + "learning_rate": 6.070326974368796e-08, + "loss": 0.5343, + "step": 6740 + }, + { + "epoch": 5.5806184428492545, + "grad_norm": 0.9590021967887878, + "learning_rate": 6.046264355366121e-08, + "loss": 0.5443, + "step": 6741 + }, + { + "epoch": 5.581446714522364, + "grad_norm": 0.9807665348052979, + "learning_rate": 6.022248939137165e-08, + "loss": 0.5516, + "step": 6742 + }, + { + "epoch": 5.582274986195472, + "grad_norm": 0.9926226735115051, + "learning_rate": 5.998280730328738e-08, + "loss": 0.5494, + "step": 6743 + }, + { + "epoch": 5.583103257868581, + "grad_norm": 0.9682310819625854, + "learning_rate": 5.974359733578406e-08, + "loss": 0.5348, + "step": 6744 + }, + { + "epoch": 5.583931529541689, + "grad_norm": 0.9812586903572083, + "learning_rate": 5.950485953514718e-08, + "loss": 0.5607, + "step": 6745 + }, + { + "epoch": 5.5847598012147985, + "grad_norm": 0.9767264723777771, + "learning_rate": 5.9266593947569786e-08, + "loss": 0.5368, + "step": 6746 + }, + { + "epoch": 5.585588072887907, + "grad_norm": 0.9910010695457458, + "learning_rate": 5.902880061915417e-08, + "loss": 0.5387, + "step": 6747 + }, + { + "epoch": 5.586416344561016, + "grad_norm": 0.9638100862503052, + "learning_rate": 5.8791479595911307e-08, + "loss": 0.559, + "step": 6748 + }, + { + "epoch": 5.587244616234125, + "grad_norm": 0.9895380735397339, + "learning_rate": 5.855463092376057e-08, + "loss": 0.5547, + "step": 6749 + }, + { + "epoch": 5.588072887907233, + "grad_norm": 0.9643979072570801, + "learning_rate": 5.8318254648529474e-08, + "loss": 0.5559, + "step": 6750 + }, + { + "epoch": 5.588901159580343, + "grad_norm": 0.9791732430458069, + "learning_rate": 5.808235081595559e-08, + "loss": 0.546, + "step": 6751 + }, + { + "epoch": 5.589729431253451, + "grad_norm": 0.9702895283699036, + "learning_rate": 5.7846919471683524e-08, + "loss": 0.548, + "step": 6752 + }, + { + "epoch": 5.59055770292656, + "grad_norm": 0.9695649147033691, + "learning_rate": 5.7611960661266565e-08, + "loss": 0.5422, + "step": 6753 + }, + { + "epoch": 5.591385974599668, + "grad_norm": 1.054394006729126, + "learning_rate": 5.737747443016778e-08, + "loss": 0.5618, + "step": 6754 + }, + { + "epoch": 5.5922142462727775, + "grad_norm": 0.9772742390632629, + "learning_rate": 5.71434608237581e-08, + "loss": 0.5541, + "step": 6755 + }, + { + "epoch": 5.593042517945886, + "grad_norm": 0.972944974899292, + "learning_rate": 5.690991988731659e-08, + "loss": 0.5494, + "step": 6756 + }, + { + "epoch": 5.593870789618995, + "grad_norm": 0.9849849343299866, + "learning_rate": 5.6676851666031265e-08, + "loss": 0.5526, + "step": 6757 + }, + { + "epoch": 5.594699061292104, + "grad_norm": 0.9597331285476685, + "learning_rate": 5.644425620499883e-08, + "loss": 0.5384, + "step": 6758 + }, + { + "epoch": 5.595527332965212, + "grad_norm": 0.9747866988182068, + "learning_rate": 5.621213354922439e-08, + "loss": 0.5418, + "step": 6759 + }, + { + "epoch": 5.596355604638322, + "grad_norm": 0.9931576251983643, + "learning_rate": 5.59804837436212e-08, + "loss": 0.5479, + "step": 6760 + }, + { + "epoch": 5.59718387631143, + "grad_norm": 0.9567369222640991, + "learning_rate": 5.5749306833011716e-08, + "loss": 0.5516, + "step": 6761 + }, + { + "epoch": 5.598012147984539, + "grad_norm": 0.9719016551971436, + "learning_rate": 5.5518602862126566e-08, + "loss": 0.5598, + "step": 6762 + }, + { + "epoch": 5.598840419657647, + "grad_norm": 0.9701929092407227, + "learning_rate": 5.528837187560421e-08, + "loss": 0.5482, + "step": 6763 + }, + { + "epoch": 5.5996686913307565, + "grad_norm": 0.9580748081207275, + "learning_rate": 5.505861391799289e-08, + "loss": 0.5246, + "step": 6764 + }, + { + "epoch": 5.600496963003865, + "grad_norm": 0.9584888219833374, + "learning_rate": 5.482932903374816e-08, + "loss": 0.546, + "step": 6765 + }, + { + "epoch": 5.601325234676974, + "grad_norm": 0.9721468687057495, + "learning_rate": 5.460051726723453e-08, + "loss": 0.5733, + "step": 6766 + }, + { + "epoch": 5.602153506350083, + "grad_norm": 0.9819844961166382, + "learning_rate": 5.437217866272548e-08, + "loss": 0.5508, + "step": 6767 + }, + { + "epoch": 5.602981778023191, + "grad_norm": 0.9763482809066772, + "learning_rate": 5.41443132644015e-08, + "loss": 0.5507, + "step": 6768 + }, + { + "epoch": 5.603810049696301, + "grad_norm": 1.024173378944397, + "learning_rate": 5.391692111635316e-08, + "loss": 0.5255, + "step": 6769 + }, + { + "epoch": 5.604638321369409, + "grad_norm": 0.975405752658844, + "learning_rate": 5.3690002262578586e-08, + "loss": 0.5416, + "step": 6770 + }, + { + "epoch": 5.605466593042518, + "grad_norm": 0.9836496114730835, + "learning_rate": 5.346355674698433e-08, + "loss": 0.5318, + "step": 6771 + }, + { + "epoch": 5.606294864715627, + "grad_norm": 0.9690279960632324, + "learning_rate": 5.323758461338535e-08, + "loss": 0.541, + "step": 6772 + }, + { + "epoch": 5.6071231363887355, + "grad_norm": 0.9432088732719421, + "learning_rate": 5.301208590550555e-08, + "loss": 0.5395, + "step": 6773 + }, + { + "epoch": 5.607951408061845, + "grad_norm": 0.9760525226593018, + "learning_rate": 5.278706066697642e-08, + "loss": 0.5264, + "step": 6774 + }, + { + "epoch": 5.608779679734953, + "grad_norm": 0.9518134593963623, + "learning_rate": 5.256250894133841e-08, + "loss": 0.5435, + "step": 6775 + }, + { + "epoch": 5.609607951408062, + "grad_norm": 0.9828734993934631, + "learning_rate": 5.233843077204037e-08, + "loss": 0.5389, + "step": 6776 + }, + { + "epoch": 5.61043622308117, + "grad_norm": 0.9697222709655762, + "learning_rate": 5.211482620243902e-08, + "loss": 0.5336, + "step": 6777 + }, + { + "epoch": 5.61126449475428, + "grad_norm": 0.9675080180168152, + "learning_rate": 5.1891695275800013e-08, + "loss": 0.5455, + "step": 6778 + }, + { + "epoch": 5.612092766427388, + "grad_norm": 0.9964091181755066, + "learning_rate": 5.166903803529688e-08, + "loss": 0.5612, + "step": 6779 + }, + { + "epoch": 5.612921038100497, + "grad_norm": 0.9672561287879944, + "learning_rate": 5.14468545240121e-08, + "loss": 0.5415, + "step": 6780 + }, + { + "epoch": 5.613749309773606, + "grad_norm": 0.9824404716491699, + "learning_rate": 5.1225144784935456e-08, + "loss": 0.5373, + "step": 6781 + }, + { + "epoch": 5.6145775814467145, + "grad_norm": 0.9557321667671204, + "learning_rate": 5.100390886096623e-08, + "loss": 0.5585, + "step": 6782 + }, + { + "epoch": 5.615405853119824, + "grad_norm": 0.9670886993408203, + "learning_rate": 5.078314679491159e-08, + "loss": 0.5509, + "step": 6783 + }, + { + "epoch": 5.616234124792932, + "grad_norm": 0.9520448446273804, + "learning_rate": 5.056285862948651e-08, + "loss": 0.539, + "step": 6784 + }, + { + "epoch": 5.617062396466041, + "grad_norm": 1.0446722507476807, + "learning_rate": 5.034304440731497e-08, + "loss": 0.5394, + "step": 6785 + }, + { + "epoch": 5.617890668139149, + "grad_norm": 0.9921750426292419, + "learning_rate": 5.012370417092932e-08, + "loss": 0.543, + "step": 6786 + }, + { + "epoch": 5.618718939812259, + "grad_norm": 0.9812518954277039, + "learning_rate": 4.990483796276896e-08, + "loss": 0.5525, + "step": 6787 + }, + { + "epoch": 5.619547211485367, + "grad_norm": 0.9780861139297485, + "learning_rate": 4.968644582518306e-08, + "loss": 0.5587, + "step": 6788 + }, + { + "epoch": 5.620375483158476, + "grad_norm": 1.1603604555130005, + "learning_rate": 4.9468527800428376e-08, + "loss": 0.5481, + "step": 6789 + }, + { + "epoch": 5.621203754831585, + "grad_norm": 0.9589557647705078, + "learning_rate": 4.925108393067007e-08, + "loss": 0.5376, + "step": 6790 + }, + { + "epoch": 5.6220320265046935, + "grad_norm": 0.9825956225395203, + "learning_rate": 4.903411425798116e-08, + "loss": 0.5446, + "step": 6791 + }, + { + "epoch": 5.622860298177803, + "grad_norm": 0.9554243087768555, + "learning_rate": 4.8817618824343884e-08, + "loss": 0.549, + "step": 6792 + }, + { + "epoch": 5.623688569850911, + "grad_norm": 0.9802545309066772, + "learning_rate": 4.860159767164752e-08, + "loss": 0.5596, + "step": 6793 + }, + { + "epoch": 5.62451684152402, + "grad_norm": 0.9786462783813477, + "learning_rate": 4.838605084169057e-08, + "loss": 0.5384, + "step": 6794 + }, + { + "epoch": 5.625345113197128, + "grad_norm": 0.9578393697738647, + "learning_rate": 4.8170978376179125e-08, + "loss": 0.5439, + "step": 6795 + }, + { + "epoch": 5.6261733848702375, + "grad_norm": 0.9585602879524231, + "learning_rate": 4.7956380316727666e-08, + "loss": 0.5611, + "step": 6796 + }, + { + "epoch": 5.627001656543346, + "grad_norm": 0.9862198829650879, + "learning_rate": 4.7742256704858825e-08, + "loss": 0.58, + "step": 6797 + }, + { + "epoch": 5.627829928216455, + "grad_norm": 0.9744497537612915, + "learning_rate": 4.7528607582003616e-08, + "loss": 0.5385, + "step": 6798 + }, + { + "epoch": 5.628658199889564, + "grad_norm": 0.9749640226364136, + "learning_rate": 4.7315432989501476e-08, + "loss": 0.5301, + "step": 6799 + }, + { + "epoch": 5.6294864715626725, + "grad_norm": 0.9528510570526123, + "learning_rate": 4.710273296859913e-08, + "loss": 0.561, + "step": 6800 + }, + { + "epoch": 5.630314743235782, + "grad_norm": 0.9534420371055603, + "learning_rate": 4.689050756045227e-08, + "loss": 0.5244, + "step": 6801 + }, + { + "epoch": 5.63114301490889, + "grad_norm": 0.94547438621521, + "learning_rate": 4.6678756806124994e-08, + "loss": 0.5582, + "step": 6802 + }, + { + "epoch": 5.631971286581999, + "grad_norm": 0.9832236766815186, + "learning_rate": 4.646748074658814e-08, + "loss": 0.525, + "step": 6803 + }, + { + "epoch": 5.632799558255107, + "grad_norm": 0.9723865389823914, + "learning_rate": 4.625667942272261e-08, + "loss": 0.5401, + "step": 6804 + }, + { + "epoch": 5.6336278299282165, + "grad_norm": 0.9863663911819458, + "learning_rate": 4.604635287531606e-08, + "loss": 0.573, + "step": 6805 + }, + { + "epoch": 5.634456101601325, + "grad_norm": 1.0028870105743408, + "learning_rate": 4.5836501145064826e-08, + "loss": 0.5348, + "step": 6806 + }, + { + "epoch": 5.635284373274434, + "grad_norm": 1.0082811117172241, + "learning_rate": 4.56271242725731e-08, + "loss": 0.5507, + "step": 6807 + }, + { + "epoch": 5.636112644947543, + "grad_norm": 0.9527307152748108, + "learning_rate": 4.5418222298353745e-08, + "loss": 0.5456, + "step": 6808 + }, + { + "epoch": 5.636940916620651, + "grad_norm": 0.9916908144950867, + "learning_rate": 4.520979526282665e-08, + "loss": 0.5275, + "step": 6809 + }, + { + "epoch": 5.637769188293761, + "grad_norm": 0.9775404334068298, + "learning_rate": 4.500184320632095e-08, + "loss": 0.5357, + "step": 6810 + }, + { + "epoch": 5.638597459966869, + "grad_norm": 0.9828300476074219, + "learning_rate": 4.479436616907362e-08, + "loss": 0.5407, + "step": 6811 + }, + { + "epoch": 5.639425731639978, + "grad_norm": 0.9787706732749939, + "learning_rate": 4.458736419122922e-08, + "loss": 0.5576, + "step": 6812 + }, + { + "epoch": 5.640254003313086, + "grad_norm": 0.9941301345825195, + "learning_rate": 4.438083731284071e-08, + "loss": 0.56, + "step": 6813 + }, + { + "epoch": 5.6410822749861955, + "grad_norm": 0.9876093864440918, + "learning_rate": 4.4174785573869174e-08, + "loss": 0.5672, + "step": 6814 + }, + { + "epoch": 5.641910546659304, + "grad_norm": 0.9657239317893982, + "learning_rate": 4.3969209014183837e-08, + "loss": 0.5469, + "step": 6815 + }, + { + "epoch": 5.642738818332413, + "grad_norm": 0.9718003869056702, + "learning_rate": 4.3764107673561485e-08, + "loss": 0.5762, + "step": 6816 + }, + { + "epoch": 5.643567090005522, + "grad_norm": 0.9725603461265564, + "learning_rate": 4.355948159168788e-08, + "loss": 0.5315, + "step": 6817 + }, + { + "epoch": 5.64439536167863, + "grad_norm": 0.9833802580833435, + "learning_rate": 4.3355330808155795e-08, + "loss": 0.5221, + "step": 6818 + }, + { + "epoch": 5.64522363335174, + "grad_norm": 0.97885662317276, + "learning_rate": 4.315165536246668e-08, + "loss": 0.5792, + "step": 6819 + }, + { + "epoch": 5.646051905024848, + "grad_norm": 0.9684316515922546, + "learning_rate": 4.294845529402986e-08, + "loss": 0.5465, + "step": 6820 + }, + { + "epoch": 5.646880176697957, + "grad_norm": 0.9782618880271912, + "learning_rate": 4.2745730642162476e-08, + "loss": 0.5149, + "step": 6821 + }, + { + "epoch": 5.647708448371065, + "grad_norm": 0.971509575843811, + "learning_rate": 4.254348144608983e-08, + "loss": 0.5299, + "step": 6822 + }, + { + "epoch": 5.6485367200441745, + "grad_norm": 0.9667693376541138, + "learning_rate": 4.2341707744945895e-08, + "loss": 0.5566, + "step": 6823 + }, + { + "epoch": 5.649364991717283, + "grad_norm": 0.9685983657836914, + "learning_rate": 4.21404095777711e-08, + "loss": 0.5441, + "step": 6824 + }, + { + "epoch": 5.650193263390392, + "grad_norm": 0.9624237418174744, + "learning_rate": 4.193958698351513e-08, + "loss": 0.5508, + "step": 6825 + }, + { + "epoch": 5.651021535063501, + "grad_norm": 0.9783890843391418, + "learning_rate": 4.173924000103552e-08, + "loss": 0.5235, + "step": 6826 + }, + { + "epoch": 5.651849806736609, + "grad_norm": 0.9971002340316772, + "learning_rate": 4.1539368669097355e-08, + "loss": 0.5415, + "step": 6827 + }, + { + "epoch": 5.652678078409719, + "grad_norm": 0.9450921416282654, + "learning_rate": 4.133997302637388e-08, + "loss": 0.5524, + "step": 6828 + }, + { + "epoch": 5.653506350082827, + "grad_norm": 0.9719640016555786, + "learning_rate": 4.1141053111446174e-08, + "loss": 0.5342, + "step": 6829 + }, + { + "epoch": 5.654334621755936, + "grad_norm": 0.9782803654670715, + "learning_rate": 4.094260896280372e-08, + "loss": 0.5652, + "step": 6830 + }, + { + "epoch": 5.655162893429045, + "grad_norm": 0.9767885208129883, + "learning_rate": 4.074464061884331e-08, + "loss": 0.5336, + "step": 6831 + }, + { + "epoch": 5.6559911651021535, + "grad_norm": 0.9558879137039185, + "learning_rate": 4.054714811786986e-08, + "loss": 0.5296, + "step": 6832 + }, + { + "epoch": 5.656819436775263, + "grad_norm": 0.943391740322113, + "learning_rate": 4.035013149809669e-08, + "loss": 0.5456, + "step": 6833 + }, + { + "epoch": 5.657647708448371, + "grad_norm": 0.9620651006698608, + "learning_rate": 4.015359079764441e-08, + "loss": 0.5292, + "step": 6834 + }, + { + "epoch": 5.65847598012148, + "grad_norm": 0.9600332379341125, + "learning_rate": 3.995752605454178e-08, + "loss": 0.5365, + "step": 6835 + }, + { + "epoch": 5.659304251794588, + "grad_norm": 0.9740662574768066, + "learning_rate": 3.976193730672595e-08, + "loss": 0.5359, + "step": 6836 + }, + { + "epoch": 5.660132523467698, + "grad_norm": 0.9533049464225769, + "learning_rate": 3.9566824592041086e-08, + "loss": 0.5269, + "step": 6837 + }, + { + "epoch": 5.660960795140806, + "grad_norm": 0.9276461601257324, + "learning_rate": 3.937218794823977e-08, + "loss": 0.512, + "step": 6838 + }, + { + "epoch": 5.661789066813915, + "grad_norm": 0.9811263680458069, + "learning_rate": 3.9178027412982435e-08, + "loss": 0.5553, + "step": 6839 + }, + { + "epoch": 5.662617338487024, + "grad_norm": 0.9731243252754211, + "learning_rate": 3.8984343023837354e-08, + "loss": 0.5612, + "step": 6840 + }, + { + "epoch": 5.6634456101601325, + "grad_norm": 0.9473799467086792, + "learning_rate": 3.8791134818280675e-08, + "loss": 0.5367, + "step": 6841 + }, + { + "epoch": 5.664273881833242, + "grad_norm": 0.9842115044593811, + "learning_rate": 3.859840283369609e-08, + "loss": 0.5595, + "step": 6842 + }, + { + "epoch": 5.66510215350635, + "grad_norm": 