diff --git a/.gitattributes b/.gitattributes index a6344aac8c09253b3b630fb776ae94478aa0275b..979e2a607767e40a8e6c384853f88b024431cbcb 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-146/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-292/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-438/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-584/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-730/tokenizer.json filter=lfs diff=lfs merge=lfs -text +checkpoint-876/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..41e17b1939e65231fb88e96ecc374a0645c5231c --- /dev/null +++ b/README.md @@ -0,0 +1,146 @@ +--- +library_name: peft +license: llama3.3 +base_model: meta-llama/Llama-3.3-70B-Instruct +tags: +- generated_from_trainer +datasets: +- ugaoo/sample_dataset_10k_each +model-index: +- name: out/sample_dataset_10k_each + 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-3.3-70B-Instruct +model_type: AutoModelForCausalLM +tokenizer_type: AutoTokenizer +trust_remote_code: true + +load_in_8bit: false +load_in_4bit: true +strict: false + +datasets: + - path: ugaoo/sample_dataset_10k_each + type: alpaca +val_set_size: 0 +output_dir: ./out/sample_dataset_10k_each + +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: sample_dataset_10k_each_llama33 +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/sample_dataset_10k_each + +This model is a fine-tuned version of [meta-llama/Llama-3.3-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct) on the ugaoo/sample_dataset_10k_each 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 +- num_devices: 3 +- gradient_accumulation_steps: 3 +- total_train_batch_size: 36 +- total_eval_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 0.15.0 +- Transformers 4.49.0 +- Pytorch 2.5.1+cu124 +- Datasets 3.4.1 +- Tokenizers 0.21.1 \ No newline at end of file diff --git a/adapter_config.json b/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a5775df151cb9649596940e8a117cb92f810dbb7 --- /dev/null +++ b/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.3-70B-Instruct", + "bias": "none", + "corda_config": null, + "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", + "down_proj", + "k_proj", + "q_proj", + "o_proj", + "v_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "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..c12ae1d95b5e5d0e20dd71bde916917d11271cde --- /dev/null +++ b/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8c1233ae8b00040dd1b4d89ecad17fab0a716b076689a2c996ceda85f168b356 +size 10829849744 diff --git a/checkpoint-146/README.md b/checkpoint-146/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1b184114a0c28ed3e4c082c18486736dc818166d --- /dev/null +++ b/checkpoint-146/README.md @@ -0,0 +1,202 @@ +--- +base_model: meta-llama/Llama-3.3-70B-Instruct +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 0.15.0 \ No newline at end of file diff --git a/checkpoint-146/adapter_config.json b/checkpoint-146/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a5775df151cb9649596940e8a117cb92f810dbb7 --- /dev/null +++ b/checkpoint-146/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.3-70B-Instruct", + "bias": "none", + "corda_config": null, + "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", + "down_proj", + "k_proj", + "q_proj", + "o_proj", + "v_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-146/adapter_model.safetensors b/checkpoint-146/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..1a5a60765389348977c2c3a5ed4798f588d14b69 --- /dev/null +++ b/checkpoint-146/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3e975129a05c70b28f97a66942ccd15b469678706e0a5dcbd801fc6f09dde7f4 +size 10829849744 diff --git a/checkpoint-146/global_step146/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-146/global_step146/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..12f4314847fac46a3bb1a06fc0d0c29ee8dd8189 --- /dev/null +++ b/checkpoint-146/global_step146/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:472e5e5351d498e987689a9ad45b6cd47c112ae76f4706d76be96d1498693cf8 +size 21659418140 diff --git a/checkpoint-146/global_step146/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-146/global_step146/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..80bd401f4528be5e9552940c64d7f1874aca7645 --- /dev/null +++ b/checkpoint-146/global_step146/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0bb224871033f2531482b0d565919039fec4125a537310ac51be0d4e64ef94ed +size 21659457372 diff --git a/checkpoint-146/global_step146/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoint-146/global_step146/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..7a45c341f32abe64f853fae960ae130237c89ddd --- /dev/null +++ b/checkpoint-146/global_step146/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e081c91fbaf83b584cbf3db39fc19a791f77e30ced26c1c6f30559d4040cbf5d +size 21659417820 diff --git a/checkpoint-146/global_step146/mp_rank_00_model_states.pt b/checkpoint-146/global_step146/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..69f67c2b16a8fa5dc394d3d5028c85bc51580b46 --- /dev/null +++ b/checkpoint-146/global_step146/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2d5ac309b07f3489df8fde31ef2e4081303af8204c41557241c1cefb4cfaf97c +size 11918643933 diff --git a/checkpoint-146/latest b/checkpoint-146/latest new file mode 100644 index 0000000000000000000000000000000000000000..66939ec377f094e431c3da374d11b80784fa15fb --- /dev/null +++ b/checkpoint-146/latest @@ -0,0 +1 @@ +global_step146 \ No newline at end of file diff --git a/checkpoint-146/rng_state_0.pth b/checkpoint-146/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..fcbeaa19dca3fd1954279ab82647895bd81ee724 --- /dev/null +++ b/checkpoint-146/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:98e65ec16cb4d84c9586835ffd7630d8bf6b54862a1656f3498e874633d8f6bf +size 14768 diff --git a/checkpoint-146/rng_state_1.pth b/checkpoint-146/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..f751fda6166ef5f7eb49f60c2343a1d2ee29de54 --- /dev/null +++ b/checkpoint-146/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ec6e80f40d3296de89de8c1e10122ff725e8fd1868c39ad129c31a9972921ac6 +size 14768 diff --git a/checkpoint-146/rng_state_2.pth b/checkpoint-146/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..14fdabc1ef81f7c9c76fb60753924c1c4cf8673f --- /dev/null +++ b/checkpoint-146/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6c54085101a9a89df27265beeb56f000c6befade309e99e7b088ffba50cafc20 +size 14768 diff --git a/checkpoint-146/scheduler.pt b/checkpoint-146/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..2dcaa7e9a444ac01e7809438be001efc2b7cafcb --- /dev/null +++ b/checkpoint-146/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:de643fc0ceb5560bbc616c4f1b96fc1c1929048c34502f11986ce6878a1b2d8e +size 1064 diff --git a/checkpoint-146/special_tokens_map.json b/checkpoint-146/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-146/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": "<|eot_id|>", + "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-146/tokenizer.json b/checkpoint-146/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-146/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-146/tokenizer_config.json b/checkpoint-146/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ca91a2ef55f4239a7af81d7c9abb05f53621a07b --- /dev/null +++ b/checkpoint-146/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "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|>", + "chat_template": "{{- bos_token }}\n{%- if custom_tools is defined %}\n {%- set tools = custom_tools %}\n{%- endif %}\n{%- if not tools_in_user_message is defined %}\n {%- set tools_in_user_message = true %}\n{%- endif %}\n{%- if not date_string is defined %}\n {%- set date_string = \"26 Jul 2024\" %}\n{%- endif %}\n{%- if not tools is defined %}\n {%- set tools = none %}\n{%- endif %}\n\n{#- This block extracts the system message, so we can slot it into the right place. #}\n{%- if messages[0]['role'] == 'system' %}\n {%- set system_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n{%- else %}\n {%- set system_message = \"\" %}\n{%- endif %}\n\n{#- System message + builtin tools #}\n{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}\n{%- if builtin_tools is defined or tools is not none %}\n {{- \"Environment: ipython\\n\" }}\n{%- endif %}\n{%- if builtin_tools is defined %}\n {{- \"Tools: \" + builtin_tools | reject('equalto', 'code_interpreter') | join(\", \") + \"\\n\\n\"}}\n{%- endif %}\n{{- \"Cutting Knowledge Date: December 2023\\n\" }}\n{{- \"Today Date: \" + date_string + \"\\n\\n\" }}\n{%- if tools is not none and not tools_in_user_message %}\n {{- \"You have access to the following functions. To call a function, please respond with JSON for a function call.\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n{%- endif %}\n{{- system_message }}\n{{- \"<|eot_id|>\" }}\n\n{#- Custom tools are passed in a user message with some extra guidance #}\n{%- if tools_in_user_message and not tools is none %}\n {#- Extract the first user message so we can plug it in here #}\n {%- if messages | length != 0 %}\n {%- set first_user_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n {%- else %}\n {{- raise_exception(\"Cannot put tools in the first user message when there's no first user message!\") }}\n{%- endif %}\n {{- '<|start_header_id|>user<|end_header_id|>\\n\\n' -}}\n {{- \"Given the following functions, please respond with a JSON for a function call \" }}\n {{- \"with its proper arguments that best answers the given prompt.\\n\\n\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n {{- first_user_message + \"<|eot_id|>\"}}\n{%- endif %}\n\n{%- for message in messages %}\n {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}\n {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n'+ message['content'] | trim + '<|eot_id|>' }}\n {%- elif 'tool_calls' in message %}\n {%- if not message.tool_calls|length == 1 %}\n {{- raise_exception(\"This model only supports single tool-calls at once!\") }}\n {%- endif %}\n {%- set tool_call = message.tool_calls[0].function %}\n {%- if builtin_tools is defined and tool_call.name in builtin_tools %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- \"<|python_tag|>\" + tool_call.name + \".call(\" }}\n {%- for arg_name, arg_val in tool_call.arguments | items %}\n {{- arg_name + '=\"' + arg_val + '\"' }}\n {%- if not loop.last %}\n {{- \", \" }}\n {%- endif %}\n {%- endfor %}\n {{- \")\" }}\n {%- else %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- '{\"name\": \"' + tool_call.name + '\", ' }}\n {{- '\"parameters\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- \"}\" }}\n {%- endif %}\n {%- if builtin_tools is defined %}\n {#- This means we're in ipython mode #}\n {{- \"<|eom_id|>\" }}\n {%- else %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n {%- elif message.role == \"tool\" or message.role == \"ipython\" %}\n {{- \"<|start_header_id|>ipython<|end_header_id|>\\n\\n\" }}\n {%- if message.content is mapping or message.content is iterable %}\n {{- message.content | tojson }}\n {%- else %}\n {{- message.content }}\n {%- endif %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}\n{%- endif %}\n", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "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-146/trainer_state.json b/checkpoint-146/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..9ebcae2eb6cebef46d2da0aeaa031ac9beb50df0 --- /dev/null +++ b/checkpoint-146/trainer_state.json @@ -0,0 +1,1055 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.9954545454545455, + "eval_steps": 500, + "global_step": 146, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.006818181818181818, + "grad_norm": 1.8943578004837036, + "learning_rate": 5.0000000000000004e-08, + "loss": 0.8676, + "step": 1 + }, + { + "epoch": 0.013636363636363636, + "grad_norm": 7.775813102722168, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.8513, + "step": 2 + }, + { + "epoch": 0.020454545454545454, + "grad_norm": 1.768803596496582, + "learning_rate": 1.5000000000000002e-07, + "loss": 0.8339, + "step": 3 + }, + { + "epoch": 0.02727272727272727, + "grad_norm": 1.7032263278961182, + "learning_rate": 2.0000000000000002e-07, + "loss": 0.8348, + "step": 4 + }, + { + "epoch": 0.03409090909090909, + "grad_norm": 1.8244796991348267, + "learning_rate": 2.5000000000000004e-07, + "loss": 0.8687, + "step": 5 + }, + { + "epoch": 0.04090909090909091, + "grad_norm": 1.7780181169509888, + "learning_rate": 3.0000000000000004e-07, + "loss": 0.8113, + "step": 6 + }, + { + "epoch": 0.04772727272727273, + "grad_norm": 1.8362191915512085, + "learning_rate": 3.5000000000000004e-07, + "loss": 0.8463, + "step": 7 + }, + { + "epoch": 0.05454545454545454, + "grad_norm": 1.6640979051589966, + "learning_rate": 4.0000000000000003e-07, + "loss": 0.8336, + "step": 8 + }, + { + "epoch": 0.06136363636363636, + "grad_norm": 1.8013464212417603, + "learning_rate": 4.5000000000000003e-07, + "loss": 0.8572, + "step": 9 + }, + { + "epoch": 0.06818181818181818, + "grad_norm": 1.7237999439239502, + "learning_rate": 5.000000000000001e-07, + "loss": 0.8501, + "step": 10 + }, + { + "epoch": 0.075, + "grad_norm": 1.989346981048584, + "learning_rate": 5.5e-07, + "loss": 0.8615, + "step": 11 + }, + { + "epoch": 0.08181818181818182, + "grad_norm": 2.0312082767486572, + "learning_rate": 6.000000000000001e-07, + "loss": 0.858, + "step": 12 + }, + { + "epoch": 0.08863636363636364, + "grad_norm": 1.6921299695968628, + "learning_rate": 6.5e-07, + "loss": 0.8577, + "step": 13 + }, + { + "epoch": 0.09545454545454546, + "grad_norm": 2.4207160472869873, + "learning_rate": 7.000000000000001e-07, + "loss": 0.8564, + "step": 14 + }, + { + "epoch": 0.10227272727272728, + "grad_norm": 1.7572585344314575, + "learning_rate": 7.5e-07, + "loss": 0.8409, + "step": 15 + }, + { + "epoch": 0.10909090909090909, + "grad_norm": 1.6135950088500977, + "learning_rate": 8.000000000000001e-07, + "loss": 0.8086, + "step": 16 + }, + { + "epoch": 0.1159090909090909, + "grad_norm": 2.226036787033081, + "learning_rate": 8.500000000000001e-07, + "loss": 0.8106, + "step": 17 + }, + { + "epoch": 0.12272727272727273, + "grad_norm": 1.7164231538772583, + "learning_rate": 9.000000000000001e-07, + "loss": 0.8494, + "step": 18 + }, + { + "epoch": 0.12954545454545455, + "grad_norm": 1.6572000980377197, + "learning_rate": 9.500000000000001e-07, + "loss": 0.8023, + "step": 19 + }, + { + "epoch": 0.13636363636363635, + "grad_norm": 1.5672118663787842, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.787, + "step": 20 + }, + { + "epoch": 0.1431818181818182, + "grad_norm": 1.5080257654190063, + "learning_rate": 1.0500000000000001e-06, + "loss": 0.8105, + "step": 21 + }, + { + "epoch": 0.15, + "grad_norm": 1.484892725944519, + "learning_rate": 1.1e-06, + "loss": 0.8258, + "step": 22 + }, + { + "epoch": 0.15681818181818183, + "grad_norm": 1.4591134786605835, + "learning_rate": 1.1500000000000002e-06, + "loss": 0.8032, + "step": 23 + }, + { + "epoch": 0.16363636363636364, + "grad_norm": 1.5506278276443481, + "learning_rate": 1.2000000000000002e-06, + "loss": 0.8236, + "step": 24 + }, + { + "epoch": 0.17045454545454544, + "grad_norm": 1.639350414276123, + "learning_rate": 1.25e-06, + "loss": 0.8185, + "step": 25 + }, + { + "epoch": 0.17727272727272728, + "grad_norm": 1.592640995979309, + "learning_rate": 1.3e-06, + "loss": 0.7898, + "step": 26 + }, + { + "epoch": 0.18409090909090908, + "grad_norm": 1.5280641317367554, + "learning_rate": 1.3500000000000002e-06, + "loss": 0.7731, + "step": 27 + }, + { + "epoch": 0.19090909090909092, + "grad_norm": 1.4015443325042725, + "learning_rate": 1.4000000000000001e-06, + "loss": 0.7956, + "step": 28 + }, + { + "epoch": 0.19772727272727272, + "grad_norm": 1.4890056848526, + "learning_rate": 1.45e-06, + "loss": 0.7838, + "step": 29 + }, + { + "epoch": 0.20454545454545456, + "grad_norm": 2.4770750999450684, + "learning_rate": 1.5e-06, + "loss": 0.7766, + "step": 30 + }, + { + "epoch": 0.21136363636363636, + "grad_norm": 1.5863685607910156, + "learning_rate": 1.5500000000000002e-06, + "loss": 0.7353, + "step": 31 + }, + { + "epoch": 0.21818181818181817, + "grad_norm": 1.058114767074585, + "learning_rate": 1.6000000000000001e-06, + "loss": 0.7237, + "step": 32 + }, + { + "epoch": 0.225, + "grad_norm": 1.5869101285934448, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.7597, + "step": 33 + }, + { + "epoch": 0.2318181818181818, + "grad_norm": 1.0012413263320923, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.7388, + "step": 34 + }, + { + "epoch": 0.23863636363636365, + "grad_norm": 1.118056058883667, + "learning_rate": 1.75e-06, + "loss": 0.7786, + "step": 35 + }, + { + "epoch": 0.24545454545454545, + "grad_norm": 1.0648820400238037, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.7502, + "step": 36 + }, + { + "epoch": 0.25227272727272726, + "grad_norm": 0.9440382719039917, + "learning_rate": 1.85e-06, + "loss": 0.6942, + "step": 37 + }, + { + "epoch": 0.2590909090909091, + "grad_norm": 0.8300187587738037, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.7312, + "step": 38 + }, + { + "epoch": 0.26590909090909093, + "grad_norm": 0.8296191692352295, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.7461, + "step": 39 + }, + { + "epoch": 0.2727272727272727, + "grad_norm": 0.7797508239746094, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.7234, + "step": 40 + }, + { + "epoch": 0.27954545454545454, + "grad_norm": 0.7335019707679749, + "learning_rate": 2.05e-06, + "loss": 0.6927, + "step": 41 + }, + { + "epoch": 0.2863636363636364, + "grad_norm": 0.7062106728553772, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.6552, + "step": 42 + }, + { + "epoch": 0.29318181818181815, + "grad_norm": 1.2342087030410767, + "learning_rate": 2.15e-06, + "loss": 0.6959, + "step": 43 + }, + { + "epoch": 0.3, + "grad_norm": 0.8967597484588623, + "learning_rate": 2.2e-06, + "loss": 0.704, + "step": 44 + }, + { + "epoch": 0.3068181818181818, + "grad_norm": 0.7228404879570007, + "learning_rate": 2.25e-06, + "loss": 0.6777, + "step": 45 + }, + { + "epoch": 0.31363636363636366, + "grad_norm": 0.6731488704681396, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.6652, + "step": 46 + }, + { + "epoch": 0.32045454545454544, + "grad_norm": 0.71966952085495, + "learning_rate": 2.35e-06, + "loss": 0.7281, + "step": 47 + }, + { + "epoch": 0.32727272727272727, + "grad_norm": 0.7057356238365173, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.6476, + "step": 48 + }, + { + "epoch": 0.3340909090909091, + "grad_norm": 0.6346054077148438, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.6865, + "step": 49 + }, + { + "epoch": 0.3409090909090909, + "grad_norm": 0.6938223838806152, + "learning_rate": 2.5e-06, + "loss": 0.6808, + "step": 50 + }, + { + "epoch": 0.3477272727272727, + "grad_norm": 0.7425184845924377, + "learning_rate": 2.55e-06, + "loss": 0.7074, + "step": 51 + }, + { + "epoch": 0.35454545454545455, + "grad_norm": 0.6743818521499634, + "learning_rate": 2.6e-06, + "loss": 0.657, + "step": 52 + }, + { + "epoch": 0.3613636363636364, + "grad_norm": 0.6785942316055298, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.6768, + "step": 53 + }, + { + "epoch": 0.36818181818181817, + "grad_norm": 0.586974024772644, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.651, + "step": 54 + }, + { + "epoch": 0.375, + "grad_norm": 0.5727331042289734, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.669, + "step": 55 + }, + { + "epoch": 0.38181818181818183, + "grad_norm": 0.7333543300628662, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.6571, + "step": 56 + }, + { + "epoch": 0.3886363636363636, + "grad_norm": 0.5710961818695068, + "learning_rate": 2.85e-06, + "loss": 0.6475, + "step": 57 + }, + { + "epoch": 0.39545454545454545, + "grad_norm": 0.5614489912986755, + "learning_rate": 2.9e-06, + "loss": 0.6314, + "step": 58 + }, + { + "epoch": 0.4022727272727273, + "grad_norm": 0.5007341504096985, + "learning_rate": 2.95e-06, + "loss": 0.6346, + "step": 59 + }, + { + "epoch": 0.4090909090909091, + "grad_norm": 0.54221510887146, + "learning_rate": 3e-06, + "loss": 0.6715, + "step": 60 + }, + { + "epoch": 0.4159090909090909, + "grad_norm": 0.5177039504051208, + "learning_rate": 3.05e-06, + "loss": 0.6459, + "step": 61 + }, + { + "epoch": 0.42272727272727273, + "grad_norm": 0.4741189181804657, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.592, + "step": 62 + }, + { + "epoch": 0.42954545454545456, + "grad_norm": 0.5637328028678894, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.6459, + "step": 63 + }, + { + "epoch": 0.43636363636363634, + "grad_norm": 0.5478869676589966, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.6447, + "step": 64 + }, + { + "epoch": 0.4431818181818182, + "grad_norm": 0.5483130216598511, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.6319, + "step": 65 + }, + { + "epoch": 0.45, + "grad_norm": 0.478081613779068, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.6102, + "step": 66 + }, + { + "epoch": 0.45681818181818185, + "grad_norm": 0.4777645766735077, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.5986, + "step": 67 + }, + { + "epoch": 0.4636363636363636, + "grad_norm": 0.5291482210159302, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.6314, + "step": 68 + }, + { + "epoch": 0.47045454545454546, + "grad_norm": 0.5224051475524902, + "learning_rate": 3.45e-06, + "loss": 0.6278, + "step": 69 + }, + { + "epoch": 0.4772727272727273, + "grad_norm": 0.6474127173423767, + "learning_rate": 3.5e-06, + "loss": 0.6398, + "step": 70 + }, + { + "epoch": 0.48409090909090907, + "grad_norm": 0.5060178637504578, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.5936, + "step": 71 + }, + { + "epoch": 0.4909090909090909, + "grad_norm": 3.0769243240356445, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.5975, + "step": 72 + }, + { + "epoch": 0.49772727272727274, + "grad_norm": 0.5319749116897583, + "learning_rate": 3.65e-06, + "loss": 0.6119, + "step": 73 + }, + { + "epoch": 0.5045454545454545, + "grad_norm": 0.6745399236679077, + "learning_rate": 3.7e-06, + "loss": 0.6213, + "step": 74 + }, + { + "epoch": 0.5113636363636364, + "grad_norm": 0.5156731009483337, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.6262, + "step": 75 + }, + { + "epoch": 0.5181818181818182, + "grad_norm": 0.9138725399971008, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.6244, + "step": 76 + }, + { + "epoch": 0.525, + "grad_norm": 0.4362037777900696, + "learning_rate": 3.85e-06, + "loss": 0.5874, + "step": 77 + }, + { + "epoch": 0.5318181818181819, + "grad_norm": 0.4674088954925537, + "learning_rate": 3.900000000000001e-06, + "loss": 0.615, + "step": 78 + }, + { + "epoch": 0.5386363636363637, + "grad_norm": 0.4590819180011749, + "learning_rate": 3.95e-06, + "loss": 0.6003, + "step": 79 + }, + { + "epoch": 0.5454545454545454, + "grad_norm": 0.5145031809806824, + "learning_rate": 4.000000000000001e-06, + "loss": 0.6079, + "step": 80 + }, + { + "epoch": 0.5522727272727272, + "grad_norm": 0.46660348773002625, + "learning_rate": 4.05e-06, + "loss": 0.6174, + "step": 81 + }, + { + "epoch": 0.5590909090909091, + "grad_norm": 0.48053789138793945, + "learning_rate": 4.1e-06, + "loss": 0.6232, + "step": 82 + }, + { + "epoch": 0.5659090909090909, + "grad_norm": 0.540793776512146, + "learning_rate": 4.15e-06, + "loss": 0.619, + "step": 83 + }, + { + "epoch": 0.5727272727272728, + "grad_norm": 0.4925222396850586, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.5838, + "step": 84 + }, + { + "epoch": 0.5795454545454546, + "grad_norm": 0.44989290833473206, + "learning_rate": 4.25e-06, + "loss": 0.5972, + "step": 85 + }, + { + "epoch": 0.5863636363636363, + "grad_norm": 0.42691388726234436, + "learning_rate": 4.3e-06, + "loss": 0.6046, + "step": 86 + }, + { + "epoch": 0.5931818181818181, + "grad_norm": 0.4512398838996887, + "learning_rate": 4.350000000000001e-06, + "loss": 0.5828, + "step": 87 + }, + { + "epoch": 0.6, + "grad_norm": 0.44399499893188477, + "learning_rate": 4.4e-06, + "loss": 0.6143, + "step": 88 + }, + { + "epoch": 0.6068181818181818, + "grad_norm": 0.46058326959609985, + "learning_rate": 4.450000000000001e-06, + "loss": 0.6117, + "step": 89 + }, + { + "epoch": 0.6136363636363636, + "grad_norm": 0.8795785903930664, + "learning_rate": 4.5e-06, + "loss": 0.6, + "step": 90 + }, + { + "epoch": 0.6204545454545455, + "grad_norm": 0.5148798227310181, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.6041, + "step": 91 + }, + { + "epoch": 0.6272727272727273, + "grad_norm": 0.48829612135887146, + "learning_rate": 4.600000000000001e-06, + "loss": 0.5492, + "step": 92 + }, + { + "epoch": 0.634090909090909, + "grad_norm": 0.5211894512176514, + "learning_rate": 4.65e-06, + "loss": 0.6044, + "step": 93 + }, + { + "epoch": 0.6409090909090909, + "grad_norm": 0.44403275847435, + "learning_rate": 4.7e-06, + "loss": 0.5929, + "step": 94 + }, + { + "epoch": 0.6477272727272727, + "grad_norm": 0.6037693619728088, + "learning_rate": 4.75e-06, + "loss": 0.6132, + "step": 95 + }, + { + "epoch": 0.6545454545454545, + "grad_norm": 0.4381515085697174, + "learning_rate": 4.800000000000001e-06, + "loss": 0.5822, + "step": 96 + }, + { + "epoch": 0.6613636363636364, + "grad_norm": 0.4997427761554718, + "learning_rate": 4.85e-06, + "loss": 0.593, + "step": 97 + }, + { + "epoch": 0.6681818181818182, + "grad_norm": 0.440571665763855, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.6072, + "step": 98 + }, + { + "epoch": 0.675, + "grad_norm": 0.5020624995231628, + "learning_rate": 4.95e-06, + "loss": 0.5861, + "step": 99 + }, + { + "epoch": 0.6818181818181818, + "grad_norm": 0.4280160963535309, + "learning_rate": 5e-06, + "loss": 0.6163, + "step": 100 + }, + { + "epoch": 0.6886363636363636, + "grad_norm": 0.4584537148475647, + "learning_rate": 4.9999795126530275e-06, + "loss": 0.6177, + "step": 101 + }, + { + "epoch": 0.6954545454545454, + "grad_norm": 0.6835049390792847, + "learning_rate": 4.999918050947891e-06, + "loss": 0.579, + "step": 102 + }, + { + "epoch": 0.7022727272727273, + "grad_norm": 0.4551607072353363, + "learning_rate": 4.999815615891943e-06, + "loss": 0.5927, + "step": 103 + }, + { + "epoch": 0.7090909090909091, + "grad_norm": 0.5893972516059875, + "learning_rate": 4.9996722091640805e-06, + "loss": 0.5775, + "step": 104 + }, + { + "epoch": 0.7159090909090909, + "grad_norm": 0.4574092924594879, + "learning_rate": 4.9994878331147225e-06, + "loss": 0.5863, + "step": 105 + }, + { + "epoch": 0.7227272727272728, + "grad_norm": 0.5354658365249634, + "learning_rate": 4.99926249076577e-06, + "loss": 0.5453, + "step": 106 + }, + { + "epoch": 0.7295454545454545, + "grad_norm": 0.4347354471683502, + "learning_rate": 4.998996185810557e-06, + "loss": 0.5913, + "step": 107 + }, + { + "epoch": 0.7363636363636363, + "grad_norm": 0.4487966299057007, + "learning_rate": 4.998688922613788e-06, + "loss": 0.5749, + "step": 108 + }, + { + "epoch": 0.7431818181818182, + "grad_norm": 0.42477577924728394, + "learning_rate": 4.9983407062114695e-06, + "loss": 0.5761, + "step": 109 + }, + { + "epoch": 0.75, + "grad_norm": 0.4801446497440338, + "learning_rate": 4.9979515423108255e-06, + "loss": 0.5977, + "step": 110 + }, + { + "epoch": 0.7568181818181818, + "grad_norm": 0.4876883029937744, + "learning_rate": 4.997521437290205e-06, + "loss": 0.5849, + "step": 111 + }, + { + "epoch": 0.7636363636363637, + "grad_norm": 0.4328872263431549, + "learning_rate": 4.997050398198977e-06, + "loss": 0.5988, + "step": 112 + }, + { + "epoch": 0.7704545454545455, + "grad_norm": 0.4332719147205353, + "learning_rate": 4.996538432757414e-06, + "loss": 0.6059, + "step": 113 + }, + { + "epoch": 0.7772727272727272, + "grad_norm": 0.4455336630344391, + "learning_rate": 4.995985549356568e-06, + "loss": 0.5725, + "step": 114 + }, + { + "epoch": 0.7840909090909091, + "grad_norm": 0.4197766184806824, + "learning_rate": 4.995391757058129e-06, + "loss": 0.5715, + "step": 115 + }, + { + "epoch": 0.7909090909090909, + "grad_norm": 0.41945499181747437, + "learning_rate": 4.99475706559428e-06, + "loss": 0.5845, + "step": 116 + }, + { + "epoch": 0.7977272727272727, + "grad_norm": 0.42414823174476624, + "learning_rate": 4.994081485367537e-06, + "loss": 0.5595, + "step": 117 + }, + { + "epoch": 0.8045454545454546, + "grad_norm": 0.4056423008441925, + "learning_rate": 4.993365027450576e-06, + "loss": 0.582, + "step": 118 + }, + { + "epoch": 0.8113636363636364, + "grad_norm": 0.4554205536842346, + "learning_rate": 4.992607703586058e-06, + "loss": 0.5591, + "step": 119 + }, + { + "epoch": 0.8181818181818182, + "grad_norm": 0.4307089149951935, + "learning_rate": 4.991809526186424e-06, + "loss": 0.5927, + "step": 120 + }, + { + "epoch": 0.825, + "grad_norm": 0.48133528232574463, + "learning_rate": 4.990970508333707e-06, + "loss": 0.5703, + "step": 121 + }, + { + "epoch": 0.8318181818181818, + "grad_norm": 0.48582738637924194, + "learning_rate": 4.990090663779305e-06, + "loss": 0.5491, + "step": 122 + }, + { + "epoch": 0.8386363636363636, + "grad_norm": 0.44926196336746216, + "learning_rate": 4.9891700069437635e-06, + "loss": 0.5803, + "step": 123 + }, + { + "epoch": 0.8454545454545455, + "grad_norm": 0.4562165141105652, + "learning_rate": 4.988208552916535e-06, + "loss": 0.5945, + "step": 124 + }, + { + "epoch": 0.8522727272727273, + "grad_norm": 0.5885360836982727, + "learning_rate": 4.987206317455734e-06, + "loss": 0.5632, + "step": 125 + }, + { + "epoch": 0.8590909090909091, + "grad_norm": 0.4525550603866577, + "learning_rate": 4.986163316987877e-06, + "loss": 0.5221, + "step": 126 + }, + { + "epoch": 0.865909090909091, + "grad_norm": 0.4545478820800781, + "learning_rate": 4.985079568607613e-06, + "loss": 0.587, + "step": 127 + }, + { + "epoch": 0.8727272727272727, + "grad_norm": 0.4364396333694458, + "learning_rate": 4.983955090077445e-06, + "loss": 0.5626, + "step": 128 + }, + { + "epoch": 0.8795454545454545, + "grad_norm": 0.4354369640350342, + "learning_rate": 4.982789899827439e-06, + "loss": 0.5674, + "step": 129 + }, + { + "epoch": 0.8863636363636364, + "grad_norm": 0.4822773337364197, + "learning_rate": 4.9815840169549216e-06, + "loss": 0.5685, + "step": 130 + }, + { + "epoch": 0.8931818181818182, + "grad_norm": 0.44673025608062744, + "learning_rate": 4.980337461224164e-06, + "loss": 0.5733, + "step": 131 + }, + { + "epoch": 0.9, + "grad_norm": 0.6294501423835754, + "learning_rate": 4.979050253066064e-06, + "loss": 0.5396, + "step": 132 + }, + { + "epoch": 0.9068181818181819, + "grad_norm": 0.42890027165412903, + "learning_rate": 4.977722413577802e-06, + "loss": 0.5718, + "step": 133 + }, + { + "epoch": 0.9136363636363637, + "grad_norm": 0.470587819814682, + "learning_rate": 4.976353964522509e-06, + "loss": 0.5705, + "step": 134 + }, + { + "epoch": 0.9204545454545454, + "grad_norm": 0.4632768929004669, + "learning_rate": 4.974944928328894e-06, + "loss": 0.5394, + "step": 135 + }, + { + "epoch": 0.9272727272727272, + "grad_norm": 0.5304691195487976, + "learning_rate": 4.973495328090891e-06, + "loss": 0.5762, + "step": 136 + }, + { + "epoch": 0.9340909090909091, + "grad_norm": 0.42036354541778564, + "learning_rate": 4.972005187567267e-06, + "loss": 0.5649, + "step": 137 + }, + { + "epoch": 0.9409090909090909, + "grad_norm": 0.45189663767814636, + "learning_rate": 4.970474531181245e-06, + "loss": 0.5554, + "step": 138 + }, + { + "epoch": 0.9477272727272728, + "grad_norm": 0.47352468967437744, + "learning_rate": 4.968903384020095e-06, + "loss": 0.5511, + "step": 139 + }, + { + "epoch": 0.9545454545454546, + "grad_norm": 0.525702953338623, + "learning_rate": 4.967291771834727e-06, + "loss": 0.5777, + "step": 140 + }, + { + "epoch": 0.9613636363636363, + "grad_norm": 0.46146854758262634, + "learning_rate": 4.965639721039267e-06, + "loss": 0.5537, + "step": 141 + }, + { + "epoch": 0.9681818181818181, + "grad_norm": 0.5200941562652588, + "learning_rate": 4.963947258710626e-06, + "loss": 0.5522, + "step": 142 + }, + { + "epoch": 0.975, + "grad_norm": 0.5260375142097473, + "learning_rate": 4.962214412588053e-06, + "loss": 0.5747, + "step": 143 + }, + { + "epoch": 0.9818181818181818, + "grad_norm": 0.518645703792572, + "learning_rate": 4.960441211072686e-06, + "loss": 0.5257, + "step": 144 + }, + { + "epoch": 0.9886363636363636, + "grad_norm": 0.4528891444206238, + "learning_rate": 4.9586276832270785e-06, + "loss": 0.5548, + "step": 145 + }, + { + "epoch": 0.9954545454545455, + "grad_norm": 0.5137557983398438, + "learning_rate": 4.9567738587747314e-06, + "loss": 0.5526, + "step": 146 + } + ], + "logging_steps": 1, + "max_steps": 876, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 146, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 9.39244756481029e+18, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-146/training_args.bin b/checkpoint-146/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..b41c805059a36d8d2e3bf3511fb0fd95b74dae74 --- /dev/null +++ b/checkpoint-146/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68945b8f9934f5870e67a2d4a20c31fa8e742da41075fa7236fb313c6197456f +size 7928 diff --git a/checkpoint-146/zero_to_fp32.py b/checkpoint-146/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-146/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-292/README.md b/checkpoint-292/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1b184114a0c28ed3e4c082c18486736dc818166d --- /dev/null +++ b/checkpoint-292/README.md @@ -0,0 +1,202 @@ +--- +base_model: meta-llama/Llama-3.3-70B-Instruct +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 0.15.0 \ No newline at end of file diff --git a/checkpoint-292/adapter_config.json b/checkpoint-292/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a5775df151cb9649596940e8a117cb92f810dbb7 --- /dev/null +++ b/checkpoint-292/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.3-70B-Instruct", + "bias": "none", + "corda_config": null, + "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", + "down_proj", + "k_proj", + "q_proj", + "o_proj", + "v_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-292/adapter_model.safetensors b/checkpoint-292/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..3b5c968b75133545c2e9f41aa1fcac00b8ece517 --- /dev/null +++ b/checkpoint-292/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8e90e91d5950a19589af805f29f6a967276fb1492527e4ca68d6fd2ebaadc770 +size 10829849744 diff --git a/checkpoint-292/global_step292/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-292/global_step292/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..9740be96ae58efca0b2734968c14463504209f5e --- /dev/null +++ b/checkpoint-292/global_step292/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:615f29d177a3b3b040f57e1ffd2a46d81c7dc8a469760958521e6e2a728b8218 +size 21659418140 diff --git a/checkpoint-292/global_step292/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-292/global_step292/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..90705ba8d44fc468ce1070f5ffbc329f611e47ed --- /dev/null +++ b/checkpoint-292/global_step292/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f2343f817bb54039d9c935d19fa81737c8bb52d6bab5595d827f94ce6783454a +size 21659457372 diff --git a/checkpoint-292/global_step292/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoint-292/global_step292/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..9aee6a7d2fe9054f2081aa57d8175853d6b95bf2 --- /dev/null +++ b/checkpoint-292/global_step292/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:18a5e079d09c135d1fe54a2a86b299d2d6eea2f8ec233b1c9666fa9bd962b01c +size 21659417820 diff --git a/checkpoint-292/global_step292/mp_rank_00_model_states.pt b/checkpoint-292/global_step292/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..28a129ae0b6a0db72897b05e8ab714ba258246dc --- /dev/null +++ b/checkpoint-292/global_step292/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7ef16a6b9833d6d94033e03f3784b45c7c370e5a99d8f969f7b22aa4cbc99533 +size 11918643933 diff --git a/checkpoint-292/latest b/checkpoint-292/latest new file mode 100644 index 0000000000000000000000000000000000000000..9e32c73ff832ec049bfa2583d79e6a0c7683d549 --- /dev/null +++ b/checkpoint-292/latest @@ -0,0 +1 @@ +global_step292 \ No newline at end of file diff --git a/checkpoint-292/rng_state_0.pth b/checkpoint-292/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..85249ccced50bf01dbab5d2b37baafb348b77a19 --- /dev/null +++ b/checkpoint-292/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:09c6e347b836048b84b3508622b283e05d7f54dd715bcb9cebba2277b5ffcffe +size 14768 diff --git a/checkpoint-292/rng_state_1.pth b/checkpoint-292/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..d7a031ca7230d2a5820d21cc143ad1c0e25ea7b0 --- /dev/null +++ b/checkpoint-292/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:07f205ee249dcb07e0bb691a6ab486f4d0c4684caaaa1910964284504aa613ce +size 14768 diff --git a/checkpoint-292/rng_state_2.pth b/checkpoint-292/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..9bcf19bc3ecb7176448a02101dd2c2d1ace9a962 --- /dev/null +++ b/checkpoint-292/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:70203b767a0d006e03400748ac173224bdb6be8829213a6adf28f4e17bf939ec +size 14768 diff --git a/checkpoint-292/scheduler.pt b/checkpoint-292/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..73dd73a7071fe8362d98cced470c2710b1b4f7c5 --- /dev/null +++ b/checkpoint-292/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:27085d7815ec28e1624f97c5e7ce294c86aa22c9b72d42d96ab3087f6dab29fe +size 1064 diff --git a/checkpoint-292/special_tokens_map.json b/checkpoint-292/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-292/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": "<|eot_id|>", + "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-292/tokenizer.json b/checkpoint-292/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-292/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-292/tokenizer_config.json b/checkpoint-292/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ca91a2ef55f4239a7af81d7c9abb05f53621a07b --- /dev/null +++ b/checkpoint-292/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "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|>", + "chat_template": "{{- bos_token }}\n{%- if custom_tools is defined %}\n {%- set tools = custom_tools %}\n{%- endif %}\n{%- if not tools_in_user_message is defined %}\n {%- set tools_in_user_message = true %}\n{%- endif %}\n{%- if not date_string is defined %}\n {%- set date_string = \"26 Jul 2024\" %}\n{%- endif %}\n{%- if not tools is defined %}\n {%- set tools = none %}\n{%- endif %}\n\n{#- This block extracts the system message, so we can slot it into the right place. #}\n{%- if messages[0]['role'] == 'system' %}\n {%- set system_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n{%- else %}\n {%- set system_message = \"\" %}\n{%- endif %}\n\n{#- System message + builtin tools #}\n{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}\n{%- if builtin_tools is defined or tools is not none %}\n {{- \"Environment: ipython\\n\" }}\n{%- endif %}\n{%- if builtin_tools is defined %}\n {{- \"Tools: \" + builtin_tools | reject('equalto', 'code_interpreter') | join(\", \") + \"\\n\\n\"}}\n{%- endif %}\n{{- \"Cutting Knowledge Date: December 2023\\n\" }}\n{{- \"Today Date: \" + date_string + \"\\n\\n\" }}\n{%- if tools is not none and not tools_in_user_message %}\n {{- \"You have access to the following functions. To call a function, please respond with JSON for a function call.\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n{%- endif %}\n{{- system_message }}\n{{- \"<|eot_id|>\" }}\n\n{#- Custom tools are passed in a user message with some extra guidance #}\n{%- if tools_in_user_message and not tools is none %}\n {#- Extract the first user message so we can plug it in here #}\n {%- if messages | length != 0 %}\n {%- set first_user_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n {%- else %}\n {{- raise_exception(\"Cannot put tools in the first user message when there's no first user message!\") }}\n{%- endif %}\n {{- '<|start_header_id|>user<|end_header_id|>\\n\\n' -}}\n {{- \"Given the following functions, please respond with a JSON for a function call \" }}\n {{- \"with its proper arguments that best answers the given prompt.\\n\\n\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n {{- first_user_message + \"<|eot_id|>\"}}\n{%- endif %}\n\n{%- for message in messages %}\n {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}\n {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n'+ message['content'] | trim + '<|eot_id|>' }}\n {%- elif 'tool_calls' in message %}\n {%- if not message.tool_calls|length == 1 %}\n {{- raise_exception(\"This model only supports single tool-calls at once!\") }}\n {%- endif %}\n {%- set tool_call = message.tool_calls[0].function %}\n {%- if builtin_tools is defined and tool_call.name in builtin_tools %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- \"<|python_tag|>\" + tool_call.name + \".call(\" }}\n {%- for arg_name, arg_val in tool_call.arguments | items %}\n {{- arg_name + '=\"' + arg_val + '\"' }}\n {%- if not loop.last %}\n {{- \", \" }}\n {%- endif %}\n {%- endfor %}\n {{- \")\" }}\n {%- else %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- '{\"name\": \"' + tool_call.name + '\", ' }}\n {{- '\"parameters\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- \"}\" }}\n {%- endif %}\n {%- if builtin_tools is defined %}\n {#- This means we're in ipython mode #}\n {{- \"<|eom_id|>\" }}\n {%- else %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n {%- elif message.role == \"tool\" or message.role == \"ipython\" %}\n {{- \"<|start_header_id|>ipython<|end_header_id|>\\n\\n\" }}\n {%- if message.content is mapping or message.content is iterable %}\n {{- message.content | tojson }}\n {%- else %}\n {{- message.content }}\n {%- endif %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}\n{%- endif %}\n", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "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-292/trainer_state.json b/checkpoint-292/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..f8bdcb00ececa09130bbf1e28eed19d3a1f64037 --- /dev/null +++ b/checkpoint-292/trainer_state.json @@ -0,0 +1,2077 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 1.9954545454545456, + "eval_steps": 500, + "global_step": 292, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.006818181818181818, + "grad_norm": 1.8943578004837036, + "learning_rate": 5.0000000000000004e-08, + "loss": 0.8676, + "step": 1 + }, + { + "epoch": 0.013636363636363636, + "grad_norm": 7.775813102722168, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.8513, + "step": 2 + }, + { + "epoch": 0.020454545454545454, + "grad_norm": 1.768803596496582, + "learning_rate": 1.5000000000000002e-07, + "loss": 0.8339, + "step": 3 + }, + { + "epoch": 0.02727272727272727, + "grad_norm": 1.7032263278961182, + "learning_rate": 2.0000000000000002e-07, + "loss": 0.8348, + "step": 4 + }, + { + "epoch": 0.03409090909090909, + "grad_norm": 1.8244796991348267, + "learning_rate": 2.5000000000000004e-07, + "loss": 0.8687, + "step": 5 + }, + { + "epoch": 0.04090909090909091, + "grad_norm": 1.7780181169509888, + "learning_rate": 3.0000000000000004e-07, + "loss": 0.8113, + "step": 6 + }, + { + "epoch": 0.04772727272727273, + "grad_norm": 1.8362191915512085, + "learning_rate": 3.5000000000000004e-07, + "loss": 0.8463, + "step": 7 + }, + { + "epoch": 0.05454545454545454, + "grad_norm": 1.6640979051589966, + "learning_rate": 4.0000000000000003e-07, + "loss": 0.8336, + "step": 8 + }, + { + "epoch": 0.06136363636363636, + "grad_norm": 1.8013464212417603, + "learning_rate": 4.5000000000000003e-07, + "loss": 0.8572, + "step": 9 + }, + { + "epoch": 0.06818181818181818, + "grad_norm": 1.7237999439239502, + "learning_rate": 5.000000000000001e-07, + "loss": 0.8501, + "step": 10 + }, + { + "epoch": 0.075, + "grad_norm": 1.989346981048584, + "learning_rate": 5.5e-07, + "loss": 0.8615, + "step": 11 + }, + { + "epoch": 0.08181818181818182, + "grad_norm": 2.0312082767486572, + "learning_rate": 6.000000000000001e-07, + "loss": 0.858, + "step": 12 + }, + { + "epoch": 0.08863636363636364, + "grad_norm": 1.6921299695968628, + "learning_rate": 6.5e-07, + "loss": 0.8577, + "step": 13 + }, + { + "epoch": 0.09545454545454546, + "grad_norm": 2.4207160472869873, + "learning_rate": 7.000000000000001e-07, + "loss": 0.8564, + "step": 14 + }, + { + "epoch": 0.10227272727272728, + "grad_norm": 1.7572585344314575, + "learning_rate": 7.5e-07, + "loss": 0.8409, + "step": 15 + }, + { + "epoch": 0.10909090909090909, + "grad_norm": 1.6135950088500977, + "learning_rate": 8.000000000000001e-07, + "loss": 0.8086, + "step": 16 + }, + { + "epoch": 0.1159090909090909, + "grad_norm": 2.226036787033081, + "learning_rate": 8.500000000000001e-07, + "loss": 0.8106, + "step": 17 + }, + { + "epoch": 0.12272727272727273, + "grad_norm": 1.7164231538772583, + "learning_rate": 9.000000000000001e-07, + "loss": 0.8494, + "step": 18 + }, + { + "epoch": 0.12954545454545455, + "grad_norm": 1.6572000980377197, + "learning_rate": 9.500000000000001e-07, + "loss": 0.8023, + "step": 19 + }, + { + "epoch": 0.13636363636363635, + "grad_norm": 1.5672118663787842, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.787, + "step": 20 + }, + { + "epoch": 0.1431818181818182, + "grad_norm": 1.5080257654190063, + "learning_rate": 1.0500000000000001e-06, + "loss": 0.8105, + "step": 21 + }, + { + "epoch": 0.15, + "grad_norm": 1.484892725944519, + "learning_rate": 1.1e-06, + "loss": 0.8258, + "step": 22 + }, + { + "epoch": 0.15681818181818183, + "grad_norm": 1.4591134786605835, + "learning_rate": 1.1500000000000002e-06, + "loss": 0.8032, + "step": 23 + }, + { + "epoch": 0.16363636363636364, + "grad_norm": 1.5506278276443481, + "learning_rate": 1.2000000000000002e-06, + "loss": 0.8236, + "step": 24 + }, + { + "epoch": 0.17045454545454544, + "grad_norm": 1.639350414276123, + "learning_rate": 1.25e-06, + "loss": 0.8185, + "step": 25 + }, + { + "epoch": 0.17727272727272728, + "grad_norm": 1.592640995979309, + "learning_rate": 1.3e-06, + "loss": 0.7898, + "step": 26 + }, + { + "epoch": 0.18409090909090908, + "grad_norm": 1.5280641317367554, + "learning_rate": 1.3500000000000002e-06, + "loss": 0.7731, + "step": 27 + }, + { + "epoch": 0.19090909090909092, + "grad_norm": 1.4015443325042725, + "learning_rate": 1.4000000000000001e-06, + "loss": 0.7956, + "step": 28 + }, + { + "epoch": 0.19772727272727272, + "grad_norm": 1.4890056848526, + "learning_rate": 1.45e-06, + "loss": 0.7838, + "step": 29 + }, + { + "epoch": 0.20454545454545456, + "grad_norm": 2.4770750999450684, + "learning_rate": 1.5e-06, + "loss": 0.7766, + "step": 30 + }, + { + "epoch": 0.21136363636363636, + "grad_norm": 1.5863685607910156, + "learning_rate": 1.5500000000000002e-06, + "loss": 0.7353, + "step": 31 + }, + { + "epoch": 0.21818181818181817, + "grad_norm": 1.058114767074585, + "learning_rate": 1.6000000000000001e-06, + "loss": 0.7237, + "step": 32 + }, + { + "epoch": 0.225, + "grad_norm": 1.5869101285934448, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.7597, + "step": 33 + }, + { + "epoch": 0.2318181818181818, + "grad_norm": 1.0012413263320923, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.7388, + "step": 34 + }, + { + "epoch": 0.23863636363636365, + "grad_norm": 1.118056058883667, + "learning_rate": 1.75e-06, + "loss": 0.7786, + "step": 35 + }, + { + "epoch": 0.24545454545454545, + "grad_norm": 1.0648820400238037, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.7502, + "step": 36 + }, + { + "epoch": 0.25227272727272726, + "grad_norm": 0.9440382719039917, + "learning_rate": 1.85e-06, + "loss": 0.6942, + "step": 37 + }, + { + "epoch": 0.2590909090909091, + "grad_norm": 0.8300187587738037, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.7312, + "step": 38 + }, + { + "epoch": 0.26590909090909093, + "grad_norm": 0.8296191692352295, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.7461, + "step": 39 + }, + { + "epoch": 0.2727272727272727, + "grad_norm": 0.7797508239746094, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.7234, + "step": 40 + }, + { + "epoch": 0.27954545454545454, + "grad_norm": 0.7335019707679749, + "learning_rate": 2.05e-06, + "loss": 0.6927, + "step": 41 + }, + { + "epoch": 0.2863636363636364, + "grad_norm": 0.7062106728553772, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.6552, + "step": 42 + }, + { + "epoch": 0.29318181818181815, + "grad_norm": 1.2342087030410767, + "learning_rate": 2.15e-06, + "loss": 0.6959, + "step": 43 + }, + { + "epoch": 0.3, + "grad_norm": 0.8967597484588623, + "learning_rate": 2.2e-06, + "loss": 0.704, + "step": 44 + }, + { + "epoch": 0.3068181818181818, + "grad_norm": 0.7228404879570007, + "learning_rate": 2.25e-06, + "loss": 0.6777, + "step": 45 + }, + { + "epoch": 0.31363636363636366, + "grad_norm": 0.6731488704681396, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.6652, + "step": 46 + }, + { + "epoch": 0.32045454545454544, + "grad_norm": 0.71966952085495, + "learning_rate": 2.35e-06, + "loss": 0.7281, + "step": 47 + }, + { + "epoch": 0.32727272727272727, + "grad_norm": 0.7057356238365173, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.6476, + "step": 48 + }, + { + "epoch": 0.3340909090909091, + "grad_norm": 0.6346054077148438, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.6865, + "step": 49 + }, + { + "epoch": 0.3409090909090909, + "grad_norm": 0.6938223838806152, + "learning_rate": 2.5e-06, + "loss": 0.6808, + "step": 50 + }, + { + "epoch": 0.3477272727272727, + "grad_norm": 0.7425184845924377, + "learning_rate": 2.55e-06, + "loss": 0.7074, + "step": 51 + }, + { + "epoch": 0.35454545454545455, + "grad_norm": 0.6743818521499634, + "learning_rate": 2.6e-06, + "loss": 0.657, + "step": 52 + }, + { + "epoch": 0.3613636363636364, + "grad_norm": 0.6785942316055298, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.6768, + "step": 53 + }, + { + "epoch": 0.36818181818181817, + "grad_norm": 0.586974024772644, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.651, + "step": 54 + }, + { + "epoch": 0.375, + "grad_norm": 0.5727331042289734, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.669, + "step": 55 + }, + { + "epoch": 0.38181818181818183, + "grad_norm": 0.7333543300628662, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.6571, + "step": 56 + }, + { + "epoch": 0.3886363636363636, + "grad_norm": 0.5710961818695068, + "learning_rate": 2.85e-06, + "loss": 0.6475, + "step": 57 + }, + { + "epoch": 0.39545454545454545, + "grad_norm": 0.5614489912986755, + "learning_rate": 2.9e-06, + "loss": 0.6314, + "step": 58 + }, + { + "epoch": 0.4022727272727273, + "grad_norm": 0.5007341504096985, + "learning_rate": 2.95e-06, + "loss": 0.6346, + "step": 59 + }, + { + "epoch": 0.4090909090909091, + "grad_norm": 0.54221510887146, + "learning_rate": 3e-06, + "loss": 0.6715, + "step": 60 + }, + { + "epoch": 0.4159090909090909, + "grad_norm": 0.5177039504051208, + "learning_rate": 3.05e-06, + "loss": 0.6459, + "step": 61 + }, + { + "epoch": 0.42272727272727273, + "grad_norm": 0.4741189181804657, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.592, + "step": 62 + }, + { + "epoch": 0.42954545454545456, + "grad_norm": 0.5637328028678894, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.6459, + "step": 63 + }, + { + "epoch": 0.43636363636363634, + "grad_norm": 0.5478869676589966, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.6447, + "step": 64 + }, + { + "epoch": 0.4431818181818182, + "grad_norm": 0.5483130216598511, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.6319, + "step": 65 + }, + { + "epoch": 0.45, + "grad_norm": 0.478081613779068, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.6102, + "step": 66 + }, + { + "epoch": 0.45681818181818185, + "grad_norm": 0.4777645766735077, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.5986, + "step": 67 + }, + { + "epoch": 0.4636363636363636, + "grad_norm": 0.5291482210159302, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.6314, + "step": 68 + }, + { + "epoch": 0.47045454545454546, + "grad_norm": 0.5224051475524902, + "learning_rate": 3.45e-06, + "loss": 0.6278, + "step": 69 + }, + { + "epoch": 0.4772727272727273, + "grad_norm": 0.6474127173423767, + "learning_rate": 3.5e-06, + "loss": 0.6398, + "step": 70 + }, + { + "epoch": 0.48409090909090907, + "grad_norm": 0.5060178637504578, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.5936, + "step": 71 + }, + { + "epoch": 0.4909090909090909, + "grad_norm": 3.0769243240356445, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.5975, + "step": 72 + }, + { + "epoch": 0.49772727272727274, + "grad_norm": 0.5319749116897583, + "learning_rate": 3.65e-06, + "loss": 0.6119, + "step": 73 + }, + { + "epoch": 0.5045454545454545, + "grad_norm": 0.6745399236679077, + "learning_rate": 3.7e-06, + "loss": 0.6213, + "step": 74 + }, + { + "epoch": 0.5113636363636364, + "grad_norm": 0.5156731009483337, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.6262, + "step": 75 + }, + { + "epoch": 0.5181818181818182, + "grad_norm": 0.9138725399971008, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.6244, + "step": 76 + }, + { + "epoch": 0.525, + "grad_norm": 0.4362037777900696, + "learning_rate": 3.85e-06, + "loss": 0.5874, + "step": 77 + }, + { + "epoch": 0.5318181818181819, + "grad_norm": 0.4674088954925537, + "learning_rate": 3.900000000000001e-06, + "loss": 0.615, + "step": 78 + }, + { + "epoch": 0.5386363636363637, + "grad_norm": 0.4590819180011749, + "learning_rate": 3.95e-06, + "loss": 0.6003, + "step": 79 + }, + { + "epoch": 0.5454545454545454, + "grad_norm": 0.5145031809806824, + "learning_rate": 4.000000000000001e-06, + "loss": 0.6079, + "step": 80 + }, + { + "epoch": 0.5522727272727272, + "grad_norm": 0.46660348773002625, + "learning_rate": 4.05e-06, + "loss": 0.6174, + "step": 81 + }, + { + "epoch": 0.5590909090909091, + "grad_norm": 0.48053789138793945, + "learning_rate": 4.1e-06, + "loss": 0.6232, + "step": 82 + }, + { + "epoch": 0.5659090909090909, + "grad_norm": 0.540793776512146, + "learning_rate": 4.15e-06, + "loss": 0.619, + "step": 83 + }, + { + "epoch": 0.5727272727272728, + "grad_norm": 0.4925222396850586, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.5838, + "step": 84 + }, + { + "epoch": 0.5795454545454546, + "grad_norm": 0.44989290833473206, + "learning_rate": 4.25e-06, + "loss": 0.5972, + "step": 85 + }, + { + "epoch": 0.5863636363636363, + "grad_norm": 0.42691388726234436, + "learning_rate": 4.3e-06, + "loss": 0.6046, + "step": 86 + }, + { + "epoch": 0.5931818181818181, + "grad_norm": 0.4512398838996887, + "learning_rate": 4.350000000000001e-06, + "loss": 0.5828, + "step": 87 + }, + { + "epoch": 0.6, + "grad_norm": 0.44399499893188477, + "learning_rate": 4.4e-06, + "loss": 0.6143, + "step": 88 + }, + { + "epoch": 0.6068181818181818, + "grad_norm": 0.46058326959609985, + "learning_rate": 4.450000000000001e-06, + "loss": 0.6117, + "step": 89 + }, + { + "epoch": 0.6136363636363636, + "grad_norm": 0.8795785903930664, + "learning_rate": 4.5e-06, + "loss": 0.6, + "step": 90 + }, + { + "epoch": 0.6204545454545455, + "grad_norm": 0.5148798227310181, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.6041, + "step": 91 + }, + { + "epoch": 0.6272727272727273, + "grad_norm": 0.48829612135887146, + "learning_rate": 4.600000000000001e-06, + "loss": 0.5492, + "step": 92 + }, + { + "epoch": 0.634090909090909, + "grad_norm": 0.5211894512176514, + "learning_rate": 4.65e-06, + "loss": 0.6044, + "step": 93 + }, + { + "epoch": 0.6409090909090909, + "grad_norm": 0.44403275847435, + "learning_rate": 4.7e-06, + "loss": 0.5929, + "step": 94 + }, + { + "epoch": 0.6477272727272727, + "grad_norm": 0.6037693619728088, + "learning_rate": 4.75e-06, + "loss": 0.6132, + "step": 95 + }, + { + "epoch": 0.6545454545454545, + "grad_norm": 0.4381515085697174, + "learning_rate": 4.800000000000001e-06, + "loss": 0.5822, + "step": 96 + }, + { + "epoch": 0.6613636363636364, + "grad_norm": 0.4997427761554718, + "learning_rate": 4.85e-06, + "loss": 0.593, + "step": 97 + }, + { + "epoch": 0.6681818181818182, + "grad_norm": 0.440571665763855, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.6072, + "step": 98 + }, + { + "epoch": 0.675, + "grad_norm": 0.5020624995231628, + "learning_rate": 4.95e-06, + "loss": 0.5861, + "step": 99 + }, + { + "epoch": 0.6818181818181818, + "grad_norm": 0.4280160963535309, + "learning_rate": 5e-06, + "loss": 0.6163, + "step": 100 + }, + { + "epoch": 0.6886363636363636, + "grad_norm": 0.4584537148475647, + "learning_rate": 4.9999795126530275e-06, + "loss": 0.6177, + "step": 101 + }, + { + "epoch": 0.6954545454545454, + "grad_norm": 0.6835049390792847, + "learning_rate": 4.999918050947891e-06, + "loss": 0.579, + "step": 102 + }, + { + "epoch": 0.7022727272727273, + "grad_norm": 0.4551607072353363, + "learning_rate": 4.999815615891943e-06, + "loss": 0.5927, + "step": 103 + }, + { + "epoch": 0.7090909090909091, + "grad_norm": 0.5893972516059875, + "learning_rate": 4.9996722091640805e-06, + "loss": 0.5775, + "step": 104 + }, + { + "epoch": 0.7159090909090909, + "grad_norm": 0.4574092924594879, + "learning_rate": 4.9994878331147225e-06, + "loss": 0.5863, + "step": 105 + }, + { + "epoch": 0.7227272727272728, + "grad_norm": 0.5354658365249634, + "learning_rate": 4.99926249076577e-06, + "loss": 0.5453, + "step": 106 + }, + { + "epoch": 0.7295454545454545, + "grad_norm": 0.4347354471683502, + "learning_rate": 4.998996185810557e-06, + "loss": 0.5913, + "step": 107 + }, + { + "epoch": 0.7363636363636363, + "grad_norm": 0.4487966299057007, + "learning_rate": 4.998688922613788e-06, + "loss": 0.5749, + "step": 108 + }, + { + "epoch": 0.7431818181818182, + "grad_norm": 0.42477577924728394, + "learning_rate": 4.9983407062114695e-06, + "loss": 0.5761, + "step": 109 + }, + { + "epoch": 0.75, + "grad_norm": 0.4801446497440338, + "learning_rate": 4.9979515423108255e-06, + "loss": 0.5977, + "step": 110 + }, + { + "epoch": 0.7568181818181818, + "grad_norm": 0.4876883029937744, + "learning_rate": 4.997521437290205e-06, + "loss": 0.5849, + "step": 111 + }, + { + "epoch": 0.7636363636363637, + "grad_norm": 0.4328872263431549, + "learning_rate": 4.997050398198977e-06, + "loss": 0.5988, + "step": 112 + }, + { + "epoch": 0.7704545454545455, + "grad_norm": 0.4332719147205353, + "learning_rate": 4.996538432757414e-06, + "loss": 0.6059, + "step": 113 + }, + { + "epoch": 0.7772727272727272, + "grad_norm": 0.4455336630344391, + "learning_rate": 4.995985549356568e-06, + "loss": 0.5725, + "step": 114 + }, + { + "epoch": 0.7840909090909091, + "grad_norm": 0.4197766184806824, + "learning_rate": 4.995391757058129e-06, + "loss": 0.5715, + "step": 115 + }, + { + "epoch": 0.7909090909090909, + "grad_norm": 0.41945499181747437, + "learning_rate": 4.99475706559428e-06, + "loss": 0.5845, + "step": 116 + }, + { + "epoch": 0.7977272727272727, + "grad_norm": 0.42414823174476624, + "learning_rate": 4.994081485367537e-06, + "loss": 0.5595, + "step": 117 + }, + { + "epoch": 0.8045454545454546, + "grad_norm": 0.4056423008441925, + "learning_rate": 4.993365027450576e-06, + "loss": 0.582, + "step": 118 + }, + { + "epoch": 0.8113636363636364, + "grad_norm": 0.4554205536842346, + "learning_rate": 4.992607703586058e-06, + "loss": 0.5591, + "step": 119 + }, + { + "epoch": 0.8181818181818182, + "grad_norm": 0.4307089149951935, + "learning_rate": 4.991809526186424e-06, + "loss": 0.5927, + "step": 120 + }, + { + "epoch": 0.825, + "grad_norm": 0.48133528232574463, + "learning_rate": 4.990970508333707e-06, + "loss": 0.5703, + "step": 121 + }, + { + "epoch": 0.8318181818181818, + "grad_norm": 0.48582738637924194, + "learning_rate": 4.990090663779305e-06, + "loss": 0.5491, + "step": 122 + }, + { + "epoch": 0.8386363636363636, + "grad_norm": 0.44926196336746216, + "learning_rate": 4.9891700069437635e-06, + "loss": 0.5803, + "step": 123 + }, + { + "epoch": 0.8454545454545455, + "grad_norm": 0.4562165141105652, + "learning_rate": 4.988208552916535e-06, + "loss": 0.5945, + "step": 124 + }, + { + "epoch": 0.8522727272727273, + "grad_norm": 0.5885360836982727, + "learning_rate": 4.987206317455734e-06, + "loss": 0.5632, + "step": 125 + }, + { + "epoch": 0.8590909090909091, + "grad_norm": 0.4525550603866577, + "learning_rate": 4.986163316987877e-06, + "loss": 0.5221, + "step": 126 + }, + { + "epoch": 0.865909090909091, + "grad_norm": 0.4545478820800781, + "learning_rate": 4.985079568607613e-06, + "loss": 0.587, + "step": 127 + }, + { + "epoch": 0.8727272727272727, + "grad_norm": 0.4364396333694458, + "learning_rate": 4.983955090077445e-06, + "loss": 0.5626, + "step": 128 + }, + { + "epoch": 0.8795454545454545, + "grad_norm": 0.4354369640350342, + "learning_rate": 4.982789899827439e-06, + "loss": 0.5674, + "step": 129 + }, + { + "epoch": 0.8863636363636364, + "grad_norm": 0.4822773337364197, + "learning_rate": 4.9815840169549216e-06, + "loss": 0.5685, + "step": 130 + }, + { + "epoch": 0.8931818181818182, + "grad_norm": 0.44673025608062744, + "learning_rate": 4.980337461224164e-06, + "loss": 0.5733, + "step": 131 + }, + { + "epoch": 0.9, + "grad_norm": 0.6294501423835754, + "learning_rate": 4.979050253066064e-06, + "loss": 0.5396, + "step": 132 + }, + { + "epoch": 0.9068181818181819, + "grad_norm": 0.42890027165412903, + "learning_rate": 4.977722413577802e-06, + "loss": 0.5718, + "step": 133 + }, + { + "epoch": 0.9136363636363637, + "grad_norm": 0.470587819814682, + "learning_rate": 4.976353964522509e-06, + "loss": 0.5705, + "step": 134 + }, + { + "epoch": 0.9204545454545454, + "grad_norm": 0.4632768929004669, + "learning_rate": 4.974944928328894e-06, + "loss": 0.5394, + "step": 135 + }, + { + "epoch": 0.9272727272727272, + "grad_norm": 0.5304691195487976, + "learning_rate": 4.973495328090891e-06, + "loss": 0.5762, + "step": 136 + }, + { + "epoch": 0.9340909090909091, + "grad_norm": 0.42036354541778564, + "learning_rate": 4.972005187567267e-06, + "loss": 0.5649, + "step": 137 + }, + { + "epoch": 0.9409090909090909, + "grad_norm": 0.45189663767814636, + "learning_rate": 4.970474531181245e-06, + "loss": 0.5554, + "step": 138 + }, + { + "epoch": 0.9477272727272728, + "grad_norm": 0.47352468967437744, + "learning_rate": 4.968903384020095e-06, + "loss": 0.5511, + "step": 139 + }, + { + "epoch": 0.9545454545454546, + "grad_norm": 0.525702953338623, + "learning_rate": 4.967291771834727e-06, + "loss": 0.5777, + "step": 140 + }, + { + "epoch": 0.9613636363636363, + "grad_norm": 0.46146854758262634, + "learning_rate": 4.965639721039267e-06, + "loss": 0.5537, + "step": 141 + }, + { + "epoch": 0.9681818181818181, + "grad_norm": 0.5200941562652588, + "learning_rate": 4.963947258710626e-06, + "loss": 0.5522, + "step": 142 + }, + { + "epoch": 0.975, + "grad_norm": 0.5260375142097473, + "learning_rate": 4.962214412588053e-06, + "loss": 0.5747, + "step": 143 + }, + { + "epoch": 0.9818181818181818, + "grad_norm": 0.518645703792572, + "learning_rate": 4.960441211072686e-06, + "loss": 0.5257, + "step": 144 + }, + { + "epoch": 0.9886363636363636, + "grad_norm": 0.4528891444206238, + "learning_rate": 4.9586276832270785e-06, + "loss": 0.5548, + "step": 145 + }, + { + "epoch": 0.9954545454545455, + "grad_norm": 0.5137557983398438, + "learning_rate": 4.9567738587747314e-06, + "loss": 0.5526, + "step": 146 + }, + { + "epoch": 1.0068181818181818, + "grad_norm": 0.8773607611656189, + "learning_rate": 4.954879768099599e-06, + "loss": 1.1, + "step": 147 + }, + { + "epoch": 1.0136363636363637, + "grad_norm": 0.39079979062080383, + "learning_rate": 4.952945442245598e-06, + "loss": 0.568, + "step": 148 + }, + { + "epoch": 1.0204545454545455, + "grad_norm": 0.45004600286483765, + "learning_rate": 4.95097091291609e-06, + "loss": 0.5383, + "step": 149 + }, + { + "epoch": 1.0272727272727273, + "grad_norm": 0.41872867941856384, + "learning_rate": 4.948956212473371e-06, + "loss": 0.5436, + "step": 150 + }, + { + "epoch": 1.0340909090909092, + "grad_norm": 0.46829020977020264, + "learning_rate": 4.946901373938132e-06, + "loss": 0.5336, + "step": 151 + }, + { + "epoch": 1.040909090909091, + "grad_norm": 0.45038115978240967, + "learning_rate": 4.944806430988927e-06, + "loss": 0.538, + "step": 152 + }, + { + "epoch": 1.0477272727272728, + "grad_norm": 0.6496132016181946, + "learning_rate": 4.942671417961615e-06, + "loss": 0.5459, + "step": 153 + }, + { + "epoch": 1.0545454545454545, + "grad_norm": 0.4842703640460968, + "learning_rate": 4.940496369848795e-06, + "loss": 0.553, + "step": 154 + }, + { + "epoch": 1.0613636363636363, + "grad_norm": 0.4338674545288086, + "learning_rate": 