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..73d7dedde9f04fd57891b99d38c4f046a905e8cd --- /dev/null +++ b/README.md @@ -0,0 +1,146 @@ +--- +library_name: peft +license: llama3.1 +base_model: nvidia/Llama-3.1-Nemotron-70B-Instruct-HF +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: nvidia/Llama-3.1-Nemotron-70B-Instruct-HF +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_Nemotron-70B +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 [nvidia/Llama-3.1-Nemotron-70B-Instruct-HF](https://huggingface.co/nvidia/Llama-3.1-Nemotron-70B-Instruct-HF) 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..604978b603be594e46459a6c8c10d206a5a03b31 --- /dev/null +++ b/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF", + "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": [ + "down_proj", + "v_proj", + "q_proj", + "gate_proj", + "o_proj", + "up_proj", + "k_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..90b76f45c94794a5d3b9e53430bdb322764d258d --- /dev/null +++ b/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f148e162b556bb3d7b017226cf92103ac687f1d14e97e9344ddd35493aeb9d61 +size 10829849744 diff --git a/checkpoint-146/README.md b/checkpoint-146/README.md new file mode 100644 index 0000000000000000000000000000000000000000..fdf619c317c2fe82074662582dbd68166b6f9d50 --- /dev/null +++ b/checkpoint-146/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3.1-Nemotron-70B-Instruct-HF +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..604978b603be594e46459a6c8c10d206a5a03b31 --- /dev/null +++ b/checkpoint-146/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF", + "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": [ + "down_proj", + "v_proj", + "q_proj", + "gate_proj", + "o_proj", + "up_proj", + "k_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..04b31f8ed9e755eb3f0768ff55b7ce178ea34ec1 --- /dev/null +++ b/checkpoint-146/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7788384f056a62b567b6635519f49d0dabb2165f7ee4fc2f8b87f42bd2e287cf +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..54bfac6b3f0c2596dd9e0c321d29aaac4decab84 --- /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 sha256:48d467edb59ac84555d968d3a18a78be524823fccff1131176a31272f1b3a27f +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..0f5033ef6069e2ae0fd00f3ca27e5f40fec04df3 --- /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 sha256:0f1669f4eec19aef022075d8d90d090de6379f4e20b661923d165e4b10890770 +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..f1cc6e516b81a50b8c658f0dddb7c8a0f1f07640 --- /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:f34f022c7156b2cf9a7d249a609736a059c2681aa705d940a626ca1d64f640cf +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..bdfd31c3cf37dc8494816237a595547eacd10adf --- /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 sha256:3df86bb653506812c686ca4be2e4fdcdf7a72e7e2b9741c2d9865bafa66b62bc +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..fdde94c29816839ec3c29d6c6461206a49911f3c --- /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\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..155120e9fc8edc87d74b6912da5051a0c4db960c --- /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": 2.020231246948242, + "learning_rate": 5.0000000000000004e-08, + "loss": 0.8736, + "step": 1 + }, + { + "epoch": 0.013636363636363636, + "grad_norm": 1.9368854761123657, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.8543, + "step": 2 + }, + { + "epoch": 0.020454545454545454, + "grad_norm": 1.908379077911377, + "learning_rate": 1.5000000000000002e-07, + "loss": 0.8339, + "step": 3 + }, + { + "epoch": 0.02727272727272727, + "grad_norm": 1.9313658475875854, + "learning_rate": 2.0000000000000002e-07, + "loss": 0.834, + "step": 4 + }, + { + "epoch": 0.03409090909090909, + "grad_norm": 1.983262062072754, + "learning_rate": 2.5000000000000004e-07, + "loss": 0.8726, + "step": 5 + }, + { + "epoch": 0.04090909090909091, + "grad_norm": 1.9994115829467773, + "learning_rate": 3.0000000000000004e-07, + "loss": 0.8135, + "step": 6 + }, + { + "epoch": 0.04772727272727273, + "grad_norm": 1.8889026641845703, + "learning_rate": 3.5000000000000004e-07, + "loss": 0.8439, + "step": 7 + }, + { + "epoch": 0.05454545454545454, + "grad_norm": 1.9258317947387695, + "learning_rate": 4.0000000000000003e-07, + "loss": 0.84, + "step": 8 + }, + { + "epoch": 0.06136363636363636, + "grad_norm": 1.9654443264007568, + "learning_rate": 4.5000000000000003e-07, + "loss": 0.8609, + "step": 9 + }, + { + "epoch": 0.06818181818181818, + "grad_norm": 1.9278813600540161, + "learning_rate": 5.000000000000001e-07, + "loss": 0.8515, + "step": 10 + }, + { + "epoch": 0.075, + "grad_norm": 1.8698863983154297, + "learning_rate": 5.5e-07, + "loss": 0.8632, + "step": 11 + }, + { + "epoch": 0.08181818181818182, + "grad_norm": 2.7078285217285156, + "learning_rate": 6.000000000000001e-07, + "loss": 0.8643, + "step": 12 + }, + { + "epoch": 0.08863636363636364, + "grad_norm": 1.959272027015686, + "learning_rate": 6.5e-07, + "loss": 0.861, + "step": 13 + }, + { + "epoch": 0.09545454545454546, + "grad_norm": 1.8952852487564087, + "learning_rate": 7.000000000000001e-07, + "loss": 0.8599, + "step": 14 + }, + { + "epoch": 0.10227272727272728, + "grad_norm": 1.9469462633132935, + "learning_rate": 7.5e-07, + "loss": 0.8467, + "step": 15 + }, + { + "epoch": 0.10909090909090909, + "grad_norm": 2.0111300945281982, + "learning_rate": 8.000000000000001e-07, + "loss": 0.8078, + "step": 16 + }, + { + "epoch": 0.1159090909090909, + "grad_norm": 1.739009141921997, + "learning_rate": 8.500000000000001e-07, + "loss": 0.8114, + "step": 17 + }, + { + "epoch": 0.12272727272727273, + "grad_norm": 1.7578809261322021, + "learning_rate": 9.000000000000001e-07, + "loss": 0.8489, + "step": 18 + }, + { + "epoch": 0.12954545454545455, + "grad_norm": 1.9787132740020752, + "learning_rate": 9.500000000000001e-07, + "loss": 0.8052, + "step": 19 + }, + { + "epoch": 0.13636363636363635, + "grad_norm": 1.6380164623260498, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.7882, + "step": 20 + }, + { + "epoch": 0.1431818181818182, + "grad_norm": 1.686877727508545, + "learning_rate": 1.0500000000000001e-06, + "loss": 0.8126, + "step": 21 + }, + { + "epoch": 0.15, + "grad_norm": 1.690590739250183, + "learning_rate": 1.1e-06, + "loss": 0.8248, + "step": 22 + }, + { + "epoch": 0.15681818181818183, + "grad_norm": 1.612255334854126, + "learning_rate": 1.1500000000000002e-06, + "loss": 0.8024, + "step": 23 + }, + { + "epoch": 0.16363636363636364, + "grad_norm": 1.6669739484786987, + "learning_rate": 1.2000000000000002e-06, + "loss": 0.818, + "step": 24 + }, + { + "epoch": 0.17045454545454544, + "grad_norm": 1.7726502418518066, + "learning_rate": 1.25e-06, + "loss": 0.8158, + "step": 25 + }, + { + "epoch": 0.17727272727272728, + "grad_norm": 1.5930036306381226, + "learning_rate": 1.3e-06, + "loss": 0.7886, + "step": 26 + }, + { + "epoch": 0.18409090909090908, + "grad_norm": 1.5480856895446777, + "learning_rate": 1.3500000000000002e-06, + "loss": 0.7747, + "step": 27 + }, + { + "epoch": 0.19090909090909092, + "grad_norm": 1.3537116050720215, + "learning_rate": 1.4000000000000001e-06, + "loss": 0.7891, + "step": 28 + }, + { + "epoch": 0.19772727272727272, + "grad_norm": 1.3412829637527466, + "learning_rate": 1.45e-06, + "loss": 0.7799, + "step": 29 + }, + { + "epoch": 0.20454545454545456, + "grad_norm": 1.237492322921753, + "learning_rate": 1.5e-06, + "loss": 0.7768, + "step": 30 + }, + { + "epoch": 0.21136363636363636, + "grad_norm": 1.3467835187911987, + "learning_rate": 1.5500000000000002e-06, + "loss": 0.7344, + "step": 31 + }, + { + "epoch": 0.21818181818181817, + "grad_norm": 1.1665314435958862, + "learning_rate": 1.6000000000000001e-06, + "loss": 0.7152, + "step": 32 + }, + { + "epoch": 0.225, + "grad_norm": 1.2461743354797363, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.758, + "step": 33 + }, + { + "epoch": 0.2318181818181818, + "grad_norm": 1.0343059301376343, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.7362, + "step": 34 + }, + { + "epoch": 0.23863636363636365, + "grad_norm": 1.112117052078247, + "learning_rate": 1.75e-06, + "loss": 0.7771, + "step": 35 + }, + { + "epoch": 0.24545454545454545, + "grad_norm": 1.104337453842163, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.7479, + "step": 36 + }, + { + "epoch": 0.25227272727272726, + "grad_norm": 0.9324578642845154, + "learning_rate": 1.85e-06, + "loss": 0.6936, + "step": 37 + }, + { + "epoch": 0.2590909090909091, + "grad_norm": 1.0733048915863037, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.7302, + "step": 38 + }, + { + "epoch": 0.26590909090909093, + "grad_norm": 0.8614453673362732, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.7466, + "step": 39 + }, + { + "epoch": 0.2727272727272727, + "grad_norm": 0.7729659080505371, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.725, + "step": 40 + }, + { + "epoch": 0.27954545454545454, + "grad_norm": 0.9341004490852356, + "learning_rate": 2.05e-06, + "loss": 0.6949, + "step": 41 + }, + { + "epoch": 0.2863636363636364, + "grad_norm": 0.6668925881385803, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.6587, + "step": 42 + }, + { + "epoch": 0.29318181818181815, + "grad_norm": 0.681847333908081, + "learning_rate": 2.15e-06, + "loss": 0.6984, + "step": 43 + }, + { + "epoch": 0.3, + "grad_norm": 0.6222082376480103, + "learning_rate": 2.2e-06, + "loss": 0.7063, + "step": 44 + }, + { + "epoch": 0.3068181818181818, + "grad_norm": 0.790607213973999, + "learning_rate": 2.25e-06, + "loss": 0.6804, + "step": 45 + }, + { + "epoch": 0.31363636363636366, + "grad_norm": 0.6110427379608154, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.6694, + "step": 46 + }, + { + "epoch": 0.32045454545454544, + "grad_norm": 0.6481993794441223, + "learning_rate": 2.35e-06, + "loss": 0.7321, + "step": 47 + }, + { + "epoch": 0.32727272727272727, + "grad_norm": 0.5973438620567322, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.6508, + "step": 48 + }, + { + "epoch": 0.3340909090909091, + "grad_norm": 0.6252740621566772, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.6911, + "step": 49 + }, + { + "epoch": 0.3409090909090909, + "grad_norm": 0.6826481819152832, + "learning_rate": 2.5e-06, + "loss": 0.6855, + "step": 50 + }, + { + "epoch": 0.3477272727272727, + "grad_norm": 0.6307975649833679, + "learning_rate": 2.55e-06, + "loss": 0.714, + "step": 51 + }, + { + "epoch": 0.35454545454545455, + "grad_norm": 0.5900976061820984, + "learning_rate": 2.6e-06, + "loss": 0.6619, + "step": 52 + }, + { + "epoch": 0.3613636363636364, + "grad_norm": 0.6203920841217041, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.6822, + "step": 53 + }, + { + "epoch": 0.36818181818181817, + "grad_norm": 0.6197589039802551, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.6584, + "step": 54 + }, + { + "epoch": 0.375, + "grad_norm": 0.5921110510826111, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.6739, + "step": 55 + }, + { + "epoch": 0.38181818181818183, + "grad_norm": 0.6215619444847107, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.6631, + "step": 56 + }, + { + "epoch": 0.3886363636363636, + "grad_norm": 0.5495648980140686, + "learning_rate": 2.85e-06, + "loss": 0.6531, + "step": 57 + }, + { + "epoch": 0.39545454545454545, + "grad_norm": 0.531947910785675, + "learning_rate": 2.9e-06, + "loss": 0.6365, + "step": 58 + }, + { + "epoch": 0.4022727272727273, + "grad_norm": 0.559112548828125, + "learning_rate": 2.95e-06, + "loss": 0.6391, + "step": 59 + }, + { + "epoch": 0.4090909090909091, + "grad_norm": 0.5377560257911682, + "learning_rate": 3e-06, + "loss": 0.6775, + "step": 60 + }, + { + "epoch": 0.4159090909090909, + "grad_norm": 0.5167352557182312, + "learning_rate": 3.05e-06, + "loss": 0.6518, + "step": 61 + }, + { + "epoch": 0.42272727272727273, + "grad_norm": 0.5483390092849731, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.5972, + "step": 62 + }, + { + "epoch": 0.42954545454545456, + "grad_norm": 0.5687050223350525, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.6515, + "step": 63 + }, + { + "epoch": 0.43636363636363634, + "grad_norm": 0.9046968817710876, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.6497, + "step": 64 + }, + { + "epoch": 0.4431818181818182, + "grad_norm": 0.5231667757034302, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.6364, + "step": 65 + }, + { + "epoch": 0.45, + "grad_norm": 0.49699342250823975, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.6144, + "step": 66 + }, + { + "epoch": 0.45681818181818185, + "grad_norm": 0.5390080213546753, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.6051, + "step": 67 + }, + { + "epoch": 0.4636363636363636, + "grad_norm": 0.5252938270568848, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.6353, + "step": 68 + }, + { + "epoch": 0.47045454545454546, + "grad_norm": 0.6291780471801758, + "learning_rate": 3.45e-06, + "loss": 0.6326, + "step": 69 + }, + { + "epoch": 0.4772727272727273, + "grad_norm": 0.5545375943183899, + "learning_rate": 3.5e-06, + "loss": 0.647, + "step": 70 + }, + { + "epoch": 0.48409090909090907, + "grad_norm": 0.47775956988334656, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.5993, + "step": 71 + }, + { + "epoch": 0.4909090909090909, + "grad_norm": 0.5016375184059143, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.6056, + "step": 72 + }, + { + "epoch": 0.49772727272727274, + "grad_norm": 0.4780517816543579, + "learning_rate": 3.65e-06, + "loss": 0.6189, + "step": 73 + }, + { + "epoch": 0.5045454545454545, + "grad_norm": 0.5763716697692871, + "learning_rate": 3.7e-06, + "loss": 0.6262, + "step": 74 + }, + { + "epoch": 0.5113636363636364, + "grad_norm": 0.5321404337882996, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.6336, + "step": 75 + }, + { + "epoch": 0.5181818181818182, + "grad_norm": 0.4641991853713989, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.6317, + "step": 76 + }, + { + "epoch": 0.525, + "grad_norm": 0.45118412375450134, + "learning_rate": 3.85e-06, + "loss": 0.5927, + "step": 77 + }, + { + "epoch": 0.5318181818181819, + "grad_norm": 0.5500645041465759, + "learning_rate": 3.900000000000001e-06, + "loss": 0.62, + "step": 78 + }, + { + "epoch": 0.5386363636363637, + "grad_norm": 0.5197616219520569, + "learning_rate": 3.95e-06, + "loss": 0.6063, + "step": 79 + }, + { + "epoch": 0.5454545454545454, + "grad_norm": 0.4672992527484894, + "learning_rate": 4.000000000000001e-06, + "loss": 0.6156, + "step": 80 + }, + { + "epoch": 0.5522727272727272, + "grad_norm": 0.5995073318481445, + "learning_rate": 4.05e-06, + "loss": 0.6219, + "step": 81 + }, + { + "epoch": 0.5590909090909091, + "grad_norm": 0.5593804717063904, + "learning_rate": 4.1e-06, + "loss": 0.6282, + "step": 82 + }, + { + "epoch": 0.5659090909090909, + "grad_norm": 0.7544977068901062, + "learning_rate": 4.15e-06, + "loss": 0.6246, + "step": 83 + }, + { + "epoch": 0.5727272727272728, + "grad_norm": 0.45378655195236206, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.5881, + "step": 84 + }, + { + "epoch": 0.5795454545454546, + "grad_norm": 0.4407907724380493, + "learning_rate": 4.25e-06, + "loss": 0.6026, + "step": 85 + }, + { + "epoch": 0.5863636363636363, + "grad_norm": 0.4333001673221588, + "learning_rate": 4.3e-06, + "loss": 0.6098, + "step": 86 + }, + { + "epoch": 0.5931818181818181, + "grad_norm": 0.45020997524261475, + "learning_rate": 4.350000000000001e-06, + "loss": 0.5858, + "step": 87 + }, + { + "epoch": 0.6, + "grad_norm": 0.424003541469574, + "learning_rate": 4.4e-06, + "loss": 0.6186, + "step": 88 + }, + { + "epoch": 0.6068181818181818, + "grad_norm": 0.427081435918808, + "learning_rate": 4.450000000000001e-06, + "loss": 0.6164, + "step": 89 + }, + { + "epoch": 0.6136363636363636, + "grad_norm": 0.4430546164512634, + "learning_rate": 4.5e-06, + "loss": 0.6049, + "step": 90 + }, + { + "epoch": 0.6204545454545455, + "grad_norm": 0.5469494462013245, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.608, + "step": 91 + }, + { + "epoch": 0.6272727272727273, + "grad_norm": 0.4481559693813324, + "learning_rate": 4.600000000000001e-06, + "loss": 0.5532, + "step": 92 + }, + { + "epoch": 0.634090909090909, + "grad_norm": 0.6563135981559753, + "learning_rate": 4.65e-06, + "loss": 0.6105, + "step": 93 + }, + { + "epoch": 0.6409090909090909, + "grad_norm": 0.4572807252407074, + "learning_rate": 4.7e-06, + "loss": 0.5976, + "step": 94 + }, + { + "epoch": 0.6477272727272727, + "grad_norm": 0.4423324167728424, + "learning_rate": 4.75e-06, + "loss": 0.616, + "step": 95 + }, + { + "epoch": 0.6545454545454545, + "grad_norm": 0.6230632066726685, + "learning_rate": 4.800000000000001e-06, + "loss": 0.5879, + "step": 96 + }, + { + "epoch": 0.6613636363636364, + "grad_norm": 0.461467981338501, + "learning_rate": 4.85e-06, + "loss": 0.5992, + "step": 97 + }, + { + "epoch": 0.6681818181818182, + "grad_norm": 0.44455233216285706, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.6138, + "step": 98 + }, + { + "epoch": 0.675, + "grad_norm": 0.47281715273857117, + "learning_rate": 4.95e-06, + "loss": 0.59, + "step": 99 + }, + { + "epoch": 0.6818181818181818, + "grad_norm": 0.46114516258239746, + "learning_rate": 5e-06, + "loss": 0.6216, + "step": 100 + }, + { + "epoch": 0.6886363636363636, + "grad_norm": 0.49900761246681213, + "learning_rate": 4.9999795126530275e-06, + "loss": 0.6222, + "step": 101 + }, + { + "epoch": 0.6954545454545454, + "grad_norm": 0.4764839708805084, + "learning_rate": 4.999918050947891e-06, + "loss": 0.5829, + "step": 102 + }, + { + "epoch": 0.7022727272727273, + "grad_norm": 0.43502742052078247, + "learning_rate": 4.999815615891943e-06, + "loss": 0.5982, + "step": 103 + }, + { + "epoch": 0.7090909090909091, + "grad_norm": 0.43318599462509155, + "learning_rate": 4.9996722091640805e-06, + "loss": 0.5827, + "step": 104 + }, + { + "epoch": 0.7159090909090909, + "grad_norm": 0.47218361496925354, + "learning_rate": 4.9994878331147225e-06, + "loss": 0.5907, + "step": 105 + }, + { + "epoch": 0.7227272727272728, + "grad_norm": 0.48492228984832764, + "learning_rate": 4.99926249076577e-06, + "loss": 0.5459, + "step": 106 + }, + { + "epoch": 0.7295454545454545, + "grad_norm": 0.4019850194454193, + "learning_rate": 4.998996185810557e-06, + "loss": 0.5957, + "step": 107 + }, + { + "epoch": 0.7363636363636363, + "grad_norm": 0.47361984848976135, + "learning_rate": 4.998688922613788e-06, + "loss": 0.579, + "step": 108 + }, + { + "epoch": 0.7431818181818182, + "grad_norm": 0.4612269401550293, + "learning_rate": 4.9983407062114695e-06, + "loss": 0.5814, + "step": 109 + }, + { + "epoch": 0.75, + "grad_norm": 0.45558398962020874, + "learning_rate": 4.9979515423108255e-06, + "loss": 0.6036, + "step": 110 + }, + { + "epoch": 0.7568181818181818, + "grad_norm": 0.41492199897766113, + "learning_rate": 4.997521437290205e-06, + "loss": 0.5891, + "step": 111 + }, + { + "epoch": 0.7636363636363637, + "grad_norm": 0.4484409689903259, + "learning_rate": 4.997050398198977e-06, + "loss": 0.6031, + "step": 112 + }, + { + "epoch": 0.7704545454545455, + "grad_norm": 0.42671382427215576, + "learning_rate": 4.996538432757414e-06, + "loss": 0.6106, + "step": 113 + }, + { + "epoch": 0.7772727272727272, + "grad_norm": 0.4222075343132019, + "learning_rate": 4.995985549356568e-06, + "loss": 0.5774, + "step": 114 + }, + { + "epoch": 0.7840909090909091, + "grad_norm": 0.421021431684494, + "learning_rate": 4.995391757058129e-06, + "loss": 0.5754, + "step": 115 + }, + { + "epoch": 0.7909090909090909, + "grad_norm": 0.4215952754020691, + "learning_rate": 4.99475706559428e-06, + "loss": 0.5895, + "step": 116 + }, + { + "epoch": 0.7977272727272727, + "grad_norm": 0.42652982473373413, + "learning_rate": 4.994081485367537e-06, + "loss": 0.5646, + "step": 117 + }, + { + "epoch": 0.8045454545454546, + "grad_norm": 0.4145742654800415, + "learning_rate": 4.993365027450576e-06, + "loss": 0.5862, + "step": 118 + }, + { + "epoch": 0.8113636363636364, + "grad_norm": 0.4754612445831299, + "learning_rate": 4.992607703586058e-06, + "loss": 0.5645, + "step": 119 + }, + { + "epoch": 0.8181818181818182, + "grad_norm": 0.4694959223270416, + "learning_rate": 4.991809526186424e-06, + "loss": 0.5989, + "step": 120 + }, + { + "epoch": 0.825, + "grad_norm": 0.4573237895965576, + "learning_rate": 4.990970508333707e-06, + "loss": 0.5769, + "step": 121 + }, + { + "epoch": 0.8318181818181818, + "grad_norm": 0.4592607915401459, + "learning_rate": 4.990090663779305e-06, + "loss": 0.5529, + "step": 122 + }, + { + "epoch": 0.8386363636363636, + "grad_norm": 0.4325920045375824, + "learning_rate": 4.9891700069437635e-06, + "loss": 0.5846, + "step": 123 + }, + { + "epoch": 0.8454545454545455, + "grad_norm": 0.44623467326164246, + "learning_rate": 4.988208552916535e-06, + "loss": 0.5997, + "step": 124 + }, + { + "epoch": 0.8522727272727273, + "grad_norm": 0.46071872115135193, + "learning_rate": 4.987206317455734e-06, + "loss": 0.5687, + "step": 125 + }, + { + "epoch": 0.8590909090909091, + "grad_norm": 0.43121206760406494, + "learning_rate": 4.986163316987877e-06, + "loss": 0.5277, + "step": 126 + }, + { + "epoch": 0.865909090909091, + "grad_norm": 0.4711630046367645, + "learning_rate": 4.985079568607613e-06, + "loss": 0.592, + "step": 127 + }, + { + "epoch": 0.8727272727272727, + "grad_norm": 0.4450221657752991, + "learning_rate": 4.983955090077445e-06, + "loss": 0.5656, + "step": 128 + }, + { + "epoch": 0.8795454545454545, + "grad_norm": 0.6546564102172852, + "learning_rate": 4.982789899827439e-06, + "loss": 0.572, + "step": 129 + }, + { + "epoch": 0.8863636363636364, + "grad_norm": 0.49622273445129395, + "learning_rate": 4.9815840169549216e-06, + "loss": 0.5719, + "step": 130 + }, + { + "epoch": 0.8931818181818182, + "grad_norm": 0.5478544235229492, + "learning_rate": 4.980337461224164e-06, + "loss": 0.5774, + "step": 131 + }, + { + "epoch": 0.9, + "grad_norm": 0.4340018033981323, + "learning_rate": 4.979050253066064e-06, + "loss": 0.5444, + "step": 132 + }, + { + "epoch": 0.9068181818181819, + "grad_norm": 0.4243793189525604, + "learning_rate": 4.977722413577802e-06, + "loss": 0.5759, + "step": 133 + }, + { + "epoch": 0.9136363636363637, + "grad_norm": 0.45201942324638367, + "learning_rate": 4.976353964522509e-06, + "loss": 0.5746, + "step": 134 + }, + { + "epoch": 0.9204545454545454, + "grad_norm": 0.42892220616340637, + "learning_rate": 4.974944928328894e-06, + "loss": 0.544, + "step": 135 + }, + { + "epoch": 0.9272727272727272, + "grad_norm": 0.43185824155807495, + "learning_rate": 4.973495328090891e-06, + "loss": 0.5803, + "step": 136 + }, + { + "epoch": 0.9340909090909091, + "grad_norm": 0.5512542724609375, + "learning_rate": 4.972005187567267e-06, + "loss": 0.5694, + "step": 137 + }, + { + "epoch": 0.9409090909090909, + "grad_norm": 0.4836059808731079, + "learning_rate": 4.970474531181245e-06, + "loss": 0.5581, + "step": 138 + }, + { + "epoch": 0.9477272727272728, + "grad_norm": 0.4508282244205475, + "learning_rate": 4.968903384020095e-06, + "loss": 0.5553, + "step": 139 + }, + { + "epoch": 0.9545454545454546, + "grad_norm": 0.4764733612537384, + "learning_rate": 4.967291771834727e-06, + "loss": 0.5798, + "step": 140 + }, + { + "epoch": 0.9613636363636363, + "grad_norm": 0.463468074798584, + "learning_rate": 4.965639721039267e-06, + "loss": 0.559, + "step": 141 + }, + { + "epoch": 0.9681818181818181, + "grad_norm": 0.4655650854110718, + "learning_rate": 4.963947258710626e-06, + "loss": 0.5576, + "step": 142 + }, + { + "epoch": 0.975, + "grad_norm": 0.4651234745979309, + "learning_rate": 4.962214412588053e-06, + "loss": 0.5792, + "step": 143 + }, + { + "epoch": 0.9818181818181818, + "grad_norm": 0.43080538511276245, + "learning_rate": 4.960441211072686e-06, + "loss": 0.5314, + "step": 144 + }, + { + "epoch": 0.9886363636363636, + "grad_norm": 0.4403238594532013, + "learning_rate": 4.9586276832270785e-06, + "loss": 0.5592, + "step": 145 + }, + { + "epoch": 0.9954545454545455, + "grad_norm": 0.43968647718429565, + "learning_rate": 4.9567738587747314e-06, + "loss": 0.5573, + "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..a03cebdd4a04fb2e9443a05c92f15b2b2db01e10 --- /dev/null +++ b/checkpoint-146/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90038f7ff377ce29f00aedab533b18bfc9515b30d2b4232d8b0d6de26ec3ea56 +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..fdf619c317c2fe82074662582dbd68166b6f9d50 --- /dev/null +++ b/checkpoint-292/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3.1-Nemotron-70B-Instruct-HF +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..604978b603be594e46459a6c8c10d206a5a03b31 --- /dev/null +++ b/checkpoint-292/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF", + "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": [ + "down_proj", + "v_proj", + "q_proj", + "gate_proj", + "o_proj", + "up_proj", + "k_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..a20c5be9cddd4c482f554b764a6f470b1e3b008f --- /dev/null +++ b/checkpoint-292/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:66adf3963e01b07292bf8f8f5c29ac5a66178487d0eb087e0d945e5405bc4efc +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..5217911ee3d09cb7b8864a7ddc5f53cec90e951d --- /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 sha256:919a2d538d32454bcc58737af12622950622cfbed641b698dd0acd668887323b +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..8ea1246c47377d43493dd2429ce2022a4e96e29a --- /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:b2441b1086a3338928d3ee0cf5a0f2fa40a4acf608a21ce89a509c1c26db5d11 +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..d58f82c456b9cee90a6387f3d0dc771bd323ecae --- /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 sha256:490baddde65257c91507ce48edfa7f927b0db84534c978afafe7f548470e5be7 +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..0272bb098ceac9ec418390b27fa53f7f72848350 --- /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 sha256:1e276f3398329a565a5c1d150e8c72fc4af4dfdde9863fa0fb4f35a1ca5f1773 +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..fdde94c29816839ec3c29d6c6461206a49911f3c --- /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\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..e54bfbddcefee1f2bcfe401ef3995d1a31beec34 --- /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": 2.020231246948242, + "learning_rate": 5.0000000000000004e-08, + "loss": 0.8736, + "step": 1 + }, + { + "epoch": 0.013636363636363636, + "grad_norm": 1.9368854761123657, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.8543, + "step": 2 + }, + { + "epoch": 0.020454545454545454, + "grad_norm": 1.908379077911377, + "learning_rate": 1.5000000000000002e-07, + "loss": 0.8339, + "step": 3 + }, + { + "epoch": 0.02727272727272727, + "grad_norm": 1.9313658475875854, + "learning_rate": 2.0000000000000002e-07, + "loss": 0.834, + "step": 4 + }, + { + "epoch": 0.03409090909090909, + "grad_norm": 1.983262062072754, + "learning_rate": 2.5000000000000004e-07, + "loss": 0.8726, + "step": 5 + }, + { + "epoch": 0.04090909090909091, + "grad_norm": 1.9994115829467773, + "learning_rate": 3.0000000000000004e-07, + "loss": 0.8135, + "step": 6 + }, + { + "epoch": 0.04772727272727273, + "grad_norm": 1.8889026641845703, + "learning_rate": 3.5000000000000004e-07, + "loss": 0.8439, + "step": 7 + }, + { + "epoch": 0.05454545454545454, + "grad_norm": 1.9258317947387695, + "learning_rate": 4.0000000000000003e-07, + "loss": 0.84, + "step": 8 + }, + { + "epoch": 0.06136363636363636, + "grad_norm": 1.9654443264007568, + "learning_rate": 4.5000000000000003e-07, + "loss": 0.8609, + "step": 9 + }, + { + "epoch": 0.06818181818181818, + "grad_norm": 1.9278813600540161, + "learning_rate": 5.000000000000001e-07, + "loss": 0.8515, + "step": 10 + }, + { + "epoch": 0.075, + "grad_norm": 1.8698863983154297, + "learning_rate": 5.5e-07, + "loss": 0.8632, + "step": 11 + }, + { + "epoch": 0.08181818181818182, + "grad_norm": 2.7078285217285156, + "learning_rate": 6.000000000000001e-07, + "loss": 0.8643, + "step": 12 + }, + { + "epoch": 0.08863636363636364, + "grad_norm": 1.959272027015686, + "learning_rate": 6.5e-07, + "loss": 0.861, + "step": 13 + }, + { + "epoch": 0.09545454545454546, + "grad_norm": 1.8952852487564087, + "learning_rate": 7.000000000000001e-07, + "loss": 0.8599, + "step": 14 + }, + { + "epoch": 0.10227272727272728, + "grad_norm": 1.9469462633132935, + "learning_rate": 7.5e-07, + "loss": 0.8467, + "step": 15 + }, + { + "epoch": 0.10909090909090909, + "grad_norm": 2.0111300945281982, + "learning_rate": 8.000000000000001e-07, + "loss": 0.8078, + "step": 16 + }, + { + "epoch": 0.1159090909090909, + "grad_norm": 1.739009141921997, + "learning_rate": 8.500000000000001e-07, + "loss": 0.8114, + "step": 17 + }, + { + "epoch": 0.12272727272727273, + "grad_norm": 1.7578809261322021, + "learning_rate": 9.000000000000001e-07, + "loss": 0.8489, + "step": 18 + }, + { + "epoch": 0.12954545454545455, + "grad_norm": 1.9787132740020752, + "learning_rate": 9.500000000000001e-07, + "loss": 0.8052, + "step": 19 + }, + { + "epoch": 0.13636363636363635, + "grad_norm": 1.6380164623260498, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.7882, + "step": 20 + }, + { + "epoch": 0.1431818181818182, + "grad_norm": 1.686877727508545, + "learning_rate": 1.0500000000000001e-06, + "loss": 0.8126, + "step": 21 + }, + { + "epoch": 0.15, + "grad_norm": 1.690590739250183, + "learning_rate": 1.1e-06, + "loss": 0.8248, + "step": 22 + }, + { + "epoch": 0.15681818181818183, + "grad_norm": 1.612255334854126, + "learning_rate": 1.1500000000000002e-06, + "loss": 0.8024, + "step": 23 + }, + { + "epoch": 0.16363636363636364, + "grad_norm": 1.6669739484786987, + "learning_rate": 1.2000000000000002e-06, + "loss": 0.818, + "step": 24 + }, + { + "epoch": 0.17045454545454544, + "grad_norm": 1.7726502418518066, + "learning_rate": 1.25e-06, + "loss": 0.8158, + "step": 25 + }, + { + "epoch": 0.17727272727272728, + "grad_norm": 1.5930036306381226, + "learning_rate": 1.3e-06, + "loss": 0.7886, + "step": 26 + }, + { + "epoch": 0.18409090909090908, + "grad_norm": 1.5480856895446777, + "learning_rate": 1.3500000000000002e-06, + "loss": 0.7747, + "step": 27 + }, + { + "epoch": 0.19090909090909092, + "grad_norm": 1.3537116050720215, + "learning_rate": 1.4000000000000001e-06, + "loss": 0.7891, + "step": 28 + }, + { + "epoch": 0.19772727272727272, + "grad_norm": 1.3412829637527466, + "learning_rate": 1.45e-06, + "loss": 0.7799, + "step": 29 + }, + { + "epoch": 0.20454545454545456, + "grad_norm": 1.237492322921753, + "learning_rate": 1.5e-06, + "loss": 0.7768, + "step": 30 + }, + { + "epoch": 0.21136363636363636, + "grad_norm": 1.3467835187911987, + "learning_rate": 1.5500000000000002e-06, + "loss": 0.7344, + "step": 31 + }, + { + "epoch": 0.21818181818181817, + "grad_norm": 1.1665314435958862, + "learning_rate": 1.6000000000000001e-06, + "loss": 0.7152, + "step": 32 + }, + { + "epoch": 0.225, + "grad_norm": 1.2461743354797363, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.758, + "step": 33 + }, + { + "epoch": 0.2318181818181818, + "grad_norm": 1.0343059301376343, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.7362, + "step": 34 + }, + { + "epoch": 0.23863636363636365, + "grad_norm": 1.112117052078247, + "learning_rate": 1.75e-06, + "loss": 0.7771, + "step": 35 + }, + { + "epoch": 0.24545454545454545, + "grad_norm": 1.104337453842163, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.7479, + "step": 36 + }, + { + "epoch": 0.25227272727272726, + "grad_norm": 0.9324578642845154, + "learning_rate": 1.85e-06, + "loss": 0.6936, + "step": 37 + }, + { + "epoch": 0.2590909090909091, + "grad_norm": 1.0733048915863037, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.7302, + "step": 38 + }, + { + "epoch": 0.26590909090909093, + "grad_norm": 0.8614453673362732, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.7466, + "step": 39 + }, + { + "epoch": 0.2727272727272727, + "grad_norm": 0.7729659080505371, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.725, + "step": 40 + }, + { + "epoch": 0.27954545454545454, + "grad_norm": 0.9341004490852356, + "learning_rate": 2.05e-06, + "loss": 0.6949, + "step": 41 + }, + { + "epoch": 0.2863636363636364, + "grad_norm": 0.6668925881385803, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.6587, + "step": 42 + }, + { + "epoch": 0.29318181818181815, + "grad_norm": 0.681847333908081, + "learning_rate": 2.15e-06, + "loss": 0.6984, + "step": 43 + }, + { + "epoch": 0.3, + "grad_norm": 0.6222082376480103, + "learning_rate": 2.2e-06, + "loss": 0.7063, + "step": 44 + }, + { + "epoch": 0.3068181818181818, + "grad_norm": 0.790607213973999, + "learning_rate": 2.25e-06, + "loss": 0.6804, + "step": 45 + }, + { + "epoch": 0.31363636363636366, + "grad_norm": 0.6110427379608154, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.6694, + "step": 46 + }, + { + "epoch": 0.32045454545454544, + "grad_norm": 0.6481993794441223, + "learning_rate": 2.35e-06, + "loss": 0.7321, + "step": 47 + }, + { + "epoch": 0.32727272727272727, + "grad_norm": 0.5973438620567322, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.6508, + "step": 48 + }, + { + "epoch": 0.3340909090909091, + "grad_norm": 0.6252740621566772, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.6911, + "step": 49 + }, + { + "epoch": 0.3409090909090909, + "grad_norm": 0.6826481819152832, + "learning_rate": 2.5e-06, + "loss": 0.6855, + "step": 50 + }, + { + "epoch": 0.3477272727272727, + "grad_norm": 0.6307975649833679, + "learning_rate": 2.55e-06, + "loss": 0.714, + "step": 51 + }, + { + "epoch": 0.35454545454545455, + "grad_norm": 0.5900976061820984, + "learning_rate": 2.6e-06, + "loss": 0.6619, + "step": 52 + }, + { + "epoch": 0.3613636363636364, + "grad_norm": 0.6203920841217041, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.6822, + "step": 53 + }, + { + "epoch": 0.36818181818181817, + "grad_norm": 0.6197589039802551, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.6584, + "step": 54 + }, + { + "epoch": 0.375, + "grad_norm": 0.5921110510826111, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.6739, + "step": 55 + }, + { + "epoch": 0.38181818181818183, + "grad_norm": 0.6215619444847107, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.6631, + "step": 56 + }, + { + "epoch": 0.3886363636363636, + "grad_norm": 0.5495648980140686, + "learning_rate": 2.85e-06, + "loss": 0.6531, + "step": 57 + }, + { + "epoch": 0.39545454545454545, + "grad_norm": 0.531947910785675, + "learning_rate": 2.9e-06, + "loss": 0.6365, + "step": 58 + }, + { + "epoch": 0.4022727272727273, + "grad_norm": 0.559112548828125, + "learning_rate": 2.95e-06, + "loss": 0.6391, + "step": 59 + }, + { + "epoch": 0.4090909090909091, + "grad_norm": 0.5377560257911682, + "learning_rate": 3e-06, + "loss": 0.6775, + "step": 60 + }, + { + "epoch": 0.4159090909090909, + "grad_norm": 0.5167352557182312, + "learning_rate": 3.05e-06, + "loss": 0.6518, + "step": 61 + }, + { + "epoch": 0.42272727272727273, + "grad_norm": 0.5483390092849731, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.5972, + "step": 62 + }, + { + "epoch": 0.42954545454545456, + "grad_norm": 0.5687050223350525, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.6515, + "step": 63 + }, + { + "epoch": 0.43636363636363634, + "grad_norm": 0.9046968817710876, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.6497, + "step": 64 + }, + { + "epoch": 0.4431818181818182, + "grad_norm": 0.5231667757034302, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.6364, + "step": 65 + }, + { + "epoch": 0.45, + "grad_norm": 0.49699342250823975, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.6144, + "step": 66 + }, + { + "epoch": 0.45681818181818185, + "grad_norm": 0.5390080213546753, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.6051, + "step": 67 + }, + { + "epoch": 0.4636363636363636, + "grad_norm": 0.5252938270568848, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.6353, + "step": 68 + }, + { + "epoch": 0.47045454545454546, + "grad_norm": 0.6291780471801758, + "learning_rate": 3.45e-06, + "loss": 0.6326, + "step": 69 + }, + { + "epoch": 0.4772727272727273, + "grad_norm": 0.5545375943183899, + "learning_rate": 3.5e-06, + "loss": 0.647, + "step": 70 + }, + { + "epoch": 0.48409090909090907, + "grad_norm": 0.47775956988334656, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.5993, + "step": 71 + }, + { + "epoch": 0.4909090909090909, + "grad_norm": 0.5016375184059143, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.6056, + "step": 72 + }, + { + "epoch": 0.49772727272727274, + "grad_norm": 0.4780517816543579, + "learning_rate": 3.65e-06, + "loss": 0.6189, + "step": 73 + }, + { + "epoch": 0.5045454545454545, + "grad_norm": 0.5763716697692871, + "learning_rate": 3.7e-06, + "loss": 0.6262, + "step": 74 + }, + { + "epoch": 0.5113636363636364, + "grad_norm": 0.5321404337882996, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.6336, + "step": 75 + }, + { + "epoch": 0.5181818181818182, + "grad_norm": 0.4641991853713989, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.6317, + "step": 76 + }, + { + "epoch": 0.525, + "grad_norm": 0.45118412375450134, + "learning_rate": 3.85e-06, + "loss": 0.5927, + "step": 77 + }, + { + "epoch": 0.5318181818181819, + "grad_norm": 0.5500645041465759, + "learning_rate": 3.900000000000001e-06, + "loss": 0.62, + "step": 78 + }, + { + "epoch": 0.5386363636363637, + "grad_norm": 0.5197616219520569, + "learning_rate": 3.95e-06, + "loss": 0.6063, + "step": 79 + }, + { + "epoch": 0.5454545454545454, + "grad_norm": 0.4672992527484894, + "learning_rate": 4.000000000000001e-06, + "loss": 0.6156, + "step": 80 + }, + { + "epoch": 0.5522727272727272, + "grad_norm": 0.5995073318481445, + "learning_rate": 4.05e-06, + "loss": 0.6219, + "step": 81 + }, + { + "epoch": 0.5590909090909091, + "grad_norm": 0.5593804717063904, + "learning_rate": 4.1e-06, + "loss": 0.6282, + "step": 82 + }, + { + "epoch": 0.5659090909090909, + "grad_norm": 0.7544977068901062, + "learning_rate": 4.15e-06, + "loss": 0.6246, + "step": 83 + }, + { + "epoch": 0.5727272727272728, + "grad_norm": 0.45378655195236206, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.5881, + "step": 84 + }, + { + "epoch": 0.5795454545454546, + "grad_norm": 0.4407907724380493, + "learning_rate": 4.25e-06, + "loss": 0.6026, + "step": 85 + }, + { + "epoch": 0.5863636363636363, + "grad_norm": 0.4333001673221588, + "learning_rate": 4.3e-06, + "loss": 0.6098, + "step": 86 + }, + { + "epoch": 0.5931818181818181, + "grad_norm": 0.45020997524261475, + "learning_rate": 4.350000000000001e-06, + "loss": 0.5858, + "step": 87 + }, + { + "epoch": 0.6, + "grad_norm": 0.424003541469574, + "learning_rate": 4.4e-06, + "loss": 0.6186, + "step": 88 + }, + { + "epoch": 0.6068181818181818, + "grad_norm": 0.427081435918808, + "learning_rate": 4.450000000000001e-06, + "loss": 0.6164, + "step": 89 + }, + { + "epoch": 0.6136363636363636, + "grad_norm": 0.4430546164512634, + "learning_rate": 4.5e-06, + "loss": 0.6049, + "step": 90 + }, + { + "epoch": 0.6204545454545455, + "grad_norm": 0.5469494462013245, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.608, + "step": 91 + }, + { + "epoch": 0.6272727272727273, + "grad_norm": 0.4481559693813324, + "learning_rate": 4.600000000000001e-06, + "loss": 0.5532, + "step": 92 + }, + { + "epoch": 0.634090909090909, + "grad_norm": 0.6563135981559753, + "learning_rate": 4.65e-06, + "loss": 0.6105, + "step": 93 + }, + { + "epoch": 0.6409090909090909, + "grad_norm": 0.4572807252407074, + "learning_rate": 4.7e-06, + "loss": 0.5976, + "step": 94 + }, + { + "epoch": 0.6477272727272727, + "grad_norm": 0.4423324167728424, + "learning_rate": 4.75e-06, + "loss": 0.616, + "step": 95 + }, + { + "epoch": 0.6545454545454545, + "grad_norm": 0.6230632066726685, + "learning_rate": 4.800000000000001e-06, + "loss": 0.5879, + "step": 96 + }, + { + "epoch": 0.6613636363636364, + "grad_norm": 0.461467981338501, + "learning_rate": 4.85e-06, + "loss": 0.5992, + "step": 97 + }, + { + "epoch": 0.6681818181818182, + "grad_norm": 0.44455233216285706, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.6138, + "step": 98 + }, + { + "epoch": 0.675, + "grad_norm": 0.47281715273857117, + "learning_rate": 4.95e-06, + "loss": 0.59, + "step": 99 + }, + { + "epoch": 0.6818181818181818, + "grad_norm": 0.46114516258239746, + "learning_rate": 5e-06, + "loss": 0.6216, + "step": 100 + }, + { + "epoch": 0.6886363636363636, + "grad_norm": 0.49900761246681213, + "learning_rate": 4.9999795126530275e-06, + "loss": 0.6222, + "step": 101 + }, + { + "epoch": 0.6954545454545454, + "grad_norm": 0.4764839708805084, + "learning_rate": 4.999918050947891e-06, + "loss": 0.5829, + "step": 102 + }, + { + "epoch": 0.7022727272727273, + "grad_norm": 0.43502742052078247, + "learning_rate": 4.999815615891943e-06, + "loss": 0.5982, + "step": 103 + }, + { + "epoch": 0.7090909090909091, + "grad_norm": 0.43318599462509155, + "learning_rate": 4.9996722091640805e-06, + "loss": 0.5827, + "step": 104 + }, + { + "epoch": 0.7159090909090909, + "grad_norm": 0.47218361496925354, + "learning_rate": 4.9994878331147225e-06, + "loss": 0.5907, + "step": 105 + }, + { + "epoch": 0.7227272727272728, + "grad_norm": 0.48492228984832764, + "learning_rate": 4.99926249076577e-06, + "loss": 0.5459, + "step": 106 + }, + { + "epoch": 0.7295454545454545, + "grad_norm": 0.4019850194454193, + "learning_rate": 4.998996185810557e-06, + "loss": 0.5957, + "step": 107 + }, + { + "epoch": 0.7363636363636363, + "grad_norm": 0.47361984848976135, + "learning_rate": 4.998688922613788e-06, + "loss": 0.579, + "step": 108 + }, + { + "epoch": 0.7431818181818182, + "grad_norm": 0.4612269401550293, + "learning_rate": 4.9983407062114695e-06, + "loss": 0.5814, + "step": 109 + }, + { + "epoch": 0.75, + "grad_norm": 0.45558398962020874, + "learning_rate": 4.9979515423108255e-06, + "loss": 0.6036, + "step": 110 + }, + { + "epoch": 0.7568181818181818, + "grad_norm": 0.41492199897766113, + "learning_rate": 4.997521437290205e-06, + "loss": 0.5891, + "step": 111 + }, + { + "epoch": 0.7636363636363637, + "grad_norm": 0.4484409689903259, + "learning_rate": 4.997050398198977e-06, + "loss": 0.6031, + "step": 112 + }, + { + "epoch": 0.7704545454545455, + "grad_norm": 0.42671382427215576, + "learning_rate": 4.996538432757414e-06, + "loss": 0.6106, + "step": 113 + }, + { + "epoch": 0.7772727272727272, + "grad_norm": 0.4222075343132019, + "learning_rate": 4.995985549356568e-06, + "loss": 0.5774, + "step": 114 + }, + { + "epoch": 0.7840909090909091, + "grad_norm": 0.421021431684494, + "learning_rate": 4.995391757058129e-06, + "loss": 0.5754, + "step": 115 + }, + { + "epoch": 0.7909090909090909, + "grad_norm": 0.4215952754020691, + "learning_rate": 4.99475706559428e-06, + "loss": 0.5895, + "step": 116 + }, + { + "epoch": 0.7977272727272727, + "grad_norm": 0.42652982473373413, + "learning_rate": 4.994081485367537e-06, + "loss": 0.5646, + "step": 117 + }, + { + "epoch": 0.8045454545454546, + "grad_norm": 0.4145742654800415, + "learning_rate": 4.993365027450576e-06, + "loss": 0.5862, + "step": 118 + }, + { + "epoch": 0.8113636363636364, + "grad_norm": 0.4754612445831299, + "learning_rate": 4.992607703586058e-06, + "loss": 0.5645, + "step": 119 + }, + { + "epoch": 0.8181818181818182, + "grad_norm": 0.4694959223270416, + "learning_rate": 4.991809526186424e-06, + "loss": 0.5989, + "step": 120 + }, + { + "epoch": 0.825, + "grad_norm": 0.4573237895965576, + "learning_rate": 4.990970508333707e-06, + "loss": 0.5769, + "step": 121 + }, + { + "epoch": 0.8318181818181818, + "grad_norm": 0.4592607915401459, + "learning_rate": 4.990090663779305e-06, + "loss": 0.5529, + "step": 122 + }, + { + "epoch": 0.8386363636363636, + "grad_norm": 0.4325920045375824, + "learning_rate": 4.9891700069437635e-06, + "loss": 0.5846, + "step": 123 + }, + { + "epoch": 0.8454545454545455, + "grad_norm": 0.44623467326164246, + "learning_rate": 4.988208552916535e-06, + "loss": 0.5997, + "step": 124 + }, + { + "epoch": 0.8522727272727273, + "grad_norm": 0.46071872115135193, + "learning_rate": 4.987206317455734e-06, + "loss": 0.5687, + "step": 125 + }, + { + "epoch": 0.8590909090909091, + "grad_norm": 0.43121206760406494, + "learning_rate": 4.986163316987877e-06, + "loss": 0.5277, + "step": 126 + }, + { + "epoch": 0.865909090909091, + "grad_norm": 0.4711630046367645, + "learning_rate": 4.985079568607613e-06, + "loss": 0.592, + "step": 127 + }, + { + "epoch": 0.8727272727272727, + "grad_norm": 0.4450221657752991, + "learning_rate": 4.983955090077445e-06, + "loss": 0.5656, + "step": 128 + }, + { + "epoch": 0.8795454545454545, + "grad_norm": 0.6546564102172852, + "learning_rate": 4.982789899827439e-06, + "loss": 0.572, + "step": 129 + }, + { + "epoch": 0.8863636363636364, + "grad_norm": 0.49622273445129395, + "learning_rate": 4.9815840169549216e-06, + "loss": 0.5719, + "step": 130 + }, + { + "epoch": 0.8931818181818182, + "grad_norm": 0.5478544235229492, + "learning_rate": 4.980337461224164e-06, + "loss": 0.5774, + "step": 131 + }, + { + "epoch": 0.9, + "grad_norm": 0.4340018033981323, + "learning_rate": 4.979050253066064e-06, + "loss": 0.5444, + "step": 132 + }, + { + "epoch": 0.9068181818181819, + "grad_norm": 0.4243793189525604, + "learning_rate": 4.977722413577802e-06, + "loss": 0.5759, + "step": 133 + }, + { + "epoch": 0.9136363636363637, + "grad_norm": 0.45201942324638367, + "learning_rate": 4.976353964522509e-06, + "loss": 0.5746, + "step": 134 + }, + { + "epoch": 0.9204545454545454, + "grad_norm": 0.42892220616340637, + "learning_rate": 4.974944928328894e-06, + "loss": 0.544, + "step": 135 + }, + { + "epoch": 0.9272727272727272, + "grad_norm": 0.43185824155807495, + "learning_rate": 4.973495328090891e-06, + "loss": 0.5803, + "step": 136 + }, + { + "epoch": 0.9340909090909091, + "grad_norm": 0.5512542724609375, + "learning_rate": 4.972005187567267e-06, + "loss": 0.5694, + "step": 137 + }, + { + "epoch": 0.9409090909090909, + "grad_norm": 0.4836059808731079, + "learning_rate": 4.970474531181245e-06, + "loss": 0.5581, + "step": 138 + }, + { + "epoch": 0.9477272727272728, + "grad_norm": 0.4508282244205475, + "learning_rate": 4.968903384020095e-06, + "loss": 0.5553, + "step": 139 + }, + { + "epoch": 0.9545454545454546, + "grad_norm": 0.4764733612537384, + "learning_rate": 4.967291771834727e-06, + "loss": 0.5798, + "step": 140 + }, + { + "epoch": 0.9613636363636363, + "grad_norm": 0.463468074798584, + "learning_rate": 4.965639721039267e-06, + "loss": 0.559, + "step": 141 + }, + { + "epoch": 0.9681818181818181, + "grad_norm": 0.4655650854110718, + "learning_rate": 4.963947258710626e-06, + "loss": 0.5576, + "step": 142 + }, + { + "epoch": 0.975, + "grad_norm": 0.4651234745979309, + "learning_rate": 4.962214412588053e-06, + "loss": 0.5792, + "step": 143 + }, + { + "epoch": 0.9818181818181818, + "grad_norm": 0.43080538511276245, + "learning_rate": 4.960441211072686e-06, + "loss": 0.5314, + "step": 144 + }, + { + "epoch": 0.9886363636363636, + "grad_norm": 0.4403238594532013, + "learning_rate": 4.9586276832270785e-06, + "loss": 0.5592, + "step": 145 + }, + { + "epoch": 0.9954545454545455, + "grad_norm": 0.43968647718429565, + "learning_rate": 4.9567738587747314e-06, + "loss": 0.5573, + "step": 146 + }, + { + "epoch": 1.0068181818181818, + "grad_norm": 0.8337184190750122, + "learning_rate": 4.954879768099599e-06, + "loss": 1.1118, + "step": 147 + }, + { + "epoch": 1.0136363636363637, + "grad_norm": 0.4042762815952301, + "learning_rate": 4.952945442245598e-06, + "loss": 0.5731, + "step": 148 + }, + { + "epoch": 1.0204545454545455, + "grad_norm": 0.5666297078132629, + "learning_rate": 4.95097091291609e-06, + "loss": 0.5415, + "step": 149 + }, + { + "epoch": 1.0272727272727273, + "grad_norm": 0.6148372888565063, + "learning_rate": 4.948956212473371e-06, + "loss": 0.5486, + "step": 150 + }, + { + "epoch": 1.0340909090909092, + "grad_norm": 0.5780720710754395, + "learning_rate": 4.946901373938132e-06, + "loss": 0.5373, + "step": 151 + }, + { + "epoch": 1.040909090909091, + "grad_norm": 0.4648835361003876, + "learning_rate": 4.944806430988927e-06, + "loss": 0.5419, + "step": 152 + }, + { + "epoch": 1.0477272727272728, + "grad_norm": 0.41131219267845154, + "learning_rate": 4.942671417961615e-06, + "loss": 0.5514, + "step": 153 + }, + { + "epoch": 1.0545454545454545, + "grad_norm": 0.4733349084854126, + "learning_rate": 4.940496369848795e-06, + "loss": 0.5563, + "step": 154 + }, + { + "epoch": 1.0613636363636363, + "grad_norm": 0.41179028153419495, + "learning_rate": 4.938281322299243e-06, + "loss": 0.5413, + "step": 155 + }, + { + "epoch": 1.0681818181818181, + "grad_norm": 0.4770941138267517, + "learning_rate": 4.936026311617316e-06, + "loss": 0.5561, + "step": 156 + }, + { + "epoch": 1.075, + "grad_norm": 0.7752397060394287, + "learning_rate": 4.933731374762361e-06, + "loss": 0.5439, + "step": 157 + }, + { + "epoch": 1.0818181818181818, + "grad_norm": 0.4323958456516266, + "learning_rate": 4.931396549348115e-06, + "loss": 0.5401, + "step": 158 + }, + { + "epoch": 1.0886363636363636, + "grad_norm": 0.4497937858104706, + "learning_rate": 4.9290218736420795e-06, + "loss": 0.523, + "step": 159 + }, + { + "epoch": 1.0954545454545455, + "grad_norm": 0.4881570339202881, + "learning_rate": 4.926607386564898e-06, + "loss": 0.5441, + "step": 160 + }, + { + "epoch": 1.1022727272727273, + "grad_norm": 0.43357932567596436, + "learning_rate": 4.9241531276897196e-06, + "loss": 0.5499, + "step": 161 + }, + { + "epoch": 1.1090909090909091, + "grad_norm": 1.1125361919403076, + "learning_rate": 4.921659137241544e-06, + "loss": 0.5265, + "step": 162 + }, + { + "epoch": 1.115909090909091, + "grad_norm": 0.43150508403778076, + "learning_rate": 4.919125456096574e-06, + "loss": 0.5193, + "step": 163 + }, + { + "epoch": 1.1227272727272728, + "grad_norm": 0.40771809220314026, + "learning_rate": 4.916552125781529e-06, + "loss": 0.538, + "step": 164 + }, + { + "epoch": 1.1295454545454546, + "grad_norm": 0.5830567479133606, + "learning_rate": 4.913939188472979e-06, + "loss": 0.5548, + "step": 165 + }, + { + "epoch": 1.1363636363636362, + "grad_norm": 0.4776526391506195, + "learning_rate": 4.911286686996648e-06, + "loss": 0.5328, + "step": 166 + }, + { + "epoch": 1.143181818181818, + "grad_norm": 0.41657742857933044, + "learning_rate": 4.908594664826708e-06, + "loss": 0.5336, + "step": 167 + }, + { + "epoch": 1.15, + "grad_norm": 0.4651948809623718, + "learning_rate": 4.905863166085076e-06, + "loss": 0.5812, + "step": 168 + }, + { + "epoch": 1.1568181818181817, + "grad_norm": 0.4399661719799042, + "learning_rate": 4.903092235540679e-06, + "loss": 0.5143, + "step": 169 + }, + { + "epoch": 1.1636363636363636, + "grad_norm": 0.4548935294151306, + "learning_rate": 4.900281918608732e-06, + "loss": 0.5425, + "step": 170 + }, + { + "epoch": 1.1704545454545454, + "grad_norm": 0.43515875935554504, + "learning_rate": 4.897432261349984e-06, + "loss": 0.534, + "step": 171 + }, + { + "epoch": 1.1772727272727272, + "grad_norm": 0.42954760789871216, + "learning_rate": 4.894543310469968e-06, + "loss": 0.5436, + "step": 172 + }, + { + "epoch": 1.184090909090909, + "grad_norm": 0.4063640534877777, + "learning_rate": 4.891615113318236e-06, + "loss": 0.5481, + "step": 173 + }, + { + "epoch": 1.190909090909091, + "grad_norm": 0.49416476488113403, + "learning_rate": 4.888647717887582e-06, + "loss": 0.5298, + "step": 174 + }, + { + "epoch": 1.1977272727272728, + "grad_norm": 1.4082964658737183, + "learning_rate": 4.8856411728132526e-06, + "loss": 0.5366, + "step": 175 + }, + { + "epoch": 1.2045454545454546, + "grad_norm": 0.44356465339660645, + "learning_rate": 4.8825955273721524e-06, + "loss": 0.5563, + "step": 176 + }, + { + "epoch": 1.2113636363636364, + "grad_norm": 0.40925371646881104, + "learning_rate": 4.879510831482039e-06, + "loss": 0.5444, + "step": 177 + }, + { + "epoch": 1.2181818181818183, + "grad_norm": 0.44379451870918274, + "learning_rate": 4.876387135700701e-06, + "loss": 0.5311, + "step": 178 + }, + { + "epoch": 1.225, + "grad_norm": 0.41487446427345276, + "learning_rate": 4.873224491225128e-06, + "loss": 0.5637, + "step": 179 + }, + { + "epoch": 1.231818181818182, + "grad_norm": 0.4520554542541504, + "learning_rate": 4.870022949890676e-06, + "loss": 0.5677, + "step": 180 + }, + { + "epoch": 1.2386363636363638, + "grad_norm": 0.4528166949748993, + "learning_rate": 4.866782564170217e-06, + "loss": 0.5095, + "step": 181 + }, + { + "epoch": 1.2454545454545454, + "grad_norm": 0.4409842789173126, + "learning_rate": 4.863503387173276e-06, + "loss": 0.5423, + "step": 182 + }, + { + "epoch": 1.2522727272727272, + "grad_norm": 0.4355996251106262, + "learning_rate": 4.860185472645161e-06, + "loss": 0.5422, + "step": 183 + }, + { + "epoch": 1.259090909090909, + "grad_norm": 0.4871126115322113, + "learning_rate": 4.856828874966086e-06, + "loss": 0.5293, + "step": 184 + }, + { + "epoch": 1.2659090909090909, + "grad_norm": 0.442348450422287, + "learning_rate": 4.853433649150276e-06, + "loss": 0.5616, + "step": 185 + }, + { + "epoch": 1.2727272727272727, + "grad_norm": 0.44014668464660645, + "learning_rate": 4.849999850845066e-06, + "loss": 0.5354, + "step": 186 + }, + { + "epoch": 1.2795454545454545, + "grad_norm": 0.4496249854564667, + "learning_rate": 4.8465275363299905e-06, + "loss": 0.5298, + "step": 187 + }, + { + "epoch": 1.2863636363636364, + "grad_norm": 0.46060171723365784, + "learning_rate": 4.84301676251586e-06, + "loss": 0.554, + "step": 188 + }, + { + "epoch": 1.2931818181818182, + "grad_norm": 0.7829347252845764, + "learning_rate": 4.839467586943825e-06, + "loss": 0.5606, + "step": 189 + }, + { + "epoch": 1.3, + "grad_norm": 0.45566731691360474, + "learning_rate": 4.835880067784441e-06, + "loss": 0.5118, + "step": 190 + }, + { + "epoch": 1.3068181818181819, + "grad_norm": 0.43211469054222107, + "learning_rate": 4.832254263836708e-06, + "loss": 0.535, + "step": 191 + }, + { + "epoch": 1.3136363636363637, + "grad_norm": 0.4350590705871582, + "learning_rate": 4.828590234527107e-06, + "loss": 0.5595, + "step": 192 + }, + { + "epoch": 1.3204545454545453, + "grad_norm": 0.4056399464607239, + "learning_rate": 4.82488803990863e-06, + "loss": 0.55, + "step": 193 + }, + { + "epoch": 1.3272727272727272, + "grad_norm": 0.4458341598510742, + "learning_rate": 4.821147740659795e-06, + "loss": 0.501, + "step": 194 + }, + { + "epoch": 1.334090909090909, + "grad_norm": 0.446504145860672, + "learning_rate": 4.817369398083648e-06, + "loss": 0.5393, + "step": 195 + }, + { + "epoch": 1.3409090909090908, + "grad_norm": 0.47838401794433594, + "learning_rate": 4.813553074106761e-06, + "loss": 0.5354, + "step": 196 + }, + { + "epoch": 1.3477272727272727, + "grad_norm": 0.4675792157649994, + "learning_rate": 4.809698831278217e-06, + "loss": 0.5202, + "step": 197 + }, + { + "epoch": 1.3545454545454545, + "grad_norm": 0.4351370632648468, + "learning_rate": 4.805806732768585e-06, + "loss": 0.5256, + "step": 198 + }, + { + "epoch": 1.3613636363636363, + "grad_norm": 0.4466511607170105, + "learning_rate": 4.801876842368882e-06, + "loss": 0.5238, + "step": 199 + }, + { + "epoch": 1.3681818181818182, + "grad_norm": 0.45672544836997986, + "learning_rate": 4.797909224489531e-06, + "loss": 0.5426, + "step": 200 + }, + { + "epoch": 1.375, + "grad_norm": 0.422638863325119, + "learning_rate": 4.793903944159303e-06, + "loss": 0.5385, + "step": 201 + }, + { + "epoch": 1.3818181818181818, + "grad_norm": 0.4254551827907562, + "learning_rate": 4.789861067024253e-06, + "loss": 0.503, + "step": 202 + }, + { + "epoch": 1.3886363636363637, + "grad_norm": 0.43860334157943726, + "learning_rate": 4.785780659346642e-06, + "loss": 0.519, + "step": 203 + }, + { + "epoch": 1.3954545454545455, + "grad_norm": 0.43294480443000793, + "learning_rate": 4.781662788003851e-06, + "loss": 0.5537, + "step": 204 + }, + { + "epoch": 1.4022727272727273, + "grad_norm": 0.45006492733955383, + "learning_rate": 4.777507520487289e-06, + "loss": 0.5318, + "step": 205 + }, + { + "epoch": 1.4090909090909092, + "grad_norm": 0.4178854823112488, + "learning_rate": 4.773314924901281e-06, + "loss": 0.5287, + "step": 206 + }, + { + "epoch": 1.415909090909091, + "grad_norm": 0.45662209391593933, + "learning_rate": 4.769085069961955e-06, + "loss": 0.5609, + "step": 207 + }, + { + "epoch": 1.4227272727272728, + "grad_norm": 0.41578733921051025, + "learning_rate": 4.764818024996117e-06, + "loss": 0.5249, + "step": 208 + }, + { + "epoch": 1.4295454545454547, + "grad_norm": 0.48305758833885193, + "learning_rate": 4.760513859940112e-06, + "loss": 0.515, + "step": 209 + }, + { + "epoch": 1.4363636363636363, + "grad_norm": 0.4564506709575653, + "learning_rate": 4.756172645338675e-06, + "loss": 0.535, + "step": 210 + }, + { + "epoch": 1.4431818181818181, + "grad_norm": 0.4499240815639496, + "learning_rate": 4.751794452343785e-06, + "loss": 0.5236, + "step": 211 + }, + { + "epoch": 1.45, + "grad_norm": 0.41851088404655457, + "learning_rate": 4.747379352713489e-06, + "loss": 0.5254, + "step": 212 + }, + { + "epoch": 1.4568181818181818, + "grad_norm": 0.4690937399864197, + "learning_rate": 4.7429274188107275e-06, + "loss": 0.5312, + "step": 213 + }, + { + "epoch": 1.4636363636363636, + "grad_norm": 0.5516958832740784, + "learning_rate": 4.738438723602154e-06, + "loss": 0.5315, + "step": 214 + }, + { + "epoch": 1.4704545454545455, + "grad_norm": 1.2040799856185913, + "learning_rate": 4.733913340656933e-06, + "loss": 0.544, + "step": 215 + }, + { + "epoch": 1.4772727272727273, + "grad_norm": 0.445425808429718, + "learning_rate": 4.729351344145536e-06, + "loss": 0.5309, + "step": 216 + }, + { + "epoch": 1.4840909090909091, + "grad_norm": 0.4186754822731018, + "learning_rate": 4.7247528088385296e-06, + "loss": 0.5138, + "step": 217 + }, + { + "epoch": 1.490909090909091, + "grad_norm": 0.43284520506858826, + "learning_rate": 4.720117810105341e-06, + "loss": 0.521, + "step": 218 + }, + { + "epoch": 1.4977272727272728, + "grad_norm": 0.4308505058288574, + "learning_rate": 4.715446423913036e-06, + "loss": 0.5436, + "step": 219 + }, + { + "epoch": 1.5045454545454544, + "grad_norm": 0.4471941888332367, + "learning_rate": 4.710738726825059e-06, + "loss": 0.5528, + "step": 220 + }, + { + "epoch": 1.5113636363636362, + "grad_norm": 0.4374965727329254, + "learning_rate": 4.705994795999991e-06, + "loss": 0.5238, + "step": 221 + }, + { + "epoch": 1.518181818181818, + "grad_norm": 0.6026428937911987, + "learning_rate": 4.701214709190277e-06, + "loss": 0.5268, + "step": 222 + }, + { + "epoch": 1.525, + "grad_norm": 0.4413348138332367, + "learning_rate": 4.696398544740955e-06, + "loss": 0.5534, + "step": 223 + }, + { + "epoch": 1.5318181818181817, + "grad_norm": 0.45917898416519165, + "learning_rate": 4.69154638158837e-06, + "loss": 0.511, + "step": 224 + }, + { + "epoch": 1.5386363636363636, + "grad_norm": 0.5339583158493042, + "learning_rate": 4.686658299258881e-06, + "loss": 0.5395, + "step": 225 + }, + { + "epoch": 1.5454545454545454, + "grad_norm": 0.4617757201194763, + "learning_rate": 4.681734377867562e-06, + "loss": 0.5425, + "step": 226 + }, + { + "epoch": 1.5522727272727272, + "grad_norm": 0.4513031840324402, + "learning_rate": 4.67677469811688e-06, + "loss": 0.5241, + "step": 227 + }, + { + "epoch": 1.559090909090909, + "grad_norm": 0.43548423051834106, + "learning_rate": 4.671779341295378e-06, + "loss": 0.5224, + "step": 228 + }, + { + "epoch": 1.565909090909091, + "grad_norm": 0.5516494512557983, + "learning_rate": 4.666748389276344e-06, + "loss": 0.5346, + "step": 229 + }, + { + "epoch": 1.5727272727272728, + "grad_norm": 0.4441164433956146, + "learning_rate": 4.661681924516466e-06, + "loss": 0.5208, + "step": 230 + }, + { + "epoch": 1.5795454545454546, + "grad_norm": 0.4057452380657196, + "learning_rate": 4.6565800300544805e-06, + "loss": 0.5173, + "step": 231 + }, + { + "epoch": 1.5863636363636364, + "grad_norm": 0.4162004590034485, + "learning_rate": 4.651442789509813e-06, + "loss": 0.5307, + "step": 232 + }, + { + "epoch": 1.5931818181818183, + "grad_norm": 0.43314146995544434, + "learning_rate": 4.646270287081208e-06, + "loss": 0.5359, + "step": 233 + }, + { + "epoch": 1.6, + "grad_norm": 0.5786082148551941, + "learning_rate": 4.641062607545347e-06, + "loss": 0.5307, + "step": 234 + }, + { + "epoch": 1.606818181818182, + "grad_norm": 0.43811941146850586, + "learning_rate": 4.6358198362554585e-06, + "loss": 0.5556, + "step": 235 + }, + { + "epoch": 1.6136363636363638, + "grad_norm": 0.4374074935913086, + "learning_rate": 4.630542059139923e-06, + "loss": 0.5253, + "step": 236 + }, + { + "epoch": 1.6204545454545456, + "grad_norm": 0.4387815594673157, + "learning_rate": 4.625229362700863e-06, + "loss": 0.534, + "step": 237 + }, + { + "epoch": 1.6272727272727274, + "grad_norm": 0.4976162016391754, + "learning_rate": 4.61988183401272e-06, + "loss": 0.5255, + "step": 238 + }, + { + "epoch": 1.634090909090909, + "grad_norm": 0.4603639543056488, + "learning_rate": 4.614499560720837e-06, + "loss": 0.511, + "step": 239 + }, + { + "epoch": 1.6409090909090909, + "grad_norm": 0.43136921525001526, + "learning_rate": 4.609082631040012e-06, + "loss": 0.5099, + "step": 240 + }, + { + "epoch": 1.6477272727272727, + "grad_norm": 0.4573090970516205, + "learning_rate": 4.603631133753061e-06, + "loss": 0.5515, + "step": 241 + }, + { + "epoch": 1.6545454545454545, + "grad_norm": 0.4292807877063751, + "learning_rate": 4.598145158209356e-06, + "loss": 0.5048, + "step": 242 + }, + { + "epoch": 1.6613636363636364, + "grad_norm": 0.6377942562103271, + "learning_rate": 4.592624794323366e-06, + "loss": 0.5333, + "step": 243 + }, + { + "epoch": 1.6681818181818182, + "grad_norm": 0.49924615025520325, + "learning_rate": 4.587070132573178e-06, + "loss": 0.5127, + "step": 244 + }, + { + "epoch": 1.675, + "grad_norm": 0.4723946154117584, + "learning_rate": 4.581481263999019e-06, + "loss": 0.5127, + "step": 245 + }, + { + "epoch": 1.6818181818181817, + "grad_norm": 1.1846898794174194, + "learning_rate": 4.575858280201761e-06, + "loss": 0.5367, + "step": 246 + }, + { + "epoch": 1.6886363636363635, + "grad_norm": 0.4306166172027588, + "learning_rate": 4.570201273341418e-06, + "loss": 0.52, + "step": 247 + }, + { + "epoch": 1.6954545454545453, + "grad_norm": 0.45308929681777954, + "learning_rate": 4.564510336135642e-06, + "loss": 0.5589, + "step": 248 + }, + { + "epoch": 1.7022727272727272, + "grad_norm": 0.4482448399066925, + "learning_rate": 4.558785561858196e-06, + "loss": 0.5161, + "step": 249 + }, + { + "epoch": 1.709090909090909, + "grad_norm": 0.44250935316085815, + "learning_rate": 4.5530270443374305e-06, + "loss": 0.5342, + "step": 250 + }, + { + "epoch": 1.7159090909090908, + "grad_norm": 0.8177540898323059, + "learning_rate": 4.547234877954741e-06, + "loss": 0.5402, + "step": 251 + }, + { + "epoch": 1.7227272727272727, + "grad_norm": 0.5169854760169983, + "learning_rate": 4.541409157643027e-06, + "loss": 0.5301, + "step": 252 + }, + { + "epoch": 1.7295454545454545, + "grad_norm": 0.4544285237789154, + "learning_rate": 4.535549978885132e-06, + "loss": 0.5314, + "step": 253 + }, + { + "epoch": 1.7363636363636363, + "grad_norm": 0.42860639095306396, + "learning_rate": 4.5296574377122765e-06, + "loss": 0.5249, + "step": 254 + }, + { + "epoch": 1.7431818181818182, + "grad_norm": 0.526492178440094, + "learning_rate": 4.5237316307024895e-06, + "loss": 0.5142, + "step": 255 + }, + { + "epoch": 1.75, + "grad_norm": 0.4208003878593445, + "learning_rate": 4.517772654979024e-06, + "loss": 0.5151, + "step": 256 + }, + { + "epoch": 1.7568181818181818, + "grad_norm": 0.4974212944507599, + "learning_rate": 4.51178060820876e-06, + "loss": 0.5161, + "step": 257 + }, + { + "epoch": 1.7636363636363637, + "grad_norm": 0.4148060083389282, + "learning_rate": 4.505755588600613e-06, + "loss": 0.5099, + "step": 258 + }, + { + "epoch": 1.7704545454545455, + "grad_norm": 0.4106168746948242, + "learning_rate": 4.499697694903915e-06, + "loss": 0.5376, + "step": 259 + }, + { + "epoch": 1.7772727272727273, + "grad_norm": 0.49017220735549927, + "learning_rate": 4.493607026406802e-06, + "loss": 0.525, + "step": 260 + }, + { + "epoch": 1.7840909090909092, + "grad_norm": 0.4141218662261963, + "learning_rate": 4.487483682934587e-06, + "loss": 0.5303, + "step": 261 + }, + { + "epoch": 1.790909090909091, + "grad_norm": 0.44826585054397583, + "learning_rate": 4.481327764848118e-06, + "loss": 0.4886, + "step": 262 + }, + { + "epoch": 1.7977272727272728, + "grad_norm": 0.42013809084892273, + "learning_rate": 4.47513937304214e-06, + "loss": 0.5204, + "step": 263 + }, + { + "epoch": 1.8045454545454547, + "grad_norm": 0.44327741861343384, + "learning_rate": 4.4689186089436365e-06, + "loss": 0.5097, + "step": 264 + }, + { + "epoch": 1.8113636363636365, + "grad_norm": 0.39498746395111084, + "learning_rate": 4.462665574510169e-06, + "loss": 0.5102, + "step": 265 + }, + { + "epoch": 1.8181818181818183, + "grad_norm": 0.48296087980270386, + "learning_rate": 4.456380372228208e-06, + "loss": 0.5299, + "step": 266 + }, + { + "epoch": 1.825, + "grad_norm": 0.4398634433746338, + "learning_rate": 4.450063105111447e-06, + "loss": 0.5419, + "step": 267 + }, + { + "epoch": 1.8318181818181818, + "grad_norm": 0.43333035707473755, + "learning_rate": 4.443713876699124e-06, + "loss": 0.5047, + "step": 268 + }, + { + "epoch": 1.8386363636363636, + "grad_norm": 0.5823659300804138, + "learning_rate": 4.4373327910543125e-06, + "loss": 0.5082, + "step": 269 + }, + { + "epoch": 1.8454545454545455, + "grad_norm": 0.44385817646980286, + "learning_rate": 4.430919952762226e-06, + "loss": 0.5171, + "step": 270 + }, + { + "epoch": 1.8522727272727273, + "grad_norm": 0.4444347023963928, + "learning_rate": 4.424475466928499e-06, + "loss": 0.5351, + "step": 271 + }, + { + "epoch": 1.8590909090909091, + "grad_norm": 0.549674391746521, + "learning_rate": 4.417999439177465e-06, + "loss": 0.498, + "step": 272 + }, + { + "epoch": 1.865909090909091, + "grad_norm": 0.4340274930000305, + "learning_rate": 4.4114919756504275e-06, + "loss": 0.5429, + "step": 273 + }, + { + "epoch": 1.8727272727272726, + "grad_norm": 0.44019854068756104, + "learning_rate": 4.404953183003916e-06, + "loss": 0.5133, + "step": 274 + }, + { + "epoch": 1.8795454545454544, + "grad_norm": 0.4473486542701721, + "learning_rate": 4.398383168407941e-06, + "loss": 0.5413, + "step": 275 + }, + { + "epoch": 1.8863636363636362, + "grad_norm": 0.44450780749320984, + "learning_rate": 4.391782039544239e-06, + "loss": 0.4998, + "step": 276 + }, + { + "epoch": 1.893181818181818, + "grad_norm": 0.4310123920440674, + "learning_rate": 4.385149904604502e-06, + "loss": 0.5134, + "step": 277 + }, + { + "epoch": 1.9, + "grad_norm": 0.48953917622566223, + "learning_rate": 4.378486872288611e-06, + "loss": 0.5102, + "step": 278 + }, + { + "epoch": 1.9068181818181817, + "grad_norm": 0.49219778180122375, + "learning_rate": 4.371793051802849e-06, + "loss": 0.5213, + "step": 279 + }, + { + "epoch": 1.9136363636363636, + "grad_norm": 0.818645715713501, + "learning_rate": 4.365068552858116e-06, + "loss": 0.5251, + "step": 280 + }, + { + "epoch": 1.9204545454545454, + "grad_norm": 0.4407028257846832, + "learning_rate": 4.358313485668124e-06, + "loss": 0.5191, + "step": 281 + }, + { + "epoch": 1.9272727272727272, + "grad_norm": 0.5301506519317627, + "learning_rate": 4.3515279609476e-06, + "loss": 0.5003, + "step": 282 + }, + { + "epoch": 1.934090909090909, + "grad_norm": 0.4673819839954376, + "learning_rate": 4.3447120899104615e-06, + "loss": 0.521, + "step": 283 + }, + { + "epoch": 1.940909090909091, + "grad_norm": 0.4106168746948242, + "learning_rate": 4.337865984268002e-06, + "loss": 0.5199, + "step": 284 + }, + { + "epoch": 1.9477272727272728, + "grad_norm": 0.43583595752716064, + "learning_rate": 4.3309897562270525e-06, + "loss": 0.5367, + "step": 285 + }, + { + "epoch": 1.9545454545454546, + "grad_norm": 0.4984910488128662, + "learning_rate": 4.324083518488151e-06, + "loss": 0.5202, + "step": 286 + }, + { + "epoch": 1.9613636363636364, + "grad_norm": 0.4516090452671051, + "learning_rate": 4.317147384243688e-06, + "loss": 0.5432, + "step": 287 + }, + { + "epoch": 1.9681818181818183, + "grad_norm": 0.43753015995025635, + "learning_rate": 4.3101814671760546e-06, + "loss": 0.5309, + "step": 288 + }, + { + "epoch": 1.975, + "grad_norm": 0.4472252130508423, + "learning_rate": 4.303185881455778e-06, + "loss": 0.5344, + "step": 289 + }, + { + "epoch": 1.981818181818182, + "grad_norm": 1.178377389907837, + "learning_rate": 4.296160741739652e-06, + "loss": 0.5283, + "step": 290 + }, + { + "epoch": 1.9886363636363638, + "grad_norm": 0.5085738301277161, + "learning_rate": 4.289106163168858e-06, + "loss": 0.5133, + "step": 291 + }, + { + "epoch": 1.9954545454545456, + "grad_norm": 0.52077716588974, + "learning_rate": 4.282022261367074e-06, + "loss": 0.5155, + "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..a03cebdd4a04fb2e9443a05c92f15b2b2db01e10 --- /dev/null +++ b/checkpoint-292/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90038f7ff377ce29f00aedab533b18bfc9515b30d2b4232d8b0d6de26ec3ea56 +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..fdf619c317c2fe82074662582dbd68166b6f9d50 --- /dev/null +++ b/checkpoint-438/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3.1-Nemotron-70B-Instruct-HF +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..604978b603be594e46459a6c8c10d206a5a03b31 --- /dev/null +++ b/checkpoint-438/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF", + "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": [ + "down_proj", + "v_proj", + "q_proj", + "gate_proj", + "o_proj", + "up_proj", + "k_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..c017c62913f712dc113091c3e075bb99ecc49db1 --- /dev/null +++ b/checkpoint-438/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:1d826f61c317f3b820b34657638305bbf1d66a25ee65187001174fd37ec661eb +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..49a2d530d01d71160c91b22961de89e8c22513e3 --- /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:eb74262d3662c991e6d75a270c1f69936a74e9141453b38651481f980c1f9348 +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..d4980e7fe073f95674dd3e6417345fe2dfe3a39c --- /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 sha256:78c9afd9aad490c045f0342429f4852c2d3bdbfa7da89257c57e9c03a546ec13 +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..2cf063fa82b4bcfd223dc432754e741db9449be1 --- /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 sha256:0441e49d8dd4fe238f9d7abb6d7ba15247148f90e635fede6551d7c0a859b67d +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..7e7924caa5e43e679e0476c0efe4b6b80839416b --- /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 sha256:55e67e16d77b12e817dada206a917c56e22928788b517810357861667516c95e +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..fdde94c29816839ec3c29d6c6461206a49911f3c --- /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\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..f2af924324afdfc6e933e24fddfd4f4b156f50c3 --- /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": 2.020231246948242, + "learning_rate": 5.0000000000000004e-08, + "loss": 0.8736, + "step": 1 + }, + { + "epoch": 0.013636363636363636, + "grad_norm": 1.9368854761123657, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.8543, + "step": 2 + }, + { + "epoch": 0.020454545454545454, + "grad_norm": 1.908379077911377, + "learning_rate": 1.5000000000000002e-07, + "loss": 0.8339, + "step": 3 + }, + { + "epoch": 0.02727272727272727, + "grad_norm": 1.9313658475875854, + "learning_rate": 2.0000000000000002e-07, + "loss": 0.834, + "step": 4 + }, + { + "epoch": 0.03409090909090909, + "grad_norm": 1.983262062072754, + "learning_rate": 2.5000000000000004e-07, + "loss": 0.8726, + "step": 5 + }, + { + "epoch": 0.04090909090909091, + "grad_norm": 1.9994115829467773, + "learning_rate": 3.0000000000000004e-07, + "loss": 0.8135, + "step": 6 + }, + { + "epoch": 0.04772727272727273, + "grad_norm": 1.8889026641845703, + "learning_rate": 3.5000000000000004e-07, + "loss": 0.8439, + "step": 7 + }, + { + "epoch": 0.05454545454545454, + "grad_norm": 1.9258317947387695, + "learning_rate": 4.0000000000000003e-07, + "loss": 0.84, + "step": 8 + }, + { + "epoch": 0.06136363636363636, + "grad_norm": 1.9654443264007568, + "learning_rate": 4.5000000000000003e-07, + "loss": 0.8609, + "step": 9 + }, + { + "epoch": 0.06818181818181818, + "grad_norm": 1.9278813600540161, + "learning_rate": 5.000000000000001e-07, + "loss": 0.8515, + "step": 10 + }, + { + "epoch": 0.075, + "grad_norm": 1.8698863983154297, + "learning_rate": 5.5e-07, + "loss": 0.8632, + "step": 11 + }, + { + "epoch": 0.08181818181818182, + "grad_norm": 2.7078285217285156, + "learning_rate": 6.000000000000001e-07, + "loss": 0.8643, + "step": 12 + }, + { + "epoch": 0.08863636363636364, + "grad_norm": 1.959272027015686, + "learning_rate": 6.5e-07, + "loss": 0.861, + "step": 13 + }, + { + "epoch": 0.09545454545454546, + "grad_norm": 1.8952852487564087, + "learning_rate": 7.000000000000001e-07, + "loss": 0.8599, + "step": 14 + }, + { + "epoch": 0.10227272727272728, + "grad_norm": 1.9469462633132935, + "learning_rate": 7.5e-07, + "loss": 0.8467, + "step": 15 + }, + { + "epoch": 0.10909090909090909, + "grad_norm": 2.0111300945281982, + "learning_rate": 8.000000000000001e-07, + "loss": 0.8078, + "step": 16 + }, + { + "epoch": 0.1159090909090909, + "grad_norm": 1.739009141921997, + "learning_rate": 8.500000000000001e-07, + "loss": 0.8114, + "step": 17 + }, + { + "epoch": 0.12272727272727273, + "grad_norm": 1.7578809261322021, + "learning_rate": 9.000000000000001e-07, + "loss": 0.8489, + "step": 18 + }, + { + "epoch": 0.12954545454545455, + "grad_norm": 1.9787132740020752, + "learning_rate": 9.500000000000001e-07, + "loss": 0.8052, + "step": 19 + }, + { + "epoch": 0.13636363636363635, + "grad_norm": 1.6380164623260498, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.7882, + "step": 20 + }, + { + "epoch": 0.1431818181818182, + "grad_norm": 1.686877727508545, + "learning_rate": 1.0500000000000001e-06, + "loss": 0.8126, + "step": 21 + }, + { + "epoch": 0.15, + "grad_norm": 1.690590739250183, + "learning_rate": 1.1e-06, + "loss": 0.8248, + "step": 22 + }, + { + "epoch": 0.15681818181818183, + "grad_norm": 1.612255334854126, + "learning_rate": 1.1500000000000002e-06, + "loss": 0.8024, + "step": 23 + }, + { + "epoch": 0.16363636363636364, + "grad_norm": 1.6669739484786987, + "learning_rate": 1.2000000000000002e-06, + "loss": 0.818, + "step": 24 + }, + { + "epoch": 0.17045454545454544, + "grad_norm": 1.7726502418518066, + "learning_rate": 1.25e-06, + "loss": 0.8158, + "step": 25 + }, + { + "epoch": 0.17727272727272728, + "grad_norm": 1.5930036306381226, + "learning_rate": 1.3e-06, + "loss": 0.7886, + "step": 26 + }, + { + "epoch": 0.18409090909090908, + "grad_norm": 1.5480856895446777, + "learning_rate": 1.3500000000000002e-06, + "loss": 0.7747, + "step": 27 + }, + { + "epoch": 0.19090909090909092, + "grad_norm": 1.3537116050720215, + "learning_rate": 1.4000000000000001e-06, + "loss": 0.7891, + "step": 28 + }, + { + "epoch": 0.19772727272727272, + "grad_norm": 1.3412829637527466, + "learning_rate": 1.45e-06, + "loss": 0.7799, + "step": 29 + }, + { + "epoch": 0.20454545454545456, + "grad_norm": 1.237492322921753, + "learning_rate": 1.5e-06, + "loss": 0.7768, + "step": 30 + }, + { + "epoch": 0.21136363636363636, + "grad_norm": 1.3467835187911987, + "learning_rate": 1.5500000000000002e-06, + "loss": 0.7344, + "step": 31 + }, + { + "epoch": 0.21818181818181817, + "grad_norm": 1.1665314435958862, + "learning_rate": 1.6000000000000001e-06, + "loss": 0.7152, + "step": 32 + }, + { + "epoch": 0.225, + "grad_norm": 1.2461743354797363, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.758, + "step": 33 + }, + { + "epoch": 0.2318181818181818, + "grad_norm": 1.0343059301376343, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.7362, + "step": 34 + }, + { + "epoch": 0.23863636363636365, + "grad_norm": 1.112117052078247, + "learning_rate": 1.75e-06, + "loss": 0.7771, + "step": 35 + }, + { + "epoch": 0.24545454545454545, + "grad_norm": 1.104337453842163, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.7479, + "step": 36 + }, + { + "epoch": 0.25227272727272726, + "grad_norm": 0.9324578642845154, + "learning_rate": 1.85e-06, + "loss": 0.6936, + "step": 37 + }, + { + "epoch": 0.2590909090909091, + "grad_norm": 1.0733048915863037, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.7302, + "step": 38 + }, + { + "epoch": 0.26590909090909093, + "grad_norm": 0.8614453673362732, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.7466, + "step": 39 + }, + { + "epoch": 0.2727272727272727, + "grad_norm": 0.7729659080505371, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.725, + "step": 40 + }, + { + "epoch": 0.27954545454545454, + "grad_norm": 0.9341004490852356, + "learning_rate": 2.05e-06, + "loss": 0.6949, + "step": 41 + }, + { + "epoch": 0.2863636363636364, + "grad_norm": 0.6668925881385803, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.6587, + "step": 42 + }, + { + "epoch": 0.29318181818181815, + "grad_norm": 0.681847333908081, + "learning_rate": 2.15e-06, + "loss": 0.6984, + "step": 43 + }, + { + "epoch": 0.3, + "grad_norm": 0.6222082376480103, + "learning_rate": 2.2e-06, + "loss": 0.7063, + "step": 44 + }, + { + "epoch": 0.3068181818181818, + "grad_norm": 0.790607213973999, + "learning_rate": 2.25e-06, + "loss": 0.6804, + "step": 45 + }, + { + "epoch": 0.31363636363636366, + "grad_norm": 0.6110427379608154, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.6694, + "step": 46 + }, + { + "epoch": 0.32045454545454544, + "grad_norm": 0.6481993794441223, + "learning_rate": 2.35e-06, + "loss": 0.7321, + "step": 47 + }, + { + "epoch": 0.32727272727272727, + "grad_norm": 0.5973438620567322, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.6508, + "step": 48 + }, + { + "epoch": 0.3340909090909091, + "grad_norm": 0.6252740621566772, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.6911, + "step": 49 + }, + { + "epoch": 0.3409090909090909, + "grad_norm": 0.6826481819152832, + "learning_rate": 2.5e-06, + "loss": 0.6855, + "step": 50 + }, + { + "epoch": 0.3477272727272727, + "grad_norm": 0.6307975649833679, + "learning_rate": 2.55e-06, + "loss": 0.714, + "step": 51 + }, + { + "epoch": 0.35454545454545455, + "grad_norm": 0.5900976061820984, + "learning_rate": 2.6e-06, + "loss": 0.6619, + "step": 52 + }, + { + "epoch": 0.3613636363636364, + "grad_norm": 0.6203920841217041, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.6822, + "step": 53 + }, + { + "epoch": 0.36818181818181817, + "grad_norm": 0.6197589039802551, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.6584, + "step": 54 + }, + { + "epoch": 0.375, + "grad_norm": 0.5921110510826111, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.6739, + "step": 55 + }, + { + "epoch": 0.38181818181818183, + "grad_norm": 0.6215619444847107, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.6631, + "step": 56 + }, + { + "epoch": 0.3886363636363636, + "grad_norm": 0.5495648980140686, + "learning_rate": 2.85e-06, + "loss": 0.6531, + "step": 57 + }, + { + "epoch": 0.39545454545454545, + "grad_norm": 0.531947910785675, + "learning_rate": 2.9e-06, + "loss": 0.6365, + "step": 58 + }, + { + "epoch": 0.4022727272727273, + "grad_norm": 0.559112548828125, + "learning_rate": 2.95e-06, + "loss": 0.6391, + "step": 59 + }, + { + "epoch": 0.4090909090909091, + "grad_norm": 0.5377560257911682, + "learning_rate": 3e-06, + "loss": 0.6775, + "step": 60 + }, + { + "epoch": 0.4159090909090909, + "grad_norm": 0.5167352557182312, + "learning_rate": 3.05e-06, + "loss": 0.6518, + "step": 61 + }, + { + "epoch": 0.42272727272727273, + "grad_norm": 0.5483390092849731, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.5972, + "step": 62 + }, + { + "epoch": 0.42954545454545456, + "grad_norm": 0.5687050223350525, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.6515, + "step": 63 + }, + { + "epoch": 0.43636363636363634, + "grad_norm": 0.9046968817710876, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.6497, + "step": 64 + }, + { + "epoch": 0.4431818181818182, + "grad_norm": 0.5231667757034302, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.6364, + "step": 65 + }, + { + "epoch": 0.45, + "grad_norm": 0.49699342250823975, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.6144, + "step": 66 + }, + { + "epoch": 0.45681818181818185, + "grad_norm": 0.5390080213546753, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.6051, + "step": 67 + }, + { + "epoch": 0.4636363636363636, + "grad_norm": 0.5252938270568848, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.6353, + "step": 68 + }, + { + "epoch": 0.47045454545454546, + "grad_norm": 0.6291780471801758, + "learning_rate": 3.45e-06, + "loss": 0.6326, + "step": 69 + }, + { + "epoch": 0.4772727272727273, + "grad_norm": 0.5545375943183899, + "learning_rate": 3.5e-06, + "loss": 0.647, + "step": 70 + }, + { + "epoch": 0.48409090909090907, + "grad_norm": 0.47775956988334656, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.5993, + "step": 71 + }, + { + "epoch": 0.4909090909090909, + "grad_norm": 0.5016375184059143, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.6056, + "step": 72 + }, + { + "epoch": 0.49772727272727274, + "grad_norm": 0.4780517816543579, + "learning_rate": 3.65e-06, + "loss": 0.6189, + "step": 73 + }, + { + "epoch": 0.5045454545454545, + "grad_norm": 0.5763716697692871, + "learning_rate": 3.7e-06, + "loss": 0.6262, + "step": 74 + }, + { + "epoch": 0.5113636363636364, + "grad_norm": 0.5321404337882996, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.6336, + "step": 75 + }, + { + "epoch": 0.5181818181818182, + "grad_norm": 0.4641991853713989, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.6317, + "step": 76 + }, + { + "epoch": 0.525, + "grad_norm": 0.45118412375450134, + "learning_rate": 3.85e-06, + "loss": 0.5927, + "step": 77 + }, + { + "epoch": 0.5318181818181819, + "grad_norm": 0.5500645041465759, + "learning_rate": 3.900000000000001e-06, + "loss": 0.62, + "step": 78 + }, + { + "epoch": 0.5386363636363637, + "grad_norm": 0.5197616219520569, + "learning_rate": 3.95e-06, + "loss": 0.6063, + "step": 79 + }, + { + "epoch": 0.5454545454545454, + "grad_norm": 0.4672992527484894, + "learning_rate": 4.000000000000001e-06, + "loss": 0.6156, + "step": 80 + }, + { + "epoch": 0.5522727272727272, + "grad_norm": 0.5995073318481445, + "learning_rate": 4.05e-06, + "loss": 0.6219, + "step": 81 + }, + { + "epoch": 0.5590909090909091, + "grad_norm": 0.5593804717063904, + "learning_rate": 4.1e-06, + "loss": 0.6282, + "step": 82 + }, + { + "epoch": 0.5659090909090909, + "grad_norm": 0.7544977068901062, + "learning_rate": 4.15e-06, + "loss": 0.6246, + "step": 83 + }, + { + "epoch": 0.5727272727272728, + "grad_norm": 0.45378655195236206, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.5881, + "step": 84 + }, + { + "epoch": 0.5795454545454546, + "grad_norm": 0.4407907724380493, + "learning_rate": 4.25e-06, + "loss": 0.6026, + "step": 85 + }, + { + "epoch": 0.5863636363636363, + "grad_norm": 0.4333001673221588, + "learning_rate": 4.3e-06, + "loss": 0.6098, + "step": 86 + }, + { + "epoch": 0.5931818181818181, + "grad_norm": 0.45020997524261475, + "learning_rate": 4.350000000000001e-06, + "loss": 0.5858, + "step": 87 + }, + { + "epoch": 0.6, + "grad_norm": 0.424003541469574, + "learning_rate": 4.4e-06, + "loss": 0.6186, + "step": 88 + }, + { + "epoch": 0.6068181818181818, + "grad_norm": 0.427081435918808, + "learning_rate": 4.450000000000001e-06, + "loss": 0.6164, + "step": 89 + }, + { + "epoch": 0.6136363636363636, + "grad_norm": 0.4430546164512634, + "learning_rate": 4.5e-06, + "loss": 0.6049, + "step": 90 + }, + { + "epoch": 0.6204545454545455, + "grad_norm": 0.5469494462013245, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.608, + "step": 91 + }, + { + "epoch": 0.6272727272727273, + "grad_norm": 0.4481559693813324, + "learning_rate": 4.600000000000001e-06, + "loss": 0.5532, + "step": 92 + }, + { + "epoch": 0.634090909090909, + "grad_norm": 0.6563135981559753, + "learning_rate": 4.65e-06, + "loss": 0.6105, + "step": 93 + }, + { + "epoch": 0.6409090909090909, + "grad_norm": 0.4572807252407074, + "learning_rate": 4.7e-06, + "loss": 0.5976, + "step": 94 + }, + { + "epoch": 0.6477272727272727, + "grad_norm": 0.4423324167728424, + "learning_rate": 4.75e-06, + "loss": 0.616, + "step": 95 + }, + { + "epoch": 0.6545454545454545, + "grad_norm": 0.6230632066726685, + "learning_rate": 4.800000000000001e-06, + "loss": 0.5879, + "step": 96 + }, + { + "epoch": 0.6613636363636364, + "grad_norm": 0.461467981338501, + "learning_rate": 4.85e-06, + "loss": 0.5992, + "step": 97 + }, + { + "epoch": 0.6681818181818182, + "grad_norm": 0.44455233216285706, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.6138, + "step": 98 + }, + { + "epoch": 0.675, + "grad_norm": 0.47281715273857117, + "learning_rate": 4.95e-06, + "loss": 0.59, + "step": 99 + }, + { + "epoch": 0.6818181818181818, + "grad_norm": 0.46114516258239746, + "learning_rate": 5e-06, + "loss": 0.6216, + "step": 100 + }, + { + "epoch": 0.6886363636363636, + "grad_norm": 0.49900761246681213, + "learning_rate": 4.9999795126530275e-06, + "loss": 0.6222, + "step": 101 + }, + { + "epoch": 0.6954545454545454, + "grad_norm": 0.4764839708805084, + "learning_rate": 4.999918050947891e-06, + "loss": 0.5829, + "step": 102 + }, + { + "epoch": 0.7022727272727273, + "grad_norm": 0.43502742052078247, + "learning_rate": 4.999815615891943e-06, + "loss": 0.5982, + "step": 103 + }, + { + "epoch": 0.7090909090909091, + "grad_norm": 0.43318599462509155, + "learning_rate": 4.9996722091640805e-06, + "loss": 0.5827, + "step": 104 + }, + { + "epoch": 0.7159090909090909, + "grad_norm": 0.47218361496925354, + "learning_rate": 4.9994878331147225e-06, + "loss": 0.5907, + "step": 105 + }, + { + "epoch": 0.7227272727272728, + "grad_norm": 0.48492228984832764, + "learning_rate": 4.99926249076577e-06, + "loss": 0.5459, + "step": 106 + }, + { + "epoch": 0.7295454545454545, + "grad_norm": 0.4019850194454193, + "learning_rate": 4.998996185810557e-06, + "loss": 0.5957, + "step": 107 + }, + { + "epoch": 0.7363636363636363, + "grad_norm": 0.47361984848976135, + "learning_rate": 4.998688922613788e-06, + "loss": 0.579, + "step": 108 + }, + { + "epoch": 0.7431818181818182, + "grad_norm": 0.4612269401550293, + "learning_rate": 4.9983407062114695e-06, + "loss": 0.5814, + "step": 109 + }, + { + "epoch": 0.75, + "grad_norm": 0.45558398962020874, + "learning_rate": 4.9979515423108255e-06, + "loss": 0.6036, + "step": 110 + }, + { + "epoch": 0.7568181818181818, + "grad_norm": 0.41492199897766113, + "learning_rate": 4.997521437290205e-06, + "loss": 0.5891, + "step": 111 + }, + { + "epoch": 0.7636363636363637, + "grad_norm": 0.4484409689903259, + "learning_rate": 4.997050398198977e-06, + "loss": 0.6031, + "step": 112 + }, + { + "epoch": 0.7704545454545455, + "grad_norm": 0.42671382427215576, + "learning_rate": 4.996538432757414e-06, + "loss": 0.6106, + "step": 113 + }, + { + "epoch": 0.7772727272727272, + "grad_norm": 0.4222075343132019, + "learning_rate": 4.995985549356568e-06, + "loss": 0.5774, + "step": 114 + }, + { + "epoch": 0.7840909090909091, + "grad_norm": 0.421021431684494, + "learning_rate": 4.995391757058129e-06, + "loss": 0.5754, + "step": 115 + }, + { + "epoch": 0.7909090909090909, + "grad_norm": 0.4215952754020691, + "learning_rate": 4.99475706559428e-06, + "loss": 0.5895, + "step": 116 + }, + { + "epoch": 0.7977272727272727, + "grad_norm": 0.42652982473373413, + "learning_rate": 4.994081485367537e-06, + "loss": 0.5646, + "step": 117 + }, + { + "epoch": 0.8045454545454546, + "grad_norm": 0.4145742654800415, + "learning_rate": 4.993365027450576e-06, + "loss": 0.5862, + "step": 118 + }, + { + "epoch": 0.8113636363636364, + "grad_norm": 0.4754612445831299, + "learning_rate": 4.992607703586058e-06, + "loss": 0.5645, + "step": 119 + }, + { + "epoch": 0.8181818181818182, + "grad_norm": 0.4694959223270416, + "learning_rate": 4.991809526186424e-06, + "loss": 0.5989, + "step": 120 + }, + { + "epoch": 0.825, + "grad_norm": 0.4573237895965576, + "learning_rate": 4.990970508333707e-06, + "loss": 0.5769, + "step": 121 + }, + { + "epoch": 0.8318181818181818, + "grad_norm": 0.4592607915401459, + "learning_rate": 4.990090663779305e-06, + "loss": 0.5529, + "step": 122 + }, + { + "epoch": 0.8386363636363636, + "grad_norm": 0.4325920045375824, + "learning_rate": 4.9891700069437635e-06, + "loss": 0.5846, + "step": 123 + }, + { + "epoch": 0.8454545454545455, + "grad_norm": 0.44623467326164246, + "learning_rate": 4.988208552916535e-06, + "loss": 0.5997, + "step": 124 + }, + { + "epoch": 0.8522727272727273, + "grad_norm": 0.46071872115135193, + "learning_rate": 4.987206317455734e-06, + "loss": 0.5687, + "step": 125 + }, + { + "epoch": 0.8590909090909091, + "grad_norm": 0.43121206760406494, + "learning_rate": 4.986163316987877e-06, + "loss": 0.5277, + "step": 126 + }, + { + "epoch": 0.865909090909091, + "grad_norm": 0.4711630046367645, + "learning_rate": 4.985079568607613e-06, + "loss": 0.592, + "step": 127 + }, + { + "epoch": 0.8727272727272727, + "grad_norm": 0.4450221657752991, + "learning_rate": 4.983955090077445e-06, + "loss": 0.5656, + "step": 128 + }, + { + "epoch": 0.8795454545454545, + "grad_norm": 0.6546564102172852, + "learning_rate": 4.982789899827439e-06, + "loss": 0.572, + "step": 129 + }, + { + "epoch": 0.8863636363636364, + "grad_norm": 0.49622273445129395, + "learning_rate": 4.9815840169549216e-06, + "loss": 0.5719, + "step": 130 + }, + { + "epoch": 0.8931818181818182, + "grad_norm": 0.5478544235229492, + "learning_rate": 4.980337461224164e-06, + "loss": 0.5774, + "step": 131 + }, + { + "epoch": 0.9, + "grad_norm": 0.4340018033981323, + "learning_rate": 4.979050253066064e-06, + "loss": 0.5444, + "step": 132 + }, + { + "epoch": 0.9068181818181819, + "grad_norm": 0.4243793189525604, + "learning_rate": 4.977722413577802e-06, + "loss": 0.5759, + "step": 133 + }, + { + "epoch": 0.9136363636363637, + "grad_norm": 0.45201942324638367, + "learning_rate": 4.976353964522509e-06, + "loss": 0.5746, + "step": 134 + }, + { + "epoch": 0.9204545454545454, + "grad_norm": 0.42892220616340637, + "learning_rate": 4.974944928328894e-06, + "loss": 0.544, + "step": 135 + }, + { + "epoch": 0.9272727272727272, + "grad_norm": 0.43185824155807495, + "learning_rate": 4.973495328090891e-06, + "loss": 0.5803, + "step": 136 + }, + { + "epoch": 0.9340909090909091, + "grad_norm": 0.5512542724609375, + "learning_rate": 4.972005187567267e-06, + "loss": 0.5694, + "step": 137 + }, + { + "epoch": 0.9409090909090909, + "grad_norm": 0.4836059808731079, + "learning_rate": 4.970474531181245e-06, + "loss": 0.5581, + "step": 138 + }, + { + "epoch": 0.9477272727272728, + "grad_norm": 0.4508282244205475, + "learning_rate": 4.968903384020095e-06, + "loss": 0.5553, + "step": 139 + }, + { + "epoch": 0.9545454545454546, + "grad_norm": 0.4764733612537384, + "learning_rate": 4.967291771834727e-06, + "loss": 0.5798, + "step": 140 + }, + { + "epoch": 0.9613636363636363, + "grad_norm": 0.463468074798584, + "learning_rate": 4.965639721039267e-06, + "loss": 0.559, + "step": 141 + }, + { + "epoch": 0.9681818181818181, + "grad_norm": 0.4655650854110718, + "learning_rate": 4.963947258710626e-06, + "loss": 0.5576, + "step": 142 + }, + { + "epoch": 0.975, + "grad_norm": 0.4651234745979309, + "learning_rate": 4.962214412588053e-06, + "loss": 0.5792, + "step": 143 + }, + { + "epoch": 0.9818181818181818, + "grad_norm": 0.43080538511276245, + "learning_rate": 4.960441211072686e-06, + "loss": 0.5314, + "step": 144 + }, + { + "epoch": 0.9886363636363636, + "grad_norm": 0.4403238594532013, + "learning_rate": 4.9586276832270785e-06, + "loss": 0.5592, + "step": 145 + }, + { + "epoch": 0.9954545454545455, + "grad_norm": 0.43968647718429565, + "learning_rate": 4.9567738587747314e-06, + "loss": 0.5573, + "step": 146 + }, + { + "epoch": 1.0068181818181818, + "grad_norm": 0.8337184190750122, + "learning_rate": 4.954879768099599e-06, + "loss": 1.1118, + "step": 147 + }, + { + "epoch": 1.0136363636363637, + "grad_norm": 0.4042762815952301, + "learning_rate": 4.952945442245598e-06, + "loss": 0.5731, + "step": 148 + }, + { + "epoch": 1.0204545454545455, + "grad_norm": 0.5666297078132629, + "learning_rate": 4.95097091291609e-06, + "loss": 0.5415, + "step": 149 + }, + { + "epoch": 1.0272727272727273, + "grad_norm": 0.6148372888565063, + "learning_rate": 4.948956212473371e-06, + "loss": 0.5486, + "step": 150 + }, + { + "epoch": 1.0340909090909092, + "grad_norm": 0.5780720710754395, + "learning_rate": 4.946901373938132e-06, + "loss": 0.5373, + "step": 151 + }, + { + "epoch": 1.040909090909091, + "grad_norm": 0.4648835361003876, + "learning_rate": 4.944806430988927e-06, + "loss": 0.5419, + "step": 152 + }, + { + "epoch": 1.0477272727272728, + "grad_norm": 0.41131219267845154, + "learning_rate": 4.942671417961615e-06, + "loss": 0.5514, + "step": 153 + }, + { + "epoch": 1.0545454545454545, + "grad_norm": 0.4733349084854126, + "learning_rate": 4.940496369848795e-06, + "loss": 0.5563, + "step": 154 + }, + { + "epoch": 1.0613636363636363, + "grad_norm": 0.41179028153419495, + "learning_rate": 4.938281322299243e-06, + "loss": 0.5413, + "step": 155 + }, + { + "epoch": 1.0681818181818181, + "grad_norm": 0.4770941138267517, + "learning_rate": 4.936026311617316e-06, + "loss": 0.5561, + "step": 156 + }, + { + "epoch": 1.075, + "grad_norm": 0.7752397060394287, + "learning_rate": 4.933731374762361e-06, + "loss": 0.5439, + "step": 157 + }, + { + "epoch": 1.0818181818181818, + "grad_norm": 0.4323958456516266, + "learning_rate": 4.931396549348115e-06, + "loss": 0.5401, + "step": 158 + }, + { + "epoch": 1.0886363636363636, + "grad_norm": 0.4497937858104706, + "learning_rate": 4.9290218736420795e-06, + "loss": 0.523, + "step": 159 + }, + { + "epoch": 1.0954545454545455, + "grad_norm": 0.4881570339202881, + "learning_rate": 4.926607386564898e-06, + "loss": 0.5441, + "step": 160 + }, + { + "epoch": 1.1022727272727273, + "grad_norm": 0.43357932567596436, + "learning_rate": 4.9241531276897196e-06, + "loss": 0.5499, + "step": 161 + }, + { + "epoch": 1.1090909090909091, + "grad_norm": 1.1125361919403076, + "learning_rate": 4.921659137241544e-06, + "loss": 0.5265, + "step": 162 + }, + { + "epoch": 1.115909090909091, + "grad_norm": 0.43150508403778076, + "learning_rate": 4.919125456096574e-06, + "loss": 0.5193, + "step": 163 + }, + { + "epoch": 1.1227272727272728, + "grad_norm": 0.40771809220314026, + "learning_rate": 4.916552125781529e-06, + "loss": 0.538, + "step": 164 + }, + { + "epoch": 1.1295454545454546, + "grad_norm": 0.5830567479133606, + "learning_rate": 4.913939188472979e-06, + "loss": 0.5548, + "step": 165 + }, + { + "epoch": 1.1363636363636362, + "grad_norm": 0.4776526391506195, + "learning_rate": 4.911286686996648e-06, + "loss": 0.5328, + "step": 166 + }, + { + "epoch": 1.143181818181818, + "grad_norm": 0.41657742857933044, + "learning_rate": 4.908594664826708e-06, + "loss": 0.5336, + "step": 167 + }, + { + "epoch": 1.15, + "grad_norm": 0.4651948809623718, + "learning_rate": 4.905863166085076e-06, + "loss": 0.5812, + "step": 168 + }, + { + "epoch": 1.1568181818181817, + "grad_norm": 0.4399661719799042, + "learning_rate": 4.903092235540679e-06, + "loss": 0.5143, + "step": 169 + }, + { + "epoch": 1.1636363636363636, + "grad_norm": 0.4548935294151306, + "learning_rate": 4.900281918608732e-06, + "loss": 0.5425, + "step": 170 + }, + { + "epoch": 1.1704545454545454, + "grad_norm": 0.43515875935554504, + "learning_rate": 4.897432261349984e-06, + "loss": 0.534, + "step": 171 + }, + { + "epoch": 1.1772727272727272, + "grad_norm": 0.42954760789871216, + "learning_rate": 4.894543310469968e-06, + "loss": 0.5436, + "step": 172 + }, + { + "epoch": 1.184090909090909, + "grad_norm": 0.4063640534877777, + "learning_rate": 4.891615113318236e-06, + "loss": 0.5481, + "step": 173 + }, + { + "epoch": 1.190909090909091, + "grad_norm": 0.49416476488113403, + "learning_rate": 4.888647717887582e-06, + "loss": 0.5298, + "step": 174 + }, + { + "epoch": 1.1977272727272728, + "grad_norm": 1.4082964658737183, + "learning_rate": 4.8856411728132526e-06, + "loss": 0.5366, + "step": 175 + }, + { + "epoch": 1.2045454545454546, + "grad_norm": 0.44356465339660645, + "learning_rate": 4.8825955273721524e-06, + "loss": 0.5563, + "step": 176 + }, + { + "epoch": 1.2113636363636364, + "grad_norm": 0.40925371646881104, + "learning_rate": 4.879510831482039e-06, + "loss": 0.5444, + "step": 177 + }, + { + "epoch": 1.2181818181818183, + "grad_norm": 0.44379451870918274, + "learning_rate": 4.876387135700701e-06, + "loss": 0.5311, + "step": 178 + }, + { + "epoch": 1.225, + "grad_norm": 0.41487446427345276, + "learning_rate": 4.873224491225128e-06, + "loss": 0.5637, + "step": 179 + }, + { + "epoch": 1.231818181818182, + "grad_norm": 0.4520554542541504, + "learning_rate": 4.870022949890676e-06, + "loss": 0.5677, + "step": 180 + }, + { + "epoch": 1.2386363636363638, + "grad_norm": 0.4528166949748993, + "learning_rate": 4.866782564170217e-06, + "loss": 0.5095, + "step": 181 + }, + { + "epoch": 1.2454545454545454, + "grad_norm": 0.4409842789173126, + "learning_rate": 4.863503387173276e-06, + "loss": 0.5423, + "step": 182 + }, + { + "epoch": 1.2522727272727272, + "grad_norm": 0.4355996251106262, + "learning_rate": 4.860185472645161e-06, + "loss": 0.5422, + "step": 183 + }, + { + "epoch": 1.259090909090909, + "grad_norm": 0.4871126115322113, + "learning_rate": 4.856828874966086e-06, + "loss": 0.5293, + "step": 184 + }, + { + "epoch": 1.2659090909090909, + "grad_norm": 0.442348450422287, + "learning_rate": 4.853433649150276e-06, + "loss": 0.5616, + "step": 185 + }, + { + "epoch": 1.2727272727272727, + "grad_norm": 0.44014668464660645, + "learning_rate": 4.849999850845066e-06, + "loss": 0.5354, + "step": 186 + }, + { + "epoch": 1.2795454545454545, + "grad_norm": 0.4496249854564667, + "learning_rate": 4.8465275363299905e-06, + "loss": 0.5298, + "step": 187 + }, + { + "epoch": 1.2863636363636364, + "grad_norm": 0.46060171723365784, + "learning_rate": 4.84301676251586e-06, + "loss": 0.554, + "step": 188 + }, + { + "epoch": 1.2931818181818182, + "grad_norm": 0.7829347252845764, + "learning_rate": 4.839467586943825e-06, + "loss": 0.5606, + "step": 189 + }, + { + "epoch": 1.3, + "grad_norm": 0.45566731691360474, + "learning_rate": 4.835880067784441e-06, + "loss": 0.5118, + "step": 190 + }, + { + "epoch": 1.3068181818181819, + "grad_norm": 0.43211469054222107, + "learning_rate": 4.832254263836708e-06, + "loss": 0.535, + "step": 191 + }, + { + "epoch": 1.3136363636363637, + "grad_norm": 0.4350590705871582, + "learning_rate": 4.828590234527107e-06, + "loss": 0.5595, + "step": 192 + }, + { + "epoch": 1.3204545454545453, + "grad_norm": 0.4056399464607239, + "learning_rate": 4.82488803990863e-06, + "loss": 0.55, + "step": 193 + }, + { + "epoch": 1.3272727272727272, + "grad_norm": 0.4458341598510742, + "learning_rate": 4.821147740659795e-06, + "loss": 0.501, + "step": 194 + }, + { + "epoch": 1.334090909090909, + "grad_norm": 0.446504145860672, + "learning_rate": 4.817369398083648e-06, + "loss": 0.5393, + "step": 195 + }, + { + "epoch": 1.3409090909090908, + "grad_norm": 0.47838401794433594, + "learning_rate": 4.813553074106761e-06, + "loss": 0.5354, + "step": 196 + }, + { + "epoch": 1.3477272727272727, + "grad_norm": 0.4675792157649994, + "learning_rate": 4.809698831278217e-06, + "loss": 0.5202, + "step": 197 + }, + { + "epoch": 1.3545454545454545, + "grad_norm": 0.4351370632648468, + "learning_rate": 4.805806732768585e-06, + "loss": 0.5256, + "step": 198 + }, + { + "epoch": 1.3613636363636363, + "grad_norm": 0.4466511607170105, + "learning_rate": 4.801876842368882e-06, + "loss": 0.5238, + "step": 199 + }, + { + "epoch": 1.3681818181818182, + "grad_norm": 0.45672544836997986, + "learning_rate": 4.797909224489531e-06, + "loss": 0.5426, + "step": 200 + }, + { + "epoch": 1.375, + "grad_norm": 0.422638863325119, + "learning_rate": 4.793903944159303e-06, + "loss": 0.5385, + "step": 201 + }, + { + "epoch": 1.3818181818181818, + "grad_norm": 0.4254551827907562, + "learning_rate": 4.789861067024253e-06, + "loss": 0.503, + "step": 202 + }, + { + "epoch": 1.3886363636363637, + "grad_norm": 0.43860334157943726, + "learning_rate": 4.785780659346642e-06, + "loss": 0.519, + "step": 203 + }, + { + "epoch": 1.3954545454545455, + "grad_norm": 0.43294480443000793, + "learning_rate": 4.781662788003851e-06, + "loss": 0.5537, + "step": 204 + }, + { + "epoch": 1.4022727272727273, + "grad_norm": 0.45006492733955383, + "learning_rate": 4.777507520487289e-06, + "loss": 0.5318, + "step": 205 + }, + { + "epoch": 1.4090909090909092, + "grad_norm": 0.4178854823112488, + "learning_rate": 4.773314924901281e-06, + "loss": 0.5287, + "step": 206 + }, + { + "epoch": 1.415909090909091, + "grad_norm": 0.45662209391593933, + "learning_rate": 4.769085069961955e-06, + "loss": 0.5609, + "step": 207 + }, + { + "epoch": 1.4227272727272728, + "grad_norm": 0.41578733921051025, + "learning_rate": 4.764818024996117e-06, + "loss": 0.5249, + "step": 208 + }, + { + "epoch": 1.4295454545454547, + "grad_norm": 0.48305758833885193, + "learning_rate": 4.760513859940112e-06, + "loss": 0.515, + "step": 209 + }, + { + "epoch": 1.4363636363636363, + "grad_norm": 0.4564506709575653, + "learning_rate": 4.756172645338675e-06, + "loss": 0.535, + "step": 210 + }, + { + "epoch": 1.4431818181818181, + "grad_norm": 0.4499240815639496, + "learning_rate": 4.751794452343785e-06, + "loss": 0.5236, + "step": 211 + }, + { + "epoch": 1.45, + "grad_norm": 0.41851088404655457, + "learning_rate": 4.747379352713489e-06, + "loss": 0.5254, + "step": 212 + }, + { + "epoch": 1.4568181818181818, + "grad_norm": 0.4690937399864197, + "learning_rate": 4.7429274188107275e-06, + "loss": 0.5312, + "step": 213 + }, + { + "epoch": 1.4636363636363636, + "grad_norm": 0.5516958832740784, + "learning_rate": 4.738438723602154e-06, + "loss": 0.5315, + "step": 214 + }, + { + "epoch": 1.4704545454545455, + "grad_norm": 1.2040799856185913, + "learning_rate": 4.733913340656933e-06, + "loss": 0.544, + "step": 215 + }, + { + "epoch": 1.4772727272727273, + "grad_norm": 0.445425808429718, + "learning_rate": 4.729351344145536e-06, + "loss": 0.5309, + "step": 216 + }, + { + "epoch": 1.4840909090909091, + "grad_norm": 0.4186754822731018, + "learning_rate": 4.7247528088385296e-06, + "loss": 0.5138, + "step": 217 + }, + { + "epoch": 1.490909090909091, + "grad_norm": 0.43284520506858826, + "learning_rate": 4.720117810105341e-06, + "loss": 0.521, + "step": 218 + }, + { + "epoch": 1.4977272727272728, + "grad_norm": 0.4308505058288574, + "learning_rate": 4.715446423913036e-06, + "loss": 0.5436, + "step": 219 + }, + { + "epoch": 1.5045454545454544, + "grad_norm": 0.4471941888332367, + "learning_rate": 4.710738726825059e-06, + "loss": 0.5528, + "step": 220 + }, + { + "epoch": 1.5113636363636362, + "grad_norm": 0.4374965727329254, + "learning_rate": 4.705994795999991e-06, + "loss": 0.5238, + "step": 221 + }, + { + "epoch": 1.518181818181818, + "grad_norm": 0.6026428937911987, + "learning_rate": 4.701214709190277e-06, + "loss": 0.5268, + "step": 222 + }, + { + "epoch": 1.525, + "grad_norm": 0.4413348138332367, + "learning_rate": 4.696398544740955e-06, + "loss": 0.5534, + "step": 223 + }, + { + "epoch": 1.5318181818181817, + "grad_norm": 0.45917898416519165, + "learning_rate": 4.69154638158837e-06, + "loss": 0.511, + "step": 224 + }, + { + "epoch": 1.5386363636363636, + "grad_norm": 0.5339583158493042, + "learning_rate": 4.686658299258881e-06, + "loss": 0.5395, + "step": 225 + }, + { + "epoch": 1.5454545454545454, + "grad_norm": 0.4617757201194763, + "learning_rate": 4.681734377867562e-06, + "loss": 0.5425, + "step": 226 + }, + { + "epoch": 1.5522727272727272, + "grad_norm": 0.4513031840324402, + "learning_rate": 4.67677469811688e-06, + "loss": 0.5241, + "step": 227 + }, + { + "epoch": 1.559090909090909, + "grad_norm": 0.43548423051834106, + "learning_rate": 4.671779341295378e-06, + "loss": 0.5224, + "step": 228 + }, + { + "epoch": 1.565909090909091, + "grad_norm": 0.5516494512557983, + "learning_rate": 4.666748389276344e-06, + "loss": 0.5346, + "step": 229 + }, + { + "epoch": 1.5727272727272728, + "grad_norm": 0.4441164433956146, + "learning_rate": 4.661681924516466e-06, + "loss": 0.5208, + "step": 230 + }, + { + "epoch": 1.5795454545454546, + "grad_norm": 0.4057452380657196, + "learning_rate": 4.6565800300544805e-06, + "loss": 0.5173, + "step": 231 + }, + { + "epoch": 1.5863636363636364, + "grad_norm": 0.4162004590034485, + "learning_rate": 4.651442789509813e-06, + "loss": 0.5307, + "step": 232 + }, + { + "epoch": 1.5931818181818183, + "grad_norm": 0.43314146995544434, + "learning_rate": 4.646270287081208e-06, + "loss": 0.5359, + "step": 233 + }, + { + "epoch": 1.6, + "grad_norm": 0.5786082148551941, + "learning_rate": 4.641062607545347e-06, + "loss": 0.5307, + "step": 234 + }, + { + "epoch": 1.606818181818182, + "grad_norm": 0.43811941146850586, + "learning_rate": 4.6358198362554585e-06, + "loss": 0.5556, + "step": 235 + }, + { + "epoch": 1.6136363636363638, + "grad_norm": 0.4374074935913086, + "learning_rate": 4.630542059139923e-06, + "loss": 0.5253, + "step": 236 + }, + { + "epoch": 1.6204545454545456, + "grad_norm": 0.4387815594673157, + "learning_rate": 4.625229362700863e-06, + "loss": 0.534, + "step": 237 + }, + { + "epoch": 1.6272727272727274, + "grad_norm": 0.4976162016391754, + "learning_rate": 4.61988183401272e-06, + "loss": 0.5255, + "step": 238 + }, + { + "epoch": 1.634090909090909, + "grad_norm": 0.4603639543056488, + "learning_rate": 4.614499560720837e-06, + "loss": 0.511, + "step": 239 + }, + { + "epoch": 1.6409090909090909, + "grad_norm": 0.43136921525001526, + "learning_rate": 4.609082631040012e-06, + "loss": 0.5099, + "step": 240 + }, + { + "epoch": 1.6477272727272727, + "grad_norm": 0.4573090970516205, + "learning_rate": 4.603631133753061e-06, + "loss": 0.5515, + "step": 241 + }, + { + "epoch": 1.6545454545454545, + "grad_norm": 0.4292807877063751, + "learning_rate": 4.598145158209356e-06, + "loss": 0.5048, + "step": 242 + }, + { + "epoch": 1.6613636363636364, + "grad_norm": 0.6377942562103271, + "learning_rate": 4.592624794323366e-06, + "loss": 0.5333, + "step": 243 + }, + { + "epoch": 1.6681818181818182, + "grad_norm": 0.49924615025520325, + "learning_rate": 4.587070132573178e-06, + "loss": 0.5127, + "step": 244 + }, + { + "epoch": 1.675, + "grad_norm": 0.4723946154117584, + "learning_rate": 4.581481263999019e-06, + "loss": 0.5127, + "step": 245 + }, + { + "epoch": 1.6818181818181817, + "grad_norm": 1.1846898794174194, + "learning_rate": 4.575858280201761e-06, + "loss": 0.5367, + "step": 246 + }, + { + "epoch": 1.6886363636363635, + "grad_norm": 0.4306166172027588, + "learning_rate": 4.570201273341418e-06, + "loss": 0.52, + "step": 247 + }, + { + "epoch": 1.6954545454545453, + "grad_norm": 0.45308929681777954, + "learning_rate": 4.564510336135642e-06, + "loss": 0.5589, + "step": 248 + }, + { + "epoch": 1.7022727272727272, + "grad_norm": 0.4482448399066925, + "learning_rate": 4.558785561858196e-06, + "loss": 0.5161, + "step": 249 + }, + { + "epoch": 1.709090909090909, + "grad_norm": 0.44250935316085815, + "learning_rate": 4.5530270443374305e-06, + "loss": 0.5342, + "step": 250 + }, + { + "epoch": 1.7159090909090908, + "grad_norm": 0.8177540898323059, + "learning_rate": 4.547234877954741e-06, + "loss": 0.5402, + "step": 251 + }, + { + "epoch": 1.7227272727272727, + "grad_norm": 0.5169854760169983, + "learning_rate": 4.541409157643027e-06, + "loss": 0.5301, + "step": 252 + }, + { + "epoch": 1.7295454545454545, + "grad_norm": 0.4544285237789154, + "learning_rate": 4.535549978885132e-06, + "loss": 0.5314, + "step": 253 + }, + { + "epoch": 1.7363636363636363, + "grad_norm": 0.42860639095306396, + "learning_rate": 4.5296574377122765e-06, + "loss": 0.5249, + "step": 254 + }, + { + "epoch": 1.7431818181818182, + "grad_norm": 0.526492178440094, + "learning_rate": 4.5237316307024895e-06, + "loss": 0.5142, + "step": 255 + }, + { + "epoch": 1.75, + "grad_norm": 0.4208003878593445, + "learning_rate": 4.517772654979024e-06, + "loss": 0.5151, + "step": 256 + }, + { + "epoch": 1.7568181818181818, + "grad_norm": 0.4974212944507599, + "learning_rate": 4.51178060820876e-06, + "loss": 0.5161, + "step": 257 + }, + { + "epoch": 1.7636363636363637, + "grad_norm": 0.4148060083389282, + "learning_rate": 4.505755588600613e-06, + "loss": 0.5099, + "step": 258 + }, + { + "epoch": 1.7704545454545455, + "grad_norm": 0.4106168746948242, + "learning_rate": 4.499697694903915e-06, + "loss": 0.5376, + "step": 259 + }, + { + "epoch": 1.7772727272727273, + "grad_norm": 0.49017220735549927, + "learning_rate": 4.493607026406802e-06, + "loss": 0.525, + "step": 260 + }, + { + "epoch": 1.7840909090909092, + "grad_norm": 0.4141218662261963, + "learning_rate": 4.487483682934587e-06, + "loss": 0.5303, + "step": 261 + }, + { + "epoch": 1.790909090909091, + "grad_norm": 0.44826585054397583, + "learning_rate": 4.481327764848118e-06, + "loss": 0.4886, + "step": 262 + }, + { + "epoch": 1.7977272727272728, + "grad_norm": 0.42013809084892273, + "learning_rate": 4.47513937304214e-06, + "loss": 0.5204, + "step": 263 + }, + { + "epoch": 1.8045454545454547, + "grad_norm": 0.44327741861343384, + "learning_rate": 4.4689186089436365e-06, + "loss": 0.5097, + "step": 264 + }, + { + "epoch": 1.8113636363636365, + "grad_norm": 0.39498746395111084, + "learning_rate": 4.462665574510169e-06, + "loss": 0.5102, + "step": 265 + }, + { + "epoch": 1.8181818181818183, + "grad_norm": 0.48296087980270386, + "learning_rate": 4.456380372228208e-06, + "loss": 0.5299, + "step": 266 + }, + { + "epoch": 1.825, + "grad_norm": 0.4398634433746338, + "learning_rate": 4.450063105111447e-06, + "loss": 0.5419, + "step": 267 + }, + { + "epoch": 1.8318181818181818, + "grad_norm": 0.43333035707473755, + "learning_rate": 4.443713876699124e-06, + "loss": 0.5047, + "step": 268 + }, + { + "epoch": 1.8386363636363636, + "grad_norm": 0.5823659300804138, + "learning_rate": 4.4373327910543125e-06, + "loss": 0.5082, + "step": 269 + }, + { + "epoch": 1.8454545454545455, + "grad_norm": 0.44385817646980286, + "learning_rate": 4.430919952762226e-06, + "loss": 0.5171, + "step": 270 + }, + { + "epoch": 1.8522727272727273, + "grad_norm": 0.4444347023963928, + "learning_rate": 4.424475466928499e-06, + "loss": 0.5351, + "step": 271 + }, + { + "epoch": 1.8590909090909091, + "grad_norm": 0.549674391746521, + "learning_rate": 4.417999439177465e-06, + "loss": 0.498, + "step": 272 + }, + { + "epoch": 1.865909090909091, + "grad_norm": 0.4340274930000305, + "learning_rate": 4.4114919756504275e-06, + "loss": 0.5429, + "step": 273 + }, + { + "epoch": 1.8727272727272726, + "grad_norm": 0.44019854068756104, + "learning_rate": 4.404953183003916e-06, + "loss": 0.5133, + "step": 274 + }, + { + "epoch": 1.8795454545454544, + "grad_norm": 0.4473486542701721, + "learning_rate": 4.398383168407941e-06, + "loss": 0.5413, + "step": 275 + }, + { + "epoch": 1.8863636363636362, + "grad_norm": 0.44450780749320984, + "learning_rate": 4.391782039544239e-06, + "loss": 0.4998, + "step": 276 + }, + { + "epoch": 1.893181818181818, + "grad_norm": 0.4310123920440674, + "learning_rate": 4.385149904604502e-06, + "loss": 0.5134, + "step": 277 + }, + { + "epoch": 1.9, + "grad_norm": 0.48953917622566223, + "learning_rate": 4.378486872288611e-06, + "loss": 0.5102, + "step": 278 + }, + { + "epoch": 1.9068181818181817, + "grad_norm": 0.49219778180122375, + "learning_rate": 4.371793051802849e-06, + "loss": 0.5213, + "step": 279 + }, + { + "epoch": 1.9136363636363636, + "grad_norm": 0.818645715713501, + "learning_rate": 4.365068552858116e-06, + "loss": 0.5251, + "step": 280 + }, + { + "epoch": 1.9204545454545454, + "grad_norm": 0.4407028257846832, + "learning_rate": 4.358313485668124e-06, + "loss": 0.5191, + "step": 281 + }, + { + "epoch": 1.9272727272727272, + "grad_norm": 0.5301506519317627, + "learning_rate": 4.3515279609476e-06, + "loss": 0.5003, + "step": 282 + }, + { + "epoch": 1.934090909090909, + "grad_norm": 0.4673819839954376, + "learning_rate": 4.3447120899104615e-06, + "loss": 0.521, + "step": 283 + }, + { + "epoch": 1.940909090909091, + "grad_norm": 0.4106168746948242, + "learning_rate": 4.337865984268002e-06, + "loss": 0.5199, + "step": 284 + }, + { + "epoch": 1.9477272727272728, + "grad_norm": 0.43583595752716064, + "learning_rate": 4.3309897562270525e-06, + "loss": 0.5367, + "step": 285 + }, + { + "epoch": 1.9545454545454546, + "grad_norm": 0.4984910488128662, + "learning_rate": 4.324083518488151e-06, + "loss": 0.5202, + "step": 286 + }, + { + "epoch": 1.9613636363636364, + "grad_norm": 0.4516090452671051, + "learning_rate": 4.317147384243688e-06, + "loss": 0.5432, + "step": 287 + }, + { + "epoch": 1.9681818181818183, + "grad_norm": 0.43753015995025635, + "learning_rate": 4.3101814671760546e-06, + "loss": 0.5309, + "step": 288 + }, + { + "epoch": 1.975, + "grad_norm": 0.4472252130508423, + "learning_rate": 4.303185881455778e-06, + "loss": 0.5344, + "step": 289 + }, + { + "epoch": 1.981818181818182, + "grad_norm": 1.178377389907837, + "learning_rate": 4.296160741739652e-06, + "loss": 0.5283, + "step": 290 + }, + { + "epoch": 1.9886363636363638, + "grad_norm": 0.5085738301277161, + "learning_rate": 4.289106163168858e-06, + "loss": 0.5133, + "step": 291 + }, + { + "epoch": 1.9954545454545456, + "grad_norm": 0.52077716588974, + "learning_rate": 4.282022261367074e-06, + "loss": 0.5155, + "step": 292 + }, + { + "epoch": 2.006818181818182, + "grad_norm": 0.8534443378448486, + "learning_rate": 4.274909152438582e-06, + "loss": 1.003, + "step": 293 + }, + { + "epoch": 2.0136363636363637, + "grad_norm": 0.46160876750946045, + "learning_rate": 4.267766952966369e-06, + "loss": 0.4958, + "step": 294 + }, + { + "epoch": 2.0204545454545455, + "grad_norm": 0.5113462209701538, + "learning_rate": 4.260595780010209e-06, + "loss": 0.4895, + "step": 295 + }, + { + "epoch": 2.0272727272727273, + "grad_norm": 0.4553380608558655, + "learning_rate": 4.2533957511047485e-06, + "loss": 0.4826, + "step": 296 + }, + { + "epoch": 2.034090909090909, + "grad_norm": 0.43557822704315186, + "learning_rate": 4.24616698425758e-06, + "loss": 0.4591, + "step": 297 + }, + { + "epoch": 2.040909090909091, + "grad_norm": 0.45537200570106506, + "learning_rate": 4.238909597947307e-06, + "loss": 0.4811, + "step": 298 + }, + { + "epoch": 2.047727272727273, + "grad_norm": 0.48833370208740234, + "learning_rate": 4.231623711121603e-06, + "loss": 0.4967, + "step": 299 + }, + { + "epoch": 2.0545454545454547, + "grad_norm": 0.45504844188690186, + "learning_rate": 4.224309443195261e-06, + "loss": 0.4964, + "step": 300 + }, + { + "epoch": 2.0613636363636365, + "grad_norm": 0.4313275218009949, + "learning_rate": 4.2169669140482365e-06, + "loss": 0.4832, + "step": 301 + }, + { + "epoch": 2.0681818181818183, + "grad_norm": 0.4298369884490967, + "learning_rate": 4.2095962440236846e-06, + "loss": 0.5071, + "step": 302 + }, + { + "epoch": 2.075, + "grad_norm": 0.4188523590564728, + "learning_rate": 4.202197553925983e-06, + "loss": 0.4859, + "step": 303 + }, + { + "epoch": 2.081818181818182, + "grad_norm": 0.39195290207862854, + "learning_rate": 4.194770965018758e-06, + "loss": 0.5081, + "step": 304 + }, + { + "epoch": 2.088636363636364, + "grad_norm": 0.45844170451164246, + "learning_rate": 4.187316599022892e-06, + "loss": 0.508, + "step": 305 + }, + { + "epoch": 2.0954545454545457, + "grad_norm": 0.6743896007537842, + "learning_rate": 4.179834578114531e-06, + "loss": 0.4748, + "step": 306 + }, + { + "epoch": 2.102272727272727, + "grad_norm": 0.4317929446697235, + "learning_rate": 4.172325024923083e-06, + "loss": 0.504, + "step": 307 + }, + { + "epoch": 2.109090909090909, + "grad_norm": 0.4857175350189209, + "learning_rate": 4.164788062529203e-06, + "loss": 0.4987, + "step": 308 + }, + { + "epoch": 2.1159090909090907, + "grad_norm": 0.4454379677772522, + "learning_rate": 4.157223814462784e-06, + "loss": 0.4938, + "step": 309 + }, + { + "epoch": 2.1227272727272726, + "grad_norm": 