0.9575071930885315, + "learning_rate": 3.8406147107375724e-08, + "loss": 0.5588, + "step": 6843 + }, + { + "epoch": 5.665930425179459, + "grad_norm": 0.9698435664176941, + "learning_rate": 3.821436767651926e-08, + "loss": 0.5503, + "step": 6844 + }, + { + "epoch": 5.666758696852567, + "grad_norm": 0.956071138381958, + "learning_rate": 3.802306457823424e-08, + "loss": 0.5304, + "step": 6845 + }, + { + "epoch": 5.667586968525677, + "grad_norm": 1.0221352577209473, + "learning_rate": 3.783223784953577e-08, + "loss": 0.5417, + "step": 6846 + }, + { + "epoch": 5.668415240198785, + "grad_norm": 1.0275334119796753, + "learning_rate": 3.7641887527347375e-08, + "loss": 0.565, + "step": 6847 + }, + { + "epoch": 5.669243511871894, + "grad_norm": 0.9793328642845154, + "learning_rate": 3.745201364849988e-08, + "loss": 0.5285, + "step": 6848 + }, + { + "epoch": 5.670071783545003, + "grad_norm": 0.9487180709838867, + "learning_rate": 3.726261624973221e-08, + "loss": 0.5432, + "step": 6849 + }, + { + "epoch": 5.6709000552181115, + "grad_norm": 0.988447368144989, + "learning_rate": 3.70736953676909e-08, + "loss": 0.5502, + "step": 6850 + }, + { + "epoch": 5.671728326891221, + "grad_norm": 0.9646620750427246, + "learning_rate": 3.6885251038930593e-08, + "loss": 0.5436, + "step": 6851 + }, + { + "epoch": 5.672556598564329, + "grad_norm": 0.9508702754974365, + "learning_rate": 3.669728329991323e-08, + "loss": 0.5523, + "step": 6852 + }, + { + "epoch": 5.673384870237438, + "grad_norm": 0.9730278253555298, + "learning_rate": 3.6509792187008894e-08, + "loss": 0.5587, + "step": 6853 + }, + { + "epoch": 5.674213141910546, + "grad_norm": 0.9707013964653015, + "learning_rate": 3.6322777736495504e-08, + "loss": 0.5634, + "step": 6854 + }, + { + "epoch": 5.6750414135836555, + "grad_norm": 0.9702714681625366, + "learning_rate": 3.613623998455856e-08, + "loss": 0.5356, + "step": 6855 + }, + { + "epoch": 5.675869685256764, + "grad_norm": 0.9663801193237305, + "learning_rate": 3.5950178967291136e-08, + "loss": 0.5381, + "step": 6856 + }, + { + "epoch": 5.676697956929873, + "grad_norm": 0.9929706454277039, + "learning_rate": 3.5764594720694435e-08, + "loss": 0.5389, + "step": 6857 + }, + { + "epoch": 5.677526228602982, + "grad_norm": 0.974050760269165, + "learning_rate": 3.5579487280677506e-08, + "loss": 0.5489, + "step": 6858 + }, + { + "epoch": 5.6783545002760905, + "grad_norm": 0.9689812660217285, + "learning_rate": 3.53948566830567e-08, + "loss": 0.5423, + "step": 6859 + }, + { + "epoch": 5.6791827719492, + "grad_norm": 0.982503354549408, + "learning_rate": 3.52107029635565e-08, + "loss": 0.528, + "step": 6860 + }, + { + "epoch": 5.680011043622308, + "grad_norm": 0.9769023060798645, + "learning_rate": 3.502702615780867e-08, + "loss": 0.5493, + "step": 6861 + }, + { + "epoch": 5.680839315295417, + "grad_norm": 0.9898412227630615, + "learning_rate": 3.4843826301353135e-08, + "loss": 0.565, + "step": 6862 + }, + { + "epoch": 5.681667586968525, + "grad_norm": 0.9600477814674377, + "learning_rate": 3.466110342963708e-08, + "loss": 0.5395, + "step": 6863 + }, + { + "epoch": 5.6824958586416345, + "grad_norm": 0.9765893220901489, + "learning_rate": 3.4478857578016664e-08, + "loss": 0.5409, + "step": 6864 + }, + { + "epoch": 5.683324130314743, + "grad_norm": 0.9853554964065552, + "learning_rate": 3.4297088781753416e-08, + "loss": 0.529, + "step": 6865 + }, + { + "epoch": 5.684152401987852, + "grad_norm": 0.9655324220657349, + "learning_rate": 3.4115797076018366e-08, + "loss": 0.5498, + "step": 6866 + }, + { + "epoch": 5.684980673660961, + "grad_norm": 0.9601819515228271, + "learning_rate": 3.393498249589039e-08, + "loss": 0.5338, + "step": 6867 + }, + { + "epoch": 5.685808945334069, + "grad_norm": 0.97408527135849, + "learning_rate": 3.3754645076354564e-08, + "loss": 0.5373, + "step": 6868 + }, + { + "epoch": 5.686637217007179, + "grad_norm": 0.9690378904342651, + "learning_rate": 3.357478485230492e-08, + "loss": 0.5524, + "step": 6869 + }, + { + "epoch": 5.687465488680287, + "grad_norm": 0.9984313249588013, + "learning_rate": 3.339540185854279e-08, + "loss": 0.5401, + "step": 6870 + }, + { + "epoch": 5.688293760353396, + "grad_norm": 0.9710806608200073, + "learning_rate": 3.321649612977679e-08, + "loss": 0.563, + "step": 6871 + }, + { + "epoch": 5.689122032026504, + "grad_norm": 0.975621223449707, + "learning_rate": 3.303806770062396e-08, + "loss": 0.5492, + "step": 6872 + }, + { + "epoch": 5.6899503036996135, + "grad_norm": 0.9607579708099365, + "learning_rate": 3.2860116605608336e-08, + "loss": 0.5114, + "step": 6873 + }, + { + "epoch": 5.690778575372722, + "grad_norm": 0.9561493396759033, + "learning_rate": 3.268264287916156e-08, + "loss": 0.5814, + "step": 6874 + }, + { + "epoch": 5.691606847045831, + "grad_norm": 0.9815540313720703, + "learning_rate": 3.2505646555623385e-08, + "loss": 0.5288, + "step": 6875 + }, + { + "epoch": 5.69243511871894, + "grad_norm": 1.1233237981796265, + "learning_rate": 3.2329127669241124e-08, + "loss": 0.5671, + "step": 6876 + }, + { + "epoch": 5.693263390392048, + "grad_norm": 0.9686160683631897, + "learning_rate": 3.215308625416913e-08, + "loss": 0.5451, + "step": 6877 + }, + { + "epoch": 5.694091662065158, + "grad_norm": 0.9839127063751221, + "learning_rate": 3.197752234446988e-08, + "loss": 0.5757, + "step": 6878 + }, + { + "epoch": 5.694919933738266, + "grad_norm": 0.9613943099975586, + "learning_rate": 3.1802435974113696e-08, + "loss": 0.5287, + "step": 6879 + }, + { + "epoch": 5.695748205411375, + "grad_norm": 0.9591472148895264, + "learning_rate": 3.162782717697793e-08, + "loss": 0.5677, + "step": 6880 + }, + { + "epoch": 5.696576477084483, + "grad_norm": 0.95570969581604, + "learning_rate": 3.14536959868475e-08, + "loss": 0.5402, + "step": 6881 + }, + { + "epoch": 5.6974047487575925, + "grad_norm": 0.979266881942749, + "learning_rate": 3.128004243741545e-08, + "loss": 0.5558, + "step": 6882 + }, + { + "epoch": 5.698233020430701, + "grad_norm": 0.9818242192268372, + "learning_rate": 3.110686656228212e-08, + "loss": 0.5359, + "step": 6883 + }, + { + "epoch": 5.69906129210381, + "grad_norm": 0.9787725210189819, + "learning_rate": 3.093416839495517e-08, + "loss": 0.5327, + "step": 6884 + }, + { + "epoch": 5.699889563776919, + "grad_norm": 0.9504302740097046, + "learning_rate": 3.0761947968850626e-08, + "loss": 0.5295, + "step": 6885 + }, + { + "epoch": 5.700717835450027, + "grad_norm": 0.97245854139328, + "learning_rate": 3.059020531729101e-08, + "loss": 0.5539, + "step": 6886 + }, + { + "epoch": 5.701546107123137, + "grad_norm": 0.992629885673523, + "learning_rate": 3.041894047350752e-08, + "loss": 0.5622, + "step": 6887 + }, + { + "epoch": 5.702374378796245, + "grad_norm": 0.9556792378425598, + "learning_rate": 3.0248153470637254e-08, + "loss": 0.5433, + "step": 6888 + }, + { + "epoch": 5.703202650469354, + "grad_norm": 0.9557904601097107, + "learning_rate": 3.0077844341727114e-08, + "loss": 0.55, + "step": 6889 + }, + { + "epoch": 5.704030922142463, + "grad_norm": 0.9659007787704468, + "learning_rate": 2.990801311972935e-08, + "loss": 0.5345, + "step": 6890 + }, + { + "epoch": 5.7048591938155715, + "grad_norm": 0.995897650718689, + "learning_rate": 2.9738659837505457e-08, + "loss": 0.5819, + "step": 6891 + }, + { + "epoch": 5.705687465488681, + "grad_norm": 0.9943006634712219, + "learning_rate": 2.9569784527823386e-08, + "loss": 0.5332, + "step": 6892 + }, + { + "epoch": 5.706515737161789, + "grad_norm": 0.9578213095664978, + "learning_rate": 2.9401387223358668e-08, + "loss": 0.5356, + "step": 6893 + }, + { + "epoch": 5.707344008834898, + "grad_norm": 0.9744815826416016, + "learning_rate": 2.9233467956694962e-08, + "loss": 0.5406, + "step": 6894 + }, + { + "epoch": 5.708172280508006, + "grad_norm": 1.1124356985092163, + "learning_rate": 2.906602676032322e-08, + "loss": 0.563, + "step": 6895 + }, + { + "epoch": 5.709000552181116, + "grad_norm": 1.042433500289917, + "learning_rate": 2.8899063666641412e-08, + "loss": 0.521, + "step": 6896 + }, + { + "epoch": 5.709828823854224, + "grad_norm": 0.9627099633216858, + "learning_rate": 2.873257870795565e-08, + "loss": 0.5238, + "step": 6897 + }, + { + "epoch": 5.710657095527333, + "grad_norm": 0.9504469037055969, + "learning_rate": 2.8566571916479047e-08, + "loss": 0.5472, + "step": 6898 + }, + { + "epoch": 5.711485367200442, + "grad_norm": 1.2675774097442627, + "learning_rate": 2.84010433243323e-08, + "loss": 0.5551, + "step": 6899 + }, + { + "epoch": 5.7123136388735505, + "grad_norm": 0.9804739356040955, + "learning_rate": 2.82359929635434e-08, + "loss": 0.552, + "step": 6900 + }, + { + "epoch": 5.71314191054666, + "grad_norm": 0.9695979952812195, + "learning_rate": 2.807142086604875e-08, + "loss": 0.5406, + "step": 6901 + }, + { + "epoch": 5.713970182219768, + "grad_norm": 0.98618084192276, + "learning_rate": 2.7907327063690927e-08, + "loss": 0.5353, + "step": 6902 + }, + { + "epoch": 5.714798453892877, + "grad_norm": 0.9667906165122986, + "learning_rate": 2.7743711588220922e-08, + "loss": 0.5552, + "step": 6903 + }, + { + "epoch": 5.715626725565985, + "grad_norm": 0.9631903171539307, + "learning_rate": 2.7580574471296474e-08, + "loss": 0.5429, + "step": 6904 + }, + { + "epoch": 5.716454997239095, + "grad_norm": 0.9845322370529175, + "learning_rate": 2.7417915744483436e-08, + "loss": 0.5389, + "step": 6905 + }, + { + "epoch": 5.717283268912203, + "grad_norm": 0.9690833687782288, + "learning_rate": 2.7255735439254416e-08, + "loss": 0.5577, + "step": 6906 + }, + { + "epoch": 5.718111540585312, + "grad_norm": 0.9540392756462097, + "learning_rate": 2.709403358699014e-08, + "loss": 0.5328, + "step": 6907 + }, + { + "epoch": 5.718939812258421, + "grad_norm": 0.9816165566444397, + "learning_rate": 2.693281021897809e-08, + "loss": 0.5563, + "step": 6908 + }, + { + "epoch": 5.7197680839315295, + "grad_norm": 0.9774061441421509, + "learning_rate": 2.6772065366413857e-08, + "loss": 0.5458, + "step": 6909 + }, + { + "epoch": 5.720596355604639, + "grad_norm": 0.9635685682296753, + "learning_rate": 2.6611799060399513e-08, + "loss": 0.5281, + "step": 6910 + }, + { + "epoch": 5.721424627277747, + "grad_norm": 0.9780781865119934, + "learning_rate": 2.6452011331945526e-08, + "loss": 0.5391, + "step": 6911 + }, + { + "epoch": 5.722252898950856, + "grad_norm": 0.9801902174949646, + "learning_rate": 2.629270221196939e-08, + "loss": 0.5486, + "step": 6912 + }, + { + "epoch": 5.723081170623964, + "grad_norm": 0.9873698949813843, + "learning_rate": 2.6133871731295612e-08, + "loss": 0.5411, + "step": 6913 + }, + { + "epoch": 5.7239094422970735, + "grad_norm": 0.9897339344024658, + "learning_rate": 2.5975519920656554e-08, + "loss": 0.5532, + "step": 6914 + }, + { + "epoch": 5.724737713970182, + "grad_norm": 0.950273871421814, + "learning_rate": 2.5817646810691875e-08, + "loss": 0.5358, + "step": 6915 + }, + { + "epoch": 5.725565985643291, + "grad_norm": 0.9888784885406494, + "learning_rate": 2.5660252431948528e-08, + "loss": 0.5451, + "step": 6916 + }, + { + "epoch": 5.7263942573164, + "grad_norm": 0.9757644534111023, + "learning_rate": 2.5503336814880762e-08, + "loss": 0.5463, + "step": 6917 + }, + { + "epoch": 5.7272225289895085, + "grad_norm": 0.9696427583694458, + "learning_rate": 2.5346899989850404e-08, + "loss": 0.5402, + "step": 6918 + }, + { + "epoch": 5.728050800662618, + "grad_norm": 0.9729555249214172, + "learning_rate": 2.5190941987126295e-08, + "loss": 0.5491, + "step": 6919 + }, + { + "epoch": 5.728879072335726, + "grad_norm": 0.9871999025344849, + "learning_rate": 2.5035462836885682e-08, + "loss": 0.5373, + "step": 6920 + }, + { + "epoch": 5.729707344008835, + "grad_norm": 0.9643590450286865, + "learning_rate": 2.4880462569210895e-08, + "loss": 0.5246, + "step": 6921 + }, + { + "epoch": 5.730535615681943, + "grad_norm": 0.9502609968185425, + "learning_rate": 2.4725941214094327e-08, + "loss": 0.5305, + "step": 6922 + }, + { + "epoch": 5.7313638873550525, + "grad_norm": 0.971699595451355, + "learning_rate": 2.4571898801433736e-08, + "loss": 0.5483, + "step": 6923 + }, + { + "epoch": 5.732192159028161, + "grad_norm": 0.9628726243972778, + "learning_rate": 2.4418335361034994e-08, + "loss": 0.5402, + "step": 6924 + }, + { + "epoch": 5.73302043070127, + "grad_norm": 0.9910405874252319, + "learning_rate": 2.4265250922611285e-08, + "loss": 0.552, + "step": 6925 + }, + { + "epoch": 5.733848702374379, + "grad_norm": 0.9654805064201355, + "learning_rate": 2.411264551578335e-08, + "loss": 0.5526, + "step": 6926 + }, + { + "epoch": 5.734676974047487, + "grad_norm": 0.9743713736534119, + "learning_rate": 2.3960519170078135e-08, + "loss": 0.5392, + "step": 6927 + }, + { + "epoch": 5.735505245720597, + "grad_norm": 0.9835450649261475, + "learning_rate": 2.3808871914931254e-08, + "loss": 0.5426, + "step": 6928 + }, + { + "epoch": 5.736333517393705, + "grad_norm": 0.975803792476654, + "learning_rate": 2.365770377968507e-08, + "loss": 0.5375, + "step": 6929 + }, + { + "epoch": 5.737161789066814, + "grad_norm": 0.9816474318504333, + "learning_rate": 2.3507014793588968e-08, + "loss": 0.5647, + "step": 6930 + }, + { + "epoch": 5.737990060739922, + "grad_norm": 0.9795608520507812, + "learning_rate": 2.3356804985799896e-08, + "loss": 0.558, + "step": 6931 + }, + { + "epoch": 5.7388183324130315, + "grad_norm": 0.9778411984443665, + "learning_rate": 2.3207074385382112e-08, + "loss": 0.5584, + "step": 6932 + }, + { + "epoch": 5.73964660408614, + "grad_norm": 0.9622780084609985, + "learning_rate": 2.3057823021307157e-08, + "loss": 0.5349, + "step": 6933 + }, + { + "epoch": 5.740474875759249, + "grad_norm": 0.9485905170440674, + "learning_rate": 2.29090509224536e-08, + "loss": 0.5404, + "step": 6934 + }, + { + "epoch": 5.741303147432358, + "grad_norm": 0.9576250910758972, + "learning_rate": 2.2760758117607305e-08, + "loss": 0.5421, + "step": 6935 + }, + { + "epoch": 5.742131419105466, + "grad_norm": 0.9702200889587402, + "learning_rate": 2.2612944635461985e-08, + "loss": 0.5401, + "step": 6936 + }, + { + "epoch": 5.742959690778576, + "grad_norm": 0.9720341563224792, + "learning_rate": 2.2465610504618097e-08, + "loss": 0.5628, + "step": 6937 + }, + { + "epoch": 5.743787962451684, + "grad_norm": 0.964290976524353, + "learning_rate": 2.231875575358311e-08, + "loss": 0.5512, + "step": 6938 + }, + { + "epoch": 5.744616234124793, + "grad_norm": 0.9775505065917969, + "learning_rate": 2.2172380410772354e-08, + "loss": 0.543, + "step": 6939 + }, + { + "epoch": 5.745444505797901, + "grad_norm": 0.9689639210700989, + "learning_rate": 2.2026484504507895e-08, + "loss": 0.5483, + "step": 6940 + }, + { + "epoch": 5.7462727774710105, + "grad_norm": 0.983322024345398, + "learning_rate": 2.1881068063019097e-08, + "loss": 0.5586, + "step": 6941 + }, + { + "epoch": 5.747101049144119, + "grad_norm": 0.9811525344848633, + "learning_rate": 2.173613111444317e-08, + "loss": 0.5456, + "step": 6942 + }, + { + "epoch": 5.747929320817228, + "grad_norm": 0.9617725014686584, + "learning_rate": 2.1591673686824078e-08, + "loss": 0.5378, + "step": 6943 + }, + { + "epoch": 5.748757592490337, + "grad_norm": 0.9601479768753052, + "learning_rate": 2.1447695808112233e-08, + "loss": 0.5432, + "step": 6944 + }, + { + "epoch": 5.749585864163445, + "grad_norm": 0.9921797513961792, + "learning_rate": 2.1304197506166746e-08, + "loss": 0.5465, + "step": 