4.938281322299243e-06, + "loss": 0.5366, + "step": 155 + }, + { + "epoch": 1.0681818181818181, + "grad_norm": 0.44630634784698486, + "learning_rate": 4.936026311617316e-06, + "loss": 0.5517, + "step": 156 + }, + { + "epoch": 1.075, + "grad_norm": 0.4606929421424866, + "learning_rate": 4.933731374762361e-06, + "loss": 0.5405, + "step": 157 + }, + { + "epoch": 1.0818181818181818, + "grad_norm": 0.48667654395103455, + "learning_rate": 4.931396549348115e-06, + "loss": 0.5376, + "step": 158 + }, + { + "epoch": 1.0886363636363636, + "grad_norm": 0.4516271650791168, + "learning_rate": 4.9290218736420795e-06, + "loss": 0.5206, + "step": 159 + }, + { + "epoch": 1.0954545454545455, + "grad_norm": 0.44719284772872925, + "learning_rate": 4.926607386564898e-06, + "loss": 0.5397, + "step": 160 + }, + { + "epoch": 1.1022727272727273, + "grad_norm": 0.40752628445625305, + "learning_rate": 4.9241531276897196e-06, + "loss": 0.5466, + "step": 161 + }, + { + "epoch": 1.1090909090909091, + "grad_norm": 0.45423609018325806, + "learning_rate": 4.921659137241544e-06, + "loss": 0.5227, + "step": 162 + }, + { + "epoch": 1.115909090909091, + "grad_norm": 0.4185822308063507, + "learning_rate": 4.919125456096574e-06, + "loss": 0.5144, + "step": 163 + }, + { + "epoch": 1.1227272727272728, + "grad_norm": 0.7753382921218872, + "learning_rate": 4.916552125781529e-06, + "loss": 0.5352, + "step": 164 + }, + { + "epoch": 1.1295454545454546, + "grad_norm": 0.47211477160453796, + "learning_rate": 4.913939188472979e-06, + "loss": 0.5505, + "step": 165 + }, + { + "epoch": 1.1363636363636362, + "grad_norm": 0.42742082476615906, + "learning_rate": 4.911286686996648e-06, + "loss": 0.529, + "step": 166 + }, + { + "epoch": 1.143181818181818, + "grad_norm": 0.41340839862823486, + "learning_rate": 4.908594664826708e-06, + "loss": 0.5292, + "step": 167 + }, + { + "epoch": 1.15, + "grad_norm": 0.44648951292037964, + "learning_rate": 4.905863166085076e-06, + "loss": 0.5757, + "step": 168 + }, + { + "epoch": 1.1568181818181817, + "grad_norm": 0.4285544753074646, + "learning_rate": 4.903092235540679e-06, + "loss": 0.5102, + "step": 169 + }, + { + "epoch": 1.1636363636363636, + "grad_norm": 0.42675015330314636, + "learning_rate": 4.900281918608732e-06, + "loss": 0.5397, + "step": 170 + }, + { + "epoch": 1.1704545454545454, + "grad_norm": 0.47636961936950684, + "learning_rate": 4.897432261349984e-06, + "loss": 0.5307, + "step": 171 + }, + { + "epoch": 1.1772727272727272, + "grad_norm": 0.4213427007198334, + "learning_rate": 4.894543310469968e-06, + "loss": 0.54, + "step": 172 + }, + { + "epoch": 1.184090909090909, + "grad_norm": 0.41153451800346375, + "learning_rate": 4.891615113318236e-06, + "loss": 0.5423, + "step": 173 + }, + { + "epoch": 1.190909090909091, + "grad_norm": 0.459185391664505, + "learning_rate": 4.888647717887582e-06, + "loss": 0.5239, + "step": 174 + }, + { + "epoch": 1.1977272727272728, + "grad_norm": 0.48279181122779846, + "learning_rate": 4.8856411728132526e-06, + "loss": 0.5336, + "step": 175 + }, + { + "epoch": 1.2045454545454546, + "grad_norm": 0.4186796247959137, + "learning_rate": 4.8825955273721524e-06, + "loss": 0.5509, + "step": 176 + }, + { + "epoch": 1.2113636363636364, + "grad_norm": 0.4345344007015228, + "learning_rate": 4.879510831482039e-06, + "loss": 0.5415, + "step": 177 + }, + { + "epoch": 1.2181818181818183, + "grad_norm": 0.461592435836792, + "learning_rate": 4.876387135700701e-06, + "loss": 0.5256, + "step": 178 + }, + { + "epoch": 1.225, + "grad_norm": 0.5960240960121155, + "learning_rate": 4.873224491225128e-06, + "loss": 0.5599, + "step": 179 + }, + { + "epoch": 1.231818181818182, + "grad_norm": 0.4662570357322693, + "learning_rate": 4.870022949890676e-06, + "loss": 0.5647, + "step": 180 + }, + { + "epoch": 1.2386363636363638, + "grad_norm": 0.4355057179927826, + "learning_rate": 4.866782564170217e-06, + "loss": 0.5056, + "step": 181 + }, + { + "epoch": 1.2454545454545454, + "grad_norm": 0.48699110746383667, + "learning_rate": 4.863503387173276e-06, + "loss": 0.5394, + "step": 182 + }, + { + "epoch": 1.2522727272727272, + "grad_norm": 0.48679423332214355, + "learning_rate": 4.860185472645161e-06, + "loss": 0.5394, + "step": 183 + }, + { + "epoch": 1.259090909090909, + "grad_norm": 0.6564393043518066, + "learning_rate": 4.856828874966086e-06, + "loss": 0.5259, + "step": 184 + }, + { + "epoch": 1.2659090909090909, + "grad_norm": 0.44762882590293884, + "learning_rate": 4.853433649150276e-06, + "loss": 0.5563, + "step": 185 + }, + { + "epoch": 1.2727272727272727, + "grad_norm": 0.5391108989715576, + "learning_rate": 4.849999850845066e-06, + "loss": 0.5326, + "step": 186 + }, + { + "epoch": 1.2795454545454545, + "grad_norm": 0.4201100170612335, + "learning_rate": 4.8465275363299905e-06, + "loss": 0.526, + "step": 187 + }, + { + "epoch": 1.2863636363636364, + "grad_norm": 0.46211543679237366, + "learning_rate": 4.84301676251586e-06, + "loss": 0.5498, + "step": 188 + }, + { + "epoch": 1.2931818181818182, + "grad_norm": 0.4325430989265442, + "learning_rate": 4.839467586943825e-06, + "loss": 0.556, + "step": 189 + }, + { + "epoch": 1.3, + "grad_norm": 0.46057143807411194, + "learning_rate": 4.835880067784441e-06, + "loss": 0.5079, + "step": 190 + }, + { + "epoch": 1.3068181818181819, + "grad_norm": 0.4920806288719177, + "learning_rate": 4.832254263836708e-06, + "loss": 0.533, + "step": 191 + }, + { + "epoch": 1.3136363636363637, + "grad_norm": 0.46693211793899536, + "learning_rate": 4.828590234527107e-06, + "loss": 0.5542, + "step": 192 + }, + { + "epoch": 1.3204545454545453, + "grad_norm": 0.4227873682975769, + "learning_rate": 4.82488803990863e-06, + "loss": 0.544, + "step": 193 + }, + { + "epoch": 1.3272727272727272, + "grad_norm": 0.44745922088623047, + "learning_rate": 4.821147740659795e-06, + "loss": 0.4962, + "step": 194 + }, + { + "epoch": 1.334090909090909, + "grad_norm": 0.46137240529060364, + "learning_rate": 4.817369398083648e-06, + "loss": 0.5359, + "step": 195 + }, + { + "epoch": 1.3409090909090908, + "grad_norm": 0.6483879685401917, + "learning_rate": 4.813553074106761e-06, + "loss": 0.5323, + "step": 196 + }, + { + "epoch": 1.3477272727272727, + "grad_norm": 0.6602574586868286, + "learning_rate": 4.809698831278217e-06, + "loss": 0.5161, + "step": 197 + }, + { + "epoch": 1.3545454545454545, + "grad_norm": 0.47758543491363525, + "learning_rate": 4.805806732768585e-06, + "loss": 0.5214, + "step": 198 + }, + { + "epoch": 1.3613636363636363, + "grad_norm": 0.4587438106536865, + "learning_rate": 4.801876842368882e-06, + "loss": 0.5208, + "step": 199 + }, + { + "epoch": 1.3681818181818182, + "grad_norm": 0.46055835485458374, + "learning_rate": 4.797909224489531e-06, + "loss": 0.5374, + "step": 200 + }, + { + "epoch": 1.375, + "grad_norm": 0.43727558851242065, + "learning_rate": 4.793903944159303e-06, + "loss": 0.5351, + "step": 201 + }, + { + "epoch": 1.3818181818181818, + "grad_norm": 0.4573572874069214, + "learning_rate": 4.789861067024253e-06, + "loss": 0.4997, + "step": 202 + }, + { + "epoch": 1.3886363636363637, + "grad_norm": 0.5921999216079712, + "learning_rate": 4.785780659346642e-06, + "loss": 0.5168, + "step": 203 + }, + { + "epoch": 1.3954545454545455, + "grad_norm": 0.47943904995918274, + "learning_rate": 4.781662788003851e-06, + "loss": 0.5513, + "step": 204 + }, + { + "epoch": 1.4022727272727273, + "grad_norm": 0.43733686208724976, + "learning_rate": 4.777507520487289e-06, + "loss": 0.5281, + "step": 205 + }, + { + "epoch": 1.4090909090909092, + "grad_norm": 0.4289107620716095, + "learning_rate": 4.773314924901281e-06, + "loss": 0.524, + "step": 206 + }, + { + "epoch": 1.415909090909091, + "grad_norm": 0.4203416109085083, + "learning_rate": 4.769085069961955e-06, + "loss": 0.5563, + "step": 207 + }, + { + "epoch": 1.4227272727272728, + "grad_norm": 0.4205201268196106, + "learning_rate": 4.764818024996117e-06, + "loss": 0.5202, + "step": 208 + }, + { + "epoch": 1.4295454545454547, + "grad_norm": 0.4296267032623291, + "learning_rate": 4.760513859940112e-06, + "loss": 0.5113, + "step": 209 + }, + { + "epoch": 1.4363636363636363, + "grad_norm": 0.3979087471961975, + "learning_rate": 4.756172645338675e-06, + "loss": 0.5313, + "step": 210 + }, + { + "epoch": 1.4431818181818181, + "grad_norm": 0.45232149958610535, + "learning_rate": 4.751794452343785e-06, + "loss": 0.5204, + "step": 211 + }, + { + "epoch": 1.45, + "grad_norm": 0.4309398829936981, + "learning_rate": 4.747379352713489e-06, + "loss": 0.5216, + "step": 212 + }, + { + "epoch": 1.4568181818181818, + "grad_norm": 0.45721709728240967, + "learning_rate": 4.7429274188107275e-06, + "loss": 0.5279, + "step": 213 + }, + { + "epoch": 1.4636363636363636, + "grad_norm": 0.46138322353363037, + "learning_rate": 4.738438723602154e-06, + "loss": 0.5287, + "step": 214 + }, + { + "epoch": 1.4704545454545455, + "grad_norm": 0.41752108931541443, + "learning_rate": 4.733913340656933e-06, + "loss": 0.5399, + "step": 215 + }, + { + "epoch": 1.4772727272727273, + "grad_norm": 0.45447197556495667, + "learning_rate": 4.729351344145536e-06, + "loss": 0.5277, + "step": 216 + }, + { + "epoch": 1.4840909090909091, + "grad_norm": 0.42524856328964233, + "learning_rate": 4.7247528088385296e-06, + "loss": 0.5103, + "step": 217 + }, + { + "epoch": 1.490909090909091, + "grad_norm": 0.46024221181869507, + "learning_rate": 4.720117810105341e-06, + "loss": 0.5172, + "step": 218 + }, + { + "epoch": 1.4977272727272728, + "grad_norm": 0.5868630409240723, + "learning_rate": 4.715446423913036e-06, + "loss": 0.5409, + "step": 219 + }, + { + "epoch": 1.5045454545454544, + "grad_norm": 0.424813836812973, + "learning_rate": 4.710738726825059e-06, + "loss": 0.5495, + "step": 220 + }, + { + "epoch": 1.5113636363636362, + "grad_norm": 0.49356943368911743, + "learning_rate": 4.705994795999991e-06, + "loss": 0.519, + "step": 221 + }, + { + "epoch": 1.518181818181818, + "grad_norm": 0.44291776418685913, + "learning_rate": 4.701214709190277e-06, + "loss": 0.5232, + "step": 222 + }, + { + "epoch": 1.525, + "grad_norm": 0.48171764612197876, + "learning_rate": 4.696398544740955e-06, + "loss": 0.5501, + "step": 223 + }, + { + "epoch": 1.5318181818181817, + "grad_norm": 0.4296175539493561, + "learning_rate": 4.69154638158837e-06, + "loss": 0.5087, + "step": 224 + }, + { + "epoch": 1.5386363636363636, + "grad_norm": 0.4715246260166168, + "learning_rate": 4.686658299258881e-06, + "loss": 0.5363, + "step": 225 + }, + { + "epoch": 1.5454545454545454, + "grad_norm": 0.44422832131385803, + "learning_rate": 4.681734377867562e-06, + "loss": 0.5371, + "step": 226 + }, + { + "epoch": 1.5522727272727272, + "grad_norm": 4.221654891967773, + "learning_rate": 4.67677469811688e-06, + "loss": 0.5191, + "step": 227 + }, + { + "epoch": 1.559090909090909, + "grad_norm": 0.41857993602752686, + "learning_rate": 4.671779341295378e-06, + "loss": 0.5191, + "step": 228 + }, + { + "epoch": 1.565909090909091, + "grad_norm": 0.4950161278247833, + "learning_rate": 4.666748389276344e-06, + "loss": 0.5311, + "step": 229 + }, + { + "epoch": 1.5727272727272728, + "grad_norm": 0.44641605019569397, + "learning_rate": 4.661681924516466e-06, + "loss": 0.5174, + "step": 230 + }, + { + "epoch": 1.5795454545454546, + "grad_norm": 0.39064037799835205, + "learning_rate": 4.6565800300544805e-06, + "loss": 0.515, + "step": 231 + }, + { + "epoch": 1.5863636363636364, + "grad_norm": 0.7764399647712708, + "learning_rate": 4.651442789509813e-06, + "loss": 0.5282, + "step": 232 + }, + { + "epoch": 1.5931818181818183, + "grad_norm": 0.44421035051345825, + "learning_rate": 4.646270287081208e-06, + "loss": 0.5324, + "step": 233 + }, + { + "epoch": 1.6, + "grad_norm": 0.5149928331375122, + "learning_rate": 4.641062607545347e-06, + "loss": 0.5264, + "step": 234 + }, + { + "epoch": 1.606818181818182, + "grad_norm": 0.4784093201160431, + "learning_rate": 4.6358198362554585e-06, + "loss": 0.552, + "step": 235 + }, + { + "epoch": 1.6136363636363638, + "grad_norm": 0.4138723313808441, + "learning_rate": 4.630542059139923e-06, + "loss": 0.5202, + "step": 236 + }, + { + "epoch": 1.6204545454545456, + "grad_norm": 0.4509594440460205, + "learning_rate": 4.625229362700863e-06, + "loss": 0.5318, + "step": 237 + }, + { + "epoch": 1.6272727272727274, + "grad_norm": 0.44979560375213623, + "learning_rate": 4.61988183401272e-06, + "loss": 0.521, + "step": 238 + }, + { + "epoch": 1.634090909090909, + "grad_norm": 0.4525575041770935, + "learning_rate": 4.614499560720837e-06, + "loss": 0.5073, + "step": 239 + }, + { + "epoch": 1.6409090909090909, + "grad_norm": 0.44126543402671814, + "learning_rate": 4.609082631040012e-06, + "loss": 0.5058, + "step": 240 + }, + { + "epoch": 1.6477272727272727, + "grad_norm": 0.887867271900177, + "learning_rate": 4.603631133753061e-06, + "loss": 0.5471, + "step": 241 + }, + { + "epoch": 1.6545454545454545, + "grad_norm": 0.4497375786304474, + "learning_rate": 4.598145158209356e-06, + "loss": 0.501, + "step": 242 + }, + { + "epoch": 1.6613636363636364, + "grad_norm": 0.45681032538414, + "learning_rate": 4.592624794323366e-06, + "loss": 0.5306, + "step": 243 + }, + { + "epoch": 1.6681818181818182, + "grad_norm": 0.5186678171157837, + "learning_rate": 4.587070132573178e-06, + "loss": 0.5097, + "step": 244 + }, + { + "epoch": 1.675, + "grad_norm": 0.44289669394493103, + "learning_rate": 4.581481263999019e-06, + "loss": 0.5085, + "step": 245 + }, + { + "epoch": 1.6818181818181817, + "grad_norm": 0.4571002125740051, + "learning_rate": 4.575858280201761e-06, + "loss": 0.5327, + "step": 246 + }, + { + "epoch": 1.6886363636363635, + "grad_norm": 0.43753257393836975, + "learning_rate": 4.570201273341418e-06, + "loss": 0.5167, + "step": 247 + }, + { + "epoch": 1.6954545454545453, + "grad_norm": 0.44994574785232544, + "learning_rate": 4.564510336135642e-06, + "loss": 0.5554, + "step": 248 + }, + { + "epoch": 1.7022727272727272, + "grad_norm": 0.6185874938964844, + "learning_rate": 4.558785561858196e-06, + "loss": 0.5115, + "step": 249 + }, + { + "epoch": 1.709090909090909, + "grad_norm": 0.4883944094181061, + "learning_rate": 4.5530270443374305e-06, + "loss": 0.5312, + "step": 250 + }, + { + "epoch": 1.7159090909090908, + "grad_norm": 0.46433499455451965, + "learning_rate": 4.547234877954741e-06, + "loss": 0.5368, + "step": 251 + }, + { + "epoch": 1.7227272727272727, + "grad_norm": 0.4270826578140259, + "learning_rate": 4.541409157643027e-06, + "loss": 0.5258, + "step": 252 + }, + { + "epoch": 1.7295454545454545, + "grad_norm": 0.4390241205692291, + "learning_rate": 4.535549978885132e-06, + "loss": 0.5259, + "step": 253 + }, + { + "epoch": 1.7363636363636363, + "grad_norm": 0.42756086587905884, + "learning_rate": 4.5296574377122765e-06, + "loss": 0.5217, + "step": 254 + }, + { + "epoch": 1.7431818181818182, + "grad_norm": 0.48124927282333374, + "learning_rate": 4.5237316307024895e-06, + "loss": 0.5095, + "step": 255 + }, + { + "epoch": 1.75, + "grad_norm": 0.4394672214984894, + "learning_rate": 4.517772654979024e-06, + "loss": 0.5116, + "step": 256 + }, + { + "epoch": 1.7568181818181818, + "grad_norm": 0.5805886387825012, + "learning_rate": 4.51178060820876e-06, + "loss": 0.5145, + "step": 257 + }, + { + "epoch": 1.7636363636363637, + "grad_norm": 0.4845556616783142, + "learning_rate": 4.505755588600613e-06, + "loss": 0.5064, + "step": 258 + }, + { + "epoch": 1.7704545454545455, + "grad_norm": 0.45300963521003723, + "learning_rate": 4.499697694903915e-06, + "loss": 0.5362, + "step": 259 + }, + { + "epoch": 1.7772727272727273, + "grad_norm": 0.45417436957359314, + "learning_rate": 4.493607026406802e-06, + "loss": 0.5234, + "step": 260 + }, + { + "epoch": 1.7840909090909092, + "grad_norm": 0.4577132761478424, + "learning_rate": 4.487483682934587e-06, + "loss": 0.5271, + "step": 261 + }, + { + "epoch": 1.790909090909091, + "grad_norm": 0.4518176019191742, + "learning_rate": 4.481327764848118e-06, + "loss": 0.4853, + "step": 262 + }, + { + "epoch": 1.7977272727272728, + "grad_norm": 0.43804875016212463, + "learning_rate": 4.47513937304214e-06, + "loss": 0.5188, + "step": 263 + }, + { + "epoch": 1.8045454545454547, + "grad_norm": 0.4401354193687439, + "learning_rate": 4.4689186089436365e-06, + "loss": 0.5079, + "step": 264 + }, + { + "epoch": 1.8113636363636365, + "grad_norm": 0.4051227569580078, + "learning_rate": 4.462665574510169e-06, + "loss": 0.5084, + "step": 265 + }, + { + "epoch": 1.8181818181818183, + "grad_norm": 0.5269004106521606, + "learning_rate": 4.456380372228208e-06, + "loss": 0.528, + "step": 266 + }, + { + "epoch": 1.825, + "grad_norm": 0.44371986389160156, + "learning_rate": 4.450063105111447e-06, + "loss": 0.5388, + "step": 267 + }, + { + "epoch": 1.8318181818181818, + "grad_norm": 0.41871729493141174, + "learning_rate": 4.443713876699124e-06, + "loss": 0.502, + "step": 268 + }, + { + "epoch": 1.8386363636363636, + "grad_norm": 0.464652419090271, + "learning_rate": 4.4373327910543125e-06, + "loss": 0.5045, + "step": 269 + }, + { + "epoch": 1.8454545454545455, + "grad_norm": 0.43198078870773315, + "learning_rate": 4.430919952762226e-06, + "loss": 0.5149, + "step": 270 + }, + { + "epoch": 1.8522727272727273, + "grad_norm": 0.42869722843170166, + "learning_rate": 4.424475466928499e-06, + "loss": 0.5329, + "step": 271 + }, + { + "epoch": 1.8590909090909091, + "grad_norm": 0.529680609703064, + "learning_rate": 4.417999439177465e-06, + "loss": 0.4939, + "step": 272 + }, + { + "epoch": 1.865909090909091, + "grad_norm": 0.4562613368034363, + "learning_rate": 4.4114919756504275e-06, + "loss": 0.5389, + "step": 273 + }, + { + "epoch": 1.8727272727272726, + "grad_norm": 0.4480995237827301, + "learning_rate": 4.404953183003916e-06, + "loss": 0.51, + "step": 274 + }, + { + "epoch": 1.8795454545454544, + "grad_norm": 0.41215088963508606, + "learning_rate": 4.398383168407941e-06, + "loss": 0.5386, + "step": 275 + }, + { + "epoch": 1.8863636363636362, + "grad_norm": 0.8426409363746643, + "learning_rate": 4.391782039544239e-06, + "loss": 0.498, + "step": 276 + }, + { + "epoch": 1.893181818181818, + "grad_norm": 0.5359936356544495, + "learning_rate": 4.385149904604502e-06, + "loss": 0.51, + "step": 277 + }, + { + "epoch": 1.9, + "grad_norm": 0.4462437033653259, + "learning_rate": 4.378486872288611e-06, + "loss": 0.5082, + "step": 278 + }, + { + "epoch": 1.9068181818181817, + "grad_norm": 0.445359468460083, + "learning_rate": 4.371793051802849e-06, + "loss": 0.5171, + "step": 279 + }, + { + "epoch": 1.9136363636363636, + "grad_norm": 0.4223516583442688, + "learning_rate": 4.365068552858116e-06, + "loss": 0.5218, + "step": 280 + }, + { + "epoch": 1.9204545454545454, + "grad_norm": 0.43929681181907654, + "learning_rate": 4.358313485668124e-06, + "loss": 0.5156, + "step": 281 + }, + { + "epoch": 1.9272727272727272, + "grad_norm": 0.42535650730133057, + "learning_rate": 4.3515279609476e-06, + "loss": 0.4964, + "step": 282 + }, + { + "epoch": 1.934090909090909, + "grad_norm": 0.46448391675949097, + "learning_rate": 4.3447120899104615e-06, + "loss": 0.5182, + "step": 283 + }, + { + "epoch": 1.940909090909091, + "grad_norm": 0.4082206189632416, + "learning_rate": 4.337865984268002e-06, + "loss": 0.5175, + "step": 284 + }, + { + "epoch": 1.9477272727272728, + "grad_norm": 0.42423564195632935, + "learning_rate": 4.3309897562270525e-06, + "loss": 0.5352, + "step": 285 + }, + { + "epoch": 1.9545454545454546, + "grad_norm": 0.5240079760551453, + "learning_rate": 4.324083518488151e-06, + "loss": 0.5157, + "step": 286 + }, + { + "epoch": 1.9613636363636364, + "grad_norm": 0.4517645835876465, + "learning_rate": 4.317147384243688e-06, + "loss": 0.5399, + "step": 287 + }, + { + "epoch": 1.9681818181818183, + "grad_norm": 0.44054290652275085, + "learning_rate": 4.3101814671760546e-06, + "loss": 0.5293, + "step": 288 + }, + { + "epoch": 1.975, + "grad_norm": 0.4256063401699066, + "learning_rate": 4.303185881455778e-06, + "loss": 0.5323, + "step": 289 + }, + { + "epoch": 1.981818181818182, + "grad_norm": 0.43106094002723694, + "learning_rate": 4.296160741739652e-06, + "loss": 0.5259, + "step": 290 + }, + { + "epoch": 1.9886363636363638, + "grad_norm": 0.43816477060317993, + "learning_rate": 4.289106163168858e-06, + "loss": 0.5098, + "step": 291 + }, + { + "epoch": 1.9954545454545456, + "grad_norm": 1.1177959442138672, + "learning_rate": 4.282022261367074e-06, + "loss": 0.513, + "step": 292 + } + ], + "logging_steps": 1, + "max_steps": 876, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 146, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 1.880633908528572e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-292/training_args.bin b/checkpoint-292/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..b41c805059a36d8d2e3bf3511fb0fd95b74dae74 --- /dev/null +++ b/checkpoint-292/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68945b8f9934f5870e67a2d4a20c31fa8e742da41075fa7236fb313c6197456f +size 7928 diff --git a/checkpoint-292/zero_to_fp32.py b/checkpoint-292/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-292/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-438/README.md b/checkpoint-438/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1b184114a0c28ed3e4c082c18486736dc818166d --- /dev/null +++ b/checkpoint-438/README.md @@ -0,0 +1,202 @@ +--- +base_model: meta-llama/Llama-3.3-70B-Instruct +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 0.15.0 \ No newline at end of file diff --git a/checkpoint-438/adapter_config.json b/checkpoint-438/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a5775df151cb9649596940e8a117cb92f810dbb7 --- /dev/null +++ b/checkpoint-438/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.3-70B-Instruct", + "bias": "none", + "corda_config": null, + "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", + "down_proj", + "k_proj", + "q_proj", + "o_proj", + "v_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-438/adapter_model.safetensors b/checkpoint-438/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..8d77841dc263011e543baf97ea31c69b16e40662 --- /dev/null +++ b/checkpoint-438/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:15f52615e9834b6fe0e1050189ab340fd5f078a2d2410237c7729632d8dd02d0 +size 10829849744 diff --git a/checkpoint-438/global_step438/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-438/global_step438/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..5f442f402647efa31a0d0c2e9a978f2307128275 --- /dev/null +++ b/checkpoint-438/global_step438/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:ab5eed27a08c921db03717b0e648563d66e65848994b8ab328803edec7adb2bd +size 21659418140 diff --git a/checkpoint-438/global_step438/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-438/global_step438/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..2cfded6bb8378d04afa43ccdae6fe70abd861d2d --- /dev/null +++ b/checkpoint-438/global_step438/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:5384071fdc83151ab49eeaebb2c46c9c6e9c103b4bdf45fed6f4aa79be8ecce2 +size 21659457372 diff --git a/checkpoint-438/global_step438/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoint-438/global_step438/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..5851c65abccf51ac68bb56b35856f4c391039251 --- /dev/null +++ b/checkpoint-438/global_step438/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:fca43757b05eb3b67056bdb802b00f6350d0ae40cc17442689af3312047a0379 +size 21659417820 diff --git a/checkpoint-438/global_step438/mp_rank_00_model_states.pt b/checkpoint-438/global_step438/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..b39c850b335b88af4e07298289b06fa1cc32368e --- /dev/null +++ b/checkpoint-438/global_step438/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ebc16ca36cd9afe83eb196df6c14831c6d14d79188aa354e0e1ea83e2f2d8501 +size 11918643933 diff --git a/checkpoint-438/latest b/checkpoint-438/latest new file mode 100644 index 0000000000000000000000000000000000000000..86d68321558878066f54480bff1ef610f49b8d18 --- /dev/null +++ b/checkpoint-438/latest @@ -0,0 +1 @@ +global_step438 \ No newline at end of file diff --git a/checkpoint-438/rng_state_0.pth b/checkpoint-438/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..231ca2f4e75467112a9c042ca2314ad9c7b9efa4 --- /dev/null +++ b/checkpoint-438/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:ed7652353f9ae35d44ea17a13928bfd35ffe7a3ad046364d548451e04f56c644 +size 14768 diff --git a/checkpoint-438/rng_state_1.pth b/checkpoint-438/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..2a814d34bc80fc854cb25407017557c13ce6a730 --- /dev/null +++ b/checkpoint-438/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d637d4cd2814421e384a5d97c1d437cba64790c630a6ca054804c1f774f16227 +size 14768 diff --git a/checkpoint-438/rng_state_2.pth b/checkpoint-438/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..c8d9993a974d3723a21c161bc7300dca674268ab --- /dev/null +++ b/checkpoint-438/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8aef82806d1b494ed5f8bb32b4d7a2bdb0673158ee4e79032b84e7d1c9ac83bc +size 14768 diff --git a/checkpoint-438/scheduler.pt b/checkpoint-438/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..4702258cfd3ceeb84477677cb7d2e1242dc5d6a4 --- /dev/null +++ b/checkpoint-438/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1eba51ec914f5bb0df236bfbfffe53ac98ad583052980e3a3d82c7a3c408a7ff +size 1064 diff --git a/checkpoint-438/special_tokens_map.json b/checkpoint-438/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-438/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": "<|eot_id|>", + "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-438/tokenizer.json b/checkpoint-438/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-438/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-438/tokenizer_config.json b/checkpoint-438/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ca91a2ef55f4239a7af81d7c9abb05f53621a07b --- /dev/null +++ b/checkpoint-438/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "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|>", + "chat_template": "{{- bos_token }}\n{%- if custom_tools is defined %}\n {%- set tools = custom_tools %}\n{%- endif %}\n{%- if not tools_in_user_message is defined %}\n {%- set tools_in_user_message = true %}\n{%- endif %}\n{%- if not date_string is defined %}\n {%- set date_string = \"26 Jul 2024\" %}\n{%- endif %}\n{%- if not tools is defined %}\n {%- set tools = none %}\n{%- endif %}\n\n{#- This block extracts the system message, so we can slot it into the right place. #}\n{%- if messages[0]['role'] == 'system' %}\n {%- set system_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n{%- else %}\n {%- set system_message = \"\" %}\n{%- endif %}\n\n{#- System message + builtin tools #}\n{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}\n{%- if builtin_tools is defined or tools is not none %}\n {{- \"Environment: ipython\\n\" }}\n{%- endif %}\n{%- if builtin_tools is defined %}\n {{- \"Tools: \" + builtin_tools | reject('equalto', 'code_interpreter') | join(\", \") + \"\\n\\n\"}}\n{%- endif %}\n{{- \"Cutting Knowledge Date: December 2023\\n\" }}\n{{- \"Today Date: \" + date_string + \"\\n\\n\" }}\n{%- if tools is not none and not tools_in_user_message %}\n {{- \"You have access to the following functions. To call a function, please respond with JSON for a function call.\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n{%- endif %}\n{{- system_message }}\n{{- \"<|eot_id|>\" }}\n\n{#- Custom tools are passed in a user message with some extra guidance #}\n{%- if tools_in_user_message and not tools is none %}\n {#- Extract the first user message so we can plug it in here #}\n {%- if messages | length != 0 %}\n {%- set first_user_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n {%- else %}\n {{- raise_exception(\"Cannot put tools in the first user message when there's no first user message!\") }}\n{%- endif %}\n {{- '<|start_header_id|>user<|end_header_id|>\\n\\n' -}}\n {{- \"Given the following functions, please respond with a JSON for a function call \" }}\n {{- \"with its proper arguments that best answers the given prompt.\\n\\n\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n {{- first_user_message + \"<|eot_id|>\"}}\n{%- endif %}\n\n{%- for message in messages %}\n {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}\n {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n'+ message['content'] | trim + '<|eot_id|>' }}\n {%- elif 'tool_calls' in message %}\n {%- if not message.tool_calls|length == 1 %}\n {{- raise_exception(\"This model only supports single tool-calls at once!\") }}\n {%- endif %}\n {%- set tool_call = message.tool_calls[0].function %}\n {%- if builtin_tools is defined and tool_call.name in builtin_tools %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- \"<|python_tag|>\" + tool_call.name + \".call(\" }}\n {%- for arg_name, arg_val in tool_call.arguments | items %}\n {{- arg_name + '=\"' + arg_val + '\"' }}\n {%- if not loop.last %}\n {{- \", \" }}\n {%- endif %}\n {%- endfor %}\n {{- \")\" }}\n {%- else %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- '{\"name\": \"' + tool_call.name + '\", ' }}\n {{- '\"parameters\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- \"}\" }}\n {%- endif %}\n {%- if builtin_tools is defined %}\n {#- This means we're in ipython mode #}\n {{- \"<|eom_id|>\" }}\n {%- else %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n {%- elif message.role == \"tool\" or message.role == \"ipython\" %}\n {{- \"<|start_header_id|>ipython<|end_header_id|>\\n\\n\" }}\n {%- if message.content is mapping or message.content is iterable %}\n {{- message.content | tojson }}\n {%- else %}\n {{- message.content }}\n {%- endif %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}\n{%- endif %}\n", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "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-438/trainer_state.json b/checkpoint-438/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..78a4961c2b7f30f6acf9e4913cbb1c7e5024f7ee --- /dev/null +++ b/checkpoint-438/trainer_state.json @@ -0,0 +1,3099 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 2.9954545454545456, + "eval_steps": 500, + "global_step": 438, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.006818181818181818, + "grad_norm": 1.8943578004837036, + "learning_rate": 5.0000000000000004e-08, + "loss": 0.8676, + "step": 1 + }, + { + "epoch": 0.013636363636363636, + "grad_norm": 7.775813102722168, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.8513, + "step": 2 + }, + { + "epoch": 0.020454545454545454, + "grad_norm": 1.768803596496582, + "learning_rate": 1.5000000000000002e-07, + "loss": 0.8339, + "step": 3 + }, + { + "epoch": 0.02727272727272727, + "grad_norm": 1.7032263278961182, + "learning_rate": 2.0000000000000002e-07, + "loss": 0.8348, + "step": 4 + }, + { + "epoch": 0.03409090909090909, + "grad_norm": 1.8244796991348267, + "learning_rate": 2.5000000000000004e-07, + "loss": 0.8687, + "step": 5 + }, + { + "epoch": 0.04090909090909091, + "grad_norm": 1.7780181169509888, + "learning_rate": 3.0000000000000004e-07, + "loss": 0.8113, + "step": 6 + }, + { + "epoch": 0.04772727272727273, + "grad_norm": 1.8362191915512085, + "learning_rate": 3.5000000000000004e-07, + "loss": 0.8463, + "step": 7 + }, + { + "epoch": 0.05454545454545454, + "grad_norm": 1.6640979051589966, + "learning_rate": 4.0000000000000003e-07, + "loss": 0.8336, + "step": 8 + }, + { + "epoch": 0.06136363636363636, + "grad_norm": 1.8013464212417603, + "learning_rate": 4.5000000000000003e-07, + "loss": 0.8572, + "step": 9 + }, + { + "epoch": 0.06818181818181818, + "grad_norm": 1.7237999439239502, + "learning_rate": 5.000000000000001e-07, + "loss": 0.8501, + "step": 10 + }, + { + "epoch": 0.075, + "grad_norm": 1.989346981048584, + "learning_rate": 5.5e-07, + "loss": 0.8615, + "step": 11 + }, + { + "epoch": 0.08181818181818182, + "grad_norm": 2.0312082767486572, + "learning_rate": 6.000000000000001e-07, + "loss": 0.858, + "step": 12 + }, + { + "epoch": 0.08863636363636364, + "grad_norm": 1.6921299695968628, + "learning_rate": 6.5e-07, + "loss": 0.8577, + "step": 13 + }, + { + "epoch": 0.09545454545454546, + "grad_norm": 2.4207160472869873, + "learning_rate": 7.000000000000001e-07, + "loss": 0.8564, + "step": 14 + }, + { + "epoch": 0.10227272727272728, + "grad_norm": 1.7572585344314575, + "learning_rate": 7.5e-07, + "loss": 0.8409, + "step": 15 + }, + { + "epoch": 0.10909090909090909, + "grad_norm": 1.6135950088500977, + "learning_rate": 8.000000000000001e-07, + "loss": 0.8086, + "step": 16 + }, + { + "epoch": 0.1159090909090909, + "grad_norm": 2.226036787033081, + "learning_rate": 8.500000000000001e-07, + "loss": 0.8106, + "step": 17 + }, + { + "epoch": 0.12272727272727273, + "grad_norm": 1.7164231538772583, + "learning_rate": 9.000000000000001e-07, + "loss": 0.8494, + "step": 18 + }, + { + "epoch": 0.12954545454545455, + "grad_norm": 1.6572000980377197, + "learning_rate": 9.500000000000001e-07, + "loss": 0.8023, + "step": 19 + }, + { + "epoch": 0.13636363636363635, + "grad_norm": 1.5672118663787842, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.787, + "step": 20 + }, + { + "epoch": 0.1431818181818182, + "grad_norm": 1.5080257654190063, + "learning_rate": 1.0500000000000001e-06, + "loss": 0.8105, + "step": 21 + }, + { + "epoch": 0.15, + "grad_norm": 1.484892725944519, + "learning_rate": 1.1e-06, + "loss": 0.8258, + "step": 22 + }, + { + "epoch": 0.15681818181818183, + "grad_norm": 1.4591134786605835, + "learning_rate": 1.1500000000000002e-06, + "loss": 0.8032, + "step": 23 + }, + { + "epoch": 0.16363636363636364, + "grad_norm": 1.5506278276443481, + "learning_rate": 1.2000000000000002e-06, + "loss": 0.8236, + "step": 24 + }, + { + "epoch": 0.17045454545454544, + "grad_norm": 1.639350414276123, + "learning_rate": 1.25e-06, + "loss": 0.8185, + "step": 25 + }, + { + "epoch": 0.17727272727272728, + "grad_norm": 1.592640995979309, + "learning_rate": 1.3e-06, + "loss": 0.7898, + "step": 26 + }, + { + "epoch": 0.18409090909090908, + "grad_norm": 1.5280641317367554, + "learning_rate": 1.3500000000000002e-06, + "loss": 0.7731, + "step": 27 + }, + { + "epoch": 0.19090909090909092, + "grad_norm": 1.4015443325042725, + "learning_rate": 1.4000000000000001e-06, + "loss": 0.7956, + "step": 28 + }, + { + "epoch": 0.19772727272727272, + "grad_norm": 1.4890056848526, + "learning_rate": 1.45e-06, + "loss": 0.7838, + "step": 29 + }, + { + "epoch": 0.20454545454545456, + "grad_norm": 2.4770750999450684, + "learning_rate": 1.5e-06, + "loss": 0.7766, + "step": 30 + }, + { + "epoch": 0.21136363636363636, + "grad_norm": 1.5863685607910156, + "learning_rate": 1.5500000000000002e-06, + "loss": 0.7353, + "step": 31 + }, + { + "epoch": 0.21818181818181817, + "grad_norm": 1.058114767074585, + "learning_rate": 1.6000000000000001e-06, + "loss": 0.7237, + "step": 32 + }, + { + "epoch": 0.225, + "grad_norm": 1.5869101285934448, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.7597, + "step": 33 + }, + { + "epoch": 0.2318181818181818, + "grad_norm": 1.0012413263320923, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.7388, + "step": 34 + }, + { + "epoch": 0.23863636363636365, + "grad_norm": 1.118056058883667, + "learning_rate": 1.75e-06, + "loss": 0.7786, + "step": 35 + }, + { + "epoch": 0.24545454545454545, + "grad_norm": 1.0648820400238037, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.7502, + "step": 36 + }, + { + "epoch": 0.25227272727272726, + "grad_norm": 0.9440382719039917, + "learning_rate": 1.85e-06, + "loss": 0.6942, + "step": 37 + }, + { + "epoch": 0.2590909090909091, + "grad_norm": 0.8300187587738037, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.7312, + "step": 38 + }, + { + "epoch": 0.26590909090909093, + "grad_norm": 0.8296191692352295, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.7461, + "step": 39 + }, + { + "epoch": 0.2727272727272727, + "grad_norm": 0.7797508239746094, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.7234, + "step": 40 + }, + { + "epoch": 0.27954545454545454, + "grad_norm": 0.7335019707679749, + "learning_rate": 2.05e-06, + "loss": 0.6927, + "step": 41 + }, + { + "epoch": 0.2863636363636364, + "grad_norm": 0.7062106728553772, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.6552, + "step": 42 + }, + { + "epoch": 0.29318181818181815, + "grad_norm": 1.2342087030410767, + "learning_rate": 2.15e-06, + "loss": 0.6959, + "step": 43 + }, + { + "epoch": 0.3, + "grad_norm": 0.8967597484588623, + "learning_rate": 2.2e-06, + "loss": 0.704, + "step": 44 + }, + { + "epoch": 0.3068181818181818, + "grad_norm": 0.7228404879570007, + "learning_rate": 2.25e-06, + "loss": 0.6777, + "step": 45 + }, + { + "epoch": 0.31363636363636366, + "grad_norm": 0.6731488704681396, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.6652, + "step": 46 + }, + { + "epoch": 0.32045454545454544, + "grad_norm": 0.71966952085495, + "learning_rate": 2.35e-06, + "loss": 0.7281, + "step": 47 + }, + { + "epoch": 0.32727272727272727, + "grad_norm": 0.7057356238365173, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.6476, + "step": 48 + }, + { + "epoch": 0.3340909090909091, + "grad_norm": 0.6346054077148438, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.6865, + "step": 49 + }, + { + "epoch": 0.3409090909090909, + "grad_norm": 0.6938223838806152, + "learning_rate": 2.5e-06, + "loss": 0.6808, + "step": 50 + }, + { + "epoch": 0.3477272727272727, + "grad_norm": 0.7425184845924377, + "learning_rate": 2.55e-06, + "loss": 0.7074, + "step": 51 + }, + { + "epoch": 0.35454545454545455, + "grad_norm": 0.6743818521499634, + "learning_rate": 2.6e-06, + "loss": 0.657, + "step": 52 + }, + { + "epoch": 0.3613636363636364, + "grad_norm": 0.6785942316055298, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.6768, + "step": 53 + }, + { + "epoch": 0.36818181818181817, + "grad_norm": 0.586974024772644, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.651, + "step": 54 + }, + { + "epoch": 0.375, + "grad_norm": 0.5727331042289734, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.669, + "step": 55 + }, + { + "epoch": 0.38181818181818183, + "grad_norm": 0.7333543300628662, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.6571, + "step": 56 + }, + { + "epoch": 0.3886363636363636, + "grad_norm": 0.5710961818695068, + "learning_rate": 2.85e-06, + "loss": 0.6475, + "step": 57 + }, + { + "epoch": 0.39545454545454545, + "grad_norm": 0.5614489912986755, + "learning_rate": 2.9e-06, + "loss": 0.6314, + "step": 58 + }, + { + "epoch": 0.4022727272727273, + "grad_norm": 0.5007341504096985, + "learning_rate": 2.95e-06, + "loss": 0.6346, + "step": 59 + }, + { + "epoch": 0.4090909090909091, + "grad_norm": 0.54221510887146, + "learning_rate": 3e-06, + "loss": 0.6715, + "step": 60 + }, + { + "epoch": 0.4159090909090909, + "grad_norm": 0.5177039504051208, + "learning_rate": 3.05e-06, + "loss": 0.6459, + "step": 61 + }, + { + "epoch": 0.42272727272727273, + "grad_norm": 0.4741189181804657, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.592, + "step": 62 + }, + { + "epoch": 0.42954545454545456, + "grad_norm": 0.5637328028678894, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.6459, + "step": 63 + }, + { + "epoch": 0.43636363636363634, + "grad_norm": 0.5478869676589966, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.6447, + "step": 64 + }, + { + "epoch": 0.4431818181818182, + "grad_norm": 0.5483130216598511, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.6319, + "step": 65 + }, + { + "epoch": 0.45, + "grad_norm": 0.478081613779068, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.6102, + "step": 66 + }, + { + "epoch": 0.45681818181818185, + "grad_norm": 0.4777645766735077, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.5986, + "step": 67 + }, + { + "epoch": 0.4636363636363636, + "grad_norm": 0.5291482210159302, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.6314, + "step": 68 + }, + { + "epoch": 0.47045454545454546, + "grad_norm": 0.5224051475524902, + "learning_rate": 3.45e-06, + "loss": 0.6278, + "step": 69 + }, + { + "epoch": 0.4772727272727273, + "grad_norm": 0.6474127173423767, + "learning_rate": 3.5e-06, + "loss": 0.6398, + "step": 70 + }, + { + "epoch": 0.48409090909090907, + "grad_norm": 0.5060178637504578, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.5936, + "step": 71 + }, + { + "epoch": 0.4909090909090909, + "grad_norm": 3.0769243240356445, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.5975, + "step": 72 + }, + { + "epoch": 0.49772727272727274, + "grad_norm": 0.5319749116897583, + "learning_rate": 3.65e-06, + "loss": 0.6119, + "step": 73 + }, + { + "epoch": 0.5045454545454545, + "grad_norm": 0.6745399236679077, + "learning_rate": 3.7e-06, + "loss": 0.6213, + "step": 74 + }, + { + "epoch": 0.5113636363636364, + "grad_norm": 0.5156731009483337, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.6262, + "step": 75 + }, + { + "epoch": 0.5181818181818182, + "grad_norm": 0.9138725399971008, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.6244, + "step": 76 + }, + { + "epoch": 0.525, + "grad_norm": 0.4362037777900696, + "learning_rate": 3.85e-06, + "loss": 0.5874, + "step": 77 + }, + { + "epoch": 0.5318181818181819, + "grad_norm": 0.4674088954925537, + "learning_rate": 3.900000000000001e-06, + "loss": 0.615, + "step": 78 + }, + { + "epoch": 0.5386363636363637, + "grad_norm": 0.4590819180011749, + "learning_rate": 3.95e-06, + "loss": 0.6003, + "step": 79 + }, + { + "epoch": 0.5454545454545454, + "grad_norm": 0.5145031809806824, + "learning_rate": 4.000000000000001e-06, + "loss": 0.6079, + "step": 80 + }, + { + "epoch": 0.5522727272727272, + "grad_norm": 0.46660348773002625, + "learning_rate": 4.05e-06, + "loss": 0.6174, + "step": 81 + }, + { + "epoch": 0.5590909090909091, + "grad_norm": 0.48053789138793945, + "learning_rate": 4.1e-06, + "loss": 0.6232, + "step": 82 + }, + { + "epoch": 0.5659090909090909, + "grad_norm": 0.540793776512146, + "learning_rate": 4.15e-06, + "loss": 0.619, + "step": 83 + }, + { + "epoch": 0.5727272727272728, + "grad_norm": 0.4925222396850586, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.5838, + "step": 84 + }, + { + "epoch": 0.5795454545454546, + "grad_norm": 0.44989290833473206, + "learning_rate": 4.25e-06, + "loss": 0.5972, + "step": 85 + }, + { + "epoch": 0.5863636363636363, + "grad_norm": 0.42691388726234436, + "learning_rate": 4.3e-06, + "loss": 0.6046, + "step": 86 + }, + { + "epoch": 0.5931818181818181, + "grad_norm": 0.4512398838996887, + "learning_rate": 4.350000000000001e-06, + "loss": 0.5828, + "step": 87 + }, + { + "epoch": 0.6, + "grad_norm": 0.44399499893188477, + "learning_rate": 4.4e-06, + "loss": 0.6143, + "step": 88 + }, + { + "epoch": 0.6068181818181818, + "grad_norm": 0.46058326959609985, + "learning_rate": 4.450000000000001e-06, + "loss": 0.6117, + "step": 89 + }, + { + "epoch": 0.6136363636363636, + "grad_norm": 0.8795785903930664, + "learning_rate": 4.5e-06, + "loss": 0.6, + "step": 90 + }, + { + "epoch": 0.6204545454545455, + "grad_norm": 0.5148798227310181, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.6041, + "step": 91 + }, + { + "epoch": 0.6272727272727273, + "grad_norm": 0.48829612135887146, + "learning_rate": 4.600000000000001e-06, + "loss": 0.5492, + "step": 92 + }, + { + "epoch": 0.634090909090909, + "grad_norm": 0.5211894512176514, + "learning_rate": 4.65e-06, + "loss": 0.6044, + "step": 93 + }, + { + "epoch": 0.6409090909090909, + "grad_norm": 0.44403275847435, + "learning_rate": 4.7e-06, + "loss": 0.5929, + "step": 94 + }, + { + "epoch": 0.6477272727272727, + "grad_norm": 0.6037693619728088, + "learning_rate": 4.75e-06, + "loss": 0.6132, + "step": 95 + }, + { + "epoch": 0.6545454545454545, + "grad_norm": 0.4381515085697174, + "learning_rate": 4.800000000000001e-06, + "loss": 0.5822, + "step": 96 + }, + { + "epoch": 0.6613636363636364, + "grad_norm": 0.4997427761554718, + "learning_rate": 4.85e-06, + "loss": 0.593, + "step": 97 + }, + { + "epoch": 0.6681818181818182, + "grad_norm": 0.440571665763855, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.6072, + "step": 98 + }, + { + "epoch": 0.675, + "grad_norm": 0.5020624995231628, + "learning_rate": 4.95e-06, + "loss": 0.5861, + "step": 99 + }, + { + "epoch": 0.6818181818181818, + "grad_norm": 0.4280160963535309, + "learning_rate": 5e-06, + "loss": 0.6163, + "step": 100 + }, + { + "epoch": 0.6886363636363636, + "grad_norm": 0.4584537148475647, + "learning_rate": 4.9999795126530275e-06, + "loss": 0.6177, + "step": 101 + }, + { + "epoch": 0.6954545454545454, + "grad_norm": 0.6835049390792847, + "learning_rate": 4.999918050947891e-06, + "loss": 0.579, + "step": 102 + }, + { + "epoch": 0.7022727272727273, + "grad_norm": 0.4551607072353363, + "learning_rate": 4.999815615891943e-06, + "loss": 0.5927, + "step": 103 + }, + { + "epoch": 0.7090909090909091, + "grad_norm": 0.5893972516059875, + "learning_rate": 4.9996722091640805e-06, + "loss": 0.5775, + "step": 104 + }, + { + "epoch": 0.7159090909090909, + "grad_norm": 0.4574092924594879, + "learning_rate": 4.9994878331147225e-06, + "loss": 0.5863, + "step": 105 + }, + { + "epoch": 0.7227272727272728, + "grad_norm": 0.5354658365249634, + "learning_rate": 4.99926249076577e-06, + "loss": 0.5453, + "step": 106 + }, + { + "epoch": 0.7295454545454545, + "grad_norm": 0.4347354471683502, + "learning_rate": 4.998996185810557e-06, + "loss": 0.5913, + "step": 107 + }, + { + "epoch": 0.7363636363636363, + "grad_norm": 0.4487966299057007, + "learning_rate": 4.998688922613788e-06, + "loss": 0.5749, + "step": 108 + }, + { + "epoch": 0.7431818181818182, + "grad_norm": 0.42477577924728394, + "learning_rate": 4.9983407062114695e-06, + "loss": 0.5761, + "step": 109 + }, + { + "epoch": 0.75, + "grad_norm": 0.4801446497440338, + "learning_rate": 4.9979515423108255e-06, + "loss": 0.5977, + "step": 110 + }, + { + "epoch": 0.7568181818181818, + "grad_norm": 0.4876883029937744, + "learning_rate": 4.997521437290205e-06, + "loss": 0.5849, + "step": 111 + }, + { + "epoch": 0.7636363636363637, + "grad_norm": 0.4328872263431549, + "learning_rate": 4.997050398198977e-06, + "loss": 0.5988, + "step": 112 + }, + { + "epoch": 0.7704545454545455, + "grad_norm": 0.4332719147205353, + "learning_rate": 4.996538432757414e-06, + "loss": 0.6059, + "step": 113 + }, + { + "epoch": 0.7772727272727272, + "grad_norm": 0.4455336630344391, + "learning_rate": 4.995985549356568e-06, + "loss": 0.5725, + "step": 114 + }, + { + "epoch": 0.7840909090909091, + "grad_norm": 0.4197766184806824, + "learning_rate": 4.995391757058129e-06, + "loss": 0.5715, + "step": 115 + }, + { + "epoch": 0.7909090909090909, + "grad_norm": 0.41945499181747437, + "learning_rate": 4.99475706559428e-06, + "loss": 0.5845, + "step": 116 + }, + { + "epoch": 0.7977272727272727, + "grad_norm": 0.42414823174476624, + "learning_rate": 4.994081485367537e-06, + "loss": 0.5595, + "step": 117 + }, + { + "epoch": 0.8045454545454546, + "grad_norm": 0.4056423008441925, + "learning_rate": 4.993365027450576e-06, + "loss": 0.582, + "step": 118 + }, + { + "epoch": 0.8113636363636364, + "grad_norm": 0.4554205536842346, + "learning_rate": 4.992607703586058e-06, + "loss": 0.5591, + "step": 119 + }, + { + "epoch": 0.8181818181818182, + "grad_norm": 0.4307089149951935, + "learning_rate": 4.991809526186424e-06, + "loss": 0.5927, + "step": 120 + }, + { + "epoch": 0.825, + "grad_norm": 0.48133528232574463, + "learning_rate": 4.990970508333707e-06, + "loss": 0.5703, + "step": 121 + }, + { + "epoch": 0.8318181818181818, + "grad_norm": 0.48582738637924194, + "learning_rate": 4.990090663779305e-06, + "loss": 0.5491, + "step": 122 + }, + { + "epoch": 0.8386363636363636, + "grad_norm": 0.44926196336746216, + "learning_rate": 4.9891700069437635e-06, + "loss": 0.5803, + "step": 123 + }, + { + "epoch": 0.8454545454545455, + "grad_norm": 0.4562165141105652, + "learning_rate": 4.988208552916535e-06, + "loss": 0.5945, + "step": 124 + }, + { + "epoch": 0.8522727272727273, + "grad_norm": 0.5885360836982727, + "learning_rate": 4.987206317455734e-06, + "loss": 0.5632, + "step": 125 + }, + { + "epoch": 0.8590909090909091, + "grad_norm": 0.4525550603866577, + "learning_rate": 4.986163316987877e-06, + "loss": 0.5221, + "step": 126 + }, + { + "epoch": 0.865909090909091, + "grad_norm": 0.4545478820800781, + "learning_rate": 4.985079568607613e-06, + "loss": 0.587, + "step": 127 + }, + { + "epoch": 0.8727272727272727, + "grad_norm": 0.4364396333694458, + "learning_rate": 4.983955090077445e-06, + "loss": 0.5626, + "step": 128 + }, + { + "epoch": 0.8795454545454545, + "grad_norm": 0.4354369640350342, + "learning_rate": 4.982789899827439e-06, + "loss": 0.5674, + "step": 129 + }, + { + "epoch": 0.8863636363636364, + "grad_norm": 0.4822773337364197, + "learning_rate": 4.9815840169549216e-06, + "loss": 0.5685, + "step": 130 + }, + { + "epoch": 0.8931818181818182, + "grad_norm": 0.44673025608062744, + "learning_rate": 4.980337461224164e-06, + "loss": 0.5733, + "step": 131 + }, + { + "epoch": 0.9, + "grad_norm": 0.6294501423835754, + "learning_rate": 4.979050253066064e-06, + "loss": 0.5396, + "step": 132 + }, + { + "epoch": 0.9068181818181819, + "grad_norm": 0.42890027165412903, + "learning_rate": 4.977722413577802e-06, + "loss": 0.5718, + "step": 133 + }, + { + "epoch": 0.9136363636363637, + "grad_norm": 0.470587819814682, + "learning_rate": 4.976353964522509e-06, + "loss": 0.5705, + "step": 134 + }, + { + "epoch": 0.9204545454545454, + "grad_norm": 0.4632768929004669, + "learning_rate": 4.974944928328894e-06, + "loss": 0.5394, + "step": 135 + }, + { + "epoch": 0.9272727272727272, + "grad_norm": 0.5304691195487976, + "learning_rate": 4.973495328090891e-06, + "loss": 0.5762, + "step": 136 + }, + { + "epoch": 0.9340909090909091, + "grad_norm": 0.42036354541778564, + "learning_rate": 4.972005187567267e-06, + "loss": 0.5649, + "step": 137 + }, + { + "epoch": 0.9409090909090909, + "grad_norm": 0.45189663767814636, + "learning_rate": 4.970474531181245e-06, + "loss": 0.5554, + "step": 138 + }, + { + "epoch": 0.9477272727272728, + "grad_norm": 0.47352468967437744, + "learning_rate": 4.968903384020095e-06, + "loss": 0.5511, + "step": 139 + }, + { + "epoch": 0.9545454545454546, + "grad_norm": 0.525702953338623, + "learning_rate": 4.967291771834727e-06, + "loss": 0.5777, + "step": 140 + }, + { + "epoch": 0.9613636363636363, + "grad_norm": 0.46146854758262634, + "learning_rate": 4.965639721039267e-06, + "loss": 0.5537, + "step": 141 + }, + { + "epoch": 0.9681818181818181, + "grad_norm": 0.5200941562652588, + "learning_rate": 4.963947258710626e-06, + "loss": 0.5522, + "step": 142 + }, + { + "epoch": 0.975, + "grad_norm": 0.5260375142097473, + "learning_rate": 4.962214412588053e-06, + "loss": 0.5747, + "step": 143 + }, + { + "epoch": 0.9818181818181818, + "grad_norm": 0.518645703792572, + "learning_rate": 4.960441211072686e-06, + "loss": 0.5257, + "step": 144 + }, + { + "epoch": 0.9886363636363636, + "grad_norm": 0.4528891444206238, + "learning_rate": 4.9586276832270785e-06, + "loss": 0.5548, + "step": 145 + }, + { + "epoch": 0.9954545454545455, + "grad_norm": 0.5137557983398438, + "learning_rate": 4.9567738587747314e-06, + "loss": 0.5526, + "step": 146 + }, + { + "epoch": 1.0068181818181818, + "grad_norm": 0.8773607611656189, + "learning_rate": 4.954879768099599e-06, + "loss": 1.1, + "step": 147 + }, + { + "epoch": 1.0136363636363637, + "grad_norm": 0.39079979062080383, + "learning_rate": 4.952945442245598e-06, + "loss": 0.568, + "step": 148 + }, + { + "epoch": 1.0204545454545455, + "grad_norm": 0.45004600286483765, + "learning_rate": 4.95097091291609e-06, + "loss": 0.5383, + "step": 149 + }, + { + "epoch": 1.0272727272727273, + "grad_norm": 0.41872867941856384, + "learning_rate": 4.948956212473371e-06, + "loss": 0.5436, + "step": 150 + }, + { + "epoch": 1.0340909090909092, + "grad_norm": 0.46829020977020264, + "learning_rate": 4.946901373938132e-06, + "loss": 0.5336, + "step": 151 + }, + { + "epoch": 1.040909090909091, + "grad_norm": 0.45038115978240967, + "learning_rate": 4.944806430988927e-06, + "loss": 0.538, + "step": 152 + }, + { + "epoch": 1.0477272727272728, + "grad_norm": 0.6496132016181946, + "learning_rate": 4.942671417961615e-06, + "loss": 0.5459, + "step": 153 + }, + { + "epoch": 1.0545454545454545, + "grad_norm": 0.4842703640460968, + "learning_rate": 4.940496369848795e-06, + "loss": 0.553, + "step": 154 + }, + { + "epoch": 1.0613636363636363, + "grad_norm": 0.4338674545288086, + "learning_rate": 4.938281322299243e-06, + "loss": 0.5366, + "step": 155 + }, + { + "epoch": 1.0681818181818181, + "grad_norm": 0.44630634784698486, + "learning_rate": 4.936026311617316e-06, + "loss": 0.5517, + "step": 156 + }, + { + "epoch": 1.075, + "grad_norm": 0.4606929421424866, + "learning_rate": 4.933731374762361e-06, + "loss": 0.5405, + "step": 157 + }, + { + "epoch": 1.0818181818181818, + "grad_norm": 0.48667654395103455, + "learning_rate": 4.931396549348115e-06, + "loss": 0.5376, + "step": 158 + }, + { + "epoch": 1.0886363636363636, + "grad_norm": 0.4516271650791168, + "learning_rate": 4.9290218736420795e-06, + "loss": 0.5206, + "step": 159 + }, + { + "epoch": 1.0954545454545455, + "grad_norm": 0.44719284772872925, + "learning_rate": 4.926607386564898e-06, + "loss": 0.5397, + "step": 160 + }, + { + "epoch": 1.1022727272727273, + "grad_norm": 0.40752628445625305, + "learning_rate": 4.9241531276897196e-06, + "loss": 0.5466, + "step": 161 + }, + { + "epoch": 1.1090909090909091, + "grad_norm": 0.45423609018325806, + "learning_rate": 4.921659137241544e-06, + "loss": 0.5227, + "step": 162 + }, + { + "epoch": 1.115909090909091, + "grad_norm": 0.4185822308063507, + "learning_rate": 4.919125456096574e-06, + "loss": 0.5144, + "step": 163 + }, + { + "epoch": 1.1227272727272728, + "grad_norm": 0.7753382921218872, + "learning_rate": 4.916552125781529e-06, + "loss": 0.5352, + "step": 164 + }, + { + "epoch": 1.1295454545454546, + "grad_norm": 0.47211477160453796, + "learning_rate": 4.913939188472979e-06, + "loss": 0.5505, + "step": 165 + }, + { + "epoch": 1.1363636363636362, + "grad_norm": 0.42742082476615906, + "learning_rate": 4.911286686996648e-06, + "loss": 0.529, + "step": 166 + }, + { + "epoch": 1.143181818181818, + "grad_norm": 0.41340839862823486, + "learning_rate": 4.908594664826708e-06, + "loss": 0.5292, + "step": 167 + }, + { + "epoch": 1.15, + "grad_norm": 0.44648951292037964, + "learning_rate": 4.905863166085076e-06, + "loss": 0.5757, + "step": 168 + }, + { + "epoch": 1.1568181818181817, + "grad_norm": 0.4285544753074646, + "learning_rate": 4.903092235540679e-06, + "loss": 0.5102, + "step": 169 + }, + { + "epoch": 1.1636363636363636, + "grad_norm": 0.42675015330314636, + "learning_rate": 4.900281918608732e-06, + "loss": 0.5397, + "step": 170 + }, + { + "epoch": 1.1704545454545454, + "grad_norm": 0.47636961936950684, + "learning_rate": 4.897432261349984e-06, + "loss": 0.5307, + "step": 171 + }, + { + "epoch": 1.1772727272727272, + "grad_norm": 0.4213427007198334, + "learning_rate": 4.894543310469968e-06, + "loss": 0.54, + "step": 172 + }, + { + "epoch": 1.184090909090909, + "grad_norm": 0.41153451800346375, + "learning_rate": 4.891615113318236e-06, + "loss": 0.5423, + "step": 173 + }, + { + "epoch": 1.190909090909091, + "grad_norm": 0.459185391664505, + "learning_rate": 4.888647717887582e-06, + "loss": 0.5239, + "step": 174 + }, + { + "epoch": 1.1977272727272728, + "grad_norm": 0.48279181122779846, + "learning_rate": 4.8856411728132526e-06, + "loss": 0.5336, + "step": 175 + }, + { + "epoch": 1.2045454545454546, + "grad_norm": 0.4186796247959137, + "learning_rate": 4.8825955273721524e-06, + "loss": 0.5509, + "step": 176 + }, + { + "epoch": 1.2113636363636364, + "grad_norm": 0.4345344007015228, + "learning_rate": 4.879510831482039e-06, + "loss": 0.5415, + "step": 177 + }, + { + "epoch": 1.2181818181818183, + "grad_norm": 0.461592435836792, + "learning_rate": 4.876387135700701e-06, + "loss": 0.5256, + "step": 178 + }, + { + "epoch": 1.225, + "grad_norm": 0.5960240960121155, + "learning_rate": 4.873224491225128e-06, + "loss": 0.5599, + "step": 179 + }, + { + "epoch": 1.231818181818182, + "grad_norm": 0.4662570357322693, + "learning_rate": 4.870022949890676e-06, + "loss": 0.5647, + "step": 180 + }, + { + "epoch": 1.2386363636363638, + "grad_norm": 0.4355057179927826, + "learning_rate": 4.866782564170217e-06, + "loss": 0.5056, + "step": 181 + }, + { + "epoch": 1.2454545454545454, + "grad_norm": 0.48699110746383667, + "learning_rate": 4.863503387173276e-06, + "loss": 0.5394, + "step": 182 + }, + { + "epoch": 1.2522727272727272, + "grad_norm": 0.48679423332214355, + "learning_rate": 4.860185472645161e-06, + "loss": 0.5394, + "step": 183 + }, + { + "epoch": 1.259090909090909, + "grad_norm": 0.6564393043518066, + "learning_rate": 4.856828874966086e-06, + "loss": 0.5259, + "step": 184 + }, + { + "epoch": 1.2659090909090909, + "grad_norm": 0.44762882590293884, + "learning_rate": 4.853433649150276e-06, + "loss": 0.5563, + "step": 185 + }, + { + "epoch": 1.2727272727272727, + "grad_norm": 0.5391108989715576, + "learning_rate": 4.849999850845066e-06, + "loss": 0.5326, + "step": 186 + }, + { + "epoch": 1.2795454545454545, + "grad_norm": 0.4201100170612335, + "learning_rate": 4.8465275363299905e-06, + "loss": 0.526, + "step": 187 + }, + { + "epoch": 1.2863636363636364, + "grad_norm": 0.46211543679237366, + "learning_rate": 4.84301676251586e-06, + "loss": 0.5498, + "step": 188 + }, + { + "epoch": 1.2931818181818182, + "grad_norm": 0.4325430989265442, + "learning_rate": 4.839467586943825e-06, + "loss": 0.556, + "step": 189 + }, + { + "epoch": 1.3, + "grad_norm": 0.46057143807411194, + "learning_rate": 4.835880067784441e-06, + "loss": 0.5079, + "step": 190 + }, + { + "epoch": 1.3068181818181819, + "grad_norm": 0.4920806288719177, + "learning_rate": 4.832254263836708e-06, + "loss": 0.533, + "step": 191 + }, + { + "epoch": 1.3136363636363637, + "grad_norm": 0.46693211793899536, + "learning_rate": 4.828590234527107e-06, + "loss": 0.5542, + "step": 192 + }, + { + "epoch": 1.3204545454545453, + "grad_norm": 0.4227873682975769, + "learning_rate": 4.82488803990863e-06, + "loss": 0.544, + "step": 193 + }, + { + "epoch": 1.3272727272727272, + "grad_norm": 0.44745922088623047, + "learning_rate": 4.821147740659795e-06, + "loss": 0.4962, + "step": 194 + }, + { + "epoch": 1.334090909090909, + "grad_norm": 0.46137240529060364, + "learning_rate": 4.817369398083648e-06, + "loss": 0.5359, + "step": 195 + }, + { + "epoch": 1.3409090909090908, + "grad_norm": 0.6483879685401917, + "learning_rate": 4.813553074106761e-06, + "loss": 0.5323, + "step": 196 + }, + { + "epoch": 1.3477272727272727, + "grad_norm": 0.6602574586868286, + "learning_rate": 4.809698831278217e-06, + "loss": 0.5161, + "step": 197 + }, + { + "epoch": 1.3545454545454545, + "grad_norm": 0.47758543491363525, + "learning_rate": 4.805806732768585e-06, + "loss": 0.5214, + "step": 198 + }, + { + "epoch": 1.3613636363636363, + "grad_norm": 0.4587438106536865, + "learning_rate": 4.801876842368882e-06, + "loss": 0.5208, + "step": 199 + }, + { + "epoch": 1.3681818181818182, + "grad_norm": 0.46055835485458374, + "learning_rate": 4.797909224489531e-06, + "loss": 0.5374, + "step": 200 + }, + { + "epoch": 1.375, + "grad_norm": 0.43727558851242065, + "learning_rate": 4.793903944159303e-06, + "loss": 0.5351, + "step": 201 + }, + { + "epoch": 1.3818181818181818, + "grad_norm": 0.4573572874069214, + "learning_rate": 4.789861067024253e-06, + "loss": 0.4997, + "step": 202 + }, + { + "epoch": 1.3886363636363637, + "grad_norm": 0.5921999216079712, + "learning_rate": 4.785780659346642e-06, + "loss": 0.5168, + "step": 203 + }, + { + "epoch": 1.3954545454545455, + "grad_norm": 0.47943904995918274, + "learning_rate": 4.781662788003851e-06, + "loss": 0.5513, + "step": 204 + }, + { + "epoch": 1.4022727272727273, + "grad_norm": 0.43733686208724976, + "learning_rate": 4.777507520487289e-06, + "loss": 0.5281, + "step": 205 + }, + { + "epoch": 1.4090909090909092, + "grad_norm": 0.4289107620716095, + "learning_rate": 4.773314924901281e-06, + "loss": 0.524, + "step": 206 + }, + { + "epoch": 1.415909090909091, + "grad_norm": 0.4203416109085083, + "learning_rate": 4.769085069961955e-06, + "loss": 0.5563, + "step": 207 + }, + { + "epoch": 1.4227272727272728, + "grad_norm": 0.4205201268196106, + "learning_rate": 4.764818024996117e-06, + "loss": 0.5202, + "step": 208 + }, + { + "epoch": 1.4295454545454547, + "grad_norm": 0.4296267032623291, + "learning_rate": 4.760513859940112e-06, + "loss": 0.5113, + "step": 209 + }, + { + "epoch": 1.4363636363636363, + "grad_norm": 0.3979087471961975, + "learning_rate": 4.756172645338675e-06, + "loss": 0.5313, + "step": 210 + }, + { + "epoch": 1.4431818181818181, + "grad_norm": 0.45232149958610535, + "learning_rate": 4.751794452343785e-06, + "loss": 0.5204, + "step": 211 + }, + { + "epoch": 1.45, + "grad_norm": 0.4309398829936981, + "learning_rate": 4.747379352713489e-06, + "loss": 0.5216, + "step": 212 + }, + { + "epoch": 1.4568181818181818, + "grad_norm": 0.45721709728240967, + "learning_rate": 4.7429274188107275e-06, + "loss": 0.5279, + "step": 213 + }, + { + "epoch": 1.4636363636363636, + "grad_norm": 0.46138322353363037, + "learning_rate": 4.738438723602154e-06, + "loss": 0.5287, + "step": 214 + }, + { + "epoch": 1.4704545454545455, + "grad_norm": 0.41752108931541443, + "learning_rate": 4.733913340656933e-06, + "loss": 0.5399, + "step": 215 + }, + { + "epoch": 1.4772727272727273, + "grad_norm": 0.45447197556495667, + "learning_rate": 4.729351344145536e-06, + "loss": 0.5277, + "step": 216 + }, + { + "epoch": 1.4840909090909091, + "grad_norm": 0.42524856328964233, + "learning_rate": 4.7247528088385296e-06, + "loss": 0.5103, + "step": 217 + }, + { + "epoch": 1.490909090909091, + "grad_norm": 0.46024221181869507, + "learning_rate": 4.720117810105341e-06, + "loss": 0.5172, + "step": 218 + }, + { + "epoch": 1.4977272727272728, + "grad_norm": 0.5868630409240723, + "learning_rate": 4.715446423913036e-06, + "loss": 0.5409, + "step": 219 + }, + { + "epoch": 1.5045454545454544, + "grad_norm": 0.424813836812973, + "learning_rate": 4.710738726825059e-06, + "loss": 0.5495, + "step": 220 + }, + { + "epoch": 1.5113636363636362, + "grad_norm": 0.49356943368911743, + "learning_rate": 4.705994795999991e-06, + "loss": 0.519, + "step": 221 + }, + { + "epoch": 1.518181818181818, + "grad_norm": 0.44291776418685913, + "learning_rate": 4.701214709190277e-06, + "loss": 0.5232, + "step": 222 + }, + { + "epoch": 1.525, + "grad_norm": 0.48171764612197876, + "learning_rate": 4.696398544740955e-06, + "loss": 0.5501, + "step": 223 + }, + { + "epoch": 1.5318181818181817, + "grad_norm": 0.4296175539493561, + "learning_rate": 4.69154638158837e-06, + "loss": 0.5087, + "step": 224 + }, + { + "epoch": 1.5386363636363636, + "grad_norm": 0.4715246260166168, + "learning_rate": 4.686658299258881e-06, + "loss": 0.5363, + "step": 225 + }, + { + "epoch": 1.5454545454545454, + "grad_norm": 0.44422832131385803, + "learning_rate": 4.681734377867562e-06, + "loss": 0.5371, + "step": 226 + }, + { + "epoch": 1.5522727272727272, + "grad_norm": 4.221654891967773, + "learning_rate": 4.67677469811688e-06, + "loss": 0.5191, + "step": 227 + }, + { + "epoch": 1.559090909090909, + "grad_norm": 0.41857993602752686, + "learning_rate": 4.671779341295378e-06, + "loss": 0.5191, + "step": 228 + }, + { + "epoch": 1.565909090909091, + "grad_norm": 0.4950161278247833, + "learning_rate": 4.666748389276344e-06, + "loss": 0.5311, + "step": 229 + }, + { + "epoch": 1.5727272727272728, + "grad_norm": 0.44641605019569397, + "learning_rate": 4.661681924516466e-06, + "loss": 0.5174, + "step": 230 + }, + { + "epoch": 1.5795454545454546, + "grad_norm": 0.39064037799835205, + "learning_rate": 4.6565800300544805e-06, + "loss": 0.515, + "step": 231 + }, + { + "epoch": 1.5863636363636364, + "grad_norm": 0.7764399647712708, + "learning_rate": 4.651442789509813e-06, + "loss": 0.5282, + "step": 232 + }, + { + "epoch": 1.5931818181818183, + "grad_norm": 0.44421035051345825, + "learning_rate": 4.646270287081208e-06, + "loss": 0.5324, + "step": 233 + }, + { + "epoch": 1.6, + "grad_norm": 0.5149928331375122, + "learning_rate": 4.641062607545347e-06, + "loss": 0.5264, + "step": 234 + }, + { + "epoch": 1.606818181818182, + "grad_norm": 0.4784093201160431, + "learning_rate": 4.6358198362554585e-06, + "loss": 0.552, + "step": 235 + }, + { + "epoch": 1.6136363636363638, + "grad_norm": 0.4138723313808441, + "learning_rate": 4.630542059139923e-06, + "loss": 0.5202, + "step": 236 + }, + { + "epoch": 1.6204545454545456, + "grad_norm": 0.4509594440460205, + "learning_rate": 4.625229362700863e-06, + "loss": 0.5318, + "step": 237 + }, + { + "epoch": 1.6272727272727274, + "grad_norm": 0.44979560375213623, + "learning_rate": 4.61988183401272e-06, + "loss": 