0.5128078460693359, + "learning_rate": 4.149632404700925e-06, + "loss": 0.4917, + "step": 310 + }, + { + "epoch": 2.1295454545454544, + "grad_norm": 0.5195722579956055, + "learning_rate": 4.142013957665903e-06, + "loss": 0.4753, + "step": 311 + }, + { + "epoch": 2.1363636363636362, + "grad_norm": 0.4242931604385376, + "learning_rate": 4.134368598223132e-06, + "loss": 0.4832, + "step": 312 + }, + { + "epoch": 2.143181818181818, + "grad_norm": 0.4419071674346924, + "learning_rate": 4.126696451679116e-06, + "loss": 0.4783, + "step": 313 + }, + { + "epoch": 2.15, + "grad_norm": 0.42852380871772766, + "learning_rate": 4.118997643779401e-06, + "loss": 0.5034, + "step": 314 + }, + { + "epoch": 2.1568181818181817, + "grad_norm": 0.7373576164245605, + "learning_rate": 4.111272300706502e-06, + "loss": 0.4934, + "step": 315 + }, + { + "epoch": 2.1636363636363636, + "grad_norm": 0.4435337781906128, + "learning_rate": 4.1035205490778505e-06, + "loss": 0.5089, + "step": 316 + }, + { + "epoch": 2.1704545454545454, + "grad_norm": 0.4714502692222595, + "learning_rate": 4.095742515943703e-06, + "loss": 0.502, + "step": 317 + }, + { + "epoch": 2.1772727272727272, + "grad_norm": 0.4732494056224823, + "learning_rate": 4.087938328785071e-06, + "loss": 0.5073, + "step": 318 + }, + { + "epoch": 2.184090909090909, + "grad_norm": 0.45177504420280457, + "learning_rate": 4.080108115511629e-06, + "loss": 0.5076, + "step": 319 + }, + { + "epoch": 2.190909090909091, + "grad_norm": 0.506157398223877, + "learning_rate": 4.072252004459612e-06, + "loss": 0.4837, + "step": 320 + }, + { + "epoch": 2.1977272727272728, + "grad_norm": 0.43044376373291016, + "learning_rate": 4.064370124389718e-06, + "loss": 0.5167, + "step": 321 + }, + { + "epoch": 2.2045454545454546, + "grad_norm": 0.45878180861473083, + "learning_rate": 4.056462604484998e-06, + "loss": 0.4936, + "step": 322 + }, + { + "epoch": 2.2113636363636364, + "grad_norm": 0.43078047037124634, + "learning_rate": 4.048529574348734e-06, + "loss": 0.4783, + "step": 323 + }, + { + "epoch": 2.2181818181818183, + "grad_norm": 0.42789366841316223, + "learning_rate": 4.040571164002319e-06, + "loss": 0.4853, + "step": 324 + }, + { + "epoch": 2.225, + "grad_norm": 0.44888222217559814, + "learning_rate": 4.032587503883124e-06, + "loss": 0.5097, + "step": 325 + }, + { + "epoch": 2.231818181818182, + "grad_norm": 0.446851521730423, + "learning_rate": 4.0245787248423614e-06, + "loss": 0.4844, + "step": 326 + }, + { + "epoch": 2.2386363636363638, + "grad_norm": 0.436567097902298, + "learning_rate": 4.0165449581429404e-06, + "loss": 0.4549, + "step": 327 + }, + { + "epoch": 2.2454545454545456, + "grad_norm": 0.4813944697380066, + "learning_rate": 4.008486335457312e-06, + "loss": 0.49, + "step": 328 + }, + { + "epoch": 2.2522727272727274, + "grad_norm": 0.4935426115989685, + "learning_rate": 4.000402988865316e-06, + "loss": 0.4995, + "step": 329 + }, + { + "epoch": 2.2590909090909093, + "grad_norm": 1.1695573329925537, + "learning_rate": 3.992295050852013e-06, + "loss": 0.5033, + "step": 330 + }, + { + "epoch": 2.265909090909091, + "grad_norm": 0.41830316185951233, + "learning_rate": 3.984162654305516e-06, + "loss": 0.4828, + "step": 331 + }, + { + "epoch": 2.2727272727272725, + "grad_norm": 0.4522007703781128, + "learning_rate": 3.976005932514807e-06, + "loss": 0.4948, + "step": 332 + }, + { + "epoch": 2.2795454545454543, + "grad_norm": 0.8304370045661926, + "learning_rate": 3.967825019167559e-06, + "loss": 0.504, + "step": 333 + }, + { + "epoch": 2.286363636363636, + "grad_norm": 0.4533236622810364, + "learning_rate": 3.959620048347938e-06, + "loss": 0.4487, + "step": 334 + }, + { + "epoch": 2.293181818181818, + "grad_norm": 0.43578940629959106, + "learning_rate": 3.951391154534415e-06, + "loss": 0.4702, + "step": 335 + }, + { + "epoch": 2.3, + "grad_norm": 0.5146592259407043, + "learning_rate": 3.943138472597549e-06, + "loss": 0.4984, + "step": 336 + }, + { + "epoch": 2.3068181818181817, + "grad_norm": 0.6412249207496643, + "learning_rate": 3.934862137797788e-06, + "loss": 0.4761, + "step": 337 + }, + { + "epoch": 2.3136363636363635, + "grad_norm": 1.3352930545806885, + "learning_rate": 3.9265622857832455e-06, + "loss": 0.499, + "step": 338 + }, + { + "epoch": 2.3204545454545453, + "grad_norm": 0.4465159475803375, + "learning_rate": 3.918239052587481e-06, + "loss": 0.4525, + "step": 339 + }, + { + "epoch": 2.327272727272727, + "grad_norm": 0.43597185611724854, + "learning_rate": 3.909892574627267e-06, + "loss": 0.4952, + "step": 340 + }, + { + "epoch": 2.334090909090909, + "grad_norm": 0.7057136297225952, + "learning_rate": 3.901522988700355e-06, + "loss": 0.4576, + "step": 341 + }, + { + "epoch": 2.340909090909091, + "grad_norm": 0.45416319370269775, + "learning_rate": 3.893130431983234e-06, + "loss": 0.4902, + "step": 342 + }, + { + "epoch": 2.3477272727272727, + "grad_norm": 0.43900424242019653, + "learning_rate": 3.884715042028882e-06, + "loss": 0.495, + "step": 343 + }, + { + "epoch": 2.3545454545454545, + "grad_norm": 0.4553733468055725, + "learning_rate": 3.876276956764509e-06, + "loss": 0.4861, + "step": 344 + }, + { + "epoch": 2.3613636363636363, + "grad_norm": 0.5471370816230774, + "learning_rate": 3.867816314489301e-06, + "loss": 0.5043, + "step": 345 + }, + { + "epoch": 2.368181818181818, + "grad_norm": 0.43938082456588745, + "learning_rate": 3.8593332538721465e-06, + "loss": 0.4678, + "step": 346 + }, + { + "epoch": 2.375, + "grad_norm": 0.4583636224269867, + "learning_rate": 3.8508279139493736e-06, + "loss": 0.4679, + "step": 347 + }, + { + "epoch": 2.381818181818182, + "grad_norm": 0.44170665740966797, + "learning_rate": 3.84230043412246e-06, + "loss": 0.4647, + "step": 348 + }, + { + "epoch": 2.3886363636363637, + "grad_norm": 0.4952428340911865, + "learning_rate": 3.833750954155757e-06, + "loss": 0.4796, + "step": 349 + }, + { + "epoch": 2.3954545454545455, + "grad_norm": 0.42052993178367615, + "learning_rate": 3.825179614174195e-06, + "loss": 0.4606, + "step": 350 + }, + { + "epoch": 2.4022727272727273, + "grad_norm": 0.5792545676231384, + "learning_rate": 3.816586554660987e-06, + "loss": 0.4805, + "step": 351 + }, + { + "epoch": 2.409090909090909, + "grad_norm": 0.4477898180484772, + "learning_rate": 3.807971916455325e-06, + "loss": 0.4793, + "step": 352 + }, + { + "epoch": 2.415909090909091, + "grad_norm": 0.45617541670799255, + "learning_rate": 3.799335840750077e-06, + "loss": 0.502, + "step": 353 + }, + { + "epoch": 2.422727272727273, + "grad_norm": 0.45021289587020874, + "learning_rate": 3.790678469089465e-06, + "loss": 0.478, + "step": 354 + }, + { + "epoch": 2.4295454545454547, + "grad_norm": 0.4506435990333557, + "learning_rate": 3.7819999433667503e-06, + "loss": 0.4828, + "step": 355 + }, + { + "epoch": 2.4363636363636365, + "grad_norm": 0.48021042346954346, + "learning_rate": 3.773300405821908e-06, + "loss": 0.4827, + "step": 356 + }, + { + "epoch": 2.4431818181818183, + "grad_norm": 0.4759495258331299, + "learning_rate": 3.764579999039293e-06, + "loss": 0.4633, + "step": 357 + }, + { + "epoch": 2.45, + "grad_norm": 0.4734678268432617, + "learning_rate": 3.7558388659453052e-06, + "loss": 0.4982, + "step": 358 + }, + { + "epoch": 2.456818181818182, + "grad_norm": 0.4502730369567871, + "learning_rate": 3.7470771498060455e-06, + "loss": 0.5009, + "step": 359 + }, + { + "epoch": 2.463636363636364, + "grad_norm": 0.444940447807312, + "learning_rate": 3.7382949942249695e-06, + "loss": 0.4994, + "step": 360 + }, + { + "epoch": 2.4704545454545457, + "grad_norm": 0.4882519245147705, + "learning_rate": 3.7294925431405306e-06, + "loss": 0.4607, + "step": 361 + }, + { + "epoch": 2.4772727272727275, + "grad_norm": 0.51338791847229, + "learning_rate": 3.720669940823827e-06, + "loss": 0.4801, + "step": 362 + }, + { + "epoch": 2.484090909090909, + "grad_norm": 0.495211124420166, + "learning_rate": 3.7118273318762275e-06, + "loss": 0.4984, + "step": 363 + }, + { + "epoch": 2.4909090909090907, + "grad_norm": 0.4894484281539917, + "learning_rate": 3.702964861227013e-06, + "loss": 0.4834, + "step": 364 + }, + { + "epoch": 2.4977272727272726, + "grad_norm": 0.5554991364479065, + "learning_rate": 3.694082674130991e-06, + "loss": 0.4874, + "step": 365 + }, + { + "epoch": 2.5045454545454544, + "grad_norm": 0.4586203396320343, + "learning_rate": 3.6851809161661206e-06, + "loss": 0.5117, + "step": 366 + }, + { + "epoch": 2.5113636363636362, + "grad_norm": 0.7342172861099243, + "learning_rate": 3.6762597332311254e-06, + "loss": 0.5184, + "step": 367 + }, + { + "epoch": 2.518181818181818, + "grad_norm": 0.46638190746307373, + "learning_rate": 3.6673192715431016e-06, + "loss": 0.47, + "step": 368 + }, + { + "epoch": 2.525, + "grad_norm": 0.4584850072860718, + "learning_rate": 3.658359677635122e-06, + "loss": 0.45, + "step": 369 + }, + { + "epoch": 2.5318181818181817, + "grad_norm": 0.4285255968570709, + "learning_rate": 3.649381098353834e-06, + "loss": 0.513, + "step": 370 + }, + { + "epoch": 2.5386363636363636, + "grad_norm": 0.49487683176994324, + "learning_rate": 3.6403836808570512e-06, + "loss": 0.5048, + "step": 371 + }, + { + "epoch": 2.5454545454545454, + "grad_norm": 0.44076815247535706, + "learning_rate": 3.631367572611348e-06, + "loss": 0.4927, + "step": 372 + }, + { + "epoch": 2.5522727272727272, + "grad_norm": 0.44529926776885986, + "learning_rate": 3.6223329213896313e-06, + "loss": 0.4863, + "step": 373 + }, + { + "epoch": 2.559090909090909, + "grad_norm": 0.5416815876960754, + "learning_rate": 3.613279875268731e-06, + "loss": 0.4808, + "step": 374 + }, + { + "epoch": 2.565909090909091, + "grad_norm": 0.43630415201187134, + "learning_rate": 3.604208582626964e-06, + "loss": 0.4792, + "step": 375 + }, + { + "epoch": 2.5727272727272728, + "grad_norm": 0.47734326124191284, + "learning_rate": 3.5951191921417063e-06, + "loss": 0.4462, + "step": 376 + }, + { + "epoch": 2.5795454545454546, + "grad_norm": 0.4540305435657501, + "learning_rate": 3.586011852786955e-06, + "loss": 0.4915, + "step": 377 + }, + { + "epoch": 2.5863636363636364, + "grad_norm": 0.47168779373168945, + "learning_rate": 3.5768867138308872e-06, + "loss": 0.47, + "step": 378 + }, + { + "epoch": 2.5931818181818183, + "grad_norm": 0.46372920274734497, + "learning_rate": 3.5677439248334133e-06, + "loss": 0.4775, + "step": 379 + }, + { + "epoch": 2.6, + "grad_norm": 0.41653329133987427, + "learning_rate": 3.5585836356437266e-06, + "loss": 0.4875, + "step": 380 + }, + { + "epoch": 2.606818181818182, + "grad_norm": 0.4495103657245636, + "learning_rate": 3.5494059963978433e-06, + "loss": 0.4637, + "step": 381 + }, + { + "epoch": 2.6136363636363638, + "grad_norm": 0.432577520608902, + "learning_rate": 3.540211157516149e-06, + "loss": 0.4991, + "step": 382 + }, + { + "epoch": 2.6204545454545456, + "grad_norm": 0.4517219662666321, + "learning_rate": 3.530999269700927e-06, + "loss": 0.4785, + "step": 383 + }, + { + "epoch": 2.6272727272727274, + "grad_norm": 0.519922137260437, + "learning_rate": 3.521770483933891e-06, + "loss": 0.5026, + "step": 384 + }, + { + "epoch": 2.634090909090909, + "grad_norm": 0.48750364780426025, + "learning_rate": 3.5125249514737093e-06, + "loss": 0.4805, + "step": 385 + }, + { + "epoch": 2.6409090909090907, + "grad_norm": 0.6416886448860168, + "learning_rate": 3.503262823853527e-06, + "loss": 0.4849, + "step": 386 + }, + { + "epoch": 2.6477272727272725, + "grad_norm": 0.5748171806335449, + "learning_rate": 3.493984252878483e-06, + "loss": 0.4885, + "step": 387 + }, + { + "epoch": 2.6545454545454543, + "grad_norm": 0.5148530006408691, + "learning_rate": 3.484689390623218e-06, + "loss": 0.484, + "step": 388 + }, + { + "epoch": 2.661363636363636, + "grad_norm": 0.526839017868042, + "learning_rate": 3.4753783894293886e-06, + "loss": 0.4758, + "step": 389 + }, + { + "epoch": 2.668181818181818, + "grad_norm": 0.7565727829933167, + "learning_rate": 3.466051401903162e-06, + "loss": 0.479, + "step": 390 + }, + { + "epoch": 2.675, + "grad_norm": 0.4598790109157562, + "learning_rate": 3.4567085809127247e-06, + "loss": 0.4972, + "step": 391 + }, + { + "epoch": 2.6818181818181817, + "grad_norm": 0.4911792278289795, + "learning_rate": 3.4473500795857674e-06, + "loss": 0.4714, + "step": 392 + }, + { + "epoch": 2.6886363636363635, + "grad_norm": 0.4214136600494385, + "learning_rate": 3.4379760513069804e-06, + "loss": 0.4906, + "step": 393 + }, + { + "epoch": 2.6954545454545453, + "grad_norm": 0.5070644021034241, + "learning_rate": 3.428586649715542e-06, + "loss": 0.4936, + "step": 394 + }, + { + "epoch": 2.702272727272727, + "grad_norm": 0.4591734707355499, + "learning_rate": 3.4191820287025916e-06, + "loss": 0.4751, + "step": 395 + }, + { + "epoch": 2.709090909090909, + "grad_norm": 0.579781711101532, + "learning_rate": 3.4097623424087196e-06, + "loss": 0.4935, + "step": 396 + }, + { + "epoch": 2.715909090909091, + "grad_norm": 0.4178840219974518, + "learning_rate": 3.4003277452214284e-06, + "loss": 0.4771, + "step": 397 + }, + { + "epoch": 2.7227272727272727, + "grad_norm": 0.5002609491348267, + "learning_rate": 3.3908783917726123e-06, + "loss": 0.4739, + "step": 398 + }, + { + "epoch": 2.7295454545454545, + "grad_norm": 0.5176806449890137, + "learning_rate": 3.381414436936018e-06, + "loss": 0.4965, + "step": 399 + }, + { + "epoch": 2.7363636363636363, + "grad_norm": 0.8039212226867676, + "learning_rate": 3.3719360358247054e-06, + "loss": 0.4822, + "step": 400 + }, + { + "epoch": 2.743181818181818, + "grad_norm": 0.48943471908569336, + "learning_rate": 3.36244334378851e-06, + "loss": 0.4815, + "step": 401 + }, + { + "epoch": 2.75, + "grad_norm": 0.43710342049598694, + "learning_rate": 3.3529365164114903e-06, + "loss": 0.4932, + "step": 402 + }, + { + "epoch": 2.756818181818182, + "grad_norm": 0.5585265159606934, + "learning_rate": 3.3434157095093846e-06, + "loss": 0.5125, + "step": 403 + }, + { + "epoch": 2.7636363636363637, + "grad_norm": 0.4626101553440094, + "learning_rate": 3.333881079127052e-06, + "loss": 0.4934, + "step": 404 + }, + { + "epoch": 2.7704545454545455, + "grad_norm": 0.4675556421279907, + "learning_rate": 3.3243327815359168e-06, + "loss": 0.4903, + "step": 405 + }, + { + "epoch": 2.7772727272727273, + "grad_norm": 0.45658671855926514, + "learning_rate": 3.314770973231408e-06, + "loss": 0.4778, + "step": 406 + }, + { + "epoch": 2.784090909090909, + "grad_norm": 0.4911910593509674, + "learning_rate": 3.305195810930393e-06, + "loss": 0.4795, + "step": 407 + }, + { + "epoch": 2.790909090909091, + "grad_norm": 0.4397219121456146, + "learning_rate": 3.2956074515686105e-06, + "loss": 0.4589, + "step": 408 + }, + { + "epoch": 2.797727272727273, + "grad_norm": 0.5184643864631653, + "learning_rate": 3.2860060522980945e-06, + "loss": 0.4827, + "step": 409 + }, + { + "epoch": 2.8045454545454547, + "grad_norm": 0.45334312319755554, + "learning_rate": 3.276391770484606e-06, + "loss": 0.4821, + "step": 410 + }, + { + "epoch": 2.8113636363636365, + "grad_norm": 0.41846391558647156, + "learning_rate": 3.266764763705046e-06, + "loss": 0.5157, + "step": 411 + }, + { + "epoch": 2.8181818181818183, + "grad_norm": 0.45868799090385437, + "learning_rate": 3.257125189744877e-06, + "loss": 0.4876, + "step": 412 + }, + { + "epoch": 2.825, + "grad_norm": 0.4785497486591339, + "learning_rate": 3.247473206595536e-06, + "loss": 0.4809, + "step": 413 + }, + { + "epoch": 2.831818181818182, + "grad_norm": 0.5683679580688477, + "learning_rate": 3.2378089724518464e-06, + "loss": 0.505, + "step": 414 + }, + { + "epoch": 2.838636363636364, + "grad_norm": 0.48336583375930786, + "learning_rate": 3.228132645709421e-06, + "loss": 0.4816, + "step": 415 + }, + { + "epoch": 2.8454545454545457, + "grad_norm": 0.4236341416835785, + "learning_rate": 3.218444384962071e-06, + "loss": 0.4885, + "step": 416 + }, + { + "epoch": 2.8522727272727275, + "grad_norm": 0.4892752766609192, + "learning_rate": 3.2087443489992043e-06, + "loss": 0.4782, + "step": 417 + }, + { + "epoch": 2.8590909090909093, + "grad_norm": 0.4632202088832855, + "learning_rate": 3.1990326968032225e-06, + "loss": 0.5211, + "step": 418 + }, + { + "epoch": 2.865909090909091, + "grad_norm": 0.8842809796333313, + "learning_rate": 3.189309587546917e-06, + "loss": 0.4716, + "step": 419 + }, + { + "epoch": 2.8727272727272726, + "grad_norm": 0.46771398186683655, + "learning_rate": 3.1795751805908578e-06, + "loss": 0.4745, + "step": 420 + }, + { + "epoch": 2.8795454545454544, + "grad_norm": 0.8178832530975342, + "learning_rate": 3.169829635480783e-06, + "loss": 0.4917, + "step": 421 + }, + { + "epoch": 2.8863636363636362, + "grad_norm": 0.43057572841644287, + "learning_rate": 3.160073111944983e-06, + "loss": 0.4926, + "step": 422 + }, + { + "epoch": 2.893181818181818, + "grad_norm": 0.4258928894996643, + "learning_rate": 3.150305769891686e-06, + "loss": 0.4698, + "step": 423 + }, + { + "epoch": 2.9, + "grad_norm": 0.4559054374694824, + "learning_rate": 3.1405277694064306e-06, + "loss": 0.4537, + "step": 424 + }, + { + "epoch": 2.9068181818181817, + "grad_norm": 0.4342961609363556, + "learning_rate": 3.13073927074945e-06, + "loss": 0.4834, + "step": 425 + }, + { + "epoch": 2.9136363636363636, + "grad_norm": 0.48813825845718384, + "learning_rate": 3.1209404343530374e-06, + "loss": 0.4853, + "step": 426 + }, + { + "epoch": 2.9204545454545454, + "grad_norm": 0.47864893078804016, + "learning_rate": 3.111131420818922e-06, + "loss": 0.4627, + "step": 427 + }, + { + "epoch": 2.9272727272727272, + "grad_norm": 0.4394248425960541, + "learning_rate": 3.1013123909156347e-06, + "loss": 0.4817, + "step": 428 + }, + { + "epoch": 2.934090909090909, + "grad_norm": 0.6629825830459595, + "learning_rate": 3.091483505575873e-06, + "loss": 0.4738, + "step": 429 + }, + { + "epoch": 2.940909090909091, + "grad_norm": 0.47470518946647644, + "learning_rate": 3.081644925893866e-06, + "loss": 0.4943, + "step": 430 + }, + { + "epoch": 2.9477272727272728, + "grad_norm": 0.44007545709609985, + "learning_rate": 3.0717968131227285e-06, + "loss": 0.4805, + "step": 431 + }, + { + "epoch": 2.9545454545454546, + "grad_norm": 0.4425961375236511, + "learning_rate": 3.061939328671824e-06, + "loss": 0.5092, + "step": 432 + }, + { + "epoch": 2.9613636363636364, + "grad_norm": 0.4215753972530365, + "learning_rate": 3.0520726341041165e-06, + "loss": 0.4616, + "step": 433 + }, + { + "epoch": 2.9681818181818183, + "grad_norm": 0.4265761077404022, + "learning_rate": 3.0421968911335196e-06, + "loss": 0.4764, + "step": 434 + }, + { + "epoch": 2.975, + "grad_norm": 0.45009252429008484, + "learning_rate": 3.032312261622255e-06, + "loss": 0.4865, + "step": 435 + }, + { + "epoch": 2.981818181818182, + "grad_norm": 0.5175682306289673, + "learning_rate": 3.0224189075781886e-06, + "loss": 0.5085, + "step": 436 + }, + { + "epoch": 2.9886363636363638, + "grad_norm": 0.4518980383872986, + "learning_rate": 3.012516991152181e-06, + "loss": 0.4568, + "step": 437 + }, + { + "epoch": 2.9954545454545456, + "grad_norm": 0.4789412021636963, + "learning_rate": 3.002606674635432e-06, + "loss": 0.4942, + "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..a03cebdd4a04fb2e9443a05c92f15b2b2db01e10 --- /dev/null +++ b/checkpoint-438/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90038f7ff377ce29f00aedab533b18bfc9515b30d2b4232d8b0d6de26ec3ea56 +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..fdf619c317c2fe82074662582dbd68166b6f9d50 --- /dev/null +++ b/checkpoint-584/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3.1-Nemotron-70B-Instruct-HF +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..604978b603be594e46459a6c8c10d206a5a03b31 --- /dev/null +++ b/checkpoint-584/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF", + "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": [ + "down_proj", + "v_proj", + "q_proj", + "gate_proj", + "o_proj", + "up_proj", + "k_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..8654b3cc5cd60ed17813c4a47a1fb7fcaa9aac73 --- /dev/null +++ b/checkpoint-584/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:30908138aad12049824ad0f7a37d791e0c443c76e99840f7d154f861cd3679f2 +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..1d3cb71456a572b85e9108dd11589acc68af7bd5 --- /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 sha256:5e72b29195270ca1a37443ae0d7c739981e788d2e2e1e43e1135372689b0018e +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..1bd02e37c8150a97f8dfa7e7e263e875f0eed22f --- /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 sha256:7be16c9289256225c9738919147d257990147344af1d60a30a5d1d4c0af830e8 +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..4cc2c465192c08869f91f9239a40bd91a8e2edeb --- /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:ad1982bc16e8b59519be7abb400470f88ed98dc2b58e1f93de1cc2d0c64799db +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..18ac249f2488869748f2368a4ef67bcaead94d26 --- /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:ff72a1b8b5b49d532fe3512cc313efb0447eefe63f0f4d0279bd2c8adf235af2 +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..fdde94c29816839ec3c29d6c6461206a49911f3c --- /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\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..7ff48a9d94d13a6c399e9d656ce4cfe4b1f72534 --- /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": 2.020231246948242, + "learning_rate": 5.0000000000000004e-08, + "loss": 0.8736, + "step": 1 + }, + { + "epoch": 0.013636363636363636, + "grad_norm": 1.9368854761123657, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.8543, + "step": 2 + }, + { + "epoch": 0.020454545454545454, + "grad_norm": 1.908379077911377, + "learning_rate": 1.5000000000000002e-07, + "loss": 0.8339, + "step": 3 + }, + { + "epoch": 0.02727272727272727, + "grad_norm": 1.9313658475875854, + "learning_rate": 2.0000000000000002e-07, + "loss": 0.834, + "step": 4 + }, + { + "epoch": 0.03409090909090909, + "grad_norm": 1.983262062072754, + "learning_rate": 2.5000000000000004e-07, + "loss": 0.8726, + "step": 5 + }, + { + "epoch": 0.04090909090909091, + "grad_norm": 1.9994115829467773, + "learning_rate": 3.0000000000000004e-07, + "loss": 0.8135, + "step": 6 + }, + { + "epoch": 0.04772727272727273, + "grad_norm": 1.8889026641845703, + "learning_rate": 3.5000000000000004e-07, + "loss": 0.8439, + "step": 7 + }, + { + "epoch": 0.05454545454545454, + "grad_norm": 1.9258317947387695, + "learning_rate": 4.0000000000000003e-07, + "loss": 0.84, + "step": 8 + }, + { + "epoch": 0.06136363636363636, + "grad_norm": 1.9654443264007568, + "learning_rate": 4.5000000000000003e-07, + "loss": 0.8609, + "step": 9 + }, + { + "epoch": 0.06818181818181818, + "grad_norm": 1.9278813600540161, + "learning_rate": 5.000000000000001e-07, + "loss": 0.8515, + "step": 10 + }, + { + "epoch": 0.075, + "grad_norm": 1.8698863983154297, + "learning_rate": 5.5e-07, + "loss": 0.8632, + "step": 11 + }, + { + "epoch": 0.08181818181818182, + "grad_norm": 2.7078285217285156, + "learning_rate": 6.000000000000001e-07, + "loss": 0.8643, + "step": 12 + }, + { + "epoch": 0.08863636363636364, + "grad_norm": 1.959272027015686, + "learning_rate": 6.5e-07, + "loss": 0.861, + "step": 13 + }, + { + "epoch": 0.09545454545454546, + "grad_norm": 1.8952852487564087, + "learning_rate": 7.000000000000001e-07, + "loss": 0.8599, + "step": 14 + }, + { + "epoch": 0.10227272727272728, + "grad_norm": 1.9469462633132935, + "learning_rate": 7.5e-07, + "loss": 0.8467, + "step": 15 + }, + { + "epoch": 0.10909090909090909, + "grad_norm": 2.0111300945281982, + "learning_rate": 8.000000000000001e-07, + "loss": 0.8078, + "step": 16 + }, + { + "epoch": 0.1159090909090909, + "grad_norm": 1.739009141921997, + "learning_rate": 8.500000000000001e-07, + "loss": 0.8114, + "step": 17 + }, + { + "epoch": 0.12272727272727273, + "grad_norm": 1.7578809261322021, + "learning_rate": 9.000000000000001e-07, + "loss": 0.8489, + "step": 18 + }, + { + "epoch": 0.12954545454545455, + "grad_norm": 1.9787132740020752, + "learning_rate": 9.500000000000001e-07, + "loss": 0.8052, + "step": 19 + }, + { + "epoch": 0.13636363636363635, + "grad_norm": 1.6380164623260498, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.7882, + "step": 20 + }, + { + "epoch": 0.1431818181818182, + "grad_norm": 1.686877727508545, + "learning_rate": 1.0500000000000001e-06, + "loss": 0.8126, + "step": 21 + }, + { + "epoch": 0.15, + "grad_norm": 1.690590739250183, + "learning_rate": 1.1e-06, + "loss": 0.8248, + "step": 22 + }, + { + "epoch": 0.15681818181818183, + "grad_norm": 1.612255334854126, + "learning_rate": 1.1500000000000002e-06, + "loss": 0.8024, + "step": 23 + }, + { + "epoch": 0.16363636363636364, + "grad_norm": 1.6669739484786987, + "learning_rate": 1.2000000000000002e-06, + "loss": 0.818, + "step": 24 + }, + { + "epoch": 0.17045454545454544, + "grad_norm": 1.7726502418518066, + "learning_rate": 1.25e-06, + "loss": 0.8158, + "step": 25 + }, + { + "epoch": 0.17727272727272728, + "grad_norm": 1.5930036306381226, + "learning_rate": 1.3e-06, + "loss": 0.7886, + "step": 26 + }, + { + "epoch": 0.18409090909090908, + "grad_norm": 1.5480856895446777, + "learning_rate": 1.3500000000000002e-06, + "loss": 0.7747, + "step": 27 + }, + { + "epoch": 0.19090909090909092, + "grad_norm": 1.3537116050720215, + "learning_rate": 1.4000000000000001e-06, + "loss": 0.7891, + "step": 28 + }, + { + "epoch": 0.19772727272727272, + "grad_norm": 1.3412829637527466, + "learning_rate": 1.45e-06, + "loss": 0.7799, + "step": 29 + }, + { + "epoch": 0.20454545454545456, + "grad_norm": 1.237492322921753, + "learning_rate": 1.5e-06, + "loss": 0.7768, + "step": 30 + }, + { + "epoch": 0.21136363636363636, + "grad_norm": 1.3467835187911987, + "learning_rate": 1.5500000000000002e-06, + "loss": 0.7344, + "step": 31 + }, + { + "epoch": 0.21818181818181817, + "grad_norm": 1.1665314435958862, + "learning_rate": 1.6000000000000001e-06, + "loss": 0.7152, + "step": 32 + }, + { + "epoch": 0.225, + "grad_norm": 1.2461743354797363, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.758, + "step": 33 + }, + { + "epoch": 0.2318181818181818, + "grad_norm": 1.0343059301376343, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.7362, + "step": 34 + }, + { + "epoch": 0.23863636363636365, + "grad_norm": 1.112117052078247, + "learning_rate": 1.75e-06, + "loss": 0.7771, + "step": 35 + }, + { + "epoch": 0.24545454545454545, + "grad_norm": 1.104337453842163, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.7479, + "step": 36 + }, + { + "epoch": 0.25227272727272726, + "grad_norm": 0.9324578642845154, + "learning_rate": 1.85e-06, + "loss": 0.6936, + "step": 37 + }, + { + "epoch": 0.2590909090909091, + "grad_norm": 1.0733048915863037, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.7302, + "step": 38 + }, + { + "epoch": 0.26590909090909093, + "grad_norm": 0.8614453673362732, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.7466, + "step": 39 + }, + { + "epoch": 0.2727272727272727, + "grad_norm": 0.7729659080505371, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.725, + "step": 40 + }, + { + "epoch": 0.27954545454545454, + "grad_norm": 0.9341004490852356, + "learning_rate": 2.05e-06, + "loss": 0.6949, + "step": 41 + }, + { + "epoch": 0.2863636363636364, + "grad_norm": 0.6668925881385803, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.6587, + "step": 42 + }, + { + "epoch": 0.29318181818181815, + "grad_norm": 0.681847333908081, + "learning_rate": 2.15e-06, + "loss": 0.6984, + "step": 43 + }, + { + "epoch": 0.3, + "grad_norm": 0.6222082376480103, + "learning_rate": 2.2e-06, + "loss": 0.7063, + "step": 44 + }, + { + "epoch": 0.3068181818181818, + "grad_norm": 0.790607213973999, + "learning_rate": 2.25e-06, + "loss": 0.6804, + "step": 45 + }, + { + "epoch": 0.31363636363636366, + "grad_norm": 0.6110427379608154, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.6694, + "step": 46 + }, + { + "epoch": 0.32045454545454544, + "grad_norm": 0.6481993794441223, + "learning_rate": 2.35e-06, + "loss": 0.7321, + "step": 47 + }, + { + "epoch": 0.32727272727272727, + "grad_norm": 0.5973438620567322, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.6508, + "step": 48 + }, + { + "epoch": 0.3340909090909091, + "grad_norm": 0.6252740621566772, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.6911, + "step": 49 + }, + { + "epoch": 0.3409090909090909, + "grad_norm": 0.6826481819152832, + "learning_rate": 2.5e-06, + "loss": 0.6855, + "step": 50 + }, + { + "epoch": 0.3477272727272727, + "grad_norm": 0.6307975649833679, + "learning_rate": 2.55e-06, + "loss": 0.714, + "step": 51 + }, + { + "epoch": 0.35454545454545455, + "grad_norm": 0.5900976061820984, + "learning_rate": 2.6e-06, + "loss": 0.6619, + "step": 52 + }, + { + "epoch": 0.3613636363636364, + "grad_norm": 0.6203920841217041, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.6822, + "step": 53 + }, + { + "epoch": 0.36818181818181817, + "grad_norm": 0.6197589039802551, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.6584, + "step": 54 + }, + { + "epoch": 0.375, + "grad_norm": 0.5921110510826111, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.6739, + "step": 55 + }, + { + "epoch": 0.38181818181818183, + "grad_norm": 0.6215619444847107, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.6631, + "step": 56 + }, + { + "epoch": 0.3886363636363636, + "grad_norm": 0.5495648980140686, + "learning_rate": 2.85e-06, + "loss": 0.6531, + "step": 57 + }, + { + "epoch": 0.39545454545454545, + "grad_norm": 0.531947910785675, + "learning_rate": 2.9e-06, + "loss": 0.6365, + "step": 58 + }, + { + "epoch": 0.4022727272727273, + "grad_norm": 0.559112548828125, + "learning_rate": 2.95e-06, + "loss": 0.6391, + "step": 59 + }, + { + "epoch": 0.4090909090909091, + "grad_norm": 0.5377560257911682, + "learning_rate": 3e-06, + "loss": 0.6775, + "step": 60 + }, + { + "epoch": 0.4159090909090909, + "grad_norm": 0.5167352557182312, + "learning_rate": 3.05e-06, + "loss": 0.6518, + "step": 61 + }, + { + "epoch": 0.42272727272727273, + "grad_norm": 0.5483390092849731, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.5972, + "step": 62 + }, + { + "epoch": 0.42954545454545456, + "grad_norm": 0.5687050223350525, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.6515, + "step": 63 + }, + { + "epoch": 0.43636363636363634, + "grad_norm": 0.9046968817710876, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.6497, + "step": 64 + }, + { + "epoch": 0.4431818181818182, + "grad_norm": 0.5231667757034302, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.6364, + "step": 65 + }, + { + "epoch": 0.45, + "grad_norm": 0.49699342250823975, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.6144, + "step": 66 + }, + { + "epoch": 0.45681818181818185, + "grad_norm": 0.5390080213546753, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.6051, + "step": 67 + }, + { + "epoch": 0.4636363636363636, + "grad_norm": 0.5252938270568848, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.6353, + "step": 68 + }, + { + "epoch": 0.47045454545454546, + "grad_norm": 0.6291780471801758, + "learning_rate": 3.45e-06, + "loss": 0.6326, + "step": 69 + }, + { + "epoch": 0.4772727272727273, + "grad_norm": 0.5545375943183899, + "learning_rate": 3.5e-06, + "loss": 0.647, + "step": 70 + }, + { + "epoch": 0.48409090909090907, + "grad_norm": 0.47775956988334656, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.5993, + "step": 71 + }, + { + "epoch": 0.4909090909090909, + "grad_norm": 0.5016375184059143, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.6056, + "step": 72 + }, + { + "epoch": 0.49772727272727274, + "grad_norm": 0.4780517816543579, + "learning_rate": 3.65e-06, + "loss": 0.6189, + "step": 73 + }, + { + "epoch": 0.5045454545454545, + "grad_norm": 0.5763716697692871, + "learning_rate": 3.7e-06, + "loss": 0.6262, + "step": 74 + }, + { + "epoch": 0.5113636363636364, + "grad_norm": 0.5321404337882996, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.6336, + "step": 75 + }, + { + "epoch": 0.5181818181818182, + "grad_norm": 0.4641991853713989, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.6317, + "step": 76 + }, + { + "epoch": 0.525, + "grad_norm": 0.45118412375450134, + "learning_rate": 3.85e-06, + "loss": 0.5927, + "step": 77 + }, + { + "epoch": 0.5318181818181819, + "grad_norm": 0.5500645041465759, + "learning_rate": 3.900000000000001e-06, + "loss": 0.62, + "step": 78 + }, + { + "epoch": 0.5386363636363637, + "grad_norm": 0.5197616219520569, + "learning_rate": 3.95e-06, + "loss": 0.6063, + "step": 79 + }, + { + "epoch": 0.5454545454545454, + "grad_norm": 0.4672992527484894, + "learning_rate": 4.000000000000001e-06, + "loss": 0.6156, + "step": 80 + }, + { + "epoch": 0.5522727272727272, + "grad_norm": 0.5995073318481445, + "learning_rate": 4.05e-06, + "loss": 0.6219, + "step": 81 + }, + { + "epoch": 0.5590909090909091, + "grad_norm": 0.5593804717063904, + "learning_rate": 4.1e-06, + "loss": 0.6282, + "step": 82 + }, + { + "epoch": 0.5659090909090909, + "grad_norm": 0.7544977068901062, + "learning_rate": 4.15e-06, + "loss": 0.6246, + "step": 83 + }, + { + "epoch": 0.5727272727272728, + "grad_norm": 0.45378655195236206, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.5881, + "step": 84 + }, + { + "epoch": 0.5795454545454546, + "grad_norm": 0.4407907724380493, + "learning_rate": 4.25e-06, + "loss": 0.6026, + "step": 85 + }, + { + "epoch": 0.5863636363636363, + "grad_norm": 0.4333001673221588, + "learning_rate": 4.3e-06, + "loss": 0.6098, + "step": 86 + }, + { + "epoch": 0.5931818181818181, + "grad_norm": 0.45020997524261475, + "learning_rate": 4.350000000000001e-06, + "loss": 0.5858, + "step": 87 + }, + { + "epoch": 0.6, + "grad_norm": 0.424003541469574, + "learning_rate": 4.4e-06, + "loss": 0.6186, + "step": 88 + }, + { + "epoch": 0.6068181818181818, + "grad_norm": 0.427081435918808, + "learning_rate": 4.450000000000001e-06, + "loss": 0.6164, + "step": 89 + }, + { + "epoch": 0.6136363636363636, + "grad_norm": 0.4430546164512634, + "learning_rate": 4.5e-06, + "loss": 0.6049, + "step": 90 + }, + { + "epoch": 0.6204545454545455, + "grad_norm": 0.5469494462013245, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.608, + "step": 91 + }, + { + "epoch": 0.6272727272727273, + "grad_norm": 0.4481559693813324, + "learning_rate": 4.600000000000001e-06, + "loss": 0.5532, + "step": 92 + }, + { + "epoch": 0.634090909090909, + "grad_norm": 0.6563135981559753, + "learning_rate": 4.65e-06, + "loss": 0.6105, + "step": 93 + }, + { + "epoch": 0.6409090909090909, + "grad_norm": 0.4572807252407074, + "learning_rate": 4.7e-06, + "loss": 0.5976, + "step": 94 + }, + { + "epoch": 0.6477272727272727, + "grad_norm": 0.4423324167728424, + "learning_rate": 4.75e-06, + "loss": 0.616, + "step": 95 + }, + { + "epoch": 0.6545454545454545, + "grad_norm": 0.6230632066726685, + "learning_rate": 4.800000000000001e-06, + "loss": 0.5879, + "step": 96 + }, + { + "epoch": 0.6613636363636364, + "grad_norm": 0.461467981338501, + "learning_rate": 4.85e-06, + "loss": 0.5992, + "step": 97 + }, + { + "epoch": 0.6681818181818182, + "grad_norm": 0.44455233216285706, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.6138, + "step": 98 + }, + { + "epoch": 0.675, + "grad_norm": 0.47281715273857117, + "learning_rate": 4.95e-06, + "loss": 0.59, + "step": 99 + }, + { + "epoch": 0.6818181818181818, + "grad_norm": 0.46114516258239746, + "learning_rate": 5e-06, + "loss": 0.6216, + "step": 100 + }, + { + "epoch": 0.6886363636363636, + "grad_norm": 0.49900761246681213, + "learning_rate": 4.9999795126530275e-06, + "loss": 0.6222, + "step": 101 + }, + { + "epoch": 0.6954545454545454, + "grad_norm": 0.4764839708805084, + "learning_rate": 4.999918050947891e-06, + "loss": 0.5829, + "step": 102 + }, + { + "epoch": 0.7022727272727273, + "grad_norm": 0.43502742052078247, + "learning_rate": 4.999815615891943e-06, + "loss": 0.5982, + "step": 103 + }, + { + "epoch": 0.7090909090909091, + "grad_norm": 0.43318599462509155, + "learning_rate": 4.9996722091640805e-06, + "loss": 0.5827, + "step": 104 + }, + { + "epoch": 0.7159090909090909, + "grad_norm": 0.47218361496925354, + "learning_rate": 4.9994878331147225e-06, + "loss": 0.5907, + "step": 105 + }, + { + "epoch": 0.7227272727272728, + "grad_norm": 0.48492228984832764, + "learning_rate": 4.99926249076577e-06, + "loss": 0.5459, + "step": 106 + }, + { + "epoch": 0.7295454545454545, + "grad_norm": 0.4019850194454193, + "learning_rate": 4.998996185810557e-06, + "loss": 0.5957, + "step": 107 + }, + { + "epoch": 0.7363636363636363, + "grad_norm": 0.47361984848976135, + "learning_rate": 4.998688922613788e-06, + "loss": 0.579, + "step": 108 + }, + { + "epoch": 0.7431818181818182, + "grad_norm": 0.4612269401550293, + "learning_rate": 4.9983407062114695e-06, + "loss": 0.5814, + "step": 109 + }, + { + "epoch": 0.75, + "grad_norm": 0.45558398962020874, + "learning_rate": 4.9979515423108255e-06, + "loss": 0.6036, + "step": 110 + }, + { + "epoch": 0.7568181818181818, + "grad_norm": 0.41492199897766113, + "learning_rate": 4.997521437290205e-06, + "loss": 0.5891, + "step": 111 + }, + { + "epoch": 0.7636363636363637, + "grad_norm": 0.4484409689903259, + "learning_rate": 4.997050398198977e-06, + "loss": 0.6031, + "step": 112 + }, + { + "epoch": 0.7704545454545455, + "grad_norm": 0.42671382427215576, + "learning_rate": 4.996538432757414e-06, + "loss": 0.6106, + "step": 113 + }, + { + "epoch": 0.7772727272727272, + "grad_norm": 0.4222075343132019, + "learning_rate": 4.995985549356568e-06, + "loss": 0.5774, + "step": 114 + }, + { + "epoch": 0.7840909090909091, + "grad_norm": 0.421021431684494, + "learning_rate": 4.995391757058129e-06, + "loss": 0.5754, + "step": 115 + }, + { + "epoch": 0.7909090909090909, + "grad_norm": 0.4215952754020691, + "learning_rate": 4.99475706559428e-06, + "loss": 0.5895, + "step": 116 + }, + { + "epoch": 0.7977272727272727, + "grad_norm": 0.42652982473373413, + "learning_rate": 4.994081485367537e-06, + "loss": 0.5646, + "step": 117 + }, + { + "epoch": 0.8045454545454546, + "grad_norm": 0.4145742654800415, + "learning_rate": 4.993365027450576e-06, + "loss": 0.5862, + "step": 118 + }, + { + "epoch": 0.8113636363636364, + "grad_norm": 0.4754612445831299, + "learning_rate": 4.992607703586058e-06, + "loss": 0.5645, + "step": 119 + }, + { + "epoch": 0.8181818181818182, + "grad_norm": 0.4694959223270416, + "learning_rate": 4.991809526186424e-06, + "loss": 0.5989, + "step": 120 + }, + { + "epoch": 0.825, + "grad_norm": 0.4573237895965576, + "learning_rate": 4.990970508333707e-06, + "loss": 0.5769, + "step": 121 + }, + { + "epoch": 0.8318181818181818, + "grad_norm": 0.4592607915401459, + "learning_rate": 4.990090663779305e-06, + "loss": 0.5529, + "step": 122 + }, + { + "epoch": 0.8386363636363636, + "grad_norm": 0.4325920045375824, + "learning_rate": 4.9891700069437635e-06, + "loss": 0.5846, + "step": 123 + }, + { + "epoch": 0.8454545454545455, + "grad_norm": 0.44623467326164246, + "learning_rate": 4.988208552916535e-06, + "loss": 0.5997, + "step": 124 + }, + { + "epoch": 0.8522727272727273, + "grad_norm": 0.46071872115135193, + "learning_rate": 4.987206317455734e-06, + "loss": 0.5687, + "step": 125 + }, + { + "epoch": 0.8590909090909091, + "grad_norm": 0.43121206760406494, + "learning_rate": 4.986163316987877e-06, + "loss": 0.5277, + "step": 126 + }, + { + "epoch": 0.865909090909091, + "grad_norm": 0.4711630046367645, + "learning_rate": 4.985079568607613e-06, + "loss": 0.592, + "step": 127 + }, + { + "epoch": 0.8727272727272727, + "grad_norm": 0.4450221657752991, + "learning_rate": 4.983955090077445e-06, + "loss": 0.5656, + "step": 128 + }, + { + "epoch": 0.8795454545454545, + "grad_norm": 0.6546564102172852, + "learning_rate": 4.982789899827439e-06, + "loss": 0.572, + "step": 129 + }, + { + "epoch": 0.8863636363636364, + "grad_norm": 0.49622273445129395, + "learning_rate": 4.9815840169549216e-06, + "loss": 0.5719, + "step": 130 + }, + { + "epoch": 0.8931818181818182, + "grad_norm": 0.5478544235229492, + "learning_rate": 4.980337461224164e-06, + "loss": 0.5774, + "step": 131 + }, + { + "epoch": 0.9, + "grad_norm": 0.4340018033981323, + "learning_rate": 4.979050253066064e-06, + "loss": 0.5444, + "step": 132 + }, + { + "epoch": 0.9068181818181819, + "grad_norm": 0.4243793189525604, + "learning_rate": 4.977722413577802e-06, + "loss": 0.5759, + "step": 133 + }, + { + "epoch": 0.9136363636363637, + "grad_norm": 0.45201942324638367, + "learning_rate": 4.976353964522509e-06, + "loss": 0.5746, + "step": 134 + }, + { + "epoch": 0.9204545454545454, + "grad_norm": 0.42892220616340637, + "learning_rate": 4.974944928328894e-06, + "loss": 0.544, + "step": 135 + }, + { + "epoch": 0.9272727272727272, + "grad_norm": 0.43185824155807495, + "learning_rate": 4.973495328090891e-06, + "loss": 0.5803, + "step": 136 + }, + { + "epoch": 0.9340909090909091, + "grad_norm": 0.5512542724609375, + "learning_rate": 4.972005187567267e-06, + "loss": 0.5694, + "step": 137 + }, + { + "epoch": 0.9409090909090909, + "grad_norm": 0.4836059808731079, + "learning_rate": 4.970474531181245e-06, + "loss": 0.5581, + "step": 138 + }, + { + "epoch": 0.9477272727272728, + "grad_norm": 0.4508282244205475, + "learning_rate": 4.968903384020095e-06, + "loss": 0.5553, + "step": 139 + }, + { + "epoch": 0.9545454545454546, + "grad_norm": 0.4764733612537384, + "learning_rate": 4.967291771834727e-06, + "loss": 0.5798, + "step": 140 + }, + { + "epoch": 0.9613636363636363, + "grad_norm": 0.463468074798584, + "learning_rate": 4.965639721039267e-06, + "loss": 0.559, + "step": 141 + }, + { + "epoch": 0.9681818181818181, + "grad_norm": 0.4655650854110718, + "learning_rate": 4.963947258710626e-06, + "loss": 0.5576, + "step": 142 + }, + { + "epoch": 0.975, + "grad_norm": 0.4651234745979309, + "learning_rate": 4.962214412588053e-06, + "loss": 0.5792, + "step": 143 + }, + { + "epoch": 0.9818181818181818, + "grad_norm": 0.43080538511276245, + "learning_rate": 4.960441211072686e-06, + "loss": 0.5314, + "step": 144 + }, + { + "epoch": 0.9886363636363636, + "grad_norm": 0.4403238594532013, + "learning_rate": 4.9586276832270785e-06, + "loss": 0.5592, + "step": 145 + }, + { + "epoch": 0.9954545454545455, + "grad_norm": 0.43968647718429565, + "learning_rate": 4.9567738587747314e-06, + "loss": 0.5573, + "step": 146 + }, + { + "epoch": 1.0068181818181818, + "grad_norm": 0.8337184190750122, + "learning_rate": 4.954879768099599e-06, + "loss": 1.1118, + "step": 147 + }, + { + "epoch": 1.0136363636363637, + "grad_norm": 0.4042762815952301, + "learning_rate": 4.952945442245598e-06, + "loss": 0.5731, + "step": 148 + }, + { + "epoch": 1.0204545454545455, + "grad_norm": 0.5666297078132629, + "learning_rate": 4.95097091291609e-06, + "loss": 0.5415, + "step": 149 + }, + { + "epoch": 1.0272727272727273, + "grad_norm": 0.6148372888565063, + "learning_rate": 4.948956212473371e-06, + "loss": 0.5486, + "step": 150 + }, + { + "epoch": 1.0340909090909092, + "grad_norm": 0.5780720710754395, + "learning_rate": 4.946901373938132e-06, + "loss": 0.5373, + "step": 151 + }, + { + "epoch": 1.040909090909091, + "grad_norm": 0.4648835361003876, + "learning_rate": 4.944806430988927e-06, + "loss": 0.5419, + "step": 152 + }, + { + "epoch": 1.0477272727272728, + "grad_norm": 0.41131219267845154, + "learning_rate": 4.942671417961615e-06, + "loss": 0.5514, + "step": 153 + }, + { + "epoch": 1.0545454545454545, + "grad_norm": 0.4733349084854126, + "learning_rate": 4.940496369848795e-06, + "loss": 0.5563, + "step": 154 + }, + { + "epoch": 1.0613636363636363, + "grad_norm": 0.41179028153419495, + "learning_rate": 4.938281322299243e-06, + "loss": 0.5413, + "step": 155 + }, + { + "epoch": 1.0681818181818181, + "grad_norm": 0.4770941138267517, + "learning_rate": 4.936026311617316e-06, + "loss": 0.5561, + "step": 156 + }, + { + "epoch": 1.075, + "grad_norm": 0.7752397060394287, + "learning_rate": 4.933731374762361e-06, + "loss": 0.5439, + "step": 157 + }, + { + "epoch": 1.0818181818181818, + "grad_norm": 0.4323958456516266, + "learning_rate": 4.931396549348115e-06, + "loss": 0.5401, + "step": 158 + }, + { + "epoch": 1.0886363636363636, + "grad_norm": 0.4497937858104706, + "learning_rate": 4.9290218736420795e-06, + "loss": 0.523, + "step": 159 + }, + { + "epoch": 1.0954545454545455, + "grad_norm": 0.4881570339202881, + "learning_rate": 4.926607386564898e-06, + "loss": 0.5441, + "step": 160 + }, + { + "epoch": 1.1022727272727273, + "grad_norm": 0.43357932567596436, + "learning_rate": 4.9241531276897196e-06, + "loss": 0.5499, + "step": 161 + }, + { + "epoch": 1.1090909090909091, + "grad_norm": 1.1125361919403076, + "learning_rate": 4.921659137241544e-06, + "loss": 0.5265, + "step": 162 + }, + { + "epoch": 1.115909090909091, + "grad_norm": 0.43150508403778076, + "learning_rate": 4.919125456096574e-06, + "loss": 0.5193, + "step": 163 + }, + { + "epoch": 1.1227272727272728, + "grad_norm": 0.40771809220314026, + "learning_rate": 4.916552125781529e-06, + "loss": 0.538, + "step": 164 + }, + { + "epoch": 1.1295454545454546, + "grad_norm": 0.5830567479133606, + "learning_rate": 4.913939188472979e-06, + "loss": 0.5548, + "step": 165 + }, + { + "epoch": 1.1363636363636362, + "grad_norm": 0.4776526391506195, + "learning_rate": 4.911286686996648e-06, + "loss": 0.5328, + "step": 166 + }, + { + "epoch": 1.143181818181818, + "grad_norm": 0.41657742857933044, + "learning_rate": 4.908594664826708e-06, + "loss": 0.5336, + "step": 167 + }, + { + "epoch": 1.15, + "grad_norm": 0.4651948809623718, + "learning_rate": 4.905863166085076e-06, + "loss": 0.5812, + "step": 168 + }, + { + "epoch": 1.1568181818181817, + "grad_norm": 0.4399661719799042, + "learning_rate": 4.903092235540679e-06, + "loss": 0.5143, + "step": 169 + }, + { + "epoch": 1.1636363636363636, + "grad_norm": 0.4548935294151306, + "learning_rate": 4.900281918608732e-06, + "loss": 0.5425, + "step": 170 + }, + { + "epoch": 1.1704545454545454, + "grad_norm": 0.43515875935554504, + "learning_rate": 4.897432261349984e-06, + "loss": 0.534, + "step": 171 + }, + { + "epoch": 1.1772727272727272, + "grad_norm": 0.42954760789871216, + "learning_rate": 4.894543310469968e-06, + "loss": 0.5436, + "step": 172 + }, + { + "epoch": 1.184090909090909, + "grad_norm": 0.4063640534877777, + "learning_rate": 4.891615113318236e-06, + "loss": 0.5481, + "step": 173 + }, + { + "epoch": 1.190909090909091, + "grad_norm": 0.49416476488113403, + "learning_rate": 4.888647717887582e-06, + "loss": 0.5298, + "step": 174 + }, + { + "epoch": 1.1977272727272728, + "grad_norm": 1.4082964658737183, + "learning_rate": 4.8856411728132526e-06, + "loss": 0.5366, + "step": 175 + }, + { + "epoch": 1.2045454545454546, + "grad_norm": 0.44356465339660645, + "learning_rate": 4.8825955273721524e-06, + "loss": 0.5563, + "step": 176 + }, + { + "epoch": 1.2113636363636364, + "grad_norm": 0.40925371646881104, + "learning_rate": 4.879510831482039e-06, + "loss": 0.5444, + "step": 177 + }, + { + "epoch": 1.2181818181818183, + "grad_norm": 0.44379451870918274, + "learning_rate": 4.876387135700701e-06, + "loss": 0.5311, + "step": 178 + }, + { + "epoch": 1.225, + "grad_norm": 0.41487446427345276, + "learning_rate": 4.873224491225128e-06, + "loss": 0.5637, + "step": 179 + }, + { + "epoch": 1.231818181818182, + "grad_norm": 0.4520554542541504, + "learning_rate": 4.870022949890676e-06, + "loss": 0.5677, + "step": 180 + }, + { + "epoch": 1.2386363636363638, + "grad_norm": 0.4528166949748993, + "learning_rate": 4.866782564170217e-06, + "loss": 0.5095, + "step": 181 + }, + { + "epoch": 1.2454545454545454, + "grad_norm": 0.4409842789173126, + "learning_rate": 4.863503387173276e-06, + "loss": 0.5423, + "step": 182 + }, + { + "epoch": 1.2522727272727272, + "grad_norm": 0.4355996251106262, + "learning_rate": 4.860185472645161e-06, + "loss": 0.5422, + "step": 183 + }, + { + "epoch": 1.259090909090909, + "grad_norm": 0.4871126115322113, + "learning_rate": 4.856828874966086e-06, + "loss": 0.5293, + "step": 184 + }, + { + "epoch": 1.2659090909090909, + "grad_norm": 0.442348450422287, + "learning_rate": 4.853433649150276e-06, + "loss": 0.5616, + "step": 185 + }, + { + "epoch": 1.2727272727272727, + "grad_norm": 0.44014668464660645, + "learning_rate": 4.849999850845066e-06, + "loss": 0.5354, + "step": 186 + }, + { + "epoch": 1.2795454545454545, + "grad_norm": 0.4496249854564667, + "learning_rate": 4.8465275363299905e-06, + "loss": 0.5298, + "step": 187 + }, + { + "epoch": 1.2863636363636364, + "grad_norm": 0.46060171723365784, + "learning_rate": 4.84301676251586e-06, + "loss": 0.554, + "step": 188 + }, + { + "epoch": 1.2931818181818182, + "grad_norm": 0.7829347252845764, + "learning_rate": 4.839467586943825e-06, + "loss": 0.5606, + "step": 189 + }, + { + "epoch": 1.3, + "grad_norm": 0.45566731691360474, + "learning_rate": 4.835880067784441e-06, + "loss": 0.5118, + "step": 190 + }, + { + "epoch": 1.3068181818181819, + "grad_norm": 0.43211469054222107, + "learning_rate": 4.832254263836708e-06, + "loss": 0.535, + "step": 191 + }, + { + "epoch": 1.3136363636363637, + "grad_norm": 0.4350590705871582, + "learning_rate": 4.828590234527107e-06, + "loss": 0.5595, + "step": 192 + }, + { + "epoch": 1.3204545454545453, + "grad_norm": 0.4056399464607239, + "learning_rate": 4.82488803990863e-06, + "loss": 0.55, + "step": 193 + }, + { + "epoch": 1.3272727272727272, + "grad_norm": 0.4458341598510742, + "learning_rate": 4.821147740659795e-06, + "loss": 0.501, + "step": 194 + }, + { + "epoch": 1.334090909090909, + "grad_norm": 0.446504145860672, + "learning_rate": 4.817369398083648e-06, + "loss": 0.5393, + "step": 195 + }, + { + "epoch": 1.3409090909090908, + "grad_norm": 0.47838401794433594, + "learning_rate": 4.813553074106761e-06, + "loss": 0.5354, + "step": 196 + }, + { + "epoch": 1.3477272727272727, + "grad_norm": 0.4675792157649994, + "learning_rate": 4.809698831278217e-06, + "loss": 0.5202, + "step": 197 + }, + { + "epoch": 1.3545454545454545, + "grad_norm": 0.4351370632648468, + "learning_rate": 4.805806732768585e-06, + "loss": 0.5256, + "step": 198 + }, + { + "epoch": 1.3613636363636363, + "grad_norm": 0.4466511607170105, + "learning_rate": 4.801876842368882e-06, + "loss": 0.5238, + "step": 199 + }, + { + "epoch": 1.3681818181818182, + "grad_norm": 0.45672544836997986, + "learning_rate": 4.797909224489531e-06, + "loss": 0.5426, + "step": 200 + }, + { + "epoch": 1.375, + "grad_norm": 0.422638863325119, + "learning_rate": 4.793903944159303e-06, + "loss": 0.5385, + "step": 201 + }, + { + "epoch": 1.3818181818181818, + "grad_norm": 0.4254551827907562, + "learning_rate": 4.789861067024253e-06, + "loss": 0.503, + "step": 202 + }, + { + "epoch": 1.3886363636363637, + "grad_norm": 0.43860334157943726, + "learning_rate": 4.785780659346642e-06, + "loss": 0.519, + "step": 203 + }, + { + "epoch": 1.3954545454545455, + "grad_norm": 0.43294480443000793, + "learning_rate": 4.781662788003851e-06, + "loss": 0.5537, + "step": 204 + }, + { + "epoch": 1.4022727272727273, + "grad_norm": 0.45006492733955383, + "learning_rate": 4.777507520487289e-06, + "loss": 0.5318, + "step": 205 + }, + { + "epoch": 1.4090909090909092, + "grad_norm": 0.4178854823112488, + "learning_rate": 4.773314924901281e-06, + "loss": 0.5287, + "step": 206 + }, + { + "epoch": 1.415909090909091, + "grad_norm": 0.45662209391593933, + "learning_rate": 4.769085069961955e-06, + "loss": 0.5609, + "step": 207 + }, + { + "epoch": 1.4227272727272728, + "grad_norm": 0.41578733921051025, + "learning_rate": 4.764818024996117e-06, + "loss": 0.5249, + "step": 208 + }, + { + "epoch": 1.4295454545454547, + "grad_norm": 0.48305758833885193, + "learning_rate": 4.760513859940112e-06, + "loss": 0.515, + "step": 209 + }, + { + "epoch": 1.4363636363636363, + "grad_norm": 0.4564506709575653, + "learning_rate": 4.756172645338675e-06, + "loss": 0.535, + "step": 210 + }, + { + "epoch": 1.4431818181818181, + "grad_norm": 0.4499240815639496, + "learning_rate": 4.751794452343785e-06, + "loss": 0.5236, + "step": 211 + }, + { + "epoch": 1.45, + "grad_norm": 0.41851088404655457, + "learning_rate": 4.747379352713489e-06, + "loss": 0.5254, + "step": 212 + }, + { + "epoch": 1.4568181818181818, + "grad_norm": 0.4690937399864197, + "learning_rate": 4.7429274188107275e-06, + "loss": 0.5312, + "step": 213 + }, + { + "epoch": 1.4636363636363636, + "grad_norm": 0.5516958832740784, + "learning_rate": 4.738438723602154e-06, + "loss": 0.5315, + "step": 214 + }, + { + "epoch": 1.4704545454545455, + "grad_norm": 1.2040799856185913, + "learning_rate": 4.733913340656933e-06, + "loss": 0.544, + "step": 215 + }, + { + "epoch": 1.4772727272727273, + "grad_norm": 0.445425808429718, + "learning_rate": 4.729351344145536e-06, + "loss": 0.5309, + "step": 216 + }, + { + "epoch": 1.4840909090909091, + "grad_norm": 0.4186754822731018, + "learning_rate": 4.7247528088385296e-06, + "loss": 0.5138, + "step": 217 + }, + { + "epoch": 1.490909090909091, + "grad_norm": 0.43284520506858826, + "learning_rate": 4.720117810105341e-06, + "loss": 0.521, + "step": 218 + }, + { + "epoch": 1.4977272727272728, + "grad_norm": 0.4308505058288574, + "learning_rate": 4.715446423913036e-06, + "loss": 0.5436, + "step": 219 + }, + { + "epoch": 1.5045454545454544, + "grad_norm": 0.4471941888332367, + "learning_rate": 4.710738726825059e-06, + "loss": 0.5528, + "step": 220 + }, + { + "epoch": 1.5113636363636362, + "grad_norm": 0.4374965727329254, + "learning_rate": 4.705994795999991e-06, + "loss": 0.5238, + "step": 221 + }, + { + "epoch": 1.518181818181818, + "grad_norm": 0.6026428937911987, + "learning_rate": 4.701214709190277e-06, + "loss": 0.5268, + "step": 222 + }, + { + "epoch": 1.525, + "grad_norm": 0.4413348138332367, + "learning_rate": 4.696398544740955e-06, + "loss": 0.5534, + "step": 223 + }, + { + "epoch": 1.5318181818181817, + "grad_norm": 0.45917898416519165, + "learning_rate": 4.69154638158837e-06, + "loss": 0.511, + "step": 224 + }, + { + "epoch": 1.5386363636363636, + "grad_norm": 0.5339583158493042, + "learning_rate": 4.686658299258881e-06, + "loss": 0.5395, + "step": 225 + }, + { + "epoch": 1.5454545454545454, + "grad_norm": 0.4617757201194763, + "learning_rate": 4.681734377867562e-06, + "loss": 0.5425, + "step": 226 + }, + { + "epoch": 1.5522727272727272, + "grad_norm": 0.4513031840324402, + "learning_rate": 4.67677469811688e-06, + "loss": 0.5241, + "step": 227 + }, + { + "epoch": 1.559090909090909, + "grad_norm": 0.43548423051834106, + "learning_rate": 4.671779341295378e-06, + "loss": 0.5224, + "step": 228 + }, + { + "epoch": 1.565909090909091, + "grad_norm": 0.5516494512557983, + "learning_rate": 4.666748389276344e-06, + "loss": 0.5346, + "step": 229 + }, + { + "epoch": 1.5727272727272728, + "grad_norm": 0.4441164433956146, + "learning_rate": 4.661681924516466e-06, + "loss": 0.5208, + "step": 230 + }, + { + "epoch": 1.5795454545454546, + "grad_norm": 0.4057452380657196, + "learning_rate": 4.6565800300544805e-06, + "loss": 0.5173, + "step": 231 + }, + { + "epoch": 1.5863636363636364, + "grad_norm": 0.4162004590034485, + "learning_rate": 4.651442789509813e-06, + "loss": 0.5307, + "step": 232 + }, + { + "epoch": 1.5931818181818183, + "grad_norm": 0.43314146995544434, + "learning_rate": 4.646270287081208e-06, + "loss": 0.5359, + "step": 233 + }, + { + "epoch": 1.6, + "grad_norm": 0.5786082148551941, + "learning_rate": 4.641062607545347e-06, + "loss": 0.5307, + "step": 234 + }, + { + "epoch": 1.606818181818182, + "grad_norm": 0.43811941146850586, + "learning_rate": 4.6358198362554585e-06, + "loss": 0.5556, + "step": 235 + }, + { + "epoch": 1.6136363636363638, + "grad_norm": 0.4374074935913086, + "learning_rate": 4.630542059139923e-06, + "loss": 0.5253, + "step": 236 + }, + { + "epoch": 1.6204545454545456, + "grad_norm": 0.4387815594673157, + "learning_rate": 4.625229362700863e-06, + "loss": 0.534, + "step": 237 + }, + { + "epoch": 1.6272727272727274, + "grad_norm": 0.4976162016391754, + "learning_rate": 4.61988183401272e-06, + "loss": 0.5255, + "step": 238 + }, + { + "epoch": 1.634090909090909, + "grad_norm": 0.4603639543056488, + "learning_rate": 4.614499560720837e-06, + "loss": 0.511, + "step": 239 + }, + { + "epoch": 1.6409090909090909, + "grad_norm": 0.43136921525001526, + "learning_rate": 4.609082631040012e-06, + "loss": 0.5099, + "step": 240 + }, + { + "epoch": 1.6477272727272727, + "grad_norm": 0.4573090970516205, + "learning_rate": 4.603631133753061e-06, + "loss": 0.5515, + "step": 241 + }, + { + "epoch": 1.6545454545454545, + "grad_norm": 0.4292807877063751, + "learning_rate": 4.598145158209356e-06, + "loss": 0.5048, + "step": 242 + }, + { + "epoch": 1.6613636363636364, + "grad_norm": 0.6377942562103271, + "learning_rate": 4.592624794323366e-06, + "loss": 0.5333, + "step": 243 + }, + { + "epoch": 1.6681818181818182, + "grad_norm": 0.49924615025520325, + "learning_rate": 4.587070132573178e-06, + "loss": 0.5127, + "step": 244 + }, + { + "epoch": 1.675, + "grad_norm": 0.4723946154117584, + "learning_rate": 4.581481263999019e-06, + "loss": 0.5127, + "step": 245 + }, + { + "epoch": 1.6818181818181817, + "grad_norm": 1.1846898794174194, + "learning_rate": 4.575858280201761e-06, + "loss": 0.5367, + "step": 246 + }, + { + "epoch": 1.6886363636363635, + "grad_norm": 0.4306166172027588, + "learning_rate": 4.570201273341418e-06, + "loss": 0.52, + "step": 247 + }, + { + "epoch": 1.6954545454545453, + "grad_norm": 0.45308929681777954, + "learning_rate": 4.564510336135642e-06, + "loss": 0.5589, + "step": 248 + }, + { + "epoch": 1.7022727272727272, + "grad_norm": 0.4482448399066925, + "learning_rate": 4.558785561858196e-06, + "loss": 0.5161, + "step": 249 + }, + { + "epoch": 1.709090909090909, + "grad_norm": 0.44250935316085815, + "learning_rate": 4.5530270443374305e-06, + "loss": 0.5342, + "step": 250 + }, + { + "epoch": 1.7159090909090908, + "grad_norm": 0.8177540898323059, + "learning_rate": 4.547234877954741e-06, + "loss": 0.5402, + "step": 251 + }, + { + "epoch": 1.7227272727272727, + "grad_norm": 0.5169854760169983, + "learning_rate": 4.541409157643027e-06, + "loss": 0.5301, + "step": 252 + }, + { + "epoch": 1.7295454545454545, + "grad_norm": 0.4544285237789154, + "learning_rate": 4.535549978885132e-06, + "loss": 0.5314, + "step": 253 + }, + { + "epoch": 1.7363636363636363, + "grad_norm": 