6945 + }, + { + "epoch": 5.750414135836555, + "grad_norm": 0.9696030020713806, + "learning_rate": 2.1161178808752624e-08, + "loss": 0.5707, + "step": 6946 + }, + { + "epoch": 5.751242407509663, + "grad_norm": 0.9966047406196594, + "learning_rate": 2.1018639743543013e-08, + "loss": 0.5373, + "step": 6947 + }, + { + "epoch": 5.752070679182772, + "grad_norm": 0.9805617332458496, + "learning_rate": 2.087658033811807e-08, + "loss": 0.5761, + "step": 6948 + }, + { + "epoch": 5.752898950855881, + "grad_norm": 0.9657739996910095, + "learning_rate": 2.0735000619964696e-08, + "loss": 0.5719, + "step": 6949 + }, + { + "epoch": 5.7537272225289895, + "grad_norm": 0.9992994070053101, + "learning_rate": 2.059390061647709e-08, + "loss": 0.5542, + "step": 6950 + }, + { + "epoch": 5.754555494202099, + "grad_norm": 0.9776666164398193, + "learning_rate": 2.0453280354957307e-08, + "loss": 0.5553, + "step": 6951 + }, + { + "epoch": 5.755383765875207, + "grad_norm": 0.949368417263031, + "learning_rate": 2.0313139862613575e-08, + "loss": 0.5304, + "step": 6952 + }, + { + "epoch": 5.756212037548316, + "grad_norm": 0.9486882090568542, + "learning_rate": 2.0173479166561705e-08, + "loss": 0.5446, + "step": 6953 + }, + { + "epoch": 5.757040309221424, + "grad_norm": 0.9690361022949219, + "learning_rate": 2.003429829382536e-08, + "loss": 0.5388, + "step": 6954 + }, + { + "epoch": 5.757868580894534, + "grad_norm": 0.9916055798530579, + "learning_rate": 1.989559727133411e-08, + "loss": 0.5744, + "step": 6955 + }, + { + "epoch": 5.758696852567642, + "grad_norm": 0.9738194346427917, + "learning_rate": 1.9757376125925652e-08, + "loss": 0.5262, + "step": 6956 + }, + { + "epoch": 5.759525124240751, + "grad_norm": 0.9450884461402893, + "learning_rate": 1.9619634884344152e-08, + "loss": 0.5351, + "step": 6957 + }, + { + "epoch": 5.76035339591386, + "grad_norm": 0.9778859615325928, + "learning_rate": 1.9482373573241897e-08, + "loss": 0.5437, + "step": 6958 + }, + { + "epoch": 5.7611816675869685, + "grad_norm": 0.9689139127731323, + "learning_rate": 1.934559221917709e-08, + "loss": 0.544, + "step": 6959 + }, + { + "epoch": 5.762009939260078, + "grad_norm": 0.9951791763305664, + "learning_rate": 1.920929084861578e-08, + "loss": 0.5347, + "step": 6960 + }, + { + "epoch": 5.762838210933186, + "grad_norm": 0.9886312484741211, + "learning_rate": 1.907346948793104e-08, + "loss": 0.573, + "step": 6961 + }, + { + "epoch": 5.763666482606295, + "grad_norm": 0.9668784737586975, + "learning_rate": 1.8938128163402957e-08, + "loss": 0.535, + "step": 6962 + }, + { + "epoch": 5.764494754279403, + "grad_norm": 0.9551658034324646, + "learning_rate": 1.880326690121892e-08, + "loss": 0.5561, + "step": 6963 + }, + { + "epoch": 5.765323025952513, + "grad_norm": 0.9872788190841675, + "learning_rate": 1.8668885727473608e-08, + "loss": 0.557, + "step": 6964 + }, + { + "epoch": 5.766151297625621, + "grad_norm": 0.9696853160858154, + "learning_rate": 1.8534984668167887e-08, + "loss": 0.524, + "step": 6965 + }, + { + "epoch": 5.76697956929873, + "grad_norm": 0.9753289222717285, + "learning_rate": 1.840156374921104e-08, + "loss": 0.5314, + "step": 6966 + }, + { + "epoch": 5.767807840971839, + "grad_norm": 0.9808300137519836, + "learning_rate": 1.826862299641824e-08, + "loss": 0.5507, + "step": 6967 + }, + { + "epoch": 5.7686361126449475, + "grad_norm": 0.9681927561759949, + "learning_rate": 1.8136162435512528e-08, + "loss": 0.5553, + "step": 6968 + }, + { + "epoch": 5.769464384318057, + "grad_norm": 0.9726253747940063, + "learning_rate": 1.800418209212368e-08, + "loss": 0.5355, + "step": 6969 + }, + { + "epoch": 5.770292655991165, + "grad_norm": 0.9678818583488464, + "learning_rate": 1.7872681991788765e-08, + "loss": 0.5417, + "step": 6970 + }, + { + "epoch": 5.771120927664274, + "grad_norm": 0.9672456979751587, + "learning_rate": 1.77416621599516e-08, + "loss": 0.5494, + "step": 6971 + }, + { + "epoch": 5.771949199337382, + "grad_norm": 1.0456998348236084, + "learning_rate": 1.761112262196385e-08, + "loss": 0.5418, + "step": 6972 + }, + { + "epoch": 5.7727774710104915, + "grad_norm": 0.9646222591400146, + "learning_rate": 1.7481063403083087e-08, + "loss": 0.5292, + "step": 6973 + }, + { + "epoch": 5.7736057426836, + "grad_norm": 0.9929711222648621, + "learning_rate": 1.7351484528475016e-08, + "loss": 0.5583, + "step": 6974 + }, + { + "epoch": 5.774434014356709, + "grad_norm": 0.9624965786933899, + "learning_rate": 1.7222386023211523e-08, + "loss": 0.5469, + "step": 6975 + }, + { + "epoch": 5.775262286029818, + "grad_norm": 0.9512826204299927, + "learning_rate": 1.709376791227235e-08, + "loss": 0.5508, + "step": 6976 + }, + { + "epoch": 5.7760905577029265, + "grad_norm": 0.9400210976600647, + "learning_rate": 1.6965630220543705e-08, + "loss": 0.5374, + "step": 6977 + }, + { + "epoch": 5.776918829376036, + "grad_norm": 1.0849382877349854, + "learning_rate": 1.6837972972818805e-08, + "loss": 0.5262, + "step": 6978 + }, + { + "epoch": 5.777747101049144, + "grad_norm": 0.9866847395896912, + "learning_rate": 1.6710796193798727e-08, + "loss": 0.5578, + "step": 6979 + }, + { + "epoch": 5.778575372722253, + "grad_norm": 0.999788761138916, + "learning_rate": 1.6584099908090457e-08, + "loss": 0.5424, + "step": 6980 + }, + { + "epoch": 5.779403644395361, + "grad_norm": 0.969163179397583, + "learning_rate": 1.6457884140208825e-08, + "loss": 0.5478, + "step": 6981 + }, + { + "epoch": 5.7802319160684705, + "grad_norm": 0.9818345308303833, + "learning_rate": 1.633214891457541e-08, + "loss": 0.5541, + "step": 6982 + }, + { + "epoch": 5.781060187741579, + "grad_norm": 0.9601730704307556, + "learning_rate": 1.6206894255518523e-08, + "loss": 0.5572, + "step": 6983 + }, + { + "epoch": 5.781888459414688, + "grad_norm": 0.9823173880577087, + "learning_rate": 1.6082120187274064e-08, + "loss": 0.5666, + "step": 6984 + }, + { + "epoch": 5.782716731087797, + "grad_norm": 0.9641292095184326, + "learning_rate": 1.595782673398466e-08, + "loss": 0.5409, + "step": 6985 + }, + { + "epoch": 5.783545002760905, + "grad_norm": 0.9899305701255798, + "learning_rate": 1.5834013919699688e-08, + "loss": 0.5336, + "step": 6986 + }, + { + "epoch": 5.784373274434015, + "grad_norm": 0.9704510569572449, + "learning_rate": 1.5710681768375812e-08, + "loss": 0.5214, + "step": 6987 + }, + { + "epoch": 5.785201546107123, + "grad_norm": 0.9482161998748779, + "learning_rate": 1.5587830303877006e-08, + "loss": 0.5383, + "step": 6988 + }, + { + "epoch": 5.786029817780232, + "grad_norm": 0.9724152684211731, + "learning_rate": 1.5465459549973692e-08, + "loss": 0.5466, + "step": 6989 + }, + { + "epoch": 5.78685808945334, + "grad_norm": 0.9514291286468506, + "learning_rate": 1.5343569530343326e-08, + "loss": 0.5448, + "step": 6990 + }, + { + "epoch": 5.7876863611264495, + "grad_norm": 0.9600810408592224, + "learning_rate": 1.5222160268570375e-08, + "loss": 0.517, + "step": 6991 + }, + { + "epoch": 5.788514632799558, + "grad_norm": 0.9695376753807068, + "learning_rate": 1.5101231788146874e-08, + "loss": 0.5271, + "step": 6992 + }, + { + "epoch": 5.789342904472667, + "grad_norm": 0.984328031539917, + "learning_rate": 1.4980784112471057e-08, + "loss": 0.5486, + "step": 6993 + }, + { + "epoch": 5.790171176145776, + "grad_norm": 0.9926164150238037, + "learning_rate": 1.486081726484817e-08, + "loss": 0.5545, + "step": 6994 + }, + { + "epoch": 5.790999447818884, + "grad_norm": 0.9973723292350769, + "learning_rate": 1.474133126849131e-08, + "loss": 0.5725, + "step": 6995 + }, + { + "epoch": 5.791827719491994, + "grad_norm": 0.9827430844306946, + "learning_rate": 1.4622326146519483e-08, + "loss": 0.533, + "step": 6996 + }, + { + "epoch": 5.792655991165102, + "grad_norm": 0.9672914743423462, + "learning_rate": 1.4503801921958993e-08, + "loss": 0.5617, + "step": 6997 + }, + { + "epoch": 5.793484262838211, + "grad_norm": 0.9774397015571594, + "learning_rate": 1.4385758617743439e-08, + "loss": 0.5382, + "step": 6998 + }, + { + "epoch": 5.794312534511319, + "grad_norm": 0.9794498085975647, + "learning_rate": 1.4268196256713163e-08, + "loss": 0.54, + "step": 6999 + }, + { + "epoch": 5.7951408061844285, + "grad_norm": 0.9872910976409912, + "learning_rate": 1.4151114861614967e-08, + "loss": 0.5647, + "step": 7000 + }, + { + "epoch": 5.795969077857537, + "grad_norm": 0.9678724408149719, + "learning_rate": 1.403451445510351e-08, + "loss": 0.5481, + "step": 7001 + }, + { + "epoch": 5.796797349530646, + "grad_norm": 0.9404424428939819, + "learning_rate": 1.3918395059739354e-08, + "loss": 0.5613, + "step": 7002 + }, + { + "epoch": 5.797625621203755, + "grad_norm": 0.9896771907806396, + "learning_rate": 1.3802756697990915e-08, + "loss": 0.5695, + "step": 7003 + }, + { + "epoch": 5.798453892876863, + "grad_norm": 0.982484757900238, + "learning_rate": 1.3687599392233075e-08, + "loss": 0.5487, + "step": 7004 + }, + { + "epoch": 5.799282164549973, + "grad_norm": 0.9674330353736877, + "learning_rate": 1.3572923164747732e-08, + "loss": 0.5403, + "step": 7005 + }, + { + "epoch": 5.800110436223081, + "grad_norm": 0.956146240234375, + "learning_rate": 1.3458728037723245e-08, + "loss": 0.568, + "step": 7006 + }, + { + "epoch": 5.80093870789619, + "grad_norm": 0.9884467720985413, + "learning_rate": 1.3345014033255831e-08, + "loss": 0.5518, + "step": 7007 + }, + { + "epoch": 5.801766979569299, + "grad_norm": 0.9704195857048035, + "learning_rate": 1.323178117334789e-08, + "loss": 0.5486, + "step": 7008 + }, + { + "epoch": 5.8025952512424075, + "grad_norm": 0.9681942462921143, + "learning_rate": 1.3119029479909117e-08, + "loss": 0.5376, + "step": 7009 + }, + { + "epoch": 5.803423522915517, + "grad_norm": 0.9738938808441162, + "learning_rate": 1.3006758974755395e-08, + "loss": 0.5413, + "step": 7010 + }, + { + "epoch": 5.804251794588625, + "grad_norm": 0.9589844346046448, + "learning_rate": 1.2894969679610737e-08, + "loss": 0.5156, + "step": 7011 + }, + { + "epoch": 5.805080066261734, + "grad_norm": 0.989353597164154, + "learning_rate": 1.2783661616104504e-08, + "loss": 0.5441, + "step": 7012 + }, + { + "epoch": 5.805908337934842, + "grad_norm": 0.9586181640625, + "learning_rate": 1.2672834805774469e-08, + "loss": 0.5468, + "step": 7013 + }, + { + "epoch": 5.806736609607952, + "grad_norm": 1.0199816226959229, + "learning_rate": 1.2562489270064314e-08, + "loss": 0.5377, + "step": 7014 + }, + { + "epoch": 5.80756488128106, + "grad_norm": 0.988461971282959, + "learning_rate": 1.2452625030325016e-08, + "loss": 0.5466, + "step": 7015 + }, + { + "epoch": 5.808393152954169, + "grad_norm": 0.9814547300338745, + "learning_rate": 1.2343242107814012e-08, + "loss": 0.5354, + "step": 7016 + }, + { + "epoch": 5.809221424627278, + "grad_norm": 0.9581533074378967, + "learning_rate": 1.2234340523696043e-08, + "loss": 0.5568, + "step": 7017 + }, + { + "epoch": 5.8100496963003865, + "grad_norm": 1.0101220607757568, + "learning_rate": 1.2125920299042303e-08, + "loss": 0.5265, + "step": 7018 + }, + { + "epoch": 5.810877967973496, + "grad_norm": 0.9737896919250488, + "learning_rate": 1.2017981454831574e-08, + "loss": 0.5324, + "step": 7019 + }, + { + "epoch": 5.811706239646604, + "grad_norm": 0.9637385606765747, + "learning_rate": 1.191052401194881e-08, + "loss": 0.5482, + "step": 7020 + }, + { + "epoch": 5.812534511319713, + "grad_norm": 0.9694854617118835, + "learning_rate": 1.1803547991185715e-08, + "loss": 0.5487, + "step": 7021 + }, + { + "epoch": 5.813362782992821, + "grad_norm": 0.9526222944259644, + "learning_rate": 1.1697053413241566e-08, + "loss": 0.563, + "step": 7022 + }, + { + "epoch": 5.814191054665931, + "grad_norm": 0.9911684989929199, + "learning_rate": 1.1591040298722101e-08, + "loss": 0.5409, + "step": 7023 + }, + { + "epoch": 5.815019326339039, + "grad_norm": 0.9701586961746216, + "learning_rate": 1.1485508668139522e-08, + "loss": 0.5541, + "step": 7024 + }, + { + "epoch": 5.815847598012148, + "grad_norm": 1.0002903938293457, + "learning_rate": 1.138045854191333e-08, + "loss": 0.5341, + "step": 7025 + }, + { + "epoch": 5.816675869685257, + "grad_norm": 0.9774640798568726, + "learning_rate": 1.1275889940369766e-08, + "loss": 0.5548, + "step": 7026 + }, + { + "epoch": 5.8175041413583655, + "grad_norm": 0.9753071665763855, + "learning_rate": 1.1171802883742088e-08, + "loss": 0.5564, + "step": 7027 + }, + { + "epoch": 5.818332413031475, + "grad_norm": 0.9886694550514221, + "learning_rate": 1.1068197392169743e-08, + "loss": 0.5726, + "step": 7028 + }, + { + "epoch": 5.819160684704583, + "grad_norm": 1.0048620700836182, + "learning_rate": 1.096507348569975e-08, + "loss": 0.5582, + "step": 7029 + }, + { + "epoch": 5.819988956377692, + "grad_norm": 0.9816103577613831, + "learning_rate": 1.0862431184285593e-08, + "loss": 0.5333, + "step": 7030 + }, + { + "epoch": 5.8208172280508, + "grad_norm": 0.9757484197616577, + "learning_rate": 1.0760270507787495e-08, + "loss": 0.5755, + "step": 7031 + }, + { + "epoch": 5.8216454997239095, + "grad_norm": 0.9705796837806702, + "learning_rate": 1.0658591475972701e-08, + "loss": 0.5659, + "step": 7032 + }, + { + "epoch": 5.822473771397018, + "grad_norm": 0.9831752777099609, + "learning_rate": 1.0557394108515195e-08, + "loss": 0.5499, + "step": 7033 + }, + { + "epoch": 5.823302043070127, + "grad_norm": 0.962212324142456, + "learning_rate": 1.0456678424995703e-08, + "loss": 0.5398, + "step": 7034 + }, + { + "epoch": 5.824130314743236, + "grad_norm": 0.9733263254165649, + "learning_rate": 1.0356444444901693e-08, + "loss": 0.5296, + "step": 7035 + }, + { + "epoch": 5.8249585864163445, + "grad_norm": 0.9691694974899292, + "learning_rate": 1.025669218762737e-08, + "loss": 0.53, + "step": 7036 + }, + { + "epoch": 5.825786858089454, + "grad_norm": 0.992228090763092, + "learning_rate": 1.0157421672473965e-08, + "loss": 0.5462, + "step": 7037 + }, + { + "epoch": 5.826615129762562, + "grad_norm": 0.9853614568710327, + "learning_rate": 1.0058632918649725e-08, + "loss": 0.5663, + "step": 7038 + }, + { + "epoch": 5.827443401435671, + "grad_norm": 0.9494875073432922, + "learning_rate": 9.9603259452688e-09, + "loss": 0.5625, + "step": 7039 + }, + { + "epoch": 5.828271673108779, + "grad_norm": 0.9839235544204712, + "learning_rate": 9.862500771352922e-09, + "loss": 0.559, + "step": 7040 + }, + { + "epoch": 5.8290999447818885, + "grad_norm": 1.0184983015060425, + "learning_rate": 9.765157415830562e-09, + "loss": 0.5203, + "step": 7041 + }, + { + "epoch": 5.829928216454997, + "grad_norm": 0.9664837121963501, + "learning_rate": 9.668295897536373e-09, + "loss": 0.5496, + "step": 7042 + }, + { + "epoch": 5.830756488128106, + "grad_norm": 0.9685920476913452, + "learning_rate": 9.571916235212587e-09, + "loss": 0.5434, + "step": 7043 + }, + { + "epoch": 5.831584759801215, + "grad_norm": 0.9576565027236938, + "learning_rate": 9.476018447507341e-09, + "loss": 0.5357, + "step": 7044 + }, + { + "epoch": 5.832413031474323, + "grad_norm": 0.9969997406005859, + "learning_rate": 9.380602552976348e-09, + "loss": 0.535, + "step": 7045 + }, + { + "epoch": 5.833241303147433, + "grad_norm": 0.9744444489479065, + "learning_rate": 9.285668570081507e-09, + "loss": 0.5378, + "step": 7046 + }, + { + "epoch": 5.834069574820541, + "grad_norm": 0.9656137824058533, + "learning_rate": 9.191216517191459e-09, + "loss": 0.5483, + "step": 7047 + }, + { + "epoch": 5.83489784649365, + "grad_norm": 