0.521, + "step": 238 + }, + { + "epoch": 1.634090909090909, + "grad_norm": 0.4525575041770935, + "learning_rate": 4.614499560720837e-06, + "loss": 0.5073, + "step": 239 + }, + { + "epoch": 1.6409090909090909, + "grad_norm": 0.44126543402671814, + "learning_rate": 4.609082631040012e-06, + "loss": 0.5058, + "step": 240 + }, + { + "epoch": 1.6477272727272727, + "grad_norm": 0.887867271900177, + "learning_rate": 4.603631133753061e-06, + "loss": 0.5471, + "step": 241 + }, + { + "epoch": 1.6545454545454545, + "grad_norm": 0.4497375786304474, + "learning_rate": 4.598145158209356e-06, + "loss": 0.501, + "step": 242 + }, + { + "epoch": 1.6613636363636364, + "grad_norm": 0.45681032538414, + "learning_rate": 4.592624794323366e-06, + "loss": 0.5306, + "step": 243 + }, + { + "epoch": 1.6681818181818182, + "grad_norm": 0.5186678171157837, + "learning_rate": 4.587070132573178e-06, + "loss": 0.5097, + "step": 244 + }, + { + "epoch": 1.675, + "grad_norm": 0.44289669394493103, + "learning_rate": 4.581481263999019e-06, + "loss": 0.5085, + "step": 245 + }, + { + "epoch": 1.6818181818181817, + "grad_norm": 0.4571002125740051, + "learning_rate": 4.575858280201761e-06, + "loss": 0.5327, + "step": 246 + }, + { + "epoch": 1.6886363636363635, + "grad_norm": 0.43753257393836975, + "learning_rate": 4.570201273341418e-06, + "loss": 0.5167, + "step": 247 + }, + { + "epoch": 1.6954545454545453, + "grad_norm": 0.44994574785232544, + "learning_rate": 4.564510336135642e-06, + "loss": 0.5554, + "step": 248 + }, + { + "epoch": 1.7022727272727272, + "grad_norm": 0.6185874938964844, + "learning_rate": 4.558785561858196e-06, + "loss": 0.5115, + "step": 249 + }, + { + "epoch": 1.709090909090909, + "grad_norm": 0.4883944094181061, + "learning_rate": 4.5530270443374305e-06, + "loss": 0.5312, + "step": 250 + }, + { + "epoch": 1.7159090909090908, + "grad_norm": 0.46433499455451965, + "learning_rate": 4.547234877954741e-06, + "loss": 0.5368, + "step": 251 + }, + { + "epoch": 1.7227272727272727, + "grad_norm": 0.4270826578140259, + "learning_rate": 4.541409157643027e-06, + "loss": 0.5258, + "step": 252 + }, + { + "epoch": 1.7295454545454545, + "grad_norm": 0.4390241205692291, + "learning_rate": 4.535549978885132e-06, + "loss": 0.5259, + "step": 253 + }, + { + "epoch": 1.7363636363636363, + "grad_norm": 0.42756086587905884, + "learning_rate": 4.5296574377122765e-06, + "loss": 0.5217, + "step": 254 + }, + { + "epoch": 1.7431818181818182, + "grad_norm": 0.48124927282333374, + "learning_rate": 4.5237316307024895e-06, + "loss": 0.5095, + "step": 255 + }, + { + "epoch": 1.75, + "grad_norm": 0.4394672214984894, + "learning_rate": 4.517772654979024e-06, + "loss": 0.5116, + "step": 256 + }, + { + "epoch": 1.7568181818181818, + "grad_norm": 0.5805886387825012, + "learning_rate": 4.51178060820876e-06, + "loss": 0.5145, + "step": 257 + }, + { + "epoch": 1.7636363636363637, + "grad_norm": 0.4845556616783142, + "learning_rate": 4.505755588600613e-06, + "loss": 0.5064, + "step": 258 + }, + { + "epoch": 1.7704545454545455, + "grad_norm": 0.45300963521003723, + "learning_rate": 4.499697694903915e-06, + "loss": 0.5362, + "step": 259 + }, + { + "epoch": 1.7772727272727273, + "grad_norm": 0.45417436957359314, + "learning_rate": 4.493607026406802e-06, + "loss": 0.5234, + "step": 260 + }, + { + "epoch": 1.7840909090909092, + "grad_norm": 0.4577132761478424, + "learning_rate": 4.487483682934587e-06, + "loss": 0.5271, + "step": 261 + }, + { + "epoch": 1.790909090909091, + "grad_norm": 0.4518176019191742, + "learning_rate": 4.481327764848118e-06, + "loss": 0.4853, + "step": 262 + }, + { + "epoch": 1.7977272727272728, + "grad_norm": 0.43804875016212463, + "learning_rate": 4.47513937304214e-06, + "loss": 0.5188, + "step": 263 + }, + { + "epoch": 1.8045454545454547, + "grad_norm": 0.4401354193687439, + "learning_rate": 4.4689186089436365e-06, + "loss": 0.5079, + "step": 264 + }, + { + "epoch": 1.8113636363636365, + "grad_norm": 0.4051227569580078, + "learning_rate": 4.462665574510169e-06, + "loss": 0.5084, + "step": 265 + }, + { + "epoch": 1.8181818181818183, + "grad_norm": 0.5269004106521606, + "learning_rate": 4.456380372228208e-06, + "loss": 0.528, + "step": 266 + }, + { + "epoch": 1.825, + "grad_norm": 0.44371986389160156, + "learning_rate": 4.450063105111447e-06, + "loss": 0.5388, + "step": 267 + }, + { + "epoch": 1.8318181818181818, + "grad_norm": 0.41871729493141174, + "learning_rate": 4.443713876699124e-06, + "loss": 0.502, + "step": 268 + }, + { + "epoch": 1.8386363636363636, + "grad_norm": 0.464652419090271, + "learning_rate": 4.4373327910543125e-06, + "loss": 0.5045, + "step": 269 + }, + { + "epoch": 1.8454545454545455, + "grad_norm": 0.43198078870773315, + "learning_rate": 4.430919952762226e-06, + "loss": 0.5149, + "step": 270 + }, + { + "epoch": 1.8522727272727273, + "grad_norm": 0.42869722843170166, + "learning_rate": 4.424475466928499e-06, + "loss": 0.5329, + "step": 271 + }, + { + "epoch": 1.8590909090909091, + "grad_norm": 0.529680609703064, + "learning_rate": 4.417999439177465e-06, + "loss": 0.4939, + "step": 272 + }, + { + "epoch": 1.865909090909091, + "grad_norm": 0.4562613368034363, + "learning_rate": 4.4114919756504275e-06, + "loss": 0.5389, + "step": 273 + }, + { + "epoch": 1.8727272727272726, + "grad_norm": 0.4480995237827301, + "learning_rate": 4.404953183003916e-06, + "loss": 0.51, + "step": 274 + }, + { + "epoch": 1.8795454545454544, + "grad_norm": 0.41215088963508606, + "learning_rate": 4.398383168407941e-06, + "loss": 0.5386, + "step": 275 + }, + { + "epoch": 1.8863636363636362, + "grad_norm": 0.8426409363746643, + "learning_rate": 4.391782039544239e-06, + "loss": 0.498, + "step": 276 + }, + { + "epoch": 1.893181818181818, + "grad_norm": 0.5359936356544495, + "learning_rate": 4.385149904604502e-06, + "loss": 0.51, + "step": 277 + }, + { + "epoch": 1.9, + "grad_norm": 0.4462437033653259, + "learning_rate": 4.378486872288611e-06, + "loss": 0.5082, + "step": 278 + }, + { + "epoch": 1.9068181818181817, + "grad_norm": 0.445359468460083, + "learning_rate": 4.371793051802849e-06, + "loss": 0.5171, + "step": 279 + }, + { + "epoch": 1.9136363636363636, + "grad_norm": 0.4223516583442688, + "learning_rate": 4.365068552858116e-06, + "loss": 0.5218, + "step": 280 + }, + { + "epoch": 1.9204545454545454, + "grad_norm": 0.43929681181907654, + "learning_rate": 4.358313485668124e-06, + "loss": 0.5156, + "step": 281 + }, + { + "epoch": 1.9272727272727272, + "grad_norm": 0.42535650730133057, + "learning_rate": 4.3515279609476e-06, + "loss": 0.4964, + "step": 282 + }, + { + "epoch": 1.934090909090909, + "grad_norm": 0.46448391675949097, + "learning_rate": 4.3447120899104615e-06, + "loss": 0.5182, + "step": 283 + }, + { + "epoch": 1.940909090909091, + "grad_norm": 0.4082206189632416, + "learning_rate": 4.337865984268002e-06, + "loss": 0.5175, + "step": 284 + }, + { + "epoch": 1.9477272727272728, + "grad_norm": 0.42423564195632935, + "learning_rate": 4.3309897562270525e-06, + "loss": 0.5352, + "step": 285 + }, + { + "epoch": 1.9545454545454546, + "grad_norm": 0.5240079760551453, + "learning_rate": 4.324083518488151e-06, + "loss": 0.5157, + "step": 286 + }, + { + "epoch": 1.9613636363636364, + "grad_norm": 0.4517645835876465, + "learning_rate": 4.317147384243688e-06, + "loss": 0.5399, + "step": 287 + }, + { + "epoch": 1.9681818181818183, + "grad_norm": 0.44054290652275085, + "learning_rate": 4.3101814671760546e-06, + "loss": 0.5293, + "step": 288 + }, + { + "epoch": 1.975, + "grad_norm": 0.4256063401699066, + "learning_rate": 4.303185881455778e-06, + "loss": 0.5323, + "step": 289 + }, + { + "epoch": 1.981818181818182, + "grad_norm": 0.43106094002723694, + "learning_rate": 4.296160741739652e-06, + "loss": 0.5259, + "step": 290 + }, + { + "epoch": 1.9886363636363638, + "grad_norm": 0.43816477060317993, + "learning_rate": 4.289106163168858e-06, + "loss": 0.5098, + "step": 291 + }, + { + "epoch": 1.9954545454545456, + "grad_norm": 1.1177959442138672, + "learning_rate": 4.282022261367074e-06, + "loss": 0.513, + "step": 292 + }, + { + "epoch": 2.006818181818182, + "grad_norm": 0.8377058506011963, + "learning_rate": 4.274909152438582e-06, + "loss": 0.9984, + "step": 293 + }, + { + "epoch": 2.0136363636363637, + "grad_norm": 0.4339721202850342, + "learning_rate": 4.267766952966369e-06, + "loss": 0.493, + "step": 294 + }, + { + "epoch": 2.0204545454545455, + "grad_norm": 0.39961451292037964, + "learning_rate": 4.260595780010209e-06, + "loss": 0.486, + "step": 295 + }, + { + "epoch": 2.0272727272727273, + "grad_norm": 0.4809112548828125, + "learning_rate": 4.2533957511047485e-06, + "loss": 0.4803, + "step": 296 + }, + { + "epoch": 2.034090909090909, + "grad_norm": 0.414700984954834, + "learning_rate": 4.24616698425758e-06, + "loss": 0.4566, + "step": 297 + }, + { + "epoch": 2.040909090909091, + "grad_norm": 0.552201509475708, + "learning_rate": 4.238909597947307e-06, + "loss": 0.4775, + "step": 298 + }, + { + "epoch": 2.047727272727273, + "grad_norm": 0.4647877812385559, + "learning_rate": 4.231623711121603e-06, + "loss": 0.4944, + "step": 299 + }, + { + "epoch": 2.0545454545454547, + "grad_norm": 0.684251606464386, + "learning_rate": 4.224309443195261e-06, + "loss": 0.495, + "step": 300 + }, + { + "epoch": 2.0613636363636365, + "grad_norm": 0.42145106196403503, + "learning_rate": 4.2169669140482365e-06, + "loss": 0.4809, + "step": 301 + }, + { + "epoch": 2.0681818181818183, + "grad_norm": 0.43771296739578247, + "learning_rate": 4.2095962440236846e-06, + "loss": 0.5033, + "step": 302 + }, + { + "epoch": 2.075, + "grad_norm": 0.4323962926864624, + "learning_rate": 4.202197553925983e-06, + "loss": 0.4843, + "step": 303 + }, + { + "epoch": 2.081818181818182, + "grad_norm": 0.40169817209243774, + "learning_rate": 4.194770965018758e-06, + "loss": 0.504, + "step": 304 + }, + { + "epoch": 2.088636363636364, + "grad_norm": 0.4206427037715912, + "learning_rate": 4.187316599022892e-06, + "loss": 0.5059, + "step": 305 + }, + { + "epoch": 2.0954545454545457, + "grad_norm": 0.4550352692604065, + "learning_rate": 4.179834578114531e-06, + "loss": 0.4717, + "step": 306 + }, + { + "epoch": 2.102272727272727, + "grad_norm": 0.4193796217441559, + "learning_rate": 4.172325024923083e-06, + "loss": 0.5018, + "step": 307 + }, + { + "epoch": 2.109090909090909, + "grad_norm": 0.550195038318634, + "learning_rate": 4.164788062529203e-06, + "loss": 0.4961, + "step": 308 + }, + { + "epoch": 2.1159090909090907, + "grad_norm": 0.43539106845855713, + "learning_rate": 4.157223814462784e-06, + "loss": 0.4909, + "step": 309 + }, + { + "epoch": 2.1227272727272726, + "grad_norm": 0.4565756022930145, + "learning_rate": 4.149632404700925e-06, + "loss": 0.49, + "step": 310 + }, + { + "epoch": 2.1295454545454544, + "grad_norm": 0.4206918179988861, + "learning_rate": 4.142013957665903e-06, + "loss": 0.4722, + "step": 311 + }, + { + "epoch": 2.1363636363636362, + "grad_norm": 0.4098123013973236, + "learning_rate": 4.134368598223132e-06, + "loss": 0.4791, + "step": 312 + }, + { + "epoch": 2.143181818181818, + "grad_norm": 0.4387701153755188, + "learning_rate": 4.126696451679116e-06, + "loss": 0.4758, + "step": 313 + }, + { + "epoch": 2.15, + "grad_norm": 0.4594134986400604, + "learning_rate": 4.118997643779401e-06, + "loss": 0.5009, + "step": 314 + }, + { + "epoch": 2.1568181818181817, + "grad_norm": 0.4581888020038605, + "learning_rate": 4.111272300706502e-06, + "loss": 0.4915, + "step": 315 + }, + { + "epoch": 2.1636363636363636, + "grad_norm": 0.5279524326324463, + "learning_rate": 4.1035205490778505e-06, + "loss": 0.5056, + "step": 316 + }, + { + "epoch": 2.1704545454545454, + "grad_norm": 0.4253177344799042, + "learning_rate": 4.095742515943703e-06, + "loss": 0.5001, + "step": 317 + }, + { + "epoch": 2.1772727272727272, + "grad_norm": 0.5294811129570007, + "learning_rate": 4.087938328785071e-06, + "loss": 0.5027, + "step": 318 + }, + { + "epoch": 2.184090909090909, + "grad_norm": 0.4642890989780426, + "learning_rate": 4.080108115511629e-06, + "loss": 0.5051, + "step": 319 + }, + { + "epoch": 2.190909090909091, + "grad_norm": 0.43700429797172546, + "learning_rate": 4.072252004459612e-06, + "loss": 0.4809, + "step": 320 + }, + { + "epoch": 2.1977272727272728, + "grad_norm": 0.4536914825439453, + "learning_rate": 4.064370124389718e-06, + "loss": 0.5129, + "step": 321 + }, + { + "epoch": 2.2045454545454546, + "grad_norm": 0.42350664734840393, + "learning_rate": 4.056462604484998e-06, + "loss": 0.4912, + "step": 322 + }, + { + "epoch": 2.2113636363636364, + "grad_norm": 0.47886520624160767, + "learning_rate": 4.048529574348734e-06, + "loss": 0.4757, + "step": 323 + }, + { + "epoch": 2.2181818181818183, + "grad_norm": 0.5205736756324768, + "learning_rate": 4.040571164002319e-06, + "loss": 0.4821, + "step": 324 + }, + { + "epoch": 2.225, + "grad_norm": 0.4465322196483612, + "learning_rate": 4.032587503883124e-06, + "loss": 0.5057, + "step": 325 + }, + { + "epoch": 2.231818181818182, + "grad_norm": 0.41364315152168274, + "learning_rate": 4.0245787248423614e-06, + "loss": 0.4809, + "step": 326 + }, + { + "epoch": 2.2386363636363638, + "grad_norm": 0.42846816778182983, + "learning_rate": 4.0165449581429404e-06, + "loss": 0.4522, + "step": 327 + }, + { + "epoch": 2.2454545454545456, + "grad_norm": 0.4604497253894806, + "learning_rate": 4.008486335457312e-06, + "loss": 0.4879, + "step": 328 + }, + { + "epoch": 2.2522727272727274, + "grad_norm": 0.45866841077804565, + "learning_rate": 4.000402988865316e-06, + "loss": 0.4969, + "step": 329 + }, + { + "epoch": 2.2590909090909093, + "grad_norm": 0.5116038918495178, + "learning_rate": 3.992295050852013e-06, + "loss": 0.5015, + "step": 330 + }, + { + "epoch": 2.265909090909091, + "grad_norm": 0.4389069080352783, + "learning_rate": 3.984162654305516e-06, + "loss": 0.4797, + "step": 331 + }, + { + "epoch": 2.2727272727272725, + "grad_norm": 0.45724064111709595, + "learning_rate": 3.976005932514807e-06, + "loss": 0.4915, + "step": 332 + }, + { + "epoch": 2.2795454545454543, + "grad_norm": 0.44370219111442566, + "learning_rate": 3.967825019167559e-06, + "loss": 0.499, + "step": 333 + }, + { + "epoch": 2.286363636363636, + "grad_norm": 0.4414970278739929, + "learning_rate": 3.959620048347938e-06, + "loss": 0.447, + "step": 334 + }, + { + "epoch": 2.293181818181818, + "grad_norm": 0.7909289598464966, + "learning_rate": 3.951391154534415e-06, + "loss": 0.4658, + "step": 335 + }, + { + "epoch": 2.3, + "grad_norm": 0.4643534719944, + "learning_rate": 3.943138472597549e-06, + "loss": 0.4941, + "step": 336 + }, + { + "epoch": 2.3068181818181817, + "grad_norm": 0.5114858150482178, + "learning_rate": 3.934862137797788e-06, + "loss": 0.4733, + "step": 337 + }, + { + "epoch": 2.3136363636363635, + "grad_norm": 0.4057219624519348, + "learning_rate": 3.9265622857832455e-06, + "loss": 0.4965, + "step": 338 + }, + { + "epoch": 2.3204545454545453, + "grad_norm": 0.4124796390533447, + "learning_rate": 3.918239052587481e-06, + "loss": 0.4508, + "step": 339 + }, + { + "epoch": 2.327272727272727, + "grad_norm": 0.43895164132118225, + "learning_rate": 3.909892574627267e-06, + "loss": 0.4939, + "step": 340 + }, + { + "epoch": 2.334090909090909, + "grad_norm": 0.47555556893348694, + "learning_rate": 3.901522988700355e-06, + "loss": 0.4541, + "step": 341 + }, + { + "epoch": 2.340909090909091, + "grad_norm": 0.44449353218078613, + "learning_rate": 3.893130431983234e-06, + "loss": 0.4883, + "step": 342 + }, + { + "epoch": 2.3477272727272727, + "grad_norm": 0.441793829202652, + "learning_rate": 3.884715042028882e-06, + "loss": 0.4911, + "step": 343 + }, + { + "epoch": 2.3545454545454545, + "grad_norm": 0.4836761951446533, + "learning_rate": 3.876276956764509e-06, + "loss": 0.4838, + "step": 344 + }, + { + "epoch": 2.3613636363636363, + "grad_norm": 0.7246759533882141, + "learning_rate": 3.867816314489301e-06, + "loss": 0.5014, + "step": 345 + }, + { + "epoch": 2.368181818181818, + "grad_norm": 0.5535604357719421, + "learning_rate": 3.8593332538721465e-06, + "loss": 0.4654, + "step": 346 + }, + { + "epoch": 2.375, + "grad_norm": 0.4468318819999695, + "learning_rate": 3.8508279139493736e-06, + "loss": 0.463, + "step": 347 + }, + { + "epoch": 2.381818181818182, + "grad_norm": 0.4505653381347656, + "learning_rate": 3.84230043412246e-06, + "loss": 0.4614, + "step": 348 + }, + { + "epoch": 2.3886363636363637, + "grad_norm": 0.4461449384689331, + "learning_rate": 3.833750954155757e-06, + "loss": 0.4777, + "step": 349 + }, + { + "epoch": 2.3954545454545455, + "grad_norm": 0.423711895942688, + "learning_rate": 3.825179614174195e-06, + "loss": 0.4581, + "step": 350 + }, + { + "epoch": 2.4022727272727273, + "grad_norm": 0.462819904088974, + "learning_rate": 3.816586554660987e-06, + "loss": 0.4782, + "step": 351 + }, + { + "epoch": 2.409090909090909, + "grad_norm": 0.44206613302230835, + "learning_rate": 3.807971916455325e-06, + "loss": 0.477, + "step": 352 + }, + { + "epoch": 2.415909090909091, + "grad_norm": 0.4178541600704193, + "learning_rate": 3.799335840750077e-06, + "loss": 0.4981, + "step": 353 + }, + { + "epoch": 2.422727272727273, + "grad_norm": 0.670315682888031, + "learning_rate": 3.790678469089465e-06, + "loss": 0.4762, + "step": 354 + }, + { + "epoch": 2.4295454545454547, + "grad_norm": 0.44383955001831055, + "learning_rate": 3.7819999433667503e-06, + "loss": 0.4807, + "step": 355 + }, + { + "epoch": 2.4363636363636365, + "grad_norm": 0.4850422143936157, + "learning_rate": 3.773300405821908e-06, + "loss": 0.479, + "step": 356 + }, + { + "epoch": 2.4431818181818183, + "grad_norm": 0.458435595035553, + "learning_rate": 3.764579999039293e-06, + "loss": 0.4628, + "step": 357 + }, + { + "epoch": 2.45, + "grad_norm": 0.43913447856903076, + "learning_rate": 3.7558388659453052e-06, + "loss": 0.4948, + "step": 358 + }, + { + "epoch": 2.456818181818182, + "grad_norm": 0.4294499456882477, + "learning_rate": 3.7470771498060455e-06, + "loss": 0.4979, + "step": 359 + }, + { + "epoch": 2.463636363636364, + "grad_norm": 0.4472196400165558, + "learning_rate": 3.7382949942249695e-06, + "loss": 0.496, + "step": 360 + }, + { + "epoch": 2.4704545454545457, + "grad_norm": 0.4681987762451172, + "learning_rate": 3.7294925431405306e-06, + "loss": 0.4591, + "step": 361 + }, + { + "epoch": 2.4772727272727275, + "grad_norm": 0.40683335065841675, + "learning_rate": 3.720669940823827e-06, + "loss": 0.4755, + "step": 362 + }, + { + "epoch": 2.484090909090909, + "grad_norm": 0.46134939789772034, + "learning_rate": 3.7118273318762275e-06, + "loss": 0.4956, + "step": 363 + }, + { + "epoch": 2.4909090909090907, + "grad_norm": 0.4279940724372864, + "learning_rate": 3.702964861227013e-06, + "loss": 0.4815, + "step": 364 + }, + { + "epoch": 2.4977272727272726, + "grad_norm": 0.4699287414550781, + "learning_rate": 3.694082674130991e-06, + "loss": 0.4846, + "step": 365 + }, + { + "epoch": 2.5045454545454544, + "grad_norm": 0.4230785369873047, + "learning_rate": 3.6851809161661206e-06, + "loss": 0.5077, + "step": 366 + }, + { + "epoch": 2.5113636363636362, + "grad_norm": 0.4509921967983246, + "learning_rate": 3.6762597332311254e-06, + "loss": 0.5134, + "step": 367 + }, + { + "epoch": 2.518181818181818, + "grad_norm": 0.45494896173477173, + "learning_rate": 3.6673192715431016e-06, + "loss": 0.4685, + "step": 368 + }, + { + "epoch": 2.525, + "grad_norm": 0.4804443120956421, + "learning_rate": 3.658359677635122e-06, + "loss": 0.4458, + "step": 369 + }, + { + "epoch": 2.5318181818181817, + "grad_norm": 0.4149218499660492, + "learning_rate": 3.649381098353834e-06, + "loss": 0.5102, + "step": 370 + }, + { + "epoch": 2.5386363636363636, + "grad_norm": 0.4423210024833679, + "learning_rate": 3.6403836808570512e-06, + "loss": 0.5034, + "step": 371 + }, + { + "epoch": 2.5454545454545454, + "grad_norm": 0.4915727376937866, + "learning_rate": 3.631367572611348e-06, + "loss": 0.4889, + "step": 372 + }, + { + "epoch": 2.5522727272727272, + "grad_norm": 0.4189188778400421, + "learning_rate": 3.6223329213896313e-06, + "loss": 0.4829, + "step": 373 + }, + { + "epoch": 2.559090909090909, + "grad_norm": 0.511681079864502, + "learning_rate": 3.613279875268731e-06, + "loss": 0.4786, + "step": 374 + }, + { + "epoch": 2.565909090909091, + "grad_norm": 0.430137038230896, + "learning_rate": 3.604208582626964e-06, + "loss": 0.4783, + "step": 375 + }, + { + "epoch": 2.5727272727272728, + "grad_norm": 0.5629504919052124, + "learning_rate": 3.5951191921417063e-06, + "loss": 0.4438, + "step": 376 + }, + { + "epoch": 2.5795454545454546, + "grad_norm": 0.4354007840156555, + "learning_rate": 3.586011852786955e-06, + "loss": 0.4869, + "step": 377 + }, + { + "epoch": 2.5863636363636364, + "grad_norm": 0.45930856466293335, + "learning_rate": 3.5768867138308872e-06, + "loss": 0.4683, + "step": 378 + }, + { + "epoch": 2.5931818181818183, + "grad_norm": 0.4965468943119049, + "learning_rate": 3.5677439248334133e-06, + "loss": 0.4739, + "step": 379 + }, + { + "epoch": 2.6, + "grad_norm": 0.39618629217147827, + "learning_rate": 3.5585836356437266e-06, + "loss": 0.486, + "step": 380 + }, + { + "epoch": 2.606818181818182, + "grad_norm": 0.504718542098999, + "learning_rate": 3.5494059963978433e-06, + "loss": 0.4608, + "step": 381 + }, + { + "epoch": 2.6136363636363638, + "grad_norm": 0.3951753079891205, + "learning_rate": 3.540211157516149e-06, + "loss": 0.4973, + "step": 382 + }, + { + "epoch": 2.6204545454545456, + "grad_norm": 0.45937463641166687, + "learning_rate": 3.530999269700927e-06, + "loss": 0.4751, + "step": 383 + }, + { + "epoch": 2.6272727272727274, + "grad_norm": 0.4644455909729004, + "learning_rate": 3.521770483933891e-06, + "loss": 0.5003, + "step": 384 + }, + { + "epoch": 2.634090909090909, + "grad_norm": 0.4427309036254883, + "learning_rate": 3.5125249514737093e-06, + "loss": 0.4779, + "step": 385 + }, + { + "epoch": 2.6409090909090907, + "grad_norm": 0.4289201498031616, + "learning_rate": 3.503262823853527e-06, + "loss": 0.48, + "step": 386 + }, + { + "epoch": 2.6477272727272725, + "grad_norm": 0.5118290185928345, + "learning_rate": 3.493984252878483e-06, + "loss": 0.4858, + "step": 387 + }, + { + "epoch": 2.6545454545454543, + "grad_norm": 0.4599657952785492, + "learning_rate": 3.484689390623218e-06, + "loss": 0.4813, + "step": 388 + }, + { + "epoch": 2.661363636363636, + "grad_norm": 0.5077105164527893, + "learning_rate": 3.4753783894293886e-06, + "loss": 0.4732, + "step": 389 + }, + { + "epoch": 2.668181818181818, + "grad_norm": 0.5006387829780579, + "learning_rate": 3.466051401903162e-06, + "loss": 0.4743, + "step": 390 + }, + { + "epoch": 2.675, + "grad_norm": 0.4769401252269745, + "learning_rate": 3.4567085809127247e-06, + "loss": 0.4941, + "step": 391 + }, + { + "epoch": 2.6818181818181817, + "grad_norm": 0.4538915455341339, + "learning_rate": 3.4473500795857674e-06, + "loss": 0.4679, + "step": 392 + }, + { + "epoch": 2.6886363636363635, + "grad_norm": 0.4370926320552826, + "learning_rate": 3.4379760513069804e-06, + "loss": 0.4866, + "step": 393 + }, + { + "epoch": 2.6954545454545453, + "grad_norm": 0.445913165807724, + "learning_rate": 3.428586649715542e-06, + "loss": 0.4914, + "step": 394 + }, + { + "epoch": 2.702272727272727, + "grad_norm": 0.45687681436538696, + "learning_rate": 3.4191820287025916e-06, + "loss": 0.474, + "step": 395 + }, + { + "epoch": 2.709090909090909, + "grad_norm": 1.1952316761016846, + "learning_rate": 3.4097623424087196e-06, + "loss": 0.4905, + "step": 396 + }, + { + "epoch": 2.715909090909091, + "grad_norm": 0.41248032450675964, + "learning_rate": 3.4003277452214284e-06, + "loss": 0.4753, + "step": 397 + }, + { + "epoch": 2.7227272727272727, + "grad_norm": 0.4349999725818634, + "learning_rate": 3.3908783917726123e-06, + "loss": 0.4712, + "step": 398 + }, + { + "epoch": 2.7295454545454545, + "grad_norm": 0.42977845668792725, + "learning_rate": 3.381414436936018e-06, + "loss": 0.4936, + "step": 399 + }, + { + "epoch": 2.7363636363636363, + "grad_norm": 0.446321040391922, + "learning_rate": 3.3719360358247054e-06, + "loss": 0.4795, + "step": 400 + }, + { + "epoch": 2.743181818181818, + "grad_norm": 0.4957369565963745, + "learning_rate": 3.36244334378851e-06, + "loss": 0.48, + "step": 401 + }, + { + "epoch": 2.75, + "grad_norm": 0.43430694937705994, + "learning_rate": 3.3529365164114903e-06, + "loss": 0.4903, + "step": 402 + }, + { + "epoch": 2.756818181818182, + "grad_norm": 0.4340098202228546, + "learning_rate": 3.3434157095093846e-06, + "loss": 0.5111, + "step": 403 + }, + { + "epoch": 2.7636363636363637, + "grad_norm": 0.4309055805206299, + "learning_rate": 3.333881079127052e-06, + "loss": 0.4909, + "step": 404 + }, + { + "epoch": 2.7704545454545455, + "grad_norm": 0.43131211400032043, + "learning_rate": 3.3243327815359168e-06, + "loss": 0.4875, + "step": 405 + }, + { + "epoch": 2.7772727272727273, + "grad_norm": 0.4794235825538635, + "learning_rate": 3.314770973231408e-06, + "loss": 0.4766, + "step": 406 + }, + { + "epoch": 2.784090909090909, + "grad_norm": 0.4632478356361389, + "learning_rate": 3.305195810930393e-06, + "loss": 0.4765, + "step": 407 + }, + { + "epoch": 2.790909090909091, + "grad_norm": 0.4836251139640808, + "learning_rate": 3.2956074515686105e-06, + "loss": 0.4573, + "step": 408 + }, + { + "epoch": 2.797727272727273, + "grad_norm": 0.44514402747154236, + "learning_rate": 3.2860060522980945e-06, + "loss": 0.4794, + "step": 409 + }, + { + "epoch": 2.8045454545454547, + "grad_norm": 0.446785569190979, + "learning_rate": 3.276391770484606e-06, + "loss": 0.482, + "step": 410 + }, + { + "epoch": 2.8113636363636365, + "grad_norm": 0.4410347044467926, + "learning_rate": 3.266764763705046e-06, + "loss": 0.5136, + "step": 411 + }, + { + "epoch": 2.8181818181818183, + "grad_norm": 0.4888826906681061, + "learning_rate": 3.257125189744877e-06, + "loss": 0.4852, + "step": 412 + }, + { + "epoch": 2.825, + "grad_norm": 0.4502141773700714, + "learning_rate": 3.247473206595536e-06, + "loss": 0.4764, + "step": 413 + }, + { + "epoch": 2.831818181818182, + "grad_norm": 0.5693864822387695, + "learning_rate": 3.2378089724518464e-06, + "loss": 0.5016, + "step": 414 + }, + { + "epoch": 2.838636363636364, + "grad_norm": 0.49588093161582947, + "learning_rate": 3.228132645709421e-06, + "loss": 0.4786, + "step": 415 + }, + { + "epoch": 2.8454545454545457, + "grad_norm": 0.5465879440307617, + "learning_rate": 3.218444384962071e-06, + "loss": 0.4857, + "step": 416 + }, + { + "epoch": 2.8522727272727275, + "grad_norm": 0.4633638858795166, + "learning_rate": 3.2087443489992043e-06, + "loss": 0.475, + "step": 417 + }, + { + "epoch": 2.8590909090909093, + "grad_norm": 0.47139137983322144, + "learning_rate": 3.1990326968032225e-06, + "loss": 0.5162, + "step": 418 + }, + { + "epoch": 2.865909090909091, + "grad_norm": 0.45505622029304504, + "learning_rate": 3.189309587546917e-06, + "loss": 0.4699, + "step": 419 + }, + { + "epoch": 2.8727272727272726, + "grad_norm": 0.4410839080810547, + "learning_rate": 3.1795751805908578e-06, + "loss": 0.4734, + "step": 420 + }, + { + "epoch": 2.8795454545454544, + "grad_norm": 0.47172659635543823, + "learning_rate": 3.169829635480783e-06, + "loss": 0.4896, + "step": 421 + }, + { + "epoch": 2.8863636363636362, + "grad_norm": 0.4332088828086853, + "learning_rate": 3.160073111944983e-06, + "loss": 0.4898, + "step": 422 + }, + { + "epoch": 2.893181818181818, + "grad_norm": 0.4303777813911438, + "learning_rate": 3.150305769891686e-06, + "loss": 0.4663, + "step": 423 + }, + { + "epoch": 2.9, + "grad_norm": 0.44356584548950195, + "learning_rate": 3.1405277694064306e-06, + "loss": 0.4516, + "step": 424 + }, + { + "epoch": 2.9068181818181817, + "grad_norm": 0.5283893942832947, + "learning_rate": 3.13073927074945e-06, + "loss": 0.4792, + "step": 425 + }, + { + "epoch": 2.9136363636363636, + "grad_norm": 0.90226149559021, + "learning_rate": 3.1209404343530374e-06, + "loss": 0.4811, + "step": 426 + }, + { + "epoch": 2.9204545454545454, + "grad_norm": 0.4658738970756531, + "learning_rate": 3.111131420818922e-06, + "loss": 0.4601, + "step": 427 + }, + { + "epoch": 2.9272727272727272, + "grad_norm": 0.4286805987358093, + "learning_rate": 3.1013123909156347e-06, + "loss": 0.4801, + "step": 428 + }, + { + "epoch": 2.934090909090909, + "grad_norm": 0.5181618332862854, + "learning_rate": 3.091483505575873e-06, + "loss": 0.4717, + "step": 429 + }, + { + "epoch": 2.940909090909091, + "grad_norm": 0.4391111135482788, + "learning_rate": 3.081644925893866e-06, + "loss": 0.4913, + "step": 430 + }, + { + "epoch": 2.9477272727272728, + "grad_norm": 0.4414638578891754, + "learning_rate": 3.0717968131227285e-06, + "loss": 0.4796, + "step": 431 + }, + { + "epoch": 2.9545454545454546, + "grad_norm": 0.4619010090827942, + "learning_rate": 3.061939328671824e-06, + "loss": 0.5059, + "step": 432 + }, + { + "epoch": 2.9613636363636364, + "grad_norm": 0.42799466848373413, + "learning_rate": 3.0520726341041165e-06, + "loss": 0.4598, + "step": 433 + }, + { + "epoch": 2.9681818181818183, + "grad_norm": 0.43869948387145996, + "learning_rate": 3.0421968911335196e-06, + "loss": 0.4733, + "step": 434 + }, + { + "epoch": 2.975, + "grad_norm": 0.552511990070343, + "learning_rate": 3.032312261622255e-06, + "loss": 0.4855, + "step": 435 + }, + { + "epoch": 2.981818181818182, + "grad_norm": 0.5187029242515564, + "learning_rate": 3.0224189075781886e-06, + "loss": 0.5067, + "step": 436 + }, + { + "epoch": 2.9886363636363638, + "grad_norm": 0.4445323944091797, + "learning_rate": 3.012516991152181e-06, + "loss": 0.4542, + "step": 437 + }, + { + "epoch": 2.9954545454545456, + "grad_norm": 0.5056352019309998, + "learning_rate": 3.002606674635432e-06, + "loss": 0.4907, + "step": 438 + } + ], + "logging_steps": 1, + "max_steps": 876, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 146, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 2.822023060576115e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-438/training_args.bin b/checkpoint-438/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..b41c805059a36d8d2e3bf3511fb0fd95b74dae74 --- /dev/null +++ b/checkpoint-438/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68945b8f9934f5870e67a2d4a20c31fa8e742da41075fa7236fb313c6197456f +size 7928 diff --git a/checkpoint-438/zero_to_fp32.py b/checkpoint-438/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-438/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-584/README.md b/checkpoint-584/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1b184114a0c28ed3e4c082c18486736dc818166d --- /dev/null +++ b/checkpoint-584/README.md @@ -0,0 +1,202 @@ +--- +base_model: meta-llama/Llama-3.3-70B-Instruct +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 0.15.0 \ No newline at end of file diff --git a/checkpoint-584/adapter_config.json b/checkpoint-584/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a5775df151cb9649596940e8a117cb92f810dbb7 --- /dev/null +++ b/checkpoint-584/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.3-70B-Instruct", + "bias": "none", + "corda_config": null, + "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", + "down_proj", + "k_proj", + "q_proj", + "o_proj", + "v_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-584/adapter_model.safetensors b/checkpoint-584/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..5dc373f77c852bc653ebd7ad419f1c0890ac4ad9 --- /dev/null +++ b/checkpoint-584/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c5a376964951f9f2b16d279d9cec3c85d3b3f54cf050d5e6b4ff8cc90971df9a +size 10829849744 diff --git a/checkpoint-584/global_step585/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-584/global_step585/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..56e5d878d810b03d4bcf7e022e095d547edb5f6a --- /dev/null +++ b/checkpoint-584/global_step585/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9de5efbbfacae168bd339580bd4392d83f893282243963d047bd95547875eda7 +size 21659418140 diff --git a/checkpoint-584/global_step585/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-584/global_step585/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..cf0659a85a016497d626039c6c345544ffa39741 --- /dev/null +++ b/checkpoint-584/global_step585/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:2aafe51e93376993a7fb19e69f338093b913653f52b0f19ecb36ead2ee290843 +size 21659457372 diff --git a/checkpoint-584/global_step585/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoint-584/global_step585/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..000738166abc2654e090363db7f3ba735ba186eb --- /dev/null +++ b/checkpoint-584/global_step585/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:965f4a1743a92c8ae6718d722ec8cdbf7358eee5eeb1bd1b5508cf7eb5652073 +size 21659417820 diff --git a/checkpoint-584/global_step585/mp_rank_00_model_states.pt b/checkpoint-584/global_step585/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..fdc43e36cb9ea0d948a0ff4b7985d6455163262d --- /dev/null +++ b/checkpoint-584/global_step585/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b33c2eb42c63e188d34e5f5365fb5b0f1c97bdc91dd11f72a0f42d0446152be6 +size 11918643933 diff --git a/checkpoint-584/latest b/checkpoint-584/latest new file mode 100644 index 0000000000000000000000000000000000000000..80109b2b35a87a444310330bcd3a0c4ea8922797 --- /dev/null +++ b/checkpoint-584/latest @@ -0,0 +1 @@ +global_step585 \ No newline at end of file diff --git a/checkpoint-584/rng_state_0.pth b/checkpoint-584/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..120ea0ac12377116bcb5a579c419860ff28af5e9 --- /dev/null +++ b/checkpoint-584/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:370fc9cc15509b4c6846b908cb2b1489fa98729f7382b8cee180ccf7c885dd76 +size 14768 diff --git a/checkpoint-584/rng_state_1.pth b/checkpoint-584/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..f9afabb4df4f400b138a4aeadad65f882719be4b --- /dev/null +++ b/checkpoint-584/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:a5e209442648eb8886aa322f383071772a598298fe9db480e31a65b54c34cfde +size 14768 diff --git a/checkpoint-584/rng_state_2.pth b/checkpoint-584/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..e6df2d6f25372ef5652096e08d7ab18fc7b3850f --- /dev/null +++ b/checkpoint-584/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:81818fb4ea344177ac0f28523fba6e69849915a3cba71daaba7137d7103c33fb +size 14768 diff --git a/checkpoint-584/scheduler.pt b/checkpoint-584/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..f2f77ee0baf79371ee95cd83f42264909eedf3e7 --- /dev/null +++ b/checkpoint-584/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:d6bd663421730ca300e5e7464cd8c61e90b489baf824489fd92c6c45d35c5dc9 +size 1064 diff --git a/checkpoint-584/special_tokens_map.json b/checkpoint-584/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-584/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": "<|eot_id|>", + "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-584/tokenizer.json b/checkpoint-584/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-584/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-584/tokenizer_config.json b/checkpoint-584/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ca91a2ef55f4239a7af81d7c9abb05f53621a07b --- /dev/null +++ b/checkpoint-584/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "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|>", + "chat_template": "{{- bos_token }}\n{%- if custom_tools is defined %}\n {%- set tools = custom_tools %}\n{%- endif %}\n{%- if not tools_in_user_message is defined %}\n {%- set tools_in_user_message = true %}\n{%- endif %}\n{%- if not date_string is defined %}\n {%- set date_string = \"26 Jul 2024\" %}\n{%- endif %}\n{%- if not tools is defined %}\n {%- set tools = none %}\n{%- endif %}\n\n{#- This block extracts the system message, so we can slot it into the right place. #}\n{%- if messages[0]['role'] == 'system' %}\n {%- set system_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n{%- else %}\n {%- set system_message = \"\" %}\n{%- endif %}\n\n{#- System message + builtin tools #}\n{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}\n{%- if builtin_tools is defined or tools is not none %}\n {{- \"Environment: ipython\\n\" }}\n{%- endif %}\n{%- if builtin_tools is defined %}\n {{- \"Tools: \" + builtin_tools | reject('equalto', 'code_interpreter') | join(\", \") + \"\\n\\n\"}}\n{%- endif %}\n{{- \"Cutting Knowledge Date: December 2023\\n\" }}\n{{- \"Today Date: \" + date_string + \"\\n\\n\" }}\n{%- if tools is not none and not tools_in_user_message %}\n {{- \"You have access to the following functions. To call a function, please respond with JSON for a function call.\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n{%- endif %}\n{{- system_message }}\n{{- \"<|eot_id|>\" }}\n\n{#- Custom tools are passed in a user message with some extra guidance #}\n{%- if tools_in_user_message and not tools is none %}\n {#- Extract the first user message so we can plug it in here #}\n {%- if messages | length != 0 %}\n {%- set first_user_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n {%- else %}\n {{- raise_exception(\"Cannot put tools in the first user message when there's no first user message!\") }}\n{%- endif %}\n {{- '<|start_header_id|>user<|end_header_id|>\\n\\n' -}}\n {{- \"Given the following functions, please respond with a JSON for a function call \" }}\n {{- \"with its proper arguments that best answers the given prompt.\\n\\n\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n {{- first_user_message + \"<|eot_id|>\"}}\n{%- endif %}\n\n{%- for message in messages %}\n {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}\n {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n'+ message['content'] | trim + '<|eot_id|>' }}\n {%- elif 'tool_calls' in message %}\n {%- if not message.tool_calls|length == 1 %}\n {{- raise_exception(\"This model only supports single tool-calls at once!\") }}\n {%- endif %}\n {%- set tool_call = message.tool_calls[0].function %}\n {%- if builtin_tools is defined and tool_call.name in builtin_tools %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- \"<|python_tag|>\" + tool_call.name + \".call(\" }}\n {%- for arg_name, arg_val in tool_call.arguments | items %}\n {{- arg_name + '=\"' + arg_val + '\"' }}\n {%- if not loop.last %}\n {{- \", \" }}\n {%- endif %}\n {%- endfor %}\n {{- \")\" }}\n {%- else %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- '{\"name\": \"' + tool_call.name + '\", ' }}\n {{- '\"parameters\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- \"}\" }}\n {%- endif %}\n {%- if builtin_tools is defined %}\n {#- This means we're in ipython mode #}\n {{- \"<|eom_id|>\" }}\n {%- else %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n {%- elif message.role == \"tool\" or message.role == \"ipython\" %}\n {{- \"<|start_header_id|>ipython<|end_header_id|>\\n\\n\" }}\n {%- if message.content is mapping or message.content is iterable %}\n {{- message.content | tojson }}\n {%- else %}\n {{- message.content }}\n {%- endif %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}\n{%- endif %}\n", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "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-584/trainer_state.json b/checkpoint-584/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..56530f76fb10d81e2dfb0e9445dcfe650be463c1 --- /dev/null +++ b/checkpoint-584/trainer_state.json @@ -0,0 +1,4121 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 3.9954545454545456, + "eval_steps": 500, + "global_step": 584, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.006818181818181818, + "grad_norm": 1.8943578004837036, + "learning_rate": 5.0000000000000004e-08, + "loss": 0.8676, + "step": 1 + }, + { + "epoch": 0.013636363636363636, + "grad_norm": 7.775813102722168, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.8513, + "step": 2 + }, + { + "epoch": 0.020454545454545454, + "grad_norm": 1.768803596496582, + "learning_rate": 1.5000000000000002e-07, + "loss": 0.8339, + "step": 3 + }, + { + "epoch": 0.02727272727272727, + "grad_norm": 1.7032263278961182, + "learning_rate": 2.0000000000000002e-07, + "loss": 0.8348, + "step": 4 + }, + { + "epoch": 0.03409090909090909, + "grad_norm": 1.8244796991348267, + "learning_rate": 2.5000000000000004e-07, + "loss": 0.8687, + "step": 5 + }, + { + "epoch": 0.04090909090909091, + "grad_norm": 1.7780181169509888, + "learning_rate": 3.0000000000000004e-07, + "loss": 0.8113, + "step": 6 + }, + { + "epoch": 0.04772727272727273, + "grad_norm": 1.8362191915512085, + "learning_rate": 3.5000000000000004e-07, + "loss": 0.8463, + "step": 7 + }, + { + "epoch": 0.05454545454545454, + "grad_norm": 1.6640979051589966, + "learning_rate": 4.0000000000000003e-07, + "loss": 0.8336, + "step": 8 + }, + { + "epoch": 0.06136363636363636, + "grad_norm": 1.8013464212417603, + "learning_rate": 4.5000000000000003e-07, + "loss": 0.8572, + "step": 9 + }, + { + "epoch": 0.06818181818181818, + "grad_norm": 1.7237999439239502, + "learning_rate": 5.000000000000001e-07, + "loss": 0.8501, + "step": 10 + }, + { + "epoch": 0.075, + "grad_norm": 1.989346981048584, + "learning_rate": 5.5e-07, + "loss": 0.8615, + "step": 11 + }, + { + "epoch": 0.08181818181818182, + "grad_norm": 2.0312082767486572, + "learning_rate": 6.000000000000001e-07, + "loss": 0.858, + "step": 12 + }, + { + "epoch": 0.08863636363636364, + "grad_norm": 1.6921299695968628, + "learning_rate": 6.5e-07, + "loss": 0.8577, + "step": 13 + }, + { + "epoch": 0.09545454545454546, + "grad_norm": 2.4207160472869873, + "learning_rate": 7.000000000000001e-07, + "loss": 0.8564, + "step": 14 + }, + { + "epoch": 0.10227272727272728, + "grad_norm": 1.7572585344314575, + "learning_rate": 7.5e-07, + "loss": 0.8409, + "step": 15 + }, + { + "epoch": 0.10909090909090909, + "grad_norm": 1.6135950088500977, + "learning_rate": 8.000000000000001e-07, + "loss": 0.8086, + "step": 16 + }, + { + "epoch": 0.1159090909090909, + "grad_norm": 2.226036787033081, + "learning_rate": 8.500000000000001e-07, + "loss": 0.8106, + "step": 17 + }, + { + "epoch": 0.12272727272727273, + "grad_norm": 1.7164231538772583, + "learning_rate": 9.000000000000001e-07, + "loss": 0.8494, + "step": 18 + }, + { + "epoch": 0.12954545454545455, + "grad_norm": 1.6572000980377197, + "learning_rate": 9.500000000000001e-07, + "loss": 0.8023, + "step": 19 + }, + { + "epoch": 0.13636363636363635, + "grad_norm": 1.5672118663787842, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.787, + "step": 20 + }, + { + "epoch": 0.1431818181818182, + "grad_norm": 1.5080257654190063, + "learning_rate": 1.0500000000000001e-06, + "loss": 0.8105, + "step": 21 + }, + { + "epoch": 0.15, + "grad_norm": 1.484892725944519, + "learning_rate": 1.1e-06, + "loss": 0.8258, + "step": 22 + }, + { + "epoch": 0.15681818181818183, + "grad_norm": 1.4591134786605835, + "learning_rate": 1.1500000000000002e-06, + "loss": 0.8032, + "step": 23 + }, + { + "epoch": 0.16363636363636364, + "grad_norm": 1.5506278276443481, + "learning_rate": 1.2000000000000002e-06, + "loss": 0.8236, + "step": 24 + }, + { + "epoch": 0.17045454545454544, + "grad_norm": 1.639350414276123, + "learning_rate": 1.25e-06, + "loss": 0.8185, + "step": 25 + }, + { + "epoch": 0.17727272727272728, + "grad_norm": 1.592640995979309, + "learning_rate": 1.3e-06, + "loss": 0.7898, + "step": 26 + }, + { + "epoch": 0.18409090909090908, + "grad_norm": 1.5280641317367554, + "learning_rate": 1.3500000000000002e-06, + "loss": 0.7731, + "step": 27 + }, + { + "epoch": 0.19090909090909092, + "grad_norm": 1.4015443325042725, + "learning_rate": 1.4000000000000001e-06, + "loss": 0.7956, + "step": 28 + }, + { + "epoch": 0.19772727272727272, + "grad_norm": 1.4890056848526, + "learning_rate": 1.45e-06, + "loss": 0.7838, + "step": 29 + }, + { + "epoch": 0.20454545454545456, + "grad_norm": 2.4770750999450684, + "learning_rate": 1.5e-06, + "loss": 0.7766, + "step": 30 + }, + { + "epoch": 0.21136363636363636, + "grad_norm": 1.5863685607910156, + "learning_rate": 1.5500000000000002e-06, + "loss": 0.7353, + "step": 31 + }, + { + "epoch": 0.21818181818181817, + "grad_norm": 1.058114767074585, + "learning_rate": 1.6000000000000001e-06, + "loss": 0.7237, + "step": 32 + }, + { + "epoch": 0.225, + "grad_norm": 1.5869101285934448, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.7597, + "step": 33 + }, + { + "epoch": 0.2318181818181818, + "grad_norm": 1.0012413263320923, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.7388, + "step": 34 + }, + { + "epoch": 0.23863636363636365, + "grad_norm": 1.118056058883667, + "learning_rate": 1.75e-06, + "loss": 0.7786, + "step": 35 + }, + { + "epoch": 0.24545454545454545, + "grad_norm": 1.0648820400238037, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.7502, + "step": 36 + }, + { + "epoch": 0.25227272727272726, + "grad_norm": 0.9440382719039917, + "learning_rate": 1.85e-06, + "loss": 0.6942, + "step": 37 + }, + { + "epoch": 0.2590909090909091, + "grad_norm": 0.8300187587738037, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.7312, + "step": 38 + }, + { + "epoch": 0.26590909090909093, + "grad_norm": 0.8296191692352295, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.7461, + "step": 39 + }, + { + "epoch": 0.2727272727272727, + "grad_norm": 0.7797508239746094, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.7234, + "step": 40 + }, + { + "epoch": 0.27954545454545454, + "grad_norm": 0.7335019707679749, + "learning_rate": 2.05e-06, + "loss": 0.6927, + "step": 41 + }, + { + "epoch": 0.2863636363636364, + "grad_norm": 0.7062106728553772, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.6552, + "step": 42 + }, + { + "epoch": 0.29318181818181815, + "grad_norm": 1.2342087030410767, + "learning_rate": 2.15e-06, + "loss": 0.6959, + "step": 43 + }, + { + "epoch": 0.3, + "grad_norm": 0.8967597484588623, + "learning_rate": 2.2e-06, + "loss": 0.704, + "step": 44 + }, + { + "epoch": 0.3068181818181818, + "grad_norm": 0.7228404879570007, + "learning_rate": 2.25e-06, + "loss": 0.6777, + "step": 45 + }, + { + "epoch": 0.31363636363636366, + "grad_norm": 0.6731488704681396, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.6652, + "step": 46 + }, + { + "epoch": 0.32045454545454544, + "grad_norm": 0.71966952085495, + "learning_rate": 2.35e-06, + "loss": 0.7281, + "step": 47 + }, + { + "epoch": 0.32727272727272727, + "grad_norm": 0.7057356238365173, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.6476, + "step": 48 + }, + { + "epoch": 0.3340909090909091, + "grad_norm": 0.6346054077148438, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.6865, + "step": 49 + }, + { + "epoch": 0.3409090909090909, + "grad_norm": 0.6938223838806152, + "learning_rate": 2.5e-06, + "loss": 0.6808, + "step": 50 + }, + { + "epoch": 0.3477272727272727, + "grad_norm": 0.7425184845924377, + "learning_rate": 2.55e-06, + "loss": 0.7074, + "step": 51 + }, + { + "epoch": 0.35454545454545455, + "grad_norm": 0.6743818521499634, + "learning_rate": 2.6e-06, + "loss": 0.657, + "step": 52 + }, + { + "epoch": 0.3613636363636364, + "grad_norm": 0.6785942316055298, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.6768, + "step": 53 + }, + { + "epoch": 0.36818181818181817, + "grad_norm": 0.586974024772644, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.651, + "step": 54 + }, + { + "epoch": 0.375, + "grad_norm": 0.5727331042289734, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.669, + "step": 55 + }, + { + "epoch": 0.38181818181818183, + "grad_norm": 0.7333543300628662, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.6571, + "step": 56 + }, + { + "epoch": 0.3886363636363636, + "grad_norm": 0.5710961818695068, + "learning_rate": 2.85e-06, + "loss": 0.6475, + "step": 57 + }, + { + "epoch": 0.39545454545454545, + "grad_norm": 0.5614489912986755, + "learning_rate": 2.9e-06, + "loss": 0.6314, + "step": 58 + }, + { + "epoch": 0.4022727272727273, + "grad_norm": 0.5007341504096985, + "learning_rate": 2.95e-06, + "loss": 0.6346, + "step": 59 + }, + { + "epoch": 0.4090909090909091, + "grad_norm": 0.54221510887146, + "learning_rate": 3e-06, + "loss": 0.6715, + "step": 60 + }, + { + "epoch": 0.4159090909090909, + "grad_norm": 0.5177039504051208, + "learning_rate": 3.05e-06, + "loss": 0.6459, + "step": 61 + }, + { + "epoch": 0.42272727272727273, + "grad_norm": 0.4741189181804657, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.592, + "step": 62 + }, + { + "epoch": 0.42954545454545456, + "grad_norm": 0.5637328028678894, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.6459, + "step": 63 + }, + { + "epoch": 0.43636363636363634, + "grad_norm": 0.5478869676589966, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.6447, + "step": 64 + }, + { + "epoch": 0.4431818181818182, + "grad_norm": 0.5483130216598511, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.6319, + "step": 65 + }, + { + "epoch": 0.45, + "grad_norm": 0.478081613779068, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.6102, + "step": 66 + }, + { + "epoch": 0.45681818181818185, + "grad_norm": 0.4777645766735077, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.5986, + "step": 67 + }, + { + "epoch": 0.4636363636363636, + "grad_norm": 0.5291482210159302, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.6314, + "step": 68 + }, + { + "epoch": 0.47045454545454546, + "grad_norm": 0.5224051475524902, + "learning_rate": 3.45e-06, + "loss": 0.6278, + "step": 69 + }, + { + "epoch": 0.4772727272727273, + "grad_norm": 0.6474127173423767, + "learning_rate": 3.5e-06, + "loss": 0.6398, + "step": 70 + }, + { + "epoch": 0.48409090909090907, + "grad_norm": 0.5060178637504578, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.5936, + "step": 71 + }, + { + "epoch": 0.4909090909090909, + "grad_norm": 3.0769243240356445, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.5975, + "step": 72 + }, + { + "epoch": 0.49772727272727274, + "grad_norm": 0.5319749116897583, + "learning_rate": 3.65e-06, + "loss": 0.6119, + "step": 73 + }, + { + "epoch": 0.5045454545454545, + "grad_norm": 0.6745399236679077, + "learning_rate": 3.7e-06, + "loss": 0.6213, + "step": 74 + }, + { + "epoch": 0.5113636363636364, + "grad_norm": 0.5156731009483337, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.6262, + "step": 75 + }, + { + "epoch": 0.5181818181818182, + "grad_norm": 0.9138725399971008, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.6244, + "step": 76 + }, + { + "epoch": 0.525, + "grad_norm": 0.4362037777900696, + "learning_rate": 3.85e-06, + "loss": 0.5874, + "step": 77 + }, + { + "epoch": 0.5318181818181819, + "grad_norm": 0.4674088954925537, + "learning_rate": 3.900000000000001e-06, + "loss": 0.615, + "step": 78 + }, + { + "epoch": 0.5386363636363637, + "grad_norm": 0.4590819180011749, + "learning_rate": 3.95e-06, + "loss": 0.6003, + "step": 79 + }, + { + "epoch": 0.5454545454545454, + "grad_norm": 0.5145031809806824, + "learning_rate": 4.000000000000001e-06, + "loss": 0.6079, + "step": 80 + }, + { + "epoch": 0.5522727272727272, + "grad_norm": 0.46660348773002625, + "learning_rate": 4.05e-06, + "loss": 0.6174, + "step": 81 + }, + { + "epoch": 0.5590909090909091, + "grad_norm": 0.48053789138793945, + "learning_rate": 4.1e-06, + "loss": 0.6232, + "step": 82 + }, + { + "epoch": 0.5659090909090909, + "grad_norm": 0.540793776512146, + "learning_rate": 4.15e-06, + "loss": 0.619, + "step": 83 + }, + { + "epoch": 0.5727272727272728, + "grad_norm": 0.4925222396850586, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.5838, + "step": 84 + }, + { + "epoch": 0.5795454545454546, + "grad_norm": 0.44989290833473206, + "learning_rate": 4.25e-06, + "loss": 0.5972, + "step": 85 + }, + { + "epoch": 0.5863636363636363, + "grad_norm": 0.42691388726234436, + "learning_rate": 4.3e-06, + "loss": 0.6046, + "step": 86 + }, + { + "epoch": 0.5931818181818181, + "grad_norm": 0.4512398838996887, + "learning_rate": 4.350000000000001e-06, + "loss": 0.5828, + "step": 87 + }, + { + "epoch": 0.6, + "grad_norm": 0.44399499893188477, + "learning_rate": 4.4e-06, + "loss": 0.6143, + "step": 88 + }, + { + "epoch": 0.6068181818181818, + "grad_norm": 0.46058326959609985, + "learning_rate": 4.450000000000001e-06, + "loss": 0.6117, + "step": 89 + }, + { + "epoch": 0.6136363636363636, + "grad_norm": 0.8795785903930664, + "learning_rate": 4.5e-06, + "loss": 0.6, + "step": 90 + }, + { + "epoch": 0.6204545454545455, + "grad_norm": 0.5148798227310181, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.6041, + "step": 91 + }, + { + "epoch": 0.6272727272727273, + "grad_norm": 0.48829612135887146, + "learning_rate": 4.600000000000001e-06, + "loss": 0.5492, + "step": 92 + }, + { + "epoch": 0.634090909090909, + "grad_norm": 0.5211894512176514, + "learning_rate": 4.65e-06, + "loss": 0.6044, + "step": 93 + }, + { + "epoch": 0.6409090909090909, + "grad_norm": 0.44403275847435, + "learning_rate": 4.7e-06, + "loss": 0.5929, + "step": 94 + }, + { + "epoch": 0.6477272727272727, + "grad_norm": 0.6037693619728088, + "learning_rate": 4.75e-06, + "loss": 0.6132, + "step": 95 + }, + { + "epoch": 0.6545454545454545, + "grad_norm": 0.4381515085697174, + "learning_rate": 4.800000000000001e-06, + "loss": 0.5822, + "step": 96 + }, + { + "epoch": 0.6613636363636364, + "grad_norm": 0.4997427761554718, + "learning_rate": 4.85e-06, + "loss": 0.593, + "step": 97 + }, + { + "epoch": 0.6681818181818182, + "grad_norm": 0.440571665763855, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.6072, + "step": 98 + }, + { + "epoch": 0.675, + "grad_norm": 0.5020624995231628, + "learning_rate": 4.95e-06, + "loss": 0.5861, + "step": 99 + }, + { + "epoch": 0.6818181818181818, + "grad_norm": 0.4280160963535309, + "learning_rate": 5e-06, + "loss": 0.6163, + "step": 100 + }, + { + "epoch": 0.6886363636363636, + "grad_norm": 0.4584537148475647, + "learning_rate": 4.9999795126530275e-06, + "loss": 0.6177, + "step": 101 + }, + { + "epoch": 0.6954545454545454, + "grad_norm": 0.6835049390792847, + "learning_rate": 4.999918050947891e-06, + "loss": 0.579, + "step": 102 + }, + { + "epoch": 0.7022727272727273, + "grad_norm": 0.4551607072353363, + "learning_rate": 4.999815615891943e-06, + "loss": 0.5927, + "step": 103 + }, + { + "epoch": 0.7090909090909091, + "grad_norm": 0.5893972516059875, + "learning_rate": 4.9996722091640805e-06, + "loss": 0.5775, + "step": 104 + }, + { + "epoch": 0.7159090909090909, + "grad_norm": 0.4574092924594879, + "learning_rate": 4.9994878331147225e-06, + "loss": 0.5863, + "step": 105 + }, + { + "epoch": 0.7227272727272728, + "grad_norm": 0.5354658365249634, + "learning_rate": 4.99926249076577e-06, + "loss": 0.5453, + "step": 106 + }, + { + "epoch": 0.7295454545454545, + "grad_norm": 0.4347354471683502, + "learning_rate": 4.998996185810557e-06, + "loss": 0.5913, + "step": 107 + }, + { + "epoch": 0.7363636363636363, + "grad_norm": 0.4487966299057007, + "learning_rate": 4.998688922613788e-06, + "loss": 0.5749, + "step": 108 + }, + { + "epoch": 0.7431818181818182, + "grad_norm": 0.42477577924728394, + "learning_rate": 4.9983407062114695e-06, + "loss": 0.5761, + "step": 109 + }, + { + "epoch": 0.75, + "grad_norm": 0.4801446497440338, + "learning_rate": 4.9979515423108255e-06, + "loss": 0.5977, + "step": 110 + }, + { + "epoch": 0.7568181818181818, + "grad_norm": 0.4876883029937744, + "learning_rate": 4.997521437290205e-06, + "loss": 0.5849, + "step": 111 + }, + { + "epoch": 0.7636363636363637, + "grad_norm": 0.4328872263431549, + "learning_rate": 4.997050398198977e-06, + "loss": 0.5988, + "step": 112 + }, + { + "epoch": 0.7704545454545455, + "grad_norm": 0.4332719147205353, + "learning_rate": 4.996538432757414e-06, + "loss": 0.6059, + "step": 113 + }, + { + "epoch": 0.7772727272727272, + "grad_norm": 0.4455336630344391, + "learning_rate": 4.995985549356568e-06, + "loss": 0.5725, + "step": 114 + }, + { + "epoch": 0.7840909090909091, + "grad_norm": 0.4197766184806824, + "learning_rate": 4.995391757058129e-06, + "loss": 0.5715, + "step": 115 + }, + { + "epoch": 0.7909090909090909, + "grad_norm": 0.41945499181747437, + "learning_rate": 4.99475706559428e-06, + "loss": 0.5845, + "step": 116 + }, + { + "epoch": 0.7977272727272727, + "grad_norm": 0.42414823174476624, + "learning_rate": 4.994081485367537e-06, + "loss": 0.5595, + "step": 117 + }, + { + "epoch": 0.8045454545454546, + "grad_norm": 0.4056423008441925, + "learning_rate": 4.993365027450576e-06, + "loss": 0.582, + "step": 118 + }, + { + "epoch": 0.8113636363636364, + "grad_norm": 0.4554205536842346, + "learning_rate": 4.992607703586058e-06, + "loss": 0.5591, + "step": 119 + }, + { + "epoch": 0.8181818181818182, + "grad_norm": 0.4307089149951935, + "learning_rate": 4.991809526186424e-06, + "loss": 0.5927, + "step": 120 + }, + { + "epoch": 0.825, + "grad_norm": 0.48133528232574463, + "learning_rate": 4.990970508333707e-06, + "loss": 0.5703, + "step": 121 + }, + { + "epoch": 0.8318181818181818, + "grad_norm": 0.48582738637924194, + "learning_rate": 4.990090663779305e-06, + "loss": 0.5491, + "step": 122 + }, + { + "epoch": 0.8386363636363636, + "grad_norm": 0.44926196336746216, + "learning_rate": 4.9891700069437635e-06, + "loss": 0.5803, + "step": 123 + }, + { + "epoch": 0.8454545454545455, + "grad_norm": 0.4562165141105652, + "learning_rate": 4.988208552916535e-06, + "loss": 0.5945, + "step": 124 + }, + { + "epoch": 0.8522727272727273, + "grad_norm": 0.5885360836982727, + "learning_rate": 4.987206317455734e-06, + "loss": 0.5632, + "step": 125 + }, + { + "epoch": 0.8590909090909091, + "grad_norm": 0.4525550603866577, + "learning_rate": 4.986163316987877e-06, + "loss": 0.5221, + "step": 126 + }, + { + "epoch": 0.865909090909091, + "grad_norm": 0.4545478820800781, + "learning_rate": 4.985079568607613e-06, + "loss": 0.587, + "step": 127 + }, + { + "epoch": 0.8727272727272727, + "grad_norm": 0.4364396333694458, + "learning_rate": 4.983955090077445e-06, + "loss": 0.5626, + "step": 128 + }, + { + "epoch": 0.8795454545454545, + "grad_norm": 0.4354369640350342, + "learning_rate": 4.982789899827439e-06, + "loss": 0.5674, + "step": 129 + }, + { + "epoch": 0.8863636363636364, + "grad_norm": 0.4822773337364197, + "learning_rate": 4.9815840169549216e-06, + "loss": 0.5685, + "step": 130 + }, + { + "epoch": 0.8931818181818182, + "grad_norm": 0.44673025608062744, + "learning_rate": 4.980337461224164e-06, + "loss": 0.5733, + "step": 131 + }, + { + "epoch": 0.9, + "grad_norm": 0.6294501423835754, + "learning_rate": 4.979050253066064e-06, + "loss": 0.5396, + "step": 132 + }, + { + "epoch": 0.9068181818181819, + "grad_norm": 0.42890027165412903, + "learning_rate": 4.977722413577802e-06, + "loss": 0.5718, + "step": 133 + }, + { + "epoch": 0.9136363636363637, + "grad_norm": 0.470587819814682, + "learning_rate": 4.976353964522509e-06, + "loss": 0.5705, + "step": 134 + }, + { + "epoch": 0.9204545454545454, + "grad_norm": 0.4632768929004669, + "learning_rate": 4.974944928328894e-06, + "loss": 0.5394, + "step": 135 + }, + { + "epoch": 0.9272727272727272, + "grad_norm": 0.5304691195487976, + "learning_rate": 4.973495328090891e-06, + "loss": 0.5762, + "step": 136 + }, + { + "epoch": 0.9340909090909091, + "grad_norm": 0.42036354541778564, + "learning_rate": 4.972005187567267e-06, + "loss": 0.5649, + "step": 137 + }, + { + "epoch": 0.9409090909090909, + "grad_norm": 0.45189663767814636, + "learning_rate": 4.970474531181245e-06, + "loss": 0.5554, + "step": 138 + }, + { + "epoch": 0.9477272727272728, + "grad_norm": 0.47352468967437744, + "learning_rate": 4.968903384020095e-06, + "loss": 0.5511, + "step": 139 + }, + { + "epoch": 0.9545454545454546, + "grad_norm": 0.525702953338623, + "learning_rate": 4.967291771834727e-06, + "loss": 0.5777, + "step": 140 + }, + { + "epoch": 0.9613636363636363, + "grad_norm": 0.46146854758262634, + "learning_rate": 4.965639721039267e-06, + "loss": 0.5537, + "step": 141 + }, + { + "epoch": 0.9681818181818181, + "grad_norm": 0.5200941562652588, + "learning_rate": 4.963947258710626e-06, + "loss": 0.5522, + "step": 142 + }, + { + "epoch": 0.975, + "grad_norm": 0.5260375142097473, + "learning_rate": 4.962214412588053e-06, + "loss": 0.5747, + "step": 143 + }, + { + "epoch": 0.9818181818181818, + "grad_norm": 0.518645703792572, + "learning_rate": 4.960441211072686e-06, + "loss": 0.5257, + "step": 144 + }, + { + "epoch": 0.9886363636363636, + "grad_norm": 0.4528891444206238, + "learning_rate": 4.9586276832270785e-06, + "loss": 0.5548, + "step": 145 + }, + { + "epoch": 0.9954545454545455, + "grad_norm": 0.5137557983398438, + "learning_rate": 4.9567738587747314e-06, + "loss": 0.5526, + "step": 146 + }, + { + "epoch": 1.0068181818181818, + "grad_norm": 0.8773607611656189, + "learning_rate": 4.954879768099599e-06, + "loss": 1.1, + "step": 147 + }, + { + "epoch": 1.0136363636363637, + "grad_norm": 0.39079979062080383, + "learning_rate": 4.952945442245598e-06, + "loss": 0.568, + "step": 148 + }, + { + "epoch": 1.0204545454545455, + "grad_norm": 0.45004600286483765, + "learning_rate": 4.95097091291609e-06, + "loss": 0.5383, + "step": 149 + }, + { + "epoch": 1.0272727272727273, + "grad_norm": 0.41872867941856384, + "learning_rate": 4.948956212473371e-06, + "loss": 0.5436, + "step": 150 + }, + { + "epoch": 1.0340909090909092, + "grad_norm": 0.46829020977020264, + "learning_rate": 4.946901373938132e-06, + "loss": 0.5336, + "step": 151 + }, + { + "epoch": 1.040909090909091, + "grad_norm": 0.45038115978240967, + "learning_rate": 4.944806430988927e-06, + "loss": 0.538, + "step": 152 + }, + { + "epoch": 1.0477272727272728, + "grad_norm": 0.6496132016181946, + "learning_rate": 4.942671417961615e-06, + "loss": 0.5459, + "step": 153 + }, + { + "epoch": 1.0545454545454545, + "grad_norm": 0.4842703640460968, + "learning_rate": 4.940496369848795e-06, + "loss": 0.553, + "step": 154 + }, + { + "epoch": 1.0613636363636363, + "grad_norm": 0.4338674545288086, + "learning_rate": 4.938281322299243e-06, + "loss": 0.5366, + "step": 155 + }, + { + "epoch": 1.0681818181818181, + "grad_norm": 0.44630634784698486, + "learning_rate": 4.936026311617316e-06, + "loss": 0.5517, + "step": 156 + }, + { + "epoch": 1.075, + "grad_norm": 0.4606929421424866, + "learning_rate": 4.933731374762361e-06, + "loss": 0.5405, + "step": 157 + }, + { + "epoch": 1.0818181818181818, + "grad_norm": 0.48667654395103455, + "learning_rate": 4.931396549348115e-06, + "loss": 0.5376, + "step": 158 + }, + { + "epoch": 1.0886363636363636, + "grad_norm": 0.4516271650791168, + "learning_rate": 4.9290218736420795e-06, + "loss": 0.5206, + "step": 159 + }, + { + "epoch": 1.0954545454545455, + "grad_norm": 0.44719284772872925, + "learning_rate": 4.926607386564898e-06, + "loss": 0.5397, + "step": 160 + }, + { + "epoch": 1.1022727272727273, + "grad_norm": 0.40752628445625305, + "learning_rate": 4.9241531276897196e-06, + "loss": 0.5466, + "step": 161 + }, + { + "epoch": 1.1090909090909091, + "grad_norm": 0.45423609018325806, + "learning_rate": 4.921659137241544e-06, + "loss": 0.5227, + "step": 162 + }, + { + "epoch": 1.115909090909091, + "grad_norm": 0.4185822308063507, + "learning_rate": 4.919125456096574e-06, + "loss": 0.5144, + "step": 163 + }, + { + "epoch": 1.1227272727272728, + "grad_norm": 0.7753382921218872, + "learning_rate": 4.916552125781529e-06, + "loss": 0.5352, + "step": 164 + }, + { + "epoch": 1.1295454545454546, + "grad_norm": 0.47211477160453796, + "learning_rate": 4.913939188472979e-06, + "loss": 0.5505, + "step": 165 + }, + { + "epoch": 1.1363636363636362, + "grad_norm": 0.42742082476615906, + "learning_rate": 4.911286686996648e-06, + "loss": 0.529, + "step": 166 + }, + { + "epoch": 1.143181818181818, + "grad_norm": 0.41340839862823486, + "learning_rate": 4.908594664826708e-06, + "loss": 0.5292, + "step": 167 + }, + { + "epoch": 1.15, + "grad_norm": 0.44648951292037964, + "learning_rate": 4.905863166085076e-06, + "loss": 0.5757, + "step": 168 + }, + { + "epoch": 1.1568181818181817, + "grad_norm": 0.4285544753074646, + "learning_rate": 4.903092235540679e-06, + "loss": 0.5102, + "step": 169 + }, + { + "epoch": 1.1636363636363636, + "grad_norm": 0.42675015330314636, + "learning_rate": 4.900281918608732e-06, + "loss": 0.5397, + "step": 170 + }, + { + "epoch": 1.1704545454545454, + "grad_norm": 0.47636961936950684, + "learning_rate": 4.897432261349984e-06, + "loss": 0.5307, + "step": 171 + }, + { + "epoch": 1.1772727272727272, + "grad_norm": 0.4213427007198334, + "learning_rate": 4.894543310469968e-06, + "loss": 0.54, + "step": 172 + }, + { + "epoch": 1.184090909090909, + "grad_norm": 0.41153451800346375, + "learning_rate": 4.891615113318236e-06, + "loss": 0.5423, + "step": 173 + }, + { + "epoch": 1.190909090909091, + "grad_norm": 0.459185391664505, + "learning_rate": 4.888647717887582e-06, + "loss": 0.5239, + "step": 174 + }, + { + "epoch": 1.1977272727272728, + "grad_norm": 0.48279181122779846, + "learning_rate": 4.8856411728132526e-06, + "loss": 0.5336, + "step": 175 + }, + { + "epoch": 1.2045454545454546, + "grad_norm": 0.4186796247959137, + "learning_rate": 4.8825955273721524e-06, + "loss": 0.5509, + "step": 176 + }, + { + "epoch": 1.2113636363636364, + "grad_norm": 0.4345344007015228, + "learning_rate": 4.879510831482039e-06, + "loss": 0.5415, + "step": 177 + }, + { + "epoch": 1.2181818181818183, + "grad_norm": 0.461592435836792, + "learning_rate": 4.876387135700701e-06, + "loss": 0.5256, + "step": 178 + }, + { + "epoch": 1.225, + "grad_norm": 0.5960240960121155, + "learning_rate": 4.873224491225128e-06, + "loss": 0.5599, + "step": 179 + }, + { + "epoch": 1.231818181818182, + "grad_norm": 0.4662570357322693, + "learning_rate": 4.870022949890676e-06, + "loss": 0.5647, + "step": 180 + }, + { + "epoch": 1.2386363636363638, + "grad_norm": 0.4355057179927826, + "learning_rate": 4.866782564170217e-06, + "loss": 0.5056, + "step": 181 + }, + { + "epoch": 1.2454545454545454, + "grad_norm": 0.48699110746383667, + "learning_rate": 4.863503387173276e-06, + "loss": 0.5394, + "step": 182 + }, + { + "epoch": 1.2522727272727272, + "grad_norm": 0.48679423332214355, + "learning_rate": 4.860185472645161e-06, + "loss": 0.5394, + "step": 183 + }, + { + "epoch": 1.259090909090909, + "grad_norm": 0.6564393043518066, + "learning_rate": 4.856828874966086e-06, + "loss": 0.5259, + "step": 184 + }, + { + "epoch": 1.2659090909090909, + "grad_norm": 0.44762882590293884, + "learning_rate": 4.853433649150276e-06, + "loss": 0.5563, + "step": 185 + }, + { + "epoch": 1.2727272727272727, + "grad_norm": 0.5391108989715576, + "learning_rate": 4.849999850845066e-06, + "loss": 0.5326, + "step": 186 + }, + { + "epoch": 1.2795454545454545, + "grad_norm": 0.4201100170612335, + "learning_rate": 4.8465275363299905e-06, + "loss": 0.526, + "step": 187 + }, + { + "epoch": 1.2863636363636364, + "grad_norm": 0.46211543679237366, + "learning_rate": 4.84301676251586e-06, + "loss": 0.5498, + "step": 188 + }, + { + "epoch": 1.2931818181818182, + "grad_norm": 0.4325430989265442, + "learning_rate": 4.839467586943825e-06, + "loss": 0.556, + "step": 189 + }, + { + "epoch": 1.3, + "grad_norm": 0.46057143807411194, + "learning_rate": 4.835880067784441e-06, + "loss": 0.5079, + "step": 190 + }, + { + "epoch": 1.3068181818181819, + "grad_norm": 0.4920806288719177, + "learning_rate": 4.832254263836708e-06, + "loss": 0.533, + "step": 191 + }, + { + "epoch": 1.3136363636363637, + "grad_norm": 0.46693211793899536, + "learning_rate": 4.828590234527107e-06, + "loss": 0.5542, + "step": 192 + }, + { + "epoch": 1.3204545454545453, + "grad_norm": 0.4227873682975769, + "learning_rate": 4.82488803990863e-06, + "loss": 0.544, + "step": 193 + }, + { + "epoch": 1.3272727272727272, + "grad_norm": 0.44745922088623047, + "learning_rate": 4.821147740659795e-06, + "loss": 0.4962, + "step": 194 + }, + { + "epoch": 1.334090909090909, + "grad_norm": 0.46137240529060364, + "learning_rate": 4.817369398083648e-06, + "loss": 0.5359, + "step": 195 + }, + { + "epoch": 1.3409090909090908, + "grad_norm": 0.6483879685401917, + "learning_rate": 4.813553074106761e-06, + "loss": 0.5323, + "step": 196 + }, + { + "epoch": 1.3477272727272727, + "grad_norm": 0.6602574586868286, + "learning_rate": 4.809698831278217e-06, + "loss": 0.5161, + "step": 197 + }, + { + "epoch": 1.3545454545454545, + "grad_norm": 0.47758543491363525, + "learning_rate": 4.805806732768585e-06, + "loss": 0.5214, + "step": 198 + }, + { + "epoch": 1.3613636363636363, + "grad_norm": 0.4587438106536865, + "learning_rate": 4.801876842368882e-06, + "loss": 0.5208, + "step": 199 + }, + { + "epoch": 1.3681818181818182, + "grad_norm": 0.46055835485458374, + "learning_rate": 4.797909224489531e-06, + "loss": 0.5374, + "step": 200 + }, + { + "epoch": 1.375, + "grad_norm": 0.43727558851242065, + "learning_rate": 4.793903944159303e-06, + "loss": 0.5351, + "step": 201 + }, + { + "epoch": 1.3818181818181818, + "grad_norm": 0.4573572874069214, + "learning_rate": 4.789861067024253e-06, + "loss": 0.4997, + "step": 202 + }, + { + "epoch": 1.3886363636363637, + "grad_norm": 0.5921999216079712, + "learning_rate": 4.785780659346642e-06, + "loss": 0.5168, + "step": 203 + }, + { + "epoch": 1.3954545454545455, + "grad_norm": 0.47943904995918274, + "learning_rate": 4.781662788003851e-06, + "loss": 0.5513, + "step": 204 + }, + { + "epoch": 1.4022727272727273, + "grad_norm": 0.43733686208724976, + "learning_rate": 4.777507520487289e-06, + "loss": 0.5281, + "step": 205 + }, + { + "epoch": 1.4090909090909092, + "grad_norm": 0.4289107620716095, + "learning_rate": 4.773314924901281e-06, + "loss": 0.524, + "step": 206 + }, + { + "epoch": 1.415909090909091, + "grad_norm": 0.4203416109085083, + "learning_rate": 4.769085069961955e-06, + "loss": 0.5563, + "step": 207 + }, + { + "epoch": 1.4227272727272728, + "grad_norm": 0.4205201268196106, + "learning_rate": 4.764818024996117e-06, + "loss": 0.5202, + "step": 208 + }, + { + "epoch": 1.4295454545454547, + "grad_norm": 0.4296267032623291, + "learning_rate": 4.760513859940112e-06, + "loss": 0.5113, + "step": 209 + }, + { + "epoch": 1.4363636363636363, + "grad_norm": 0.3979087471961975, + "learning_rate": 4.756172645338675e-06, + "loss": 0.5313, + "step": 210 + }, + { + "epoch": 1.4431818181818181, + "grad_norm": 0.45232149958610535, + "learning_rate": 4.751794452343785e-06, + "loss": 0.5204, + "step": 211 + }, + { + "epoch": 1.45, + "grad_norm": 0.4309398829936981, + "learning_rate": 4.747379352713489e-06, + "loss": 0.5216, + "step": 212 + }, + { + "epoch": 1.4568181818181818, + "grad_norm": 0.45721709728240967, + "learning_rate": 4.7429274188107275e-06, + "loss": 0.5279, + "step": 213 + }, + { + "epoch": 1.4636363636363636, + "grad_norm": 0.46138322353363037, + "learning_rate": 4.738438723602154e-06, + "loss": 0.5287, + "step": 214 + }, + { + "epoch": 1.4704545454545455, + "grad_norm": 0.41752108931541443, + "learning_rate": 4.733913340656933e-06, + "loss": 0.5399, + "step": 215 + }, + { + "epoch": 1.4772727272727273, + "grad_norm": 0.45447197556495667, + "learning_rate": 4.729351344145536e-06, + "loss": 0.5277, + "step": 216 + }, + { + "epoch": 1.4840909090909091, + "grad_norm": 0.42524856328964233, + "learning_rate": 4.7247528088385296e-06, + "loss": 0.5103, + "step": 217 + }, + { + "epoch": 1.490909090909091, + "grad_norm": 0.46024221181869507, + "learning_rate": 4.720117810105341e-06, + "loss": 0.5172, + "step": 218 + }, + { + "epoch": 1.4977272727272728, + "grad_norm": 0.5868630409240723, + "learning_rate": 4.715446423913036e-06, + "loss": 0.5409, + "step": 219 + }, + { + "epoch": 1.5045454545454544, + "grad_norm": 0.424813836812973, + "learning_rate": 4.710738726825059e-06, + "loss": 0.5495, + "step": 220 + }, + { + "epoch": 1.5113636363636362, + "grad_norm": 0.49356943368911743, + "learning_rate": 4.705994795999991e-06, + "loss": 0.519, + "step": 221 + }, + { + "epoch": 1.518181818181818, + "grad_norm": 0.44291776418685913, + "learning_rate": 4.701214709190277e-06, + "loss": 0.5232, + "step": 222 + }, + { + "epoch": 1.525, + "grad_norm": 0.48171764612197876, + "learning_rate": 4.696398544740955e-06, + "loss": 0.5501, + "step": 223 + }, + { + "epoch": 1.5318181818181817, + "grad_norm": 0.4296175539493561, + "learning_rate": 4.69154638158837e-06, + "loss": 0.5087, + "step": 224 + }, + { + "epoch": 1.5386363636363636, + "grad_norm": 0.4715246260166168, + "learning_rate": 4.686658299258881e-06, + "loss": 0.5363, + "step": 225 + }, + { + "epoch": 1.5454545454545454, + "grad_norm": 0.44422832131385803, + "learning_rate": 4.681734377867562e-06, + "loss": 0.5371, + "step": 226 + }, + { + "epoch": 1.5522727272727272, + "grad_norm": 4.221654891967773, + "learning_rate": 4.67677469811688e-06, + "loss": 0.5191, + "step": 227 + }, + { + "epoch": 1.559090909090909, + "grad_norm": 0.41857993602752686, + "learning_rate": 4.671779341295378e-06, + "loss": 0.5191, + "step": 228 + }, + { + "epoch": 1.565909090909091, + "grad_norm": 0.4950161278247833, + "learning_rate": 4.666748389276344e-06, + "loss": 0.5311, + "step": 229 + }, + { + "epoch": 1.5727272727272728, + "grad_norm": 0.44641605019569397, + "learning_rate": 4.661681924516466e-06, + "loss": 0.5174, + "step": 230 + }, + { + "epoch": 1.5795454545454546, + "grad_norm": 0.39064037799835205, + "learning_rate": 4.6565800300544805e-06, + "loss": 0.515, + "step": 231 + }, + { + "epoch": 1.5863636363636364, + "grad_norm": 0.7764399647712708, + "learning_rate": 4.651442789509813e-06, + "loss": 0.5282, + "step": 232 + }, + { + "epoch": 1.5931818181818183, + "grad_norm": 0.44421035051345825, + "learning_rate": 4.646270287081208e-06, + "loss": 0.5324, + "step": 233 + }, + { + "epoch": 1.6, + "grad_norm": 0.5149928331375122, + "learning_rate": 4.641062607545347e-06, + "loss": 0.5264, + "step": 234 + }, + { + "epoch": 1.606818181818182, + "grad_norm": 0.4784093201160431, + "learning_rate": 4.6358198362554585e-06, + "loss": 0.552, + "step": 235 + }, + { + "epoch": 1.6136363636363638, + "grad_norm": 0.4138723313808441, + "learning_rate": 4.630542059139923e-06, + "loss": 0.5202, + "step": 236 + }, + { + "epoch": 1.6204545454545456, + "grad_norm": 0.4509594440460205, + "learning_rate": 4.625229362700863e-06, + "loss": 0.5318, + "step": 237 + }, + { + "epoch": 1.6272727272727274, + "grad_norm": 0.44979560375213623, + "learning_rate": 4.61988183401272e-06, + "loss": 0.521, + "step": 238 + }, + { + "epoch": 1.634090909090909, + "grad_norm": 0.4525575041770935, + "learning_rate": 4.614499560720837e-06, + "loss": 0.5073, + "step": 239 + }, + { + "epoch": 1.6409090909090909, + "grad_norm": 0.44126543402671814, + "learning_rate": 4.609082631040012e-06, + "loss": 0.5058, + "step": 240 + }, + { + "epoch": 1.6477272727272727, + "grad_norm": 0.887867271900177, + "learning_rate": 4.603631133753061e-06, + "loss": 0.5471, + "step": 241 + }, + { + "epoch": 1.6545454545454545, + "grad_norm": 0.4497375786304474, + "learning_rate": 4.598145158209356e-06, + "loss": 0.501, + "step": 242 + }, + { + "epoch": 1.6613636363636364, + "grad_norm": 0.45681032538414, + "learning_rate": 4.592624794323366e-06, + "loss": 0.5306, + "step": 243 + }, + { + "epoch": 1.6681818181818182, + "grad_norm": 0.5186678171157837, + "learning_rate": 4.587070132573178e-06, + "loss": 0.5097, + "step": 244 + }, + { + "epoch": 1.675, + "grad_norm": 0.44289669394493103, + "learning_rate": 4.581481263999019e-06, + "loss": 0.5085, + "step": 245 + }, + { + "epoch": 1.6818181818181817, + "grad_norm": 0.4571002125740051, + "learning_rate": 4.575858280201761e-06, + "loss": 0.5327, + "step": 246 + }, + { + "epoch": 1.6886363636363635, + "grad_norm": 0.43753257393836975, + "learning_rate": 4.570201273341418e-06, + "loss": 0.5167, + "step": 247 + }, + { + "epoch": 1.6954545454545453, + "grad_norm": 0.44994574785232544, + "learning_rate": 4.564510336135642e-06, + "loss": 0.5554, + "step": 248 + }, + { + "epoch": 1.7022727272727272, + "grad_norm": 0.6185874938964844, + "learning_rate": 4.558785561858196e-06, + "loss": 0.5115, + "step": 249 + }, + { + "epoch": 1.709090909090909, + "grad_norm": 0.4883944094181061, + "learning_rate": 4.5530270443374305e-06, + "loss": 0.5312, + "step": 250 + }, + { + "epoch": 1.7159090909090908, + "grad_norm": 0.46433499455451965, + "learning_rate": 4.547234877954741e-06, + "loss": 0.5368, + "step": 251 + }, + { + "epoch": 1.7227272727272727, + "grad_norm": 0.4270826578140259, + "learning_rate": 4.541409157643027e-06, + "loss": 0.5258, + "step": 252 + }, + { + "epoch": 1.7295454545454545, + "grad_norm": 0.4390241205692291, + "learning_rate": 4.535549978885132e-06, + "loss": 0.5259, + "step": 253 + }, + { + "epoch": 1.7363636363636363, + "grad_norm": 0.42756086587905884, + "learning_rate": 4.5296574377122765e-06, + "loss": 0.5217, + "step": 254 + }, + { + "epoch": 1.7431818181818182, + "grad_norm": 0.48124927282333374, + "learning_rate": 4.5237316307024895e-06, + "loss": 0.5095, + "step": 255 + }, + { + "epoch": 1.75, + "grad_norm": 0.4394672214984894, + "learning_rate": 4.517772654979024e-06, + "loss": 0.5116, + "step": 256 + }, + { + "epoch": 1.7568181818181818, + "grad_norm": 0.5805886387825012, + "learning_rate": 4.51178060820876e-06, + "loss": 0.5145, + "step": 257 + }, + { + "epoch": 1.7636363636363637, + "grad_norm": 0.4845556616783142, + "learning_rate": 4.505755588600613e-06, + "loss": 0.5064, + "step": 258 + }, + { + "epoch": 1.7704545454545455, + "grad_norm": 0.45300963521003723, + "learning_rate": 4.499697694903915e-06, + "loss": 0.5362, + "step": 259 + }, + { + "epoch": 1.7772727272727273, + "grad_norm": 0.45417436957359314, + "learning_rate": 4.493607026406802e-06, + "loss": 0.5234, + "step": 260 + }, + { + "epoch": 1.7840909090909092, + "grad_norm": 0.4577132761478424, + "learning_rate": 4.487483682934587e-06, + "loss": 0.5271, + "step": 261 + }, + { + "epoch": 1.790909090909091, + "grad_norm": 0.4518176019191742, + "learning_rate": 4.481327764848118e-06, + "loss": 0.4853, + "step": 262 + }, + { + "epoch": 1.7977272727272728, + "grad_norm": 0.43804875016212463, + "learning_rate": 4.47513937304214e-06, + "loss": 0.5188, + "step": 263 + }, + { + "epoch": 1.8045454545454547, + "grad_norm": 0.4401354193687439, + "learning_rate": 4.4689186089436365e-06, + "loss": 0.5079, + "step": 264 + }, + { + "epoch": 1.8113636363636365, + "grad_norm": 0.4051227569580078, + "learning_rate": 4.462665574510169e-06, + "loss": 0.5084, + "step": 265 + }, + { + "epoch": 1.8181818181818183, + "grad_norm": 0.5269004106521606, + "learning_rate": 4.456380372228208e-06, + "loss": 0.528, + "step": 266 + }, + { + "epoch": 1.825, + "grad_norm": 0.44371986389160156, + "learning_rate": 4.450063105111447e-06, + "loss": 0.5388, + "step": 267 + }, + { + "epoch": 1.8318181818181818, + "grad_norm": 0.41871729493141174, + "learning_rate": 4.443713876699124e-06, + "loss": 0.502, + "step": 268 + }, + { + "epoch": 1.8386363636363636, + "grad_norm": 0.464652419090271, + "learning_rate": 4.4373327910543125e-06, + "loss": 0.5045, + "step": 269 + }, + { + "epoch": 1.8454545454545455, + "grad_norm": 0.43198078870773315, + "learning_rate": 4.430919952762226e-06, + "loss": 0.5149, + "step": 270 + }, + { + "epoch": 1.8522727272727273, + "grad_norm": 0.42869722843170166, + "learning_rate": 4.424475466928499e-06, + "loss": 0.5329, + "step": 271 + }, + { + "epoch": 1.8590909090909091, + "grad_norm": 0.529680609703064, + "learning_rate": 4.417999439177465e-06, + "loss": 0.4939, + "step": 272 + }, + { + "epoch": 1.865909090909091, + "grad_norm": 0.4562613368034363, + "learning_rate": 4.4114919756504275e-06, + "loss": 0.5389, + "step": 273 + }, + { + "epoch": 1.8727272727272726, + "grad_norm": 0.4480995237827301, + "learning_rate": 4.404953183003916e-06, + "loss": 0.51, + "step": 274 + }, + { + "epoch": 1.8795454545454544, + "grad_norm": 0.41215088963508606, + "learning_rate": 4.398383168407941e-06, + "loss": 0.5386, + "step": 275 + }, + { + "epoch": 1.8863636363636362, + "grad_norm": 0.8426409363746643, + "learning_rate": 4.391782039544239e-06, + "loss": 0.498, + "step": 276 + }, + { + "epoch": 1.893181818181818, + "grad_norm": 0.5359936356544495, + "learning_rate": 4.385149904604502e-06, + "loss": 0.51, + "step": 277 + }, + { + "epoch": 1.9, + "grad_norm": 0.4462437033653259, + "learning_rate": 4.378486872288611e-06, + "loss": 0.5082, + "step": 278 + }, + { + "epoch": 1.9068181818181817, + "grad_norm": 0.445359468460083, + "learning_rate": 4.371793051802849e-06, + "loss": 0.5171, + "step": 279 + }, + { + "epoch": 1.9136363636363636, + "grad_norm": 0.4223516583442688, + "learning_rate": 4.365068552858116e-06, + "loss": 0.5218, + "step": 280 + }, + { + "epoch": 1.9204545454545454, + "grad_norm": 0.43929681181907654, + "learning_rate": 4.358313485668124e-06, + "loss": 0.5156, + "step": 281 + }, + { + "epoch": 1.9272727272727272, + "grad_norm": 0.42535650730133057, + "learning_rate": 4.3515279609476e-06, + "loss": 0.4964, + "step": 282 + }, + { + "epoch": 1.934090909090909, + "grad_norm": 0.46448391675949097, + "learning_rate": 4.3447120899104615e-06, + "loss": 0.5182, + "step": 283 + }, + { + "epoch": 1.940909090909091, + "grad_norm": 0.4082206189632416, + "learning_rate": 4.337865984268002e-06, + "loss": 0.5175, + "step": 284 + }, + { + "epoch": 1.9477272727272728, + "grad_norm": 0.42423564195632935, + "learning_rate": 4.3309897562270525e-06, + "loss": 0.5352, + "step": 285 + }, + { + "epoch": 1.9545454545454546, + "grad_norm": 0.5240079760551453, + "learning_rate": 4.324083518488151e-06, + "loss": 0.5157, + "step": 286 + }, + { + "epoch": 1.9613636363636364, + "grad_norm": 0.4517645835876465, + "learning_rate": 4.317147384243688e-06, + "loss": 0.5399, + "step": 287 + }, + { + "epoch": 1.9681818181818183, + "grad_norm": 0.44054290652275085, + "learning_rate": 4.3101814671760546e-06, + "loss": 0.5293, + "step": 288 + }, + { + "epoch": 1.975, + "grad_norm": 0.4256063401699066, + "learning_rate": 4.303185881455778e-06, + "loss": 0.5323, + "step": 289 + }, + { + "epoch": 1.981818181818182, + "grad_norm": 0.43106094002723694, + "learning_rate": 4.296160741739652e-06, + "loss": 0.5259, + "step": 290 + }, + { + "epoch": 1.9886363636363638, + "grad_norm": 0.43816477060317993, + "learning_rate": 4.289106163168858e-06, + "loss": 0.5098, + "step": 291 + }, + { + "epoch": 1.9954545454545456, + "grad_norm": 1.1177959442138672, + "learning_rate": 4.282022261367074e-06, + "loss": 0.513, + "step": 292 + }, + { + "epoch": 2.006818181818182, + "grad_norm": 0.8377058506011963, + "learning_rate": 4.274909152438582e-06, + "loss": 0.9984, + "step": 293 + }, + { + "epoch": 2.0136363636363637, + "grad_norm": 0.4339721202850342, + "learning_rate": 4.267766952966369e-06, + "loss": 0.493, + "step": 294 + }, + { + "epoch": 2.0204545454545455, + "grad_norm": 0.39961451292037964, + "learning_rate": 4.260595780010209e-06, + "loss": 0.486, + "step": 295 + }, + { + "epoch": 2.0272727272727273, + "grad_norm": 0.4809112548828125, + "learning_rate": 4.2533957511047485e-06, + "loss": 0.4803, + "step": 296 + }, + { + "epoch": 2.034090909090909, + "grad_norm": 0.414700984954834, + "learning_rate": 4.24616698425758e-06, + "loss": 0.4566, + "step": 297 + }, + { + "epoch": 2.040909090909091, + "grad_norm": 0.552201509475708, + "learning_rate": 4.238909597947307e-06, + "loss": 0.4775, + "step": 298 + }, + { + "epoch": 2.047727272727273, + "grad_norm": 0.4647877812385559, + "learning_rate": 4.231623711121603e-06, + "loss": 0.4944, + "step": 299 + }, + { + "epoch": 2.0545454545454547, + "grad_norm": 0.684251606464386, + "learning_rate": 4.224309443195261e-06, + "loss": 0.495, + "step": 300 + }, + { + "epoch": 2.0613636363636365, + "grad_norm": 0.42145106196403503, + "learning_rate": 4.2169669140482365e-06, + "loss": 0.4809, + "step": 301 + }, + { + "epoch": 2.0681818181818183, + "grad_norm": 0.43771296739578247, + "learning_rate": 4.2095962440236846e-06, + "loss": 0.5033, + "step": 302 + }, + { + "epoch": 2.075, + "grad_norm": 0.4323962926864624, + "learning_rate": 4.202197553925983e-06, + "loss": 0.4843, + "step": 303 + }, + { + "epoch": 2.081818181818182, + "grad_norm": 0.40169817209243774, + "learning_rate": 4.194770965018758e-06, + "loss": 0.504, + "step": 304 + }, + { + "epoch": 2.088636363636364, + "grad_norm": 0.4206427037715912, + "learning_rate": 4.187316599022892e-06, + "loss": 0.5059, + "step": 305 + }, + { + "epoch": 2.0954545454545457, + "grad_norm": 0.4550352692604065, + "learning_rate": 4.179834578114531e-06, + "loss": 0.4717, + "step": 306 + }, + { + "epoch": 2.102272727272727, + "grad_norm": 0.4193796217441559, + "learning_rate": 4.172325024923083e-06, + "loss": 0.5018, + "step": 307 + }, + { + "epoch": 2.109090909090909, + "grad_norm": 0.550195038318634, + "learning_rate": 4.164788062529203e-06, + "loss": 0.4961, + "step": 308 + }, + { + "epoch": 2.1159090909090907, + "grad_norm": 0.43539106845855713, + "learning_rate": 4.157223814462784e-06, + "loss": 0.4909, + "step": 309 + }, + { + "epoch": 2.1227272727272726, + "grad_norm": 0.4565756022930145, + "learning_rate": 4.149632404700925e-06, + "loss": 0.49, + "step": 310 + }, + { + "epoch": 2.1295454545454544, + "grad_norm": 0.4206918179988861, + "learning_rate": 4.142013957665903e-06, + "loss": 0.4722, + "step": 311 + }, + { + "epoch": 2.1363636363636362, + "grad_norm": 0.4098123013973236, + "learning_rate": 4.134368598223132e-06, + "loss": 0.4791, + "step": 312 + }, + { + "epoch": 2.143181818181818, + "grad_norm": 0.4387701153755188, + "learning_rate": 4.126696451679116e-06, + "loss": 0.4758, + "step": 313 + }, + { + "epoch": 2.15, + "grad_norm": 0.4594134986400604, + "learning_rate": 4.118997643779401e-06, + "loss": 0.5009, + "step": 314 + }, + { + "epoch": 2.1568181818181817, + "grad_norm": 0.4581888020038605, + "learning_rate": 4.111272300706502e-06, + "loss": 0.4915, + "step": 315 + }, + { + "epoch": 2.1636363636363636, + "grad_norm": 0.5279524326324463, + "learning_rate": 4.1035205490778505e-06, + "loss": 0.5056, + "step": 316 + }, + { + "epoch": 2.1704545454545454, + "grad_norm": 0.4253177344799042, + "learning_rate": 4.095742515943703e-06, + "loss": 0.5001, + "step": 317 + }, + { + "epoch": 2.1772727272727272, + "grad_norm": 0.5294811129570007, + "learning_rate": 4.087938328785071e-06, + "loss": 0.5027, + "step": 318 + }, + { + "epoch": 2.184090909090909, + "grad_norm": 0.4642890989780426, + "learning_rate": 4.080108115511629e-06, + "loss": 0.5051, + "step": 319 + }, + { + "epoch": 2.190909090909091, + "grad_norm": 0.43700429797172546, + "learning_rate": 4.072252004459612e-06, + "loss": 0.4809, + "step": 320 + }, + { + "epoch": 2.1977272727272728, + "grad_norm": 0.4536914825439453, + "learning_rate": 4.064370124389718e-06, + "loss": 0.5129, + "step": 321 + }, + { + "epoch": 2.2045454545454546, + "grad_norm": 0.42350664734840393, + "learning_rate": 4.056462604484998e-06, + "loss": 0.4912, + "step": 322 + }, + { + "epoch": 2.2113636363636364, + "grad_norm": 0.47886520624160767, + "learning_rate": 4.048529574348734e-06, + "loss": 0.4757, + "step": 323 + }, + { + "epoch": 2.2181818181818183, + "grad_norm": 0.5205736756324768, + "learning_rate": 4.040571164002319e-06, + "loss": 0.4821, + "step": 324 + }, + { + "epoch": 2.225, + "grad_norm": 0.4465322196483612, + "learning_rate": 4.032587503883124e-06, + "loss": 0.5057, + "step": 325 + }, + { + "epoch": 2.231818181818182, + "grad_norm": 0.41364315152168274, + "learning_rate": 4.0245787248423614e-06, + "loss": 0.4809, + "step": 326 + }, + { + "epoch": 2.2386363636363638, + "grad_norm": 0.42846816778182983, + "learning_rate": 4.0165449581429404e-06, + "loss": 0.4522, + "step": 327 + }, + { + "epoch": 2.2454545454545456, + "grad_norm": 0.4604497253894806, + "learning_rate": 4.008486335457312e-06, + "loss": 0.4879, + "step": 328 + }, + { + "epoch": 2.2522727272727274, + "grad_norm": 0.45866841077804565, + "learning_rate": 4.000402988865316e-06, + "loss": 0.4969, + "step": 329 + }, + { + "epoch": 2.2590909090909093, + "grad_norm": 0.5116038918495178, + "learning_rate": 3.992295050852013e-06, + "loss": 0.5015, + "step": 330 + }, + { + "epoch": 2.265909090909091, + "grad_norm": 0.4389069080352783, + "learning_rate": 3.984162654305516e-06, + "loss": 0.4797, + "step": 331 + }, + { + "epoch": 2.2727272727272725, + "grad_norm": 0.45724064111709595, + "learning_rate": 3.976005932514807e-06, + "loss": 0.4915, + "step": 332 + }, + { + "epoch": 2.2795454545454543, + "grad_norm": 0.44370219111442566, + "learning_rate": 3.967825019167559e-06, + "loss": 0.499, + "step": 333 + }, + { + "epoch": 2.286363636363636, + "grad_norm": 0.4414970278739929, + "learning_rate": 3.959620048347938e-06, + "loss": 0.447, + "step": 334 + }, + { + "epoch": 2.293181818181818, + "grad_norm": 0.7909289598464966, + "learning_rate": 3.951391154534415e-06, + "loss": 0.4658, + "step": 335 + }, + { + "epoch": 2.3, + "grad_norm": 0.4643534719944, + "learning_rate": 3.943138472597549e-06, + "loss": 0.4941, + "step": 336 + }, + { + "epoch": 2.3068181818181817, + "grad_norm": 0.5114858150482178, + "learning_rate": 3.934862137797788e-06, + "loss": 0.4733, + "step": 337 + }, + { + "epoch": 2.3136363636363635, + "grad_norm": 0.4057219624519348, + "learning_rate": 3.9265622857832455e-06, + "loss": 0.4965, + "step": 338 + }, + { + "epoch": 2.3204545454545453, + "grad_norm": 0.4124796390533447, + "learning_rate": 3.918239052587481e-06, + "loss": 0.4508, + "step": 339 + }, + { + "epoch": 2.327272727272727, + "grad_norm": 0.43895164132118225, + "learning_rate": 3.909892574627267e-06, + "loss": 0.4939, + "step": 340 + }, + { + "epoch": 2.334090909090909, + "grad_norm": 0.47555556893348694, + "learning_rate": 3.901522988700355e-06, + "loss": 0.4541, + "step": 341 + }, + { + "epoch": 2.340909090909091, + "grad_norm": 0.44449353218078613, + "learning_rate": 3.893130431983234e-06, + "loss": 0.4883, + "step": 342 + }, + { + "epoch": 2.3477272727272727, + "grad_norm": 0.441793829202652, + "learning_rate": 3.884715042028882e-06, + "loss": 0.4911, + "step": 343 + }, + { + "epoch": 2.3545454545454545, + "grad_norm": 0.4836761951446533, + "learning_rate": 3.876276956764509e-06, + "loss": 0.4838, + "step": 344 + }, + { + "epoch": 2.3613636363636363, + "grad_norm": 0.7246759533882141, + "learning_rate": 3.867816314489301e-06, + "loss": 0.5014, + "step": 345 + }, + { + "epoch": 2.368181818181818, + "grad_norm": 0.5535604357719421, + "learning_rate": 3.8593332538721465e-06, + "loss": 0.4654, + "step": 346 + }, + { + "epoch": 2.375, + "grad_norm": 0.4468318819999695, + "learning_rate": 3.8508279139493736e-06, + "loss": 0.463, + "step": 347 + }, + { + "epoch": 2.381818181818182, + "grad_norm": 0.4505653381347656, + "learning_rate": 3.84230043412246e-06, + "loss": 0.4614, + "step": 348 + }, + { + "epoch": 2.3886363636363637, + "grad_norm": 0.4461449384689331, + "learning_rate": 3.833750954155757e-06, + "loss": 0.4777, + "step": 349 + }, + { + "epoch": 2.3954545454545455, + "grad_norm": 0.423711895942688, + "learning_rate": 3.825179614174195e-06, + "loss": 0.4581, + "step": 350 + }, + { + "epoch": 2.4022727272727273, + "grad_norm": 0.462819904088974, + "learning_rate": 3.816586554660987e-06, + "loss": 0.4782, + "step": 351 + }, + { + "epoch": 2.409090909090909, + "grad_norm": 0.44206613302230835, + "learning_rate": 3.807971916455325e-06, + "loss": 0.477, + "step": 352 + }, + { + "epoch": 2.415909090909091, + "grad_norm": 0.4178541600704193, + "learning_rate": 3.799335840750077e-06, + "loss": 0.4981, + "step": 353 + }, + { + "epoch": 2.422727272727273, + "grad_norm": 0.670315682888031, + "learning_rate": 3.790678469089465e-06, + "loss": 0.4762, + "step": 354 + }, + { + "epoch": 2.4295454545454547, + "grad_norm": 0.44383955001831055, + "learning_rate": 3.7819999433667503e-06, + "loss": 0.4807, + "step": 355 + }, + { + "epoch": 2.4363636363636365, + "grad_norm": 0.4850422143936157, + "learning_rate": 3.773300405821908e-06, + "loss": 0.479, + "step": 356 + }, + { + "epoch": 2.4431818181818183, + "grad_norm": 0.458435595035553, + "learning_rate": 3.764579999039293e-06, + "loss": 0.4628, + "step": 357 + }, + { + "epoch": 2.45, + "grad_norm": 0.43913447856903076, + "learning_rate": 3.7558388659453052e-06, + "loss": 0.4948, + "step": 358 + }, + { + "epoch": 2.456818181818182, + "grad_norm": 0.4294499456882477, + "learning_rate": 3.7470771498060455e-06, + "loss": 0.4979, + "step": 359 + }, + { + "epoch": 2.463636363636364, + "grad_norm": 0.4472196400165558, + "learning_rate": 3.7382949942249695e-06, + "loss": 0.496, + "step": 360 + }, + { + "epoch": 2.4704545454545457, + "grad_norm": 0.4681987762451172, + "learning_rate": 3.7294925431405306e-06, + "loss": 0.4591, + "step": 361 + }, + { + "epoch": 2.4772727272727275, + "grad_norm": 0.40683335065841675, + "learning_rate": 3.720669940823827e-06, + "loss": 0.4755, + "step": 362 + }, + { + "epoch": 2.484090909090909, + "grad_norm": 0.46134939789772034, + "learning_rate": 3.7118273318762275e-06, + "loss": 0.4956, + "step": 363 + }, + { + "epoch": 2.4909090909090907, + "grad_norm": 0.4279940724372864, + "learning_rate": 3.702964861227013e-06, + "loss": 0.4815, + "step": 364 + }, + { + "epoch": 2.4977272727272726, + "grad_norm": 0.4699287414550781, + "learning_rate": 3.694082674130991e-06, + "loss": 0.4846, + "step": 365 + }, + { + "epoch": 2.5045454545454544, + "grad_norm": 0.4230785369873047, + "learning_rate": 3.6851809161661206e-06, + "loss": 0.5077, + "step": 366 + }, + { + "epoch": 2.5113636363636362, + "grad_norm": 0.4509921967983246, + "learning_rate": 3.6762597332311254e-06, + "loss": 0.5134, + "step": 367 + }, + { + "epoch": 2.518181818181818, + "grad_norm": 0.45494896173477173, + "learning_rate": 3.6673192715431016e-06, + "loss": 0.4685, + "step": 368 + }, + { + "epoch": 2.525, + "grad_norm": 0.4804443120956421, + "learning_rate": 3.658359677635122e-06, + "loss": 0.4458, + "step": 369 + }, + { + "epoch": 2.5318181818181817, + "grad_norm": 0.4149218499660492, + "learning_rate": 3.649381098353834e-06, + "loss": 0.5102, + "step": 370 + }, + { + "epoch": 2.5386363636363636, + "grad_norm": 0.4423210024833679, + "learning_rate": 3.6403836808570512e-06, + "loss": 0.5034, + "step": 371 + }, + { + "epoch": 2.5454545454545454, + "grad_norm": 0.4915727376937866, + "learning_rate": 3.631367572611348e-06, + "loss": 0.4889, + "step": 372 + }, + { + "epoch": 2.5522727272727272, + "grad_norm": 0.4189188778400421, + "learning_rate": 3.6223329213896313e-06, + "loss": 0.4829, + "step": 373 + }, + { + "epoch": 2.559090909090909, + "grad_norm": 0.511681079864502, + "learning_rate": 3.613279875268731e-06, + "loss": 0.4786, + "step": 374 + }, + { + "epoch": 2.565909090909091, + "grad_norm": 0.430137038230896, + "learning_rate": 3.604208582626964e-06, + "loss": 0.4783, + "step": 375 + }, + { + "epoch": 2.5727272727272728, + "grad_norm": 0.5629504919052124, + "learning_rate": 3.5951191921417063e-06, + "loss": 0.4438, + "step": 376 + }, + { + "epoch": 2.5795454545454546, + "grad_norm": 0.4354007840156555, + "learning_rate": 3.586011852786955e-06, + "loss": 0.4869, + "step": 377 + }, + { + "epoch": 2.5863636363636364, + "grad_norm": 0.45930856466293335, + "learning_rate": 3.5768867138308872e-06, + "loss": 0.4683, + "step": 378 + }, + { + "epoch": 2.5931818181818183, + "grad_norm": 0.4965468943119049, + "learning_rate": 3.5677439248334133e-06, + "loss": 0.4739, + "step": 379 + }, + { + "epoch": 2.6, + "grad_norm": 0.39618629217147827, + "learning_rate": 3.5585836356437266e-06, + "loss": 0.486, + "step": 380 + }, + { + "epoch": 2.606818181818182, + "grad_norm": 0.504718542098999, + "learning_rate": 3.5494059963978433e-06, + "loss": 0.4608, + "step": 381 + }, + { + "epoch": 2.6136363636363638, + "grad_norm": 0.3951753079891205, + "learning_rate": 3.540211157516149e-06, + "loss": 0.4973, + "step": 382 + }, + { + "epoch": 2.6204545454545456, + "grad_norm": 0.45937463641166687, + "learning_rate": 3.530999269700927e-06, + "loss": 0.4751, + "step": 383 + }, + { + "epoch": 2.6272727272727274, + "grad_norm": 0.4644455909729004, + "learning_rate": 3.521770483933891e-06, + "loss": 0.5003, + "step": 384 + }, + { + "epoch": 2.634090909090909, + "grad_norm": 0.4427309036254883, + "learning_rate": 3.5125249514737093e-06, + "loss": 0.4779, + "step": 385 + }, + { + "epoch": 2.6409090909090907, + "grad_norm": 0.4289201498031616, + "learning_rate": 3.503262823853527e-06, + "loss": 0.48, + "step": 386 + }, + { + "epoch": 2.6477272727272725, + "grad_norm": 0.5118290185928345, + "learning_rate": 3.493984252878483e-06, + "loss": 0.4858, + "step": 387 + }, + { + "epoch": 2.6545454545454543, + "grad_norm": 0.4599657952785492, + "learning_rate": 3.484689390623218e-06, + "loss": 0.4813, + "step": 388 + }, + { + "epoch": 2.661363636363636, + "grad_norm": 0.5077105164527893, + "learning_rate": 3.4753783894293886e-06, + "loss": 0.4732, + "step": 389 + }, + { + "epoch": 2.668181818181818, + "grad_norm": 0.5006387829780579, + "learning_rate": 3.466051401903162e-06, + "loss": 0.4743, + "step": 390 + }, + { + "epoch": 2.675, + "grad_norm": 0.4769401252269745, + "learning_rate": 3.4567085809127247e-06, + "loss": 0.4941, + "step": 391 + }, + { + "epoch": 2.6818181818181817, + "grad_norm": 0.4538915455341339, + "learning_rate": 3.4473500795857674e-06, + "loss": 0.4679, + "step": 392 + }, + { + "epoch": 2.6886363636363635, + "grad_norm": 0.4370926320552826, + "learning_rate": 3.4379760513069804e-06, + "loss": 0.4866, + "step": 393 + }, + { + "epoch": 2.6954545454545453, + "grad_norm": 0.445913165807724, + "learning_rate": 3.428586649715542e-06, + "loss": 0.4914, + "step": 394 + }, + { + "epoch": 2.702272727272727, + "grad_norm": 0.45687681436538696, + "learning_rate": 3.4191820287025916e-06, + "loss": 0.474, + "step": 395 + }, + { + "epoch": 2.709090909090909, + "grad_norm": 1.1952316761016846, + "learning_rate": 3.4097623424087196e-06, + "loss": 0.4905, + "step": 396 + }, + { + "epoch": 2.715909090909091, + "grad_norm": 0.41248032450675964, + "learning_rate": 3.4003277452214284e-06, + "loss": 0.4753, + "step": 397 + }, + { + "epoch": 2.7227272727272727, + "grad_norm": 0.4349999725818634, + "learning_rate": 3.3908783917726123e-06, + "loss": 0.4712, + "step": 398 + }, + { + "epoch": 2.7295454545454545, + "grad_norm": 0.42977845668792725, + "learning_rate": 3.381414436936018e-06, + "loss": 0.4936, + "step": 399 + }, + { + "epoch": 2.7363636363636363, + "grad_norm": 0.446321040391922, + "learning_rate": 3.3719360358247054e-06, + "loss": 0.4795, + "step": 400 + }, + { + "epoch": 2.743181818181818, + "grad_norm": 0.4957369565963745, + "learning_rate": 3.36244334378851e-06, + "loss": 0.48, + "step": 401 + }, + { + "epoch": 2.75, + "grad_norm": 0.43430694937705994, + "learning_rate": 3.3529365164114903e-06, + "loss": 0.4903, + "step": 402 + }, + { + "epoch": 2.756818181818182, + "grad_norm": 0.4340098202228546, + "learning_rate": 3.3434157095093846e-06, + "loss": 0.5111, + "step": 403 + }, + { + "epoch": 2.7636363636363637, + "grad_norm": 0.4309055805206299, + "learning_rate": 3.333881079127052e-06, + "loss": 0.4909, + "step": 404 + }, + { + "epoch": 2.7704545454545455, + "grad_norm": 0.43131211400032043, + "learning_rate": 3.3243327815359168e-06, + "loss": 0.4875, + "step": 405 + }, + { + "epoch": 2.7772727272727273, + "grad_norm": 0.4794235825538635, + "learning_rate": 3.314770973231408e-06, + "loss": 0.4766, + "step": 406 + }, + { + "epoch": 2.784090909090909, + "grad_norm": 0.4632478356361389, + "learning_rate": 3.305195810930393e-06, + "loss": 0.4765, + "step": 407 + }, + { + "epoch": 2.790909090909091, + "grad_norm": 0.4836251139640808, + "learning_rate": 3.2956074515686105e-06, + "loss": 0.4573, + "step": 408 + }, + { + "epoch": 2.797727272727273, + "grad_norm": 0.44514402747154236, + "learning_rate": 3.2860060522980945e-06, + "loss": 0.4794, + "step": 409 + }, + { + "epoch": 2.8045454545454547, + "grad_norm": 0.446785569190979, + "learning_rate": 3.276391770484606e-06, + "loss": 0.482, + "step": 410 + }, + { + "epoch": 2.8113636363636365, + "grad_norm": 0.4410347044467926, + "learning_rate": 3.266764763705046e-06, + "loss": 0.5136, + "step": 411 + }, + { + "epoch": 2.8181818181818183, + "grad_norm": 0.4888826906681061, + "learning_rate": 3.257125189744877e-06, + "loss": 0.4852, + "step": 412 + }, + { + "epoch": 2.825, + "grad_norm": 0.4502141773700714, + "learning_rate": 3.247473206595536e-06, + "loss": 0.4764, + "step": 413 + }, + { + "epoch": 2.831818181818182, + "grad_norm": 0.5693864822387695, + "learning_rate": 3.2378089724518464e-06, + "loss": 0.5016, + "step": 414 + }, + { + "epoch": 2.838636363636364, + "grad_norm": 0.49588093161582947, + "learning_rate": 3.228132645709421e-06, + "loss": 0.4786, + "step": 415 + }, + { + "epoch": 2.8454545454545457, + "grad_norm": 0.5465879440307617, + "learning_rate": 3.218444384962071e-06, + "loss": 0.4857, + "step": 416 + }, + { + "epoch": 2.8522727272727275, + "grad_norm": 0.4633638858795166, + "learning_rate": 3.2087443489992043e-06, + "loss": 0.475, + "step": 417 + }, + { + "epoch": 2.8590909090909093, + "grad_norm": 0.47139137983322144, + "learning_rate": 3.1990326968032225e-06, + "loss": 0.5162, + "step": 418 + }, + { + "epoch": 2.865909090909091, + "grad_norm": 0.45505622029304504, + "learning_rate": 3.189309587546917e-06, + "loss": 0.4699, + "step": 419 + }, + { + "epoch": 2.8727272727272726, + "grad_norm": 0.4410839080810547, + "learning_rate": 3.1795751805908578e-06, + "loss": 0.4734, + "step": 420 + }, + { + "epoch": 2.8795454545454544, + "grad_norm": 0.47172659635543823, + "learning_rate": 3.169829635480783e-06, + "loss": 0.4896, + "step": 421 + }, + { + "epoch": 2.8863636363636362, + "grad_norm": 0.4332088828086853, + "learning_rate": 3.160073111944983e-06, + "loss": 0.4898, + "step": 422 + }, + { + "epoch": 2.893181818181818, + "grad_norm": 0.4303777813911438, + "learning_rate": 3.150305769891686e-06, + "loss": 0.4663, + "step": 423 + }, + { + "epoch": 2.9, + "grad_norm": 0.44356584548950195, + "learning_rate": 3.1405277694064306e-06, + "loss": 0.4516, + "step": 424 + }, + { + "epoch": 2.9068181818181817, + "grad_norm": 0.5283893942832947, + "learning_rate": 3.13073927074945e-06, + "loss": 0.4792, + "step": 425 + }, + { + "epoch": 2.9136363636363636, + "grad_norm": 0.90226149559021, + "learning_rate": 3.1209404343530374e-06, + "loss": 0.4811, + "step": 426 + }, + { + "epoch": 2.9204545454545454, + "grad_norm": 0.4658738970756531, + "learning_rate": 3.111131420818922e-06, + "loss": 0.4601, + "step": 427 + }, + { + "epoch": 2.9272727272727272, + "grad_norm": 0.4286805987358093, + "learning_rate": 3.1013123909156347e-06, + "loss": 0.4801, + "step": 428 + }, + { + "epoch": 2.934090909090909, + "grad_norm": 0.5181618332862854, + "learning_rate": 3.091483505575873e-06, + "loss": 0.4717, + "step": 429 + }, + { + "epoch": 2.940909090909091, + "grad_norm": 0.4391111135482788, + "learning_rate": 3.081644925893866e-06, + "loss": 0.4913, + "step": 430 + }, + { + "epoch": 2.9477272727272728, + "grad_norm": 0.4414638578891754, + "learning_rate": 3.0717968131227285e-06, + "loss": 0.4796, + "step": 431 + }, + { + "epoch": 2.9545454545454546, + "grad_norm": 0.4619010090827942, + "learning_rate": 3.061939328671824e-06, + "loss": 0.5059, + "step": 432 + }, + { + "epoch": 2.9613636363636364, + "grad_norm": 0.42799466848373413, + "learning_rate": 3.0520726341041165e-06, + "loss": 0.4598, + "step": 433 + }, + { + "epoch": 2.9681818181818183, + "grad_norm": 0.43869948387145996, + "learning_rate": 3.0421968911335196e-06, + "loss": 0.4733, + "step": 434 + }, + { + "epoch": 2.975, + "grad_norm": 0.552511990070343, + "learning_rate": 3.032312261622255e-06, + "loss": 0.4855, + "step": 435 + }, + { + "epoch": 2.981818181818182, + "grad_norm": 0.5187029242515564, + "learning_rate": 3.0224189075781886e-06, + "loss": 0.5067, + "step": 436 + }, + { + "epoch": 2.9886363636363638, + "grad_norm": 0.4445323944091797, + "learning_rate": 3.012516991152181e-06, + "loss": 0.4542, + "step": 437 + }, + { + "epoch": 2.9954545454545456, + "grad_norm": 0.5056352019309998, + "learning_rate": 3.002606674635432e-06, + "loss": 0.4907, + "step": 438 + }, + { + "epoch": 3.006818181818182, + "grad_norm": 0.4359378218650818, + "learning_rate": 2.9926881204568153e-06, + "loss": 0.9747, + "step": 439 + }, + { + "epoch": 3.0136363636363637, + "grad_norm": 0.41525086760520935, + "learning_rate": 2.9827614911802205e-06, + "loss": 0.4672, + "step": 440 + }, + { + "epoch": 3.0204545454545455, + "grad_norm": 0.45512208342552185, + "learning_rate": 2.972826949501884e-06, + "loss": 0.4607, + "step": 441 + }, + { + "epoch": 3.0272727272727273, + "grad_norm": 0.4749738574028015, + "learning_rate": 2.9628846582477305e-06, + "loss": 0.4471, + "step": 442 + }, + { + "epoch": 3.034090909090909, + "grad_norm": 0.42964112758636475, + "learning_rate": 2.9529347803706943e-06, + "loss": 0.4437, + "step": 443 + }, + { + "epoch": 3.040909090909091, + "grad_norm": 0.42203956842422485, + "learning_rate": 2.9429774789480576e-06, + "loss": 0.444, + "step": 444 + }, + { + "epoch": 3.047727272727273, + "grad_norm": 0.4311574697494507, + "learning_rate": 2.9330129171787704e-06, + "loss": 0.4942, + "step": 445 + }, + { + "epoch": 3.0545454545454547, + "grad_norm": 0.4946717917919159, + "learning_rate": 2.923041258380779e-06, + "loss": 0.4505, + "step": 446 + }, + { + "epoch": 3.0613636363636365, + "grad_norm": 0.45294976234436035, + "learning_rate": 2.9130626659883537e-06, + "loss": 0.4537, + "step": 447 + }, + { + "epoch": 3.0681818181818183, + "grad_norm": 0.4274294376373291, + "learning_rate": 2.9030773035493997e-06, + "loss": 0.4532, + "step": 448 + }, + { + "epoch": 3.075, + "grad_norm": 0.4929022789001465, + "learning_rate": 2.893085334722786e-06, + "loss": 0.4403, + "step": 449 + }, + { + "epoch": 3.081818181818182, + "grad_norm": 0.4446672797203064, + "learning_rate": 2.883086923275658e-06, + "loss": 0.4572, + "step": 450 + }, + { + "epoch": 3.088636363636364, + "grad_norm": 0.45900678634643555, + "learning_rate": 2.8730822330807556e-06, + "loss": 0.4435, + "step": 451 + }, + { + "epoch": 3.0954545454545457, + "grad_norm": 0.4808853566646576, + "learning_rate": 2.8630714281137263e-06, + "loss": 0.4668, + "step": 452 + }, + { + "epoch": 3.102272727272727, + "grad_norm": 0.460186630487442, + "learning_rate": 2.853054672450437e-06, + "loss": 0.4393, + "step": 453 + }, + { + "epoch": 3.109090909090909, + "grad_norm": 0.4464828670024872, + "learning_rate": 2.8430321302642887e-06, + "loss": 0.4374, + "step": 454 + }, + { + "epoch": 3.1159090909090907, + "grad_norm": 0.4616970717906952, + "learning_rate": 2.8330039658235194e-06, + "loss": 0.4508, + "step": 455 + }, + { + "epoch": 3.1227272727272726, + "grad_norm": 0.45340001583099365, + "learning_rate": 2.8229703434885165e-06, + "loss": 0.4488, + "step": 456 + }, + { + "epoch": 3.1295454545454544, + "grad_norm": 0.4715891480445862, + "learning_rate": 2.8129314277091224e-06, + "loss": 0.4642, + "step": 457 + }, + { + "epoch": 3.1363636363636362, + "grad_norm": 0.4530370533466339, + "learning_rate": 2.8028873830219373e-06, + "loss": 0.4502, + "step": 458 + }, + { + "epoch": 3.143181818181818, + "grad_norm": 0.4896863102912903, + "learning_rate": 2.7928383740476247e-06, + "loss": 0.4559, + "step": 459 + }, + { + "epoch": 3.15, + "grad_norm": 0.45334747433662415, + "learning_rate": 2.7827845654882112e-06, + "loss": 0.4462, + "step": 460 + }, + { + "epoch": 3.1568181818181817, + "grad_norm": 0.4454493820667267, + "learning_rate": 2.7727261221243875e-06, + "loss": 0.4726, + "step": 461 + }, + { + "epoch": 3.1636363636363636, + "grad_norm": 0.46723124384880066, + "learning_rate": 2.76266320881281e-06, + "loss": 0.476, + "step": 462 + }, + { + "epoch": 3.1704545454545454, + "grad_norm": 0.45423537492752075, + "learning_rate": 2.7525959904833955e-06, + "loss": 0.4269, + "step": 463 + }, + { + "epoch": 3.1772727272727272, + "grad_norm": 0.44072285294532776, + "learning_rate": 2.7425246321366205e-06, + "loss": 0.4556, + "step": 464 + }, + { + "epoch": 3.184090909090909, + "grad_norm": 0.44829070568084717, + "learning_rate": 2.7324492988408146e-06, + "loss": 0.4479, + "step": 465 + }, + { + "epoch": 3.190909090909091, + "grad_norm": 0.47372356057167053, + "learning_rate": 2.7223701557294574e-06, + "loss": 0.4302, + "step": 466 + }, + { + "epoch": 3.1977272727272728, + "grad_norm": 0.4265802800655365, + "learning_rate": 2.712287367998471e-06, + "loss": 0.4557, + "step": 467 + }, + { + "epoch": 3.2045454545454546, + "grad_norm": 0.5714879631996155, + "learning_rate": 2.702201100903511e-06, + "loss": 0.437, + "step": 468 + }, + { + "epoch": 3.2113636363636364, + "grad_norm": 0.4389351010322571, + "learning_rate": 2.692111519757261e-06, + "loss": 0.4371, + "step": 469 + }, + { + "epoch": 3.2181818181818183, + "grad_norm": 0.44851404428482056, + "learning_rate": 2.6820187899267203e-06, + "loss": 0.4711, + "step": 470 + }, + { + "epoch": 3.225, + "grad_norm": 0.5391034483909607, + "learning_rate": 2.671923076830496e-06, + "loss": 0.4698, + "step": 471 + }, + { + "epoch": 3.231818181818182, + "grad_norm": 0.5056214332580566, + "learning_rate": 2.6618245459360896e-06, + "loss": 0.4712, + "step": 472 + }, + { + "epoch": 3.2386363636363638, + "grad_norm": 0.47290515899658203, + "learning_rate": 2.651723362757186e-06, + "loss": 0.4515, + "step": 473 + }, + { + "epoch": 3.2454545454545456, + "grad_norm": 0.4664989411830902, + "learning_rate": 2.641619692850941e-06, + "loss": 0.4602, + "step": 474 + }, + { + "epoch": 3.2522727272727274, + "grad_norm": 0.5000811219215393, + "learning_rate": 2.631513701815267e-06, + "loss": 0.4521, + "step": 475 + }, + { + "epoch": 3.2590909090909093, + "grad_norm": 0.4550754427909851, + "learning_rate": 2.6214055552861213e-06, + "loss": 0.471, + "step": 476 + }, + { + "epoch": 3.265909090909091, + "grad_norm": 0.4234600365161896, + "learning_rate": 2.611295418934786e-06, + "loss": 0.4486, + "step": 477 + }, + { + "epoch": 3.2727272727272725, + "grad_norm": 0.49546584486961365, + "learning_rate": 2.6011834584651597e-06, + "loss": 0.4367, + "step": 478 + }, + { + "epoch": 3.2795454545454543, + "grad_norm": 0.42106184363365173, + "learning_rate": 2.591069839611036e-06, + "loss": 0.4366, + "step": 479 + }, + { + "epoch": 3.286363636363636, + "grad_norm": 0.4530644118785858, + "learning_rate": 2.5809547281333904e-06, + "loss": 0.4729, + "step": 480 + }, + { + "epoch": 3.293181818181818, + "grad_norm": 0.4612523019313812, + "learning_rate": 2.570838289817661e-06, + "loss": 0.4531, + "step": 481 + }, + { + "epoch": 3.3, + "grad_norm": 0.4771486222743988, + "learning_rate": 2.560720690471033e-06, + "loss": 0.4656, + "step": 482 + }, + { + "epoch": 3.3068181818181817, + "grad_norm": 0.45801112055778503, + "learning_rate": 2.5506020959197218e-06, + "loss": 0.4443, + "step": 483 + }, + { + "epoch": 3.3136363636363635, + "grad_norm": 0.4690552353858948, + "learning_rate": 2.5404826720062544e-06, + "loss": 0.4442, + "step": 484 + }, + { + "epoch": 3.3204545454545453, + "grad_norm": 0.4829693138599396, + "learning_rate": 2.5303625845867475e-06, + "loss": 0.4658, + "step": 485 + }, + { + "epoch": 3.327272727272727, + "grad_norm": 0.4420918822288513, + "learning_rate": 2.5202419995281966e-06, + "loss": 0.4486, + "step": 486 + }, + { + "epoch": 3.334090909090909, + "grad_norm": 0.4645867347717285, + "learning_rate": 2.5101210827057516e-06, + "loss": 0.4536, + "step": 487 + }, + { + "epoch": 3.340909090909091, + "grad_norm": 0.45904067158699036, + "learning_rate": 2.5e-06, + "loss": 0.4777, + "step": 488 + }, + { + "epoch": 3.3477272727272727, + "grad_norm": 0.47885990142822266, + "learning_rate": 2.4898789172942492e-06, + "loss": 0.4507, + "step": 489 + }, + { + "epoch": 3.3545454545454545, + "grad_norm": 0.48885077238082886, + "learning_rate": 2.4797580004718038e-06, + "loss": 0.4492, + "step": 490 + }, + { + "epoch": 3.3613636363636363, + "grad_norm": 0.43438634276390076, + "learning_rate": 2.4696374154132533e-06, + "loss": 0.4622, + "step": 491 + }, + { + "epoch": 3.368181818181818, + "grad_norm": 0.6276775598526001, + "learning_rate": 2.4595173279937464e-06, + "loss": 0.4488, + "step": 492 + }, + { + "epoch": 3.375, + "grad_norm": 0.5427472591400146, + "learning_rate": 2.4493979040802786e-06, + "loss": 0.4482, + "step": 493 + }, + { + "epoch": 3.381818181818182, + "grad_norm": 0.4438450336456299, + "learning_rate": 2.4392793095289677e-06, + "loss": 0.4434, + "step": 494 + }, + { + "epoch": 3.3886363636363637, + "grad_norm": 0.49523183703422546, + "learning_rate": 2.42916171018234e-06, + "loss": 0.4645, + "step": 495 + }, + { + "epoch": 3.3954545454545455, + "grad_norm": 0.4602353274822235, + "learning_rate": 2.419045271866611e-06, + "loss": 0.4569, + "step": 496 + }, + { + "epoch": 3.4022727272727273, + "grad_norm": 0.4849902391433716, + "learning_rate": 2.408930160388965e-06, + "loss": 0.4538, + "step": 497 + }, + { + "epoch": 3.409090909090909, + "grad_norm": 0.6242562532424927, + "learning_rate": 2.3988165415348416e-06, + "loss": 0.4509, + "step": 498 + }, + { + "epoch": 3.415909090909091, + "grad_norm": 0.45550772547721863, + "learning_rate": 2.388704581065215e-06, + "loss": 0.4396, + "step": 499 + }, + { + "epoch": 3.422727272727273, + "grad_norm": 0.428011953830719, + "learning_rate": 2.3785944447138804e-06, + "loss": 0.4506, + "step": 500 + }, + { + "epoch": 3.4295454545454547, + "grad_norm": 0.47410571575164795, + "learning_rate": 2.368486298184733e-06, + "loss": 0.4573, + "step": 501 + }, + { + "epoch": 3.4363636363636365, + "grad_norm": 0.45392513275146484, + "learning_rate": 2.358380307149059e-06, + "loss": 0.4607, + "step": 502 + }, + { + "epoch": 3.4431818181818183, + "grad_norm": 0.4263063669204712, + "learning_rate": 2.348276637242814e-06, + "loss": 0.4574, + "step": 503 + }, + { + "epoch": 3.45, + "grad_norm": 0.5994442701339722, + "learning_rate": 2.3381754540639108e-06, + "loss": 0.4483, + "step": 504 + }, + { + "epoch": 3.456818181818182, + "grad_norm": 0.45459091663360596, + "learning_rate": 2.328076923169504e-06, + "loss": 0.4445, + "step": 505 + }, + { + "epoch": 3.463636363636364, + "grad_norm": 0.48127374053001404, + "learning_rate": 2.31798121007328e-06, + "loss": 0.4455, + "step": 506 + }, + { + "epoch": 3.4704545454545457, + "grad_norm": 0.7549276947975159, + "learning_rate": 2.3078884802427394e-06, + "loss": 0.4475, + "step": 507 + }, + { + "epoch": 3.4772727272727275, + "grad_norm": 0.4759480953216553, + "learning_rate": 2.29779889909649e-06, + "loss": 0.4512, + "step": 508 + }, + { + "epoch": 3.484090909090909, + "grad_norm": 0.45603302121162415, + "learning_rate": 2.2877126320015295e-06, + "loss": 0.4441, + "step": 509 + }, + { + "epoch": 3.4909090909090907, + "grad_norm": 0.47070786356925964, + "learning_rate": 2.2776298442705434e-06, + "loss": 0.4402, + "step": 510 + }, + { + "epoch": 3.4977272727272726, + "grad_norm": 0.4690212905406952, + "learning_rate": 2.267550701159186e-06, + "loss": 0.4517, + "step": 511 + }, + { + "epoch": 3.5045454545454544, + "grad_norm": 0.8859854340553284, + "learning_rate": 2.25747536786338e-06, + "loss": 0.4374, + "step": 512 + }, + { + "epoch": 3.5113636363636362, + "grad_norm": 0.4920756220817566, + "learning_rate": 2.247404009516605e-06, + "loss": 0.446, + "step": 513 + }, + { + "epoch": 3.518181818181818, + "grad_norm": 0.49091637134552, + "learning_rate": 2.2373367911871904e-06, + "loss": 0.438, + "step": 514 + }, + { + "epoch": 3.525, + "grad_norm": 0.8683479428291321, + "learning_rate": 2.227273877875613e-06, + "loss": 0.4435, + "step": 515 + }, + { + "epoch": 3.5318181818181817, + "grad_norm": 0.4606627821922302, + "learning_rate": 2.2172154345117896e-06, + "loss": 0.4517, + "step": 516 + }, + { + "epoch": 3.5386363636363636, + "grad_norm": 0.48613184690475464, + "learning_rate": 2.207161625952376e-06, + "loss": 0.4449, + "step": 517 + }, + { + "epoch": 3.5454545454545454, + "grad_norm": 0.5018280148506165, + "learning_rate": 2.1971126169780636e-06, + "loss": 0.4414, + "step": 518 + }, + { + "epoch": 3.5522727272727272, + "grad_norm": 0.435921847820282, + "learning_rate": 2.1870685722908784e-06, + "loss": 0.4542, + "step": 519 + }, + { + "epoch": 3.559090909090909, + "grad_norm": 0.48534268140792847, + "learning_rate": 2.1770296565114847e-06, + "loss": 0.4373, + "step": 520 + }, + { + "epoch": 3.565909090909091, + "grad_norm": 0.49759384989738464, + "learning_rate": 2.166996034176482e-06, + "loss": 0.4518, + "step": 521 + }, + { + "epoch": 3.5727272727272728, + "grad_norm": 0.524696409702301, + "learning_rate": 2.1569678697357126e-06, + "loss": 0.4668, + "step": 522 + }, + { + "epoch": 3.5795454545454546, + "grad_norm": 0.4403073191642761, + "learning_rate": 2.1469453275495634e-06, + "loss": 0.44, + "step": 523 + }, + { + "epoch": 3.5863636363636364, + "grad_norm": 0.5214805006980896, + "learning_rate": 2.136928571886275e-06, + "loss": 0.437, + "step": 524 + }, + { + "epoch": 3.5931818181818183, + "grad_norm": 0.4685129225254059, + "learning_rate": 2.126917766919245e-06, + "loss": 0.4477, + "step": 525 + }, + { + "epoch": 3.6, + "grad_norm": 0.500602126121521, + "learning_rate": 2.1169130767243424e-06, + "loss": 0.455, + "step": 526 + }, + { + "epoch": 3.606818181818182, + "grad_norm": 0.5242429375648499, + "learning_rate": 2.1069146652772142e-06, + "loss": 0.4398, + "step": 527 + }, + { + "epoch": 3.6136363636363638, + "grad_norm": 0.6797722578048706, + "learning_rate": 2.0969226964506007e-06, + "loss": 0.4393, + "step": 528 + }, + { + "epoch": 3.6204545454545456, + "grad_norm": 0.44578859210014343, + "learning_rate": 2.0869373340116467e-06, + "loss": 0.458, + "step": 529 + }, + { + "epoch": 3.6272727272727274, + "grad_norm": 0.46465393900871277, + "learning_rate": 2.0769587416192212e-06, + "loss": 0.4461, + "step": 530 + }, + { + "epoch": 3.634090909090909, + "grad_norm": 0.48858389258384705, + "learning_rate": 2.066987082821231e-06, + "loss": 0.4623, + "step": 531 + }, + { + "epoch": 3.6409090909090907, + "grad_norm": 0.4606386423110962, + "learning_rate": 2.0570225210519433e-06, + "loss": 0.4538, + "step": 532 + }, + { + "epoch": 3.6477272727272725, + "grad_norm": 0.46266230940818787, + "learning_rate": 2.047065219629306e-06, + "loss": 0.4635, + "step": 533 + }, + { + "epoch": 3.6545454545454543, + "grad_norm": 0.5587124824523926, + "learning_rate": 2.0371153417522703e-06, + "loss": 0.4614, + "step": 534 + }, + { + "epoch": 3.661363636363636, + "grad_norm": 0.6143503785133362, + "learning_rate": 2.0271730504981165e-06, + "loss": 0.4408, + "step": 535 + }, + { + "epoch": 3.668181818181818, + "grad_norm": 0.5167360305786133, + "learning_rate": 2.0172385088197804e-06, + "loss": 0.4539, + "step": 536 + }, + { + "epoch": 3.675, + "grad_norm": 0.47937509417533875, + "learning_rate": 2.007311879543185e-06, + "loss": 0.4587, + "step": 537 + }, + { + "epoch": 3.6818181818181817, + "grad_norm": 0.4938436448574066, + "learning_rate": 1.9973933253645684e-06, + "loss": 0.4448, + "step": 538 + }, + { + "epoch": 3.6886363636363635, + "grad_norm": 0.4234243631362915, + "learning_rate": 1.9874830088478196e-06, + "loss": 0.4584, + "step": 539 + }, + { + "epoch": 3.6954545454545453, + "grad_norm": 0.5978733897209167, + "learning_rate": 1.9775810924218126e-06, + "loss": 0.4647, + "step": 540 + }, + { + "epoch": 3.702272727272727, + "grad_norm": 0.49155011773109436, + "learning_rate": 1.967687738377746e-06, + "loss": 0.4496, + "step": 541 + }, + { + "epoch": 3.709090909090909, + "grad_norm": 0.4360124468803406, + "learning_rate": 1.9578031088664812e-06, + "loss": 0.466, + "step": 542 + }, + { + "epoch": 3.715909090909091, + "grad_norm": 0.4867744743824005, + "learning_rate": 1.9479273658958852e-06, + "loss": 0.4362, + "step": 543 + }, + { + "epoch": 3.7227272727272727, + "grad_norm": 0.49178293347358704, + "learning_rate": 1.9380606713281773e-06, + "loss": 0.4501, + "step": 544 + }, + { + "epoch": 3.7295454545454545, + "grad_norm": 0.5098608136177063, + "learning_rate": 1.928203186877273e-06, + "loss": 0.4655, + "step": 545 + }, + { + "epoch": 3.7363636363636363, + "grad_norm": 0.4781307578086853, + "learning_rate": 1.9183550741061354e-06, + "loss": 0.4464, + "step": 546 + }, + { + "epoch": 3.743181818181818, + "grad_norm": 0.49883827567100525, + "learning_rate": 1.9085164944241275e-06, + "loss": 0.4366, + "step": 547 + }, + { + "epoch": 3.75, + "grad_norm": 0.5124484300613403, + "learning_rate": 1.8986876090843668e-06, + "loss": 0.4625, + "step": 548 + }, + { + "epoch": 3.756818181818182, + "grad_norm": 0.46270111203193665, + "learning_rate": 1.8888685791810784e-06, + "loss": 0.4149, + "step": 549 + }, + { + "epoch": 3.7636363636363637, + "grad_norm": 0.5264699459075928, + "learning_rate": 1.8790595656469628e-06, + "loss": 0.4373, + "step": 550 + }, + { + "epoch": 3.7704545454545455, + "grad_norm": 0.46891146898269653, + "learning_rate": 1.86926072925055e-06, + "loss": 0.4437, + "step": 551 + }, + { + "epoch": 3.7772727272727273, + "grad_norm": 0.4786033630371094, + "learning_rate": 1.8594722305935691e-06, + "loss": 0.4364, + "step": 552 + }, + { + "epoch": 3.784090909090909, + "grad_norm": 0.45475709438323975, + "learning_rate": 1.8496942301083142e-06, + "loss": 0.4515, + "step": 553 + }, + { + "epoch": 3.790909090909091, + "grad_norm": 0.4658668637275696, + "learning_rate": 1.8399268880550174e-06, + "loss": 0.4553, + "step": 554 + }, + { + "epoch": 3.797727272727273, + "grad_norm": 0.5587930679321289, + "learning_rate": 1.8301703645192178e-06, + "loss": 0.4412, + "step": 555 + }, + { + "epoch": 3.8045454545454547, + "grad_norm": 0.4532659947872162, + "learning_rate": 1.8204248194091429e-06, + "loss": 0.4529, + "step": 556 + }, + { + "epoch": 3.8113636363636365, + "grad_norm": 0.4606110751628876, + "learning_rate": 1.8106904124530839e-06, + "loss": 0.437, + "step": 557 + }, + { + "epoch": 3.8181818181818183, + "grad_norm": 0.4482369124889374, + "learning_rate": 1.800967303196778e-06, + "loss": 0.4309, + "step": 558 + }, + { + "epoch": 3.825, + "grad_norm": 0.533805787563324, + "learning_rate": 1.7912556510007967e-06, + "loss": 0.4393, + "step": 559 + }, + { + "epoch": 3.831818181818182, + "grad_norm": 0.5079528093338013, + "learning_rate": 1.7815556150379298e-06, + "loss": 0.4585, + "step": 560 + }, + { + "epoch": 3.838636363636364, + "grad_norm": 0.5496546626091003, + "learning_rate": 1.77186735429058e-06, + "loss": 0.4293, + "step": 561 + }, + { + "epoch": 3.8454545454545457, + "grad_norm": 0.49827462434768677, + "learning_rate": 1.7621910275481544e-06, + "loss": 0.4457, + "step": 562 + }, + { + "epoch": 3.8522727272727275, + "grad_norm": 0.5003693699836731, + "learning_rate": 1.7525267934044642e-06, + "loss": 0.4579, + "step": 563 + }, + { + "epoch": 3.8590909090909093, + "grad_norm": 0.47428634762763977, + "learning_rate": 1.7428748102551237e-06, + "loss": 0.452, + "step": 564 + }, + { + "epoch": 3.865909090909091, + "grad_norm": 0.4468045234680176, + "learning_rate": 1.7332352362949546e-06, + "loss": 0.4558, + "step": 565 + }, + { + "epoch": 3.8727272727272726, + "grad_norm": 0.4439336955547333, + "learning_rate": 1.7236082295153948e-06, + "loss": 0.4351, + "step": 566 + }, + { + "epoch": 3.8795454545454544, + "grad_norm": 0.46482470631599426, + "learning_rate": 1.7139939477019057e-06, + "loss": 0.4462, + "step": 567 + }, + { + "epoch": 3.8863636363636362, + "grad_norm": 1.1179648637771606, + "learning_rate": 1.7043925484313911e-06, + "loss": 0.4435, + "step": 568 + }, + { + "epoch": 3.893181818181818, + "grad_norm": 0.46164315938949585, + "learning_rate": 1.6948041890696076e-06, + "loss": 0.4446, + "step": 569 + }, + { + "epoch": 3.9, + "grad_norm": 0.4524129629135132, + "learning_rate": 1.685229026768593e-06, + "loss": 0.4567, + "step": 570 + }, + { + "epoch": 3.9068181818181817, + "grad_norm": 0.45722272992134094, + "learning_rate": 1.6756672184640847e-06, + "loss": 0.4368, + "step": 571 + }, + { + "epoch": 3.9136363636363636, + "grad_norm": 0.49532264471054077, + "learning_rate": 1.6661189208729492e-06, + "loss": 0.4572, + "step": 572 + }, + { + "epoch": 3.9204545454545454, + "grad_norm": 0.44752320647239685, + "learning_rate": 1.6565842904906154e-06, + "loss": 0.43, + "step": 573 + }, + { + "epoch": 3.9272727272727272, + "grad_norm": 0.5931842923164368, + "learning_rate": 1.6470634835885097e-06, + "loss": 0.4334, + "step": 574 + }, + { + "epoch": 3.934090909090909, + "grad_norm": 0.5873560905456543, + "learning_rate": 1.6375566562114903e-06, + "loss": 0.4374, + "step": 575 + }, + { + "epoch": 3.940909090909091, + "grad_norm": 0.4484315812587738, + "learning_rate": 1.6280639641752944e-06, + "loss": 0.4402, + "step": 576 + }, + { + "epoch": 3.9477272727272728, + "grad_norm": 0.4635726809501648, + "learning_rate": 1.6185855630639818e-06, + "loss": 0.4485, + "step": 577 + }, + { + "epoch": 3.9545454545454546, + "grad_norm": 0.7416269779205322, + "learning_rate": 1.6091216082273875e-06, + "loss": 0.4655, + "step": 578 + }, + { + "epoch": 3.9613636363636364, + "grad_norm": 0.4709416329860687, + "learning_rate": 1.5996722547785722e-06, + "loss": 0.4551, + "step": 579 + }, + { + "epoch": 3.9681818181818183, + "grad_norm": 0.45414504408836365, + "learning_rate": 1.5902376575912815e-06, + "loss": 0.4478, + "step": 580 + }, + { + "epoch": 3.975, + "grad_norm": 0.45831212401390076, + "learning_rate": 1.580817971297409e-06, + "loss": 0.462, + "step": 581 + }, + { + "epoch": 3.981818181818182, + "grad_norm": 0.4817596971988678, + "learning_rate": 1.5714133502844591e-06, + "loss": 0.4554, + "step": 582 + }, + { + "epoch": 3.9886363636363638, + "grad_norm": 0.48830825090408325, + "learning_rate": 1.56202394869302e-06, + "loss": 0.4382, + "step": 583 + }, + { + "epoch": 3.9954545454545456, + "grad_norm": 0.4884580075740814, + "learning_rate": 1.5526499204142332e-06, + "loss": 0.4527, + "step": 584 + } + ], + "logging_steps": 1, + "max_steps": 876, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 146, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 3.763412212623658e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-584/training_args.bin b/checkpoint-584/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..b41c805059a36d8d2e3bf3511fb0fd95b74dae74 --- /dev/null +++ b/checkpoint-584/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68945b8f9934f5870e67a2d4a20c31fa8e742da41075fa7236fb313c6197456f +size 7928 diff --git a/checkpoint-584/zero_to_fp32.py b/checkpoint-584/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-584/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-730/README.md b/checkpoint-730/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1b184114a0c28ed3e4c082c18486736dc818166d --- /dev/null +++ b/checkpoint-730/README.md @@ -0,0 +1,202 @@ +--- +base_model: meta-llama/Llama-3.3-70B-Instruct +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 0.15.0 \ No newline at end of file diff --git a/checkpoint-730/adapter_config.json b/checkpoint-730/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a5775df151cb9649596940e8a117cb92f810dbb7 --- /dev/null +++ b/checkpoint-730/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.3-70B-Instruct", + "bias": "none", + "corda_config": null, + "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", + "down_proj", + "k_proj", + "q_proj", + "o_proj", + "v_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-730/adapter_model.safetensors b/checkpoint-730/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c5546fe334955062d14b08793118cc0dba2ca4eb --- /dev/null +++ b/checkpoint-730/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b682b3cdd236a43cd84b908b2783a3f781c2600cd0736c94064bf9ec829367f5 +size 10829849744 diff --git a/checkpoint-730/global_step731/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-730/global_step731/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..bac8cbf71378c8ad6dfe138659c20b483925338d --- /dev/null +++ b/checkpoint-730/global_step731/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:3849a45d2c501669f263d1f047c346f653489b2da16054e323a682c3f985be08 +size 21659418140 diff --git a/checkpoint-730/global_step731/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-730/global_step731/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..a1b49680371da96d81638510940e820f094d98ab --- /dev/null +++ b/checkpoint-730/global_step731/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:efcadf6b426c20a0c39d7267334df731338c69e15258f94edb7b73087409640f +size 21659457372 diff --git a/checkpoint-730/global_step731/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoint-730/global_step731/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..d820dc04a46985c3ef369ffa5e8930cdd1d28c82 --- /dev/null +++ b/checkpoint-730/global_step731/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c6e0d3ad35888617dd43e875a8e94e09a551c35cc2271c9f8e6d512ea28aa6c +size 21659417820 diff --git a/checkpoint-730/global_step731/mp_rank_00_model_states.pt b/checkpoint-730/global_step731/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..e729334fb454c4afde5d866030fb25e3aa2be4ef --- /dev/null +++ b/checkpoint-730/global_step731/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:deb33b07357fd64de762b997a62c18131a3db77f923991af6976c45b597788fa +size 11918643933 diff --git a/checkpoint-730/latest b/checkpoint-730/latest new file mode 100644 index 0000000000000000000000000000000000000000..171216d66aba7b132ebd0851de2c93970e71b355 --- /dev/null +++ b/checkpoint-730/latest @@ -0,0 +1 @@ +global_step731 \ No newline at end of file diff --git a/checkpoint-730/rng_state_0.pth b/checkpoint-730/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..e62d4a6555d773ddf4cf49653fc121066c0265f4 --- /dev/null +++ b/checkpoint-730/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:eed5221112b22a5bf0549bb3d98ee0f29f662cff6ebdc1f5519215686d873ab8 +size 14768 diff --git a/checkpoint-730/rng_state_1.pth b/checkpoint-730/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..b95665b5b03bcb20f79335f9f4c9e33146eb2a26 --- /dev/null +++ b/checkpoint-730/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:4960339ed54c923c85ef56210d4a796868ebc576e8d2d7c5f087a412a7100121 +size 14768 diff --git a/checkpoint-730/rng_state_2.pth b/checkpoint-730/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..b24ff4acc9ecae5ab2a38027104498af1a87bc75 --- /dev/null +++ b/checkpoint-730/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f9fccfbfb163c3cee177d2e0d65128248f41479e37a33c0b5de7a224155cee9a +size 14768 diff --git a/checkpoint-730/scheduler.pt b/checkpoint-730/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..ed3c5aa5fda44f0e5730e822b947c090b8a3ec1f --- /dev/null +++ b/checkpoint-730/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:e5eff5ea5b6d63a833ba68fecbde9efd05e3945f157a54fbcfd3c537675e3504 +size 1064 diff --git a/checkpoint-730/special_tokens_map.json b/checkpoint-730/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-730/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": "<|eot_id|>", + "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-730/tokenizer.json b/checkpoint-730/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-730/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-730/tokenizer_config.json b/checkpoint-730/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ca91a2ef55f4239a7af81d7c9abb05f53621a07b --- /dev/null +++ b/checkpoint-730/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "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|>", + "chat_template": "{{- bos_token }}\n{%- if custom_tools is defined %}\n {%- set tools = custom_tools %}\n{%- endif %}\n{%- if not tools_in_user_message is defined %}\n {%- set tools_in_user_message = true %}\n{%- endif %}\n{%- if not date_string is defined %}\n {%- set date_string = \"26 Jul 2024\" %}\n{%- endif %}\n{%- if not tools is defined %}\n {%- set tools = none %}\n{%- endif %}\n\n{#- This block extracts the system message, so we can slot it into the right place. #}\n{%- if messages[0]['role'] == 'system' %}\n {%- set system_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n{%- else %}\n {%- set system_message = \"\" %}\n{%- endif %}\n\n{#- System message + builtin tools #}\n{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}\n{%- if builtin_tools is defined or tools is not none %}\n {{- \"Environment: ipython\\n\" }}\n{%- endif %}\n{%- if builtin_tools is defined %}\n {{- \"Tools: \" + builtin_tools | reject('equalto', 'code_interpreter') | join(\", \") + \"\\n\\n\"}}\n{%- endif %}\n{{- \"Cutting Knowledge Date: December 2023\\n\" }}\n{{- \"Today Date: \" + date_string + \"\\n\\n\" }}\n{%- if tools is not none and not tools_in_user_message %}\n {{- \"You have access to the following functions. To call a function, please respond with JSON for a function call.