0.42860639095306396, + "learning_rate": 4.5296574377122765e-06, + "loss": 0.5249, + "step": 254 + }, + { + "epoch": 1.7431818181818182, + "grad_norm": 0.526492178440094, + "learning_rate": 4.5237316307024895e-06, + "loss": 0.5142, + "step": 255 + }, + { + "epoch": 1.75, + "grad_norm": 0.4208003878593445, + "learning_rate": 4.517772654979024e-06, + "loss": 0.5151, + "step": 256 + }, + { + "epoch": 1.7568181818181818, + "grad_norm": 0.4974212944507599, + "learning_rate": 4.51178060820876e-06, + "loss": 0.5161, + "step": 257 + }, + { + "epoch": 1.7636363636363637, + "grad_norm": 0.4148060083389282, + "learning_rate": 4.505755588600613e-06, + "loss": 0.5099, + "step": 258 + }, + { + "epoch": 1.7704545454545455, + "grad_norm": 0.4106168746948242, + "learning_rate": 4.499697694903915e-06, + "loss": 0.5376, + "step": 259 + }, + { + "epoch": 1.7772727272727273, + "grad_norm": 0.49017220735549927, + "learning_rate": 4.493607026406802e-06, + "loss": 0.525, + "step": 260 + }, + { + "epoch": 1.7840909090909092, + "grad_norm": 0.4141218662261963, + "learning_rate": 4.487483682934587e-06, + "loss": 0.5303, + "step": 261 + }, + { + "epoch": 1.790909090909091, + "grad_norm": 0.44826585054397583, + "learning_rate": 4.481327764848118e-06, + "loss": 0.4886, + "step": 262 + }, + { + "epoch": 1.7977272727272728, + "grad_norm": 0.42013809084892273, + "learning_rate": 4.47513937304214e-06, + "loss": 0.5204, + "step": 263 + }, + { + "epoch": 1.8045454545454547, + "grad_norm": 0.44327741861343384, + "learning_rate": 4.4689186089436365e-06, + "loss": 0.5097, + "step": 264 + }, + { + "epoch": 1.8113636363636365, + "grad_norm": 0.39498746395111084, + "learning_rate": 4.462665574510169e-06, + "loss": 0.5102, + "step": 265 + }, + { + "epoch": 1.8181818181818183, + "grad_norm": 0.48296087980270386, + "learning_rate": 4.456380372228208e-06, + "loss": 0.5299, + "step": 266 + }, + { + "epoch": 1.825, + "grad_norm": 0.4398634433746338, + "learning_rate": 4.450063105111447e-06, + "loss": 0.5419, + "step": 267 + }, + { + "epoch": 1.8318181818181818, + "grad_norm": 0.43333035707473755, + "learning_rate": 4.443713876699124e-06, + "loss": 0.5047, + "step": 268 + }, + { + "epoch": 1.8386363636363636, + "grad_norm": 0.5823659300804138, + "learning_rate": 4.4373327910543125e-06, + "loss": 0.5082, + "step": 269 + }, + { + "epoch": 1.8454545454545455, + "grad_norm": 0.44385817646980286, + "learning_rate": 4.430919952762226e-06, + "loss": 0.5171, + "step": 270 + }, + { + "epoch": 1.8522727272727273, + "grad_norm": 0.4444347023963928, + "learning_rate": 4.424475466928499e-06, + "loss": 0.5351, + "step": 271 + }, + { + "epoch": 1.8590909090909091, + "grad_norm": 0.549674391746521, + "learning_rate": 4.417999439177465e-06, + "loss": 0.498, + "step": 272 + }, + { + "epoch": 1.865909090909091, + "grad_norm": 0.4340274930000305, + "learning_rate": 4.4114919756504275e-06, + "loss": 0.5429, + "step": 273 + }, + { + "epoch": 1.8727272727272726, + "grad_norm": 0.44019854068756104, + "learning_rate": 4.404953183003916e-06, + "loss": 0.5133, + "step": 274 + }, + { + "epoch": 1.8795454545454544, + "grad_norm": 0.4473486542701721, + "learning_rate": 4.398383168407941e-06, + "loss": 0.5413, + "step": 275 + }, + { + "epoch": 1.8863636363636362, + "grad_norm": 0.44450780749320984, + "learning_rate": 4.391782039544239e-06, + "loss": 0.4998, + "step": 276 + }, + { + "epoch": 1.893181818181818, + "grad_norm": 0.4310123920440674, + "learning_rate": 4.385149904604502e-06, + "loss": 0.5134, + "step": 277 + }, + { + "epoch": 1.9, + "grad_norm": 0.48953917622566223, + "learning_rate": 4.378486872288611e-06, + "loss": 0.5102, + "step": 278 + }, + { + "epoch": 1.9068181818181817, + "grad_norm": 0.49219778180122375, + "learning_rate": 4.371793051802849e-06, + "loss": 0.5213, + "step": 279 + }, + { + "epoch": 1.9136363636363636, + "grad_norm": 0.818645715713501, + "learning_rate": 4.365068552858116e-06, + "loss": 0.5251, + "step": 280 + }, + { + "epoch": 1.9204545454545454, + "grad_norm": 0.4407028257846832, + "learning_rate": 4.358313485668124e-06, + "loss": 0.5191, + "step": 281 + }, + { + "epoch": 1.9272727272727272, + "grad_norm": 0.5301506519317627, + "learning_rate": 4.3515279609476e-06, + "loss": 0.5003, + "step": 282 + }, + { + "epoch": 1.934090909090909, + "grad_norm": 0.4673819839954376, + "learning_rate": 4.3447120899104615e-06, + "loss": 0.521, + "step": 283 + }, + { + "epoch": 1.940909090909091, + "grad_norm": 0.4106168746948242, + "learning_rate": 4.337865984268002e-06, + "loss": 0.5199, + "step": 284 + }, + { + "epoch": 1.9477272727272728, + "grad_norm": 0.43583595752716064, + "learning_rate": 4.3309897562270525e-06, + "loss": 0.5367, + "step": 285 + }, + { + "epoch": 1.9545454545454546, + "grad_norm": 0.4984910488128662, + "learning_rate": 4.324083518488151e-06, + "loss": 0.5202, + "step": 286 + }, + { + "epoch": 1.9613636363636364, + "grad_norm": 0.4516090452671051, + "learning_rate": 4.317147384243688e-06, + "loss": 0.5432, + "step": 287 + }, + { + "epoch": 1.9681818181818183, + "grad_norm": 0.43753015995025635, + "learning_rate": 4.3101814671760546e-06, + "loss": 0.5309, + "step": 288 + }, + { + "epoch": 1.975, + "grad_norm": 0.4472252130508423, + "learning_rate": 4.303185881455778e-06, + "loss": 0.5344, + "step": 289 + }, + { + "epoch": 1.981818181818182, + "grad_norm": 1.178377389907837, + "learning_rate": 4.296160741739652e-06, + "loss": 0.5283, + "step": 290 + }, + { + "epoch": 1.9886363636363638, + "grad_norm": 0.5085738301277161, + "learning_rate": 4.289106163168858e-06, + "loss": 0.5133, + "step": 291 + }, + { + "epoch": 1.9954545454545456, + "grad_norm": 0.52077716588974, + "learning_rate": 4.282022261367074e-06, + "loss": 0.5155, + "step": 292 + }, + { + "epoch": 2.006818181818182, + "grad_norm": 0.8534443378448486, + "learning_rate": 4.274909152438582e-06, + "loss": 1.003, + "step": 293 + }, + { + "epoch": 2.0136363636363637, + "grad_norm": 0.46160876750946045, + "learning_rate": 4.267766952966369e-06, + "loss": 0.4958, + "step": 294 + }, + { + "epoch": 2.0204545454545455, + "grad_norm": 0.5113462209701538, + "learning_rate": 4.260595780010209e-06, + "loss": 0.4895, + "step": 295 + }, + { + "epoch": 2.0272727272727273, + "grad_norm": 0.4553380608558655, + "learning_rate": 4.2533957511047485e-06, + "loss": 0.4826, + "step": 296 + }, + { + "epoch": 2.034090909090909, + "grad_norm": 0.43557822704315186, + "learning_rate": 4.24616698425758e-06, + "loss": 0.4591, + "step": 297 + }, + { + "epoch": 2.040909090909091, + "grad_norm": 0.45537200570106506, + "learning_rate": 4.238909597947307e-06, + "loss": 0.4811, + "step": 298 + }, + { + "epoch": 2.047727272727273, + "grad_norm": 0.48833370208740234, + "learning_rate": 4.231623711121603e-06, + "loss": 0.4967, + "step": 299 + }, + { + "epoch": 2.0545454545454547, + "grad_norm": 0.45504844188690186, + "learning_rate": 4.224309443195261e-06, + "loss": 0.4964, + "step": 300 + }, + { + "epoch": 2.0613636363636365, + "grad_norm": 0.4313275218009949, + "learning_rate": 4.2169669140482365e-06, + "loss": 0.4832, + "step": 301 + }, + { + "epoch": 2.0681818181818183, + "grad_norm": 0.4298369884490967, + "learning_rate": 4.2095962440236846e-06, + "loss": 0.5071, + "step": 302 + }, + { + "epoch": 2.075, + "grad_norm": 0.4188523590564728, + "learning_rate": 4.202197553925983e-06, + "loss": 0.4859, + "step": 303 + }, + { + "epoch": 2.081818181818182, + "grad_norm": 0.39195290207862854, + "learning_rate": 4.194770965018758e-06, + "loss": 0.5081, + "step": 304 + }, + { + "epoch": 2.088636363636364, + "grad_norm": 0.45844170451164246, + "learning_rate": 4.187316599022892e-06, + "loss": 0.508, + "step": 305 + }, + { + "epoch": 2.0954545454545457, + "grad_norm": 0.6743896007537842, + "learning_rate": 4.179834578114531e-06, + "loss": 0.4748, + "step": 306 + }, + { + "epoch": 2.102272727272727, + "grad_norm": 0.4317929446697235, + "learning_rate": 4.172325024923083e-06, + "loss": 0.504, + "step": 307 + }, + { + "epoch": 2.109090909090909, + "grad_norm": 0.4857175350189209, + "learning_rate": 4.164788062529203e-06, + "loss": 0.4987, + "step": 308 + }, + { + "epoch": 2.1159090909090907, + "grad_norm": 0.4454379677772522, + "learning_rate": 4.157223814462784e-06, + "loss": 0.4938, + "step": 309 + }, + { + "epoch": 2.1227272727272726, + "grad_norm": 0.5128078460693359, + "learning_rate": 4.149632404700925e-06, + "loss": 0.4917, + "step": 310 + }, + { + "epoch": 2.1295454545454544, + "grad_norm": 0.5195722579956055, + "learning_rate": 4.142013957665903e-06, + "loss": 0.4753, + "step": 311 + }, + { + "epoch": 2.1363636363636362, + "grad_norm": 0.4242931604385376, + "learning_rate": 4.134368598223132e-06, + "loss": 0.4832, + "step": 312 + }, + { + "epoch": 2.143181818181818, + "grad_norm": 0.4419071674346924, + "learning_rate": 4.126696451679116e-06, + "loss": 0.4783, + "step": 313 + }, + { + "epoch": 2.15, + "grad_norm": 0.42852380871772766, + "learning_rate": 4.118997643779401e-06, + "loss": 0.5034, + "step": 314 + }, + { + "epoch": 2.1568181818181817, + "grad_norm": 0.7373576164245605, + "learning_rate": 4.111272300706502e-06, + "loss": 0.4934, + "step": 315 + }, + { + "epoch": 2.1636363636363636, + "grad_norm": 0.4435337781906128, + "learning_rate": 4.1035205490778505e-06, + "loss": 0.5089, + "step": 316 + }, + { + "epoch": 2.1704545454545454, + "grad_norm": 0.4714502692222595, + "learning_rate": 4.095742515943703e-06, + "loss": 0.502, + "step": 317 + }, + { + "epoch": 2.1772727272727272, + "grad_norm": 0.4732494056224823, + "learning_rate": 4.087938328785071e-06, + "loss": 0.5073, + "step": 318 + }, + { + "epoch": 2.184090909090909, + "grad_norm": 0.45177504420280457, + "learning_rate": 4.080108115511629e-06, + "loss": 0.5076, + "step": 319 + }, + { + "epoch": 2.190909090909091, + "grad_norm": 0.506157398223877, + "learning_rate": 4.072252004459612e-06, + "loss": 0.4837, + "step": 320 + }, + { + "epoch": 2.1977272727272728, + "grad_norm": 0.43044376373291016, + "learning_rate": 4.064370124389718e-06, + "loss": 0.5167, + "step": 321 + }, + { + "epoch": 2.2045454545454546, + "grad_norm": 0.45878180861473083, + "learning_rate": 4.056462604484998e-06, + "loss": 0.4936, + "step": 322 + }, + { + "epoch": 2.2113636363636364, + "grad_norm": 0.43078047037124634, + "learning_rate": 4.048529574348734e-06, + "loss": 0.4783, + "step": 323 + }, + { + "epoch": 2.2181818181818183, + "grad_norm": 0.42789366841316223, + "learning_rate": 4.040571164002319e-06, + "loss": 0.4853, + "step": 324 + }, + { + "epoch": 2.225, + "grad_norm": 0.44888222217559814, + "learning_rate": 4.032587503883124e-06, + "loss": 0.5097, + "step": 325 + }, + { + "epoch": 2.231818181818182, + "grad_norm": 0.446851521730423, + "learning_rate": 4.0245787248423614e-06, + "loss": 0.4844, + "step": 326 + }, + { + "epoch": 2.2386363636363638, + "grad_norm": 0.436567097902298, + "learning_rate": 4.0165449581429404e-06, + "loss": 0.4549, + "step": 327 + }, + { + "epoch": 2.2454545454545456, + "grad_norm": 0.4813944697380066, + "learning_rate": 4.008486335457312e-06, + "loss": 0.49, + "step": 328 + }, + { + "epoch": 2.2522727272727274, + "grad_norm": 0.4935426115989685, + "learning_rate": 4.000402988865316e-06, + "loss": 0.4995, + "step": 329 + }, + { + "epoch": 2.2590909090909093, + "grad_norm": 1.1695573329925537, + "learning_rate": 3.992295050852013e-06, + "loss": 0.5033, + "step": 330 + }, + { + "epoch": 2.265909090909091, + "grad_norm": 0.41830316185951233, + "learning_rate": 3.984162654305516e-06, + "loss": 0.4828, + "step": 331 + }, + { + "epoch": 2.2727272727272725, + "grad_norm": 0.4522007703781128, + "learning_rate": 3.976005932514807e-06, + "loss": 0.4948, + "step": 332 + }, + { + "epoch": 2.2795454545454543, + "grad_norm": 0.8304370045661926, + "learning_rate": 3.967825019167559e-06, + "loss": 0.504, + "step": 333 + }, + { + "epoch": 2.286363636363636, + "grad_norm": 0.4533236622810364, + "learning_rate": 3.959620048347938e-06, + "loss": 0.4487, + "step": 334 + }, + { + "epoch": 2.293181818181818, + "grad_norm": 0.43578940629959106, + "learning_rate": 3.951391154534415e-06, + "loss": 0.4702, + "step": 335 + }, + { + "epoch": 2.3, + "grad_norm": 0.5146592259407043, + "learning_rate": 3.943138472597549e-06, + "loss": 0.4984, + "step": 336 + }, + { + "epoch": 2.3068181818181817, + "grad_norm": 0.6412249207496643, + "learning_rate": 3.934862137797788e-06, + "loss": 0.4761, + "step": 337 + }, + { + "epoch": 2.3136363636363635, + "grad_norm": 1.3352930545806885, + "learning_rate": 3.9265622857832455e-06, + "loss": 0.499, + "step": 338 + }, + { + "epoch": 2.3204545454545453, + "grad_norm": 0.4465159475803375, + "learning_rate": 3.918239052587481e-06, + "loss": 0.4525, + "step": 339 + }, + { + "epoch": 2.327272727272727, + "grad_norm": 0.43597185611724854, + "learning_rate": 3.909892574627267e-06, + "loss": 0.4952, + "step": 340 + }, + { + "epoch": 2.334090909090909, + "grad_norm": 0.7057136297225952, + "learning_rate": 3.901522988700355e-06, + "loss": 0.4576, + "step": 341 + }, + { + "epoch": 2.340909090909091, + "grad_norm": 0.45416319370269775, + "learning_rate": 3.893130431983234e-06, + "loss": 0.4902, + "step": 342 + }, + { + "epoch": 2.3477272727272727, + "grad_norm": 0.43900424242019653, + "learning_rate": 3.884715042028882e-06, + "loss": 0.495, + "step": 343 + }, + { + "epoch": 2.3545454545454545, + "grad_norm": 0.4553733468055725, + "learning_rate": 3.876276956764509e-06, + "loss": 0.4861, + "step": 344 + }, + { + "epoch": 2.3613636363636363, + "grad_norm": 0.5471370816230774, + "learning_rate": 3.867816314489301e-06, + "loss": 0.5043, + "step": 345 + }, + { + "epoch": 2.368181818181818, + "grad_norm": 0.43938082456588745, + "learning_rate": 3.8593332538721465e-06, + "loss": 0.4678, + "step": 346 + }, + { + "epoch": 2.375, + "grad_norm": 0.4583636224269867, + "learning_rate": 3.8508279139493736e-06, + "loss": 0.4679, + "step": 347 + }, + { + "epoch": 2.381818181818182, + "grad_norm": 0.44170665740966797, + "learning_rate": 3.84230043412246e-06, + "loss": 0.4647, + "step": 348 + }, + { + "epoch": 2.3886363636363637, + "grad_norm": 0.4952428340911865, + "learning_rate": 3.833750954155757e-06, + "loss": 0.4796, + "step": 349 + }, + { + "epoch": 2.3954545454545455, + "grad_norm": 0.42052993178367615, + "learning_rate": 3.825179614174195e-06, + "loss": 0.4606, + "step": 350 + }, + { + "epoch": 2.4022727272727273, + "grad_norm": 0.5792545676231384, + "learning_rate": 3.816586554660987e-06, + "loss": 0.4805, + "step": 351 + }, + { + "epoch": 2.409090909090909, + "grad_norm": 0.4477898180484772, + "learning_rate": 3.807971916455325e-06, + "loss": 0.4793, + "step": 352 + }, + { + "epoch": 2.415909090909091, + "grad_norm": 0.45617541670799255, + "learning_rate": 3.799335840750077e-06, + "loss": 0.502, + "step": 353 + }, + { + "epoch": 2.422727272727273, + "grad_norm": 0.45021289587020874, + "learning_rate": 3.790678469089465e-06, + "loss": 0.478, + "step": 354 + }, + { + "epoch": 2.4295454545454547, + "grad_norm": 0.4506435990333557, + "learning_rate": 3.7819999433667503e-06, + "loss": 0.4828, + "step": 355 + }, + { + "epoch": 2.4363636363636365, + "grad_norm": 0.48021042346954346, + "learning_rate": 3.773300405821908e-06, + "loss": 0.4827, + "step": 356 + }, + { + "epoch": 2.4431818181818183, + "grad_norm": 0.4759495258331299, + "learning_rate": 3.764579999039293e-06, + "loss": 0.4633, + "step": 357 + }, + { + "epoch": 2.45, + "grad_norm": 0.4734678268432617, + "learning_rate": 3.7558388659453052e-06, + "loss": 0.4982, + "step": 358 + }, + { + "epoch": 2.456818181818182, + "grad_norm": 0.4502730369567871, + "learning_rate": 3.7470771498060455e-06, + "loss": 0.5009, + "step": 359 + }, + { + "epoch": 2.463636363636364, + "grad_norm": 0.444940447807312, + "learning_rate": 3.7382949942249695e-06, + "loss": 0.4994, + "step": 360 + }, + { + "epoch": 2.4704545454545457, + "grad_norm": 0.4882519245147705, + "learning_rate": 3.7294925431405306e-06, + "loss": 0.4607, + "step": 361 + }, + { + "epoch": 2.4772727272727275, + "grad_norm": 0.51338791847229, + "learning_rate": 3.720669940823827e-06, + "loss": 0.4801, + "step": 362 + }, + { + "epoch": 2.484090909090909, + "grad_norm": 0.495211124420166, + "learning_rate": 3.7118273318762275e-06, + "loss": 0.4984, + "step": 363 + }, + { + "epoch": 2.4909090909090907, + "grad_norm": 0.4894484281539917, + "learning_rate": 3.702964861227013e-06, + "loss": 0.4834, + "step": 364 + }, + { + "epoch": 2.4977272727272726, + "grad_norm": 0.5554991364479065, + "learning_rate": 3.694082674130991e-06, + "loss": 0.4874, + "step": 365 + }, + { + "epoch": 2.5045454545454544, + "grad_norm": 0.4586203396320343, + "learning_rate": 3.6851809161661206e-06, + "loss": 0.5117, + "step": 366 + }, + { + "epoch": 2.5113636363636362, + "grad_norm": 0.7342172861099243, + "learning_rate": 3.6762597332311254e-06, + "loss": 0.5184, + "step": 367 + }, + { + "epoch": 2.518181818181818, + "grad_norm": 0.46638190746307373, + "learning_rate": 3.6673192715431016e-06, + "loss": 0.47, + "step": 368 + }, + { + "epoch": 2.525, + "grad_norm": 0.4584850072860718, + "learning_rate": 3.658359677635122e-06, + "loss": 0.45, + "step": 369 + }, + { + "epoch": 2.5318181818181817, + "grad_norm": 0.4285255968570709, + "learning_rate": 3.649381098353834e-06, + "loss": 0.513, + "step": 370 + }, + { + "epoch": 2.5386363636363636, + "grad_norm": 0.49487683176994324, + "learning_rate": 3.6403836808570512e-06, + "loss": 0.5048, + "step": 371 + }, + { + "epoch": 2.5454545454545454, + "grad_norm": 0.44076815247535706, + "learning_rate": 3.631367572611348e-06, + "loss": 0.4927, + "step": 372 + }, + { + "epoch": 2.5522727272727272, + "grad_norm": 0.44529926776885986, + "learning_rate": 3.6223329213896313e-06, + "loss": 0.4863, + "step": 373 + }, + { + "epoch": 2.559090909090909, + "grad_norm": 0.5416815876960754, + "learning_rate": 3.613279875268731e-06, + "loss": 0.4808, + "step": 374 + }, + { + "epoch": 2.565909090909091, + "grad_norm": 0.43630415201187134, + "learning_rate": 3.604208582626964e-06, + "loss": 0.4792, + "step": 375 + }, + { + "epoch": 2.5727272727272728, + "grad_norm": 0.47734326124191284, + "learning_rate": 3.5951191921417063e-06, + "loss": 0.4462, + "step": 376 + }, + { + "epoch": 2.5795454545454546, + "grad_norm": 0.4540305435657501, + "learning_rate": 3.586011852786955e-06, + "loss": 0.4915, + "step": 377 + }, + { + "epoch": 2.5863636363636364, + "grad_norm": 0.47168779373168945, + "learning_rate": 3.5768867138308872e-06, + "loss": 0.47, + "step": 378 + }, + { + "epoch": 2.5931818181818183, + "grad_norm": 0.46372920274734497, + "learning_rate": 3.5677439248334133e-06, + "loss": 0.4775, + "step": 379 + }, + { + "epoch": 2.6, + "grad_norm": 0.41653329133987427, + "learning_rate": 3.5585836356437266e-06, + "loss": 0.4875, + "step": 380 + }, + { + "epoch": 2.606818181818182, + "grad_norm": 0.4495103657245636, + "learning_rate": 3.5494059963978433e-06, + "loss": 0.4637, + "step": 381 + }, + { + "epoch": 2.6136363636363638, + "grad_norm": 0.432577520608902, + "learning_rate": 3.540211157516149e-06, + "loss": 0.4991, + "step": 382 + }, + { + "epoch": 2.6204545454545456, + "grad_norm": 0.4517219662666321, + "learning_rate": 3.530999269700927e-06, + "loss": 0.4785, + "step": 383 + }, + { + "epoch": 2.6272727272727274, + "grad_norm": 0.519922137260437, + "learning_rate": 3.521770483933891e-06, + "loss": 0.5026, + "step": 384 + }, + { + "epoch": 2.634090909090909, + "grad_norm": 0.48750364780426025, + "learning_rate": 3.5125249514737093e-06, + "loss": 0.4805, + "step": 385 + }, + { + "epoch": 2.6409090909090907, + "grad_norm": 0.6416886448860168, + "learning_rate": 3.503262823853527e-06, + "loss": 0.4849, + "step": 386 + }, + { + "epoch": 2.6477272727272725, + "grad_norm": 0.5748171806335449, + "learning_rate": 3.493984252878483e-06, + "loss": 0.4885, + "step": 387 + }, + { + "epoch": 2.6545454545454543, + "grad_norm": 0.5148530006408691, + "learning_rate": 3.484689390623218e-06, + "loss": 0.484, + "step": 388 + }, + { + "epoch": 2.661363636363636, + "grad_norm": 0.526839017868042, + "learning_rate": 3.4753783894293886e-06, + "loss": 0.4758, + "step": 389 + }, + { + "epoch": 2.668181818181818, + "grad_norm": 0.7565727829933167, + "learning_rate": 3.466051401903162e-06, + "loss": 0.479, + "step": 390 + }, + { + "epoch": 2.675, + "grad_norm": 0.4598790109157562, + "learning_rate": 3.4567085809127247e-06, + "loss": 0.4972, + "step": 391 + }, + { + "epoch": 2.6818181818181817, + "grad_norm": 0.4911792278289795, + "learning_rate": 3.4473500795857674e-06, + "loss": 0.4714, + "step": 392 + }, + { + "epoch": 2.6886363636363635, + "grad_norm": 0.4214136600494385, + "learning_rate": 3.4379760513069804e-06, + "loss": 0.4906, + "step": 393 + }, + { + "epoch": 2.6954545454545453, + "grad_norm": 0.5070644021034241, + "learning_rate": 3.428586649715542e-06, + "loss": 0.4936, + "step": 394 + }, + { + "epoch": 2.702272727272727, + "grad_norm": 0.4591734707355499, + "learning_rate": 3.4191820287025916e-06, + "loss": 0.4751, + "step": 395 + }, + { + "epoch": 2.709090909090909, + "grad_norm": 0.579781711101532, + "learning_rate": 3.4097623424087196e-06, + "loss": 0.4935, + "step": 396 + }, + { + "epoch": 2.715909090909091, + "grad_norm": 0.4178840219974518, + "learning_rate": 3.4003277452214284e-06, + "loss": 0.4771, + "step": 397 + }, + { + "epoch": 2.7227272727272727, + "grad_norm": 0.5002609491348267, + "learning_rate": 3.3908783917726123e-06, + "loss": 0.4739, + "step": 398 + }, + { + "epoch": 2.7295454545454545, + "grad_norm": 0.5176806449890137, + "learning_rate": 3.381414436936018e-06, + "loss": 0.4965, + "step": 399 + }, + { + "epoch": 2.7363636363636363, + "grad_norm": 0.8039212226867676, + "learning_rate": 3.3719360358247054e-06, + "loss": 0.4822, + "step": 400 + }, + { + "epoch": 2.743181818181818, + "grad_norm": 0.48943471908569336, + "learning_rate": 3.36244334378851e-06, + "loss": 0.4815, + "step": 401 + }, + { + "epoch": 2.75, + "grad_norm": 0.43710342049598694, + "learning_rate": 3.3529365164114903e-06, + "loss": 0.4932, + "step": 402 + }, + { + "epoch": 2.756818181818182, + "grad_norm": 0.5585265159606934, + "learning_rate": 3.3434157095093846e-06, + "loss": 0.5125, + "step": 403 + }, + { + "epoch": 2.7636363636363637, + "grad_norm": 0.4626101553440094, + "learning_rate": 3.333881079127052e-06, + "loss": 0.4934, + "step": 404 + }, + { + "epoch": 2.7704545454545455, + "grad_norm": 0.4675556421279907, + "learning_rate": 3.3243327815359168e-06, + "loss": 0.4903, + "step": 405 + }, + { + "epoch": 2.7772727272727273, + "grad_norm": 0.45658671855926514, + "learning_rate": 3.314770973231408e-06, + "loss": 0.4778, + "step": 406 + }, + { + "epoch": 2.784090909090909, + "grad_norm": 0.4911910593509674, + "learning_rate": 3.305195810930393e-06, + "loss": 0.4795, + "step": 407 + }, + { + "epoch": 2.790909090909091, + "grad_norm": 0.4397219121456146, + "learning_rate": 3.2956074515686105e-06, + "loss": 0.4589, + "step": 408 + }, + { + "epoch": 2.797727272727273, + "grad_norm": 0.5184643864631653, + "learning_rate": 3.2860060522980945e-06, + "loss": 0.4827, + "step": 409 + }, + { + "epoch": 2.8045454545454547, + "grad_norm": 0.45334312319755554, + "learning_rate": 3.276391770484606e-06, + "loss": 0.4821, + "step": 410 + }, + { + "epoch": 2.8113636363636365, + "grad_norm": 0.41846391558647156, + "learning_rate": 3.266764763705046e-06, + "loss": 0.5157, + "step": 411 + }, + { + "epoch": 2.8181818181818183, + "grad_norm": 0.45868799090385437, + "learning_rate": 3.257125189744877e-06, + "loss": 0.4876, + "step": 412 + }, + { + "epoch": 2.825, + "grad_norm": 0.4785497486591339, + "learning_rate": 3.247473206595536e-06, + "loss": 0.4809, + "step": 413 + }, + { + "epoch": 2.831818181818182, + "grad_norm": 0.5683679580688477, + "learning_rate": 3.2378089724518464e-06, + "loss": 0.505, + "step": 414 + }, + { + "epoch": 2.838636363636364, + "grad_norm": 0.48336583375930786, + "learning_rate": 3.228132645709421e-06, + "loss": 0.4816, + "step": 415 + }, + { + "epoch": 2.8454545454545457, + "grad_norm": 0.4236341416835785, + "learning_rate": 3.218444384962071e-06, + "loss": 0.4885, + "step": 416 + }, + { + "epoch": 2.8522727272727275, + "grad_norm": 0.4892752766609192, + "learning_rate": 3.2087443489992043e-06, + "loss": 0.4782, + "step": 417 + }, + { + "epoch": 2.8590909090909093, + "grad_norm": 0.4632202088832855, + "learning_rate": 3.1990326968032225e-06, + "loss": 0.5211, + "step": 418 + }, + { + "epoch": 2.865909090909091, + "grad_norm": 0.8842809796333313, + "learning_rate": 3.189309587546917e-06, + "loss": 0.4716, + "step": 419 + }, + { + "epoch": 2.8727272727272726, + "grad_norm": 0.46771398186683655, + "learning_rate": 3.1795751805908578e-06, + "loss": 0.4745, + "step": 420 + }, + { + "epoch": 2.8795454545454544, + "grad_norm": 0.8178832530975342, + "learning_rate": 3.169829635480783e-06, + "loss": 0.4917, + "step": 421 + }, + { + "epoch": 2.8863636363636362, + "grad_norm": 0.43057572841644287, + "learning_rate": 3.160073111944983e-06, + "loss": 0.4926, + "step": 422 + }, + { + "epoch": 2.893181818181818, + "grad_norm": 0.4258928894996643, + "learning_rate": 3.150305769891686e-06, + "loss": 0.4698, + "step": 423 + }, + { + "epoch": 2.9, + "grad_norm": 0.4559054374694824, + "learning_rate": 3.1405277694064306e-06, + "loss": 0.4537, + "step": 424 + }, + { + "epoch": 2.9068181818181817, + "grad_norm": 0.4342961609363556, + "learning_rate": 3.13073927074945e-06, + "loss": 0.4834, + "step": 425 + }, + { + "epoch": 2.9136363636363636, + "grad_norm": 0.48813825845718384, + "learning_rate": 3.1209404343530374e-06, + "loss": 0.4853, + "step": 426 + }, + { + "epoch": 2.9204545454545454, + "grad_norm": 0.47864893078804016, + "learning_rate": 3.111131420818922e-06, + "loss": 0.4627, + "step": 427 + }, + { + "epoch": 2.9272727272727272, + "grad_norm": 0.4394248425960541, + "learning_rate": 3.1013123909156347e-06, + "loss": 0.4817, + "step": 428 + }, + { + "epoch": 2.934090909090909, + "grad_norm": 0.6629825830459595, + "learning_rate": 3.091483505575873e-06, + "loss": 0.4738, + "step": 429 + }, + { + "epoch": 2.940909090909091, + "grad_norm": 0.47470518946647644, + "learning_rate": 3.081644925893866e-06, + "loss": 0.4943, + "step": 430 + }, + { + "epoch": 2.9477272727272728, + "grad_norm": 0.44007545709609985, + "learning_rate": 3.0717968131227285e-06, + "loss": 0.4805, + "step": 431 + }, + { + "epoch": 2.9545454545454546, + "grad_norm": 0.4425961375236511, + "learning_rate": 3.061939328671824e-06, + "loss": 0.5092, + "step": 432 + }, + { + "epoch": 2.9613636363636364, + "grad_norm": 0.4215753972530365, + "learning_rate": 3.0520726341041165e-06, + "loss": 0.4616, + "step": 433 + }, + { + "epoch": 2.9681818181818183, + "grad_norm": 0.4265761077404022, + "learning_rate": 3.0421968911335196e-06, + "loss": 0.4764, + "step": 434 + }, + { + "epoch": 2.975, + "grad_norm": 0.45009252429008484, + "learning_rate": 3.032312261622255e-06, + "loss": 0.4865, + "step": 435 + }, + { + "epoch": 2.981818181818182, + "grad_norm": 0.5175682306289673, + "learning_rate": 3.0224189075781886e-06, + "loss": 0.5085, + "step": 436 + }, + { + "epoch": 2.9886363636363638, + "grad_norm": 0.4518980383872986, + "learning_rate": 3.012516991152181e-06, + "loss": 0.4568, + "step": 437 + }, + { + "epoch": 2.9954545454545456, + "grad_norm": 0.4789412021636963, + "learning_rate": 3.002606674635432e-06, + "loss": 0.4942, + "step": 438 + }, + { + "epoch": 3.006818181818182, + "grad_norm": 0.495550274848938, + "learning_rate": 2.9926881204568153e-06, + "loss": 0.9793, + "step": 439 + }, + { + "epoch": 3.0136363636363637, + "grad_norm": 0.47357404232025146, + "learning_rate": 2.9827614911802205e-06, + "loss": 0.4689, + "step": 440 + }, + { + "epoch": 3.0204545454545455, + "grad_norm": 0.4523085355758667, + "learning_rate": 2.972826949501884e-06, + "loss": 0.4632, + "step": 441 + }, + { + "epoch": 3.0272727272727273, + "grad_norm": 0.4381036162376404, + "learning_rate": 2.9628846582477305e-06, + "loss": 0.4508, + "step": 442 + }, + { + "epoch": 3.034090909090909, + "grad_norm": 0.5367632508277893, + "learning_rate": 2.9529347803706943e-06, + "loss": 0.4472, + "step": 443 + }, + { + "epoch": 3.040909090909091, + "grad_norm": 0.5390172600746155, + "learning_rate": 2.9429774789480576e-06, + "loss": 0.4475, + "step": 444 + }, + { + "epoch": 3.047727272727273, + "grad_norm": 0.44644543528556824, + "learning_rate": 2.9330129171787704e-06, + "loss": 0.497, + "step": 445 + }, + { + "epoch": 3.0545454545454547, + "grad_norm": 0.46184423565864563, + "learning_rate": 2.923041258380779e-06, + "loss": 0.4545, + "step": 446 + }, + { + "epoch": 3.0613636363636365, + "grad_norm": 0.45963627099990845, + "learning_rate": 2.9130626659883537e-06, + "loss": 0.4571, + "step": 447 + }, + { + "epoch": 3.0681818181818183, + "grad_norm": 0.4504947364330292, + "learning_rate": 2.9030773035493997e-06, + "loss": 0.4563, + "step": 448 + }, + { + "epoch": 3.075, + "grad_norm": 0.4393884241580963, + "learning_rate": 2.893085334722786e-06, + "loss": 0.4433, + "step": 449 + }, + { + "epoch": 3.081818181818182, + "grad_norm": 0.7640544772148132, + "learning_rate": 2.883086923275658e-06, + "loss": 0.4598, + "step": 450 + }, + { + "epoch": 3.088636363636364, + "grad_norm": 0.44159695506095886, + "learning_rate": 2.8730822330807556e-06, + "loss": 0.4443, + "step": 451 + }, + { + "epoch": 3.0954545454545457, + "grad_norm": 0.4525455832481384, + "learning_rate": 2.8630714281137263e-06, + "loss": 0.4692, + "step": 452 + }, + { + "epoch": 3.102272727272727, + "grad_norm": 0.44391655921936035, + "learning_rate": 2.853054672450437e-06, + "loss": 0.4412, + "step": 453 + }, + { + "epoch": 3.109090909090909, + "grad_norm": 0.45541912317276, + "learning_rate": 2.8430321302642887e-06, + "loss": 0.4417, + "step": 454 + }, + { + "epoch": 3.1159090909090907, + "grad_norm": 0.43590644001960754, + "learning_rate": 2.8330039658235194e-06, + "loss": 0.4529, + "step": 455 + }, + { + "epoch": 3.1227272727272726, + "grad_norm": 0.48097532987594604, + "learning_rate": 2.8229703434885165e-06, + "loss": 0.4519, + "step": 456 + }, + { + "epoch": 3.1295454545454544, + "grad_norm": 0.43121981620788574, + "learning_rate": 2.8129314277091224e-06, + "loss": 0.4663, + "step": 457 + }, + { + "epoch": 3.1363636363636362, + "grad_norm": 0.46553748846054077, + "learning_rate": 2.8028873830219373e-06, + "loss": 0.4521, + "step": 458 + }, + { + "epoch": 3.143181818181818, + "grad_norm": 0.449290007352829, + "learning_rate": 2.7928383740476247e-06, + "loss": 0.4586, + "step": 459 + }, + { + "epoch": 3.15, + "grad_norm": 0.679371178150177, + "learning_rate": 2.7827845654882112e-06, + "loss": 0.4476, + "step": 460 + }, + { + "epoch": 3.1568181818181817, + "grad_norm": 0.49499261379241943, + "learning_rate": 2.7727261221243875e-06, + "loss": 0.4752, + "step": 461 + }, + { + "epoch": 3.1636363636363636, + "grad_norm": 0.4675534963607788, + "learning_rate": 2.76266320881281e-06, + "loss": 0.478, + "step": 462 + }, + { + "epoch": 3.1704545454545454, + "grad_norm": 0.44854965806007385, + "learning_rate": 2.7525959904833955e-06, + "loss": 0.4289, + "step": 463 + }, + { + "epoch": 3.1772727272727272, + "grad_norm": 0.4300600290298462, + "learning_rate": 2.7425246321366205e-06, + "loss": 0.4588, + "step": 464 + }, + { + "epoch": 3.184090909090909, + "grad_norm": 0.46728187799453735, + "learning_rate": 2.7324492988408146e-06, + "loss": 0.4498, + "step": 465 + }, + { + "epoch": 3.190909090909091, + "grad_norm": 0.46769359707832336, + "learning_rate": 2.7223701557294574e-06, + "loss": 0.4324, + "step": 466 + }, + { + "epoch": 3.1977272727272728, + "grad_norm": 0.4275517463684082, + "learning_rate": 2.712287367998471e-06, + "loss": 0.4574, + "step": 467 + }, + { + "epoch": 3.2045454545454546, + "grad_norm": 0.5888628959655762, + "learning_rate": 2.702201100903511e-06, + "loss": 0.4397, + "step": 468 + }, + { + "epoch": 3.2113636363636364, + "grad_norm": 0.4613579511642456, + "learning_rate": 2.692111519757261e-06, + "loss": 0.4391, + "step": 469 + }, + { + "epoch": 3.2181818181818183, + "grad_norm": 0.4581657946109772, + "learning_rate": 2.6820187899267203e-06, + "loss": 0.4734, + "step": 470 + }, + { + "epoch": 3.225, + "grad_norm": 0.4940917491912842, + "learning_rate": 2.671923076830496e-06, + "loss": 0.4726, + "step": 471 + }, + { + "epoch": 3.231818181818182, + "grad_norm": 0.7280719876289368, + "learning_rate": 2.6618245459360896e-06, + "loss": 0.4741, + "step": 472 + }, + { + "epoch": 3.2386363636363638, + "grad_norm": 0.519322395324707, + "learning_rate": 2.651723362757186e-06, + "loss": 0.4549, + "step": 473 + }, + { + "epoch": 3.2454545454545456, + "grad_norm": 0.47500962018966675, + "learning_rate": 2.641619692850941e-06, + "loss": 0.4598, + "step": 474 + }, + { + "epoch": 3.2522727272727274, + "grad_norm": 0.4661007821559906, + "learning_rate": 2.631513701815267e-06, + "loss": 0.4536, + "step": 475 + }, + { + "epoch": 3.2590909090909093, + "grad_norm": 0.48142874240875244, + "learning_rate": 2.6214055552861213e-06, + "loss": 0.474, + "step": 476 + }, + { + "epoch": 3.265909090909091, + "grad_norm": 0.5227756500244141, + "learning_rate": 2.611295418934786e-06, + "loss": 0.4516, + "step": 477 + }, + { + "epoch": 3.2727272727272725, + "grad_norm": 0.4693076014518738, + "learning_rate": 2.6011834584651597e-06, + "loss": 0.4409, + "step": 478 + }, + { + "epoch": 3.2795454545454543, + "grad_norm": 0.44575586915016174, + "learning_rate": 2.591069839611036e-06, + "loss": 0.4382, + "step": 479 + }, + { + "epoch": 3.286363636363636, + "grad_norm": 0.4569104313850403, + "learning_rate": 2.5809547281333904e-06, + "loss": 0.476, + "step": 480 + }, + { + "epoch": 3.293181818181818, + "grad_norm": 0.4542261064052582, + "learning_rate": 2.570838289817661e-06, + "loss": 0.4542, + "step": 481 + }, + { + "epoch": 3.3, + "grad_norm": 0.5086669921875, + "learning_rate": 2.560720690471033e-06, + "loss": 0.4673, + "step": 482 + }, + { + "epoch": 3.3068181818181817, + "grad_norm": 0.46243855357170105, + "learning_rate": 2.5506020959197218e-06, + "loss": 0.4462, + "step": 483 + }, + { + "epoch": 3.3136363636363635, + "grad_norm": 0.4877456724643707, + "learning_rate": 2.5404826720062544e-06, + "loss": 0.4463, + "step": 484 + }, + { + "epoch": 3.3204545454545453, + "grad_norm": 0.4592823088169098, + "learning_rate": 2.5303625845867475e-06, + "loss": 0.4674, + "step": 485 + }, + { + "epoch": 3.327272727272727, + "grad_norm": 0.4744127690792084, + "learning_rate": 2.5202419995281966e-06, + "loss": 0.4502, + "step": 486 + }, + { + "epoch": 3.334090909090909, + "grad_norm": 0.4267658591270447, + "learning_rate": 2.5101210827057516e-06, + "loss": 0.457, + "step": 487 + }, + { + "epoch": 3.340909090909091, + "grad_norm": 0.5804760456085205, + "learning_rate": 2.5e-06, + "loss": 0.4809, + "step": 488 + }, + { + "epoch": 3.3477272727272727, + "grad_norm": 0.5594801306724548, + "learning_rate": 2.4898789172942492e-06, + "loss": 0.4522, + "step": 489 + }, + { + "epoch": 3.3545454545454545, + "grad_norm": 0.4550947844982147, + "learning_rate": 2.4797580004718038e-06, + "loss": 0.4528, + "step": 490 + }, + { + "epoch": 3.3613636363636363, + "grad_norm": 0.430431991815567, + "learning_rate": 2.4696374154132533e-06, + "loss": 0.4657, + "step": 491 + }, + { + "epoch": 3.368181818181818, + "grad_norm": 0.4511764645576477, + "learning_rate": 2.4595173279937464e-06, + "loss": 0.4494, + "step": 492 + }, + { + "epoch": 3.375, + "grad_norm": 0.47958365082740784, + "learning_rate": 2.4493979040802786e-06, + "loss": 0.4508, + "step": 493 + }, + { + "epoch": 3.381818181818182, + "grad_norm": 0.4632541835308075, + "learning_rate": 2.4392793095289677e-06, + "loss": 0.4444, + "step": 494 + }, + { + "epoch": 3.3886363636363637, + "grad_norm": 0.46840110421180725, + "learning_rate": 2.42916171018234e-06, + "loss": 0.4676, + "step": 495 + }, + { + "epoch": 3.3954545454545455, + "grad_norm": 0.49999886751174927, + "learning_rate": 2.419045271866611e-06, + "loss": 0.4589, + "step": 496 + }, + { + "epoch": 3.4022727272727273, + "grad_norm": 0.44237902760505676, + "learning_rate": 2.408930160388965e-06, + "loss": 0.4569, + "step": 497 + }, + { + "epoch": 3.409090909090909, + "grad_norm": 0.4862026572227478, + "learning_rate": 2.3988165415348416e-06, + "loss": 0.4534, + "step": 498 + }, + { + "epoch": 3.415909090909091, + "grad_norm": 0.48773396015167236, + "learning_rate": 2.388704581065215e-06, + "loss": 0.4411, + "step": 499 + }, + { + "epoch": 3.422727272727273, + "grad_norm": 0.4380181133747101, + "learning_rate": 2.3785944447138804e-06, + "loss": 0.4529, + "step": 500 + }, + { + "epoch": 3.4295454545454547, + "grad_norm": 0.4616672992706299, + "learning_rate": 2.368486298184733e-06, + "loss": 0.4596, + "step": 501 + }, + { + "epoch": 3.4363636363636365, + "grad_norm": 0.4307732582092285, + "learning_rate": 2.358380307149059e-06, + "loss": 0.4626, + "step": 502 + }, + { + "epoch": 3.4431818181818183, + "grad_norm": 0.4402921795845032, + "learning_rate": 2.348276637242814e-06, + "loss": 0.4606, + "step": 503 + }, + { + "epoch": 3.45, + "grad_norm": 0.4782681465148926, + "learning_rate": 2.3381754540639108e-06, + "loss": 0.4508, + "step": 504 + }, + { + "epoch": 3.456818181818182, + "grad_norm": 0.43719062209129333, + "learning_rate": 2.328076923169504e-06, + "loss": 0.4476, + "step": 505 + }, + { + "epoch": 3.463636363636364, + "grad_norm": 0.4890111982822418, + "learning_rate": 2.31798121007328e-06, + "loss": 0.4484, + "step": 506 + }, + { + "epoch": 3.4704545454545457, + "grad_norm": 0.47682395577430725, + "learning_rate": 2.3078884802427394e-06, + "loss": 0.4494, + "step": 507 + }, + { + "epoch": 3.4772727272727275, + "grad_norm": 0.566819429397583, + "learning_rate": 2.29779889909649e-06, + "loss": 0.4539, + "step": 508 + }, + { + "epoch": 3.484090909090909, + "grad_norm": 0.4956795871257782, + "learning_rate": 2.2877126320015295e-06, + "loss": 0.4476, + "step": 509 + }, + { + "epoch": 3.4909090909090907, + "grad_norm": 0.5630896687507629, + "learning_rate": 2.2776298442705434e-06, + "loss": 0.4436, + "step": 510 + }, + { + "epoch": 3.4977272727272726, + "grad_norm": 0.9436272382736206, + "learning_rate": 2.267550701159186e-06, + "loss": 0.4558, + "step": 511 + }, + { + "epoch": 3.5045454545454544, + "grad_norm": 0.43885400891304016, + "learning_rate": 2.25747536786338e-06, + "loss": 0.4375, + "step": 512 + }, + { + "epoch": 3.5113636363636362, + "grad_norm": 0.762102484703064, + "learning_rate": 2.247404009516605e-06, + "loss": 0.4491, + "step": 513 + }, + { + "epoch": 3.518181818181818, + "grad_norm": 0.46883025765419006, + "learning_rate": 2.2373367911871904e-06, + "loss": 0.4393, + "step": 514 + }, + { + "epoch": 3.525, + "grad_norm": 0.4758882522583008, + "learning_rate": 2.227273877875613e-06, + "loss": 0.4446, + "step": 515 + }, + { + "epoch": 3.5318181818181817, + "grad_norm": 0.4799712300300598, + "learning_rate": 2.2172154345117896e-06, + "loss": 0.455, + "step": 516 + }, + { + "epoch": 3.5386363636363636, + "grad_norm": 0.46632879972457886, + "learning_rate": 2.207161625952376e-06, + "loss": 0.4481, + "step": 517 + }, + { + "epoch": 3.5454545454545454, + "grad_norm": 0.6286160945892334, + "learning_rate": 2.1971126169780636e-06, + "loss": 0.4436, + "step": 518 + }, + { + "epoch": 3.5522727272727272, + "grad_norm": 0.4679605960845947, + "learning_rate": 2.1870685722908784e-06, + "loss": 0.4575, + "step": 519 + }, + { + "epoch": 3.559090909090909, + "grad_norm": 0.495728075504303, + "learning_rate": 2.1770296565114847e-06, + "loss": 0.4399, + "step": 520 + }, + { + "epoch": 3.565909090909091, + "grad_norm": 0.4634799361228943, + "learning_rate": 2.166996034176482e-06, + "loss": 0.4553, + "step": 521 + }, + { + "epoch": 3.5727272727272728, + "grad_norm": 0.4464644491672516, + "learning_rate": 2.1569678697357126e-06, + "loss": 0.4693, + "step": 522 + }, + { + "epoch": 3.5795454545454546, + "grad_norm": 0.4530065655708313, + "learning_rate": 2.1469453275495634e-06, + "loss": 0.4418, + "step": 523 + }, + { + "epoch": 3.5863636363636364, + "grad_norm": 0.4806428551673889, + "learning_rate": 2.136928571886275e-06, + "loss": 0.4412, + "step": 524 + }, + { + "epoch": 3.5931818181818183, + "grad_norm": 0.46114400029182434, + "learning_rate": 2.126917766919245e-06, + "loss": 0.4502, + "step": 525 + }, + { + "epoch": 3.6, + "grad_norm": 0.7423635125160217, + "learning_rate": 2.1169130767243424e-06, + "loss": 0.4573, + "step": 526 + }, + { + "epoch": 3.606818181818182, + "grad_norm": 0.47871625423431396, + "learning_rate": 2.1069146652772142e-06, + "loss": 0.4427, + "step": 527 + }, + { + "epoch": 3.6136363636363638, + "grad_norm": 0.45554301142692566, + "learning_rate": 2.0969226964506007e-06, + "loss": 0.4429, + "step": 528 + }, + { + "epoch": 3.6204545454545456, + "grad_norm": 0.44822901487350464, + "learning_rate": 2.0869373340116467e-06, + "loss": 0.4616, + "step": 529 + }, + { + "epoch": 3.6272727272727274, + "grad_norm": 0.4922308027744293, + "learning_rate": 2.0769587416192212e-06, + "loss": 0.4482, + "step": 530 + }, + { + "epoch": 3.634090909090909, + "grad_norm": 0.4700371026992798, + "learning_rate": 2.066987082821231e-06, + "loss": 0.4652, + "step": 531 + }, + { + "epoch": 3.6409090909090907, + "grad_norm": 0.5513899326324463, + "learning_rate": 2.0570225210519433e-06, + "loss": 0.4542, + "step": 532 + }, + { + "epoch": 3.6477272727272725, + "grad_norm": 0.47292518615722656, + "learning_rate": 2.047065219629306e-06, + "loss": 0.4675, + "step": 533 + }, + { + "epoch": 3.6545454545454543, + "grad_norm": 0.5037868618965149, + "learning_rate": 2.0371153417522703e-06, + "loss": 0.4637, + "step": 534 + }, + { + "epoch": 3.661363636363636, + "grad_norm": 0.4578365087509155, + "learning_rate": 2.0271730504981165e-06, + "loss": 0.4432, + "step": 535 + }, + { + "epoch": 3.668181818181818, + "grad_norm": 0.518247127532959, + "learning_rate": 2.0172385088197804e-06, + "loss": 0.4574, + "step": 536 + }, + { + "epoch": 3.675, + "grad_norm": 0.5211175680160522, + "learning_rate": 2.007311879543185e-06, + "loss": 0.4607, + "step": 537 + }, + { + "epoch": 3.6818181818181817, + "grad_norm": 0.4695160686969757, + "learning_rate": 1.9973933253645684e-06, + "loss": 0.447, + "step": 538 + }, + { + "epoch": 3.6886363636363635, + "grad_norm": 0.5875166654586792, + "learning_rate": 1.9874830088478196e-06, + "loss": 0.4615, + "step": 539 + }, + { + "epoch": 3.6954545454545453, + "grad_norm": 0.4762527644634247, + "learning_rate": 1.9775810924218126e-06, + "loss": 0.4677, + "step": 540 + }, + { + "epoch": 3.702272727272727, + "grad_norm": 0.47306928038597107, + "learning_rate": 1.967687738377746e-06, + "loss": 0.4526, + "step": 541 + }, + { + "epoch": 3.709090909090909, + "grad_norm": 0.44723549485206604, + "learning_rate": 1.9578031088664812e-06, + "loss": 0.469, + "step": 542 + }, + { + "epoch": 3.715909090909091, + "grad_norm": 0.4447214901447296, + "learning_rate": 1.9479273658958852e-06, + "loss": 0.4375, + "step": 543 + }, + { + "epoch": 3.7227272727272727, + "grad_norm": 0.5299004316329956, + "learning_rate": 1.9380606713281773e-06, + "loss": 0.4544, + "step": 544 + }, + { + "epoch": 3.7295454545454545, + "grad_norm": 0.4843369424343109, + "learning_rate": 1.928203186877273e-06, + "loss": 0.4692, + "step": 545 + }, + { + "epoch": 3.7363636363636363, + "grad_norm": 0.4805986285209656, + "learning_rate": 1.9183550741061354e-06, + "loss": 0.4498, + "step": 546 + }, + { + "epoch": 3.743181818181818, + "grad_norm": 0.6539422273635864, + "learning_rate": 1.9085164944241275e-06, + "loss": 0.4398, + "step": 547 + }, + { + "epoch": 3.75, + "grad_norm": 0.47688615322113037, + "learning_rate": 1.8986876090843668e-06, + "loss": 0.4642, + "step": 548 + }, + { + "epoch": 3.756818181818182, + "grad_norm": 0.4808521270751953, + "learning_rate": 1.8888685791810784e-06, + "loss": 0.4166, + "step": 549 + }, + { + "epoch": 3.7636363636363637, + "grad_norm": 0.49430814385414124, + "learning_rate": 1.8790595656469628e-06, + "loss": 0.4404, + "step": 550 + }, + { + "epoch": 3.7704545454545455, + "grad_norm": 0.5284092426300049, + "learning_rate": 1.86926072925055e-06, + "loss": 0.445, + "step": 551 + }, + { + "epoch": 3.7772727272727273, + "grad_norm": 0.4650699496269226, + "learning_rate": 1.8594722305935691e-06, + "loss": 0.4381, + "step": 552 + }, + { + "epoch": 3.784090909090909, + "grad_norm": 0.4954371452331543, + "learning_rate": 1.8496942301083142e-06, + "loss": 0.4534, + "step": 553 + }, + { + "epoch": 3.790909090909091, + "grad_norm": 0.4756702482700348, + "learning_rate": 1.8399268880550174e-06, + "loss": 0.4563, + "step": 554 + }, + { + "epoch": 3.797727272727273, + "grad_norm": 0.4566079378128052, + "learning_rate": 1.8301703645192178e-06, + "loss": 0.4422, + "step": 555 + }, + { + "epoch": 3.8045454545454547, + "grad_norm": 1.6503181457519531, + "learning_rate": 1.8204248194091429e-06, + "loss": 0.4548, + "step": 556 + }, + { + "epoch": 3.8113636363636365, + "grad_norm": 0.5488978028297424, + "learning_rate": 1.8106904124530839e-06, + "loss": 0.4398, + "step": 557 + }, + { + "epoch": 3.8181818181818183, + "grad_norm": 0.44948896765708923, + "learning_rate": 1.800967303196778e-06, + "loss": 0.4331, + "step": 558 + }, + { + "epoch": 3.825, + "grad_norm": 0.49597036838531494, + "learning_rate": 1.7912556510007967e-06, + "loss": 0.4409, + "step": 559 + }, + { + "epoch": 3.831818181818182, + "grad_norm": 0.5093109011650085, + "learning_rate": 1.7815556150379298e-06, + "loss": 0.4615, + "step": 560 + }, + { + "epoch": 3.838636363636364, + "grad_norm": 0.4719341993331909, + "learning_rate": 1.77186735429058e-06, + "loss": 0.4316, + "step": 561 + }, + { + "epoch": 3.8454545454545457, + "grad_norm": 0.5281686782836914, + "learning_rate": 1.7621910275481544e-06, + "loss": 0.4474, + "step": 562 + }, + { + "epoch": 3.8522727272727275, + "grad_norm": 0.4955421984195709, + "learning_rate": 1.7525267934044642e-06, + "loss": 0.462, + "step": 563 + }, + { + "epoch": 3.8590909090909093, + "grad_norm": 0.4997479021549225, + "learning_rate": 1.7428748102551237e-06, + "loss": 0.4546, + "step": 564 + }, + { + "epoch": 3.865909090909091, + "grad_norm": 0.47572800517082214, + "learning_rate": 1.7332352362949546e-06, + "loss": 0.4587, + "step": 565 + }, + { + "epoch": 3.8727272727272726, + "grad_norm": 0.5156065225601196, + "learning_rate": 1.7236082295153948e-06, + "loss": 0.4375, + "step": 566 + }, + { + "epoch": 3.8795454545454544, + "grad_norm": 0.44926485419273376, + "learning_rate": 1.7139939477019057e-06, + "loss": 0.45, + "step": 567 + }, + { + "epoch": 3.8863636363636362, + "grad_norm": 0.5433095693588257, + "learning_rate": 1.7043925484313911e-06, + "loss": 0.4464, + "step": 568 + }, + { + "epoch": 3.893181818181818, + "grad_norm": 0.4719317555427551, + "learning_rate": 1.6948041890696076e-06, + "loss": 0.4467, + "step": 569 + }, + { + "epoch": 3.9, + "grad_norm": 0.4679432511329651, + "learning_rate": 1.685229026768593e-06, + "loss": 0.4577, + "step": 570 + }, + { + "epoch": 3.9068181818181817, + "grad_norm": 0.5053668022155762, + "learning_rate": 1.6756672184640847e-06, + "loss": 0.4379, + "step": 571 + }, + { + "epoch": 3.9136363636363636, + "grad_norm": 0.4880099594593048, + "learning_rate": 1.6661189208729492e-06, + "loss": 0.4585, + "step": 572 + }, + { + "epoch": 3.9204545454545454, + "grad_norm": 0.44481492042541504, + "learning_rate": 1.6565842904906154e-06, + "loss": 0.4319, + "step": 573 + }, + { + "epoch": 3.9272727272727272, + "grad_norm": 0.4998759329319, + "learning_rate": 1.6470634835885097e-06, + "loss": 0.4349, + "step": 574 + }, + { + "epoch": 3.934090909090909, + "grad_norm": 0.4702436625957489, + "learning_rate": 1.6375566562114903e-06, + "loss": 0.4409, + "step": 575 + }, + { + "epoch": 3.940909090909091, + "grad_norm": 1.1932134628295898, + "learning_rate": 1.6280639641752944e-06, + "loss": 0.4435, + "step": 576 + }, + { + "epoch": 3.9477272727272728, + "grad_norm": 0.46057313680648804, + "learning_rate": 1.6185855630639818e-06, + "loss": 0.451, + "step": 577 + }, + { + "epoch": 3.9545454545454546, + "grad_norm": 0.5858541131019592, + "learning_rate": 1.6091216082273875e-06, + "loss": 0.4662, + "step": 578 + }, + { + "epoch": 3.9613636363636364, + "grad_norm": 0.46951207518577576, + "learning_rate": 1.5996722547785722e-06, + "loss": 0.4561, + "step": 579 + }, + { + "epoch": 3.9681818181818183, + "grad_norm": 0.4747212529182434, + "learning_rate": 1.5902376575912815e-06, + "loss": 0.4507, + "step": 580 + }, + { + "epoch": 3.975, + "grad_norm": 0.5081924200057983, + "learning_rate": 1.580817971297409e-06, + "loss": 0.4649, + "step": 581 + }, + { + "epoch": 3.981818181818182, + "grad_norm": 0.4900786280632019, + "learning_rate": 1.5714133502844591e-06, + "loss": 0.4587, + "step": 582 + }, + { + "epoch": 3.9886363636363638, + "grad_norm": 0.4827897250652313, + "learning_rate": 1.56202394869302e-06, + "loss": 0.441, + "step": 583 + }, + { + "epoch": 3.9954545454545456, + "grad_norm": 0.499607652425766, + "learning_rate": 1.5526499204142332e-06, + "loss": 0.4559, + "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..a03cebdd4a04fb2e9443a05c92f15b2b2db01e10 --- /dev/null +++ b/checkpoint-584/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90038f7ff377ce29f00aedab533b18bfc9515b30d2b4232d8b0d6de26ec3ea56 +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..fdf619c317c2fe82074662582dbd68166b6f9d50 --- /dev/null +++ b/checkpoint-730/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3.1-Nemotron-70B-Instruct-HF +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..604978b603be594e46459a6c8c10d206a5a03b31 --- /dev/null +++ b/checkpoint-730/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF", + "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": [ + "down_proj", + "v_proj", + "q_proj", + "gate_proj", + "o_proj", + "up_proj", + "k_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..17a81119e765c7738784661e08a18eba4b7451ac --- /dev/null +++ b/checkpoint-730/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:b1a4de7edd478a2cdee0eb8819d5478638bfae5099dfcd2c2fc972ee58817d95 +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..7fe277fadd55803c15612e3726052b4eb4159f1c --- /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 sha256:22daac2b4952212b04da8f5498c314bcf56ca5a41ffc70664011c693ec1126df +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..9f14a4dc1e61e7538c355b065feb384722fd2b92 --- /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 sha256:555c9640a4068d0b80b819687182338025f6573dd9c3fa80de48ebcc6ea6dd2d +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..3186d6093c86ba95f2c4e719a24fee26efda05f2 --- /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 sha256:203695ed3d3c8e5b4cf5b87b7012a48cb2fcbfd59cad143ee37fed6936f2ef19 +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..3135b295411936b5f006675251f4428b6c2cd4b6 --- /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:ef4f08b609397dcb436bff3c0caac996b156de466d36557c58fc2f4eaa8ed8b9 +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..fdde94c29816839ec3c29d6c6461206a49911f3c --- /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\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..af02ef91bcfbc7b6485f86eaf33c0affa5cc4db2 --- /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": 2.020231246948242, + "learning_rate": 5.0000000000000004e-08, + "loss": 0.8736, + "step": 1 + }, + { + "epoch": 0.013636363636363636, + "grad_norm": 1.9368854761123657, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.8543, + "step": 2 + }, + { + "epoch": 0.020454545454545454, + "grad_norm": 1.908379077911377, + "learning_rate": 1.5000000000000002e-07, + "loss": 0.8339, + "step": 3 + }, + { + "epoch": 0.02727272727272727, + "grad_norm": 1.9313658475875854, + "learning_rate": 2.0000000000000002e-07, + "loss": 0.834, + "step": 4 + }, + { + "epoch": 0.03409090909090909, + "grad_norm": 1.983262062072754, + "learning_rate": 2.5000000000000004e-07, + "loss": 0.8726, + "step": 5 + }, + { + "epoch": 0.04090909090909091, + "grad_norm": 1.9994115829467773, + "learning_rate": 3.0000000000000004e-07, + "loss": 0.8135, + "step": 6 + }, + { + "epoch": 0.04772727272727273, + "grad_norm": 1.8889026641845703, + "learning_rate": 3.5000000000000004e-07, + "loss": 0.8439, + "step": 7 + }, + { + "epoch": 0.05454545454545454, + "grad_norm": 1.9258317947387695, + "learning_rate": 4.0000000000000003e-07, + "loss": 0.84, + "step": 8 + }, + { + "epoch": 0.06136363636363636, + "grad_norm": 1.9654443264007568, + "learning_rate": 4.5000000000000003e-07, + "loss": 0.8609, + "step": 9 + }, + { + "epoch": 0.06818181818181818, + "grad_norm": 1.9278813600540161, + "learning_rate": 5.000000000000001e-07, + "loss": 0.8515, + "step": 10 + }, + { + "epoch": 0.075, + "grad_norm": 1.8698863983154297, + "learning_rate": 5.5e-07, + "loss": 0.8632, + "step": 11 + }, + { + "epoch": 0.08181818181818182, + "grad_norm": 2.7078285217285156, + "learning_rate": 6.000000000000001e-07, + "loss": 0.8643, + "step": 12 + }, + { + "epoch": 0.08863636363636364, + "grad_norm": 1.959272027015686, + "learning_rate": 6.5e-07, + "loss": 0.861, + "step": 13 + }, + { + "epoch": 0.09545454545454546, + "grad_norm": 1.8952852487564087, + "learning_rate": 7.000000000000001e-07, + "loss": 0.8599, + "step": 14 + }, + { + "epoch": 0.10227272727272728, + "grad_norm": 1.9469462633132935, + "learning_rate": 7.5e-07, + "loss": 0.8467, + "step": 15 + }, + { + "epoch": 0.10909090909090909, + "grad_norm": 2.0111300945281982, + "learning_rate": 8.000000000000001e-07, + "loss": 0.8078, + "step": 16 + }, + { + "epoch": 0.1159090909090909, + "grad_norm": 1.739009141921997, + "learning_rate": 8.500000000000001e-07, + "loss": 0.8114, + "step": 17 + }, + { + "epoch": 0.12272727272727273, + "grad_norm": 1.7578809261322021, + "learning_rate": 9.000000000000001e-07, + "loss": 0.8489, + "step": 18 + }, + { + "epoch": 0.12954545454545455, + "grad_norm": 1.9787132740020752, + "learning_rate": 9.500000000000001e-07, + "loss": 0.8052, + "step": 19 + }, + { + "epoch": 0.13636363636363635, + "grad_norm": 1.6380164623260498, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.7882, + "step": 20 + }, + { + "epoch": 0.1431818181818182, + "grad_norm": 1.686877727508545, + "learning_rate": 1.0500000000000001e-06, + "loss": 0.8126, + "step": 21 + }, + { + "epoch": 0.15, + "grad_norm": 1.690590739250183, + "learning_rate": 1.1e-06, + "loss": 0.8248, + "step": 22 + }, + { + "epoch": 0.15681818181818183, + "grad_norm": 1.612255334854126, + "learning_rate": 1.1500000000000002e-06, + "loss": 0.8024, + "step": 23 + }, + { + "epoch": 0.16363636363636364, + "grad_norm": 1.6669739484786987, + "learning_rate": 1.2000000000000002e-06, + "loss": 0.818, + "step": 24 + }, + { + "epoch": 0.17045454545454544, + "grad_norm": 1.7726502418518066, + "learning_rate": 1.25e-06, + "loss": 0.8158, + "step": 25 + }, + { + "epoch": 0.17727272727272728, + "grad_norm": 1.5930036306381226, + "learning_rate": 1.3e-06, + "loss": 0.7886, + "step": 26 + }, + { + "epoch": 0.18409090909090908, + "grad_norm": 1.5480856895446777, + "learning_rate": 1.3500000000000002e-06, + "loss": 0.7747, + "step": 27 + }, + { + "epoch": 0.19090909090909092, + "grad_norm": 1.3537116050720215, + "learning_rate": 1.4000000000000001e-06, + "loss": 0.7891, + "step": 28 + }, + { + "epoch": 0.19772727272727272, + "grad_norm": 1.3412829637527466, + "learning_rate": 1.45e-06, + "loss": 0.7799, + "step": 29 + }, + { + "epoch": 0.20454545454545456, + "grad_norm": 1.237492322921753, + "learning_rate": 1.5e-06, + "loss": 0.7768, + "step": 30 + }, + { + "epoch": 0.21136363636363636, + "grad_norm": 1.3467835187911987, + "learning_rate": 1.5500000000000002e-06, + "loss": 0.7344, + "step": 31 + }, + { + "epoch": 0.21818181818181817, + "grad_norm": 1.1665314435958862, + "learning_rate": 1.6000000000000001e-06, + "loss": 0.7152, + "step": 32 + }, + { + "epoch": 0.225, + "grad_norm": 1.2461743354797363, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.758, + "step": 33 + }, + { + "epoch": 0.2318181818181818, + "grad_norm": 1.0343059301376343, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.7362, + "step": 34 + }, + { + "epoch": 0.23863636363636365, + "grad_norm": 1.112117052078247, + "learning_rate": 1.75e-06, + "loss": 0.7771, + "step": 35 + }, + { + "epoch": 0.24545454545454545, + "grad_norm": 1.104337453842163, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.7479, + "step": 36 + }, + { + "epoch": 0.25227272727272726, + "grad_norm": 0.9324578642845154, + "learning_rate": 1.85e-06, + "loss": 0.6936, + "step": 37 + }, + { + "epoch": 0.2590909090909091, + "grad_norm": 1.0733048915863037, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.7302, + "step": 38 + }, + { + "epoch": 0.26590909090909093, + "grad_norm": 0.8614453673362732, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.7466, + "step": 39 + }, + { + "epoch": 0.2727272727272727, + "grad_norm": 0.7729659080505371, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.725, + "step": 40 + }, + { + "epoch": 0.27954545454545454, + "grad_norm": 0.9341004490852356, + "learning_rate": 2.05e-06, + "loss": 0.6949, + "step": 41 + }, + { + "epoch": 0.2863636363636364, + "grad_norm": 0.6668925881385803, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.6587, + "step": 42 + }, + { + "epoch": 0.29318181818181815, + "grad_norm": 0.681847333908081, + "learning_rate": 2.15e-06, + "loss": 0.6984, + "step": 43 + }, + { + "epoch": 0.3, + "grad_norm": 0.6222082376480103, + "learning_rate": 2.2e-06, + "loss": 0.7063, + "step": 44 + }, + { + "epoch": 0.3068181818181818, + "grad_norm": 0.790607213973999, + "learning_rate": 2.25e-06, + "loss": 0.6804, + "step": 45 + }, + { + "epoch": 0.31363636363636366, + "grad_norm": 0.6110427379608154, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.6694, + "step": 46 + }, + { + "epoch": 0.32045454545454544, + "grad_norm": 0.6481993794441223, + "learning_rate": 2.35e-06, + "loss": 0.7321, + "step": 47 + }, + { + "epoch": 0.32727272727272727, + "grad_norm": 0.5973438620567322, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.6508, + "step": 48 + }, + { + "epoch": 0.3340909090909091, + "grad_norm": 