0.9839324951171875, + "learning_rate": 9.09724641258214e-09, + "loss": 0.5622, + "step": 7048 + }, + { + "epoch": 5.835726118166758, + "grad_norm": 0.9730606079101562, + "learning_rate": 9.00375827443567e-09, + "loss": 0.5304, + "step": 7049 + }, + { + "epoch": 5.8365543898398675, + "grad_norm": 0.9569813013076782, + "learning_rate": 8.910752120841193e-09, + "loss": 0.5475, + "step": 7050 + }, + { + "epoch": 5.837382661512976, + "grad_norm": 0.9814510345458984, + "learning_rate": 8.81822796979459e-09, + "loss": 0.5379, + "step": 7051 + }, + { + "epoch": 5.838210933186085, + "grad_norm": 0.956279456615448, + "learning_rate": 8.72618583919821e-09, + "loss": 0.5352, + "step": 7052 + }, + { + "epoch": 5.839039204859194, + "grad_norm": 0.9415268301963806, + "learning_rate": 8.634625746861692e-09, + "loss": 0.5461, + "step": 7053 + }, + { + "epoch": 5.839867476532302, + "grad_norm": 0.9894720315933228, + "learning_rate": 8.543547710500866e-09, + "loss": 0.589, + "step": 7054 + }, + { + "epoch": 5.840695748205412, + "grad_norm": 1.0055619478225708, + "learning_rate": 8.452951747738581e-09, + "loss": 0.5441, + "step": 7055 + }, + { + "epoch": 5.84152401987852, + "grad_norm": 0.9444437623023987, + "learning_rate": 8.362837876104147e-09, + "loss": 0.5346, + "step": 7056 + }, + { + "epoch": 5.842352291551629, + "grad_norm": 0.9920063614845276, + "learning_rate": 8.273206113033893e-09, + "loss": 0.5533, + "step": 7057 + }, + { + "epoch": 5.843180563224737, + "grad_norm": 0.9872284531593323, + "learning_rate": 8.184056475870616e-09, + "loss": 0.5523, + "step": 7058 + }, + { + "epoch": 5.8440088348978465, + "grad_norm": 0.9557161331176758, + "learning_rate": 8.09538898186385e-09, + "loss": 0.5358, + "step": 7059 + }, + { + "epoch": 5.844837106570955, + "grad_norm": 0.9707210659980774, + "learning_rate": 8.007203648170147e-09, + "loss": 0.5551, + "step": 7060 + }, + { + "epoch": 5.845665378244064, + "grad_norm": 0.9481123089790344, + "learning_rate": 7.919500491852528e-09, + "loss": 0.5406, + "step": 7061 + }, + { + "epoch": 5.846493649917173, + "grad_norm": 0.9758118987083435, + "learning_rate": 7.832279529880748e-09, + "loss": 0.5336, + "step": 7062 + }, + { + "epoch": 5.847321921590281, + "grad_norm": 0.9930400848388672, + "learning_rate": 7.745540779131033e-09, + "loss": 0.5294, + "step": 7063 + }, + { + "epoch": 5.848150193263391, + "grad_norm": 0.9815714359283447, + "learning_rate": 7.659284256386901e-09, + "loss": 0.543, + "step": 7064 + }, + { + "epoch": 5.848978464936499, + "grad_norm": 0.9778390526771545, + "learning_rate": 7.573509978337779e-09, + "loss": 0.5492, + "step": 7065 + }, + { + "epoch": 5.849806736609608, + "grad_norm": 1.0015196800231934, + "learning_rate": 7.488217961580391e-09, + "loss": 0.5467, + "step": 7066 + }, + { + "epoch": 5.850635008282717, + "grad_norm": 0.9793812036514282, + "learning_rate": 7.403408222617925e-09, + "loss": 0.5402, + "step": 7067 + }, + { + "epoch": 5.8514632799558255, + "grad_norm": 0.9751242995262146, + "learning_rate": 7.31908077786031e-09, + "loss": 0.5561, + "step": 7068 + }, + { + "epoch": 5.852291551628935, + "grad_norm": 0.9758898019790649, + "learning_rate": 7.2352356436242165e-09, + "loss": 0.5443, + "step": 7069 + }, + { + "epoch": 5.853119823302043, + "grad_norm": 0.9417216777801514, + "learning_rate": 7.151872836132501e-09, + "loss": 0.5524, + "step": 7070 + }, + { + "epoch": 5.853948094975152, + "grad_norm": 0.9571611881256104, + "learning_rate": 7.068992371515593e-09, + "loss": 0.5414, + "step": 7071 + }, + { + "epoch": 5.85477636664826, + "grad_norm": 0.9850255250930786, + "learning_rate": 6.9865942658098316e-09, + "loss": 0.5338, + "step": 7072 + }, + { + "epoch": 5.85560463832137, + "grad_norm": 0.9601150751113892, + "learning_rate": 6.904678534958298e-09, + "loss": 0.5581, + "step": 7073 + }, + { + "epoch": 5.856432909994478, + "grad_norm": 0.9871248602867126, + "learning_rate": 6.823245194811368e-09, + "loss": 0.5316, + "step": 7074 + }, + { + "epoch": 5.857261181667587, + "grad_norm": 0.9434945583343506, + "learning_rate": 6.7422942611253265e-09, + "loss": 0.5383, + "step": 7075 + }, + { + "epoch": 5.858089453340696, + "grad_norm": 0.9655126333236694, + "learning_rate": 6.661825749563477e-09, + "loss": 0.5443, + "step": 7076 + }, + { + "epoch": 5.8589177250138045, + "grad_norm": 0.9629698395729065, + "learning_rate": 6.581839675695589e-09, + "loss": 0.5416, + "step": 7077 + }, + { + "epoch": 5.859745996686914, + "grad_norm": 0.9848405718803406, + "learning_rate": 6.502336054998449e-09, + "loss": 0.5486, + "step": 7078 + }, + { + "epoch": 5.860574268360022, + "grad_norm": 0.9730758666992188, + "learning_rate": 6.423314902855582e-09, + "loss": 0.5403, + "step": 7079 + }, + { + "epoch": 5.861402540033131, + "grad_norm": 0.9618121385574341, + "learning_rate": 6.344776234555872e-09, + "loss": 0.547, + "step": 7080 + }, + { + "epoch": 5.862230811706239, + "grad_norm": 0.9769609570503235, + "learning_rate": 6.266720065296605e-09, + "loss": 0.5952, + "step": 7081 + }, + { + "epoch": 5.863059083379349, + "grad_norm": 0.97953861951828, + "learning_rate": 6.18914641018098e-09, + "loss": 0.5433, + "step": 7082 + }, + { + "epoch": 5.863887355052457, + "grad_norm": 0.9867740869522095, + "learning_rate": 6.1120552842183764e-09, + "loss": 0.5378, + "step": 7083 + }, + { + "epoch": 5.864715626725566, + "grad_norm": 0.9563124179840088, + "learning_rate": 6.035446702325198e-09, + "loss": 0.5497, + "step": 7084 + }, + { + "epoch": 5.865543898398675, + "grad_norm": 1.0118049383163452, + "learning_rate": 5.959320679324587e-09, + "loss": 0.5424, + "step": 7085 + }, + { + "epoch": 5.8663721700717835, + "grad_norm": 0.9843424558639526, + "learning_rate": 5.883677229946428e-09, + "loss": 0.562, + "step": 7086 + }, + { + "epoch": 5.867200441744893, + "grad_norm": 1.0272270441055298, + "learning_rate": 5.8085163688267885e-09, + "loss": 0.5397, + "step": 7087 + }, + { + "epoch": 5.868028713418001, + "grad_norm": 0.9695208072662354, + "learning_rate": 5.7338381105087605e-09, + "loss": 0.5518, + "step": 7088 + }, + { + "epoch": 5.86885698509111, + "grad_norm": 0.9898740649223328, + "learning_rate": 5.659642469441895e-09, + "loss": 0.5294, + "step": 7089 + }, + { + "epoch": 5.869685256764218, + "grad_norm": 0.9693114161491394, + "learning_rate": 5.585929459982209e-09, + "loss": 0.5288, + "step": 7090 + }, + { + "epoch": 5.8705135284373275, + "grad_norm": 0.9432626366615295, + "learning_rate": 5.51269909639246e-09, + "loss": 0.5412, + "step": 7091 + }, + { + "epoch": 5.871341800110436, + "grad_norm": 0.9695253372192383, + "learning_rate": 5.439951392842147e-09, + "loss": 0.5485, + "step": 7092 + }, + { + "epoch": 5.872170071783545, + "grad_norm": 0.9403347969055176, + "learning_rate": 5.367686363407232e-09, + "loss": 0.5214, + "step": 7093 + }, + { + "epoch": 5.872998343456654, + "grad_norm": 0.9907011985778809, + "learning_rate": 5.2959040220704194e-09, + "loss": 0.5386, + "step": 7094 + }, + { + "epoch": 5.8738266151297625, + "grad_norm": 0.9865439534187317, + "learning_rate": 5.224604382720877e-09, + "loss": 0.5281, + "step": 7095 + }, + { + "epoch": 5.874654886802872, + "grad_norm": 0.9828709959983826, + "learning_rate": 5.153787459154236e-09, + "loss": 0.5611, + "step": 7096 + }, + { + "epoch": 5.87548315847598, + "grad_norm": 0.9787101745605469, + "learning_rate": 5.083453265073146e-09, + "loss": 0.5526, + "step": 7097 + }, + { + "epoch": 5.876311430149089, + "grad_norm": 0.9709711670875549, + "learning_rate": 5.013601814086444e-09, + "loss": 0.5601, + "step": 7098 + }, + { + "epoch": 5.877139701822197, + "grad_norm": 0.9540531039237976, + "learning_rate": 4.9442331197097074e-09, + "loss": 0.5397, + "step": 7099 + }, + { + "epoch": 5.8779679734953065, + "grad_norm": 0.9769091606140137, + "learning_rate": 4.875347195365532e-09, + "loss": 0.5413, + "step": 7100 + }, + { + "epoch": 5.878796245168415, + "grad_norm": 0.9591909050941467, + "learning_rate": 4.806944054382423e-09, + "loss": 0.5586, + "step": 7101 + }, + { + "epoch": 5.879624516841524, + "grad_norm": 0.9611213803291321, + "learning_rate": 4.739023709995627e-09, + "loss": 0.5356, + "step": 7102 + }, + { + "epoch": 5.880452788514633, + "grad_norm": 0.9750087261199951, + "learning_rate": 4.671586175347132e-09, + "loss": 0.5317, + "step": 7103 + }, + { + "epoch": 5.881281060187741, + "grad_norm": 0.9924328923225403, + "learning_rate": 4.604631463485665e-09, + "loss": 0.5635, + "step": 7104 + }, + { + "epoch": 5.882109331860851, + "grad_norm": 0.9783445596694946, + "learning_rate": 4.538159587366142e-09, + "loss": 0.5616, + "step": 7105 + }, + { + "epoch": 5.882937603533959, + "grad_norm": 0.9920464158058167, + "learning_rate": 4.472170559850497e-09, + "loss": 0.5397, + "step": 7106 + }, + { + "epoch": 5.883765875207068, + "grad_norm": 0.9603058099746704, + "learning_rate": 4.4066643937068495e-09, + "loss": 0.5495, + "step": 7107 + }, + { + "epoch": 5.884594146880176, + "grad_norm": 0.9678807854652405, + "learning_rate": 4.3416411016097835e-09, + "loss": 0.5359, + "step": 7108 + }, + { + "epoch": 5.8854224185532855, + "grad_norm": 0.9531580805778503, + "learning_rate": 4.277100696140901e-09, + "loss": 0.5449, + "step": 7109 + }, + { + "epoch": 5.886250690226394, + "grad_norm": 0.9751785397529602, + "learning_rate": 4.213043189788546e-09, + "loss": 0.5563, + "step": 7110 + }, + { + "epoch": 5.887078961899503, + "grad_norm": 0.9814699292182922, + "learning_rate": 4.1494685949466925e-09, + "loss": 0.5208, + "step": 7111 + }, + { + "epoch": 5.887907233572612, + "grad_norm": 0.9473446011543274, + "learning_rate": 4.0863769239166126e-09, + "loss": 0.5519, + "step": 7112 + }, + { + "epoch": 5.88873550524572, + "grad_norm": 0.9983262419700623, + "learning_rate": 4.023768188906041e-09, + "loss": 0.5251, + "step": 7113 + }, + { + "epoch": 5.88956377691883, + "grad_norm": 0.9547951817512512, + "learning_rate": 3.961642402029175e-09, + "loss": 0.5597, + "step": 7114 + }, + { + "epoch": 5.890392048591938, + "grad_norm": 1.050804853439331, + "learning_rate": 3.8999995753066785e-09, + "loss": 0.5296, + "step": 7115 + }, + { + "epoch": 5.891220320265047, + "grad_norm": 0.9743812084197998, + "learning_rate": 3.838839720665954e-09, + "loss": 0.5148, + "step": 7116 + }, + { + "epoch": 5.892048591938155, + "grad_norm": 0.963163435459137, + "learning_rate": 3.77816284994087e-09, + "loss": 0.5381, + "step": 7117 + }, + { + "epoch": 5.8928768636112645, + "grad_norm": 0.9762606620788574, + "learning_rate": 3.717968974871755e-09, + "loss": 0.5563, + "step": 7118 + }, + { + "epoch": 5.893705135284373, + "grad_norm": 0.9654428958892822, + "learning_rate": 3.6582581071054057e-09, + "loss": 0.5469, + "step": 7119 + }, + { + "epoch": 5.894533406957482, + "grad_norm": 1.0135955810546875, + "learning_rate": 3.5990302581959125e-09, + "loss": 0.5372, + "step": 7120 + }, + { + "epoch": 5.895361678630591, + "grad_norm": 0.9559961557388306, + "learning_rate": 3.540285439602442e-09, + "loss": 0.5373, + "step": 7121 + }, + { + "epoch": 5.896189950303699, + "grad_norm": 0.9662891030311584, + "learning_rate": 3.482023662692291e-09, + "loss": 0.5621, + "step": 7122 + }, + { + "epoch": 5.897018221976809, + "grad_norm": 0.987074613571167, + "learning_rate": 3.4242449387383856e-09, + "loss": 0.5541, + "step": 7123 + }, + { + "epoch": 5.897846493649917, + "grad_norm": 0.9992825984954834, + "learning_rate": 3.3669492789201173e-09, + "loss": 0.5523, + "step": 7124 + }, + { + "epoch": 5.898674765323026, + "grad_norm": 0.98309326171875, + "learning_rate": 3.3101366943238954e-09, + "loss": 0.5251, + "step": 7125 + }, + { + "epoch": 5.899503036996135, + "grad_norm": 0.960849404335022, + "learning_rate": 3.2538071959425934e-09, + "loss": 0.5621, + "step": 7126 + }, + { + "epoch": 5.9003313086692435, + "grad_norm": 0.9476284384727478, + "learning_rate": 3.1979607946749924e-09, + "loss": 0.5752, + "step": 7127 + }, + { + "epoch": 5.901159580342353, + "grad_norm": 0.9730549454689026, + "learning_rate": 3.1425975013271715e-09, + "loss": 0.5308, + "step": 7128 + }, + { + "epoch": 5.901987852015461, + "grad_norm": 0.961501955986023, + "learning_rate": 3.0877173266113947e-09, + "loss": 0.5386, + "step": 7129 + }, + { + "epoch": 5.90281612368857, + "grad_norm": 0.9586473107337952, + "learning_rate": 3.0333202811461125e-09, + "loss": 0.5401, + "step": 7130 + }, + { + "epoch": 5.903644395361678, + "grad_norm": 0.9700204730033875, + "learning_rate": 2.9794063754570725e-09, + "loss": 0.5334, + "step": 7131 + }, + { + "epoch": 5.904472667034788, + "grad_norm": 0.9568300247192383, + "learning_rate": 2.9259756199762068e-09, + "loss": 0.5356, + "step": 7132 + }, + { + "epoch": 5.905300938707896, + "grad_norm": 0.967390775680542, + "learning_rate": 2.873028025041358e-09, + "loss": 0.5484, + "step": 7133 + }, + { + "epoch": 5.906129210381005, + "grad_norm": 1.0474945306777954, + "learning_rate": 2.8205636008976634e-09, + "loss": 0.5268, + "step": 7134 + }, + { + "epoch": 5.906957482054114, + "grad_norm": 0.9590755701065063, + "learning_rate": 2.7685823576967252e-09, + "loss": 0.5375, + "step": 7135 + }, + { + "epoch": 5.9077857537272225, + "grad_norm": 0.9741535782814026, + "learning_rate": 2.717084305496054e-09, + "loss": 0.5403, + "step": 7136 + }, + { + "epoch": 5.908614025400332, + "grad_norm": 0.965206503868103, + "learning_rate": 2.6660694542601786e-09, + "loss": 0.5364, + "step": 7137 + }, + { + "epoch": 5.90944229707344, + "grad_norm": 0.9916989207267761, + "learning_rate": 2.615537813860092e-09, + "loss": 0.5469, + "step": 7138 + }, + { + "epoch": 5.910270568746549, + "grad_norm": 0.9577746987342834, + "learning_rate": 2.565489394072973e-09, + "loss": 0.5493, + "step": 7139 + }, + { + "epoch": 5.911098840419657, + "grad_norm": 0.9726204872131348, + "learning_rate": 2.51592420458302e-09, + "loss": 0.546, + "step": 7140 + }, + { + "epoch": 5.911927112092767, + "grad_norm": 0.9797086119651794, + "learning_rate": 2.4668422549803395e-09, + "loss": 0.5541, + "step": 7141 + }, + { + "epoch": 5.912755383765875, + "grad_norm": 0.9471781849861145, + "learning_rate": 2.4182435547620565e-09, + "loss": 0.5444, + "step": 7142 + }, + { + "epoch": 5.913583655438984, + "grad_norm": 0.959069013595581, + "learning_rate": 2.3701281133314823e-09, + "loss": 0.5528, + "step": 7143 + }, + { + "epoch": 5.914411927112093, + "grad_norm": 0.9871560335159302, + "learning_rate": 2.32249593999867e-09, + "loss": 0.5431, + "step": 7144 + }, + { + "epoch": 5.9152401987852015, + "grad_norm": 0.9931521415710449, + "learning_rate": 2.275347043979581e-09, + "loss": 0.5461, + "step": 7145 + }, + { + "epoch": 5.916068470458311, + "grad_norm": 0.982261061668396, + "learning_rate": 2.2286814343974726e-09, + "loss": 0.5292, + "step": 7146 + }, + { + "epoch": 5.916896742131419, + "grad_norm": 0.9529389142990112, + "learning_rate": 2.1824991202815116e-09, + "loss": 0.5393, + "step": 7147 + }, + { + "epoch": 5.917725013804528, + "grad_norm": 0.9705958366394043, + "learning_rate": 2.1368001105678824e-09, + "loss": 0.5805, + "step": 7148 + }, + { + "epoch": 5.918553285477636, + "grad_norm": 0.9528070092201233, + "learning_rate": 2.091584414098402e-09, + "loss": 0.557, + "step": 7149 + }, + { + "epoch": 5.9193815571507455, + "grad_norm": 0.9753689169883728, + "learning_rate": 2.046852039621905e-09, + "loss": 0.5504, + "step": 7150 + }, + { + "epoch": 