\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n{%- endif %}\n{{- system_message }}\n{{- \"<|eot_id|>\" }}\n\n{#- Custom tools are passed in a user message with some extra guidance #}\n{%- if tools_in_user_message and not tools is none %}\n {#- Extract the first user message so we can plug it in here #}\n {%- if messages | length != 0 %}\n {%- set first_user_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n {%- else %}\n {{- raise_exception(\"Cannot put tools in the first user message when there's no first user message!\") }}\n{%- endif %}\n {{- '<|start_header_id|>user<|end_header_id|>\\n\\n' -}}\n {{- \"Given the following functions, please respond with a JSON for a function call \" }}\n {{- \"with its proper arguments that best answers the given prompt.\\n\\n\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n {{- first_user_message + \"<|eot_id|>\"}}\n{%- endif %}\n\n{%- for message in messages %}\n {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}\n {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n'+ message['content'] | trim + '<|eot_id|>' }}\n {%- elif 'tool_calls' in message %}\n {%- if not message.tool_calls|length == 1 %}\n {{- raise_exception(\"This model only supports single tool-calls at once!\") }}\n {%- endif %}\n {%- set tool_call = message.tool_calls[0].function %}\n {%- if builtin_tools is defined and tool_call.name in builtin_tools %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- \"<|python_tag|>\" + tool_call.name + \".call(\" }}\n {%- for arg_name, arg_val in tool_call.arguments | items %}\n {{- arg_name + '=\"' + arg_val + '\"' }}\n {%- if not loop.last %}\n {{- \", \" }}\n {%- endif %}\n {%- endfor %}\n {{- \")\" }}\n {%- else %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- '{\"name\": \"' + tool_call.name + '\", ' }}\n {{- '\"parameters\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- \"}\" }}\n {%- endif %}\n {%- if builtin_tools is defined %}\n {#- This means we're in ipython mode #}\n {{- \"<|eom_id|>\" }}\n {%- else %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n {%- elif message.role == \"tool\" or message.role == \"ipython\" %}\n {{- \"<|start_header_id|>ipython<|end_header_id|>\\n\\n\" }}\n {%- if message.content is mapping or message.content is iterable %}\n {{- message.content | tojson }}\n {%- else %}\n {{- message.content }}\n {%- endif %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}\n{%- endif %}\n", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "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-730/trainer_state.json b/checkpoint-730/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..39b83b47b41d1e5d660cfdb5577caf05f1f70190 --- /dev/null +++ b/checkpoint-730/trainer_state.json @@ -0,0 +1,5143 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 4.995454545454545, + "eval_steps": 500, + "global_step": 730, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.006818181818181818, + "grad_norm": 1.8943578004837036, + "learning_rate": 5.0000000000000004e-08, + "loss": 0.8676, + "step": 1 + }, + { + "epoch": 0.013636363636363636, + "grad_norm": 7.775813102722168, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.8513, + "step": 2 + }, + { + "epoch": 0.020454545454545454, + "grad_norm": 1.768803596496582, + "learning_rate": 1.5000000000000002e-07, + "loss": 0.8339, + "step": 3 + }, + { + "epoch": 0.02727272727272727, + "grad_norm": 1.7032263278961182, + "learning_rate": 2.0000000000000002e-07, + "loss": 0.8348, + "step": 4 + }, + { + "epoch": 0.03409090909090909, + "grad_norm": 1.8244796991348267, + "learning_rate": 2.5000000000000004e-07, + "loss": 0.8687, + "step": 5 + }, + { + "epoch": 0.04090909090909091, + "grad_norm": 1.7780181169509888, + "learning_rate": 3.0000000000000004e-07, + "loss": 0.8113, + "step": 6 + }, + { + "epoch": 0.04772727272727273, + "grad_norm": 1.8362191915512085, + "learning_rate": 3.5000000000000004e-07, + "loss": 0.8463, + "step": 7 + }, + { + "epoch": 0.05454545454545454, + "grad_norm": 1.6640979051589966, + "learning_rate": 4.0000000000000003e-07, + "loss": 0.8336, + "step": 8 + }, + { + "epoch": 0.06136363636363636, + "grad_norm": 1.8013464212417603, + "learning_rate": 4.5000000000000003e-07, + "loss": 0.8572, + "step": 9 + }, + { + "epoch": 0.06818181818181818, + "grad_norm": 1.7237999439239502, + "learning_rate": 5.000000000000001e-07, + "loss": 0.8501, + "step": 10 + }, + { + "epoch": 0.075, + "grad_norm": 1.989346981048584, + "learning_rate": 5.5e-07, + "loss": 0.8615, + "step": 11 + }, + { + "epoch": 0.08181818181818182, + "grad_norm": 2.0312082767486572, + "learning_rate": 6.000000000000001e-07, + "loss": 0.858, + "step": 12 + }, + { + "epoch": 0.08863636363636364, + "grad_norm": 1.6921299695968628, + "learning_rate": 6.5e-07, + "loss": 0.8577, + "step": 13 + }, + { + "epoch": 0.09545454545454546, + "grad_norm": 2.4207160472869873, + "learning_rate": 7.000000000000001e-07, + "loss": 0.8564, + "step": 14 + }, + { + "epoch": 0.10227272727272728, + "grad_norm": 1.7572585344314575, + "learning_rate": 7.5e-07, + "loss": 0.8409, + "step": 15 + }, + { + "epoch": 0.10909090909090909, + "grad_norm": 1.6135950088500977, + "learning_rate": 8.000000000000001e-07, + "loss": 0.8086, + "step": 16 + }, + { + "epoch": 0.1159090909090909, + "grad_norm": 2.226036787033081, + "learning_rate": 8.500000000000001e-07, + "loss": 0.8106, + "step": 17 + }, + { + "epoch": 0.12272727272727273, + "grad_norm": 1.7164231538772583, + "learning_rate": 9.000000000000001e-07, + "loss": 0.8494, + "step": 18 + }, + { + "epoch": 0.12954545454545455, + "grad_norm": 1.6572000980377197, + "learning_rate": 9.500000000000001e-07, + "loss": 0.8023, + "step": 19 + }, + { + "epoch": 0.13636363636363635, + "grad_norm": 1.5672118663787842, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.787, + "step": 20 + }, + { + "epoch": 0.1431818181818182, + "grad_norm": 1.5080257654190063, + "learning_rate": 1.0500000000000001e-06, + "loss": 0.8105, + "step": 21 + }, + { + "epoch": 0.15, + "grad_norm": 1.484892725944519, + "learning_rate": 1.1e-06, + "loss": 0.8258, + "step": 22 + }, + { + "epoch": 0.15681818181818183, + "grad_norm": 1.4591134786605835, + "learning_rate": 1.1500000000000002e-06, + "loss": 0.8032, + "step": 23 + }, + { + "epoch": 0.16363636363636364, + "grad_norm": 1.5506278276443481, + "learning_rate": 1.2000000000000002e-06, + "loss": 0.8236, + "step": 24 + }, + { + "epoch": 0.17045454545454544, + "grad_norm": 1.639350414276123, + "learning_rate": 1.25e-06, + "loss": 0.8185, + "step": 25 + }, + { + "epoch": 0.17727272727272728, + "grad_norm": 1.592640995979309, + "learning_rate": 1.3e-06, + "loss": 0.7898, + "step": 26 + }, + { + "epoch": 0.18409090909090908, + "grad_norm": 1.5280641317367554, + "learning_rate": 1.3500000000000002e-06, + "loss": 0.7731, + "step": 27 + }, + { + "epoch": 0.19090909090909092, + "grad_norm": 1.4015443325042725, + "learning_rate": 1.4000000000000001e-06, + "loss": 0.7956, + "step": 28 + }, + { + "epoch": 0.19772727272727272, + "grad_norm": 1.4890056848526, + "learning_rate": 1.45e-06, + "loss": 0.7838, + "step": 29 + }, + { + "epoch": 0.20454545454545456, + "grad_norm": 2.4770750999450684, + "learning_rate": 1.5e-06, + "loss": 0.7766, + "step": 30 + }, + { + "epoch": 0.21136363636363636, + "grad_norm": 1.5863685607910156, + "learning_rate": 1.5500000000000002e-06, + "loss": 0.7353, + "step": 31 + }, + { + "epoch": 0.21818181818181817, + "grad_norm": 1.058114767074585, + "learning_rate": 1.6000000000000001e-06, + "loss": 0.7237, + "step": 32 + }, + { + "epoch": 0.225, + "grad_norm": 1.5869101285934448, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.7597, + "step": 33 + }, + { + "epoch": 0.2318181818181818, + "grad_norm": 1.0012413263320923, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.7388, + "step": 34 + }, + { + "epoch": 0.23863636363636365, + "grad_norm": 1.118056058883667, + "learning_rate": 1.75e-06, + "loss": 0.7786, + "step": 35 + }, + { + "epoch": 0.24545454545454545, + "grad_norm": 1.0648820400238037, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.7502, + "step": 36 + }, + { + "epoch": 0.25227272727272726, + "grad_norm": 0.9440382719039917, + "learning_rate": 1.85e-06, + "loss": 0.6942, + "step": 37 + }, + { + "epoch": 0.2590909090909091, + "grad_norm": 0.8300187587738037, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.7312, + "step": 38 + }, + { + "epoch": 0.26590909090909093, + "grad_norm": 0.8296191692352295, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.7461, + "step": 39 + }, + { + "epoch": 0.2727272727272727, + "grad_norm": 0.7797508239746094, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.7234, + "step": 40 + }, + { + "epoch": 0.27954545454545454, + "grad_norm": 0.7335019707679749, + "learning_rate": 2.05e-06, + "loss": 0.6927, + "step": 41 + }, + { + "epoch": 0.2863636363636364, + "grad_norm": 0.7062106728553772, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.6552, + "step": 42 + }, + { + "epoch": 0.29318181818181815, + "grad_norm": 1.2342087030410767, + "learning_rate": 2.15e-06, + "loss": 0.6959, + "step": 43 + }, + { + "epoch": 0.3, + "grad_norm": 0.8967597484588623, + "learning_rate": 2.2e-06, + "loss": 0.704, + "step": 44 + }, + { + "epoch": 0.3068181818181818, + "grad_norm": 0.7228404879570007, + "learning_rate": 2.25e-06, + "loss": 0.6777, + "step": 45 + }, + { + "epoch": 0.31363636363636366, + "grad_norm": 0.6731488704681396, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.6652, + "step": 46 + }, + { + "epoch": 0.32045454545454544, + "grad_norm": 0.71966952085495, + "learning_rate": 2.35e-06, + "loss": 0.7281, + "step": 47 + }, + { + "epoch": 0.32727272727272727, + "grad_norm": 0.7057356238365173, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.6476, + "step": 48 + }, + { + "epoch": 0.3340909090909091, + "grad_norm": 0.6346054077148438, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.6865, + "step": 49 + }, + { + "epoch": 0.3409090909090909, + "grad_norm": 0.6938223838806152, + "learning_rate": 2.5e-06, + "loss": 0.6808, + "step": 50 + }, + { + "epoch": 0.3477272727272727, + "grad_norm": 0.7425184845924377, + "learning_rate": 2.55e-06, + "loss": 0.7074, + "step": 51 + }, + { + "epoch": 0.35454545454545455, + "grad_norm": 0.6743818521499634, + "learning_rate": 2.6e-06, + "loss": 0.657, + "step": 52 + }, + { + "epoch": 0.3613636363636364, + "grad_norm": 0.6785942316055298, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.6768, + "step": 53 + }, + { + "epoch": 0.36818181818181817, + "grad_norm": 0.586974024772644, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.651, + "step": 54 + }, + { + "epoch": 0.375, + "grad_norm": 0.5727331042289734, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.669, + "step": 55 + }, + { + "epoch": 0.38181818181818183, + "grad_norm": 0.7333543300628662, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.6571, + "step": 56 + }, + { + "epoch": 0.3886363636363636, + "grad_norm": 0.5710961818695068, + "learning_rate": 2.85e-06, + "loss": 0.6475, + "step": 57 + }, + { + "epoch": 0.39545454545454545, + "grad_norm": 0.5614489912986755, + "learning_rate": 2.9e-06, + "loss": 0.6314, + "step": 58 + }, + { + "epoch": 0.4022727272727273, + "grad_norm": 0.5007341504096985, + "learning_rate": 2.95e-06, + "loss": 0.6346, + "step": 59 + }, + { + "epoch": 0.4090909090909091, + "grad_norm": 0.54221510887146, + "learning_rate": 3e-06, + "loss": 0.6715, + "step": 60 + }, + { + "epoch": 0.4159090909090909, + "grad_norm": 0.5177039504051208, + "learning_rate": 3.05e-06, + "loss": 0.6459, + "step": 61 + }, + { + "epoch": 0.42272727272727273, + "grad_norm": 0.4741189181804657, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.592, + "step": 62 + }, + { + "epoch": 0.42954545454545456, + "grad_norm": 0.5637328028678894, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.6459, + "step": 63 + }, + { + "epoch": 0.43636363636363634, + "grad_norm": 0.5478869676589966, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.6447, + "step": 64 + }, + { + "epoch": 0.4431818181818182, + "grad_norm": 0.5483130216598511, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.6319, + "step": 65 + }, + { + "epoch": 0.45, + "grad_norm": 0.478081613779068, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.6102, + "step": 66 + }, + { + "epoch": 0.45681818181818185, + "grad_norm": 0.4777645766735077, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.5986, + "step": 67 + }, + { + "epoch": 0.4636363636363636, + "grad_norm": 0.5291482210159302, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.6314, + "step": 68 + }, + { + "epoch": 0.47045454545454546, + "grad_norm": 0.5224051475524902, + "learning_rate": 3.45e-06, + "loss": 0.6278, + "step": 69 + }, + { + "epoch": 0.4772727272727273, + "grad_norm": 0.6474127173423767, + "learning_rate": 3.5e-06, + "loss": 0.6398, + "step": 70 + }, + { + "epoch": 0.48409090909090907, + "grad_norm": 0.5060178637504578, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.5936, + "step": 71 + }, + { + "epoch": 0.4909090909090909, + "grad_norm": 3.0769243240356445, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.5975, + "step": 72 + }, + { + "epoch": 0.49772727272727274, + "grad_norm": 0.5319749116897583, + "learning_rate": 3.65e-06, + "loss": 0.6119, + "step": 73 + }, + { + "epoch": 0.5045454545454545, + "grad_norm": 0.6745399236679077, + "learning_rate": 3.7e-06, + "loss": 0.6213, + "step": 74 + }, + { + "epoch": 0.5113636363636364, + "grad_norm": 0.5156731009483337, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.6262, + "step": 75 + }, + { + "epoch": 0.5181818181818182, + "grad_norm": 0.9138725399971008, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.6244, + "step": 76 + }, + { + "epoch": 0.525, + "grad_norm": 0.4362037777900696, + "learning_rate": 3.85e-06, + "loss": 0.5874, + "step": 77 + }, + { + "epoch": 0.5318181818181819, + "grad_norm": 0.4674088954925537, + "learning_rate": 3.900000000000001e-06, + "loss": 0.615, + "step": 78 + }, + { + "epoch": 0.5386363636363637, + "grad_norm": 0.4590819180011749, + "learning_rate": 3.95e-06, + "loss": 0.6003, + "step": 79 + }, + { + "epoch": 0.5454545454545454, + "grad_norm": 0.5145031809806824, + "learning_rate": 4.000000000000001e-06, + "loss": 0.6079, + "step": 80 + }, + { + "epoch": 0.5522727272727272, + "grad_norm": 0.46660348773002625, + "learning_rate": 4.05e-06, + "loss": 0.6174, + "step": 81 + }, + { + "epoch": 0.5590909090909091, + "grad_norm": 0.48053789138793945, + "learning_rate": 4.1e-06, + "loss": 0.6232, + "step": 82 + }, + { + "epoch": 0.5659090909090909, + "grad_norm": 0.540793776512146, + "learning_rate": 4.15e-06, + "loss": 0.619, + "step": 83 + }, + { + "epoch": 0.5727272727272728, + "grad_norm": 0.4925222396850586, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.5838, + "step": 84 + }, + { + "epoch": 0.5795454545454546, + "grad_norm": 0.44989290833473206, + "learning_rate": 4.25e-06, + "loss": 0.5972, + "step": 85 + }, + { + "epoch": 0.5863636363636363, + "grad_norm": 0.42691388726234436, + "learning_rate": 4.3e-06, + "loss": 0.6046, + "step": 86 + }, + { + "epoch": 0.5931818181818181, + "grad_norm": 0.4512398838996887, + "learning_rate": 4.350000000000001e-06, + "loss": 0.5828, + "step": 87 + }, + { + "epoch": 0.6, + "grad_norm": 0.44399499893188477, + "learning_rate": 4.4e-06, + "loss": 0.6143, + "step": 88 + }, + { + "epoch": 0.6068181818181818, + "grad_norm": 0.46058326959609985, + "learning_rate": 4.450000000000001e-06, + "loss": 0.6117, + "step": 89 + }, + { + "epoch": 0.6136363636363636, + "grad_norm": 0.8795785903930664, + "learning_rate": 4.5e-06, + "loss": 0.6, + "step": 90 + }, + { + "epoch": 0.6204545454545455, + "grad_norm": 0.5148798227310181, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.6041, + "step": 91 + }, + { + "epoch": 0.6272727272727273, + "grad_norm": 0.48829612135887146, + "learning_rate": 4.600000000000001e-06, + "loss": 0.5492, + "step": 92 + }, + { + "epoch": 0.634090909090909, + "grad_norm": 0.5211894512176514, + "learning_rate": 4.65e-06, + "loss": 0.6044, + "step": 93 + }, + { + "epoch": 0.6409090909090909, + "grad_norm": 0.44403275847435, + "learning_rate": 4.7e-06, + "loss": 0.5929, + "step": 94 + }, + { + "epoch": 0.6477272727272727, + "grad_norm": 0.6037693619728088, + "learning_rate": 4.75e-06, + "loss": 0.6132, + "step": 95 + }, + { + "epoch": 0.6545454545454545, + "grad_norm": 0.4381515085697174, + "learning_rate": 4.800000000000001e-06, + "loss": 0.5822, + "step": 96 + }, + { + "epoch": 0.6613636363636364, + "grad_norm": 0.4997427761554718, + "learning_rate": 4.85e-06, + "loss": 0.593, + "step": 97 + }, + { + "epoch": 0.6681818181818182, + "grad_norm": 0.440571665763855, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.6072, + "step": 98 + }, + { + "epoch": 0.675, + "grad_norm": 0.5020624995231628, + "learning_rate": 4.95e-06, + "loss": 0.5861, + "step": 99 + }, + { + "epoch": 0.6818181818181818, + "grad_norm": 0.4280160963535309, + "learning_rate": 5e-06, + "loss": 0.6163, + "step": 100 + }, + { + "epoch": 0.6886363636363636, + "grad_norm": 0.4584537148475647, + "learning_rate": 4.9999795126530275e-06, + "loss": 0.6177, + "step": 101 + }, + { + "epoch": 0.6954545454545454, + "grad_norm": 0.6835049390792847, + "learning_rate": 4.999918050947891e-06, + "loss": 0.579, + "step": 102 + }, + { + "epoch": 0.7022727272727273, + "grad_norm": 0.4551607072353363, + "learning_rate": 4.999815615891943e-06, + "loss": 0.5927, + "step": 103 + }, + { + "epoch": 0.7090909090909091, + "grad_norm": 0.5893972516059875, + "learning_rate": 4.9996722091640805e-06, + "loss": 0.5775, + "step": 104 + }, + { + "epoch": 0.7159090909090909, + "grad_norm": 0.4574092924594879, + "learning_rate": 4.9994878331147225e-06, + "loss": 0.5863, + "step": 105 + }, + { + "epoch": 0.7227272727272728, + "grad_norm": 0.5354658365249634, + "learning_rate": 4.99926249076577e-06, + "loss": 0.5453, + "step": 106 + }, + { + "epoch": 0.7295454545454545, + "grad_norm": 0.4347354471683502, + "learning_rate": 4.998996185810557e-06, + "loss": 0.5913, + "step": 107 + }, + { + "epoch": 0.7363636363636363, + "grad_norm": 0.4487966299057007, + "learning_rate": 4.998688922613788e-06, + "loss": 0.5749, + "step": 108 + }, + { + "epoch": 0.7431818181818182, + "grad_norm": 0.42477577924728394, + "learning_rate": 4.9983407062114695e-06, + "loss": 0.5761, + "step": 109 + }, + { + "epoch": 0.75, + "grad_norm": 0.4801446497440338, + "learning_rate": 4.9979515423108255e-06, + "loss": 0.5977, + "step": 110 + }, + { + "epoch": 0.7568181818181818, + "grad_norm": 0.4876883029937744, + "learning_rate": 4.997521437290205e-06, + "loss": 0.5849, + "step": 111 + }, + { + "epoch": 0.7636363636363637, + "grad_norm": 0.4328872263431549, + "learning_rate": 4.997050398198977e-06, + "loss": 0.5988, + "step": 112 + }, + { + "epoch": 0.7704545454545455, + "grad_norm": 0.4332719147205353, + "learning_rate": 4.996538432757414e-06, + "loss": 0.6059, + "step": 113 + }, + { + "epoch": 0.7772727272727272, + "grad_norm": 0.4455336630344391, + "learning_rate": 4.995985549356568e-06, + "loss": 0.5725, + "step": 114 + }, + { + "epoch": 0.7840909090909091, + "grad_norm": 0.4197766184806824, + "learning_rate": 4.995391757058129e-06, + "loss": 0.5715, + "step": 115 + }, + { + "epoch": 0.7909090909090909, + "grad_norm": 0.41945499181747437, + "learning_rate": 4.99475706559428e-06, + "loss": 0.5845, + "step": 116 + }, + { + "epoch": 0.7977272727272727, + "grad_norm": 0.42414823174476624, + "learning_rate": 4.994081485367537e-06, + "loss": 0.5595, + "step": 117 + }, + { + "epoch": 0.8045454545454546, + "grad_norm": 0.4056423008441925, + "learning_rate": 4.993365027450576e-06, + "loss": 0.582, + "step": 118 + }, + { + "epoch": 0.8113636363636364, + "grad_norm": 0.4554205536842346, + "learning_rate": 4.992607703586058e-06, + "loss": 0.5591, + "step": 119 + }, + { + "epoch": 0.8181818181818182, + "grad_norm": 0.4307089149951935, + "learning_rate": 4.991809526186424e-06, + "loss": 0.5927, + "step": 120 + }, + { + "epoch": 0.825, + "grad_norm": 0.48133528232574463, + "learning_rate": 4.990970508333707e-06, + "loss": 0.5703, + "step": 121 + }, + { + "epoch": 0.8318181818181818, + "grad_norm": 0.48582738637924194, + "learning_rate": 4.990090663779305e-06, + "loss": 0.5491, + "step": 122 + }, + { + "epoch": 0.8386363636363636, + "grad_norm": 0.44926196336746216, + "learning_rate": 4.9891700069437635e-06, + "loss": 0.5803, + "step": 123 + }, + { + "epoch": 0.8454545454545455, + "grad_norm": 0.4562165141105652, + "learning_rate": 4.988208552916535e-06, + "loss": 0.5945, + "step": 124 + }, + { + "epoch": 0.8522727272727273, + "grad_norm": 0.5885360836982727, + "learning_rate": 4.987206317455734e-06, + "loss": 0.5632, + "step": 125 + }, + { + "epoch": 0.8590909090909091, + "grad_norm": 0.4525550603866577, + "learning_rate": 4.986163316987877e-06, + "loss": 0.5221, + "step": 126 + }, + { + "epoch": 0.865909090909091, + "grad_norm": 0.4545478820800781, + "learning_rate": 4.985079568607613e-06, + "loss": 0.587, + "step": 127 + }, + { + "epoch": 0.8727272727272727, + "grad_norm": 0.4364396333694458, + "learning_rate": 4.983955090077445e-06, + "loss": 0.5626, + "step": 128 + }, + { + "epoch": 0.8795454545454545, + "grad_norm": 0.4354369640350342, + "learning_rate": 4.982789899827439e-06, + "loss": 0.5674, + "step": 129 + }, + { + "epoch": 0.8863636363636364, + "grad_norm": 0.4822773337364197, + "learning_rate": 4.9815840169549216e-06, + "loss": 0.5685, + "step": 130 + }, + { + "epoch": 0.8931818181818182, + "grad_norm": 0.44673025608062744, + "learning_rate": 4.980337461224164e-06, + "loss": 0.5733, + "step": 131 + }, + { + "epoch": 0.9, + "grad_norm": 0.6294501423835754, + "learning_rate": 4.979050253066064e-06, + "loss": 0.5396, + "step": 132 + }, + { + "epoch": 0.9068181818181819, + "grad_norm": 0.42890027165412903, + "learning_rate": 4.977722413577802e-06, + "loss": 0.5718, + "step": 133 + }, + { + "epoch": 0.9136363636363637, + "grad_norm": 0.470587819814682, + "learning_rate": 4.976353964522509e-06, + "loss": 0.5705, + "step": 134 + }, + { + "epoch": 0.9204545454545454, + "grad_norm": 0.4632768929004669, + "learning_rate": 4.974944928328894e-06, + "loss": 0.5394, + "step": 135 + }, + { + "epoch": 0.9272727272727272, + "grad_norm": 0.5304691195487976, + "learning_rate": 4.973495328090891e-06, + "loss": 0.5762, + "step": 136 + }, + { + "epoch": 0.9340909090909091, + "grad_norm": 0.42036354541778564, + "learning_rate": 4.972005187567267e-06, + "loss": 0.5649, + "step": 137 + }, + { + "epoch": 0.9409090909090909, + "grad_norm": 0.45189663767814636, + "learning_rate": 4.970474531181245e-06, + "loss": 0.5554, + "step": 138 + }, + { + "epoch": 0.9477272727272728, + "grad_norm": 0.47352468967437744, + "learning_rate": 4.968903384020095e-06, + "loss": 0.5511, + "step": 139 + }, + { + "epoch": 0.9545454545454546, + "grad_norm": 0.525702953338623, + "learning_rate": 4.967291771834727e-06, + "loss": 0.5777, + "step": 140 + }, + { + "epoch": 0.9613636363636363, + "grad_norm": 0.46146854758262634, + "learning_rate": 4.965639721039267e-06, + "loss": 0.5537, + "step": 141 + }, + { + "epoch": 0.9681818181818181, + "grad_norm": 0.5200941562652588, + "learning_rate": 4.963947258710626e-06, + "loss": 0.5522, + "step": 142 + }, + { + "epoch": 0.975, + "grad_norm": 0.5260375142097473, + "learning_rate": 4.962214412588053e-06, + "loss": 0.5747, + "step": 143 + }, + { + "epoch": 0.9818181818181818, + "grad_norm": 0.518645703792572, + "learning_rate": 4.960441211072686e-06, + "loss": 0.5257, + "step": 144 + }, + { + "epoch": 0.9886363636363636, + "grad_norm": 0.4528891444206238, + "learning_rate": 4.9586276832270785e-06, + "loss": 0.5548, + "step": 145 + }, + { + "epoch": 0.9954545454545455, + "grad_norm": 0.5137557983398438, + "learning_rate": 4.9567738587747314e-06, + "loss": 0.5526, + "step": 146 + }, + { + "epoch": 1.0068181818181818, + "grad_norm": 0.8773607611656189, + "learning_rate": 4.954879768099599e-06, + "loss": 1.1, + "step": 147 + }, + { + "epoch": 1.0136363636363637, + "grad_norm": 0.39079979062080383, + "learning_rate": 4.952945442245598e-06, + "loss": 0.568, + "step": 148 + }, + { + "epoch": 1.0204545454545455, + "grad_norm": 0.45004600286483765, + "learning_rate": 4.95097091291609e-06, + "loss": 0.5383, + "step": 149 + }, + { + "epoch": 1.0272727272727273, + "grad_norm": 0.41872867941856384, + "learning_rate": 4.948956212473371e-06, + "loss": 0.5436, + "step": 150 + }, + { + "epoch": 1.0340909090909092, + "grad_norm": 0.46829020977020264, + "learning_rate": 4.946901373938132e-06, + "loss": 0.5336, + "step": 151 + }, + { + "epoch": 1.040909090909091, + "grad_norm": 0.45038115978240967, + "learning_rate": 4.944806430988927e-06, + "loss": 0.538, + "step": 152 + }, + { + "epoch": 1.0477272727272728, + "grad_norm": 0.6496132016181946, + "learning_rate": 4.942671417961615e-06, + "loss": 0.5459, + "step": 153 + }, + { + "epoch": 1.0545454545454545, + "grad_norm": 0.4842703640460968, + "learning_rate": 4.940496369848795e-06, + "loss": 0.553, + "step": 154 + }, + { + "epoch": 1.0613636363636363, + "grad_norm": 0.4338674545288086, + "learning_rate": 4.938281322299243e-06, + "loss": 0.5366, + "step": 155 + }, + { + "epoch": 1.0681818181818181, + "grad_norm": 0.44630634784698486, + "learning_rate": 4.936026311617316e-06, + "loss": 0.5517, + "step": 156 + }, + { + "epoch": 1.075, + "grad_norm": 0.4606929421424866, + "learning_rate": 4.933731374762361e-06, + "loss": 0.5405, + "step": 157 + }, + { + "epoch": 1.0818181818181818, + "grad_norm": 0.48667654395103455, + "learning_rate": 4.931396549348115e-06, + "loss": 0.5376, + "step": 158 + }, + { + "epoch": 1.0886363636363636, + "grad_norm": 0.4516271650791168, + "learning_rate": 4.9290218736420795e-06, + "loss": 0.5206, + "step": 159 + }, + { + "epoch": 1.0954545454545455, + "grad_norm": 0.44719284772872925, + "learning_rate": 4.926607386564898e-06, + "loss": 0.5397, + "step": 160 + }, + { + "epoch": 1.1022727272727273, + "grad_norm": 0.40752628445625305, + "learning_rate": 4.9241531276897196e-06, + "loss": 0.5466, + "step": 161 + }, + { + "epoch": 1.1090909090909091, + "grad_norm": 0.45423609018325806, + "learning_rate": 4.921659137241544e-06, + "loss": 0.5227, + "step": 162 + }, + { + "epoch": 1.115909090909091, + "grad_norm": 0.4185822308063507, + "learning_rate": 4.919125456096574e-06, + "loss": 0.5144, + "step": 163 + }, + { + "epoch": 1.1227272727272728, + "grad_norm": 0.7753382921218872, + "learning_rate": 4.916552125781529e-06, + "loss": 0.5352, + "step": 164 + }, + { + "epoch": 1.1295454545454546, + "grad_norm": 0.47211477160453796, + "learning_rate": 4.913939188472979e-06, + "loss": 0.5505, + "step": 165 + }, + { + "epoch": 1.1363636363636362, + "grad_norm": 0.42742082476615906, + "learning_rate": 4.911286686996648e-06, + "loss": 0.529, + "step": 166 + }, + { + "epoch": 1.143181818181818, + "grad_norm": 0.41340839862823486, + "learning_rate": 4.908594664826708e-06, + "loss": 0.5292, + "step": 167 + }, + { + "epoch": 1.15, + "grad_norm": 0.44648951292037964, + "learning_rate": 4.905863166085076e-06, + "loss": 0.5757, + "step": 168 + }, + { + "epoch": 1.1568181818181817, + "grad_norm": 0.4285544753074646, + "learning_rate": 4.903092235540679e-06, + "loss": 0.5102, + "step": 169 + }, + { + "epoch": 1.1636363636363636, + "grad_norm": 0.42675015330314636, + "learning_rate": 4.900281918608732e-06, + "loss": 0.5397, + "step": 170 + }, + { + "epoch": 1.1704545454545454, + "grad_norm": 0.47636961936950684, + "learning_rate": 4.897432261349984e-06, + "loss": 0.5307, + "step": 171 + }, + { + "epoch": 1.1772727272727272, + "grad_norm": 0.4213427007198334, + "learning_rate": 4.894543310469968e-06, + "loss": 0.54, + "step": 172 + }, + { + "epoch": 1.184090909090909, + "grad_norm": 0.41153451800346375, + "learning_rate": 4.891615113318236e-06, + "loss": 0.5423, + "step": 173 + }, + { + "epoch": 1.190909090909091, + "grad_norm": 0.459185391664505, + "learning_rate": 4.888647717887582e-06, + "loss": 0.5239, + "step": 174 + }, + { + "epoch": 1.1977272727272728, + "grad_norm": 0.48279181122779846, + "learning_rate": 4.8856411728132526e-06, + "loss": 0.5336, + "step": 175 + }, + { + "epoch": 1.2045454545454546, + "grad_norm": 0.4186796247959137, + "learning_rate": 4.8825955273721524e-06, + "loss": 0.5509, + "step": 176 + }, + { + "epoch": 1.2113636363636364, + "grad_norm": 0.4345344007015228, + "learning_rate": 4.879510831482039e-06, + "loss": 0.5415, + "step": 177 + }, + { + "epoch": 1.2181818181818183, + "grad_norm": 0.461592435836792, + "learning_rate": 4.876387135700701e-06, + "loss": 0.5256, + "step": 178 + }, + { + "epoch": 1.225, + "grad_norm": 0.5960240960121155, + "learning_rate": 4.873224491225128e-06, + "loss": 0.5599, + "step": 179 + }, + { + "epoch": 1.231818181818182, + "grad_norm": 0.4662570357322693, + "learning_rate": 4.870022949890676e-06, + "loss": 0.5647, + "step": 180 + }, + { + "epoch": 1.2386363636363638, + "grad_norm": 0.4355057179927826, + "learning_rate": 4.866782564170217e-06, + "loss": 0.5056, + "step": 181 + }, + { + "epoch": 1.2454545454545454, + "grad_norm": 0.48699110746383667, + "learning_rate": 4.863503387173276e-06, + "loss": 0.5394, + "step": 182 + }, + { + "epoch": 1.2522727272727272, + "grad_norm": 0.48679423332214355, + "learning_rate": 4.860185472645161e-06, + "loss": 0.5394, + "step": 183 + }, + { + "epoch": 1.259090909090909, + "grad_norm": 0.6564393043518066, + "learning_rate": 4.856828874966086e-06, + "loss": 0.5259, + "step": 184 + }, + { + "epoch": 1.2659090909090909, + "grad_norm": 0.44762882590293884, + "learning_rate": 4.853433649150276e-06, + "loss": 0.5563, + "step": 185 + }, + { + "epoch": 1.2727272727272727, + "grad_norm": 0.5391108989715576, + "learning_rate": 4.849999850845066e-06, + "loss": 0.5326, + "step": 186 + }, + { + "epoch": 1.2795454545454545, + "grad_norm": 0.4201100170612335, + "learning_rate": 4.8465275363299905e-06, + "loss": 0.526, + "step": 187 + }, + { + "epoch": 1.2863636363636364, + "grad_norm": 0.46211543679237366, + "learning_rate": 4.84301676251586e-06, + "loss": 0.5498, + "step": 188 + }, + { + "epoch": 1.2931818181818182, + "grad_norm": 0.4325430989265442, + "learning_rate": 4.839467586943825e-06, + "loss": 0.556, + "step": 189 + }, + { + "epoch": 1.3, + "grad_norm": 0.46057143807411194, + "learning_rate": 4.835880067784441e-06, + "loss": 0.5079, + "step": 190 + }, + { + "epoch": 1.3068181818181819, + "grad_norm": 0.4920806288719177, + "learning_rate": 4.832254263836708e-06, + "loss": 0.533, + "step": 191 + }, + { + "epoch": 1.3136363636363637, + "grad_norm": 0.46693211793899536, + "learning_rate": 4.828590234527107e-06, + "loss": 0.5542, + "step": 192 + }, + { + "epoch": 1.3204545454545453, + "grad_norm": 0.4227873682975769, + "learning_rate": 4.82488803990863e-06, + "loss": 0.544, + "step": 193 + }, + { + "epoch": 1.3272727272727272, + "grad_norm": 0.44745922088623047, + "learning_rate": 4.821147740659795e-06, + "loss": 0.4962, + "step": 194 + }, + { + "epoch": 1.334090909090909, + "grad_norm": 0.46137240529060364, + "learning_rate": 4.817369398083648e-06, + "loss": 0.5359, + "step": 195 + }, + { + "epoch": 1.3409090909090908, + "grad_norm": 0.6483879685401917, + "learning_rate": 4.813553074106761e-06, + "loss": 0.5323, + "step": 196 + }, + { + "epoch": 1.3477272727272727, + "grad_norm": 0.6602574586868286, + "learning_rate": 4.809698831278217e-06, + "loss": 0.5161, + "step": 197 + }, + { + "epoch": 1.3545454545454545, + "grad_norm": 0.47758543491363525, + "learning_rate": 4.805806732768585e-06, + "loss": 0.5214, + "step": 198 + }, + { + "epoch": 1.3613636363636363, + "grad_norm": 0.4587438106536865, + "learning_rate": 4.801876842368882e-06, + "loss": 0.5208, + "step": 199 + }, + { + "epoch": 1.3681818181818182, + "grad_norm": 0.46055835485458374, + "learning_rate": 4.797909224489531e-06, + "loss": 0.5374, + "step": 200 + }, + { + "epoch": 1.375, + "grad_norm": 0.43727558851242065, + "learning_rate": 4.793903944159303e-06, + "loss": 0.5351, + "step": 201 + }, + { + "epoch": 1.3818181818181818, + "grad_norm": 0.4573572874069214, + "learning_rate": 4.789861067024253e-06, + "loss": 0.4997, + "step": 202 + }, + { + "epoch": 1.3886363636363637, + "grad_norm": 0.5921999216079712, + "learning_rate": 4.785780659346642e-06, + "loss": 0.5168, + "step": 203 + }, + { + "epoch": 1.3954545454545455, + "grad_norm": 0.47943904995918274, + "learning_rate": 4.781662788003851e-06, + "loss": 0.5513, + "step": 204 + }, + { + "epoch": 1.4022727272727273, + "grad_norm": 0.43733686208724976, + "learning_rate": 4.777507520487289e-06, + "loss": 0.5281, + "step": 205 + }, + { + "epoch": 1.4090909090909092, + "grad_norm": 0.4289107620716095, + "learning_rate": 4.773314924901281e-06, + "loss": 0.524, + "step": 206 + }, + { + "epoch": 1.415909090909091, + "grad_norm": 0.4203416109085083, + "learning_rate": 4.769085069961955e-06, + "loss": 0.5563, + "step": 207 + }, + { + "epoch": 1.4227272727272728, + "grad_norm": 0.4205201268196106, + "learning_rate": 4.764818024996117e-06, + "loss": 0.5202, + "step": 208 + }, + { + "epoch": 1.4295454545454547, + "grad_norm": 0.4296267032623291, + "learning_rate": 4.760513859940112e-06, + "loss": 0.5113, + "step": 209 + }, + { + "epoch": 1.4363636363636363, + "grad_norm": 0.3979087471961975, + "learning_rate": 4.756172645338675e-06, + "loss": 0.5313, + "step": 210 + }, + { + "epoch": 1.4431818181818181, + "grad_norm": 0.45232149958610535, + "learning_rate": 4.751794452343785e-06, + "loss": 0.5204, + "step": 211 + }, + { + "epoch": 1.45, + "grad_norm": 0.4309398829936981, + "learning_rate": 4.747379352713489e-06, + "loss": 0.5216, + "step": 212 + }, + { + "epoch": 1.4568181818181818, + "grad_norm": 0.45721709728240967, + "learning_rate": 4.7429274188107275e-06, + "loss": 0.5279, + "step": 213 + }, + { + "epoch": 1.4636363636363636, + "grad_norm": 0.46138322353363037, + "learning_rate": 4.738438723602154e-06, + "loss": 0.5287, + "step": 214 + }, + { + "epoch": 1.4704545454545455, + "grad_norm": 0.41752108931541443, + "learning_rate": 4.733913340656933e-06, + "loss": 0.5399, + "step": 215 + }, + { + "epoch": 1.4772727272727273, + "grad_norm": 0.45447197556495667, + "learning_rate": 4.729351344145536e-06, + "loss": 0.5277, + "step": 216 + }, + { + "epoch": 1.4840909090909091, + "grad_norm": 0.42524856328964233, + "learning_rate": 4.7247528088385296e-06, + "loss": 0.5103, + "step": 217 + }, + { + "epoch": 1.490909090909091, + "grad_norm": 0.46024221181869507, + "learning_rate": 4.720117810105341e-06, + "loss": 0.5172, + "step": 218 + }, + { + "epoch": 1.4977272727272728, + "grad_norm": 0.5868630409240723, + "learning_rate": 4.715446423913036e-06, + "loss": 0.5409, + "step": 219 + }, + { + "epoch": 1.5045454545454544, + "grad_norm": 0.424813836812973, + "learning_rate": 4.710738726825059e-06, + "loss": 0.5495, + "step": 220 + }, + { + "epoch": 1.5113636363636362, + "grad_norm": 0.49356943368911743, + "learning_rate": 4.705994795999991e-06, + "loss": 0.519, + "step": 221 + }, + { + "epoch": 1.518181818181818, + "grad_norm": 0.44291776418685913, + "learning_rate": 4.701214709190277e-06, + "loss": 0.5232, + "step": 222 + }, + { + "epoch": 1.525, + "grad_norm": 0.48171764612197876, + "learning_rate": 4.696398544740955e-06, + "loss": 0.5501, + "step": 223 + }, + { + "epoch": 1.5318181818181817, + "grad_norm": 0.4296175539493561, + "learning_rate": 4.69154638158837e-06, + "loss": 0.5087, + "step": 224 + }, + { + "epoch": 1.5386363636363636, + "grad_norm": 0.4715246260166168, + "learning_rate": 4.686658299258881e-06, + "loss": 0.5363, + "step": 225 + }, + { + "epoch": 1.5454545454545454, + "grad_norm": 0.44422832131385803, + "learning_rate": 4.681734377867562e-06, + "loss": 0.5371, + "step": 226 + }, + { + "epoch": 1.5522727272727272, + "grad_norm": 4.221654891967773, + "learning_rate": 4.67677469811688e-06, + "loss": 0.5191, + "step": 227 + }, + { + "epoch": 1.559090909090909, + "grad_norm": 0.41857993602752686, + "learning_rate": 4.671779341295378e-06, + "loss": 0.5191, + "step": 228 + }, + { + "epoch": 1.565909090909091, + "grad_norm": 0.4950161278247833, + "learning_rate": 4.666748389276344e-06, + "loss": 0.5311, + "step": 229 + }, + { + "epoch": 1.5727272727272728, + "grad_norm": 0.44641605019569397, + "learning_rate": 4.661681924516466e-06, + "loss": 0.5174, + "step": 230 + }, + { + "epoch": 1.5795454545454546, + "grad_norm": 0.39064037799835205, + "learning_rate": 4.6565800300544805e-06, + "loss": 0.515, + "step": 231 + }, + { + "epoch": 1.5863636363636364, + "grad_norm": 0.7764399647712708, + "learning_rate": 4.651442789509813e-06, + "loss": 0.5282, + "step": 232 + }, + { + "epoch": 1.5931818181818183, + "grad_norm": 0.44421035051345825, + "learning_rate": 4.646270287081208e-06, + "loss": 0.5324, + "step": 233 + }, + { + "epoch": 1.6, + "grad_norm": 0.5149928331375122, + "learning_rate": 4.641062607545347e-06, + "loss": 0.5264, + "step": 234 + }, + { + "epoch": 1.606818181818182, + "grad_norm": 0.4784093201160431, + "learning_rate": 4.6358198362554585e-06, + "loss": 0.552, + "step": 235 + }, + { + "epoch": 1.6136363636363638, + "grad_norm": 0.4138723313808441, + "learning_rate": 4.630542059139923e-06, + "loss": 0.5202, + "step": 236 + }, + { + "epoch": 1.6204545454545456, + "grad_norm": 0.4509594440460205, + "learning_rate": 4.625229362700863e-06, + "loss": 0.5318, + "step": 237 + }, + { + "epoch": 1.6272727272727274, + "grad_norm": 0.44979560375213623, + "learning_rate": 4.61988183401272e-06, + "loss": 0.521, + "step": 238 + }, + { + "epoch": 1.634090909090909, + "grad_norm": 0.4525575041770935, + "learning_rate": 4.614499560720837e-06, + "loss": 0.5073, + "step": 239 + }, + { + "epoch": 1.6409090909090909, + "grad_norm": 0.44126543402671814, + "learning_rate": 4.609082631040012e-06, + "loss": 0.5058, + "step": 240 + }, + { + "epoch": 1.6477272727272727, + "grad_norm": 0.887867271900177, + "learning_rate": 4.603631133753061e-06, + "loss": 0.5471, + "step": 241 + }, + { + "epoch": 1.6545454545454545, + "grad_norm": 0.4497375786304474, + "learning_rate": 4.598145158209356e-06, + "loss": 0.501, + "step": 242 + }, + { + "epoch": 1.6613636363636364, + "grad_norm": 0.45681032538414, + "learning_rate": 4.592624794323366e-06, + "loss": 0.5306, + "step": 243 + }, + { + "epoch": 1.6681818181818182, + "grad_norm": 0.5186678171157837, + "learning_rate": 4.587070132573178e-06, + "loss": 0.5097, + "step": 244 + }, + { + "epoch": 1.675, + "grad_norm": 0.44289669394493103, + "learning_rate": 4.581481263999019e-06, + "loss": 0.5085, + "step": 245 + }, + { + "epoch": 1.6818181818181817, + "grad_norm": 0.4571002125740051, + "learning_rate": 4.575858280201761e-06, + "loss": 0.5327, + "step": 246 + }, + { + "epoch": 1.6886363636363635, + "grad_norm": 0.43753257393836975, + "learning_rate": 4.570201273341418e-06, + "loss": 0.5167, + "step": 247 + }, + { + "epoch": 1.6954545454545453, + "grad_norm": 0.44994574785232544, + "learning_rate": 4.564510336135642e-06, + "loss": 0.5554, + "step": 248 + }, + { + "epoch": 1.7022727272727272, + "grad_norm": 0.6185874938964844, + "learning_rate": 4.558785561858196e-06, + "loss": 0.5115, + "step": 249 + }, + { + "epoch": 1.709090909090909, + "grad_norm": 0.4883944094181061, + "learning_rate": 4.5530270443374305e-06, + "loss": 0.5312, + "step": 250 + }, + { + "epoch": 1.7159090909090908, + "grad_norm": 0.46433499455451965, + "learning_rate": 4.547234877954741e-06, + "loss": 0.5368, + "step": 251 + }, + { + "epoch": 1.7227272727272727, + "grad_norm": 0.4270826578140259, + "learning_rate": 4.541409157643027e-06, + "loss": 0.5258, + "step": 252 + }, + { + "epoch": 1.7295454545454545, + "grad_norm": 0.4390241205692291, + "learning_rate": 4.535549978885132e-06, + "loss": 0.5259, + "step": 253 + }, + { + "epoch": 1.7363636363636363, + "grad_norm": 0.42756086587905884, + "learning_rate": 4.5296574377122765e-06, + "loss": 0.5217, + "step": 254 + }, + { + "epoch": 1.7431818181818182, + "grad_norm": 0.48124927282333374, + "learning_rate": 4.5237316307024895e-06, + "loss": 0.5095, + "step": 255 + }, + { + "epoch": 1.75, + "grad_norm": 0.4394672214984894, + "learning_rate": 4.517772654979024e-06, + "loss": 0.5116, + "step": 256 + }, + { + "epoch": 1.7568181818181818, + "grad_norm": 0.5805886387825012, + "learning_rate": 4.51178060820876e-06, + "loss": 0.5145, + "step": 257 + }, + { + "epoch": 1.7636363636363637, + "grad_norm": 0.4845556616783142, + "learning_rate": 4.505755588600613e-06, + "loss": 0.5064, + "step": 258 + }, + { + "epoch": 1.7704545454545455, + "grad_norm": 0.45300963521003723, + "learning_rate": 4.499697694903915e-06, + "loss": 0.5362, + "step": 259 + }, + { + "epoch": 1.7772727272727273, + "grad_norm": 0.45417436957359314, + "learning_rate": 4.493607026406802e-06, + "loss": 0.5234, + "step": 260 + }, + { + "epoch": 1.7840909090909092, + "grad_norm": 0.4577132761478424, + "learning_rate": 4.487483682934587e-06, + "loss": 0.5271, + "step": 261 + }, + { + "epoch": 1.790909090909091, + "grad_norm": 0.4518176019191742, + "learning_rate": 4.481327764848118e-06, + "loss": 0.4853, + "step": 262 + }, + { + "epoch": 1.7977272727272728, + "grad_norm": 0.43804875016212463, + "learning_rate": 4.47513937304214e-06, + "loss": 0.5188, + "step": 263 + }, + { + "epoch": 1.8045454545454547, + "grad_norm": 0.4401354193687439, + "learning_rate": 4.4689186089436365e-06, + "loss": 0.5079, + "step": 264 + }, + { + "epoch": 1.8113636363636365, + "grad_norm": 0.4051227569580078, + "learning_rate": 4.462665574510169e-06, + "loss": 0.5084, + "step": 265 + }, + { + "epoch": 1.8181818181818183, + "grad_norm": 0.5269004106521606, + "learning_rate": 4.456380372228208e-06, + "loss": 0.528, + "step": 266 + }, + { + "epoch": 1.825, + "grad_norm": 0.44371986389160156, + "learning_rate": 4.450063105111447e-06, + "loss": 0.5388, + "step": 267 + }, + { + "epoch": 1.8318181818181818, + "grad_norm": 0.41871729493141174, + "learning_rate": 4.443713876699124e-06, + "loss": 0.502, + "step": 268 + }, + { + "epoch": 1.8386363636363636, + "grad_norm": 0.464652419090271, + "learning_rate": 4.4373327910543125e-06, + "loss": 0.5045, + "step": 269 + }, + { + "epoch": 1.8454545454545455, + "grad_norm": 0.43198078870773315, + "learning_rate": 4.430919952762226e-06, + "loss": 0.5149, + "step": 270 + }, + { + "epoch": 1.8522727272727273, + "grad_norm": 0.42869722843170166, + "learning_rate": 4.424475466928499e-06, + "loss": 0.5329, + "step": 271 + }, + { + "epoch": 1.8590909090909091, + "grad_norm": 0.529680609703064, + "learning_rate": 4.417999439177465e-06, + "loss": 0.4939, + "step": 272 + }, + { + "epoch": 1.865909090909091, + "grad_norm": 0.4562613368034363, + "learning_rate": 4.4114919756504275e-06, + "loss": 0.5389, + "step": 273 + }, + { + "epoch": 1.8727272727272726, + "grad_norm": 0.4480995237827301, + "learning_rate": 4.404953183003916e-06, + "loss": 0.51, + "step": 274 + }, + { + "epoch": 1.8795454545454544, + "grad_norm": 0.41215088963508606, + "learning_rate": 4.398383168407941e-06, + "loss": 0.5386, + "step": 275 + }, + { + "epoch": 1.8863636363636362, + "grad_norm": 0.8426409363746643, + "learning_rate": 4.391782039544239e-06, + "loss": 0.498, + "step": 276 + }, + { + "epoch": 1.893181818181818, + "grad_norm": 0.5359936356544495, + "learning_rate": 4.385149904604502e-06, + "loss": 0.51, + "step": 277 + }, + { + "epoch": 1.9, + "grad_norm": 0.4462437033653259, + "learning_rate": 4.378486872288611e-06, + "loss": 0.5082, + "step": 278 + }, + { + "epoch": 1.9068181818181817, + "grad_norm": 0.445359468460083, + "learning_rate": 4.371793051802849e-06, + "loss": 0.5171, + "step": 279 + }, + { + "epoch": 1.9136363636363636, + "grad_norm": 0.4223516583442688, + "learning_rate": 4.365068552858116e-06, + "loss": 0.5218, + "step": 280 + }, + { + "epoch": 1.9204545454545454, + "grad_norm": 0.43929681181907654, + "learning_rate": 4.358313485668124e-06, + "loss": 0.5156, + "step": 281 + }, + { + "epoch": 1.9272727272727272, + "grad_norm": 0.42535650730133057, + "learning_rate": 4.3515279609476e-06, + "loss": 0.4964, + "step": 282 + }, + { + "epoch": 1.934090909090909, + "grad_norm": 0.46448391675949097, + "learning_rate": 4.3447120899104615e-06, + "loss": 0.5182, + "step": 283 + }, + { + "epoch": 1.940909090909091, + "grad_norm": 0.4082206189632416, + "learning_rate": 4.337865984268002e-06, + "loss": 0.5175, + "step": 284 + }, + { + "epoch": 1.9477272727272728, + "grad_norm": 0.42423564195632935, + "learning_rate": 4.3309897562270525e-06, + "loss": 0.5352, + "step": 285 + }, + { + "epoch": 1.9545454545454546, + "grad_norm": 0.5240079760551453, + "learning_rate": 4.324083518488151e-06, + "loss": 0.5157, + "step": 286 + }, + { + "epoch": 1.9613636363636364, + "grad_norm": 0.4517645835876465, + "learning_rate": 4.317147384243688e-06, + "loss": 0.5399, + "step": 287 + }, + { + "epoch": 1.9681818181818183, + "grad_norm": 0.44054290652275085, + "learning_rate": 4.3101814671760546e-06, + "loss": 0.5293, + "step": 288 + }, + { + "epoch": 1.975, + "grad_norm": 0.4256063401699066, + "learning_rate": 4.303185881455778e-06, + "loss": 0.5323, + "step": 289 + }, + { + "epoch": 1.981818181818182, + "grad_norm": 0.43106094002723694, + "learning_rate": 4.296160741739652e-06, + "loss": 0.5259, + "step": 290 + }, + { + "epoch": 1.9886363636363638, + "grad_norm": 0.43816477060317993, + "learning_rate": 4.289106163168858e-06, + "loss": 0.5098, + "step": 291 + }, + { + "epoch": 1.9954545454545456, + "grad_norm": 1.1177959442138672, + "learning_rate": 4.282022261367074e-06, + "loss": 0.513, + "step": 292 + }, + { + "epoch": 2.006818181818182, + "grad_norm": 0.8377058506011963, + "learning_rate": 4.274909152438582e-06, + "loss": 0.9984, + "step": 293 + }, + { + "epoch": 2.0136363636363637, + "grad_norm": 0.4339721202850342, + "learning_rate": 4.267766952966369e-06, + "loss": 0.493, + "step": 294 + }, + { + "epoch": 2.0204545454545455, + "grad_norm": 0.39961451292037964, + "learning_rate": 4.260595780010209e-06, + "loss": 0.486, + "step": 295 + }, + { + "epoch": 2.0272727272727273, + "grad_norm": 0.4809112548828125, + "learning_rate": 4.2533957511047485e-06, + "loss": 0.4803, + "step": 296 + }, + { + "epoch": 2.034090909090909, + "grad_norm": 0.414700984954834, + "learning_rate": 4.24616698425758e-06, + "loss": 0.4566, + "step": 297 + }, + { + "epoch": 2.040909090909091, + "grad_norm": 0.552201509475708, + "learning_rate": 4.238909597947307e-06, + "loss": 0.4775, + "step": 298 + }, + { + "epoch": 2.047727272727273, + "grad_norm": 0.4647877812385559, + "learning_rate": 4.231623711121603e-06, + "loss": 0.4944, + "step": 299 + }, + { + "epoch": 2.0545454545454547, + "grad_norm": 0.684251606464386, + "learning_rate": 4.224309443195261e-06, + "loss": 0.495, + "step": 300 + }, + { + "epoch": 2.0613636363636365, + "grad_norm": 0.42145106196403503, + "learning_rate": 4.2169669140482365e-06, + "loss": 0.4809, + "step": 301 + }, + { + "epoch": 2.0681818181818183, + "grad_norm": 0.43771296739578247, + "learning_rate": 4.2095962440236846e-06, + "loss": 0.5033, + "step": 302 + }, + { + "epoch": 2.075, + "grad_norm": 0.4323962926864624, + "learning_rate": 4.202197553925983e-06, + "loss": 0.4843, + "step": 303 + }, + { + "epoch": 2.081818181818182, + "grad_norm": 0.40169817209243774, + "learning_rate": 4.194770965018758e-06, + "loss": 0.504, + "step": 304 + }, + { + "epoch": 2.088636363636364, + "grad_norm": 0.4206427037715912, + "learning_rate": 4.187316599022892e-06, + "loss": 0.5059, + "step": 305 + }, + { + "epoch": 2.0954545454545457, + "grad_norm": 0.4550352692604065, + "learning_rate": 4.179834578114531e-06, + "loss": 0.4717, + "step": 306 + }, + { + "epoch": 2.102272727272727, + "grad_norm": 0.4193796217441559, + "learning_rate": 4.172325024923083e-06, + "loss": 0.5018, + "step": 307 + }, + { + "epoch": 2.109090909090909, + "grad_norm": 0.550195038318634, + "learning_rate": 4.164788062529203e-06, + "loss": 0.4961, + "step": 308 + }, + { + "epoch": 2.1159090909090907, + "grad_norm": 0.43539106845855713, + "learning_rate": 4.157223814462784e-06, + "loss": 0.4909, + "step": 309 + }, + { + "epoch": 2.1227272727272726, + "grad_norm": 0.4565756022930145, + "learning_rate": 4.149632404700925e-06, + "loss": 0.49, + "step": 310 + }, + { + "epoch": 2.1295454545454544, + "grad_norm": 0.4206918179988861, + "learning_rate": 4.142013957665903e-06, + "loss": 0.4722, + "step": 311 + }, + { + "epoch": 2.1363636363636362, + "grad_norm": 0.4098123013973236, + "learning_rate": 4.134368598223132e-06, + "loss": 0.4791, + "step": 312 + }, + { + "epoch": 2.143181818181818, + "grad_norm": 0.4387701153755188, + "learning_rate": 4.126696451679116e-06, + "loss": 0.4758, + "step": 313 + }, + { + "epoch": 2.15, + "grad_norm": 0.4594134986400604, + "learning_rate": 4.118997643779401e-06, + "loss": 0.5009, + "step": 314 + }, + { + "epoch": 2.1568181818181817, + "grad_norm": 0.4581888020038605, + "learning_rate": 4.111272300706502e-06, + "loss": 0.4915, + "step": 315 + }, + { + "epoch": 2.1636363636363636, + "grad_norm": 0.5279524326324463, + "learning_rate": 4.1035205490778505e-06, + "loss": 0.5056, + "step": 316 + }, + { + "epoch": 2.1704545454545454, + "grad_norm": 0.4253177344799042, + "learning_rate": 4.095742515943703e-06, + "loss": 0.5001, + "step": 317 + }, + { + "epoch": 2.1772727272727272, + "grad_norm": 0.5294811129570007, + "learning_rate": 4.087938328785071e-06, + "loss": 0.5027, + "step": 318 + }, + { + "epoch": 2.184090909090909, + "grad_norm": 0.4642890989780426, + "learning_rate": 4.080108115511629e-06, + "loss": 0.5051, + "step": 319 + }, + { + "epoch": 2.190909090909091, + "grad_norm": 0.43700429797172546, + "learning_rate": 4.072252004459612e-06, + "loss": 0.4809, + "step": 320 + }, + { + "epoch": 2.1977272727272728, + "grad_norm": 0.4536914825439453, + "learning_rate": 4.064370124389718e-06, + "loss": 0.5129, + "step": 321 + }, + { + "epoch": 2.2045454545454546, + "grad_norm": 0.42350664734840393, + "learning_rate": 4.056462604484998e-06, + "loss": 0.4912, + "step": 322 + }, + { + "epoch": 2.2113636363636364, + "grad_norm": 0.47886520624160767, + "learning_rate": 4.048529574348734e-06, + "loss": 0.4757, + "step": 323 + }, + { + "epoch": 2.2181818181818183, + "grad_norm": 0.5205736756324768, + "learning_rate": 4.040571164002319e-06, + "loss": 0.4821, + "step": 324 + }, + { + "epoch": 2.225, + "grad_norm": 0.4465322196483612, + "learning_rate": 4.032587503883124e-06, + "loss": 0.5057, + "step": 325 + }, + { + "epoch": 2.231818181818182, + "grad_norm": 0.41364315152168274, + "learning_rate": 4.0245787248423614e-06, + "loss": 0.4809, + "step": 326 + }, + { + "epoch": 2.2386363636363638, + "grad_norm": 0.42846816778182983, + "learning_rate": 4.0165449581429404e-06, + "loss": 0.4522, + "step": 327 + }, + { + "epoch": 2.2454545454545456, + "grad_norm": 0.4604497253894806, + "learning_rate": 4.008486335457312e-06, + "loss": 0.4879, + "step": 328 + }, + { + "epoch": 2.2522727272727274, + "grad_norm": 0.45866841077804565, + "learning_rate": 4.000402988865316e-06, + "loss": 0.4969, + "step": 329 + }, + { + "epoch": 2.2590909090909093, + "grad_norm": 0.5116038918495178, + "learning_rate": 3.992295050852013e-06, + "loss": 0.5015, + "step": 330 + }, + { + "epoch": 2.265909090909091, + "grad_norm": 0.4389069080352783, + "learning_rate": 3.984162654305516e-06, + "loss": 0.4797, + "step": 331 + }, + { + "epoch": 2.2727272727272725, + "grad_norm": 0.45724064111709595, + "learning_rate": 3.976005932514807e-06, + "loss": 0.4915, + "step": 332 + }, + { + "epoch": 2.2795454545454543, + "grad_norm": 0.44370219111442566, + "learning_rate": 3.967825019167559e-06, + "loss": 0.499, + "step": 333 + }, + { + "epoch": 2.286363636363636, + "grad_norm": 0.4414970278739929, + "learning_rate": 3.959620048347938e-06, + "loss": 0.447, + "step": 334 + }, + { + "epoch": 2.293181818181818, + "grad_norm": 0.7909289598464966, + "learning_rate": 3.951391154534415e-06, + "loss": 0.4658, + "step": 335 + }, + { + "epoch": 2.3, + "grad_norm": 0.4643534719944, + "learning_rate": 3.943138472597549e-06, + "loss": 0.4941, + "step": 336 + }, + { + "epoch": 2.3068181818181817, + "grad_norm": 0.5114858150482178, + "learning_rate": 3.934862137797788e-06, + "loss": 0.4733, + "step": 337 + }, + { + "epoch": 2.3136363636363635, + "grad_norm": 0.4057219624519348, + "learning_rate": 3.9265622857832455e-06, + "loss": 0.4965, + "step": 338 + }, + { + "epoch": 2.3204545454545453, + "grad_norm": 0.4124796390533447, + "learning_rate": 3.918239052587481e-06, + "loss": 0.4508, + "step": 339 + }, + { + "epoch": 2.327272727272727, + "grad_norm": 0.43895164132118225, + "learning_rate": 3.909892574627267e-06, + "loss": 0.4939, + "step": 340 + }, + { + "epoch": 2.334090909090909, + "grad_norm": 0.47555556893348694, + "learning_rate": 3.901522988700355e-06, + "loss": 0.4541, + "step": 341 + }, + { + "epoch": 2.340909090909091, + "grad_norm": 0.44449353218078613, + "learning_rate": 3.893130431983234e-06, + "loss": 0.4883, + "step": 342 + }, + { + "epoch": 2.3477272727272727, + "grad_norm": 0.441793829202652, + "learning_rate": 3.884715042028882e-06, + "loss": 0.4911, + "step": 343 + }, + { + "epoch": 2.3545454545454545, + "grad_norm": 0.4836761951446533, + "learning_rate": 3.876276956764509e-06, + "loss": 0.4838, + "step": 344 + }, + { + "epoch": 2.3613636363636363, + "grad_norm": 0.7246759533882141, + "learning_rate": 3.867816314489301e-06, + "loss": 0.5014, + "step": 345 + }, + { + "epoch": 2.368181818181818, + "grad_norm": 0.5535604357719421, + "learning_rate": 3.8593332538721465e-06, + "loss": 0.4654, + "step": 346 + }, + { + "epoch": 2.375, + "grad_norm": 0.4468318819999695, + "learning_rate": 3.8508279139493736e-06, + "loss": 0.463, + "step": 347 + }, + { + "epoch": 2.381818181818182, + "grad_norm": 0.4505653381347656, + "learning_rate": 3.84230043412246e-06, + "loss": 0.4614, + "step": 348 + }, + { + "epoch": 2.3886363636363637, + "grad_norm": 0.4461449384689331, + "learning_rate": 3.833750954155757e-06, + "loss": 0.4777, + "step": 349 + }, + { + "epoch": 2.3954545454545455, + "grad_norm": 0.423711895942688, + "learning_rate": 3.825179614174195e-06, + "loss": 0.4581, + "step": 350 + }, + { + "epoch": 2.4022727272727273, + "grad_norm": 0.462819904088974, + "learning_rate": 3.816586554660987e-06, + "loss": 0.4782, + "step": 351 + }, + { + "epoch": 2.409090909090909, + "grad_norm": 0.44206613302230835, + "learning_rate": 3.807971916455325e-06, + "loss": 0.477, + "step": 352 + }, + { + "epoch": 2.415909090909091, + "grad_norm": 0.4178541600704193, + "learning_rate": 3.799335840750077e-06, + "loss": 0.4981, + "step": 353 + }, + { + "epoch": 2.422727272727273, + "grad_norm": 0.670315682888031, + "learning_rate": 3.790678469089465e-06, + "loss": 0.4762, + "step": 354 + }, + { + "epoch": 2.4295454545454547, + "grad_norm": 0.44383955001831055, + "learning_rate": 3.7819999433667503e-06, + "loss": 0.4807, + "step": 355 + }, + { + "epoch": 2.4363636363636365, + "grad_norm": 0.4850422143936157, + "learning_rate": 3.773300405821908e-06, + "loss": 0.479, + "step": 356 + }, + { + "epoch": 2.4431818181818183, + "grad_norm": 0.458435595035553, + "learning_rate": 3.764579999039293e-06, + "loss": 0.4628, + "step": 357 + }, + { + "epoch": 2.45, + "grad_norm": 0.43913447856903076, + "learning_rate": 3.7558388659453052e-06, + "loss": 0.4948, + "step": 358 + }, + { + "epoch": 2.456818181818182, + "grad_norm": 0.4294499456882477, + "learning_rate": 3.7470771498060455e-06, + "loss": 0.4979, + "step": 359 + }, + { + "epoch": 2.463636363636364, + "grad_norm": 0.4472196400165558, + "learning_rate": 3.7382949942249695e-06, + "loss": 0.496, + "step": 360 + }, + { + "epoch": 2.4704545454545457, + "grad_norm": 0.4681987762451172, + "learning_rate": 3.7294925431405306e-06, + "loss": 0.4591, + "step": 361 + }, + { + "epoch": 2.4772727272727275, + "grad_norm": 0.40683335065841675, + "learning_rate": 3.720669940823827e-06, + "loss": 0.4755, + "step": 362 + }, + { + "epoch": 2.484090909090909, + "grad_norm": 0.46134939789772034, + "learning_rate": 3.7118273318762275e-06, + "loss": 0.4956, + "step": 363 + }, + { + "epoch": 2.4909090909090907, + "grad_norm": 0.4279940724372864, + "learning_rate": 3.702964861227013e-06, + "loss": 0.4815, + "step": 364 + }, + { + "epoch": 2.4977272727272726, + "grad_norm": 0.4699287414550781, + "learning_rate": 3.694082674130991e-06, + "loss": 0.4846, + "step": 365 + }, + { + "epoch": 2.5045454545454544, + "grad_norm": 0.4230785369873047, + "learning_rate": 3.6851809161661206e-06, + "loss": 0.5077, + "step": 366 + }, + { + "epoch": 2.5113636363636362, + "grad_norm": 0.4509921967983246, + "learning_rate": 3.6762597332311254e-06, + "loss": 0.5134, + "step": 367 + }, + { + "epoch": 2.518181818181818, + "grad_norm": 0.45494896173477173, + "learning_rate": 3.6673192715431016e-06, + "loss": 0.4685, + "step": 368 + }, + { + "epoch": 2.525, + "grad_norm": 0.4804443120956421, + "learning_rate": 3.658359677635122e-06, + "loss": 0.4458, + "step": 369 + }, + { + "epoch": 2.5318181818181817, + "grad_norm": 0.4149218499660492, + "learning_rate": 3.649381098353834e-06, + "loss": 0.5102, + "step": 370 + }, + { + "epoch": 2.5386363636363636, + "grad_norm": 0.4423210024833679, + "learning_rate": 3.6403836808570512e-06, + "loss": 0.5034, + "step": 371 + }, + { + "epoch": 2.5454545454545454, + "grad_norm": 0.4915727376937866, + "learning_rate": 3.631367572611348e-06, + "loss": 0.4889, + "step": 372 + }, + { + "epoch": 2.5522727272727272, + "grad_norm": 0.4189188778400421, + "learning_rate": 3.6223329213896313e-06, + "loss": 0.4829, + "step": 373 + }, + { + "epoch": 2.559090909090909, + "grad_norm": 0.511681079864502, + "learning_rate": 3.613279875268731e-06, + "loss": 0.4786, + "step": 374 + }, + { + "epoch": 2.565909090909091, + "grad_norm": 0.430137038230896, + "learning_rate": 3.604208582626964e-06, + "loss": 0.4783, + "step": 375 + }, + { + "epoch": 2.5727272727272728, + "grad_norm": 0.5629504919052124, + "learning_rate": 3.5951191921417063e-06, + "loss": 0.4438, + "step": 376 + }, + { + "epoch": 2.5795454545454546, + "grad_norm": 0.4354007840156555, + "learning_rate": 3.586011852786955e-06, + "loss": 0.4869, + "step": 377 + }, + { + "epoch": 2.5863636363636364, + "grad_norm": 0.45930856466293335, + "learning_rate": 3.5768867138308872e-06, + "loss": 0.4683, + "step": 378 + }, + { + "epoch": 2.5931818181818183, + "grad_norm": 0.4965468943119049, + "learning_rate": 3.5677439248334133e-06, + "loss": 0.4739, + "step": 379 + }, + { + "epoch": 2.6, + "grad_norm": 0.39618629217147827, + "learning_rate": 3.5585836356437266e-06, + "loss": 0.486, + "step": 380 + }, + { + "epoch": 2.606818181818182, + "grad_norm": 0.504718542098999, + "learning_rate": 3.5494059963978433e-06, + "loss": 0.4608, + "step": 381 + }, + { + "epoch": 2.6136363636363638, + "grad_norm": 0.3951753079891205, + "learning_rate": 3.540211157516149e-06, + "loss": 0.4973, + "step": 382 + }, + { + "epoch": 2.6204545454545456, + "grad_norm": 0.45937463641166687, + "learning_rate": 3.530999269700927e-06, + "loss": 0.4751, + "step": 383 + }, + { + "epoch": 2.6272727272727274, + "grad_norm": 0.4644455909729004, + "learning_rate": 3.521770483933891e-06, + "loss": 0.5003, + "step": 384 + }, + { + "epoch": 2.634090909090909, + "grad_norm": 0.4427309036254883, + "learning_rate": 3.5125249514737093e-06, + "loss": 0.4779, + "step": 385 + }, + { + "epoch": 2.6409090909090907, + "grad_norm": 0.4289201498031616, + "learning_rate": 3.503262823853527e-06, + "loss": 0.48, + "step": 386 + }, + { + "epoch": 2.6477272727272725, + "grad_norm": 0.5118290185928345, + "learning_rate": 3.493984252878483e-06, + "loss": 0.4858, + "step": 387 + }, + { + "epoch": 2.6545454545454543, + "grad_norm": 0.4599657952785492, + "learning_rate": 3.484689390623218e-06, + "loss": 0.4813, + "step": 388 + }, + { + "epoch": 2.661363636363636, + "grad_norm": 0.5077105164527893, + "learning_rate": 3.4753783894293886e-06, + "loss": 0.4732, + "step": 389 + }, + { + "epoch": 2.668181818181818, + "grad_norm": 0.5006387829780579, + "learning_rate": 3.466051401903162e-06, + "loss": 0.4743, + "step": 390 + }, + { + "epoch": 2.675, + "grad_norm": 0.4769401252269745, + "learning_rate": 3.4567085809127247e-06, + "loss": 0.4941, + "step": 391 + }, + { + "epoch": 2.6818181818181817, + "grad_norm": 0.4538915455341339, + "learning_rate": 3.4473500795857674e-06, + "loss": 0.4679, + "step": 392 + }, + { + "epoch": 2.6886363636363635, + "grad_norm": 0.4370926320552826, + "learning_rate": 3.4379760513069804e-06, + "loss": 0.4866, + "step": 393 + }, + { + "epoch": 2.6954545454545453, + "grad_norm": 0.445913165807724, + "learning_rate": 3.428586649715542e-06, + "loss": 0.4914, + "step": 394 + }, + { + "epoch": 2.702272727272727, + "grad_norm": 0.45687681436538696, + "learning_rate": 3.4191820287025916e-06, + "loss": 0.474, + "step": 395 + }, + { + "epoch": 2.709090909090909, + "grad_norm": 1.1952316761016846, + "learning_rate": 3.4097623424087196e-06, + "loss": 0.4905, + "step": 396 + }, + { + "epoch": 2.715909090909091, + "grad_norm": 0.41248032450675964, + "learning_rate": 3.4003277452214284e-06, + "loss": 0.4753, + "step": 397 + }, + { + "epoch": 2.7227272727272727, + "grad_norm": 0.4349999725818634, + "learning_rate": 3.3908783917726123e-06, + "loss": 0.4712, + "step": 398 + }, + { + "epoch": 2.7295454545454545, + "grad_norm": 0.42977845668792725, + "learning_rate": 3.381414436936018e-06, + "loss": 0.4936, + "step": 399 + }, + { + "epoch": 2.7363636363636363, + "grad_norm": 0.446321040391922, + "learning_rate": 3.3719360358247054e-06, + "loss": 0.4795, + "step": 400 + }, + { + "epoch": 2.743181818181818, + "grad_norm": 0.4957369565963745, + "learning_rate": 3.36244334378851e-06, + "loss": 0.48, + "step": 401 + }, + { + "epoch": 2.75, + "grad_norm": 0.43430694937705994, + "learning_rate": 3.3529365164114903e-06, + "loss": 0.4903, + "step": 402 + }, + { + "epoch": 2.756818181818182, + "grad_norm": 0.4340098202228546, + "learning_rate": 3.3434157095093846e-06, + "loss": 0.5111, + "step": 403 + }, + { + "epoch": 2.7636363636363637, + "grad_norm": 0.4309055805206299, + "learning_rate": 