0.6252740621566772, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.6911, + "step": 49 + }, + { + "epoch": 0.3409090909090909, + "grad_norm": 0.6826481819152832, + "learning_rate": 2.5e-06, + "loss": 0.6855, + "step": 50 + }, + { + "epoch": 0.3477272727272727, + "grad_norm": 0.6307975649833679, + "learning_rate": 2.55e-06, + "loss": 0.714, + "step": 51 + }, + { + "epoch": 0.35454545454545455, + "grad_norm": 0.5900976061820984, + "learning_rate": 2.6e-06, + "loss": 0.6619, + "step": 52 + }, + { + "epoch": 0.3613636363636364, + "grad_norm": 0.6203920841217041, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.6822, + "step": 53 + }, + { + "epoch": 0.36818181818181817, + "grad_norm": 0.6197589039802551, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.6584, + "step": 54 + }, + { + "epoch": 0.375, + "grad_norm": 0.5921110510826111, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.6739, + "step": 55 + }, + { + "epoch": 0.38181818181818183, + "grad_norm": 0.6215619444847107, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.6631, + "step": 56 + }, + { + "epoch": 0.3886363636363636, + "grad_norm": 0.5495648980140686, + "learning_rate": 2.85e-06, + "loss": 0.6531, + "step": 57 + }, + { + "epoch": 0.39545454545454545, + "grad_norm": 0.531947910785675, + "learning_rate": 2.9e-06, + "loss": 0.6365, + "step": 58 + }, + { + "epoch": 0.4022727272727273, + "grad_norm": 0.559112548828125, + "learning_rate": 2.95e-06, + "loss": 0.6391, + "step": 59 + }, + { + "epoch": 0.4090909090909091, + "grad_norm": 0.5377560257911682, + "learning_rate": 3e-06, + "loss": 0.6775, + "step": 60 + }, + { + "epoch": 0.4159090909090909, + "grad_norm": 0.5167352557182312, + "learning_rate": 3.05e-06, + "loss": 0.6518, + "step": 61 + }, + { + "epoch": 0.42272727272727273, + "grad_norm": 0.5483390092849731, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.5972, + "step": 62 + }, + { + "epoch": 0.42954545454545456, + "grad_norm": 0.5687050223350525, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.6515, + "step": 63 + }, + { + "epoch": 0.43636363636363634, + "grad_norm": 0.9046968817710876, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.6497, + "step": 64 + }, + { + "epoch": 0.4431818181818182, + "grad_norm": 0.5231667757034302, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.6364, + "step": 65 + }, + { + "epoch": 0.45, + "grad_norm": 0.49699342250823975, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.6144, + "step": 66 + }, + { + "epoch": 0.45681818181818185, + "grad_norm": 0.5390080213546753, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.6051, + "step": 67 + }, + { + "epoch": 0.4636363636363636, + "grad_norm": 0.5252938270568848, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.6353, + "step": 68 + }, + { + "epoch": 0.47045454545454546, + "grad_norm": 0.6291780471801758, + "learning_rate": 3.45e-06, + "loss": 0.6326, + "step": 69 + }, + { + "epoch": 0.4772727272727273, + "grad_norm": 0.5545375943183899, + "learning_rate": 3.5e-06, + "loss": 0.647, + "step": 70 + }, + { + "epoch": 0.48409090909090907, + "grad_norm": 0.47775956988334656, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.5993, + "step": 71 + }, + { + "epoch": 0.4909090909090909, + "grad_norm": 0.5016375184059143, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.6056, + "step": 72 + }, + { + "epoch": 0.49772727272727274, + "grad_norm": 0.4780517816543579, + "learning_rate": 3.65e-06, + "loss": 0.6189, + "step": 73 + }, + { + "epoch": 0.5045454545454545, + "grad_norm": 0.5763716697692871, + "learning_rate": 3.7e-06, + "loss": 0.6262, + "step": 74 + }, + { + "epoch": 0.5113636363636364, + "grad_norm": 0.5321404337882996, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.6336, + "step": 75 + }, + { + "epoch": 0.5181818181818182, + "grad_norm": 0.4641991853713989, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.6317, + "step": 76 + }, + { + "epoch": 0.525, + "grad_norm": 0.45118412375450134, + "learning_rate": 3.85e-06, + "loss": 0.5927, + "step": 77 + }, + { + "epoch": 0.5318181818181819, + "grad_norm": 0.5500645041465759, + "learning_rate": 3.900000000000001e-06, + "loss": 0.62, + "step": 78 + }, + { + "epoch": 0.5386363636363637, + "grad_norm": 0.5197616219520569, + "learning_rate": 3.95e-06, + "loss": 0.6063, + "step": 79 + }, + { + "epoch": 0.5454545454545454, + "grad_norm": 0.4672992527484894, + "learning_rate": 4.000000000000001e-06, + "loss": 0.6156, + "step": 80 + }, + { + "epoch": 0.5522727272727272, + "grad_norm": 0.5995073318481445, + "learning_rate": 4.05e-06, + "loss": 0.6219, + "step": 81 + }, + { + "epoch": 0.5590909090909091, + "grad_norm": 0.5593804717063904, + "learning_rate": 4.1e-06, + "loss": 0.6282, + "step": 82 + }, + { + "epoch": 0.5659090909090909, + "grad_norm": 0.7544977068901062, + "learning_rate": 4.15e-06, + "loss": 0.6246, + "step": 83 + }, + { + "epoch": 0.5727272727272728, + "grad_norm": 0.45378655195236206, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.5881, + "step": 84 + }, + { + "epoch": 0.5795454545454546, + "grad_norm": 0.4407907724380493, + "learning_rate": 4.25e-06, + "loss": 0.6026, + "step": 85 + }, + { + "epoch": 0.5863636363636363, + "grad_norm": 0.4333001673221588, + "learning_rate": 4.3e-06, + "loss": 0.6098, + "step": 86 + }, + { + "epoch": 0.5931818181818181, + "grad_norm": 0.45020997524261475, + "learning_rate": 4.350000000000001e-06, + "loss": 0.5858, + "step": 87 + }, + { + "epoch": 0.6, + "grad_norm": 0.424003541469574, + "learning_rate": 4.4e-06, + "loss": 0.6186, + "step": 88 + }, + { + "epoch": 0.6068181818181818, + "grad_norm": 0.427081435918808, + "learning_rate": 4.450000000000001e-06, + "loss": 0.6164, + "step": 89 + }, + { + "epoch": 0.6136363636363636, + "grad_norm": 0.4430546164512634, + "learning_rate": 4.5e-06, + "loss": 0.6049, + "step": 90 + }, + { + "epoch": 0.6204545454545455, + "grad_norm": 0.5469494462013245, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.608, + "step": 91 + }, + { + "epoch": 0.6272727272727273, + "grad_norm": 0.4481559693813324, + "learning_rate": 4.600000000000001e-06, + "loss": 0.5532, + "step": 92 + }, + { + "epoch": 0.634090909090909, + "grad_norm": 0.6563135981559753, + "learning_rate": 4.65e-06, + "loss": 0.6105, + "step": 93 + }, + { + "epoch": 0.6409090909090909, + "grad_norm": 0.4572807252407074, + "learning_rate": 4.7e-06, + "loss": 0.5976, + "step": 94 + }, + { + "epoch": 0.6477272727272727, + "grad_norm": 0.4423324167728424, + "learning_rate": 4.75e-06, + "loss": 0.616, + "step": 95 + }, + { + "epoch": 0.6545454545454545, + "grad_norm": 0.6230632066726685, + "learning_rate": 4.800000000000001e-06, + "loss": 0.5879, + "step": 96 + }, + { + "epoch": 0.6613636363636364, + "grad_norm": 0.461467981338501, + "learning_rate": 4.85e-06, + "loss": 0.5992, + "step": 97 + }, + { + "epoch": 0.6681818181818182, + "grad_norm": 0.44455233216285706, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.6138, + "step": 98 + }, + { + "epoch": 0.675, + "grad_norm": 0.47281715273857117, + "learning_rate": 4.95e-06, + "loss": 0.59, + "step": 99 + }, + { + "epoch": 0.6818181818181818, + "grad_norm": 0.46114516258239746, + "learning_rate": 5e-06, + "loss": 0.6216, + "step": 100 + }, + { + "epoch": 0.6886363636363636, + "grad_norm": 0.49900761246681213, + "learning_rate": 4.9999795126530275e-06, + "loss": 0.6222, + "step": 101 + }, + { + "epoch": 0.6954545454545454, + "grad_norm": 0.4764839708805084, + "learning_rate": 4.999918050947891e-06, + "loss": 0.5829, + "step": 102 + }, + { + "epoch": 0.7022727272727273, + "grad_norm": 0.43502742052078247, + "learning_rate": 4.999815615891943e-06, + "loss": 0.5982, + "step": 103 + }, + { + "epoch": 0.7090909090909091, + "grad_norm": 0.43318599462509155, + "learning_rate": 4.9996722091640805e-06, + "loss": 0.5827, + "step": 104 + }, + { + "epoch": 0.7159090909090909, + "grad_norm": 0.47218361496925354, + "learning_rate": 4.9994878331147225e-06, + "loss": 0.5907, + "step": 105 + }, + { + "epoch": 0.7227272727272728, + "grad_norm": 0.48492228984832764, + "learning_rate": 4.99926249076577e-06, + "loss": 0.5459, + "step": 106 + }, + { + "epoch": 0.7295454545454545, + "grad_norm": 0.4019850194454193, + "learning_rate": 4.998996185810557e-06, + "loss": 0.5957, + "step": 107 + }, + { + "epoch": 0.7363636363636363, + "grad_norm": 0.47361984848976135, + "learning_rate": 4.998688922613788e-06, + "loss": 0.579, + "step": 108 + }, + { + "epoch": 0.7431818181818182, + "grad_norm": 0.4612269401550293, + "learning_rate": 4.9983407062114695e-06, + "loss": 0.5814, + "step": 109 + }, + { + "epoch": 0.75, + "grad_norm": 0.45558398962020874, + "learning_rate": 4.9979515423108255e-06, + "loss": 0.6036, + "step": 110 + }, + { + "epoch": 0.7568181818181818, + "grad_norm": 0.41492199897766113, + "learning_rate": 4.997521437290205e-06, + "loss": 0.5891, + "step": 111 + }, + { + "epoch": 0.7636363636363637, + "grad_norm": 0.4484409689903259, + "learning_rate": 4.997050398198977e-06, + "loss": 0.6031, + "step": 112 + }, + { + "epoch": 0.7704545454545455, + "grad_norm": 0.42671382427215576, + "learning_rate": 4.996538432757414e-06, + "loss": 0.6106, + "step": 113 + }, + { + "epoch": 0.7772727272727272, + "grad_norm": 0.4222075343132019, + "learning_rate": 4.995985549356568e-06, + "loss": 0.5774, + "step": 114 + }, + { + "epoch": 0.7840909090909091, + "grad_norm": 0.421021431684494, + "learning_rate": 4.995391757058129e-06, + "loss": 0.5754, + "step": 115 + }, + { + "epoch": 0.7909090909090909, + "grad_norm": 0.4215952754020691, + "learning_rate": 4.99475706559428e-06, + "loss": 0.5895, + "step": 116 + }, + { + "epoch": 0.7977272727272727, + "grad_norm": 0.42652982473373413, + "learning_rate": 4.994081485367537e-06, + "loss": 0.5646, + "step": 117 + }, + { + "epoch": 0.8045454545454546, + "grad_norm": 0.4145742654800415, + "learning_rate": 4.993365027450576e-06, + "loss": 0.5862, + "step": 118 + }, + { + "epoch": 0.8113636363636364, + "grad_norm": 0.4754612445831299, + "learning_rate": 4.992607703586058e-06, + "loss": 0.5645, + "step": 119 + }, + { + "epoch": 0.8181818181818182, + "grad_norm": 0.4694959223270416, + "learning_rate": 4.991809526186424e-06, + "loss": 0.5989, + "step": 120 + }, + { + "epoch": 0.825, + "grad_norm": 0.4573237895965576, + "learning_rate": 4.990970508333707e-06, + "loss": 0.5769, + "step": 121 + }, + { + "epoch": 0.8318181818181818, + "grad_norm": 0.4592607915401459, + "learning_rate": 4.990090663779305e-06, + "loss": 0.5529, + "step": 122 + }, + { + "epoch": 0.8386363636363636, + "grad_norm": 0.4325920045375824, + "learning_rate": 4.9891700069437635e-06, + "loss": 0.5846, + "step": 123 + }, + { + "epoch": 0.8454545454545455, + "grad_norm": 0.44623467326164246, + "learning_rate": 4.988208552916535e-06, + "loss": 0.5997, + "step": 124 + }, + { + "epoch": 0.8522727272727273, + "grad_norm": 0.46071872115135193, + "learning_rate": 4.987206317455734e-06, + "loss": 0.5687, + "step": 125 + }, + { + "epoch": 0.8590909090909091, + "grad_norm": 0.43121206760406494, + "learning_rate": 4.986163316987877e-06, + "loss": 0.5277, + "step": 126 + }, + { + "epoch": 0.865909090909091, + "grad_norm": 0.4711630046367645, + "learning_rate": 4.985079568607613e-06, + "loss": 0.592, + "step": 127 + }, + { + "epoch": 0.8727272727272727, + "grad_norm": 0.4450221657752991, + "learning_rate": 4.983955090077445e-06, + "loss": 0.5656, + "step": 128 + }, + { + "epoch": 0.8795454545454545, + "grad_norm": 0.6546564102172852, + "learning_rate": 4.982789899827439e-06, + "loss": 0.572, + "step": 129 + }, + { + "epoch": 0.8863636363636364, + "grad_norm": 0.49622273445129395, + "learning_rate": 4.9815840169549216e-06, + "loss": 0.5719, + "step": 130 + }, + { + "epoch": 0.8931818181818182, + "grad_norm": 0.5478544235229492, + "learning_rate": 4.980337461224164e-06, + "loss": 0.5774, + "step": 131 + }, + { + "epoch": 0.9, + "grad_norm": 0.4340018033981323, + "learning_rate": 4.979050253066064e-06, + "loss": 0.5444, + "step": 132 + }, + { + "epoch": 0.9068181818181819, + "grad_norm": 0.4243793189525604, + "learning_rate": 4.977722413577802e-06, + "loss": 0.5759, + "step": 133 + }, + { + "epoch": 0.9136363636363637, + "grad_norm": 0.45201942324638367, + "learning_rate": 4.976353964522509e-06, + "loss": 0.5746, + "step": 134 + }, + { + "epoch": 0.9204545454545454, + "grad_norm": 0.42892220616340637, + "learning_rate": 4.974944928328894e-06, + "loss": 0.544, + "step": 135 + }, + { + "epoch": 0.9272727272727272, + "grad_norm": 0.43185824155807495, + "learning_rate": 4.973495328090891e-06, + "loss": 0.5803, + "step": 136 + }, + { + "epoch": 0.9340909090909091, + "grad_norm": 0.5512542724609375, + "learning_rate": 4.972005187567267e-06, + "loss": 0.5694, + "step": 137 + }, + { + "epoch": 0.9409090909090909, + "grad_norm": 0.4836059808731079, + "learning_rate": 4.970474531181245e-06, + "loss": 0.5581, + "step": 138 + }, + { + "epoch": 0.9477272727272728, + "grad_norm": 0.4508282244205475, + "learning_rate": 4.968903384020095e-06, + "loss": 0.5553, + "step": 139 + }, + { + "epoch": 0.9545454545454546, + "grad_norm": 0.4764733612537384, + "learning_rate": 4.967291771834727e-06, + "loss": 0.5798, + "step": 140 + }, + { + "epoch": 0.9613636363636363, + "grad_norm": 0.463468074798584, + "learning_rate": 4.965639721039267e-06, + "loss": 0.559, + "step": 141 + }, + { + "epoch": 0.9681818181818181, + "grad_norm": 0.4655650854110718, + "learning_rate": 4.963947258710626e-06, + "loss": 0.5576, + "step": 142 + }, + { + "epoch": 0.975, + "grad_norm": 0.4651234745979309, + "learning_rate": 4.962214412588053e-06, + "loss": 0.5792, + "step": 143 + }, + { + "epoch": 0.9818181818181818, + "grad_norm": 0.43080538511276245, + "learning_rate": 4.960441211072686e-06, + "loss": 0.5314, + "step": 144 + }, + { + "epoch": 0.9886363636363636, + "grad_norm": 0.4403238594532013, + "learning_rate": 4.9586276832270785e-06, + "loss": 0.5592, + "step": 145 + }, + { + "epoch": 0.9954545454545455, + "grad_norm": 0.43968647718429565, + "learning_rate": 4.9567738587747314e-06, + "loss": 0.5573, + "step": 146 + }, + { + "epoch": 1.0068181818181818, + "grad_norm": 0.8337184190750122, + "learning_rate": 4.954879768099599e-06, + "loss": 1.1118, + "step": 147 + }, + { + "epoch": 1.0136363636363637, + "grad_norm": 0.4042762815952301, + "learning_rate": 4.952945442245598e-06, + "loss": 0.5731, + "step": 148 + }, + { + "epoch": 1.0204545454545455, + "grad_norm": 0.5666297078132629, + "learning_rate": 4.95097091291609e-06, + "loss": 0.5415, + "step": 149 + }, + { + "epoch": 1.0272727272727273, + "grad_norm": 0.6148372888565063, + "learning_rate": 4.948956212473371e-06, + "loss": 0.5486, + "step": 150 + }, + { + "epoch": 1.0340909090909092, + "grad_norm": 0.5780720710754395, + "learning_rate": 4.946901373938132e-06, + "loss": 0.5373, + "step": 151 + }, + { + "epoch": 1.040909090909091, + "grad_norm": 0.4648835361003876, + "learning_rate": 4.944806430988927e-06, + "loss": 0.5419, + "step": 152 + }, + { + "epoch": 1.0477272727272728, + "grad_norm": 0.41131219267845154, + "learning_rate": 4.942671417961615e-06, + "loss": 0.5514, + "step": 153 + }, + { + "epoch": 1.0545454545454545, + "grad_norm": 0.4733349084854126, + "learning_rate": 4.940496369848795e-06, + "loss": 0.5563, + "step": 154 + }, + { + "epoch": 1.0613636363636363, + "grad_norm": 0.41179028153419495, + "learning_rate": 4.938281322299243e-06, + "loss": 0.5413, + "step": 155 + }, + { + "epoch": 1.0681818181818181, + "grad_norm": 0.4770941138267517, + "learning_rate": 4.936026311617316e-06, + "loss": 0.5561, + "step": 156 + }, + { + "epoch": 1.075, + "grad_norm": 0.7752397060394287, + "learning_rate": 4.933731374762361e-06, + "loss": 0.5439, + "step": 157 + }, + { + "epoch": 1.0818181818181818, + "grad_norm": 0.4323958456516266, + "learning_rate": 4.931396549348115e-06, + "loss": 0.5401, + "step": 158 + }, + { + "epoch": 1.0886363636363636, + "grad_norm": 0.4497937858104706, + "learning_rate": 4.9290218736420795e-06, + "loss": 0.523, + "step": 159 + }, + { + "epoch": 1.0954545454545455, + "grad_norm": 0.4881570339202881, + "learning_rate": 4.926607386564898e-06, + "loss": 0.5441, + "step": 160 + }, + { + "epoch": 1.1022727272727273, + "grad_norm": 0.43357932567596436, + "learning_rate": 4.9241531276897196e-06, + "loss": 0.5499, + "step": 161 + }, + { + "epoch": 1.1090909090909091, + "grad_norm": 1.1125361919403076, + "learning_rate": 4.921659137241544e-06, + "loss": 0.5265, + "step": 162 + }, + { + "epoch": 1.115909090909091, + "grad_norm": 0.43150508403778076, + "learning_rate": 4.919125456096574e-06, + "loss": 0.5193, + "step": 163 + }, + { + "epoch": 1.1227272727272728, + "grad_norm": 0.40771809220314026, + "learning_rate": 4.916552125781529e-06, + "loss": 0.538, + "step": 164 + }, + { + "epoch": 1.1295454545454546, + "grad_norm": 0.5830567479133606, + "learning_rate": 4.913939188472979e-06, + "loss": 0.5548, + "step": 165 + }, + { + "epoch": 1.1363636363636362, + "grad_norm": 0.4776526391506195, + "learning_rate": 4.911286686996648e-06, + "loss": 0.5328, + "step": 166 + }, + { + "epoch": 1.143181818181818, + "grad_norm": 0.41657742857933044, + "learning_rate": 4.908594664826708e-06, + "loss": 0.5336, + "step": 167 + }, + { + "epoch": 1.15, + "grad_norm": 0.4651948809623718, + "learning_rate": 4.905863166085076e-06, + "loss": 0.5812, + "step": 168 + }, + { + "epoch": 1.1568181818181817, + "grad_norm": 0.4399661719799042, + "learning_rate": 4.903092235540679e-06, + "loss": 0.5143, + "step": 169 + }, + { + "epoch": 1.1636363636363636, + "grad_norm": 0.4548935294151306, + "learning_rate": 4.900281918608732e-06, + "loss": 0.5425, + "step": 170 + }, + { + "epoch": 1.1704545454545454, + "grad_norm": 0.43515875935554504, + "learning_rate": 4.897432261349984e-06, + "loss": 0.534, + "step": 171 + }, + { + "epoch": 1.1772727272727272, + "grad_norm": 0.42954760789871216, + "learning_rate": 4.894543310469968e-06, + "loss": 0.5436, + "step": 172 + }, + { + "epoch": 1.184090909090909, + "grad_norm": 0.4063640534877777, + "learning_rate": 4.891615113318236e-06, + "loss": 0.5481, + "step": 173 + }, + { + "epoch": 1.190909090909091, + "grad_norm": 0.49416476488113403, + "learning_rate": 4.888647717887582e-06, + "loss": 0.5298, + "step": 174 + }, + { + "epoch": 1.1977272727272728, + "grad_norm": 1.4082964658737183, + "learning_rate": 4.8856411728132526e-06, + "loss": 0.5366, + "step": 175 + }, + { + "epoch": 1.2045454545454546, + "grad_norm": 0.44356465339660645, + "learning_rate": 4.8825955273721524e-06, + "loss": 0.5563, + "step": 176 + }, + { + "epoch": 1.2113636363636364, + "grad_norm": 0.40925371646881104, + "learning_rate": 4.879510831482039e-06, + "loss": 0.5444, + "step": 177 + }, + { + "epoch": 1.2181818181818183, + "grad_norm": 0.44379451870918274, + "learning_rate": 4.876387135700701e-06, + "loss": 0.5311, + "step": 178 + }, + { + "epoch": 1.225, + "grad_norm": 0.41487446427345276, + "learning_rate": 4.873224491225128e-06, + "loss": 0.5637, + "step": 179 + }, + { + "epoch": 1.231818181818182, + "grad_norm": 0.4520554542541504, + "learning_rate": 4.870022949890676e-06, + "loss": 0.5677, + "step": 180 + }, + { + "epoch": 1.2386363636363638, + "grad_norm": 0.4528166949748993, + "learning_rate": 4.866782564170217e-06, + "loss": 0.5095, + "step": 181 + }, + { + "epoch": 1.2454545454545454, + "grad_norm": 0.4409842789173126, + "learning_rate": 4.863503387173276e-06, + "loss": 0.5423, + "step": 182 + }, + { + "epoch": 1.2522727272727272, + "grad_norm": 0.4355996251106262, + "learning_rate": 4.860185472645161e-06, + "loss": 0.5422, + "step": 183 + }, + { + "epoch": 1.259090909090909, + "grad_norm": 0.4871126115322113, + "learning_rate": 4.856828874966086e-06, + "loss": 0.5293, + "step": 184 + }, + { + "epoch": 1.2659090909090909, + "grad_norm": 0.442348450422287, + "learning_rate": 4.853433649150276e-06, + "loss": 0.5616, + "step": 185 + }, + { + "epoch": 1.2727272727272727, + "grad_norm": 0.44014668464660645, + "learning_rate": 4.849999850845066e-06, + "loss": 0.5354, + "step": 186 + }, + { + "epoch": 1.2795454545454545, + "grad_norm": 0.4496249854564667, + "learning_rate": 4.8465275363299905e-06, + "loss": 0.5298, + "step": 187 + }, + { + "epoch": 1.2863636363636364, + "grad_norm": 0.46060171723365784, + "learning_rate": 4.84301676251586e-06, + "loss": 0.554, + "step": 188 + }, + { + "epoch": 1.2931818181818182, + "grad_norm": 0.7829347252845764, + "learning_rate": 4.839467586943825e-06, + "loss": 0.5606, + "step": 189 + }, + { + "epoch": 1.3, + "grad_norm": 0.45566731691360474, + "learning_rate": 4.835880067784441e-06, + "loss": 0.5118, + "step": 190 + }, + { + "epoch": 1.3068181818181819, + "grad_norm": 0.43211469054222107, + "learning_rate": 4.832254263836708e-06, + "loss": 0.535, + "step": 191 + }, + { + "epoch": 1.3136363636363637, + "grad_norm": 0.4350590705871582, + "learning_rate": 4.828590234527107e-06, + "loss": 0.5595, + "step": 192 + }, + { + "epoch": 1.3204545454545453, + "grad_norm": 0.4056399464607239, + "learning_rate": 4.82488803990863e-06, + "loss": 0.55, + "step": 193 + }, + { + "epoch": 1.3272727272727272, + "grad_norm": 0.4458341598510742, + "learning_rate": 4.821147740659795e-06, + "loss": 0.501, + "step": 194 + }, + { + "epoch": 1.334090909090909, + "grad_norm": 0.446504145860672, + "learning_rate": 4.817369398083648e-06, + "loss": 0.5393, + "step": 195 + }, + { + "epoch": 1.3409090909090908, + "grad_norm": 0.47838401794433594, + "learning_rate": 4.813553074106761e-06, + "loss": 0.5354, + "step": 196 + }, + { + "epoch": 1.3477272727272727, + "grad_norm": 0.4675792157649994, + "learning_rate": 4.809698831278217e-06, + "loss": 0.5202, + "step": 197 + }, + { + "epoch": 1.3545454545454545, + "grad_norm": 0.4351370632648468, + "learning_rate": 4.805806732768585e-06, + "loss": 0.5256, + "step": 198 + }, + { + "epoch": 1.3613636363636363, + "grad_norm": 0.4466511607170105, + "learning_rate": 4.801876842368882e-06, + "loss": 0.5238, + "step": 199 + }, + { + "epoch": 1.3681818181818182, + "grad_norm": 0.45672544836997986, + "learning_rate": 4.797909224489531e-06, + "loss": 0.5426, + "step": 200 + }, + { + "epoch": 1.375, + "grad_norm": 0.422638863325119, + "learning_rate": 4.793903944159303e-06, + "loss": 0.5385, + "step": 201 + }, + { + "epoch": 1.3818181818181818, + "grad_norm": 0.4254551827907562, + "learning_rate": 4.789861067024253e-06, + "loss": 0.503, + "step": 202 + }, + { + "epoch": 1.3886363636363637, + "grad_norm": 0.43860334157943726, + "learning_rate": 4.785780659346642e-06, + "loss": 0.519, + "step": 203 + }, + { + "epoch": 1.3954545454545455, + "grad_norm": 0.43294480443000793, + "learning_rate": 4.781662788003851e-06, + "loss": 0.5537, + "step": 204 + }, + { + "epoch": 1.4022727272727273, + "grad_norm": 0.45006492733955383, + "learning_rate": 4.777507520487289e-06, + "loss": 0.5318, + "step": 205 + }, + { + "epoch": 1.4090909090909092, + "grad_norm": 0.4178854823112488, + "learning_rate": 4.773314924901281e-06, + "loss": 0.5287, + "step": 206 + }, + { + "epoch": 1.415909090909091, + "grad_norm": 0.45662209391593933, + "learning_rate": 4.769085069961955e-06, + "loss": 0.5609, + "step": 207 + }, + { + "epoch": 1.4227272727272728, + "grad_norm": 0.41578733921051025, + "learning_rate": 4.764818024996117e-06, + "loss": 0.5249, + "step": 208 + }, + { + "epoch": 1.4295454545454547, + "grad_norm": 0.48305758833885193, + "learning_rate": 4.760513859940112e-06, + "loss": 0.515, + "step": 209 + }, + { + "epoch": 1.4363636363636363, + "grad_norm": 0.4564506709575653, + "learning_rate": 4.756172645338675e-06, + "loss": 0.535, + "step": 210 + }, + { + "epoch": 1.4431818181818181, + "grad_norm": 0.4499240815639496, + "learning_rate": 4.751794452343785e-06, + "loss": 0.5236, + "step": 211 + }, + { + "epoch": 1.45, + "grad_norm": 0.41851088404655457, + "learning_rate": 4.747379352713489e-06, + "loss": 0.5254, + "step": 212 + }, + { + "epoch": 1.4568181818181818, + "grad_norm": 0.4690937399864197, + "learning_rate": 4.7429274188107275e-06, + "loss": 0.5312, + "step": 213 + }, + { + "epoch": 1.4636363636363636, + "grad_norm": 0.5516958832740784, + "learning_rate": 4.738438723602154e-06, + "loss": 0.5315, + "step": 214 + }, + { + "epoch": 1.4704545454545455, + "grad_norm": 1.2040799856185913, + "learning_rate": 4.733913340656933e-06, + "loss": 0.544, + "step": 215 + }, + { + "epoch": 1.4772727272727273, + "grad_norm": 0.445425808429718, + "learning_rate": 4.729351344145536e-06, + "loss": 0.5309, + "step": 216 + }, + { + "epoch": 1.4840909090909091, + "grad_norm": 0.4186754822731018, + "learning_rate": 4.7247528088385296e-06, + "loss": 0.5138, + "step": 217 + }, + { + "epoch": 1.490909090909091, + "grad_norm": 0.43284520506858826, + "learning_rate": 4.720117810105341e-06, + "loss": 0.521, + "step": 218 + }, + { + "epoch": 1.4977272727272728, + "grad_norm": 0.4308505058288574, + "learning_rate": 4.715446423913036e-06, + "loss": 0.5436, + "step": 219 + }, + { + "epoch": 1.5045454545454544, + "grad_norm": 0.4471941888332367, + "learning_rate": 4.710738726825059e-06, + "loss": 0.5528, + "step": 220 + }, + { + "epoch": 1.5113636363636362, + "grad_norm": 0.4374965727329254, + "learning_rate": 4.705994795999991e-06, + "loss": 0.5238, + "step": 221 + }, + { + "epoch": 1.518181818181818, + "grad_norm": 0.6026428937911987, + "learning_rate": 4.701214709190277e-06, + "loss": 0.5268, + "step": 222 + }, + { + "epoch": 1.525, + "grad_norm": 0.4413348138332367, + "learning_rate": 4.696398544740955e-06, + "loss": 0.5534, + "step": 223 + }, + { + "epoch": 1.5318181818181817, + "grad_norm": 0.45917898416519165, + "learning_rate": 4.69154638158837e-06, + "loss": 0.511, + "step": 224 + }, + { + "epoch": 1.5386363636363636, + "grad_norm": 0.5339583158493042, + "learning_rate": 4.686658299258881e-06, + "loss": 0.5395, + "step": 225 + }, + { + "epoch": 1.5454545454545454, + "grad_norm": 0.4617757201194763, + "learning_rate": 4.681734377867562e-06, + "loss": 0.5425, + "step": 226 + }, + { + "epoch": 1.5522727272727272, + "grad_norm": 0.4513031840324402, + "learning_rate": 4.67677469811688e-06, + "loss": 0.5241, + "step": 227 + }, + { + "epoch": 1.559090909090909, + "grad_norm": 0.43548423051834106, + "learning_rate": 4.671779341295378e-06, + "loss": 0.5224, + "step": 228 + }, + { + "epoch": 1.565909090909091, + "grad_norm": 0.5516494512557983, + "learning_rate": 4.666748389276344e-06, + "loss": 0.5346, + "step": 229 + }, + { + "epoch": 1.5727272727272728, + "grad_norm": 0.4441164433956146, + "learning_rate": 4.661681924516466e-06, + "loss": 0.5208, + "step": 230 + }, + { + "epoch": 1.5795454545454546, + "grad_norm": 0.4057452380657196, + "learning_rate": 4.6565800300544805e-06, + "loss": 0.5173, + "step": 231 + }, + { + "epoch": 1.5863636363636364, + "grad_norm": 0.4162004590034485, + "learning_rate": 4.651442789509813e-06, + "loss": 0.5307, + "step": 232 + }, + { + "epoch": 1.5931818181818183, + "grad_norm": 0.43314146995544434, + "learning_rate": 4.646270287081208e-06, + "loss": 0.5359, + "step": 233 + }, + { + "epoch": 1.6, + "grad_norm": 0.5786082148551941, + "learning_rate": 4.641062607545347e-06, + "loss": 0.5307, + "step": 234 + }, + { + "epoch": 1.606818181818182, + "grad_norm": 0.43811941146850586, + "learning_rate": 4.6358198362554585e-06, + "loss": 0.5556, + "step": 235 + }, + { + "epoch": 1.6136363636363638, + "grad_norm": 0.4374074935913086, + "learning_rate": 4.630542059139923e-06, + "loss": 0.5253, + "step": 236 + }, + { + "epoch": 1.6204545454545456, + "grad_norm": 0.4387815594673157, + "learning_rate": 4.625229362700863e-06, + "loss": 0.534, + "step": 237 + }, + { + "epoch": 1.6272727272727274, + "grad_norm": 0.4976162016391754, + "learning_rate": 4.61988183401272e-06, + "loss": 0.5255, + "step": 238 + }, + { + "epoch": 1.634090909090909, + "grad_norm": 0.4603639543056488, + "learning_rate": 4.614499560720837e-06, + "loss": 0.511, + "step": 239 + }, + { + "epoch": 1.6409090909090909, + "grad_norm": 0.43136921525001526, + "learning_rate": 4.609082631040012e-06, + "loss": 0.5099, + "step": 240 + }, + { + "epoch": 1.6477272727272727, + "grad_norm": 0.4573090970516205, + "learning_rate": 4.603631133753061e-06, + "loss": 0.5515, + "step": 241 + }, + { + "epoch": 1.6545454545454545, + "grad_norm": 0.4292807877063751, + "learning_rate": 4.598145158209356e-06, + "loss": 0.5048, + "step": 242 + }, + { + "epoch": 1.6613636363636364, + "grad_norm": 0.6377942562103271, + "learning_rate": 4.592624794323366e-06, + "loss": 0.5333, + "step": 243 + }, + { + "epoch": 1.6681818181818182, + "grad_norm": 0.49924615025520325, + "learning_rate": 4.587070132573178e-06, + "loss": 0.5127, + "step": 244 + }, + { + "epoch": 1.675, + "grad_norm": 0.4723946154117584, + "learning_rate": 4.581481263999019e-06, + "loss": 0.5127, + "step": 245 + }, + { + "epoch": 1.6818181818181817, + "grad_norm": 1.1846898794174194, + "learning_rate": 4.575858280201761e-06, + "loss": 0.5367, + "step": 246 + }, + { + "epoch": 1.6886363636363635, + "grad_norm": 0.4306166172027588, + "learning_rate": 4.570201273341418e-06, + "loss": 0.52, + "step": 247 + }, + { + "epoch": 1.6954545454545453, + "grad_norm": 0.45308929681777954, + "learning_rate": 4.564510336135642e-06, + "loss": 0.5589, + "step": 248 + }, + { + "epoch": 1.7022727272727272, + "grad_norm": 0.4482448399066925, + "learning_rate": 4.558785561858196e-06, + "loss": 0.5161, + "step": 249 + }, + { + "epoch": 1.709090909090909, + "grad_norm": 0.44250935316085815, + "learning_rate": 4.5530270443374305e-06, + "loss": 0.5342, + "step": 250 + }, + { + "epoch": 1.7159090909090908, + "grad_norm": 0.8177540898323059, + "learning_rate": 4.547234877954741e-06, + "loss": 0.5402, + "step": 251 + }, + { + "epoch": 1.7227272727272727, + "grad_norm": 0.5169854760169983, + "learning_rate": 4.541409157643027e-06, + "loss": 0.5301, + "step": 252 + }, + { + "epoch": 1.7295454545454545, + "grad_norm": 0.4544285237789154, + "learning_rate": 4.535549978885132e-06, + "loss": 0.5314, + "step": 253 + }, + { + "epoch": 1.7363636363636363, + "grad_norm": 0.42860639095306396, + "learning_rate": 4.5296574377122765e-06, + "loss": 0.5249, + "step": 254 + }, + { + "epoch": 1.7431818181818182, + "grad_norm": 0.526492178440094, + "learning_rate": 4.5237316307024895e-06, + "loss": 0.5142, + "step": 255 + }, + { + "epoch": 1.75, + "grad_norm": 0.4208003878593445, + "learning_rate": 4.517772654979024e-06, + "loss": 0.5151, + "step": 256 + }, + { + "epoch": 1.7568181818181818, + "grad_norm": 0.4974212944507599, + "learning_rate": 4.51178060820876e-06, + "loss": 0.5161, + "step": 257 + }, + { + "epoch": 1.7636363636363637, + "grad_norm": 0.4148060083389282, + "learning_rate": 4.505755588600613e-06, + "loss": 0.5099, + "step": 258 + }, + { + "epoch": 1.7704545454545455, + "grad_norm": 0.4106168746948242, + "learning_rate": 4.499697694903915e-06, + "loss": 0.5376, + "step": 259 + }, + { + "epoch": 1.7772727272727273, + "grad_norm": 0.49017220735549927, + "learning_rate": 4.493607026406802e-06, + "loss": 0.525, + "step": 260 + }, + { + "epoch": 1.7840909090909092, + "grad_norm": 0.4141218662261963, + "learning_rate": 4.487483682934587e-06, + "loss": 0.5303, + "step": 261 + }, + { + "epoch": 1.790909090909091, + "grad_norm": 0.44826585054397583, + "learning_rate": 4.481327764848118e-06, + "loss": 0.4886, + "step": 262 + }, + { + "epoch": 1.7977272727272728, + "grad_norm": 0.42013809084892273, + "learning_rate": 4.47513937304214e-06, + "loss": 0.5204, + "step": 263 + }, + { + "epoch": 1.8045454545454547, + "grad_norm": 0.44327741861343384, + "learning_rate": 4.4689186089436365e-06, + "loss": 0.5097, + "step": 264 + }, + { + "epoch": 1.8113636363636365, + "grad_norm": 0.39498746395111084, + "learning_rate": 4.462665574510169e-06, + "loss": 0.5102, + "step": 265 + }, + { + "epoch": 1.8181818181818183, + "grad_norm": 0.48296087980270386, + "learning_rate": 4.456380372228208e-06, + "loss": 0.5299, + "step": 266 + }, + { + "epoch": 1.825, + "grad_norm": 0.4398634433746338, + "learning_rate": 4.450063105111447e-06, + "loss": 0.5419, + "step": 267 + }, + { + "epoch": 1.8318181818181818, + "grad_norm": 0.43333035707473755, + "learning_rate": 4.443713876699124e-06, + "loss": 0.5047, + "step": 268 + }, + { + "epoch": 1.8386363636363636, + "grad_norm": 0.5823659300804138, + "learning_rate": 4.4373327910543125e-06, + "loss": 0.5082, + "step": 269 + }, + { + "epoch": 1.8454545454545455, + "grad_norm": 0.44385817646980286, + "learning_rate": 4.430919952762226e-06, + "loss": 0.5171, + "step": 270 + }, + { + "epoch": 1.8522727272727273, + "grad_norm": 0.4444347023963928, + "learning_rate": 4.424475466928499e-06, + "loss": 0.5351, + "step": 271 + }, + { + "epoch": 1.8590909090909091, + "grad_norm": 0.549674391746521, + "learning_rate": 4.417999439177465e-06, + "loss": 0.498, + "step": 272 + }, + { + "epoch": 1.865909090909091, + "grad_norm": 0.4340274930000305, + "learning_rate": 4.4114919756504275e-06, + "loss": 0.5429, + "step": 273 + }, + { + "epoch": 1.8727272727272726, + "grad_norm": 0.44019854068756104, + "learning_rate": 4.404953183003916e-06, + "loss": 0.5133, + "step": 274 + }, + { + "epoch": 1.8795454545454544, + "grad_norm": 0.4473486542701721, + "learning_rate": 4.398383168407941e-06, + "loss": 0.5413, + "step": 275 + }, + { + "epoch": 1.8863636363636362, + "grad_norm": 0.44450780749320984, + "learning_rate": 4.391782039544239e-06, + "loss": 0.4998, + "step": 276 + }, + { + "epoch": 1.893181818181818, + "grad_norm": 0.4310123920440674, + "learning_rate": 4.385149904604502e-06, + "loss": 0.5134, + "step": 277 + }, + { + "epoch": 1.9, + "grad_norm": 0.48953917622566223, + "learning_rate": 4.378486872288611e-06, + "loss": 0.5102, + "step": 278 + }, + { + "epoch": 1.9068181818181817, + "grad_norm": 0.49219778180122375, + "learning_rate": 4.371793051802849e-06, + "loss": 0.5213, + "step": 279 + }, + { + "epoch": 1.9136363636363636, + "grad_norm": 0.818645715713501, + "learning_rate": 4.365068552858116e-06, + "loss": 0.5251, + "step": 280 + }, + { + "epoch": 1.9204545454545454, + "grad_norm": 0.4407028257846832, + "learning_rate": 4.358313485668124e-06, + "loss": 0.5191, + "step": 281 + }, + { + "epoch": 1.9272727272727272, + "grad_norm": 0.5301506519317627, + "learning_rate": 4.3515279609476e-06, + "loss": 0.5003, + "step": 282 + }, + { + "epoch": 1.934090909090909, + "grad_norm": 0.4673819839954376, + "learning_rate": 4.3447120899104615e-06, + "loss": 0.521, + "step": 283 + }, + { + "epoch": 1.940909090909091, + "grad_norm": 0.4106168746948242, + "learning_rate": 4.337865984268002e-06, + "loss": 0.5199, + "step": 284 + }, + { + "epoch": 1.9477272727272728, + "grad_norm": 0.43583595752716064, + "learning_rate": 4.3309897562270525e-06, + "loss": 0.5367, + "step": 285 + }, + { + "epoch": 1.9545454545454546, + "grad_norm": 0.4984910488128662, + "learning_rate": 4.324083518488151e-06, + "loss": 0.5202, + "step": 286 + }, + { + "epoch": 1.9613636363636364, + "grad_norm": 0.4516090452671051, + "learning_rate": 4.317147384243688e-06, + "loss": 0.5432, + "step": 287 + }, + { + "epoch": 1.9681818181818183, + "grad_norm": 0.43753015995025635, + "learning_rate": 4.3101814671760546e-06, + "loss": 0.5309, + "step": 288 + }, + { + "epoch": 1.975, + "grad_norm": 0.4472252130508423, + "learning_rate": 4.303185881455778e-06, + "loss": 0.5344, + "step": 289 + }, + { + "epoch": 1.981818181818182, + "grad_norm": 1.178377389907837, + "learning_rate": 4.296160741739652e-06, + "loss": 0.5283, + "step": 290 + }, + { + "epoch": 1.9886363636363638, + "grad_norm": 0.5085738301277161, + "learning_rate": 4.289106163168858e-06, + "loss": 0.5133, + "step": 291 + }, + { + "epoch": 1.9954545454545456, + "grad_norm": 0.52077716588974, + "learning_rate": 4.282022261367074e-06, + "loss": 0.5155, + "step": 292 + }, + { + "epoch": 2.006818181818182, + "grad_norm": 0.8534443378448486, + "learning_rate": 4.274909152438582e-06, + "loss": 1.003, + "step": 293 + }, + { + "epoch": 2.0136363636363637, + "grad_norm": 0.46160876750946045, + "learning_rate": 4.267766952966369e-06, + "loss": 0.4958, + "step": 294 + }, + { + "epoch": 2.0204545454545455, + "grad_norm": 0.5113462209701538, + "learning_rate": 4.260595780010209e-06, + "loss": 0.4895, + "step": 295 + }, + { + "epoch": 2.0272727272727273, + "grad_norm": 0.4553380608558655, + "learning_rate": 4.2533957511047485e-06, + "loss": 0.4826, + "step": 296 + }, + { + "epoch": 2.034090909090909, + "grad_norm": 0.43557822704315186, + "learning_rate": 4.24616698425758e-06, + "loss": 0.4591, + "step": 297 + }, + { + "epoch": 2.040909090909091, + "grad_norm": 0.45537200570106506, + "learning_rate": 4.238909597947307e-06, + "loss": 0.4811, + "step": 298 + }, + { + "epoch": 2.047727272727273, + "grad_norm": 0.48833370208740234, + "learning_rate": 4.231623711121603e-06, + "loss": 0.4967, + "step": 299 + }, + { + "epoch": 2.0545454545454547, + "grad_norm": 0.45504844188690186, + "learning_rate": 4.224309443195261e-06, + "loss": 0.4964, + "step": 300 + }, + { + "epoch": 2.0613636363636365, + "grad_norm": 0.4313275218009949, + "learning_rate": 4.2169669140482365e-06, + "loss": 0.4832, + "step": 301 + }, + { + "epoch": 2.0681818181818183, + "grad_norm": 0.4298369884490967, + "learning_rate": 4.2095962440236846e-06, + "loss": 0.5071, + "step": 302 + }, + { + "epoch": 2.075, + "grad_norm": 0.4188523590564728, + "learning_rate": 4.202197553925983e-06, + "loss": 0.4859, + "step": 303 + }, + { + "epoch": 2.081818181818182, + "grad_norm": 0.39195290207862854, + "learning_rate": 4.194770965018758e-06, + "loss": 0.5081, + "step": 304 + }, + { + "epoch": 2.088636363636364, + "grad_norm": 0.45844170451164246, + "learning_rate": 4.187316599022892e-06, + "loss": 0.508, + "step": 305 + }, + { + "epoch": 2.0954545454545457, + "grad_norm": 0.6743896007537842, + "learning_rate": 4.179834578114531e-06, + "loss": 0.4748, + "step": 306 + }, + { + "epoch": 2.102272727272727, + "grad_norm": 0.4317929446697235, + "learning_rate": 4.172325024923083e-06, + "loss": 0.504, + "step": 307 + }, + { + "epoch": 2.109090909090909, + "grad_norm": 0.4857175350189209, + "learning_rate": 4.164788062529203e-06, + "loss": 0.4987, + "step": 308 + }, + { + "epoch": 2.1159090909090907, + "grad_norm": 0.4454379677772522, + "learning_rate": 4.157223814462784e-06, + "loss": 0.4938, + "step": 309 + }, + { + "epoch": 2.1227272727272726, + "grad_norm": 0.5128078460693359, + "learning_rate": 4.149632404700925e-06, + "loss": 0.4917, + "step": 310 + }, + { + "epoch": 2.1295454545454544, + "grad_norm": 0.5195722579956055, + "learning_rate": 4.142013957665903e-06, + "loss": 0.4753, + "step": 311 + }, + { + "epoch": 2.1363636363636362, + "grad_norm": 0.4242931604385376, + "learning_rate": 4.134368598223132e-06, + "loss": 0.4832, + "step": 312 + }, + { + "epoch": 2.143181818181818, + "grad_norm": 0.4419071674346924, + "learning_rate": 4.126696451679116e-06, + "loss": 0.4783, + "step": 313 + }, + { + "epoch": 2.15, + "grad_norm": 0.42852380871772766, + "learning_rate": 4.118997643779401e-06, + "loss": 0.5034, + "step": 314 + }, + { + "epoch": 2.1568181818181817, + "grad_norm": 0.7373576164245605, + "learning_rate": 4.111272300706502e-06, + "loss": 0.4934, + "step": 315 + }, + { + "epoch": 2.1636363636363636, + "grad_norm": 0.4435337781906128, + "learning_rate": 4.1035205490778505e-06, + "loss": 0.5089, + "step": 316 + }, + { + "epoch": 2.1704545454545454, + "grad_norm": 0.4714502692222595, + "learning_rate": 4.095742515943703e-06, + "loss": 0.502, + "step": 317 + }, + { + "epoch": 2.1772727272727272, + "grad_norm": 0.4732494056224823, + "learning_rate": 4.087938328785071e-06, + "loss": 0.5073, + "step": 318 + }, + { + "epoch": 2.184090909090909, + "grad_norm": 0.45177504420280457, + "learning_rate": 4.080108115511629e-06, + "loss": 0.5076, + "step": 319 + }, + { + "epoch": 2.190909090909091, + "grad_norm": 0.506157398223877, + "learning_rate": 4.072252004459612e-06, + "loss": 0.4837, + "step": 320 + }, + { + "epoch": 2.1977272727272728, + "grad_norm": 0.43044376373291016, + "learning_rate": 4.064370124389718e-06, + "loss": 0.5167, + "step": 321 + }, + { + "epoch": 2.2045454545454546, + "grad_norm": 0.45878180861473083, + "learning_rate": 4.056462604484998e-06, + "loss": 0.4936, + "step": 322 + }, + { + "epoch": 2.2113636363636364, + "grad_norm": 0.43078047037124634, + "learning_rate": 4.048529574348734e-06, + "loss": 0.4783, + "step": 323 + }, + { + "epoch": 2.2181818181818183, + "grad_norm": 0.42789366841316223, + "learning_rate": 4.040571164002319e-06, + "loss": 0.4853, + "step": 324 + }, + { + "epoch": 2.225, + "grad_norm": 0.44888222217559814, + "learning_rate": 4.032587503883124e-06, + "loss": 0.5097, + "step": 325 + }, + { + "epoch": 2.231818181818182, + "grad_norm": 0.446851521730423, + "learning_rate": 4.0245787248423614e-06, + "loss": 0.4844, + "step": 326 + }, + { + "epoch": 2.2386363636363638, + "grad_norm": 0.436567097902298, + "learning_rate": 4.0165449581429404e-06, + "loss": 0.4549, + "step": 327 + }, + { + "epoch": 2.2454545454545456, + "grad_norm": 0.4813944697380066, + "learning_rate": 4.008486335457312e-06, + "loss": 0.49, + "step": 328 + }, + { + "epoch": 2.2522727272727274, + "grad_norm": 0.4935426115989685, + "learning_rate": 4.000402988865316e-06, + "loss": 0.4995, + "step": 329 + }, + { + "epoch": 2.2590909090909093, + "grad_norm": 1.1695573329925537, + "learning_rate": 3.992295050852013e-06, + "loss": 0.5033, + "step": 330 + }, + { + "epoch": 2.265909090909091, + "grad_norm": 0.41830316185951233, + "learning_rate": 3.984162654305516e-06, + "loss": 0.4828, + "step": 331 + }, + { + "epoch": 2.2727272727272725, + "grad_norm": 0.4522007703781128, + "learning_rate": 3.976005932514807e-06, + "loss": 0.4948, + "step": 332 + }, + { + "epoch": 2.2795454545454543, + "grad_norm": 0.8304370045661926, + "learning_rate": 3.967825019167559e-06, + "loss": 0.504, + "step": 333 + }, + { + "epoch": 2.286363636363636, + "grad_norm": 0.4533236622810364, + "learning_rate": 3.959620048347938e-06, + "loss": 0.4487, + "step": 334 + }, + { + "epoch": 2.293181818181818, + "grad_norm": 0.43578940629959106, + "learning_rate": 3.951391154534415e-06, + "loss": 0.4702, + "step": 335 + }, + { + "epoch": 2.3, + "grad_norm": 0.5146592259407043, + "learning_rate": 3.943138472597549e-06, + "loss": 0.4984, + "step": 336 + }, + { + "epoch": 2.3068181818181817, + "grad_norm": 0.6412249207496643, + "learning_rate": 3.934862137797788e-06, + "loss": 0.4761, + "step": 337 + }, + { + "epoch": 2.3136363636363635, + "grad_norm": 1.3352930545806885, + "learning_rate": 3.9265622857832455e-06, + "loss": 0.499, + "step": 338 + }, + { + "epoch": 2.3204545454545453, + "grad_norm": 0.4465159475803375, + "learning_rate": 3.918239052587481e-06, + "loss": 0.4525, + "step": 339 + }, + { + "epoch": 2.327272727272727, + "grad_norm": 0.43597185611724854, + "learning_rate": 3.909892574627267e-06, + "loss": 0.4952, + "step": 340 + }, + { + "epoch": 2.334090909090909, + "grad_norm": 0.7057136297225952, + "learning_rate": 3.901522988700355e-06, + "loss": 0.4576, + "step": 341 + }, + { + "epoch": 2.340909090909091, + "grad_norm": 0.45416319370269775, + "learning_rate": 3.893130431983234e-06, + "loss": 0.4902, + "step": 342 + }, + { + "epoch": 2.3477272727272727, + "grad_norm": 0.43900424242019653, + "learning_rate": 3.884715042028882e-06, + "loss": 0.495, + "step": 343 + }, + { + "epoch": 2.3545454545454545, + "grad_norm": 0.4553733468055725, + "learning_rate": 3.876276956764509e-06, + "loss": 0.4861, + "step": 344 + }, + { + "epoch": 2.3613636363636363, + "grad_norm": 0.5471370816230774, + "learning_rate": 3.867816314489301e-06, + "loss": 0.5043, + "step": 345 + }, + { + "epoch": 2.368181818181818, + "grad_norm": 0.43938082456588745, + "learning_rate": 3.8593332538721465e-06, + "loss": 0.4678, + "step": 346 + }, + { + "epoch": 2.375, + "grad_norm": 0.4583636224269867, + "learning_rate": 3.8508279139493736e-06, + "loss": 0.4679, + "step": 347 + }, + { + "epoch": 2.381818181818182, + "grad_norm": 0.44170665740966797, + "learning_rate": 3.84230043412246e-06, + "loss": 0.4647, + "step": 348 + }, + { + "epoch": 2.3886363636363637, + "grad_norm": 0.4952428340911865, + "learning_rate": 3.833750954155757e-06, + "loss": 0.4796, + "step": 349 + }, + { + "epoch": 2.3954545454545455, + "grad_norm": 0.42052993178367615, + "learning_rate": 3.825179614174195e-06, + "loss": 0.4606, + "step": 350 + }, + { + "epoch": 2.4022727272727273, + "grad_norm": 0.5792545676231384, + "learning_rate": 3.816586554660987e-06, + "loss": 0.4805, + "step": 351 + }, + { + "epoch": 2.409090909090909, + "grad_norm": 0.4477898180484772, + "learning_rate": 3.807971916455325e-06, + "loss": 0.4793, + "step": 352 + }, + { + "epoch": 2.415909090909091, + "grad_norm": 0.45617541670799255, + "learning_rate": 3.799335840750077e-06, + "loss": 0.502, + "step": 353 + }, + { + "epoch": 2.422727272727273, + "grad_norm": 0.45021289587020874, + "learning_rate": 3.790678469089465e-06, + "loss": 0.478, + "step": 354 + }, + { + "epoch": 2.4295454545454547, + "grad_norm": 0.4506435990333557, + "learning_rate": 3.7819999433667503e-06, + "loss": 0.4828, + "step": 355 + }, + { + "epoch": 2.4363636363636365, + "grad_norm": 0.48021042346954346, + "learning_rate": 3.773300405821908e-06, + "loss": 0.4827, + "step": 356 + }, + { + "epoch": 2.4431818181818183, + "grad_norm": 0.4759495258331299, + "learning_rate": 3.764579999039293e-06, + "loss": 0.4633, + "step": 357 + }, + { + "epoch": 2.45, + "grad_norm": 0.4734678268432617, + "learning_rate": 3.7558388659453052e-06, + "loss": 0.4982, + "step": 358 + }, + { + "epoch": 2.456818181818182, + "grad_norm": 0.4502730369567871, + "learning_rate": 3.7470771498060455e-06, + "loss": 0.5009, + "step": 359 + }, + { + "epoch": 2.463636363636364, + "grad_norm": 0.444940447807312, + "learning_rate": 3.7382949942249695e-06, + "loss": 0.4994, + "step": 360 + }, + { + "epoch": 2.4704545454545457, + "grad_norm": 0.4882519245147705, + "learning_rate": 3.7294925431405306e-06, + "loss": 0.4607, + "step": 361 + }, + { + "epoch": 2.4772727272727275, + "grad_norm": 0.51338791847229, + "learning_rate": 3.720669940823827e-06, + "loss": 0.4801, + "step": 362 + }, + { + "epoch": 2.484090909090909, + "grad_norm": 0.495211124420166, + "learning_rate": 3.7118273318762275e-06, + "loss": 0.4984, + "step": 363 + }, + { + "epoch": 2.4909090909090907, + "grad_norm": 0.4894484281539917, + "learning_rate": 3.702964861227013e-06, + "loss": 0.4834, + "step": 364 + }, + { + "epoch": 2.4977272727272726, + "grad_norm": 0.5554991364479065, + "learning_rate": 3.694082674130991e-06, + "loss": 0.4874, + "step": 365 + }, + { + "epoch": 2.5045454545454544, + "grad_norm": 0.4586203396320343, + "learning_rate": 3.6851809161661206e-06, + "loss": 0.5117, + "step": 366 + }, + { + "epoch": 2.5113636363636362, + "grad_norm": 0.7342172861099243, + "learning_rate": 3.6762597332311254e-06, + "loss": 0.5184, + "step": 367 + }, + { + "epoch": 2.518181818181818, + "grad_norm": 0.46638190746307373, + "learning_rate": 3.6673192715431016e-06, + "loss": 0.47, + "step": 368 + }, + { + "epoch": 2.525, + "grad_norm": 0.4584850072860718, + "learning_rate": 3.658359677635122e-06, + "loss": 0.45, + "step": 369 + }, + { + "epoch": 2.5318181818181817, + "grad_norm": 0.4285255968570709, + "learning_rate": 3.649381098353834e-06, + "loss": 0.513, + "step": 370 + }, + { + "epoch": 2.5386363636363636, + "grad_norm": 0.49487683176994324, + "learning_rate": 3.6403836808570512e-06, + "loss": 0.5048, + "step": 371 + }, + { + "epoch": 2.5454545454545454, + "grad_norm": 0.44076815247535706, + "learning_rate": 3.631367572611348e-06, + "loss": 0.4927, + "step": 372 + }, + { + "epoch": 2.5522727272727272, + "grad_norm": 0.44529926776885986, + "learning_rate": 3.6223329213896313e-06, + "loss": 0.4863, + "step": 373 + }, + { + "epoch": 2.559090909090909, + "grad_norm": 0.5416815876960754, + "learning_rate": 3.613279875268731e-06, + "loss": 0.4808, + "step": 374 + }, + { + "epoch": 2.565909090909091, + "grad_norm": 0.43630415201187134, + "learning_rate": 3.604208582626964e-06, + "loss": 0.4792, + "step": 375 + }, + { + "epoch": 2.5727272727272728, + "grad_norm": 0.47734326124191284, + "learning_rate": 3.5951191921417063e-06, + "loss": 0.4462, + "step": 376 + }, + { + "epoch": 2.5795454545454546, + "grad_norm": 0.4540305435657501, + "learning_rate": 3.586011852786955e-06, + "loss": 0.4915, + "step": 377 + }, + { + "epoch": 2.5863636363636364, + "grad_norm": 0.47168779373168945, + "learning_rate": 3.5768867138308872e-06, + "loss": 0.47, + "step": 378 + }, + { + "epoch": 2.5931818181818183, + "grad_norm": 0.46372920274734497, + "learning_rate": 3.5677439248334133e-06, + "loss": 0.4775, + "step": 379 + }, + { + "epoch": 2.6, + "grad_norm": 0.41653329133987427, + "learning_rate": 3.5585836356437266e-06, + "loss": 0.4875, + "step": 380 + }, + { + "epoch": 2.606818181818182, + "grad_norm": 0.4495103657245636, + "learning_rate": 3.5494059963978433e-06, + "loss": 0.4637, + "step": 381 + }, + { + "epoch": 2.6136363636363638, + "grad_norm": 0.432577520608902, + "learning_rate": 3.540211157516149e-06, + "loss": 0.4991, + "step": 382 + }, + { + "epoch": 2.6204545454545456, + "grad_norm": 0.4517219662666321, + "learning_rate": 3.530999269700927e-06, + "loss": 0.4785, + "step": 383 + }, + { + "epoch": 2.6272727272727274, + "grad_norm": 0.519922137260437, + "learning_rate": 3.521770483933891e-06, + "loss": 0.5026, + "step": 384 + }, + { + "epoch": 2.634090909090909, + "grad_norm": 0.48750364780426025, + "learning_rate": 3.5125249514737093e-06, + "loss": 0.4805, + "step": 385 + }, + { + "epoch": 2.6409090909090907, + "grad_norm": 0.6416886448860168, + "learning_rate": 3.503262823853527e-06, + "loss": 0.4849, + "step": 386 + }, + { + "epoch": 2.6477272727272725, + "grad_norm": 0.5748171806335449, + "learning_rate": 3.493984252878483e-06, + "loss": 0.4885, + "step": 387 + }, + { + "epoch": 2.6545454545454543, + "grad_norm": 0.5148530006408691, + "learning_rate": 3.484689390623218e-06, + "loss": 0.484, + "step": 388 + }, + { + "epoch": 2.661363636363636, + "grad_norm": 0.526839017868042, + "learning_rate": 3.4753783894293886e-06, + "loss": 0.4758, + "step": 389 + }, + { + "epoch": 2.668181818181818, + "grad_norm": 0.7565727829933167, + "learning_rate": 3.466051401903162e-06, + "loss": 0.479, + "step": 390 + }, + { + "epoch": 2.675, + "grad_norm": 0.4598790109157562, + "learning_rate": 3.4567085809127247e-06, + "loss": 0.4972, + "step": 391 + }, + { + "epoch": 2.6818181818181817, + "grad_norm": 0.4911792278289795, + "learning_rate": 3.4473500795857674e-06, + "loss": 0.4714, + "step": 392 + }, + { + "epoch": 2.6886363636363635, + "grad_norm": 0.4214136600494385, + "learning_rate": 3.4379760513069804e-06, + "loss": 0.4906, + "step": 393 + }, + { + "epoch": 2.6954545454545453, + "grad_norm": 0.5070644021034241, + "learning_rate": 3.428586649715542e-06, + "loss": 0.4936, + "step": 394 + }, + { + "epoch": 2.702272727272727, + "grad_norm": 0.4591734707355499, + "learning_rate": 3.4191820287025916e-06, + "loss": 0.4751, + "step": 395 + }, + { + "epoch": 2.709090909090909, + "grad_norm": 0.579781711101532, + "learning_rate": 3.4097623424087196e-06, + "loss": 0.4935, + "step": 396 + }, + { + "epoch": 2.715909090909091, + "grad_norm": 0.4178840219974518, + "learning_rate": 3.4003277452214284e-06, + "loss": 0.4771, + "step": 397 + }, + { + "epoch": 2.7227272727272727, + "grad_norm": 0.5002609491348267, + "learning_rate": 3.3908783917726123e-06, + "loss": 0.4739, + "step": 398 + }, + { + "epoch": 2.7295454545454545, + "grad_norm": 0.5176806449890137, + "learning_rate": 3.381414436936018e-06, + "loss": 0.4965, + "step": 399 + }, + { + "epoch": 2.7363636363636363, + "grad_norm": 0.8039212226867676, + "learning_rate": 3.3719360358247054e-06, + "loss": 0.4822, + "step": 400 + }, + { + "epoch": 2.743181818181818, + "grad_norm": 0.48943471908569336, + "learning_rate": 3.36244334378851e-06, + "loss": 0.4815, + "step": 401 + }, + { + "epoch": 2.75, + "grad_norm": 0.43710342049598694, + "learning_rate": 3.3529365164114903e-06, + "loss": 0.4932, + "step": 402 + }, + { + "epoch": 2.756818181818182, + "grad_norm": 0.5585265159606934, + "learning_rate": 3.3434157095093846e-06, + "loss": 0.5125, + "step": 403 + }, + { + "epoch": 2.7636363636363637, + "grad_norm": 0.4626101553440094, + "learning_rate": 3.333881079127052e-06, + "loss": 0.4934, + "step": 404 + }, + { + "epoch": 2.7704545454545455, + "grad_norm": 0.4675556421279907, + "learning_rate": 3.3243327815359168e-06, + "loss": 0.4903, + "step": 405 + }, + { + "epoch": 2.7772727272727273, + "grad_norm": 0.45658671855926514, + "learning_rate": 3.314770973231408e-06, + "loss": 0.4778, + "step": 406 + }, + { + "epoch": 2.784090909090909, + "grad_norm": 0.4911910593509674, + "learning_rate": 3.305195810930393e-06, + "loss": 0.4795, + "step": 407 + }, + { + "epoch": 2.790909090909091, + "grad_norm": 0.4397219121456146, + "learning_rate": 3.2956074515686105e-06, + "loss": 0.4589, + "step": 408 + }, + { + "epoch": 2.797727272727273, + "grad_norm": 0.5184643864631653, + "learning_rate": 3.2860060522980945e-06, + "loss": 0.4827, + "step": 409 + }, + { + "epoch": 2.8045454545454547, + "grad_norm": 0.45334312319755554, + "learning_rate": 3.276391770484606e-06, + "loss": 0.4821, + "step": 410 + }, + { + "epoch": 2.8113636363636365, + "grad_norm": 0.41846391558647156, + "learning_rate": 3.266764763705046e-06, + "loss": 0.5157, + "step": 411 + }, + { + "epoch": 2.8181818181818183, + "grad_norm": 0.45868799090385437, + "learning_rate": 3.257125189744877e-06, + "loss": 0.4876, + "step": 412 + }, + { + "epoch": 2.825, + "grad_norm": 0.4785497486591339, + "learning_rate": 3.247473206595536e-06, + "loss": 0.4809, + "step": 413 + }, + { + "epoch": 2.831818181818182, + "grad_norm": 0.5683679580688477, + "learning_rate": 3.2378089724518464e-06, + "loss": 0.505, + "step": 414 + }, + { + "epoch": 2.838636363636364, + "grad_norm": 0.48336583375930786, + "learning_rate": 3.228132645709421e-06, + "loss": 0.4816, + "step": 415 + }, + { + "epoch": 2.8454545454545457, + "grad_norm": 0.4236341416835785, + "learning_rate": 3.218444384962071e-06, + "loss": 0.4885, + "step": 416 + }, + { + "epoch": 2.8522727272727275, + "grad_norm": 0.4892752766609192, + "learning_rate": 3.2087443489992043e-06, + "loss": 0.4782, + "step": 417 + }, + { + "epoch": 2.8590909090909093, + "grad_norm": 0.4632202088832855, + "learning_rate": 3.1990326968032225e-06, + "loss": 0.5211, + "step": 418 + }, + { + "epoch": 2.865909090909091, + "grad_norm": 0.8842809796333313, + "learning_rate": 3.189309587546917e-06, + "loss": 0.4716, + "step": 419 + }, + { + "epoch": 2.8727272727272726, + "grad_norm": 0.46771398186683655, + "learning_rate": 3.1795751805908578e-06, + "loss": 0.4745, + "step": 420 + }, + { + "epoch": 2.8795454545454544, + "grad_norm": 0.8178832530975342, + "learning_rate": 3.169829635480783e-06, + "loss": 0.4917, + "step": 421 + }, + { + "epoch": 2.8863636363636362, + "grad_norm": 0.43057572841644287, + "learning_rate": 3.160073111944983e-06, + "loss": 0.4926, + "step": 422 + }, + { + "epoch": 2.893181818181818, + "grad_norm": 0.4258928894996643, + "learning_rate": 3.150305769891686e-06, + "loss": 0.4698, + "step": 423 + }, + { + "epoch": 2.9, + "grad_norm": 0.4559054374694824, + "learning_rate": 3.1405277694064306e-06, + "loss": 0.4537, + "step": 424 + }, + { + "epoch": 2.9068181818181817, + "grad_norm": 0.4342961609363556, + "learning_rate": 3.13073927074945e-06, + "loss": 0.4834, + "step": 425 + }, + { + "epoch": 2.9136363636363636, + "grad_norm": 0.48813825845718384, + "learning_rate": 3.1209404343530374e-06, + "loss": 0.4853, + "step": 426 + }, + { + "epoch": 2.9204545454545454, + "grad_norm": 0.47864893078804016, + "learning_rate": 3.111131420818922e-06, + "loss": 0.4627, + "step": 427 + }, + { + "epoch": 2.9272727272727272, + "grad_norm": 0.4394248425960541, + "learning_rate": 3.1013123909156347e-06, + "loss": 0.4817, + "step": 428 + }, + { + "epoch": 2.934090909090909, + "grad_norm": 0.6629825830459595, + "learning_rate": 3.091483505575873e-06, + "loss": 0.4738, + "step": 429 + }, + { + "epoch": 2.940909090909091, + "grad_norm": 0.47470518946647644, + "learning_rate": 3.081644925893866e-06, + "loss": 0.4943, + "step": 430 + }, + { + "epoch": 2.9477272727272728, + "grad_norm": 0.44007545709609985, + "learning_rate": 3.0717968131227285e-06, + "loss": 0.4805, + "step": 431 + }, + { + "epoch": 2.9545454545454546, + "grad_norm": 0.4425961375236511, + "learning_rate": 3.061939328671824e-06, + "loss": 0.5092, + "step": 432 + }, + { + "epoch": 2.9613636363636364, + "grad_norm": 0.4215753972530365, + "learning_rate": 3.0520726341041165e-06, + "loss": 0.4616, + "step": 433 + }, + { + "epoch": 2.9681818181818183, + "grad_norm": 0.4265761077404022, + "learning_rate": 3.0421968911335196e-06, + "loss": 0.4764, + "step": 434 + }, + { + "epoch": 2.975, + "grad_norm": 0.45009252429008484, + "learning_rate": 3.032312261622255e-06, + "loss": 0.4865, + "step": 435 + }, + { + "epoch": 2.981818181818182, + "grad_norm": 0.5175682306289673, + "learning_rate": 3.0224189075781886e-06, + "loss": 0.5085, + "step": 436 + }, + { + "epoch": 2.9886363636363638, + "grad_norm": 0.4518980383872986, + "learning_rate": 3.012516991152181e-06, + "loss": 0.4568, + "step": 437 + }, + { + "epoch": 2.9954545454545456, + "grad_norm": 0.4789412021636963, + "learning_rate": 3.002606674635432e-06, + "loss": 0.4942, + "step": 438 + }, + { + "epoch": 3.006818181818182, + "grad_norm": 0.495550274848938, + "learning_rate": 2.9926881204568153e-06, + "loss": 0.9793, + "step": 439 + }, + { + "epoch": 3.0136363636363637, + "grad_norm": 0.47357404232025146, + "learning_rate": 2.9827614911802205e-06, + "loss": 0.4689, + "step": 440 + }, + { + "epoch": 3.0204545454545455, + "grad_norm": 0.4523085355758667, + "learning_rate": 2.972826949501884e-06, + "loss": 0.4632, + "step": 441 + }, + { + "epoch": 3.0272727272727273, + "grad_norm": 0.4381036162376404, + "learning_rate": 2.9628846582477305e-06, + "loss": 0.4508, + "step": 442 + }, + { + "epoch": 3.034090909090909, + "grad_norm": 0.5367632508277893, + "learning_rate": 2.9529347803706943e-06, + "loss": 0.4472, + "step": 443 + }, + { + "epoch": 3.040909090909091, + "grad_norm": 0.5390172600746155, + "learning_rate": 2.9429774789480576e-06, + "loss": 