5.920209828823854, + "grad_norm": 0.9820945262908936, + "learning_rate": 2.002602995794245e-09, + "loss": 0.535, + "step": 7151 + }, + { + "epoch": 5.921038100496963, + "grad_norm": 0.9580931663513184, + "learning_rate": 1.958837291176907e-09, + "loss": 0.5439, + "step": 7152 + }, + { + "epoch": 5.921866372170072, + "grad_norm": 0.9746168255805969, + "learning_rate": 1.9155549342378397e-09, + "loss": 0.5619, + "step": 7153 + }, + { + "epoch": 5.9226946438431805, + "grad_norm": 0.9795688986778259, + "learning_rate": 1.8727559333522883e-09, + "loss": 0.5576, + "step": 7154 + }, + { + "epoch": 5.92352291551629, + "grad_norm": 0.9887169599533081, + "learning_rate": 1.8304402968011282e-09, + "loss": 0.5292, + "step": 7155 + }, + { + "epoch": 5.924351187189398, + "grad_norm": 0.9666093587875366, + "learning_rate": 1.7886080327719767e-09, + "loss": 0.5393, + "step": 7156 + }, + { + "epoch": 5.925179458862507, + "grad_norm": 0.97262042760849, + "learning_rate": 1.7472591493594704e-09, + "loss": 0.5395, + "step": 7157 + }, + { + "epoch": 5.926007730535615, + "grad_norm": 0.9713893532752991, + "learning_rate": 1.7063936545635983e-09, + "loss": 0.5399, + "step": 7158 + }, + { + "epoch": 5.9268360022087245, + "grad_norm": 0.993696928024292, + "learning_rate": 1.6660115562919243e-09, + "loss": 0.5559, + "step": 7159 + }, + { + "epoch": 5.927664273881833, + "grad_norm": 0.9678178429603577, + "learning_rate": 1.6261128623579203e-09, + "loss": 0.5342, + "step": 7160 + }, + { + "epoch": 5.928492545554942, + "grad_norm": 0.9986034035682678, + "learning_rate": 1.5866975804812445e-09, + "loss": 0.5362, + "step": 7161 + }, + { + "epoch": 5.929320817228051, + "grad_norm": 0.9905020594596863, + "learning_rate": 1.5477657182888517e-09, + "loss": 0.5375, + "step": 7162 + }, + { + "epoch": 5.930149088901159, + "grad_norm": 1.0065239667892456, + "learning_rate": 1.5093172833136049e-09, + "loss": 0.5244, + "step": 7163 + }, + { + "epoch": 5.930977360574269, + "grad_norm": 0.9741962552070618, + "learning_rate": 1.4713522829948313e-09, + "loss": 0.5516, + "step": 7164 + }, + { + "epoch": 5.931805632247377, + "grad_norm": 0.9660084247589111, + "learning_rate": 1.4338707246783212e-09, + "loss": 0.5484, + "step": 7165 + }, + { + "epoch": 5.932633903920486, + "grad_norm": 0.9778664112091064, + "learning_rate": 1.3968726156163293e-09, + "loss": 0.5154, + "step": 7166 + }, + { + "epoch": 5.933462175593594, + "grad_norm": 0.9807467460632324, + "learning_rate": 1.3603579629681284e-09, + "loss": 0.5638, + "step": 7167 + }, + { + "epoch": 5.9342904472667035, + "grad_norm": 0.9695494771003723, + "learning_rate": 1.3243267737983455e-09, + "loss": 0.55, + "step": 7168 + }, + { + "epoch": 5.935118718939812, + "grad_norm": 0.9637518525123596, + "learning_rate": 1.2887790550791812e-09, + "loss": 0.5236, + "step": 7169 + }, + { + "epoch": 5.935946990612921, + "grad_norm": 0.9720240235328674, + "learning_rate": 1.253714813688467e-09, + "loss": 0.5448, + "step": 7170 + }, + { + "epoch": 5.93677526228603, + "grad_norm": 0.9662929773330688, + "learning_rate": 1.2191340564107756e-09, + "loss": 0.5475, + "step": 7171 + }, + { + "epoch": 5.937603533959138, + "grad_norm": 0.9746811389923096, + "learning_rate": 1.185036789937699e-09, + "loss": 0.5546, + "step": 7172 + }, + { + "epoch": 5.938431805632248, + "grad_norm": 0.9598021507263184, + "learning_rate": 1.151423020865905e-09, + "loss": 0.5412, + "step": 7173 + }, + { + "epoch": 5.939260077305356, + "grad_norm": 0.9442312717437744, + "learning_rate": 1.1182927557001898e-09, + "loss": 0.5475, + "step": 7174 + }, + { + "epoch": 5.940088348978465, + "grad_norm": 0.9720962047576904, + "learning_rate": 1.0856460008504265e-09, + "loss": 0.5358, + "step": 7175 + }, + { + "epoch": 5.940916620651573, + "grad_norm": 0.944737434387207, + "learning_rate": 1.0534827626335066e-09, + "loss": 0.5456, + "step": 7176 + }, + { + "epoch": 5.9417448923246825, + "grad_norm": 0.9579403400421143, + "learning_rate": 1.0218030472727848e-09, + "loss": 0.5381, + "step": 7177 + }, + { + "epoch": 5.942573163997791, + "grad_norm": 0.9504916667938232, + "learning_rate": 9.906068608978026e-10, + "loss": 0.5325, + "step": 7178 + }, + { + "epoch": 5.9434014356709, + "grad_norm": 1.000918984413147, + "learning_rate": 9.598942095451202e-10, + "loss": 0.5312, + "step": 7179 + }, + { + "epoch": 5.944229707344009, + "grad_norm": 0.9837747812271118, + "learning_rate": 9.296650991572065e-10, + "loss": 0.5334, + "step": 7180 + }, + { + "epoch": 5.945057979017117, + "grad_norm": 0.992518961429596, + "learning_rate": 8.999195355829937e-10, + "loss": 0.5487, + "step": 7181 + }, + { + "epoch": 5.945886250690227, + "grad_norm": 0.9739290475845337, + "learning_rate": 8.706575245781557e-10, + "loss": 0.5352, + "step": 7182 + }, + { + "epoch": 5.946714522363335, + "grad_norm": 0.9569925665855408, + "learning_rate": 8.418790718042747e-10, + "loss": 0.5467, + "step": 7183 + }, + { + "epoch": 5.947542794036444, + "grad_norm": 0.9718315005302429, + "learning_rate": 8.135841828302293e-10, + "loss": 0.5348, + "step": 7184 + }, + { + "epoch": 5.948371065709553, + "grad_norm": 0.9414976239204407, + "learning_rate": 7.857728631302519e-10, + "loss": 0.549, + "step": 7185 + }, + { + "epoch": 5.9491993373826615, + "grad_norm": 0.960256040096283, + "learning_rate": 7.58445118085871e-10, + "loss": 0.5645, + "step": 7186 + }, + { + "epoch": 5.950027609055771, + "grad_norm": 1.2102229595184326, + "learning_rate": 7.316009529848011e-10, + "loss": 0.5457, + "step": 7187 + }, + { + "epoch": 5.950855880728879, + "grad_norm": 0.9590608477592468, + "learning_rate": 7.05240373021221e-10, + "loss": 0.5431, + "step": 7188 + }, + { + "epoch": 5.951684152401988, + "grad_norm": 0.9723589420318604, + "learning_rate": 6.793633832952174e-10, + "loss": 0.522, + "step": 7189 + }, + { + "epoch": 5.952512424075096, + "grad_norm": 0.980858564376831, + "learning_rate": 6.539699888141738e-10, + "loss": 0.5187, + "step": 7190 + }, + { + "epoch": 5.953340695748206, + "grad_norm": 0.9443047642707825, + "learning_rate": 6.290601944913821e-10, + "loss": 0.5326, + "step": 7191 + }, + { + "epoch": 5.954168967421314, + "grad_norm": 0.9618434309959412, + "learning_rate": 6.046340051465982e-10, + "loss": 0.5583, + "step": 7192 + }, + { + "epoch": 5.954997239094423, + "grad_norm": 0.9642929434776306, + "learning_rate": 5.806914255060414e-10, + "loss": 0.5121, + "step": 7193 + }, + { + "epoch": 5.955825510767532, + "grad_norm": 0.9700872898101807, + "learning_rate": 5.572324602023948e-10, + "loss": 0.5423, + "step": 7194 + }, + { + "epoch": 5.9566537824406405, + "grad_norm": 0.992155909538269, + "learning_rate": 5.342571137748054e-10, + "loss": 0.5454, + "step": 7195 + }, + { + "epoch": 5.95748205411375, + "grad_norm": 0.9873687028884888, + "learning_rate": 5.11765390668606e-10, + "loss": 0.5327, + "step": 7196 + }, + { + "epoch": 5.958310325786858, + "grad_norm": 0.9675857424736023, + "learning_rate": 4.897572952358709e-10, + "loss": 0.5457, + "step": 7197 + }, + { + "epoch": 5.959138597459967, + "grad_norm": 0.9508792757987976, + "learning_rate": 4.682328317351381e-10, + "loss": 0.5548, + "step": 7198 + }, + { + "epoch": 5.959966869133075, + "grad_norm": 0.9641818404197693, + "learning_rate": 4.471920043308542e-10, + "loss": 0.5313, + "step": 7199 + }, + { + "epoch": 5.9607951408061846, + "grad_norm": 0.9656306505203247, + "learning_rate": 4.266348170944845e-10, + "loss": 0.5394, + "step": 7200 + }, + { + "epoch": 5.961623412479293, + "grad_norm": 0.9801570773124695, + "learning_rate": 4.0656127400340306e-10, + "loss": 0.5718, + "step": 7201 + }, + { + "epoch": 5.962451684152402, + "grad_norm": 0.9776313304901123, + "learning_rate": 3.869713789417251e-10, + "loss": 0.5473, + "step": 7202 + }, + { + "epoch": 5.963279955825511, + "grad_norm": 0.9702569246292114, + "learning_rate": 3.678651357003071e-10, + "loss": 0.5279, + "step": 7203 + }, + { + "epoch": 5.9641082274986195, + "grad_norm": 0.9656468033790588, + "learning_rate": 3.492425479756367e-10, + "loss": 0.5512, + "step": 7204 + }, + { + "epoch": 5.964936499171729, + "grad_norm": 0.9543492794036865, + "learning_rate": 3.311036193709427e-10, + "loss": 0.5584, + "step": 7205 + }, + { + "epoch": 5.965764770844837, + "grad_norm": 1.001214861869812, + "learning_rate": 3.134483533961952e-10, + "loss": 0.5515, + "step": 7206 + }, + { + "epoch": 5.966593042517946, + "grad_norm": 0.9717075824737549, + "learning_rate": 2.962767534672728e-10, + "loss": 0.5377, + "step": 7207 + }, + { + "epoch": 5.967421314191054, + "grad_norm": 0.9800747632980347, + "learning_rate": 2.795888229070731e-10, + "loss": 0.5096, + "step": 7208 + }, + { + "epoch": 5.9682495858641635, + "grad_norm": 0.9655010104179382, + "learning_rate": 2.6338456494412466e-10, + "loss": 0.5486, + "step": 7209 + }, + { + "epoch": 5.969077857537272, + "grad_norm": 0.9708304405212402, + "learning_rate": 2.476639827142524e-10, + "loss": 0.5468, + "step": 7210 + }, + { + "epoch": 5.969906129210381, + "grad_norm": 0.9659552574157715, + "learning_rate": 2.3242707925863474e-10, + "loss": 0.5404, + "step": 7211 + }, + { + "epoch": 5.97073440088349, + "grad_norm": 0.9564220905303955, + "learning_rate": 2.1767385752630156e-10, + "loss": 0.5533, + "step": 7212 + }, + { + "epoch": 5.9715626725565985, + "grad_norm": 0.9549928903579712, + "learning_rate": 2.0340432037108116e-10, + "loss": 0.548, + "step": 7213 + }, + { + "epoch": 5.972390944229708, + "grad_norm": 1.0049433708190918, + "learning_rate": 1.8961847055437577e-10, + "loss": 0.5451, + "step": 7214 + }, + { + "epoch": 5.973219215902816, + "grad_norm": 0.9905382990837097, + "learning_rate": 1.763163107434962e-10, + "loss": 0.5605, + "step": 7215 + }, + { + "epoch": 5.974047487575925, + "grad_norm": 0.9783766865730286, + "learning_rate": 1.6349784351249454e-10, + "loss": 0.5532, + "step": 7216 + }, + { + "epoch": 5.974875759249033, + "grad_norm": 1.049691081047058, + "learning_rate": 1.5116307134160902e-10, + "loss": 0.564, + "step": 7217 + }, + { + "epoch": 5.9757040309221425, + "grad_norm": 0.9859261512756348, + "learning_rate": 1.3931199661698647e-10, + "loss": 0.5468, + "step": 7218 + }, + { + "epoch": 5.976532302595251, + "grad_norm": 0.9845983386039734, + "learning_rate": 1.2794462163234765e-10, + "loss": 0.566, + "step": 7219 + }, + { + "epoch": 5.97736057426836, + "grad_norm": 0.9913472533226013, + "learning_rate": 1.1706094858676686e-10, + "loss": 0.5615, + "step": 7220 + }, + { + "epoch": 5.978188845941469, + "grad_norm": 0.9817662239074707, + "learning_rate": 1.066609795866147e-10, + "loss": 0.5409, + "step": 7221 + }, + { + "epoch": 5.979017117614577, + "grad_norm": 0.9485713243484497, + "learning_rate": 9.674471664361529e-11, + "loss": 0.5434, + "step": 7222 + }, + { + "epoch": 5.979845389287687, + "grad_norm": 0.9645923376083374, + "learning_rate": 8.731216167651157e-11, + "loss": 0.5369, + "step": 7223 + }, + { + "epoch": 5.980673660960795, + "grad_norm": 0.9520633816719055, + "learning_rate": 7.836331651078777e-11, + "loss": 0.5121, + "step": 7224 + }, + { + "epoch": 5.981501932633904, + "grad_norm": 0.9586499333381653, + "learning_rate": 6.989818287783667e-11, + "loss": 0.5591, + "step": 7225 + }, + { + "epoch": 5.982330204307012, + "grad_norm": 0.9719886183738708, + "learning_rate": 6.19167624155148e-11, + "loss": 0.529, + "step": 7226 + }, + { + "epoch": 5.9831584759801215, + "grad_norm": 0.9657121896743774, + "learning_rate": 5.4419056668419956e-11, + "loss": 0.5417, + "step": 7227 + }, + { + "epoch": 5.98398674765323, + "grad_norm": 0.9681416153907776, + "learning_rate": 4.740506708678094e-11, + "loss": 0.54, + "step": 7228 + }, + { + "epoch": 5.984815019326339, + "grad_norm": 0.9593567252159119, + "learning_rate": 4.087479502784541e-11, + "loss": 0.5399, + "step": 7229 + }, + { + "epoch": 5.985643290999448, + "grad_norm": 0.9805291295051575, + "learning_rate": 3.4828241755602286e-11, + "loss": 0.5371, + "step": 7230 + }, + { + "epoch": 5.986471562672556, + "grad_norm": 0.9544205665588379, + "learning_rate": 2.9265408439671516e-11, + "loss": 0.5221, + "step": 7231 + }, + { + "epoch": 5.987299834345666, + "grad_norm": 0.9567616581916809, + "learning_rate": 2.4186296156136768e-11, + "loss": 0.5485, + "step": 7232 + }, + { + "epoch": 5.988128106018774, + "grad_norm": 0.9742857217788696, + "learning_rate": 1.9590905888378086e-11, + "loss": 0.5295, + "step": 7233 + }, + { + "epoch": 5.988956377691883, + "grad_norm": 1.0573118925094604, + "learning_rate": 1.547923852540656e-11, + "loss": 0.5368, + "step": 7234 + }, + { + "epoch": 5.989784649364991, + "grad_norm": 0.9670606851577759, + "learning_rate": 1.1851294862419426e-11, + "loss": 0.5383, + "step": 7235 + }, + { + "epoch": 5.9906129210381005, + "grad_norm": 0.9531207084655762, + "learning_rate": 8.707075601632753e-12, + "loss": 0.5401, + "step": 7236 + }, + { + "epoch": 5.991441192711209, + "grad_norm": 0.949664294719696, + "learning_rate": 6.046581351171199e-12, + "loss": 0.5578, + "step": 7237 + }, + { + "epoch": 5.992269464384318, + "grad_norm": 0.9601742029190063, + "learning_rate": 3.869812626455804e-12, + "loss": 0.5321, + "step": 7238 + }, + { + "epoch": 5.993097736057427, + "grad_norm": 0.9694385528564453, + "learning_rate": 2.1767698479835398e-12, + "loss": 0.5274, + "step": 7239 + }, + { + "epoch": 5.993926007730535, + "grad_norm": 0.9623011350631714, + "learning_rate": 9.674533438253086e-13, + "loss": 0.5278, + "step": 7240 + }, + { + "epoch": 5.994754279403645, + "grad_norm": 0.9480747580528259, + "learning_rate": 2.4186334768305787e-13, + "loss": 0.5492, + "step": 7241 + }, + { + "epoch": 5.995582551076753, + "grad_norm": 0.9750016927719116, + "learning_rate": 0.0, + "loss": 0.5547, + "step": 7242 + } + ], + "logging_steps": 1, + "max_steps": 7242, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 1207, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 1.8285137635615703e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-7242/training_args.bin b/checkpoint-7242/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..7017eeafdf1f43b26668f0e0e5710ca52345a005 --- /dev/null +++ b/checkpoint-7242/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fdc878d2541c09353dd68f4f8a2464140b28fd84e1bc48b59914a6e84328e85e +size 8056 diff --git a/checkpoint-7242/zero_to_fp32.py b/checkpoint-7242/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-7242/zero_to_fp32.py @@ -0,0 +1,604 @@ +#!