3.333881079127052e-06, + "loss": 0.4909, + "step": 404 + }, + { + "epoch": 2.7704545454545455, + "grad_norm": 0.43131211400032043, + "learning_rate": 3.3243327815359168e-06, + "loss": 0.4875, + "step": 405 + }, + { + "epoch": 2.7772727272727273, + "grad_norm": 0.4794235825538635, + "learning_rate": 3.314770973231408e-06, + "loss": 0.4766, + "step": 406 + }, + { + "epoch": 2.784090909090909, + "grad_norm": 0.4632478356361389, + "learning_rate": 3.305195810930393e-06, + "loss": 0.4765, + "step": 407 + }, + { + "epoch": 2.790909090909091, + "grad_norm": 0.4836251139640808, + "learning_rate": 3.2956074515686105e-06, + "loss": 0.4573, + "step": 408 + }, + { + "epoch": 2.797727272727273, + "grad_norm": 0.44514402747154236, + "learning_rate": 3.2860060522980945e-06, + "loss": 0.4794, + "step": 409 + }, + { + "epoch": 2.8045454545454547, + "grad_norm": 0.446785569190979, + "learning_rate": 3.276391770484606e-06, + "loss": 0.482, + "step": 410 + }, + { + "epoch": 2.8113636363636365, + "grad_norm": 0.4410347044467926, + "learning_rate": 3.266764763705046e-06, + "loss": 0.5136, + "step": 411 + }, + { + "epoch": 2.8181818181818183, + "grad_norm": 0.4888826906681061, + "learning_rate": 3.257125189744877e-06, + "loss": 0.4852, + "step": 412 + }, + { + "epoch": 2.825, + "grad_norm": 0.4502141773700714, + "learning_rate": 3.247473206595536e-06, + "loss": 0.4764, + "step": 413 + }, + { + "epoch": 2.831818181818182, + "grad_norm": 0.5693864822387695, + "learning_rate": 3.2378089724518464e-06, + "loss": 0.5016, + "step": 414 + }, + { + "epoch": 2.838636363636364, + "grad_norm": 0.49588093161582947, + "learning_rate": 3.228132645709421e-06, + "loss": 0.4786, + "step": 415 + }, + { + "epoch": 2.8454545454545457, + "grad_norm": 0.5465879440307617, + "learning_rate": 3.218444384962071e-06, + "loss": 0.4857, + "step": 416 + }, + { + "epoch": 2.8522727272727275, + "grad_norm": 0.4633638858795166, + "learning_rate": 3.2087443489992043e-06, + "loss": 0.475, + "step": 417 + }, + { + "epoch": 2.8590909090909093, + "grad_norm": 0.47139137983322144, + "learning_rate": 3.1990326968032225e-06, + "loss": 0.5162, + "step": 418 + }, + { + "epoch": 2.865909090909091, + "grad_norm": 0.45505622029304504, + "learning_rate": 3.189309587546917e-06, + "loss": 0.4699, + "step": 419 + }, + { + "epoch": 2.8727272727272726, + "grad_norm": 0.4410839080810547, + "learning_rate": 3.1795751805908578e-06, + "loss": 0.4734, + "step": 420 + }, + { + "epoch": 2.8795454545454544, + "grad_norm": 0.47172659635543823, + "learning_rate": 3.169829635480783e-06, + "loss": 0.4896, + "step": 421 + }, + { + "epoch": 2.8863636363636362, + "grad_norm": 0.4332088828086853, + "learning_rate": 3.160073111944983e-06, + "loss": 0.4898, + "step": 422 + }, + { + "epoch": 2.893181818181818, + "grad_norm": 0.4303777813911438, + "learning_rate": 3.150305769891686e-06, + "loss": 0.4663, + "step": 423 + }, + { + "epoch": 2.9, + "grad_norm": 0.44356584548950195, + "learning_rate": 3.1405277694064306e-06, + "loss": 0.4516, + "step": 424 + }, + { + "epoch": 2.9068181818181817, + "grad_norm": 0.5283893942832947, + "learning_rate": 3.13073927074945e-06, + "loss": 0.4792, + "step": 425 + }, + { + "epoch": 2.9136363636363636, + "grad_norm": 0.90226149559021, + "learning_rate": 3.1209404343530374e-06, + "loss": 0.4811, + "step": 426 + }, + { + "epoch": 2.9204545454545454, + "grad_norm": 0.4658738970756531, + "learning_rate": 3.111131420818922e-06, + "loss": 0.4601, + "step": 427 + }, + { + "epoch": 2.9272727272727272, + "grad_norm": 0.4286805987358093, + "learning_rate": 3.1013123909156347e-06, + "loss": 0.4801, + "step": 428 + }, + { + "epoch": 2.934090909090909, + "grad_norm": 0.5181618332862854, + "learning_rate": 3.091483505575873e-06, + "loss": 0.4717, + "step": 429 + }, + { + "epoch": 2.940909090909091, + "grad_norm": 0.4391111135482788, + "learning_rate": 3.081644925893866e-06, + "loss": 0.4913, + "step": 430 + }, + { + "epoch": 2.9477272727272728, + "grad_norm": 0.4414638578891754, + "learning_rate": 3.0717968131227285e-06, + "loss": 0.4796, + "step": 431 + }, + { + "epoch": 2.9545454545454546, + "grad_norm": 0.4619010090827942, + "learning_rate": 3.061939328671824e-06, + "loss": 0.5059, + "step": 432 + }, + { + "epoch": 2.9613636363636364, + "grad_norm": 0.42799466848373413, + "learning_rate": 3.0520726341041165e-06, + "loss": 0.4598, + "step": 433 + }, + { + "epoch": 2.9681818181818183, + "grad_norm": 0.43869948387145996, + "learning_rate": 3.0421968911335196e-06, + "loss": 0.4733, + "step": 434 + }, + { + "epoch": 2.975, + "grad_norm": 0.552511990070343, + "learning_rate": 3.032312261622255e-06, + "loss": 0.4855, + "step": 435 + }, + { + "epoch": 2.981818181818182, + "grad_norm": 0.5187029242515564, + "learning_rate": 3.0224189075781886e-06, + "loss": 0.5067, + "step": 436 + }, + { + "epoch": 2.9886363636363638, + "grad_norm": 0.4445323944091797, + "learning_rate": 3.012516991152181e-06, + "loss": 0.4542, + "step": 437 + }, + { + "epoch": 2.9954545454545456, + "grad_norm": 0.5056352019309998, + "learning_rate": 3.002606674635432e-06, + "loss": 0.4907, + "step": 438 + }, + { + "epoch": 3.006818181818182, + "grad_norm": 0.4359378218650818, + "learning_rate": 2.9926881204568153e-06, + "loss": 0.9747, + "step": 439 + }, + { + "epoch": 3.0136363636363637, + "grad_norm": 0.41525086760520935, + "learning_rate": 2.9827614911802205e-06, + "loss": 0.4672, + "step": 440 + }, + { + "epoch": 3.0204545454545455, + "grad_norm": 0.45512208342552185, + "learning_rate": 2.972826949501884e-06, + "loss": 0.4607, + "step": 441 + }, + { + "epoch": 3.0272727272727273, + "grad_norm": 0.4749738574028015, + "learning_rate": 2.9628846582477305e-06, + "loss": 0.4471, + "step": 442 + }, + { + "epoch": 3.034090909090909, + "grad_norm": 0.42964112758636475, + "learning_rate": 2.9529347803706943e-06, + "loss": 0.4437, + "step": 443 + }, + { + "epoch": 3.040909090909091, + "grad_norm": 0.42203956842422485, + "learning_rate": 2.9429774789480576e-06, + "loss": 0.444, + "step": 444 + }, + { + "epoch": 3.047727272727273, + "grad_norm": 0.4311574697494507, + "learning_rate": 2.9330129171787704e-06, + "loss": 0.4942, + "step": 445 + }, + { + "epoch": 3.0545454545454547, + "grad_norm": 0.4946717917919159, + "learning_rate": 2.923041258380779e-06, + "loss": 0.4505, + "step": 446 + }, + { + "epoch": 3.0613636363636365, + "grad_norm": 0.45294976234436035, + "learning_rate": 2.9130626659883537e-06, + "loss": 0.4537, + "step": 447 + }, + { + "epoch": 3.0681818181818183, + "grad_norm": 0.4274294376373291, + "learning_rate": 2.9030773035493997e-06, + "loss": 0.4532, + "step": 448 + }, + { + "epoch": 3.075, + "grad_norm": 0.4929022789001465, + "learning_rate": 2.893085334722786e-06, + "loss": 0.4403, + "step": 449 + }, + { + "epoch": 3.081818181818182, + "grad_norm": 0.4446672797203064, + "learning_rate": 2.883086923275658e-06, + "loss": 0.4572, + "step": 450 + }, + { + "epoch": 3.088636363636364, + "grad_norm": 0.45900678634643555, + "learning_rate": 2.8730822330807556e-06, + "loss": 0.4435, + "step": 451 + }, + { + "epoch": 3.0954545454545457, + "grad_norm": 0.4808853566646576, + "learning_rate": 2.8630714281137263e-06, + "loss": 0.4668, + "step": 452 + }, + { + "epoch": 3.102272727272727, + "grad_norm": 0.460186630487442, + "learning_rate": 2.853054672450437e-06, + "loss": 0.4393, + "step": 453 + }, + { + "epoch": 3.109090909090909, + "grad_norm": 0.4464828670024872, + "learning_rate": 2.8430321302642887e-06, + "loss": 0.4374, + "step": 454 + }, + { + "epoch": 3.1159090909090907, + "grad_norm": 0.4616970717906952, + "learning_rate": 2.8330039658235194e-06, + "loss": 0.4508, + "step": 455 + }, + { + "epoch": 3.1227272727272726, + "grad_norm": 0.45340001583099365, + "learning_rate": 2.8229703434885165e-06, + "loss": 0.4488, + "step": 456 + }, + { + "epoch": 3.1295454545454544, + "grad_norm": 0.4715891480445862, + "learning_rate": 2.8129314277091224e-06, + "loss": 0.4642, + "step": 457 + }, + { + "epoch": 3.1363636363636362, + "grad_norm": 0.4530370533466339, + "learning_rate": 2.8028873830219373e-06, + "loss": 0.4502, + "step": 458 + }, + { + "epoch": 3.143181818181818, + "grad_norm": 0.4896863102912903, + "learning_rate": 2.7928383740476247e-06, + "loss": 0.4559, + "step": 459 + }, + { + "epoch": 3.15, + "grad_norm": 0.45334747433662415, + "learning_rate": 2.7827845654882112e-06, + "loss": 0.4462, + "step": 460 + }, + { + "epoch": 3.1568181818181817, + "grad_norm": 0.4454493820667267, + "learning_rate": 2.7727261221243875e-06, + "loss": 0.4726, + "step": 461 + }, + { + "epoch": 3.1636363636363636, + "grad_norm": 0.46723124384880066, + "learning_rate": 2.76266320881281e-06, + "loss": 0.476, + "step": 462 + }, + { + "epoch": 3.1704545454545454, + "grad_norm": 0.45423537492752075, + "learning_rate": 2.7525959904833955e-06, + "loss": 0.4269, + "step": 463 + }, + { + "epoch": 3.1772727272727272, + "grad_norm": 0.44072285294532776, + "learning_rate": 2.7425246321366205e-06, + "loss": 0.4556, + "step": 464 + }, + { + "epoch": 3.184090909090909, + "grad_norm": 0.44829070568084717, + "learning_rate": 2.7324492988408146e-06, + "loss": 0.4479, + "step": 465 + }, + { + "epoch": 3.190909090909091, + "grad_norm": 0.47372356057167053, + "learning_rate": 2.7223701557294574e-06, + "loss": 0.4302, + "step": 466 + }, + { + "epoch": 3.1977272727272728, + "grad_norm": 0.4265802800655365, + "learning_rate": 2.712287367998471e-06, + "loss": 0.4557, + "step": 467 + }, + { + "epoch": 3.2045454545454546, + "grad_norm": 0.5714879631996155, + "learning_rate": 2.702201100903511e-06, + "loss": 0.437, + "step": 468 + }, + { + "epoch": 3.2113636363636364, + "grad_norm": 0.4389351010322571, + "learning_rate": 2.692111519757261e-06, + "loss": 0.4371, + "step": 469 + }, + { + "epoch": 3.2181818181818183, + "grad_norm": 0.44851404428482056, + "learning_rate": 2.6820187899267203e-06, + "loss": 0.4711, + "step": 470 + }, + { + "epoch": 3.225, + "grad_norm": 0.5391034483909607, + "learning_rate": 2.671923076830496e-06, + "loss": 0.4698, + "step": 471 + }, + { + "epoch": 3.231818181818182, + "grad_norm": 0.5056214332580566, + "learning_rate": 2.6618245459360896e-06, + "loss": 0.4712, + "step": 472 + }, + { + "epoch": 3.2386363636363638, + "grad_norm": 0.47290515899658203, + "learning_rate": 2.651723362757186e-06, + "loss": 0.4515, + "step": 473 + }, + { + "epoch": 3.2454545454545456, + "grad_norm": 0.4664989411830902, + "learning_rate": 2.641619692850941e-06, + "loss": 0.4602, + "step": 474 + }, + { + "epoch": 3.2522727272727274, + "grad_norm": 0.5000811219215393, + "learning_rate": 2.631513701815267e-06, + "loss": 0.4521, + "step": 475 + }, + { + "epoch": 3.2590909090909093, + "grad_norm": 0.4550754427909851, + "learning_rate": 2.6214055552861213e-06, + "loss": 0.471, + "step": 476 + }, + { + "epoch": 3.265909090909091, + "grad_norm": 0.4234600365161896, + "learning_rate": 2.611295418934786e-06, + "loss": 0.4486, + "step": 477 + }, + { + "epoch": 3.2727272727272725, + "grad_norm": 0.49546584486961365, + "learning_rate": 2.6011834584651597e-06, + "loss": 0.4367, + "step": 478 + }, + { + "epoch": 3.2795454545454543, + "grad_norm": 0.42106184363365173, + "learning_rate": 2.591069839611036e-06, + "loss": 0.4366, + "step": 479 + }, + { + "epoch": 3.286363636363636, + "grad_norm": 0.4530644118785858, + "learning_rate": 2.5809547281333904e-06, + "loss": 0.4729, + "step": 480 + }, + { + "epoch": 3.293181818181818, + "grad_norm": 0.4612523019313812, + "learning_rate": 2.570838289817661e-06, + "loss": 0.4531, + "step": 481 + }, + { + "epoch": 3.3, + "grad_norm": 0.4771486222743988, + "learning_rate": 2.560720690471033e-06, + "loss": 0.4656, + "step": 482 + }, + { + "epoch": 3.3068181818181817, + "grad_norm": 0.45801112055778503, + "learning_rate": 2.5506020959197218e-06, + "loss": 0.4443, + "step": 483 + }, + { + "epoch": 3.3136363636363635, + "grad_norm": 0.4690552353858948, + "learning_rate": 2.5404826720062544e-06, + "loss": 0.4442, + "step": 484 + }, + { + "epoch": 3.3204545454545453, + "grad_norm": 0.4829693138599396, + "learning_rate": 2.5303625845867475e-06, + "loss": 0.4658, + "step": 485 + }, + { + "epoch": 3.327272727272727, + "grad_norm": 0.4420918822288513, + "learning_rate": 2.5202419995281966e-06, + "loss": 0.4486, + "step": 486 + }, + { + "epoch": 3.334090909090909, + "grad_norm": 0.4645867347717285, + "learning_rate": 2.5101210827057516e-06, + "loss": 0.4536, + "step": 487 + }, + { + "epoch": 3.340909090909091, + "grad_norm": 0.45904067158699036, + "learning_rate": 2.5e-06, + "loss": 0.4777, + "step": 488 + }, + { + "epoch": 3.3477272727272727, + "grad_norm": 0.47885990142822266, + "learning_rate": 2.4898789172942492e-06, + "loss": 0.4507, + "step": 489 + }, + { + "epoch": 3.3545454545454545, + "grad_norm": 0.48885077238082886, + "learning_rate": 2.4797580004718038e-06, + "loss": 0.4492, + "step": 490 + }, + { + "epoch": 3.3613636363636363, + "grad_norm": 0.43438634276390076, + "learning_rate": 2.4696374154132533e-06, + "loss": 0.4622, + "step": 491 + }, + { + "epoch": 3.368181818181818, + "grad_norm": 0.6276775598526001, + "learning_rate": 2.4595173279937464e-06, + "loss": 0.4488, + "step": 492 + }, + { + "epoch": 3.375, + "grad_norm": 0.5427472591400146, + "learning_rate": 2.4493979040802786e-06, + "loss": 0.4482, + "step": 493 + }, + { + "epoch": 3.381818181818182, + "grad_norm": 0.4438450336456299, + "learning_rate": 2.4392793095289677e-06, + "loss": 0.4434, + "step": 494 + }, + { + "epoch": 3.3886363636363637, + "grad_norm": 0.49523183703422546, + "learning_rate": 2.42916171018234e-06, + "loss": 0.4645, + "step": 495 + }, + { + "epoch": 3.3954545454545455, + "grad_norm": 0.4602353274822235, + "learning_rate": 2.419045271866611e-06, + "loss": 0.4569, + "step": 496 + }, + { + "epoch": 3.4022727272727273, + "grad_norm": 0.4849902391433716, + "learning_rate": 2.408930160388965e-06, + "loss": 0.4538, + "step": 497 + }, + { + "epoch": 3.409090909090909, + "grad_norm": 0.6242562532424927, + "learning_rate": 2.3988165415348416e-06, + "loss": 0.4509, + "step": 498 + }, + { + "epoch": 3.415909090909091, + "grad_norm": 0.45550772547721863, + "learning_rate": 2.388704581065215e-06, + "loss": 0.4396, + "step": 499 + }, + { + "epoch": 3.422727272727273, + "grad_norm": 0.428011953830719, + "learning_rate": 2.3785944447138804e-06, + "loss": 0.4506, + "step": 500 + }, + { + "epoch": 3.4295454545454547, + "grad_norm": 0.47410571575164795, + "learning_rate": 2.368486298184733e-06, + "loss": 0.4573, + "step": 501 + }, + { + "epoch": 3.4363636363636365, + "grad_norm": 0.45392513275146484, + "learning_rate": 2.358380307149059e-06, + "loss": 0.4607, + "step": 502 + }, + { + "epoch": 3.4431818181818183, + "grad_norm": 0.4263063669204712, + "learning_rate": 2.348276637242814e-06, + "loss": 0.4574, + "step": 503 + }, + { + "epoch": 3.45, + "grad_norm": 0.5994442701339722, + "learning_rate": 2.3381754540639108e-06, + "loss": 0.4483, + "step": 504 + }, + { + "epoch": 3.456818181818182, + "grad_norm": 0.45459091663360596, + "learning_rate": 2.328076923169504e-06, + "loss": 0.4445, + "step": 505 + }, + { + "epoch": 3.463636363636364, + "grad_norm": 0.48127374053001404, + "learning_rate": 2.31798121007328e-06, + "loss": 0.4455, + "step": 506 + }, + { + "epoch": 3.4704545454545457, + "grad_norm": 0.7549276947975159, + "learning_rate": 2.3078884802427394e-06, + "loss": 0.4475, + "step": 507 + }, + { + "epoch": 3.4772727272727275, + "grad_norm": 0.4759480953216553, + "learning_rate": 2.29779889909649e-06, + "loss": 0.4512, + "step": 508 + }, + { + "epoch": 3.484090909090909, + "grad_norm": 0.45603302121162415, + "learning_rate": 2.2877126320015295e-06, + "loss": 0.4441, + "step": 509 + }, + { + "epoch": 3.4909090909090907, + "grad_norm": 0.47070786356925964, + "learning_rate": 2.2776298442705434e-06, + "loss": 0.4402, + "step": 510 + }, + { + "epoch": 3.4977272727272726, + "grad_norm": 0.4690212905406952, + "learning_rate": 2.267550701159186e-06, + "loss": 0.4517, + "step": 511 + }, + { + "epoch": 3.5045454545454544, + "grad_norm": 0.8859854340553284, + "learning_rate": 2.25747536786338e-06, + "loss": 0.4374, + "step": 512 + }, + { + "epoch": 3.5113636363636362, + "grad_norm": 0.4920756220817566, + "learning_rate": 2.247404009516605e-06, + "loss": 0.446, + "step": 513 + }, + { + "epoch": 3.518181818181818, + "grad_norm": 0.49091637134552, + "learning_rate": 2.2373367911871904e-06, + "loss": 0.438, + "step": 514 + }, + { + "epoch": 3.525, + "grad_norm": 0.8683479428291321, + "learning_rate": 2.227273877875613e-06, + "loss": 0.4435, + "step": 515 + }, + { + "epoch": 3.5318181818181817, + "grad_norm": 0.4606627821922302, + "learning_rate": 2.2172154345117896e-06, + "loss": 0.4517, + "step": 516 + }, + { + "epoch": 3.5386363636363636, + "grad_norm": 0.48613184690475464, + "learning_rate": 2.207161625952376e-06, + "loss": 0.4449, + "step": 517 + }, + { + "epoch": 3.5454545454545454, + "grad_norm": 0.5018280148506165, + "learning_rate": 2.1971126169780636e-06, + "loss": 0.4414, + "step": 518 + }, + { + "epoch": 3.5522727272727272, + "grad_norm": 0.435921847820282, + "learning_rate": 2.1870685722908784e-06, + "loss": 0.4542, + "step": 519 + }, + { + "epoch": 3.559090909090909, + "grad_norm": 0.48534268140792847, + "learning_rate": 2.1770296565114847e-06, + "loss": 0.4373, + "step": 520 + }, + { + "epoch": 3.565909090909091, + "grad_norm": 0.49759384989738464, + "learning_rate": 2.166996034176482e-06, + "loss": 0.4518, + "step": 521 + }, + { + "epoch": 3.5727272727272728, + "grad_norm": 0.524696409702301, + "learning_rate": 2.1569678697357126e-06, + "loss": 0.4668, + "step": 522 + }, + { + "epoch": 3.5795454545454546, + "grad_norm": 0.4403073191642761, + "learning_rate": 2.1469453275495634e-06, + "loss": 0.44, + "step": 523 + }, + { + "epoch": 3.5863636363636364, + "grad_norm": 0.5214805006980896, + "learning_rate": 2.136928571886275e-06, + "loss": 0.437, + "step": 524 + }, + { + "epoch": 3.5931818181818183, + "grad_norm": 0.4685129225254059, + "learning_rate": 2.126917766919245e-06, + "loss": 0.4477, + "step": 525 + }, + { + "epoch": 3.6, + "grad_norm": 0.500602126121521, + "learning_rate": 2.1169130767243424e-06, + "loss": 0.455, + "step": 526 + }, + { + "epoch": 3.606818181818182, + "grad_norm": 0.5242429375648499, + "learning_rate": 2.1069146652772142e-06, + "loss": 0.4398, + "step": 527 + }, + { + "epoch": 3.6136363636363638, + "grad_norm": 0.6797722578048706, + "learning_rate": 2.0969226964506007e-06, + "loss": 0.4393, + "step": 528 + }, + { + "epoch": 3.6204545454545456, + "grad_norm": 0.44578859210014343, + "learning_rate": 2.0869373340116467e-06, + "loss": 0.458, + "step": 529 + }, + { + "epoch": 3.6272727272727274, + "grad_norm": 0.46465393900871277, + "learning_rate": 2.0769587416192212e-06, + "loss": 0.4461, + "step": 530 + }, + { + "epoch": 3.634090909090909, + "grad_norm": 0.48858389258384705, + "learning_rate": 2.066987082821231e-06, + "loss": 0.4623, + "step": 531 + }, + { + "epoch": 3.6409090909090907, + "grad_norm": 0.4606386423110962, + "learning_rate": 2.0570225210519433e-06, + "loss": 0.4538, + "step": 532 + }, + { + "epoch": 3.6477272727272725, + "grad_norm": 0.46266230940818787, + "learning_rate": 2.047065219629306e-06, + "loss": 0.4635, + "step": 533 + }, + { + "epoch": 3.6545454545454543, + "grad_norm": 0.5587124824523926, + "learning_rate": 2.0371153417522703e-06, + "loss": 0.4614, + "step": 534 + }, + { + "epoch": 3.661363636363636, + "grad_norm": 0.6143503785133362, + "learning_rate": 2.0271730504981165e-06, + "loss": 0.4408, + "step": 535 + }, + { + "epoch": 3.668181818181818, + "grad_norm": 0.5167360305786133, + "learning_rate": 2.0172385088197804e-06, + "loss": 0.4539, + "step": 536 + }, + { + "epoch": 3.675, + "grad_norm": 0.47937509417533875, + "learning_rate": 2.007311879543185e-06, + "loss": 0.4587, + "step": 537 + }, + { + "epoch": 3.6818181818181817, + "grad_norm": 0.4938436448574066, + "learning_rate": 1.9973933253645684e-06, + "loss": 0.4448, + "step": 538 + }, + { + "epoch": 3.6886363636363635, + "grad_norm": 0.4234243631362915, + "learning_rate": 1.9874830088478196e-06, + "loss": 0.4584, + "step": 539 + }, + { + "epoch": 3.6954545454545453, + "grad_norm": 0.5978733897209167, + "learning_rate": 1.9775810924218126e-06, + "loss": 0.4647, + "step": 540 + }, + { + "epoch": 3.702272727272727, + "grad_norm": 0.49155011773109436, + "learning_rate": 1.967687738377746e-06, + "loss": 0.4496, + "step": 541 + }, + { + "epoch": 3.709090909090909, + "grad_norm": 0.4360124468803406, + "learning_rate": 1.9578031088664812e-06, + "loss": 0.466, + "step": 542 + }, + { + "epoch": 3.715909090909091, + "grad_norm": 0.4867744743824005, + "learning_rate": 1.9479273658958852e-06, + "loss": 0.4362, + "step": 543 + }, + { + "epoch": 3.7227272727272727, + "grad_norm": 0.49178293347358704, + "learning_rate": 1.9380606713281773e-06, + "loss": 0.4501, + "step": 544 + }, + { + "epoch": 3.7295454545454545, + "grad_norm": 0.5098608136177063, + "learning_rate": 1.928203186877273e-06, + "loss": 0.4655, + "step": 545 + }, + { + "epoch": 3.7363636363636363, + "grad_norm": 0.4781307578086853, + "learning_rate": 1.9183550741061354e-06, + "loss": 0.4464, + "step": 546 + }, + { + "epoch": 3.743181818181818, + "grad_norm": 0.49883827567100525, + "learning_rate": 1.9085164944241275e-06, + "loss": 0.4366, + "step": 547 + }, + { + "epoch": 3.75, + "grad_norm": 0.5124484300613403, + "learning_rate": 1.8986876090843668e-06, + "loss": 0.4625, + "step": 548 + }, + { + "epoch": 3.756818181818182, + "grad_norm": 0.46270111203193665, + "learning_rate": 1.8888685791810784e-06, + "loss": 0.4149, + "step": 549 + }, + { + "epoch": 3.7636363636363637, + "grad_norm": 0.5264699459075928, + "learning_rate": 1.8790595656469628e-06, + "loss": 0.4373, + "step": 550 + }, + { + "epoch": 3.7704545454545455, + "grad_norm": 0.46891146898269653, + "learning_rate": 1.86926072925055e-06, + "loss": 0.4437, + "step": 551 + }, + { + "epoch": 3.7772727272727273, + "grad_norm": 0.4786033630371094, + "learning_rate": 1.8594722305935691e-06, + "loss": 0.4364, + "step": 552 + }, + { + "epoch": 3.784090909090909, + "grad_norm": 0.45475709438323975, + "learning_rate": 1.8496942301083142e-06, + "loss": 0.4515, + "step": 553 + }, + { + "epoch": 3.790909090909091, + "grad_norm": 0.4658668637275696, + "learning_rate": 1.8399268880550174e-06, + "loss": 0.4553, + "step": 554 + }, + { + "epoch": 3.797727272727273, + "grad_norm": 0.5587930679321289, + "learning_rate": 1.8301703645192178e-06, + "loss": 0.4412, + "step": 555 + }, + { + "epoch": 3.8045454545454547, + "grad_norm": 0.4532659947872162, + "learning_rate": 1.8204248194091429e-06, + "loss": 0.4529, + "step": 556 + }, + { + "epoch": 3.8113636363636365, + "grad_norm": 0.4606110751628876, + "learning_rate": 1.8106904124530839e-06, + "loss": 0.437, + "step": 557 + }, + { + "epoch": 3.8181818181818183, + "grad_norm": 0.4482369124889374, + "learning_rate": 1.800967303196778e-06, + "loss": 0.4309, + "step": 558 + }, + { + "epoch": 3.825, + "grad_norm": 0.533805787563324, + "learning_rate": 1.7912556510007967e-06, + "loss": 0.4393, + "step": 559 + }, + { + "epoch": 3.831818181818182, + "grad_norm": 0.5079528093338013, + "learning_rate": 1.7815556150379298e-06, + "loss": 0.4585, + "step": 560 + }, + { + "epoch": 3.838636363636364, + "grad_norm": 0.5496546626091003, + "learning_rate": 1.77186735429058e-06, + "loss": 0.4293, + "step": 561 + }, + { + "epoch": 3.8454545454545457, + "grad_norm": 0.49827462434768677, + "learning_rate": 1.7621910275481544e-06, + "loss": 0.4457, + "step": 562 + }, + { + "epoch": 3.8522727272727275, + "grad_norm": 0.5003693699836731, + "learning_rate": 1.7525267934044642e-06, + "loss": 0.4579, + "step": 563 + }, + { + "epoch": 3.8590909090909093, + "grad_norm": 0.47428634762763977, + "learning_rate": 1.7428748102551237e-06, + "loss": 0.452, + "step": 564 + }, + { + "epoch": 3.865909090909091, + "grad_norm": 0.4468045234680176, + "learning_rate": 1.7332352362949546e-06, + "loss": 0.4558, + "step": 565 + }, + { + "epoch": 3.8727272727272726, + "grad_norm": 0.4439336955547333, + "learning_rate": 1.7236082295153948e-06, + "loss": 0.4351, + "step": 566 + }, + { + "epoch": 3.8795454545454544, + "grad_norm": 0.46482470631599426, + "learning_rate": 1.7139939477019057e-06, + "loss": 0.4462, + "step": 567 + }, + { + "epoch": 3.8863636363636362, + "grad_norm": 1.1179648637771606, + "learning_rate": 1.7043925484313911e-06, + "loss": 0.4435, + "step": 568 + }, + { + "epoch": 3.893181818181818, + "grad_norm": 0.46164315938949585, + "learning_rate": 1.6948041890696076e-06, + "loss": 0.4446, + "step": 569 + }, + { + "epoch": 3.9, + "grad_norm": 0.4524129629135132, + "learning_rate": 1.685229026768593e-06, + "loss": 0.4567, + "step": 570 + }, + { + "epoch": 3.9068181818181817, + "grad_norm": 0.45722272992134094, + "learning_rate": 1.6756672184640847e-06, + "loss": 0.4368, + "step": 571 + }, + { + "epoch": 3.9136363636363636, + "grad_norm": 0.49532264471054077, + "learning_rate": 1.6661189208729492e-06, + "loss": 0.4572, + "step": 572 + }, + { + "epoch": 3.9204545454545454, + "grad_norm": 0.44752320647239685, + "learning_rate": 1.6565842904906154e-06, + "loss": 0.43, + "step": 573 + }, + { + "epoch": 3.9272727272727272, + "grad_norm": 0.5931842923164368, + "learning_rate": 1.6470634835885097e-06, + "loss": 0.4334, + "step": 574 + }, + { + "epoch": 3.934090909090909, + "grad_norm": 0.5873560905456543, + "learning_rate": 1.6375566562114903e-06, + "loss": 0.4374, + "step": 575 + }, + { + "epoch": 3.940909090909091, + "grad_norm": 0.4484315812587738, + "learning_rate": 1.6280639641752944e-06, + "loss": 0.4402, + "step": 576 + }, + { + "epoch": 3.9477272727272728, + "grad_norm": 0.4635726809501648, + "learning_rate": 1.6185855630639818e-06, + "loss": 0.4485, + "step": 577 + }, + { + "epoch": 3.9545454545454546, + "grad_norm": 0.7416269779205322, + "learning_rate": 1.6091216082273875e-06, + "loss": 0.4655, + "step": 578 + }, + { + "epoch": 3.9613636363636364, + "grad_norm": 0.4709416329860687, + "learning_rate": 1.5996722547785722e-06, + "loss": 0.4551, + "step": 579 + }, + { + "epoch": 3.9681818181818183, + "grad_norm": 0.45414504408836365, + "learning_rate": 1.5902376575912815e-06, + "loss": 0.4478, + "step": 580 + }, + { + "epoch": 3.975, + "grad_norm": 0.45831212401390076, + "learning_rate": 1.580817971297409e-06, + "loss": 0.462, + "step": 581 + }, + { + "epoch": 3.981818181818182, + "grad_norm": 0.4817596971988678, + "learning_rate": 1.5714133502844591e-06, + "loss": 0.4554, + "step": 582 + }, + { + "epoch": 3.9886363636363638, + "grad_norm": 0.48830825090408325, + "learning_rate": 1.56202394869302e-06, + "loss": 0.4382, + "step": 583 + }, + { + "epoch": 3.9954545454545456, + "grad_norm": 0.4884580075740814, + "learning_rate": 1.5526499204142332e-06, + "loss": 0.4527, + "step": 584 + }, + { + "epoch": 4.006818181818182, + "grad_norm": 1.0956987142562866, + "learning_rate": 1.5432914190872757e-06, + "loss": 0.9264, + "step": 585 + }, + { + "epoch": 4.013636363636364, + "grad_norm": 0.45760378241539, + "learning_rate": 1.5339485980968383e-06, + "loss": 0.4399, + "step": 586 + }, + { + "epoch": 4.0204545454545455, + "grad_norm": 0.44204893708229065, + "learning_rate": 1.5246216105706124e-06, + "loss": 0.4373, + "step": 587 + }, + { + "epoch": 4.027272727272727, + "grad_norm": 0.8188589215278625, + "learning_rate": 1.5153106093767827e-06, + "loss": 0.4356, + "step": 588 + }, + { + "epoch": 4.034090909090909, + "grad_norm": 0.4039295017719269, + "learning_rate": 1.506015747121518e-06, + "loss": 0.4215, + "step": 589 + }, + { + "epoch": 4.040909090909091, + "grad_norm": 0.4564852714538574, + "learning_rate": 1.4967371761464738e-06, + "loss": 0.4402, + "step": 590 + }, + { + "epoch": 4.047727272727273, + "grad_norm": 0.4643855094909668, + "learning_rate": 1.4874750485262917e-06, + "loss": 0.4374, + "step": 591 + }, + { + "epoch": 4.054545454545455, + "grad_norm": 0.561626672744751, + "learning_rate": 1.4782295160661103e-06, + "loss": 0.4323, + "step": 592 + }, + { + "epoch": 4.0613636363636365, + "grad_norm": 0.4655170440673828, + "learning_rate": 1.469000730299074e-06, + "loss": 0.4278, + "step": 593 + }, + { + "epoch": 4.068181818181818, + "grad_norm": 0.5915263891220093, + "learning_rate": 1.4597888424838519e-06, + "loss": 0.4208, + "step": 594 + }, + { + "epoch": 4.075, + "grad_norm": 0.44420668482780457, + "learning_rate": 1.450594003602158e-06, + "loss": 0.4424, + "step": 595 + }, + { + "epoch": 4.081818181818182, + "grad_norm": 0.49877864122390747, + "learning_rate": 1.4414163643562755e-06, + "loss": 0.4206, + "step": 596 + }, + { + "epoch": 4.088636363636364, + "grad_norm": 0.4505457580089569, + "learning_rate": 1.4322560751665873e-06, + "loss": 0.4345, + "step": 597 + }, + { + "epoch": 4.095454545454546, + "grad_norm": 0.47181791067123413, + "learning_rate": 1.4231132861691128e-06, + "loss": 0.4207, + "step": 598 + }, + { + "epoch": 4.1022727272727275, + "grad_norm": 0.4665491282939911, + "learning_rate": 1.4139881472130453e-06, + "loss": 0.4221, + "step": 599 + }, + { + "epoch": 4.109090909090909, + "grad_norm": 0.46456652879714966, + "learning_rate": 1.4048808078582943e-06, + "loss": 0.4162, + "step": 600 + }, + { + "epoch": 4.115909090909091, + "grad_norm": 0.437472939491272, + "learning_rate": 1.3957914173730366e-06, + "loss": 0.4379, + "step": 601 + }, + { + "epoch": 4.122727272727273, + "grad_norm": 0.5029657483100891, + "learning_rate": 1.3867201247312697e-06, + "loss": 0.4402, + "step": 602 + }, + { + "epoch": 4.129545454545455, + "grad_norm": 0.8025555610656738, + "learning_rate": 1.3776670786103685e-06, + "loss": 0.4449, + "step": 603 + }, + { + "epoch": 4.136363636363637, + "grad_norm": 0.5241179466247559, + "learning_rate": 1.3686324273886531e-06, + "loss": 0.4318, + "step": 604 + }, + { + "epoch": 4.1431818181818185, + "grad_norm": 0.4537895917892456, + "learning_rate": 1.359616319142949e-06, + "loss": 0.4354, + "step": 605 + }, + { + "epoch": 4.15, + "grad_norm": 0.5133662223815918, + "learning_rate": 1.3506189016461674e-06, + "loss": 0.4363, + "step": 606 + }, + { + "epoch": 4.156818181818182, + "grad_norm": 0.4720538854598999, + "learning_rate": 1.341640322364878e-06, + "loss": 0.4332, + "step": 607 + }, + { + "epoch": 4.163636363636364, + "grad_norm": 1.3719831705093384, + "learning_rate": 1.3326807284568984e-06, + "loss": 0.43, + "step": 608 + }, + { + "epoch": 4.170454545454546, + "grad_norm": 0.4677439332008362, + "learning_rate": 1.323740266768875e-06, + "loss": 0.4369, + "step": 609 + }, + { + "epoch": 4.177272727272728, + "grad_norm": 0.5197097659111023, + "learning_rate": 1.3148190838338804e-06, + "loss": 0.3977, + "step": 610 + }, + { + "epoch": 4.184090909090909, + "grad_norm": 0.8679460287094116, + "learning_rate": 1.3059173258690102e-06, + "loss": 0.4474, + "step": 611 + }, + { + "epoch": 4.190909090909091, + "grad_norm": 0.446285218000412, + "learning_rate": 1.2970351387729875e-06, + "loss": 0.438, + "step": 612 + }, + { + "epoch": 4.197727272727272, + "grad_norm": 0.48193076252937317, + "learning_rate": 1.2881726681237727e-06, + "loss": 0.4403, + "step": 613 + }, + { + "epoch": 4.204545454545454, + "grad_norm": 0.4694555997848511, + "learning_rate": 1.2793300591761742e-06, + "loss": 0.4022, + "step": 614 + }, + { + "epoch": 4.211363636363636, + "grad_norm": 0.46118268370628357, + "learning_rate": 1.27050745685947e-06, + "loss": 0.442, + "step": 615 + }, + { + "epoch": 4.218181818181818, + "grad_norm": 0.6086122989654541, + "learning_rate": 1.2617050057750322e-06, + "loss": 0.4272, + "step": 616 + }, + { + "epoch": 4.225, + "grad_norm": 0.4840640723705292, + "learning_rate": 1.252922850193955e-06, + "loss": 0.4263, + "step": 617 + }, + { + "epoch": 4.2318181818181815, + "grad_norm": 0.46273577213287354, + "learning_rate": 1.2441611340546958e-06, + "loss": 0.4254, + "step": 618 + }, + { + "epoch": 4.238636363636363, + "grad_norm": 0.501190721988678, + "learning_rate": 1.2354200009607081e-06, + "loss": 0.435, + "step": 619 + }, + { + "epoch": 4.245454545454545, + "grad_norm": 0.4813865125179291, + "learning_rate": 1.2266995941780934e-06, + "loss": 0.4267, + "step": 620 + }, + { + "epoch": 4.252272727272727, + "grad_norm": 0.5107240676879883, + "learning_rate": 1.2180000566332503e-06, + "loss": 0.3868, + "step": 621 + }, + { + "epoch": 4.259090909090909, + "grad_norm": 0.49328625202178955, + "learning_rate": 1.2093215309105352e-06, + "loss": 0.4228, + "step": 622 + }, + { + "epoch": 4.265909090909091, + "grad_norm": 0.7121614217758179, + "learning_rate": 1.2006641592499233e-06, + "loss": 0.4262, + "step": 623 + }, + { + "epoch": 4.2727272727272725, + "grad_norm": 0.5118347406387329, + "learning_rate": 1.192028083544675e-06, + "loss": 0.4228, + "step": 624 + }, + { + "epoch": 4.279545454545454, + "grad_norm": 0.45035621523857117, + "learning_rate": 1.1834134453390136e-06, + "loss": 0.4239, + "step": 625 + }, + { + "epoch": 4.286363636363636, + "grad_norm": 0.45246437191963196, + "learning_rate": 1.1748203858258056e-06, + "loss": 0.4268, + "step": 626 + }, + { + "epoch": 4.293181818181818, + "grad_norm": 0.48875924944877625, + "learning_rate": 1.166249045844243e-06, + "loss": 0.4227, + "step": 627 + }, + { + "epoch": 4.3, + "grad_norm": 0.4633196294307709, + "learning_rate": 1.1576995658775405e-06, + "loss": 0.4192, + "step": 628 + }, + { + "epoch": 4.306818181818182, + "grad_norm": 0.5269569158554077, + "learning_rate": 1.1491720860506273e-06, + "loss": 0.4212, + "step": 629 + }, + { + "epoch": 4.3136363636363635, + "grad_norm": 0.4277689754962921, + "learning_rate": 1.140666746127854e-06, + "loss": 0.4123, + "step": 630 + }, + { + "epoch": 4.320454545454545, + "grad_norm": 0.49047568440437317, + "learning_rate": 1.1321836855107007e-06, + "loss": 0.4128, + "step": 631 + }, + { + "epoch": 4.327272727272727, + "grad_norm": 0.45740997791290283, + "learning_rate": 1.1237230432354912e-06, + "loss": 0.4284, + "step": 632 + }, + { + "epoch": 4.334090909090909, + "grad_norm": 0.7805290818214417, + "learning_rate": 1.1152849579711187e-06, + "loss": 0.4487, + "step": 633 + }, + { + "epoch": 4.340909090909091, + "grad_norm": 0.6119694113731384, + "learning_rate": 1.1068695680167665e-06, + "loss": 0.4442, + "step": 634 + }, + { + "epoch": 4.347727272727273, + "grad_norm": 0.48440369963645935, + "learning_rate": 1.0984770112996463e-06, + "loss": 0.4223, + "step": 635 + }, + { + "epoch": 4.3545454545454545, + "grad_norm": 0.4387761056423187, + "learning_rate": 1.0901074253727338e-06, + "loss": 0.434, + "step": 636 + }, + { + "epoch": 4.361363636363636, + "grad_norm": 0.499102920293808, + "learning_rate": 1.0817609474125195e-06, + "loss": 0.427, + "step": 637 + }, + { + "epoch": 4.368181818181818, + "grad_norm": 0.541165292263031, + "learning_rate": 1.0734377142167549e-06, + "loss": 0.4236, + "step": 638 + }, + { + "epoch": 4.375, + "grad_norm": 0.5833367705345154, + "learning_rate": 1.065137862202213e-06, + "loss": 0.4171, + "step": 639 + }, + { + "epoch": 4.381818181818182, + "grad_norm": 0.45729443430900574, + "learning_rate": 1.0568615274024521e-06, + "loss": 0.4241, + "step": 640 + }, + { + "epoch": 4.388636363636364, + "grad_norm": 0.4729260206222534, + "learning_rate": 1.0486088454655856e-06, + "loss": 0.4164, + "step": 641 + }, + { + "epoch": 4.3954545454545455, + "grad_norm": 0.5857828855514526, + "learning_rate": 1.0403799516520619e-06, + "loss": 0.4219, + "step": 642 + }, + { + "epoch": 4.402272727272727, + "grad_norm": 0.5247046947479248, + "learning_rate": 1.0321749808324425e-06, + "loss": 0.4138, + "step": 643 + }, + { + "epoch": 4.409090909090909, + "grad_norm": 0.5515534281730652, + "learning_rate": 1.0239940674851943e-06, + "loss": 0.4613, + "step": 644 + }, + { + "epoch": 4.415909090909091, + "grad_norm": 0.4807690680027008, + "learning_rate": 1.0158373456944856e-06, + "loss": 0.4146, + "step": 645 + }, + { + "epoch": 4.422727272727273, + "grad_norm": 0.5009557604789734, + "learning_rate": 1.0077049491479874e-06, + "loss": 0.4031, + "step": 646 + }, + { + "epoch": 4.429545454545455, + "grad_norm": 0.4809257686138153, + "learning_rate": 9.995970111346842e-07, + "loss": 0.4293, + "step": 647 + }, + { + "epoch": 4.4363636363636365, + "grad_norm": 0.4761544466018677, + "learning_rate": 9.915136645426885e-07, + "loss": 0.428, + "step": 648 + }, + { + "epoch": 4.443181818181818, + "grad_norm": 0.44973552227020264, + "learning_rate": 9.834550418570602e-07, + "loss": 0.4346, + "step": 649 + }, + { + "epoch": 4.45, + "grad_norm": 0.4652273654937744, + "learning_rate": 9.754212751576386e-07, + "loss": 0.4255, + "step": 650 + }, + { + "epoch": 4.456818181818182, + "grad_norm": 0.5163391828536987, + "learning_rate": 9.67412496116876e-07, + "loss": 0.4008, + "step": 651 + }, + { + "epoch": 4.463636363636364, + "grad_norm": 0.525343120098114, + "learning_rate": 9.594288359976817e-07, + "loss": 0.4233, + "step": 652 + }, + { + "epoch": 4.470454545454546, + "grad_norm": 0.4707981050014496, + "learning_rate": 9.514704256512669e-07, + "loss": 0.4473, + "step": 653 + }, + { + "epoch": 4.4772727272727275, + "grad_norm": 0.4338839650154114, + "learning_rate": 9.435373955150032e-07, + "loss": 0.4216, + "step": 654 + }, + { + "epoch": 4.484090909090909, + "grad_norm": 0.4526594281196594, + "learning_rate": 9.35629875610283e-07, + "loss": 0.4307, + "step": 655 + }, + { + "epoch": 4.490909090909091, + "grad_norm": 0.5091291069984436, + "learning_rate": 9.277479955403887e-07, + "loss": 0.4449, + "step": 656 + }, + { + "epoch": 4.497727272727273, + "grad_norm": 0.4887843430042267, + "learning_rate": 9.198918844883714e-07, + "loss": 0.434, + "step": 657 + }, + { + "epoch": 4.504545454545455, + "grad_norm": 0.46774518489837646, + "learning_rate": 9.120616712149291e-07, + "loss": 0.426, + "step": 658 + }, + { + "epoch": 4.511363636363637, + "grad_norm": 0.47825315594673157, + "learning_rate": 9.042574840562982e-07, + "loss": 0.4372, + "step": 659 + }, + { + "epoch": 4.5181818181818185, + "grad_norm": 0.46589818596839905, + "learning_rate": 8.964794509221508e-07, + "loss": 0.4378, + "step": 660 + }, + { + "epoch": 4.525, + "grad_norm": 0.4720231592655182, + "learning_rate": 8.887276992934976e-07, + "loss": 0.4348, + "step": 661 + }, + { + "epoch": 4.531818181818182, + "grad_norm": 0.47262752056121826, + "learning_rate": 8.810023562206e-07, + "loss": 0.4195, + "step": 662 + }, + { + "epoch": 4.538636363636364, + "grad_norm": 0.975907564163208, + "learning_rate": 8.733035483208841e-07, + "loss": 0.4382, + "step": 663 + }, + { + "epoch": 4.545454545454545, + "grad_norm": 0.47100895643234253, + "learning_rate": 8.656314017768694e-07, + "loss": 0.4213, + "step": 664 + }, + { + "epoch": 4.552272727272728, + "grad_norm": 0.517842710018158, + "learning_rate": 8.579860423340977e-07, + "loss": 0.432, + "step": 665 + }, + { + "epoch": 4.559090909090909, + "grad_norm": 0.4891679286956787, + "learning_rate": 8.503675952990756e-07, + "loss": 0.442, + "step": 666 + }, + { + "epoch": 4.565909090909091, + "grad_norm": 0.48658767342567444, + "learning_rate": 8.427761855372169e-07, + "loss": 0.4255, + "step": 667 + }, + { + "epoch": 4.572727272727272, + "grad_norm": 0.4577118158340454, + "learning_rate": 8.352119374707979e-07, + "loss": 0.4199, + "step": 668 + }, + { + "epoch": 4.579545454545455, + "grad_norm": 0.6099198460578918, + "learning_rate": 8.276749750769186e-07, + "loss": 0.4255, + "step": 669 + }, + { + "epoch": 4.586363636363636, + "grad_norm": 0.5071508884429932, + "learning_rate": 8.20165421885469e-07, + "loss": 0.4384, + "step": 670 + }, + { + "epoch": 4.593181818181818, + "grad_norm": 0.5968758463859558, + "learning_rate": 8.126834009771079e-07, + "loss": 0.4247, + "step": 671 + }, + { + "epoch": 4.6, + "grad_norm": 0.4893784821033478, + "learning_rate": 8.052290349812419e-07, + "loss": 0.4368, + "step": 672 + }, + { + "epoch": 4.6068181818181815, + "grad_norm": 0.4897981584072113, + "learning_rate": 7.978024460740169e-07, + "loss": 0.4423, + "step": 673 + }, + { + "epoch": 4.613636363636363, + "grad_norm": 0.49245044589042664, + "learning_rate": 7.904037559763162e-07, + "loss": 0.4508, + "step": 674 + }, + { + "epoch": 4.620454545454545, + "grad_norm": 0.48101088404655457, + "learning_rate": 7.83033085951764e-07, + "loss": 0.4268, + "step": 675 + }, + { + "epoch": 4.627272727272727, + "grad_norm": 0.6497372388839722, + "learning_rate": 7.756905568047393e-07, + "loss": 0.4374, + "step": 676 + }, + { + "epoch": 4.634090909090909, + "grad_norm": 0.47810637950897217, + "learning_rate": 7.683762888783977e-07, + "loss": 0.4265, + "step": 677 + }, + { + "epoch": 4.640909090909091, + "grad_norm": 0.47082048654556274, + "learning_rate": 7.610904020526938e-07, + "loss": 0.4229, + "step": 678 + }, + { + "epoch": 4.6477272727272725, + "grad_norm": 0.4664064645767212, + "learning_rate": 7.538330157424212e-07, + "loss": 0.4418, + "step": 679 + }, + { + "epoch": 4.654545454545454, + "grad_norm": 0.44690611958503723, + "learning_rate": 7.466042488952521e-07, + "loss": 0.4288, + "step": 680 + }, + { + "epoch": 4.661363636363636, + "grad_norm": 0.4714190661907196, + "learning_rate": 7.394042199897916e-07, + "loss": 0.4476, + "step": 681 + }, + { + "epoch": 4.668181818181818, + "grad_norm": 0.49196168780326843, + "learning_rate": 7.322330470336314e-07, + "loss": 0.4224, + "step": 682 + }, + { + "epoch": 4.675, + "grad_norm": 0.5213456153869629, + "learning_rate": 7.250908475614185e-07, + "loss": 0.4044, + "step": 683 + }, + { + "epoch": 4.681818181818182, + "grad_norm": 0.4465913772583008, + "learning_rate": 7.179777386329276e-07, + "loss": 0.411, + "step": 684 + }, + { + "epoch": 4.6886363636363635, + "grad_norm": 0.49001193046569824, + "learning_rate": 7.108938368311424e-07, + "loss": 0.4556, + "step": 685 + }, + { + "epoch": 4.695454545454545, + "grad_norm": 0.4741291105747223, + "learning_rate": 7.038392582603481e-07, + "loss": 0.4537, + "step": 686 + }, + { + "epoch": 4.702272727272727, + "grad_norm": 0.48216235637664795, + "learning_rate": 6.968141185442229e-07, + "loss": 0.4187, + "step": 687 + }, + { + "epoch": 4.709090909090909, + "grad_norm": 0.45541542768478394, + "learning_rate": 6.898185328239468e-07, + "loss": 0.4094, + "step": 688 + }, + { + "epoch": 4.715909090909091, + "grad_norm": 0.4746590256690979, + "learning_rate": 6.828526157563126e-07, + "loss": 0.4269, + "step": 689 + }, + { + "epoch": 4.722727272727273, + "grad_norm": 0.45999908447265625, + "learning_rate": 6.759164815118493e-07, + "loss": 0.4325, + "step": 690 + }, + { + "epoch": 4.7295454545454545, + "grad_norm": 0.5204467177391052, + "learning_rate": 6.690102437729481e-07, + "loss": 0.4407, + "step": 691 + }, + { + "epoch": 4.736363636363636, + "grad_norm": 0.45999932289123535, + "learning_rate": 6.621340157319998e-07, + "loss": 0.4365, + "step": 692 + }, + { + "epoch": 4.743181818181818, + "grad_norm": 0.4780198931694031, + "learning_rate": 6.552879100895396e-07, + "loss": 0.4207, + "step": 693 + }, + { + "epoch": 4.75, + "grad_norm": 0.4917505979537964, + "learning_rate": 6.484720390524008e-07, + "loss": 0.4299, + "step": 694 + }, + { + "epoch": 4.756818181818182, + "grad_norm": 0.49415823817253113, + "learning_rate": 6.416865143318757e-07, + "loss": 0.4356, + "step": 695 + }, + { + "epoch": 4.763636363636364, + "grad_norm": 0.46500974893569946, + "learning_rate": 6.349314471418849e-07, + "loss": 0.4465, + "step": 696 + }, + { + "epoch": 4.7704545454545455, + "grad_norm": 0.4365505874156952, + "learning_rate": 6.282069481971514e-07, + "loss": 0.4205, + "step": 697 + }, + { + "epoch": 4.777272727272727, + "grad_norm": 0.4652706980705261, + "learning_rate": 6.2151312771139e-07, + "loss": 0.4385, + "step": 698 + }, + { + "epoch": 4.784090909090909, + "grad_norm": 0.4932256042957306, + "learning_rate": 6.148500953954992e-07, + "loss": 0.4056, + "step": 699 + }, + { + "epoch": 4.790909090909091, + "grad_norm": 0.8236297965049744, + "learning_rate": 6.082179604557617e-07, + "loss": 0.4147, + "step": 700 + }, + { + "epoch": 4.797727272727273, + "grad_norm": 0.6776428818702698, + "learning_rate": 6.016168315920593e-07, + "loss": 0.4111, + "step": 701 + }, + { + "epoch": 4.804545454545455, + "grad_norm": 0.4967111647129059, + "learning_rate": 5.950468169960846e-07, + "loss": 0.4467, + "step": 702 + }, + { + "epoch": 4.8113636363636365, + "grad_norm": 0.48729217052459717, + "learning_rate": 5.885080243495731e-07, + "loss": 0.4398, + "step": 703 + }, + { + "epoch": 4.818181818181818, + "grad_norm": 0.5257642865180969, + "learning_rate": 5.820005608225345e-07, + "loss": 0.4076, + "step": 704 + }, + { + "epoch": 4.825, + "grad_norm": 0.5564248561859131, + "learning_rate": 5.755245330715014e-07, + "loss": 0.4304, + "step": 705 + }, + { + "epoch": 4.831818181818182, + "grad_norm": 0.4611823260784149, + "learning_rate": 5.690800472377747e-07, + "loss": 0.4357, + "step": 706 + }, + { + "epoch": 4.838636363636364, + "grad_norm": 0.46857500076293945, + "learning_rate": 5.626672089456887e-07, + "loss": 0.4373, + "step": 707 + }, + { + "epoch": 4.845454545454546, + "grad_norm": 0.49044084548950195, + "learning_rate": 5.562861233008774e-07, + "loss": 0.4299, + "step": 708 + }, + { + "epoch": 4.8522727272727275, + "grad_norm": 0.4881190061569214, + "learning_rate": 5.499368948885528e-07, + "loss": 0.4306, + "step": 709 + }, + { + "epoch": 4.859090909090909, + "grad_norm": 0.4680595397949219, + "learning_rate": 5.436196277717928e-07, + "loss": 0.4145, + "step": 710 + }, + { + "epoch": 4.865909090909091, + "grad_norm": 0.45758605003356934, + "learning_rate": 5.373344254898313e-07, + "loss": 0.4294, + "step": 711 + }, + { + "epoch": 4.872727272727273, + "grad_norm": 0.4999527335166931, + "learning_rate": 5.310813910563645e-07, + "loss": 0.422, + "step": 712 + }, + { + "epoch": 4.879545454545455, + "grad_norm": 0.4778689444065094, + "learning_rate": 5.24860626957861e-07, + "loss": 0.4252, + "step": 713 + }, + { + "epoch": 4.886363636363637, + "grad_norm": 0.4799656569957733, + "learning_rate": 5.186722351518822e-07, + "loss": 0.4309, + "step": 714 + }, + { + "epoch": 4.8931818181818185, + "grad_norm": 0.4888315200805664, + "learning_rate": 5.125163170654138e-07, + "loss": 0.4123, + "step": 715 + }, + { + "epoch": 4.9, + "grad_norm": 0.49795007705688477, + "learning_rate": 5.063929735931985e-07, + "loss": 0.3995, + "step": 716 + }, + { + "epoch": 4.906818181818182, + "grad_norm": 0.5609537959098816, + "learning_rate": 5.003023050960865e-07, + "loss": 0.417, + "step": 717 + }, + { + "epoch": 4.913636363636364, + "grad_norm": 0.5369165539741516, + "learning_rate": 4.94244411399388e-07, + "loss": 0.4143, + "step": 718 + }, + { + "epoch": 4.920454545454545, + "grad_norm": 0.48569849133491516, + "learning_rate": 4.882193917912398e-07, + "loss": 0.4603, + "step": 719 + }, + { + "epoch": 4.927272727272728, + "grad_norm": 0.5146452188491821, + "learning_rate": 4.822273450209767e-07, + "loss": 0.406, + "step": 720 + }, + { + "epoch": 4.934090909090909, + "grad_norm": 0.4748017191886902, + "learning_rate": 4.7626836929751035e-07, + "loss": 0.4496, + "step": 721 + }, + { + "epoch": 4.940909090909091, + "grad_norm": 0.48057058453559875, + "learning_rate": 4.703425622877239e-07, + "loss": 0.4237, + "step": 722 + }, + { + "epoch": 4.947727272727272, + "grad_norm": 0.4811655580997467, + "learning_rate": 4.6445002111486866e-07, + "loss": 0.4429, + "step": 723 + }, + { + "epoch": 4.954545454545455, + "grad_norm": 0.45915886759757996, + "learning_rate": 4.5859084235697236e-07, + "loss": 0.4407, + "step": 724 + }, + { + "epoch": 4.961363636363636, + "grad_norm": 0.47130313515663147, + "learning_rate": 4.527651220452589e-07, + "loss": 0.4177, + "step": 725 + }, + { + "epoch": 4.968181818181818, + "grad_norm": 0.49092698097229004, + "learning_rate": 4.469729556625704e-07, + "loss": 0.4204, + "step": 726 + }, + { + "epoch": 4.975, + "grad_norm": 0.4968926012516022, + "learning_rate": 4.412144381418049e-07, + "loss": 0.4517, + "step": 727 + }, + { + "epoch": 4.9818181818181815, + "grad_norm": 0.4605294466018677, + "learning_rate": 4.354896638643591e-07, + "loss": 0.4377, + "step": 728 + }, + { + "epoch": 4.988636363636363, + "grad_norm": 0.5322148203849792, + "learning_rate": 4.2979872665858266e-07, + "loss": 0.4263, + "step": 729 + }, + { + "epoch": 4.995454545454545, + "grad_norm": 0.5127450823783875, + "learning_rate": 4.2414171979824e-07, + "loss": 0.4043, + "step": 730 + } + ], + "logging_steps": 1, + "max_steps": 876, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 146, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 4.704801364671201e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-730/training_args.bin b/checkpoint-730/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..b41c805059a36d8d2e3bf3511fb0fd95b74dae74 --- /dev/null +++ b/checkpoint-730/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68945b8f9934f5870e67a2d4a20c31fa8e742da41075fa7236fb313c6197456f +size 7928 diff --git a/checkpoint-730/zero_to_fp32.py b/checkpoint-730/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-730/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-876/README.md b/checkpoint-876/README.md new file mode 100644 index 0000000000000000000000000000000000000000..1b184114a0c28ed3e4c082c18486736dc818166d --- /dev/null +++ b/checkpoint-876/README.md @@ -0,0 +1,202 @@ +--- +base_model: meta-llama/Llama-3.3-70B-Instruct +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 0.15.0 \ No newline at end of file diff --git a/checkpoint-876/adapter_config.json b/checkpoint-876/adapter_config.json new file mode 100644 index 0000000000000000000000000000000000000000..a5775df151cb9649596940e8a117cb92f810dbb7 --- /dev/null +++ b/checkpoint-876/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "meta-llama/Llama-3.3-70B-Instruct", + "bias": "none", + "corda_config": null, + "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", + "down_proj", + "k_proj", + "q_proj", + "o_proj", + "v_proj", + "up_proj" + ], + "task_type": "CAUSAL_LM", + "trainable_token_indices": null, + "use_dora": false, + "use_rslora": false +} \ No newline at end of file diff --git a/checkpoint-876/adapter_model.safetensors b/checkpoint-876/adapter_model.safetensors new file mode 100644 index 0000000000000000000000000000000000000000..c12ae1d95b5e5d0e20dd71bde916917d11271cde --- /dev/null +++ b/checkpoint-876/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8c1233ae8b00040dd1b4d89ecad17fab0a716b076689a2c996ceda85f168b356 +size 10829849744 diff --git a/checkpoint-876/global_step877/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt b/checkpoint-876/global_step877/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..1481787635b3e2c235c9d464931e8afd2ff2f009 --- /dev/null +++ b/checkpoint-876/global_step877/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:ea82a52257bac04bc5ea7249a037f707b1786196c9a515f5c8666df7f9dfdabc +size 21659418140 diff --git a/checkpoint-876/global_step877/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt b/checkpoint-876/global_step877/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..b0051e99f40185234c84a646d85d28cbf3c7f2d3 --- /dev/null +++ b/checkpoint-876/global_step877/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68f6324297e7830f54e2deb23ba0b61b60a56283a3050e7c4e61d77ff3b50f47 +size 21659457372 diff --git a/checkpoint-876/global_step877/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt b/checkpoint-876/global_step877/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..53554a617c2f0f0c57c4af6c04bd8fb6df443929 --- /dev/null +++ b/checkpoint-876/global_step877/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:06428e513cfefb7b69e99717f66259affb054c05ac837cce9f75683e0929d391 +size 21659417820 diff --git a/checkpoint-876/global_step877/mp_rank_00_model_states.pt b/checkpoint-876/global_step877/mp_rank_00_model_states.pt new file mode 100644 index 0000000000000000000000000000000000000000..8cb9b3f501fceba92ed1f1805038c30a0cc61f03 --- /dev/null +++ b/checkpoint-876/global_step877/mp_rank_00_model_states.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:93bcf19b8b2c35032f91b7fd64c8add10ba3cd9cde4ed9c2a86f1a25392c18a0 +size 11918643933 diff --git a/checkpoint-876/latest b/checkpoint-876/latest new file mode 100644 index 0000000000000000000000000000000000000000..46a92d7bc9ba694d5bdfb259769b86d881b01f2f --- /dev/null +++ b/checkpoint-876/latest @@ -0,0 +1 @@ +global_step877 \ No newline at end of file diff --git a/checkpoint-876/rng_state_0.pth b/checkpoint-876/rng_state_0.pth new file mode 100644 index 0000000000000000000000000000000000000000..d5e73ff487bb99ae1fc29c342e999f1aaaea1f64 --- /dev/null +++ b/checkpoint-876/rng_state_0.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6aed1695117d67da4d78129c6d902b6cdd73704abdd04ac9c0a0b9df735d6316 +size 14768 diff --git a/checkpoint-876/rng_state_1.pth b/checkpoint-876/rng_state_1.pth new file mode 100644 index 0000000000000000000000000000000000000000..e71d379204f4ba0031dde399b1fd01db448a829a --- /dev/null +++ b/checkpoint-876/rng_state_1.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:97c9621f2c2b0048ace164d7f1ece6f1f6635c249f032eb995fdcea1fcea7790 +size 14768 diff --git a/checkpoint-876/rng_state_2.pth b/checkpoint-876/rng_state_2.pth new file mode 100644 index 0000000000000000000000000000000000000000..b8894dd5c8321259444a95493a27db3af11089eb --- /dev/null +++ b/checkpoint-876/rng_state_2.pth @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b53edfe6640d93b5d63a04adb42c11b9d04be91dd0432b8f3227956601724a56 +size 14768 diff --git a/checkpoint-876/scheduler.pt b/checkpoint-876/scheduler.pt new file mode 100644 index 0000000000000000000000000000000000000000..8881c93e92699c31473ce99b324703015ed20c19 --- /dev/null +++ b/checkpoint-876/scheduler.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:071d4e0b3293b75b61ab20008bc97e7ffeac51a8765dccc85ebc03bc2a386b81 +size 1064 diff --git a/checkpoint-876/special_tokens_map.json b/checkpoint-876/special_tokens_map.json new file mode 100644 index 0000000000000000000000000000000000000000..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /dev/null +++ b/checkpoint-876/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": "<|eot_id|>", + "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-876/tokenizer.json b/checkpoint-876/tokenizer.json new file mode 100644 index 0000000000000000000000000000000000000000..1c1d8d5c9024994f1d3b00f9662b8dd89ca13cf2 --- /dev/null +++ b/checkpoint-876/tokenizer.json @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b +size 17209920 diff --git a/checkpoint-876/tokenizer_config.json b/checkpoint-876/tokenizer_config.json new file mode 100644 index 0000000000000000000000000000000000000000..ca91a2ef55f4239a7af81d7c9abb05f53621a07b --- /dev/null +++ b/checkpoint-876/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "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|>", + "chat_template": "{{- bos_token }}\n{%- if custom_tools is defined %}\n {%- set tools = custom_tools %}\n{%- endif %}\n{%- if not tools_in_user_message is defined %}\n {%- set tools_in_user_message = true %}\n{%- endif %}\n{%- if not date_string is defined %}\n {%- set date_string = \"26 Jul 2024\" %}\n{%- endif %}\n{%- if not tools is defined %}\n {%- set tools = none %}\n{%- endif %}\n\n{#- This block extracts the system message, so we can slot it into the right place. #}\n{%- if messages[0]['role'] == 'system' %}\n {%- set system_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n{%- else %}\n {%- set system_message = \"\" %}\n{%- endif %}\n\n{#- System message + builtin tools #}\n{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}\n{%- if builtin_tools is defined or tools is not none %}\n {{- \"Environment: ipython\\n\" }}\n{%- endif %}\n{%- if builtin_tools is defined %}\n {{- \"Tools: \" + builtin_tools | reject('equalto', 'code_interpreter') | join(\", \") + \"\\n\\n\"}}\n{%- endif %}\n{{- \"Cutting Knowledge Date: December 2023\\n\" }}\n{{- \"Today Date: \" + date_string + \"\\n\\n\" }}\n{%- if tools is not none and not tools_in_user_message %}\n {{- \"You have access to the following functions. To call a function, please respond with JSON for a function call.\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n{%- endif %}\n{{- system_message }}\n{{- \"<|eot_id|>\" }}\n\n{#- Custom tools are passed in a user message with some extra guidance #}\n{%- if tools_in_user_message and not tools is none %}\n {#- Extract the first user message so we can plug it in here #}\n {%- if messages | length != 0 %}\n {%- set first_user_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n {%- else %}\n {{- raise_exception(\"Cannot put tools in the first user message when there's no first user message!\") }}\n{%- endif %}\n {{- '<|start_header_id|>user<|end_header_id|>\\n\\n' -}}\n {{- \"Given the following functions, please respond with a JSON for a function call \" }}\n {{- \"with its proper arguments that best answers the given prompt.\\n\\n\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n {{- first_user_message + \"<|eot_id|>\"}}\n{%- endif %}\n\n{%- for message in messages %}\n {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}\n {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n'+ message['content'] | trim + '<|eot_id|>' }}\n {%- elif 'tool_calls' in message %}\n {%- if not message.tool_calls|length == 1 %}\n {{- raise_exception(\"This model only supports single tool-calls at once!