0.4475, + "step": 444 + }, + { + "epoch": 3.047727272727273, + "grad_norm": 0.44644543528556824, + "learning_rate": 2.9330129171787704e-06, + "loss": 0.497, + "step": 445 + }, + { + "epoch": 3.0545454545454547, + "grad_norm": 0.46184423565864563, + "learning_rate": 2.923041258380779e-06, + "loss": 0.4545, + "step": 446 + }, + { + "epoch": 3.0613636363636365, + "grad_norm": 0.45963627099990845, + "learning_rate": 2.9130626659883537e-06, + "loss": 0.4571, + "step": 447 + }, + { + "epoch": 3.0681818181818183, + "grad_norm": 0.4504947364330292, + "learning_rate": 2.9030773035493997e-06, + "loss": 0.4563, + "step": 448 + }, + { + "epoch": 3.075, + "grad_norm": 0.4393884241580963, + "learning_rate": 2.893085334722786e-06, + "loss": 0.4433, + "step": 449 + }, + { + "epoch": 3.081818181818182, + "grad_norm": 0.7640544772148132, + "learning_rate": 2.883086923275658e-06, + "loss": 0.4598, + "step": 450 + }, + { + "epoch": 3.088636363636364, + "grad_norm": 0.44159695506095886, + "learning_rate": 2.8730822330807556e-06, + "loss": 0.4443, + "step": 451 + }, + { + "epoch": 3.0954545454545457, + "grad_norm": 0.4525455832481384, + "learning_rate": 2.8630714281137263e-06, + "loss": 0.4692, + "step": 452 + }, + { + "epoch": 3.102272727272727, + "grad_norm": 0.44391655921936035, + "learning_rate": 2.853054672450437e-06, + "loss": 0.4412, + "step": 453 + }, + { + "epoch": 3.109090909090909, + "grad_norm": 0.45541912317276, + "learning_rate": 2.8430321302642887e-06, + "loss": 0.4417, + "step": 454 + }, + { + "epoch": 3.1159090909090907, + "grad_norm": 0.43590644001960754, + "learning_rate": 2.8330039658235194e-06, + "loss": 0.4529, + "step": 455 + }, + { + "epoch": 3.1227272727272726, + "grad_norm": 0.48097532987594604, + "learning_rate": 2.8229703434885165e-06, + "loss": 0.4519, + "step": 456 + }, + { + "epoch": 3.1295454545454544, + "grad_norm": 0.43121981620788574, + "learning_rate": 2.8129314277091224e-06, + "loss": 0.4663, + "step": 457 + }, + { + "epoch": 3.1363636363636362, + "grad_norm": 0.46553748846054077, + "learning_rate": 2.8028873830219373e-06, + "loss": 0.4521, + "step": 458 + }, + { + "epoch": 3.143181818181818, + "grad_norm": 0.449290007352829, + "learning_rate": 2.7928383740476247e-06, + "loss": 0.4586, + "step": 459 + }, + { + "epoch": 3.15, + "grad_norm": 0.679371178150177, + "learning_rate": 2.7827845654882112e-06, + "loss": 0.4476, + "step": 460 + }, + { + "epoch": 3.1568181818181817, + "grad_norm": 0.49499261379241943, + "learning_rate": 2.7727261221243875e-06, + "loss": 0.4752, + "step": 461 + }, + { + "epoch": 3.1636363636363636, + "grad_norm": 0.4675534963607788, + "learning_rate": 2.76266320881281e-06, + "loss": 0.478, + "step": 462 + }, + { + "epoch": 3.1704545454545454, + "grad_norm": 0.44854965806007385, + "learning_rate": 2.7525959904833955e-06, + "loss": 0.4289, + "step": 463 + }, + { + "epoch": 3.1772727272727272, + "grad_norm": 0.4300600290298462, + "learning_rate": 2.7425246321366205e-06, + "loss": 0.4588, + "step": 464 + }, + { + "epoch": 3.184090909090909, + "grad_norm": 0.46728187799453735, + "learning_rate": 2.7324492988408146e-06, + "loss": 0.4498, + "step": 465 + }, + { + "epoch": 3.190909090909091, + "grad_norm": 0.46769359707832336, + "learning_rate": 2.7223701557294574e-06, + "loss": 0.4324, + "step": 466 + }, + { + "epoch": 3.1977272727272728, + "grad_norm": 0.4275517463684082, + "learning_rate": 2.712287367998471e-06, + "loss": 0.4574, + "step": 467 + }, + { + "epoch": 3.2045454545454546, + "grad_norm": 0.5888628959655762, + "learning_rate": 2.702201100903511e-06, + "loss": 0.4397, + "step": 468 + }, + { + "epoch": 3.2113636363636364, + "grad_norm": 0.4613579511642456, + "learning_rate": 2.692111519757261e-06, + "loss": 0.4391, + "step": 469 + }, + { + "epoch": 3.2181818181818183, + "grad_norm": 0.4581657946109772, + "learning_rate": 2.6820187899267203e-06, + "loss": 0.4734, + "step": 470 + }, + { + "epoch": 3.225, + "grad_norm": 0.4940917491912842, + "learning_rate": 2.671923076830496e-06, + "loss": 0.4726, + "step": 471 + }, + { + "epoch": 3.231818181818182, + "grad_norm": 0.7280719876289368, + "learning_rate": 2.6618245459360896e-06, + "loss": 0.4741, + "step": 472 + }, + { + "epoch": 3.2386363636363638, + "grad_norm": 0.519322395324707, + "learning_rate": 2.651723362757186e-06, + "loss": 0.4549, + "step": 473 + }, + { + "epoch": 3.2454545454545456, + "grad_norm": 0.47500962018966675, + "learning_rate": 2.641619692850941e-06, + "loss": 0.4598, + "step": 474 + }, + { + "epoch": 3.2522727272727274, + "grad_norm": 0.4661007821559906, + "learning_rate": 2.631513701815267e-06, + "loss": 0.4536, + "step": 475 + }, + { + "epoch": 3.2590909090909093, + "grad_norm": 0.48142874240875244, + "learning_rate": 2.6214055552861213e-06, + "loss": 0.474, + "step": 476 + }, + { + "epoch": 3.265909090909091, + "grad_norm": 0.5227756500244141, + "learning_rate": 2.611295418934786e-06, + "loss": 0.4516, + "step": 477 + }, + { + "epoch": 3.2727272727272725, + "grad_norm": 0.4693076014518738, + "learning_rate": 2.6011834584651597e-06, + "loss": 0.4409, + "step": 478 + }, + { + "epoch": 3.2795454545454543, + "grad_norm": 0.44575586915016174, + "learning_rate": 2.591069839611036e-06, + "loss": 0.4382, + "step": 479 + }, + { + "epoch": 3.286363636363636, + "grad_norm": 0.4569104313850403, + "learning_rate": 2.5809547281333904e-06, + "loss": 0.476, + "step": 480 + }, + { + "epoch": 3.293181818181818, + "grad_norm": 0.4542261064052582, + "learning_rate": 2.570838289817661e-06, + "loss": 0.4542, + "step": 481 + }, + { + "epoch": 3.3, + "grad_norm": 0.5086669921875, + "learning_rate": 2.560720690471033e-06, + "loss": 0.4673, + "step": 482 + }, + { + "epoch": 3.3068181818181817, + "grad_norm": 0.46243855357170105, + "learning_rate": 2.5506020959197218e-06, + "loss": 0.4462, + "step": 483 + }, + { + "epoch": 3.3136363636363635, + "grad_norm": 0.4877456724643707, + "learning_rate": 2.5404826720062544e-06, + "loss": 0.4463, + "step": 484 + }, + { + "epoch": 3.3204545454545453, + "grad_norm": 0.4592823088169098, + "learning_rate": 2.5303625845867475e-06, + "loss": 0.4674, + "step": 485 + }, + { + "epoch": 3.327272727272727, + "grad_norm": 0.4744127690792084, + "learning_rate": 2.5202419995281966e-06, + "loss": 0.4502, + "step": 486 + }, + { + "epoch": 3.334090909090909, + "grad_norm": 0.4267658591270447, + "learning_rate": 2.5101210827057516e-06, + "loss": 0.457, + "step": 487 + }, + { + "epoch": 3.340909090909091, + "grad_norm": 0.5804760456085205, + "learning_rate": 2.5e-06, + "loss": 0.4809, + "step": 488 + }, + { + "epoch": 3.3477272727272727, + "grad_norm": 0.5594801306724548, + "learning_rate": 2.4898789172942492e-06, + "loss": 0.4522, + "step": 489 + }, + { + "epoch": 3.3545454545454545, + "grad_norm": 0.4550947844982147, + "learning_rate": 2.4797580004718038e-06, + "loss": 0.4528, + "step": 490 + }, + { + "epoch": 3.3613636363636363, + "grad_norm": 0.430431991815567, + "learning_rate": 2.4696374154132533e-06, + "loss": 0.4657, + "step": 491 + }, + { + "epoch": 3.368181818181818, + "grad_norm": 0.4511764645576477, + "learning_rate": 2.4595173279937464e-06, + "loss": 0.4494, + "step": 492 + }, + { + "epoch": 3.375, + "grad_norm": 0.47958365082740784, + "learning_rate": 2.4493979040802786e-06, + "loss": 0.4508, + "step": 493 + }, + { + "epoch": 3.381818181818182, + "grad_norm": 0.4632541835308075, + "learning_rate": 2.4392793095289677e-06, + "loss": 0.4444, + "step": 494 + }, + { + "epoch": 3.3886363636363637, + "grad_norm": 0.46840110421180725, + "learning_rate": 2.42916171018234e-06, + "loss": 0.4676, + "step": 495 + }, + { + "epoch": 3.3954545454545455, + "grad_norm": 0.49999886751174927, + "learning_rate": 2.419045271866611e-06, + "loss": 0.4589, + "step": 496 + }, + { + "epoch": 3.4022727272727273, + "grad_norm": 0.44237902760505676, + "learning_rate": 2.408930160388965e-06, + "loss": 0.4569, + "step": 497 + }, + { + "epoch": 3.409090909090909, + "grad_norm": 0.4862026572227478, + "learning_rate": 2.3988165415348416e-06, + "loss": 0.4534, + "step": 498 + }, + { + "epoch": 3.415909090909091, + "grad_norm": 0.48773396015167236, + "learning_rate": 2.388704581065215e-06, + "loss": 0.4411, + "step": 499 + }, + { + "epoch": 3.422727272727273, + "grad_norm": 0.4380181133747101, + "learning_rate": 2.3785944447138804e-06, + "loss": 0.4529, + "step": 500 + }, + { + "epoch": 3.4295454545454547, + "grad_norm": 0.4616672992706299, + "learning_rate": 2.368486298184733e-06, + "loss": 0.4596, + "step": 501 + }, + { + "epoch": 3.4363636363636365, + "grad_norm": 0.4307732582092285, + "learning_rate": 2.358380307149059e-06, + "loss": 0.4626, + "step": 502 + }, + { + "epoch": 3.4431818181818183, + "grad_norm": 0.4402921795845032, + "learning_rate": 2.348276637242814e-06, + "loss": 0.4606, + "step": 503 + }, + { + "epoch": 3.45, + "grad_norm": 0.4782681465148926, + "learning_rate": 2.3381754540639108e-06, + "loss": 0.4508, + "step": 504 + }, + { + "epoch": 3.456818181818182, + "grad_norm": 0.43719062209129333, + "learning_rate": 2.328076923169504e-06, + "loss": 0.4476, + "step": 505 + }, + { + "epoch": 3.463636363636364, + "grad_norm": 0.4890111982822418, + "learning_rate": 2.31798121007328e-06, + "loss": 0.4484, + "step": 506 + }, + { + "epoch": 3.4704545454545457, + "grad_norm": 0.47682395577430725, + "learning_rate": 2.3078884802427394e-06, + "loss": 0.4494, + "step": 507 + }, + { + "epoch": 3.4772727272727275, + "grad_norm": 0.566819429397583, + "learning_rate": 2.29779889909649e-06, + "loss": 0.4539, + "step": 508 + }, + { + "epoch": 3.484090909090909, + "grad_norm": 0.4956795871257782, + "learning_rate": 2.2877126320015295e-06, + "loss": 0.4476, + "step": 509 + }, + { + "epoch": 3.4909090909090907, + "grad_norm": 0.5630896687507629, + "learning_rate": 2.2776298442705434e-06, + "loss": 0.4436, + "step": 510 + }, + { + "epoch": 3.4977272727272726, + "grad_norm": 0.9436272382736206, + "learning_rate": 2.267550701159186e-06, + "loss": 0.4558, + "step": 511 + }, + { + "epoch": 3.5045454545454544, + "grad_norm": 0.43885400891304016, + "learning_rate": 2.25747536786338e-06, + "loss": 0.4375, + "step": 512 + }, + { + "epoch": 3.5113636363636362, + "grad_norm": 0.762102484703064, + "learning_rate": 2.247404009516605e-06, + "loss": 0.4491, + "step": 513 + }, + { + "epoch": 3.518181818181818, + "grad_norm": 0.46883025765419006, + "learning_rate": 2.2373367911871904e-06, + "loss": 0.4393, + "step": 514 + }, + { + "epoch": 3.525, + "grad_norm": 0.4758882522583008, + "learning_rate": 2.227273877875613e-06, + "loss": 0.4446, + "step": 515 + }, + { + "epoch": 3.5318181818181817, + "grad_norm": 0.4799712300300598, + "learning_rate": 2.2172154345117896e-06, + "loss": 0.455, + "step": 516 + }, + { + "epoch": 3.5386363636363636, + "grad_norm": 0.46632879972457886, + "learning_rate": 2.207161625952376e-06, + "loss": 0.4481, + "step": 517 + }, + { + "epoch": 3.5454545454545454, + "grad_norm": 0.6286160945892334, + "learning_rate": 2.1971126169780636e-06, + "loss": 0.4436, + "step": 518 + }, + { + "epoch": 3.5522727272727272, + "grad_norm": 0.4679605960845947, + "learning_rate": 2.1870685722908784e-06, + "loss": 0.4575, + "step": 519 + }, + { + "epoch": 3.559090909090909, + "grad_norm": 0.495728075504303, + "learning_rate": 2.1770296565114847e-06, + "loss": 0.4399, + "step": 520 + }, + { + "epoch": 3.565909090909091, + "grad_norm": 0.4634799361228943, + "learning_rate": 2.166996034176482e-06, + "loss": 0.4553, + "step": 521 + }, + { + "epoch": 3.5727272727272728, + "grad_norm": 0.4464644491672516, + "learning_rate": 2.1569678697357126e-06, + "loss": 0.4693, + "step": 522 + }, + { + "epoch": 3.5795454545454546, + "grad_norm": 0.4530065655708313, + "learning_rate": 2.1469453275495634e-06, + "loss": 0.4418, + "step": 523 + }, + { + "epoch": 3.5863636363636364, + "grad_norm": 0.4806428551673889, + "learning_rate": 2.136928571886275e-06, + "loss": 0.4412, + "step": 524 + }, + { + "epoch": 3.5931818181818183, + "grad_norm": 0.46114400029182434, + "learning_rate": 2.126917766919245e-06, + "loss": 0.4502, + "step": 525 + }, + { + "epoch": 3.6, + "grad_norm": 0.7423635125160217, + "learning_rate": 2.1169130767243424e-06, + "loss": 0.4573, + "step": 526 + }, + { + "epoch": 3.606818181818182, + "grad_norm": 0.47871625423431396, + "learning_rate": 2.1069146652772142e-06, + "loss": 0.4427, + "step": 527 + }, + { + "epoch": 3.6136363636363638, + "grad_norm": 0.45554301142692566, + "learning_rate": 2.0969226964506007e-06, + "loss": 0.4429, + "step": 528 + }, + { + "epoch": 3.6204545454545456, + "grad_norm": 0.44822901487350464, + "learning_rate": 2.0869373340116467e-06, + "loss": 0.4616, + "step": 529 + }, + { + "epoch": 3.6272727272727274, + "grad_norm": 0.4922308027744293, + "learning_rate": 2.0769587416192212e-06, + "loss": 0.4482, + "step": 530 + }, + { + "epoch": 3.634090909090909, + "grad_norm": 0.4700371026992798, + "learning_rate": 2.066987082821231e-06, + "loss": 0.4652, + "step": 531 + }, + { + "epoch": 3.6409090909090907, + "grad_norm": 0.5513899326324463, + "learning_rate": 2.0570225210519433e-06, + "loss": 0.4542, + "step": 532 + }, + { + "epoch": 3.6477272727272725, + "grad_norm": 0.47292518615722656, + "learning_rate": 2.047065219629306e-06, + "loss": 0.4675, + "step": 533 + }, + { + "epoch": 3.6545454545454543, + "grad_norm": 0.5037868618965149, + "learning_rate": 2.0371153417522703e-06, + "loss": 0.4637, + "step": 534 + }, + { + "epoch": 3.661363636363636, + "grad_norm": 0.4578365087509155, + "learning_rate": 2.0271730504981165e-06, + "loss": 0.4432, + "step": 535 + }, + { + "epoch": 3.668181818181818, + "grad_norm": 0.518247127532959, + "learning_rate": 2.0172385088197804e-06, + "loss": 0.4574, + "step": 536 + }, + { + "epoch": 3.675, + "grad_norm": 0.5211175680160522, + "learning_rate": 2.007311879543185e-06, + "loss": 0.4607, + "step": 537 + }, + { + "epoch": 3.6818181818181817, + "grad_norm": 0.4695160686969757, + "learning_rate": 1.9973933253645684e-06, + "loss": 0.447, + "step": 538 + }, + { + "epoch": 3.6886363636363635, + "grad_norm": 0.5875166654586792, + "learning_rate": 1.9874830088478196e-06, + "loss": 0.4615, + "step": 539 + }, + { + "epoch": 3.6954545454545453, + "grad_norm": 0.4762527644634247, + "learning_rate": 1.9775810924218126e-06, + "loss": 0.4677, + "step": 540 + }, + { + "epoch": 3.702272727272727, + "grad_norm": 0.47306928038597107, + "learning_rate": 1.967687738377746e-06, + "loss": 0.4526, + "step": 541 + }, + { + "epoch": 3.709090909090909, + "grad_norm": 0.44723549485206604, + "learning_rate": 1.9578031088664812e-06, + "loss": 0.469, + "step": 542 + }, + { + "epoch": 3.715909090909091, + "grad_norm": 0.4447214901447296, + "learning_rate": 1.9479273658958852e-06, + "loss": 0.4375, + "step": 543 + }, + { + "epoch": 3.7227272727272727, + "grad_norm": 0.5299004316329956, + "learning_rate": 1.9380606713281773e-06, + "loss": 0.4544, + "step": 544 + }, + { + "epoch": 3.7295454545454545, + "grad_norm": 0.4843369424343109, + "learning_rate": 1.928203186877273e-06, + "loss": 0.4692, + "step": 545 + }, + { + "epoch": 3.7363636363636363, + "grad_norm": 0.4805986285209656, + "learning_rate": 1.9183550741061354e-06, + "loss": 0.4498, + "step": 546 + }, + { + "epoch": 3.743181818181818, + "grad_norm": 0.6539422273635864, + "learning_rate": 1.9085164944241275e-06, + "loss": 0.4398, + "step": 547 + }, + { + "epoch": 3.75, + "grad_norm": 0.47688615322113037, + "learning_rate": 1.8986876090843668e-06, + "loss": 0.4642, + "step": 548 + }, + { + "epoch": 3.756818181818182, + "grad_norm": 0.4808521270751953, + "learning_rate": 1.8888685791810784e-06, + "loss": 0.4166, + "step": 549 + }, + { + "epoch": 3.7636363636363637, + "grad_norm": 0.49430814385414124, + "learning_rate": 1.8790595656469628e-06, + "loss": 0.4404, + "step": 550 + }, + { + "epoch": 3.7704545454545455, + "grad_norm": 0.5284092426300049, + "learning_rate": 1.86926072925055e-06, + "loss": 0.445, + "step": 551 + }, + { + "epoch": 3.7772727272727273, + "grad_norm": 0.4650699496269226, + "learning_rate": 1.8594722305935691e-06, + "loss": 0.4381, + "step": 552 + }, + { + "epoch": 3.784090909090909, + "grad_norm": 0.4954371452331543, + "learning_rate": 1.8496942301083142e-06, + "loss": 0.4534, + "step": 553 + }, + { + "epoch": 3.790909090909091, + "grad_norm": 0.4756702482700348, + "learning_rate": 1.8399268880550174e-06, + "loss": 0.4563, + "step": 554 + }, + { + "epoch": 3.797727272727273, + "grad_norm": 0.4566079378128052, + "learning_rate": 1.8301703645192178e-06, + "loss": 0.4422, + "step": 555 + }, + { + "epoch": 3.8045454545454547, + "grad_norm": 1.6503181457519531, + "learning_rate": 1.8204248194091429e-06, + "loss": 0.4548, + "step": 556 + }, + { + "epoch": 3.8113636363636365, + "grad_norm": 0.5488978028297424, + "learning_rate": 1.8106904124530839e-06, + "loss": 0.4398, + "step": 557 + }, + { + "epoch": 3.8181818181818183, + "grad_norm": 0.44948896765708923, + "learning_rate": 1.800967303196778e-06, + "loss": 0.4331, + "step": 558 + }, + { + "epoch": 3.825, + "grad_norm": 0.49597036838531494, + "learning_rate": 1.7912556510007967e-06, + "loss": 0.4409, + "step": 559 + }, + { + "epoch": 3.831818181818182, + "grad_norm": 0.5093109011650085, + "learning_rate": 1.7815556150379298e-06, + "loss": 0.4615, + "step": 560 + }, + { + "epoch": 3.838636363636364, + "grad_norm": 0.4719341993331909, + "learning_rate": 1.77186735429058e-06, + "loss": 0.4316, + "step": 561 + }, + { + "epoch": 3.8454545454545457, + "grad_norm": 0.5281686782836914, + "learning_rate": 1.7621910275481544e-06, + "loss": 0.4474, + "step": 562 + }, + { + "epoch": 3.8522727272727275, + "grad_norm": 0.4955421984195709, + "learning_rate": 1.7525267934044642e-06, + "loss": 0.462, + "step": 563 + }, + { + "epoch": 3.8590909090909093, + "grad_norm": 0.4997479021549225, + "learning_rate": 1.7428748102551237e-06, + "loss": 0.4546, + "step": 564 + }, + { + "epoch": 3.865909090909091, + "grad_norm": 0.47572800517082214, + "learning_rate": 1.7332352362949546e-06, + "loss": 0.4587, + "step": 565 + }, + { + "epoch": 3.8727272727272726, + "grad_norm": 0.5156065225601196, + "learning_rate": 1.7236082295153948e-06, + "loss": 0.4375, + "step": 566 + }, + { + "epoch": 3.8795454545454544, + "grad_norm": 0.44926485419273376, + "learning_rate": 1.7139939477019057e-06, + "loss": 0.45, + "step": 567 + }, + { + "epoch": 3.8863636363636362, + "grad_norm": 0.5433095693588257, + "learning_rate": 1.7043925484313911e-06, + "loss": 0.4464, + "step": 568 + }, + { + "epoch": 3.893181818181818, + "grad_norm": 0.4719317555427551, + "learning_rate": 1.6948041890696076e-06, + "loss": 0.4467, + "step": 569 + }, + { + "epoch": 3.9, + "grad_norm": 0.4679432511329651, + "learning_rate": 1.685229026768593e-06, + "loss": 0.4577, + "step": 570 + }, + { + "epoch": 3.9068181818181817, + "grad_norm": 0.5053668022155762, + "learning_rate": 1.6756672184640847e-06, + "loss": 0.4379, + "step": 571 + }, + { + "epoch": 3.9136363636363636, + "grad_norm": 0.4880099594593048, + "learning_rate": 1.6661189208729492e-06, + "loss": 0.4585, + "step": 572 + }, + { + "epoch": 3.9204545454545454, + "grad_norm": 0.44481492042541504, + "learning_rate": 1.6565842904906154e-06, + "loss": 0.4319, + "step": 573 + }, + { + "epoch": 3.9272727272727272, + "grad_norm": 0.4998759329319, + "learning_rate": 1.6470634835885097e-06, + "loss": 0.4349, + "step": 574 + }, + { + "epoch": 3.934090909090909, + "grad_norm": 0.4702436625957489, + "learning_rate": 1.6375566562114903e-06, + "loss": 0.4409, + "step": 575 + }, + { + "epoch": 3.940909090909091, + "grad_norm": 1.1932134628295898, + "learning_rate": 1.6280639641752944e-06, + "loss": 0.4435, + "step": 576 + }, + { + "epoch": 3.9477272727272728, + "grad_norm": 0.46057313680648804, + "learning_rate": 1.6185855630639818e-06, + "loss": 0.451, + "step": 577 + }, + { + "epoch": 3.9545454545454546, + "grad_norm": 0.5858541131019592, + "learning_rate": 1.6091216082273875e-06, + "loss": 0.4662, + "step": 578 + }, + { + "epoch": 3.9613636363636364, + "grad_norm": 0.46951207518577576, + "learning_rate": 1.5996722547785722e-06, + "loss": 0.4561, + "step": 579 + }, + { + "epoch": 3.9681818181818183, + "grad_norm": 0.4747212529182434, + "learning_rate": 1.5902376575912815e-06, + "loss": 0.4507, + "step": 580 + }, + { + "epoch": 3.975, + "grad_norm": 0.5081924200057983, + "learning_rate": 1.580817971297409e-06, + "loss": 0.4649, + "step": 581 + }, + { + "epoch": 3.981818181818182, + "grad_norm": 0.4900786280632019, + "learning_rate": 1.5714133502844591e-06, + "loss": 0.4587, + "step": 582 + }, + { + "epoch": 3.9886363636363638, + "grad_norm": 0.4827897250652313, + "learning_rate": 1.56202394869302e-06, + "loss": 0.441, + "step": 583 + }, + { + "epoch": 3.9954545454545456, + "grad_norm": 0.499607652425766, + "learning_rate": 1.5526499204142332e-06, + "loss": 0.4559, + "step": 584 + }, + { + "epoch": 4.006818181818182, + "grad_norm": 1.06562340259552, + "learning_rate": 1.5432914190872757e-06, + "loss": 0.928, + "step": 585 + }, + { + "epoch": 4.013636363636364, + "grad_norm": 0.5927309393882751, + "learning_rate": 1.5339485980968383e-06, + "loss": 0.4446, + "step": 586 + }, + { + "epoch": 4.0204545454545455, + "grad_norm": 0.46158862113952637, + "learning_rate": 1.5246216105706124e-06, + "loss": 0.4394, + "step": 587 + }, + { + "epoch": 4.027272727272727, + "grad_norm": 0.46964529156684875, + "learning_rate": 1.5153106093767827e-06, + "loss": 0.4364, + "step": 588 + }, + { + "epoch": 4.034090909090909, + "grad_norm": 0.4228534698486328, + "learning_rate": 1.506015747121518e-06, + "loss": 0.4234, + "step": 589 + }, + { + "epoch": 4.040909090909091, + "grad_norm": 0.5211766362190247, + "learning_rate": 1.4967371761464738e-06, + "loss": 0.4428, + "step": 590 + }, + { + "epoch": 4.047727272727273, + "grad_norm": 0.4786047637462616, + "learning_rate": 1.4874750485262917e-06, + "loss": 0.4416, + "step": 591 + }, + { + "epoch": 4.054545454545455, + "grad_norm": 2.0014617443084717, + "learning_rate": 1.4782295160661103e-06, + "loss": 0.4338, + "step": 592 + }, + { + "epoch": 4.0613636363636365, + "grad_norm": 0.5174881219863892, + "learning_rate": 1.469000730299074e-06, + "loss": 0.431, + "step": 593 + }, + { + "epoch": 4.068181818181818, + "grad_norm": 0.4879564344882965, + "learning_rate": 1.4597888424838519e-06, + "loss": 0.4236, + "step": 594 + }, + { + "epoch": 4.075, + "grad_norm": 0.4452327787876129, + "learning_rate": 1.450594003602158e-06, + "loss": 0.4447, + "step": 595 + }, + { + "epoch": 4.081818181818182, + "grad_norm": 0.6323989033699036, + "learning_rate": 1.4414163643562755e-06, + "loss": 0.4241, + "step": 596 + }, + { + "epoch": 4.088636363636364, + "grad_norm": 0.532921314239502, + "learning_rate": 1.4322560751665873e-06, + "loss": 0.4388, + "step": 597 + }, + { + "epoch": 4.095454545454546, + "grad_norm": 0.49364256858825684, + "learning_rate": 1.4231132861691128e-06, + "loss": 0.4244, + "step": 598 + }, + { + "epoch": 4.1022727272727275, + "grad_norm": 0.47677481174468994, + "learning_rate": 1.4139881472130453e-06, + "loss": 0.4235, + "step": 599 + }, + { + "epoch": 4.109090909090909, + "grad_norm": 0.4729495644569397, + "learning_rate": 1.4048808078582943e-06, + "loss": 0.4177, + "step": 600 + }, + { + "epoch": 4.115909090909091, + "grad_norm": 0.44931760430336, + "learning_rate": 1.3957914173730366e-06, + "loss": 0.4429, + "step": 601 + }, + { + "epoch": 4.122727272727273, + "grad_norm": 0.5063349604606628, + "learning_rate": 1.3867201247312697e-06, + "loss": 0.4432, + "step": 602 + }, + { + "epoch": 4.129545454545455, + "grad_norm": 0.5439795255661011, + "learning_rate": 1.3776670786103685e-06, + "loss": 0.4455, + "step": 603 + }, + { + "epoch": 4.136363636363637, + "grad_norm": 0.4854280650615692, + "learning_rate": 1.3686324273886531e-06, + "loss": 0.4334, + "step": 604 + }, + { + "epoch": 4.1431818181818185, + "grad_norm": 0.47003456950187683, + "learning_rate": 1.359616319142949e-06, + "loss": 0.438, + "step": 605 + }, + { + "epoch": 4.15, + "grad_norm": 0.6236258149147034, + "learning_rate": 1.3506189016461674e-06, + "loss": 0.4356, + "step": 606 + }, + { + "epoch": 4.156818181818182, + "grad_norm": 0.9478527903556824, + "learning_rate": 1.341640322364878e-06, + "loss": 0.4353, + "step": 607 + }, + { + "epoch": 4.163636363636364, + "grad_norm": 0.48970767855644226, + "learning_rate": 1.3326807284568984e-06, + "loss": 0.4319, + "step": 608 + }, + { + "epoch": 4.170454545454546, + "grad_norm": 0.5102277994155884, + "learning_rate": 1.323740266768875e-06, + "loss": 0.438, + "step": 609 + }, + { + "epoch": 4.177272727272728, + "grad_norm": 0.48640403151512146, + "learning_rate": 1.3148190838338804e-06, + "loss": 0.4001, + "step": 610 + }, + { + "epoch": 4.184090909090909, + "grad_norm": 0.4933972656726837, + "learning_rate": 1.3059173258690102e-06, + "loss": 0.451, + "step": 611 + }, + { + "epoch": 4.190909090909091, + "grad_norm": 0.4632134437561035, + "learning_rate": 1.2970351387729875e-06, + "loss": 0.4396, + "step": 612 + }, + { + "epoch": 4.197727272727272, + "grad_norm": 0.5672054290771484, + "learning_rate": 1.2881726681237727e-06, + "loss": 0.4433, + "step": 613 + }, + { + "epoch": 4.204545454545454, + "grad_norm": 0.4600079655647278, + "learning_rate": 1.2793300591761742e-06, + "loss": 0.4038, + "step": 614 + }, + { + "epoch": 4.211363636363636, + "grad_norm": 0.44386839866638184, + "learning_rate": 1.27050745685947e-06, + "loss": 0.4442, + "step": 615 + }, + { + "epoch": 4.218181818181818, + "grad_norm": 0.793684184551239, + "learning_rate": 1.2617050057750322e-06, + "loss": 0.4309, + "step": 616 + }, + { + "epoch": 4.225, + "grad_norm": 0.4929650127887726, + "learning_rate": 1.252922850193955e-06, + "loss": 0.4304, + "step": 617 + }, + { + "epoch": 4.2318181818181815, + "grad_norm": 0.5073612332344055, + "learning_rate": 1.2441611340546958e-06, + "loss": 0.4271, + "step": 618 + }, + { + "epoch": 4.238636363636363, + "grad_norm": 0.4528202414512634, + "learning_rate": 1.2354200009607081e-06, + "loss": 0.4384, + "step": 619 + }, + { + "epoch": 4.245454545454545, + "grad_norm": 0.46402111649513245, + "learning_rate": 1.2266995941780934e-06, + "loss": 0.43, + "step": 620 + }, + { + "epoch": 4.252272727272727, + "grad_norm": 0.567254364490509, + "learning_rate": 1.2180000566332503e-06, + "loss": 0.3905, + "step": 621 + }, + { + "epoch": 4.259090909090909, + "grad_norm": 0.5210692286491394, + "learning_rate": 1.2093215309105352e-06, + "loss": 0.4262, + "step": 622 + }, + { + "epoch": 4.265909090909091, + "grad_norm": 1.3842746019363403, + "learning_rate": 1.2006641592499233e-06, + "loss": 0.4268, + "step": 623 + }, + { + "epoch": 4.2727272727272725, + "grad_norm": 0.49402037262916565, + "learning_rate": 1.192028083544675e-06, + "loss": 0.4244, + "step": 624 + }, + { + "epoch": 4.279545454545454, + "grad_norm": 0.4855121374130249, + "learning_rate": 1.1834134453390136e-06, + "loss": 0.4274, + "step": 625 + }, + { + "epoch": 4.286363636363636, + "grad_norm": 0.5327234864234924, + "learning_rate": 1.1748203858258056e-06, + "loss": 0.429, + "step": 626 + }, + { + "epoch": 4.293181818181818, + "grad_norm": 0.482337087392807, + "learning_rate": 1.166249045844243e-06, + "loss": 0.4253, + "step": 627 + }, + { + "epoch": 4.3, + "grad_norm": 0.9390714168548584, + "learning_rate": 1.1576995658775405e-06, + "loss": 0.4224, + "step": 628 + }, + { + "epoch": 4.306818181818182, + "grad_norm": 0.5063149929046631, + "learning_rate": 1.1491720860506273e-06, + "loss": 0.4243, + "step": 629 + }, + { + "epoch": 4.3136363636363635, + "grad_norm": 0.4565083980560303, + "learning_rate": 1.140666746127854e-06, + "loss": 0.4137, + "step": 630 + }, + { + "epoch": 4.320454545454545, + "grad_norm": 0.48249998688697815, + "learning_rate": 1.1321836855107007e-06, + "loss": 0.4143, + "step": 631 + }, + { + "epoch": 4.327272727272727, + "grad_norm": 0.5024334788322449, + "learning_rate": 1.1237230432354912e-06, + "loss": 0.4302, + "step": 632 + }, + { + "epoch": 4.334090909090909, + "grad_norm": 0.4903137981891632, + "learning_rate": 1.1152849579711187e-06, + "loss": 0.4522, + "step": 633 + }, + { + "epoch": 4.340909090909091, + "grad_norm": 0.47344261407852173, + "learning_rate": 1.1068695680167665e-06, + "loss": 0.4487, + "step": 634 + }, + { + "epoch": 4.347727272727273, + "grad_norm": 0.5437475442886353, + "learning_rate": 1.0984770112996463e-06, + "loss": 0.4255, + "step": 635 + }, + { + "epoch": 4.3545454545454545, + "grad_norm": 0.4634874165058136, + "learning_rate": 1.0901074253727338e-06, + "loss": 0.4345, + "step": 636 + }, + { + "epoch": 4.361363636363636, + "grad_norm": 0.5247130393981934, + "learning_rate": 1.0817609474125195e-06, + "loss": 0.4296, + "step": 637 + }, + { + "epoch": 4.368181818181818, + "grad_norm": 0.503746747970581, + "learning_rate": 1.0734377142167549e-06, + "loss": 0.4251, + "step": 638 + }, + { + "epoch": 4.375, + "grad_norm": 0.5096430778503418, + "learning_rate": 1.065137862202213e-06, + "loss": 0.4186, + "step": 639 + }, + { + "epoch": 4.381818181818182, + "grad_norm": 0.4813026785850525, + "learning_rate": 1.0568615274024521e-06, + "loss": 0.4265, + "step": 640 + }, + { + "epoch": 4.388636363636364, + "grad_norm": 0.45982855558395386, + "learning_rate": 1.0486088454655856e-06, + "loss": 0.4188, + "step": 641 + }, + { + "epoch": 4.3954545454545455, + "grad_norm": 0.5047855377197266, + "learning_rate": 1.0403799516520619e-06, + "loss": 0.4231, + "step": 642 + }, + { + "epoch": 4.402272727272727, + "grad_norm": 0.46901944279670715, + "learning_rate": 1.0321749808324425e-06, + "loss": 0.4173, + "step": 643 + }, + { + "epoch": 4.409090909090909, + "grad_norm": 0.48727908730506897, + "learning_rate": 1.0239940674851943e-06, + "loss": 0.4631, + "step": 644 + }, + { + "epoch": 4.415909090909091, + "grad_norm": 0.6558519005775452, + "learning_rate": 1.0158373456944856e-06, + "loss": 0.4177, + "step": 645 + }, + { + "epoch": 4.422727272727273, + "grad_norm": 0.4979148805141449, + "learning_rate": 1.0077049491479874e-06, + "loss": 0.4046, + "step": 646 + }, + { + "epoch": 4.429545454545455, + "grad_norm": 0.5404463410377502, + "learning_rate": 9.995970111346842e-07, + "loss": 0.4314, + "step": 647 + }, + { + "epoch": 4.4363636363636365, + "grad_norm": 0.7192148566246033, + "learning_rate": 9.915136645426885e-07, + "loss": 0.429, + "step": 648 + }, + { + "epoch": 4.443181818181818, + "grad_norm": 0.47399184107780457, + "learning_rate": 9.834550418570602e-07, + "loss": 0.4392, + "step": 649 + }, + { + "epoch": 4.45, + "grad_norm": 0.48218435049057007, + "learning_rate": 9.754212751576386e-07, + "loss": 0.4282, + "step": 650 + }, + { + "epoch": 4.456818181818182, + "grad_norm": 0.5437904596328735, + "learning_rate": 9.67412496116876e-07, + "loss": 0.4031, + "step": 651 + }, + { + "epoch": 4.463636363636364, + "grad_norm": 0.45721930265426636, + "learning_rate": 9.594288359976817e-07, + "loss": 0.4268, + "step": 652 + }, + { + "epoch": 4.470454545454546, + "grad_norm": 0.48765209317207336, + "learning_rate": 9.514704256512669e-07, + "loss": 0.4491, + "step": 653 + }, + { + "epoch": 4.4772727272727275, + "grad_norm": 0.45831185579299927, + "learning_rate": 9.435373955150032e-07, + "loss": 0.423, + "step": 654 + }, + { + "epoch": 4.484090909090909, + "grad_norm": 0.50241619348526, + "learning_rate": 9.35629875610283e-07, + "loss": 0.4334, + "step": 655 + }, + { + "epoch": 4.490909090909091, + "grad_norm": 0.4767427444458008, + "learning_rate": 9.277479955403887e-07, + "loss": 0.4475, + "step": 656 + }, + { + "epoch": 4.497727272727273, + "grad_norm": 0.48379838466644287, + "learning_rate": 9.198918844883714e-07, + "loss": 0.4353, + "step": 657 + }, + { + "epoch": 4.504545454545455, + "grad_norm": 0.5098893046379089, + "learning_rate": 9.120616712149291e-07, + "loss": 0.4285, + "step": 658 + }, + { + "epoch": 4.511363636363637, + "grad_norm": 0.5217880010604858, + "learning_rate": 9.042574840562982e-07, + "loss": 0.4395, + "step": 659 + }, + { + "epoch": 4.5181818181818185, + "grad_norm": 0.449683278799057, + "learning_rate": 8.964794509221508e-07, + "loss": 0.4382, + "step": 660 + }, + { + "epoch": 4.525, + "grad_norm": 0.5583493113517761, + "learning_rate": 8.887276992934976e-07, + "loss": 0.4378, + "step": 661 + }, + { + "epoch": 4.531818181818182, + "grad_norm": 0.503844678401947, + "learning_rate": 8.810023562206e-07, + "loss": 0.4222, + "step": 662 + }, + { + "epoch": 4.538636363636364, + "grad_norm": 0.46325281262397766, + "learning_rate": 8.733035483208841e-07, + "loss": 0.441, + "step": 663 + }, + { + "epoch": 4.545454545454545, + "grad_norm": 0.5824127197265625, + "learning_rate": 8.656314017768694e-07, + "loss": 0.4224, + "step": 664 + }, + { + "epoch": 4.552272727272728, + "grad_norm": 0.5601843595504761, + "learning_rate": 8.579860423340977e-07, + "loss": 0.4348, + "step": 665 + }, + { + "epoch": 4.559090909090909, + "grad_norm": 0.45808762311935425, + "learning_rate": 8.503675952990756e-07, + "loss": 0.4445, + "step": 666 + }, + { + "epoch": 4.565909090909091, + "grad_norm": 0.494069367647171, + "learning_rate": 8.427761855372169e-07, + "loss": 0.4271, + "step": 667 + }, + { + "epoch": 4.572727272727272, + "grad_norm": 0.5897441506385803, + "learning_rate": 8.352119374707979e-07, + "loss": 0.4218, + "step": 668 + }, + { + "epoch": 4.579545454545455, + "grad_norm": 0.5399529337882996, + "learning_rate": 8.276749750769186e-07, + "loss": 0.4295, + "step": 669 + }, + { + "epoch": 4.586363636363636, + "grad_norm": 0.44742587208747864, + "learning_rate": 8.20165421885469e-07, + "loss": 0.4406, + "step": 670 + }, + { + "epoch": 4.593181818181818, + "grad_norm": 0.4736122786998749, + "learning_rate": 8.126834009771079e-07, + "loss": 0.4255, + "step": 671 + }, + { + "epoch": 4.6, + "grad_norm": 0.4699339270591736, + "learning_rate": 8.052290349812419e-07, + "loss": 0.4388, + "step": 672 + }, + { + "epoch": 4.6068181818181815, + "grad_norm": 1.0179539918899536, + "learning_rate": 7.978024460740169e-07, + "loss": 0.445, + "step": 673 + }, + { + "epoch": 4.613636363636363, + "grad_norm": 0.4838411211967468, + "learning_rate": 7.904037559763162e-07, + "loss": 0.4518, + "step": 674 + }, + { + "epoch": 4.620454545454545, + "grad_norm": 0.4670931398868561, + "learning_rate": 7.83033085951764e-07, + "loss": 0.4294, + "step": 675 + }, + { + "epoch": 4.627272727272727, + "grad_norm": 0.47013846039772034, + "learning_rate": 7.756905568047393e-07, + "loss": 0.4395, + "step": 676 + }, + { + "epoch": 4.634090909090909, + "grad_norm": 0.875905454158783, + "learning_rate": 7.683762888783977e-07, + "loss": 0.4288, + "step": 677 + }, + { + "epoch": 4.640909090909091, + "grad_norm": 0.47206172347068787, + "learning_rate": 7.610904020526938e-07, + "loss": 0.4244, + "step": 678 + }, + { + "epoch": 4.6477272727272725, + "grad_norm": 0.6244398355484009, + "learning_rate": 7.538330157424212e-07, + "loss": 0.4444, + "step": 679 + }, + { + "epoch": 4.654545454545454, + "grad_norm": 0.5075065493583679, + "learning_rate": 7.466042488952521e-07, + "loss": 0.433, + "step": 680 + }, + { + "epoch": 4.661363636363636, + "grad_norm": 0.7290046811103821, + "learning_rate": 7.394042199897916e-07, + "loss": 0.4509, + "step": 681 + }, + { + "epoch": 4.668181818181818, + "grad_norm": 0.4735564589500427, + "learning_rate": 7.322330470336314e-07, + "loss": 0.425, + "step": 682 + }, + { + "epoch": 4.675, + "grad_norm": 0.4939814805984497, + "learning_rate": 7.250908475614185e-07, + "loss": 0.4058, + "step": 683 + }, + { + "epoch": 4.681818181818182, + "grad_norm": 0.44935867190361023, + "learning_rate": 7.179777386329276e-07, + "loss": 0.4132, + "step": 684 + }, + { + "epoch": 4.6886363636363635, + "grad_norm": 0.48104918003082275, + "learning_rate": 7.108938368311424e-07, + "loss": 0.4581, + "step": 685 + }, + { + "epoch": 4.695454545454545, + "grad_norm": 0.5192425847053528, + "learning_rate": 7.038392582603481e-07, + "loss": 0.4542, + "step": 686 + }, + { + "epoch": 4.702272727272727, + "grad_norm": 0.5787845253944397, + "learning_rate": 6.968141185442229e-07, + "loss": 0.421, + "step": 687 + }, + { + "epoch": 4.709090909090909, + "grad_norm": 0.4911254942417145, + "learning_rate": 6.898185328239468e-07, + "loss": 0.412, + "step": 688 + }, + { + "epoch": 4.715909090909091, + "grad_norm": 0.5646374821662903, + "learning_rate": 6.828526157563126e-07, + "loss": 0.428, + "step": 689 + }, + { + "epoch": 4.722727272727273, + "grad_norm": 0.4580458402633667, + "learning_rate": 6.759164815118493e-07, + "loss": 0.4342, + "step": 690 + }, + { + "epoch": 4.7295454545454545, + "grad_norm": 0.5442156791687012, + "learning_rate": 6.690102437729481e-07, + "loss": 0.442, + "step": 691 + }, + { + "epoch": 4.736363636363636, + "grad_norm": 0.4839988052845001, + "learning_rate": 6.621340157319998e-07, + "loss": 0.4384, + "step": 692 + }, + { + "epoch": 4.743181818181818, + "grad_norm": 0.5340309143066406, + "learning_rate": 6.552879100895396e-07, + "loss": 0.4224, + "step": 693 + }, + { + "epoch": 4.75, + "grad_norm": 0.5196749567985535, + "learning_rate": 6.484720390524008e-07, + "loss": 0.4317, + "step": 694 + }, + { + "epoch": 4.756818181818182, + "grad_norm": 0.9252122044563293, + "learning_rate": 6.416865143318757e-07, + "loss": 0.4375, + "step": 695 + }, + { + "epoch": 4.763636363636364, + "grad_norm": 0.7175036668777466, + "learning_rate": 6.349314471418849e-07, + "loss": 0.4475, + "step": 696 + }, + { + "epoch": 4.7704545454545455, + "grad_norm": 0.5099776387214661, + "learning_rate": 6.282069481971514e-07, + "loss": 0.4231, + "step": 697 + }, + { + "epoch": 4.777272727272727, + "grad_norm": 0.6147667169570923, + "learning_rate": 6.2151312771139e-07, + "loss": 0.4437, + "step": 698 + }, + { + "epoch": 4.784090909090909, + "grad_norm": 0.45040184259414673, + "learning_rate": 6.148500953954992e-07, + "loss": 0.4086, + "step": 699 + }, + { + "epoch": 4.790909090909091, + "grad_norm": 0.45515206456184387, + "learning_rate": 6.082179604557617e-07, + "loss": 0.4147, + "step": 700 + }, + { + "epoch": 4.797727272727273, + "grad_norm": 0.5819535255432129, + "learning_rate": 6.016168315920593e-07, + "loss": 0.4126, + "step": 701 + }, + { + "epoch": 4.804545454545455, + "grad_norm": 0.5001727342605591, + "learning_rate": 5.950468169960846e-07, + "loss": 0.4498, + "step": 702 + }, + { + "epoch": 4.8113636363636365, + "grad_norm": 0.5675827860832214, + "learning_rate": 5.885080243495731e-07, + "loss": 0.443, + "step": 703 + }, + { + "epoch": 4.818181818181818, + "grad_norm": 0.4939325749874115, + "learning_rate": 5.820005608225345e-07, + "loss": 0.4102, + "step": 704 + }, + { + "epoch": 4.825, + "grad_norm": 0.5586816072463989, + "learning_rate": 5.755245330715014e-07, + "loss": 0.4322, + "step": 705 + }, + { + "epoch": 4.831818181818182, + "grad_norm": 0.4761182367801666, + "learning_rate": 5.690800472377747e-07, + "loss": 0.4386, + "step": 706 + }, + { + "epoch": 4.838636363636364, + "grad_norm": 0.4821193218231201, + "learning_rate": 5.626672089456887e-07, + "loss": 0.4397, + "step": 707 + }, + { + "epoch": 4.845454545454546, + "grad_norm": 0.5768219828605652, + "learning_rate": 5.562861233008774e-07, + "loss": 0.4328, + "step": 708 + }, + { + "epoch": 4.8522727272727275, + "grad_norm": 0.49105873703956604, + "learning_rate": 5.499368948885528e-07, + "loss": 0.4331, + "step": 709 + }, + { + "epoch": 4.859090909090909, + "grad_norm": 0.5111908316612244, + "learning_rate": 5.436196277717928e-07, + "loss": 0.4165, + "step": 710 + }, + { + "epoch": 4.865909090909091, + "grad_norm": 0.49336686730384827, + "learning_rate": 5.373344254898313e-07, + "loss": 0.4326, + "step": 711 + }, + { + "epoch": 4.872727272727273, + "grad_norm": 0.5113787651062012, + "learning_rate": 5.310813910563645e-07, + "loss": 0.4259, + "step": 712 + }, + { + "epoch": 4.879545454545455, + "grad_norm": 1.0626885890960693, + "learning_rate": 5.24860626957861e-07, + "loss": 0.4279, + "step": 713 + }, + { + "epoch": 4.886363636363637, + "grad_norm": 0.5634323954582214, + "learning_rate": 5.186722351518822e-07, + "loss": 0.4315, + "step": 714 + }, + { + "epoch": 4.8931818181818185, + "grad_norm": 0.5127065181732178, + "learning_rate": 5.125163170654138e-07, + "loss": 0.4142, + "step": 715 + }, + { + "epoch": 4.9, + "grad_norm": 0.4855251610279083, + "learning_rate": 5.063929735931985e-07, + "loss": 0.4024, + "step": 716 + }, + { + "epoch": 4.906818181818182, + "grad_norm": 0.9671811461448669, + "learning_rate": 5.003023050960865e-07, + "loss": 0.4201, + "step": 717 + }, + { + "epoch": 4.913636363636364, + "grad_norm": 0.5237546563148499, + "learning_rate": 4.94244411399388e-07, + "loss": 0.4175, + "step": 718 + }, + { + "epoch": 4.920454545454545, + "grad_norm": 0.46599382162094116, + "learning_rate": 4.882193917912398e-07, + "loss": 0.4611, + "step": 719 + }, + { + "epoch": 4.927272727272728, + "grad_norm": 0.5131009221076965, + "learning_rate": 4.822273450209767e-07, + "loss": 0.4081, + "step": 720 + }, + { + "epoch": 4.934090909090909, + "grad_norm": 0.504088819026947, + "learning_rate": 4.7626836929751035e-07, + "loss": 0.4533, + "step": 721 + }, + { + "epoch": 4.940909090909091, + "grad_norm": 0.5421441197395325, + "learning_rate": 4.703425622877239e-07, + "loss": 0.4275, + "step": 722 + }, + { + "epoch": 4.947727272727272, + "grad_norm": 0.5401121377944946, + "learning_rate": 4.6445002111486866e-07, + "loss": 0.4449, + "step": 723 + }, + { + "epoch": 4.954545454545455, + "grad_norm": 0.46559831500053406, + "learning_rate": 4.5859084235697236e-07, + "loss": 0.4449, + "step": 724 + }, + { + "epoch": 4.961363636363636, + "grad_norm": 0.5192095041275024, + "learning_rate": 4.527651220452589e-07, + "loss": 0.4192, + "step": 725 + }, + { + "epoch": 4.968181818181818, + "grad_norm": 0.4992344379425049, + "learning_rate": 4.469729556625704e-07, + "loss": 0.4212, + "step": 726 + }, + { + "epoch": 4.975, + "grad_norm": 0.5019513964653015, + "learning_rate": 4.412144381418049e-07, + "loss": 0.4549, + "step": 727 + }, + { + "epoch": 4.9818181818181815, + "grad_norm": 0.47997716069221497, + "learning_rate": 4.354896638643591e-07, + "loss": 0.4397, + "step": 728 + }, + { + "epoch": 4.988636363636363, + "grad_norm": 0.5492618680000305, + "learning_rate": 4.2979872665858266e-07, + "loss": 0.4301, + "step": 729 + }, + { + "epoch": 4.995454545454545, + "grad_norm": 0.48399415612220764, + "learning_rate": 4.2414171979824e-07, + "loss": 0.4062, + "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..a03cebdd4a04fb2e9443a05c92f15b2b2db01e10 --- /dev/null +++ b/checkpoint-730/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90038f7ff377ce29f00aedab533b18bfc9515b30d2b4232d8b0d6de26ec3ea56 +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..fdf619c317c2fe82074662582dbd68166b6f9d50 --- /dev/null +++ b/checkpoint-876/README.md @@ -0,0 +1,202 @@ +--- +base_model: nvidia/Llama-3.1-Nemotron-70B-Instruct-HF +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..604978b603be594e46459a6c8c10d206a5a03b31 --- /dev/null +++ b/checkpoint-876/adapter_config.json @@ -0,0 +1,42 @@ +{ + "alpha_pattern": {}, + "auto_mapping": null, + "base_model_name_or_path": "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF", + "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": [ + "down_proj", + "v_proj", + "q_proj", + "gate_proj", + "o_proj", + "up_proj", + "k_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..90b76f45c94794a5d3b9e53430bdb322764d258d --- /dev/null +++ b/checkpoint-876/adapter_model.safetensors @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f148e162b556bb3d7b017226cf92103ac687f1d14e97e9344ddd35493aeb9d61 +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..199dfab44b11eb27ff7d09cdd577f4a6197d7320 --- /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:a3e21aa346562e9111bc8cfe60070e23981aa016528ba38f85fdf8958c57d1cd +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..bfc1025caa9476fa31691ab54d911af88a635c23 --- /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 sha256:685622461e030a97d054030f66166df9e910f4ff524c5ccdb7982537891f8a27 +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..dfae16ec7e2b98a23d991beb25006d90116e2f88 --- /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 sha256:9a3116abdaab8cd81053a0abb1b560dac12b91f94b4e3770a39abeb1023de4a3 +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..cd1ec6f1c6579e69a0f495ad5ad6db4bbb13fceb --- /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 sha256:7b18bf2e97bb5073a4b6cb036078503a125ceddfc766829bb51bc1c6261ec810 +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..fdde94c29816839ec3c29d6c6461206a49911f3c --- /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\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..d28d6f86b8ecd98caa90a196f092446d843ee657 --- /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": 2.020231246948242, + "learning_rate": 5.0000000000000004e-08, + "loss": 0.8736, + "step": 1 + }, + { + "epoch": 0.013636363636363636, + "grad_norm": 1.9368854761123657, + "learning_rate": 1.0000000000000001e-07, + "loss": 0.8543, + "step": 2 + }, + { + "epoch": 0.020454545454545454, + "grad_norm": 1.908379077911377, + "learning_rate": 1.5000000000000002e-07, + "loss": 0.8339, + "step": 3 + }, + { + "epoch": 0.02727272727272727, + "grad_norm": 1.9313658475875854, + "learning_rate": 2.0000000000000002e-07, + "loss": 0.834, + "step": 4 + }, + { + "epoch": 0.03409090909090909, + "grad_norm": 1.983262062072754, + "learning_rate": 2.5000000000000004e-07, + "loss": 0.8726, + "step": 5 + }, + { + "epoch": 0.04090909090909091, + "grad_norm": 1.9994115829467773, + "learning_rate": 3.0000000000000004e-07, + "loss": 0.8135, + "step": 6 + }, + { + "epoch": 0.04772727272727273, + "grad_norm": 1.8889026641845703, + "learning_rate": 3.5000000000000004e-07, + "loss": 0.8439, + "step": 7 + }, + { + "epoch": 0.05454545454545454, + "grad_norm": 1.9258317947387695, + "learning_rate": 4.0000000000000003e-07, + "loss": 0.84, + "step": 8 + }, + { + "epoch": 0.06136363636363636, + "grad_norm": 1.9654443264007568, + "learning_rate": 4.5000000000000003e-07, + "loss": 0.8609, + "step": 9 + }, + { + "epoch": 0.06818181818181818, + "grad_norm": 1.9278813600540161, + "learning_rate": 5.000000000000001e-07, + "loss": 0.8515, + "step": 10 + }, + { + "epoch": 0.075, + "grad_norm": 1.8698863983154297, + "learning_rate": 5.5e-07, + "loss": 0.8632, + "step": 11 + }, + { + "epoch": 0.08181818181818182, + "grad_norm": 2.7078285217285156, + "learning_rate": 6.000000000000001e-07, + "loss": 0.8643, + "step": 12 + }, + { + "epoch": 0.08863636363636364, + "grad_norm": 1.959272027015686, + "learning_rate": 6.5e-07, + "loss": 0.861, + "step": 13 + }, + { + "epoch": 0.09545454545454546, + "grad_norm": 1.8952852487564087, + "learning_rate": 7.000000000000001e-07, + "loss": 0.8599, + "step": 14 + }, + { + "epoch": 0.10227272727272728, + "grad_norm": 1.9469462633132935, + "learning_rate": 7.5e-07, + "loss": 0.8467, + "step": 15 + }, + { + "epoch": 0.10909090909090909, + "grad_norm": 2.0111300945281982, + "learning_rate": 8.000000000000001e-07, + "loss": 0.8078, + "step": 16 + }, + { + "epoch": 0.1159090909090909, + "grad_norm": 1.739009141921997, + "learning_rate": 8.500000000000001e-07, + "loss": 0.8114, + "step": 17 + }, + { + "epoch": 0.12272727272727273, + "grad_norm": 1.7578809261322021, + "learning_rate": 9.000000000000001e-07, + "loss": 0.8489, + "step": 18 + }, + { + "epoch": 0.12954545454545455, + "grad_norm": 1.9787132740020752, + "learning_rate": 9.500000000000001e-07, + "loss": 0.8052, + "step": 19 + }, + { + "epoch": 0.13636363636363635, + "grad_norm": 1.6380164623260498, + "learning_rate": 1.0000000000000002e-06, + "loss": 0.7882, + "step": 20 + }, + { + "epoch": 0.1431818181818182, + "grad_norm": 1.686877727508545, + "learning_rate": 1.0500000000000001e-06, + "loss": 0.8126, + "step": 21 + }, + { + "epoch": 0.15, + "grad_norm": 1.690590739250183, + "learning_rate": 1.1e-06, + "loss": 0.8248, + "step": 22 + }, + { + "epoch": 0.15681818181818183, + "grad_norm": 1.612255334854126, + "learning_rate": 1.1500000000000002e-06, + "loss": 0.8024, + "step": 23 + }, + { + "epoch": 0.16363636363636364, + "grad_norm": 1.6669739484786987, + "learning_rate": 1.2000000000000002e-06, + "loss": 0.818, + "step": 24 + }, + { + "epoch": 0.17045454545454544, + "grad_norm": 1.7726502418518066, + "learning_rate": 1.25e-06, + "loss": 0.8158, + "step": 25 + }, + { + "epoch": 0.17727272727272728, + "grad_norm": 1.5930036306381226, + "learning_rate": 1.3e-06, + "loss": 0.7886, + "step": 26 + }, + { + "epoch": 0.18409090909090908, + "grad_norm": 1.5480856895446777, + "learning_rate": 1.3500000000000002e-06, + "loss": 0.7747, + "step": 27 + }, + { + "epoch": 0.19090909090909092, + "grad_norm": 1.3537116050720215, + "learning_rate": 1.4000000000000001e-06, + "loss": 0.7891, + "step": 28 + }, + { + "epoch": 0.19772727272727272, + "grad_norm": 1.3412829637527466, + "learning_rate": 1.45e-06, + "loss": 0.7799, + "step": 29 + }, + { + "epoch": 0.20454545454545456, + "grad_norm": 1.237492322921753, + "learning_rate": 1.5e-06, + "loss": 0.7768, + "step": 30 + }, + { + "epoch": 0.21136363636363636, + "grad_norm": 1.3467835187911987, + "learning_rate": 1.5500000000000002e-06, + "loss": 0.7344, + "step": 31 + }, + { + "epoch": 0.21818181818181817, + "grad_norm": 1.1665314435958862, + "learning_rate": 1.6000000000000001e-06, + "loss": 0.7152, + "step": 32 + }, + { + "epoch": 0.225, + "grad_norm": 1.2461743354797363, + "learning_rate": 1.6500000000000003e-06, + "loss": 0.758, + "step": 33 + }, + { + "epoch": 0.2318181818181818, + "grad_norm": 1.0343059301376343, + "learning_rate": 1.7000000000000002e-06, + "loss": 0.7362, + "step": 34 + }, + { + "epoch": 0.23863636363636365, + "grad_norm": 1.112117052078247, + "learning_rate": 1.75e-06, + "loss": 0.7771, + "step": 35 + }, + { + "epoch": 0.24545454545454545, + "grad_norm": 1.104337453842163, + "learning_rate": 1.8000000000000001e-06, + "loss": 0.7479, + "step": 36 + }, + { + "epoch": 0.25227272727272726, + "grad_norm": 0.9324578642845154, + "learning_rate": 1.85e-06, + "loss": 0.6936, + "step": 37 + }, + { + "epoch": 0.2590909090909091, + "grad_norm": 1.0733048915863037, + "learning_rate": 1.9000000000000002e-06, + "loss": 0.7302, + "step": 38 + }, + { + "epoch": 0.26590909090909093, + "grad_norm": 0.8614453673362732, + "learning_rate": 1.9500000000000004e-06, + "loss": 0.7466, + "step": 39 + }, + { + "epoch": 0.2727272727272727, + "grad_norm": 0.7729659080505371, + "learning_rate": 2.0000000000000003e-06, + "loss": 0.725, + "step": 40 + }, + { + "epoch": 0.27954545454545454, + "grad_norm": 0.9341004490852356, + "learning_rate": 2.05e-06, + "loss": 0.6949, + "step": 41 + }, + { + "epoch": 0.2863636363636364, + "grad_norm": 0.6668925881385803, + "learning_rate": 2.1000000000000002e-06, + "loss": 0.6587, + "step": 42 + }, + { + "epoch": 0.29318181818181815, + "grad_norm": 0.681847333908081, + "learning_rate": 2.15e-06, + "loss": 0.6984, + "step": 43 + }, + { + "epoch": 0.3, + "grad_norm": 0.6222082376480103, + "learning_rate": 2.2e-06, + "loss": 0.7063, + "step": 44 + }, + { + "epoch": 0.3068181818181818, + "grad_norm": 0.790607213973999, + "learning_rate": 2.25e-06, + "loss": 0.6804, + "step": 45 + }, + { + "epoch": 0.31363636363636366, + "grad_norm": 0.6110427379608154, + "learning_rate": 2.3000000000000004e-06, + "loss": 0.6694, + "step": 46 + }, + { + "epoch": 0.32045454545454544, + "grad_norm": 0.6481993794441223, + "learning_rate": 2.35e-06, + "loss": 0.7321, + "step": 47 + }, + { + "epoch": 0.32727272727272727, + "grad_norm": 0.5973438620567322, + "learning_rate": 2.4000000000000003e-06, + "loss": 0.6508, + "step": 48 + }, + { + "epoch": 0.3340909090909091, + "grad_norm": 0.6252740621566772, + "learning_rate": 2.4500000000000003e-06, + "loss": 0.6911, + "step": 49 + }, + { + "epoch": 0.3409090909090909, + "grad_norm": 0.6826481819152832, + "learning_rate": 2.5e-06, + "loss": 0.6855, + "step": 50 + }, + { + "epoch": 0.3477272727272727, + "grad_norm": 0.6307975649833679, + "learning_rate": 2.55e-06, + "loss": 0.714, + "step": 51 + }, + { + "epoch": 0.35454545454545455, + "grad_norm": 0.5900976061820984, + "learning_rate": 2.6e-06, + "loss": 0.6619, + "step": 52 + }, + { + "epoch": 0.3613636363636364, + "grad_norm": 0.6203920841217041, + "learning_rate": 2.6500000000000005e-06, + "loss": 0.6822, + "step": 53 + }, + { + "epoch": 0.36818181818181817, + "grad_norm": 0.6197589039802551, + "learning_rate": 2.7000000000000004e-06, + "loss": 0.6584, + "step": 54 + }, + { + "epoch": 0.375, + "grad_norm": 0.5921110510826111, + "learning_rate": 2.7500000000000004e-06, + "loss": 0.6739, + "step": 55 + }, + { + "epoch": 0.38181818181818183, + "grad_norm": 0.6215619444847107, + "learning_rate": 2.8000000000000003e-06, + "loss": 0.6631, + "step": 56 + }, + { + "epoch": 0.3886363636363636, + "grad_norm": 0.5495648980140686, + "learning_rate": 2.85e-06, + "loss": 0.6531, + "step": 57 + }, + { + "epoch": 0.39545454545454545, + "grad_norm": 0.531947910785675, + "learning_rate": 2.9e-06, + "loss": 0.6365, + "step": 58 + }, + { + "epoch": 0.4022727272727273, + "grad_norm": 0.559112548828125, + "learning_rate": 2.95e-06, + "loss": 0.6391, + "step": 59 + }, + { + "epoch": 0.4090909090909091, + "grad_norm": 0.5377560257911682, + "learning_rate": 3e-06, + "loss": 0.6775, + "step": 60 + }, + { + "epoch": 0.4159090909090909, + "grad_norm": 0.5167352557182312, + "learning_rate": 3.05e-06, + "loss": 0.6518, + "step": 61 + }, + { + "epoch": 0.42272727272727273, + "grad_norm": 0.5483390092849731, + "learning_rate": 3.1000000000000004e-06, + "loss": 0.5972, + "step": 62 + }, + { + "epoch": 0.42954545454545456, + "grad_norm": 0.5687050223350525, + "learning_rate": 3.1500000000000003e-06, + "loss": 0.6515, + "step": 63 + }, + { + "epoch": 0.43636363636363634, + "grad_norm": 0.9046968817710876, + "learning_rate": 3.2000000000000003e-06, + "loss": 0.6497, + "step": 64 + }, + { + "epoch": 0.4431818181818182, + "grad_norm": 0.5231667757034302, + "learning_rate": 3.2500000000000002e-06, + "loss": 0.6364, + "step": 65 + }, + { + "epoch": 0.45, + "grad_norm": 0.49699342250823975, + "learning_rate": 3.3000000000000006e-06, + "loss": 0.6144, + "step": 66 + }, + { + "epoch": 0.45681818181818185, + "grad_norm": 0.5390080213546753, + "learning_rate": 3.3500000000000005e-06, + "loss": 0.6051, + "step": 67 + }, + { + "epoch": 0.4636363636363636, + "grad_norm": 0.5252938270568848, + "learning_rate": 3.4000000000000005e-06, + "loss": 0.6353, + "step": 68 + }, + { + "epoch": 0.47045454545454546, + "grad_norm": 0.6291780471801758, + "learning_rate": 3.45e-06, + "loss": 0.6326, + "step": 69 + }, + { + "epoch": 0.4772727272727273, + "grad_norm": 0.5545375943183899, + "learning_rate": 3.5e-06, + "loss": 0.647, + "step": 70 + }, + { + "epoch": 0.48409090909090907, + "grad_norm": 0.47775956988334656, + "learning_rate": 3.5500000000000003e-06, + "loss": 0.5993, + "step": 71 + }, + { + "epoch": 0.4909090909090909, + "grad_norm": 0.5016375184059143, + "learning_rate": 3.6000000000000003e-06, + "loss": 0.6056, + "step": 72 + }, + { + "epoch": 0.49772727272727274, + "grad_norm": 0.4780517816543579, + "learning_rate": 3.65e-06, + "loss": 0.6189, + "step": 73 + }, + { + "epoch": 0.5045454545454545, + "grad_norm": 0.5763716697692871, + "learning_rate": 3.7e-06, + "loss": 0.6262, + "step": 74 + }, + { + "epoch": 0.5113636363636364, + "grad_norm": 0.5321404337882996, + "learning_rate": 3.7500000000000005e-06, + "loss": 0.6336, + "step": 75 + }, + { + "epoch": 0.5181818181818182, + "grad_norm": 0.4641991853713989, + "learning_rate": 3.8000000000000005e-06, + "loss": 0.6317, + "step": 76 + }, + { + "epoch": 0.525, + "grad_norm": 0.45118412375450134, + "learning_rate": 3.85e-06, + "loss": 0.5927, + "step": 77 + }, + { + "epoch": 0.5318181818181819, + "grad_norm": 0.5500645041465759, + "learning_rate": 3.900000000000001e-06, + "loss": 0.62, + "step": 78 + }, + { + "epoch": 0.5386363636363637, + "grad_norm": 0.5197616219520569, + "learning_rate": 3.95e-06, + "loss": 0.6063, + "step": 79 + }, + { + "epoch": 0.5454545454545454, + "grad_norm": 0.4672992527484894, + "learning_rate": 4.000000000000001e-06, + "loss": 0.6156, + "step": 80 + }, + { + "epoch": 0.5522727272727272, + "grad_norm": 0.5995073318481445, + "learning_rate": 4.05e-06, + "loss": 0.6219, + "step": 81 + }, + { + "epoch": 0.5590909090909091, + "grad_norm": 0.5593804717063904, + "learning_rate": 4.1e-06, + "loss": 0.6282, + "step": 82 + }, + { + "epoch": 0.5659090909090909, + "grad_norm": 0.7544977068901062, + "learning_rate": 4.15e-06, + "loss": 0.6246, + "step": 83 + }, + { + "epoch": 0.5727272727272728, + "grad_norm": 0.45378655195236206, + "learning_rate": 4.2000000000000004e-06, + "loss": 0.5881, + "step": 84 + }, + { + "epoch": 0.5795454545454546, + "grad_norm": 0.4407907724380493, + "learning_rate": 4.25e-06, + "loss": 0.6026, + "step": 85 + }, + { + "epoch": 0.5863636363636363, + "grad_norm": 0.4333001673221588, + "learning_rate": 4.3e-06, + "loss": 0.6098, + "step": 86 + }, + { + "epoch": 0.5931818181818181, + "grad_norm": 0.45020997524261475, + "learning_rate": 4.350000000000001e-06, + "loss": 0.5858, + "step": 87 + }, + { + "epoch": 0.6, + "grad_norm": 0.424003541469574, + "learning_rate": 4.4e-06, + "loss": 0.6186, + "step": 88 + }, + { + "epoch": 0.6068181818181818, + "grad_norm": 0.427081435918808, + "learning_rate": 4.450000000000001e-06, + "loss": 0.6164, + "step": 89 + }, + { + "epoch": 0.6136363636363636, + "grad_norm": 0.4430546164512634, + "learning_rate": 4.5e-06, + "loss": 0.6049, + "step": 90 + }, + { + "epoch": 0.6204545454545455, + "grad_norm": 0.5469494462013245, + "learning_rate": 4.5500000000000005e-06, + "loss": 0.608, + "step": 91 + }, + { + "epoch": 0.6272727272727273, + "grad_norm": 0.4481559693813324, + "learning_rate": 4.600000000000001e-06, + "loss": 0.5532, + "step": 92 + }, + { + "epoch": 0.634090909090909, + "grad_norm": 0.6563135981559753, + "learning_rate": 4.65e-06, + "loss": 0.6105, + "step": 93 + }, + { + "epoch": 0.6409090909090909, + "grad_norm": 0.4572807252407074, + "learning_rate": 4.7e-06, + "loss": 0.5976, + "step": 94 + }, + { + "epoch": 0.6477272727272727, + "grad_norm": 0.4423324167728424, + "learning_rate": 