/usr/bin/env python + +# Copyright (c) Microsoft Corporation. +# SPDX-License-Identifier: Apache-2.0 + +# DeepSpeed Team + +# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets +# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in +# the future. Once extracted, the weights don't require DeepSpeed and can be used in any +# application. +# +# example: python zero_to_fp32.py . pytorch_model.bin + +import argparse +import torch +import glob +import math +import os +import re +from collections import OrderedDict +from dataclasses import dataclass + +# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with +# DeepSpeed data structures it has to be available in the current python environment. +from deepspeed.utils import logger +from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS, + FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES, + FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS) + + +@dataclass +class zero_model_state: + buffers: dict() + param_shapes: dict() + shared_params: list + ds_version: int + frozen_param_shapes: dict() + frozen_param_fragments: dict() + + +debug = 0 + +# load to cpu +device = torch.device('cpu') + + +def atoi(text): + return int(text) if text.isdigit() else text + + +def natural_keys(text): + ''' + alist.sort(key=natural_keys) sorts in human order + http://nedbatchelder.com/blog/200712/human_sorting.html + (See Toothy's implementation in the comments) + ''' + return [atoi(c) for c in re.split(r'(\d+)', text)] + + +def get_model_state_file(checkpoint_dir, zero_stage): + if not os.path.isdir(checkpoint_dir): + raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist") + + # there should be only one file + if zero_stage <= 2: + file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt") + elif zero_stage == 3: + file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt") + + if not os.path.exists(file): + raise FileNotFoundError(f"can't find model states file at '{file}'") + + return file + + +def get_checkpoint_files(checkpoint_dir, glob_pattern): + # XXX: need to test that this simple glob rule works for multi-node setup too + ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys) + + if len(ckpt_files) == 0: + raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'") + + return ckpt_files + + +def get_optim_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt") + + +def get_model_state_files(checkpoint_dir): + return get_checkpoint_files(checkpoint_dir, "*_model_states.pt") + + +def parse_model_states(files): + zero_model_states = [] + for file in files: + state_dict = torch.load(file, map_location=device) + + if BUFFER_NAMES not in state_dict: + raise ValueError(f"{file} is not a model state checkpoint") + buffer_names = state_dict[BUFFER_NAMES] + if debug: + print("Found buffers:", buffer_names) + + # recover just the buffers while restoring them to fp32 if they were saved in fp16 + buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names} + param_shapes = state_dict[PARAM_SHAPES] + + # collect parameters that are included in param_shapes + param_names = [] + for s in param_shapes: + for name in s.keys(): + param_names.append(name) + + # update with frozen parameters + frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None) + if frozen_param_shapes is not None: + if debug: + print(f"Found frozen_param_shapes: {frozen_param_shapes}") + param_names += list(frozen_param_shapes.keys()) + + # handle shared params + shared_params = [[k, v] for k, v in state_dict["shared_params"].items()] + + ds_version = state_dict.get(DS_VERSION, None) + + frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None) + + z_model_state = zero_model_state(buffers=buffers, + param_shapes=param_shapes, + shared_params=shared_params, + ds_version=ds_version, + frozen_param_shapes=frozen_param_shapes, + frozen_param_fragments=frozen_param_fragments) + zero_model_states.append(z_model_state) + + return zero_model_states + + +def parse_optim_states(files, ds_checkpoint_dir): + + total_files = len(files) + state_dicts = [] + for f in files: + state_dict = torch.load(f, map_location=device) + # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights + # and also handle the case where it was already removed by another helper script + state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None) + state_dicts.append(state_dict) + + if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]: + raise ValueError(f"{files[0]} is not a zero checkpoint") + zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE] + world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT] + + # For ZeRO-2 each param group can have different partition_count as data parallelism for expert + # parameters can be different from data parallelism for non-expert parameters. So we can just + # use the max of the partition_count to get the dp world_size. + + if type(world_size) is list: + world_size = max(world_size) + + if world_size != total_files: + raise ValueError( + f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. " + "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes." + ) + + # the groups are named differently in each stage + if zero_stage <= 2: + fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS + elif zero_stage == 3: + fp32_groups_key = FP32_FLAT_GROUPS + else: + raise ValueError(f"unknown zero stage {zero_stage}") + + if zero_stage <= 2: + fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))] + elif zero_stage == 3: + # if there is more than one param group, there will be multiple flattened tensors - one + # flattened tensor per group - for simplicity merge them into a single tensor + # + # XXX: could make the script more memory efficient for when there are multiple groups - it + # will require matching the sub-lists of param_shapes for each param group flattened tensor + + fp32_flat_groups = [ + torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts)) + ] + + return zero_stage, world_size, fp32_flat_groups + + +def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters): + """ + Returns fp32 state_dict reconstructed from ds checkpoint + + Args: + - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are) + + """ + print(f"Processing zero checkpoint '{ds_checkpoint_dir}'") + + optim_files = get_optim_files(ds_checkpoint_dir) + zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir) + print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}") + + model_files = get_model_state_files(ds_checkpoint_dir) + + zero_model_states = parse_model_states(model_files) + print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}') + + if zero_stage <= 2: + return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + elif zero_stage == 3: + return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters) + + +def _zero2_merge_frozen_params(state_dict, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + frozen_param_fragments = zero_model_states[0].frozen_param_fragments + + if debug: + num_elem = sum(s.numel() for s in frozen_param_shapes.values()) + print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in frozen_param_fragments.values()]) + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + state_dict[name] = frozen_param_fragments[name] + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _has_callable(obj, fn): + attr = getattr(obj, fn, None) + return callable(attr) + + +def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + + # Reconstruction protocol: + # + # XXX: document this + + if debug: + for i in range(world_size): + for j in range(len(fp32_flat_groups[0])): + print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}") + + # XXX: memory usage doubles here (zero2) + num_param_groups = len(fp32_flat_groups[0]) + merged_single_partition_of_fp32_groups = [] + for i in range(num_param_groups): + merged_partitions = [sd[i] for sd in fp32_flat_groups] + full_single_fp32_vector = torch.cat(merged_partitions, 0) + merged_single_partition_of_fp32_groups.append(full_single_fp32_vector) + avail_numel = sum( + [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups]) + + if debug: + wanted_params = sum([len(shapes) for shapes in param_shapes]) + wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes]) + # not asserting if there is a mismatch due to possible padding + print(f"Have {avail_numel} numels to process.") + print(f"Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + total_numel = 0 + total_params = 0 + for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups): + offset = 0 + avail_numel = full_single_fp32_vector.numel() + for name, shape in shapes.items(): + + unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape) + total_numel += unpartitioned_numel + total_params += 1 + + if debug: + print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ") + state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape) + offset += unpartitioned_numel + + # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and + # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex + # paddings performed in the code it's almost impossible to predict the exact numbers w/o the + # live optimizer object, so we are checking that the numbers are within the right range + align_to = 2 * world_size + + def zero2_align(x): + return align_to * math.ceil(x / align_to) + + if debug: + print(f"original offset={offset}, avail_numel={avail_numel}") + + offset = zero2_align(offset) + avail_numel = zero2_align(avail_numel) + + if debug: + print(f"aligned offset={offset}, avail_numel={avail_numel}") + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero2_merge_frozen_params(state_dict, zero_model_states) + + _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def zero3_partitioned_param_info(unpartitioned_numel, world_size): + remainder = unpartitioned_numel % world_size + padding_numel = (world_size - remainder) if remainder else 0 + partitioned_numel = math.ceil(unpartitioned_numel / world_size) + return partitioned_numel, padding_numel + + +def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states): + if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0: + return + + if debug: + for i in range(world_size): + num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values()) + print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}') + + frozen_param_shapes = zero_model_states[0].frozen_param_shapes + wanted_params = len(frozen_param_shapes) + wanted_numel = sum(s.numel() for s in frozen_param_shapes.values()) + avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size + print(f'Frozen params: Have {avail_numel} numels to process.') + print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params') + + total_params = 0 + total_numel = 0 + for name, shape in zero_model_states[0].frozen_param_shapes.items(): + total_params += 1 + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + + param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states) + state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape) + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements") + + +def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states): + param_shapes = zero_model_states[0].param_shapes + avail_numel = fp32_flat_groups[0].numel() * world_size + # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each + # param, re-consolidating each param, while dealing with padding if any + + # merge list of dicts, preserving order + param_shapes = {k: v for d in param_shapes for k, v in d.items()} + + if debug: + for i in range(world_size): + print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}") + + wanted_params = len(param_shapes) + wanted_numel = sum(shape.numel() for shape in param_shapes.values()) + # not asserting if there is a mismatch due to possible padding + avail_numel = fp32_flat_groups[0].numel() * world_size + print(f"Trainable params: Have {avail_numel} numels to process.") + print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.") + + # params + # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support + # out-of-core computing solution + offset = 0 + total_numel = 0 + total_params = 0 + for name, shape in param_shapes.items(): + + unpartitioned_numel = shape.numel() + total_numel += unpartitioned_numel + total_params += 1 + + partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size) + + if debug: + print( + f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}" + ) + + # XXX: memory usage doubles here + state_dict[name] = torch.cat( + tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)), + 0).narrow(0, 0, unpartitioned_numel).view(shape) + offset += partitioned_numel + + offset *= world_size + + # Sanity check + if offset != avail_numel: + raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong") + + print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements") + + +def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states, + exclude_frozen_parameters): + state_dict = OrderedDict() + + # buffers + buffers = zero_model_states[0].buffers + state_dict.update(buffers) + if debug: + print(f"added {len(buffers)} buffers") + + if not exclude_frozen_parameters: + _zero3_merge_frozen_params(state_dict, world_size, zero_model_states) + + _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states) + + # recover shared parameters + for pair in zero_model_states[0].shared_params: + if pair[1] in state_dict: + state_dict[pair[0]] = state_dict[pair[1]] + + return state_dict + + +def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with + ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example + via a model hub. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + + Returns: + - pytorch ``state_dict`` + + Note: this approach may not work if your application doesn't have sufficient free CPU memory and + you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with + the checkpoint. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint + # do the training and checkpoint saving + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu + model = model.cpu() # move to cpu + model.load_state_dict(state_dict) + # submit to model hub or save the model to share with others + + In this example the ``model`` will no longer be usable in the deepspeed context of the same + application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead. + + """ + if tag is None: + latest_path = os.path.join(checkpoint_dir, 'latest') + if os.path.isfile(latest_path): + with open(latest_path, 'r') as fd: + tag = fd.read().strip() + else: + raise ValueError(f"Unable to find 'latest' file at {latest_path}") + + ds_checkpoint_dir = os.path.join(checkpoint_dir, tag) + + if not os.path.isdir(ds_checkpoint_dir): + raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist") + + return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters) + + +def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False): + """ + Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be + loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed. + + Args: + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + - ``exclude_frozen_parameters``: exclude frozen parameters + """ + + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters) + print(f"Saving fp32 state dict to {output_file}") + torch.save(state_dict, output_file) + + +def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None): + """ + 1. Put the provided model to cpu + 2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` + 3. Load it into the provided model + + Args: + - ``model``: the model object to update + - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``) + - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14`` + + Returns: + - ``model`: modified model + + Make sure you have plenty of CPU memory available before you call this function. If you don't + have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it + conveniently placed for you in the checkpoint folder. + + A typical usage might be :: + + from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint + model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir) + # submit to model hub or save the model to share with others + + Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context + of the same application. i.e. you will need to re-initialize the