\") }}\n {%- endif %}\n {%- set tool_call = message.tool_calls[0].function %}\n {%- if builtin_tools is defined and tool_call.name in builtin_tools %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- \"<|python_tag|>\" + tool_call.name + \".call(\" }}\n {%- for arg_name, arg_val in tool_call.arguments | items %}\n {{- arg_name + '=\"' + arg_val + '\"' }}\n {%- if not loop.last %}\n {{- \", \" }}\n {%- endif %}\n {%- endfor %}\n {{- \")\" }}\n {%- else %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- '{\"name\": \"' + tool_call.name + '\", ' }}\n {{- '\"parameters\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- \"}\" }}\n {%- endif %}\n {%- if builtin_tools is defined %}\n {#- This means we're in ipython mode #}\n {{- \"<|eom_id|>\" }}\n {%- else %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n {%- elif message.role == \"tool\" or message.role == \"ipython\" %}\n {{- \"<|start_header_id|>ipython<|end_header_id|>\\n\\n\" }}\n {%- if message.content is mapping or message.content is iterable %}\n {{- message.content | tojson }}\n {%- else %}\n {{- message.content }}\n {%- endif %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}\n{%- endif %}\n", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "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-876/trainer_state.json b/checkpoint-876/trainer_state.json new file mode 100644 index 0000000000000000000000000000000000000000..fa1b883eecfb12197c97eb6c48316930a25a3dbd --- /dev/null +++ b/checkpoint-876/trainer_state.json @@ -0,0 +1,6165 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 5.995454545454545, + "eval_steps": 500, + "global_step": 876, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.006818181818181818, + "grad_norm": 1.8943578004837036, + "learning_rate": 5.0000000000000004e-08, + "loss": 0.8676, + "step": 1 + }, + { + "epoch": 0.013636363636363636, + "grad_norm": 7.775813102722168, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.8513, + "step": 2 + }, + { + "epoch": 0.020454545454545454, + "grad_norm": 1.768803596496582, + "learning_rate": 1.5000000000000002e-07, + "loss": 0.8339, + "step": 3 + }, + { + "epoch": 0.02727272727272727, + "grad_norm": 1.7032263278961182, + "learning_rate": 2.0000000000000002e-07, + "loss": 0.8348, + "step": 4 + }, + { + "epoch": 0.03409090909090909, + "grad_norm": 1.8244796991348267, + "learning_rate": 2.5000000000000004e-07, + "loss": 0.8687, + "step": 5 + }, + { + "epoch": 0.04090909090909091, + "grad_norm": 1.7780181169509888, + "learning_rate": 3.0000000000000004e-07, + "loss": 0.8113, + "step": 6 + }, + { + "epoch": 0.04772727272727273, + "grad_norm": 1.8362191915512085, + "learning_rate": 3.5000000000000004e-07, + "loss": 0.8463, + "step": 7 + }, + { + "epoch": 0.05454545454545454, + "grad_norm": 1.6640979051589966, + "learning_rate": 4.0000000000000003e-07, + "loss": 0.8336, + "step": 8 + }, + { + "epoch": 0.06136363636363636, + "grad_norm": 1.8013464212417603, + "learning_rate": 4.5000000000000003e-07, + "loss": 0.8572, + "step": 9 + }, + { + "epoch": 0.06818181818181818, + "grad_norm": 1.7237999439239502, + "learning_rate": 5.000000000000001e-07, + "loss": 0.8501, + "step": 10 + }, + { + "epoch": 0.075, + "grad_norm": 1.989346981048584, + "learning_rate": 5.5e-07, + "loss": 0.8615, + "step": 11 + }, + { + "epoch": 0.08181818181818182, + "grad_norm": 2.0312082767486572, + "learning_rate": 6.000000000000001e-07, + "loss": 0.858, + "step": 12 + }, + { + "epoch": 0.08863636363636364, + "grad_norm": 1.6921299695968628, + "learning_rate": 6.5e-07, + "loss": 0.8577, + "step": 13 + }, + { + "epoch": 0.09545454545454546, + "grad_norm": 2.4207160472869873, + "learning_rate": 7.000000000000001e-07, + "loss": 0.8564, + "step": 14 + }, + { + "epoch": 0.10227272727272728, + "grad_norm": 1.7572585344314575, + "learning_rate": 7.5e-07, + "loss": 0.8409, + "step": 15 + }, + { + "epoch": 0.10909090909090909, + "grad_norm": 1.6135950088500977, + "learning_rate": 8.000000000000001e-07, + "loss": 0.8086, + "step": 16 + }, + { + "epoch": 0.1159090909090909, + "grad_norm": 2.226036787033081, + "learning_rate": 8.500000000000001e-07, + "loss": 0.8106, + "step": 17 + }, + { + "epoch": 0.12272727272727273, + "grad_norm": 1.7164231538772583, + "learning_rate": 9.000000000000001e-07, + "loss": 0.8494, + "step": 18 + }, + { + "epoch": 0.12954545454545455, + "grad_norm": 1.6572000980377197, + "learning_rate": 9.500000000000001e-07, + "loss": 0.8023, + "step": 19 + }, + { + "epoch": 0.13636363636363635, + "grad_norm": 1.5672118663787842, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.787, + "step": 20 + }, + { + "epoch": 0.1431818181818182, + "grad_norm": 1.5080257654190063, + "learning_rate": 1.0500000000000001e-06, + "loss": 0.8105, + "step": 21 + }, + { + "epoch": 0.15, + "grad_norm": 1.484892725944519, + "learning_rate": 1.1e-06, + "loss": 0.8258, + "step": 22 + }, + { + "epoch": 0.15681818181818183, + "grad_norm": 1.4591134786605835, + "learning_rate": 1.1500000000000002e-06, + "loss": 0.8032, + "step": 23 + }, + { + "epoch": 0.16363636363636364, + "grad_norm": 1.5506278276443481, + "learning_rate": 1.2000000000000002e-06, + "loss": 0.8236, + "step": 24 + }, + { + "epoch": 0.17045454545454544, + "grad_norm": 1.639350414276123, + "learning_rate": 1.25e-06, + "loss": 0.8185, + "step": 25 + }, + { + "epoch": 0.17727272727272728, + "grad_norm": 1.592640995979309, + "learning_rate": 1.3e-06, + "loss": 0.7898, + "step": 26 + }, + { + "epoch": 0.18409090909090908, + "grad_norm": 1.5280641317367554, + "learning_rate": 1.3500000000000002e-06, + "loss": 0.7731, + "step": 27 + }, + { + "epoch": 0.19090909090909092, + "grad_norm": 1.4015443325042725, + "learning_rate": 1.4000000000000001e-06, + "loss": 0.7956, + "step": 28 + }, + { + "epoch": 0.19772727272727272, + "grad_norm": 1.4890056848526, + "learning_rate": 1.45e-06, + "loss": 0.7838, + "step": 29 + }, + { + "epoch": 0.20454545454545456, + "grad_norm": 2.4770750999450684, + "learning_rate": 1.5e-06, + "loss": 0.7766, + "step": 30 + }, + { + "epoch": 0.21136363636363636, + "grad_norm": 1.5863685607910156, + "learning_rate": 1.5500000000000002e-06, + "loss": 0.7353, + "step": 31 + }, + { + "epoch": 0.21818181818181817, + "grad_norm": 1.058114767074585, + "learning_rate": 1.6000000000000001e-06, + "loss": 0.7237, + "step": 32 + }, + { + "epoch": 0.225, + "grad_norm": 1.5869101285934448, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.7597, + "step": 33 + }, + { + "epoch": 0.2318181818181818, + "grad_norm": 1.0012413263320923, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.7388, + "step": 34 + }, + { + "epoch": 0.23863636363636365, + "grad_norm": 1.118056058883667, + "learning_rate": 1.75e-06, + "loss": 0.7786, + "step": 35 + }, + { + "epoch": 0.24545454545454545, + "grad_norm": 1.0648820400238037, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.7502, + "step": 36 + }, + { + "epoch": 0.25227272727272726, + "grad_norm": 0.9440382719039917, + "learning_rate": 1.85e-06, + "loss": 0.6942, + "step": 37 + }, + { + "epoch": 0.2590909090909091, + "grad_norm": 0.8300187587738037, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.7312, + "step": 38 + }, + { + "epoch": 0.26590909090909093, + "grad_norm": 0.8296191692352295, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.7461, + "step": 39 + }, + { + "epoch": 0.2727272727272727, + "grad_norm": 0.7797508239746094, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.7234, + "step": 40 + }, + { + "epoch": 0.27954545454545454, + "grad_norm": 0.7335019707679749, + "learning_rate": 2.05e-06, + "loss": 0.6927, + "step": 41 + }, + { + "epoch": 0.2863636363636364, + "grad_norm": 0.7062106728553772, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.6552, + "step": 42 + }, + { + "epoch": 0.29318181818181815, + "grad_norm": 1.2342087030410767, + "learning_rate": 2.15e-06, + "loss": 0.6959, + "step": 43 + }, + { + "epoch": 0.3, + "grad_norm": 0.8967597484588623, + "learning_rate": 2.2e-06, + "loss": 0.704, + "step": 44 + }, + { + "epoch": 0.3068181818181818, + "grad_norm": 0.7228404879570007, + "learning_rate": 2.25e-06, + "loss": 0.6777, + "step": 45 + }, + { + "epoch": 0.31363636363636366, + "grad_norm": 0.6731488704681396, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.6652, + "step": 46 + }, + { + "epoch": 0.32045454545454544, + "grad_norm": 0.71966952085495, + "learning_rate": 2.35e-06, + "loss": 0.7281, + "step": 47 + }, + { + "epoch": 0.32727272727272727, + "grad_norm": 0.7057356238365173, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.6476, + "step": 48 + }, + { + "epoch": 0.3340909090909091, + "grad_norm": 0.6346054077148438, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.6865, + "step": 49 + }, + { + "epoch": 0.3409090909090909, + "grad_norm": 0.6938223838806152, + "learning_rate": 2.5e-06, + "loss": 0.6808, + "step": 50 + }, + { + "epoch": 0.3477272727272727, + "grad_norm": 0.7425184845924377, + "learning_rate": 2.55e-06, + "loss": 0.7074, + "step": 51 + }, + { + "epoch": 0.35454545454545455, + "grad_norm": 0.6743818521499634, + "learning_rate": 2.6e-06, + "loss": 0.657, + "step": 52 + }, + { + "epoch": 0.3613636363636364, + "grad_norm": 0.6785942316055298, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.6768, + "step": 53 + }, + { + "epoch": 0.36818181818181817, + "grad_norm": 0.586974024772644, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.651, + "step": 54 + }, + { + "epoch": 0.375, + "grad_norm": 0.5727331042289734, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.669, + "step": 55 + }, + { + "epoch": 0.38181818181818183, + "grad_norm": 0.7333543300628662, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.6571, + "step": 56 + }, + { + "epoch": 0.3886363636363636, + "grad_norm": 0.5710961818695068, + "learning_rate": 2.85e-06, + "loss": 0.6475, + "step": 57 + }, + { + "epoch": 0.39545454545454545, + "grad_norm": 0.5614489912986755, + "learning_rate": 2.9e-06, + "loss": 0.6314, + "step": 58 + }, + { + "epoch": 0.4022727272727273, + "grad_norm": 0.5007341504096985, + "learning_rate": 2.95e-06, + "loss": 0.6346, + "step": 59 + }, + { + "epoch": 0.4090909090909091, + "grad_norm": 0.54221510887146, + "learning_rate": 3e-06, + "loss": 0.6715, + "step": 60 + }, + { + "epoch": 0.4159090909090909, + "grad_norm": 0.5177039504051208, + "learning_rate": 3.05e-06, + "loss": 0.6459, + "step": 61 + }, + { + "epoch": 0.42272727272727273, + "grad_norm": 0.4741189181804657, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.592, + "step": 62 + }, + { + "epoch": 0.42954545454545456, + "grad_norm": 0.5637328028678894, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.6459, + "step": 63 + }, + { + "epoch": 0.43636363636363634, + "grad_norm": 0.5478869676589966, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.6447, + "step": 64 + }, + { + "epoch": 0.4431818181818182, + "grad_norm": 0.5483130216598511, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.6319, + "step": 65 + }, + { + "epoch": 0.45, + "grad_norm": 0.478081613779068, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.6102, + "step": 66 + }, + { + "epoch": 0.45681818181818185, + "grad_norm": 0.4777645766735077, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.5986, + "step": 67 + }, + { + "epoch": 0.4636363636363636, + "grad_norm": 0.5291482210159302, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.6314, + "step": 68 + }, + { + "epoch": 0.47045454545454546, + "grad_norm": 0.5224051475524902, + "learning_rate": 3.45e-06, + "loss": 0.6278, + "step": 69 + }, + { + "epoch": 0.4772727272727273, + "grad_norm": 0.6474127173423767, + "learning_rate": 3.5e-06, + "loss": 0.6398, + "step": 70 + }, + { + "epoch": 0.48409090909090907, + "grad_norm": 0.5060178637504578, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.5936, + "step": 71 + }, + { + "epoch": 0.4909090909090909, + "grad_norm": 3.0769243240356445, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.5975, + "step": 72 + }, + { + "epoch": 0.49772727272727274, + "grad_norm": 0.5319749116897583, + "learning_rate": 3.65e-06, + "loss": 0.6119, + "step": 73 + }, + { + "epoch": 0.5045454545454545, + "grad_norm": 0.6745399236679077, + "learning_rate": 3.7e-06, + "loss": 0.6213, + "step": 74 + }, + { + "epoch": 0.5113636363636364, + "grad_norm": 0.5156731009483337, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.6262, + "step": 75 + }, + { + "epoch": 0.5181818181818182, + "grad_norm": 0.9138725399971008, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.6244, + "step": 76 + }, + { + "epoch": 0.525, + "grad_norm": 0.4362037777900696, + "learning_rate": 3.85e-06, + "loss": 0.5874, + "step": 77 + }, + { + "epoch": 0.5318181818181819, + "grad_norm": 0.4674088954925537, + "learning_rate": 3.900000000000001e-06, + "loss": 0.615, + "step": 78 + }, + { + "epoch": 0.5386363636363637, + "grad_norm": 0.4590819180011749, + "learning_rate": 3.95e-06, + "loss": 0.6003, + "step": 79 + }, + { + "epoch": 0.5454545454545454, + "grad_norm": 0.5145031809806824, + "learning_rate": 4.000000000000001e-06, + "loss": 0.6079, + "step": 80 + }, + { + "epoch": 0.5522727272727272, + "grad_norm": 0.46660348773002625, + "learning_rate": 4.05e-06, + "loss": 0.6174, + "step": 81 + }, + { + "epoch": 0.5590909090909091, + "grad_norm": 0.48053789138793945, + "learning_rate": 4.1e-06, + "loss": 0.6232, + "step": 82 + }, + { + "epoch": 0.5659090909090909, + "grad_norm": 0.540793776512146, + "learning_rate": 4.15e-06, + "loss": 0.619, + "step": 83 + }, + { + "epoch": 0.5727272727272728, + "grad_norm": 0.4925222396850586, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.5838, + "step": 84 + }, + { + "epoch": 0.5795454545454546, + "grad_norm": 0.44989290833473206, + "learning_rate": 4.25e-06, + "loss": 0.5972, + "step": 85 + }, + { + "epoch": 0.5863636363636363, + "grad_norm": 0.42691388726234436, + "learning_rate": 4.3e-06, + "loss": 0.6046, + "step": 86 + }, + { + "epoch": 0.5931818181818181, + "grad_norm": 0.4512398838996887, + "learning_rate": 4.350000000000001e-06, + "loss": 0.5828, + "step": 87 + }, + { + "epoch": 0.6, + "grad_norm": 0.44399499893188477, + "learning_rate": 4.4e-06, + "loss": 0.6143, + "step": 88 + }, + { + "epoch": 0.6068181818181818, + "grad_norm": 0.46058326959609985, + "learning_rate": 4.450000000000001e-06, + "loss": 0.6117, + "step": 89 + }, + { + "epoch": 0.6136363636363636, + "grad_norm": 0.8795785903930664, + "learning_rate": 4.5e-06, + "loss": 0.6, + "step": 90 + }, + { + "epoch": 0.6204545454545455, + "grad_norm": 0.5148798227310181, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.6041, + "step": 91 + }, + { + "epoch": 0.6272727272727273, + "grad_norm": 0.48829612135887146, + "learning_rate": 4.600000000000001e-06, + "loss": 0.5492, + "step": 92 + }, + { + "epoch": 0.634090909090909, + "grad_norm": 0.5211894512176514, + "learning_rate": 4.65e-06, + "loss": 0.6044, + "step": 93 + }, + { + "epoch": 0.6409090909090909, + "grad_norm": 0.44403275847435, + "learning_rate": 4.7e-06, + "loss": 0.5929, + "step": 94 + }, + { + "epoch": 0.6477272727272727, + "grad_norm": 0.6037693619728088, + "learning_rate": 4.75e-06, + "loss": 0.6132, + "step": 95 + }, + { + "epoch": 0.6545454545454545, + "grad_norm": 0.4381515085697174, + "learning_rate": 4.800000000000001e-06, + "loss": 0.5822, + "step": 96 + }, + { + "epoch": 0.6613636363636364, + "grad_norm": 0.4997427761554718, + "learning_rate": 4.85e-06, + "loss": 0.593, + "step": 97 + }, + { + "epoch": 0.6681818181818182, + "grad_norm": 0.440571665763855, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.6072, + "step": 98 + }, + { + "epoch": 0.675, + "grad_norm": 0.5020624995231628, + "learning_rate": 4.95e-06, + "loss": 0.5861, + "step": 99 + }, + { + "epoch": 0.6818181818181818, + "grad_norm": 0.4280160963535309, + "learning_rate": 5e-06, + "loss": 0.6163, + "step": 100 + }, + { + "epoch": 0.6886363636363636, + "grad_norm": 0.4584537148475647, + "learning_rate": 4.9999795126530275e-06, + "loss": 0.6177, + "step": 101 + }, + { + "epoch": 0.6954545454545454, + "grad_norm": 0.6835049390792847, + "learning_rate": 4.999918050947891e-06, + "loss": 0.579, + "step": 102 + }, + { + "epoch": 0.7022727272727273, + "grad_norm": 0.4551607072353363, + "learning_rate": 4.999815615891943e-06, + "loss": 0.5927, + "step": 103 + }, + { + "epoch": 0.7090909090909091, + "grad_norm": 0.5893972516059875, + "learning_rate": 4.9996722091640805e-06, + "loss": 0.5775, + "step": 104 + }, + { + "epoch": 0.7159090909090909, + "grad_norm": 0.4574092924594879, + "learning_rate": 4.9994878331147225e-06, + "loss": 0.5863, + "step": 105 + }, + { + "epoch": 0.7227272727272728, + "grad_norm": 0.5354658365249634, + "learning_rate": 4.99926249076577e-06, + "loss": 0.5453, + "step": 106 + }, + { + "epoch": 0.7295454545454545, + "grad_norm": 0.4347354471683502, + "learning_rate": 4.998996185810557e-06, + "loss": 0.5913, + "step": 107 + }, + { + "epoch": 0.7363636363636363, + "grad_norm": 0.4487966299057007, + "learning_rate": 4.998688922613788e-06, + "loss": 0.5749, + "step": 108 + }, + { + "epoch": 0.7431818181818182, + "grad_norm": 0.42477577924728394, + "learning_rate": 4.9983407062114695e-06, + "loss": 0.5761, + "step": 109 + }, + { + "epoch": 0.75, + "grad_norm": 0.4801446497440338, + "learning_rate": 4.9979515423108255e-06, + "loss": 0.5977, + "step": 110 + }, + { + "epoch": 0.7568181818181818, + "grad_norm": 0.4876883029937744, + "learning_rate": 4.997521437290205e-06, + "loss": 0.5849, + "step": 111 + }, + { + "epoch": 0.7636363636363637, + "grad_norm": 0.4328872263431549, + "learning_rate": 4.997050398198977e-06, + "loss": 0.5988, + "step": 112 + }, + { + "epoch": 0.7704545454545455, + "grad_norm": 0.4332719147205353, + "learning_rate": 4.996538432757414e-06, + "loss": 0.6059, + "step": 113 + }, + { + "epoch": 0.7772727272727272, + "grad_norm": 0.4455336630344391, + "learning_rate": 4.995985549356568e-06, + "loss": 0.5725, + "step": 114 + }, + { + "epoch": 0.7840909090909091, + "grad_norm": 0.4197766184806824, + "learning_rate": 4.995391757058129e-06, + "loss": 0.5715, + "step": 115 + }, + { + "epoch": 0.7909090909090909, + "grad_norm": 0.41945499181747437, + "learning_rate": 4.99475706559428e-06, + "loss": 0.5845, + "step": 116 + }, + { + "epoch": 0.7977272727272727, + "grad_norm": 0.42414823174476624, + "learning_rate": 4.994081485367537e-06, + "loss": 0.5595, + "step": 117 + }, + { + "epoch": 0.8045454545454546, + "grad_norm": 0.4056423008441925, + "learning_rate": 4.993365027450576e-06, + "loss": 0.582, + "step": 118 + }, + { + "epoch": 0.8113636363636364, + "grad_norm": 0.4554205536842346, + "learning_rate": 4.992607703586058e-06, + "loss": 0.5591, + "step": 119 + }, + { + "epoch": 0.8181818181818182, + "grad_norm": 0.4307089149951935, + "learning_rate": 4.991809526186424e-06, + "loss": 0.5927, + "step": 120 + }, + { + "epoch": 0.825, + "grad_norm": 0.48133528232574463, + "learning_rate": 4.990970508333707e-06, + "loss": 0.5703, + "step": 121 + }, + { + "epoch": 0.8318181818181818, + "grad_norm": 0.48582738637924194, + "learning_rate": 4.990090663779305e-06, + "loss": 0.5491, + "step": 122 + }, + { + "epoch": 0.8386363636363636, + "grad_norm": 0.44926196336746216, + "learning_rate": 4.9891700069437635e-06, + "loss": 0.5803, + "step": 123 + }, + { + "epoch": 0.8454545454545455, + "grad_norm": 0.4562165141105652, + "learning_rate": 4.988208552916535e-06, + "loss": 0.5945, + "step": 124 + }, + { + "epoch": 0.8522727272727273, + "grad_norm": 0.5885360836982727, + "learning_rate": 4.987206317455734e-06, + "loss": 0.5632, + "step": 125 + }, + { + "epoch": 0.8590909090909091, + "grad_norm": 0.4525550603866577, + "learning_rate": 4.986163316987877e-06, + "loss": 0.5221, + "step": 126 + }, + { + "epoch": 0.865909090909091, + "grad_norm": 0.4545478820800781, + "learning_rate": 4.985079568607613e-06, + "loss": 0.587, + "step": 127 + }, + { + "epoch": 0.8727272727272727, + "grad_norm": 0.4364396333694458, + "learning_rate": 4.983955090077445e-06, + "loss": 0.5626, + "step": 128 + }, + { + "epoch": 0.8795454545454545, + "grad_norm": 0.4354369640350342, + "learning_rate": 4.982789899827439e-06, + "loss": 0.5674, + "step": 129 + }, + { + "epoch": 0.8863636363636364, + "grad_norm": 0.4822773337364197, + "learning_rate": 4.9815840169549216e-06, + "loss": 0.5685, + "step": 130 + }, + { + "epoch": 0.8931818181818182, + "grad_norm": 0.44673025608062744, + "learning_rate": 4.980337461224164e-06, + "loss": 0.5733, + "step": 131 + }, + { + "epoch": 0.9, + "grad_norm": 0.6294501423835754, + "learning_rate": 4.979050253066064e-06, + "loss": 0.5396, + "step": 132 + }, + { + "epoch": 0.9068181818181819, + "grad_norm": 0.42890027165412903, + "learning_rate": 4.977722413577802e-06, + "loss": 0.5718, + "step": 133 + }, + { + "epoch": 0.9136363636363637, + "grad_norm": 0.470587819814682, + "learning_rate": 4.976353964522509e-06, + "loss": 0.5705, + "step": 134 + }, + { + "epoch": 0.9204545454545454, + "grad_norm": 0.4632768929004669, + "learning_rate": 4.974944928328894e-06, + "loss": 0.5394, + "step": 135 + }, + { + "epoch": 0.9272727272727272, + "grad_norm": 0.5304691195487976, + "learning_rate": 4.973495328090891e-06, + "loss": 0.5762, + "step": 136 + }, + { + "epoch": 0.9340909090909091, + "grad_norm": 0.42036354541778564, + "learning_rate": 4.972005187567267e-06, + "loss": 0.5649, + "step": 137 + }, + { + "epoch": 0.9409090909090909, + "grad_norm": 0.45189663767814636, + "learning_rate": 4.970474531181245e-06, + "loss": 0.5554, + "step": 138 + }, + { + "epoch": 0.9477272727272728, + "grad_norm": 0.47352468967437744, + "learning_rate": 4.968903384020095e-06, + "loss": 0.5511, + "step": 139 + }, + { + "epoch": 0.9545454545454546, + "grad_norm": 0.525702953338623, + "learning_rate": 4.967291771834727e-06, + "loss": 0.5777, + "step": 140 + }, + { + "epoch": 0.9613636363636363, + "grad_norm": 0.46146854758262634, + "learning_rate": 4.965639721039267e-06, + "loss": 0.5537, + "step": 141 + }, + { + "epoch": 0.9681818181818181, + "grad_norm": 0.5200941562652588, + "learning_rate": 4.963947258710626e-06, + "loss": 0.5522, + "step": 142 + }, + { + "epoch": 0.975, + "grad_norm": 0.5260375142097473, + "learning_rate": 4.962214412588053e-06, + "loss": 0.5747, + "step": 143 + }, + { + "epoch": 0.9818181818181818, + "grad_norm": 0.518645703792572, + "learning_rate": 4.960441211072686e-06, + "loss": 0.5257, + "step": 144 + }, + { + "epoch": 0.9886363636363636, + "grad_norm": 0.4528891444206238, + "learning_rate": 4.9586276832270785e-06, + "loss": 0.5548, + "step": 145 + }, + { + "epoch": 0.9954545454545455, + "grad_norm": 0.5137557983398438, + "learning_rate": 4.9567738587747314e-06, + "loss": 0.5526, + "step": 146 + }, + { + "epoch": 1.0068181818181818, + "grad_norm": 0.8773607611656189, + "learning_rate": 4.954879768099599e-06, + "loss": 1.1, + "step": 147 + }, + { + "epoch": 1.0136363636363637, + "grad_norm": 0.39079979062080383, + "learning_rate": 4.952945442245598e-06, + "loss": 0.568, + "step": 148 + }, + { + "epoch": 1.0204545454545455, + "grad_norm": 0.45004600286483765, + "learning_rate": 4.95097091291609e-06, + "loss": 0.5383, + "step": 149 + }, + { + "epoch": 1.0272727272727273, + "grad_norm": 0.41872867941856384, + "learning_rate": 4.948956212473371e-06, + "loss": 0.5436, + "step": 150 + }, + { + "epoch": 1.0340909090909092, + "grad_norm": 0.46829020977020264, + "learning_rate": 4.946901373938132e-06, + "loss": 0.5336, + "step": 151 + }, + { + "epoch": 1.040909090909091, + "grad_norm": 0.45038115978240967, + "learning_rate": 4.944806430988927e-06, + "loss": 0.538, + "step": 152 + }, + { + "epoch": 1.0477272727272728, + "grad_norm": 0.6496132016181946, + "learning_rate": 4.942671417961615e-06, + "loss": 0.5459, + "step": 153 + }, + { + "epoch": 1.0545454545454545, + "grad_norm": 0.4842703640460968, + "learning_rate": 4.940496369848795e-06, + "loss": 0.553, + "step": 154 + }, + { + "epoch": 1.0613636363636363, + "grad_norm": 0.4338674545288086, + "learning_rate": 4.938281322299243e-06, + "loss": 0.5366, + "step": 155 + }, + { + "epoch": 1.0681818181818181, + "grad_norm": 0.44630634784698486, + "learning_rate": 4.936026311617316e-06, + "loss": 0.5517, + "step": 156 + }, + { + "epoch": 1.075, + "grad_norm": 0.4606929421424866, + "learning_rate": 4.933731374762361e-06, + "loss": 0.5405, + "step": 157 + }, + { + "epoch": 1.0818181818181818, + "grad_norm": 0.48667654395103455, + "learning_rate": 4.931396549348115e-06, + "loss": 0.5376, + "step": 158 + }, + { + "epoch": 1.0886363636363636, + "grad_norm": 0.4516271650791168, + "learning_rate": 4.9290218736420795e-06, + "loss": 0.5206, + "step": 159 + }, + { + "epoch": 1.0954545454545455, + "grad_norm": 0.44719284772872925, + "learning_rate": 4.926607386564898e-06, + "loss": 0.5397, + "step": 160 + }, + { + "epoch": 1.1022727272727273, + "grad_norm": 0.40752628445625305, + "learning_rate": 4.9241531276897196e-06, + "loss": 0.5466, + "step": 161 + }, + { + "epoch": 1.1090909090909091, + "grad_norm": 0.45423609018325806, + "learning_rate": 4.921659137241544e-06, + "loss": 0.5227, + "step": 162 + }, + { + "epoch": 1.115909090909091, + "grad_norm": 0.4185822308063507, + "learning_rate": 4.919125456096574e-06, + "loss": 0.5144, + "step": 163 + }, + { + "epoch": 1.1227272727272728, + "grad_norm": 0.7753382921218872, + "learning_rate": 4.916552125781529e-06, + "loss": 0.5352, + "step": 164 + }, + { + "epoch": 1.1295454545454546, + "grad_norm": 0.47211477160453796, + "learning_rate": 4.913939188472979e-06, + "loss": 0.5505, + "step": 165 + }, + { + "epoch": 1.1363636363636362, + "grad_norm": 0.42742082476615906, + "learning_rate": 4.911286686996648e-06, + "loss": 0.529, + "step": 166 + }, + { + "epoch": 1.143181818181818, + "grad_norm": 0.41340839862823486, + "learning_rate": 4.908594664826708e-06, + "loss": 0.5292, + "step": 167 + }, + { + "epoch": 1.15, + "grad_norm": 0.44648951292037964, + "learning_rate": 4.905863166085076e-06, + "loss": 0.5757, + "step": 168 + }, + { + "epoch": 1.1568181818181817, + "grad_norm": 0.4285544753074646, + "learning_rate": 4.903092235540679e-06, + "loss": 0.5102, + "step": 169 + }, + { + "epoch": 1.1636363636363636, + "grad_norm": 0.42675015330314636, + "learning_rate": 4.900281918608732e-06, + "loss": 0.5397, + "step": 170 + }, + { + "epoch": 1.1704545454545454, + "grad_norm": 0.47636961936950684, + "learning_rate": 4.897432261349984e-06, + "loss": 0.5307, + "step": 171 + }, + { + "epoch": 1.1772727272727272, + "grad_norm": 0.4213427007198334, + "learning_rate": 4.894543310469968e-06, + "loss": 0.54, + "step": 172 + }, + { + "epoch": 1.184090909090909, + "grad_norm": 0.41153451800346375, + "learning_rate": 4.891615113318236e-06, + "loss": 0.5423, + "step": 173 + }, + { + "epoch": 1.190909090909091, + "grad_norm": 0.459185391664505, + "learning_rate": 4.888647717887582e-06, + "loss": 0.5239, + "step": 174 + }, + { + "epoch": 1.1977272727272728, + "grad_norm": 0.48279181122779846, + "learning_rate": 4.8856411728132526e-06, + "loss": 0.5336, + "step": 175 + }, + { + "epoch": 1.2045454545454546, + "grad_norm": 0.4186796247959137, + "learning_rate": 4.8825955273721524e-06, + "loss": 0.5509, + "step": 176 + }, + { + "epoch": 1.2113636363636364, + "grad_norm": 0.4345344007015228, + "learning_rate": 4.879510831482039e-06, + "loss": 0.5415, + "step": 177 + }, + { + "epoch": 1.2181818181818183, + "grad_norm": 0.461592435836792, + "learning_rate": 4.876387135700701e-06, + "loss": 0.5256, + "step": 178 + }, + { + "epoch": 1.225, + "grad_norm": 0.5960240960121155, + "learning_rate": 4.873224491225128e-06, + "loss": 0.5599, + "step": 179 + }, + { + "epoch": 1.231818181818182, + "grad_norm": 0.4662570357322693, + "learning_rate": 4.870022949890676e-06, + "loss": 0.5647, + "step": 180 + }, + { + "epoch": 1.2386363636363638, + "grad_norm": 0.4355057179927826, + "learning_rate": 4.866782564170217e-06, + "loss": 0.5056, + "step": 181 + }, + { + "epoch": 1.2454545454545454, + "grad_norm": 0.48699110746383667, + "learning_rate": 4.863503387173276e-06, + "loss": 0.5394, + "step": 182 + }, + { + "epoch": 1.2522727272727272, + "grad_norm": 0.48679423332214355, + "learning_rate": 4.860185472645161e-06, + "loss": 0.5394, + "step": 183 + }, + { + "epoch": 1.259090909090909, + "grad_norm": 0.6564393043518066, + "learning_rate": 4.856828874966086e-06, + "loss": 0.5259, + "step": 184 + }, + { + "epoch": 1.2659090909090909, + "grad_norm": 0.44762882590293884, + "learning_rate": 4.853433649150276e-06, + "loss": 0.5563, + "step": 185 + }, + { + "epoch": 1.2727272727272727, + "grad_norm": 0.5391108989715576, + "learning_rate": 4.849999850845066e-06, + "loss": 0.5326, + "step": 186 + }, + { + "epoch": 1.2795454545454545, + "grad_norm": 0.4201100170612335, + "learning_rate": 4.8465275363299905e-06, + "loss": 0.526, + "step": 187 + }, + { + "epoch": 1.2863636363636364, + "grad_norm": 0.46211543679237366, + "learning_rate": 4.84301676251586e-06, + "loss": 0.5498, + "step": 188 + }, + { + "epoch": 1.2931818181818182, + "grad_norm": 0.4325430989265442, + "learning_rate": 4.839467586943825e-06, + "loss": 0.556, + "step": 189 + }, + { + "epoch": 1.3, + "grad_norm": 0.46057143807411194, + "learning_rate": 4.835880067784441e-06, + "loss": 0.5079, + "step": 190 + }, + { + "epoch": 1.3068181818181819, + "grad_norm": 0.4920806288719177, + "learning_rate": 4.832254263836708e-06, + "loss": 0.533, + "step": 191 + }, + { + "epoch": 1.3136363636363637, + "grad_norm": 0.46693211793899536, + "learning_rate": 4.828590234527107e-06, + "loss": 0.5542, + "step": 192 + }, + { + "epoch": 1.3204545454545453, + "grad_norm": 0.4227873682975769, + "learning_rate": 4.82488803990863e-06, + "loss": 0.544, + "step": 193 + }, + { + "epoch": 1.3272727272727272, + "grad_norm": 0.44745922088623047, + "learning_rate": 4.821147740659795e-06, + "loss": 0.4962, + "step": 194 + }, + { + "epoch": 1.334090909090909, + "grad_norm": 0.46137240529060364, + "learning_rate": 4.817369398083648e-06, + "loss": 0.5359, + "step": 195 + }, + { + "epoch": 1.3409090909090908, + "grad_norm": 0.6483879685401917, + "learning_rate": 4.813553074106761e-06, + "loss": 0.5323, + "step": 196 + }, + { + "epoch": 1.3477272727272727, + "grad_norm": 0.6602574586868286, + "learning_rate": 4.809698831278217e-06, + "loss": 0.5161, + "step": 197 + }, + { + "epoch": 1.3545454545454545, + "grad_norm": 0.47758543491363525, + "learning_rate": 4.805806732768585e-06, + "loss": 0.5214, + "step": 198 + }, + { + "epoch": 1.3613636363636363, + "grad_norm": 0.4587438106536865, + "learning_rate": 4.801876842368882e-06, + "loss": 0.5208, + "step": 199 + }, + { + "epoch": 1.3681818181818182, + "grad_norm": 0.46055835485458374, + "learning_rate": 4.797909224489531e-06, + "loss": 0.5374, + "step": 200 + }, + { + "epoch": 1.375, + "grad_norm": 0.43727558851242065, + "learning_rate": 4.793903944159303e-06, + "loss": 0.5351, + "step": 201 + }, + { + "epoch": 1.3818181818181818, + "grad_norm": 0.4573572874069214, + "learning_rate": 4.789861067024253e-06, + "loss": 0.4997, + "step": 202 + }, + { + "epoch": 1.3886363636363637, + "grad_norm": 0.5921999216079712, + "learning_rate": 4.785780659346642e-06, + "loss": 0.5168, + "step": 203 + }, + { + "epoch": 1.3954545454545455, + "grad_norm": 0.47943904995918274, + "learning_rate": 4.781662788003851e-06, + "loss": 0.5513, + "step": 204 + }, + { + "epoch": 1.4022727272727273, + "grad_norm": 0.43733686208724976, + "learning_rate": 4.777507520487289e-06, + "loss": 0.5281, + "step": 205 + }, + { + "epoch": 1.4090909090909092, + "grad_norm": 0.4289107620716095, + "learning_rate": 4.773314924901281e-06, + "loss": 0.524, + "step": 206 + }, + { + "epoch": 1.415909090909091, + "grad_norm": 0.4203416109085083, + "learning_rate": 4.769085069961955e-06, + "loss": 0.5563, + "step": 207 + }, + { + "epoch": 1.4227272727272728, + "grad_norm": 0.4205201268196106, + "learning_rate": 4.764818024996117e-06, + "loss": 0.5202, + "step": 208 + }, + { + "epoch": 1.4295454545454547, + "grad_norm": 0.4296267032623291, + "learning_rate": 4.760513859940112e-06, + "loss": 0.5113, + "step": 209 + }, + { + "epoch": 1.4363636363636363, + "grad_norm": 0.3979087471961975, + "learning_rate": 4.756172645338675e-06, + "loss": 0.5313, + "step": 210 + }, + { + "epoch": 1.4431818181818181, + "grad_norm": 0.45232149958610535, + "learning_rate": 4.751794452343785e-06, + "loss": 0.5204, + "step": 211 + }, + { + "epoch": 1.45, + "grad_norm": 0.4309398829936981, + "learning_rate": 4.747379352713489e-06, + "loss": 0.5216, + "step": 212 + }, + { + "epoch": 1.4568181818181818, + "grad_norm": 0.45721709728240967, + "learning_rate": 4.7429274188107275e-06, + "loss": 0.5279, + "step": 213 + }, + { + "epoch": 1.4636363636363636, + "grad_norm": 0.46138322353363037, + "learning_rate": 4.738438723602154e-06, + "loss": 0.5287, + "step": 214 + }, + { + "epoch": 1.4704545454545455, + "grad_norm": 0.41752108931541443, + "learning_rate": 4.733913340656933e-06, + "loss": 0.5399, + "step": 215 + }, + { + "epoch": 1.4772727272727273, + "grad_norm": 0.45447197556495667, + "learning_rate": 4.729351344145536e-06, + "loss": 0.5277, + "step": 216 + }, + { + "epoch": 1.4840909090909091, + "grad_norm": 0.42524856328964233, + "learning_rate": 4.7247528088385296e-06, + "loss": 0.5103, + "step": 217 + }, + { + "epoch": 1.490909090909091, + "grad_norm": 0.46024221181869507, + "learning_rate": 4.720117810105341e-06, + "loss": 0.5172, + "step": 218 + }, + { + "epoch": 1.4977272727272728, + "grad_norm": 0.5868630409240723, + "learning_rate": 4.715446423913036e-06, + "loss": 0.5409, + "step": 219 + }, + { + "epoch": 1.5045454545454544, + "grad_norm": 0.424813836812973, + "learning_rate": 4.710738726825059e-06, + "loss": 0.5495, + "step": 220 + }, + { + "epoch": 1.5113636363636362, + "grad_norm": 0.49356943368911743, + "learning_rate": 4.705994795999991e-06, + "loss": 0.519, + "step": 221 + }, + { + "epoch": 1.518181818181818, + "grad_norm": 0.44291776418685913, + "learning_rate": 4.701214709190277e-06, + "loss": 0.5232, + "step": 222 + }, + { + "epoch": 1.525, + "grad_norm": 0.48171764612197876, + "learning_rate": 4.696398544740955e-06, + "loss": 0.5501, + "step": 223 + }, + { + "epoch": 1.5318181818181817, + "grad_norm": 0.4296175539493561, + "learning_rate": 4.69154638158837e-06, + "loss": 0.5087, + "step": 224 + }, + { + "epoch": 1.5386363636363636, + "grad_norm": 0.4715246260166168, + "learning_rate": 4.686658299258881e-06, + "loss": 0.5363, + "step": 225 + }, + { + "epoch": 1.5454545454545454, + "grad_norm": 0.44422832131385803, + "learning_rate": 4.681734377867562e-06, + "loss": 0.5371, + "step": 226 + }, + { + "epoch": 1.5522727272727272, + "grad_norm": 4.221654891967773, + "learning_rate": 4.67677469811688e-06, + "loss": 0.5191, + "step": 227 + }, + { + "epoch": 1.559090909090909, + "grad_norm": 0.41857993602752686, + "learning_rate": 4.671779341295378e-06, + "loss": 0.5191, + "step": 228 + }, + { + "epoch": 1.565909090909091, + "grad_norm": 0.4950161278247833, + "learning_rate": 4.666748389276344e-06, + "loss": 0.5311, + "step": 229 + }, + { + "epoch": 1.5727272727272728, + "grad_norm": 0.44641605019569397, + "learning_rate": 4.661681924516466e-06, + "loss": 0.5174, + "step": 230 + }, + { + "epoch": 1.5795454545454546, + "grad_norm": 0.39064037799835205, + "learning_rate": 4.6565800300544805e-06, + "loss": 0.515, + "step": 231 + }, + { + "epoch": 1.5863636363636364, + "grad_norm": 0.7764399647712708, + "learning_rate": 4.651442789509813e-06, + "loss": 0.5282, + "step": 232 + }, + { + "epoch": 1.5931818181818183, + "grad_norm": 0.44421035051345825, + "learning_rate": 4.646270287081208e-06, + "loss": 0.5324, + "step": 233 + }, + { + "epoch": 1.6, + "grad_norm": 0.5149928331375122, + "learning_rate": 4.641062607545347e-06, + "loss": 0.5264, + "step": 234 + }, + { + "epoch": 1.606818181818182, + "grad_norm": 0.4784093201160431, + "learning_rate": 4.6358198362554585e-06, + "loss": 0.552, + "step": 235 + }, + { + "epoch": 1.6136363636363638, + "grad_norm": 0.4138723313808441, + "learning_rate": 4.630542059139923e-06, + "loss": 0.5202, + "step": 236 + }, + { + "epoch": 1.6204545454545456, + "grad_norm": 0.4509594440460205, + "learning_rate": 4.625229362700863e-06, + "loss": 0.5318, + "step": 237 + }, + { + "epoch": 1.6272727272727274, + "grad_norm": 0.44979560375213623, + "learning_rate": 4.61988183401272e-06, + "loss": 0.521, + "step": 238 + }, + { + "epoch": 1.634090909090909, + "grad_norm": 0.4525575041770935, + "learning_rate": 4.614499560720837e-06, + "loss": 0.5073, + "step": 239 + }, + { + "epoch": 1.6409090909090909, + "grad_norm": 0.44126543402671814, + "learning_rate": 4.609082631040012e-06, + "loss": 0.5058, + "step": 240 + }, + { + "epoch": 1.6477272727272727, + "grad_norm": 0.887867271900177, + "learning_rate": 4.603631133753061e-06, + "loss": 0.5471, + "step": 241 + }, + { + "epoch": 1.6545454545454545, + "grad_norm": 0.4497375786304474, + "learning_rate": 4.598145158209356e-06, + "loss": 0.501, + "step": 242 + }, + { + "epoch": 1.6613636363636364, + "grad_norm": 0.45681032538414, + "learning_rate": 4.592624794323366e-06, + "loss": 0.5306, + "step": 243 + }, + { + "epoch": 1.6681818181818182, + "grad_norm": 0.5186678171157837, + "learning_rate": 4.587070132573178e-06, + "loss": 0.5097, + "step": 244 + }, + { + "epoch": 1.675, + "grad_norm": 0.44289669394493103, + "learning_rate": 4.581481263999019e-06, + "loss": 0.5085, + "step": 245 + }, + { + "epoch": 1.6818181818181817, + "grad_norm": 0.4571002125740051, + "learning_rate": 4.575858280201761e-06, + "loss": 0.5327, + "step": 246 + }, + { + "epoch": 1.6886363636363635, + "grad_norm": 0.43753257393836975, + "learning_rate": 4.570201273341418e-06, + "loss": 0.5167, + "step": 247 + }, + { + "epoch": 1.6954545454545453, + "grad_norm": 0.44994574785232544, + "learning_rate": 4.564510336135642e-06, + "loss": 0.5554, + "step": 248 + }, + { + "epoch": 1.7022727272727272, + "grad_norm": 0.6185874938964844, + "learning_rate": 4.558785561858196e-06, + "loss": 0.5115, + "step": 249 + }, + { + "epoch": 1.709090909090909, + "grad_norm": 0.4883944094181061, + "learning_rate": 4.5530270443374305e-06, + "loss": 0.5312, + "step": 250 + }, + { + "epoch": 1.7159090909090908, + "grad_norm": 0.46433499455451965, + "learning_rate": 4.547234877954741e-06, + "loss": 0.5368, + "step": 251 + }, + { + "epoch": 1.7227272727272727, + "grad_norm": 0.4270826578140259, + "learning_rate": 4.541409157643027e-06, + "loss": 0.5258, + "step": 252 + }, + { + "epoch": 1.7295454545454545, + "grad_norm": 0.4390241205692291, + "learning_rate": 4.535549978885132e-06, + "loss": 0.5259, + "step": 253 + }, + { + "epoch": 1.7363636363636363, + "grad_norm": 0.42756086587905884, + "learning_rate": 4.5296574377122765e-06, + "loss": 0.5217, + "step": 254 + }, + { + "epoch": 1.7431818181818182, + "grad_norm": 0.48124927282333374, + "learning_rate": 4.5237316307024895e-06, + "loss": 0.5095, + "step": 255 + }, + { + "epoch": 1.75, + "grad_norm": 0.4394672214984894, + "learning_rate": 4.517772654979024e-06, + "loss": 0.5116, + "step": 256 + }, + { + "epoch": 1.7568181818181818, + "grad_norm": 0.5805886387825012, + "learning_rate": 4.51178060820876e-06, + "loss": 0.5145, + "step": 257 + }, + { + "epoch": 1.7636363636363637, + "grad_norm": 0.4845556616783142, + "learning_rate": 4.505755588600613e-06, + "loss": 0.5064, + "step": 258 + }, + { + "epoch": 1.7704545454545455, + "grad_norm": 0.45300963521003723, + "learning_rate": 4.499697694903915e-06, + "loss": 0.5362, + "step": 259 + }, + { + "epoch": 1.7772727272727273, + "grad_norm": 0.45417436957359314, + "learning_rate": 4.493607026406802e-06, + "loss": 0.5234, + "step": 260 + }, + { + "epoch": 1.7840909090909092, + "grad_norm": 0.4577132761478424, + "learning_rate": 4.487483682934587e-06, + "loss": 0.5271, + "step": 261 + }, + { + "epoch": 1.790909090909091, + "grad_norm": 0.4518176019191742, + "learning_rate": 4.481327764848118e-06, + "loss": 0.4853, + "step": 262 + }, + { + "epoch": 1.7977272727272728, + "grad_norm": 0.43804875016212463, + "learning_rate": 4.47513937304214e-06, + "loss": 0.5188, + "step": 263 + }, + { + "epoch": 1.8045454545454547, + "grad_norm": 0.4401354193687439, + "learning_rate": 4.4689186089436365e-06, + "loss": 0.5079, + "step": 264 + }, + { + "epoch": 1.8113636363636365, + "grad_norm": 0.4051227569580078, + "learning_rate": 4.462665574510169e-06, + "loss": 0.5084, + "step": 265 + }, + { + "epoch": 1.8181818181818183, + "grad_norm": 0.5269004106521606, + "learning_rate": 4.456380372228208e-06, + "loss": 0.528, + "step": 266 + }, + { + "epoch": 1.825, + "grad_norm": 0.44371986389160156, + "learning_rate": 4.450063105111447e-06, + "loss": 0.5388, + "step": 267 + }, + { + "epoch": 1.8318181818181818, + "grad_norm": 0.41871729493141174, + "learning_rate": 4.443713876699124e-06, + "loss": 0.502, + "step": 268 + }, + { + "epoch": 1.8386363636363636, + "grad_norm": 0.464652419090271, + "learning_rate": 4.4373327910543125e-06, + "loss": 0.5045, + "step": 269 + }, + { + "epoch": 1.8454545454545455, + "grad_norm": 0.43198078870773315, + "learning_rate": 4.430919952762226e-06, + "loss": 0.5149, + "step": 270 + }, + { + "epoch": 1.8522727272727273, + "grad_norm": 0.42869722843170166, + "learning_rate": 4.424475466928499e-06, + "loss": 0.5329, + "step": 271 + }, + { + "epoch": 1.8590909090909091, + "grad_norm": 0.529680609703064, + "learning_rate": 4.417999439177465e-06, + "loss": 0.4939, + "step": 272 + }, + { + "epoch": 1.865909090909091, + "grad_norm": 0.4562613368034363, + "learning_rate": 4.4114919756504275e-06, + "loss": 0.5389, + "step": 273 + }, + { + "epoch": 1.8727272727272726, + "grad_norm": 0.4480995237827301, + "learning_rate": 4.404953183003916e-06, + "loss": 0.51, + "step": 274 + }, + { + "epoch": 1.8795454545454544, + "grad_norm": 0.41215088963508606, + "learning_rate": 4.398383168407941e-06, + "loss": 0.5386, + "step": 275 + }, + { + "epoch": 1.8863636363636362, + "grad_norm": 0.8426409363746643, + "learning_rate": 4.391782039544239e-06, + "loss": 0.498, + "step": 276 + }, + { + "epoch": 1.893181818181818, + "grad_norm": 0.5359936356544495, + "learning_rate": 4.385149904604502e-06, + "loss": 0.51, + "step": 277 + }, + { + "epoch": 1.9, + "grad_norm": 0.4462437033653259, + "learning_rate": 4.378486872288611e-06, + "loss": 0.5082, + "step": 278 + }, + { + "epoch": 1.9068181818181817, + "grad_norm": 0.445359468460083, + "learning_rate": 4.371793051802849e-06, + "loss": 0.5171, + "step": 279 + }, + { + "epoch": 1.9136363636363636, + "grad_norm": 0.4223516583442688, + "learning_rate": 4.365068552858116e-06, + "loss": 0.5218, + "step": 280 + }, + { + "epoch": 1.9204545454545454, + "grad_norm": 0.43929681181907654, + "learning_rate": 4.358313485668124e-06, + "loss": 0.5156, + "step": 281 + }, + { + "epoch": 1.9272727272727272, + "grad_norm": 0.42535650730133057, + "learning_rate": 4.3515279609476e-06, + "loss": 0.4964, + "step": 282 + }, + { + "epoch": 1.934090909090909, + "grad_norm": 0.46448391675949097, + "learning_rate": 4.3447120899104615e-06, + "loss": 0.5182, + "step": 283 + }, + { + "epoch": 1.940909090909091, + "grad_norm": 0.4082206189632416, + "learning_rate": 4.337865984268002e-06, + "loss": 0.5175, + "step": 284 + }, + { + "epoch": 1.9477272727272728, + "grad_norm": 0.42423564195632935, + "learning_rate": 4.3309897562270525e-06, + "loss": 0.5352, + "step": 285 + }, + { + "epoch": 1.9545454545454546, + "grad_norm": 0.5240079760551453, + "learning_rate": 4.324083518488151e-06, + "loss": 0.5157, + "step": 286 + }, + { + "epoch": 1.9613636363636364, + "grad_norm": 0.4517645835876465, + "learning_rate": 4.317147384243688e-06, + "loss": 0.5399, + "step": 287 + }, + { + "epoch": 1.9681818181818183, + "grad_norm": 0.44054290652275085, + "learning_rate": 4.3101814671760546e-06, + "loss": 0.5293, + "step": 288 + }, + { + "epoch": 1.975, + "grad_norm": 0.4256063401699066, + "learning_rate": 4.303185881455778e-06, + "loss": 0.5323, + "step": 289 + }, + { + "epoch": 1.981818181818182, + "grad_norm": 0.43106094002723694, + "learning_rate": 4.296160741739652e-06, + "loss": 0.5259, + "step": 290 + }, + { + "epoch": 1.9886363636363638, + "grad_norm": 0.43816477060317993, + "learning_rate": 4.289106163168858e-06, + "loss": 0.5098, + "step": 291 + }, + { + "epoch": 1.9954545454545456, + "grad_norm": 1.1177959442138672, + "learning_rate": 4.282022261367074e-06, + "loss": 0.513, + "step": 292 + }, + { + "epoch": 2.006818181818182, + "grad_norm": 0.8377058506011963, + "learning_rate": 4.274909152438582e-06, + "loss": 0.9984, + "step": 293 + }, + { + "epoch": 2.0136363636363637, + "grad_norm": 0.4339721202850342, + "learning_rate": 4.267766952966369e-06, + "loss": 0.493, + "step": 294 + }, + { + "epoch": 2.0204545454545455, + "grad_norm": 0.39961451292037964, + "learning_rate": 4.260595780010209e-06, + "loss": 0.486, + "step": 295 + }, + { + "epoch": 2.0272727272727273, + "grad_norm": 0.4809112548828125, + "learning_rate": 4.2533957511047485e-06, + "loss": 0.4803, + "step": 296 + }, + { + "epoch": 2.034090909090909, + "grad_norm": 0.414700984954834, + "learning_rate": 4.24616698425758e-06, + "loss": 0.4566, + "step": 297 + }, + { + "epoch": 2.040909090909091, + "grad_norm": 0.552201509475708, + "learning_rate": 4.238909597947307e-06, + "loss": 0.4775, + "step": 298 + }, + { + "epoch": 2.047727272727273, + "grad_norm": 0.4647877812385559, + "learning_rate": 4.231623711121603e-06, + "loss": 0.4944, + "step": 299 + }, + { + "epoch": 2.0545454545454547, + "grad_norm": 0.684251606464386, + "learning_rate": 4.224309443195261e-06, + "loss": 0.495, + "step": 300 + }, + { + "epoch": 2.0613636363636365, + "grad_norm": 0.42145106196403503, + "learning_rate": 4.2169669140482365e-06, + "loss": 0.4809, + "step": 301 + }, + { + "epoch": 2.0681818181818183, + "grad_norm": 0.43771296739578247, + "learning_rate": 4.2095962440236846e-06, + "loss": 0.5033, + "step": 302 + }, + { + "epoch": 2.075, + "grad_norm": 0.4323962926864624, + "learning_rate": 4.202197553925983e-06, + "loss": 0.4843, + "step": 303 + }, + { + "epoch": 2.081818181818182, + "grad_norm": 0.40169817209243774, + "learning_rate": 4.194770965018758e-06, + "loss": 0.504, + "step": 304 + }, + { + "epoch": 2.088636363636364, + "grad_norm": 0.4206427037715912, + "learning_rate": 4.187316599022892e-06, + "loss": 0.5059, + "step": 305 + }, + { + "epoch": 2.0954545454545457, + "grad_norm": 0.4550352692604065, + "learning_rate": 4.179834578114531e-06, + "loss": 0.4717, + "step": 306 + }, + { + "epoch": 2.102272727272727, + "grad_norm": 0.4193796217441559, + "learning_rate": 4.172325024923083e-06, + "loss": 0.5018, + "step": 307 + }, + { + "epoch": 2.109090909090909, + "grad_norm": 0.550195038318634, + "learning_rate": 4.164788062529203e-06, + "loss": 0.4961, + "step": 308 + }, + { + "epoch": 2.1159090909090907, + "grad_norm": 0.43539106845855713, + "learning_rate": 4.157223814462784e-06, + "loss": 0.4909, + "step": 309 + }, + { + "epoch": 2.1227272727272726, + "grad_norm": 0.4565756022930145, + "learning_rate": 4.149632404700925e-06, + "loss": 0.49, + "step": 310 + }, + { + "epoch": 2.1295454545454544, + "grad_norm": 0.4206918179988861, + "learning_rate": 4.142013957665903e-06, + "loss": 0.4722, + "step": 311 + }, + { + "epoch": 2.1363636363636362, + "grad_norm": 0.4098123013973236, + "learning_rate": 4.134368598223132e-06, + "loss": 0.4791, + "step": 312 + }, + { + "epoch": 2.143181818181818, + "grad_norm": 0.4387701153755188, + "learning_rate": 4.126696451679116e-06, + "loss": 0.4758, + "step": 313 + }, + { + "epoch": 2.15, + "grad_norm": 0.4594134986400604, + "learning_rate": 4.118997643779401e-06, + "loss": 0.5009, + "step": 314 + }, + { + "epoch": 2.1568181818181817, + "grad_norm": 0.4581888020038605, + "learning_rate": 4.111272300706502e-06, + "loss": 0.4915, + "step": 315 + }, + { + "epoch": 2.1636363636363636, + "grad_norm": 0.5279524326324463, + "learning_rate": 4.1035205490778505e-06, + "loss": 0.5056, + "step": 316 + }, + { + "epoch": 2.1704545454545454, + "grad_norm": 0.4253177344799042, + "learning_rate": 4.095742515943703e-06, + "loss": 0.5001, + "step": 317 + }, + { + "epoch": 2.1772727272727272, + "grad_norm": 0.5294811129570007, + "learning_rate": 4.087938328785071e-06, + "loss": 0.5027, + "step": 318 + }, + { + "epoch": 2.184090909090909, + "grad_norm": 0.4642890989780426, + "learning_rate": 4.080108115511629e-06, + "loss": 0.5051, + "step": 319 + }, + { + "epoch": 2.190909090909091, + "grad_norm": 0.43700429797172546, + "learning_rate": 4.072252004459612e-06, + "loss": 0.4809, + "step": 320 + }, + { + "epoch": 2.1977272727272728, + "grad_norm": 0.4536914825439453, + "learning_rate": 4.064370124389718e-06, + "loss": 0.5129, + "step": 321 + }, + { + "epoch": 2.2045454545454546, + "grad_norm": 0.42350664734840393, + "learning_rate": 4.056462604484998e-06, + "loss": 0.4912, + "step": 322 + }, + { + "epoch": 2.2113636363636364, + "grad_norm": 0.47886520624160767, + "learning_rate": 4.048529574348734e-06, + "loss": 0.4757, + "step": 323 + }, + { + "epoch": 2.2181818181818183, + "grad_norm": 0.5205736756324768, + "learning_rate": 4.040571164002319e-06, + "loss": 0.4821, + "step": 324 + }, + { + "epoch": 2.225, + "grad_norm": 0.4465322196483612, + "learning_rate": 4.032587503883124e-06, + "loss": 0.5057, + "step": 325 + }, + { + "epoch": 2.231818181818182, + "grad_norm": 0.41364315152168274, + "learning_rate": 4.0245787248423614e-06, + "loss": 0.4809, + "step": 326 + }, + { + "epoch": 2.2386363636363638, + "grad_norm": 0.42846816778182983, + "learning_rate": 4.0165449581429404e-06, + "loss": 0.4522, + "step": 327 + }, + { + "epoch": 2.2454545454545456, + "grad_norm": 0.4604497253894806, + "learning_rate": 4.008486335457312e-06, + "loss": 0.4879, + "step": 328 + }, + { + "epoch": 2.2522727272727274, + "grad_norm": 0.45866841077804565, + "learning_rate": 4.000402988865316e-06, + "loss": 0.4969, + "step": 329 + }, + { + "epoch": 2.2590909090909093, + "grad_norm": 0.5116038918495178, + "learning_rate": 3.992295050852013e-06, + "loss": 0.5015, + "step": 330 + }, + { + "epoch": 2.265909090909091, + "grad_norm": 0.4389069080352783, + "learning_rate": 3.984162654305516e-06, + "loss": 0.4797, + "step": 331 + }, + { + "epoch": 2.2727272727272725, + "grad_norm": 0.45724064111709595, + "learning_rate": 3.976005932514807e-06, + "loss": 0.4915, + "step": 332 + }, + { + "epoch": 2.2795454545454543, + "grad_norm": 0.44370219111442566, + "learning_rate": 3.967825019167559e-06, + "loss": 0.499, + "step": 333 + }, + { + "epoch": 2.286363636363636, + "grad_norm": 0.4414970278739929, + "learning_rate": 3.959620048347938e-06, + "loss": 0.447, + "step": 334 + }, + { + "epoch": 2.293181818181818, + "grad_norm": 0.7909289598464966, + "learning_rate": 3.951391154534415e-06, + "loss": 0.4658, + "step": 335 + }, + { + "epoch": 2.3, + "grad_norm": 0.4643534719944, + "learning_rate": 3.943138472597549e-06, + "loss": 0.4941, + "step": 336 + }, + { + "epoch": 2.3068181818181817, + "grad_norm": 0.5114858150482178, + "learning_rate": 3.934862137797788e-06, + "loss": 0.4733, + "step": 337 + }, + { + "epoch": 2.3136363636363635, + "grad_norm": 0.4057219624519348, + "learning_rate": 3.9265622857832455e-06, + "loss": 0.4965, + "step": 338 + }, + { + "epoch": 2.3204545454545453, + "grad_norm": 0.4124796390533447, + "learning_rate": 3.918239052587481e-06, + "loss": 0.4508, + "step": 339 + }, + { + "epoch": 2.327272727272727, + "grad_norm": 0.43895164132118225, + "learning_rate": 3.909892574627267e-06, + "loss": 0.4939, + "step": 340 + }, + { + "epoch": 2.334090909090909, + "grad_norm": 0.47555556893348694, + "learning_rate": 3.901522988700355e-06, + "loss": 0.4541, + "step": 341 + }, + { + "epoch": 2.340909090909091, + "grad_norm": 0.44449353218078613, + "learning_rate": 3.893130431983234e-06, + "loss": 0.4883, + "step": 342 + }, + { + "epoch": 2.3477272727272727, + "grad_norm": 0.441793829202652, + "learning_rate": 3.884715042028882e-06, + "loss": 0.4911, + "step": 343 + }, + { + "epoch": 2.3545454545454545, + "grad_norm": 0.4836761951446533, + "learning_rate": 3.876276956764509e-06, + "loss": 0.4838, + "step": 344 + }, + { + "epoch": 2.3613636363636363, + "grad_norm": 0.7246759533882141, + "learning_rate": 3.867816314489301e-06, + "loss": 0.5014, + "step": 345 + }, + { + "epoch": 2.368181818181818, + "grad_norm": 0.5535604357719421, + "learning_rate": 3.8593332538721465e-06, + "loss": 0.4654, + "step": 346 + }, + { + "epoch": 2.375, + "grad_norm": 0.4468318819999695, + "learning_rate": 3.8508279139493736e-06, + "loss": 0.463, + "step": 347 + }, + { + "epoch": 2.381818181818182, + "grad_norm": 0.4505653381347656, + "learning_rate": 3.84230043412246e-06, + "loss": 0.4614, + "step": 348 + }, + { + "epoch": 2.3886363636363637, + "grad_norm": 0.4461449384689331, + "learning_rate": 3.833750954155757e-06, + "loss": 0.4777, + "step": 349 + }, + { + "epoch": 2.3954545454545455, + "grad_norm": 0.423711895942688, + "learning_rate": 3.825179614174195e-06, + "loss": 0.4581, + "step": 350 + }, + { + "epoch": 2.4022727272727273, + "grad_norm": 0.462819904088974, + "learning_rate": 3.816586554660987e-06, + "loss": 0.4782, + "step": 351 + }, + { + "epoch": 2.409090909090909, + "grad_norm": 0.44206613302230835, + "learning_rate": 3.807971916455325e-06, + "loss": 0.477, + "step": 352 + }, + { + "epoch": 2.415909090909091, + "grad_norm": 0.4178541600704193, + "learning_rate": 3.799335840750077e-06, + "loss": 0.4981, + "step": 353 + }, + { + "epoch": 2.422727272727273, + "grad_norm": 0.670315682888031, + "learning_rate": 3.790678469089465e-06, + "loss": 0.4762, + "step": 354 + }, + { + "epoch": 2.4295454545454547, + "grad_norm": 0.44383955001831055, + "learning_rate": 3.7819999433667503e-06, + "loss": 0.4807, + "step": 355 + }, + { + "epoch": 2.4363636363636365, + "grad_norm": 0.4850422143936157, + "learning_rate": 3.773300405821908e-06, + "loss": 0.479, + "step": 356 + }, + { + "epoch": 2.4431818181818183, + "grad_norm": 0.458435595035553, + "learning_rate": 3.764579999039293e-06, + "loss": 0.4628, + "step": 357 + }, + { + "epoch": 2.45, + "grad_norm": 0.43913447856903076, + "learning_rate": 3.7558388659453052e-06, + "loss": 0.4948, + "step": 358 + }, + { + "epoch": 2.456818181818182, + "grad_norm": 0.4294499456882477, + "learning_rate": 3.7470771498060455e-06, + "loss": 0.4979, + "step": 359 + }, + { + "epoch": 2.463636363636364, + "grad_norm": 0.4472196400165558, + "learning_rate": 3.7382949942249695e-06, + "loss": 0.496, + "step": 360 + }, + { + "epoch": 2.4704545454545457, + "grad_norm": 0.4681987762451172, + "learning_rate": 3.7294925431405306e-06, + "loss": 0.4591, + "step": 361 + }, + { + "epoch": 2.4772727272727275, + "grad_norm": 0.40683335065841675, + "learning_rate": 3.720669940823827e-06, + "loss": 0.4755, + "step": 362 + }, + { + "epoch": 2.484090909090909, + "grad_norm": 0.46134939789772034, + "learning_rate": 3.7118273318762275e-06, + "loss": 0.4956, + "step": 363 + }, + { + "epoch": 2.4909090909090907, + "grad_norm": 0.4279940724372864, + "learning_rate": 3.702964861227013e-06, + "loss": 0.4815, + "step": 364 + }, + { + "epoch": 2.4977272727272726, + "grad_norm": 0.4699287414550781, + "learning_rate": 3.694082674130991e-06, + "loss": 0.4846, + "step": 365 + }, + { + "epoch": 2.5045454545454544, + "grad_norm": 0.4230785369873047, + "learning_rate": 3.6851809161661206e-06, + "loss": 0.5077, + "step": 366 + }, + { + "epoch": 2.5113636363636362, + "grad_norm": 0.4509921967983246, + "learning_rate": 3.6762597332311254e-06, + "loss": 0.5134, + "step": 367 + }, + { + "epoch": 2.518181818181818, + "grad_norm": 0.45494896173477173, + "learning_rate": 3.6673192715431016e-06, + "loss": 0.4685, + "step": 368 + }, + { + "epoch": 2.525, + "grad_norm": 0.4804443120956421, + "learning_rate": 3.658359677635122e-06, + "loss": 0.4458, + "step": 369 + }, + { + "epoch": 2.5318181818181817, + "grad_norm": 0.4149218499660492, + "learning_rate": 3.649381098353834e-06, + "loss": 0.5102, + "step": 370 + }, + { + "epoch": 2.5386363636363636, + "grad_norm": 0.4423210024833679, + "learning_rate": 3.6403836808570512e-06, + "loss": 0.5034, + "step": 371 + }, + { + "epoch": 2.5454545454545454, + "grad_norm": 0.4915727376937866, + "learning_rate": 3.631367572611348e-06, + "loss": 0.4889, + "step": 372 + }, + { + "epoch": 2.5522727272727272, + "grad_norm": 0.4189188778400421, + "learning_rate": 3.6223329213896313e-06, + "loss": 0.4829, + "step": 373 + }, + { + "epoch": 2.559090909090909, + "grad_norm": 0.511681079864502, + "learning_rate": 3.613279875268731e-06, + "loss": 0.4786, + "step": 374 + }, + { + "epoch": 2.565909090909091, + "grad_norm": 0.430137038230896, + "learning_rate": 3.604208582626964e-06, + "loss": 0.4783, + "step": 375 + }, + { + "epoch": 2.5727272727272728, + "grad_norm": 0.5629504919052124, + "learning_rate": 3.5951191921417063e-06, + "loss": 0.4438, + "step": 376 + }, + { + "epoch": 2.5795454545454546, + "grad_norm": 0.4354007840156555, + "learning_rate": 3.586011852786955e-06, + "loss": 0.4869, + "step": 377 + }, + { + "epoch": 2.5863636363636364, + "grad_norm": 0.45930856466293335, + "learning_rate": 3.5768867138308872e-06, + "loss": 0.4683, + "step": 378 + }, + { + "epoch": 2.5931818181818183, + "grad_norm": 0.4965468943119049, + "learning_rate": 3.5677439248334133e-06, + "loss": 0.4739, + "step": 379 + }, + { + "epoch": 2.6, + "grad_norm": 0.39618629217147827, + "learning_rate": 3.5585836356437266e-06, + "loss": 0.486, + "step": 380 + }, + { + "epoch": 2.606818181818182, + "grad_norm": 0.504718542098999, + "learning_rate": 3.5494059963978433e-06, + "loss": 0.4608, + "step": 381 + }, + { + "epoch": 2.6136363636363638, + "grad_norm": 0.3951753079891205, + "learning_rate": 3.540211157516149e-06, + "loss": 0.4973, + "step": 382 + }, + { + "epoch": 2.6204545454545456, + "grad_norm": 0.45937463641166687, + "learning_rate": 3.530999269700927e-06, + "loss": 0.4751, + "step": 383 + }, + { + "epoch": 2.6272727272727274, + "grad_norm": 0.4644455909729004, + "learning_rate": 3.521770483933891e-06, + "loss": 0.5003, + "step": 384 + }, + { + "epoch": 2.634090909090909, + "grad_norm": 0.4427309036254883, + "learning_rate": 3.5125249514737093e-06, + "loss": 0.4779, + "step": 385 + }, + { + "epoch": 2.6409090909090907, + "grad_norm": 0.4289201498031616, + "learning_rate": 3.503262823853527e-06, + "loss": 0.48, + "step": 386 + }, + { + "epoch": 2.6477272727272725, + "grad_norm": 0.5118290185928345, + "learning_rate": 3.493984252878483e-06, + "loss": 0.4858, + "step": 387 + }, + { + "epoch": 2.6545454545454543, + "grad_norm": 0.4599657952785492, + "learning_rate": 3.484689390623218e-06, + "loss": 0.4813, + "step": 388 + }, + { + "epoch": 2.661363636363636, + "grad_norm": 0.5077105164527893, + "learning_rate": 3.4753783894293886e-06, + "loss": 0.4732, + "step": 389 + }, + { + "epoch": 2.668181818181818, + "grad_norm": 0.5006387829780579, + "learning_rate": 3.466051401903162e-06, + "loss": 0.4743, + "step": 390 + }, + { + "epoch": 2.675, + "grad_norm": 0.4769401252269745, + "learning_rate": 3.4567085809127247e-06, + "loss": 0.4941, + "step": 391 + }, + { + "epoch": 2.6818181818181817, + "grad_norm": 0.4538915455341339, + "learning_rate": 3.4473500795857674e-06, + "loss": 0.4679, + "step": 392 + }, + { + "epoch": 2.6886363636363635, + "grad_norm": 0.4370926320552826, + "learning_rate": 3.4379760513069804e-06, + "loss": 0.4866, + "step": 393 + }, + { + "epoch": 2.6954545454545453, + "grad_norm": 0.445913165807724, + "learning_rate": 3.428586649715542e-06, + "loss": 0.4914, + "step": 394 + }, + { + "epoch": 2.702272727272727, + "grad_norm": 0.45687681436538696, + "learning_rate": 3.4191820287025916e-06, + "loss": 0.474, + "step": 395 + }, + { + "epoch": 2.709090909090909, + "grad_norm": 1.1952316761016846, + "learning_rate": 3.4097623424087196e-06, + "loss": 0.4905, + "step": 396 + }, + { + "epoch": 2.715909090909091, + "grad_norm": 0.41248032450675964, + "learning_rate": 3.4003277452214284e-06, + "loss": 0.4753, + "step": 397 + }, + { + "epoch": 2.7227272727272727, + "grad_norm": 0.4349999725818634, + "learning_rate": 3.3908783917726123e-06, + "loss": 0.4712, + "step": 398 + }, + { + "epoch": 2.7295454545454545, + "grad_norm": 0.42977845668792725, + "learning_rate": 3.381414436936018e-06, + "loss": 0.4936, + "step": 399 + }, + { + "epoch": 2.7363636363636363, + "grad_norm": 0.446321040391922, + "learning_rate": 3.3719360358247054e-06, + "loss": 0.4795, + "step": 400 + }, + { + "epoch": 2.743181818181818, + "grad_norm": 0.4957369565963745, + "learning_rate": 3.36244334378851e-06, + "loss": 0.48, + "step": 401 + }, + { + "epoch": 2.75, + "grad_norm": 0.43430694937705994, + "learning_rate": 3.3529365164114903e-06, + "loss": 0.4903, + "step": 402 + }, + { + "epoch": 2.756818181818182, + "grad_norm": 0.4340098202228546, + "learning_rate": 3.3434157095093846e-06, + "loss": 0.5111, + "step": 403 + }, + { + "epoch": 2.7636363636363637, + "grad_norm": 0.4309055805206299, + "learning_rate": 3.333881079127052e-06, + "loss": 0.4909, + "step": 404 + }, + { + "epoch": 2.7704545454545455, + "grad_norm": 0.43131211400032043, + "learning_rate": 3.3243327815359168e-06, + "loss": 0.4875, + "step": 405 + }, + { + "epoch": 2.7772727272727273, + "grad_norm": 0.4794235825538635, + "learning_rate": 3.314770973231408e-06, + "loss": 0.4766, + "step": 406 + }, + { + "epoch": 2.784090909090909, + "grad_norm": 0.4632478356361389, + "learning_rate": 3.305195810930393e-06, + "loss": 0.4765, + "step": 407 + }, + { + "epoch": 2.790909090909091, + "grad_norm": 0.4836251139640808, + "learning_rate": 3.2956074515686105e-06, + "loss": 0.4573, + "step": 408 + }, + { + "epoch": 2.797727272727273, + "grad_norm": 0.44514402747154236, + "learning_rate": 3.2860060522980945e-06, + "loss": 0.4794, + "step": 409 + }, + { + "epoch": 2.8045454545454547, + "grad_norm": 0.446785569190979, + "learning_rate": 3.276391770484606e-06, + "loss": 0.482, + "step": 410 + }, + { + "epoch": 2.8113636363636365, + "grad_norm": 0.4410347044467926, + "learning_rate": 3.266764763705046e-06, + "loss": 0.5136, + "step": 411 + }, + { + "epoch": 2.8181818181818183, + "grad_norm": 0.4888826906681061, + "learning_rate": 3.257125189744877e-06, + "loss": 0.4852, + "step": 412 + }, + { + "epoch": 2.825, + "grad_norm": 0.4502141773700714, + "learning_rate": 3.247473206595536e-06, + "loss": 0.4764, + "step": 413 + }, + { + "epoch": 2.831818181818182, + "grad_norm": 0.5693864822387695, + "learning_rate": 3.2378089724518464e-06, + "loss": 0.5016, + "step": 414 + }, + { + "epoch": 2.838636363636364, + "grad_norm": 0.49588093161582947, + "learning_rate": 3.228132645709421e-06, + "loss": 0.4786, + "step": 415 + }, + { + "epoch": 2.8454545454545457, + "grad_norm": 0.5465879440307617, + "learning_rate": 3.218444384962071e-06, + "loss": 0.4857, + "step": 416 + }, + { + "epoch": 2.8522727272727275, + "grad_norm": 0.4633638858795166, + "learning_rate": 3.2087443489992043e-06, + "loss": 0.475, + "step": 417 + }, + { + "epoch": 2.8590909090909093, + "grad_norm": 0.47139137983322144, + "learning_rate": 3.1990326968032225e-06, + "loss": 0.5162, + "step": 418 + }, + { + "epoch": 2.865909090909091, + "grad_norm": 0.45505622029304504, + "learning_rate": 3.189309587546917e-06, + "loss": 0.4699, + "step": 419 + }, + { + "epoch": 2.8727272727272726, + "grad_norm": 0.4410839080810547, + "learning_rate": 3.1795751805908578e-06, + "loss": 0.4734, + "step": 420 + }, + { + "epoch": 2.8795454545454544, + "grad_norm": 0.47172659635543823, + "learning_rate": 3.169829635480783e-06, + "loss": 0.4896, + "step": 421 + }, + { + "epoch": 2.8863636363636362, + "grad_norm": 0.4332088828086853, + "learning_rate": 3.160073111944983e-06, + "loss": 0.4898, + "step": 422 + }, + { + "epoch": 2.893181818181818, + "grad_norm": 0.4303777813911438, + "learning_rate": 3.150305769891686e-06, + "loss": 0.4663, + "step": 423 + }, + { + "epoch": 2.9, + "grad_norm": 0.44356584548950195, + "learning_rate": 3.1405277694064306e-06, + "loss": 0.4516, + "step": 424 + }, + { + "epoch": 2.9068181818181817, + "grad_norm": 0.5283893942832947, + "learning_rate": 3.13073927074945e-06, + "loss": 0.4792, + "step": 425 + }, + { + "epoch": 2.9136363636363636, + "grad_norm": 0.90226149559021, + "learning_rate": 3.1209404343530374e-06, + "loss": 0.4811, + "step": 426 + }, + { + "epoch": 2.9204545454545454, + "grad_norm": 0.4658738970756531, + "learning_rate": 3.111131420818922e-06, + "loss": 0.4601, + "step": 427 + }, + { + "epoch": 2.9272727272727272, + "grad_norm": 0.4286805987358093, + "learning_rate": 3.1013123909156347e-06, + "loss": 0.4801, + "step": 428 + }, + { + "epoch": 2.934090909090909, + "grad_norm": 0.5181618332862854, + "learning_rate": 3.091483505575873e-06, + "loss": 0.4717, + "step": 429 + }, + { + "epoch": 2.940909090909091, + "grad_norm": 0.4391111135482788, + "learning_rate": 3.081644925893866e-06, + "loss": 0.4913, + "step": 430 + }, + { + "epoch": 2.9477272727272728, + "grad_norm": 0.4414638578891754, + "learning_rate": 3.0717968131227285e-06, + "loss": 0.4796, + "step": 431 + }, + { + "epoch": 2.9545454545454546, + "grad_norm": 0.4619010090827942, + "learning_rate": 3.061939328671824e-06, + "loss": 0.5059, + "step": 432 + }, + { + "epoch": 2.9613636363636364, + "grad_norm": 0.42799466848373413, + "learning_rate": 3.0520726341041165e-06, + "loss": 0.4598, + "step": 433 + }, + { + "epoch": 2.9681818181818183, + "grad_norm": 0.43869948387145996, + "learning_rate": 3.0421968911335196e-06, + "loss": 0.4733, + "step": 434 + }, + { + "epoch": 2.975, + "grad_norm": 0.552511990070343, + "learning_rate": 3.032312261622255e-06, + "loss": 0.4855, + "step": 435 + }, + { + "epoch": 2.981818181818182, + "grad_norm": 0.5187029242515564, + "learning_rate": 3.0224189075781886e-06, + "loss": 0.5067, + "step": 436 + }, + { + "epoch": 2.9886363636363638, + "grad_norm": 0.4445323944091797, + "learning_rate": 3.012516991152181e-06, + "loss": 0.4542, + "step": 437 + }, + { + "epoch": 2.9954545454545456, + "grad_norm": 0.5056352019309998, + "learning_rate": 3.002606674635432e-06, + "loss": 0.4907, + "step": 438 + }, + { + "epoch": 3.006818181818182, + "grad_norm": 0.4359378218650818, + "learning_rate": 2.9926881204568153e-06, + "loss": 0.9747, + "step": 439 + }, + { + "epoch": 3.0136363636363637, + "grad_norm": 0.41525086760520935, + "learning_rate": 2.9827614911802205e-06, + "loss": 0.4672, + "step": 440 + }, + { + "epoch": 3.0204545454545455, + "grad_norm": 0.45512208342552185, + "learning_rate": 2.972826949501884e-06, + "loss": 0.4607, + "step": 441 + }, + { + "epoch": 3.0272727272727273, + "grad_norm": 0.4749738574028015, + "learning_rate": 2.9628846582477305e-06, + "loss": 0.4471, + "step": 442 + }, + { + "epoch": 3.034090909090909, + "grad_norm": 0.42964112758636475, + "learning_rate": 2.9529347803706943e-06, + "loss": 0.4437, + "step": 443 + }, + { + "epoch": 3.040909090909091, + "grad_norm": 0.42203956842422485, + "learning_rate": 2.9429774789480576e-06, + "loss": 0.444, + "step": 444 + }, + { + "epoch": 3.047727272727273, + "grad_norm": 0.4311574697494507, + "learning_rate": 2.9330129171787704e-06, + "loss": 0.4942, + "step": 445 + }, + { + "epoch": 3.0545454545454547, + "grad_norm": 0.4946717917919159, + "learning_rate": 2.923041258380779e-06, + "loss": 0.4505, + "step": 446 + }, + { + "epoch": 3.0613636363636365, + "grad_norm": 0.45294976234436035, + "learning_rate": 2.9130626659883537e-06, + "loss": 0.4537, + "step": 447 + }, + { + "epoch": 3.0681818181818183, + "grad_norm": 0.4274294376373291, + "learning_rate": 2.9030773035493997e-06, + "loss": 0.4532, + "step": 448 + }, + { + "epoch": 3.075, + "grad_norm": 0.4929022789001465, + "learning_rate": 2.893085334722786e-06, + "loss": 0.4403, + "step": 449 + }, + { + "epoch": 3.081818181818182, + "grad_norm": 0.4446672797203064, + "learning_rate": 2.883086923275658e-06, + "loss": 0.4572, + "step": 450 + }, + { + "epoch": 3.088636363636364, + "grad_norm": 0.45900678634643555, + "learning_rate": 2.8730822330807556e-06, + "loss": 0.4435, + "step": 451 + }, + { + "epoch": 3.0954545454545457, + "grad_norm": 0.4808853566646576, + "learning_rate": 2.8630714281137263e-06, + "loss": 0.4668, + "step": 452 + }, + { + "epoch": 3.102272727272727, + "grad_norm": 0.460186630487442, + "learning_rate": 2.853054672450437e-06, + "loss": 0.4393, + "step": 453 + }, + { + "epoch": 3.109090909090909, + "grad_norm": 0.4464828670024872, + "learning_rate": 2.8430321302642887e-06, + "loss": 0.4374, + "step": 454 + }, + { + "epoch": 3.1159090909090907, + "grad_norm": 0.4616970717906952, + "learning_rate": 2.8330039658235194e-06, + "loss": 0.4508, + "step": 455 + }, + { + "epoch": 3.1227272727272726, + "grad_norm": 0.45340001583099365, + "learning_rate": 2.8229703434885165e-06, + "loss": 0.4488, + "step": 456 + }, + { + "epoch": 3.1295454545454544, + "grad_norm": 0.4715891480445862, + "learning_rate": 2.8129314277091224e-06, + "loss": 0.4642, + "step": 457 + }, + { + "epoch": 3.1363636363636362, + "grad_norm": 0.4530370533466339, + "learning_rate": 2.8028873830219373e-06, + "loss": 0.4502, + "step": 458 + }, + { + "epoch": 3.143181818181818, + "grad_norm": 0.4896863102912903, + "learning_rate": 2.7928383740476247e-06, + "loss": 0.4559, + "step": 459 + }, + { + "epoch": 3.15, + "grad_norm": 0.45334747433662415, + "learning_rate": 2.7827845654882112e-06, + "loss": 0.4462, + "step": 460 + }, + { + "epoch": 3.1568181818181817, + "grad_norm": 0.4454493820667267, + "learning_rate": 2.7727261221243875e-06, + "loss": 0.4726, + "step": 461 + }, + { + "epoch": 3.1636363636363636, + "grad_norm": 0.46723124384880066, + "learning_rate": 2.76266320881281e-06, + "loss": 0.476, + "step": 462 + }, + { + "epoch": 3.1704545454545454, + "grad_norm": 0.45423537492752075, + "learning_rate": 2.7525959904833955e-06, + "loss": 0.4269, + "step": 463 + }, + { + "epoch": 3.1772727272727272, + "grad_norm": 0.44072285294532776, + "learning_rate": 2.7425246321366205e-06, + "loss": 0.4556, + "step": 464 + }, + { + "epoch": 3.184090909090909, + "grad_norm": 0.44829070568084717, + "learning_rate": 2.7324492988408146e-06, + "loss": 0.4479, + "step": 465 + }, + { + "epoch": 3.190909090909091, + "grad_norm": 0.47372356057167053, + "learning_rate": 2.7223701557294574e-06, + "loss": 0.4302, + "step": 466 + }, + { + "epoch": 3.1977272727272728, + "grad_norm": 0.4265802800655365, + "learning_rate": 2.712287367998471e-06, + "loss": 0.4557, + "step": 467 + }, + { + "epoch": 3.2045454545454546, + "grad_norm": 0.5714879631996155, + "learning_rate": 2.702201100903511e-06, + "loss": 0.437, + "step": 468 + }, + { + "epoch": 3.2113636363636364, + "grad_norm": 0.4389351010322571, + "learning_rate": 2.692111519757261e-06, + "loss": 0.4371, + "step": 469 + }, + { + "epoch": 3.2181818181818183, + "grad_norm": 0.44851404428482056, + "learning_rate": 2.6820187899267203e-06, + "loss": 0.4711, + "step": 470 + }, + { + "epoch": 3.225, + "grad_norm": 0.5391034483909607, + "learning_rate": 2.671923076830496e-06, + "loss": 0.4698, + "step": 471 + }, + { + "epoch": 3.231818181818182, + "grad_norm": 0.5056214332580566, + "learning_rate": 2.6618245459360896e-06, + "loss": 0.4712, + "step": 472 + }, + { + "epoch": 3.2386363636363638, + "grad_norm": 0.47290515899658203, + "learning_rate": 2.651723362757186e-06, + "loss": 0.4515, + "step": 473 + }, + { + "epoch": 3.2454545454545456, + "grad_norm": 0.4664989411830902, + "learning_rate": 2.641619692850941e-06, + "loss": 0.4602, + "step": 474 + }, + { + "epoch": 3.2522727272727274, + "grad_norm": 0.5000811219215393, + "learning_rate": 2.631513701815267e-06, + "loss": 0.4521, + "step": 475 + }, + { + "epoch": 3.2590909090909093, + "grad_norm": 0.4550754427909851, + "learning_rate": 2.6214055552861213e-06, + "loss": 0.471, + "step": 476 + }, + { + "epoch": 3.265909090909091, + "grad_norm": 0.4234600365161896, + "learning_rate": 2.611295418934786e-06, + "loss": 0.4486, + "step": 477 + }, + { + "epoch": 3.2727272727272725, + "grad_norm": 0.49546584486961365, + "learning_rate": 2.6011834584651597e-06, + "loss": 0.4367, + "step": 478 + }, + { + "epoch": 3.2795454545454543, + "grad_norm": 0.42106184363365173, + "learning_rate": 2.591069839611036e-06, + "loss": 0.4366, + "step": 479 + }, + { + "epoch": 3.286363636363636, + "grad_norm": 0.4530644118785858, + "learning_rate": 2.5809547281333904e-06, + "loss": 0.4729, + "step": 480 + }, + { + "epoch": 3.293181818181818, + "grad_norm": 