4.75e-06, + "loss": 0.616, + "step": 95 + }, + { + "epoch": 0.6545454545454545, + "grad_norm": 0.6230632066726685, + "learning_rate": 4.800000000000001e-06, + "loss": 0.5879, + "step": 96 + }, + { + "epoch": 0.6613636363636364, + "grad_norm": 0.461467981338501, + "learning_rate": 4.85e-06, + "loss": 0.5992, + "step": 97 + }, + { + "epoch": 0.6681818181818182, + "grad_norm": 0.44455233216285706, + "learning_rate": 4.9000000000000005e-06, + "loss": 0.6138, + "step": 98 + }, + { + "epoch": 0.675, + "grad_norm": 0.47281715273857117, + "learning_rate": 4.95e-06, + "loss": 0.59, + "step": 99 + }, + { + "epoch": 0.6818181818181818, + "grad_norm": 0.46114516258239746, + "learning_rate": 5e-06, + "loss": 0.6216, + "step": 100 + }, + { + "epoch": 0.6886363636363636, + "grad_norm": 0.49900761246681213, + "learning_rate": 4.9999795126530275e-06, + "loss": 0.6222, + "step": 101 + }, + { + "epoch": 0.6954545454545454, + "grad_norm": 0.4764839708805084, + "learning_rate": 4.999918050947891e-06, + "loss": 0.5829, + "step": 102 + }, + { + "epoch": 0.7022727272727273, + "grad_norm": 0.43502742052078247, + "learning_rate": 4.999815615891943e-06, + "loss": 0.5982, + "step": 103 + }, + { + "epoch": 0.7090909090909091, + "grad_norm": 0.43318599462509155, + "learning_rate": 4.9996722091640805e-06, + "loss": 0.5827, + "step": 104 + }, + { + "epoch": 0.7159090909090909, + "grad_norm": 0.47218361496925354, + "learning_rate": 4.9994878331147225e-06, + "loss": 0.5907, + "step": 105 + }, + { + "epoch": 0.7227272727272728, + "grad_norm": 0.48492228984832764, + "learning_rate": 4.99926249076577e-06, + "loss": 0.5459, + "step": 106 + }, + { + "epoch": 0.7295454545454545, + "grad_norm": 0.4019850194454193, + "learning_rate": 4.998996185810557e-06, + "loss": 0.5957, + "step": 107 + }, + { + "epoch": 0.7363636363636363, + "grad_norm": 0.47361984848976135, + "learning_rate": 4.998688922613788e-06, + "loss": 0.579, + "step": 108 + }, + { + "epoch": 0.7431818181818182, + "grad_norm": 0.4612269401550293, + "learning_rate": 4.9983407062114695e-06, + "loss": 0.5814, + "step": 109 + }, + { + "epoch": 0.75, + "grad_norm": 0.45558398962020874, + "learning_rate": 4.9979515423108255e-06, + "loss": 0.6036, + "step": 110 + }, + { + "epoch": 0.7568181818181818, + "grad_norm": 0.41492199897766113, + "learning_rate": 4.997521437290205e-06, + "loss": 0.5891, + "step": 111 + }, + { + "epoch": 0.7636363636363637, + "grad_norm": 0.4484409689903259, + "learning_rate": 4.997050398198977e-06, + "loss": 0.6031, + "step": 112 + }, + { + "epoch": 0.7704545454545455, + "grad_norm": 0.42671382427215576, + "learning_rate": 4.996538432757414e-06, + "loss": 0.6106, + "step": 113 + }, + { + "epoch": 0.7772727272727272, + "grad_norm": 0.4222075343132019, + "learning_rate": 4.995985549356568e-06, + "loss": 0.5774, + "step": 114 + }, + { + "epoch": 0.7840909090909091, + "grad_norm": 0.421021431684494, + "learning_rate": 4.995391757058129e-06, + "loss": 0.5754, + "step": 115 + }, + { + "epoch": 0.7909090909090909, + "grad_norm": 0.4215952754020691, + "learning_rate": 4.99475706559428e-06, + "loss": 0.5895, + "step": 116 + }, + { + "epoch": 0.7977272727272727, + "grad_norm": 0.42652982473373413, + "learning_rate": 4.994081485367537e-06, + "loss": 0.5646, + "step": 117 + }, + { + "epoch": 0.8045454545454546, + "grad_norm": 0.4145742654800415, + "learning_rate": 4.993365027450576e-06, + "loss": 0.5862, + "step": 118 + }, + { + "epoch": 0.8113636363636364, + "grad_norm": 0.4754612445831299, + "learning_rate": 4.992607703586058e-06, + "loss": 0.5645, + "step": 119 + }, + { + "epoch": 0.8181818181818182, + "grad_norm": 0.4694959223270416, + "learning_rate": 4.991809526186424e-06, + "loss": 0.5989, + "step": 120 + }, + { + "epoch": 0.825, + "grad_norm": 0.4573237895965576, + "learning_rate": 4.990970508333707e-06, + "loss": 0.5769, + "step": 121 + }, + { + "epoch": 0.8318181818181818, + "grad_norm": 0.4592607915401459, + "learning_rate": 4.990090663779305e-06, + "loss": 0.5529, + "step": 122 + }, + { + "epoch": 0.8386363636363636, + "grad_norm": 0.4325920045375824, + "learning_rate": 4.9891700069437635e-06, + "loss": 0.5846, + "step": 123 + }, + { + "epoch": 0.8454545454545455, + "grad_norm": 0.44623467326164246, + "learning_rate": 4.988208552916535e-06, + "loss": 0.5997, + "step": 124 + }, + { + "epoch": 0.8522727272727273, + "grad_norm": 0.46071872115135193, + "learning_rate": 4.987206317455734e-06, + "loss": 0.5687, + "step": 125 + }, + { + "epoch": 0.8590909090909091, + "grad_norm": 0.43121206760406494, + "learning_rate": 4.986163316987877e-06, + "loss": 0.5277, + "step": 126 + }, + { + "epoch": 0.865909090909091, + "grad_norm": 0.4711630046367645, + "learning_rate": 4.985079568607613e-06, + "loss": 0.592, + "step": 127 + }, + { + "epoch": 0.8727272727272727, + "grad_norm": 0.4450221657752991, + "learning_rate": 4.983955090077445e-06, + "loss": 0.5656, + "step": 128 + }, + { + "epoch": 0.8795454545454545, + "grad_norm": 0.6546564102172852, + "learning_rate": 4.982789899827439e-06, + "loss": 0.572, + "step": 129 + }, + { + "epoch": 0.8863636363636364, + "grad_norm": 0.49622273445129395, + "learning_rate": 4.9815840169549216e-06, + "loss": 0.5719, + "step": 130 + }, + { + "epoch": 0.8931818181818182, + "grad_norm": 0.5478544235229492, + "learning_rate": 4.980337461224164e-06, + "loss": 0.5774, + "step": 131 + }, + { + "epoch": 0.9, + "grad_norm": 0.4340018033981323, + "learning_rate": 4.979050253066064e-06, + "loss": 0.5444, + "step": 132 + }, + { + "epoch": 0.9068181818181819, + "grad_norm": 0.4243793189525604, + "learning_rate": 4.977722413577802e-06, + "loss": 0.5759, + "step": 133 + }, + { + "epoch": 0.9136363636363637, + "grad_norm": 0.45201942324638367, + "learning_rate": 4.976353964522509e-06, + "loss": 0.5746, + "step": 134 + }, + { + "epoch": 0.9204545454545454, + "grad_norm": 0.42892220616340637, + "learning_rate": 4.974944928328894e-06, + "loss": 0.544, + "step": 135 + }, + { + "epoch": 0.9272727272727272, + "grad_norm": 0.43185824155807495, + "learning_rate": 4.973495328090891e-06, + "loss": 0.5803, + "step": 136 + }, + { + "epoch": 0.9340909090909091, + "grad_norm": 0.5512542724609375, + "learning_rate": 4.972005187567267e-06, + "loss": 0.5694, + "step": 137 + }, + { + "epoch": 0.9409090909090909, + "grad_norm": 0.4836059808731079, + "learning_rate": 4.970474531181245e-06, + "loss": 0.5581, + "step": 138 + }, + { + "epoch": 0.9477272727272728, + "grad_norm": 0.4508282244205475, + "learning_rate": 4.968903384020095e-06, + "loss": 0.5553, + "step": 139 + }, + { + "epoch": 0.9545454545454546, + "grad_norm": 0.4764733612537384, + "learning_rate": 4.967291771834727e-06, + "loss": 0.5798, + "step": 140 + }, + { + "epoch": 0.9613636363636363, + "grad_norm": 0.463468074798584, + "learning_rate": 4.965639721039267e-06, + "loss": 0.559, + "step": 141 + }, + { + "epoch": 0.9681818181818181, + "grad_norm": 0.4655650854110718, + "learning_rate": 4.963947258710626e-06, + "loss": 0.5576, + "step": 142 + }, + { + "epoch": 0.975, + "grad_norm": 0.4651234745979309, + "learning_rate": 4.962214412588053e-06, + "loss": 0.5792, + "step": 143 + }, + { + "epoch": 0.9818181818181818, + "grad_norm": 0.43080538511276245, + "learning_rate": 4.960441211072686e-06, + "loss": 0.5314, + "step": 144 + }, + { + "epoch": 0.9886363636363636, + "grad_norm": 0.4403238594532013, + "learning_rate": 4.9586276832270785e-06, + "loss": 0.5592, + "step": 145 + }, + { + "epoch": 0.9954545454545455, + "grad_norm": 0.43968647718429565, + "learning_rate": 4.9567738587747314e-06, + "loss": 0.5573, + "step": 146 + }, + { + "epoch": 1.0068181818181818, + "grad_norm": 0.8337184190750122, + "learning_rate": 4.954879768099599e-06, + "loss": 1.1118, + "step": 147 + }, + { + "epoch": 1.0136363636363637, + "grad_norm": 0.4042762815952301, + "learning_rate": 4.952945442245598e-06, + "loss": 0.5731, + "step": 148 + }, + { + "epoch": 1.0204545454545455, + "grad_norm": 0.5666297078132629, + "learning_rate": 4.95097091291609e-06, + "loss": 0.5415, + "step": 149 + }, + { + "epoch": 1.0272727272727273, + "grad_norm": 0.6148372888565063, + "learning_rate": 4.948956212473371e-06, + "loss": 0.5486, + "step": 150 + }, + { + "epoch": 1.0340909090909092, + "grad_norm": 0.5780720710754395, + "learning_rate": 4.946901373938132e-06, + "loss": 0.5373, + "step": 151 + }, + { + "epoch": 1.040909090909091, + "grad_norm": 0.4648835361003876, + "learning_rate": 4.944806430988927e-06, + "loss": 0.5419, + "step": 152 + }, + { + "epoch": 1.0477272727272728, + "grad_norm": 0.41131219267845154, + "learning_rate": 4.942671417961615e-06, + "loss": 0.5514, + "step": 153 + }, + { + "epoch": 1.0545454545454545, + "grad_norm": 0.4733349084854126, + "learning_rate": 4.940496369848795e-06, + "loss": 0.5563, + "step": 154 + }, + { + "epoch": 1.0613636363636363, + "grad_norm": 0.41179028153419495, + "learning_rate": 4.938281322299243e-06, + "loss": 0.5413, + "step": 155 + }, + { + "epoch": 1.0681818181818181, + "grad_norm": 0.4770941138267517, + "learning_rate": 4.936026311617316e-06, + "loss": 0.5561, + "step": 156 + }, + { + "epoch": 1.075, + "grad_norm": 0.7752397060394287, + "learning_rate": 4.933731374762361e-06, + "loss": 0.5439, + "step": 157 + }, + { + "epoch": 1.0818181818181818, + "grad_norm": 0.4323958456516266, + "learning_rate": 4.931396549348115e-06, + "loss": 0.5401, + "step": 158 + }, + { + "epoch": 1.0886363636363636, + "grad_norm": 0.4497937858104706, + "learning_rate": 4.9290218736420795e-06, + "loss": 0.523, + "step": 159 + }, + { + "epoch": 1.0954545454545455, + "grad_norm": 0.4881570339202881, + "learning_rate": 4.926607386564898e-06, + "loss": 0.5441, + "step": 160 + }, + { + "epoch": 1.1022727272727273, + "grad_norm": 0.43357932567596436, + "learning_rate": 4.9241531276897196e-06, + "loss": 0.5499, + "step": 161 + }, + { + "epoch": 1.1090909090909091, + "grad_norm": 1.1125361919403076, + "learning_rate": 4.921659137241544e-06, + "loss": 0.5265, + "step": 162 + }, + { + "epoch": 1.115909090909091, + "grad_norm": 0.43150508403778076, + "learning_rate": 4.919125456096574e-06, + "loss": 0.5193, + "step": 163 + }, + { + "epoch": 1.1227272727272728, + "grad_norm": 0.40771809220314026, + "learning_rate": 4.916552125781529e-06, + "loss": 0.538, + "step": 164 + }, + { + "epoch": 1.1295454545454546, + "grad_norm": 0.5830567479133606, + "learning_rate": 4.913939188472979e-06, + "loss": 0.5548, + "step": 165 + }, + { + "epoch": 1.1363636363636362, + "grad_norm": 0.4776526391506195, + "learning_rate": 4.911286686996648e-06, + "loss": 0.5328, + "step": 166 + }, + { + "epoch": 1.143181818181818, + "grad_norm": 0.41657742857933044, + "learning_rate": 4.908594664826708e-06, + "loss": 0.5336, + "step": 167 + }, + { + "epoch": 1.15, + "grad_norm": 0.4651948809623718, + "learning_rate": 4.905863166085076e-06, + "loss": 0.5812, + "step": 168 + }, + { + "epoch": 1.1568181818181817, + "grad_norm": 0.4399661719799042, + "learning_rate": 4.903092235540679e-06, + "loss": 0.5143, + "step": 169 + }, + { + "epoch": 1.1636363636363636, + "grad_norm": 0.4548935294151306, + "learning_rate": 4.900281918608732e-06, + "loss": 0.5425, + "step": 170 + }, + { + "epoch": 1.1704545454545454, + "grad_norm": 0.43515875935554504, + "learning_rate": 4.897432261349984e-06, + "loss": 0.534, + "step": 171 + }, + { + "epoch": 1.1772727272727272, + "grad_norm": 0.42954760789871216, + "learning_rate": 4.894543310469968e-06, + "loss": 0.5436, + "step": 172 + }, + { + "epoch": 1.184090909090909, + "grad_norm": 0.4063640534877777, + "learning_rate": 4.891615113318236e-06, + "loss": 0.5481, + "step": 173 + }, + { + "epoch": 1.190909090909091, + "grad_norm": 0.49416476488113403, + "learning_rate": 4.888647717887582e-06, + "loss": 0.5298, + "step": 174 + }, + { + "epoch": 1.1977272727272728, + "grad_norm": 1.4082964658737183, + "learning_rate": 4.8856411728132526e-06, + "loss": 0.5366, + "step": 175 + }, + { + "epoch": 1.2045454545454546, + "grad_norm": 0.44356465339660645, + "learning_rate": 4.8825955273721524e-06, + "loss": 0.5563, + "step": 176 + }, + { + "epoch": 1.2113636363636364, + "grad_norm": 0.40925371646881104, + "learning_rate": 4.879510831482039e-06, + "loss": 0.5444, + "step": 177 + }, + { + "epoch": 1.2181818181818183, + "grad_norm": 0.44379451870918274, + "learning_rate": 4.876387135700701e-06, + "loss": 0.5311, + "step": 178 + }, + { + "epoch": 1.225, + "grad_norm": 0.41487446427345276, + "learning_rate": 4.873224491225128e-06, + "loss": 0.5637, + "step": 179 + }, + { + "epoch": 1.231818181818182, + "grad_norm": 0.4520554542541504, + "learning_rate": 4.870022949890676e-06, + "loss": 0.5677, + "step": 180 + }, + { + "epoch": 1.2386363636363638, + "grad_norm": 0.4528166949748993, + "learning_rate": 4.866782564170217e-06, + "loss": 0.5095, + "step": 181 + }, + { + "epoch": 1.2454545454545454, + "grad_norm": 0.4409842789173126, + "learning_rate": 4.863503387173276e-06, + "loss": 0.5423, + "step": 182 + }, + { + "epoch": 1.2522727272727272, + "grad_norm": 0.4355996251106262, + "learning_rate": 4.860185472645161e-06, + "loss": 0.5422, + "step": 183 + }, + { + "epoch": 1.259090909090909, + "grad_norm": 0.4871126115322113, + "learning_rate": 4.856828874966086e-06, + "loss": 0.5293, + "step": 184 + }, + { + "epoch": 1.2659090909090909, + "grad_norm": 0.442348450422287, + "learning_rate": 4.853433649150276e-06, + "loss": 0.5616, + "step": 185 + }, + { + "epoch": 1.2727272727272727, + "grad_norm": 0.44014668464660645, + "learning_rate": 4.849999850845066e-06, + "loss": 0.5354, + "step": 186 + }, + { + "epoch": 1.2795454545454545, + "grad_norm": 0.4496249854564667, + "learning_rate": 4.8465275363299905e-06, + "loss": 0.5298, + "step": 187 + }, + { + "epoch": 1.2863636363636364, + "grad_norm": 0.46060171723365784, + "learning_rate": 4.84301676251586e-06, + "loss": 0.554, + "step": 188 + }, + { + "epoch": 1.2931818181818182, + "grad_norm": 0.7829347252845764, + "learning_rate": 4.839467586943825e-06, + "loss": 0.5606, + "step": 189 + }, + { + "epoch": 1.3, + "grad_norm": 0.45566731691360474, + "learning_rate": 4.835880067784441e-06, + "loss": 0.5118, + "step": 190 + }, + { + "epoch": 1.3068181818181819, + "grad_norm": 0.43211469054222107, + "learning_rate": 4.832254263836708e-06, + "loss": 0.535, + "step": 191 + }, + { + "epoch": 1.3136363636363637, + "grad_norm": 0.4350590705871582, + "learning_rate": 4.828590234527107e-06, + "loss": 0.5595, + "step": 192 + }, + { + "epoch": 1.3204545454545453, + "grad_norm": 0.4056399464607239, + "learning_rate": 4.82488803990863e-06, + "loss": 0.55, + "step": 193 + }, + { + "epoch": 1.3272727272727272, + "grad_norm": 0.4458341598510742, + "learning_rate": 4.821147740659795e-06, + "loss": 0.501, + "step": 194 + }, + { + "epoch": 1.334090909090909, + "grad_norm": 0.446504145860672, + "learning_rate": 4.817369398083648e-06, + "loss": 0.5393, + "step": 195 + }, + { + "epoch": 1.3409090909090908, + "grad_norm": 0.47838401794433594, + "learning_rate": 4.813553074106761e-06, + "loss": 0.5354, + "step": 196 + }, + { + "epoch": 1.3477272727272727, + "grad_norm": 0.4675792157649994, + "learning_rate": 4.809698831278217e-06, + "loss": 0.5202, + "step": 197 + }, + { + "epoch": 1.3545454545454545, + "grad_norm": 0.4351370632648468, + "learning_rate": 4.805806732768585e-06, + "loss": 0.5256, + "step": 198 + }, + { + "epoch": 1.3613636363636363, + "grad_norm": 0.4466511607170105, + "learning_rate": 4.801876842368882e-06, + "loss": 0.5238, + "step": 199 + }, + { + "epoch": 1.3681818181818182, + "grad_norm": 0.45672544836997986, + "learning_rate": 4.797909224489531e-06, + "loss": 0.5426, + "step": 200 + }, + { + "epoch": 1.375, + "grad_norm": 0.422638863325119, + "learning_rate": 4.793903944159303e-06, + "loss": 0.5385, + "step": 201 + }, + { + "epoch": 1.3818181818181818, + "grad_norm": 0.4254551827907562, + "learning_rate": 4.789861067024253e-06, + "loss": 0.503, + "step": 202 + }, + { + "epoch": 1.3886363636363637, + "grad_norm": 0.43860334157943726, + "learning_rate": 4.785780659346642e-06, + "loss": 0.519, + "step": 203 + }, + { + "epoch": 1.3954545454545455, + "grad_norm": 0.43294480443000793, + "learning_rate": 4.781662788003851e-06, + "loss": 0.5537, + "step": 204 + }, + { + "epoch": 1.4022727272727273, + "grad_norm": 0.45006492733955383, + "learning_rate": 4.777507520487289e-06, + "loss": 0.5318, + "step": 205 + }, + { + "epoch": 1.4090909090909092, + "grad_norm": 0.4178854823112488, + "learning_rate": 4.773314924901281e-06, + "loss": 0.5287, + "step": 206 + }, + { + "epoch": 1.415909090909091, + "grad_norm": 0.45662209391593933, + "learning_rate": 4.769085069961955e-06, + "loss": 0.5609, + "step": 207 + }, + { + "epoch": 1.4227272727272728, + "grad_norm": 0.41578733921051025, + "learning_rate": 4.764818024996117e-06, + "loss": 0.5249, + "step": 208 + }, + { + "epoch": 1.4295454545454547, + "grad_norm": 0.48305758833885193, + "learning_rate": 4.760513859940112e-06, + "loss": 0.515, + "step": 209 + }, + { + "epoch": 1.4363636363636363, + "grad_norm": 0.4564506709575653, + "learning_rate": 4.756172645338675e-06, + "loss": 0.535, + "step": 210 + }, + { + "epoch": 1.4431818181818181, + "grad_norm": 0.4499240815639496, + "learning_rate": 4.751794452343785e-06, + "loss": 0.5236, + "step": 211 + }, + { + "epoch": 1.45, + "grad_norm": 0.41851088404655457, + "learning_rate": 4.747379352713489e-06, + "loss": 0.5254, + "step": 212 + }, + { + "epoch": 1.4568181818181818, + "grad_norm": 0.4690937399864197, + "learning_rate": 4.7429274188107275e-06, + "loss": 0.5312, + "step": 213 + }, + { + "epoch": 1.4636363636363636, + "grad_norm": 0.5516958832740784, + "learning_rate": 4.738438723602154e-06, + "loss": 0.5315, + "step": 214 + }, + { + "epoch": 1.4704545454545455, + "grad_norm": 1.2040799856185913, + "learning_rate": 4.733913340656933e-06, + "loss": 0.544, + "step": 215 + }, + { + "epoch": 1.4772727272727273, + "grad_norm": 0.445425808429718, + "learning_rate": 4.729351344145536e-06, + "loss": 0.5309, + "step": 216 + }, + { + "epoch": 1.4840909090909091, + "grad_norm": 0.4186754822731018, + "learning_rate": 4.7247528088385296e-06, + "loss": 0.5138, + "step": 217 + }, + { + "epoch": 1.490909090909091, + "grad_norm": 0.43284520506858826, + "learning_rate": 4.720117810105341e-06, + "loss": 0.521, + "step": 218 + }, + { + "epoch": 1.4977272727272728, + "grad_norm": 0.4308505058288574, + "learning_rate": 4.715446423913036e-06, + "loss": 0.5436, + "step": 219 + }, + { + "epoch": 1.5045454545454544, + "grad_norm": 0.4471941888332367, + "learning_rate": 4.710738726825059e-06, + "loss": 0.5528, + "step": 220 + }, + { + "epoch": 1.5113636363636362, + "grad_norm": 0.4374965727329254, + "learning_rate": 4.705994795999991e-06, + "loss": 0.5238, + "step": 221 + }, + { + "epoch": 1.518181818181818, + "grad_norm": 0.6026428937911987, + "learning_rate": 4.701214709190277e-06, + "loss": 0.5268, + "step": 222 + }, + { + "epoch": 1.525, + "grad_norm": 0.4413348138332367, + "learning_rate": 4.696398544740955e-06, + "loss": 0.5534, + "step": 223 + }, + { + "epoch": 1.5318181818181817, + "grad_norm": 0.45917898416519165, + "learning_rate": 4.69154638158837e-06, + "loss": 0.511, + "step": 224 + }, + { + "epoch": 1.5386363636363636, + "grad_norm": 0.5339583158493042, + "learning_rate": 4.686658299258881e-06, + "loss": 0.5395, + "step": 225 + }, + { + "epoch": 1.5454545454545454, + "grad_norm": 0.4617757201194763, + "learning_rate": 4.681734377867562e-06, + "loss": 0.5425, + "step": 226 + }, + { + "epoch": 1.5522727272727272, + "grad_norm": 0.4513031840324402, + "learning_rate": 4.67677469811688e-06, + "loss": 0.5241, + "step": 227 + }, + { + "epoch": 1.559090909090909, + "grad_norm": 0.43548423051834106, + "learning_rate": 4.671779341295378e-06, + "loss": 0.5224, + "step": 228 + }, + { + "epoch": 1.565909090909091, + "grad_norm": 0.5516494512557983, + "learning_rate": 4.666748389276344e-06, + "loss": 0.5346, + "step": 229 + }, + { + "epoch": 1.5727272727272728, + "grad_norm": 0.4441164433956146, + "learning_rate": 4.661681924516466e-06, + "loss": 0.5208, + "step": 230 + }, + { + "epoch": 1.5795454545454546, + "grad_norm": 0.4057452380657196, + "learning_rate": 4.6565800300544805e-06, + "loss": 0.5173, + "step": 231 + }, + { + "epoch": 1.5863636363636364, + "grad_norm": 0.4162004590034485, + "learning_rate": 4.651442789509813e-06, + "loss": 0.5307, + "step": 232 + }, + { + "epoch": 1.5931818181818183, + "grad_norm": 0.43314146995544434, + "learning_rate": 4.646270287081208e-06, + "loss": 0.5359, + "step": 233 + }, + { + "epoch": 1.6, + "grad_norm": 0.5786082148551941, + "learning_rate": 4.641062607545347e-06, + "loss": 0.5307, + "step": 234 + }, + { + "epoch": 1.606818181818182, + "grad_norm": 0.43811941146850586, + "learning_rate": 4.6358198362554585e-06, + "loss": 0.5556, + "step": 235 + }, + { + "epoch": 1.6136363636363638, + "grad_norm": 0.4374074935913086, + "learning_rate": 4.630542059139923e-06, + "loss": 0.5253, + "step": 236 + }, + { + "epoch": 1.6204545454545456, + "grad_norm": 0.4387815594673157, + "learning_rate": 4.625229362700863e-06, + "loss": 0.534, + "step": 237 + }, + { + "epoch": 1.6272727272727274, + "grad_norm": 0.4976162016391754, + "learning_rate": 4.61988183401272e-06, + "loss": 0.5255, + "step": 238 + }, + { + "epoch": 1.634090909090909, + "grad_norm": 0.4603639543056488, + "learning_rate": 4.614499560720837e-06, + "loss": 0.511, + "step": 239 + }, + { + "epoch": 1.6409090909090909, + "grad_norm": 0.43136921525001526, + "learning_rate": 4.609082631040012e-06, + "loss": 0.5099, + "step": 240 + }, + { + "epoch": 1.6477272727272727, + "grad_norm": 0.4573090970516205, + "learning_rate": 4.603631133753061e-06, + "loss": 0.5515, + "step": 241 + }, + { + "epoch": 1.6545454545454545, + "grad_norm": 0.4292807877063751, + "learning_rate": 4.598145158209356e-06, + "loss": 0.5048, + "step": 242 + }, + { + "epoch": 1.6613636363636364, + "grad_norm": 0.6377942562103271, + "learning_rate": 4.592624794323366e-06, + "loss": 0.5333, + "step": 243 + }, + { + "epoch": 1.6681818181818182, + "grad_norm": 0.49924615025520325, + "learning_rate": 4.587070132573178e-06, + "loss": 0.5127, + "step": 244 + }, + { + "epoch": 1.675, + "grad_norm": 0.4723946154117584, + "learning_rate": 4.581481263999019e-06, + "loss": 0.5127, + "step": 245 + }, + { + "epoch": 1.6818181818181817, + "grad_norm": 1.1846898794174194, + "learning_rate": 4.575858280201761e-06, + "loss": 0.5367, + "step": 246 + }, + { + "epoch": 1.6886363636363635, + "grad_norm": 0.4306166172027588, + "learning_rate": 4.570201273341418e-06, + "loss": 0.52, + "step": 247 + }, + { + "epoch": 1.6954545454545453, + "grad_norm": 0.45308929681777954, + "learning_rate": 4.564510336135642e-06, + "loss": 0.5589, + "step": 248 + }, + { + "epoch": 1.7022727272727272, + "grad_norm": 0.4482448399066925, + "learning_rate": 4.558785561858196e-06, + "loss": 0.5161, + "step": 249 + }, + { + "epoch": 1.709090909090909, + "grad_norm": 0.44250935316085815, + "learning_rate": 4.5530270443374305e-06, + "loss": 0.5342, + "step": 250 + }, + { + "epoch": 1.7159090909090908, + "grad_norm": 0.8177540898323059, + "learning_rate": 4.547234877954741e-06, + "loss": 0.5402, + "step": 251 + }, + { + "epoch": 1.7227272727272727, + "grad_norm": 0.5169854760169983, + "learning_rate": 4.541409157643027e-06, + "loss": 0.5301, + "step": 252 + }, + { + "epoch": 1.7295454545454545, + "grad_norm": 0.4544285237789154, + "learning_rate": 4.535549978885132e-06, + "loss": 0.5314, + "step": 253 + }, + { + "epoch": 1.7363636363636363, + "grad_norm": 0.42860639095306396, + "learning_rate": 4.5296574377122765e-06, + "loss": 0.5249, + "step": 254 + }, + { + "epoch": 1.7431818181818182, + "grad_norm": 0.526492178440094, + "learning_rate": 4.5237316307024895e-06, + "loss": 0.5142, + "step": 255 + }, + { + "epoch": 1.75, + "grad_norm": 0.4208003878593445, + "learning_rate": 4.517772654979024e-06, + "loss": 0.5151, + "step": 256 + }, + { + "epoch": 1.7568181818181818, + "grad_norm": 0.4974212944507599, + "learning_rate": 4.51178060820876e-06, + "loss": 0.5161, + "step": 257 + }, + { + "epoch": 1.7636363636363637, + "grad_norm": 0.4148060083389282, + "learning_rate": 4.505755588600613e-06, + "loss": 0.5099, + "step": 258 + }, + { + "epoch": 1.7704545454545455, + "grad_norm": 0.4106168746948242, + "learning_rate": 4.499697694903915e-06, + "loss": 0.5376, + "step": 259 + }, + { + "epoch": 1.7772727272727273, + "grad_norm": 0.49017220735549927, + "learning_rate": 4.493607026406802e-06, + "loss": 0.525, + "step": 260 + }, + { + "epoch": 1.7840909090909092, + "grad_norm": 0.4141218662261963, + "learning_rate": 4.487483682934587e-06, + "loss": 0.5303, + "step": 261 + }, + { + "epoch": 1.790909090909091, + "grad_norm": 0.44826585054397583, + "learning_rate": 4.481327764848118e-06, + "loss": 0.4886, + "step": 262 + }, + { + "epoch": 1.7977272727272728, + "grad_norm": 0.42013809084892273, + "learning_rate": 4.47513937304214e-06, + "loss": 0.5204, + "step": 263 + }, + { + "epoch": 1.8045454545454547, + "grad_norm": 0.44327741861343384, + "learning_rate": 4.4689186089436365e-06, + "loss": 0.5097, + "step": 264 + }, + { + "epoch": 1.8113636363636365, + "grad_norm": 0.39498746395111084, + "learning_rate": 4.462665574510169e-06, + "loss": 0.5102, + "step": 265 + }, + { + "epoch": 1.8181818181818183, + "grad_norm": 0.48296087980270386, + "learning_rate": 4.456380372228208e-06, + "loss": 0.5299, + "step": 266 + }, + { + "epoch": 1.825, + "grad_norm": 0.4398634433746338, + "learning_rate": 4.450063105111447e-06, + "loss": 0.5419, + "step": 267 + }, + { + "epoch": 1.8318181818181818, + "grad_norm": 0.43333035707473755, + "learning_rate": 4.443713876699124e-06, + "loss": 0.5047, + "step": 268 + }, + { + "epoch": 1.8386363636363636, + "grad_norm": 0.5823659300804138, + "learning_rate": 4.4373327910543125e-06, + "loss": 0.5082, + "step": 269 + }, + { + "epoch": 1.8454545454545455, + "grad_norm": 0.44385817646980286, + "learning_rate": 4.430919952762226e-06, + "loss": 0.5171, + "step": 270 + }, + { + "epoch": 1.8522727272727273, + "grad_norm": 0.4444347023963928, + "learning_rate": 4.424475466928499e-06, + "loss": 0.5351, + "step": 271 + }, + { + "epoch": 1.8590909090909091, + "grad_norm": 0.549674391746521, + "learning_rate": 4.417999439177465e-06, + "loss": 0.498, + "step": 272 + }, + { + "epoch": 1.865909090909091, + "grad_norm": 0.4340274930000305, + "learning_rate": 4.4114919756504275e-06, + "loss": 0.5429, + "step": 273 + }, + { + "epoch": 1.8727272727272726, + "grad_norm": 0.44019854068756104, + "learning_rate": 4.404953183003916e-06, + "loss": 0.5133, + "step": 274 + }, + { + "epoch": 1.8795454545454544, + "grad_norm": 0.4473486542701721, + "learning_rate": 4.398383168407941e-06, + "loss": 0.5413, + "step": 275 + }, + { + "epoch": 1.8863636363636362, + "grad_norm": 0.44450780749320984, + "learning_rate": 4.391782039544239e-06, + "loss": 0.4998, + "step": 276 + }, + { + "epoch": 1.893181818181818, + "grad_norm": 0.4310123920440674, + "learning_rate": 4.385149904604502e-06, + "loss": 0.5134, + "step": 277 + }, + { + "epoch": 1.9, + "grad_norm": 0.48953917622566223, + "learning_rate": 4.378486872288611e-06, + "loss": 0.5102, + "step": 278 + }, + { + "epoch": 1.9068181818181817, + "grad_norm": 0.49219778180122375, + "learning_rate": 4.371793051802849e-06, + "loss": 0.5213, + "step": 279 + }, + { + "epoch": 1.9136363636363636, + "grad_norm": 0.818645715713501, + "learning_rate": 4.365068552858116e-06, + "loss": 0.5251, + "step": 280 + }, + { + "epoch": 1.9204545454545454, + "grad_norm": 0.4407028257846832, + "learning_rate": 4.358313485668124e-06, + "loss": 0.5191, + "step": 281 + }, + { + "epoch": 1.9272727272727272, + "grad_norm": 0.5301506519317627, + "learning_rate": 4.3515279609476e-06, + "loss": 0.5003, + "step": 282 + }, + { + "epoch": 1.934090909090909, + "grad_norm": 0.4673819839954376, + "learning_rate": 4.3447120899104615e-06, + "loss": 0.521, + "step": 283 + }, + { + "epoch": 1.940909090909091, + "grad_norm": 0.4106168746948242, + "learning_rate": 4.337865984268002e-06, + "loss": 0.5199, + "step": 284 + }, + { + "epoch": 1.9477272727272728, + "grad_norm": 0.43583595752716064, + "learning_rate": 4.3309897562270525e-06, + "loss": 0.5367, + "step": 285 + }, + { + "epoch": 1.9545454545454546, + "grad_norm": 0.4984910488128662, + "learning_rate": 4.324083518488151e-06, + "loss": 0.5202, + "step": 286 + }, + { + "epoch": 1.9613636363636364, + "grad_norm": 0.4516090452671051, + "learning_rate": 4.317147384243688e-06, + "loss": 0.5432, + "step": 287 + }, + { + "epoch": 1.9681818181818183, + "grad_norm": 0.43753015995025635, + "learning_rate": 4.3101814671760546e-06, + "loss": 0.5309, + "step": 288 + }, + { + "epoch": 1.975, + "grad_norm": 0.4472252130508423, + "learning_rate": 4.303185881455778e-06, + "loss": 0.5344, + "step": 289 + }, + { + "epoch": 1.981818181818182, + "grad_norm": 1.178377389907837, + "learning_rate": 4.296160741739652e-06, + "loss": 0.5283, + "step": 290 + }, + { + "epoch": 1.9886363636363638, + "grad_norm": 0.5085738301277161, + "learning_rate": 4.289106163168858e-06, + "loss": 0.5133, + "step": 291 + }, + { + "epoch": 1.9954545454545456, + "grad_norm": 0.52077716588974, + "learning_rate": 4.282022261367074e-06, + "loss": 0.5155, + "step": 292 + }, + { + "epoch": 2.006818181818182, + "grad_norm": 0.8534443378448486, + "learning_rate": 4.274909152438582e-06, + "loss": 1.003, + "step": 293 + }, + { + "epoch": 2.0136363636363637, + "grad_norm": 0.46160876750946045, + "learning_rate": 4.267766952966369e-06, + "loss": 0.4958, + "step": 294 + }, + { + "epoch": 2.0204545454545455, + "grad_norm": 0.5113462209701538, + "learning_rate": 4.260595780010209e-06, + "loss": 0.4895, + "step": 295 + }, + { + "epoch": 2.0272727272727273, + "grad_norm": 0.4553380608558655, + "learning_rate": 4.2533957511047485e-06, + "loss": 0.4826, + "step": 296 + }, + { + "epoch": 2.034090909090909, + "grad_norm": 0.43557822704315186, + "learning_rate": 4.24616698425758e-06, + "loss": 0.4591, + "step": 297 + }, + { + "epoch": 2.040909090909091, + "grad_norm": 0.45537200570106506, + "learning_rate": 4.238909597947307e-06, + "loss": 0.4811, + "step": 298 + }, + { + "epoch": 2.047727272727273, + "grad_norm": 0.48833370208740234, + "learning_rate": 4.231623711121603e-06, + "loss": 0.4967, + "step": 299 + }, + { + "epoch": 2.0545454545454547, + "grad_norm": 0.45504844188690186, + "learning_rate": 4.224309443195261e-06, + "loss": 0.4964, + "step": 300 + }, + { + "epoch": 2.0613636363636365, + "grad_norm": 0.4313275218009949, + "learning_rate": 4.2169669140482365e-06, + "loss": 0.4832, + "step": 301 + }, + { + "epoch": 2.0681818181818183, + "grad_norm": 0.4298369884490967, + "learning_rate": 4.2095962440236846e-06, + "loss": 0.5071, + "step": 302 + }, + { + "epoch": 2.075, + "grad_norm": 0.4188523590564728, + "learning_rate": 4.202197553925983e-06, + "loss": 0.4859, + "step": 303 + }, + { + "epoch": 2.081818181818182, + "grad_norm": 0.39195290207862854, + "learning_rate": 4.194770965018758e-06, + "loss": 0.5081, + "step": 304 + }, + { + "epoch": 2.088636363636364, + "grad_norm": 0.45844170451164246, + "learning_rate": 4.187316599022892e-06, + "loss": 0.508, + "step": 305 + }, + { + "epoch": 2.0954545454545457, + "grad_norm": 0.6743896007537842, + "learning_rate": 4.179834578114531e-06, + "loss": 0.4748, + "step": 306 + }, + { + "epoch": 2.102272727272727, + "grad_norm": 0.4317929446697235, + "learning_rate": 4.172325024923083e-06, + "loss": 0.504, + "step": 307 + }, + { + "epoch": 2.109090909090909, + "grad_norm": 0.4857175350189209, + "learning_rate": 4.164788062529203e-06, + "loss": 0.4987, + "step": 308 + }, + { + "epoch": 2.1159090909090907, + "grad_norm": 0.4454379677772522, + "learning_rate": 4.157223814462784e-06, + "loss": 0.4938, + "step": 309 + }, + { + "epoch": 2.1227272727272726, + "grad_norm": 0.5128078460693359, + "learning_rate": 4.149632404700925e-06, + "loss": 0.4917, + "step": 310 + }, + { + "epoch": 2.1295454545454544, + "grad_norm": 0.5195722579956055, + "learning_rate": 4.142013957665903e-06, + "loss": 0.4753, + "step": 311 + }, + { + "epoch": 2.1363636363636362, + "grad_norm": 0.4242931604385376, + "learning_rate": 4.134368598223132e-06, + "loss": 0.4832, + "step": 312 + }, + { + "epoch": 2.143181818181818, + "grad_norm": 0.4419071674346924, + "learning_rate": 4.126696451679116e-06, + "loss": 0.4783, + "step": 313 + }, + { + "epoch": 2.15, + "grad_norm": 0.42852380871772766, + "learning_rate": 4.118997643779401e-06, + "loss": 0.5034, + "step": 314 + }, + { + "epoch": 2.1568181818181817, + "grad_norm": 0.7373576164245605, + "learning_rate": 4.111272300706502e-06, + "loss": 0.4934, + "step": 315 + }, + { + "epoch": 2.1636363636363636, + "grad_norm": 0.4435337781906128, + "learning_rate": 4.1035205490778505e-06, + "loss": 0.5089, + "step": 316 + }, + { + "epoch": 2.1704545454545454, + "grad_norm": 0.4714502692222595, + "learning_rate": 4.095742515943703e-06, + "loss": 0.502, + "step": 317 + }, + { + "epoch": 2.1772727272727272, + "grad_norm": 0.4732494056224823, + "learning_rate": 4.087938328785071e-06, + "loss": 0.5073, + "step": 318 + }, + { + "epoch": 2.184090909090909, + "grad_norm": 0.45177504420280457, + "learning_rate": 4.080108115511629e-06, + "loss": 0.5076, + "step": 319 + }, + { + "epoch": 2.190909090909091, + "grad_norm": 0.506157398223877, + "learning_rate": 4.072252004459612e-06, + "loss": 0.4837, + "step": 320 + }, + { + "epoch": 2.1977272727272728, + "grad_norm": 0.43044376373291016, + "learning_rate": 4.064370124389718e-06, + "loss": 0.5167, + "step": 321 + }, + { + "epoch": 2.2045454545454546, + "grad_norm": 0.45878180861473083, + "learning_rate": 4.056462604484998e-06, + "loss": 0.4936, + "step": 322 + }, + { + "epoch": 2.2113636363636364, + "grad_norm": 0.43078047037124634, + "learning_rate": 4.048529574348734e-06, + "loss": 0.4783, + "step": 323 + }, + { + "epoch": 2.2181818181818183, + "grad_norm": 0.42789366841316223, + "learning_rate": 4.040571164002319e-06, + "loss": 0.4853, + "step": 324 + }, + { + "epoch": 2.225, + "grad_norm": 0.44888222217559814, + "learning_rate": 4.032587503883124e-06, + "loss": 0.5097, + "step": 325 + }, + { + "epoch": 2.231818181818182, + "grad_norm": 0.446851521730423, + "learning_rate": 4.0245787248423614e-06, + "loss": 0.4844, + "step": 326 + }, + { + "epoch": 2.2386363636363638, + "grad_norm": 0.436567097902298, + "learning_rate": 4.0165449581429404e-06, + "loss": 0.4549, + "step": 327 + }, + { + "epoch": 2.2454545454545456, + "grad_norm": 0.4813944697380066, + "learning_rate": 4.008486335457312e-06, + "loss": 0.49, + "step": 328 + }, + { + "epoch": 2.2522727272727274, + "grad_norm": 0.4935426115989685, + "learning_rate": 4.000402988865316e-06, + "loss": 0.4995, + "step": 329 + }, + { + "epoch": 2.2590909090909093, + "grad_norm": 1.1695573329925537, + "learning_rate": 3.992295050852013e-06, + "loss": 0.5033, + "step": 330 + }, + { + "epoch": 2.265909090909091, + "grad_norm": 0.41830316185951233, + "learning_rate": 3.984162654305516e-06, + "loss": 0.4828, + "step": 331 + }, + { + "epoch": 2.2727272727272725, + "grad_norm": 0.4522007703781128, + "learning_rate": 3.976005932514807e-06, + "loss": 0.4948, + "step": 332 + }, + { + "epoch": 2.2795454545454543, + "grad_norm": 0.8304370045661926, + "learning_rate": 3.967825019167559e-06, + "loss": 0.504, + "step": 333 + }, + { + "epoch": 2.286363636363636, + "grad_norm": 0.4533236622810364, + "learning_rate": 3.959620048347938e-06, + "loss": 0.4487, + "step": 334 + }, + { + "epoch": 2.293181818181818, + "grad_norm": 0.43578940629959106, + "learning_rate": 3.951391154534415e-06, + "loss": 0.4702, + "step": 335 + }, + { + "epoch": 2.3, + "grad_norm": 0.5146592259407043, + "learning_rate": 3.943138472597549e-06, + "loss": 0.4984, + "step": 336 + }, + { + "epoch": 2.3068181818181817, + "grad_norm": 0.6412249207496643, + "learning_rate": 3.934862137797788e-06, + "loss": 0.4761, + "step": 337 + }, + { + "epoch": 2.3136363636363635, + "grad_norm": 1.3352930545806885, + "learning_rate": 3.9265622857832455e-06, + "loss": 0.499, + "step": 338 + }, + { + "epoch": 2.3204545454545453, + "grad_norm": 0.4465159475803375, + "learning_rate": 3.918239052587481e-06, + "loss": 0.4525, + "step": 339 + }, + { + "epoch": 2.327272727272727, + "grad_norm": 0.43597185611724854, + "learning_rate": 3.909892574627267e-06, + "loss": 0.4952, + "step": 340 + }, + { + "epoch": 2.334090909090909, + "grad_norm": 0.7057136297225952, + "learning_rate": 3.901522988700355e-06, + "loss": 0.4576, + "step": 341 + }, + { + "epoch": 2.340909090909091, + "grad_norm": 0.45416319370269775, + "learning_rate": 3.893130431983234e-06, + "loss": 0.4902, + "step": 342 + }, + { + "epoch": 2.3477272727272727, + "grad_norm": 0.43900424242019653, + "learning_rate": 3.884715042028882e-06, + "loss": 0.495, + "step": 343 + }, + { + "epoch": 2.3545454545454545, + "grad_norm": 0.4553733468055725, + "learning_rate": 3.876276956764509e-06, + "loss": 0.4861, + "step": 344 + }, + { + "epoch": 2.3613636363636363, + "grad_norm": 0.5471370816230774, + "learning_rate": 3.867816314489301e-06, + "loss": 0.5043, + "step": 345 + }, + { + "epoch": 2.368181818181818, + "grad_norm": 0.43938082456588745, + "learning_rate": 3.8593332538721465e-06, + "loss": 0.4678, + "step": 346 + }, + { + "epoch": 2.375, + "grad_norm": 0.4583636224269867, + "learning_rate": 3.8508279139493736e-06, + "loss": 0.4679, + "step": 347 + }, + { + "epoch": 2.381818181818182, + "grad_norm": 0.44170665740966797, + "learning_rate": 3.84230043412246e-06, + "loss": 0.4647, + "step": 348 + }, + { + "epoch": 2.3886363636363637, + "grad_norm": 0.4952428340911865, + "learning_rate": 3.833750954155757e-06, + "loss": 0.4796, + "step": 349 + }, + { + "epoch": 2.3954545454545455, + "grad_norm": 0.42052993178367615, + "learning_rate": 3.825179614174195e-06, + "loss": 0.4606, + "step": 350 + }, + { + "epoch": 2.4022727272727273, + "grad_norm": 0.5792545676231384, + "learning_rate": 3.816586554660987e-06, + "loss": 0.4805, + "step": 351 + }, + { + "epoch": 2.409090909090909, + "grad_norm": 0.4477898180484772, + "learning_rate": 3.807971916455325e-06, + "loss": 0.4793, + "step": 352 + }, + { + "epoch": 2.415909090909091, + "grad_norm": 0.45617541670799255, + "learning_rate": 3.799335840750077e-06, + "loss": 0.502, + "step": 353 + }, + { + "epoch": 2.422727272727273, + "grad_norm": 0.45021289587020874, + "learning_rate": 3.790678469089465e-06, + "loss": 0.478, + "step": 354 + }, + { + "epoch": 2.4295454545454547, + "grad_norm": 0.4506435990333557, + "learning_rate": 3.7819999433667503e-06, + "loss": 0.4828, + "step": 355 + }, + { + "epoch": 2.4363636363636365, + "grad_norm": 0.48021042346954346, + "learning_rate": 3.773300405821908e-06, + "loss": 0.4827, + "step": 356 + }, + { + "epoch": 2.4431818181818183, + "grad_norm": 0.4759495258331299, + "learning_rate": 3.764579999039293e-06, + "loss": 0.4633, + "step": 357 + }, + { + "epoch": 2.45, + "grad_norm": 0.4734678268432617, + "learning_rate": 3.7558388659453052e-06, + "loss": 0.4982, + "step": 358 + }, + { + "epoch": 2.456818181818182, + "grad_norm": 0.4502730369567871, + "learning_rate": 3.7470771498060455e-06, + "loss": 0.5009, + "step": 359 + }, + { + "epoch": 2.463636363636364, + "grad_norm": 0.444940447807312, + "learning_rate": 3.7382949942249695e-06, + "loss": 0.4994, + "step": 360 + }, + { + "epoch": 2.4704545454545457, + "grad_norm": 0.4882519245147705, + "learning_rate": 3.7294925431405306e-06, + "loss": 0.4607, + "step": 361 + }, + { + "epoch": 2.4772727272727275, + "grad_norm": 0.51338791847229, + "learning_rate": 3.720669940823827e-06, + "loss": 0.4801, + "step": 362 + }, + { + "epoch": 2.484090909090909, + "grad_norm": 0.495211124420166, + "learning_rate": 3.7118273318762275e-06, + "loss": 0.4984, + "step": 363 + }, + { + "epoch": 2.4909090909090907, + "grad_norm": 0.4894484281539917, + "learning_rate": 3.702964861227013e-06, + "loss": 0.4834, + "step": 364 + }, + { + "epoch": 2.4977272727272726, + "grad_norm": 0.5554991364479065, + "learning_rate": 3.694082674130991e-06, + "loss": 0.4874, + "step": 365 + }, + { + "epoch": 2.5045454545454544, + "grad_norm": 0.4586203396320343, + "learning_rate": 3.6851809161661206e-06, + "loss": 0.5117, + "step": 366 + }, + { + "epoch": 2.5113636363636362, + "grad_norm": 0.7342172861099243, + "learning_rate": 3.6762597332311254e-06, + "loss": 0.5184, + "step": 367 + }, + { + "epoch": 2.518181818181818, + "grad_norm": 0.46638190746307373, + "learning_rate": 3.6673192715431016e-06, + "loss": 0.47, + "step": 368 + }, + { + "epoch": 2.525, + "grad_norm": 0.4584850072860718, + "learning_rate": 3.658359677635122e-06, + "loss": 0.45, + "step": 369 + }, + { + "epoch": 2.5318181818181817, + "grad_norm": 0.4285255968570709, + "learning_rate": 3.649381098353834e-06, + "loss": 0.513, + "step": 370 + }, + { + "epoch": 2.5386363636363636, + "grad_norm": 0.49487683176994324, + "learning_rate": 3.6403836808570512e-06, + "loss": 0.5048, + "step": 371 + }, + { + "epoch": 2.5454545454545454, + "grad_norm": 0.44076815247535706, + "learning_rate": 3.631367572611348e-06, + "loss": 0.4927, + "step": 372 + }, + { + "epoch": 2.5522727272727272, + "grad_norm": 0.44529926776885986, + "learning_rate": 3.6223329213896313e-06, + "loss": 0.4863, + "step": 373 + }, + { + "epoch": 2.559090909090909, + "grad_norm": 0.5416815876960754, + "learning_rate": 3.613279875268731e-06, + "loss": 0.4808, + "step": 374 + }, + { + "epoch": 2.565909090909091, + "grad_norm": 0.43630415201187134, + "learning_rate": 3.604208582626964e-06, + "loss": 0.4792, + "step": 375 + }, + { + "epoch": 2.5727272727272728, + "grad_norm": 0.47734326124191284, + "learning_rate": 3.5951191921417063e-06, + "loss": 0.4462, + "step": 376 + }, + { + "epoch": 2.5795454545454546, + "grad_norm": 0.4540305435657501, + "learning_rate": 3.586011852786955e-06, + "loss": 0.4915, + "step": 377 + }, + { + "epoch": 2.5863636363636364, + "grad_norm": 0.47168779373168945, + "learning_rate": 3.5768867138308872e-06, + "loss": 0.47, + "step": 378 + }, + { + "epoch": 2.5931818181818183, + "grad_norm": 0.46372920274734497, + "learning_rate": 3.5677439248334133e-06, + "loss": 0.4775, + "step": 379 + }, + { + "epoch": 2.6, + "grad_norm": 0.41653329133987427, + "learning_rate": 3.5585836356437266e-06, + "loss": 0.4875, + "step": 380 + }, + { + "epoch": 2.606818181818182, + "grad_norm": 0.4495103657245636, + "learning_rate": 3.5494059963978433e-06, + "loss": 0.4637, + "step": 381 + }, + { + "epoch": 2.6136363636363638, + "grad_norm": 0.432577520608902, + "learning_rate": 3.540211157516149e-06, + "loss": 0.4991, + "step": 382 + }, + { + "epoch": 2.6204545454545456, + "grad_norm": 0.4517219662666321, + "learning_rate": 3.530999269700927e-06, + "loss": 0.4785, + "step": 383 + }, + { + "epoch": 2.6272727272727274, + "grad_norm": 0.519922137260437, + "learning_rate": 3.521770483933891e-06, + "loss": 0.5026, + "step": 384 + }, + { + "epoch": 2.634090909090909, + "grad_norm": 0.48750364780426025, + "learning_rate": 3.5125249514737093e-06, + "loss": 0.4805, + "step": 385 + }, + { + "epoch": 2.6409090909090907, + "grad_norm": 0.6416886448860168, + "learning_rate": 3.503262823853527e-06, + "loss": 0.4849, + "step": 386 + }, + { + "epoch": 2.6477272727272725, + "grad_norm": 0.5748171806335449, + "learning_rate": 3.493984252878483e-06, + "loss": 0.4885, + "step": 387 + }, + { + "epoch": 2.6545454545454543, + "grad_norm": 0.5148530006408691, + "learning_rate": 3.484689390623218e-06, + "loss": 0.484, + "step": 388 + }, + { + "epoch": 2.661363636363636, + "grad_norm": 0.526839017868042, + "learning_rate": 3.4753783894293886e-06, + "loss": 0.4758, + "step": 389 + }, + { + "epoch": 2.668181818181818, + "grad_norm": 0.7565727829933167, + "learning_rate": 3.466051401903162e-06, + "loss": 0.479, + "step": 390 + }, + { + "epoch": 2.675, + "grad_norm": 0.4598790109157562, + "learning_rate": 3.4567085809127247e-06, + "loss": 0.4972, + "step": 391 + }, + { + "epoch": 2.6818181818181817, + "grad_norm": 0.4911792278289795, + "learning_rate": 3.4473500795857674e-06, + "loss": 0.4714, + "step": 392 + }, + { + "epoch": 2.6886363636363635, + "grad_norm": 0.4214136600494385, + "learning_rate": 3.4379760513069804e-06, + "loss": 0.4906, + "step": 393 + }, + { + "epoch": 2.6954545454545453, + "grad_norm": 0.5070644021034241, + "learning_rate": 3.428586649715542e-06, + "loss": 0.4936, + "step": 394 + }, + { + "epoch": 2.702272727272727, + "grad_norm": 0.4591734707355499, + "learning_rate": 3.4191820287025916e-06, + "loss": 0.4751, + "step": 395 + }, + { + "epoch": 2.709090909090909, + "grad_norm": 0.579781711101532, + "learning_rate": 3.4097623424087196e-06, + "loss": 0.4935, + "step": 396 + }, + { + "epoch": 2.715909090909091, + "grad_norm": 0.4178840219974518, + "learning_rate": 3.4003277452214284e-06, + "loss": 0.4771, + "step": 397 + }, + { + "epoch": 2.7227272727272727, + "grad_norm": 0.5002609491348267, + "learning_rate": 3.3908783917726123e-06, + "loss": 0.4739, + "step": 398 + }, + { + "epoch": 2.7295454545454545, + "grad_norm": 0.5176806449890137, + "learning_rate": 3.381414436936018e-06, + "loss": 0.4965, + "step": 399 + }, + { + "epoch": 2.7363636363636363, + "grad_norm": 0.8039212226867676, + "learning_rate": 3.3719360358247054e-06, + "loss": 0.4822, + "step": 400 + }, + { + "epoch": 2.743181818181818, + "grad_norm": 0.48943471908569336, + "learning_rate": 3.36244334378851e-06, + "loss": 0.4815, + "step": 401 + }, + { + "epoch": 2.75, + "grad_norm": 0.43710342049598694, + "learning_rate": 3.3529365164114903e-06, + "loss": 0.4932, + "step": 402 + }, + { + "epoch": 2.756818181818182, + "grad_norm": 0.5585265159606934, + "learning_rate": 3.3434157095093846e-06, + "loss": 0.5125, + "step": 403 + }, + { + "epoch": 2.7636363636363637, + "grad_norm": 0.4626101553440094, + "learning_rate": 3.333881079127052e-06, + "loss": 0.4934, + "step": 404 + }, + { + "epoch": 2.7704545454545455, + "grad_norm": 0.4675556421279907, + "learning_rate": 3.3243327815359168e-06, + "loss": 0.4903, + "step": 405 + }, + { + "epoch": 2.7772727272727273, + "grad_norm": 0.45658671855926514, + "learning_rate": 3.314770973231408e-06, + "loss": 0.4778, + "step": 406 + }, + { + "epoch": 2.784090909090909, + "grad_norm": 0.4911910593509674, + "learning_rate": 3.305195810930393e-06, + "loss": 0.4795, + "step": 407 + }, + { + "epoch": 2.790909090909091, + "grad_norm": 0.4397219121456146, + "learning_rate": 3.2956074515686105e-06, + "loss": 0.4589, + "step": 408 + }, + { + "epoch": 2.797727272727273, + "grad_norm": 0.5184643864631653, + "learning_rate": 3.2860060522980945e-06, + "loss": 0.4827, + "step": 409 + }, + { + "epoch": 2.8045454545454547, + "grad_norm": 0.45334312319755554, + "learning_rate": 3.276391770484606e-06, + "loss": 0.4821, + "step": 410 + }, + { + "epoch": 2.8113636363636365, + "grad_norm": 0.41846391558647156, + "learning_rate": 3.266764763705046e-06, + "loss": 0.5157, + "step": 411 + }, + { + "epoch": 2.8181818181818183, + "grad_norm": 0.45868799090385437, + "learning_rate": 3.257125189744877e-06, + "loss": 0.4876, + "step": 412 + }, + { + "epoch": 2.825, + "grad_norm": 0.4785497486591339, + "learning_rate": 3.247473206595536e-06, + "loss": 0.4809, + "step": 413 + }, + { + "epoch": 2.831818181818182, + "grad_norm": 0.5683679580688477, + "learning_rate": 3.2378089724518464e-06, + "loss": 0.505, + "step": 414 + }, + { + "epoch": 2.838636363636364, + "grad_norm": 0.48336583375930786, + "learning_rate": 3.228132645709421e-06, + "loss": 0.4816, + "step": 415 + }, + { + "epoch": 2.8454545454545457, + "grad_norm": 0.4236341416835785, + "learning_rate": 3.218444384962071e-06, + "loss": 0.4885, + "step": 416 + }, + { + "epoch": 2.8522727272727275, + "grad_norm": 0.4892752766609192, + "learning_rate": 3.2087443489992043e-06, + "loss": 0.4782, + "step": 417 + }, + { + "epoch": 2.8590909090909093, + "grad_norm": 0.4632202088832855, + "learning_rate": 3.1990326968032225e-06, + "loss": 0.5211, + "step": 418 + }, + { + "epoch": 2.865909090909091, + "grad_norm": 0.8842809796333313, + "learning_rate": 3.189309587546917e-06, + "loss": 0.4716, + "step": 419 + }, + { + "epoch": 2.8727272727272726, + "grad_norm": 0.46771398186683655, + "learning_rate": 3.1795751805908578e-06, + "loss": 0.4745, + "step": 420 + }, + { + "epoch": 2.8795454545454544, + "grad_norm": 0.8178832530975342, + "learning_rate": 3.169829635480783e-06, + "loss": 0.4917, + "step": 421 + }, + { + "epoch": 2.8863636363636362, + "grad_norm": 0.43057572841644287, + "learning_rate": 3.160073111944983e-06, + "loss": 0.4926, + "step": 422 + }, + { + "epoch": 2.893181818181818, + "grad_norm": 0.4258928894996643, + "learning_rate": 3.150305769891686e-06, + "loss": 0.4698, + "step": 423 + }, + { + "epoch": 2.9, + "grad_norm": 0.4559054374694824, + "learning_rate": 3.1405277694064306e-06, + "loss": 0.4537, + "step": 424 + }, + { + "epoch": 2.9068181818181817, + "grad_norm": 0.4342961609363556, + "learning_rate": 3.13073927074945e-06, + "loss": 0.4834, + "step": 425 + }, + { + "epoch": 2.9136363636363636, + "grad_norm": 0.48813825845718384, + "learning_rate": 3.1209404343530374e-06, + "loss": 0.4853, + "step": 426 + }, + { + "epoch": 2.9204545454545454, + "grad_norm": 0.47864893078804016, + "learning_rate": 3.111131420818922e-06, + "loss": 0.4627, + "step": 427 + }, + { + "epoch": 2.9272727272727272, + "grad_norm": 0.4394248425960541, + "learning_rate": 3.1013123909156347e-06, + "loss": 0.4817, + "step": 428 + }, + { + "epoch": 2.934090909090909, + "grad_norm": 0.6629825830459595, + "learning_rate": 3.091483505575873e-06, + "loss": 0.4738, + "step": 429 + }, + { + "epoch": 2.940909090909091, + "grad_norm": 0.47470518946647644, + "learning_rate": 3.081644925893866e-06, + "loss": 0.4943, + "step": 430 + }, + { + "epoch": 2.9477272727272728, + "grad_norm": 0.44007545709609985, + "learning_rate": 3.0717968131227285e-06, + "loss": 0.4805, + "step": 431 + }, + { + "epoch": 2.9545454545454546, + "grad_norm": 0.4425961375236511, + "learning_rate": 3.061939328671824e-06, + "loss": 0.5092, + "step": 432 + }, + { + "epoch": 2.9613636363636364, + "grad_norm": 0.4215753972530365, + "learning_rate": 3.0520726341041165e-06, + "loss": 0.4616, + "step": 433 + }, + { + "epoch": 2.9681818181818183, + "grad_norm": 0.4265761077404022, + "learning_rate": 3.0421968911335196e-06, + "loss": 0.4764, + "step": 434 + }, + { + "epoch": 2.975, + "grad_norm": 0.45009252429008484, + "learning_rate": 3.032312261622255e-06, + "loss": 0.4865, + "step": 435 + }, + { + "epoch": 2.981818181818182, + "grad_norm": 0.5175682306289673, + "learning_rate": 3.0224189075781886e-06, + "loss": 0.5085, + "step": 436 + }, + { + "epoch": 2.9886363636363638, + "grad_norm": 0.4518980383872986, + "learning_rate": 3.012516991152181e-06, + "loss": 0.4568, + "step": 437 + }, + { + "epoch": 2.9954545454545456, + "grad_norm": 0.4789412021636963, + "learning_rate": 3.002606674635432e-06, + "loss": 0.4942, + "step": 438 + }, + { + "epoch": 3.006818181818182, + "grad_norm": 0.495550274848938, + "learning_rate": 2.9926881204568153e-06, + "loss": 0.9793, + "step": 439 + }, + { + "epoch": 3.0136363636363637, + "grad_norm": 0.47357404232025146, + "learning_rate": 2.9827614911802205e-06, + "loss": 0.4689, + "step": 440 + }, + { + "epoch": 3.0204545454545455, + "grad_norm": 0.4523085355758667, + "learning_rate": 2.972826949501884e-06, + "loss": 0.4632, + "step": 441 + }, + { + "epoch": 3.0272727272727273, + "grad_norm": 0.4381036162376404, + "learning_rate": 2.9628846582477305e-06, + "loss": 0.4508, + "step": 442 + }, + { + "epoch": 3.034090909090909, + "grad_norm": 0.5367632508277893, + "learning_rate": 2.9529347803706943e-06, + "loss": 0.4472, + "step": 443 + }, + { + "epoch": 3.040909090909091, + "grad_norm": 0.5390172600746155, + "learning_rate": 2.9429774789480576e-06, + "loss": 0.4475, + "step": 444 + }, + { + "epoch": 3.047727272727273, + "grad_norm": 0.44644543528556824, + "learning_rate": 2.9330129171787704e-06, + "loss": 0.497, + "step": 445 + }, + { + "epoch": 3.0545454545454547, + "grad_norm": 0.46184423565864563, + "learning_rate": 2.923041258380779e-06, + "loss": 0.4545, + "step": 446 + }, + { + "epoch": 3.0613636363636365, + "grad_norm": 0.45963627099990845, + "learning_rate": 2.9130626659883537e-06, + "loss": 0.4571, + "step": 447 + }, + { + "epoch": 3.0681818181818183, + "grad_norm": 0.4504947364330292, + "learning_rate": 2.9030773035493997e-06, + "loss": 0.4563, + "step": 448 + }, + { + "epoch": 3.075, + "grad_norm": 0.4393884241580963, + "learning_rate": 2.893085334722786e-06, + "loss": 0.4433, + "step": 449 + }, + { + "epoch": 3.081818181818182, + "grad_norm": 0.7640544772148132, + "learning_rate": 2.883086923275658e-06, + "loss": 0.4598, + "step": 450 + }, + { + "epoch": 3.088636363636364, + "grad_norm": 0.44159695506095886, + "learning_rate": 2.8730822330807556e-06, + "loss": 0.4443, + "step": 451 + }, + { + "epoch": 3.0954545454545457, + "grad_norm": 0.4525455832481384, + "learning_rate": 2.8630714281137263e-06, + "loss": 0.4692, + "step": 452 + }, + { + "epoch": 3.102272727272727, + "grad_norm": 0.44391655921936035, + "learning_rate": 2.853054672450437e-06, + "loss": 0.4412, + "step": 453 + }, + { + "epoch": 3.109090909090909, + "grad_norm": 0.45541912317276, + "learning_rate": 2.8430321302642887e-06, + "loss": 0.4417, + "step": 454 + }, + { + "epoch": 3.1159090909090907, + "grad_norm": 0.43590644001960754, + "learning_rate": 2.8330039658235194e-06, + "loss": 0.4529, + "step": 455 + }, + { + "epoch": 3.1227272727272726, + "grad_norm": 0.48097532987594604, + "learning_rate": 2.8229703434885165e-06, + "loss": 0.4519, + "step": 456 + }, + { + "epoch": 3.1295454545454544, + "grad_norm": 0.43121981620788574, + "learning_rate": 2.8129314277091224e-06, + "loss": 0.4663, + "step": 457 + }, + { + "epoch": 3.1363636363636362, + "grad_norm": 0.46553748846054077, + "learning_rate": 2.8028873830219373e-06, + "loss": 0.4521, + "step": 458 + }, + { + "epoch": 3.143181818181818, + "grad_norm": 0.449290007352829, + "learning_rate": 2.7928383740476247e-06, + "loss": 0.4586, + "step": 459 + }, + { + "epoch": 3.15, + "grad_norm": 0.679371178150177, + "learning_rate": 2.7827845654882112e-06, + "loss": 0.4476, + "step": 460 + }, + { + "epoch": 3.1568181818181817, + "grad_norm": 0.49499261379241943, + "learning_rate": 2.7727261221243875e-06, + "loss": 0.4752, + "step": 461 + }, + { + "epoch": 3.1636363636363636, + "grad_norm": 0.4675534963607788, + "learning_rate": 2.76266320881281e-06, + "loss": 0.478, + "step": 462 + }, + { + "epoch": 3.1704545454545454, + "grad_norm": 0.44854965806007385, + "learning_rate": 2.7525959904833955e-06, + "loss": 0.4289, + "step": 463 + }, + { + "epoch": 3.1772727272727272, + "grad_norm": 0.4300600290298462, + "learning_rate": 2.7425246321366205e-06, + "loss": 0.4588, + "step": 464 + }, + { + "epoch": 3.184090909090909, + "grad_norm": 0.46728187799453735, + "learning_rate": 2.7324492988408146e-06, + "loss": 0.4498, + "step": 465 + }, + { + "epoch": 3.190909090909091, + "grad_norm": 0.46769359707832336, + "learning_rate": 2.7223701557294574e-06, + "loss": 0.4324, + "step": 466 + }, + { + "epoch": 3.1977272727272728, + "grad_norm": 0.4275517463684082, + "learning_rate": 2.712287367998471e-06, + "loss": 0.4574, + "step": 467 + }, + { + "epoch": 3.2045454545454546, + "grad_norm": 0.5888628959655762, + "learning_rate": 2.702201100903511e-06, + "loss": 0.4397, + "step": 468 + }, + { + "epoch": 3.2113636363636364, + "grad_norm": 0.4613579511642456, + "learning_rate": 2.692111519757261e-06, + "loss": 0.4391, + "step": 469 + }, + { + "epoch": 3.2181818181818183, + "grad_norm": 0.4581657946109772, + "learning_rate": 2.6820187899267203e-06, + "loss": 0.4734, + "step": 470 + }, + { + "epoch": 3.225, + "grad_norm": 0.4940917491912842, + "learning_rate": 2.671923076830496e-06, + "loss": 0.4726, + "step": 471 + }, + { + "epoch": 3.231818181818182, + "grad_norm": 0.7280719876289368, + "learning_rate": 2.6618245459360896e-06, + "loss": 0.4741, + "step": 472 + }, + { + "epoch": 3.2386363636363638, + "grad_norm": 0.519322395324707, + "learning_rate": 2.651723362757186e-06, + "loss": 0.4549, + "step": 473 + }, + { + "epoch": 3.2454545454545456, + "grad_norm": 0.47500962018966675, + "learning_rate": 2.641619692850941e-06, + "loss": 0.4598, + "step": 474 + }, + { + "epoch": 3.2522727272727274, + "grad_norm": 0.4661007821559906, + "learning_rate": 2.631513701815267e-06, + "loss": 0.4536, + "step": 475 + }, + { + "epoch": 3.2590909090909093, + "grad_norm": 0.48142874240875244, + "learning_rate": 2.6214055552861213e-06, + "loss": 0.474, + "step": 476 + }, + { + "epoch": 3.265909090909091, + "grad_norm": 0.5227756500244141, + "learning_rate": 2.611295418934786e-06, + "loss": 0.4516, + "step": 477 + }, + { + "epoch": 3.2727272727272725, + "grad_norm": 0.4693076014518738, + "learning_rate": 2.6011834584651597e-06, + "loss": 0.4409, + "step": 478 + }, + { + "epoch": 3.2795454545454543, + "grad_norm": 0.44575586915016174, + "learning_rate": 2.591069839611036e-06, + "loss": 0.4382, + "step": 479 + }, + { + "epoch": 3.286363636363636, + "grad_norm": 0.4569104313850403, + "learning_rate": 2.5809547281333904e-06, + "loss": 0.476, + "step": 480 + }, + { + "epoch": 3.293181818181818, + "grad_norm": 0.4542261064052582, + "learning_rate": 2.570838289817661e-06, + "loss": 0.4542, + "step": 481 + }, + { + "epoch": 3.3, + "grad_norm": 0.5086669921875, + "learning_rate": 2.560720690471033e-06, + "loss": 0.4673, + "step": 482 + }, + { + "epoch": 3.3068181818181817, + "grad_norm": 0.46243855357170105, + "learning_rate": 2.5506020959197218e-06, + "loss": 0.4462, + "step": 483 + }, + { + "epoch": 3.3136363636363635, + "grad_norm": 0.4877456724643707, + "learning_rate": 2.5404826720062544e-06, + "loss": 0.4463, + "step": 484 + }, + { + "epoch": 3.3204545454545453, + "grad_norm": 0.4592823088169098, + "learning_rate": 2.5303625845867475e-06, + "loss": 0.4674, + "step": 485 + }, + { + "epoch": 3.327272727272727, + "grad_norm": 0.4744127690792084, + "learning_rate": 2.5202419995281966e-06, + "loss": 0.4502, + "step": 486 + }, + { + "epoch": 3.334090909090909, + "grad_norm": 0.4267658591270447, + "learning_rate": 2.5101210827057516e-06, + "loss": 0.457, + "step": 487 + }, + { + "epoch": 3.340909090909091, + "grad_norm": 0.5804760456085205, + "learning_rate": 2.5e-06, + "loss": 0.4809, + "step": 488 + }, + { + "epoch": 3.3477272727272727, + "grad_norm": 0.5594801306724548, + "learning_rate": 2.4898789172942492e-06, + "loss": 0.4522, + "step": 489 + }, + { + "epoch": 3.3545454545454545, + "grad_norm": 0.4550947844982147, + "learning_rate": 2.4797580004718038e-06, + "loss": 0.4528, + "step": 490 + }, + { + "epoch": 3.3613636363636363, + "grad_norm": 0.430431991815567, + "learning_rate": 2.4696374154132533e-06, + "loss": 0.4657, + "step": 491 + }, + { + "epoch": 3.368181818181818, + "grad_norm": 0.4511764645576477, + "learning_rate": 2.4595173279937464e-06, + "loss": 0.4494, + "step": 492 + }, + { + "epoch": 3.375, + "grad_norm": 0.47958365082740784, + "learning_rate": 2.4493979040802786e-06, + "loss": 0.4508, + "step": 493 + }, + { + "epoch": 3.381818181818182, + "grad_norm": 0.4632541835308075, + "learning_rate": 2.4392793095289677e-06, + "loss": 0.4444, + "step": 494 + }, + { + "epoch": 3.3886363636363637, + "grad_norm": 0.46840110421180725, + "learning_rate": 2.42916171018234e-06, + "loss": 0.4676, + "step": 495 + }, + { + "epoch": 3.3954545454545455, + "grad_norm": 0.49999886751174927, + "learning_rate": 2.419045271866611e-06, + "loss": 0.4589, + "step": 496 + }, + { + "epoch": 3.4022727272727273, + "grad_norm": 0.44237902760505676, + "learning_rate": 2.408930160388965e-06, + "loss": 0.4569, + "step": 497 + }, + { + "epoch": 3.409090909090909, + "grad_norm": 0.4862026572227478, + "learning_rate": 2.3988165415348416e-06, + "loss": 0.4534, + "step": 498 + }, + { + "epoch": 3.415909090909091, + "grad_norm": 0.48773396015167236, + "learning_rate": 2.388704581065215e-06, + "loss": 0.4411, + "step": 499 + }, + { + "epoch": 3.422727272727273, + "grad_norm": 0.4380181133747101, + "learning_rate": 2.3785944447138804e-06, + "loss": 0.4529, + "step": 500 + }, + { + "epoch": 3.4295454545454547, + "grad_norm": 0.4616672992706299, + "learning_rate": 2.368486298184733e-06, + "loss": 0.4596, + "step": 501 + }, + { + "epoch": 3.4363636363636365, + "grad_norm": 0.4307732582092285, + "learning_rate": 2.358380307149059e-06, + "loss": 0.4626, + "step": 502 + }, + { + "epoch": 3.4431818181818183, + "grad_norm": 0.4402921795845032, + "learning_rate": 2.348276637242814e-06, + "loss": 0.4606, + "step": 503 + }, + { + "epoch": 3.45, + "grad_norm": 0.4782681465148926, + "learning_rate": 2.3381754540639108e-06, + "loss": 0.4508, + "step": 504 + }, + { + "epoch": 3.456818181818182, + "grad_norm": 0.43719062209129333, + "learning_rate": 2.328076923169504e-06, + "loss": 0.4476, + "step": 505 + }, + { + "epoch": 3.463636363636364, + "grad_norm": 0.4890111982822418, + "learning_rate": 2.31798121007328e-06, + "loss": 0.4484, + "step": 506 + }, + { + "epoch": 3.4704545454545457, + "grad_norm": 0.47682395577430725, + "learning_rate": 2.3078884802427394e-06, + "loss": 0.4494, + "step": 507 + }, + { + "epoch": 3.4772727272727275, + "grad_norm": 0.566819429397583, + "learning_rate": 2.29779889909649e-06, + "loss": 0.4539, + "step": 508 + }, + { + "epoch": 3.484090909090909, + "grad_norm": 0.4956795871257782, + "learning_rate": 2.2877126320015295e-06, + "loss": 0.4476, + "step": 509 + }, + { + "epoch": 3.4909090909090907, + "grad_norm": 0.5630896687507629, + "learning_rate": 2.2776298442705434e-06, + "loss": 0.4436, + "step": 510 + }, + { + "epoch": 3.4977272727272726, + "grad_norm": 0.9436272382736206, + "learning_rate": 2.267550701159186e-06, + "loss": 0.4558, + "step": 511 + }, + { + "epoch": 3.5045454545454544, + "grad_norm": 0.43885400891304016, + "learning_rate": 2.25747536786338e-06, + "loss": 0.4375, + "step": 512 + }, + { + "epoch": 3.5113636363636362, + "grad_norm": 0.762102484703064, + "learning_rate": 2.247404009516605e-06, + "loss": 0.4491, + "step": 513 + }, + { + "epoch": 3.518181818181818, + "grad_norm": 0.46883025765419006, + "learning_rate": 2.2373367911871904e-06, + "loss": 0.4393, + "step": 514 + }, + { + "epoch": 3.525, + "grad_norm": 0.4758882522583008, + "learning_rate": 2.227273877875613e-06, + "loss": 0.4446, + "step": 515 + }, + { + "epoch": 3.5318181818181817, + "grad_norm": 0.4799712300300598, + "learning_rate": 2.2172154345117896e-06, + "loss": 0.455, + "step": 516 + }, + { + "epoch": 3.5386363636363636, + "grad_norm": 0.46632879972457886, + "learning_rate": 2.207161625952376e-06, + "loss": 0.4481, + "step": 517 + }, + { + "epoch": 3.5454545454545454, + "grad_norm": 0.6286160945892334, + "learning_rate": 2.1971126169780636e-06, + "loss": 0.4436, + "step": 518 + }, + { + "epoch": 3.5522727272727272, + "grad_norm": 0.4679605960845947, + "learning_rate": 2.1870685722908784e-06, + "loss": 0.4575, + "step": 519 + }, + { + "epoch": 3.559090909090909, + "grad_norm": 0.495728075504303, + "learning_rate": 2.1770296565114847e-06, + "loss": 0.4399, + "step": 520 + }, + { + "epoch": 3.565909090909091, + "grad_norm": 0.4634799361228943, + "learning_rate": 2.166996034176482e-06, + "loss": 0.4553, + "step": 521 + }, + { + "epoch": 3.5727272727272728, + "grad_norm": 0.4464644491672516, + "learning_rate": 2.1569678697357126e-06, + "loss": 0.4693, + "step": 522 + }, + { + "epoch": 3.5795454545454546, + "grad_norm": 0.4530065655708313, + "learning_rate": 2.1469453275495634e-06, + "loss": 0.4418, + "step": 523 + }, + { + "epoch": 3.5863636363636364, + "grad_norm": 0.4806428551673889, + "learning_rate": 2.136928571886275e-06, + "loss": 0.4412, + "step": 524 + }, + { + "epoch": 3.5931818181818183, + "grad_norm": 0.46114400029182434, + "learning_rate": 2.126917766919245e-06, + "loss": 0.4502, + "step": 525 + }, + { + "epoch": 3.6, + "grad_norm": 0.7423635125160217, + "learning_rate": 2.1169130767243424e-06, + "loss": 0.4573, + "step": 526 + }, + { + "epoch": 3.606818181818182, + "grad_norm": 0.47871625423431396, + "learning_rate": 2.1069146652772142e-06, + "loss": 0.4427, + "step": 527 + }, + { + "epoch": 3.6136363636363638, + "grad_norm": 0.45554301142692566, + "learning_rate": 2.0969226964506007e-06, + "loss": 0.4429, + "step": 528 + }, + { + "epoch": 3.6204545454545456, + "grad_norm": 0.44822901487350464, + "learning_rate": 2.0869373340116467e-06, + "loss": 0.4616, + "step": 529 + }, + { + "epoch": 3.6272727272727274, + "grad_norm": 0.4922308027744293, + "learning_rate": 2.0769587416192212e-06, + "loss": 0.4482, + "step": 530 + }, + { + "epoch": 3.634090909090909, + "grad_norm": 0.4700371026992798, + "learning_rate": 2.066987082821231e-06, + "loss": 0.4652, + "step": 531 + }, + { + "epoch": 3.6409090909090907, + "grad_norm": 0.5513899326324463, + "learning_rate": 2.0570225210519433e-06, + "loss": 0.4542, + "step": 532 + }, + { + "epoch": 3.6477272727272725, + "grad_norm": 0.47292518615722656, + "learning_rate": 2.047065219629306e-06, + "loss": 0.4675, + "step": 533 + }, + { + "epoch": 3.6545454545454543, + "grad_norm": 0.5037868618965149, + "learning_rate": 2.0371153417522703e-06, + "loss": 0.4637, + "step": 534 + }, + { + "epoch": 3.661363636363636, + "grad_norm": 0.4578365087509155, + "learning_rate": 2.0271730504981165e-06, + "loss": 0.4432, + "step": 535 + }, + { + "epoch": 3.668181818181818, + "grad_norm": 0.518247127532959, + "learning_rate": 2.0172385088197804e-06, + "loss": 0.4574, + "step": 536 + }, + { + "epoch": 3.675, + "grad_norm": 0.5211175680160522, + "learning_rate": 2.007311879543185e-06, + "loss": 0.4607, + "step": 537 + }, + { + "epoch": 3.6818181818181817, + "grad_norm": 0.4695160686969757, + "learning_rate": 1.9973933253645684e-06, + "loss": 0.447, + "step": 538 + }, + { + "epoch": 3.6886363636363635, + "grad_norm": 0.5875166654586792, + "learning_rate": 1.9874830088478196e-06, + "loss": 0.4615, + "step": 539 + }, + { + "epoch": 3.6954545454545453, + "grad_norm": 0.4762527644634247, + "learning_rate": 1.9775810924218126e-06, + "loss": 0.4677, + "step": 540 + }, + { + "epoch": 3.702272727272727, + "grad_norm": 0.47306928038597107, + "learning_rate": 1.967687738377746e-06, + "loss": 0.4526, + "step": 541 + }, + { + "epoch": 3.709090909090909, + "grad_norm": 0.44723549485206604, + "learning_rate": 1.9578031088664812e-06, + "loss": 0.469, + "step": 542 + }, + { + "epoch": 3.715909090909091, + "grad_norm": 0.4447214901447296, + "learning_rate": 1.9479273658958852e-06, + "loss": 0.4375, + "step": 543 + }, + { + "epoch": 3.7227272727272727, + "grad_norm": 0.5299004316329956, + "learning_rate": 1.9380606713281773e-06, + "loss": 0.4544, + "step": 544 + }, + { + "epoch": 3.7295454545454545, + "grad_norm": 0.4843369424343109, + "learning_rate": 1.928203186877273e-06, + "loss": 0.4692, + "step": 545 + }, + { + "epoch": 3.7363636363636363, + "grad_norm": 0.4805986285209656, + "learning_rate": 1.9183550741061354e-06, + "loss": 0.4498, + "step": 546 + }, + { + "epoch": 3.743181818181818, + "grad_norm": 0.6539422273635864, + "learning_rate": 1.9085164944241275e-06, + "loss": 0.4398, + "step": 547 + }, + { + "epoch": 3.75, + "grad_norm": 0.47688615322113037, + "learning_rate": 1.8986876090843668e-06, + "loss": 0.4642, + "step": 548 + }, + { + "epoch": 3.756818181818182, + "grad_norm": 0.4808521270751953, + "learning_rate": 1.8888685791810784e-06, + "loss": 0.4166, + "step": 549 + }, + { + "epoch": 3.7636363636363637, + "grad_norm": 0.49430814385414124, + "learning_rate": 1.8790595656469628e-06, + "loss": 0.4404, + "step": 550 + }, + { + "epoch": 3.7704545454545455, + "grad_norm": 0.5284092426300049, + "learning_rate": 1.86926072925055e-06, + "loss": 0.445, + "step": 551 + }, + { + "epoch": 3.7772727272727273, + "grad_norm": 0.4650699496269226, + "learning_rate": 1.8594722305935691e-06, + "loss": 0.4381, + "step": 552 + }, + { + "epoch": 3.784090909090909, + "grad_norm": 0.4954371452331543, + "learning_rate": 1.8496942301083142e-06, + "loss": 0.4534, + "step": 553 + }, + { + "epoch": 3.790909090909091, + "grad_norm": 0.4756702482700348, + "learning_rate": 1.8399268880550174e-06, + "loss": 0.4563, + "step": 554 + }, + { + "epoch": 3.797727272727273, + "grad_norm": 0.4566079378128052, + "learning_rate": 1.8301703645192178e-06, + "loss": 0.4422, + "step": 555 + }, + { + "epoch": 3.8045454545454547, + "grad_norm": 1.6503181457519531, + "learning_rate": 1.8204248194091429e-06, + "loss": 0.4548, + "step": 556 + }, + { + "epoch": 3.8113636363636365, + "grad_norm": 0.5488978028297424, + "learning_rate": 1.8106904124530839e-06, + "loss": 0.4398, + "step": 557 + }, + { + "epoch": 3.8181818181818183, + "grad_norm": 0.44948896765708923, + "learning_rate": 1.800967303196778e-06, + "loss": 0.4331, + "step": 558 + }, + { + "epoch": 3.825, + "grad_norm": 0.49597036838531494, + "learning_rate": 1.7912556510007967e-06, + "loss": 0.4409, + "step": 559 + }, + { + "epoch": 3.831818181818182, + "grad_norm": 0.5093109011650085, + "learning_rate": 1.7815556150379298e-06, + "loss": 0.4615, + "step": 560 + }, + { + "epoch": 3.838636363636364, + "grad_norm": 0.4719341993331909, + "learning_rate": 1.77186735429058e-06, + "loss": 0.4316, + "step": 561 + }, + { + "epoch": 3.8454545454545457, + "grad_norm": 0.5281686782836914, + "learning_rate": 1.7621910275481544e-06, + "loss": 0.4474, + "step": 562 + }, + { + "epoch": 3.8522727272727275, + "grad_norm": 0.4955421984195709, + "learning_rate": 1.7525267934044642e-06, + "loss": 0.462, + "step": 563 + }, + { + "epoch": 3.8590909090909093, + "grad_norm": 0.4997479021549225, + "learning_rate": 1.7428748102551237e-06, + "loss": 0.4546, + "step": 564 + }, + { + "epoch": 3.865909090909091, + "grad_norm": 0.47572800517082214, + "learning_rate": 1.7332352362949546e-06, + "loss": 0.4587, + "step": 565 + }, + { + "epoch": 3.8727272727272726, + "grad_norm": 0.5156065225601196, + "learning_rate": 1.7236082295153948e-06, + "loss": 0.4375, + "step": 566 + }, + { + "epoch": 3.8795454545454544, + "grad_norm": 0.44926485419273376, + "learning_rate": 1.7139939477019057e-06, + "loss": 0.45, + "step": 567 + }, + { + "epoch": 3.8863636363636362, + "grad_norm": 0.5433095693588257, + "learning_rate": 1.7043925484313911e-06, + "loss": 0.4464, + "step": 568 + }, + { + "epoch": 3.893181818181818, + "grad_norm": 0.4719317555427551, + "learning_rate": 1.6948041890696076e-06, + "loss": 0.4467, + "step": 569 + }, + { + "epoch": 3.9, + "grad_norm": 0.4679432511329651, + "learning_rate": 1.685229026768593e-06, + "loss": 0.4577, + "step": 570 + }, + { + "epoch": 3.9068181818181817, + "grad_norm": 0.5053668022155762, + "learning_rate": 1.6756672184640847e-06, + "loss": 0.4379, + "step": 571 + }, + { + "epoch": 3.9136363636363636, + "grad_norm": 0.4880099594593048, + "learning_rate": 1.6661189208729492e-06, + "loss": 0.4585, + "step": 572 + }, + { + "epoch": 3.9204545454545454, + "grad_norm": 0.44481492042541504, + "learning_rate": 1.6565842904906154e-06, + "loss": 0.4319, + "step": 573 + }, + { + "epoch": 3.9272727272727272, + "grad_norm": 0.4998759329319, + "learning_rate": 1.6470634835885097e-06, + "loss": 0.4349, + "step": 574 + }, + { + "epoch": 3.934090909090909, + "grad_norm": 0.4702436625957489, + "learning_rate": 1.6375566562114903e-06, + "loss": 0.4409, + "step": 575 + }, + { + "epoch": 3.940909090909091, + "grad_norm": 1.1932134628295898, + "learning_rate": 1.6280639641752944e-06, + "loss": 0.4435, + "step": 576 + }, + { + "epoch": 3.9477272727272728, + "grad_norm": 0.46057313680648804, + "learning_rate": 1.6185855630639818e-06, + "loss": 0.451, + "step": 577 + }, + { + "epoch": 3.9545454545454546, + "grad_norm": 0.5858541131019592, + "learning_rate": 1.6091216082273875e-06, + "loss": 0.4662, + "step": 578 + }, + { + "epoch": 3.9613636363636364, + "grad_norm": 0.46951207518577576, + "learning_rate": 1.5996722547785722e-06, + "loss": 0.4561, + "step": 579 + }, + { + "epoch": 3.9681818181818183, + "grad_norm": 0.4747212529182434, + "learning_rate": 1.5902376575912815e-06, + "loss": 0.4507, + "step": 580 + }, + { + "epoch": 3.975, + "grad_norm": 0.5081924200057983, + "learning_rate": 1.580817971297409e-06, + "loss": 0.4649, + "step": 581 + }, + { + "epoch": 3.981818181818182, + "grad_norm": 0.4900786280632019, + "learning_rate": 1.5714133502844591e-06, + "loss": 0.4587, + "step": 582 + }, + { + "epoch": 3.9886363636363638, + "grad_norm": 0.4827897250652313, + "learning_rate": 1.56202394869302e-06, + "loss": 0.441, + "step": 583 + }, + { + "epoch": 3.9954545454545456, + "grad_norm": 0.499607652425766, + "learning_rate": 1.5526499204142332e-06, + "loss": 0.4559, + "step": 584 + }, + { + "epoch": 4.006818181818182, + "grad_norm": 1.06562340259552, + "learning_rate": 1.5432914190872757e-06, + "loss": 0.928, + "step": 585 + }, + { + "epoch": 4.013636363636364, + "grad_norm": 0.5927309393882751, + "learning_rate": 1.5339485980968383e-06, + "loss": 0.4446, + "step": 586 + }, + { + "epoch": 4.0204545454545455, + "grad_norm": 0.46158862113952637, + "learning_rate": 1.5246216105706124e-06, + "loss": 0.4394, + "step": 587 + }, + { + "epoch": 4.027272727272727, + "grad_norm": 0.46964529156684875, + "learning_rate": 1.5153106093767827e-06, + "loss": 0.4364, + "step": 588 + }, + { + "epoch": 4.034090909090909, + "grad_norm": 0.4228534698486328, + "learning_rate": 1.506015747121518e-06, + "loss": 0.4234, + "step": 589 + }, + { + "epoch": 4.040909090909091, + "grad_norm": 0.5211766362190247, + "learning_rate": 1.4967371761464738e-06, + "loss": 0.4428, + "step": 590 + }, + { + "epoch": 4.047727272727273, + "grad_norm": 0.4786047637462616, + "learning_rate": 1.4874750485262917e-06, + "loss": 0.4416, + "step": 591 + }, + { + "epoch": 4.054545454545455, + "grad_norm": 2.0014617443084717, + "learning_rate": 1.4782295160661103e-06, + "loss": 0.4338, + "step": 592 + }, + { + "epoch": 4.0613636363636365, + "grad_norm": 0.5174881219863892, + "learning_rate": 1.469000730299074e-06, + "loss": 0.431, + "step": 593 + }, + { + "epoch": 4.068181818181818, + "grad_norm": 0.4879564344882965, + "learning_rate": 1.4597888424838519e-06, + "loss": 0.4236, + "step": 594 + }, + { + "epoch": 4.075, + "grad_norm": 0.4452327787876129, + "learning_rate": 1.450594003602158e-06, + "loss": 0.4447, + "step": 595 + }, + { + "epoch": 4.081818181818182, + "grad_norm": 0.6323989033699036, + "learning_rate": 1.4414163643562755e-06, + "loss": 0.4241, + "step": 596 + }, + { + "epoch": 4.088636363636364, + "grad_norm": 0.532921314239502, + "learning_rate": 1.4322560751665873e-06, + "loss": 0.4388, + "step": 597 + }, + { + "epoch": 4.095454545454546, + "grad_norm": 0.49364256858825684, + "learning_rate": 1.4231132861691128e-06, + "loss": 0.4244, + "step": 598 + }, + { + "epoch": 4.1022727272727275, + "grad_norm": 0.47677481174468994, + "learning_rate": 1.4139881472130453e-06, + "loss": 0.4235, + "step": 599 + }, + { + "epoch": 4.109090909090909, + "grad_norm": 0.4729495644569397, + "learning_rate": 1.4048808078582943e-06, + "loss": 0.4177, + "step": 600 + }, + { + "epoch": 4.115909090909091, + "grad_norm": 0.44931760430336, + "learning_rate": 1.3957914173730366e-06, + "loss": 0.4429, + "step": 601 + }, + { + "epoch": 4.122727272727273, + "grad_norm": 0.5063349604606628, + "learning_rate": 1.3867201247312697e-06, + "loss": 0.4432, + "step": 602 + }, + { + "epoch": 4.129545454545455, + "grad_norm": 0.5439795255661011, + "learning_rate": 1.3776670786103685e-06, + "loss": 0.4455, + "step": 603 + }, + { + "epoch": 4.136363636363637, + "grad_norm": 0.4854280650615692, + "learning_rate": 1.3686324273886531e-06, + "loss": 0.4334, + "step": 604 + }, + { + "epoch": 4.1431818181818185, + "grad_norm": 0.47003456950187683, + "learning_rate": 1.359616319142949e-06, + "loss": 0.438, + "step": 605 + }, + { + "epoch": 4.15, + "grad_norm": 0.6236258149147034, + "learning_rate": 1.3506189016461674e-06, + "loss": 0.4356, + "step": 606 + }, + { + "epoch": 4.156818181818182, + "grad_norm": 0.9478527903556824, + "learning_rate": 1.341640322364878e-06, + "loss": 0.4353, + "step": 607 + }, + { + "epoch": 4.163636363636364, + "grad_norm": 0.48970767855644226, + "learning_rate": 1.3326807284568984e-06, + "loss": 0.4319, + "step": 608 + }, + { + "epoch": 4.170454545454546, + "grad_norm": 0.5102277994155884, + "learning_rate": 1.323740266768875e-06, + "loss": 0.438, + "step": 609 + }, + { + "epoch": 4.177272727272728, + "grad_norm": 0.48640403151512146, + "learning_rate": 1.3148190838338804e-06, + "loss": 0.4001, + "step": 610 + }, + { + "epoch": 4.184090909090909, + "grad_norm": 0.4933972656726837, + "learning_rate": 1.3059173258690102e-06, + "loss": 0.451, + "step": 611 + }, + { + "epoch": 4.190909090909091, + "grad_norm": 0.4632134437561035, + "learning_rate": 1.2970351387729875e-06, + "loss": 0.4396, + "step": 612 + }, + { + "epoch": 4.197727272727272, + "grad_norm": 0.5672054290771484, + "learning_rate": 1.2881726681237727e-06, + "loss": 0.4433, + "step": 613 + }, + { + "epoch": 4.204545454545454, + "grad_norm": 0.4600079655647278, + "learning_rate": 1.2793300591761742e-06, + "loss": 0.4038, + "step": 614 + }, + { + "epoch": 4.211363636363636, + "grad_norm": 0.44386839866638184, + "learning_rate": 1.27050745685947e-06, + "loss": 0.4442, + "step": 615 + }, + { + "epoch": 4.218181818181818, + "grad_norm": 0.793684184551239, + "learning_rate": 1.2617050057750322e-06, + "loss": 0.4309, + "step": 616 + }, + { + "epoch": 4.225, + "grad_norm": 0.4929650127887726, + "learning_rate": 1.252922850193955e-06, + "loss": 0.4304, + "step": 617 + }, + { + "epoch": 4.2318181818181815, + "grad_norm": 0.5073612332344055, + "learning_rate": 1.2441611340546958e-06, + "loss": 0.4271, + "step": 618 + }, + { + "epoch": 4.238636363636363, + "grad_norm": 0.4528202414512634, + "learning_rate": 1.2354200009607081e-06, + "loss": 0.4384, + "step": 619 + }, + { + "epoch": 4.245454545454545, + "grad_norm": 0.46402111649513245, + "learning_rate": 1.2266995941780934e-06, + "loss": 0.43, + "step": 620 + }, + { + "epoch": 4.252272727272727, + "grad_norm": 0.567254364490509, + "learning_rate": 1.2180000566332503e-06, + "loss": 0.3905, + "step": 621 + }, + { + "epoch": 4.259090909090909, + "grad_norm": 0.5210692286491394, + "learning_rate": 1.2093215309105352e-06, + "loss": 0.4262, + "step": 622 + }, + { + "epoch": 4.265909090909091, + "grad_norm": 1.3842746019363403, + "learning_rate": 1.2006641592499233e-06, + "loss": 0.4268, + "step": 623 + }, + { + "epoch": 4.2727272727272725, + "grad_norm": 0.49402037262916565, + "learning_rate": 1.192028083544675e-06, + "loss": 0.4244, + "step": 624 + }, + { + "epoch": 4.279545454545454, + "grad_norm": 0.4855121374130249, + "learning_rate": 1.1834134453390136e-06, + "loss": 0.4274, + "step": 625 + }, + { + "epoch": 4.286363636363636, + "grad_norm": 0.5327234864234924, + "learning_rate": 1.1748203858258056e-06, + "loss": 0.429, + "step": 626 + }, + { + "epoch": 4.293181818181818, + "grad_norm": 0.482337087392807, + "learning_rate": 1.166249045844243e-06, + "loss": 0.4253, + "step": 627 + }, + { + "epoch": 4.3, + "grad_norm": 0.9390714168548584, + "learning_rate": 1.1576995658775405e-06, + "loss": 0.4224, + "step": 628 + }, + { + "epoch": 4.306818181818182, + "grad_norm": 0.5063149929046631, + "learning_rate": 1.1491720860506273e-06, + "loss": 0.4243, + "step": 629 + }, + { + "epoch": 4.3136363636363635, + "grad_norm": 0.4565083980560303, + "learning_rate": 1.140666746127854e-06, + "loss": 0.4137, + "step": 630 + }, + { + "epoch": 4.320454545454545, + "grad_norm": 0.48249998688697815, + "learning_rate": 1.1321836855107007e-06, + "loss": 0.4143, + "step": 631 + }, + { + "epoch": 4.327272727272727, + "grad_norm": 0.5024334788322449, + "learning_rate": 1.1237230432354912e-06, + "loss": 0.4302, + "step": 632 + }, + { + "epoch": 4.334090909090909, + "grad_norm": 0.4903137981891632, + "learning_rate": 1.1152849579711187e-06, + "loss": 0.4522, + "step": 633 + }, + { + "epoch": 4.340909090909091, + "grad_norm": 0.47344261407852173, + "learning_rate": 1.1068695680167665e-06, + "loss": 0.4487, + "step": 634 + }, + { + "epoch": 4.347727272727273, + "grad_norm": 0.5437475442886353, + "learning_rate": 1.0984770112996463e-06, + "loss": 0.4255, + "step": 635 + }, + { + "epoch": 4.3545454545454545, + "grad_norm": 0.4634874165058136, + "learning_rate": 1.0901074253727338e-06, + "loss": 0.4345, + "step": 636 + }, + { + "epoch": 4.361363636363636, + "grad_norm": 0.5247130393981934, + "learning_rate": 1.0817609474125195e-06, + "loss": 0.4296, + "step": 637 + }, + { + "epoch": 4.368181818181818, + "grad_norm": 0.503746747970581, + "learning_rate": 1.0734377142167549e-06, + "loss": 0.4251, + "step": 638 + }, + { + "epoch": 4.375, + "grad_norm": 0.5096430778503418, + "learning_rate": 1.065137862202213e-06, + "loss": 0.4186, + "step": 639 + }, + { + "epoch": 4.381818181818182, + "grad_norm": 0.4813026785850525, + "learning_rate": 1.0568615274024521e-06, + "loss": 0.4265, + "step": 640 + }, + { + "epoch": 4.388636363636364, + "grad_norm": 0.45982855558395386, + "learning_rate": 1.0486088454655856e-06, + "loss": 0.4188, + "step": 641 + }, + { + "epoch": 4.3954545454545455, + "grad_norm": 0.5047855377197266, + "learning_rate": 1.0403799516520619e-06, + "loss": 0.4231, + "step": 642 + }, + { + "epoch": 4.402272727272727, + "grad_norm": 0.46901944279670715, + "learning_rate": 1.0321749808324425e-06, + "loss": 0.4173, + "step": 643 + }, + { + "epoch": 4.409090909090909, + "grad_norm": 0.48727908730506897, + "learning_rate": 1.0239940674851943e-06, + "loss": 0.4631, + "step": 644 + }, + { + "epoch": 4.415909090909091, + "grad_norm": 0.6558519005775452, + "learning_rate": 1.0158373456944856e-06, + "loss": 0.4177, + "step": 645 + }, + { + "epoch": 4.422727272727273, + "grad_norm": 0.4979148805141449, + "learning_rate": 1.0077049491479874e-06, + "loss": 0.4046, + "step": 646 + }, + { + "epoch": 4.429545454545455, + "grad_norm": 0.5404463410377502, + "learning_rate": 9.995970111346842e-07, + "loss": 0.4314, + "step": 647 + }, + { + "epoch": 4.4363636363636365, + "grad_norm": 0.7192148566246033, + "learning_rate": 9.915136645426885e-07, + "loss": 0.429, + "step": 648 + }, + { + "epoch": 4.443181818181818, + "grad_norm": 0.47399184107780457, + "learning_rate": 9.834550418570602e-07, + "loss": 0.4392, + "step": 649 + }, + { + "epoch": 4.45, + "grad_norm": 0.48218435049057007, + "learning_rate": 9.754212751576386e-07, + "loss": 0.4282, + "step": 650 + }, + { + "epoch": 4.456818181818182, + "grad_norm": 0.5437904596328735, + "learning_rate": 9.67412496116876e-07, + "loss": 0.4031, + "step": 651 + }, + { + "epoch": 4.463636363636364, + "grad_norm": 0.45721930265426636, + "learning_rate": 9.594288359976817e-07, + "loss": 0.4268, + "step": 652 + }, + { + "epoch": 4.470454545454546, + "grad_norm": 0.48765209317207336, + "learning_rate": 9.514704256512669e-07, + "loss": 0.4491, + "step": 653 + }, + { + "epoch": 4.4772727272727275, + "grad_norm": 0.45831185579299927, + "learning_rate": 9.435373955150032e-07, + "loss": 0.423, + "step": 654 + }, + { + "epoch": 4.484090909090909, + "grad_norm": 0.50241619348526, + "learning_rate": 9.35629875610283e-07, + "loss": 0.4334, + "step": 655 + }, + { + "epoch": 4.490909090909091, + "grad_norm": 0.4767427444458008, + "learning_rate": 9.277479955403887e-07, + "loss": 0.4475, + "step": 656 + }, + { + "epoch": 4.497727272727273, + "grad_norm": 0.48379838466644287, + "learning_rate": 9.198918844883714e-07, + "loss": 0.4353, + "step": 657 + }, + { + "epoch": 4.504545454545455, + "grad_norm": 0.5098893046379089, + "learning_rate": 9.120616712149291e-07, + "loss": 0.4285, + "step": 658 + }, + { + "epoch": 4.511363636363637, + "grad_norm": 0.5217880010604858, + "learning_rate": 9.042574840562982e-07, + "loss": 0.4395, + "step": 659 + }, + { + "epoch": 4.5181818181818185, + "grad_norm": 0.449683278799057, + "learning_rate": 8.964794509221508e-07, + "loss": 0.4382, + "step": 660 + }, + { + "epoch": 4.525, + "grad_norm": 0.5583493113517761, + "learning_rate": 8.887276992934976e-07, + "loss": 0.4378, + "step": 661 + }, + { + "epoch": 4.531818181818182, + "grad_norm": 0.503844678401947, + "learning_rate": 8.810023562206e-07, + "loss": 0.4222, + "step": 662 + }, + { + "epoch": 4.538636363636364, + "grad_norm": 0.46325281262397766, + "learning_rate": 8.733035483208841e-07, + "loss": 0.441, + "step": 663 + }, + { + "epoch": 4.545454545454545, + "grad_norm": 0.5824127197265625, + "learning_rate": 8.656314017768694e-07, + "loss": 0.4224, + "step": 664 + }, + { + "epoch": 4.552272727272728, + "grad_norm": 0.5601843595504761, + "learning_rate": 8.579860423340977e-07, + "loss": 0.4348, + "step": 665 + }, + { + "epoch": 4.559090909090909, + "grad_norm": 0.45808762311935425, + "learning_rate": 8.503675952990756e-07, + "loss": 0.4445, + "step": 666 + }, + { + "epoch": 4.565909090909091, + "grad_norm": 0.494069367647171, + "learning_rate": 8.427761855372169e-07, + "loss": 0.4271, + "step": 667 + }, + { + "epoch": 4.572727272727272, + "grad_norm": 0.5897441506385803, + "learning_rate": 8.352119374707979e-07, + "loss": 0.4218, + "step": 668 + }, + { + "epoch": 4.579545454545455, + "grad_norm": 0.5399529337882996, + "learning_rate": 8.276749750769186e-07, + "loss": 0.4295, + "step": 669 + }, + { + "epoch": 4.586363636363636, + "grad_norm": 0.44742587208747864, + "learning_rate": 8.20165421885469e-07, + "loss": 0.4406, + "step": 670 + }, + { + "epoch": 4.593181818181818, + "grad_norm": 0.4736122786998749, + "learning_rate": 8.126834009771079e-07, + "loss": 0.4255, + "step": 671 + }, + { + "epoch": 4.6, + "grad_norm": 0.4699339270591736, + "learning_rate": 8.052290349812419e-07, + "loss": 0.4388, + "step": 672 + }, + { + "epoch": 4.6068181818181815, + "grad_norm": 1.0179539918899536, + "learning_rate": 7.978024460740169e-07, + "loss": 0.445, + "step": 673 + }, + { + "epoch": 4.613636363636363, + "grad_norm": 0.4838411211967468, + "learning_rate": 7.904037559763162e-07, + "loss": 0.4518, + "step": 674 + }, + { + "epoch": 4.620454545454545, + "grad_norm": 0.4670931398868561, + "learning_rate": 7.83033085951764e-07, + "loss": 0.4294, + "step": 675 + }, + { + "epoch": 4.627272727272727, + "grad_norm": 0.47013846039772034, + "learning_rate": 7.756905568047393e-07, + "loss": 0.4395, + "step": 676 + }, + { + "epoch": 4.634090909090909, + "grad_norm": 0.875905454158783, + "learning_rate": 7.683762888783977e-07, + "loss": 0.4288, + "step": 677 + }, + { + "epoch": 4.640909090909091, + "grad_norm": 0.47206172347068787, + "learning_rate": 7.610904020526938e-07, + "loss": 0.4244, + "step": 678 + }, + { + "epoch": 4.6477272727272725, + "grad_norm": 0.6244398355484009, + "learning_rate": 7.538330157424212e-07, + "loss": 0.4444, + "step": 679 + }, + { + "epoch": 4.654545454545454, + "grad_norm": 0.5075065493583679, + "learning_rate": 7.466042488952521e-07, + "loss": 0.433, + "step": 680 + }, + { + "epoch": 4.661363636363636, + "grad_norm": 0.7290046811103821, + "learning_rate": 7.394042199897916e-07, + "loss": 0.4509, + "step": 681 + }, + { + "epoch": 4.668181818181818, + "grad_norm": 0.4735564589500427, + "learning_rate": 7.322330470336314e-07, + "loss": 0.425, + "step": 682 + }, + { + "epoch": 4.675, + "grad_norm": 0.4939814805984497, + "learning_rate": 7.250908475614185e-07, + "loss": 0.4058, + "step": 683 + }, + { + "epoch": 4.681818181818182, + "grad_norm": 0.44935867190361023, + "learning_rate": 7.179777386329276e-07, + "loss": 0.4132, + "step": 684 + }, + { + "epoch": 4.6886363636363635, + "grad_norm": 0.48104918003082275, + "learning_rate": 7.108938368311424e-07, + "loss": 0.4581, + "step": 685 + }, + { + "epoch": 4.695454545454545, + "grad_norm": 0.5192425847053528, + "learning_rate": 7.038392582603481e-07, + "loss": 0.4542, + "step": 686 + }, + { + "epoch": 4.702272727272727, + "grad_norm": 0.5787845253944397, + "learning_rate": 6.968141185442229e-07, + "loss": 0.421, + "step": 687 + }, + { + "epoch": 4.709090909090909, + "grad_norm": 0.4911254942417145, + "learning_rate": 6.898185328239468e-07, + "loss": 0.412, + "step": 688 + }, + { + "epoch": 4.715909090909091, + "grad_norm": 0.5646374821662903, + "learning_rate": 6.828526157563126e-07, + "loss": 0.428, + "step": 689 + }, + { + "epoch": 4.722727272727273, + "grad_norm": 0.4580458402633667, + "learning_rate": 6.759164815118493e-07, + "loss": 0.4342, + "step": 690 + }, + { + "epoch": 4.7295454545454545, + "grad_norm": 0.5442156791687012, + "learning_rate": 6.690102437729481e-07, + "loss": 0.442, + "step": 691 + }, + { + "epoch": 4.736363636363636, + "grad_norm": 0.4839988052845001, + "learning_rate": 6.621340157319998e-07, + "loss": 0.4384, + "step": 692 + }, + { + "epoch": 4.743181818181818, + "grad_norm": 0.5340309143066406, + "learning_rate": 6.552879100895396e-07, + "loss": 0.4224, + "step": 693 + }, + { + "epoch": 4.75, + "grad_norm": 0.5196749567985535, + "learning_rate": 6.484720390524008e-07, + "loss": 0.4317, + "step": 694 + }, + { + "epoch": 4.756818181818182, + "grad_norm": 0.9252122044563293, + "learning_rate": 6.416865143318757e-07, + "loss": 0.4375, + "step": 695 + }, + { + "epoch": 4.763636363636364, + "grad_norm": 0.7175036668777466, + "learning_rate": 6.349314471418849e-07, + "loss": 0.4475, + "step": 696 + }, + { + "epoch": 4.7704545454545455, + "grad_norm": 0.5099776387214661, + "learning_rate": 6.282069481971514e-07, + "loss": 0.4231, + "step": 697 + }, + { + "epoch": 4.777272727272727, + "grad_norm": 0.6147667169570923, + "learning_rate": 6.2151312771139e-07, + "loss": 0.4437, + "step": 698 + }, + { + "epoch": 4.784090909090909, + "grad_norm": 0.45040184259414673, + "learning_rate": 6.148500953954992e-07, + "loss": 0.4086, + "step": 699 + }, + { + "epoch": 4.790909090909091, + "grad_norm": 0.45515206456184387, + "learning_rate": 6.082179604557617e-07, + "loss": 0.4147, + "step": 700 + }, + { + "epoch": 4.797727272727273, + "grad_norm": 0.5819535255432129, + "learning_rate": 6.016168315920593e-07, + "loss": 0.4126, + "step": 701 + }, + { + "epoch": 4.804545454545455, + "grad_norm": 0.5001727342605591, + "learning_rate": 5.950468169960846e-07, + "loss": 0.4498, + "step": 702 + }, + { + "epoch": 4.8113636363636365, + "grad_norm": 0.5675827860832214, + "learning_rate": 5.885080243495731e-07, + "loss": 0.443, + "step": 703 + }, + { + "epoch": 4.818181818181818, + "grad_norm": 0.4939325749874115, + "learning_rate": 5.820005608225345e-07, + "loss": 0.4102, + "step": 704 + }, + { + "epoch": 4.825, + "grad_norm": 0.5586816072463989, + "learning_rate": 5.755245330715014e-07, + "loss": 0.4322, + "step": 705 + }, + { + "epoch": 4.831818181818182, + "grad_norm": 0.4761182367801666, + "learning_rate": 5.690800472377747e-07, + "loss": 0.4386, + "step": 706 + }, + { + "epoch": 4.838636363636364, + "grad_norm": 0.4821193218231201, + "learning_rate": 5.626672089456887e-07, + "loss": 0.4397, + "step": 707 + }, + { + "epoch": 4.845454545454546, + "grad_norm": 0.5768219828605652, + "learning_rate": 5.562861233008774e-07, + "loss": 0.4328, + "step": 708 + }, + { + "epoch": 4.8522727272727275, + "grad_norm": 0.49105873703956604, + "learning_rate": 5.499368948885528e-07, + "loss": 0.4331, + "step": 709 + }, + { + "epoch": 4.859090909090909, + "grad_norm": 0.5111908316612244, + "learning_rate": 5.436196277717928e-07, + "loss": 0.4165, + "step": 710 + }, + { + "epoch": 4.865909090909091, + "grad_norm": 0.49336686730384827, + "learning_rate": 5.373344254898313e-07, + "loss": 0.4326, + "step": 711 + }, + { + "epoch": 4.872727272727273, + "grad_norm": 0.5113787651062012, + "learning_rate": 5.310813910563645e-07, + "loss": 0.4259, + "step": 712 + }, + { + "epoch": 4.879545454545455, + "grad_norm": 1.0626885890960693, + "learning_rate": 5.24860626957861e-07, + "loss": 0.4279, + "step": 713 + }, + { + "epoch": 4.886363636363637, + "grad_norm": 0.5634323954582214, + "learning_rate": 5.186722351518822e-07, + "loss": 0.4315, + "step": 714 + }, + { + "epoch": 4.8931818181818185, + "grad_norm": 0.5127065181732178, + "learning_rate": 5.125163170654138e-07, + "loss": 0.4142, + "step": 715 + }, + { + "epoch": 4.9, + "grad_norm": 0.4855251610279083, + "learning_rate": 5.063929735931985e-07, + "loss": 0.4024, + "step": 716 + }, + { + "epoch": 4.906818181818182, + "grad_norm": 0.9671811461448669, + "learning_rate": 5.003023050960865e-07, + "loss": 0.4201, + "step": 717 + }, + { + "epoch": 4.913636363636364, + "grad_norm": 0.5237546563148499, + "learning_rate": 4.94244411399388e-07, + "loss": 0.4175, + "step": 718 + }, + { + "epoch": 4.920454545454545, + "grad_norm": 0.46599382162094116, + "learning_rate": 4.882193917912398e-07, + "loss": 0.4611, + "step": 719 + }, + { + "epoch": 4.927272727272728, + "grad_norm": 0.5131009221076965, + "learning_rate": 4.822273450209767e-07, + "loss": 0.4081, + "step": 720 + }, + { + "epoch": 4.934090909090909, + "grad_norm": 0.504088819026947, + "learning_rate": 4.7626836929751035e-07, + "loss": 0.4533, + "step": 721 + }, + { + "epoch": 4.940909090909091, + "grad_norm": 0.5421441197395325, + "learning_rate": 4.703425622877239e-07, + "loss": 0.4275, + "step": 722 + }, + { + "epoch": 4.947727272727272, + "grad_norm": 0.5401121377944946, + "learning_rate": 4.6445002111486866e-07, + "loss": 0.4449, + "step": 723 + }, + { + "epoch": 4.954545454545455, + "grad_norm": 0.46559831500053406, + "learning_rate": 4.5859084235697236e-07, + "loss": 0.4449, + "step": 724 + }, + { + "epoch": 4.961363636363636, + "grad_norm": 0.5192095041275024, + "learning_rate": 4.527651220452589e-07, + "loss": 0.4192, + "step": 725 + }, + { + "epoch": 4.968181818181818, + "grad_norm": 0.4992344379425049, + "learning_rate": 4.469729556625704e-07, + "loss": 0.4212, + "step": 726 + }, + { + "epoch": 4.975, + "grad_norm": 0.5019513964653015, + "learning_rate": 4.412144381418049e-07, + "loss": 0.4549, + "step": 727 + }, + { + "epoch": 4.9818181818181815, + "grad_norm": 0.47997716069221497, + "learning_rate": 4.354896638643591e-07, + "loss": 0.4397, + "step": 728 + }, + { + "epoch": 4.988636363636363, + "grad_norm": 0.5492618680000305, + "learning_rate": 4.2979872665858266e-07, + "loss": 0.4301, + "step": 729 + }, + { + "epoch": 4.995454545454545, + "grad_norm": 0.48399415612220764, + "learning_rate": 4.2414171979824e-07, + "loss": 0.4062, + "step": 730 + }, + { + "epoch": 5.006818181818182, + "grad_norm": 1.002280831336975, + "learning_rate": 4.1851873600098154e-07, + "loss": 0.7908, + "step": 731 + }, + { + "epoch": 5.013636363636364, + "grad_norm": 0.45833030343055725, + "learning_rate": 4.129298674268226e-07, + "loss": 0.4289, + "step": 732 + }, + { + "epoch": 5.0204545454545455, + "grad_norm": 0.4525989294052124, + "learning_rate": 4.073752056766342e-07, + "loss": 0.4246, + "step": 733 + }, + { + "epoch": 5.027272727272727, + "grad_norm": 0.4920106828212738, + "learning_rate": 4.0185484179064427e-07, + "loss": 0.4061, + "step": 734 + }, + { + "epoch": 5.034090909090909, + "grad_norm": 0.483296275138855, + "learning_rate": 3.9636886624694e-07, + "loss": 0.4023, + "step": 735 + }, + { + "epoch": 5.040909090909091, + "grad_norm": 0.4520129859447479, + "learning_rate": 3.9091736895998907e-07, + "loss": 0.4306, + "step": 736 + }, + { + "epoch": 5.047727272727273, + "grad_norm": 0.5209662914276123, + "learning_rate": 3.855004392791645e-07, + "loss": 0.3731, + "step": 737 + }, + { + "epoch": 5.054545454545455, + "grad_norm": 0.4719908833503723, + "learning_rate": 3.801181659872805e-07, + "loss": 0.411, + "step": 738 + }, + { + "epoch": 5.0613636363636365, + "grad_norm": 0.7588437795639038, + "learning_rate": 3.7477063729913804e-07, + "loss": 0.4267, + "step": 739 + }, + { + "epoch": 5.068181818181818, + "grad_norm": 0.45445874333381653, + "learning_rate": 3.6945794086007706e-07, + "loss": 0.4088, + "step": 740 + }, + { + "epoch": 5.075, + "grad_norm": 0.4535755217075348, + "learning_rate": 3.6418016374454247e-07, + "loss": 0.4142, + "step": 741 + }, + { + "epoch": 5.081818181818182, + "grad_norm": 0.4743533432483673, + "learning_rate": 3.5893739245465465e-07, + "loss": 0.425, + "step": 742 + }, + { + "epoch": 5.088636363636364, + "grad_norm": 0.4621371030807495, + "learning_rate": 3.537297129187925e-07, + "loss": 0.4381, + "step": 743 + }, + { + "epoch": 5.095454545454546, + "grad_norm": 0.5395707488059998, + "learning_rate": 3.485572104901869e-07, + "loss": 0.428, + "step": 744 + }, + { + "epoch": 5.1022727272727275, + "grad_norm": 0.4829772114753723, + "learning_rate": 3.4341996994551957e-07, + "loss": 0.4342, + "step": 745 + }, + { + "epoch": 5.109090909090909, + "grad_norm": 0.5233829021453857, + "learning_rate": 3.383180754835344e-07, + "loss": 0.4383, + "step": 746 + }, + { + "epoch": 5.115909090909091, + "grad_norm": 0.46988317370414734, + "learning_rate": 3.3325161072365636e-07, + "loss": 0.4152, + "step": 747 + }, + { + "epoch": 5.122727272727273, + "grad_norm": 0.5108771920204163, + "learning_rate": 3.2822065870462216e-07, + "loss": 0.3991, + "step": 748 + }, + { + "epoch": 5.129545454545455, + "grad_norm": 0.501301646232605, + "learning_rate": 3.232253018831208e-07, + "loss": 0.4073, + "step": 749 + }, + { + "epoch": 5.136363636363637, + "grad_norm": 1.0080974102020264, + "learning_rate": 3.182656221324384e-07, + "loss": 0.4231, + "step": 750 + }, + { + "epoch": 5.1431818181818185, + "grad_norm": 0.5004135966300964, + "learning_rate": 3.133417007411188e-07, + "loss": 0.4269, + "step": 751 + }, + { + "epoch": 5.15, + "grad_norm": 0.48254939913749695, + "learning_rate": 3.08453618411631e-07, + "loss": 0.4329, + "step": 752 + }, + { + "epoch": 5.156818181818182, + "grad_norm": 0.4400676488876343, + "learning_rate": 3.036014552590455e-07, + "loss": 0.4183, + "step": 753 + }, + { + "epoch": 5.163636363636364, + "grad_norm": 0.6217389106750488, + "learning_rate": 2.98785290809723e-07, + "loss": 0.419, + "step": 754 + }, + { + "epoch": 5.170454545454546, + "grad_norm": 0.48898404836654663, + "learning_rate": 2.940052040000091e-07, + "loss": 0.4161, + "step": 755 + }, + { + "epoch": 5.177272727272728, + "grad_norm": 0.5164027214050293, + "learning_rate": 2.892612731749414e-07, + "loss": 0.4266, + "step": 756 + }, + { + "epoch": 5.184090909090909, + "grad_norm": 0.4923621416091919, + "learning_rate": 2.8455357608696497e-07, + "loss": 0.4202, + "step": 757 + }, + { + "epoch": 5.190909090909091, + "grad_norm": 0.5050857067108154, + "learning_rate": 2.798821898946588e-07, + "loss": 0.4308, + "step": 758 + }, + { + "epoch": 5.197727272727272, + "grad_norm": 0.5368906855583191, + "learning_rate": 2.7524719116147154e-07, + "loss": 0.424, + "step": 759 + }, + { + "epoch": 5.204545454545454, + "grad_norm": 0.5245484709739685, + "learning_rate": 2.706486558544644e-07, + "loss": 0.4156, + "step": 760 + }, + { + "epoch": 5.211363636363636, + "grad_norm": 0.5602469444274902, + "learning_rate": 2.6608665934306775e-07, + "loss": 0.4163, + "step": 761 + }, + { + "epoch": 5.218181818181818, + "grad_norm": 0.4578344523906708, + "learning_rate": 2.615612763978462e-07, + "loss": 0.4222, + "step": 762 + }, + { + "epoch": 5.225, + "grad_norm": 0.5114238262176514, + "learning_rate": 2.570725811892727e-07, + "loss": 0.4312, + "step": 763 + }, + { + "epoch": 5.2318181818181815, + "grad_norm": 0.4622151255607605, + "learning_rate": 2.52620647286512e-07, + "loss": 0.4287, + "step": 764 + }, + { + "epoch": 5.238636363636363, + "grad_norm": 5.495304107666016, + "learning_rate": 2.4820554765621534e-07, + "loss": 0.4084, + "step": 765 + }, + { + "epoch": 5.245454545454545, + "grad_norm": 0.502539873123169, + "learning_rate": 2.438273546613257e-07, + "loss": 0.4126, + "step": 766 + }, + { + "epoch": 5.252272727272727, + "grad_norm": 0.543896496295929, + "learning_rate": 2.394861400598894e-07, + "loss": 0.4187, + "step": 767 + }, + { + "epoch": 5.259090909090909, + "grad_norm": 1.6630384922027588, + "learning_rate": 2.3518197500388278e-07, + "loss": 0.4157, + "step": 768 + }, + { + "epoch": 5.265909090909091, + "grad_norm": 0.4552000164985657, + "learning_rate": 2.3091493003804476e-07, + "loss": 0.4271, + "step": 769 + }, + { + "epoch": 5.2727272727272725, + "grad_norm": 0.46218186616897583, + "learning_rate": 2.2668507509871957e-07, + "loss": 0.42, + "step": 770 + }, + { + "epoch": 5.279545454545454, + "grad_norm": 0.5455384254455566, + "learning_rate": 2.2249247951271174e-07, + "loss": 0.4245, + "step": 771 + }, + { + "epoch": 5.286363636363636, + "grad_norm": 0.47305184602737427, + "learning_rate": 2.1833721199614992e-07, + "loss": 0.4368, + "step": 772 + }, + { + "epoch": 5.293181818181818, + "grad_norm": 0.5162175297737122, + "learning_rate": 2.1421934065335909e-07, + "loss": 0.4232, + "step": 773 + }, + { + "epoch": 5.3, + "grad_norm": 0.46097734570503235, + "learning_rate": 2.1013893297574777e-07, + "loss": 0.4271, + "step": 774 + }, + { + "epoch": 5.306818181818182, + "grad_norm": 1.0151066780090332, + "learning_rate": 2.0609605584069741e-07, + "loss": 0.4137, + "step": 775 + }, + { + "epoch": 5.3136363636363635, + "grad_norm": 0.46064189076423645, + "learning_rate": 2.020907755104698e-07, + "loss": 0.4252, + "step": 776 + }, + { + "epoch": 5.320454545454545, + "grad_norm": 0.5077064633369446, + "learning_rate": 1.9812315763111823e-07, + "loss": 0.4199, + "step": 777 + }, + { + "epoch": 5.327272727272727, + "grad_norm": 0.4949788749217987, + "learning_rate": 1.9419326723141534e-07, + "loss": 0.4038, + "step": 778 + }, + { + "epoch": 5.334090909090909, + "grad_norm": 0.4672517776489258, + "learning_rate": 1.9030116872178317e-07, + "loss": 0.4141, + "step": 779 + }, + { + "epoch": 5.340909090909091, + "grad_norm": 0.5564689636230469, + "learning_rate": 1.864469258932397e-07, + "loss": 0.4096, + "step": 780 + }, + { + "epoch": 5.347727272727273, + "grad_norm": 0.4476451575756073, + "learning_rate": 1.8263060191635317e-07, + "loss": 0.418, + "step": 781 + }, + { + "epoch": 5.3545454545454545, + "grad_norm": 0.5125645995140076, + "learning_rate": 1.788522593402059e-07, + "loss": 0.4163, + "step": 782 + }, + { + "epoch": 5.361363636363636, + "grad_norm": 0.46515682339668274, + "learning_rate": 1.7511196009137087e-07, + "loss": 0.4232, + "step": 783 + }, + { + "epoch": 5.368181818181818, + "grad_norm": 0.48179587721824646, + "learning_rate": 1.7140976547289438e-07, + "loss": 0.4295, + "step": 784 + }, + { + "epoch": 5.375, + "grad_norm": 0.4904819428920746, + "learning_rate": 1.6774573616329336e-07, + "loss": 0.415, + "step": 785 + }, + { + "epoch": 5.381818181818182, + "grad_norm": 0.49641627073287964, + "learning_rate": 1.6411993221555928e-07, + "loss": 0.4174, + "step": 786 + }, + { + "epoch": 5.388636363636364, + "grad_norm": 0.4949050545692444, + "learning_rate": 1.605324130561753e-07, + "loss": 0.4135, + "step": 787 + }, + { + "epoch": 5.3954545454545455, + "grad_norm": 0.5043060183525085, + "learning_rate": 1.5698323748414123e-07, + "loss": 0.4438, + "step": 788 + }, + { + "epoch": 5.402272727272727, + "grad_norm": 0.4563829004764557, + "learning_rate": 1.5347246367000994e-07, + "loss": 0.4486, + "step": 789 + }, + { + "epoch": 5.409090909090909, + "grad_norm": 0.505873441696167, + "learning_rate": 1.5000014915493467e-07, + "loss": 0.412, + "step": 790 + }, + { + "epoch": 5.415909090909091, + "grad_norm": 0.5464740991592407, + "learning_rate": 1.4656635084972475e-07, + "loss": 0.4345, + "step": 791 + }, + { + "epoch": 5.422727272727273, + "grad_norm": 0.4924403727054596, + "learning_rate": 1.4317112503391433e-07, + "loss": 0.4235, + "step": 792 + }, + { + "epoch": 5.429545454545455, + "grad_norm": 0.5023462176322937, + "learning_rate": 1.398145273548396e-07, + "loss": 0.4247, + "step": 793 + }, + { + "epoch": 5.4363636363636365, + "grad_norm": 0.530540943145752, + "learning_rate": 1.3649661282672478e-07, + "loss": 0.4339, + "step": 794 + }, + { + "epoch": 5.443181818181818, + "grad_norm": 0.4698546826839447, + "learning_rate": 1.3321743582978303e-07, + "loss": 0.4136, + "step": 795 + }, + { + "epoch": 5.45, + "grad_norm": 0.48455655574798584, + "learning_rate": 1.2997705010932394e-07, + "loss": 0.4033, + "step": 796 + }, + { + "epoch": 5.456818181818182, + "grad_norm": 0.48737505078315735, + "learning_rate": 1.2677550877487232e-07, + "loss": 0.4039, + "step": 797 + }, + { + "epoch": 5.463636363636364, + "grad_norm": 0.46107780933380127, + "learning_rate": 1.2361286429929953e-07, + "loss": 0.4264, + "step": 798 + }, + { + "epoch": 5.470454545454546, + "grad_norm": 0.5445685386657715, + "learning_rate": 1.20489168517961e-07, + "loss": 0.4236, + "step": 799 + }, + { + "epoch": 5.4772727272727275, + "grad_norm": 0.4544489085674286, + "learning_rate": 1.1740447262784782e-07, + "loss": 0.4283, + "step": 800 + }, + { + "epoch": 5.484090909090909, + "grad_norm": 0.6632809638977051, + "learning_rate": 1.1435882718674823e-07, + "loss": 0.4315, + "step": 801 + }, + { + "epoch": 5.490909090909091, + "grad_norm": 0.5885090231895447, + "learning_rate": 1.1135228211241827e-07, + "loss": 0.4235, + "step": 802 + }, + { + "epoch": 5.497727272727273, + "grad_norm": 0.505340576171875, + "learning_rate": 1.0838488668176383e-07, + "loss": 0.415, + "step": 803 + }, + { + "epoch": 5.504545454545455, + "grad_norm": 0.5026459097862244, + "learning_rate": 1.054566895300324e-07, + "loss": 0.4219, + "step": 804 + }, + { + "epoch": 5.511363636363637, + "grad_norm": 0.5613178610801697, + "learning_rate": 1.0256773865001679e-07, + "loss": 0.419, + "step": 805 + }, + { + "epoch": 5.5181818181818185, + "grad_norm": 0.45813271403312683, + "learning_rate": 9.97180813912682e-08, + "loss": 0.4025, + "step": 806 + }, + { + "epoch": 5.525, + "grad_norm": 0.48044487833976746, + "learning_rate": 9.690776445932082e-08, + "loss": 0.4215, + "step": 807 + }, + { + "epoch": 5.531818181818182, + "grad_norm": 0.5943904519081116, + "learning_rate": 9.413683391492456e-08, + "loss": 0.4339, + "step": 808 + }, + { + "epoch": 5.538636363636364, + "grad_norm": 0.4397302567958832, + "learning_rate": 9.140533517329214e-08, + "loss": 0.4173, + "step": 809 + }, + { + "epoch": 5.545454545454545, + "grad_norm": 0.48651400208473206, + "learning_rate": 8.871331300335322e-08, + "loss": 0.4091, + "step": 810 + }, + { + "epoch": 5.552272727272728, + "grad_norm": 0.5245802998542786, + "learning_rate": 8.606081152702145e-08, + "loss": 0.4263, + "step": 811 + }, + { + "epoch": 5.559090909090909, + "grad_norm": 0.6158308982849121, + "learning_rate": 8.344787421847216e-08, + "loss": 0.4295, + "step": 812 + }, + { + "epoch": 5.565909090909091, + "grad_norm": 0.7189339995384216, + "learning_rate": 8.087454390342725e-08, + "loss": 0.4313, + "step": 813 + }, + { + "epoch": 5.572727272727272, + "grad_norm": 0.5823023915290833, + "learning_rate": 7.834086275845587e-08, + "loss": 0.43, + "step": 814 + }, + { + "epoch": 5.579545454545455, + "grad_norm": 0.490366667509079, + "learning_rate": 7.584687231028121e-08, + "loss": 0.4312, + "step": 815 + }, + { + "epoch": 5.586363636363636, + "grad_norm": 0.501454770565033, + "learning_rate": 7.339261343510207e-08, + "loss": 0.4241, + "step": 816 + }, + { + "epoch": 5.593181818181818, + "grad_norm": 0.5100012421607971, + "learning_rate": 7.097812635792095e-08, + "loss": 0.4255, + "step": 817 + }, + { + "epoch": 5.6, + "grad_norm": 0.5133018493652344, + "learning_rate": 6.860345065188512e-08, + "loss": 0.4142, + "step": 818 + }, + { + "epoch": 5.6068181818181815, + "grad_norm": 0.48468226194381714, + "learning_rate": 6.626862523763905e-08, + "loss": 0.4064, + "step": 819 + }, + { + "epoch": 5.613636363636363, + "grad_norm": 0.7193787097930908, + "learning_rate": 6.397368838268497e-08, + "loss": 0.425, + "step": 820 + }, + { + "epoch": 5.620454545454545, + "grad_norm": 0.4811626672744751, + "learning_rate": 6.171867770075723e-08, + "loss": 0.4162, + "step": 821 + }, + { + "epoch": 5.627272727272727, + "grad_norm": 0.49769723415374756, + "learning_rate": 5.9503630151205025e-08, + "loss": 0.4428, + "step": 822 + }, + { + "epoch": 5.634090909090909, + "grad_norm": 0.49450811743736267, + "learning_rate": 5.7328582038386483e-08, + "loss": 0.4158, + "step": 823 + }, + { + "epoch": 5.640909090909091, + "grad_norm": 0.48033422231674194, + "learning_rate": 5.519356901107359e-08, + "loss": 0.4087, + "step": 824 + }, + { + "epoch": 5.6477272727272725, + "grad_norm": 0.4670649766921997, + "learning_rate": 5.309862606186877e-08, + "loss": 0.4125, + "step": 825 + }, + { + "epoch": 5.654545454545454, + "grad_norm": 0.4971778690814972, + "learning_rate": 5.104378752663008e-08, + "loss": 0.4055, + "step": 826 + }, + { + "epoch": 5.661363636363636, + "grad_norm": 0.5656769275665283, + "learning_rate": 4.902908708391024e-08, + "loss": 0.4184, + "step": 827 + }, + { + "epoch": 5.668181818181818, + "grad_norm": 0.5054930448532104, + "learning_rate": 4.705455775440237e-08, + "loss": 0.4074, + "step": 828 + }, + { + "epoch": 5.675, + "grad_norm": 1.0643500089645386, + "learning_rate": 4.5120231900400715e-08, + "loss": 0.4471, + "step": 829 + }, + { + "epoch": 5.681818181818182, + "grad_norm": 0.6868470907211304, + "learning_rate": 4.3226141225268804e-08, + "loss": 0.4358, + "step": 830 + }, + { + "epoch": 5.6886363636363635, + "grad_norm": 0.48121461272239685, + "learning_rate": 4.1372316772921584e-08, + "loss": 0.4272, + "step": 831 + }, + { + "epoch": 5.695454545454545, + "grad_norm": 0.5572350025177002, + "learning_rate": 3.955878892731441e-08, + "loss": 0.4086, + "step": 832 + }, + { + "epoch": 5.702272727272727, + "grad_norm": 0.4539232850074768, + "learning_rate": 3.778558741194677e-08, + "loss": 0.3917, + "step": 833 + }, + { + "epoch": 5.709090909090909, + "grad_norm": 0.600853443145752, + "learning_rate": 3.605274128937464e-08, + "loss": 0.4223, + "step": 834 + }, + { + "epoch": 5.715909090909091, + "grad_norm": 0.514529287815094, + "learning_rate": 3.436027896073307e-08, + "loss": 0.4104, + "step": 835 + }, + { + "epoch": 5.722727272727273, + "grad_norm": 0.4706551730632782, + "learning_rate": 3.270822816527325e-08, + "loss": 0.4301, + "step": 836 + }, + { + "epoch": 5.7295454545454545, + "grad_norm": 0.4552484154701233, + "learning_rate": 3.109661597990532e-08, + "loss": 0.4121, + "step": 837 + }, + { + "epoch": 5.736363636363636, + "grad_norm": 0.4661986827850342, + "learning_rate": 2.9525468818755455e-08, + "loss": 0.4026, + "step": 838 + }, + { + "epoch": 5.743181818181818, + "grad_norm": 0.5309687256813049, + "learning_rate": 2.7994812432733664e-08, + "loss": 0.43, + "step": 839 + }, + { + "epoch": 5.75, + "grad_norm": 0.5013267397880554, + "learning_rate": 2.6504671909109993e-08, + "loss": 0.417, + "step": 840 + }, + { + "epoch": 5.756818181818182, + "grad_norm": 0.5112711191177368, + "learning_rate": 2.5055071671105936e-08, + "loss": 0.4127, + "step": 841 + }, + { + "epoch": 5.763636363636364, + "grad_norm": 0.5376843214035034, + "learning_rate": 2.3646035477491726e-08, + "loss": 0.437, + "step": 842 + }, + { + "epoch": 5.7704545454545455, + "grad_norm": 0.5034119486808777, + "learning_rate": 2.2277586422198017e-08, + "loss": 0.4045, + "step": 843 + }, + { + "epoch": 5.777272727272727, + "grad_norm": 0.594643235206604, + "learning_rate": 2.094974693393731e-08, + "loss": 0.4518, + "step": 844 + }, + { + "epoch": 5.784090909090909, + "grad_norm": 0.48406273126602173, + "learning_rate": 1.9662538775836182e-08, + "loss": 0.4301, + "step": 845 + }, + { + "epoch": 5.790909090909091, + "grad_norm": 0.47127723693847656, + "learning_rate": 1.841598304507891e-08, + "loss": 0.4284, + "step": 846 + }, + { + "epoch": 5.797727272727273, + "grad_norm": 0.5004918575286865, + "learning_rate": 1.7210100172561084e-08, + "loss": 0.4282, + "step": 847 + }, + { + "epoch": 5.804545454545455, + "grad_norm": 0.48565441370010376, + "learning_rate": 1.6044909922555973e-08, + "loss": 0.4096, + "step": 848 + }, + { + "epoch": 5.8113636363636365, + "grad_norm": 0.5008643269538879, + "learning_rate": 1.4920431392388412e-08, + "loss": 0.4178, + "step": 849 + }, + { + "epoch": 5.818181818181818, + "grad_norm": 0.5197197794914246, + "learning_rate": 1.383668301212393e-08, + "loss": 0.4232, + "step": 850 + }, + { + "epoch": 5.825, + "grad_norm": 0.4915326237678528, + "learning_rate": 1.2793682544266216e-08, + "loss": 0.4116, + "step": 851 + }, + { + "epoch": 5.831818181818182, + "grad_norm": 0.46172675490379333, + "learning_rate": 1.1791447083465136e-08, + "loss": 0.4407, + "step": 852 + }, + { + "epoch": 5.838636363636364, + "grad_norm": 0.6167759299278259, + "learning_rate": 1.0829993056236942e-08, + "loss": 0.4113, + "step": 853 + }, + { + "epoch": 5.845454545454546, + "grad_norm": 0.4805396795272827, + "learning_rate": 9.90933622069562e-09, + "loss": 0.4293, + "step": 854 + }, + { + "epoch": 5.8522727272727275, + "grad_norm": 0.6547632813453674, + "learning_rate": 9.029491666293911e-09, + "loss": 0.3919, + "step": 855 + }, + { + "epoch": 5.859090909090909, + "grad_norm": 0.4851185977458954, + "learning_rate": 8.190473813576571e-09, + "loss": 0.4148, + "step": 856 + }, + { + "epoch": 5.865909090909091, + "grad_norm": 0.47095951437950134, + "learning_rate": 7.3922964139433455e-09, + "loss": 0.4234, + "step": 857 + }, + { + "epoch": 5.872727272727273, + "grad_norm": 0.4624910354614258, + "learning_rate": 6.634972549423857e-09, + "loss": 0.4479, + "step": 858 + }, + { + "epoch": 5.879545454545455, + "grad_norm": 0.492658793926239, + "learning_rate": 5.918514632463901e-09, + "loss": 0.4471, + "step": 859 + }, + { + "epoch": 5.886363636363637, + "grad_norm": 0.5247507691383362, + "learning_rate": 5.242934405720879e-09, + "loss": 0.4438, + "step": 860 + }, + { + "epoch": 5.8931818181818185, + "grad_norm": 0.512353241443634, + "learning_rate": 4.6082429418720095e-09, + "loss": 0.3983, + "step": 861 + }, + { + "epoch": 5.9, + "grad_norm": 0.5170801877975464, + "learning_rate": 4.01445064343281e-09, + "loss": 0.4168, + "step": 862 + }, + { + "epoch": 5.906818181818182, + "grad_norm": 0.7755751013755798, + "learning_rate": 3.4615672425861167e-09, + "loss": 0.4051, + "step": 863 + }, + { + "epoch": 5.913636363636364, + "grad_norm": 0.4807862937450409, + "learning_rate": 2.9496018010233275e-09, + "loss": 0.4054, + "step": 864 + }, + { + "epoch": 5.920454545454545, + "grad_norm": 0.5142385959625244, + "learning_rate": 2.478562709795074e-09, + "loss": 0.4296, + "step": 865 + }, + { + "epoch": 5.927272727272728, + "grad_norm": 0.6004032492637634, + "learning_rate": 2.048457689174943e-09, + "loss": 0.4272, + "step": 866 + }, + { + "epoch": 5.934090909090909, + "grad_norm": 0.5010991096496582, + "learning_rate": 1.6592937885312466e-09, + "loss": 0.4082, + "step": 867 + }, + { + "epoch": 5.940909090909091, + "grad_norm": 0.45423299074172974, + "learning_rate": 1.3110773862126669e-09, + "loss": 0.4345, + "step": 868 + }, + { + "epoch": 5.947727272727272, + "grad_norm": 0.5379949808120728, + "learning_rate": 1.0038141894436192e-09, + "loss": 0.4303, + "step": 869 + }, + { + "epoch": 5.954545454545455, + "grad_norm": 0.535897433757782, + "learning_rate": 7.375092342298828e-10, + "loss": 0.421, + "step": 870 + }, + { + "epoch": 5.961363636363636, + "grad_norm": 0.5135968923568726, + "learning_rate": 5.121668852775541e-10, + "loss": 0.4105, + "step": 871 + }, + { + "epoch": 5.968181818181818, + "grad_norm": 0.45730048418045044, + "learning_rate": 3.277908359194948e-10, + "loss": 0.4371, + "step": 872 + }, + { + "epoch": 5.975, + "grad_norm": 0.4870580732822418, + "learning_rate": 1.8438410805732277e-10, + "loss": 0.4138, + "step": 873 + }, + { + "epoch": 5.9818181818181815, + "grad_norm": 0.48615431785583496, + "learning_rate": 8.194905210923143e-11, + "loss": 0.4151, + "step": 874 + }, + { + "epoch": 5.988636363636363, + "grad_norm": 0.4868053197860718, + "learning_rate": 2.0487346973629975e-11, + "loss": 0.4373, + "step": 875 + }, + { + "epoch": 5.995454545454545, + "grad_norm": 0.5503740906715393, + "learning_rate": 0.0, + "loss": 0.394, + "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..a03cebdd4a04fb2e9443a05c92f15b2b2db01e10 --- /dev/null +++ b/checkpoint-876/training_args.bin @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:90038f7ff377ce29f00aedab533b18bfc9515b30d2b4232d8b0d6de26ec3ea56 +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..dd4e5ad4ad4c9bafb071dccdbf016fd2b2c567d4 --- /dev/null +++ b/config.json @@ -0,0 +1,52 @@ +{ + "_attn_implementation_autoset": true, + "_name_or_path": "nvidia/Llama-3.1-Nemotron-70B-Instruct-HF", + "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..fdde94c29816839ec3c29d6c6461206a49911f3c --- /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\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" +}