deepspeed engine, since + ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it. + + """ + logger.info(f"Extracting fp32 weights") + state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag) + + logger.info(f"Overwriting model with fp32 weights") + model = model.cpu() + model.load_state_dict(state_dict, strict=False) + + return model + + +if __name__ == "__main__": + + parser = argparse.ArgumentParser() + parser.add_argument("checkpoint_dir", + type=str, + help="path to the desired checkpoint folder, e.g., path/checkpoint-12") + parser.add_argument( + "output_file", + type=str, + help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)") + parser.add_argument("-t", + "--tag", + type=str, + default=None, + help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1") + parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters") + parser.add_argument("-d", "--debug", action='store_true', help="enable debug") + args = parser.parse_args() + + debug = args.debug + + convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, + args.output_file, + tag=args.tag, + exclude_frozen_parameters=args.exclude_frozen_parameters) diff --git a/config.json b/config.json new file mode 100644 index 0000000000000000000000000000000000000000..bc1fab23ab76663db86132a9d92014a99265286d --- /dev/null +++ b/config.json @@ -0,0 +1,52 @@ +{ + "_attn_implementation_autoset": true, + "_name_or_path": "meta-llama/Llama-3.1-8B", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128001, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 4096, + "initializer_range": 0.02, + "intermediate_size": 14336, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 32, + "num_hidden_layers": 32, + "num_key_value_heads": 8, + "pretraining_tp": 1, + "quantization_config": { + "_load_in_4bit": true, + "_load_in_8bit": false, + "bnb_4bit_compute_dtype": "bfloat16", + "bnb_4bit_quant_storage": "bfloat16", + "bnb_4bit_quant_type": "nf4", + "bnb_4bit_use_double_quant": true, + "llm_int8_enable_fp32_cpu_offload": false, + "llm_int8_has_fp16_weight": false, + "llm_int8_skip_modules": null, + "llm_int8_threshold": 6.0, + "load_in_4bit": true, + "load_in_8bit": false, + "quant_method": "bitsandbytes" + }, + "rms_norm_eps": 1e-05, + "rope_scaling": { + "factor": 8.0, + "high_freq_factor": 4.0, + "low_freq_factor": 1.0, + "original_max_position_embeddings": 8192, + "rope_type": "llama3" + }, + "rope_theta": 500000.0, + "tie_word_embeddings": false, + "torch_dtype": "bfloat16", + "transformers_version": "4.49.0", + "use_cache": false, + "vocab_size": 128256 +} diff --git a/special_tokens_map.json b/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..e5b39b6305d89284b04934011c68dbb26bf588ca --- /dev/null +++ b/special_tokens_map.json @@ -0,0 +1,23 @@ +{ + "bos_token": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "eos_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + }, + "pad_token": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false + } +} diff --git a/tokenizer.json b/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/tokenizer_config.json b/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..81dd14db6632ad5b35b9d447732e37ac074873a5 --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,2063 @@ +{ + "added_tokens_decoder": { + "128000": { + "content": "<|begin_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128001": { + "content": "<|end_of_text|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128002": { + "content": "<|reserved_special_token_0|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128003": { + "content": "<|reserved_special_token_1|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128004": { + "content": "<|finetune_right_pad_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128005": { + "content": "<|reserved_special_token_2|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128006": { + "content": "<|start_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128007": { + "content": "<|end_header_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128008": { + "content": "<|eom_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128009": { + "content": "<|eot_id|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128010": { + "content": "<|python_tag|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128011": { + "content": "<|reserved_special_token_3|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128012": { + "content": "<|reserved_special_token_4|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128013": { + "content": "<|reserved_special_token_5|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128014": { + "content": "<|reserved_special_token_6|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128015": { + "content": "<|reserved_special_token_7|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128016": { + "content": "<|reserved_special_token_8|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128017": { + "content": "<|reserved_special_token_9|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128018": { + "content": "<|reserved_special_token_10|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128019": { + "content": "<|reserved_special_token_11|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128020": { + "content": "<|reserved_special_token_12|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128021": { + "content": "<|reserved_special_token_13|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128022": { + "content": "<|reserved_special_token_14|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128023": { + "content": "<|reserved_special_token_15|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128024": { + "content": "<|reserved_special_token_16|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128025": { + "content": "<|reserved_special_token_17|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128026": { + "content": "<|reserved_special_token_18|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128027": { + "content": "<|reserved_special_token_19|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128028": { + "content": "<|reserved_special_token_20|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128029": { + "content": "<|reserved_special_token_21|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128030": { + "content": "<|reserved_special_token_22|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128031": { + "content": "<|reserved_special_token_23|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128032": { + "content": "<|reserved_special_token_24|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128033": { + "content": "<|reserved_special_token_25|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128034": { + "content": "<|reserved_special_token_26|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128035": { + "content": "<|reserved_special_token_27|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128036": { + "content": "<|reserved_special_token_28|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128037": { + "content": "<|reserved_special_token_29|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128038": { + "content": "<|reserved_special_token_30|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128039": { + "content": "<|reserved_special_token_31|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128040": { + "content": "<|reserved_special_token_32|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128041": { + "content": "<|reserved_special_token_33|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128042": { + "content": "<|reserved_special_token_34|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128043": { + "content": "<|reserved_special_token_35|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128044": { + "content": "<|reserved_special_token_36|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128045": { + "content": "<|reserved_special_token_37|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128046": { + "content": "<|reserved_special_token_38|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128047": { + "content": "<|reserved_special_token_39|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128048": { + "content": "<|reserved_special_token_40|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128049": { + "content": "<|reserved_special_token_41|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128050": { + "content": "<|reserved_special_token_42|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128051": { + "content": "<|reserved_special_token_43|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128052": { + "content": "<|reserved_special_token_44|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128053": { + "content": "<|reserved_special_token_45|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128054": { + "content": "<|reserved_special_token_46|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128055": { + "content": "<|reserved_special_token_47|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128056": { + "content": "<|reserved_special_token_48|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128057": { + "content": "<|reserved_special_token_49|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128058": { + "content": "<|reserved_special_token_50|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128059": { + "content": "<|reserved_special_token_51|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128060": { + "content": "<|reserved_special_token_52|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128061": { + "content": "<|reserved_special_token_53|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128062": { + "content": "<|reserved_special_token_54|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128063": { + "content": "<|reserved_special_token_55|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128064": { + "content": "<|reserved_special_token_56|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128065": { + "content": "<|reserved_special_token_57|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128066": { + "content": "<|reserved_special_token_58|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128067": { + "content": "<|reserved_special_token_59|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128068": { + "content": "<|reserved_special_token_60|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128069": { + "content": "<|reserved_special_token_61|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128070": { + "content": "<|reserved_special_token_62|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128071": { + "content": "<|reserved_special_token_63|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128072": { + "content": "<|reserved_special_token_64|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128073": { + "content": "<|reserved_special_token_65|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128074": { + "content": "<|reserved_special_token_66|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128075": { + "content": "<|reserved_special_token_67|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128076": { + "content": "<|reserved_special_token_68|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128077": { + "content": "<|reserved_special_token_69|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128078": { + "content": "<|reserved_special_token_70|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128079": { + "content": "<|reserved_special_token_71|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128080": { + "content": "<|reserved_special_token_72|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128081": { + "content": "<|reserved_special_token_73|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128082": { + "content": "<|reserved_special_token_74|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128083": { + "content": "<|reserved_special_token_75|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128084": { + "content": "<|reserved_special_token_76|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128085": { + "content": "<|reserved_special_token_77|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128086": { + "content": "<|reserved_special_token_78|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128087": { + "content": "<|reserved_special_token_79|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128088": { + "content": "<|reserved_special_token_80|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128089": { + "content": "<|reserved_special_token_81|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128090": { + "content": "<|reserved_special_token_82|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128091": { + "content": "<|reserved_special_token_83|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128092": { + "content": "<|reserved_special_token_84|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128093": { + "content": "<|reserved_special_token_85|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128094": { + "content": "<|reserved_special_token_86|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128095": { + "content": "<|reserved_special_token_87|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128096": { + "content": "<|reserved_special_token_88|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128097": { + "content": "<|reserved_special_token_89|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128098": { + "content": "<|reserved_special_token_90|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128099": { + "content": "<|reserved_special_token_91|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128100": { + "content": "<|reserved_special_token_92|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128101": { + "content": "<|reserved_special_token_93|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128102": { + "content": "<|reserved_special_token_94|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128103": { + "content": "<|reserved_special_token_95|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128104": { + "content": "<|reserved_special_token_96|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128105": { + "content": "<|reserved_special_token_97|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128106": { + "content": "<|reserved_special_token_98|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128107": { + "content": "<|reserved_special_token_99|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128108": { + "content": "<|reserved_special_token_100|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128109": { + "content": "<|reserved_special_token_101|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128110": { + "content": "<|reserved_special_token_102|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128111": { + "content": "<|reserved_special_token_103|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128112": { + "content": "<|reserved_special_token_104|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128113": { + "content": "<|reserved_special_token_105|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128114": { + "content": "<|reserved_special_token_106|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