0.4612523019313812, + "learning_rate": 2.570838289817661e-06, + "loss": 0.4531, + "step": 481 + }, + { + "epoch": 3.3, + "grad_norm": 0.4771486222743988, + "learning_rate": 2.560720690471033e-06, + "loss": 0.4656, + "step": 482 + }, + { + "epoch": 3.3068181818181817, + "grad_norm": 0.45801112055778503, + "learning_rate": 2.5506020959197218e-06, + "loss": 0.4443, + "step": 483 + }, + { + "epoch": 3.3136363636363635, + "grad_norm": 0.4690552353858948, + "learning_rate": 2.5404826720062544e-06, + "loss": 0.4442, + "step": 484 + }, + { + "epoch": 3.3204545454545453, + "grad_norm": 0.4829693138599396, + "learning_rate": 2.5303625845867475e-06, + "loss": 0.4658, + "step": 485 + }, + { + "epoch": 3.327272727272727, + "grad_norm": 0.4420918822288513, + "learning_rate": 2.5202419995281966e-06, + "loss": 0.4486, + "step": 486 + }, + { + "epoch": 3.334090909090909, + "grad_norm": 0.4645867347717285, + "learning_rate": 2.5101210827057516e-06, + "loss": 0.4536, + "step": 487 + }, + { + "epoch": 3.340909090909091, + "grad_norm": 0.45904067158699036, + "learning_rate": 2.5e-06, + "loss": 0.4777, + "step": 488 + }, + { + "epoch": 3.3477272727272727, + "grad_norm": 0.47885990142822266, + "learning_rate": 2.4898789172942492e-06, + "loss": 0.4507, + "step": 489 + }, + { + "epoch": 3.3545454545454545, + "grad_norm": 0.48885077238082886, + "learning_rate": 2.4797580004718038e-06, + "loss": 0.4492, + "step": 490 + }, + { + "epoch": 3.3613636363636363, + "grad_norm": 0.43438634276390076, + "learning_rate": 2.4696374154132533e-06, + "loss": 0.4622, + "step": 491 + }, + { + "epoch": 3.368181818181818, + "grad_norm": 0.6276775598526001, + "learning_rate": 2.4595173279937464e-06, + "loss": 0.4488, + "step": 492 + }, + { + "epoch": 3.375, + "grad_norm": 0.5427472591400146, + "learning_rate": 2.4493979040802786e-06, + "loss": 0.4482, + "step": 493 + }, + { + "epoch": 3.381818181818182, + "grad_norm": 0.4438450336456299, + "learning_rate": 2.4392793095289677e-06, + "loss": 0.4434, + "step": 494 + }, + { + "epoch": 3.3886363636363637, + "grad_norm": 0.49523183703422546, + "learning_rate": 2.42916171018234e-06, + "loss": 0.4645, + "step": 495 + }, + { + "epoch": 3.3954545454545455, + "grad_norm": 0.4602353274822235, + "learning_rate": 2.419045271866611e-06, + "loss": 0.4569, + "step": 496 + }, + { + "epoch": 3.4022727272727273, + "grad_norm": 0.4849902391433716, + "learning_rate": 2.408930160388965e-06, + "loss": 0.4538, + "step": 497 + }, + { + "epoch": 3.409090909090909, + "grad_norm": 0.6242562532424927, + "learning_rate": 2.3988165415348416e-06, + "loss": 0.4509, + "step": 498 + }, + { + "epoch": 3.415909090909091, + "grad_norm": 0.45550772547721863, + "learning_rate": 2.388704581065215e-06, + "loss": 0.4396, + "step": 499 + }, + { + "epoch": 3.422727272727273, + "grad_norm": 0.428011953830719, + "learning_rate": 2.3785944447138804e-06, + "loss": 0.4506, + "step": 500 + }, + { + "epoch": 3.4295454545454547, + "grad_norm": 0.47410571575164795, + "learning_rate": 2.368486298184733e-06, + "loss": 0.4573, + "step": 501 + }, + { + "epoch": 3.4363636363636365, + "grad_norm": 0.45392513275146484, + "learning_rate": 2.358380307149059e-06, + "loss": 0.4607, + "step": 502 + }, + { + "epoch": 3.4431818181818183, + "grad_norm": 0.4263063669204712, + "learning_rate": 2.348276637242814e-06, + "loss": 0.4574, + "step": 503 + }, + { + "epoch": 3.45, + "grad_norm": 0.5994442701339722, + "learning_rate": 2.3381754540639108e-06, + "loss": 0.4483, + "step": 504 + }, + { + "epoch": 3.456818181818182, + "grad_norm": 0.45459091663360596, + "learning_rate": 2.328076923169504e-06, + "loss": 0.4445, + "step": 505 + }, + { + "epoch": 3.463636363636364, + "grad_norm": 0.48127374053001404, + "learning_rate": 2.31798121007328e-06, + "loss": 0.4455, + "step": 506 + }, + { + "epoch": 3.4704545454545457, + "grad_norm": 0.7549276947975159, + "learning_rate": 2.3078884802427394e-06, + "loss": 0.4475, + "step": 507 + }, + { + "epoch": 3.4772727272727275, + "grad_norm": 0.4759480953216553, + "learning_rate": 2.29779889909649e-06, + "loss": 0.4512, + "step": 508 + }, + { + "epoch": 3.484090909090909, + "grad_norm": 0.45603302121162415, + "learning_rate": 2.2877126320015295e-06, + "loss": 0.4441, + "step": 509 + }, + { + "epoch": 3.4909090909090907, + "grad_norm": 0.47070786356925964, + "learning_rate": 2.2776298442705434e-06, + "loss": 0.4402, + "step": 510 + }, + { + "epoch": 3.4977272727272726, + "grad_norm": 0.4690212905406952, + "learning_rate": 2.267550701159186e-06, + "loss": 0.4517, + "step": 511 + }, + { + "epoch": 3.5045454545454544, + "grad_norm": 0.8859854340553284, + "learning_rate": 2.25747536786338e-06, + "loss": 0.4374, + "step": 512 + }, + { + "epoch": 3.5113636363636362, + "grad_norm": 0.4920756220817566, + "learning_rate": 2.247404009516605e-06, + "loss": 0.446, + "step": 513 + }, + { + "epoch": 3.518181818181818, + "grad_norm": 0.49091637134552, + "learning_rate": 2.2373367911871904e-06, + "loss": 0.438, + "step": 514 + }, + { + "epoch": 3.525, + "grad_norm": 0.8683479428291321, + "learning_rate": 2.227273877875613e-06, + "loss": 0.4435, + "step": 515 + }, + { + "epoch": 3.5318181818181817, + "grad_norm": 0.4606627821922302, + "learning_rate": 2.2172154345117896e-06, + "loss": 0.4517, + "step": 516 + }, + { + "epoch": 3.5386363636363636, + "grad_norm": 0.48613184690475464, + "learning_rate": 2.207161625952376e-06, + "loss": 0.4449, + "step": 517 + }, + { + "epoch": 3.5454545454545454, + "grad_norm": 0.5018280148506165, + "learning_rate": 2.1971126169780636e-06, + "loss": 0.4414, + "step": 518 + }, + { + "epoch": 3.5522727272727272, + "grad_norm": 0.435921847820282, + "learning_rate": 2.1870685722908784e-06, + "loss": 0.4542, + "step": 519 + }, + { + "epoch": 3.559090909090909, + "grad_norm": 0.48534268140792847, + "learning_rate": 2.1770296565114847e-06, + "loss": 0.4373, + "step": 520 + }, + { + "epoch": 3.565909090909091, + "grad_norm": 0.49759384989738464, + "learning_rate": 2.166996034176482e-06, + "loss": 0.4518, + "step": 521 + }, + { + "epoch": 3.5727272727272728, + "grad_norm": 0.524696409702301, + "learning_rate": 2.1569678697357126e-06, + "loss": 0.4668, + "step": 522 + }, + { + "epoch": 3.5795454545454546, + "grad_norm": 0.4403073191642761, + "learning_rate": 2.1469453275495634e-06, + "loss": 0.44, + "step": 523 + }, + { + "epoch": 3.5863636363636364, + "grad_norm": 0.5214805006980896, + "learning_rate": 2.136928571886275e-06, + "loss": 0.437, + "step": 524 + }, + { + "epoch": 3.5931818181818183, + "grad_norm": 0.4685129225254059, + "learning_rate": 2.126917766919245e-06, + "loss": 0.4477, + "step": 525 + }, + { + "epoch": 3.6, + "grad_norm": 0.500602126121521, + "learning_rate": 2.1169130767243424e-06, + "loss": 0.455, + "step": 526 + }, + { + "epoch": 3.606818181818182, + "grad_norm": 0.5242429375648499, + "learning_rate": 2.1069146652772142e-06, + "loss": 0.4398, + "step": 527 + }, + { + "epoch": 3.6136363636363638, + "grad_norm": 0.6797722578048706, + "learning_rate": 2.0969226964506007e-06, + "loss": 0.4393, + "step": 528 + }, + { + "epoch": 3.6204545454545456, + "grad_norm": 0.44578859210014343, + "learning_rate": 2.0869373340116467e-06, + "loss": 0.458, + "step": 529 + }, + { + "epoch": 3.6272727272727274, + "grad_norm": 0.46465393900871277, + "learning_rate": 2.0769587416192212e-06, + "loss": 0.4461, + "step": 530 + }, + { + "epoch": 3.634090909090909, + "grad_norm": 0.48858389258384705, + "learning_rate": 2.066987082821231e-06, + "loss": 0.4623, + "step": 531 + }, + { + "epoch": 3.6409090909090907, + "grad_norm": 0.4606386423110962, + "learning_rate": 2.0570225210519433e-06, + "loss": 0.4538, + "step": 532 + }, + { + "epoch": 3.6477272727272725, + "grad_norm": 0.46266230940818787, + "learning_rate": 2.047065219629306e-06, + "loss": 0.4635, + "step": 533 + }, + { + "epoch": 3.6545454545454543, + "grad_norm": 0.5587124824523926, + "learning_rate": 2.0371153417522703e-06, + "loss": 0.4614, + "step": 534 + }, + { + "epoch": 3.661363636363636, + "grad_norm": 0.6143503785133362, + "learning_rate": 2.0271730504981165e-06, + "loss": 0.4408, + "step": 535 + }, + { + "epoch": 3.668181818181818, + "grad_norm": 0.5167360305786133, + "learning_rate": 2.0172385088197804e-06, + "loss": 0.4539, + "step": 536 + }, + { + "epoch": 3.675, + "grad_norm": 0.47937509417533875, + "learning_rate": 2.007311879543185e-06, + "loss": 0.4587, + "step": 537 + }, + { + "epoch": 3.6818181818181817, + "grad_norm": 0.4938436448574066, + "learning_rate": 1.9973933253645684e-06, + "loss": 0.4448, + "step": 538 + }, + { + "epoch": 3.6886363636363635, + "grad_norm": 0.4234243631362915, + "learning_rate": 1.9874830088478196e-06, + "loss": 0.4584, + "step": 539 + }, + { + "epoch": 3.6954545454545453, + "grad_norm": 0.5978733897209167, + "learning_rate": 1.9775810924218126e-06, + "loss": 0.4647, + "step": 540 + }, + { + "epoch": 3.702272727272727, + "grad_norm": 0.49155011773109436, + "learning_rate": 1.967687738377746e-06, + "loss": 0.4496, + "step": 541 + }, + { + "epoch": 3.709090909090909, + "grad_norm": 0.4360124468803406, + "learning_rate": 1.9578031088664812e-06, + "loss": 0.466, + "step": 542 + }, + { + "epoch": 3.715909090909091, + "grad_norm": 0.4867744743824005, + "learning_rate": 1.9479273658958852e-06, + "loss": 0.4362, + "step": 543 + }, + { + "epoch": 3.7227272727272727, + "grad_norm": 0.49178293347358704, + "learning_rate": 1.9380606713281773e-06, + "loss": 0.4501, + "step": 544 + }, + { + "epoch": 3.7295454545454545, + "grad_norm": 0.5098608136177063, + "learning_rate": 1.928203186877273e-06, + "loss": 0.4655, + "step": 545 + }, + { + "epoch": 3.7363636363636363, + "grad_norm": 0.4781307578086853, + "learning_rate": 1.9183550741061354e-06, + "loss": 0.4464, + "step": 546 + }, + { + "epoch": 3.743181818181818, + "grad_norm": 0.49883827567100525, + "learning_rate": 1.9085164944241275e-06, + "loss": 0.4366, + "step": 547 + }, + { + "epoch": 3.75, + "grad_norm": 0.5124484300613403, + "learning_rate": 1.8986876090843668e-06, + "loss": 0.4625, + "step": 548 + }, + { + "epoch": 3.756818181818182, + "grad_norm": 0.46270111203193665, + "learning_rate": 1.8888685791810784e-06, + "loss": 0.4149, + "step": 549 + }, + { + "epoch": 3.7636363636363637, + "grad_norm": 0.5264699459075928, + "learning_rate": 1.8790595656469628e-06, + "loss": 0.4373, + "step": 550 + }, + { + "epoch": 3.7704545454545455, + "grad_norm": 0.46891146898269653, + "learning_rate": 1.86926072925055e-06, + "loss": 0.4437, + "step": 551 + }, + { + "epoch": 3.7772727272727273, + "grad_norm": 0.4786033630371094, + "learning_rate": 1.8594722305935691e-06, + "loss": 0.4364, + "step": 552 + }, + { + "epoch": 3.784090909090909, + "grad_norm": 0.45475709438323975, + "learning_rate": 1.8496942301083142e-06, + "loss": 0.4515, + "step": 553 + }, + { + "epoch": 3.790909090909091, + "grad_norm": 0.4658668637275696, + "learning_rate": 1.8399268880550174e-06, + "loss": 0.4553, + "step": 554 + }, + { + "epoch": 3.797727272727273, + "grad_norm": 0.5587930679321289, + "learning_rate": 1.8301703645192178e-06, + "loss": 0.4412, + "step": 555 + }, + { + "epoch": 3.8045454545454547, + "grad_norm": 0.4532659947872162, + "learning_rate": 1.8204248194091429e-06, + "loss": 0.4529, + "step": 556 + }, + { + "epoch": 3.8113636363636365, + "grad_norm": 0.4606110751628876, + "learning_rate": 1.8106904124530839e-06, + "loss": 0.437, + "step": 557 + }, + { + "epoch": 3.8181818181818183, + "grad_norm": 0.4482369124889374, + "learning_rate": 1.800967303196778e-06, + "loss": 0.4309, + "step": 558 + }, + { + "epoch": 3.825, + "grad_norm": 0.533805787563324, + "learning_rate": 1.7912556510007967e-06, + "loss": 0.4393, + "step": 559 + }, + { + "epoch": 3.831818181818182, + "grad_norm": 0.5079528093338013, + "learning_rate": 1.7815556150379298e-06, + "loss": 0.4585, + "step": 560 + }, + { + "epoch": 3.838636363636364, + "grad_norm": 0.5496546626091003, + "learning_rate": 1.77186735429058e-06, + "loss": 0.4293, + "step": 561 + }, + { + "epoch": 3.8454545454545457, + "grad_norm": 0.49827462434768677, + "learning_rate": 1.7621910275481544e-06, + "loss": 0.4457, + "step": 562 + }, + { + "epoch": 3.8522727272727275, + "grad_norm": 0.5003693699836731, + "learning_rate": 1.7525267934044642e-06, + "loss": 0.4579, + "step": 563 + }, + { + "epoch": 3.8590909090909093, + "grad_norm": 0.47428634762763977, + "learning_rate": 1.7428748102551237e-06, + "loss": 0.452, + "step": 564 + }, + { + "epoch": 3.865909090909091, + "grad_norm": 0.4468045234680176, + "learning_rate": 1.7332352362949546e-06, + "loss": 0.4558, + "step": 565 + }, + { + "epoch": 3.8727272727272726, + "grad_norm": 0.4439336955547333, + "learning_rate": 1.7236082295153948e-06, + "loss": 0.4351, + "step": 566 + }, + { + "epoch": 3.8795454545454544, + "grad_norm": 0.46482470631599426, + "learning_rate": 1.7139939477019057e-06, + "loss": 0.4462, + "step": 567 + }, + { + "epoch": 3.8863636363636362, + "grad_norm": 1.1179648637771606, + "learning_rate": 1.7043925484313911e-06, + "loss": 0.4435, + "step": 568 + }, + { + "epoch": 3.893181818181818, + "grad_norm": 0.46164315938949585, + "learning_rate": 1.6948041890696076e-06, + "loss": 0.4446, + "step": 569 + }, + { + "epoch": 3.9, + "grad_norm": 0.4524129629135132, + "learning_rate": 1.685229026768593e-06, + "loss": 0.4567, + "step": 570 + }, + { + "epoch": 3.9068181818181817, + "grad_norm": 0.45722272992134094, + "learning_rate": 1.6756672184640847e-06, + "loss": 0.4368, + "step": 571 + }, + { + "epoch": 3.9136363636363636, + "grad_norm": 0.49532264471054077, + "learning_rate": 1.6661189208729492e-06, + "loss": 0.4572, + "step": 572 + }, + { + "epoch": 3.9204545454545454, + "grad_norm": 0.44752320647239685, + "learning_rate": 1.6565842904906154e-06, + "loss": 0.43, + "step": 573 + }, + { + "epoch": 3.9272727272727272, + "grad_norm": 0.5931842923164368, + "learning_rate": 1.6470634835885097e-06, + "loss": 0.4334, + "step": 574 + }, + { + "epoch": 3.934090909090909, + "grad_norm": 0.5873560905456543, + "learning_rate": 1.6375566562114903e-06, + "loss": 0.4374, + "step": 575 + }, + { + "epoch": 3.940909090909091, + "grad_norm": 0.4484315812587738, + "learning_rate": 1.6280639641752944e-06, + "loss": 0.4402, + "step": 576 + }, + { + "epoch": 3.9477272727272728, + "grad_norm": 0.4635726809501648, + "learning_rate": 1.6185855630639818e-06, + "loss": 0.4485, + "step": 577 + }, + { + "epoch": 3.9545454545454546, + "grad_norm": 0.7416269779205322, + "learning_rate": 1.6091216082273875e-06, + "loss": 0.4655, + "step": 578 + }, + { + "epoch": 3.9613636363636364, + "grad_norm": 0.4709416329860687, + "learning_rate": 1.5996722547785722e-06, + "loss": 0.4551, + "step": 579 + }, + { + "epoch": 3.9681818181818183, + "grad_norm": 0.45414504408836365, + "learning_rate": 1.5902376575912815e-06, + "loss": 0.4478, + "step": 580 + }, + { + "epoch": 3.975, + "grad_norm": 0.45831212401390076, + "learning_rate": 1.580817971297409e-06, + "loss": 0.462, + "step": 581 + }, + { + "epoch": 3.981818181818182, + "grad_norm": 0.4817596971988678, + "learning_rate": 1.5714133502844591e-06, + "loss": 0.4554, + "step": 582 + }, + { + "epoch": 3.9886363636363638, + "grad_norm": 0.48830825090408325, + "learning_rate": 1.56202394869302e-06, + "loss": 0.4382, + "step": 583 + }, + { + "epoch": 3.9954545454545456, + "grad_norm": 0.4884580075740814, + "learning_rate": 1.5526499204142332e-06, + "loss": 0.4527, + "step": 584 + }, + { + "epoch": 4.006818181818182, + "grad_norm": 1.0956987142562866, + "learning_rate": 1.5432914190872757e-06, + "loss": 0.9264, + "step": 585 + }, + { + "epoch": 4.013636363636364, + "grad_norm": 0.45760378241539, + "learning_rate": 1.5339485980968383e-06, + "loss": 0.4399, + "step": 586 + }, + { + "epoch": 4.0204545454545455, + "grad_norm": 0.44204893708229065, + "learning_rate": 1.5246216105706124e-06, + "loss": 0.4373, + "step": 587 + }, + { + "epoch": 4.027272727272727, + "grad_norm": 0.8188589215278625, + "learning_rate": 1.5153106093767827e-06, + "loss": 0.4356, + "step": 588 + }, + { + "epoch": 4.034090909090909, + "grad_norm": 0.4039295017719269, + "learning_rate": 1.506015747121518e-06, + "loss": 0.4215, + "step": 589 + }, + { + "epoch": 4.040909090909091, + "grad_norm": 0.4564852714538574, + "learning_rate": 1.4967371761464738e-06, + "loss": 0.4402, + "step": 590 + }, + { + "epoch": 4.047727272727273, + "grad_norm": 0.4643855094909668, + "learning_rate": 1.4874750485262917e-06, + "loss": 0.4374, + "step": 591 + }, + { + "epoch": 4.054545454545455, + "grad_norm": 0.561626672744751, + "learning_rate": 1.4782295160661103e-06, + "loss": 0.4323, + "step": 592 + }, + { + "epoch": 4.0613636363636365, + "grad_norm": 0.4655170440673828, + "learning_rate": 1.469000730299074e-06, + "loss": 0.4278, + "step": 593 + }, + { + "epoch": 4.068181818181818, + "grad_norm": 0.5915263891220093, + "learning_rate": 1.4597888424838519e-06, + "loss": 0.4208, + "step": 594 + }, + { + "epoch": 4.075, + "grad_norm": 0.44420668482780457, + "learning_rate": 1.450594003602158e-06, + "loss": 0.4424, + "step": 595 + }, + { + "epoch": 4.081818181818182, + "grad_norm": 0.49877864122390747, + "learning_rate": 1.4414163643562755e-06, + "loss": 0.4206, + "step": 596 + }, + { + "epoch": 4.088636363636364, + "grad_norm": 0.4505457580089569, + "learning_rate": 1.4322560751665873e-06, + "loss": 0.4345, + "step": 597 + }, + { + "epoch": 4.095454545454546, + "grad_norm": 0.47181791067123413, + "learning_rate": 1.4231132861691128e-06, + "loss": 0.4207, + "step": 598 + }, + { + "epoch": 4.1022727272727275, + "grad_norm": 0.4665491282939911, + "learning_rate": 1.4139881472130453e-06, + "loss": 0.4221, + "step": 599 + }, + { + "epoch": 4.109090909090909, + "grad_norm": 0.46456652879714966, + "learning_rate": 1.4048808078582943e-06, + "loss": 0.4162, + "step": 600 + }, + { + "epoch": 4.115909090909091, + "grad_norm": 0.437472939491272, + "learning_rate": 1.3957914173730366e-06, + "loss": 0.4379, + "step": 601 + }, + { + "epoch": 4.122727272727273, + "grad_norm": 0.5029657483100891, + "learning_rate": 1.3867201247312697e-06, + "loss": 0.4402, + "step": 602 + }, + { + "epoch": 4.129545454545455, + "grad_norm": 0.8025555610656738, + "learning_rate": 1.3776670786103685e-06, + "loss": 0.4449, + "step": 603 + }, + { + "epoch": 4.136363636363637, + "grad_norm": 0.5241179466247559, + "learning_rate": 1.3686324273886531e-06, + "loss": 0.4318, + "step": 604 + }, + { + "epoch": 4.1431818181818185, + "grad_norm": 0.4537895917892456, + "learning_rate": 1.359616319142949e-06, + "loss": 0.4354, + "step": 605 + }, + { + "epoch": 4.15, + "grad_norm": 0.5133662223815918, + "learning_rate": 1.3506189016461674e-06, + "loss": 0.4363, + "step": 606 + }, + { + "epoch": 4.156818181818182, + "grad_norm": 0.4720538854598999, + "learning_rate": 1.341640322364878e-06, + "loss": 0.4332, + "step": 607 + }, + { + "epoch": 4.163636363636364, + "grad_norm": 1.3719831705093384, + "learning_rate": 1.3326807284568984e-06, + "loss": 0.43, + "step": 608 + }, + { + "epoch": 4.170454545454546, + "grad_norm": 0.4677439332008362, + "learning_rate": 1.323740266768875e-06, + "loss": 0.4369, + "step": 609 + }, + { + "epoch": 4.177272727272728, + "grad_norm": 0.5197097659111023, + "learning_rate": 1.3148190838338804e-06, + "loss": 0.3977, + "step": 610 + }, + { + "epoch": 4.184090909090909, + "grad_norm": 0.8679460287094116, + "learning_rate": 1.3059173258690102e-06, + "loss": 0.4474, + "step": 611 + }, + { + "epoch": 4.190909090909091, + "grad_norm": 0.446285218000412, + "learning_rate": 1.2970351387729875e-06, + "loss": 0.438, + "step": 612 + }, + { + "epoch": 4.197727272727272, + "grad_norm": 0.48193076252937317, + "learning_rate": 1.2881726681237727e-06, + "loss": 0.4403, + "step": 613 + }, + { + "epoch": 4.204545454545454, + "grad_norm": 0.4694555997848511, + "learning_rate": 1.2793300591761742e-06, + "loss": 0.4022, + "step": 614 + }, + { + "epoch": 4.211363636363636, + "grad_norm": 0.46118268370628357, + "learning_rate": 1.27050745685947e-06, + "loss": 0.442, + "step": 615 + }, + { + "epoch": 4.218181818181818, + "grad_norm": 0.6086122989654541, + "learning_rate": 1.2617050057750322e-06, + "loss": 0.4272, + "step": 616 + }, + { + "epoch": 4.225, + "grad_norm": 0.4840640723705292, + "learning_rate": 1.252922850193955e-06, + "loss": 0.4263, + "step": 617 + }, + { + "epoch": 4.2318181818181815, + "grad_norm": 0.46273577213287354, + "learning_rate": 1.2441611340546958e-06, + "loss": 0.4254, + "step": 618 + }, + { + "epoch": 4.238636363636363, + "grad_norm": 0.501190721988678, + "learning_rate": 1.2354200009607081e-06, + "loss": 0.435, + "step": 619 + }, + { + "epoch": 4.245454545454545, + "grad_norm": 0.4813865125179291, + "learning_rate": 1.2266995941780934e-06, + "loss": 0.4267, + "step": 620 + }, + { + "epoch": 4.252272727272727, + "grad_norm": 0.5107240676879883, + "learning_rate": 1.2180000566332503e-06, + "loss": 0.3868, + "step": 621 + }, + { + "epoch": 4.259090909090909, + "grad_norm": 0.49328625202178955, + "learning_rate": 1.2093215309105352e-06, + "loss": 0.4228, + "step": 622 + }, + { + "epoch": 4.265909090909091, + "grad_norm": 0.7121614217758179, + "learning_rate": 1.2006641592499233e-06, + "loss": 0.4262, + "step": 623 + }, + { + "epoch": 4.2727272727272725, + "grad_norm": 0.5118347406387329, + "learning_rate": 1.192028083544675e-06, + "loss": 0.4228, + "step": 624 + }, + { + "epoch": 4.279545454545454, + "grad_norm": 0.45035621523857117, + "learning_rate": 1.1834134453390136e-06, + "loss": 0.4239, + "step": 625 + }, + { + "epoch": 4.286363636363636, + "grad_norm": 0.45246437191963196, + "learning_rate": 1.1748203858258056e-06, + "loss": 0.4268, + "step": 626 + }, + { + "epoch": 4.293181818181818, + "grad_norm": 0.48875924944877625, + "learning_rate": 1.166249045844243e-06, + "loss": 0.4227, + "step": 627 + }, + { + "epoch": 4.3, + "grad_norm": 0.4633196294307709, + "learning_rate": 1.1576995658775405e-06, + "loss": 0.4192, + "step": 628 + }, + { + "epoch": 4.306818181818182, + "grad_norm": 0.5269569158554077, + "learning_rate": 1.1491720860506273e-06, + "loss": 0.4212, + "step": 629 + }, + { + "epoch": 4.3136363636363635, + "grad_norm": 0.4277689754962921, + "learning_rate": 1.140666746127854e-06, + "loss": 0.4123, + "step": 630 + }, + { + "epoch": 4.320454545454545, + "grad_norm": 0.49047568440437317, + "learning_rate": 1.1321836855107007e-06, + "loss": 0.4128, + "step": 631 + }, + { + "epoch": 4.327272727272727, + "grad_norm": 0.45740997791290283, + "learning_rate": 1.1237230432354912e-06, + "loss": 0.4284, + "step": 632 + }, + { + "epoch": 4.334090909090909, + "grad_norm": 0.7805290818214417, + "learning_rate": 1.1152849579711187e-06, + "loss": 0.4487, + "step": 633 + }, + { + "epoch": 4.340909090909091, + "grad_norm": 0.6119694113731384, + "learning_rate": 1.1068695680167665e-06, + "loss": 0.4442, + "step": 634 + }, + { + "epoch": 4.347727272727273, + "grad_norm": 0.48440369963645935, + "learning_rate": 1.0984770112996463e-06, + "loss": 0.4223, + "step": 635 + }, + { + "epoch": 4.3545454545454545, + "grad_norm": 0.4387761056423187, + "learning_rate": 1.0901074253727338e-06, + "loss": 0.434, + "step": 636 + }, + { + "epoch": 4.361363636363636, + "grad_norm": 0.499102920293808, + "learning_rate": 1.0817609474125195e-06, + "loss": 0.427, + "step": 637 + }, + { + "epoch": 4.368181818181818, + "grad_norm": 0.541165292263031, + "learning_rate": 1.0734377142167549e-06, + "loss": 0.4236, + "step": 638 + }, + { + "epoch": 4.375, + "grad_norm": 0.5833367705345154, + "learning_rate": 1.065137862202213e-06, + "loss": 0.4171, + "step": 639 + }, + { + "epoch": 4.381818181818182, + "grad_norm": 0.45729443430900574, + "learning_rate": 1.0568615274024521e-06, + "loss": 0.4241, + "step": 640 + }, + { + "epoch": 4.388636363636364, + "grad_norm": 0.4729260206222534, + "learning_rate": 1.0486088454655856e-06, + "loss": 0.4164, + "step": 641 + }, + { + "epoch": 4.3954545454545455, + "grad_norm": 0.5857828855514526, + "learning_rate": 1.0403799516520619e-06, + "loss": 0.4219, + "step": 642 + }, + { + "epoch": 4.402272727272727, + "grad_norm": 0.5247046947479248, + "learning_rate": 1.0321749808324425e-06, + "loss": 0.4138, + "step": 643 + }, + { + "epoch": 4.409090909090909, + "grad_norm": 0.5515534281730652, + "learning_rate": 1.0239940674851943e-06, + "loss": 0.4613, + "step": 644 + }, + { + "epoch": 4.415909090909091, + "grad_norm": 0.4807690680027008, + "learning_rate": 1.0158373456944856e-06, + "loss": 0.4146, + "step": 645 + }, + { + "epoch": 4.422727272727273, + "grad_norm": 0.5009557604789734, + "learning_rate": 1.0077049491479874e-06, + "loss": 0.4031, + "step": 646 + }, + { + "epoch": 4.429545454545455, + "grad_norm": 0.4809257686138153, + "learning_rate": 9.995970111346842e-07, + "loss": 0.4293, + "step": 647 + }, + { + "epoch": 4.4363636363636365, + "grad_norm": 0.4761544466018677, + "learning_rate": 9.915136645426885e-07, + "loss": 0.428, + "step": 648 + }, + { + "epoch": 4.443181818181818, + "grad_norm": 0.44973552227020264, + "learning_rate": 9.834550418570602e-07, + "loss": 0.4346, + "step": 649 + }, + { + "epoch": 4.45, + "grad_norm": 0.4652273654937744, + "learning_rate": 9.754212751576386e-07, + "loss": 0.4255, + "step": 650 + }, + { + "epoch": 4.456818181818182, + "grad_norm": 0.5163391828536987, + "learning_rate": 9.67412496116876e-07, + "loss": 0.4008, + "step": 651 + }, + { + "epoch": 4.463636363636364, + "grad_norm": 0.525343120098114, + "learning_rate": 9.594288359976817e-07, + "loss": 0.4233, + "step": 652 + }, + { + "epoch": 4.470454545454546, + "grad_norm": 0.4707981050014496, + "learning_rate": 9.514704256512669e-07, + "loss": 0.4473, + "step": 653 + }, + { + "epoch": 4.4772727272727275, + "grad_norm": 0.4338839650154114, + "learning_rate": 9.435373955150032e-07, + "loss": 0.4216, + "step": 654 + }, + { + "epoch": 4.484090909090909, + "grad_norm": 0.4526594281196594, + "learning_rate": 9.35629875610283e-07, + "loss": 0.4307, + "step": 655 + }, + { + "epoch": 4.490909090909091, + "grad_norm": 0.5091291069984436, + "learning_rate": 9.277479955403887e-07, + "loss": 0.4449, + "step": 656 + }, + { + "epoch": 4.497727272727273, + "grad_norm": 0.4887843430042267, + "learning_rate": 9.198918844883714e-07, + "loss": 0.434, + "step": 657 + }, + { + "epoch": 4.504545454545455, + "grad_norm": 0.46774518489837646, + "learning_rate": 9.120616712149291e-07, + "loss": 0.426, + "step": 658 + }, + { + "epoch": 4.511363636363637, + "grad_norm": 0.47825315594673157, + "learning_rate": 9.042574840562982e-07, + "loss": 0.4372, + "step": 659 + }, + { + "epoch": 4.5181818181818185, + "grad_norm": 0.46589818596839905, + "learning_rate": 8.964794509221508e-07, + "loss": 0.4378, + "step": 660 + }, + { + "epoch": 4.525, + "grad_norm": 0.4720231592655182, + "learning_rate": 8.887276992934976e-07, + "loss": 0.4348, + "step": 661 + }, + { + "epoch": 4.531818181818182, + "grad_norm": 0.47262752056121826, + "learning_rate": 8.810023562206e-07, + "loss": 0.4195, + "step": 662 + }, + { + "epoch": 4.538636363636364, + "grad_norm": 0.975907564163208, + "learning_rate": 8.733035483208841e-07, + "loss": 0.4382, + "step": 663 + }, + { + "epoch": 4.545454545454545, + "grad_norm": 0.47100895643234253, + "learning_rate": 8.656314017768694e-07, + "loss": 0.4213, + "step": 664 + }, + { + "epoch": 4.552272727272728, + "grad_norm": 0.517842710018158, + "learning_rate": 8.579860423340977e-07, + "loss": 0.432, + "step": 665 + }, + { + "epoch": 4.559090909090909, + "grad_norm": 0.4891679286956787, + "learning_rate": 8.503675952990756e-07, + "loss": 0.442, + "step": 666 + }, + { + "epoch": 4.565909090909091, + "grad_norm": 0.48658767342567444, + "learning_rate": 8.427761855372169e-07, + "loss": 0.4255, + "step": 667 + }, + { + "epoch": 4.572727272727272, + "grad_norm": 0.4577118158340454, + "learning_rate": 8.352119374707979e-07, + "loss": 0.4199, + "step": 668 + }, + { + "epoch": 4.579545454545455, + "grad_norm": 0.6099198460578918, + "learning_rate": 8.276749750769186e-07, + "loss": 0.4255, + "step": 669 + }, + { + "epoch": 4.586363636363636, + "grad_norm": 0.5071508884429932, + "learning_rate": 8.20165421885469e-07, + "loss": 0.4384, + "step": 670 + }, + { + "epoch": 4.593181818181818, + "grad_norm": 0.5968758463859558, + "learning_rate": 8.126834009771079e-07, + "loss": 0.4247, + "step": 671 + }, + { + "epoch": 4.6, + "grad_norm": 0.4893784821033478, + "learning_rate": 8.052290349812419e-07, + "loss": 0.4368, + "step": 672 + }, + { + "epoch": 4.6068181818181815, + "grad_norm": 0.4897981584072113, + "learning_rate": 7.978024460740169e-07, + "loss": 0.4423, + "step": 673 + }, + { + "epoch": 4.613636363636363, + "grad_norm": 0.49245044589042664, + "learning_rate": 7.904037559763162e-07, + "loss": 0.4508, + "step": 674 + }, + { + "epoch": 4.620454545454545, + "grad_norm": 0.48101088404655457, + "learning_rate": 7.83033085951764e-07, + "loss": 0.4268, + "step": 675 + }, + { + "epoch": 4.627272727272727, + "grad_norm": 0.6497372388839722, + "learning_rate": 7.756905568047393e-07, + "loss": 0.4374, + "step": 676 + }, + { + "epoch": 4.634090909090909, + "grad_norm": 0.47810637950897217, + "learning_rate": 7.683762888783977e-07, + "loss": 0.4265, + "step": 677 + }, + { + "epoch": 4.640909090909091, + "grad_norm": 0.47082048654556274, + "learning_rate": 7.610904020526938e-07, + "loss": 0.4229, + "step": 678 + }, + { + "epoch": 4.6477272727272725, + "grad_norm": 0.4664064645767212, + "learning_rate": 7.538330157424212e-07, + "loss": 0.4418, + "step": 679 + }, + { + "epoch": 4.654545454545454, + "grad_norm": 0.44690611958503723, + "learning_rate": 7.466042488952521e-07, + "loss": 0.4288, + "step": 680 + }, + { + "epoch": 4.661363636363636, + "grad_norm": 0.4714190661907196, + "learning_rate": 7.394042199897916e-07, + "loss": 0.4476, + "step": 681 + }, + { + "epoch": 4.668181818181818, + "grad_norm": 0.49196168780326843, + "learning_rate": 7.322330470336314e-07, + "loss": 0.4224, + "step": 682 + }, + { + "epoch": 4.675, + "grad_norm": 0.5213456153869629, + "learning_rate": 7.250908475614185e-07, + "loss": 0.4044, + "step": 683 + }, + { + "epoch": 4.681818181818182, + "grad_norm": 0.4465913772583008, + "learning_rate": 7.179777386329276e-07, + "loss": 0.411, + "step": 684 + }, + { + "epoch": 4.6886363636363635, + "grad_norm": 0.49001193046569824, + "learning_rate": 7.108938368311424e-07, + "loss": 0.4556, + "step": 685 + }, + { + "epoch": 4.695454545454545, + "grad_norm": 0.4741291105747223, + "learning_rate": 7.038392582603481e-07, + "loss": 0.4537, + "step": 686 + }, + { + "epoch": 4.702272727272727, + "grad_norm": 0.48216235637664795, + "learning_rate": 6.968141185442229e-07, + "loss": 0.4187, + "step": 687 + }, + { + "epoch": 4.709090909090909, + "grad_norm": 0.45541542768478394, + "learning_rate": 6.898185328239468e-07, + "loss": 0.4094, + "step": 688 + }, + { + "epoch": 4.715909090909091, + "grad_norm": 0.4746590256690979, + "learning_rate": 6.828526157563126e-07, + "loss": 0.4269, + "step": 689 + }, + { + "epoch": 4.722727272727273, + "grad_norm": 0.45999908447265625, + "learning_rate": 6.759164815118493e-07, + "loss": 0.4325, + "step": 690 + }, + { + "epoch": 4.7295454545454545, + "grad_norm": 0.5204467177391052, + "learning_rate": 6.690102437729481e-07, + "loss": 0.4407, + "step": 691 + }, + { + "epoch": 4.736363636363636, + "grad_norm": 0.45999932289123535, + "learning_rate": 6.621340157319998e-07, + "loss": 0.4365, + "step": 692 + }, + { + "epoch": 4.743181818181818, + "grad_norm": 0.4780198931694031, + "learning_rate": 6.552879100895396e-07, + "loss": 0.4207, + "step": 693 + }, + { + "epoch": 4.75, + "grad_norm": 0.4917505979537964, + "learning_rate": 6.484720390524008e-07, + "loss": 0.4299, + "step": 694 + }, + { + "epoch": 4.756818181818182, + "grad_norm": 0.49415823817253113, + "learning_rate": 6.416865143318757e-07, + "loss": 0.4356, + "step": 695 + }, + { + "epoch": 4.763636363636364, + "grad_norm": 0.46500974893569946, + "learning_rate": 6.349314471418849e-07, + "loss": 0.4465, + "step": 696 + }, + { + "epoch": 4.7704545454545455, + "grad_norm": 0.4365505874156952, + "learning_rate": 6.282069481971514e-07, + "loss": 0.4205, + "step": 697 + }, + { + "epoch": 4.777272727272727, + "grad_norm": 0.4652706980705261, + "learning_rate": 6.2151312771139e-07, + "loss": 0.4385, + "step": 698 + }, + { + "epoch": 4.784090909090909, + "grad_norm": 0.4932256042957306, + "learning_rate": 6.148500953954992e-07, + "loss": 0.4056, + "step": 699 + }, + { + "epoch": 4.790909090909091, + "grad_norm": 0.8236297965049744, + "learning_rate": 6.082179604557617e-07, + "loss": 0.4147, + "step": 700 + }, + { + "epoch": 4.797727272727273, + "grad_norm": 0.6776428818702698, + "learning_rate": 6.016168315920593e-07, + "loss": 0.4111, + "step": 701 + }, + { + "epoch": 4.804545454545455, + "grad_norm": 0.4967111647129059, + "learning_rate": 5.950468169960846e-07, + "loss": 0.4467, + "step": 702 + }, + { + "epoch": 4.8113636363636365, + "grad_norm": 0.48729217052459717, + "learning_rate": 5.885080243495731e-07, + "loss": 0.4398, + "step": 703 + }, + { + "epoch": 4.818181818181818, + "grad_norm": 0.5257642865180969, + "learning_rate": 5.820005608225345e-07, + "loss": 0.4076, + "step": 704 + }, + { + "epoch": 4.825, + "grad_norm": 0.5564248561859131, + "learning_rate": 5.755245330715014e-07, + "loss": 0.4304, + "step": 705 + }, + { + "epoch": 4.831818181818182, + "grad_norm": 0.4611823260784149, + "learning_rate": 5.690800472377747e-07, + "loss": 0.4357, + "step": 706 + }, + { + "epoch": 4.838636363636364, + "grad_norm": 0.46857500076293945, + "learning_rate": 5.626672089456887e-07, + "loss": 0.4373, + "step": 707 + }, + { + "epoch": 4.845454545454546, + "grad_norm": 0.49044084548950195, + "learning_rate": 5.562861233008774e-07, + "loss": 0.4299, + "step": 708 + }, + { + "epoch": 4.8522727272727275, + "grad_norm": 0.4881190061569214, + "learning_rate": 5.499368948885528e-07, + "loss": 0.4306, + "step": 709 + }, + { + "epoch": 4.859090909090909, + "grad_norm": 0.4680595397949219, + "learning_rate": 5.436196277717928e-07, + "loss": 0.4145, + "step": 710 + }, + { + "epoch": 4.865909090909091, + "grad_norm": 0.45758605003356934, + "learning_rate": 5.373344254898313e-07, + "loss": 0.4294, + "step": 711 + }, + { + "epoch": 4.872727272727273, + "grad_norm": 0.4999527335166931, + "learning_rate": 5.310813910563645e-07, + "loss": 0.422, + "step": 712 + }, + { + "epoch": 4.879545454545455, + "grad_norm": 0.4778689444065094, + "learning_rate": 5.24860626957861e-07, + "loss": 0.4252, + "step": 713 + }, + { + "epoch": 4.886363636363637, + "grad_norm": 0.4799656569957733, + "learning_rate": 5.186722351518822e-07, + "loss": 0.4309, + "step": 714 + }, + { + "epoch": 4.8931818181818185, + "grad_norm": 0.4888315200805664, + "learning_rate": 5.125163170654138e-07, + "loss": 0.4123, + "step": 715 + }, + { + "epoch": 4.9, + "grad_norm": 0.49795007705688477, + "learning_rate": 5.063929735931985e-07, + "loss": 0.3995, + "step": 716 + }, + { + "epoch": 4.906818181818182, + "grad_norm": 0.5609537959098816, + "learning_rate": 5.003023050960865e-07, + "loss": 0.417, + "step": 717 + }, + { + "epoch": 4.913636363636364, + "grad_norm": 0.5369165539741516, + "learning_rate": 4.94244411399388e-07, + "loss": 0.4143, + "step": 718 + }, + { + "epoch": 4.920454545454545, + "grad_norm": 0.48569849133491516, + "learning_rate": 4.882193917912398e-07, + "loss": 0.4603, + "step": 719 + }, + { + "epoch": 4.927272727272728, + "grad_norm": 0.5146452188491821, + "learning_rate": 4.822273450209767e-07, + "loss": 0.406, + "step": 720 + }, + { + "epoch": 4.934090909090909, + "grad_norm": 0.4748017191886902, + "learning_rate": 4.7626836929751035e-07, + "loss": 0.4496, + "step": 721 + }, + { + "epoch": 4.940909090909091, + "grad_norm": 0.48057058453559875, + "learning_rate": 4.703425622877239e-07, + "loss": 0.4237, + "step": 722 + }, + { + "epoch": 4.947727272727272, + "grad_norm": 0.4811655580997467, + "learning_rate": 4.6445002111486866e-07, + "loss": 0.4429, + "step": 723 + }, + { + "epoch": 4.954545454545455, + "grad_norm": 0.45915886759757996, + "learning_rate": 4.5859084235697236e-07, + "loss": 0.4407, + "step": 724 + }, + { + "epoch": 4.961363636363636, + "grad_norm": 0.47130313515663147, + "learning_rate": 4.527651220452589e-07, + "loss": 0.4177, + "step": 725 + }, + { + "epoch": 4.968181818181818, + "grad_norm": 0.49092698097229004, + "learning_rate": 4.469729556625704e-07, + "loss": 0.4204, + "step": 726 + }, + { + "epoch": 4.975, + "grad_norm": 0.4968926012516022, + "learning_rate": 4.412144381418049e-07, + "loss": 0.4517, + "step": 727 + }, + { + "epoch": 4.9818181818181815, + "grad_norm": 0.4605294466018677, + "learning_rate": 4.354896638643591e-07, + "loss": 0.4377, + "step": 728 + }, + { + "epoch": 4.988636363636363, + "grad_norm": 0.5322148203849792, + "learning_rate": 4.2979872665858266e-07, + "loss": 0.4263, + "step": 729 + }, + { + "epoch": 4.995454545454545, + "grad_norm": 0.5127450823783875, + "learning_rate": 4.2414171979824e-07, + "loss": 0.4043, + "step": 730 + }, + { + "epoch": 5.006818181818182, + "grad_norm": 0.9531456232070923, + "learning_rate": 4.1851873600098154e-07, + "loss": 0.7871, + "step": 731 + }, + { + "epoch": 5.013636363636364, + "grad_norm": 0.4543856978416443, + "learning_rate": 4.129298674268226e-07, + "loss": 0.4263, + "step": 732 + }, + { + "epoch": 5.0204545454545455, + "grad_norm": 0.4518638551235199, + "learning_rate": 4.073752056766342e-07, + "loss": 0.4227, + "step": 733 + }, + { + "epoch": 5.027272727272727, + "grad_norm": 0.46829870343208313, + "learning_rate": 4.0185484179064427e-07, + "loss": 0.4028, + "step": 734 + }, + { + "epoch": 5.034090909090909, + "grad_norm": 0.4828706979751587, + "learning_rate": 3.9636886624694e-07, + "loss": 0.4014, + "step": 735 + }, + { + "epoch": 5.040909090909091, + "grad_norm": 0.44380271434783936, + "learning_rate": 3.9091736895998907e-07, + "loss": 0.4275, + "step": 736 + }, + { + "epoch": 5.047727272727273, + "grad_norm": 0.4777894914150238, + "learning_rate": 3.855004392791645e-07, + "loss": 0.3705, + "step": 737 + }, + { + "epoch": 5.054545454545455, + "grad_norm": 0.47452741861343384, + "learning_rate": 3.801181659872805e-07, + "loss": 0.4085, + "step": 738 + }, + { + "epoch": 5.0613636363636365, + "grad_norm": 0.459635853767395, + "learning_rate": 3.7477063729913804e-07, + "loss": 0.4242, + "step": 739 + }, + { + "epoch": 5.068181818181818, + "grad_norm": 0.43549954891204834, + "learning_rate": 3.6945794086007706e-07, + "loss": 0.4063, + "step": 740 + }, + { + "epoch": 5.075, + "grad_norm": 0.777911365032196, + "learning_rate": 3.6418016374454247e-07, + "loss": 0.4126, + "step": 741 + }, + { + "epoch": 5.081818181818182, + "grad_norm": 0.45801717042922974, + "learning_rate": 3.5893739245465465e-07, + "loss": 0.4215, + "step": 742 + }, + { + "epoch": 5.088636363636364, + "grad_norm": 0.45145151019096375, + "learning_rate": 3.537297129187925e-07, + "loss": 0.435, + "step": 743 + }, + { + "epoch": 5.095454545454546, + "grad_norm": 0.47909021377563477, + "learning_rate": 3.485572104901869e-07, + "loss": 0.4252, + "step": 744 + }, + { + "epoch": 5.1022727272727275, + "grad_norm": 0.5086826086044312, + "learning_rate": 3.4341996994551957e-07, + "loss": 0.4329, + "step": 745 + }, + { + "epoch": 5.109090909090909, + "grad_norm": 0.4911805987358093, + "learning_rate": 3.383180754835344e-07, + "loss": 0.4362, + "step": 746 + }, + { + "epoch": 5.115909090909091, + "grad_norm": 0.47065818309783936, + "learning_rate": 3.3325161072365636e-07, + "loss": 0.4137, + "step": 747 + }, + { + "epoch": 5.122727272727273, + "grad_norm": 0.49130091071128845, + "learning_rate": 3.2822065870462216e-07, + "loss": 0.3959, + "step": 748 + }, + { + "epoch": 5.129545454545455, + "grad_norm": 0.49359121918678284, + "learning_rate": 3.232253018831208e-07, + "loss": 0.4035, + "step": 749 + }, + { + "epoch": 5.136363636363637, + "grad_norm": 0.6447023153305054, + "learning_rate": 3.182656221324384e-07, + "loss": 0.4213, + "step": 750 + }, + { + "epoch": 5.1431818181818185, + "grad_norm": 0.4835764169692993, + "learning_rate": 3.133417007411188e-07, + "loss": 0.4251, + "step": 751 + }, + { + "epoch": 5.15, + "grad_norm": 0.4686364233493805, + "learning_rate": 3.08453618411631e-07, + "loss": 0.4309, + "step": 752 + }, + { + "epoch": 5.156818181818182, + "grad_norm": 0.4404386281967163, + "learning_rate": 3.036014552590455e-07, + "loss": 0.4154, + "step": 753 + }, + { + "epoch": 5.163636363636364, + "grad_norm": 0.4862096309661865, + "learning_rate": 2.98785290809723e-07, + "loss": 0.4167, + "step": 754 + }, + { + "epoch": 5.170454545454546, + "grad_norm": 0.49654898047447205, + "learning_rate": 2.940052040000091e-07, + "loss": 0.4134, + "step": 755 + }, + { + "epoch": 5.177272727272728, + "grad_norm": 0.5038694143295288, + "learning_rate": 2.892612731749414e-07, + "loss": 0.4255, + "step": 756 + }, + { + "epoch": 5.184090909090909, + "grad_norm": 0.5015192031860352, + "learning_rate": 2.8455357608696497e-07, + "loss": 0.4189, + "step": 757 + }, + { + "epoch": 5.190909090909091, + "grad_norm": 0.47371554374694824, + "learning_rate": 2.798821898946588e-07, + "loss": 0.4279, + "step": 758 + }, + { + "epoch": 5.197727272727272, + "grad_norm": 0.4441058337688446, + "learning_rate": 2.7524719116147154e-07, + "loss": 0.4208, + "step": 759 + }, + { + "epoch": 5.204545454545454, + "grad_norm": 0.4669174551963806, + "learning_rate": 2.706486558544644e-07, + "loss": 0.4119, + "step": 760 + }, + { + "epoch": 5.211363636363636, + "grad_norm": 0.4658101499080658, + "learning_rate": 2.6608665934306775e-07, + "loss": 0.4137, + "step": 761 + }, + { + "epoch": 5.218181818181818, + "grad_norm": 0.4596281945705414, + "learning_rate": 2.615612763978462e-07, + "loss": 0.4194, + "step": 762 + }, + { + "epoch": 5.225, + "grad_norm": 0.5125681161880493, + "learning_rate": 2.570725811892727e-07, + "loss": 0.4307, + "step": 763 + }, + { + "epoch": 5.2318181818181815, + "grad_norm": 0.4651528298854828, + "learning_rate": 2.52620647286512e-07, + "loss": 0.4245, + "step": 764 + }, + { + "epoch": 5.238636363636363, + "grad_norm": 0.712844967842102, + "learning_rate": 2.4820554765621534e-07, + "loss": 0.4046, + "step": 765 + }, + { + "epoch": 5.245454545454545, + "grad_norm": 0.49198994040489197, + "learning_rate": 2.438273546613257e-07, + "loss": 0.4103, + "step": 766 + }, + { + "epoch": 5.252272727272727, + "grad_norm": 0.4724070429801941, + "learning_rate": 2.394861400598894e-07, + "loss": 0.4148, + "step": 767 + }, + { + "epoch": 5.259090909090909, + "grad_norm": 0.5466293692588806, + "learning_rate": 2.3518197500388278e-07, + "loss": 0.4133, + "step": 768 + }, + { + "epoch": 5.265909090909091, + "grad_norm": 0.47871315479278564, + "learning_rate": 2.3091493003804476e-07, + "loss": 0.4248, + "step": 769 + }, + { + "epoch": 5.2727272727272725, + "grad_norm": 0.4563087821006775, + "learning_rate": 2.2668507509871957e-07, + "loss": 0.4173, + "step": 770 + }, + { + "epoch": 5.279545454545454, + "grad_norm": 0.47406792640686035, + "learning_rate": 2.2249247951271174e-07, + "loss": 0.4223, + "step": 771 + }, + { + "epoch": 5.286363636363636, + "grad_norm": 0.5211136341094971, + "learning_rate": 2.1833721199614992e-07, + "loss": 0.4335, + "step": 772 + }, + { + "epoch": 5.293181818181818, + "grad_norm": 0.5313299298286438, + "learning_rate": 2.1421934065335909e-07, + "loss": 0.4204, + "step": 773 + }, + { + "epoch": 5.3, + "grad_norm": 0.48682135343551636, + "learning_rate": 2.1013893297574777e-07, + "loss": 0.4249, + "step": 774 + }, + { + "epoch": 5.306818181818182, + "grad_norm": 0.5159818530082703, + "learning_rate": 2.0609605584069741e-07, + "loss": 0.4112, + "step": 775 + }, + { + "epoch": 5.3136363636363635, + "grad_norm": 0.4795214831829071, + "learning_rate": 2.020907755104698e-07, + "loss": 0.4219, + "step": 776 + }, + { + "epoch": 5.320454545454545, + "grad_norm": 0.4812486469745636, + "learning_rate": 1.9812315763111823e-07, + "loss": 0.4191, + "step": 777 + }, + { + "epoch": 5.327272727272727, + "grad_norm": 0.487095445394516, + "learning_rate": 1.9419326723141534e-07, + "loss": 0.4005, + "step": 778 + }, + { + "epoch": 5.334090909090909, + "grad_norm": 0.44063085317611694, + "learning_rate": 1.9030116872178317e-07, + "loss": 0.4108, + "step": 779 + }, + { + "epoch": 5.340909090909091, + "grad_norm": 0.4788059592247009, + "learning_rate": 1.864469258932397e-07, + "loss": 0.4072, + "step": 780 + }, + { + "epoch": 5.347727272727273, + "grad_norm": 0.4548879861831665, + "learning_rate": 1.8263060191635317e-07, + "loss": 0.4173, + "step": 781 + }, + { + "epoch": 5.3545454545454545, + "grad_norm": 0.480910986661911, + "learning_rate": 1.788522593402059e-07, + "loss": 0.4144, + "step": 782 + }, + { + "epoch": 5.361363636363636, + "grad_norm": 0.4492330849170685, + "learning_rate": 1.7511196009137087e-07, + "loss": 0.4193, + "step": 783 + }, + { + "epoch": 5.368181818181818, + "grad_norm": 0.4803030490875244, + "learning_rate": 1.7140976547289438e-07, + "loss": 0.4263, + "step": 784 + }, + { + "epoch": 5.375, + "grad_norm": 0.4628124535083771, + "learning_rate": 1.6774573616329336e-07, + "loss": 0.4152, + "step": 785 + }, + { + "epoch": 5.381818181818182, + "grad_norm": 0.454474538564682, + "learning_rate": 1.6411993221555928e-07, + "loss": 0.4152, + "step": 786 + }, + { + "epoch": 5.388636363636364, + "grad_norm": 0.6482933163642883, + "learning_rate": 1.605324130561753e-07, + "loss": 0.413, + "step": 787 + }, + { + "epoch": 5.3954545454545455, + "grad_norm": 0.5180197954177856, + "learning_rate": 1.5698323748414123e-07, + "loss": 0.4422, + "step": 788 + }, + { + "epoch": 5.402272727272727, + "grad_norm": 0.46103981137275696, + "learning_rate": 1.5347246367000994e-07, + "loss": 0.4467, + "step": 789 + }, + { + "epoch": 5.409090909090909, + "grad_norm": 0.47345975041389465, + "learning_rate": 1.5000014915493467e-07, + "loss": 0.4096, + "step": 790 + }, + { + "epoch": 5.415909090909091, + "grad_norm": 0.6126012206077576, + "learning_rate": 1.4656635084972475e-07, + "loss": 0.432, + "step": 791 + }, + { + "epoch": 5.422727272727273, + "grad_norm": 0.45674052834510803, + "learning_rate": 1.4317112503391433e-07, + "loss": 0.4225, + "step": 792 + }, + { + "epoch": 5.429545454545455, + "grad_norm": 0.6501134037971497, + "learning_rate": 1.398145273548396e-07, + "loss": 0.4222, + "step": 793 + }, + { + "epoch": 5.4363636363636365, + "grad_norm": 0.499102920293808, + "learning_rate": 1.3649661282672478e-07, + "loss": 0.4308, + "step": 794 + }, + { + "epoch": 5.443181818181818, + "grad_norm": 0.4972813129425049, + "learning_rate": 1.3321743582978303e-07, + "loss": 0.4114, + "step": 795 + }, + { + "epoch": 5.45, + "grad_norm": 1.076395034790039, + "learning_rate": 1.2997705010932394e-07, + "loss": 0.4009, + "step": 796 + }, + { + "epoch": 5.456818181818182, + "grad_norm": 0.4762725234031677, + "learning_rate": 1.2677550877487232e-07, + "loss": 0.4012, + "step": 797 + }, + { + "epoch": 5.463636363636364, + "grad_norm": 0.47878921031951904, + "learning_rate": 1.2361286429929953e-07, + "loss": 0.4252, + "step": 798 + }, + { + "epoch": 5.470454545454546, + "grad_norm": 0.5394876599311829, + "learning_rate": 1.20489168517961e-07, + "loss": 0.421, + "step": 799 + }, + { + "epoch": 5.4772727272727275, + "grad_norm": 0.47245123982429504, + "learning_rate": 1.1740447262784782e-07, + "loss": 0.4265, + "step": 800 + }, + { + "epoch": 5.484090909090909, + "grad_norm": 0.45624515414237976, + "learning_rate": 1.1435882718674823e-07, + "loss": 0.4296, + "step": 801 + }, + { + "epoch": 5.490909090909091, + "grad_norm": 0.4749613106250763, + "learning_rate": 1.1135228211241827e-07, + "loss": 0.4202, + "step": 802 + }, + { + "epoch": 5.497727272727273, + "grad_norm": 0.4714772403240204, + "learning_rate": 1.0838488668176383e-07, + "loss": 0.4134, + "step": 803 + }, + { + "epoch": 5.504545454545455, + "grad_norm": 0.4895274341106415, + "learning_rate": 1.054566895300324e-07, + "loss": 0.4187, + "step": 804 + }, + { + "epoch": 5.511363636363637, + "grad_norm": 0.4979075789451599, + "learning_rate": 1.0256773865001679e-07, + "loss": 0.4181, + "step": 805 + }, + { + "epoch": 5.5181818181818185, + "grad_norm": 0.4756965935230255, + "learning_rate": 9.97180813912682e-08, + "loss": 0.4016, + "step": 806 + }, + { + "epoch": 5.525, + "grad_norm": 0.5017597079277039, + "learning_rate": 9.690776445932082e-08, + "loss": 0.4203, + "step": 807 + }, + { + "epoch": 5.531818181818182, + "grad_norm": 0.47631561756134033, + "learning_rate": 9.413683391492456e-08, + "loss": 0.4325, + "step": 808 + }, + { + "epoch": 5.538636363636364, + "grad_norm": 0.4681398570537567, + "learning_rate": 9.140533517329214e-08, + "loss": 0.4154, + "step": 809 + }, + { + "epoch": 5.545454545454545, + "grad_norm": 0.5909439325332642, + "learning_rate": 8.871331300335322e-08, + "loss": 0.4075, + "step": 810 + }, + { + "epoch": 5.552272727272728, + "grad_norm": 0.48725610971450806, + "learning_rate": 8.606081152702145e-08, + "loss": 0.4231, + "step": 811 + }, + { + "epoch": 5.559090909090909, + "grad_norm": 0.47185027599334717, + "learning_rate": 8.344787421847216e-08, + "loss": 0.4252, + "step": 812 + }, + { + "epoch": 5.565909090909091, + "grad_norm": 0.45850270986557007, + "learning_rate": 8.087454390342725e-08, + "loss": 0.428, + "step": 813 + }, + { + "epoch": 5.572727272727272, + "grad_norm": 0.7553633451461792, + "learning_rate": 7.834086275845587e-08, + "loss": 0.4284, + "step": 814 + }, + { + "epoch": 5.579545454545455, + "grad_norm": 0.48967990279197693, + "learning_rate": 7.584687231028121e-08, + "loss": 0.4284, + "step": 815 + }, + { + "epoch": 5.586363636363636, + "grad_norm": 0.4823754131793976, + "learning_rate": 7.339261343510207e-08, + "loss": 0.4191, + "step": 816 + }, + { + "epoch": 5.593181818181818, + "grad_norm": 0.4952954053878784, + "learning_rate": 7.097812635792095e-08, + "loss": 0.4237, + "step": 817 + }, + { + "epoch": 5.6, + "grad_norm": 0.5452063679695129, + "learning_rate": 6.860345065188512e-08, + "loss": 0.4106, + "step": 818 + }, + { + "epoch": 5.6068181818181815, + "grad_norm": 0.46396100521087646, + "learning_rate": 6.626862523763905e-08, + "loss": 0.4049, + "step": 819 + }, + { + "epoch": 5.613636363636363, + "grad_norm": 0.4779062569141388, + "learning_rate": 6.397368838268497e-08, + "loss": 0.4227, + "step": 820 + }, + { + "epoch": 5.620454545454545, + "grad_norm": 0.49288275837898254, + "learning_rate": 6.171867770075723e-08, + "loss": 0.4134, + "step": 821 + }, + { + "epoch": 5.627272727272727, + "grad_norm": 0.46809956431388855, + "learning_rate": 5.9503630151205025e-08, + "loss": 0.4396, + "step": 822 + }, + { + "epoch": 5.634090909090909, + "grad_norm": 0.4755077362060547, + "learning_rate": 5.7328582038386483e-08, + "loss": 0.4132, + "step": 823 + }, + { + "epoch": 5.640909090909091, + "grad_norm": 0.4648922085762024, + "learning_rate": 5.519356901107359e-08, + "loss": 0.4075, + "step": 824 + }, + { + "epoch": 5.6477272727272725, + "grad_norm": 0.4905151426792145, + "learning_rate": 5.309862606186877e-08, + "loss": 0.4087, + "step": 825 + }, + { + "epoch": 5.654545454545454, + "grad_norm": 0.5978060364723206, + "learning_rate": 5.104378752663008e-08, + "loss": 0.4048, + "step": 826 + }, + { + "epoch": 5.661363636363636, + "grad_norm": 0.47629672288894653, + "learning_rate": 4.902908708391024e-08, + "loss": 0.4174, + "step": 827 + }, + { + "epoch": 5.668181818181818, + "grad_norm": 0.48573756217956543, + "learning_rate": 4.705455775440237e-08, + "loss": 0.4056, + "step": 828 + }, + { + "epoch": 5.675, + "grad_norm": 0.4741798937320709, + "learning_rate": 4.5120231900400715e-08, + "loss": 0.4438, + "step": 829 + }, + { + "epoch": 5.681818181818182, + "grad_norm": 0.49907681345939636, + "learning_rate": 4.3226141225268804e-08, + "loss": 0.4335, + "step": 830 + }, + { + "epoch": 5.6886363636363635, + "grad_norm": 0.5794979929924011, + "learning_rate": 4.1372316772921584e-08, + "loss": 0.4255, + "step": 831 + }, + { + "epoch": 5.695454545454545, + "grad_norm": 0.46549633145332336, + "learning_rate": 3.955878892731441e-08, + "loss": 0.4064, + "step": 832 + }, + { + "epoch": 5.702272727272727, + "grad_norm": 0.6486341953277588, + "learning_rate": 3.778558741194677e-08, + "loss": 0.3878, + "step": 833 + }, + { + "epoch": 5.709090909090909, + "grad_norm": 0.5156870484352112, + "learning_rate": 3.605274128937464e-08, + "loss": 0.4195, + "step": 834 + }, + { + "epoch": 5.715909090909091, + "grad_norm": 0.4981493353843689, + "learning_rate": 3.436027896073307e-08, + "loss": 0.4073, + "step": 835 + }, + { + "epoch": 5.722727272727273, + "grad_norm": 0.5782902240753174, + "learning_rate": 3.270822816527325e-08, + "loss": 0.4278, + "step": 836 + }, + { + "epoch": 5.7295454545454545, + "grad_norm": 0.45053598284721375, + "learning_rate": 3.109661597990532e-08, + "loss": 0.4083, + "step": 837 + }, + { + "epoch": 5.736363636363636, + "grad_norm": 0.45860233902931213, + "learning_rate": 2.9525468818755455e-08, + "loss": 0.4005, + "step": 838 + }, + { + "epoch": 5.743181818181818, + "grad_norm": 0.48783180117607117, + "learning_rate": 2.7994812432733664e-08, + "loss": 0.4283, + "step": 839 + }, + { + "epoch": 5.75, + "grad_norm": 0.517410397529602, + "learning_rate": 2.6504671909109993e-08, + "loss": 0.4151, + "step": 840 + }, + { + "epoch": 5.756818181818182, + "grad_norm": 0.6215181946754456, + "learning_rate": 2.5055071671105936e-08, + "loss": 0.4096, + "step": 841 + }, + { + "epoch": 5.763636363636364, + "grad_norm": 0.46469247341156006, + "learning_rate": 2.3646035477491726e-08, + "loss": 0.4344, + "step": 842 + }, + { + "epoch": 5.7704545454545455, + "grad_norm": 0.47195473313331604, + "learning_rate": 2.2277586422198017e-08, + "loss": 0.4018, + "step": 843 + }, + { + "epoch": 5.777272727272727, + "grad_norm": 0.5247776508331299, + "learning_rate": 2.094974693393731e-08, + "loss": 0.448, + "step": 844 + }, + { + "epoch": 5.784090909090909, + "grad_norm": 0.4573938846588135, + "learning_rate": 1.9662538775836182e-08, + "loss": 0.428, + "step": 845 + }, + { + "epoch": 5.790909090909091, + "grad_norm": 0.6799011826515198, + "learning_rate": 1.841598304507891e-08, + "loss": 0.4275, + "step": 846 + }, + { + "epoch": 5.797727272727273, + "grad_norm": 0.4752521216869354, + "learning_rate": 1.7210100172561084e-08, + "loss": 0.4259, + "step": 847 + }, + { + "epoch": 5.804545454545455, + "grad_norm": 0.48902708292007446, + "learning_rate": 1.6044909922555973e-08, + "loss": 0.408, + "step": 848 + }, + { + "epoch": 5.8113636363636365, + "grad_norm": 0.5219933986663818, + "learning_rate": 1.4920431392388412e-08, + "loss": 0.4159, + "step": 849 + }, + { + "epoch": 5.818181818181818, + "grad_norm": 0.5393177270889282, + "learning_rate": 1.383668301212393e-08, + "loss": 0.4214, + "step": 850 + }, + { + "epoch": 5.825, + "grad_norm": 0.49679329991340637, + "learning_rate": 1.2793682544266216e-08, + "loss": 0.4086, + "step": 851 + }, + { + "epoch": 5.831818181818182, + "grad_norm": 0.5389913320541382, + "learning_rate": 1.1791447083465136e-08, + "loss": 0.4377, + "step": 852 + }, + { + "epoch": 5.838636363636364, + "grad_norm": 0.5072413682937622, + "learning_rate": 1.0829993056236942e-08, + "loss": 0.4084, + "step": 853 + }, + { + "epoch": 5.845454545454546, + "grad_norm": 0.5199616551399231, + "learning_rate": 9.90933622069562e-09, + "loss": 0.426, + "step": 854 + }, + { + "epoch": 5.8522727272727275, + "grad_norm": 0.5511140823364258, + "learning_rate": 9.029491666293911e-09, + "loss": 0.3893, + "step": 855 + }, + { + "epoch": 5.859090909090909, + "grad_norm": 0.5552690625190735, + "learning_rate": 8.190473813576571e-09, + "loss": 0.4125, + "step": 856 + }, + { + "epoch": 5.865909090909091, + "grad_norm": 0.47848770022392273, + "learning_rate": 7.3922964139433455e-09, + "loss": 0.4205, + "step": 857 + }, + { + "epoch": 5.872727272727273, + "grad_norm": 4.603728294372559, + "learning_rate": 6.634972549423857e-09, + "loss": 0.4449, + "step": 858 + }, + { + "epoch": 5.879545454545455, + "grad_norm": 0.5062354207038879, + "learning_rate": 5.918514632463901e-09, + "loss": 0.4443, + "step": 859 + }, + { + "epoch": 5.886363636363637, + "grad_norm": 0.5135608911514282, + "learning_rate": 5.242934405720879e-09, + "loss": 0.4404, + "step": 860 + }, + { + "epoch": 5.8931818181818185, + "grad_norm": 0.5170268416404724, + "learning_rate": 4.6082429418720095e-09, + "loss": 0.3964, + "step": 861 + }, + { + "epoch": 5.9, + "grad_norm": 0.6158047318458557, + "learning_rate": 4.01445064343281e-09, + "loss": 0.4144, + "step": 862 + }, + { + "epoch": 5.906818181818182, + "grad_norm": 0.4549572765827179, + "learning_rate": 3.4615672425861167e-09, + "loss": 0.4034, + "step": 863 + }, + { + "epoch": 5.913636363636364, + "grad_norm": 0.4846479594707489, + "learning_rate": 2.9496018010233275e-09, + "loss": 0.4022, + "step": 864 + }, + { + "epoch": 5.920454545454545, + "grad_norm": 0.5221410393714905, + "learning_rate": 2.478562709795074e-09, + "loss": 0.4272, + "step": 865 + }, + { + "epoch": 5.927272727272728, + "grad_norm": 0.4993201494216919, + "learning_rate": 2.048457689174943e-09, + "loss": 0.4245, + "step": 866 + }, + { + "epoch": 5.934090909090909, + "grad_norm": 0.45889461040496826, + "learning_rate": 1.6592937885312466e-09, + "loss": 0.4059, + "step": 867 + }, + { + "epoch": 5.940909090909091, + "grad_norm": 0.48042938113212585, + "learning_rate": 1.3110773862126669e-09, + "loss": 0.4311, + "step": 868 + }, + { + "epoch": 5.947727272727272, + "grad_norm": 0.5720022916793823, + "learning_rate": 1.0038141894436192e-09, + "loss": 0.427, + "step": 869 + }, + { + "epoch": 5.954545454545455, + "grad_norm": 0.4770817756652832, + "learning_rate": 7.375092342298828e-10, + "loss": 0.418, + "step": 870 + }, + { + "epoch": 5.961363636363636, + "grad_norm": 0.4689801037311554, + "learning_rate": 5.121668852775541e-10, + "loss": 0.4092, + "step": 871 + }, + { + "epoch": 5.968181818181818, + "grad_norm": 0.46112075448036194, + "learning_rate": 3.277908359194948e-10, + "loss": 0.435, + "step": 872 + }, + { + "epoch": 5.975, + "grad_norm": 0.45823365449905396, + "learning_rate": 1.8438410805732277e-10, + "loss": 0.4122, + "step": 873 + }, + { + "epoch": 5.9818181818181815, + "grad_norm": 0.5599728226661682, + "learning_rate": 8.194905210923143e-11, + "loss": 0.4134, + "step": 874 + }, + { + "epoch": 5.988636363636363, + "grad_norm": 0.4715409278869629, + "learning_rate": 2.0487346973629975e-11, + "loss": 0.437, + "step": 875 + }, + { + "epoch": 5.995454545454545, + "grad_norm": 0.5697607398033142, + "learning_rate": 0.0, + "loss": 0.3933, + "step": 876 + } + ], + "logging_steps": 1, + "max_steps": 876, + "num_input_tokens_seen": 0, + "num_train_epochs": 6, + "save_steps": 146, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 5.646190516718744e+19, + "train_batch_size": 4, + "trial_name": null, + "trial_params": null +} diff --git a/checkpoint-876/training_args.bin b/checkpoint-876/training_args.bin new file mode 100644 index 0000000000000000000000000000000000000000..b41c805059a36d8d2e3bf3511fb0fd95b74dae74 --- /dev/null +++ b/checkpoint-876/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:68945b8f9934f5870e67a2d4a20c31fa8e742da41075fa7236fb313c6197456f +size 7928 diff --git a/checkpoint-876/zero_to_fp32.py b/checkpoint-876/zero_to_fp32.py new file mode 100644 index 0000000000000000000000000000000000000000..24cc342e78d1a006c782b3a4cd68d9ce786d8fd8 --- /dev/null +++ b/checkpoint-876/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..9d764aead97a826f7aae3b9cfcfe2606e1d2eeec --- /dev/null +++ b/config.json @@ -0,0 +1,52 @@ +{ + "_attn_implementation_autoset": true, + "_name_or_path": "meta-llama/Llama-3.3-70B-Instruct", + "architectures": [ + "LlamaForCausalLM" + ], + "attention_bias": false, + "attention_dropout": 0.0, + "bos_token_id": 128000, + "eos_token_id": 128009, + "head_dim": 128, + "hidden_act": "silu", + "hidden_size": 8192, + "initializer_range": 0.02, + "intermediate_size": 28672, + "max_position_embeddings": 131072, + "mlp_bias": false, + "model_type": "llama", + "num_attention_heads": 64, + "num_hidden_layers": 80, + "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..278b7f0f84be865c4687700ee7b3c63d89a51e18 --- /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": "<|eot_id|>", + "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..ca91a2ef55f4239a7af81d7c9abb05f53621a07b --- /dev/null +++ b/tokenizer_config.json @@ -0,0 +1,2064 @@ +{ + "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|>", + "chat_template": "{{- bos_token }}\n{%- if custom_tools is defined %}\n {%- set tools = custom_tools %}\n{%- endif %}\n{%- if not tools_in_user_message is defined %}\n {%- set tools_in_user_message = true %}\n{%- endif %}\n{%- if not date_string is defined %}\n {%- set date_string = \"26 Jul 2024\" %}\n{%- endif %}\n{%- if not tools is defined %}\n {%- set tools = none %}\n{%- endif %}\n\n{#- This block extracts the system message, so we can slot it into the right place. #}\n{%- if messages[0]['role'] == 'system' %}\n {%- set system_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n{%- else %}\n {%- set system_message = \"\" %}\n{%- endif %}\n\n{#- System message + builtin tools #}\n{{- \"<|start_header_id|>system<|end_header_id|>\\n\\n\" }}\n{%- if builtin_tools is defined or tools is not none %}\n {{- \"Environment: ipython\\n\" }}\n{%- endif %}\n{%- if builtin_tools is defined %}\n {{- \"Tools: \" + builtin_tools | reject('equalto', 'code_interpreter') | join(\", \") + \"\\n\\n\"}}\n{%- endif %}\n{{- \"Cutting Knowledge Date: December 2023\\n\" }}\n{{- \"Today Date: \" + date_string + \"\\n\\n\" }}\n{%- if tools is not none and not tools_in_user_message %}\n {{- \"You have access to the following functions. To call a function, please respond with JSON for a function call.\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n{%- endif %}\n{{- system_message }}\n{{- \"<|eot_id|>\" }}\n\n{#- Custom tools are passed in a user message with some extra guidance #}\n{%- if tools_in_user_message and not tools is none %}\n {#- Extract the first user message so we can plug it in here #}\n {%- if messages | length != 0 %}\n {%- set first_user_message = messages[0]['content']|trim %}\n {%- set messages = messages[1:] %}\n {%- else %}\n {{- raise_exception(\"Cannot put tools in the first user message when there's no first user message!\") }}\n{%- endif %}\n {{- '<|start_header_id|>user<|end_header_id|>\\n\\n' -}}\n {{- \"Given the following functions, please respond with a JSON for a function call \" }}\n {{- \"with its proper arguments that best answers the given prompt.\\n\\n\" }}\n {{- 'Respond in the format {\"name\": function name, \"parameters\": dictionary of argument name and its value}.' }}\n {{- \"Do not use variables.\\n\\n\" }}\n {%- for t in tools %}\n {{- t | tojson(indent=4) }}\n {{- \"\\n\\n\" }}\n {%- endfor %}\n {{- first_user_message + \"<|eot_id|>\"}}\n{%- endif %}\n\n{%- for message in messages %}\n {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}\n {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\\n\\n'+ message['content'] | trim + '<|eot_id|>' }}\n {%- elif 'tool_calls' in message %}\n {%- if not message.tool_calls|length == 1 %}\n {{- raise_exception(\"This model only supports single tool-calls at once!\") }}\n {%- endif %}\n {%- set tool_call = message.tool_calls[0].function %}\n {%- if builtin_tools is defined and tool_call.name in builtin_tools %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- \"<|python_tag|>\" + tool_call.name + \".call(\" }}\n {%- for arg_name, arg_val in tool_call.arguments | items %}\n {{- arg_name + '=\"' + arg_val + '\"' }}\n {%- if not loop.last %}\n {{- \", \" }}\n {%- endif %}\n {%- endfor %}\n {{- \")\" }}\n {%- else %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' -}}\n {{- '{\"name\": \"' + tool_call.name + '\", ' }}\n {{- '\"parameters\": ' }}\n {{- tool_call.arguments | tojson }}\n {{- \"}\" }}\n {%- endif %}\n {%- if builtin_tools is defined %}\n {#- This means we're in ipython mode #}\n {{- \"<|eom_id|>\" }}\n {%- else %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n {%- elif message.role == \"tool\" or message.role == \"ipython\" %}\n {{- \"<|start_header_id|>ipython<|end_header_id|>\\n\\n\" }}\n {%- if message.content is mapping or message.content is iterable %}\n {{- message.content | tojson }}\n {%- else %}\n {{- message.content }}\n {%- endif %}\n {{- \"<|eot_id|>\" }}\n {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n {{- '<|start_header_id|>assistant<|end_header_id|>\\n\\n' }}\n{%- endif %}\n", + "clean_up_tokenization_spaces": true, + "eos_token": "<|eot_id|>", + "extra_special_tokens": {}, + "model_input_names": [ + "input_ids", + "attention_mask" + ], + "model_max_length": 131072, + "pad_token": "<|end_of_text|>", + "tokenizer_class": "PreTrainedTokenizer" +}