128115": { + "content": "<|reserved_special_token_107|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128116": { + "content": "<|reserved_special_token_108|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128117": { + "content": "<|reserved_special_token_109|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128118": { + "content": "<|reserved_special_token_110|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128119": { + "content": "<|reserved_special_token_111|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128120": { + "content": "<|reserved_special_token_112|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128121": { + "content": "<|reserved_special_token_113|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128122": { + "content": "<|reserved_special_token_114|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128123": { + "content": "<|reserved_special_token_115|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128124": { + "content": "<|reserved_special_token_116|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128125": { + "content": "<|reserved_special_token_117|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128126": { + "content": "<|reserved_special_token_118|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128127": { + "content": "<|reserved_special_token_119|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128128": { + "content": "<|reserved_special_token_120|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128129": { + "content": "<|reserved_special_token_121|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128130": { + "content": "<|reserved_special_token_122|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128131": { + "content": "<|reserved_special_token_123|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128132": { + "content": "<|reserved_special_token_124|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128133": { + "content": "<|reserved_special_token_125|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128134": { + "content": "<|reserved_special_token_126|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128135": { + "content": "<|reserved_special_token_127|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128136": { + "content": "<|reserved_special_token_128|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128137": { + "content": "<|reserved_special_token_129|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128138": { + "content": "<|reserved_special_token_130|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128139": { + "content": "<|reserved_special_token_131|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128140": { + "content": "<|reserved_special_token_132|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128141": { + "content": "<|reserved_special_token_133|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128142": { + "content": "<|reserved_special_token_134|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128143": { + "content": "<|reserved_special_token_135|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128144": { + "content": "<|reserved_special_token_136|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128145": { + "content": "<|reserved_special_token_137|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128146": { + "content": "<|reserved_special_token_138|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128147": { + "content": "<|reserved_special_token_139|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128148": { + "content": "<|reserved_special_token_140|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128149": { + "content": "<|reserved_special_token_141|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128150": { + "content": "<|reserved_special_token_142|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128151": { + "content": "<|reserved_special_token_143|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128152": { + "content": "<|reserved_special_token_144|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128153": { + "content": "<|reserved_special_token_145|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128154": { + "content": "<|reserved_special_token_146|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128155": { + "content": "<|reserved_special_token_147|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128156": { + "content": "<|reserved_special_token_148|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128157": { + "content": "<|reserved_special_token_149|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128158": { + "content": "<|reserved_special_token_150|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128159": { + "content": "<|reserved_special_token_151|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128160": { + "content": "<|reserved_special_token_152|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128161": { + "content": "<|reserved_special_token_153|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128162": { + "content": "<|reserved_special_token_154|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128163": { + "content": "<|reserved_special_token_155|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128164": { + "content": "<|reserved_special_token_156|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128165": { + "content": "<|reserved_special_token_157|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128166": { + "content": "<|reserved_special_token_158|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128167": { + "content": "<|reserved_special_token_159|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128168": { + "content": "<|reserved_special_token_160|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128169": { + "content": "<|reserved_special_token_161|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128170": { + "content": "<|reserved_special_token_162|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128171": { + "content": "<|reserved_special_token_163|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128172": { + "content": "<|reserved_special_token_164|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128173": { + "content": "<|reserved_special_token_165|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128174": { + "content": "<|reserved_special_token_166|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128175": { + "content": "<|reserved_special_token_167|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128176": { + "content": "<|reserved_special_token_168|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128177": { + "content": "<|reserved_special_token_169|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128178": { + "content": "<|reserved_special_token_170|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128179": { + "content": "<|reserved_special_token_171|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128180": { + "content": "<|reserved_special_token_172|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128181": { + "content": "<|reserved_special_token_173|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128182": { + "content": "<|reserved_special_token_174|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128183": { + "content": "<|reserved_special_token_175|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128184": { + "content": "<|reserved_special_token_176|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128185": { + "content": "<|reserved_special_token_177|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128186": { + "content": "<|reserved_special_token_178|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128187": { + "content": "<|reserved_special_token_179|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128188": { + "content": "<|reserved_special_token_180|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128189": { + "content": "<|reserved_special_token_181|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128190": { + "content": "<|reserved_special_token_182|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128191": { + "content": "<|reserved_special_token_183|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128192": { + "content": "<|reserved_special_token_184|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128193": { + "content": "<|reserved_special_token_185|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128194": { + "content": "<|reserved_special_token_186|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128195": { + "content": "<|reserved_special_token_187|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128196": { + "content": "<|reserved_special_token_188|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128197": { + "content": "<|reserved_special_token_189|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128198": { + "content": "<|reserved_special_token_190|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128199": { + "content": "<|reserved_special_token_191|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128200": { + "content": "<|reserved_special_token_192|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128201": { + "content": "<|reserved_special_token_193|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128202": { + "content": "<|reserved_special_token_194|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128203": { + "content": "<|reserved_special_token_195|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128204": { + "content": "<|reserved_special_token_196|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128205": { + "content": "<|reserved_special_token_197|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128206": { + "content": "<|reserved_special_token_198|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128207": { + "content": "<|reserved_special_token_199|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128208": { + "content": "<|reserved_special_token_200|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128209": { + "content": "<|reserved_special_token_201|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128210": { + "content": "<|reserved_special_token_202|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128211": { + "content": "<|reserved_special_token_203|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128212": { + "content": "<|reserved_special_token_204|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128213": { + "content": "<|reserved_special_token_205|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128214": { + "content": "<|reserved_special_token_206|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128215": { + "content": "<|reserved_special_token_207|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128216": { + "content": "<|reserved_special_token_208|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128217": { + "content": "<|reserved_special_token_209|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128218": { + "content": "<|reserved_special_token_210|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128219": { + "content": "<|reserved_special_token_211|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128220": { + "content": "<|reserved_special_token_212|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128221": { + "content": "<|reserved_special_token_213|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128222": { + "content": "<|reserved_special_token_214|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128223": { + "content": "<|reserved_special_token_215|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128224": { + "content": "<|reserved_special_token_216|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128225": { + "content": "<|reserved_special_token_217|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128226": { + "content": "<|reserved_special_token_218|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128227": { + "content": "<|reserved_special_token_219|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128228": { + "content": "<|reserved_special_token_220|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128229": { + "content": "<|reserved_special_token_221|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128230": { + "content": "<|reserved_special_token_222|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128231": { + "content": "<|reserved_special_token_223|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128232": { + "content": "<|reserved_special_token_224|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128233": { + "content": "<|reserved_special_token_225|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128234": { + "content": "<|reserved_special_token_226|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128235": { + "content": "<|reserved_special_token_227|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128236": { + "content": "<|reserved_special_token_228|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128237": { + "content": "<|reserved_special_token_229|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128238": { + "content": "<|reserved_special_token_230|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128239": { + "content": "<|reserved_special_token_231|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128240": { + "content": "<|reserved_special_token_232|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128241": { + "content": "<|reserved_special_token_233|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128242": { + "content": "<|reserved_special_token_234|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128243": { + "content": "<|reserved_special_token_235|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128244": { + "content": "<|reserved_special_token_236|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128245": { + "content": "<|reserved_special_token_237|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128246": { + "content": "<|reserved_special_token_238|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128247": { + "content": "<|reserved_special_token_239|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128248": { + "content": "<|reserved_special_token_240|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128249": { + "content": "<|reserved_special_token_241|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128250": { + "content": "<|reserved_special_token_242|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128251": { + "content": "<|reserved_special_token_243|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128252": { + "content": "<|reserved_special_token_244|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128253": { + "content": "<|reserved_special_token_245|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128254": { + "content": "<|reserved_special_token_246|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + }, + "128255": { + "content": "<|reserved_special_token_247|>", + "lstrip": false, + "normalized": false, + "rstrip": false, + "single_word": false, + "special": true + } + }, + "bos_token": "<|begin_of_text|>", + "clean_up_tokenization_spaces": true, + "eos_token": "<|end_of_text|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +}